Dispense Di Fisica Moderna

Anno 2010-2011 FISICA MODERNA - ELENCO FOTOCOPIE TESTI BARROW “Chimica fisica” Cap. 2 - teoria cinetica ENGE-WEHR-RICHARDS “Introduction to Atomic Physics” Cap. | - the atomic view of matter Cap. 2 - the atomic view of electricity Cap. 3 - the atomic view of radiation Cap. 4 - the atomic models of Rutherford and Bohr Cap. 5 - waves and particles DEKKER - “Solid State Physics” Cap. 2 — the specific heat of solids and lattice vibrations RICHTMYER-KENNARD-COOPER “Modern Physics” Cap. 5 - the origin of quantum theory Cap. 7 - X-rays EISBERG-RESNICK “Quantum Physiscs” Cap. 4 — Bohr’s model of the atom. ALONSO - FINN “III-Quantum and statistical physics” Cap. 3 — atoms with one electron (solo cenni, vedere quanto fatto sugli appunti) Cap. 10 - classical statistical mechanics Cap, 12 - thermal properties of gasesSupponeie che venga concordato che i punto di conglamento normale delTacgua abbia una tee ter 100 ino ain lta. ose ese alas I pert soma elie 31 Una on unpre porn sont #053 dt fo Sd ae ete cme se es i lnciiontetandos io, © ape di Cole Uist sans Be mente medeine a Nel 1846 Graham rigor i dati segvent peri tempo di efusione di vas as rspeto ala: Oo: lenge Wat sf accordano ala legge esiometic di Graham? 20, Tas In che misra que Bibliografia Newitt RG, “Te Dizovery of Boy's Law" Chom. Ee am Tale stulal deps pene, presion sche & Boje 8 902 be Sees, pa. ome Piao he Bau Min, p. 2. Peton Univeriy rt Pion ae eae i al ee pe a aire} SA Ade Tet om Pha Cons, op.) meg, Longman ‘£0 Ud Lonea, 99" Un sompario mol normal tt somnita iegt ned tes enn freak infra! su Bape dl ea Molt eos Pace uae eyo, Pon fe Sado of Sonn Beri pt Mipourtos, W., Knows E, 4 History of the Thermometer The Johor Hopes Pas Bae {Si is neg fe ata Gop pe pi oat et temmoaut ete ea dae Ha Ja, Darton, Ppa, ts Meanrenen od ovr Sc ; shin Goorin, Now York et, wo wa 4 BS Use eos ete ces ee it 6, Rastas Ms Therma, 2 od K.. we 2 fv dK. $. Pas €L Breve 3, Mere ak mm Ne Veni ens pe Spe i ntratnt Crit Tate op 9 icraw fil ook Company, New Yt, 12 Une acs al comporamente PY 8a gan (rr J.B, Coan). Ry Hatem, HT Compartons of Egptins of late in finely Desert ‘BET Reiatins Chom. Bac 4831S (ST) em tS in ety ering ~ Capitolo 2 > SARLOR Teoria cinetica-molecolare dei gas 2.1, Modello cinetico-molecolare dei gas [Nel Capitolo 1 lo studio del comportamento fisico dei gas # stato affrontato dal punto di ta empirico senza preoceuparsi di rispondere a semplici domande del tipo: perché un ies segue la legge di Boyle? Perché obbedisce alla lege di Gay Lussac’ Perché possede una erta viseositd? E cosi via, In questo captolo si cercher, invece, di approfondire la natura ei gts, ¢ diventerd facile rispondere a domande simil La teoria cinetica-molesolare, comungue, non viene introdotta soltanto per fornire una azione delle lege dei ge il motivo principale per cui @ stata sviluppata& di mettre in evidenza, di defnire, gli aspetti quantitativi del mondo molecolare. La natura dei gee non é direttamente deducibile dalle propreta che vengono misurate. B necessario servisi di questi dati in modo indiretto. Di norma, si procede formulando delle {potes sulle caraterstiche intrinseche dei gas e po sulle basi di queste, si deducono Je propriet fiche. {La validité delle ipotesifatte viene verficata confrontando le proprieté dedottee quelle mi- ‘sate, Un insieme di definiion! di carattristche intrinseche dei gas eosttuisce un cosid- detto modelo Leleggi le proprietA dei gas desrite nel Capitolo | possono essere studiate adottando come ‘modello, ad evempio, un gas costituito da un grande numero di piccole particele,chiamate rmolecole, che si muovono caoticamente urtandosi fra loro e urtando le pareti del recipiente ‘Non @ilcaso di riportare qui le nozionichimiche che stanno alla base del concetto di molecole. Sari suffcientericordare che dea che un composto chimicofosse costiuito da atom hha subito una lenta evoluzione ed & stata Universalmente accettta soltanto nel 1800, insleme ai concetti di atomo © molecola, L'applicazione della teoria atomistica in eampo chimico HE dimostrata molto valida per spiegare la composiione e la trasformazione di sostanze himiche; non ba foraito, invece informazioni sulla misura, la forma e le proprieth delle singole molecole. La sola scquisizione del concetto di molecola,tuttavia, &servta da base per lo studio det comportamento dei gas; mediante V'applicazione della cosiddetta teoria cinetica-molecolare dei gas si sono raccolte numerose informazioni circa le propriet delle singole mole~ cole. Boltzmann, Maxwell e Clausius nel tardo "800, diedero con il loro lavoro, il massimo contributo allo sviluppo di questa teoria. ‘Mediante la teoria ciuetica-molecolare verranno desunte, in questo capitolo, alcune proprieth dei gas. In taluni casi a teoria verrd esposta in forma semplifiats, per non appesan Tin le eatatterstche con operazioni matematiche troppo complesse. In un secondo tempo, ‘erranno approfonditealcane parti semplificate di tale teoria per completare Ianalisidetta- slat del moto delle molecole in un g Timodellocinetic-molecolare di un gas pué essere espresso postulando le seguenticondizioni: 1) Un ges é costituito da un grande numero di particlle 0 molecote assai piocole rispetto alla Toro distenza e alle dimensioni del recipient. 2) Le molecole sono soggette ad un movimento eaotico continuo, assolutameate casuale, 53) Gli wrt fra Je molecole e contro le pareti del reipiente sono perfettamente elastci © alVatto della colisjone non vi ¢ aleuna perdita di energia traslazionale per conversione in cnergia. interna. La prime cosa de fare, utiizzando questo modell, 2 di stabilre se esso consenta 0 meno di rieavare le proprctd osservabili nei g2sIn base al modelo cinetcomolcalare& posible cleolare Is presone eer da. N roecoledimasuam content nun eipentecubico di lato Gut dele moecle com {ole paret dl repent sono responsbil dell presion cel as exerci veso Testerna Per mantenere, quid, un volume catane io equiv si ova opporte una pressions gue, ma opposia, deta ves interno, che pad eters, af esempl,seapliceeecte la resitnza offer dalle pre del reipeote Consideamo, per comiacare, una tla delle molecol.Indichamo ta sua veloc € Girezione con'w,seito ia graselo per contraddsinguere una grandezavetoile. I Simbotow indicherd, vce, i veloc scalare Il etre w pod eer decomposto elle sue Somponent vw ¢u eke son0 perpendicola alle pare del respeate Come sede la Fig. 21. (Note chew, ad expo, id nformaziou sul dietionepotendo ever pos ‘vo o negativo. Questo compara un moto pesto o egitivo sla drezone dele. Daa parte, ebe € lo scale de vttor w,pud asumere solo valor post). Consiesam ore, efeto dela compooente di vlad lngo Vase dele, ud una mole cola, Come ristato tal eet, ln moleola uted prima uo dle pass leet, perpendicoareallase dle x, e dopo aver rimbslato urter, di conseguega, In pare Sposa L'inseme dk et i, secon lta cineca molest prods spree as 1s fora eersitatn dato dunn moleola sll parce di un recipient si pub cleolre bilizzano i secods lege de moto di Newton, Gur lege aera che i oa ok Gui arbi a quant doco una parcel, che muove in una cera deine Uguale ‘la forea agente ula parte, nel stesa dvecoae. La quant di moto ooo ila pts tet savin ala pce 4 di Fig 21. mi, Dopo orto molesoas muove di auovo on, et immu, ma conus valor i wre Gund anche dma, boro oppest, coms trata evidente dala Fig 22. Cod la pares opponeaimptt Una fora che determing ‘un cambiamento della quanta mote pas a 2a, in dieson prpendicelare ala parce gw 2. Rappresenaione dale coordinate dl velcsh Input pr In dersione ela pression clot. Fie 22. inveso a come state of aetna matt whens ‘numero di uit per secondo che si ‘percor rono in 1 suna distanza pari a, poiché la dstanza percorsa fra una colisione ela succes siva sulla parete A 82 il numero degli urti per secondo su A @u,2 1. La velceité con cui cambis il momento, cid la variazione di momento per secondo & perci: Mee i Irma, Sem Fy tu (Questa @ Ia forza che la parete 4 oppone all'urto di una molecola.Poiché la forza per unit di area rappresenta Ie pressione, la pressione reltiva alla parete 4 @ data di 2am) ey dove V=P di volume del recipient. Si pud ora ammettere che la pressione sia identica per tutte le pare e agisca su queste perpendicolarmente. Possiamo cosi fare a meno delle festrzioni legate alla parete A e a specifche direzion,eliminando la notazione vettoriale. ‘Avremo pertanto: eae 8) Consideriamo ora invece, che nel recipiente cubico vi siano N molecole. Secondo il nostro rmodello.queste molecole agiscono l'una indipendentemente dallaltra portando ciascuna tun eontributo alla pressione secondo un termine simile all’Eq. (3]. (Se esse perd interagiscono, dando luogo uri occasionali fra loro, il fatto che la quanti di moto, nella sua com- Ponente drezionale, debba essere conservato, Comporta un risultato ideatico a quello che Fi avredbe se le partcelle fossero completamente indipendenti). La somma dei contributi i pressione dele N molecole si pud indieare come: ray “ Possiamo considerare il cato in cui tute le molecole abbiano la stessa massa. II modello da ‘oi assunto prevede che le molecole si muovano in tutte Ie direioni con velocia diverse © cosi uw, e af saranno diversi per ciascuna molecola. Ses indica il valor medio di ut per le N molecole di gas con u} — si pud sviluppare Eq. [4 allo, seopo di definite la presione per mantenere N molecole di gxs confinateentro il volume ¥, nel modo seguente PFE Nr (3) Una relazione fr la pressione, dovuta alle N molecole,e la velocitd media delle molecole é, comungue, pid conveniente La relazione fra il quadrato delle componenti della velociti, precedenteniente vista, ¢ il uadrato delia velocitd medesima @ expressa dalla seguente relazione: 2 Nat wand tub yed ol run grande numero dimolecole che si muovono in ogni dtezione ia modo del tutto casuale dig it 7 « perso ase (8) (i not che w? e sono dele grandezze sealari,cioé non vettoriali, che non indicano, pertanto, a direzione dele veloc moleeolar, ma solo il logo modula). La sostuzione del Eq, [8] nell Eq, [5] da la relazione desiderata fra P e u*; cioé Ys Na z Pat oppure PY = Nit (9)‘questa importante equazione rappresenta esattamente quanto volevamo ottenere, ciot spigare il signiicato della pressione di un gas sulla base dei postulati del modell cine ‘Uco-molecolare del paragrafo 2.1. Questa relaziove, per ora, non pud essere confrontata con J legl empiriche dei gas, ma nel prossimo paragrafo verranno esposte delle altre considerazioni ebe consentiranno un con- fronto del genere 2.3, Energia cinetica e temperatura ‘Spesso, come nel presente caso, & pit conveniente impiegare Venergiacinetica invece della velocti delle molecole, L'energia cinetica media di una molecola di un gas viene indiesta ccon Ke. Questa grandezza @ legata al quadrato della velocita molecolare dellespressione apm 110) Leequszione {9} pud alloraesere sviluppata dopo averia posta nella forma: PV = 28 NU mid) a pv 3 nie om) Le relazioni empiriche del Capitolo 1 5 riferivano a quantita molasi di gas. Le relazioni pre Senti,invece, si riferiscooo, come ad esempio, nel caso Ke, alle singole molecole. Per con siderare un insieme di molecole & necessaio ricorrere al numero di Avagadro che & i unero oi molcne contenu in una mol © 7 i numero di molezle, em quello dei ven 02) Lequazione (11] pub ora essere serita Prada) 13) Se introduciemo, inoltre, il termine KE per indicare energia cinetca di un numero di molecole pari a quelio di Avogadro, si pud riscrivere I'Eq, (I1] nel modo segueate: KE 4) ‘A questo punto & bene ricordare Iespressione empirica RT « confrontare questa espressione con 1'Eq. [14] ricavata mediante la teoria cinetice-mole- colsre. La diserepanza apparente fra Je due relazioni non present® alcun problems neppure dal punto di vista storico. Calore e temperatura, infati, furono spesso associate, sia pure ualtativamente, all'des di agitazione e moto delle partcelle io un materiale ealdo. Era percidgivstifcato far collimareIespressioneteoricacon la legge sperimentale ponendo 2/3 KE vuguale @ RT, o: KE. 3 Bar us) Pertanto se energia traslarionale di un numero di molecole pari al numero di Avogadro, cio di | mote di gas, ha il valore di 3/2 RT, allora Ia legge dei gas, riassunta nella PY = n RT 1 pud deducre anche dai postulats dlia'teoria clnetice-molecolare Verranno ora rieavate aleune dele relazioni empiriche del Capitolo |. L'espressone PY = '=2/3 nKE, insieme con il postulato KE = 3/2 RT, permettono di ricavare, ad esemy la i Gay Lussac. Inoltre, a patto che i postulati della tori cinetica- nerewatiete ‘molecolare e da ogni altra propriciacaralteristica delle molecols. Questa espresione, Dud valee per dve gas diversi, alla stesa temperatura e pressione, sltanto se volumi ugual Gi quests gas contengono lo stesso numero di moli o molecole. Viene cosi dedota ¢ confer. mata T'potesi di Avogedro. oe La legge ai Dalton, inoltre, & una conseguenza diretta dei postulatioriginali poiché in que fii si affermava che le molecole non interagiscono ¢ occupano un volume trasvurabile. In {ale caso le molecole di un certo tipo di gas non avranno alcun effetto su quelle di un altro fas, Per una miscela di gas, perianto, sevondo la teoria cinetica-moleolare, Ia pressione folsle P dove essere la somma delle presioni di ciascun componente, cit: PaPi+Py 2.4. Valori numerici per le energie ¢ per le velocita molecolari In questo paragrafo si cercherd di definite aloune proprieta dele molecle che costituiscono ‘un gas, in particolae la veloc e Tenergia. E stato pia messo in evidenza che i postulati (ualitaivi dela teoria cinetiea molecolare descrivono in modo sulficientemente accurato il fhondo molecolare e consentono di ricavare le lepgi dei gas ideal Usa informazione quantitativa,invece, deriva dal fatto che Tenergia cinetica di wn numero i moleeole pari al numero di Avogadro deve essere 3/2 RT. S-valore i R, dato nel paragrafo 1.4, pari a 8,3143 J grado"! mole” permette poi di rica ‘are il contributo energia peril moto trasazionale di 1 mole di un qualsiasi gas ideale a 28°C: ar=3.639 0992205 mate" rgia cinetiea media di una molecola di gas come: 16) Poiché buona parte della nostra successva esposizione riguarda le energie dei singoli atomi cc molscole,& biileintrodurre ua nuova costate chiamata costante di Boltzmans, cos! efit kek, a = 1,3806 x 10-893 grado a ‘La costante di Boltzmann si pud allora considerare, la costante dei gas riferita alle singole imolecole, mentre Penergia cinetica media per molecola, si pud indicare co K us} 3 far che a 25°C ha un valore mumerico pari a: Be = (1,380 x 10°) 298) = 4.11 x 1077 F B cific allo stato attuale apprezzare quale importanza abbiano questi valor delenergia Cineica; o® diventera pid chiaro in seguito. B opportuno quindi, per ora rifrist a delle propria molecolari dl pil immediate comprensione, ad etempio, la velocitl eon cui Te olecole sl muovono, - Leenerpia cineticn di un mumero di molecole pari al numero di Avogadro si pud serivere: (19) fame cnn ti relation al netacolare u! = 3 RT/M oppur BRE Vim PP on I termine //W7, dal significato non facilmente intubile, rappresenta la radice quadrata della veloctd quadratica media, la sua determinazione implica che tutte Je velocita molecolari vengano elevate al quadrato, che se ne facea la media einfine si valuta radice quadrata del valor medio. Ci8 consente di ottenere una grandezza che &diversa da une semplice \elocité media. La dlferenza da essa tutavia, come vedremo nel paragrafo 2.6, €limitata ern, alor i verano onder come inser dle veo [Nel easo di N; a 25°C, per esempio, M ha in unit SI il valore 0,02802 kg e quindi avremo: __ PSsaTERTD Ue 0,02802 <5i5me = 1150 mph 21) La Tab, 2.1. riporta i dati relatvi ad alte semplici molecole PPoiché 'energiacinetica, a una certa temperatura, la stessa per tute le moleeoe, indipendentemente dalla massa, le molecole leggere avraano velocita pit grandi delle molecole pesanti, Si pud notare inolte, che come per le velocité molecolar, che possono essere definite mediante le loro comporenti nelle te dizezioni perpendicolai, cod u! mid +1? tu lun uguale procedimento pud essere applicato al valor medio delle energie cinetiche medi Li ed ik +h ik + brah fim ik + ih + 4 id oppure en. + ©, +, (2) Poiché energa di queste componenti medie & ugusle, ¢ vale KE= 3/247 we segue ©, =.= B~ 1, = G,= ym her (23) Le tre direion pependicolc in cui sono state scompote le velocth el eerie veagon0 chiamate gradi di libertd. Possiamo cosi affermare che l'energia traslazionale media dana ‘molcol per grado a ibertd¢ V2 KT. Vedet in sega come questa affemacione ia at fxmente applicable ¢ sia una gvids importante allo studio delle enerpe moet ‘Tabella 2.1. Velocith medie di molecole di gas fuguale a 0,921 VF a 25 °C (298 K) ¢ 1000°C (1273 K)}. 25 F 100 -e Gas as ‘aph() ae mph) a 170 3860, 2660 180 He 1260 220 20 5830 #0 30 1320 m0 70 ™ 40 1060 370 2190 o “0 350 310 2050 ©, x0 0 70 no. a, soo a0 an 1300 Ht x 00 4“ 110 He 180 400 370 80 1 spt ta 2.9. Vistribuzione di velocita Moiecolari Lugo una aimensione ‘Avendo considerato ¢ tabulato alcune veloctl medie molecolari& ora opportuno iavesti- {gare pit 2 fondo sui valor specific! delle veloctd molecolari da cui derivano i valori medi BA trovatl Es distrbusione di Boltzmonn & la relazione fondamentale per trattare question: relative al numero di molecole che abbiano velocit ed energie diverse. La deduzione di questa impor- fante relazione viene dffeita a dopo aver studiato le regole quantiche, che come si vedr, fs the smallest particles that can combine to form compounds, and the existence E(ioleeOTes of ai eementhand “integral molecules SSC SL Seas al can exist in GRE SRE HENGITONTS SATE Toihout proof) avery important generaliza- ToACERoWirag Avogacto’s law, that af he same femperature and pressure eqval toliones of all paver conta ie aie Rune of MioeciTes. FOR Ts Taw and is concepts of toms and molecules, Avogadro showed that the ammonia- Froducing reaction required that nitrogen gas consist of diatomic molecules and that oxygen must als be diatornc to account for the water-vapor reaction. He further concluded thal water must consist of @ union of two atoms of hydrogen and one atom of oxygen. Unfortunately. the ideas advocated by Avogadro received little notice even when revived by Ampere in 1814, ‘The notion that hydrogen and other gases ere composed oF diatomic molecules was ridiculed by Dalton and others, who vould rot conceive of a combination of atoms ofthe same kind. They asked. “it two hydrogen atoms in a container filled with this gn» can cffmg together.ety not allingtopetheranieondenselaccin This is indeed a very Food Guestion. Science was not able to give a satisfactory answer until over a century later. (See the dscustion ofthe covalent bond in Section 11-2) In the next two sections inthis chapter we wil describe some of the methods which were and stil are used (0 determine the relative masses of atoms. The results obtained by the analytical chemists using these several methods during the first half of the nineteenth century werc often contradictory. They frequently obtained different values for the atomic mass of the same element. By the 1850's inconsistencies were so numerous that many felt thatthe atomic theory of matter would have to be discarded. However, the contradictions were resolved in 1858 by the Italian chemist Cannizzagp, who had an intimate know tage of the then known methods for determining atomie masses and a broad grasp of the whole field of chemistry. He showed that Avogadro really had provided a rational basis for finding atomiie masses, and that the inconsistent results obtained by various experimenters resulted from’ a Tack of clear diss tinction between atomic masses, equivalent masses, and molecilar masses, The” ‘WieWs OF Canmiazaro Teceived the approval of the scientific world when they were adopted by the international conference on atomic masses which met in Karl ruhe, Germany. in 1869. This, then, isthe year in which the Fundamental ideas of modern chemistry were widely accepted 1-3 MOLECULAR masses ‘After Cannizzaro had clarified and established some of the basic definitions in chemisizy. Avogadro's law opened the door to one of the methods for determining molecular masses. No one had any idea of what the mass of a single molecule was, but once there was a way of isolating equal numbers of diferent Kinds of molecules, THE cou bE deterinined, The imolecule was found to be the lightest molecule, and the hydrogen atom proved to be the lightest atom. In 1815 Proust had proposed thal the. relative atomic tras of hytrogen be ably BES ae Fae and noses Ba SaTE maHES ch were nearly imtegers But, for reasons to be discussed later, it turned out that the atomic masses of many oT~ “the Heavier aOR Tlegers. HVRTORET appERTEN TS be a poor basis for the system, and more nearly ttegral atomic masses for all atoms could be obtained by making a Reaver alam The Basico The system. —For many years physicists and chemists used two different systems, both Sraverage mass value of 16, In T9GT ihe (wo sales were replaced Gy 8 cortmon EGRET SEN ich the mass of the carbon sotone C6 pro 6 neutrons) is sel (o_12._ On this scale, the atomic mass of hydrogen is not "EST ay Sas nea TRE Felatve molecular and atomic masses are all dimeusigaless aon, IF about four pasts by weight of hydrogen were combined with £2 parts by weight of oxygen, about 36 parts by weight of water vapor can be formed, according to the familiar equation, 2H, + 0, = 2H,0. Chemists as well as physicists measure amounts of material in grams or in moles. AgGGTD¥s Uefined as the amount ofa given substance containing the same number of molecules as the number of atoms in 12 grams of '?C. One mole of PC is 12 ‘one mole of carbon in its natural mixture of isotopes is TEDITIS grams, one mole of HzO is 18.0153 grams, etc. ‘The number of atoms in 12 grams of !2C (number of molecules ja amole) is called avogadro’s num onthe Arogadro con: iemportance in physics and phyweal a fate that (he mole dnd the Avogadro constant have been defined in terms of 12 grams of '#C, This is the modern practice. However, these definitions do not always conform to the MKSA system of units." Therefore, we will con occasion use the kilomole, kmole, which is 1000 moles. Obviously a kilomole of carbon 12 has 9 mass of 12 kilograms.) ‘The value of the Avogadro constant was of relatively minor importance to chemistry in the carly nineteenth century and its magnitude was not even estimated until Loschmidt did so in 1865. We will discuss Perrin’s method of determining it later in this chapter. Here is an interesting case where knowing the existence of a number was more important than knowing its magnitude as. for example, in determining the relative masses of the atoms involved in the ammonia-producing hydrogen-nitrogen reaction previously described, The value of the Avogadro constant is by modern measurements Ng = (6.02217 + 0.00004) x 107° particles per mole, Only after the magnitude of the Avogadro constant was known could the absolute mass of an atomic particle be computed. Since MeltC atoms of Fa Gass OEY TE Ba The as (in grams) of one unit is clearly Lu = 1/N, = (1.66054 + 0.00001) x 107 It follows from Avogadro's law that the volume of a mole of a gas is the same for all gases. The normal volume of a perfect gas or the standard molar + For students not familiar with the MKSA system, a brie? review is given in Appendix 18 The atomic view of matter | 1-4 volume of an ideal gas, Vo, is the volume occupied by a mole of the gas at a pressure of | standard atmosphere and-a temperature of °C. The value of Vas 2.24136 + 000030" x 10°F! perl. (S24 2) 1-4 ATOMIC Masses ‘Avogadro's law provided a systematic method for determining molecular masses. but a large amount of quantitative data on the formation of various compounds were required before the atomic masses of the known elements could be determined. The situation is somewhat like the following: Suppose that man 4 pays man B $1.00 in coin, using no coin smaller than quarters, and that we wish to determine how this is done. He may do this in any one of four ways: a) one $1,00 coin, ') two 50¢ coins, ‘) one S0y and two 25¢ coins, 4d) four 25¢ coins. If man B now pays man C 25. possibilitcs (a) and (b) are eliminated, but there is still a doubt as to how the original transaction was made. By careful observation of further transactions of those who spend the original $1.00, it could be determined just what coins A must have had originally ‘An aid to the solution of this puzzle was the empisical discovery by Dulong and Petit, in 1819. that for most elements in the solid state the specific heat per ‘mole at constant volume is about. ieTaw of Dulong and Pefinlis @ rough independent determination of atomic masses by dividing this constant by the specific heat measured in calories per gram. We shall discuss the theoretical basis of this law in Section 3-15. The masses of atoms are now measured on the '#C scale in mass spectrom= eters with an almost fantastic precision. (Uncertainties of about one part per nillion are not uncommon.) See Section 2-8, ‘+ TFitseems strange that this and some other constants are given withthe uncertainty expressed to more than one significant figure, refer to the article, “Probable Values of the General Physical Constants,” by R. T. Birge, (Phys. Rec. Suppl. 1 (1929) p. 6) ‘We note here only that the concepts of probable errors and significant figures do not correspond completely. If the probable error can be determined to less than 10 percent of itself, then more than one significant Figure is required to express it Principal articles containing the values of various constants are: “Values for the Physical Constants Recommended by NAS-NRC," Nat. Bur. Std, (U.S.) Tech, News Bull. 47 (1963), p- 175: "World Sets Atomic Definition of Time,” Nat, Bur. Std. (U.S.) Tech. News Bull. 48 (1964), p. 209; Mechtly, E. A., The Jazernational System of Units, NASA SP-7012. Washington, D.C.: U.S. Government Printing Office (1964); Cohen, E.R, and J. W. M. Du Mond, “Our Knowledge of the Fundamental Constants of Physics and Chemisiry in 1965," Rev. Mod. Phys. 37 (1968), p. 537 1-6 | Physical evidence for the atomic view of matter 9 1-8 PERIODIC TABLE Probably the most significant discovery in all chemistry. aside from the atomic nature of matter, was the periodic properties of the elements, now depicted in the familiar periodic table of the elements (Appendix 2). The chemical properties of this (able are probably familiar to most readers of this book: the physical properties will be discussed later. The table was proposed independently by Meyer and by Mendeldev in 1869. Its usefulness lay both in its regularities and in its irregularities. One interesting irregularity in the original table was that in order to have the elements fall in positions consistent with their chemical propertiss, it was necessary to leave numerous spaces unoccupied, Mendeléev suggested that these spaces would be filled with as yet undiscovered elements. Using his table, he was able to describe in considerable detail the properties these elements could be expected to have when they were discovered. It was nearly one hundred years before all the predictions that Mendeléev made were fulfilled Reflect, for a moment, on the vast simplification that the chemical dis- coveries here outlined provide. Looking about us, we sce innumerable kinds of materials, The atomic view indicates that these materials are of discrete kinds whose number, however large, is not uncountable. The discovery of elements is further simplification in that the many materials we encounter are shown to be composed of only about one hundred chemically distinguishable materials, many of which arc rare, It turns out that even these clements are not a heterog~ encous group but are subject to further classification into a periodic table, The problems of chemistry are many; however, itis easy to see that things are much simpler chan might at frst appear. 1-6 PHYSICAL EVIDENCE FOR THE ATOMIC VIEW OF MATTER In our discussion thus far, all atomic properties have been inferred from studies fof gross matter. In 1827 the English botanist Robert Brown observed that ricroscopie polien a er-appearto-dinceaboat mn ramdOm™ (ESM, AT fist the phenomenon was ascribed to the motions of living inatter In time, however, it was found that any kind of fine particles.suspendcal ina. quid performed such. nernetual-danse,- Eventually it was realized that the imoleoules of @Tiquid are in constant motion and that the suspended particles recoiled (Brownian movement), when hit by the molecules of theliquid. However, long before the equations for Brownian movement were lrived early in the ‘twentieth century, the particles of matter were thought of as moving about in a random manner and undergoing frequent collisions. Such processes are decidedly in the domain of physics, How can the principles of mechanics be applied to molecular collisions? ‘The simplest state of matter to consider was a gas. The ideal gas law. for nn moles of 1 gts pl = nRT. where Ris the universal GATT pet Fle10. The atomic view of matter 1 1-7 and p, V, and T-are the pressure, volume, and temperature, respectively. This law Waa well-established enipirical relationship. and it derivation Was one of the objectives of physics, The application of classical physics to the mechanics of ht gases is called the kinetic These gaat ATSUGH Daniel Bern had some SUCCES TN UEVETORTAE Tis THEOHY as EAM as 1738, the principal contributions that led to its establishment were made between 1850 and 1900 by Clausius, Maxwell, Boltzmann, and Gibbs. 1-7 KINETIC THEORY OF GASES: MOLAR HEAT CAPACITY Early in our study of physics, we investigated the mechanics of bodies that can be regarded as particles. The study of extended bodies was treated by introducing certain averages, and the translational problem of extended bodies was solved by introducing the concept of a center of mass that moves as though it were a particle. The study of rotational properties of extended bodies was similarly facilitated by the introduction of another average property of the body. its moment of inertia. Jn the kinetic theory of gases, we assume that pressure, volamme, temperature, etc, are averages of properties of all the molecules of a as. Kinetic theory isa large and elegant subject. We can convey its spirit by deriving the ideal gas law and a few other relationships. y Fig. 1-1 Container with gas of V particles. Consider a rectangular container, the edges of which are parallel to the x, ys, and z-axes and have dimensions a, 6, and c, as shown in Fig. 1-1. There are N identical particles, each of mass m, in the box moving in random directions ‘with a wide range of speeds. These identical particles may be, but are not necessarily, atoms or molecules. ‘We assume that the particles ar x io electing tard SOnpSTEE To COMiTons withthe plane walls of the container. .Neale ‘iinor Torees si y fees, we shall consider that the only forces acting on i sulting from collisions with the “27 | Riletie eEarY Of gaEES! motsrnett Capsery wall, We number the particles 1,2: ies+ Ne Figure 1-4 shows the ith SERRE hose velocity i, This velocity may be broken into the rectangular Camporent un fra a8 shown We assume alton ate paect clastic such that when a particle strikes a wall the velocity component rial FSTIETT thats perpendicular to that wal is reversed in direction but unchanged in magnitude. ‘The other tvo velocity components remain unchanged. Thus it the particle strikes side 4, the x-component of its momentum is changed from “4+mnv,,to —mv,,. ‘The net change in the x-component of its momentum is (—mog) = (emg) =, — 2 the wall Gxperiences_ an impulsive Torce. This Tapulaive Tore Is waknowh ‘JaGAUsE We Cannot estimate the Uime of contactam-a meanmaal way. For- {Ghately, is Rar THE Impulsive force but the average Torce Trent FepeuTed hits that we seek. Since collisions with the top, bottom, far, and near sides have no effect on v,,, and since collisions with ends A and B merely reverse the direction of v,, we sce that the time interval between successive hits on side A Sa eee ae AMRIT OS e caompenent aT ee eleeRy alo SSSR Rentoe snd ae HAO HE ERIS TORE, OTe wall On the ith particle to be fie (mo) _ 2m _ mos na a 2 ‘This fore is equa in magnitude and opposite in dieton to the force af the sf mat 1-2) : Ge) be abe V a where V is the volume of the container. ‘The pressure we have computed is due to but one, the ith, particle. The pressure Irom each particle is computed the same way. Adding the pressures from the N identical particles, we have rom the N identical particles, we have _. p= Dp Che To evaluate the sum on the right-hand side of the equation we note (see Fig. 1-1) that ~ aa oh + og + On (4) | Since this equation hols for eee ‘of the pariicles, we can add the corresponding equations and obtain Sas bone Eos as)We now define the mea, yeild, to be the average of the sum of the eines oe eS ea? w- (z a) 6) Applying this definition to all terms in Eq. (1-5), we find that it becomes Hed+gee, 4) and, substituting terms from Eq, (1-6) into Eq. (1-3), we get was p= Nea, Hf p= DN: (8) ‘Since we assume these velocities to be completely random in direction and magnitude, the three mean square velocity components must be equal or #2 = @ =F. This assumption enables us to deduce from Eq, (1-7) that a = 38 = 3 = 3 ws ‘The square root of the quantity o is called the root-mean-square speed, or relocity, tom: Substituting &2 from Eq, (1-9) into Eq. (1-8), we get as BY = jNmi2. (1-10) ‘When the pactcles in the container are the molecules of a gas, then the umber of moles 1 of the gas in the container equals the total number of molecules N init divided by the number of molecules in a mole Ng, the Avogadro Constant. Therefore we have 1 = NjN, or N= Ny. Since the product of the mass of a molecule and the Avogadro constant is the molecular mass M, twe can express the total mass of the gas in the box as Nm = nNgn = Mf When this result is substituted in Eq, (I=10) i becomes GY = hte, > (et) “hiss not the result we sought, pF = nRT, s0 we have as yet no justification for the many assumptions we have made. The result is interesting, however, because it contains the p¥’ term, and the term }Mv? has a familiar look. If wwe write quit = 3Undte>), (12) the quantity ia parenibsics isle total translational kinetic energy of the molecules. This ene he gas_U if the molecules do not have rotational 1nd if jt can be assum energy, Thane Torees_ of attraction or repulsion exist which could give rise to molecular potential z iy combining Eqs. (I-11) and (1=12), we obfaim gre > 13) We compare this resull with the ideal-gas law p¥’ = mR and find that our Sate aee ae for an ideal gas; that is, the temperature is a measure of the internal encrgy of the gas.* Let us now see how this result compares with measurements on real cases. “The change in internal energy with respect to temperature of one mole of an ideal gas at constant volume is called the molar leat capacity: C,. Therefore and its value is Bese ea eel cal a1 =iralx x LSA! 2 2,97 cal - mole“! «°K Co Rm 3 mmole «°K * 418) ‘The experimental values of C, for several gases at room temperature are given in Table I-I. Note that three values agree very closely with the computed Value but that the others are quite different. Both the agreements and the dis- ‘agreements are interesting. The values which agree are those of monatomic gases, which come closest to the ideal gas, We shall have more fo say about the apparent disagreements. The point here is to recall that in our discussion of kinetic theory we assumed that our molecules were isolated clastic spheres ‘Our result apparently does not apply to diatomic dumbbells or to more com= plicated molecules. + Temperature can he defined by the aid of a constant-volume gas theemometer as being proportional lo the pressure (From p¥ = RT). This apnroach assumes that the medium is an ideal gas which, strictly speaking, does not exist, although some igases come close 10 it. more satisfactory definition of temperature was given by Cord Kelvin in 1848._ Is based on the heat transfers in the isothermic parts of the eyele of a Camot engine and is independent of the working substance. (For example. Sle'U Ingard and W. L. Kraushasr, Mechanics, Marrer, and Waces. Reading. Mass. ‘Addison-Wesley, 1960, p. 534.)14 The stomic Table 1-1 Cyof gates Gas | Cn eatsmole tk"! Helium, He ‘ 3.00 ‘Argon, A t 3.00 Mercury, He 3.00 Hydrogen, Ha 482 Oxyeen, O32 497 Chlorine, Cle 601 Ether, 308 4-8 EQUIPARTITION OF ENERGY “The agreement we have observed for monatomic molecules would not have teen possible had there not been the number 3 in the expression Cy = 3/2. Looking back over our derivation, we find that the 3 entered into the ealeulation From the statement 9? = 30d, that ji, because, the, molecule. Was.tiee to Me ee ase apeison for the average kinetic eneey SO UIRIaUGT of The molecules of & gas is composed of three equal pars, R/2 fer degree of absolute temperature astociated with each coordinate. The Principle of eguipartitin of energy states that if a molecule can have eneray| Fociated with several coordinates, the average ergy associaed with cach SSoamnmte TE sane Because of this principle, the number of soardinates SEER IS HAI UE position and configuration of a body is called the number ofits degrees of sea. NA monaldimie molecule requiresCtifS) coordinates to specify its positing A Bald datomis molegule requires thee postion coordinates and A pecessare-tOapeuy is configuration. If we assume the second ator is at a xed Ustance from the firs. its location is specified as being on a sphere suith the first atom at its center. It requires but (wo additional coordinates to Spesily where on this sphere the second atom lies. Thus the addition of « serch atom adds two degrees of fecdom to the molecule, IP our derivation Gri amc fod ben aay ome ad fan moleutee we would. have ar gst or 3RI2. We find that SRi2 = 4.95 cal mole~'-"K~'. which agrees closely with the measured molar heat capacities of such diatomic molecules as hydrogen and oxygen, a8 shown iTable 1-1, Six coordinates are enough to specify the postion of any rigid molecule, however complex, but if the molecules are composed of vibrating atoms, freedom may hecome very large... This “aeeounts Tor the Targe mola fuller discussion of capacities requires the {Hieory reveals much, both qualitatively and quaniiat ee ee a ee With the help of independent data from molar-heat capacities, we have found that the kinetic theory of matter provides a quantitative mechanical model for both the ideal-gas law and molar-heat eapacits. The qualitative result, Eq, (I-14), applies only for the ideal gas, as demonstrated in the sub- sequent discussion of specific heat, However, Eq. (I-11) has more general validity. Using pV’ = nT, we find from,Eq, (1-11) that ue = 3RT.> (ty Substituting for the molecular mass M = Nam, where mr is the mass of one ‘molecule, we get ee» weiner ED (18) “The new constant & = RIN ys called BoltzmanW® Constant or the gas constant per molecule. Tevilie te k= 138002 x 10 TR Equation (1-18) Fhows that the average kine EAeTgy OF translation of the molecules of a gas depends only on the absolute temperature. The average cnergy per degree of freedom is clearly 447. ee “Ai a given temperature, the lighter molecules have the greater speeds. As calculated from Eq, (1-18), the root-mean-square speed of hydrogen molecules St room temperature is about 1800 m s~!, or more than {mis 1-9 MAXWELL'S SPEED DISTRIBUTION LAW We have foun tht cn bs ome rom he empes Speed thus determined Tone of the important average properties ofa gas. But the average doesn't tell the whole story. For many physical and chemical probleme itis important to know the speed distribution of the molecules. For Pence, we may ask how many molecules have more than twice the roots smennesquare specd tin, = (08)" or how many have speeds in the interval 0.79 62. t0.080 ring. This can be found from Maxwell's distribution aw, which sve shal proceed (0 derive. Figure 1-2 isa plot of this distribution Function as pplied to Ny pas ata temperature of T = 273°K (°C). Ey oma ae ah E3 com t Ae ui Bs ii SE ob it aaa ar Be Mean speed mem! cules at O°. Fig. 1-2. Distnbution of speeds fr nit ygen mi16 Theatomie view of matter | 1-9 The deri assumptions: “Assumption 1. Space is isotropic. Therefore the x-components of velocity anit the same isibaTion asthe components along any other axis (including the negative x-axis) Assumption 2, The distribution of x-velocities is independent of the veloci components ¢, anf ‘Assumption 3. The average energy of the molecules is given by E ion of Maxwell's distribution law is based on the following We start by writing the distribution function for the x-components of velocity as. ANC.) = NF} dey (1-19) Here AN(c,)/N is the fraction of the total number of molecules 1V with velocity in the smail interval c, to ¢, + Sey. This fraction, divided by Ag, is the dis tribution function f, which is written in terms of v2 instead of v,, ‘This will simplify the following derivation and does not introduce any new restrictions fon the function beyond what is included in assumption no. 1 Further, because of assumption no. 1, we can write AN(6,) = NICO} Ao. (1-20) AN(o.) = NJ(02) dv, (1-21) where the function isthe same as in Eq. (I-19), but with different argument. By writing the distribution function for x-velocity as a function of v2 only land not also of e3 or 3, we have used assumption no. 2, (Other parameters, as for instance the temperatute, may of course enter into the arguments of these functions.) Figure 1-3 illstrates the velocity distribution of a sample gas. Note that the Figure represents efocity space with coordinates r, ry. and v,, and that the Tocations of the particles (x, r, 2) are not depicted. The gas can be enclosed for instance in a cubieal container and in thermal equilibrium with its was. Every time a molecule bounces off a wall, the corresponding velocity component changes sign and the corresponding point in Fig. 1-3 changes position, The ‘overall picture, however, remains the same, ‘According to Eq, (1-19), the product /(e2) Ar, can be interpreted as the robabilty for finding a given molecule tw the velociy interval o, 10 t+ A0,. ‘The probability for finding that the same molecule at the same j of time algo has j-velocity Th ThE InCerealV TOF “evelacity-in-the interval $B, i enreeacennan (1-22) N 1-3 (8) Velocity disibution of the molecutas in sample of gas. (b) Volume element Invelocty space for speeds between v and v + dv. “The volume element in velocity space Ar, Av, Ae, is shown in Fig. 1-3 at the tip of the velocity vector 0. We now imagine @ new coordinate system placed with the aaxis along wand with the new 3 and z-axcs perpendicular fo it. Tn this new system, the fraction of molecules in a volume element AY” of the same size as Av, dx, Av, but oriented differently willbe ANG, 0,0) NW In the rotated system e, = O-and o, = 0 and the functions f(O) can b Bye constr A Since Ege. (1-28) and (I-28) describe the same fraction “SapreseT TT TWO ferent coordinate systems, We can write AREY = MoD se Lo?) (0) FOMAV. -23) with? = kbp + oh “SWE wow prove that the only type of function f that satisfies Eg. (1-24) igtheen ‘We denote with” the TerVatve OF f With respect (0 ifs argu Tent and take the partial of (1~24) with respect to v,. We obtain AL?) = SOME) LOD. Dividing by eq. U-24}-we oblain — Cw’ . fe) ~ fod By a similar procedure, including Takin and 13, we obtain a FO) £0) £0) Loy 2 w) ~ jw 1 LP a ial derivatives with respect to r, pe! | berexvlell's peed distinvon ta —J 17 seek TE418 ris atone new Ore ‘Now comes an argument that is used very often in solving separable differential equations. The function F(e?) = J"@D/F(e2) depends only upon vs, tneaning that it does not vary if c% is fixed and v3 of of varies. The only way in which Eq, (1-25) can be fulfilled for all possible values of v2, vy and then that each fraction is equal to the same constant (call it ~B), 80 that 02) 1 af SO) f dod) “The solution ofthis i Inf = —fvi + const, which gives fH sje Pt = Ae“. (1-26) is now apparent APO Sadar imaginary constant will give ani pose ’We can now write the form of the distribution function for the x-component of velocity: ) N Ae PADD (27) where 4 and f remain to be determined. We postpone the determination of dnd find A By imposing the obvious condition that the probability that the ma — ‘This procedure, which is called pormall 1 yields A = (Bim)'?. (The w mee where we have replaced A with (fjr)'7%—This exponential function shows that the most probable (or “ullest”) volume element AY = Av, Av, Av, of specified Felocity “size™ is at the origin where v = 0, If we move the same velocity Volume’ clement away from the origin, fewer velocity vectors will terminate within it. One reason for this is that as we move the volume element away from the origin, we are becoming specific concerning the direction of motion (of the molecules, (1-28) I we move from the distribution of velocities to the distribution of speeds, wwe can drop the restriction on direction of motion. Molecules with speeds between Vand V+ AV lie in a spherical shell in velocity space. To get the speed distribution function, we replace AV = Ae, Av, Ae, by AV m= Arr? dr. where v tow means speed rather than velocity. distribution of speeds t0 be - CEN WIN = An(piny? vem (1-29) “The dstibuon ot yeas T-FOONGET avi a the org. The reson is that for a given AV, our speed volume element at the origin is small and includes ‘more velocity space as the speed V increases. Tt remains to determine the constant ft. For this purpose we use assumption no, 3 and write > f (hmv? pan(Bjno%eP do = KT. 5 ‘Again by using a (able of integrals, we find that CG (1-30) Inserted into Eg. (I-29), this gives BNO) Lay ( m_\*P s2-mer28t Ay, 133 i= ete) a0 ‘Thisis Maxwell's distribution law, which is illustrated by one examplein Fig. !~2, “The most probable speed v,, and the average speed p can be found very simply by the aid of Eq. (1-31); they are Og = V 213 Pym = 0.817 Poms (32) and 8 = VB/3M Ome 0.921 Pompe With Cay = (BKTIn)" Crom Eq. (1-18). ‘Manwell’s distribution of speeds was employed to calculate other gas properties and was indirectly verified in terms of these secondary properties. A Girect experimental verification was obtained by Zartman and Ko in 1930. ‘They used an oven, shown in Fig. 1-4, containing bismuth vapor at a known hhigh temperature (827°C). Bismuth molecules streamed from a slit in the oven into an evacuated region above.* The beam was made unidirectional by 0-33) + Fast-moving molecules escape from the oven more often than slow ones. Com- putation shows that if the oven is al a temperature 7, the root-mean-square speed of Reaping moteeules is the same as the root-mean-square speed within an oven at a higher temperature, 47/3. ‘We then find Maxwell'sYor Theatortic view! matory ie = bem Sf im Fig. 1-4 Diagram of apparatus used by Zertman and Ko. another slit, which admitted only properly directed molecules. Above the slit was a cylindrical drum that could be rotated in the vacuum about a horizontal axis Perpendicular to the paper. A slit along one side of the drum had to be in a particular position to enable the beam of molecules to enter it, When the ‘Grum was stationed so that the beam could enter, the beam moved along a diameter of the drum and was deposited on a glass plate G mounted on the inside surface of the drum opposite the slit, During the experiment, the drum ‘was rotated at a constant angular velocity so that short bursts of molecules were ‘admitted on each rotation. Because the speeds of the molecules varied, some Grossed the diameter quickly and others took much more time, and since the Grum was turning while the molecules were moving across it, they struck the glass plate at different places, Thus the distribution of speeds was translated by the apparatus into a distribution in space around the inside of the drum, as indicated by the variation in the darkening of the glass where the bismuth was deposited. The thickness of the deposit was measured optically, and comparison of the experimental distribution of speeds with Maxwell's theoretical distribution expression showed excellent agreement. 4-10 COLLISION PROBABILIT MEAN FREE PATH If molecules were truly geometrical points, no collisions would take place between them, Actual molecules, however, are of finite size, and for the purposes of this discussion we are assuming that a molecule is a rigid, perfectly Elastic sphere. A collision between two molecules is considered to take place whenever one molecule makes contact with another. Let us refer to one of the colliding molecules as the target molecule, of radius r,, and to the other as the Dullet molecule, of radius r5, Then a collision occurs whenever the distance between the centers of the molecules becomes equal to the sum of their radii, y+ ry as in Fig. 1-5(a). ‘When we are considering collisions of molecules of a given gas with other molecules of the same gas, the radii r, and r, are equal and there is no difference between target molecules and bullet molecules. In many instances, however. ‘we wish (o consider collisions between different kinds of particles, and so we shall speak of the target molecules as though they differed from the bullet molecules. Aes iii n Stan 1-5 For mathematical convesience, the actual colision depicted in part (a) may be represented by the equivalent colsion shown in (b) Since it is only the center-to-center distance that determines 2 collision, it does not matier whether the target is large and the bullet is small, or vice versa We may therefore replace an actual collision with the equivalent collision shown in Fig, 1-5(b), in which the bullet molecule has been considered to shrink to @ geometrical point and the target molecule to expand to a sphere of radius 7, + fr ‘Now consider a thin layer of material of dimensions J, /, and dx. The layer contains (equivalent) target molecules only, and to begin with we assume that these are at rest. We then imagine that a very large number V of bullet molecules are incident normally on the face of the layer like a blast of pellets from a shotgun, in such a way that they are distributed over the face. If the thickness of the layer is so small that no target molecule can side behind another, the layer presents to the bullet molecules the appearance shown in Fig, 1-6, where the shaded circles represent the target molecules and the black dots the bullet molecules. Most of the bullet molecules will pass through the layer, but some will collide with target molecules. The ratio of the number of collisions, di, to the total number of bullet molecules, WY, is equal to the ratio of the area presented by the target molecules to the total area presented by the layer: AN _ target area W ™ ‘otal areaFig. 1-6 Target with equivalent target molecules (shaded) and bullot molecules (dots). ‘The target area o of a single (equivalent) molecule is omer tn) a3 “This areas called the collision cross section of one (equivalent) molecule. The total target area is the product of this and the number of molecules in the layer. If there are 7 target molecules per unit volume, this number is n!? dx, so the total target area is nol? dx, ‘The total area of the layer is, 50 aN _ nol? dx fa tel Om no ds. (1-35) In the preceding equation the quantity dN/WV is the fractional number of molecules that undergo collisions and therefore this ratio is simply the probability of a collision. (Strictly, this should have a negative sign because dV Molecules are removed from the stream of bullets.) In the beginning of this Uiscussion the eross section was thought of as an actual area presented by target molecule, but this was soon replaced by an equivalent area, As we shall see later in quantum mechanical systems, e.g, atoms and molecules, we cannot ‘zocept the concept of a sharp boundary as shown in Fig. 1-5. However, we tan use Eq, (1-35) to define the cross section ¢, which then becomes a measure of the probability for collision. Tf'N bullet molecules per unit area are incident normally on the face of layer of material containing stationary molecules having the macroscopic cross section ng, then N, the number transmitted undeflected per unit area through a finite thickness x, can be found by integrating Eq. (I-35). We then have "LN [no ds, \-3 4-10 y~Rollidrorrprobebney! medrese pute—23 — and obtain N= Ngew™*. 37) ‘This exponential equation is plotted as a solid line in Fig, 1~7. o 2 etd Fig, 1-7 Number of molecules W passing though » target layer undetected versus ‘thickness x of layer. ° . ° < Let us next follow in imagination a single bullet molecule as it makes its ‘way through a very thick target along the zigzag path shown in Fig. 1-8. We ‘wish to oblain an expression for the average distance traveled between collisions. known as the mean free path, L. This can be deduced from the results above by 1 type of reasoning that is common and useful in problems of this sort, : 0D C Fig.1-8 Molecular ree paths ‘When molecules are passing through the thin layer of material in Fig. 1-6, the number removed from the beam by collisions is small compared with the original number, and we can say that N molecules have each traversed a thick ness dx of material and that in the process a number dV of collisions have taken place. The total distance traveled by all of the W molecules is then N dv. We now make the hypothesis that the number of collisions made by a single moleculein traversing the same total distance N dx is equal to the number of collisions made by N molecules. each traversing a distance dx. Then from Eq. (1~35), the total number of collisions made by the single molecule in a total path length Nadzis AN = Nua dx. (1-38) ‘The mean free path of the molecule is equal to the total path length divided by the number of collisions, or total path length total number of collisions * From the expressions above for the total path length and the total number of collisions, we have Ndx 1 a -3 Nnodx no C9 ‘The concept of mean free path may be visualized by thinking of a man shooting a riffe aimlessly into a forest. Most of the bullets will hit trees, but some bullets will travel much farther than others. It is easy to see that the average distance the bullets go will depend inversely on both the denseness of the woods and the size of the trees. In the above analysis, we assumed that the target molecules were at rest. ‘This assumption is valid for a bullet molecule going through a solid. If, however, we consider a gas in which both the target and the bullet molecules are moving randomly, the mean free path will decrease because now there are not only head-on collisions as before, but also “sideswipes” with targets moving across the line of travel of the bullet. tis found that the mean free path of a molecule of an ideal gas having a Maxwellian distribution of speeds is 701 (1-40) 4-11 FARADAY'S LAW OF ELECTROLYSIS—SKEPTICISM ‘Another line of argument supporting the atomic view of matter eame from the ‘work of Faraday, In 1833 he observed that if the same electric charge is made to traverse different electrolytes, the masses of the materials deposited on the lectrodes are proportional to the chemical equivalent mass of the materials, The quantity of electricity required to deposit a mole of univalent ions in clectrolysis is called the Faraday constant, F, and is equal to 9.64867 x 10° Coulombs. Like the law of multiple proportions proposed by Dalton, this also implied atomicity of matter. Faraday’s law, however, brings electricity into the picture and implies that both electricity and matter are atomic. We have traced a few highlights of the development of the atomic view of matter through most of the nineteenth century, but since no one had ever seen - Lo Md Pe veri. of __Jtomi___y of ir a molecule. the entire theory was still regarded with skepticism, Maxwell, who proposed the distribution of speeds already discussed, did his greatest work in Clectrical theory. It was he who Found the relationship between electricity and light, and it is because of his work that we often call light “electromagnetic radiation,” In his comprehensive book on electricity and magnetism (1873), after explaining Faraday’s laws of electrolysis on the basis of the atomic theory of matter and electricity, Maxwell says, “It is extremely improbable that when ‘we come to understand the true nature of electrolysis we shall retain in any form the theory of molecular charges, for then we shall have obtained secure basis fon which lo form a true theory of electric currents and so become independent (of these provisional theories.” ‘As late as 1908 the physical chemist Wilhelm Ostwald and the physicist Ernst Mach opposed the atomic theory of matter. Their skepticism is an interesting question in epistemology. These scientists were unwilling to accept purely indirect evidence. Mach makes their position clear in the following analogy: A long elastic rod held in a vise may be made to execute slow. per- ceivable vibrations, If the rod is shortened, the vibrations become a blur in which individual motions of the rod cannot be followed. If the rod is shortened further, the blur may be visually unobservable but a tone is heard. I the rod is, made so short that we no longer experience a physical sensation from its behavior, we may still think of it as vibrating when struck. This, according to Mach, is a safe extrapolation of our ideas because it proceeds from the directly observable to the indirectly observable. Those who were skeptical about the atomic theory objected to the fact that the evidence was entirely inirect. The ‘experiments described in the next section provided the observable events which made the indirect evidence we have given acceptable 10 everyone, 1-42 pennin’s VERIFICATION oF THE ATomic view or warren : Nay Credit for removing the remaining skepticism of atomic theory goes to the French physical chemist Perrin, He tested the hypothesis that the suspended particles which dance about in a stationary liquid in Brownian movement behave like large gas molecules. For his experiments, Perrin prepared a water suspension of particles which met a stringent set of requirements. They had to ‘be large enough to be seen individually, but small enough to have an appreciable thermal motion which could be measured; and they had to be of known uniform size and mass. Further, the concentration of the particles in the suspension hhad to be s0 low that the force effects between them could be neglected. In short, the particles had to be directly observable and conform to the assumptions of the kinetic theory of gases. Perrin was able to obtain a suspension of ‘particles that met these requirements by centrifuginga water mixture of powdered gamboge, a gum resin more dense than water. The centrifuge separated the particles uccording to size, After drawing off a portion of the mixture where26 The atomic view of ma the magnitude of the particle size was suitable for his experiments, he could Centrifuge again and again until the size of the remaining particles was nearly Gniform, Although gamboge is more dense than water, Perrin observed that the particles did not settle out of still water. They assumed a distribution in height, with more particles per unit volume near the bottom of the container than at the top. He measured this distribution as a function of height. Fig. 1-9. Density distribution in along column of gas in a gravitational el. ‘To derive the distribution equation, consider for the moment, instead of ‘a water suspension, a vertical column of gas (Fig. 1-9) with cross-sectional area and at a uniform temperature. Through this column, let us take a horizontal Shee of thickness dy. If the weight of the gas above this slide is w, then the ‘Weight ofthe gas above the bottom of the slice willbe w + div = wr + mana dy. ‘where mg is the weight of a molecule, 1 is the average number of molecules per init volume, and a dy is the volume of the slice. (Note that we are employing fan atomic view of the gas.) The difference of these weights per unit area, tmgndy isthe pressure diference due to the gas inthe slice thats.dp = —mgn dy ‘The minus sign denotes that the pressure decreases as the height increases. Since the molecules in the column of gas have weight, the number of molecules per unit volume at a low fevl is grester than at 2 higher one. Because of this Bifference in concentration, there is a corresponding difference in the number of molecular collisions per unit time at the two levels, and the column of gas comes to dynamic equilibrium, In this state, the weight of the molecules in any fayer is just balanced by the net upward force caused by the difference between the number of molecular impacts per unit time on the lower and upper hor ontal surfaces of the layer. We shall next obtain an expression for the difference in pressure between (wo levels produced by a difference in molecular coneen- trations. “The equation pV = RT holds for one mole of any gas that can be regarded 4s ideal. According to atomic theory. the number of molecules in a mole is the ‘Avogadro constant N,.. We can obtain an expression that contains the concen tration of the molecules by dividing the ideal-gas law equation by the Avogadro constant. Thus we have PVIN, = RTIN, oF p= mT, where = MY i the moles conentation, This cnenitin 8 ction ofthe pressure and therfore, since k and T-are constant, dp = singe k and Tare constant, dp = KT When this expression for the difference of pressure due to the dilference in molecular coneeptrations in the layer of height dis equated to the diference Caused by the weight ofthe layer, we obtain dn 8 ay, AT dn =—mgndy or n kT (at) Here m isthe mas ofa single molecule “The elation between the molecular concentration myst) = Oand naty = ‘can be found by integrating Eq. (1-41). We have ? "dn _ (h_ mg a (a-L- ie n= moe mn?, (1-42) which gives Since the posure of» gas is dirty proportional othe number of molecules per ant voume, Eq (1-2) can be rewrien in ferns of presses. The resulting ecuton ten id he low of eprint Bes the dston to the force of gravity. ans at cae Equation (12) must be modified slightly to make it applicable to water suspensions, The fective weght of & particle suspended in-a Nil i the feastant of iy welgt and Arehinedes” buoyant force, The volume of @ fare of mass mand desiy pis pnd the mass OF 80 equal Yolume of Taguid having a deny pis mip Therefore, the buoyant force om this parle whey submerged tn he lui is mgp'p, and Hs eeive weigh Becomes mg ~ ng? = mg (22). a ? When we replace the actual weight mg in Eq. (1~42) by the effective weight and reinsert instead of the Boltzmann constant RIN, = Mamata sm) pRT we obtain n= no exp [ (1-43) “Tis she equation for sementrion eqvilbriom of sispension a 2 rs of Brownian movement. * "28 The ator Perrin measured a series of 1's at a series of /'s in a very dilute suspension ‘The seults verified the sedimentation equation, Equally important, his neasurements yielded a value for N, since all other quantities in the equation mrcre known, ‘The very existence of the Avogadro constant implies the correct: ness of the atomic theory. I is imeresting that Einstein, the greatest contributor to the development fof modern physics, also had a part in the final establishment of the aorie fheosy In" 1905 he derived an equation which describes how a suspended Dortile should migrate in a random manner through a liquid. is express® Trvolved the Avogadro constant, and Perrin’s verification of the Einstelt sveiula confirmed the value of this constant. The results showed that small particles suspended in a stationary liquid do rove about in the manner Pre~ Gicted by the molecular-kinetic theory of gxses. ‘Altogether, Perrin used four completely independent types of measurements: cach of which was an observable verification of atomic theory and each of Sisieh gave a quantitative estimate of the Avogadro constant. Since the publ Millon of his results in 1908, no one has seriously doubted the atomic theory ‘of matter, 4-43 BOLTZMANN'S DISTRIBUTION LAW Maxwell's speed-distribution law (Eq. 1-21) and the law of atmospheres {Ea 1-42) have something in common. In both formulas the factor exp(~ EIA) Gobears, where E in one case isthe kinetic energy 40" of a molecule and i) a pether is the potential energy mgh of a molecule in a gravitational field. The ire arity is by no means accidental, Both formulas referred to can be derived from a general law of statistical mechani called olsun dito ios: CE me (-44) “te aw dats withthe distribution of molecules or other particles over diferent The a det UNE Ep being te difernce i energis between these sae er es Yibuon over the states, ete dstbuton that would exist Tithow any energy ference, must be uniform tein cop pecans eles A the a Selosed box with walls at temperate 7. ola UTE, patanel fle, the dstibution of molecules would be the ae age) This means that there ouTd be the same number i re (er uy poiion between = Oand das nthe tate sented sae vc the equally Tage volume element between and + dh TF ya postion 10 pe prtational field and aintn the terperatre 7 on ye irene in potential energy between te two tates defined sbove ai be Eq = mgh. Inserting this into Eq, (1-44), we obtain Ea. (1-42) Strictly SpERGRE The Kinetic energy should also have been included in the description of the energy states. In the particular example discussed. the kinetic energy distribution is the same in the two volume elements. (The temperature is the same.) Therefore we need not consider the kinetic enerey. ‘Maxwell's speed-distribution law can be derived from Bq. (1-44) by pos tulating that the a prior’ distribution of molecules in velocity space is uniform. This means that the density of dots in Fig. 1-%a) would be uniform, without the influence of the kinetic-energy difference on the exponential of fq. (1-44. Boltzmann's distribution law can be derived by maximizing the entropy of a systein of particles with a constant total energy.” It is beyond the seape of this book to include this derivation. PROBLEMS. tel Ancarly chemist wishes fo determine the atomic weight of nitrogen. He assumes that the atomic weight of oxygen is exactly 16, and he prepares four oxides of nitrogen which are distinetly different compounds (see data below ‘Oxygen, parts Nitrogen. parts ‘by weight by weight Product oxide 863 197 “ a 500 285 B 300 38 ¢ 108.2 186 D 4) Show that these data demonstrate the law of multiple proportions. (it First ind the masses of one of these elements that unite with a unit mass of the ther.) ‘b) The chemical formulas of the products cannat be determined com- Dletely from the above data, but, by assuming that nature is simple, one rmty Propose several possible sets of product formulas. Write out several possible ets 0 product formulas. ¢) Calculate the atomic weight of nitrogen for each se. 1-2 Compute the molar heat capacity at constant volume of a gas composed of molecules that are rigid three-dimensional structures. 1-3 Given tit p represents the density of « gas, show that Come = So. 1-4 ay Compute the arithmetic mean speed and the root-mean-sq land the root-mean-square. speed for teach of the following distributions ofthe speeds of eight particles 1. Al eight have speeds of 10 ms”! 2, Two have speeds of 3 ms”, four have speeds of 6 ms"! and wo have speeds of 10 ms~! 2 Sen for instance, D. Ls Livesey, Atomic and Macear Physics, Waltham, Mas Blaisdell, 1966, p. 13, " ‘namThnmieh —tma k 3, One has a speed of 3 ms}, three have speeds of 6 ms~!, and four have speeds of 10 ms“! 4, Four are at rest and four have speeds of 10 ms“! by In each case decide whether the shape of the graph of the speed distriburion would be the same as that of the translational kinetic-energy’ distribution, assuming that each particle has the same mass. 1-5 The speed distribution function ofa group of W particles is given by AN, = av dt. where dl; is the number of particles that have speeds between vand e ‘+ do, and ‘2 is a constant. No particle has a speed ereater than V, and the speeds range from 010 ¥. a) Draw a graph of the distribution function, that is, plot (Nid) versus ¢. 6) Find the constant ain terms of N and V. ) Compute the average speed, the root-mean-square speed, and the most probable speed in terms of V. {) What percent of the particles have speeds between the average speed and 7? between the root-mean-square speed and V7 1-6 a) Show from Eq. (1-31) that the most probable speed in a Maxwellian distribution is given by tq = VTi = V2RTIM, and by then show from Eq. (1-18) that ty = 2/3 tga. (Hint: Determine the ‘condition for which the ordinate quantity in Fig. 1-2 is maximum.) 1-7 Show from Eg. (I-31) that the average speed in a Maxwell c= ff emaen [° anna a distribution is Some definite integrals are f eee de f, wer de = Ya, f ster dy = GW np8)a 2 Fnahan 9 1-8 Show from Eq. (I-31) that the root-mean-square speed ins Maxwellian di tribution is 34 TI. cme [{{ ane af fanaa 1-9 The speed of propagation of « sound wave in ai at 27°C is about 348 ms“! Find the ratio of this speed to the rms speed of nitrogen molecules at this temperature, 1-10 The drum of @ Zartman-Ke apparatus (Fig. 1-4), has a radius of 8 em and rotates at 6000 rpm, The oven containg mercury atoms at a temperature of 600K. Two atoms of mercury, one with the most probable speed at oven temperature and the other with the rms speed at the same temperature, leave the foven and enter the rotating drum, These (wo atoms are then deposited on the 13 1-16 7 lass plate atthe far side ofthe drum, What is the separation of these two atoins fon the glass plate? (Atomic weight of mercury = 200.6.) ‘An object can escape from the surface of the earth if its speed is greater than J/2eR, where q is the acceleration caused by gravity and R is the radius of the earth, a) Using a radius of 6.4 > 10% m, calculate this escape speed, ) Explain why oxygen and nitrogen remain in the earth's atmosphere while hydrogen does not 4) At what temperature will the rms speed of oxygen molecules he twice their tins speed at 27°C? b) At what temperature will the rms speed of nitrogen molecules equal the rms speed of oxygen molecules at 27°C? 1) To what temperature would an ideal gas in which the “particles” are buse- balls have to be heated so their rms speed in a Maxwellian distribution would equal that ofa fast ball having a speed of 30.5 ms"? (The mass of @ baseball is 14400 Find the rms speed, the average speed, and the most probable spced of the molecules of gaseous hydrogen at a temperature of (a) 20°C and (b) 120°C. Assuming that the energy E of a molecule is only translational kinetic energy {4} show from Eq, (1-31) that the fractional number of molecules which have cenergs in the range AE is ay 2 (LV pu No aaa) © b) From the enerey distribution of part a, show tat te met profile vera is KT. ©) What is the ratio of the average translational kinetic energy to the ‘most probable translational kinetic energy? Henry ‘Show that the kinetic energy of translation of a molecule having the most praheble speed in 4 Maxwellian distribution is equal to &T. ‘A neutron is fundamental particle, Like ordinary gas molecules, neutrons have 4 distribution of speeds, and this distribution is of prime importance in the theory of nuclear reactors. Quantitatively, a thermal neutron is usually delined as one having the most probable speed of a Maxwellian distribution at 20°C. Find a} the kinetic energy of translation and b) the speed of a thermal neutron. €) A thermal neutron is sometimes called a “kT neutron.” Why? (The mass of & ‘neutcon is 1.0087 u.) Assume that hydrogen atoms in the atmosphere of the sun obey the Maxwellian spec! distribution, a) Given that the temperature is 6000 K, ealeulate the Kinetic energy of one of these atoms moving with the most probable speed in the disiibution, b) Calculate the speed of this atom, ‘When the atoms in a deuterium gas have an average translational kinetic enerey ‘of 12 » 107" J, they can approach one another so closely that nuclear fusion ‘will eecur, a) What isthe speed of a deuterium atom Raving this kinetic enerey? b) To what temperature would the deuterium gas have to be heated so that the rmg speed of the alonss would equal the speed in the preceding part? (Deuterium is hydrogen having an atomic weight of 2.014.)ENCE 243 - INTRODUCTION TO ATOMIC PHYSICS HARALD A, ENGE Massachusotts Institute of Technology M, RUSSELL WEHR Drexel (Emoritus) JAMES A. RICHARDS — State University-of New York, Delhi CAR Zad QUANTIZ “eS ee “yt SEZIONE STAT. Buin BIOMOLECOLE Hy ADDISON-WESLEY PUBLISHING COMPANY Reading, Massachusetts » Menlo Park, Califomia - London - Don Mils, Ontario32 The atomic view of matter 1-20 &) What is the total kintic energy of translation of the atoms in 4 moles of helium at a temperature of 27°C7 6) What would be the answer for the same ‘amount of another ideal gas? 1-21 The microscopic cros section fora certain bullet and particle is o when they are electrically neutral, Would the effective value of o increase or decrease if the bullet and particle carried electric charges a) of like sign, b) of unlike sign? 1-22 a) Show that r, the number of molecules per unit volume of an ideal gas, is fBven by n = pN,IRT, where Nis the Avogadro constant. b) Find the number fof molecules in I'm? of an ideal exs under standard conditions. ©) What isthe ‘number of molecules in 1 m? of an ideal gas at a pressure of two atmospheres and a temperature of 47°C? (1 atmosphere = 1.013 x 10°N m7), 1-23 &) If the pressure is kept constant, at what temperature will the mean free path. Of the molecules of a given mass of an ideal gas be tvice that at 27°C? 6) If the temperature is kept constant, at what pressure in millimeters of mercury will the mean free path of the molecules of a given mass of an ideal gas be 1000 times (greater than that at a pressure of 1 atm? 1-24 The molecular diameter of all diatomic gases is approximately 2 x 10~#? m, 8) Find the mean free path for 2 Maxwellian distribution of speeds of the molecules of hydrogen gas when at a pressure of 1 atm and a temperature of 20°C. (Data for caleulating the macroscopic cross section are given in Problem 1-22) b) How many collisions per second would a molecule that is always ‘moving with the average speed in a Maxwellian distribution make in the pre~ ‘ceding case? (The time of contact during collisions is negligible.) ¢) What isthe Tatio of the mean free path ia part a) to the wavelength of green light, 2 = 500 x 10°! m? 1-25 A beam of bullet particles is incident normally on a layer of material containing ‘Stationery target particles, Find a) the fraction of the incident beam transmitted fand the fraction which experienced colisions in a layer whose thickness equals the mean free path and b) the thickness of the layer in terms of the mean free path required to reduce the transmitted beam to one-half the inteasity of the Incident bear. 1-26 In one of his experiments, using a water suspension of gamboge at 20°C, Perrin cbserved an average of 49 particles per unit area in a very shallow layer at one level and 14 particles per unit area at a level 60 higher. The density of the ‘gamboge was 1.194 gem~® and each particle was a spherical grain having & radius of 0.212 u. (Lz = 10"®m) Find a) the mass of each particle, b) the ‘Avogadro constant, and c) the molecular weight of a particle if each grain is regarded as a single giant molecule, Use the results from parts a) and b) to calculate). CHAPTER 2 The Atomic View of Electricity 2-4 ELECTRICAL DISCHARGES ‘We have already considered how Faraday’s law of eletroylsis implies that both ratter and electicity are atomic. Ta spite of Maxwell's skepticism, itis sere Sie to eplin the fat tat the pasag of one faaday of ty tour Sy aauRS WaT UT nates a ces en in : Oo ene ree rae te line of investigation and consider the passage of electricity through gases. ‘Although Benjamin Franklin's very dangerous experiment with kite and key ‘was not a particularly convincing one, it nevertheless led to the correct conclu sion that lightning is the discharge of electricity through a gas (air). Every electric spark is an example of this process. Since sparks are one of the most dramatic electric effects, it is natural that they should have been a subject of early study. ‘The passage of electricity through gases is a very complicated process and 1 great deal has been learned from it. There are many ways in which the character of an electrical discharge can be altered, but here we shall direct our attention to the effect of gas pressure. A typical discharge tube is shown in Fig. 2-1, This system has a gauge which measures the gas pressure and a pumping system which varies the pressure. Electrodes are sealed into the ends of the tube so that an electric field can be established between them. ‘When the pressure in the tube is atmospheric, a very large electric field is ‘required to produce a discharge (about 3 x 10° Vm" for air). The discharge isa violent spark as the gas suddenly changes from being an excellent insulator to being a good conductor. As the pressure is reduced, the discharges are more easily established (Fig, 2-2), until, at very low pressures, they again become 334 The atomic view of electricity | 2-4 4 Vacwum pump Fig.2-1 Gos dicharge apparatus. difficult to start. Discharges start most easily at a pressure of about 2 mm of ‘mercury (although this will depend upon the kind of gas and the geometry of the electrodes). As the pressure is reduced, the discharge changes in character. With air in the tube, the bright spark changes to a purple glow filling the whole ‘tube, and with neon, one obtains the red glow seen in many advertising signs. ‘On further lowering of the pressure, the glow assumes a remarkable and complicated structure, with striations and dark spaces. At very low pressures the glow of the gas becomes dim and a new effect appears—the glass itseif begins to glow. Ifthe bulb has within it a device which is hinged so that it can be made Starting potential, V 24 6 8 10 2 i Presure, mm of mercury Fig. 2-2 Typical starting potential curve of gaseous discharge. to move into or out of the region between the electrodes by tipping the entire bulb (Sin Fig, 2-1), then another effect may be seen. The greenish glow of the lass, which appears everywhere between the electrodes when the object is ‘out of the way, is partly obliterated when S is swung between the electrodes. If the object S has some distinctive shape, it may be seen clearly that itis casting ‘a shadow. The shadow is on the side of S that is away from the negative eleo- 2-2 | Nonrelativistic charged-particle ballistics 35 trode of cathode. If the cathode is small, the shadow is rather sharp. It is a imple deduction that the greenish glow is caused by some kind of rays from the cathode that cannot penetrate the obstruction S. These rays are called cathode rays. Many years ago it was observed that these rays could be deflected ‘by both electric and magnetic fields, and the direction of these deflections showed ‘that the rays were negatively charged. Sir J. J. Thomson undertook a quantitative study of cathode rays in 1897. He was able to show that all cathode rays or corpuscles possess a common property. He showed that the ratio of their charge to their mass, q/m, was a constant. His measurements did not establish that all the rays have identical charges or identical masses, although this is the simplest interpretation of his results, He did, however, discover a unique characteristic of these rays and he is regarded as the discoverer of a fundamental particle of electricity, the electron. 2.2 NONRELATIVISTIC CHARGED-PARTICLE BALLISTICS Before discussing one of the methods by which g/m can be measured, let us review some basic facts about the motion of charged particles in electromagnetic. fields, When a particle having charge +-q isin an electric field of intensity E, the particle expetiences a force in the direction of the field, of magnitude Fm géE. ok @y If all other forces on the particle are negligible compared with this one, ‘the particle will undergo accelerated motion, and we have by Newton's second law Bigg, (2-2) nt @2) where p is the momentum of the particle. Unless the velocity of the particle is, very high, we can in practice use nonrelativistic mechanics and write dp/dt = ma, yielding Eo @) When a particle of charge q moves in a magnetic field of induction B with 1 velocity , it experiences a force that is perpendicular to the plane formed by the vectors B and »: FaqxB 9% e4 + This equation and those following are valid in any consistent system of unit [No conversion factors for units need be introduced provided all are electrostatic unit all are electromagnetic units, ofall are meter-kllogram-second-ampere units. This will be true ofall equations in this book except in cases where units peculiar to atomic physics, such as angstrom units or electron volts, are specified.i i 36 Theatomic view of electricity | 2-3 Af the velocity vector is parallel to the magnetic field, clearly no force results and therefore there is no change in the motion, When the velocity vector ‘points in any other direetion, we can break it up into two components, one parallel to the field v), and one perpendicular to the field v,. The component , gives rise to a force F = qv,B, perpendicular to v,, and this force therefore produces a circular motion. If the component 0, is nonzero, the result is a helix rather than a circular orbit. Both components and the total speed are ‘numerically constant, but », continuously changes direction. By use of Newton's second law, we find qu.B = mo3/R, gBR = my ey We have here used the formula for the centri tal acceleration a, = 02/R. Equation (2-5) is valid also in relativistic mechanios (Appendix 3), when m is taken as the relativistic mass. This is true because the speed does not change and therefore the mass is constant. Fig. 2-3 Thomaon’s apparatus for measuring the ratio g/m for eathode rays. 2-8 THOMSON'S MEASUREMENT OF a/m We are now ready to consider how Thomson measured the ratio of charge to mass, g/m, for what he called “cathode corpuscles.” His apparatus (Fig. 2-3 ‘consisted of a highly evacuated glass tube into which several les were sealed, Electrode Cis the cathode from which the rays emerged. Electrode Fir the anode, which was maintained at a high positive potential so that a discharge of cathode rays passed to it. Most of the rays hit 4, but there was a small hole in A through which some of i ipepmmee Thee aye sere te Festricted by an electrode 4’ in which there was another hole, ‘Thus a narrow Tar of The Tays passed into the region of the two plates P and P’. After passing between the plates, the rays struck the end_of the (Ube, where they caused fluorescent material at Sto BOW. "The deflection plates P and P’ were separated a known amount, so that ‘when they were at a known difference of potential the electric field between 2-3 | Thomson's measurement of g/m 37 tion of cathode rays. them could be computed. We shall assume that the field was uniform for a distance L between the plates and’ zero outside tem. When the upper plate P ‘was made poiitive, the electric Feld dellecied the negative cathode rays upward MORE 2a, the cathode rays are assumed t ewer The Tepion betwee te plates at the origin O with a velocity v,, Because there is no force component in the x-direction, », remains constant." The general equation for displacement in uniformly accelerated motion is fot tot + 4a, Gene (2-6) eterna & Finca EB) Eq, (2-6) to the horizontal direction, we obtain Apel ae @) Between the plates the rays experince an upward acceleration, ° obtained from Eq. (2-3). The electris field £ is constant, since the fringing- field zone is neglected, and it is equal to the potential difference between the deflection plates divided by their separation. Hence the general displacement tquation in the vertical direction becomes Elimination of ¢ between Eqs. (2-7) and (2-9), yields the equation for the + _Stitty speaking, this is not true. When the particles leave the region of the plates ‘with a yedisplacement as shown, they will experience a decelerating force in the 2 rection. (The field lines in the fringing field bulge out rather than being vertical as shown.) The effect is of the exact magnitude required to conserve the energy of the particles; that is, the speed after the deflector equals the speed before the deflector. If, 21) melts = GPP ‘This is another equation relating e/m, and p,. Tt could have been used with Eq, @-18) to give e/m,. Thomson could have measured the potential difference ‘between the cathode and anode and been spared either the electric or magnetic daflcction of the beam in the vicinity of P and P’. Indeed, other methods of measuring e/m, utilize this principle. = As shown in Appendix 3, relativistic eects are important when 2c? fs not nes- Tibi cal with being the veloc of light. ‘The errors introduced here (ia Ed. 3713) oy nny clascal mechanics ae ofthe order of magnitude of oc or about 1 percent. 24 | Electronic charge 41 Focusing, Plates for etn orzgntal defection Pita fo Contry "\* Accelerating verial defection id node cotnode ear | Fig. 2-6 Basic elements of eathode-ray tubs. Cathode-ray tubes such as Thomson used have been developed into im- ‘portant modern electronic components. Electrostatic deflection of an electron ‘beam is used in the cathode-ray tube of modern oscilloscopes. Such tubes ‘usually have two sets of deflecting plates (Fig. 2-6), so that the electron beam ccan be deflected right and left as well as up and down. These tubes utilize the fact that the deflection is proportional to the electric field between the plates, as shown by Eq. (2-12). Television tubes, on the other hand, commonly utilize magnetic deflection to cause the beam to sweep over the face of the picture area. “Anyone can demonstrate for himself that electric and magnetic fields deflect electron beams. Holding a strong permanent magnet near the face of a television picture produces weird distortions. Rubbing the face of a picture tube or even the plastic protective window with wool, silk, or nylon will produce strong electric fields when the humidity is low. Neither the magnetic nor electric fields thus produced are uniform or perpendicular to the beam, and the deflections they produce are striking in their unpredictability. 2-4 ELECTRONIC CHARGE Although the measurement of e/m, indicated the identity of electrons, another ‘measurement was required before ¢ and m, could be known separately. This was first made with precision in 1909 by R. A. Millikan, who perfected a technique suggested by J. J. Thomson and H. A. Wilson. Both the charge ¢ and the mass m, of an electron are incredibly small ‘quantities. The mass of any body can be determined from the measurement of the force acting on it when it is accelerated, Even if a single electron could be isolated for study, no instrument could measure its mass directly. Similarly, the charge on a body can be determined by measuring the force it experiences in an electric field, This method does not require the isolation of a single electron ‘and, since very intense electric fields can be created, a measurable force can be produced. ‘An experiment to measure e must be carried out with a body having so few charges that the change of one charge makes a noticeable difference. Since the42 The atomic view of experiment must be done with very little charge, the foree the body experiences ‘vill be small even though a large electric field is utilized. If the force on the ‘charged body is very small, then the body itself must be very light. The force of gravity is always with us, and if the small electric force is not to be masked by 2 large gravitational force, then the mass of the body must be both small and Known. If the body is small enough that the electric force on its charges is of ‘the same order of magnitude as the gravitational force it experiences, then it ‘may be that the gravitational force will be a useful standard of comparison rather than an annoying handicap. ‘Millikan used a drop of oil as his test body. It was selected from a mist 24 | Electroniecharge 43 el! matter mecca 6 mr Biy Bre | | te Ero > Cm) * ¥ W he (© Fling) Rising Fig. 2:7. Forces acting on an ail drap (equlibrkim conditions). produced by an ordinary atomizer. The drop was so small that it could not be measured optically, but with a microscope it could be seen_as a bright spot ‘because Te scattered I 2 Tike a min im bright sunlight, “When such a drop falls under the influence of gravity, itis hindered by the air i rough, “The way 1m vis inidered by air had been described 2” rho found that at low velocities such a body experienced a resisting TOPO proportional to its velocity, or R= ke. em) “The proportionality constant k was found by Stokes to be Of the body. (This law assumes that the resisting medium is homogeneous. ‘A more complicated law must be used if the size of the body is of the same order of magnitude as the mean free path of the molecules of the medium.) This is a friction equation very different from that introduced in mechanics to describe the force between two sliding bodies. In that case we assumed that the friction force depended only on the nature of the sliding surfaces and the normal force pressing the surfaces together. Hence in mechanics we discussed @ force Which did not depend on the speed of the motion. In the problem of a box sliding against friction down an inclined plane, the friction produced a constant force opposing the motion, but the acceleration was constant and the velocity jncreased continuously. A’body subject to a frictional foros like that given by Stokes’ law will behave very differently. ce Meglows "A falling droplet of oil is acted on by its weight w, the buoyant force: SUE TORE KS KE (hie 2=2) The xesulin force Fis os a Few-B- ko. k= Gear, (2-24) the resisting force is zero, and the resultant Soitially the velocity 2 is 2020, downward force equals w — B. The drop therefore bas an initial acceleration. As its downward velocity increases, the Boca FPO (ol aE ee ane ley fiir the resisting force increases. ‘and eventually reaches value such that the resultant force is zero. The drop ‘then Falls With'a constant velocity elled its terminal velocity, vy Since F = 0 when v= 0, We have from Eq. 22), —— (225) Let p be the density of the oil and p, the density of the air. Then w= tarp, B= Srr°p.9, 226 and inserting the value of k from Eq, (2-23), we get $0? (p — pg = Garey 2 [All of the quantities in this equation except r are known or measurable, ‘We can therefore solve for the drop radius r and hence can express the proportionality constant k in terms of known or measurable quantities. The result is °o, 2 lit = val With this constant known, we can use velocity measurement to determine forces, In the experiment, the oil drop is situated between two horizontal plates where a known strong electric field may be directed upward or downward for may be turned off (Fig. 2-8). The droplet has a small electric charge q 18% (2-28) Atomizer Fig. 2-8. Milken’ ol-drop experiment.pl reds © erage OE a 44 The atomic view of electricity | 2-4 ‘which may be minus or plus, depending on whether it has an excess or deficiency of electrons. The droplet gets this charge from rubbing against the nozzle of the atomizer and from encounters with stray charges left in the air by cosmic rays, deliberately produced by x-rays, or by bringing a radioactive material nearby, In the electric field the drop will experience a force gB, which can always be directed upward by the proper choice of the direction of £. The ‘experimenter can turn E on and off so that the drop falls by gravity or rises because of a dominating electric force in the region between the plates but never touches either. ‘The microscope with which the drop’s movements are followed is equipped with two horizontal hairlines whose separation represents a known distance along the vertical line in which the drop travels. By timing the trips of the drop over this known distance, the terminal velocities of the drop are found. The velocities of fall, v, (positive down), are all the same, since oil does not evap- orate noticeably, and therefore the weight of the drop is constant. The velocity of rise, vg (positive up), however, depends on the charge g and upon B, The resultant force on the drop while it is rising (Fig. 2-7(b)) is Fe gk + B-w-bo. 2-29) When the terminal velocity og is reached, the resultant force is zero, so aE = w— 3+ kop 2-30) But from Eq. (2-25), w — B = ko, so finally Gari + op. 31) Since these terminal velocities are constant, they are relatively easy to measure. Equation (2-31) permits the evaluation of g, the charge on the drop. In the oil-drop experiment, the value of e, is determined for a particular drop with the electric field off, and a whole series of v,’s for the same drop is observed with the field on. If we knew that the electronic charge was unique and that thx vas oly on gs one Tapes Ea-_Sipwonlé a is ie SES aC ce See the nature he leironie charge was not known, RET tot pe ati he fund toe egal mulls of one a SE ar ot are. e Thus be stable the SRS pie rtrrs or leans charges oid COncnaed Tom 1 iat ~ fikan made observations on oil drops of different sizes and also on drops of mereury. In one instance a drop was watched continuously for eighteen hhours. The sets of observations always gave the same value of the electronic, charge or “atom” of electricity. ‘The best modern determination of ¢ is (1.602192 + 0.000007) x 107! C. 26 | Positiverays 45 Ete smo) 2-5 MASS OF THE ELECTRON; AVOGADRO CONSTANT ~~? Since e/m, and ¢ are now known, it is only simple arithmetic to find the mass of the electron to be = = 610956 0.00005) x 1077? kg. ‘Still another basic atomic constant may now be calculated with precision by using the value othe electronic charge. The Faraday constants the amount Sf material hough an electrolyte, Dividing te Faraday constant by gives Ge HIRE OF Seaton WHEY Nave parispated ta is Tango or The 2-6 POSITIVE RAYS ‘After the particle of negative electricity, the electron, had been identified, it was reasonable to ask about positive electricity. The search was made in @ discharge tube very similar to that which disclosed cathode rays. In 1886, Goldstein observed that if the cathode of a discharge tube had slots in it, there appeared streaks of light in the gas on the side away from the anode. These channels of light, first called “canal rays,” were easily shown to be due to charged particles. They moved in the direction of the electric field which was producing the discharge, and they were defiected by electric and magnetic fields in directions that proved that their charge was positive. Attempts were ‘made to measure g/m, the ratio of the charge to the mass, of these positive rays. It was soon discovered that g/m for positive rays was much less than for clectrons and that it depended on the kind of residual gas in the tube. The velocities of these positive rays were found to be nonuniform and much less than electron velocities. ‘Thomson devised a different method for measuring g/m of these positive rays having nonuniform velocities. Figure 2-9 shows the apparatus he used. ‘The main discharge took place in the large bulb A at the left, where K is the cathode and D is the anode. The gas under study was slowly admitted through te tube at Z and was simultaneously pumped out at F. Thus a very low gas pressure was maintained, Most of the positive rays produced in the bulb hit the cathode and heated it, The cathode had a “canal” through it, so that some of the positive rays passed into the right half of the apparatus. Just to the right of the cathode are M and J, the poles of an electromagnet. The pole pieces of ‘this magnet were electrically insulated by sheets, J, so that the magnetic pole pieces could also be used as the plates of a capacitor for the establishment of an clectric field. With neither electric nor magnetic fields, the positive rays passed straight through the chamber C to the sensitive layer at S. This layer was either ‘the emulsion on a photographic plate or a fluorescent sereen. ‘The beam was46 The atomic view of electricity | 2-6 Fig. 2-10 Formation of postve-ray parabola. ‘well defined because of the narrow tunnel in the cathode through which it had to pass. Instead of crossed fields as in the electron apparatus, this apparatus has its fields perpendicular to the rays but parallel to each other. ‘The electric field is directed downward and the magnetic induction is upward, so thet in Fig. 2-10 the electric force is toward the bottom of the page along the y-axis and the magnetic force is out of the page toward the reader along the z-axis. ‘Let a positively charged particle of unknown g/m enter the region between P and P’ in Fig. 2-10 with an unknown velocity v, along the x-axis. Then, 26 | Positiverays 47 according o Ba, (12), the deflection of the parce on the sree 5 dus to the dente fld is yo BEE xo), ‘and, according to Eq, (2-18), the deflection at S caused by the magnetic field is 2 BE (Ean). ‘tose two equations together are the parametic equations of a parabola, Tree we, Re pormeter Shoe oy is fen for diferent parties of the same type, the pattern on the screen is not a point but a locus of points. Elim- een o beowean hee equations leads to BEL (H+) », G+)» (2-32) 33) 2-34) which is the equation of a parabola, Fig. 2:11. The paraboles of neon. ‘Some actual parabolas obtained by Thomson’s method are shown in Fig. 2-11. Examination of the figure reveals several things: Positive rays have distinct values of g/m, as is shown by the fact that the traces are clearly parab- ‘las, ‘That a single-experiment discloses several values of g/m is evident from the fact that there are several parabolas. It is apparent that the method is not capable of great precision because the parabolas are not sharp. “Thomson asstumed that each particle of the positive rays carried a charge ‘equal and opposite to the electronic charge, and he attributed the divergent parabolas to differences in mass. He assumed that the positive rays were positive because each had lost one electron. Thomson could identify particular parab- ‘las with particular ions (charged atoms or molecules are called ions). Thus for ‘atomic hydrogen, he could verify that the gjm he measured was equal to the43 The atomic view of electricity | 2-7 ‘value one would expect from dividing the electronic charge by the mass_per Mom (the atomic mass of hydrogen divided by Avogadro's number). ‘The ‘avon that positive rays move more slowly than electrons and have lower values Ur aym than electrons is now clear: The positive rays are much more massive, ‘The largest gfm for positive rays is that for the lightest element, hydrogen. From the value of g/m it was found that the mass of the hydrogen ian or proton Fe 18362 times the mass of an electron, Electrons contribute only a small ‘amount to the mass of material objects, 2-7 ISOTOPES. “The most striking thing that was shown by the Thomson parabolas was that Zeatain chemically pore gases had more than one value of g/m. Most notable cere the case of neon, of atomic mass 20.2, Neon exhibited a parabola situated Te eonrespond to a particle of atomic mass 20, but it elso had a parabola which fedieated an atomic mass of 22, Since the next heavier element, sodium, has sr atomie mass of 23.0, efforts to explain away the unexpected value of g/m failed at fst, Finally, it was concluded that there must be two kinds of neon, teh different masses but chemically identical. The proof of this interpretation ‘was given by Aston, one of Thomson’s students. on used a principle which we discussed in Chapter 1. We pointed out there that the average kinetic energy of a molecule in a gas is 3K7/2. Dilfeent fas molecules mixed together in a container must be at the same temperate, eer ence the average kinetic energy of each kind of molecule must be the ste. if the two gases have different molecular masses, the lighter molecules sarnt have the higher average velocity, and these will make more collisions per ‘ave time with the walls of the container than the heavier molecules. Therefore TTthece molecules are allowed to diffuse through a porous plug from a container ite another vessel, the lighter molecules will have a higher probability of passing through than the heavier, slower ones. “Aston took chemically pure Pon gas and passed part of it through such a plug. Since one such pass aosom- plishet only a slight separation, the process hed to be repeated many tines, vie ended with two very small amounts of gas. One fraction had been through the plug many times and the other had been “eft ebind” many times. He vee pared the atomic mass of each fraction and found values of 20.15 for the Tenmer and 20.20 forthe latter. The difference was not great, butt was enough othe that there ae indeed two kinds of neon (at least). Many other elements ave since been shown to exist in forms that are chemically identical but viterent in mass. Such diffrent forms of an element are called isotopes of the dae. Thus Dalton’s belie that all the atoms of an element were physically {identical in every Way Was not correct. The discovery of isotopes solved several problems. It explained the two jparabolas observed by Thomson. It also gave a Topical explanation of the Fact 2-8 | Mass spectroscopy 49 aero eee ee emt ea Ne**) and of neon of atomic mass 22 (Ne), then there is some proportion of eee et 2-8 MASS SPECTROSCOPY ‘A detailed search for the isotopes of all the elements 3¢ elements required a more precise technique. Aston built the first of many instruments called mass spectrograph or mss petomeesin 1919 Aston introduce th ies of velocity foes, which is explained below. His first instrument could separate ions with a difference in g/m of about 1 percent, and he could determi ‘the mass to about 0.1 percent. i ro foamed these gues ’A modern mass spctrometer has dtestional rection focusing as well a velo focuing, These et can bes be exained wih refrene fo a spent ment, As an example, Fig. 2-12 shows an instrument bull perees strument built by A. O. Nier and In later work, he improved considerably upon Fig. 2-12 Principle of operation of Nie’ spectrograph, Pei ions are produced in an ion sours by im elect smite so inendecnt nent. The Tos are acre through potent Siero out 0 Wa ome ncaa ns iar se peng enon spon at 3, ih nm ta pei cf 0.0005 in. ‘The diverging bundle of rays (pace orbits) converge after Passing through an electrostatic defistor to st Sy and again after pang + LW. Aston, Phil. Mag. 38 (1918), p. 707 YAO. Nier and T. R. Roberts, Phys. Rev. 8 (1951), p. 50750 The atomic view of electricity | 2-8 through a magnetic defector to slit Sy, behind whichis sensitive current detector (electron multipin. in the magnetic deflector, one can show by simple goometry Why dretional focusing takes plae. Starting fom S) and following the ray that goes t0 the reo ke ental tay, we se that it goes a longer distance through the magnetic fiat tne the central ray does. Its therefore defected through larger ange tad consequently converges with the oentral ray. ‘The electosatc defector Srartessunlany, but the elect is stronger because the ray moving closest to Tee Secaie plats has higher potential energy, and therefore Tower kinetic nergy (aside the deflector), and is deflsted more. ational focusing as explained above, i vital in the operation ofa mas spectupaptn Good resolving power (ability to separate diferent masses) can bane only tall particles of the same mase converge to a sharp line on Sie Se and all patles ofa aiflerent mass converge to another line sptialy separated from the fit ‘We have assumed inthe discussion of diretional focusing that all particles have the same energy. Tiss appronimatly £0, because they have been accl- raed through the same poteaual (240 KV), For simplicity we assume that Si'have nit charge In Seton's fon soure, which was a gaedicharge tube, there was quite eferncr in fon energy. In a modem spectrometer extteme twa taken to produce ions of s nearly the sme energy as possible. However, Sain he exec high revolving power ofa moder instrument, it would st Te ponnbeto er considerable broadening ofthe image lines if velocity focusing were ot oe “We shal not dneuss the details of velocity focusing, but shall give an outing sfike prncple, The electrostatic defector separates particles of different kinetic energy or mo* (compare Eq. 2-12), The magnetic deflector separates particles of differeit momenta or mo (compare Eq. 2-18). For the jeauumeat in Bg 2-12, the ial postion or pont of impact ofa particle on the fan ofthe detector sit Scan then be exprested as ons?) om) mote” moty ‘Here m = mo + Am and v = 0 + Av, and a and b are constants, character ie forthe instrument. We sosume that the electric and magnetic fields TSS, ln at such that a particle with mass mg and velocity owl go through the outer of Sy (4) = 0) The fst term in Eq, (2-35) aries fom the electo- static deflector, and the second term is caused by the magnet. By partial Sierentiaons, we get from Eq (2-35) a ay @35) 236 Be aye (2-37) am im Problems 61 ‘The deflecting powers (a and 6) for the two deflectors are chosen such that y/o = 0; thatis, b = 2a. Thisis the condition for velocity focusing. Clearly, the total deflecting power for a change of mass (mass dispersion) is still nonzero so that the different masses are separated, In Fig. 2-12, slit Sy is used to limit the angle of divergence of the bundle of rays accepted by the spectrograph, and S; is used to limit the range of velocities, accepted for a given mass. Both slits have much wider openings than S, and S,. With the mass spectrometer discussed above, ions with mass differences of less than one-hundredth of 1 percent (Amim = 10") could be separated, and the differences could be measured to an accuracy of about one per million. [Newer and larger spectrographs have since been made in which both resolving ppower and accuracy have been increased by almost a factor of 100.* ‘The atomic masses determined in a mass spectrograph are always measured relative to other ions of nearly equal mass, Since #*C is the standard (soe the next section), one always attempts to measure the mass difference between an. unknown mass and a hydrocarbon compound of the same mass number:t This is called the doublet method of mass spectroscopy. An exanaple is} 8C Hy — 0 = 36381.5 + 0.9 uu (micro-mass units). This is an equation with two unknowns, the mass of Ht and the mass of O**, (The mass of 12C is 12.0000 by definition.) By measuring other doublets with different combinations of the same isotopes, possibly also involving others, one ‘can determine the masses of these isotopes on the ##C scale, often to within a fow micro-mass units, that is, with a precision of better than one point per nillion. PROBLEMS: 2-1 What is the ratio ofthe electric force on a charged particle in an electri feld of 20 Vem"? to the force of gravity on the particle if itis a) an electron? b) a proton? ) Is the weight negligible compared with the electric force? 2-2 An electron moving in a vertical plane with a speed of 5.0 x 107 ms enters a region where there is a uniform electric field of 20 V'em~* directed upward, Find the electron’s coordinates referent to the point of entry and the direction of its motion at atime 4 x 10~® slater if it enters the feld a) horizontally, b) at 37° above the horizontal, and c) at 37° below the horizontal. * See, for instance, H. Matsuda, $, Fukumoto, and T. Matsuo, Proceedings of ‘the Third Intemational Conference on Atomic Masses, R. C. Barter, editor. University cof Manitoba Press (1967), p. 733. The superscript 12 on the isotope symbol #2C is the mass number, which is equal t0 the number of nucleons (protons and neutrons) in the nucleus. The mass number of a ‘molecule is the sum of the mass numbers ofits constituent atoms, KS. Quisenberry, T. T. Scolman, and A. O. Niet, Phys. Rev. 102 (1956), 1071.52 The atomic view of electricity 2-3 If the charged particle in Problem 2-2 were a proton instead of an electron, what ‘must be the magnitude and direction ofthe electric field so that the answers for the proton would be the same as they were for the electron? 2-4 The dimensions of some parts of atypical commercial eathode-ray tube are given in Fig. 2-13. If electrons start from rest at the cathode, what is their velocity 1%, atthe origin O for an accelerating voltage of 1136 V between the anode and cathode? ia om Figure 2-13 \ 2-5 Given that the potential difference between the deflecting plates P and P* is SO V in the cathode-ray tube in Problem 2-4, a) find the y-coordinate and the direction ‘of motion of the electrons when x = . b) What is the total deflection on the screen S? 2-6 A large, plane metal plate is mounted horizontally ata distance of 0.80cm above ‘another similar horizontal plate. They are charged to a potential dilference of 40 Y, the upper plate being postive. An electron is projected horizontally with a velocity of 10° ms“! from a point O which is midway between the plates. 8) Find the x-coordinate ofthe point at which the electron strikes a plate. b) Com pute the tangent of the angle that gives the direction of the electron's motion sit strikes the plate, ) What isthe change in kinetic energy of the electron in going from O to the plate? d) What would the answer to part c) be ifthe elec- ton had no initial velocity at 0? 2-7 a) Through what potential difference would a deuterium ion have to be acceler- ‘ated from rest in @ vacuum so that it would have a speed of 847 x 10° ms ) What would have tobe the magnitude and direction of the smallest magnetic indoction that would constrain the moving deuterium fon to a circular path in an ‘evacuated toroidal tube I m in diameter? 2-8 A uniform electric field of intensity 40 x 10* V mis perpendicular to a uni orm magnetic field of flux density 2 x 107? T. (A tesla Tis a weber per square meter, Wb mr) An electron moving perpendicularly to both fields experiences zo net force. 8) Show in a diagram the relative orientation of the electric field vector, the magnetic induction vector, and the velocity of the electron. ) Cal- ‘culate the speed ofthe electron. ©) What is the radius of the electron orbit when the electric Held is removed? 2-9 A charged particle enters the region between two very large parallel meal plates, ‘the particle's velocity being parallel to the plates when the particle enters. The 210 a2 23 a4 nas 246 Problems 53 plates are separated a distance of 2 cm and the potential difference between the plates is 2000 V. When the particle has penetrated Sem into the space between, the plates, it fs found that the particle has been deflected 0.6 em. If a per ‘pendicular magnetic field of fx density 0.1 Tis impressed simultaneously with the electric field, the particle is found to undergo no defection at all. Calculate the charge-to-mass ratio for this particle. ‘A.cathode-ray tube is placed in a uniform magnetic induction B with the axis of the tube parallel to the lines of force. Given that electrons emerging from the ‘gun witha velocity v make an angle @with the axis as they pass through the origin ©, show a) that their trajectory is helix, b) that they will touch the axis again at the time ¢ = 2m,/Be,€) that the coordinate ofthe point touching is x = 2m ‘08 8/Be, and d) that for small values of 6, the coordinate ofthe point of evossing or touching the axis is independent of . ) The arrangement in this problem is, called a magnetic lens. Why? f) How do the trajectories of the electrons passing through the origin at an angle 8 above the axis difler from those directed at an. angle 6 below the axis? : Electrons are accelerated through a potential difference of 1000 V in an electron ‘gua and leave the narrow hole inthe anode as a narrow diverging beam, What ‘magnitude of axial magnetic induction is required to focus the beam on a screen 50cm from the hole. (Hint: See Problem 2-10.) ‘What isthe final velocity ofan electron accelerated through a potential diflerence of 1136 V if it has an initial velocity of 10? m s"*? ‘Two positive ions having the same charge q but different masses, m, and ma, are ‘accelerated horizontally from rest through a potential difference V. They then enter a region where there is a uniform electric field £ directed upward. 8) Show that, if the ion beam entered the feld along the z-axis, then the value of the y-coordinate foreach ion at any time is y = Ex?/4V. b) Can this arrange- ‘ment be used for isotope separation? ‘Two positive ions having the same charge g but different masses, my and mi, are accelerated horizontally from rest through a potential difference V. They then enter a region where there is uniform magnetic induction B normal to the plane of the motion. a) Show that, ifthe beam entered the magnetic feld along the a-axis, the value ofthe »-coordinate for a small deflection of each at time ris y= Beale, ) Can this arrangement be used for isotope separation? Particles with charge g and mass m are injected into a homogeneous magnetic field having induction B. When their velocities are intially perpendicular to the field the particles travel in cieular orbits, Derive an expression for the frequency of revolution of the particles and show that the frequency is independent ofthe velocity. [A charged oil drop falls 4.0 mm in 1600s at constant speed in ar inthe absence ff an electric eld. The relative density of the ol is 0.80, that of the airs 1.30 > ‘and the viscosity of the air is 1.81 x 10-8 Nsm-*, Find a) the radius of the drop and b) the mass of the drop. ) If the drop carries one electronic54 27 218 249 2-20 221 2B 24 “The atomic view of electriotty ‘unit of charge and is in an elects field of 2000 V cm, what isthe ratio ofthe force ofthe electric field om the drop to its weight? ‘When the oil drop in Problem 2-16 was in aconstant electric eld of 2000 V em=, several different times of rite over the distance of 4.0 mm were observed, The ‘measured times were 36.1, 11.5, 174, 7.55, and 23.9 s. Caloulate a) the velocity of fall under gravity, b) the velocity of rise in each case, and c) the sum of the ‘elocity in part a) and each velocity in part b). d) Show thatthe sums in part) fre integral multiples (owo significant figures) of some number and interpret this result. ¢) Caloulate the value of the electronic charge from these data, In an experiment to count and “weigh” atoms, it is found that a current of (0,800 A flowing through a copper sulfate solution for 1800 s deposits 0.473 of ‘copper. The atomic mass of copper is 63.54, its valence is 2, and the electronic Charge is 1.60 x 10-!® C, Using only the data given inthis problem find a) the ‘number of electronic charges carried by the ions which deposited as copper Atoms, b) the number of copper atoms deposited, c) the mass of a copper atom, @) the number of atoms in a gram-atomic mass of copper, ¢) the number of clectronic charges carried by a gram-equivalent mass of copper ions, f) the ‘number of Coulombs required to deposit a gramequivalent mass of copper, and {) the mass of a hydrogen atom, given that its atomic weight is 1.008. ‘What must be the direction of the electric field # and the magnetic iAdustion B in Fig, 2-10 so that the segment of the positive-ion parabola will be in a) the lower right quadrant, b) the upper right quadrant, andc) the upper left quadrant, as viewed from the right of the diagram? 42) If the ion beam in Fig. 2-10 contains two types of ions having equal charges but different masses, which of the two parabolic segments will have those of greater mass? b) Ifthe masses are equal but the charges different, which segment ‘will contain those having the larger charge? For a particular parabola in Thomson's mass spectrograms, what physical ‘quantity is different forthe ions that land close to the origin than for those landing farther away? Why does this difference exist, in spite of the fact that a constant voltage was used for the gas discharge? Io a mass spectrometer, ions having the same charge q but different masoes m, ‘and m, arecoelerated from rest through a potential difference V. A narrow beam Of these ions then enters # magnetic feld having « magnetic induction B which is perpendicular to the motion ofthe particles. Derive a simple expression in terms ‘Of the given quantities for the ratio of the radii ofthe trajectories ofthe two types | fof ions in the magnetic fet. Copper has two isotopes whose masses are 62.9 and 64.9, respectively. What is the percent of abundance of each in ordinary copper having an atomic mass of es? ‘Uranium hexafluoride, UF, is a gaseous compound at 100°C. Given a mixture of two such bexailuorides, ofe of the isotope *9*U and the other of 29°U, find the root-mean-square speed of each molecule. Comment on the possibility of separating these by a diffusion proces. CHAPTER 3 The Atomic View of Radiation 344 INTRODUCTION All sources of light consist of matter which is excited in one way or another. ‘The firefly excites his body matter by some obscure chemical process; the matter of the sun is excited by heat. But ever since Heinrich Hertz demonstrated the validity of Maxwell's theory of electromagnetic radiation, we have known that the ultimate source of radiation is an accelerated electric charge. We cannot begin the story of radiation with Maxwell, however, if we are to appreciate one of the most dramatic demonstrations of the scientific method, 3-2 PARTICLES OR WAVES Certain Greeks of ancient times argued that since a blind man reaches out to "feel his way about, the seeing man must reach out with is eyes. ‘They thought of light as a kind of tentacle emitted by the eye yet retaining contact with the eye so that information about objects touched was conveyed to the mind. Such 1 view obviously fails to explain why a man cannot see at night unless there is an outside source of illumination. It was realized long ago that light consists of something which goes out ‘from certain “sources,” bounces off objects, and may finally enter the eye. In the seventeenth century there were two views on the nature of the “something” that was bouncing about. Newton defended the premise that light “consists of a stream of fast-moving elastic particles of very small mass. His view accounted for the law of reflection, which states that the angle of incidence is equal to the angle of reflection. (This is the way perfectly clastic balls bounce from the sidewalk.) He accounted for the law of refraction by arguing that ‘when particles of light are very near any optically dense medium like glass, they are attracted to it, and this attraction increases the component of the veloci 55 .56 The stomie view of radiation | 3-3 the light travels through the medium faster than it does in free space and has its direction altered toward the normal. Christian Huygens, on the other hand, supported the view that light consists of waves. The most impressive argument in his favor at that time was that two light beams can cross through each other without “colliding.” He too explained reflection and refraction, His explanation of refraction was that when a wave front penetrates an optically dense medium at an angle, the wave moves more slowly. This slowing of the wave front causes the wave’s direction of advance to be altered toward the normal 3-3 ELECTRICITY AND LIGHT Afr a century of neglect, the undulatory theory of light was revived by the versatile English scientist Thomas Young. In 1801, he showed that only the principle of interference of waves could explain the colors of thin films and of striated surfaces. During the next half century, further experimental work, ‘especially by the French physicists Fresnel, Arago, Malus, Cornu, Fizeau, and Foucault, showed that the particle theory of light was not tenable. This work reached its culmination in 1864 when James Clerk Maxwell announced the results of his efforts to put the laws of electricity into good mathematical form, He had succeeded in this formulation and found in addition an important by-product: the laws could be combined into the mathematical form of the ‘wave equation for electromagnetic waves. He shoved, furthermore, that the ‘velocity of these waves is the velocity of light! ‘Thus in one dramatic move he ‘put the theory of electricity in order and incorporated all optics into that theory. Huygens’ view completely displaced Newton’s when Foucault found the velocity of light in an optically dense medium less than its velocity in free space, and Maxwell's theory was verified in 1888 when Hertz demonstrated that oscillating currents in an electric circuit can radiate energy through space to another similar circuit. Hertz used a circuit containing inductance and capac- itance, hence capable of oscillating. Whenever a spark jumped across a gap in the active (transmitting) cireuit, electromagnetic waves were radiated from the region in which the electric discharge occurred. (Modifications of this first ‘transmitter were used for radio communication until the advent of vacuum tubes) The passive or receiving cireuit was a loop of wire containing a gap. When energy was transferred from one circuit to the other, sparks jumped across the receiver gap. Hertz's experiments showed that the radiation generated by electric circuits obeyed the known laws of optics. It thus appeared that the theory of light was in a satisfactory and elegant state, Yet the last word on this subject had not been said. Hertz noted that the induced spark was more easily produced when the terminals of the receiving gap were illuminated by light from the sparks in the transmitter gap. This, cflect was studied more fully by one of Hertz’s students, Hallwachs, who 34 | Electrodynamics 57 showed that a negatively charged clean plate of zinc loses its charge when illuminated by ultraviolet light. Thus Hertz’s verification of Maxwell's wave ‘theory of light led almost simultancously to the discovery of the photoelectric ‘effect which, as we shall see, led in turn to a profound reinterpretation of the wave theory of radiation. 3-4 ELECTRODYNAMICS Maxwell's equations and the theory of propagation of electromagnetic waves are given in all texts on electricity and magnetism and will not be treated in detail here. A brief review is in order, however. Consider a charged particle executing sinusoidal motion with. angular frequency «, for instance, by being forced to move up and down in a vertical antenna. ‘The charge produces an electric field, E = gi4negr?, which can be measured near the antenna, but because of the r* dependenoe, diminishes rapidly as the distance increases. In addition to this electrostatic near field, the oscillating particle produces an electromagnetic field, the magnitude of which decreases only as r™!, This field moves as a wave with a velocity which in free space is ¢ = (Cy)! Here 6p = 885415 x 101? Fm? is the permittivity of free space andy = 4x x'10~" Hm” is the permeability of free space. This wave moves out from the antenna in all direetions but with an intensity that varies from zero in the direction of the antenna to maximum in the direction perpendicular to the antenna The electric and magnetic fild intensities in the electromagnetic wave are perpendicular to each other, and both are perpendicular to the direction of propagation of the wave. This is illustrated in Fig. 3-1, ia which the source, the antenna, isto the far left in the xy-plane and is centered on the x-axis, In considering atomic radiations, we replace the antenna currents with electrons revolving in atomic orbits. However, as we shal see later, the classical theory of radiation has to be modified in the atomic case. Fig. 3-1 A plane-polarized electromagnetic wave of wavelongth 2 showing the relation ofthe vectors E, and o,58 The atomic view of radiation | 3-5 ‘The energy transmitted by the electromagnetic waves in a radiation field ‘may be specified either in terms of the intensity or of the energy density of the wave motion. The intensity of the radiation is defined as the energy transmitted in unit time through a unit area normal to the direction of propagation of the waves. This is Poynting’s vector. The MKSA unit of intensity is watts per square meter. The energy density or volume density of the radiation is defined as the amount of radiant energy in a unit volume of space. The MKSA unit of energy density is joules per cubic meter. It is evident that the energy density is equal to the intensity divided by the velocity of propagation of the wave. ‘The term energy density is particularly useful in discussing the radiation within ‘a heated enclosure. 3-5 THE UNITY OF RADIATION From previous studies, the readers of this book are aware that the many forms of radiation, heat, light, radar, radio, etc, differ from one another in frequency Frogumey, Nae of Photon Wavelength See ‘dition energy, &V’ angstroms x02] sot Game roe or os —2 Xeanit, XU 10> fie — +p" 10 co) fa to ‘oa i, fp a eo, 8 10" anf 1 1o—1 nanometn,am Lesa fis — “0? 1044 Tio -}10 . ‘oH Po ef 104 itl test] 10 ae aoe x02 ——|-—— 7 oo) — or Pa aa oA center, om 1 w Shortwave 201° mater, ro 012 10191 kilometes, km oi 10 Long wave sani [on fos Fig. 3-2. The spectrum of electromagnetic radiation, 3-5 | Thermal radiation 59 bbut not in kind, The so-called “kinds” of radiation are characterized by the techniques used to produce and detect them actually, they all travel through fee space with the same velocity and should all be understood in terms of the same theory. The tremendous range of the electromagnetic spectrum is shown in Fig. 3-2 (photon energy, mentioned in the figure, will be discussed later in this chapter). The classal theory of Maxwell applies to all these radiations and all are due ultimately to the acceleration of electrical charges. Except for differences of frequency, an observation made on one “kind” of radiation ust also be true of all other kinds. 3-6 THERMAL RADIATION Information about the nature of all radiation may be obtained from a study of any of the “kinds” of radiation. We now consider the radiation from heated bodies, since that investigation has proved particularly fruitful. We all know that a'body will emit visible radiation if it is hot enough. A close relation between temperature and radiation is further implied by the fact that a white- hot body is hotter than a red-hot one. We might explore this matter further by passing the radiation from @ hot body through some dispersive instrument such as a prism or grating spectrometer. If we measure the radiant energy ‘emitted by a hot body (No. 1) for a whole series of radiant frequencies, we might obtain a graph similar to the dashed curve in Fig. 3-3. The ordinate of this ‘curve is called the monochromatic emittance, W,, which is the ainount of energy radiated per unit time per unit area of emitter in a wavelength range d2; the abscissa is the wavelength rather than the frequency. Repeating the same experiment for another body (No. 2) of a different material but at the same temperature, we might now obtain the dotted curve of Fig. 3-3. It is clear from the figure that at most wavelengths the first body is a more efficient emitter at the given temperature than the second, Although the two curves differ, they hhave the same general character. ‘They come to their highest points at about Fig. 3-3 Tho radiation spectrum of soveral hot boos.60 The atomic the same wavelength. Upon studying a great variety of substances all at the same temperature, we would obtain a great variety of emission curves, but ‘none of these would ever have a greater monochromatic emittance, Wy, than the envelope curve, shown as a solid line in Fig. 3-3. It appears that this curve ‘may have a significance which does not depend on the nature of the emitting material. Let us attempt to find or make an emitter which has an emission ccurve identical to the solid curve of Fig. 3 3-7 EMISSION AND ABSORPTION OF RADIATION It may be wondered why itis, ifthe surfaces of all bodies are continually emitting radiant energy, that all bodies do not eventually radiate away all their internal nergy and cool down to a temperature of absolute zero. The answer is that ‘they would do so if energy were not supplied to them in some way. In the case of the filament of an electric lamp, energy is supplied electrically to make up for the energy radiated, As soon as this energy supply is cut off, these bodies do, in fact, cool down very quickly to room temperature. The reason that they do not cool further is that their surroundings (the walls, and other objects in the room) are also radiating, and some of this radiant energy is intercepted, absorbed, and converted into internal energy. The same thing is true of all other objects in the room—each is both emitting and absorbing radiant energy simultaneously. If any object is hotter than its surroundings, its rate of emission will exceed its rate of absorption. There will thus be a net loss of energy and the body will cool down unless heated by some other method. Ifa body is at « ower temperature than its surroundings, its rate of absorption will be larger than its rate of emission and its temperature will rise. When the body is at the same temperature as its surroundings, the two rates become equal, there is ‘no net gain or loss of energy, and no change in temperature. Figure 3-3 shows that the emittance of a surface Wis different at different wavelengths and the paragraph above implies that it is greater for higher ‘temperatures. To simplify our next discussion consider an infinitesimal band of wavelengths and several opaque bodies in thermal equilibrium with each other and their surroundings. Because the bodies are opaque they will not ‘transmit radiation and therefore, in general, part of the incident radiation will be reflected and the remainder will be absorbed. The fraction absorbed, called the absorptance a, plus the fraction reflected, called the reflectance r, must be unity of, since the surfaces may be different, tne ath ete on since the bodies are in thermal equilibrium with their surroundings, they will be “bathed” in radiation of uniform intensity, If this is not obvious, it may be helpful to think of the bodies as being tiny specks near each other ‘but too small to cast shadows on each other. We can now write the last sentence 3-8 | Black-body radiation 61 quantitatively, The total radiation in a time Ar from body No. 1, which has fan area AA, and a radiant emittance WV, is W, MA, At. The absorption by the same body in the same time is a,fA4, At. For the condition of thermal equilibrium to exist these must be equal. ‘Therefore we have W, OA, Bt = al AA, At, GoD and similarly for another body, Wy bds At = ap! Ada Bt ete 6) Dividing Eq. (3-2) by Eq. (3-3), we obtain Wri op Waa, oe tate eet aeTiat(o ates a Since the number of specks or kinds of surface has not been restricted, it becomes evident that W/a for any substance must be a constant (which may, of course, still depend on wavelength and temperature). ‘We have just proved that a body or surface which is a good emitter (high value of W) must be a good absorber (high value of a) and conversely. If wwe could find a perfect absorber, we would necessarily have found the best possible emitter, the graph of which is shown as 6 in Fig. 3-3. 2-8 BLACK-BODY RADIATION In acoustics, an open window is taken to be a perfect absorber of sound, since an ‘open window refleets virtually no sound back into the room. In optics, there are few things darker than the keyhole of a windowless closet, since what litte light gets into the closet bounces around against absorbing surfaces before it is redirected out the keyhole, Painting the inside of the closet black may increase the darkness of the keyhole, but the essential darkness of the hole is due to ‘the geometry of the cavity rather than to the absorptivity of its surfaces. A. small hole in a cavity of opaque material is the most perfect absorber of radiant energy man has found, Conversely, a small hole in a cavity is the most perfect ‘emitter man has devised. We can conclude this from the proof above or we ‘can understand it more thoroughly from the following. If we look into a hole in heated cavity, we can see the radiation from the inside wall just opposite to the hole. In addition, we see some radiation from other parts of the inside of the cavity which was directed toward the spot of wall we are looking at and is reflected to us by that spot. The absorption and emission of radiation ‘by @ hole in a hollow tungsten cylinder is shown in Fig. 3-4. ‘The light streak across the center of the figure is an incandescent filament maintained at a constant color temperature for comparison purposes. When the cylinder is cold, the hole is darker than any other part and actually appears black, but when the cylinder is heated sufficiently, the hole is brighter than the body of62 The atomic view of radiation | 3-8 Fig. 9-4 Radiation from a hollow tungsten cylinder. the tube and matches the reference filament. Such a hollow absorber-emitter is called a black body. Since a is equal to unity for black body, from Eq. (3-4) we obtain HW This relation is called Kirchhoff's law of radiation: The ratio of the radiant emittance of a surface 10 its absorptance is the same for all surfaces at a given temperature and is equal to the radiant emittance of a black body at the same temperature. ‘We now discuss the multiple-reftection situation of a radiant cavity quantitatively. Consider the radiation traveling back and forth in an isothermal cavity formed between two plane parallel sheets of different materials, as shown in Fig. 3-5. Let us now follow the history of the radiation emitted from a unit os Fig. 9-8. Radiation traveling betwoen two diferent materials 348 | Black-body radiation 63 area of each face in a time interval At, which is just long enough to permit the radiation to travel across the space to the other face. It is reflected there with some loss of energy, returns to the first face is reflected there with further loss, and so on, Table 3-I gives the values and the direction of travel of the com- Table 24 Initially emitted by AY, Initially emitted by Wat war == aihAt = = a = a,)ih,ar = al = ay Ar == a8 ~ 05)? he = ah = a3))HAr | a a= a)Wbe + = a ~ ade d= ad ~ a)tyAe 2 = ad ~ ap ibr = = @° = 37d a ponents in the radiation streams per unit area between the walls after several intervals of Ar have elapsed, ‘While these successive traverses of the initial radiation from each face are ‘occurring, both faces continue to emit. Therefore, when the steady state has ‘been reached, there are simultaneous columns of thermal radiation going back and forth in the space between the faces. The total radiation streaming in one direction, say to the right, is simply the sum of all the components in the direction +. ‘Therefore, the total effective emittance toward the right is given by W, At = W, At + (1 ~ a4) (I = a3), At + (1 a)°(1 ~ a) Or fest = a)W, Ar + (l= a)*(1 = a3) Oe + (1 = a9 = 3)? Ate Co Let x = (1 —a,)(I — @,). Then substituting this in the preceding equation we obtain WeeWIlt xt rte} + md - al tx4 tee} BD “The series within the pairs of brackets is a simple geometric progression whose Timit, since 0 < x < I, is 1/{1 — x). In terms of the absorptances, this limit is Oe 2371-0 and ad Waa + aay o-8)64 ‘The atomic view of radiation | 3-6 Because the system is isothermal, we can obtain the following relations from Kirchhoff’s radiation law (Eq. 3-5): Wo= ai, and W = aM os) Substituting the values from Eq. G-8) and Eq. (3-9) in Eq. (3-7), we have uta = 41) yy, G10) a, F aa = 8s “This equation shows that the radiation to the right (it could just as well have been in any other direction) is effectively radiated from the left surface as ‘though from a black body. If a tunnel, which is so small it does not subtract ‘a significant portion of the radiation in the cavity, is bored through the right- hhand face, then the leakage radiation will be black body-radiation. It is to be noted that the derivation contained no assumptions about either the nature of thermal radiation or of the kinds of surfaces inside the enclosure. ‘We now know how to make a black body and have achieved the goal we set at the end of Section 3-6. "The reader can demonstrate for himself that black bodies can even be made from bright objects. A bundle of sewing needies held with their points directed toward the eye looks remarkably black, A pile of razor blades at least Yr-inch thick also looks black when viewed from the sharp side. In these cases the incident radiation is completely absorbed as a result of all the partial absorp- tions experienced at the many successive partial reflections it undergoes in traveling down into the relatively deep, narrow spaces between the needles or the blades. We now return to the question of the spectrum of the radiation emitted ‘by a hot body. If we take a black body as our sample, we can measure the emission from the hole as we did the material samples in getting the data for Fig. 3-3. This experiment shows that the emission of the black body gives at once the smooth solid curve of Fig. 3-3, which, unlike the other curves in the figure, is independent of the material used to make the emitter. This confirms what we might have suspected before, that the solid curve portrays a general Gharacteristic of thermal radiation at a given temperature. A study of this curve Should give information about radiation itself. With consideration of the taterial composing the cavity eliminated, the remaining important variable is the temperature of the radiation source. Mathematically, the total energy radiated per unit time pet unit area of emitter is proportional to the area under the curve, and Stefan found empirically that this area is directly proportional to the fourth power of the absolute temperature, W, = ort, on where o is called the Stefan-Boltzmann's constant. Its value is 6 = (5.6696 + 0.0010) x 10°* Wm"? *K~* 3-9 | Wien end Rayleigh-Jeans' laws 65 Equation (3-11) was first derived theoretically by Boltzmann and is therefore called Stefan-Boltzmann’s law. In the following two sections, we shall derive the equation giving the spectral distribution of black-body radiation (Planck's Jaw). Equation (3-11) can then be found by integration (Problem 3-5). Boltz- mann derived the Jaw by a thermodynamic argument™ before Planck's work. ‘A body that is not black will radiate less energy than the black body. Combining Eqs. (3-5) and (3-11), we get, in general, W = aT". G12) ‘The coefficient a, called socal emissivity or absorptance, is tabulated in all standard handbooks of physics. The thermodynamic method employed by Boltzmann can be pushed a litle further in predicting properties of the spectral distribution of black-body radiation, Wien found that as the temperature of any black body is changed the curve retains its general shape, but that the maximum of the curve shifts with temperature so that the wavelength of the most intense radiation is inversely proportional to the absolute temperature, oF Jaq = const/T. This is a special case of Wien's displacement law, which states that at corresponding ‘wavelengths the monochromatic energy density of the radiation in the cavity of a black body varies directly as the fifth power of the absolute temperature. The relation defining corresponding wavelengths at temperatures T, and T, is 4,7, = 22T;. The displacement law enables us to predict the entire curve ‘at any temperature, given the entire curve at one particular temperature. Neither of these radiation laws, however, treats the basic problem of why the energy radiated from a black body has this particular wavelength distribution. 3-8 WIEN AND RAYLEIGH-JEANS’ LAWS ‘A comparison of the experimental black-body radiation curves of Fig. 3-6 and the Maxwell distribution of speeds in a gas shown in Fig. 1-2 shows a remarkable similarity. Wien noted this similarity and tried to fit a function such as Maxwell had derived for the speed distribution to the black-body wavelength distribution. ‘There is more than the similarity of the curves to justify this approach. If the molecules of the black body are thermally agitated, then their distribution of speeds may be somewhat like that derived by Maxwell. The accelerations of these molecules should be related to their velocities. These molecules contain charges which are therefore thermally accelerated, and we have shown that classical electrodynamics indicates that radiation results from accelerated charges, This argument is hardly rigorous, but it is a plausible explanation of the relationship between the similar curves. The expression that Wien obtained + See, for instance, F. K, Richtmyer, E. H. Kennard, and T. Lauritsen, Introduction 10 Modern Physies, New York: McGraw-Hill, 1955).66 The atomic view of radiation | 3-9 180% 1449" fe energy density ¢y, relative Monochrom esas oe “Evite Wavelength, microns Fig. 3-6. The distibuion of ‘Siferent temperatures gy in the spectrum of the radiation trom a black body st for the monochromatic energy density y, within an isothermal blackbody enclosure in the wavelength range 2 to 2 + di is a vedi = Se aa, G13) ‘where 2is the wavelength, Tis the absolute temperature, and e isthe base of natural logarithms. This formula is essentially empirical and contains two adjustment constants c, and cy, called the first and second radiation constants, respectively. Wien chose these constants so that the it he obtained was rather ood except at long wavelengths. The graph of Wien's law is shown dashed in Fig. 3-7. The points represent the experimental data of Lummer and Pringsheim.* Buta “pretty good ft” is not good enough, and a formula which is essentially empirical tells us nothing about the nature of radiation, ‘Lord Rayleigh set out to derive the radiation distribution law in a rigorous sway. He argued that the law of equiparttion of energies (Section 1-8) should apply to the electromagnetic field in an enclosure. A standing wave of « given frequency interacts with oscillating atoms in the wall. Energy can flow in either direction from the wall to the field or conversely. The “temperature” of the field is therefore the same as the temperature of the walls. A one- dimensional harmonic oscillator has one degree of freedom representing its + Lammer and Pringsheim, Verkandl, Dew, Physik. Ges. 1 (1899), pp. 23 and 2153 2 (1900), p. 163. 38 | Plonek’s awe 8, Wiew's awe Wavelength, microns Fig. 3-7. Graphs of the racition laws, Points roprasent experimental data, kinetic energy, and, since the average potential energy equals its average kinetic energy (GkT), the total average energy is kT. A standing wave of electromagnetic radiation with given direction of polarization represents a one- dimensional oscillator and should therefore have an average energy of kT. ‘This is the basis for the derivation of Rayleigh-Jeans’ radiation law. The law gives the energy content in an. isothermal enclosure in a wavelength interval 2to 2 + di, To derive it, we need to know how many different standing waves, called modes of radiation, there are in this wavelength interval. Each mode is then assumed to have an average energy of KT. ‘Let us first look at the simpler problem of calculating the number of modes of oscillations in a given frequency interval on a vibrating string. The x-axis is along the string, which is clamped at x = O and x = L. We know from ‘mechanics that standing transverse waves can be expressed as y = Asin kx sin of, G-14) with, KL = nx, sy where n is a positive integer. The wave number k, the wavelength 2, the angular Frequency @, the frequency », and the wave velocity c on the string are related by the following equations: k= trp, w=2m, c= fk =x G19 A mode of oscillation is by definition a standing wave with a given value of n. Our objective now is to count the number of possible values of m in a given frequency interval. We simply differentiate Eq. (3-15) and find G-1n62 The atomic view of radiation | 3-9 By substituting dk = —(2n/?2) di, and neglecting the minus sign, we can also write an = aa. Gs) For example, if the string is L = 100 em, Eq. (3-18) states that there can be 20 standing waves of different wavelengths in the interval 2= 1 em to 24 di= liom. Let us now attack the three-dimensional problem, which, in principle, is very like the one we have discussed, We shall assume that the walls are per- feetly reflective. If this were stritly teue, there would be no exchange of energy between the wall and the electromagnetic field. However, we can imagine that the walls are almost perfectly reflective, and this modification will not change ‘our calculation of the number of modes. Assume that the cavity is cubical with sides L. We can then find solutions of Maxwell's equation of the form Ey = Eq cos k,x sin kyy sin kz sin of, G19) with, (ed + 1 + BY”, (3-20) “The expressions for Ey E,, Ha Hy and H, are similar to Ea. 3-19). This is the three-dimensional version of EQ. (3-14), The a-dependence is a cosine funetion in Eq, (3-19) but will be a sine function with the same wave number ki, in the expression for E, and E,, The boundary conditions are now such that the components of the E-veetor parallel to a given surface must be 2570 at that surface. This gives kL = nm, kL =n, 21) kL =n, where ny np and n, are positive integers. Combined with Eq. (3-20), this gives = oak + nh + nD. om ‘A given set of integral values of 7, n,, and n, represents a mode of electro~ magnetic oscillation with @ given set of node planes. Another set of integers rig My m, corresponds to a different set of node planes. For each set there can. be two independent directions of polarization. There are therefore two independent modes for each set 1g, My, and 1, We now ask how many different sets of rn,» and n, result in an angular frequency below a given value @ (Eq. 3-22). ‘The answer can be found most ‘easily by plotting in a cartesian coordinate system the values of ny, %,, and tte along the three axes (Fig. 3-8). Each volume element in the first quadrant of this coordinate system contains one point corresponding to two modes of electromagnetic radiation, The positive numbers n,, mand n, are components of a vector m, the maximum length of which is limited when there is a limit for the angular frequency «@, The relationship between the two is given by Eq. (6-22) as Figg, = OL. G23) ‘The total number of points, each of which represents two modes of radiation, can then be found by counting the number of unit volumes in the first quadrant ‘within the distance mag from the origin. This is equal to one-eighth the volume Fig. 3-8 Eoch point in this diagram represents s possible sot 5 ay 998 Pa of a sphere with radius mya. The number of states with angular frequency lower than @ is therefore 14 (2 @V _ 8 eon 4, (2) - oF Se 33" \ne) ~ 3nd Sa” where V = Lis the volume of the box. The number of modes in the wavelength interval between 2 and 2+ di is found by differentiation. Neglecting the minus sign, we find that aN = & aw. 6-25) ‘The energy density is now simply the number of modes per unit volume multi- plied by kT: var = SAT an, 6-2670 The stomic view of radiation | 310 This is Rayleigh-Jeans’ law, which is plotted in Fig. 3-7 for comparison with the experimental data. At first glance this law appears vastly inferior to Wien! Although it fits well for long wavelengths, at short wavelengths or high frequencies it heads toward infinity in what has been dramatically called the “ultraviolet catastrophe.” Theoretically, however, the Rayleigh law must be taken far more seriously than Wien’s. It was derived rigorously on the basis of classical physics. It involves no arbitrary constants, and where it does fit the experimental curve, it fits exactly. Whereas the failure of the Wien law ‘was “too bad,” the failure of the Rayleigh law presented a crisis. It indicated that classical’ theory was unable to account for an important experimental ‘observation. ‘This was the situation to which Max Planck directed himself. $10 PLANCK'S LAW: EMISSION QUANTIZED Planck's first step was essentially empirical. He found that by changing Wien's formula slightly to G27) and by adjusting Wien's constants, he got a formula that reduced to the Rayleigh formula at long wavelengths and that fitted the experimental curve everywhere, He knew that he had found a correct formula and that it ought to be derivable. Planck’s position was a little like that of a student who has peeked at the answer in the back of the book and is now faced with the task of showing how that answer can be logically computed. Planck tried by every method he could conctive to derive this correct formula from classical physics. He was finally forced to conclude that there was no flaw in Rayleigh’s derivation and that the flaw must lie in classical theory itself. Planck had to eliminate the “ultraviolet catastrophe” which came into Rayleigh’s derivation because of the assumption that the radiation standing ‘waves had an infinite number of modes of vibration (Eq. 3-24 with 1 ~ 0). Each of these was assumed to have an average energy KI’ and therefore the total energy content of the radiation field would be infinite. This is clearly unreasonable. Perhaps the high-frequency modes are not excited, or are they harder to excite? Let us take another look at the assumption of equipartition. In effect, Rayleigh assumed that the energy content in one mode of oxillation could have any value from zero up and that the probability of finding an oscillator with energy £ is given by Boltzmann's formula (Eq. 1~44). This gives an average energy ] p= (EEO aE jem ge 3-28) 340 | scion quantized 71 ‘Another way of expressing this is to say that the modes can be in different ‘nergy states. Rayleigh assumed a continuum of states with a population ‘proportional to the probability given by Boltzmann's law (see Fig. 3-9) z e Bair bit Papyetit au au 0 0 “Probability Probability @ oy Figs. 3-9a and 3-96 (a) Probebilty P of finding an energy content Ein a given made of ‘ecilation according to classical theory (Rayleigh). () (2), butonergy content can ‘nly take on ciserete values, & = mu (Planck). As drawn, tis fs an energy-fevel diagram “tit the horizontal coordinate indicating rlauve population in each level. Planck postulated that each mode could only receive energy from the walls in discrete steps u. The energy content of one mode would then be 0, u, 2u, 34 etc, or in general mu, where m is an integer. Let us see what this line of reasoning leads to. (The method at this point is following a procedure used by Boltzmann in 1877 to determine the distribution of kinetic energy among the molecules in a gas.) The number of oscillators having the energy m, is gem G29) ‘The energy contributed by the ”, oscillators is obviously muy, = munge™, 30) ‘Therefore the average energy E of an oscillator is : — E = Tazo munoe esp xno Moe Since m is an integer, Eq. (3-16) becomes Bn OE MEM 4 2ue™ OT + Bem OMT 4 + om TP OWT Gg oT gwaT poe72 The atomic view of radiation | 310 Let x “7, Then Eq, (3-17) can be written Bet + 33) ‘The limits of these convergent series can be found by the usual methods (note that x < I). The convergence limit of the series in the numerator is 1/(1 — x)?. This can be checked by expanding (1 — x)? according to the binomial theorem. The denominator is a simple geometric progression converging to I/(1 — 2). Substituting these limits in Eq. (3-33), we have Wax? _ owe Le id= 3) Tx” Gi) -7 When x is replaced by its equivalent, the result is G34) (3-35) If we now multiply Eq. (3-35) by the number of modes of vibration in a unit volume of cavity space, from Eq. (3-25), we obtain the energy density in a wavelength range di, ped ai Ber Recall that in this derivation the energy of an oscillator has been assumed to be an integer, m, times some small energy, u. Classical physics says the ‘energy may have any value. This is equivalent to saying u may be exceedingly small and, in the limit, approach zero, If we set u = 0, Eq. (3-35) is indeter- ‘inate, 0/0. If we apply 'Hospital’s rule, differentiating both numerator and denominator with respect to u before letting w = 0, we find that B= kr, 37) which is in complete agreement with Rayleigh's classical assumption of continuous energy. As we have seen, however, this assumption does not lead to the correct radiation law. The relation given in Eq, (3-36) begins to look like Planck’s empirical law (Eq. 3-27, if wis not zero. Indeed, the denominators of these two equations ‘become identical (except for the minus one) if @ value of u is chosen so that the powers of the exponential terms are the same. ‘To obtain this value of u, we let %, a. 636) (3-38) 39) 3:10 | Planck's law: emission quantized 73 In this last equation, ¢ is the free-space velocity of light and is the frequency of the oscillator and therefore also the frequency of the radiation it emits. If ‘we replace the constants (cgk/e) by another constant f, we have he ua hy, 3-40) Fi G40) When the value of w from Eq. (3-40) is substituted in Eq. (3-36), we obtain Planck’s law for the energy density of black-body or cavity radiation. This lawis Bxchi* Wa = a oan ‘This equation does agree with the experimental results, It is plotted in Fig. 3-7 (solid curve) for a temperature T = 1646 °K. ‘When we introduced the idea of discrete energy steps u, we did not indicate that u would be different for the different modes. However, this is what our analysis has shown to be the case. Going back to Fig. 3-5(0), we can now ‘explain qualitatively how Planck avoids the ultraviolet catastrophe. For small ‘wavelengths, high frequencies, the distance from E = 0 to the first discrete nergy level u is so large that the probability for a mode to exist in that state of excitation becomes diminishingly small. All modes of very high frequency therefore have E = 0. This is also immediately evident trom Eq. (3-35). ‘The new constant Ais called the Planck constant. We have seen that it ‘could be determined from Wien’s constant c, but it can also be evaluated from the photoclectric effect discussed later in this chapter. Its value is = (6662620 + 0.00005) x 10°°* J-s. (Note that the units are those of angular momentum.) ‘Thus Planck was led to his startling, nonclassical assumption thet the energy states of an oscillator must be an integral multiple of the product of the constant h and the frequency v of the electromagnetic radiation it emits. If E represents the smallest permissible energy change, Planck’s famous quantum* equation is E=W. 4) Planck introduced the quantum concept in 1900, and it eventually led 10 the conclusion that radiation is not emitted in continuous amounts but in discrete bundles of energy each equal to hv. These bundles or packets of radiant energy are now called quanta or photons. This was the-beginning of the atomic theory of radiation, which has grown to become the quantum theory. It is obvious, however, that quanta of radiation of different frequencies have different “izes” (energies), and that they are atomic only in the sense that they are * Quantum is the Latin word for how muck or how great.74 Thestomle view of radiation | 3-11 discrete, Planck thought at first that his ad hoe® hypothesis applied only to the ‘oscillators and, possibly, to the emitted radiation in their immediate neighbor- hhood and that, at most, it was a slight modification of Maxwell's theory of radiation. However, we shall ce that he initiated a series of events which have changed our whole concept of the interaction of electromagnetic radiation wwith matter. 3-11 PHOTOELECTRIC EFFECT ‘We now turn from thermal radiation to another portion of the electromagnetic spectrum and consider an effect which is due to radiation of higher frequency. ‘We mentioned earlier that even before the discovery of the electron, Hallwachs observed that zinc irradiated with ultraviolet light lost negative charge. He proposed that somehow the radiation caused the zinc to eject negative charge. Tn 1899 Lenard showed that the radiation caused the metal to emit electrons. ® Quarts Fig. 9-10 Apparatus for investigating the photoslectrc effect. This phenomenon, called the photoelectric effect, can be studied in detail with the apparatus shown in Fig. 3-10. In this figure, S is a source of radiation of variable and known frequency v and intensity Z, F is an emitting electrode of the material being studied, and C is a collecting electrode... Both electrodes are ‘contained in an evacuated glass envelope with a quartz window that permits the passage of ultraviolet and visible light. The electric circuit allows the electrodes to be maintained at different known potentials and permits the ‘measurement of any current between the clectrodes. We first make the collecting electrode positive with respect to the emitting electrode, so that any ‘electrons ejected will be quickly swept away from the emitter. About 10 V is enough to do this but not enough to free electrons from the negative electrode by positive-ion bombardment as was the case in the early cathode-ray tubes. If the tube is dark, no electrons are emitted and the microammeter indicates no ccurrent. If ultraviolet light is allowed to fall on the emitting electrode, electrons * Ad hoc means lieraly to this, and is used to describe a hypothesis which is appli= cable to but one (this) situation, 311 | Photoslectric effect 75 are liberated and the current is measured by the microammeter. It is found that the rate of electron emission is proportional to the light intensity. By holding the frequency y of the light and the accelerating potential V constant, wwe can obtain data like that represented in Fig. 3-11. Photodestrie current Light intenaiey, Fe Fig. 9-11 Photoolectrc current as funetion of tha intonsty of the light, ‘The frequency ‘ol the light and the accolrating potential are kept constant. It is hardly surprising that if a little light liberates a few electrons, then more light liberates many. If we vary either the frequency of the light or the ‘material irradiated, only the slope of the line changes. ‘We can now experiment by keeping the light intensity constant and varying ‘the frequenoy. The graphs of these data are shown in Fig. 3-12, where 4 and B represent two different irradiated materials. The significant thing about these curves is that for every substance irradiated there is a limiting frequency below which no photoelectrons are produced. This frequency, called the threshold frequency, Yo, is a characteristic of the material irradiated. The wavelength of light corresponding to the threshold frequency is the threshold wavelength, 4g. No photoelectrons are emitted for wavelengths greater than this. ‘The existence of a threshold frequency is difficult to explain on the basis, of the wave theory of light. If we think of light as consisting of a pulsating electromagnetic field, we can imagine that that field is sometimes directed so as to tend to eject electrons from a metallic surface, We might even feel it Photoletse current Frequency —= 3.42 Photoslectrc current for two matorsle ab a function of the frequency of the Tih. The intensity ofthe light and the eccelorating potential ae kept constant.76 The atomic view of radiation | 3-11 reasonable that certain frequencies of light would resonate with the electrons of the metal so that, for a particular metal, there might be preferred light frequencies which would cause emission more efficiently. The striking thing about these data is that for each material there is a frequency below which no photoelectrons are emitted and above which they are emitted, This effect is independent of the intensity of the light. Tn 1905 Einstein proposed a daring but simple explanation. He centered attention on the energy aspect of the situation. Whereas Planck had proposed that radiation was composed of energy bundles only in the neighborhood of the emitter, Einstein proposed that these energy bundles preserve their identity throughout’ their life. Instead of spreading out like water waves, Einstein ‘conceived that the emitted energy bundle stays together, and carries an amount of energy equal to Av. For Einstein, the significance of the light frequency was not so much an indication of the frequency of a pulsating electric field as it was ‘a measure of the energy of a bundle of light called a photon. His interpretation of the data of Fig. 3-12 would be that a quantum of light below the threshold frequency just does not have enough energy to remove an electron from the metal, but light above that frequency does. ‘The threshold frequency is dependent on the nature of the material irradiated because there is for each material a certain minimum energy necessary to liberate an electron, The photoelectric work function ot threshold energy, Wo, of a material is the minimum energy required to free a photoelectron from ‘that material In a third photoelectric experiment let us hold both the frequency and intensity of the light constant. Tho variable is the potential difference across the photoelectric cell. Starting with the collector at about 10 volts positive, wwe reduce this potential to zero and then run it negative until the photocurrent stops entirely, Curve 1, of Fig. 3-13 shows the type of curve we might expect for this particular substance. This curve requires careful interpretation. Photolet| 3.41 | Photoslectric effect 77 ‘When the potential difference across the tube is about 10 V or more, all the emitted electrons travel across the tube. This stream of charges is called the saturation current, and it is obvious that an increase in the potential of ‘the collector cannot cause an increase in current. As the accelerating potential js reduced from positive values through zero to negative values, the tube current reduces because of the applied retarding potential. Eventually this potential is large enough to stop the current completely. “The stopping potential V, is the value of the retarding potential difference that is just sufficient to halt the most energetic photoelectron emitted. Therefore the product of the stopping potential and the electronic charge, Ve, is equal to the maximum kinetic energy that an emitted electron can have. Since this stopping potential has a definite value, it indicates that the emitted electrons have a definite upper limit to their kinetic energy. Doubling the intensity of the light doubles the current at each potential, as in J; of Fig. 3-11, but the stopping potential is independent of the intensity. Stopping potential, Ve Fig. 3-14 Stopping potenti as 8 function of the frequency of the light. Results sre Indapandent of intone If, however, the experiment is repeated with a series of different light frequencies, it is found that the stopping potential increases linearly with the frequency. This is best shown by plotting the stopping potential against the frequency, as shown in Fig. 3-14. Below the threshold frequency no electrons are emitted and the stopping potential is of course zero; however, as the frequency is increased above the threshold, the stopping potential increases linearly with the frequency. To sec how fully the data of Fig. 314 confirm the Einstein photon interpretation of the photoelectric effect, we now interpret the graph of Fig. 3-14 as he would have. For light frequency between zero and the threshold fre~ queney there are no photoelectrons produced, since the incident photons have -—— 0 ——+ Accleating potential, V Fig, 3-13. Photostecric current 98 @ function of the accelerating potential for light of ferent imeneiies having a two-t0-one rato. The frequency ofthe light is constant.78 ‘Theatomic view of radiation | 3-11 less energy than the work function of the material. For light above the threshold frequency, photoelectrons are emitted. ‘The energy of these emitted electrons may vary greatly. According to the Einstein view, however, there must be an upper limit to the energy of the emitted photoelectrons. No photoelectrons ‘can have energy in excess of the energy of the incoming photon less the minimum energy to free an electron, the work function. Since the photon energy is proportional to frequency and since the stopping potential is a measure of the maximum kinetic energy of the emitted photoelectrons, the ‘graph of stopping potential against frequency should be a straight line. This is precisely what Fig. 3-14 shows. The quantitative check of the Einstein interpretation is that the slope of the straight line provides a method of determining the way in which photon energy depends on photon frequency, which is the Planck constant. ‘A final and decisive experiment consists of making the irradiating light extremely weak. In this case, the number of photoelectrons is very small and special techniques are required to detect them. The significant result of this ‘experiment is that dim light causes emission of photoelectrons which, however few, are emitted instantaneously and with the same maximum kinetic energy as for bright light of the same frequency. According to the wave theory of light, pulsating electromagnetic fields spread out from their source. Dim light corresponds to waves of small amplitude and small energy. If dim light spreads over a surface, conservation of energy requires that either no photoelectrons should be emitted or the electrons must store energy over long periods of time before gathering enough energy to become free of the metal, The fact that high-energy photoelectrons appear immediately can be explained only by assuming that the light energy falls on the surface in concentrated bundles. According to Einstein, the dim light consists of a few photons each having energy depending only on the light frequency. This energy is not spread over the surface uniformly as required by the wave theory. A photon that is absorbed gives all its energy to one electron, and that electron will be emitted violently even though the number of such events is small. ‘Ifa man dives into a swimming pool, his energy is partly converted into ‘waves which agitate other swimmers in the pool. If it were observed that when ‘a man dives into a pool another swimmer is suddenly ejected from the poot ‘onto a diving board, we would be forced to conclude that the energy provided by the diver did not spread out in an expanding wavefront, but was somehow transferred in concentrated form to the ejected swimmer. A swimming party held in such a superquantum pool would be a very odd affair compared with cone in a classical pool. To summarize Einstein's interpretation of the photoelectric effect, we ‘equate the energy of an incident photon of frequency v to the sum of the work function of the emitter (W¥ = fy) and the maximum kinetic energy that an 343 | Theelectron volt: photon energies 79 jected photoelectron can acquire. We then have fv hag + Ariba 08 Sm hy = hive, Vee = hv ~ Wo, G4) ‘This is a linear equation. It is the equation of the graph in Fig. 3-14. It is evident that the slope of the curve ¥, plotted as a function of v is equal to Ale, that vo is the vaxis intercept, and that Wole is the intercept of the extra~ polated curve on the V;-axis. Einstein’s photoelectric equation was first verified ‘with precision in 1916, eleven years after it was proposed, by Millikan, who made careful measurements of the photoemission from many different substances. 3:12 SUMMARY OF THE ATOMIC VIEW OF RADIATION We introduced this chapter by outlining the disagreement between Newton and Huygens over the nature of light. We described how Maxwell strengthened Huygens’ wave theory when he showed that electromagnetic waves were consequence of the laws of electricity and magnetism. We reported that Hertz demonstrated that electric circuits could be made to produce the electromagnetic ‘waves Maxwell predicted. But we also mentioned that Hertz observed that he could produce sparks more easily when his spark gap was illuminated. ‘Thus Hertz's work, which supported Maxwel’s theory, also contained the first observation of the photoelectric effect. Although’ the wave theory of light accounts beautifully for many optical phenomena, it fails to account for either the black-body radiation or the photoelectric effect, where light appears to possess marked particle aspects. It is hardly satisfactory to regard light as a wave motion part of the time and @ particle phenomenon at other times. We shall return to the question of the resolution of this conflict of viewpoint in Chapter 5. At this point itis clear, however, thatthe resolution ofthis paradox can never eliminate the idea that light is emitted and absorbed in bundles of energy called photons, and that now radiation must join matter and electricity in having a basically stomic character. The fact that radiant energy is quantized is a radical departure from classical physies and will require us to re-examine the whole eneray concept fram the quantum point of view. Establishing this fact has been the main business of this chapter. Before closing the chapter, however, ‘we shall consider some related topics. 3.18 THE ELECTRON VOL HOTON ENERGIES In the Einstein equation, we measured the maximum kinetic energy of the emitted electrons by noting the potential energy difference (e¥.), which was\ i i | ' \ | 80 The stomic view of radiation | 3-43 equivalent to the electron kinetic energy. This method of determining and expressing clectron energies is a particularly convenient one, and it suggests a new unit of energy. This new unit of energy is called the electron volt, eV, which is defined as the amount of energy equal to the change in kinetic energy of one electronic charge when it moves through a potential difference of one volt, Sinoe the electron volt is an energy unit i isin the same category as the joule, the foot-pound, the British thermal unit, and the kilowatt-hour. Energies in joules can be converted to electron volts by dividing by €.= 1.60 x 107%. In this case e, is not a charge but a conversion factor hhaving the units of joules per electron volt. The Einstein equation, Eq. 3-43), is valid in any consistent system of units. If we choose MISA units and divide Eq. (5-3) by the factor e,, we obtain the same relation in electron volts: hy _ Wo e-44) Exioas) (numerically equal to ¥,) = In words, this equation states that the maximum kinetic energy of a photo- clectron in electron volts equals the energy of the photon in electron volts ‘minus the work function in electron volts. When the electron volt is too small a unit, it is convenient to use 10* eV = 1 keV (kilo-electron-volt), and 10° eV = 1 MeV (million-clectron- volt). Tt is also useful to express photon energies in electron volts. In terms of photon frequency, 6:63 x 107 eV = 414 x 0-MyeV, 3-45) Or, since ¥ = c/A, we get 4 x 107" x 3 x 10 a ae 10-t 49 with 2 in meters. If the wavelength is expressed in angstroms instead of meters, wwe have E= v gm LAX 10 ey, 47) We shail use Eq. (3.47) frequently. Example. Light having a wavelength of 5000 A falls on 2 material having a photoelectric work function of 1.90 eV. Find (a) the energy of the photon ia €¥, (b) the kinetic energy of the most energetic photoelectron in eV and in joules, and (c) the stopping potential. 3-415 | Specific heat of solids 81 Solution. 1) From Eq. (3-47), E = 4Bpf = 247 eV. 'b) The law of conservation of energy gives maximum kinetic energy = photon energy ~ work function, By = 247 eV ~ 1.90 eV = 057 eV. Also Ey = 057 x 1.60 x 107! = 9.11 x 107295, O57eV___ 2 gs7v. o Y= T leet change 3:14 THERMIONIC EMISSION “The subjects discussed in this and the next section are, strictly speaking, not parts of “The Atomic View of Radiation,” but their relations to the topics discussed earlier in this Chapter are such that, pedagogically as well as historically, they belong here. ‘We have already considered two ways in which electrons can be released from a metal. Ta Chapter 2 we discussed the discharge of electricity through gases. ‘The electrons that participate in cold-cathode discharges are obtained from the cathode while itis bombarded by the positive ions produced in the residual gas in the tube, In this chapter we have considered another emission process, called photoelectric emission. There is still another kind of electron ‘emission, which we shall now discuss briefly. If a metal is heated, the thermal agitation of the matter may give electrons enough energy to exceed the work function of the material, Thus the space around a heated metal is found to contain many electrons. A study of this effect shows that the thermionic work function is very nearly the same as the Photoelectric work function—a most satisfying result Since a fine wire can be heated easily by passing an electric current through it, thermionic emission is one of the most convenient electron sources. Most “radio-type vacuum tubes use a heated cathode as ther electron source, ‘Thermal © “emission of negative charges from a hot wire in a vacuum was first observed 'by Edison in 1883 when he was making incandescent lamps, and such thermal ‘mission is called the Edison effect. In 1899, J. J. Thomson, showed that the thermions in this effect are electrons. 248 SPECIFIC HEAT OF SOLIDS In Section 1-4, we mentioned an experimental law discovered by Dulong and Petit in 1819 stating that the product of the atomic mass and specific heat is| | | 2 sromic view of radiation | 3-45 the same for all elementary (solid) substances. It can be restated to read: “The specific heat per mole is the same for all solid substances.” This law was based on early measurements of the specific heat for a series of substances. All these measurements were taken at room temperature. The value found for the specific heat per mole was approximately 6 cal-molet °C. ‘As discussed in Section 1-7, the temperature of gas is a direct measure of its translation kinetic energy. Similarly, the temperature of a solid is a measure of the kinetic energy of the atoms as they perform vibrations about the equilibrium positions in the lattice structure. At the interphase between the solid and the gas there will exist an equilibrium of no net energy flow if the temperature Of the solid is the same as the temperature of the gas. Since the gas molecules and the atoms in the solid are constantly colliding with each other, they have, in equilibrium, the same average kinetic energy. This means that each atom in the solid has an average kinetic energy of 3KT: Since these atoms are vibrating as harmonic oscillators, they will also on the average have potential energy. It is well known from mechanics that, in the harmonic oscillator, the average potential energy equals the average kinetic energy. Therefore the total energy per atom is 3 AT. The total energy content per mole is therefore E = 3kN,T = 3.7, 6-48) with R= KN, being the gas constant, The specific heat per mole is therefore ¢, = aBJaT = 3R = 5.94 cal-mole“* °K~, 49) in agreement with Dulong and Petit’s aw. Tn this simple derivation we have assumed that the substances undergo no cchange in phase, that is, change of the crystalline structure of the solid, during heating. We have also disregarded any effect resulting from changes of energy Of the so-called “free” electrons in metals. As further discussed in Chapter 11, ‘only a very small fraction of these free electrons increase their energy as the solid is heated. Therefore the effect on the specific heat of this energy increase is negligibly small. It was very soon shown that Dulong and Petit’s law for some substances failed quite badly at room temperature and that for all substances the specific | heat decreases with the absolute temperature. Figure 3-15 shows as examples the specific heats of diamond, silicon, aluminum, and lead as functions of the absolute temperature. As the temperature increases, the specific heat for all these substances approaches approximately the value 3R predicted above. However, our simple derivation fails to account for the decrease towards absolute zero. Einstein in 1907* was able to explain the basic features of the ‘curves shown in Fig. 3-15. He assumed that the energy of oscillations of the vibrating atoms is quantized in exactly the same way as the modes of electro- A. Binstein, Ann, Physik 22 (1907), p. 180. ca mole “IK 0 00 3 i200 T%« Fig. 3-15. Vorltion of specific heat at constant volume with temperature, Reproduced by permission from Richtmyer, Kennard, and Lauttsen, Vntroduction to Modern Physics, New York: McGraw-Hill 1956, p. 410. magnetic oscillation in an isothermal enclosure (Section 3-10). Accordingly, the average energy per oscillator is not 3 kT but (3-50) (compare Eq. 3-38). The frequency vis now the vibrational frequency ofthe stom in the lattice of the erstal, We find the total energy per mole by ‘uliplying Ey by Avogadro's constant N,. The spesfic heat per mole is then ae on (im)? Seo 8 ary (i) - ‘The frequency of vibration of an atom ina solid can be estimated in vatious ways, For iastance, by measuring the compressibility of the slid, one can tstimate the “spring constant” forthe restoring forces; and, ofcourse, the mass Of the atom is known. By using such techniques, one was able to show that Einstein's theory was in qualitative agreement with experimentally observed specifi heats The aseurnption of one single vibrational frequensy is mn oversimplifction. Debye® improved Einstein's formula by assuming that the vibrational fre- «quencies hada broad spectrum up to a maximum value and that there sre to 6-51) + P. Debye, Ann. Physik 39 (1912), p. 789,types of vibrations, one transverse and one longitudinal, However, of central importance also in Debye’s derivation is the average energy per degree of freedom as given by Planck’s formula. Thus Einstein and Debye showed that the concept of quantization of energy is not limited to electromagnetic radiation in an isothermal cavity, but also to mechanical harmonic oscillators. In the next chapter, we shall see how Niels Bohr found that the energies of electrons in atoms are quantized and also that their angular momenta are quantized. It fell on Enwin Schrodinger (Chapter 6) with some help of others to tie all these loose ends together in a neat theory that we call wave mechanics. PROBLEMS 341 The esplacement » at tne # of mass m which is moving with simple harmonic rotion of amplitude A and frequency i given by x = A sin 2nvz a) Show that its Kinetic energy Tis 2e?7842m sin? 21. 6) Show that th time average of T over a whole ejele and the time average of F over a whole cycle are each ‘equal to oneal the total eneay ofthe oscillator, 2277242, 3-2 Show that Planck’ radiation law, Eq. 41), reduces to Wien’ law for short ‘wavelengths and to the Rayloih-Jeans' law for long ones. (int: Express the exponential term asa series to obtain the second of these laws.) 343 The cavity ofa black-body radiator is inthe shape of a cube measuring 2 em on a side and the black body is at a temperature of 1500°K. a) Find the number of ‘modes of vibration per nit volume inthe cavity i the wavelength band between 14995 A and 5005 A. ¥) Calulate the radiant energy in the whole cavity in tis ‘wavelength band of 10 A (2) by assigning an energy of KT to each mode of Vibration and @) from Planck's radiation law. 3-4 Show that the energy density of black-body radiation expressed in terms of freaueacy is Beh ve" Sear) 3-5 Show thatthe total energy density in black-body radiation over the whole range of frequencies from 0 to co is identical in form to the Stefan-Boltzmann law for total radiation, A useful definite integral is = Side _ xt lo e115 3-6 a) Show that the maximum energy density in the spectrum of black-body radiation occurs at a wavelength 4, which can be found from the equation (— siS)e* = 1, where x = ch/AgkT: b) Find the roots of this equation, and then show that 4,7 = 2.90 x 1073 m °K. 3-1 At the surface of the earth a 1-cm? area oriented at right angles to the sun's rays receives about 0.13 J of radiant energy each second, Assume that the sun is a ‘lack-body radiator. What isthe surface temperature of the sun? (The radius ‘of the sun is about 7 x 10* m and the earth is about 1.49 x 10° kin from the sun.) Problems 85 3-8 A tungsten sphere 0.5 em in radius is suspended within a large evacuated enclosure whose walls are at 300°K. Tungsten is not a black body but has an average emissive power that is 0.35 that of a black body. What power input is required to maintain the sphere at a temperature of 3000°K if heat conduction along the supports is neglected? 3-9 Find the peroent change inthe total energy radiated per unit time by a black body if the absolute temperature ofthe black body is increased by a) 100 percent, b) 10, percent, ¢) 1 percent, and d) 0.1 percent. 3-10 At what wavelength does the maximum intensity of the radiation from a black ‘body occur at a temperature of 300°K? 1000°K? 6000°K? 3-11 If5 percent of the energy supplied to an incandescent light bulb is radiated as visibie light, how many visible quanta are emitted per second by a 100-watt ‘bulb? Assume the wavelength ofall the visible light to be 5600 A. 3-12 In order that an object be visible to the naked eye, the intensity of light entering the eye from the object must be atleast 1.5 x 10-1 J+m7?+s"!, What is the ‘minimum rate at which photons must enter the eye s0 that an object is visible, siven that the diameter of the pupil is 0.7 cm? Assume a wavelength of $600 A. 3-13 What is the energy in eV of a photon having a wavelength of 912 A? 3-14 Show that Planck's constant has the same physical units as angular momentum, 3.15 A mass of 10 g hangs from a spring witha force constant of 25 N m=". Assume that this oscillator is quantized just asthe radiation oscillators are. a) What is the minimum energy that can be supplied to the mass? b) If the mass at rest absorbs the minimum energy of part a), what i the resulting amplitude? ©) How ‘many quanta must the mass absorb in order to have an amplitude of 10 em? 3-16 Particles of a certain system can have energies of E,2E, or 3E, where E = 0,025 €. a) What are the ratios of the number of particles in each of the upper states to the number of particle inthe lowest state when the system is in equilibrium at 290°K? b) Whats the average energy ofa particle in the equilibrium distribution of these states? 3:17 Ina problem in Chapter 1, it vas found that a thermal neutron has a velocity of 2200 ms“. a) What is @ thermal neutron’s kinetic energy in eV? b) What is the increase in kinetic energy in eV of each water molecule in a stream that goes down a 450-fe (137.1 m). waterfall? 3-18 The fissioning of a 35U atom yields 200 MeV of energy. How many such atoms must fission to provide an amount of energy equal (0 that required to lift a mosquito one inch? The mass of an average mosquito is 0.90 me. 3-19 The visible light from a 40-watt incandescent bulb is incident normally on a potassium surface $0 cm from the bulb. a) How long will t take a potassium atom to absorb 2.0 eV of energy which is its photoelectric work function? Consider the bulb a point source that radiates 7.5 percent ofthe input power as visible light. Assume that the absorbing area of the potassium atom is equivalent to a circular disk having a diameter of 50. b) How long would it take the potassium atom to absorb 2.0 eV of energy itis illuminated by full moonlight? ‘The illumination at 50 em from a 40-watt bulb is about 740 times greater than fall moonlight. Te 2OS23 ~ 23-20 3.21 3.22 323 324 3.25 3.26 ‘The light-sensitive compound on most photographic films is silver bromide, ‘AgBr. We shall assume that a film is exposed when the light energy absorbed dissociates this molecule into its atoms. (The actual process is more complex, but the quantitative result does not differ greatly.) The energy or heat of dis- sociation of AgBr is 23.9 kcal - mole“. Find a) the energy in eV, b) the wave- Tength, and c) the frequency of the photon that is just able to dissociate a mole- ‘cle of silver bromide. d) What is the energy in eV of a quantum of radiation having a frequency of 100 MHz? e) Explain the fact that light from a firefly can expose a photographic film, whereas the radiation from a television station transmitting 50,000 watts at 100 MHz cannot. f) Will photographic films stored jn a light-tight container be ruined (exposed) by radio waves constantly passing through them? Explain, ‘When a certain photoelectric surface is illuminated with light of different wavelengths, the following stopping potentials are observed: 3, A 3660 405043604920 $460 5790 VaV 148 LIS 093 0620.36 0.28 Plot the stopping potential as ordinate against the frequency of the light as abscissa. Determine a the threshold frequency, b) the threshold wavelength, ©) the Photoelectric work function of the material, and d) the value of the Planck constant h (the value of e being known) “The photoelectric work function of potassium is 2.0 eV. If light having a wave- Tength of 3600 A falls on potassium, find a) the stopping potential, b) the kinetic energy in eV of the most energetic electrons ejected, and ¢) the velocities of these electrons. What will be the change in the stopping potential for photoelectrons emitted from a surface if the wavelength of the incident light is reduced from 4000 A to 3980 A? (Assume that the decrease in wavelength is so small that it may be ‘considered a differential.) ‘The threshold wavelength for photoelectric emission from a certain material is (6525 A. Find the stopping potential when the material is irradiated with a) Tight having a wavelength of 4000 A, and b) with ight having twice the frequency and three times the intensity of that in the previous part. c) If a material having double the work function were used, what would be the answers to parts a)and b)? Light of wavelength 4000 A liberates photoelectrons from a certain metal, The photoelectrons now enter a uniform magnetic field having an induction of 10°* 7. ‘The electrons move normal to the field fines so that they travel circular paths. ‘The largest circular path has a radius of 5.14 cm. Find the work function for the metal 'A surface is irradiated with monochromatic light of variable wavelength, Above 4a wavelength of 5000 A, no photoelectrons are emitted from the surface. With fan unknown wavelength a stopping potential of 3V is necessary to eliminate the photoelectric current. What is the unknown wavelength? SEZIONE STATI AGGREGATI E BIOMOLECOLE CHAPTER 4 The Atomic Models of Rutherford and Bohr 41 INTRODUCTION We have traced how matter, electricity, and radiation came to be regarded as atomic in character, We have established the existence of some elementary particles that are more fundamental than the chemical elements. Electrons, for example, are common to all elements and are a common building block of all matter. Our discussion of positive rays and mass spectroscopy showed that matter also has positive constituents which are much more massive than electrons. ‘Thomson, who made the first quantitative measurements on electrons and positive rays, assumed that a normal chemical atom consists of a ‘mixture of constituents. This mixture came to be called the “plum-pudding” atomic model: The atom was regarded as a heavy positive sphere of charge seasoned with enough electron plums to make it electrically neutral. 42 PROBING THE ATOM WITH ALPHA PARTICLES ‘A very different atomic model was indiéated by experiments performed by ‘Rutherford and his associates in 1911. ‘We shall discuss radioactivity at some length in Chapter 12, but in order to be able to discuss the Rutherford experiments a few observations ‘need to tbe made now. Certain atoms are unstable and fly apart of their own accord. |The nature of these disintegrations depends on the element that is disintegrating, ‘but in every case the fragments ejected consist cither of electrons, here called eta rays, or of doubly ionized helium atoms, called alpha particles. These " ‘disintegration fragments usually are ejected with high energies from a radio- ‘active substance and are often accompanied by very short-wavelength photon tadiation called gamma rays. Radium, for example, is an excellent source of high-energy alpha particles. These alpha particles can travel through a few centimeters of air before they are stopped, and in a vacuum they travel long 7distances without losing energy. When they strike certain materials, they cause visible fluorescent light flashes.* Rutherford studied how these alpha particles from radium were absorbed ‘by matter. He found they were absorbed by sheets of metal a few hundredths of a millimeter thick, but that they could readily pass through gold foil several . ten-thousandths of a millimeter thick. Rutherford’s apparatus is shown sche- ‘matically in Fig. 4-1. Radium was placed in a cavity at the end of a narrow ‘tunnel in a lead block. Alpha particles were emitted in random directions by this source and the lead absorbed all except those emitted along the axis of the tunnel. In this way Rutherford obtained a collimated beam of alpha particles which streamed toward the gold foil. Particles deflected by an angle 0 by passage through the foil produced flashes on a ZnS fluorescent screen, and these flashes were observed in a low-power microscope. Fluorescent Radiata e-particles “4 298 Gala teil Fig. 41 Schematic disgrem of Rutherford’s alpha-panticie scattering app ‘Lead block tus. Many of the alpha particles did go straight through the foil or were deflected only by very small amounts. Amazingly, however, some alpha particles were deflected through very large angles. A few even retumed to the side ofthe gold fol from which they came. Rutherford’s astonishment at this is evident in his comment, “It was quite the most incredible event that has ever happened to me in my life. It was almost as incredible as if you fired a 1Sinch shell at a piece of tissue paper and it came back and hit you.” This observed scattering through large angles was contrary to predictions based on + Radiolight watch dials are painted with a paste containing fluorescent material ‘and a trace of radium. Under a microscope the glow of the dial can be seen to be a ‘multitude of flashes, called scintillations, which remind one of the twinkling of the stars on a summer night. The effect may be seen through a 4-power magnifier, but itis better to use two or three times this magnification. The light is less likely to appear continuous if there is very little radioactive material on the dial. Observations must be made in a completely darkened room and it may be necessary to wait about five ‘minutes for the eyes to become dark-adapted. This time delay will also permit any phosphorescence to fade and eventually die out. the Thomson model of the atom. Let us examine these predictions in some detail. “Assume that an alpha particle of charge ze (with 2 = 2) collides with a ‘Thomeon atom of radius R. Can it be turned around in its path if the forces acting are only electrostatic forces? Before it starts on its way back, it must te momentarily at rest, and this means that the kinetic energy has been con verted into potential energy. (Since the alpha particle is much lighter than the ‘tom itis colliding with, we neglect the kinetic energy resulting from center-of- thas motion.) ‘The potential energy for a particle of charge ze at the surface ‘of a uniform-charge cloud of radius R and charge Ze is V. = 2Ze"/4nt0R. ‘At the center of the charge cloud the potential energy is 7 1) Inserting z = 2, Z = 80 (for example), R © 1A = 107° m, and the known ‘values for e and >, we get after converting to electron volts V_ = 3456 eV. ‘We have completely neglected the effect of the electrons, which, of course, ‘neutralize the positive charge and therefore substantially reduce its effect. Jadeed, only local fluctuations of charge within the atom can produce the forees we are looking for. Since the kinetic energy of the alpha particles used jn the experiments by Rutherford and his associates was about 5 MeV, it is ‘lear that a single collision with a Thomson atom could not result in anything but a very small angle of deflection. A collision with an electron cannot produce much deflection either, because of the fact that the electron is 7000 times lighter than the alpha particle. What about multiple collisions? "A simple but adequate theory of multiple collisions predicts that the number of particles deflected into a small solid angle centered about an angle of deflection @ should be N@ = NO". a2) ‘This is a Gaussian distribution, or standard distribution with & being the Mandard deviation determining the width of the curve. Two of Rutherford'’s fsvociates, Geiger and Marsden, measured the angular distribution of alpha particles scattered from a 0.4 thick gold foil. They found that about I in 50,000 was scattered through an angle of more than 90 degrees. Geiger late delermined the probable angle of deflection as 0, = 0.87 degree. The probable Single 0, defined as the angle which has a $0 percent likelihood of being receded, relates to the standard deviation as 0, = 0.67450. This gives ¢ = 129 for the experiment described above.* By inserting @ = 90° and = Tris amusing to note that Rutherford in his very famous paper on scattering 0 alpha particles by matter refers to the probable angle as “the most probable ansie o | WUeuibne* The most probable angle, thats, the maximum in the standard distribution seeany @ = 0. ‘Thus even great scientists err, and sometimes worse than this! Th reference is E, Rutherford, Phil. Mag. 21 (1911), p. 669.29°, we find from Eq. (4-2), N(90) = N(@) x 10-19°, This is such an incredibly small mimber that it clearly is not worth the effort to work out the integrals to determine the fraction of the beam of particles deflected by more than 90 degrees. 4-3 RUTHERFORD’S NUCLEAR ATOM Rutherford concluded that the large deflections must come about by single encounters between alpha particles and atoms and that therefore the atoms ‘must be seats of much more intense electric fields than those that would seem to be consistent with the Thomson model. He assumed that the positive massive part of the atom was concentrated in a very small volume at the center of the atom. This core, now called the nucleus, is surrounded by a cloud of electrons, which makes the entire atom electrically neutral. Because the atom is mostly empty space, many of the alpha particles go through the foil with practically no deviation. But an alpha particle that passes close to a nucleus experiences a very large force exerted by the massive positive core and is deflected through a large angle in a single encounter. From Eq. (4-1) and the example following it, it is clear that the radius of the nucleus must be more than 1000 times smaller than the radius of the atom if this is to stop an alpha particle and turn it around in its path. Collisions with electrons result in but very small deflections and can therefore be neglected. We shall study the mechanics of a single encounter between an alpha particle and a nucleus, with the objective of finding a formula to replace Eq. (4-2) for the angular distribution of alpha particles scattered by a foil SL, Asymptotes of trajectory Trajectory of ‘particle Fig. 4-2. Scattering of an alpha particle by the Rutherford nuclear atom. ‘ igure 4-2 shows the trajectory of an alpha particle approaching the nucleus ‘with an impact parameter b. This is the distance between the nucleus and the straight-line path the alpha particle would have taken in the absence of an jnteraction, We use a polar coordinate system (r,9), as indicated in Fig. 4-2, ‘and write the law of conservation of energy, im () + mae? (2) 4 ey 2 Nad “2 Natl” anegr ~ 2 ‘here vp is the initial asymptotic velocity of the alpha particle, We have assumed that the nucleus is so much heavier than the alpha particle that we can neglect ‘recoil. ‘The law of conservation of angular momentum. 3s 3) lds mr? 4B mogb ae a dd _ tb ab _ vob a4 a a We now introduce for convenience a parameter 2g which is equal to the distance of closest approach in a head-on collision: 2Ze = 22e “5 4 = Begmod oe © Combining Eqs. (4-3), (4-4), and (4-5), we get - “gen 1 go ee (t- 32-2) Gd ‘The double sign arises from the square!root. The radial velocity dr/dt is clearly fist negative, then changes sign to positive atthe position of minimum distance to the nucleus (Fig. 4-2). By again combining Eq. (4-6) with Eq. (4-4), we ~ #4 0-4-3) (7) + ‘This is a differential equation describing the orbit in polar coordinates (4). ‘Using a table of integrals, we find for the first part of the orbit from r = co ‘to point 4 in Fig. 42 t= sometime (-2-9)" — Arc cos ~ e ; +8) Gx ay?We have integrated from oo to r and used the minus sign in Eq. (4-7). At the point of minimum r, we have drjdr = 0, which inserted into Eq. (4-6) gives 14H ao ror at point A. The anle 6 corespending t this points, by Ea, 48), 5 ya B= Aveo ce as. a O44 sees Gs aa as) ‘The orbit is symmetric about point A, since ddr numerically depends upon r only and not upon ¢ as well (Eq. 47). The radius vector r therefore sweeps through a total angle 26,. From Fig. 4-2, we see that the total angle of deflection is Om 2by = 2AWEOS Ge; (410) From this we find cos up and easily sin? aa a1) ere Combining Eqs. (4-11) and (4-12), we get 6 b= qeot!. geots 13) This, then, is a simple relationship between the impact parameter b and the total angle of deflection 6. ‘Our objective was to find a formula for the angular distribution of particles scattered by a foil representing a large number of scattering centers (nuclei). We can best do that by calculating the collision cross section (Section 1-10) for the process. First we shall modify this concept slightly. In Section 1-10, ‘the eross section for a collision process represented a target arca. A hit in this area resulted in deflection of the “bullet” particle, Nothing was said about the direction of the scattered particle after the collision. Figure 4-2 indicates a ring, displaced to the right for clarity, representing a target area for all alpha particles with impact parameters between b and b+ db. This area is da = 2nb db. aH All particles aimed at that area will have a scattering angle between 9 and @ + dd, where 6 is given by Eq. (4-13), and d8 by the differential of same: dba 449 2 sin? 6/2 ‘The angles @ and @ + d9 define two cones with the horizontal line through the nucleus as their axis (Fig, 4-2). The solid angle between the two cones is dQ = 2n-sin 0 dO, and this is therefore the solid angle into which particles aimed St the target area do will be scattered. By substitution we now get from © Bg. (4-14), neglecting the minus sign, (15) go = Toes $i2dd _ gP2asinodd) Fd a 4p sin? 0/2 Asin 72 A sin* 62 ae Finally, we insert q from Eq, (4-5) and get the Rutherford cross-section formula da PDE aa dQ. Garegm*c sin® 0/2” ‘The interpretation of this equation is that each nucleus represents a target fatea do for deflecting a particle by an angle 0 into a solid angle d®. Let the ‘target foil have n atoms per unit volume and a thickness Ax, and let the detector tbe at an angle 6 subtending a solid angle AQ with the target. By using Eq (1-35), we find the fraction of alpha particles from the beam scattered to the detector to be AN ‘do 8 (a) Ao Ax, 418) vith the diflerential cross section de/d0 given by Ea. (4-17. ‘Rutherford’s formbla gives infinile cross section for 9 = 0. This corre sponds to b> 0 and in reality, of eourse, we do not have infinite impact Parameters. Indeed, we could have set an upper limit for as approximate The radius of the atom, because outside the atom the electric field from the ‘nucleus is shielded by the electrons. ‘Neglecting the region close to @ = 0, we see that the angular distributior ‘of alpha particles is described by the factor (sin 0/2)~*. This term does no rary neatly 30 fast with 8 as does Eq. (4-2). Rutherford found that all availabl data agreed with his predictions and thus was. ‘the nuclear model of the aton torn, The year: 1911. ‘In 1913, Geiger and Marsden* published the results of a very exhaustiv test of Rutherford’s theory. They used silver and gold foils as targets for th alpha particles and recorded the rate of scintillations as a function of angh N #0 H. Geiger and E. Marsden, Phil. Mag. 25 (1913), p. 605,over a very wide angular range.’ Fig. 4-3 shows their result. In this graph, according to Rutherford’s formula, all points for a given experiment should fall on a straight line with a slope of 45°. The agreement is very good, considering in particular the fact that the counting rate, measured by eye, varied over almost four decades. " Silver 4-3 Number AW of scattered slphapariles vous seaterng angle & (logarithm Fe ae cathe’ Rehanmes Konto, sn Caution, Invodoeton to ‘Modern Physics, New York: McGraw-Hill © 1955, p. 146. 7 Geiger and Marsden, and later Chadwick, also attempted to measure absolute values for the scattering rate (ie., the cross section) such that they could thereby determine the nuclear charge Z. They concluded that Z was approximately half the atomic mass number. Much more accurate measurements of Z for a large number of elements were made about this, time by ‘Moseley. His experiments with x-rays emitted by the various elements showed az eel with the atomic number; that is, the number giving the place of the element in the Periodic Table. iscussed i detail in Chapter 10. al 4-4 SIZE OF THE NUCLEUS ‘The detailed studies of alpha scattering carried out in Rutherford’s laboratory on many elements soon revealed that the simple (sin 8/2)" law does not always hold. The discrepancies occur for light elements, particularly for large angles of deflection. This was immediately interpreted as a breakdown of the r~? force law at small distances. Put differently, it revealed that the nucleus is not a point charge and that at small distances forces other than the Coulomb force are acting. It was possible to estimate that the radius at which these Forces affected the motion ofthe alpha particles was of the order of § x 10"!¢m for medium-mass atoms. We now know that the mucleus consists of protons and neutrons held together by very short-range but very strong foress. An alpha particle approaching the nuclear surface experiences these forees and may be scattered or broken up into iis components, two protons and two neutrons. The study of these processes is a part of nuclear physics, discussed briefly in Chapter 12 4-5 SPECTRA Most readers of this book have studied light and know that spectrographs are instruments which analyze light according to its distribution of frequency or color, ‘These instruments always have an entrance slit, a dispersive component that may be a prism or grating, and a detector that is usually a photographic plate, The instrument forms an image of the slit on the detector, the focusing being accomplished with a system of lenses, or in the case ofa reflective-grating spectrometer by the curvature of the grating. Separated images are formed for each frequency of light present, so that light which is continuous in its frequency distribution forms a wide image that is a continuous suocession of Sit images. Light which is discontinuous in frequency distribution forms discrete set of slit images that are called spectral lines. igure 4-A(a) depicts a reflective grating made by seribing fine lines in & highly reflective metallic surface. A parallel beam of light falling in on the vuirar at an angle 6, with the normal will be reflected—to zeroth order—such That it leaves at an angle 8, as shown. A first-order maximum in the diffraction coming light reflection © © ‘Fig. 4-4 (a) Reflection of paral light froma grating. (b) Detall ofthe tst-order reflection Path 2 one wavelength longer than 1.pattern will occur at some angle 0, and if the grating has a large number of lines, this maximum will be sharp. igure 4-4(b) shows in more detail two rays falling in on the grating. The points where these rays hit the grating are sources of light radiating in all directions. To get constructive interference when the two rays later are focused to one point on the detector, the wave fronts must be perpendicular to the rection of motion. This will be the case if the difference in path length from wave front A to wave front B is a whole number of wavelengths; that is, (sin 8, — sin 0) = ni, (9) where 1 is an integer. For n = 0, we get 0 = 0, independent of 2, Form = 1 or higher, we see that sin @ depends on 2, The grating therefore disperses light according to the wavelengths of its components. Figure 4-5 shows the layout of a concave grating spectrometer. The reflective grating is bent with a radius of curvature 2R. The entrance slit for the light and detector plates are placed on a circle with radius R. It can then ‘be shown that sharp images of the slit will be produced, by the focusing action of the mirror grating, on the detector. Spectroscopy with prism spectrometers, as well as with grating spectrom- ters, was a well-developed science in the latter part ofthe last century. It was found that the light from any element in gaseous form produces a discontinuous line spectrum. Each element has its own characteristic frequency distribution Grating Central image Fig. 4-5 Concave grating spectrometer. Images of te st are formed on the curve photogre in = 3 Je puts by the curvature ofthe grating (eden 28) Repredssod by pension fiom Jenkins and White. Fundamentts 9” Onc Now Yo McGraw- Hill © 1957, p. 348. oreo eae " ‘or spectrum, so that each element can be identified by the light that it emits, ‘The most dramatic instance of such identification occurred when the element thelium was “discovered” in the spectrum of the sun before it was chemically isolated here on earth e ‘The emission of a spectral line by gas atoms could presumably be explained somewhat as follows on the basis of the Thomson model of the atom: An “Glectron is bound to the atom by electric and possibly other forces and is Sormally at rest in an equilibrium position. Displaced from that position by a tallision between two atoms it will execute oscillations. Since itis charged, it vill radiate energy of frequency equal to the frequency of oscillations. If the ~ Sacillations are not strictly harmonic (ie. proportional to sin a), it will also © fadiate higher harmonics. The puzzle is that no atom radiates spectral lines | ith frequencies being simple harmonics of a fundamental. All elements display a great multitude of lines spread over a very large frequency spectrum. ‘Certain regularities are found (next section) but nothing that supports the above picture. _ Ri"after Rutherford’s discovery of the nucleus, it was immediately suggested that the electrons move in orbits about the nucleus and that the electrostatic attraction provided the centripetal force. The difficulty with this model is that the electron is continuously accelerating (centripetal acceleration), and accord | ing to classical electromagnetic theory an accelerating charge e radiates power Gta rate P = €a°/6xcgc? watts, where ais the acceleration. A simple calcula~ © tion shows that the electrons in orbits about the nucleus would lose energy and = spiral in towards the nucleus so fast that in about 10~®s all matter would have collapsed This set the stage for one of the most brilliant natural philosophers of "modem times, Niels Bohr (Section 47). Before we present his simple yet © evolutionary, theory, we shall discuss in some detail the experimental data on ich it was based. y 46 THE HYDROGEN SPECTRUM “The obvious place to start the study of spectra is with the spectrum of hydrogen. “ft is fot surprising that this lightest element has the simplest spectrum and ‘probably the simplest structure. Part of the hydrogen spectrum is shown. in “Fig. 4-6. The regularity of these spectral lines is immediately evident, and it “appears obvious that there is some interrelationship among them. Tn 1884, a = wise high-school mathematics teacher by the name of Balmer took the wave- © Jenigths of these lines as a problem in numbers. He set out to find a formula 7 ‘which would show their interrelation. He hit upon & formula which could be ‘iad to give these wavelengths very precisely. The Balmer formula is 3645.6n? Atangstroms) = SES.45 678 i ml om ma Continous ‘pectrum oe n=08628 (in A) 48618 48405 4101.7 ‘3645.6 (convergence ed ‘ie’ Vino Uitrevilet, a Fig. 4.8 Daum of the Balmer cris of atomic hysrogen. (The wavolngts sr the Each different wavelength is obtained by putting into the formula different values of the running integers n, which are n = 3, n= 4, n = 5, ete _The success of the Balmer formula led Rydberg to attempt a formulation which would apply to heavier elements. He proposed an equation of the form taa-—%_, are where ¥ is the wave mumber,* R is the Rydberg constant which is equal to 1.09737 x 10’ m~', and n is a running integer. A and « are adjustment constants which depend on the element and the part of the spectrum or spectral seties to which the formula is applied. Rydberg found that this formula, which. can be regarded as a generalization of the Balmer formula, could be fitted to ‘many spectral series, and further that the value of R was nearly the same when the formula was applied to different elements. In 1908, Ritz noted that the wave numbers of many spectral lines are the differences between the wave numbers of other spectral lines, and that the 4 term of the Rydberg formula was really a particular value of a term, like the second term of the Rydberg formula. Using this “combination principle,” Ritz rewrote the Rydberg formula as oe (n+ BP ea? (421) (4-22) “The wave number J, a8 used in spectroscopy, is the number of waves per unit length and die from the more conventional wave umber k used in Chaper 3 by factor 2x. One might suppose tha the logical quantity to use fr the reciprocal form ‘of wavelength would be the frequency, c/3. In spectroscopy, the wave number is used because in order to compute the frequency without losing the remarkable precision of wavelength measurements, it would be necessary to know the velocity of light to an ‘equal precision. Wave numbers can be computed without knowing the velocity of light land so they retain all the accuracy of spectroscopic wavelength measurements. How- ver, the wave number isnot an absolute constant fora given spectral line because its wavelength depends upon the index of refraction of the medium in which the measurements are made. ‘The wavelength in air is corrected to vacaurn by means ofthe relation Jeae = Hin Where iis the index of refraction of air forthe particular wavelength. Tn {He visible region, vc is approximately 2.5 * 10 pereent greater than Aa 6 where « and f are adjustment constants which depend on the element, For different spectral series of a given clement, m takes on different integral values. The different lines within a series are computed by changing the running integer n. It is easily shown that when «= B = 0 and m = 2, Eq. (4-22) reduces to the Balmer formula for bydrogen. Tn the same year, 1908, Paschen found another hydrogen series of lines in the infrared region to which Eq. (4-22) could be fitted by making « = i = 0, m= 3, and n= 4, 5, 6, etc. Thus, both the then-known hydrogen series ‘could be represented by 23 ‘This gives the Balmer series when m = 2 and n = 3, 4, 5, etc., and correct predicts the Paschen series for m = 3 and n = 4, 5, 6, ete 4-7 THE BOHR MODEL AND THEORY OF THE ATOM Equation (4-23) represented the entire known hydrogen spectrum with grea precision, but it was an empirical formula. In spite of many attempts it defi Serivation on the basis of known physical laws. In 1913, Niels Bohr succeed jn deriving this important relation, but only by breaking with concepts « ‘classical physics believed to be well established. ‘Bohr extended Rutherford’s model of the atom. He retained the sma core or nucleus of the atom and proposed that the electrons, known to be part of the atom, move in orbits around the nucleus. In the case of hydroge Bohr proposed that the nucleus consisted of one proton with one electrc evolving about it. This is a planctary model of the atom where the hea positive nucleus is like the sun and the light, negative electron is like the plan earth. In this model, hydrogen is'a tiny, one-planet solar system with #l ‘gravitational force of the solar systém replaced by the electrostatic force | attraction between the oppositely charged particles. The general equations f the gravitational force and the electrostatic force are, respectively, MM ang p=. 2 ane 7? F co ‘Both forces are inversely proportional to the square of the distance betwe the particles. The planets of the solar system have elliptical orbits which « ‘nearly circular, Bohr assumed that the planetary electron of hydrogen mo “Gn a circular orbit, which makes the analysis of the classical aspects of | problem straightforward, Let » be the tangential speed of a mass Mf" tha fevolving around a very large mass M in a circular orbit or radius r. Revoluti ‘occurs around the center of mass of the system which, in effect, is at the cenof the large, massive body. The centripetal force acting on M’ is the gravitational force of attraction due to M. Thus we have MM’ F Ma (425) from which we obtain (4-26) In Bohr’s model of the atom, an electron of charge , mass m,, and tangential speed » revolves in a circular orbit of radius r around a massive nucleus having a positive charge Ze. In this case, too, the center of the orbit is essentially at the center of the heavy nucleus. The centripetal force acting on the orbiting electron is the electrostatic force of attraction of the nuclear charge, and therefore the force equation is Fad tL ng = an fey r From this equation we find that ’ (428) Freomer (For hydrogen, the atomic number Z equals one. We include Z for generality.) Each of the equations (4-26) and (4-28) provides a relationship between the variables » and r. If one is known, the other can be found. ‘In the gravitational case, any pair of values of v and of r which satisfy Eq. (4-26) may actually ‘occur. In the electrical case, classical physics imposes no limitation on the number of solutions there can be for Eq. (4-28). For the case of the hydrogen atom, Bohr introduced a restrictive condition which is known as the first Bohr postulate, He assumed that not all the possible orbits that can be computed from Eq. (4-28) are found in hydrogen. Bohr’s first postulate is that only those “orbits occur for which the angular momenta of the planetary electron are integral ‘multiples of h/2n, that is, nh/2n. Here n is any integer and h is Planck’s constant. Bobr’s first postulate introduces the integer idea that appears in the Ritz formula and also introduces Planck’s constant, which we have seen plays an ‘important role in the atomic view of radiation. Stated mathematically, this first postulate is nh mor = ™ = nh, (4-29) a (4-29) where n = 1, 2,3,..-and h, called ‘*h-bar” is Planck’s constant divided by 2x. ‘The orbiting electron in hydrogen must simultaneously satisfy the conditions expressed by Eqs. (4-28) and (4-29). After eliminating v between these two or are “permitted” in the equations, we find that the orbits which hydrogen atom are only those that have radii = folie -= (430) xmze With numerical values inserted, this gives 0.529 n? x 10m = 0.529 n?A. (4-308) Thus, the “first Bohr orbit” has a radius of 0.529 A; the second has a radius of 2.116 A, ete. Tn discussing the energy of a planetary electron, we shall use the usual convention that the electron has no potential energy when it is infinitely far from its nucleus. The potential energy of a negative electronic charge at a “© distance r from the nucleus of positive charge Ze is then Ze Aregr ‘Note that the potential energy of the electron in this case is zero at infinity ‘and negative elsewhere. We can use Eq. (4-28) to find its kinetic energy: Zee Breor ‘The total energy of the planetary ciectron is the sum of the potential and kinetic energies: (431) 4-32) Ze Ze* Ze* Bat B= scsop anegr ~~ Bntgr We have computed the total enery as function ofr. But we have seen that” can have only those values even by Eq, 30) Using this equation to eliminate r, we find that, E (433) (434) cating the energy ofthe een, we ae not proce. We hve ssumed bp salog E the corey ote en odin senane rot, Snes La en ene ass ma ofthe leon, eater has mos othe Yt he ot ES is cee youid rome es to cons oe came fi fo ee oscar ol Ther core ir han ic ates tay rob sh ws ean ofr dy may be neglected if the mass of the other body is taken to be the “reduced mass,” body may be neglected the mas te omames eMfime he all + db) Setpoint ec as may te ele BE i Ea 30 aot cr i ic oem ae woke (ra #7)to have, ‘The integer m is called the fora! or principal quantum number and it cean have any of the series of values, 1, 2,3,... the energies of the states. When 7 is large, the energy is large, that is, less negative than for small integers. ‘The energy required to remove an electron from a particular state to infinity is called the binding energy of that state. Tt is numerically equal to Ey Figure 4-7 shows the potential energy and the total energy as a function of (Eq. 4-33). The allowed orbit radii (Eq. 4-30) are indicated, and the corresponding values of the total energy are shown. The energies are given in clectron volts, found by inserting numerical values for the constants in Eq. (4-34) and converting from joules to eV. In general, we get 13.6 aoe gE, = —BSev. (4-34) ‘We now consider how Bohr used this set of energies to account for the hydrogen spectrum, In Chapter 3 we described how classical electrodynamics predicts that energy will be radiated whenever a charged particle is accelerated. ‘We were careful to point out that the acceleration could be due to a change of direction of motion as well as due to a change of speed. According to classical theory, an orbital electron should radiate energy because of its centripetal acceleration. In order to preserve his atomic model of planetary electron orbits, Bohr had to devise a theory which would violate this classical prediction since, according to it, any electron that separated from the nucleus would soon radiate away its energy and fall back into the nucleus. Bohr’s second break with classical physics is contained in his second postulate, which states that no electron radiates energy so long as it remains in one of the orbital energy states; Energy 520 2116 azo_A, T nl }rotat energy 1"? =136) Fig. 4-7 Energy relations in the hydrogen atom. ‘ The values of n determine and that radiation occurs only when an electron goes from a higher energy state fod lower one, the energy of the quantum of radiation, hv, being equal to the ergy difference of the states, Let the quantum number = a represent & her energy state and n= my represent a lower energy state (my <1) then ‘the second Bohr postulate can be written as hy = Ey, — Ene (4-35) Substituting for the energies from Eq. (4-34), we have for the frequency of the ‘emitted radiation me*Z? (1 ya mek (L - 4-36 Beh? (a an ‘of, in terms of the wave number, 1 metZ? (1 i se! mete z 4-37 ic” Begite (i a) a where is the speed of light ina vacuum, Comparing Eq. (4-37) with Eq. (4-23) shows that both have the same form Equally impressive is the fact that the constant factor of the Bohr formula | jsthe Rydberg constant, R. Again comparing Eqs. (4-37) and (4-23), we find ince Z = 1 for hydrogen, R= 0973731 x 107 m7" apie (4-38) ‘The R given here is Re, which would be correct if the mass of the nucleus = Were infinite compared with the mass of n electron. Ifthe motion ofthe nucleus Fe taken into account, m, must be repliced by the reduced mass. Therefore, general, R = R,/(1 + m,{M). This accounts for the slight variation of R “from element to element noted by Rydberg. Itis a triumph of the Bohr model “and theory that the slight differences between the spectra of ordinary hydrogen = and its isotope, heavy hydrogen (deuterium), can be attributed to the influence ‘ofthe nuclear mass. In fact, heavy hydrogen was discovered spectroscopically by Urey in 1932. © “The Bohr formula gives the Balmer series for m = 2 and the Paschen Series for m, = 3, as we have seen before. But the Bohr theory places no ‘estrictions on 7m, and his result suggested that there might be additional hhydrogen series not yet found experimentally. In 1916 Lyman found a series © in the far ultraviolet, in 1922 Brackett found a new series in the infrared, and Fin 1924 Pfand located another in the same region, Table 4-1 summarizes the “five hydrogen series.Table 4-1 The spectral saties of hydrogen a ipa quanto ial gus aera, et Values of m, | Name of ser PMN Continue ey ae sna ~oaquasaspens! 1 a=: + 0.54 5 ws ij {PEELE — Plan anes 985 2 t Bracket sees ie 4 wed [= 1 Paschen 5 LF] {1 “see a 1 ae tits se 4-8 HYDROGEN ENERGY LEVELS ee ‘The discovery of sharp energy states of the hydrogen atom was a finding that. hhad very far-reaching consequences in physics. All atomic, molecular, and nuclear systems have discrete energy levels; indeed, macroscopic systems can. exist only in states of discrete energies, but these energy levels are so closely spaced that they cannot be experimentally resolved. Because of the general interest of this subject, we shall here look at the energy levels of hydrogen some detail ‘These levels can be represented graphically as shown in Fig, 4-8. The quantum numbers are shown at the left and the corresponding energies of hydrogen in electron volts are given at the right. In this array of energies, the higher (less negative) energies are at the top, while the lower (more negative) are toward the bottom. In a normal unexcited hydrogen atom, the electron is in its lowest energy state at the bottom, with n = 1. An electron in this ground state is stable and remains in this state continuously without emitting or absorbing energy. The “excitement” begins when the electron absorbs energy. in some way. There are a variety of ways in which this may be brought about. If the hydrogen is in an electric discharge, a free electron which has been accelerated by the electric field may hit* it. If the hydrogen is heated, the clectron may be excited by a thermal-motion collision. If the hydrogen is ~ illuminated, ‘it may absorb energy from a photon. Suppose the electron in hydrogen absorbs about 20 eV of energy in one of these ways. This is enough ‘energy to lift the electron to n = 0 (13.6 eV) with 6.4 eV left over. In this ‘case, the electron is made entirely free of its home nucleus and is given 6.4 eV. of kinetic energy besides. If the electron absorbs just 13.6 eV, it is merely freed from its home nucleus and drifts about with only its thermal kinetic energy. In either of these cases the remaining nucleus is an ion. If the energy of a bombarding electron is less than that required for ionization but equal to or © 1 } oy ___ —____- 16 © Fig. 4-8. Energy-level diagram of the hydrogen atom. iqreater than that needed to raise an electron in an atom to one of its permitted ‘Energy levels, then the atomic electron will absorb just enough energy to put - into some higher energy state. After the bombarding electron has transferred ‘enough energy to the atom to excite it, the electron will leave the encounter, Canrying away any excess as kinetic energy. The excitation energy of any level fh electron volts is numerically equal to the excitation potential of that state ‘in volts. “After excitation, the atomic electron returns (0 its normal state, If it was “it may jump from 4 to 1 in one step. It may also go 4, 2, 1 Ee or 3, 1 or 43,21. in each step of the return trip, the electron must lose Fan amount of energy equal to the difference of the energy levels. The only + Weuse the word “hit” loosely. We saw in the disrussion of alpha-particle scattering that a collision between charged bodies does not involve physical contact in the ‘usual sense,mechanism available for this energy loss is through the emission of electromagnetic radiation. Thus, in Fig. 47 we have represented graphically the second Bohr postulate, given in Eq. (4-35). When we see the light from a hydrogen discharge, we are “seeing” the electrons go from exicted states to lower states. The electron transitions which end on m = 1 constitute the Lyman series, on n= 2, the Balmer series, on n = 3, the Paschen series, etc. From the cenergy-level diagram we can see that the Lyman transitions involve the largest changes of energy, produce the highest frequencies, and provide the “bluest” (ultraviolet) light. ‘The shaded region at the top of Fig. 4-7 represents the fact that electrons at infinity may have kinetic energy, so that their energy there is not zero but positive. If the electron of hydrogen is completely removed from its nucleus, then one of these electrons at infinity having any energy may fall into any one of the energy levels. Such an electron undergoes a change of energy equal to its energy at infinity minus the negative energy of the level to which it falls. ‘The “double negative” in the last sentence enables us to conclude that the energy radiated by such a transition is the sum of the electron’s kinetic energy at infinity and that involved in the transition from = co to the final level ‘The energy radiated will have a value greater than that involved in the transition from m = co to the final level. Since there is a wide distribution of the energies among the electrons at infinity, there is a continuous spectrum below the short-wavelength convergence limit of any series. 49 IONIZATION POTENTIALS Confirmation of the enetgy level concept is convincingly given by a consideration of ionization potential. Consider first a radio-type tube, Fig. 49, which con- s only a filament-heated cathode and a plate as anode. When the plate is positive with respect to the cathode, electrons will move across the tube to the anode. This current is limited by two factors. First, the number of electrons emitted per unit area from a cathode depends upon cathode composition and temperature, In the remainiler of this discussion we shall assume that the cathode is operated hot enough so that the tube current is not significantly limited by cathode emission. ‘The second factor is the effect of the electrons Fig. 4-9 Apparatus for determining ionization potentil. ace eurent) 979 in the region between the electrodes upon those emerging from the cathode. “The concentration of negative charge in the inter-electrode region, is called ‘space charge, and it lowers the potential in the vicinity of the emitting surface. Thdeed, the potential in this region usually falls below the potential of the emitter, Thus, although the plate or accelerating potential is still positive, this ~ decrease in potential due to space charge will reduce the electron current fecause the potential barrier will turn back electrons emitted with low kinetic energy. (The electrons in thermionic emission have a distribution of speeds similar to that of the molecules of a gas.) However, those high-energy electrons that got beyond this barrier caused by space charge arrive at the plate with the ‘same energy they would have had if the space charge had been absent, since the total potential difference V between the electrodes is independent of the = “space charge. The space charge limited current is not a linear function of the ‘potential difference as in the case of an ohmic resistor; itis found to be proportional to the three-halves power of the potential difference. This is known as the Child-Langmuir Law. In mathematical form, it is = V9”. The value of k depends upon the geometry of the tube and the volume density of tte charges between the electrodes. ‘The ionization potential of a gas is determined by introducing some of the gis into a tube such as that shown in Fig. 4-9, and then measuring the plate ‘cutrent as a function of plate voltage. As the potential difference is increased, {itis found that above some particular value of the potential the current increases ‘much more rapidly than it does below that value, as shown in Fig. 4-10. When the plate potential reaches this critical value, the electrons arriving at the anode have acquired enough energy to knock electrons off the atoms of the gas close {fo'this electrode. ‘The positive ions produced when the voltage equals or ‘exceeds the critical value neutralize some of the negative space charge. Thus jonization causes « marked increase in the tube current. If the gas under study ig Kydrogen, the ionization potential is found to be 13.6 V. It is a remarkable ‘confirmation of the energy-level idea that there should be excellent agreementfor the ionization potential as measured by the very different techniques of spectroscopy and electronics. Since the electrons are emitted from the cathode with an initial velocity distribution, some of them acquire enough energy to produce ionization at lower accelerating potentials than others. This accounts for the curved section joining the two straight-line portions of the grapt Fig. 4-10. Except for this short curved part, the break in the line is quite abrupt. This sudden change of slope would not occur if several low-energy electrons. ‘could combine in their “efforts” to ionize the atom of hydrogen. 4-10 RESONANCE POTENTIALS The experiment just described was set up to measure the potential through which electrons must be accelerated before they can lift orbital electrons from their lowest energy state (ground state) to infinity. This ionization was detected by an increase in the current through the tube. But before the bombarding electrons have enough energy to take the atom apart by removing an electron, they have enough energy to lift an electron to an excited state. Orbital electrons in the gas can be transferred from their lowest energy state to any of the higher states. As implied earlier, the quantum conditions which require an orbital electron to emit only certain frequencies as radiation apply also to the absorption process. These electrons can absorb only energies represented by transitions between energy levels. If an orbital electron is hit by a bombarding electron with insufficient energy to produce an energy transition, the orbital electron absorbs no energy from the bombarding electron and the collision is perfectly elastic. If the orbital electron is hit by a high-energy bombarding electron, then the orbital electron can absorb energy by making a transition. This leaves the bombarding electron with that much less energy. Such a collision, is inelastic, since the bombarding electron is left with less energy than it had before the collision. Such an inelastic collision puts the orbital electron in one of the excited states, and hence it can radiate energy in returning to a lower state, Consider again the ionization experiment. As the potential difference across the tube is slowly increased, the electrons from the heated cathode are accelerated to higher and higher velocities. At low speeds, these electrons make completely elastic collisions with the electrons of the gas, so that they are deviated but not slowed by the collision process. As the potential difference across the tube is increased, however, a potential is reached where energy can be transferred to an orbital electron. For the purpose of discussion consider hydrogen, which has an ionization potential of 13.6 V. A look at Fig. 4-8 discloses that the least amount of energy the orbital electron in the ground state can absorb is 13.6 eV — 3.4 eV or 10.2 eV. A bombarding electron with 10.2 eV of energy can “resonate” with hydrogen and transfer its energy to the hydrogen. This produces no ionization, so the current through the tube is not 4, but after making such collisions, the bombarding electrons proceed ycrogen will not glow visibly because this resonance radiation is one line o the Lyman series, which is in the ultraviolet region; however ultraviolet roscopy confirms that the radiation is there. ; Fee demonstrate the resonance phenomenon electronically, we need a more elaborate tube, such as that used by Franck and Hertz" who formed the experiment in 1913 using mercury vapor as the gas. A Pergo clesrons around is nucleus and has therefore a mich more Complex energy-level diagram than hydrogen. The important fact, however, fe that it, like hydrogen, has discrete energy levels to which the atom can be ‘ited by electron bombardment. Mercury vapor is chosen because it is tnonatomie whereas hydrogen is ordinarily a diatomic molecule whose dissocia- Hon energy is 4.5eV. I?'the gas were hydrogen, most ofthe bombarding electrons Would give up their energy to excite molecular energy states and also to dis soeiate the molecules after the accelerating potential reached 4.5V. This Complex situation would conceal the effects of atomic hydrogen in this type of iment. Pare acial parts of such a tube are shown schematically in Fig. 4-11, From the standpoint of electronics, the effect of resonance is that the bombarding electrons are stowed down, so we need a device which will measure the gnergy of the bombarding electrons after they have made collisions. Suppose hat the anode of the ionization tube is perforated or made of wire mesh. In this case, some of the bombarding electrons will pass through the electrode father than hit it, We now need to know the energy with which the bombarding rons arrive at the anode. slstron a ortion of he photolectc eft, we measured the energy of photoelectrons by making them move against the foree action of an electric Feld, and the energy of the photoelectrons was fiven by the stopping potential. Here we use much the same technique and insert into the tube another electrode teyond the anode. This collector electrode is maintained less postive than the Cathode Anode Fig. 411. Apparatus for determining resonance potential of gas. > J. Franck and G. Hertz, Verhandl. Deut, Phys. Ges. 16 (1914), p. 512.anode, say 0.5 V, so that any electrons that pass through the anode will be slowed by the field between the electrodes. Electrons which reach this last electrode must have passed the anode with an energy of at least 0.5 eV. ‘The experimental procedure consists of measuring the collector current as a function of the anode potential with respect to the cathode, and typical results are shown in Fig. 4-12. From V = Oto V = 0.5V, there is no collector current, since no electrons can reach the anode if they have less than 0.5 eV of energy. Above V = 0.5 V, the collector current rises because the number of electrons having at least this minimum energy increases. When V reaches the 4.9-eV resonance potential of the gas, the collector current begins to decrease because — some of the bombarding electrons are slowed by inelastic collisions with orbital electrons in the gas. The current rises again as V is further increased since, in the stronger field, bombarding electrons can make inelastic collisions early and still undergo enough acceleration to surmount the 0.5-V barrier. ° 5 1520 10 Anode voltage Fig. 4-12 Resonance potential curve for mercury. The second dip is not due to a new energy transition, because very few electrons ever get enough energy to excite the next transition. The second dip occurs at twice the resonance potential and is caused by the bombarding clectrons suffering two inelastic collisions of the same kind. ‘Thus each peak of the curve signifies more collisions and cach peak is an integral multiple of the resonance potential. ‘The separation of successive resonance peaks is 4.9 V for meroury*. ‘Since mercury ionizes at 10.4 V, the third resonance peak will be masked by other effects if the tube shown in Fig. 4-11 is used. Actually, a more complicated tube which differentiates the resonance and ionization effecs is used. This important experimental detail does not in any way alter the principle discussed in this section. Franck and Hertz studied the radiation given off by the mercury vapor bombarded by the electrons and found a well-known mercury line at 2 = 2537 A. ‘The corresponding photon energy is given by Eq. (3-47) as 1.24 x 10* = 4gev. 2537 ‘The radiation from the mercury vapor was detected only when the anode = voltage exceeded 4.9 V. Clearly, the energy level in mercury at 4.9 eV cannot = be excited by multiple hits from electrons; it has to be reached in one single step. This is clear evidence that mercury cannot exist with an intermediate tcnergy content and is the strongest proof yet devised of Bohr's idea of discrete ‘levels. Fig. 4-13 Energy-lovel diagram illustrating absorption and emission of energy by mercury. Figure 4-13 shows the ground state and fist Lected state of mercury in an endrgy level diagram where the ground-state energy has been chosen as refer- «Ghee The absorption and emission processes of the Franck-Hertz experiment are indicated by arrows. © 4-11 PHOTON ABSORPTION th the previous two sections we have discussed how an atom can absorb = iserete amounts of energy from bombarding electrons. An atom may also | ghsorb energy from photons, but there is an important difference. Absorbed = photons disappear entirely. “A photon with more energy than the ionization Grergy of an atom can always be absorbed because the excess energy will Sppear as Kinetic energy of the photoelectron. A photon with less than the Jonization energy cannot be absorbed unless its energy is equal to one of theexcitation energies of the absorbing atom.* Consider hydrogen again. Or- dinarily, the probability of finding a hydrogen atom in an excited state is very small; therefore, we assume that itis always in its ground state. Thus hydrogen. can only absorb photons whose wavelengths correspond to those emitted in the far ultraviolet, the Lyman series. Hydrogen atoms are therefore transparent to visible and infrared light. If we pass radiation of all wavelengths through hydrogen and analyze the transmitted light by means of a spectrograph, we find the transmitted intensity reduced for the Lyman wavelengths. Such a spectrum, having a bright background and dark lines, is called an absorption spectrum. Because the atoms that have been excited by the absorption of radiation re-emit photons in random directions upon returning to the ground state, there is a decrease of intensity along the direction of the transmitted radiation, ‘The absorption lines observed are actually very faint bright lines that appear dark by contrast. There are certain advantages of studying absorption spectra. For many atoms, as for hydrogen, the absorption spectrum is simpler than the emission spectrum. For hydrogen, as mentioned above, one cin observe absorption for the Lyman series only. From these lines, one can directly construct an energy level diagram. According to our present simple theory, this energy-level diagram appears to be complete. It is not, because selection rules (see below) do not allow absorptive transitions to all states, in particular for many-electron atoms. However, absorption spectra give the most direct information on the levels that do connect with the ground state by photon emission or by absorption. ‘The determination of elements on the sun is a dramatic example of absorption spectroscopy. The sun is a hot body which emits a continuous spectram ‘of photons. As these photons pass through the outer atmosphere of the sun, wavelengths which are characteristic of the gases present are absorbed. Thus the continuous spectrum of the light from the sun is crossed with (relatively) dark lines which were first observed by Fraunhofer in 1815. Most of the Fraunhofer lines correspond to the wavelengths of elements found on the earth. The absorption lines of the Balmer series of hydrogen are especially prominent in the spectrum of the sun. The Balmer lines of hydrogen and the visible-region lines of other elements have rather long wavelengths and can be absorbed only by excited atoms of the respective elements. These higher energy states are produced in the following way. Although the gaseous atmosphere of the sun is cooler than its surface, the temperature of the gas is still so hhigh that a large number of atoms have sufficient kinetic energy to excite * This statement is only approximately true, First, an excited state has a certain ‘width, so that the energy of the incident photon can vary within some narrow limits, and absorption still takes place (see Chapter 8). Secondly, in a process called the Raman effect, a photon of energy higher than a given energy level can with finite but low probability excite the atom to that state, with the excess energy being carried off in the form of a reduced-energy photon. ‘other atoms by collision, Thus these atoms are raised above the lowest ltraviolet, For many years one set of Fraunhofer lines in the visible region vould not be associated with any known element. It was presumed to be due to anew “sun element” which was appropriately named helium. This hypothesis ‘yas confirmed when helium was isolated on the earth and its emission spectrum vas found to correspond with the previously unidentified Fraunhofer lines. Many of the spectral series that are characteristic of molecular structure ‘are in the infrared region. Absorption spectroscopy is the only feasible way af investigating the structure of these molecules that would be dissociated by ‘feing excited in an electric arc in an attempt to produce emission lines. © Molecular spectra are further discussed in Chapter 11 : 4-412 A CRITICISM OF BOHR’S THEORY - Max Planck discovered that energy was exchanged between a heated body and © the surrounding electromagnetic field in quantized steps. Niels Bohr discovered = “quantization of angular momentum and quantization of the energy content of © the atom. He interpreted the observed discrete frequencies of emitted radiation = as arising from “quantum jumps” between discrete atomic energy levels with the frequency v given by iv = AE. Bohr's theory, or parts of it, has been | superseded by wave mechanics, in which the quantization of angular momentum “nd energy is shown to result from the wave nature of matter. The principal shortcomings of Bohr’s theory are the following: 1, The emission spectrum from the hydrogen atom exhibits more details, =e, splitting of the lines, when recorded in a spectrograph with high resolving power, By assuming elliptic orbits in addition to circular ones, and by sep- ) prom en (a) Before heating Bed () After heating Fig. 5-6 (a) Electron scattering from a block of nickel, ‘enystal of nickel. 1nd (b) election diffraction from aFig. 5-7 (a) Electron diffraction pattern of alt io cera teay ieee axe diferent but the similarities are evident. Davi re evident. Davsson and Geter, usin destone iad of ry, pated an experiment ae i vt y oT Compton on difactng sok xray fom a mancmade gat a Co 20 rays fro ade grating at grazing ‘With the complete verification of the dee i ication of jroglie hypothesis, we have arrived st pt we ou atone word hs Sang ae Compon tering sl jat waves have particle aspects and the deBroglie hypoth ws. that particles have wave characteristics. ae In the following two section shall di the ns we shall discuss the mathematical theory of we groups and in the rest of the chapter some of the physical implications. Werte now Gvtig fom a een of te hon evetnment of Jem physics and going into 0 more formal development of wave mechanics. I ec hl temp es he ls mre ly thn ington did when he humorously suggested thatthe primordial entity i really a “wavicle.” ee ne 5-6 WAVE GROUPS ° . Although the interference effects compel us to aocept the fas like aa re ene a waves oad parce ti sl ie to see how the wave extended over a region of space does not conflict with the small particle located at a specific position. This apparent conflict can be resolved by the formation of what is called a wave group or wave packet. In essence this is a wave which extends over a very limited region of space. It is formed by applying the principle of superposition to many waves, When many waves travel through the same medium, the resultant wave is the summation (in Cour case it will be algebraic) of all the waves. In acoustics such a summation fof two waves of slightly different frequencies produces beats, which are 0) ~ @2 = ftuctuations in the amplitude of the resultant wave, We shall see that when recat more frequencies aze added, the regions where the amplitude i arg can be made smaller. ratty the disoussion ofthe superposition we will sea one-dimension@) waves that is, wave whose displacement, y, depends on the positon slong 2 Trae. andthe time fA sinusoidal wave with amplitude A may be written as y = Asin 2x & - ») = Asin (kx — 0), (5-20) “where we have introduced k, the propagation number which equals 25/2, and Tee sagular frequency @ = 2nv. Even with one-dimensional conditions is citrate to draw the wave, since there are two independent variables. What Ye diy do is hold one fixed and vary the other. Thus we ether dram the displacement as a function of postion at some fixed time (equivalent toa Srapshot) of the displacement as a function of time at some fixed position. In _ Ta llowing development, we will use time as the variable. ‘The results ‘would vee ee came using the position as the variable if ¢ were replaced by = and © ‘were replaced by k “rhe tame dependence of a sinusoidal wave with amplitude 4 can, be concidcred ne a component of a vector of length A which is rotating with 2 Oy ‘The resultant wave is then y = 2A 008 Hfdie x — do £] 008 (kx — 1). (5-26) “The second cosine function times A is one of the original waves. ‘The cosine © ametion in front modulates this wave with another cosine function of wave length n/dl and a frequency (da/x). From the product of these we can get the velocity of propagation of the modulation, which is the group velocity ty: z) (do) _ do w= (GG) oie ay “The group velocity depends on the way in which the frequency varies with the Fropbgation number, Since we have considered only two phase waves, it i veeie that ¢, wil differ for diffrent pairs of phase waves; that i, the group Ports may be a fonction of frequency. When many waves are added, the derivative of Eq. (5-27) is evaluated for the central frequency of those used in the summa If the wave group is to represent a particle, then it is necessary ‘that the speed of the group and the speed of the particle be the same. IF these speeds ‘Mffered, the particle would soon be in a region where the amplitude of the "wave is negligible and the wave would not give a useful indication of the position of the particle. Using the mass-energy relation from relativity, we tan find the particle velocity in terms of its momentum and energy: = pim = perfec? = peE. (5-28) ‘To get the group velocity in the same terms we use p = h/A and E = hv which are written as (5-29)* The group velocity can then be written as do dk aE dp” To evaluate this we need the energy-momentum relationship from reat Eq. (A3-48); EP = pict + mic Differentiating this, we obtain 2E dE = Spe* dp. ‘The group velocity becomes 631) ‘The particle velocity (Eq. 5-28) thus equals the group velocity (Eq. 5-31), ‘We therefore see that the choices that were made for the frequency and wave- ength to be associated with the particle also lead to the satisfying result that the wave group and the particle have the same speed.* Thus it is possible to have a wave motion which has the particle characteristic of being in a small region of space and which will move with the particle's speed. 5-7 FOURIER ANALYSIS: ORTHOGONALITY Most readers will be familiar with the mathematics of Fourier series used to deseribe periodic functions, For instance, a furetion of x with a period L can bbe expanded in a Fourier series as follows 2anx nn 10) = J (Ancos 7 +B, 2a) L L %, (5-32) For example, the transverse displacement of the plucked guitar string mentioned in Section 5-3 can be described by this function. We then think of the guitar string as continuing indefinitely, and the simplest description is obtained by making the period L equal to twice the length of the string. Thus the length of the string is half the wavelength of the fundamental (lowest frequency) for which n = 1. We further have 4, = 0 for all values of n because the cosine functions do not satisfy the boundary requirement that the displacement is zero for x = Oand x = L. Therefore B, is the amplitude of the fundamental, B, is the amplitude of the second harmonic, Bs is the amplitude of the third trarmonic, etc. + It isa curious fact that if we had used the classical expression E = Fy = 4m? = ‘p?[2m, we would also have found agreement between group and particle velocities. Equation (5-32) can be written as cet, Fs) 3) "The proof of this statement is left to one of the problems. Note that the ‘integer n now runs from —20 to +00. We thereby get positive and negative fxponentials under the summation sign as needed to produce cosine and sine “functions. Assuming that the coefficients 4, and B, in Eq. (5-32) are real, as, zequired for instance in the case of a plucked string, the coefficient C, will 'be complex. Tn order to find the coefficients C,, we assume that the function f(x) is ven, for instance, in tabular form or in analytic forms broken up over several parts of the period. Examples of the latter are sawtooth functions and square - Graves well known in electronics. If f(x) is known in such a form that it can te integrated, we find the coefficients C, in the following way. We multiply ‘each side of Eq. (5-33) with e~?*™/" and integrate over x from —Z/2 to +Z/2: XG [io peveomas = -ua “The function in the integrand on the right-hand side of Eq. (5-34) is a sum of ‘a sine function and a cosine function, both with periods equal to L/(n — m), ‘except when n = m, in which case the function is unity. For n % m, the inte- - grals from 1/2 to +L/2 are integrals over a whole number of wavelengths, and they therefore vanish, We can then write [oermtan 9 cet yoartenle " L Jz (5-35) ‘We have here transposed the two sides of Eq, (5-34) and written C, instead ‘of C,, because only the term in the series for which n= m remains on the right-hand side of Eq. (5-34). Equation (5-33) can be used to describe any periodic function with period of length L and with the coefficients C determined by the aid of Ea. (5-35). For instance, this analysis can be used to describe the transverse displacements of a plucked guitar string, as described above, because the function is defined ‘ver a limited range in x and we can therefore arbitrarily call this range half the period. ‘The function (5-33) can also be used, for instance, to describe the "pressure oscillation in air near a piano in which middle C has been struck, ‘The independent variable x is then time, and is the period. The coefiicients‘C will be slowly decaying functions of time with a different decay rate for the various harmonics.* ‘The functions e?*™4, with n being an integer running from ton = +00, are orthogonal functions in the interval x = —/2 to x Two functions $,(x) and $(x) are orthogonal in the interval if which we recognize as being the propagation number k = 22/4, We further obtain Ak = 2nAn/L with An = 1. For an expression such as Eq. (5-33) to be meaningful, the series must ‘converge. This means that the coefficients C,, in general, will decrease as n increases. However, one can get an arbitrarily good description of f(x) with | atruncated series, that is, with a finite number of terms, The longer the interval fy the more terms are needed, and cach individual C, will then be smaller. ‘Coming back to Eq. (5-37), we see that for f(x) to remain finite, the product ‘of L and C, must remain finite, so C, approaches zero as L goes to infinity. ‘We therefore introduce a new function to replace the product LC,: ok) = 2x)? LC. (5-39) 0 41/2. atox ff eee for ném. 636, Here is the complex conjugate of dq, that is, @y is produced by changing the sign of all terms containing i = /—1. The orthogonality of the functions used in Fourier analysis is, of course, necessary for Eq. (5-34) to reduce to Eq. (5-38). It is not only sine and cosine fanetions and complex exponentials that form orthogonal sets of functions. In later chapters we shall discuss further the fact that the solutions of: ‘Schrédinger’s wave equation in general form orthogonal sets, and that arbitrary functions can be expanded in terms of these functions in a similar manner as that expressed by Eq. (5-32) or (5-33). Another feature of the trigonometric functions or complex exponentials is that they form a complete set. This means that an infinite series, such as Eq, (5-33), can always be found to describe f(x) exactly, provided f(x) is finite and has a limited number of discontinuities in the interval x = —Z/2 tox = +L/2. For the proof of this statement, we have to refer to a textbook in mathematics. In the discussion above, we have defined the function f(x) in an interval 1/2 to +L/2, and itis clear from the formula that the function will repeat itself in successive intervals of length L. Now we see that we can increase the length L of the interval to include any region that we may be interested in, and it would not matter if f(x) repeated itself outside these limits. Indeed, there | is no reason why we should not let L go to infinity. In order to do so, we first write Eq. (5-33) as The factor 22)" is included for symmetry reasons, as will be immediately evident. Making this substitution in Eq. (5-37) and converting the sum to = integrals, we obtain Fea) = ay" foe ak 6-40) This is a Fourier series, in which the wavelength does not change in discrete jumps as in Eq. (5-33) or (5-32) but rather varies continuously (or stated “Gifferently, the wavelength changes by infinitesimal steps). ‘The amplitude function g(k) is found very simply by substituting Eq. (5-39) into Eq. (5-35), ‘changing the subscript from m to n. The result is ah) = nyt J fener dx 4) "The function g(K) is called the Fourier transform of f(x), and the two functions F(a) and g(k) are called Fourier inverse functions. ‘As an example, consider the Gaussian distribution function Fe) ‘where o is the standard deviation or root-mean-square deviation from the jncan value Z = 0. This function is plotted in Fig. 5-10. We find the Fourier ~ transform from Eq, (5-41) to be (3-42) Slo) = "YEE mak, 3h where we have introduced a new variable k = InnjL, (5-38) afk) = 21g Mg? f eBlog dy We are oversimplifying a bit. ‘The middle C key strikes not one but three strings in the piano. The three strings are intentionally tuned to slightly different resonance Dag Bg 12 fr e728 cos kx de. 6-43) frequencies (fundamentals): otherwise, the piano would sound dull, Also, the “har- monies” do not have exactly two or three times the frequency of the fundamental because of stiffness of the strings. ‘We have here written e~M* = cos kx — isin kx and made use of the fact that an antisymmetric function (ie., the one containing sin kx) integrates out tozero. The last integral can be found in a table of definite integrals*. Inserted’ into Eq. (5-43), it 1M ght Balt, ak) This transform is also plotted in Fig. 5-10. (44) 23 —| Fig. 5-10 The Gaussion function f(x) and its Fourior tansfor 9(K) which is also Gaussian, We can write the exponential in Eq. (5-44) as e~"*%, ‘The standard deviation a, for the transform can be found by comparing Eqs. (5-42) and (5-44). We find that o = Vo. 6-43) Consider the Gaussian (Eq. 5-42) as an example of a deBroglie pilot wave. ‘The transform (Eq. 5-44) then represents the spectrum of propagation numbers Je necessary to describe this wave or the wavelength spectrum with 2 = 2x/k. If we use the standard deviations as measures of the width of the wave train and its transform, we find from Eq. (5-45) that Ax Ak ~ (6-46) Using the widths 2V2 ¢ and 2V'2/o indicated under the curves in Fig. 5-10, wwe obtain Ax Ak = 8, (6-47) which can be compared with Eq. (5-25). 5-8 WAVE-PARTICLE DUALITY ‘The wave-group idea is an attempt to resolve the conflict contained in the fact that some explanations of experiments require the use of waves while others require the use of particles. This interpretation means that an electron is group of matter waves and a photon is a group of electromagnetic waves, + Forinstance, in Handbook of Chemistry and Physies, Chemical Rubber Publishing Company, Cleveland (any year) ‘that i, there are only waves in nature but they form wave packets which we call particles. As indicated in Section 54, the trajectories of the particles are due to the refractions of the waves. Macroscopic objects consist of many of these wave goups and the interactions between objects are simply interactions of waves. ‘Although this “reality” of the waves seems reasonable, there are some difficulties associated with it. When a wave is incident on a boundary between ‘two media, it generally splits into a reflected wave and a refracted wave. The incident wave group becomes two groups. If the wave group is to be an electron, ‘where the boundary would be produced by a change in electric potential energy, ‘we know that the electron does not split. It is difficult to see how the two wave groups, one reflected and one refracted, can be the single electron. In addition, the Coulomb force, which is stated for point charges and acts between electrons, + canmot be handled conveniently with this interpretation. ‘Another possible interpretation was presented by deBroglie. In this the | patticles are “real” and the associated waves are pilot waves which guide the ~ particle and are to be looked upon as probability waves. The amplitude of ~ these probability waves at a certain position is a measure of the probability of | the particle being at that position. The pilot waves are abstract quantities and the word “wave” is used as it is, for instance, in the phrases “wave of enthu- ‘siasm” and “crime wave.” Since the waves are not directly observable, there is no necessity for a medium. This interpretation removes the difficulty of the ‘wave group concept when the wave is partially reflected and partially refracted. © For the probability waves, the amplitude of the reflected wave determines the ‘probability that the particle reflects from the boundary and the amplitude of the refracted wave determines the probability that the particle penetrates the = boundary. A single particle will do only one or the other, but a large number of particles will divide according to the probabilities. This pilot wave interpretation, which means that a light beam is a stream ‘of photons, also runs into difficulty. Suppose that we pass a light beam through “a pair of slits separated by a small distance, as in the interference experiments of “Young. The interference pattern produced on a screen consists of alternate “ight and dark bands. The positions of these bands are determined by the differences of the paths of the two waves proceeding from the slits to the ‘screen, ‘The alternate light and dark bands represent alternate large and small “© probabilities of a photon arriving at the respective positions on the screen. If one of the slits is covered, the interference pattern changes, which means the "probabilities of where the photon will artive at the screen changes. For example, ‘the photon may now have a large probability of arriving at a position where the previous probability was very small. If the photons are real, however, they ‘must go through one slit or the other. Consequently, as the photon goes ‘through one slit, its motion is influenced by the other slit and the photon is able to know whether the slit through which it did not pass is open or not.“This gives some “intelligence” to the particles which does not seem reasonable: We will return to this problem in Section 5-10. ‘The resolution of the conflict between waves and particles lies in an appraisal of what we mean by a wave and a particle. Both of these terms, when applied to the fundamental entities, are abstractions of the human mind which are arrived at by extrapolation from the macroscopic world of grains of sand and waves on strings. The following is another very clever and useful trick of the human mind. A hollow rubber ball has its center of-gravity at its center. Discussion of most, but not all, of the motion of the ball can be greatly facilitated by regarding the ball as a point mass with all its mass at the center of gravity. The center of gravity has no objective reality, and if someone cuts ‘the ball open, points to the center and says, “Ha! You see, there is no mass there,” we reply that the center of gravity makes a poor description of what is at the center of the ball, but that it continues to be useful in describing the ‘motion of the ball. No one description of the ball can ever completely represent what the reality of the ball is. In the same way the particle description and the wave description are each incomplete in attempting to describe physical reality. ‘The mistake of those who say that interference shows that light i a wave phenomenon is a verbal mistake that is made every day. We point to a map and say this és the United States. What we mean is that this diagram on a piece of paper is a scale representation of many of the physical and political features of the United States. We know that the real United States cannot be folded, rolled up, or burned. We know that the states are not different in color, only a few square inches in area, and completely flat. ‘The map is a lever, useful, elegant model invented by the human mind. More may be learned about some aspects of the United States in one hour of map study than in a lifetime of looking at the real United States. We do not scoff at maps ‘because they are unreal, we admire them as useful descriptions. Both the wave and the particle are models we have constructed in attempting to describe matter. Quite naturally, we do not expect either model to give a complete description. Some properties, such as interference, are contained in fone model, the wave, while other properties, such as mass, are contained in the other model, the particle, The two models complement each other in that together they give a description of matter. Thus we should say that the electrons are waves and particles, not waves or particles. The same statement can be made about electromagnetic radiation. During an experiment the particular ‘model which is used is determined by the apparatus used. Even though we admit that waves and particles are not “real,” it is very awkward to talk about experimental procedures and results in such @ way as to indicate this. Consequently, we will still make statements which seem to imply that particles exist; for example, we say that the intensity of the wave is 1a measure of the probability of the location of the particle. This is for convenience only. Both the wave and the particle are incomplete models and both Me necessary for a description of all the properties of matter which are no conflict in experimentally determined. With this interpretation, there the dual nature of matter or electromagnetic radiation, 5-9 THE HEISENBERG UNCERTAINTY PRINCIPLE ‘An important consequence of the wave-particle duality can be developed from ‘he wave group analysis of Sections 5-6 and 5-7. In an attempt to get a wave {har had a limited extent, we added many waves to form a wave group. If we Correlate the wave model to the particle model by assuming that the amplitude St the group measures the probability of the particle being at that position, we seat there is still no certainty in knowing the location of the particle. It Jd be anywhere in the group. cone p decrease the uncertainty in location of the particle, we have to reduce the size of the group, Ax. I we rearrange Eq. (5-25) into Ax Ak = 4n, (5-48) ‘the size of the group can be reduced by increasing the spread of propagation umbers, Ak. Tt appears that we may climinate the uncertainty in position af the particle by using an infinite spread of propagation numbers. We see from &— 2np[h that the momentum of the particle is determined by the propagation number of the wave, If we use an infinite spread of propagation vambers, we wil have an infinite spread in the momentum of the particle, When we decrease the uncertainty of the particle's position, we increase the Tneertainty of the particle's momentum, If we put Eq. (5-48) in terms of ‘momentum, we have Ax Ap = 2h. “The coefficient off in this expression was obtaied through our simplification Gresing waves with the same amplitudes and with a uniform spread of fre- ent number for this < Guencies. A different choice of these would give a different num! «Ueficient. If we choose to characterize the uncertainties Ax and ‘Ap by root. ~ Shean-square deviations rather than the widths of the central maximum as in Fig. 5-9, we find Ax Ap, © h. (5-49) smpare Eq. ($-46) with Eq. (5-49)- the ‘abe 1) sided to the momentum to indicate thatit isthe momentum associated with the x-displacement. Equation (5-49) is interpreted as indicating ‘hat the uncertainty in position of the particle times the uncertainty in the fmociated momentum is approximately A. The words associated momentum fine used because in the three-dimensional case there are also momentum Components in the )~ and 2- Cine /mole degree functions of temperature for copper; note that at low temperatures their difference becomes very small ea and that both go to zero at T= 0. It is essentially the temperature 41. The temperature vat variation of the specific heat at [71 ne Cy and “Cy for copper. [By permission constant volume which will be from M. W. Zemansky, Heat and Thermo- Giscussed in the present chapter. dynamics, ded, McGraw, New York, It may be noted that ifno direct 198 p.237) | compressibility data are available for the temperature range of interest, cone frequently employs the relationship C=C ymyT) with = apVIKCy 27) The quantity y is called the Griineisen constant and is practically independent of temperature. Thus by calculating y at some arbitrary temperature from available data, one may obtain an approximation for Cy at other temperatures from a knowledge of the coefficient of volume expansion. «Sep, for example M.W. Zemansky, Heat and Thermals, 248, MeCi86- a, New Von 9, p22. TE Geinciten, Handbuch der Physi, 10, 1 $9(1926); se aon J.C. Shier. Ph TF Grtinelsen, |a SPECIFIC HEAT AND LATTICE VIBRATIONS — (Chap. 2 From the atomic point of view one may distinguish between various contributions to the specific heat of solids. In the first place, there is the contribution resulting from the atomic vibrations in the crystal; an increase in temperature is associated with a more vigorous motion of the atoms, which requires an input of energy. Second, in metals and in semiconductors there is an additional contribution to the specific heat from the electronic system. Usually this contribution is small relative to that of the lattice vibrations, as explained in Chapter 9. As the temperature is raised from absolute zero, the 0 specific heat increases rather rapidly from zero and finally levels off toa nearly constant value. For Gs clements, the value at high temper- t atures is about 6 cal mole degree". This is known as the law of Dulong and Petit. Anomaliesin thespecificheat curvesareobserved in the ferromagnetic metals; for example, in ickel,iron,andcobalt, a peak is observed in the vicinity Fig. 2-2. C, in cal mole-* degree" for of the ferromagnetic Curie tem- nithel as function “of the absolute perature (see Fig.2-2). The height temperature. ‘of the peak is of the same ‘order of magnitude as the normal specific heat. The peak is associated with the transition from the ferromagnetic (ordered) to the paramagnetic (disordered) state. Similar peaks occur in the specific heat curves of alloys which exhibit order-disorder transitions, and in ferroelectric materials. These anomalies are discussed in the relevant chapters; in the present chapter the discussion is confined to the specific heat associated with atomic vibrations. 0» 300" G00" B00” 6001000 Yat =r 2-2, The various theories of the lattice specific heat In Sec. 2-10 it will be shown that the vibrational energy of a linear chain of N atoms may be expressed as the energy of N harmonic oscillators. Extending the arguments employed there to the three-dimensional case, cone is led to the conclusion that: The vibrational energy of a erystal containing N atoms is qa AAT TS aS ego of IN ane Tear with The energy of a systems of 3N harmonte osciMators. distinction Between the various theories is based on their differences in the proposed frequency spectrum of the oscillators. The central problem jv the theory of the specific heat is therefore the calculation of the See, 2.2] SPECIFIC HEAT AND LATTICE VIBRATIONS 35 Wavelength and ‘of the possible qhrodes ot VibratTOp of the tal under consideration. The different approaches if problem ‘will be outlined below. —— With regard to the harmonic oscillator representation referred to above, the following qualitative remarks may provide some clarification. Suppose it were possible to fix the position of all the nuclei in a crystal such that they are all in their equilibrium position. If one of the nuclei were now displaced over a distance small compared with the shortest interatomie distances, and then set free again, the displaced atom would carry out harmonic vibrations about its equilibrium position, and its energy of vibration would be the same as that of three one-dimensional harmonic oscillators, one for each direction of motion. Applying the same reasoning to the other atoms in the crystal, one arrives at a system of 3 harmonic oscillators representing the vibrations of the crystal as a whole. 2-3, The breakdown of the classical theory The energy of a harmonic oscillator of natural angular frequency may be written Pim motgsl 28) where the first term on the right represents the kinetic energy (p'is the momentum) and the second term represents the potential energy (7 is the deflection from the equilibrium position). It is well known that the average energy of a harmonic oscillator according to classical statistical mechanics is given by (Oe [ec a] cde ED 025) where 7 is the absolute temperature and & is Boltzmann's constant. It is important to note that the frequency does not enter in this result. In other words, the vibrational energy of a crystal of N atoms is classical always equal to CEH BMD (2-10) independent of the assumed frequency distribution of the oscillators used in the model, Now, as Jong as the volume of the solid is kept constant, (@-10) is the only temperature-dependent contribution to the total energy ‘of the system. Thus, for a solid containing one type of atoms and putting 1 equal to the number of Avogadro, one obtains for the specific heat per gram atom, 5.96 cal degree"! mole? (211) where Ris the gas constant. Similarly, if the solid consists of N atoms A and N atoms B, the specific heat per mole would be 6R, ete. The result36 SPECIFIC HEAT AND LATTICE VIBRATIONS — [Chap.2 obiained is in quantitative agreement with experiment (if sources of the specific heat other than lattice vibrations are subtracted) at high temperatures only, In other words, it does not explain the decrease of Te speaTRC— Theat at low temperatures, as observed for all solids. This discrepancy is essentially removed when quantum theory is used, as will be seen below. It may be noted that the classical theory led to similar difficulties in the specific heat of molecules. 2-4. Einstein’s theory of the specific heat A great step forward toward an understanding of the specific heat curves at low temperatures was made by Einstein in 1906.* Although the physical model employed by Einstein was oversimplified, his results definitely indicated that quantum theory contained the answer to the difficulty encountered in the classical theory. He assumed that a solid clement, containing N atoms, could be represented by 3N harmonic TSO We STURGES THE eT pls at he atoms Vibrate independently of each Olfer, their frequencies being the same because of their assumed identical surroundings. For the average energy of an oscillator Einstein made use of a result obtained by Planck in 1900, in connection with the theory of black-body radiation. According to Planck, a harmonic oscillator does not haveacontinuous energy spectrum, as assumed in the classical theory, but can accept only energy values equal to an integer times fy, where h is Planck’s constant. The possible energy levels of an oscillator may thus be represented by* By replacing the integrals appearing in (2-9) by summations, one thus obtains for the average energy the expression 0,1,2,3, (412) co = Sines Seer ery ocala hs expen, et consi the denominator ieee Differentiating with respect to 1/kT, one obtains as on hye" ayer) ~~ 2,8 = — ewe It is observed that the expression in the center is identical with the A. Einstein, Ann, Physik, 22, 180, 800 (1906); 34, 170 (1911), «See any Introduction to Modern Physics Sec. 24] SPECIFIC HEAT AND LATTICE VIBRATIONS 7 numerator in (2-13). Substitution into (2-13) thus leads to the well-known Planck formula for the average energy of an oscillator at a temperature T: a nD = a (2-14) We emphasize “that in contrast with (2-9), this expres contains he frequency of the oscilator- The temparture depos 3) i iltrated in Pig. 23, showing (oIRT as fonction of W/kT. Note thatat high emperatrese) ET, in agreement ith the astal theory. However. at low temper oar » oT aa hen Fig. 2:3. The average energy in unite of KT of a harmonie oscillator of frequency » asa function of hr/AT, according to Planck atures, («) decreases exponentially to zero. In the Einstein model, the vibrational energy of a solid element containing N atoms is thus equal to -—~. ia The specific heat at constant volume is therefore per mole a hy y ur o=g a) 216) en Before discussing this result, it may be remarked that according to quantummechanies, the possible energy levels of a harmonic oscitlator are given by ale Dh m= 0,12, (2.17) rather than by (2-12)5 This has the effect of shifting all energy levels by ‘the constant amount of Av/2, and instead of (2-14), one obtains hy hy -3+ yur] (218) ‘The first term is called the zero-point energy of the oscillator becau (= fof for T = 0. Thus, according to quantum mechanics, the atoms ‘have vibrational energy even at absolute zero, The expression for the specific heat is not altered by this result, because Cy. is determined by the derivative of (¢) with respect to 7. With regard to (2-16) itis observed that for AT'S> hr, this expression rediices in first approximation to the classical result (2-11). At low temperatures, however, the specific heat decreases. To discuss this See ere38 SPECIFIC HEAT AND LATTICE VIBRATIONS {Chap.2 behavior, it is convenient to introduce the Einstein temperature Op, defined by (2-19) Expression (2-16) may then be writeilt the form & n) Ea F,(% (2-20) where Fy is called the Einstein function; it determines the ratio of the ver 1 L ! o 24 6 8 16 1d 4a 48 aa a0 70 Fig, 4. The Debye and Einstein functions as function of 710. specific heat at a temperature 7 and the classical (high-temperature) value 3K. The Einstein function is represented in Fig. 2-4, together with the Debye function, which will be discussed in Sec. 2-6: We see that the curve obtained has the same appearance as the observed specific heat curves, On the other hand, the Einstein curve deviates from the experimentally observed ones in the region.of low temperatures, Experimentally, it i found that for most solids the lattice specific heat at very low temperatures (liquid Relium) is proportional to T*. However, for T1. The wavelengths and frequencies of the possible vibrations represented by (2-22) are A,=2L/n and “A, (2-23) The frequency spectrum is discrete, one frequency corresponding to each integer value n. Note that for the one-dimensional string the frequency Spectrum corresponds to an infinite number of equidistant fines, as illustrated in Fig. 2-Sa. The number of possible modes of vibration in a frequency interval dy is, on the average, equal to dn = (2L{e,) dv (2-24) In the three-dimensional case, the wave equation reads a ae wet pt iat a ha ©P. Debye. Ann. Pheik 20 720 10177 (225)40 SPECIFIC HEAT AND LATTICE VIBRATIONS — {Chap. 2 Assuming 2 continuous medium in the shape of a cube of edge L and assuming the faces of the cube to be fixed, the possible standing wave solutions are, in analogy with (2-22), ux = A sin (ngx/L) sin (n,ny/L) sin (n,rz]L) c08 mot (2-26) Where now mm my and n, ate positive integers 31. Substituting. this solution into the differential equation (2-25), one obtains the following expression for the possible modes of vibration: (ILENE + mp + nb) — Antv®fck = nti? 27 ‘Thus the possible wavelengths and frequencies are determined by three integers in this case. Let us now ask the question: What is the number 2 | 201 a t I Fig. 2-5. (a) Frequency spectrum for a finite continuous string, according to (2-24). (b) Frequency spectrum for a three-dimen, sional continuum, according to (2-30), ta) ‘by of possible modes of vibrations Z() dv in the frequency interval between Y and ¥-+ dr? To answer this, considera network of points, each point being determined by three Cartesian postive integer coordinates mr ma and n,. Writing " Rem nk tn + of = atalet 228) itis evident that the number of points in a shell between R and R-+ dR is equal to? A4nR* aR (229) 7 Each point occupies on the average a unit volume in the integer spoce. Sec. 2-5] SPECIFIC HEAT AND LATTICE VIBRATIONS 41 Now, each point corresponds to a set of three integers m,. mys. and each set of integers determines, according to (2-26), a possible mode of vibration; hence (2-29) immediately gives the number of possible modes of vibration in a given range. Expressing R in terms of » in (2-29) one thus finds Z(0) dv = (be fe? ae 2:30) where V is the volume of the solid. For a perfect continuum, the possible Frequencies vary between 0 and 2, the number of such possible vibrations increasing with the square of the frequency (see Fig. 2-5b). This situation holds, for example, in the case of electromagnetic waves in a box of volume V. Expression (2-30) is therefore basic in the theory of black-body radiation. In the case of elastic waves, we may distinguish between transverse and longitudinal waves. In general, the velocities of propagation, say c, and qn Fespectively, will not be equal. To set up an expression for Z(v) din this case one should keep in mind that for each frequency or wavelength there are two transverse modes and one longitudinal mode.* Thus, instead of (2-30) one obtains 20) dh 31) How this expression has been used in the theory of the specific heat of solids will be discussed in the following. two sections, 2-6. The Debye approximation ‘One may wonder what the discussion of the preceding section could have to do with the specific heat of crystals, which are by no means continuous but are built up of atoms, i.e., of discrete ‘mass points.” The reason is the following: Consider an elastic wave propagated in a crystal of volume V. As long as the wavelength of the wave is large compared with the interatomic distances, the crystal “looks like” a continuum from the point of view of the wave. The essential assumption of Debye is now that this continuum model may be employed for all possible vibrational modes of the erystal. Furthermore, the fact that the crystal actually consists of atoms is taken into account by limiting the total number of vibrational modes to 3N (see Sec. 2-2), N being the total number of atoms. In other words, the frequency spectrum corresponding to a perfect continuum is cut off so as to comply with a total of 3N modes (see Fig. 2+6a), The Debye cut-off procedure leads to a maximum ‘In the longitudinal modes, the deflection is along the direction of propagation: inthe transverse modes the deflection is perpendicular to the direction of propagation, which gives two independent components.2 SPECIFIC HEAT AND LATTICE VIBRATIONS — [Chap. 2 Freavency 1» (the Debye frequency) common to the transverse and longitudinal modes ; it is defined by {°20) de = anv G +a)itre iN (2:32) 2 ON (2 iy! 3 7 (3 +4) > Ne (233) ‘where Z0) as given by (231) has been used. It should be voted that tis Procedure assumes thatthe velocities c, and care indevendeoy rine 2) t 2) “ by) Fie 24. The Debye cuc-of takes place at the Debye frequency Yur common to the tansverse and longitudinal modes (2) Born's procedure, the eutcoff takes place at a common minimum avclength, corresponding to the maximum frequencies», and >, for the transverse and longitudinal modes respectively (6). Now, that e¢ < 6 Navelength, asin the continuum, It willbe seen in Sec. 29 that ths is not Correct for actual crystals, The order of magnitude of »,, may be obtained by Jaking N/¥=x 10* perem? and using for the velocity of sound 10% cm sec“! This gives »»)= 10" per second, This corresponds to a iminimum wavelength of the order of one Angstrom, indicating that the confnuum theory may beat faut, especially in the high-frequency region, Associating with each vibrational mode a harmonic ostillator of the Same frequency, one finds from (2-31) and Planck's formula (214) for the vibrational energy of the crystal, B= [2 20) sea dr = 9N (234) here 2 = hvikT and xq = hvy/&T. Here, asin the Einstein theory, it is convenient to introduce a characteristic temperature; thus one defizes the Debye temperature as 9, = brik @35) Sec, 2-6] SPECIFIC HEAT AND LATTICE VIBRATIONS 4a ‘The upper limit of integration is then equal to xq == 0,7. It is observed that for high temperatures (7 5>0,), x is small compared with unity for the whole range of integration. In that case, the denominator of the integrand in (2-34) may be replaced in first approximation by v. This yields for the specific heat, Cy Peer Nk for T 30, 4 result identical with the classical theory. In the case of very low temperatures, such that TO,» the upper limit of integration in (2-34) may be replaced by infinity. Now,* atx beeai so that E= Wt NkT(TIO,% for T 1, onecan employ the boundary condition (2-52) also in the case of a linear chain. Imagine, for example, an infinite one-dimensional lattice divided into macroscopic sections of length L = Na. From the physical point of view, each section should have the same properties as a circular chain of length L, because as long as N> 1, each atom would “see” the same atomic configuration, the interaction between the atoms being confined to very small distances. 2-10. The equivalence of a vibrational mode and a harmonic oscillator In Sec. 2.2 it was pointed out that the central problem of the specific heat of solids is the determination of the possible modes of vibration of the lattice under consideration. Once the answer to this question has been obtained, the vibrational energy of the solid is calculated on the assumption that the energy corresponding to a particular mode is the same as that of a harmonic oscillator of the same frequency. In the present section we shall show for the simple one-dimensional lattice of identical atoms that this identification is justified, For a general treatment of the three-dimensional case we refer to the literature.!® Ie is well known that the energy of a harmonic oscillator of mass M and angular frequency w may be written E=P/2M + Moye Where y is the deflection and p = M dyldr is the momentum. In terms of alone, we may write E= iMdyldt) + Motyt/2 @s7) We shall now show that the energy associated with a vibrational mode can indeed be written in the form (2-57). Let us consider a mode corresponding to a standing wave sin gna cos wt. The kinetic energy of the Particles in the lattice resulting from this vibrational mode is equal to Fm = im () = mot sintor Zinta (258) 6M, Born and M. Goppert-Mayer, Handbuch der Physik Vol. 24/2; see also F. Scite, Modern Theory of Solids, McGraw-Hill, New York, 1940, p. 125,32 SPECIFIC HEAT AND LATTICE VIBRATIONS (Chap. 2 ‘where mis the mass per atom and the summation extends over all particles in the chain, The potential energy of the system due to the vibrational mode q is a function of all coordinates x,; let it be denoted by Ws Xp ve » Sq 0). The force exerted on patticle mis then, in accordance with (2-42), WV dy (Cima + Xe — 2e9) (2-59) from this one may arrive at the following expression for the potential energy Ha Sapo Xue) = APE — Kya — Fees) 2-60) (Note that cach of the mixed terms appears twice in the summation, providing agreement between the last two equations; this may readily be verified by writing out the sum explicitly.) Substituting the standing wave solution into (2-60), one obtains after some manipulation, ¥ = Y sin? (gad) cos" ot & sin? gna on Making use of the relation between w and g as given by (2-44), one may write V = dmo? cost wt & sin® gna (2-62) The total vibrational energy resulting from mode q is obtained by adding. (2-58) and (2-62), leading to E= tmtS with S— x sintgna 2-63) Note that this expression is independent of time. Suppose now we identify (2-62) with the potential energy of a harmonic oscillator, ie, with the last term in (2-57). This requires evidently » (mS{4)* cos cot 264) If the vibrational mode were indeed equivalent with a harmonic oscillator, the kinetic energy should be, according to (2-57) and (2-64), Eun = IM (dylde)? = yma? sin* wt ‘This expression is identical with (2-58), which proves the sought equivalence. The average energy associated with ‘a particular mode of vibration of angular frequency a is thus given by Planck’s formula (2:14), ie, hoy =a (265) Sec. 2-10] SPECIFIC HEAT AND LATTICE VIBRATIONS 3 The number of quanta n, associated with the vibrational mode of wave vector g at a temperature T is Cohen, = (eek — )-# (2-66) ‘The quanta are commonly referred to as phonons of frequency a, in analogy with photons in the case of electromagnetic radiation. The concept of a phonon is convenient in the discussion of interaction of electrons with lattice vibrations in the theory of electrical conductivity. A phonon, like @ photon, has particle aspects in the sense that one can. associate with it a certain energy fy, = hw, as well as @ momentum P= hye, where c,is the velocity of propagation of the vibrational mode. ‘Thus the “collision” between a phonon and an electron may be treated as a collision between two particles for which the conservation laws of ‘energy and momentum hold. 2-11, The specific heat of a one-dimensional lattice of identical atoms From the results obtained in the preceding sections it is a simple ‘matter to derive an expression for the specific heat of a one-dimensional lattice of identical atoms. In the standing-wave representation the number ‘of modes in the wave vector interval dg is, according to (2-56), equal to L dal where L is the length of the chain. The wave vector is confined between O and 7/a. The vibrational energy of the lattice at a temperature T is thus given by L pate fax ep Galery i Ce where the summation over the possible wave vectors defined by (2-50) has been approximated by an integral. Employing the relation between , and q as given by (2-44), one may replace dg by y 2de 2d : do” = Gag c08 Ga Ae OA ee 2L pom fio da Fee eee ‘The specific heat as function of temperature may be obi ferentiating with respect to T. The result is represented by the lower curve in Fig. 2-11 for a critical temperature 0 = hdma/k = 200°K. It is of interest to compare this result with the continuum theory corresponding, to the Debye approximation in one dimension. According to (2-24) thecc sa SPECIFIC HEAT AND LATTICE VIBRATIONS (Chap. > \ seq, 212) SPECIFIC HEAT AND LATTICE VIBRATIONS 35 number of vibrational modes in the range di of a continuous string of \ are the amplitudes corresponding to particles ‘of mass M and m, respec ' Tenath Lis equal to Lolscy, where the velocity of PrortgaTT ty isa \ are Sattituton of the sation into 2-72) yields the following two | she ee plying this model to a equations: | ‘go cat/mole des string of atoms by 2 suitable cut-off (Mot — 2f)4 + 2Bfe0sqa =O | fof the frequency’ spectrum, one obtains for the vibrational energy (mot — 2f)B + 2Afcosqa =0 m8) as Vf aes “This system has nonvanishing solutions fr A and Bonly ifthe determinant E went | Of the coefficients of A and B vanishes, ie. Bp GolkT)— & | 70) [ont yous | es | ! se vpetinitoe inthis ce ® | | ayeosge (ont — 27) | Toe are by theft that the mum \ she ges forthe gure of he equeney the lowing Poss: | tper of modes is equal to the number ne 3 are fl | Lit 1a Asia)" | of particles N in the string, i.e. wtas(heg 1 1V_ ie mI pat the string, ie | (b+) 405 iy a em | (Elna [om aio = | Since « should be positive, each value . | Tae oe eT of oF leads to a single value for eo = TeKi a Neal a | Thus in contrast to the monoatomic ig tate Cone repens the na ; | Jattice, there are now two angular 2 aptm? Mie eat vero fora monostome Note thal 086 T & different from frequencies «,. and «2. corresponding spe heat veing to@s, cure that appearingin 2-69), THe specific \ to a single value of the wave vector q- ly? befers to the Debye theory, represent” heat ‘calculated on the basis of (2-70) | Ina plot of «» versus q (Fig. 2-13) this wy See open ye nt re | in Pele branched the ne | ast mk eos 2H again for a entcal temperature \ leads ding too is called the Taq 0B i ‘i935 Rada = 200°K. | feoustical branch, the one associated a | | swith «is the optical branch, These with Or ches will now be discussed on Fig, 243. The opel 000, | tvo bran retion that > m. For Oe dower | 43.12. The vibrational modes of a diatomic linear lattice considera diatom atic in one dimension as illustrated in Fig 215 | 7 Owe obtain Sure) branches corresponding 10 the distance between nearest neighbors ‘will be denoted by a. The particles | 1 ype 2 diatomic linear lattice \ fare numbered in such a way that Co jy (b+ Hl and for q=0'@™ | are Mian numbers have amass My 9 + *§ myo 4 0 i mM the odd ones m. In analogy with kom Mom | (0-82) we now have the following equations of motion, assuming Flg:212 ‘Atinear chain of equidistant =— | From the form of (2-76) itis observed that here, a in the ‘monatomic case, | the frequency is a periodie function of the wave vesto™ The first zone | rye sts the values of 10 the range between — 7122 and +n/2a as searest neighbor interaction only: maspoints M and m | u " thus in ip, 213. For q = den/za, the two angular frequencies 2° Mig = (tea + Fae — Be) | evidently mi, Sigg + Kane — 28 7 | sa = Jing + Sts Dead en \ cg = Qf and o. = AMIE for a= afta (278) | ye uy to sive these equations by runing waves of he 8 y ge oe | sae capi carves fran vere gt nse a F212 ry Ae and nei = BE 9 | The come are in she wee he egune gp Bee ean etor of a particular mode of vibration; A and B | The Wrbetines, The existence of a “forbidden” frequency (er energy) |58 SPECIFIC HEAT AND LATTICE VIBRATIONS (Chap. 2 An investigation of NaCI has been made by Kellermann,!* using ionic and repulsive forces between the particles, Figure 2-16a gives the vibr tional spectrum of NaCl obtained by Kellerman, arid the difference with continuum spectrum as used in the Debye theory is obvious. The Debye temperature 6p as function of T calculated by Kellermann is given by the circles in Fig. 2-16b. It is observed that the theory is in remarkably good agreement with the curve obtained experimentally. REFERENCES M. Born and M. Goppert-Mayer, Handbuch der Physik, 24 (2) (1933). M, Born and K. Huang, Dynamical Theory of Crystal Lattices, Oxford, New York, 1954, M. Blackman, Reports on Progress in Physics, 8, 11 (1941). L. Brillouin, Wave Propagation in Periodic Structures, Dover, New York, 1953. ‘A, Eucken, Handbuch der Experimental Physik, 8 (1) (1928). P.H. Keesom and N, Pearlman, Encyclopedia of Physics, 14, Springer, Berlin, 1956. J. de Launay, in F. Seitz and D. Turnbull (eds.), Solid State Physics, Vol. 2, Academic Press, New York, 1956. E, Schrédinger, Handbuch der Physik, 10 (1933). F. Seitz, Modern Theory of Solids, McGraw Hill, New York, 1940, Chap. 3. PROBLEMS 2-1, (a) Give a derivation of expression (2-4) for the difference botween'C, and Cy. (b) Calculate C, — Cy. per mole of sodium at room temperature if at this temperature the compressibility of sodium is 12.3 x 10-#em# dyne-t and the linear coefficient of expansion is 6.22 x 10°; compare the result with C, Also calculate the Griineisen constant for Na. Gy for a monatomic gas. 2-2. The possible energy levels of a rigid rotator according to quantum mechanics are given by E, = (f#/2J)a(n -+ 1) where J is the moment of inertia and =0, 1,2, .... For the molecules H, and Cl, calculate the ‘energy difference between the ground state and the first excited state for rotation about an axis perpendicular to the line joining the nuclei. (Answers. Resp., 14.7 x 10"? and 0.06 x 10-Fev.) Also estimate the value of E, — Ey for rotation about the line joining the nuclei and show 1 E,W, Kellerman, Phil Tran. A238, 513(1940); Proc, Roy. Soc. London), A178, 17 98h, Chap. 2] SPECIFIC HEAT AND LATTICE VIBRATIONS 59 that this rotation does not in general contribute to the rotational specific heat. At which temperatures for Hy and Cl, do quantum effects enter in the rotational specific heat? If it is given that the number of possible states corresponding to an energy level E, for a rotator is equal to 2n(n + 1), show on the basis of statistical mechanics that the rotational specific heat for a molecule such as Cl, at room temperature is R cal per mole. (Hint: According to statistical mechanies the average energy at T E EZ n exp (—EqlkT IVE Zy exp (—E,/kT)} where Z, is the number of possible states associated with E,. For the problem” under consideration one can replace the summations by integrals.) 2-3. Discuss in some detail the specific heat of a diatomic molecule (including translation, rotation, and vibration). What is the value of CIC in various temperature regions? 2-4, Consider an array of N similar atoms, the separation between nearest neighbors being a. Discuss the specific heat of the system on the basis of the Debye approximation and show that at low temperatures the specific heat is proportional to T. 2-5. Discuss the specific heat of a two-dimensional square lattice with a nearest neighbor separation a on the basis of the Debye approximation. Show that at low temperatures the specific heat is proportional oP, 2-6. Consider a cavity filled with black-body radiation in equi with a temperature bath T. As is well known, the energy of fa unit volume w is a function only of T; also, the radi p=w]3. In a p-V diagram, carry out a Carnot cycle with this “gas” first expand isothermally from V, to V», then expand adiabatically such that the temperature drops slightly from T — AT; finally, return to the starting point by isothermal and adiabatic compression. By making use of a well-known theorem about the efficiency of transforming heat into work, show that the energy density w is proportional to T*. Explain why the specific heat of the radiation gas is always proportional to T°, whereas for a solidin the Debye approximation this is true only at low temperatures. 2-7. Discuss in some detail the analogy between the mechanical properties of an array of equidistant similar atoms and a low-pass electric filter. (See, for example, Brillouin, op. ci.) 28, Discuss the specific heat of a solid on the’ basis of the cut-off procedure suggested by Born (Sec. 2-7) and show that one arrives at an expression of the type (2-41).Introduction F. K. Richimyer Late Professor of Physics Cornell Unicersity E. H. Kennard Late Professor of Physics Cornell University Fohn N. Cooper Professor of Physics Naval Postgraduate School 0 Modern Physies SIXTH EDITION TATA McGRAW-HILL PUBLISHING COMPANY LTD. Now Delhichapter five The Origin of the Quantum Theory Relativity ‘and quantum thetry represent the' two great conceptual remodelings of classical physics which were required to bring physics inlo ‘the modern era... The quantum theory, first proposed by Planck in’ 1900, ‘arose oul of the inability of classical physies lo explain the experimentally observed distribution of energy in the spectram of ¢:blackbody... By assuming ‘hat electromagnetic radiation was emilled by harmonic oscillator, the energy of which could change only by discrete energy. jumps of amount hy, Planck ‘war able to-bring theory and-experiment:inlo accord. In the following 25 ‘Years striking confirmations of the quantum hypathesis came from’ several ‘areas of physics. 5.1 Thermal Radiation | Ahot body e jiant the quantity and quality of which depend | on the temperature. Thus, the rate at which an incandescent filament aFS Sosa qe wy saieanh 51 dortorpesomeopus St th esas ban JorsiogaRtaRY: ¥ oF pouyEOD 4 Kpog UET=Kq _f Petra nonmpes oms- som “osmyuradainy sis in 40 sansopo we tt _ ABsoud YS Risa Sey 8! 60 37Ot Gen Bf SuOye Soups qeIN Aq poritaio MOnePss Aq:soejink WIG # Jo JuoNy ut yank poonpoid Me9p Alsou ree ou, ad ey i SU SRL i ea sere te a MEF uONUIBES TEAS aE Ty ATMOS AU SoS SPE HPN oper woeR seo 0 sjjea.ouy Aq: paris, ABseum PR susie ueHodiay sex, omenjaus ue wipjey cot eips Eine g i ke Ai oon, J bOgUPEAL i Tameka Stine! soiig fui voy, ox03 ytd Oy py' Sui pag Sus Youle ‘8 Samiedo Hoa (BS OM. “alters soBconS 17. wiory aiotd Yerepea BuqiOsqE hq F 2m9 27 win Joiioy ouiosoq ua pinom aioqsosqo oetq jo at. ani f 80 2 ue. tees Sisey oti’ “eoqionqe sopiNs ony sonponur prnos Saou Saipkesy yelp wei, s>19018 arom som IuyAEN woneIpe fo ese atp jt oidaiexs-aoyt “Sotareaxpomsorn Jo mel puodoe ayy ayoporn on 0. Seen Bong 04 210M syoouieres om i ‘katy Jo jooug— toe SPL AHROK OA $b paren BOS Tenor monimprd Aura 7s Ta? 7 ‘uoiempes Ui ejuamdopoxop jeowen Weeden kuursoztiow = ‘VS Hpaaldy “cep ‘oustihowdes a UoneIper or Ife m-atp Oo : se emcee niga er ori Shoda ejdan kiquqsnaas stor evsosed 1 opus uohempes Pes 28) Senso0q yealoiuy Jeeds jos q308 sip, Jo. aetieojous jousayy on ay sim om r EDT pouciny jo ove3 yeqSOO OTSA Sey ATR Ses mO eT “porntie #6 0298 yows poquoeqe 8 Cove iusrper tents Fe Senkoog Hsin “Sinbioduig uf Sine) ion Suis soypNoU axe Kowp Twtaanh Stngsoeqo pity Sunuimo oxy saipoq saith wieduio Ie 19 ‘mou £4 poquoaqy fipstiied 420s} 18 s1so1poq uodn une; ton eipeyy BE Ca ‘onbedo 94 03 snows 36 uous deusp ore Yorn sive(qo.kq AlUO dax%op ajqeioardde ue jim. ATueurpio s woHeIpes Hong TRIG d0j/Sba6: guises soneonb:feroAdg same ee ES RC MDM A TINT TTT. searqan, sotno0q amaiy partis Bue Simosnducy oy orvotou im Aqpides moeoion wore tia « reg wapoe 0} wrRepemy gr120 Introduction lo Modern Physice 5.2, where the data of ‘curve serves to represent blackbody radiation at AeT = const = 2.998 x 10-1 m-°K. forall tecnperatures—a special case of Wiens di wus, by reasoning ba E we dy teasoning based_on thermodyoami nergy arbitrary unit Werclength Fig 2 Diwibtion of emgy ia ont wavelength atrval 2k Ex ke incoay actfam "at out temperatures fon Kevin ie. 52, ig: 5.1, obtained in 1890 shin, are presented in a dideent form. Asa all temperatures follows that if i. is the wavelength at which eis maxim It 63) nt lass, ‘the problem of The Origin ofthe Quantum Theory 121 alt Fig. 5.2 Experimental verification of the Wien displgcement law for blackbody radiation. deserve mention. Win derived a foimula based on special assumptions regarding emission and absorption of radiation; he found that the energy density U; dh for wavelengths between» and d+ di hacia = era a where e is the speed of light in free space. By proper choice of the ‘undetermined constants C; and C; Wien's radiation law can be made to fit the data of Fig. 5.2 except at higher values of XT, where it predicts too low a value, In 1900 Rayleigh applied the principle of equipartition of energy to blaékbody radiation. This led, with a contribution from Jeans, to a formola for e. which fitted the high-\7 part of the ¢ curve but rose to infinity with decreasing XT. ‘The reasoning employed by Rayleigh and Jeans, discussed below, has applications in modern physics, 5.3 Degressof Erecdominan Enclosure ‘A. typical vibrating system, such as a violin string or organ pipe, is capable of many modes of vibration. i icours pus ‘poydde .oq ysnur y/(k 80042) = 7/3*u worrpuc eL, “poa[OAUT of0 S2KEA'JO Stas 119 YR 20j “aqno euo!uOUNp-L03mH 18 0} popuarxe aq weo SoxvA Hurpueys Suonzysuoo Jo POT SILL, 2X © (w curd 4 2 900) 2Y. = (g 900 + 0,800) fut tu BC = (sus + © 900) = (#800 + » 800) + a co es neg oa ‘ppe pu fyoeo orenbs “40 pus *u 305 oAjos 9M JE /(G 809 22) = 7/e KE X/(@ 900 #2) = 7/24 onc om “CN /29g) UB = (7/2=U) US 200s pu 9600 1900 ay pire » s00/¢ = *y ay) sep st amNY amp MOS -¢ miu a0xtn, pue wonooatp 2 om TOL pareposse sTPSuajeaem ste Mos oze o10m 9809 Te|NONIOT yp 20y fspce 2 97 YEA 92g = 2 ofBuY soyEMT YOR ‘suns uo “abedoud otp Jo ou0 syiosaudos Mosse Savor omy ge'S "SE BT “Sp UopATK tee Jo soquinu fesdoquy uv aq sn uonsoup uoneiedosd om Zuope pi Aue Jo uonoefoad orp “Kypwoyioods ‘sinJupPAea PoAoye ay UO ON me synd. 10 sea. otf ¥e sopou 9g 220m THN WORTPHIOD OUT, Ear ean TS ioy wopenbs op Sf Siyy ‘sxedoquy ove *u pow fu xT Ree us (7) = mioy aff} Jo are toryenbe ean amp Jo suomnos poxorfs amp “7 30.0 = & -oquo # p aueyd o, In reality, the vention density i an isothermal enclosure is ordinarily much less hi radiation derecity dve to thermal agitation ofthe molecule of « said seNOGeK U is about 7.5 X 10-+ J/mt, a8 compared with « x. fon. At temperatures above $6 The Rayleigh scheme SOrrK the comparison would be reversed. [afi i Tice, Te harhigh thas fr oweting, the | i eld So eps bal nob ny! tn ln than(srs) “iso § selene syd poe =e ore soxoruos 200q sndgip wospeay momen eg o10u 134 oy APFOHNTUC PN ASO Se SAN wap ae TT eT, rs) OP Taw VN = NV “(exs) ‘bat 841 Bos ey evmod Ay poysoaadon ssoyeyposo Jo sequin ayy “ung bo yeUINY Jo sms» UL “OTOH, 7 ‘ou ‘Tfeurs Rion STAY I sosdyT oso "ag +» pu » S018s0U9 4WO.0 HD ruts woomtoq #a4 ond op anogsHONy Angas 40) umeip oq sosduyjo yore oma 7 Tyres soumor Alsou ayy ah AEA w fone Wt oBU “(g's Bug) STUNTPTOCT TN SPRTRTOT ETT] “wu05 quoropp v oyu ejnuso ax MOTD OF yuaZTOATOD ‘5 af ‘si0yeqT}80 ayy Jo SoxBraNE am UI ATO payseiayU are Om eoUIS ‘Al = NP J WYN Yous ueysti0o Ayyfeuorsodord wv st Y exo (ers) 4p =P ssv2VN = NP aqoy dp + d pued voomoq ummuowiom pur zp + 2 pur z uoomeq soyvurp1000 him VP sorenfee yo zoqurnu oq) pug om SuOsUOE ooede-areyd Om 01 GL) “be Jo wondenU0 » kq “(O° *98) VONNANEP wae IO STeaxeyy ain 01 Surpioooe Kau oF peynqunstp ox dnow# axque om «co sexsi 9x9 1nq SpppEM soyeA sores: Tenpratpu ue Jo ABs0U0 op Zeumpqunbs peuzaify uy axe e.e7}oe0 Yons Jo xoquin OBE] © LITA es) Asoayy Axequomep £q pure donqea om sey « wonsqta Jo Sousnboyy B44 ars) i ‘| ABzou9 eror SI EN, -ug/ad = <(p/zp)uek 8 KBzo00 opaury sy ‘auersuoy e940} amp Banou9p ¢ 'cr9f sH0} 1 UL uoTIBM oq WED sorEITES OM Jo ABI0U9 TEAUEIOM Ber Aono umpurmd on fo wibH0 AL ORL 1P/zp wt = d umyuomout st pue z aq wameoyderp 1 49] ‘uonoUL oquomrey opdans mi soyeaqia yey w ssEUH JO sOVEIT80 JEaUH| B JepIEUOD AS — apm moms em PF TOT 95 “wont? oq ne Azoomn pwonsHeNS om Jo uN ~luomp joug w uondame ay Supunwpin reveNy On eytoutal [Rossee aut Syeonsespoyoxq7eq) uondansee mot © Sune zyFe ko o8 Sau ur popaocons 9 “ejmatzoy aou sy on pray pinom Kaoory oy yey oF ‘supveqtoH0 Jo 28 » Some ABLou9 30 uoNgLnsT 2if Jo Ksoom Fone our Appom oy anos oy uot, "STOR Tey waz} ee once eae we pmo ER PETTY PSA HOMY.os9 yo EHO} 8.UOIAA IWIN AP HUGE PaaoN uMeGlINY. UE eu = ‘toeaung pe oem kqwnstoode fpoqraeryas aioaeoes cee 0061 8t “('s) ‘Ba “Aussuop uonerpes otf 30) FIMEIOS s,ua4NK O1 IRAE 38 ~ por sex oq dondamave yuor0g%p © Jo seq otn U0, Foo Top KR erro ro TE TOU Pip TIASAMOY FOUN -wa9} sue -xBprkey ox 1 poy seu 24 43:00 Jo wonnrednbs oy ely FostoP ‘KBroU0 ofes0aw sx 203 Jy ONyeA [eorsseIO axp Pousnese’ uoMH et J] “3mm -siaduiny waag# © ye zoyjoeo ue yo £Bz0u9 oferonw om Jo waafqoud om OF Peonpor sn sea unzoods poqyontg ain Jo woqqosd aug, ouonony 819 30 siopeToe0 ax Jo ABrouo afei0K0 ain pow « Kouonbay Ku yo tot expel Jo Aypo9p stp womseq ones anTUyEp © oq prnom axon “umTTS yo aes on up “FAT Toes Ty fo Te Be UT Fae A a 240 0} your pot Azoomn anousemorioje Jo sweat £q UoResTToAuy “ssovetneo osm Aq posto are se om q wonepes jo won dioege poe tows on get poe wove epror] ah on zeps Aqenuess “erouonbouy [Te Jo eIOAEIBOSO urVTLOD sansopus jemsampost wo Jo sfex om yer ougomn Kew one SOTATSS 3 Ta ETS PUG WON aONGT aay FUIOAG ERAT TUT PUTT OT AT AOD Sem pos “14 J0 Kioom anoudeMORDep ayy Oy wonwyuES Teuy On ©% pouroes per soxem onouFeMONDd| uo ZUDH jo swuounIedeO aH], SID wwomuodxs qua jomoosesyaa09 uy en} © pune} md -wo9 tmoysoud ip souzeA ye ann uorrenouny forped w Suronporat ke "LL ecg wsrpeyy 0} vrmpomu } .< \128 Inlvoduction to Modern Physics Fig, 5.5 The momentum- 304 (gts) “bal WoUL Buy yeNp Jo Azepunog souut oy) oy Jadosd ABiou0 om Zaye ekg werpeHY 9) vorEmpenT Os132 Introduction to Modern Physics ‘The difficulties with the theory of electromagnetic waves weie overcome _ when the theory of wave mechanics was applied to the electromagnetic field itself. It should be emphasized that Planck’s revolutionary assumptions were not based upon an extension of the ordinary lines of reasoning of classical physics. Quite the contrary; they represented an empirical ‘modification of classical ideas made in order to bring the theoretical deductions into harmony with experiment. Had the magnitude of the quantum of energy turned out to be not hy but something independent of the frequency, the new theory might well have taken the form of @ simple atomicity of energy, similar to the atomicity of electricity repre sented by the electronic charge. Such is not the ease, however. Rather, it is the new universal constant h that represents the essentially new clement introduced into physics by the quantum theory. We shall find 4k playing an important part in a wide variety of atomie phenomena. AW 5.8 Planck's Radiation Law ‘Planck derived his new radiation formula by considering the inter bet m and electric. oscil- 1 ich he imagin ist_in the walls of the enclosure, A more direct and equally satisfactory procedure is to combine Planck's new expression for the mean energy of an oscillator with the analysis of the electromagnetic field by the method of Rayleigh and Jeans, in which the Various modes of oscillation of the field inside an enclosure are treated as if they were oscillators. In Sec. 5.3, Eq. (6.9), we found that there would be 8 dh/A* such modes of oscillation or degrees of freedom per unit volume in the wavelength range \ to \ + dd. If we multiply this number by # as given by Eq. (5.18), we obtain a x hy Uy dd = br Let us substitute here » = ¢/d, ¢ being the speed of light in vacuum. ‘Thus we obtain, as Planck's new radiation law, [ees (Strictly speaking, we should have used for @ the value given by Eq. 6.19), which agrees with the value obtained from wave mechanics. ‘The effect of this change would be to add in Us a term independent of tempera- (6.20) | | | ‘The Origin of the Quantum Theory 188 ture. Since only changes in U are perceptible, this term would be with- on ek to Wins formula near one end ofthe trum and to the RaylighJesns formule neat sna Fig. 5.6 is shown a comparison between the several spectral-en tuibution formulas and the experimental data. ‘The circles show observa- - tions by Coblentz on the energy distribution in the spectrum of a black~ | body ot 1600°K. The reson thatthe cure fr Planck Coomola dome fr Hoyt aut Seow contrat Se mnek's formula has the forin that we found to be roquired by the pre prise ee es aon Fig. 5.6 Comparison ofthe three radiation Jaws with experiment at 1600°K.‘own ka porvooop st pucoos lod onjns omy Jo vare aun oF PHUOAIPP sryeq) rmuou orp Jo wonsaTE om ul wTuowoW oxy enn, 2 200 & tomsodoud us spuspiom fwatiou Jo ove oxf, wy wet sso] sr aOepS omy oy yeutou umquouion amp Jo qwouorTuioo amp ‘esomqNIN “Yr AORpsNE fm) uo g $00/t epmyutout Jo voiv so8ze] v Aq poaspoos st (UWS. Fd weg) eke oth 01 soqnoqpuadsad uacexp voxe yin w soes0u9 emp &B:0u0 Sheet, “payed se wap sTaoKTIpeL am yey xu ‘eeoddng 7 -o on pono aunssoid - ‘ouojoromp ‘soonstiodxo pw umywowtom ery seaioas yf ‘SokyINe oy Poquone sue sone ot JI 200}sns off preAoy Sywunou peioanp Faq venta v ae tumuoWow otf ‘puooes zed wimuotION jo mpun m ‘A909 JO ro tpt Suoqe ‘spoyans op Jo waze kum youo 07 dn Sug soxwa SN, £ “Guz TO HTIN GH ose n setae yaroaue oq ur Riss asus uwsar omy a pe TuOMT, “Apoq wJO sovjans otf Uo A7feuou Sie} NMEA w UE YorreTpes JO maT w Bsoddng “TopDIpoy Sono oF ona My AUT pew onary TVS Rs0ouL uormpEy 199185019 - VS = PUOday. — sy pow ‘94g “2 0 seu ut possosdxa ueysu0s wojag dyn maT (1's) “bat porDAo.er oxee om POU F/T — AT Irene yun 20d poretpes sowod oxy smug, “Aayinop £8040 oyedoxddo sip unt 9/2 asa 9 Tearanay « 20 \ UeAgs © Uf 30 ‘TOD “ware yun sod poreypes samod og “(s"vs) pue (HVS) Sbat “T'VS “99S UL UmOUS ST EY “oma poquos -ayd v worn sy 2 18 porerper KB:on9 ona Jo vonD omy SupMOKR 9AIN2 fe tpem Supe 1's “Sta ur WAONS st APOqADEIT ¥ Jo Yonge Testoods pomyeuon auf, "St /ranyea om sey pesBonty mp pu LAK /yP = = 930M La? fe. wy apa ~) sae? 7 HTT ANTP RIT THOT TT wep 4 pee woutog o8cs om 07) Aap KBz000 omy OF SID UT "XH TEL ovonbaxp Jo sa wy sey Yous] aM STEN oF S|TENP HOUO SAL ser | ano, wmpcond nn fo 840 PH (38 1) come 2 wiv popes Kis yo woe pt Aone wen ead ave [ose ware- a e/est St n-0t X 95299 = 1 2n]84 1p PIPL sporpaur soyI0q pure ‘soA9MoY ‘AO APPANEIOS srovep reason jo Coenen e013 k mn se-OT X S59 = Youeccap fou ox Pourerqo yurd (19) Jaded ous ya -umouy onsale 1 jo somea any wonyn or ejnmy xp aia 7 Jo toner poasaqo en [ Aewoduoo } EL aWODE|AeNP WOH 2 ae waa} wuvusntog-tneag om oq "0p sn wt NS at ong (ozs) = aon? 61) ome = Ns] ORL = 0 uweyum eesyd Yerpoyy 0} won-mpomyy HEFig. 5A ‘obliqueness of incidence in the ratio cos* 8, and the resulting pressure, if the radiation is entirely absorbed, is P= weost 6 ‘The same expression holds for the pressure caused by the emission of @ beam at an angle 8 or for the additional pressure caused by the occurrence of a reflected beam. If an incident beam is specularly reflected from a surface at the angle of incidence 8, the total pressure on the surface is 210 cost 8. Finally, let radiation be streaming toward a surface and also away from it with equal intensities in all directions, as in an isothermal enclosure, Such a distribution of radiation is equivalent to a large number of, beams of plane waves, all of equal intensity, with their directions of propagation distributed equally in direction. Let there be /V beams in all, and let the energy density due to any one of them be w. ‘Then the total energy density U just in front of the surface and the pressure p on itare, respectively, U=Nw Ala) p = Zw cost 0 = wP cost 0 GA.1B) ‘To find this latter sum, imagine lines drawn outward from a point con the surface to represent the various directions of the beams, whether moving toward the surface or away from it, and then about 0 as center draw a hemispherical surface of unit radius with its base on the surface (Gee Fig. 58.2, where only two of the lines are shown). Froia the hemisphere cut out a ring-shaped clement of area QS by means of two cones of semiangle @ and @-+ dé, drawn from O as apex and with the normal ‘The edge of {sa circle of perimeter 2 sin &, and its width is do; hence its area is 2 sin @ d®, whereas the area of the whole ‘Now the lines of approach of the IV beams of radi here in points equally dis- | The Origin ofthe Quantum Theory 187 tributed over its surface. Hence, if we let dN denote the number of ‘these lines that pass through the ring-shaped clement, dV will be to Nin the ratio of the area of the ring to the area of the hemisphere, so that aN _ 2n sin w te = sin 0.40 The value of cos* @ is the same for all the dIV beams. Hence their contribution to E cos* @ is cost 6 dN or, from the last equation, NV cost 8 sin 6d, ‘Thus Joos « (The limit is +/2 because directions all around the normal OP are includ J cost oan = N [7" cost osin @d0= 3N (6A.2) | inthering.) For the pressure we thus obtain, from Eq. ($A.1),p = 30N, GAs) “Thus the pressure on the walls of an isothermal end third of the radiant-en i interior poi It will be useful to calculate also the total energy brought up to the surface. Since half of the NV waves are moving toward the surface and each wave delivers eneray ew 008 6, the total energy brought up to unit ‘area per second is HN 5" cx c08 0 sin 0 do = tow = 4c by GA.1e). In an isothermal enclosure an equal amount of energy is, carried away from the surface. It can be shown in the same way that in an isothermal enclosure in which the energy density is U, energy +4€U per second passes in each direction across unit area of any imaginary ‘surface drawn inside the enclosure. Fig. 52ae aha oun 4 vst roma: o 3] ROSSETTRSTRGT Ge Kyou seh eee ong STR, 7 Pi xt syrepuno "go ommend ion om doy Si wt sass yoosexa waxy pos 9 one OY elo yoy Jo yunoure uy 3094 Jo uondaowgo we q pormodaoooe sy uowuEd ousoqiont tay of 6X9 wre) daeutpuo wo ut sv ant pan om ug oDeds bun on ponds wwoy on aoqearnbo st 7 KBsoc9 ou, “ag om Tewmoraog ax Aq (Gry Ia) woosFep bony wo poyabsasdas AOR OATES, (vs) (a — "19 F =" ooh (ayia ¢ = puzsnge Goranson. C0 1019 0 pane ra, uem ac my gp Fa esiont ey Sung] 2 oF Suyrasoo;oamnpoA pyun Ange “pooadn oxom oy wore om en09, es Sree al por MOST SAT FD Seis PAT Ce ys May OOS TON OTT TRS 1 Ieyaiae jo ojak> Bixa0no} om sepipdbs kou Lea 07h “0.04 %g way sp tg 04 ( Wlo3} ayer auIUS amp ae Amesed oq TM NORMTPEL amp pr 18 ig a ew pao ou wONEpEL JN AiO ‘ures ata st sz9ip. yum tg wo3y g ys nostp "y SayeIND VOR PIPE ‘8d wonerpes di 30 a Ko rows am sop weesienp ed oat PS FE [eur -ouyfuo uoneipes ReweuryOR eS FAs ° @ o @) @ Tg zopuxo ox wou, “(¢-VS ta) Ly sampesodtuey ve poureruTEHE plas ig aansojua yeuLiayyost payenaeaa ue sn Suuado ue ausoddo put pazoaooun.g aujuado ayn yam pooeyd.oq.sepuiso simp 4271_soA09 Bur “yooyor Apooyrad v Xq [14 18 pos9A09 aq ubo yoy DeEq aIp UO Huw “avat 20 nie eur Yo yo Boon eg v pur ‘vonauy nome Suro pus. yoy 01 snoraodan say] UOISId B OK] OF SOLA Zour sige sea. zepurio NT Te ar SOMBIE AUTRION ap JO aTAT vi POW WONUPET STONE SURI UE OT SPSATTSUIEY AUT TO SOT OF PRHTTAT a o80d und sip. io oumesoduin ipia doniepes kpoqyoeyq jo AaysuoqU [20% Sup Jo UoNWUPA amp Joy pj YeoRBIONTT rT WIE OG- UNIS UL EVE sv) WE= 2 vous, -(x?'7 Sf = 2180 68) v7 £4 porwes condos Agiquy oq mYSCERANA pp abe oures oy PEA ohteop oN o Asana ype 0 Kysnop ac at pu" (xp °f = af ve 08) wp Kg pinowop og VP sfihiojsand Jo odode w shim KpSaoMG Sh ely pee A8iom amp we] “Appyurndes psuajaaws yous sj SpIsy uOREpA sures OTL wv) uonenbe om ‘hq ormesodwer omes ox Suavy ounsopous yeuomost we'mt /) kyyeuap K2zouo amp OF povoyss &1 'y souEN IH ywepeL MY S| YOR ‘soups POLI, ue yo woxe yim £qpuosos sod poyyime A3soue ayy “otouoquany ntyg wnpopy 0} wormpomsy eer140 Inlroductin lo Modern Physics sure_at certain new temperature Ts. If the expansion during this second proces was tery small Seoaay TORT the change is toa by dT and the corresponding change in energy density Us — Uy Since P= 4U, we have then canes ap = yw An ion to pass through 0 into™B. The compression is supposed to take place so slowly, however, that the radiation density remains constant at a value only infinitesimally in excess of i this second isothermal process, radiant energy in amount Hs len ‘4 Theniston having redched Ble point Pa The ‘The net external work done during this eyle i represented by the area PPPs Fig 9A It wo ame the change of prewar to ave been very small, this area equals (or —n) dp. Calling the net external work dW, we have, therefore, ° ! dW = (0: — 0) dp = 402 — 0) dU by (5A.7). Hence, by the usual rule for a Carnot eycle, aw MT ar Boh Th and, using the value found for dW and also Eq. (5A.6), we have au _ 4dr Guat Thus, dropping the subscript, Integrated, this equation gives log U = 4 log T + const or =r Gas) where a is a constant, not yet known. From Eq. (5A.4) we have thea The Origin of the Quanlam Theory 141 also, for the emissive power or radiant emittance R=er]" (GA9a) om tea (GA.9b) "Thus both the energy density of the radiation within an isothermal enclosure and the lolal emisss powerof the absolute temperature T._ This the Stefan-Boltzmann law. 54.3 Reflection from a Moving Mirror In the preceding discussion no attention was paid to'the spectral distribution of the radiation. The question presents itself, however, whether ‘the same law can be applied also to the separate wavelengths. "In order ‘to investigate this question, we needto know ‘what happens’ to the spectral distribution ‘of a beam of radiation when it is rellected Troi a ‘mirror that ia in motion, such as the piston in the ideal apparatus that ‘was described in the last soction. ‘Let us consider first the effect of such motion upon e monochromatic ‘beam, For this purpose we employ Huygens’ principle. In Fig. 5A.5a, ‘MM represents @ mirror moving with a component of velocity V perpen dicular to its plane. AB and DE represent parts of two incident waves which are one wavelength, or a distance 2, apart and are falling on the. oe wey ap 4 apa ~wav= =(avidp— =P vat soup “V1 81 Yon Jo yNUTe Two) oMp ‘KSs0N9 posoyono 947) Jo ssuodxe otf 18 aUOp Sf SOK sy, “anseord woeIpEL o7 onp 29104, amp Aq. 44 wo, 200p.91 1p VIE = ap d/ 20K Yp coum ® proscno weaowr dofaid om, wows ‘quowou Xue ¥6 Xsuap SuwURANO A @+e) us Pits Cerys) Su owwo ,y wosy poddoxp sejrorpuodiod v st 0,y Jf ‘2000H] -m poods om 28 ,W 1 1 WW O4} poaour ios:yur axp “HS JO 2 poods am 8,¥ On ¥ moyy yuOM F OY ‘oxoMAKAIN. “gd o1FUe om BY» OBOyR ays) i) Wav = QV apts yey ‘oySueEn seis Aq ‘sopsiiny ‘suo aT septa IyBEE aie “ava pv Kor Vooatoq FEY Fu!G Va BOO {VV eqeue 3k, “ad AW sonal Lg pasnsaen g VV kes omy Boop papaah ¥ CAG wonysod nou oxy ws Yoga ‘out amy HF 28 YOETIOD c amp ‘av £4 umows 84 ‘9 oyu ox 16 porooyar wooq Seq 3 eI “FHO|SOT Joye] BYP AAA Tp JO VORIOd ames aL, TAH ‘8 uonod snooiieyirersuY asoys ‘sour oy UO {Te on SurUNBeq rent onea 0 Jo gy-szed omy ‘ge-ys “Sta Uj “oAeK ous om J0 suORIEd OAS ~seoone ow soplsuo “g PUR,,y UPOI}2q LORNA PUOIIS v UTE;qO OY, ous aus x souaya (6 UIST = ‘9 US Ag — eA andy xp moxy snoraqo 1 1] "ede ,X aoumnstp ® “9 wonoayas Jo a[8ue ‘ue 48 zousu om SuIAvo] AOU oze PUR PoIDOYOs 99g OABY YONA BALK umes. om 30 sind, are gq pUC gD ‘9 soUapIoNy Jo o|Fue we ye JoNTUL enekyg weno 0} vorpmpomnE BPE144 Inlrodaction to Modern Physica and, after integration, og 8 = — log It + const (6.150) a 6.158) € denoting « constant. Now it can be shown by thermodynamic reasoning that an expansion of the type considered here cannot destroy the blackbody property of the radiation. For, at a certain instant, suppose that the expansion has reduced the total energy density to Us, and let 7s be the temperature of ‘an enclosure in which the density has this same valué. Suppose that in the cylinder there were more radiation per unit olume’of wavelengths near some value }’ than at the same wavelengths in the enclosure and leas radiation near some other wavelength \”. It would then be possible to cause a little radiation to-pass from the cylinder into a second enclosure at a temperature Tr; slightly above 7; by covering the opening in the base of the cylinder with a plate transmitting wavelengths near x’ but reflecting all others and putting the cylinder into commistication with the second enclosure through this opening. In a similar way, enough radiation near X” could be passed inlo the eylinder from an enclogure at a slightly lower temperature Ty’ to restore the total energy, to Us, ‘Then the radiation could be compressed back to U;, the changes in U and land the amount of the work done being just the reverse of these quantities during the expansion. Finally, putting the cylinder again into communication with the enclosure at 7}, we could allow the spectral distribution to be restored to that proper to a blackbody at, 7; hut without. any net transfer of energy between cylinder afd enclosure, since the total energy density has already been restored to that, corresponding to T's. ‘Thus we should have performed a cyclic, operation, the only effset of which is to transfer heat energy ftom an enclosure at Ty’ to one at a higher temperature 77’. But this is inconsistent with the second Jaw of thermodynamics. ‘Thus black radiation must remain black during any slow.adiabatic expansion, or compression. Its density and temperature, however, decrease. In the case under discussion, the last equation, in combination with Eq. (5A.8), gives 1 Tey ps GAB), To determine the effect on the spectral distribution, we must find the average rate at which wavelengths are increased. Suppose, first, that the, walls of the cylinder and the piston reflect specularly. . Then ‘he Origin ofthe Quinlan Theory — 146 LN Fig. 546 any ray preserves its angle of inc spite of repeated reflections (sce Fig. component of velocit strikes the pis ing tq-(SA.14), of the piston, tation to the axis of the eylinder 8 in 5A.6).and has, therefore, a constant 'Y © e08 0 perpendicular to the piston 0 ‘The va (© cos @)/21 times a second, and its wavelength, i accord increases each time by (2VA cos #)/e, V being Uni Thus its wavelength increaps ata rae, We sek a _ 6008 82VA cos 6 _ Vi a BRE = Teast Uerays of wavelength } are oqually distributed in direction, the average Yalue of cos forall of thom is 4 according tae GCS average valu of dh/d for these rays” MF GA?) hones the Vy nat a7 aad since V = dla . To simplify the calculation, i nt ion, suppose now that there is a dia folleting spat onthe wall Then al the raya ake ore in the various dretions, and all rays of wavelets) will undergo the average change of wavelongth just calculated. (this conclusion fon he confirmed by a more completa analysis for which we have no spoce) The last equation cam be intogeated thus: —_ DL tdl td (A.170) og 4 = log t+ const : (6A.178) (SA.1T¢) by (6A.16), ‘Thus we have the conclusion pEnea tee the conclusion that each spentral component nge in ways x thy“Me000E pe ‘0007 “000T JO sarmesoduny 403 (Yy 9002 0% OOS) HoH#ox oyqysrA om Ur st Yor KPoqyouI 8 £q popes KSxou9 amp Jo vonaeIy am L's “ALY Woy SOUT, “TT “ep over mONy x pu y Put ws6y ot OL X 67 = BSSF = oy Hout ot x 62 = BT = y HR 5 Lo e/E 1-01 X 9°2) = 2A08 YoY ‘uneqUY pure euaqny jo minis yruoutedss sun kq y tusrea0o MueMAIOR 2uN Jo onqea oy Pu y Jo anyeA sy porndioo ATeUYFHO YOUNG “OT shosT/aete uy Ges) Pg yo oso ‘kg 4 pur ‘9 “y Jo suiiey ut » quEjsu09 uRjeng oy Jo anfea oth PULL “6 WLER'S = #304 81200 49/44 = = Jo sme) uy (2°g “Sg) oamo “ox Jo unuReM om TOM MOUS “e -uonsonb wy qumsuy am ae osoyds om Kq porerpes somod jer amp pus sumendui om put “easno Kpoqypeyg otf ay Hoy Org) a ‘nol swwap-ysepiey op oF pur Ly Kay wR “(s)he BRT kaso, wmpuomd 2g fo bug 2s ‘san s,uotyy 01 soonpas (02's) “bg “no wonvyper s yoUB|g WIA MOUS “z -y 0067 =" {erp 1095 orp twoay uns ayp Jo amyuradisay sowjins amp aveusAL (2) “201/A\ OOPL 3! (8K 8,uns 04) 7 foULOU donpINe ‘8 uo vase un sod somod) qUPIsUOD sejos UR UY OHH ay IOI} PIED oN mosy mH OT X $1 St GOYA WOT X 2 SrIpes Jo Apoqyoeiq ¥ Xyjenuassa ar Suyunsse “une om yo sumyeioduey eoopns on owwunes (0) sw9190-4e. qo 0q S102 wrod yo 10 pea 9auno afdusv ‘eomoaqe we, ASurese portold stsoner asamp J0 290 31 says, TONGAN TOT snyen Ta om FL LT SL MUTT PHT LT _ ‘8 vein Burkes Ky passasdea oq os woo ay AOTTURMONP TTT se Wao % toNNpoLOD SIU, “AIMEE amnjosye ayn JO anod wh amp oF fouonrodoud aq 04 pinoy aq | Sane NN Lay“ OF eLOR:Od ond fyessaaut syyBu9peAem Te Inq MAOPENEM oUNeS ayn 1 20U poseazoD are soimozoduiny quozogip ows 3e "2 20°%7 30 NWA 9H} “SOK (SVS) ba, a ayy st 9 O24 ka. 177 Jo sua uy UaAI ‘Kpoqyoe|q # Jo s9Mod aN rvs) vn fa er SNP MPT ROT a HIG TT ‘sus sopurpdo a4 uf quosoud "yp ur worvelpes om yuo YEA porzeys 2Av4 ysTUE 9m *(ST'¥s) “ba on dn poj yey quaunasw-omp UL 40y £91 KBsoUD [BIO ax se oxyex auius atp ur pue “paseaioop aun aiies atp ye 81 yp ut ABIOUD OWL, - LN (rvs) bes fou, ; tN (arys) nap EN TR ‘aroun yp 4X pe 802 ‘poyuas asnf ynsox oy 0} Simproane‘oBues sp Jo stam a4y SoS ty, camecadem on ss0a01 tolyn vowwedre oneipe uy AT STTEEETY WARTS We TH suuajon YN sed Typ Ty Kons SupaerIOD Wp FY SPR ma TT ToONTOT UO TOES Tne Ay STAT HON oT TUTUITHATG WOH OTL VE edyg wnpon oF worenpesTut —9PT: F. K. Richtinyer Late Professor of Physics > Cornell University E. H. Kennard Late Professor of Physics Cornell University John.N. Cooper Professor of Physics Nesal Postgraduate School to Modern Physies : SIXTH EDITION car — X RAYS TATA McGRAW-HILL PUBLISHING COMPANY LTD. New Dalhi RicHT Bp chapter seven | Xerayis importance tothe entire world of research in pure science. Within 3 months afler Rénlgen's fortuitous discovery, z-rays were being pul fo practical use in a hospital in Vienna in conneclin with surgical operations. The use of this new aid to. surgery soon spread rapidly, Since Rinlgen's time, | z-rays have completely revolutionized cerlain phases of medical practice. | However, had Rénigen deliberately set about to discover some means of | tssisting surgeons in reducing fractures, il is almost cerlain that he would never have been working wilh Use esacialed tubes, induction coils, and the Probably no subject in, all science-illustrales beller than z-rays the like, which led lo his famous discovery. In many olher fields of applied science, both biological arid physical, important uses have been found for x-rays, Transcending these uses in applied science are the applications of z-rays to such problems as the alomié and the molecular structure of matter and the mechanism of the interaction of im1; 9889]04 portdate omp yo Apyuapasdopus paqfasyzoo.0q pines yuozn9 wonoop amp snyy, “Aiijeq ¥ wos} quaLMS a[qesnipe we oq s2insqa ‘Sexpoq pogEIsepe aBreypsprskeryx “9 -ydes2ooud ajoqaid jo suxour hq pomdys watuny se ‘som agree uy eae faq, -Ppy anoudeus w dq poreoper 94 tela e samen uaouyun otj—sKex Mou asamp Jo sofsodond omy pngs 0} popes ord so ve wo ‘Asaconp sy Jo eouriodany amy Sate, “sku opoyrea amp £q yonsys 9704 s]fou $50 ou aioga army oSzegpmp aq of auod ramp Ye polwufe aseeomory ou posneo orys Kouase ox yey JToeMY pooutAuos uooe womUoY Ssmuredds ox wosy Kea ar z o|quasosqo em 2ousss0i0NY ONL, ,"2qm s8teTpmp OM PYEAo} pou 2q Joo aM 40 apw porwaN omy JOTOG om Ajrenbo ssosozong pue KpuenTag dn syysq aprueco-ouneyd-amuey qi, Polsex woes saded v,, yey) Uoreasosqo JeUapISOR ay peu 94, “wOOs Pouayrep 9st: 9 rayo 8 Aq Jato Yous wosy powered ‘umesp-oq eo sound Jo soqurey yons uu eo: ayn Jo swe ayy 70 YEnosIp seed “asst: SuoUie’ “yor, usLOAp' aq" souv[d jorfesed queysrpinbo Jo yos v yeysiuo Lue yonorys,s8ipSino poquiod OH “S8eig “TM Aq Posodoud sem Susu? [wyekso w kq vonaeiiyp {Jo seo0ud amp 12 Buyyoo} Jo ken [duis y “uouowoudyd wou oyp JO ‘Apms otp dn yoo, sore gauy Auew yuowodxe jnyssooons szoxp Jo one] pus ‘Suiddwy ‘youpous £q juowsounouue oy Sumosoy APywIpouLNT ney 380g #2 “pey wo} dip Jo Soadse jue odui osu oy JO BwIOs 07 uoLssnosIP ano auyooo ays 94 “AuoH.298 Soyo) Oy UL sfersks0 wy sqno9four 20 ‘swoye Jo suoweduesre om Jo Apmis amp (2) Sma ‘uondiosqe ‘Suyorw.s uo syioupddss ut surwoq snoauodowoy Jo asn ax pus exyoods Jo pms am ‘skusex Uy (1) :dn_pouado sono ye 910m wonedNseAut Jo Spjoy AaU ONL ‘sorx_uyei0 nou ¥ Jo SuxausGoq ox poyseur wounds aL 1PisAs9 Jo Anomwiks you294x0 o1p, Aq poyso3ns sw.‘s9p20 euorsuoup-2osip Joyner ur poduerre axe joys{40 v Jo, swore otf yeiN (z) pu, ‘sqpdusjoaom. atuyap, Jo SABA oNoUeMIONDOAD ore SKeI-x IMP (qT) tquouniodxo amp, 2uLktzopun soyeinysod om) omg} porsoddns aynsox sty, “at 31-0 X £8°H PUR J.Z"T Uoomyoq BuLATeA SiAUDZOARA wieaq KBIEK OTP UL jiatead 3194 910", yey popnppuc ‘pouad quopious ay7, o} yoodsou THU so[fue snoizea ye paruotio opua|q aurz Jo [msK30 8 Jo sqdusSon0yd Jo so1z08 ! wos} fone] pue ‘Surddiuy ‘youpeyy ‘reded peurd0 mp, UL Sq) peiorpoad pey oney se yenf ‘suonoonp. jemeds uureqi99 ut eras9, 9 Aq porousmp uo0q poy urvoq Kes-x quoprout omy yt) Surjvorpur ‘sjods poSueire Kyroyndos yng soyuyoy Kawa a.Uid aep wo yuasaud rekig wepoW a vorpmpemT — 9LT178 Inlfoduction fo Modern Physice Incident wavefront ‘erystal oocare when 1 = 2d sa #. interfere constructively with that scattered by the first, the path difference AOB must be an integer number of wavelengths, whence Ay = 2d sin @ ay where n is an integer called the order of the reflection, Equation (7.1) together with the requirement that @ = 6" constitutes Bragg's law for xray “reflection.” Suppose, noi, that a parallel wave train containing a continuous spectrum of wavelengths is incident upon a crystal, as represented by the parallel arrows a, 6, ¢, d in Fig. 7.4. In the figure, traces of five families of Bragg planes are shown, numbered 1, 2, 3, 4, 5, with their characteristic spacings di, dz... . . Many other families of planes might be imagined, ‘some perpendicular and some not perpendicular to the plane of the paper. Suppose that in tho incident beam there is a wavelength d; such that nds = 2d; sin 64, where n is an integer, ds is the distance between'the set of planes numbered 2, and 4 is the glancing angle between the direction of the incident radiation and these planes.. Then there will be reflected from this grouip of planes a beam A, of wavelength Xs, which will proceed in the direction of the arrow A. Similarly, we may have reflected beams B,C, D,”. ...'in different directions in the plane of the paper, and also many other beams reflected from other families of planes in directions not in the plane-of the pager.” Each Laue spot in the experiment of Friedrich and Knipping may be interpreted as produced by such a reflected beam. In general, the most intense spots corréspond to reflections from Bragg planes containing the greatest numberof atoms on each plane. E Ma 7.8. The X-ray’ Spectrometer W. H, Braggand his eon W. L. Bragg were responsible for the early Xray 179 from target T pass through two narrow slits Sand S, and fall at glancing angle 0 on the cleavage face of a crystal K—rock salt, calcite; mica, gypoum, quartz, etc.—mounted on a table D, the angular position of which can be read. ‘The rellected beam, which makes an angle 26 with the incident beam, enters an ionization chamber C by means of which the intensity can be measured. For photographic registration, the ionization chamber can be replaced by a photographic plate PP (Fig. 7.6). With the crystal set at a glancing angle 6, the reflected beam will strike the plate at L (or at L’, if the crystal is reversed). From the position 0 at which the direct beam strikes the plate, the distances OL and OA, and hence the angle 28, can be determined. The distance d between the reflecting planes of a crystal such as NaCl is-determined as follows. In the rock-salt crystal, Na and Cl ions occupy alternate positions in a cubic lattice similar to that in Fig. 7.4, which represents one plane of atoms. Taking the atomic weight of Ci as 35.45 and of Na as 22.99, we find the molecular weight of NaCl to be 58.44. Therefore, 58.4 kg of the NaCl contains 2V, atoms, where Ni is Avogadro's. number. .Thus we find for the number of atoms n in @ . @) Fig: 74j, ho xray spectrometer using (4) Foniaatnn dates.Bites soy Bk “Sy ty tacts ad eae eect oa oe ‘Bis dep parttatg ret ace eeeeectcs OF Seabed it petit 19958 Seei at je jeatacbice sth RE Hi ‘pits stictidyfubs dintiqaddd Meee ot tH Mio aa (ati ttt ttioe _Bise toys bt 0 a es to ea, ai oatioan Boo Sat A ites enol pote ssodnid 618 soit oHbiioitadtibie SBady, thse ag pid HiFi apobabiys aie wid 36 SqLAS|DARA vith Siti foupsod oith jo sipedd otf) bes ip. ‘posit peisAio- out sin aktaa ro ond tea conde ty Odewls soo ‘uorerpes ohutionfsotoni Sunes ath pojtiotsdas ekuaye w[ayEkid tusioyrp Jo} waipeont dirjactisod osin Wi ‘iSAsdot ‘fps PootqAndS Hug ‘peokds Spud fionoutany © poi jiALi3 ows 48EN Rapopol oqo cpio itesogIp 48 paitiodd vita! oy omp, Bulag eouasoyip Luo Ot “STEAKS ZoTNO THEA pouTthgs siam y°) “By OF feuNE KIN “kpanoodsad *y ze't ‘ETL L6'6 Patbiqo pte "9 pow “9 "*p soy sopr0-puosoe otf} ioj gm pi "9 ped “9 “to sou ain oj {=u Serge ‘spsuapeave itor pomnduios off +19 pue tq “18 sod att jo suorpoopor "2 959 wou oniacnsiep dei Sepnons ‘wmnoods kes ro oy tonnglasp k8ubaa oof an 8SBRSG 22 SE Sap ithe Soy ana wopOP er ehony =ppio-puboge se #2 puw “ig “o eurvar ons’ poadadiniuy Siig.’ “isa ‘oth ofqnop, Kpewanxordde ye poassaqo sea “to. pure "Wg: ‘to “ompxwa ‘aaayy Jo dnosd puoss ¥_*.9°ST PUP ‘9°IT ‘6°6 JO 9 so{SuE earjoodsas any 3 poareago su "19 puy ‘ig “tp ‘oumxem som jo dno y “pouiego sea 2-1 Sty wy wots yom) TEENS dno y “Ginmpront droys w OF soe ureyia0 ye sso ang ojfue mise AfuOsun Axwa rout pip ArysueIat Ker oy) IU punos puw g a[Guv Fujsueys ysurede yuosNd uonEzTUOS JO- damo om ponold aff “avg Jo sdans uy Joqureyo uoryeztuo! aq pUS BV jo sdoqe uy [ershzo amp porwr Beg “H{ “A “TPrsAs0 4[e9-yON w Jo 200 aBeavopp om Uo yueplout yedsey mnume|d @ wiOsy BXEi-X Jo TIVE ® TEAL uopoppoy onsyexos1049 opemorysoULTY 92 suo | 92966 | ax‘aves = 60, | oro0s: | ozsesz otris“oseo ‘wnat, ¥0"0 osvsszy | Zo over ors ‘naend so | usrseor |. s¥'ez0e t0090 ‘IPD two | asverez |: oorvtec DPN Se PO we 0t wn OLX 2onx ox . seme | PPO hig = vioq og |__| ___| Pus 204049 | 9,91 yo p 220dg Ouypo4s) S ‘cdconongands foie sup poem syeancso ousos fo ssonds Pures: 1-2 219° spnysyp9 soufop sod nowsuedxo xeon om pup sqonds Sunes sro yA sme Kdoosaxseds ere uy pom Atuoumtios speyskso OWS MEY TL OBL “MeO TED sso] quoared ozo’ sf uN X MT WWI} UAOUA Mou ST IT “NX SH'6ZOE 9 0 D.8T 18 B10[¥o Jo eonds Fund ayy Surqer Aq pouyop ‘HI ¥ Jo suse, uy poysodax wooq uoryo oauq smSI[eAD Kex-x ‘OiO}OIIGL, “WAOTY ame qeakio [wordy v Jo IY! TooojOM pue KrsuO am pre zquINE soxpoSoay wor} oeposd o20u st UOT %0T UL ered toy w Jo ord & ‘yuA paredmoo oq eo stpfuoppaua Kes-x ‘efuyyexd peyeAz0 Foren kg “Y 28% = MOT X' 28S = P FTN -wnbo ony osm mory’ “:p/1 =, u uw ‘Buol wi swOye Jo KOI v UT suoye Jo sequina amp. ef p/L “aqno ap Jo 28po sp) Su0pe- 7x90 amp ‘PUB. Moye 2u0 Jo 22yu90 2x7 WaeMoq douEIsEp Om SI PIT “TOBN auTTEISAID Jo Avsuap omy “,t1/3y 501g = 4 |s0qM-"HH'RS/A7NZ YOR IPO JO. T eye Wepoy 1 uoNEMpaT — ORT( i | M cris lines a min t j i Fa i - * Lseries lines { “ | K tevies lines { a a a a Wavelength, A 1 Fig. 7.8 Wavelengths of some of the more prominent characteristi lines; ‘oaly @ small feaction of Land M lines are shown. ment the characteristic spectrum has a few lines at a relatively short wavelength, then an extended region in which there are no x-ray lines, followed in turn by a region in which there are a large number of lines, and so forth. The shortest wavelength and most penetrating lines ‘emitted by an element are grouped in a narrow wavelength interval and are called the K lines; those in the ‘next shorter wavelength region are called the L lines; for the heavier elements there are groups at still longer wavelengths called the M and N lines. “1 Moseley’s Law The K series of each element, as photographed by Moseley, appeared ‘8 two lines, the stronger of which was named Ka and the weaker KB.ci te “ionepr ansubeindS“Suionpoud'uy dn oat no * wipe amy auqn' k8u0ng yso8ie] yf sy baboon om Joins aonin ue jo "37 Kou ondtiny ange ony Ave "24 eoUbs8 gy 0 01 Tetons ret ‘pei me in ps aon “9p yew poaoiye yunyy pad omen St6q oy _-Ayenpesd yo.sypes Aso, ayy’ 2pre mpSudjeana-du0] 3qj. uO: gesioq“poagep" yp’ a1 oouaiogp Teniioiod oqny wei? w 49, param “mauapAua’ yeoriots! ogy “ot “prin ‘pbuajeone-joys & SysIx9 aim soNDIBP TeRUROd oes. soy“ “Ayoverpowayy oo oxy ons 294mn9 aM Jo somo} DOM NP +X pur x wsomyoq woo YSUIPAwA OF} UF payrtCS semod amp: st py ox -oqy Amsex omp_o1 pordde soouazapp Ton ,“Eeod, Ino} ios (tinsBum) wresyjom Jo umnnveds snonepUod 9xp, 10F fof) 20 MEP BIST 9M Smous TTL an¥Ly “Ponuepr ‘Kyrenitoswa, are mou “9p ye8im Jo Nios ye wo Foods fonimpiiGo om Jo soanyeay [esoued OWL. “sony onsLmoEseG sms se [oA sw poyNID eXeI-x Jo wnM20de Snonupuoo peoig ¥ aXeaye sy ox0xp 98:04 ¥ Uo [oj SuONoOTS ABN GSK, TOYA. “wnsjdeds Kus-x snonunuoo om Moxy omteD «ij CB0U Jo suoxoyd 58 porjae sf uonsipes ononseuionop exp swoqoddy amy 105 y1oddng uuninoedg heey enonupieg sus #2 “ajeqean st ojduree y2nous a8201 w poptaoad “cmnnoods Kexx y zedoad amp Jo moj aq uy yor] eANMGEp v opqUTCAR fey yum “9P Mow v Jo KroAcoep om 40) uNBIp Luw “yom s,LopsoHy OMS soAqy ‘Teuonemr 7.8m amp uF juoW2y> amp Jo souasexd ayy Jo jooud oxysnfonco ‘te popisfax Ajfesoued 1 quomale uv jo umuoads onsumoeseys some PERT Jo vonearesqo on “Fy soqumU omoye seq uINIuOYANE YIN pre Sapstar sea gy quaui9pD oT VONVEATIEGO otp gw UORNGLUOS JOE Img “sisimoyo Aq poisons wo0q poy se “(z0T'6e = 4a 40) unio x0yoq poonyd oq Pmoys (g¢6°6¢ = 14 30) UOSTE yey pouITUOO gem 4t Ayenung “1g soquinu quowap 9 (129g = 4a 38) [PYM TEM woNeT (e6'8s = 44 90) vyeqoo rem punoy sea ay -2Iqe ofpoued I THA ‘onsoutoo ur yeoie;u yuau3 Jo oxo Horqe sua[qoId [OIOADE JO UOHN|OSOL OMT sem rom 8,Xoj9S0HY Jo suonwoydds yreyrodany yexy om JO 9UQ “(96 20g) ayog Aq Podopeasp Azoory w Jo suisoy uy a[qereaw Apoyerpomtunt 3s0uHTs Beak STONY Toomtdan »,koppe0yy jo uonsuepdzs uy “ou waa? v 40] yaMsU00 are 2 Bue y soy “myuouiemseom jeyuamDdxo ©} sy pood kiquuoseas oavd (1) 3@—Z)y =« ‘wi20 9mp Jo woRFUNbo we ywq punos seat 35 sour] JoMO IO Ser nony ab 8 soul ot Una aotal MAasie aT MOE ae LE pissin iy « ot ’ 4 ; Pah Le 1-01 x fosonbay oot aomabg, ¥ phon siaiony ey TH lt — Z)n01 X B¥e0 = « one ot ef Kppsopp Arak pojuaseadas oq v0 Z soqunu omens Jo quomop Te 10} arf Myf otn) Jo KouoMboxy oy], “yayd £epas0yy © pores sf sequin oywoye Jo totouny v s¥ porofd sf ou Kui-x we jo (queysuoo © Aq portdriquar {tpensny Kouishbixy oy Jo yoor sfenbe amp yore ty yded y “soquinu oftuce Jo Honotity se Sout gy pre my O41 JO ‘s91819U9 amp Jo yoos oxeiibe om Jo r0}d amp saioys oz amy (-AqAwoUIOD preadn yq3y[e v oAny Sout of Jey; tLkOys SABy syuoMIEINFeOUT OBID0Kg) “amy 1yaENe w kpreOH sppELK Joqumnd ofeuoye Jo uoNoUNy w sb oxy Kase uoay8 Aue ysourpe Jo Kotonboxy 2tp Jo yo0r arenbe ox Jo 1ofd w “poopty “anime off Jo sequmu ofmoye ayy 04 [oR Odosd Aprwom suk ony Oy ema Jo Kouanbosy-3in Jo 1001 azenbe ax erp poxeccomtp opssoy ‘ontzodasd oymtoye Amour Jo onsyojouiayo os Ayorpoued omy Jo uoNworput oust aZOTL, “paseauouy sy 398ri otjy jo zequinu oTtiOTe omy] Su sesedtsSp sift fenoods ‘UoAyS v Jo TBBUsfoAwA OTH JOM ‘SMoNe g'2 “Biz sv ‘puitio} Kepesoyy' -ouN dy 9 WOM TpSUopeARa soRNO| v oY eXeaTe OUI MY om. uaKHO.TaALT Aue sog (xojdmdo os of gy se daoiry KTreuisuo Surf aM set pue ‘selqnop Ayemoe sf ouy my oy JWI MAONE BALDY S}UOLHdENSwRH Joye) rtd MPO OF MorempaEY yerSOO" Keays der ‘186 Introduction to Modern Physics ofan electron. Whenever ¢0 is negligible compared with hv, Eqs. (6.11) and (7.3) take the eame forme. 2 4": * Thus the quantum: theory. readily offers explanation of the short- wavelength limit of; the:continuous'x-ray spéctram,;4 topic with which’ | classical physics never dealt euccesefully.: +." . 2., The intensity ‘of the continuous spectiGm increases at all wave: lengths as the ‘potential difference across the tbe is increased. The total x-ray power emitted is roughly proportional to the square of the potential difference 'V-actdss the tube. and tothe atomic-number, Z of the target. material. For ordinary x-ray: potential-differences the effi» ciency & of x-ray produetion, defined! as the ratid of x-ray power einitied to the power dissipated by the eléctron beam, js given approximately by 10 Pile se geeks an Abine nent shoe yew vo 6B Lh OLOMEN ho: sca : ; 7 plgcie Hoa, HA ACER bain Mie oF tb : Most of the'incident electrons in.an x-ray tube:do.not Jose all their s kinetic energy’ in a single collision: -A typical electron undergoes several | collisions in a "thick target, “emitting: quanta’ which’ have ‘wavelengths Jonget than dal PXEE ENS Starke AML NE oer 0 ifference V across the x-ray tube is increased, , “fhe “T, cirve reaches its midviinumn “moves “Ulrey tould tor the data of Fig. 7.11 that eh ‘he aeleggt a whi akin covey “s | is added" However, “data “obtaizbd hi tlhe af aot flwaye t t aaa Ta ts Ot Surprising When one considers fey oe oe bates ‘in the coutiovous . : the complexity “of the phenomenon.” Some: of the incident electrons a ear enclose netiate into the target and radiate photogs; which may be absorbed sem onapum Sf RRR acs eens Yt before they escape, and so.forth. °°. #5. AUPE : pod eT ae Minne hein nace a ‘Thus far we have ‘been discussing the contintous\x-rays from: a goes toa single photon, thie resulting fiequeticy rx and the cozre- thick target. “However, if the target is guficiutly-‘thin'—say of very sponding wavelength “Awa should; according to the Planck-Finstein. | thin gold foil—only a few of the eléctrons collide with atozis in it, most theory, be given by ; ace s of them passing through the target undsviated. This, slowly movi clectrons will not be present in a thin target to the sauigAdee\as if 4 * thick one. . Accordingly; wo expect that a greatet proportioaof He energy in the continuous spectrum from thin targets should lie fea’ Ie Kafe: 2 i howek Timit than from thick. targets. This is in agreement with expétiment.” {? j ‘ at, Tn. the ‘continuous spectrum, from a.tery thin target, expiciment ne and his collabora: indicates, in agreement with the wave-mechanical computation’of Som- Fs Ero’ data Bat Agia 2 ehenee ab nek aman merfeld, thatthe maximuin of the energy-distribution curve occurs al the ‘ofa tot dad ‘thin phehOlBetion-of The Buortoway length Vimit is oe : limiting wavelingths dese: dae On the short Way ide'ot ASiet the birve thé tnoerse phbloeleciric effet rey of the ji ae i. drops abruptly 'to the ‘axis of ‘abscissas, ‘whereas. toward, longer, Waves” | dlaition is converted "ate : ‘die’ Photocled it falls nearly in proportion to 1/34, as illustratedsin Figis7.42, «The effect Failiant energy is converted, ‘at Tea inietic energy curves for a thick target, asin Fig. 7.11, can be regarded s tising troit’- Tore) alee om peer nome om uno ae wn noe (ys £= oe as ee f= ‘a 9404 prnoy ef“ soniod Potoyiede ony “worse omy upeinosms 4 -seyper Jo°oroyds v. 2040. (2°2) “bay SonesBorm Le” 206 = 9 ww HEIKO tunmTONE-y ‘suoR soap presyoey Puy prea oii moo, Ajeuney om s'¢ axaqia Aq.w0ai8 st q 3v oxem porey “aos on jo Axgruoyaypoi0> on, “1 —.# we. ‘eymododmo9 £ om 20) og 800 op jenbo st (6'2). "ba, ut # ue sytouodmoo.2 amp, 204° “odaneAR yp wo Penbo aze yorya ‘syuauOdmo £ paw x oy poazoser oq UOT Uw 3 J01994 oinoop quoprouy oy], “oumpd zz amp xy SEW FON PUL 2 79].0 ‘uoqmoaueo 8 3¥ ‘uox}29[9 ue UO JWapIoNY S| BONEpeX pozLBjodun Jy : soaaitnson dL cy) or va 7 sf Ayouoqnr yuapiouy omp. on gf 4 AyysuorMt posonavo’ Tp Jo Orie om, “Aqysuazur oLnoope our Jo orenbe orp 0} [euonsodord sy AjYsuoyuT sour eo sara 43 Jo opmyusear amp oxeq 94 ‘poyngnoqne of 1/3e—"= 2 woJA “® pus z woeaged sidue 2m a # pure qf 04 o8:0p poyeD|2008 om MOI, douNSKP om oY 4 OFA D) #t-4 apnaytdare Jo 7 amyod ye 301008 2mnoep ap:mpMA soxvm dnouseMonseTe semper osmY pepo" Te ‘taco onausemdxoe[ [eotswep © Suppioooy “uoReIpEI OPIOUT amp JO IstOU OLjOO IO ONT 81 3 OIOyA “wi/Z2— = *D Aq UOAIB st UONDOPP OY JO riecpead cea oraisinl Ste cates noe amen ee smiojrod nonoop oi ‘aonoep oes) ofFuye & iE mowTY OARA SI JL or sony. ‘i wan og finn 12 Kany eoppar om. conoes =» Semin ‘eae ot aten neste a ir Sqnouieg ove! °* ann Ka won eee “aaa | p te gs: “ p Herb Beg) vonsdnp's oy Uy soon oN SE YN spe Z-oM SuOTe mmoxoap pezejod-ousd w pozspisuoo ‘uosuioy, sp op Aq posodod ‘Aioom ysiy sy, “uouedsa pus jeonezooM TOG SSoqpreasox uuu’ jo yoolqo ain uooq seq pur soIsKyd WOpoUI Jo S9L0aNp, uy qiod quvpiodmy we podeid seq Souonucs Avi-x Jo uouamouayd oy, propesnyy ‘Suysarrn9g Ko1-X 62 ‘eamno uonnqursp-A8:009 9m7 Jo uNUIeM JO yeod Oqy'C} ‘ptiodsox109 plnous Yor HoIsTTIOD Jo adka aqnqoud joou ¥ 9q PIO 939K, gs ponoadso oq 1yBtuL 44 79810) w YoNS UT “IY Jo FON[EA sNOLeA THI ‘SEL “SU UT 4m se qons ‘FAIS AreyuoMo|> AueMI Jo uoHysodsodne oy aaa on em ccoon oaQ4 tai. uinayoods Kas-x srontnueD ‘omy wy KBious Jo vonngune, EZ “Fe “3 ~ inky epee) orENpOET HET190 Inlroduction lo Modern Physics - ‘The ratio of the power scattered to, the primary intensity is called the scallering cross'seclion (or scallering coefficient). of the free. -tletrom, deienated by ¢¢ Caley 7 “ste The scattering coefficient has the dimensions of sea; of the radiation incident on a unit area, the electron scatters the aiouitt which would fall on the area ¢,, which has the numerical Yalue 0.666 X 10-* m?, or 0.666 barn (1 barn = 10cm? = 10! m*): The Thomson (or classical) scattering coellicient is independent-of wavelength. « If each tlectroti'of atom scatters independently, the atomic scattering cross section is the product of the atomic number of the atom and the electron cross section; ¢. = Zo, ‘Thomson's theory predicts not only the intensity of the scattered beam but also its polarization. At @= 0 and @ = 180°, the scattered ‘beam froin an unpolarized incident beam is inpolarized, but at @ = 90° ‘the scattered beam is plane-polarized with electric vector in the y direction. ‘When a beam of x-rays of initial intensity J» passes through a thin layer of thickness Az of any material, scattering by electrons in the layer changes the intensity by an amount —Al = ele Az, 2 (20) ‘nhere ¢ is called the linear ecallering coeficient of the material.’ If each OF the electtdag ofthe material were to scatter independently of the others, the linear scattering coefficient would be. 3s. anne ( a) cay here n is the autmber of electrons per unit volume, The ratio of « to the density p of the scatterer is the mass seatlering coefficient a, . For substances suth ‘ss gases, thé lied scattering cbéfhi- cient: depends; ‘of’ coutsé, ‘oh the détisity, but o4'='0/p ié constant: The first quantitative estiniate of the nilinber of lectroris fii ait atom was made-by Barkla and his coworkers in 1909 by applying ‘Thomson's theory to data om the scattering of x-rays by carbon. They passed a beam of Mo Ka x-rays through a graphite absorber aud after correcting for photoelectric. absorption found that the value of em is approximately 0.02 m*/kg for carbon. Ife = neq then em. nei/p, and the number of electrons par Kilogram is “8°8 0.08 tai ag wy ere THEE TO mi/electwon !The number of carbon atoms per kilogram is the tatio of Avogadro's j humber N, to thé atoniie weight A, oF ‘Ns 36.02 X10 atomi/kmoie a> “Trkg/kmole = 5 X 10 atoms /kg / Fale 2p oa at Ae Ae sella sel oo Werelengh 6X a Fig. 710 Rida Go sxitringeottceats e of vaioua atoms 1 Zea ‘alia en world porta tsa ve sealed depen Si es sion vn: (Rependacd by permission of D. Van Campa, Tne; fron Big MES in AH Compan and 8. Alison Xray Thay end Ep2% Paw ureoq {uoprony om Jo ywT sw oures ox oM0 ‘earouenbauf omy poutey” wo9.uiwoq poses otf 48m Puno} oGf -qUOproIt diom skeb-x oF eUIO;o ~ouomt yoy uodn suioye uogreo’ kq" poioyyeoe sorouonbosy Kes-x ox Jo Siuomamseots [nyereo Jo sinsz ay paystiqnd uoydmog ez6q Uy “£81009 J9MO] Jo 28 suo}oyd pazoytee amp ofesoAN arp. uO IeYA sor[deH ‘own ‘ureoq qweprouy oMp WEEN Paquosqe peas 1out ow skes-x posoTjeOe detp aaoys pey ssoMO uw KexH ‘F161 O10j9q -sofounbasj YK 3" oom 81 44 SoAoMoHT “uinz29de spoUseworoo|P amNUD om 4940 POAIDAAO #1 uoneiper quapPouy aif jo Kouonboxy ou 48 Suuzorye8 yong. --kouoNbad) sures oup 38 omppased sioijonjs:pozessjsoon ath teoaWasuspiouy om Jo Aususm ounseye oifs-£q uonour oyuomey, aydinns:mzo}od 04, waAtsp ; 34 suoD2[9 oIWOTe, “Fationeoe Jo .Aroom yeoIsseID e1p 0} “BUpIO0y tHduapoaen wwepeey ‘Suton oe DOME ¢ oun We ye (y ZOL0 = W) eXenx Hy ont Je Suveniooe UL, ST's Oud aL «teks fo mawerg Tupanmg uordwey ory “so120.iy fqpenumysqne uo9q ovey pinous uorre|noqRo Sa “CY 202°0) = OW UIG.s2%pex (y-¢1) uoHTMIpEL By no. poen 10 soquint omOye saHAt Jo sose;;008 w Po}O9|ee Oy PE * -WOyw uoddeD tod suoroope Jo zoquinu oy Jo quDuoansToUr [RysSOZINE 8.UPLOG UL pospoaut 970m soojoyo smoynai0y yeyR 8998 2u0 HTL “Fy Jo soAIND otf WON, “(c've “20g) eysIN pue UPI, « ©} anp Suuerye08 Jo ymoMnven feoquEYoou-tmuENb ‘oNsLAREpA am eG Povorpod sy stp Suopoawm 320s 4B 2 UL asw—I0OP OL sbuysayoos yor hoy : 58 0} poizos Suypq wovjo suorzop stmoye axp jo Sutsorjeas aanpeisdocs i ot "uBPLKeYT £q 300 poxr0M sem Bussey v8 yUaIDYOO Jo Au0=IN BULL, “uz : 91 Aaysusnuy poxoecs om pu 7 0} jsuoniodosd kjpywurxosdde 4 wore mH-meip9Hy “¥961 NOIKdoD, "yisuspeaem. B18 pudsee omy huodwer yoou qa (0 ~ Dyz00 = ve siom sqp2tsjoxva 20800I 205 :kpyuosayoo soyyeDs suos}oo}9 (97) 2ouUt ath 825 48 ‘onyea jpems Kx2A wv mo} soaeoz0Ur Y Sy “aunso0 s0Ud!9}ZqU 94th _pnituco o8 pus ‘orayd oures orp uy soi 20 a10m posodiodne azw auor}2of> euoqudion £q pozenyeos eaxva otp tktpuopuodopur soyyeneso800] ou fath, mole Ue ur suoNs9p teomiog seoxieeP om yRIA opqesedinoD ‘soulooat KeeHA “Afreossseyp pouroydee Aya a1 Qursoyyeoe oaeoo% aH], “(HTL Sed) "22 wet Je7ROs8 Bf *2 woes sToI SuUoT ee sImO,e aif} 200%, Prose (ev, # awoIoIpeeD BuSHee ayn eyaMITS JAN 30} yetN Pemoe Y £0 <'X 204 syustoujooo Supoyiwos jo eyuouramnevony Joye] ~ *¥ 30 3194 ynoqe sdequod sy puv amg;ou tmoye Cuoseruiod £4 peer, le 3 a a 3 2. g gs 3 z | 3 Z a 2 z 3 3 i 3 seadocg ‘N *f pu wus “M4 -Y aat¥) up 2¥ 800 “dofweipad pasertooe oyp ty| 1 194 Iniadacion lo Wearw Physica second somewhat lower. On the basis of the quantum theory of radiation, Compton derived a relation which quantitatively predicted scattering at the observed lower frequency. This Complon effeel represents ~ ‘one of the most conclusive evidences of the particle properties of electro magitetic radiation and probably.convinced more physicists that quantum mechanics should be accepted than any other experimient. Compton used a graphite block to scatter Mo Ka radiation and measured the wavelength of the scattéred photons with a Bragg spectrometer (Fig. 7.15). At each finite scattering angle, he observed two peaks in the scattered beam, one at the incident wavelength 0.707 A and the second at a wavelength longer by an amount ad dependent on the scattering angle 4, according to the relation AX = 0.024(1 — cos #) A. Comipton subsequently showed that ad is independent of the scattering material. To explain the occurrence of the shifted component, Compton boldly applied the quantum picture of radiant energy. Hé assumed that the scattering process could be treated as an elastic collision between a photon and an electron, governed by the two laws of mechanics, the conservation of energy and: the conservation of momentum. Let-an incident photon of energy hy collide with an electron initially at rest. The photon is scattered through an angle 8, while.th® electron recoils in a direction ¢ (Fig. 7.16). The kinetic energy K given to the electron is (m— moet by Eq. (2.12). If vy is the frequency of the scattered | photon, the conservation of energy requires that the sum of the kinetic ‘eiergy of the electron and the energy of the scattered photon be equal Fig. 7-16 Elastic collision of ajthgton with” _lectroa initislly at reat. aaa to the energy of the incident photon, or hy = ht (mm mde (a2) i tal to i hy divided by Each photon carries shomentum equal to its energy hy divided b the speed of light, or h/\. Since momentum is a vector quantity and is conserved, the «andy componente (Fig. 7.16) must abey the equations hy hn 608.0 + p cos ¢ so (7.130) a rab) i reel (Tie pls sigh appears where p is the momentum of the ‘recoil electron. (- The ; with sin ¢ because it is assumed all angles are positive ‘When measured counterclockwise from the # axis; thus in Fig: 7.16 ¢ is a negative angle.) If'we introduce the wavélength of the two photond throtigh'»/¢ = 1/ and recall that p ='mi/ JT er voy ern) tay — 44 = (eu — w) = y (1-2) “bat Aq wa4r8 sy suoxycof9 ooas axp Jo ,9(¢u — uw) = {Som onoury ou, “stpog q pur vosEAA “UL “D Aq punoy puv 10} Poyoo] Atrdmiosd arom A200" s,uordmog £q pororpasd suomao[9 pooos 94, fuos207 poooy uordwoy T1'2 “voy2s0u9 puw eyuomom penfay Jo oSuer © ona anq 9898 48 ATRIOS you a2¥ suoHoopp SuusayyNOe you YEN 1905 2H Jo sumioy up Kpo}oE;sHN poureydxo aq wvo out poyrpomtun sip oF Poredmioo oun, pegrpour omy jo mipworq soywo19 Aqenueysqns ony, - a “suonogd Kvi-x ppordy 30} o1qu Sa a 91.1 80 [Yous o8 st pus (972) “bay ut wore Sumooax amp Jo ssEOE om Xq ot Smopidox. £q vals st anys yTUOPAEM Peroipaid 944 nor{}00 ‘Hons uy “yooes on (jevekzo aantua we uox9 30) oye asta ue 0} Anan Anuoteugne PP ‘wori29pe punog wv rp¥a. woroUd w yo wostjoo w 40 pt=Ht af Fe rdsonos uoyoyd eys Jo suz2y uy poureidxs oq uo oum poyrpowin SAL “Asem oxy ee maduojpans owes oup. que oun ospe anq 20, poy Pou © A[uo you powedde asotp yom peyoMNodsa 8,noyduNo UL (orn) (0 00 — 1) Sa uo 21 Ur (9T'2) “bg Sunumas £q ptmoy Xppeos oq ues sv ‘Posres st uonond su=prour om Jo AB.ou9 om sv-Aqpides soeeanoct voqoayd Posoaieos puv auopiout useaoq sonanogyp ASsoua ayy “puey som otf UO, “uorond auoprony ox Jo £3z0u9 ayp Jo qUopuadopuy st nq aya Buty woe om no spuadep “(91'2) “by “pSuepaum ur anys uoydmoy aq, H/4 8p soem 7901 Jo Botord e Jo mBuepava uovdmey ax ayy ‘soponsed somo on wap! ap uoaye of [nyeen sompamios st 3]. “uon}0opp ox Jo aus LBIoUD ye at Sr Tunbo kfsoue wpym uoioyd v Jo yiBuapeAex aif 01 spuodsousoo 3 tosh “9 910 J0 ybroaton wordy stp se wAoUy st 2u/y KanuENb 9a, “suonoyd Jo sonstiaronreyp ajonied oy Siwenox Suuazeo8 wonduiocy -synsox [wyuouedeD om mit jurueorse WwelTeor2 UE $f Acorn s,uerdwor kq payorpard ayys ySuaP—AEA om ‘NN, oe Y © 9 — pozseoo = ut (9800 — DFR ex m yy 2 PML MepoH oF worRmpomUT —96r0 for the velocit ig can be‘computed, reased further, the any direction of the unmodified Jine erat large angles of scat- Marrower. Eveatuall that of the incident ‘Beai ie Compton line, soon Sane way “Ta Classical theory, "an électro- ‘Certain character at eye .. According to the new view, sucha conoéption’ of the’ electromagnetic field is valid at’best as an epproximation. anh . ae ’ It is possible for the haracteristics under different ‘circum. tion of thie field takes the form with the Compton'éffect. “Here photon thereby changing’ the mpmeiitum of id discoritiniouisly.: “The change: oMmornen- ‘bierVable and mieasueablé by the physicist ‘Problems 1. _ Find the grating space for KBr and KCI, which have densities of 2750 and 1984 kg/m! respectively. Ans: 3.30 4; 3.14.4 . 2. "Find the wavelength for which first-order Bragg reflection occurs at 2 20° glancing angle from calcite, At which angle does second-orderLee sire D 1 yo fe wosid 9 o[8tv te ve syooos toon amp 10m AON *# ous 18 70 soxout uostooje floods otf Suter r208 Uordiiog w UL IT “OZ “isiimy sig 30y 2/4 opted aife pttu “aoxya9f9 poder au JO A8xou0 orp “aol pos wos amp J6 ABioto ofp puyy “praacyoeg 1ysreNT Posonre0s s pus tio péor{ uondeps tit sets LoroUd ASH-ITS'O VY “6E may = 9 atom 800. WET aay ay t sf A3z0u0 om nt WIA wojogd ¥ sO “BT ADIN S26 Fy © pity “AeTH OT = 473 @) “too Hoses 9x Jo £83049 onoU om st HY oxDya Yeo E + 2 os as IeTIOO SII OF SMB] UONEAIDStOD vip Ardy (0) VI AoE 0} UOISTTIOO sit et abit sayy & TTA uoIsy[oo uo-puoy B soxeUr ay AI:0US YBiy Jo wooyd Y “LT “@LD) ba came “OL 09°98 {2H oO X ST SUV . cwosnosfo ou J6 fie poser am; pare uogd qnoprour otp Jo Aouonboxy oma punt “uoyoqd wuaprour amp Jo emp oraa thdaoppaea ¥ a0y wopoyd poserteod oy, "390848 ATfeRTCE Noi ue fq 406 YSnosyy poroyyere sf «y KB30U9 Jo UOjoyd Kes-oyms0D Y “ST ‘vox oop ay Jo mnyuouoM om Bf d Bx0qA ein Moys ‘Furteyye08 uojdmog suosz9pun aday, = # 809 Pd FY) ‘) Aq waa8 sy woroa]9 uoxdmon © 205 offre [ooss oy wy MONS “FT APT 98e 2UY -yox gu Aqjenqur uosyoop9 we oy ABs0u9 84 JO JTeY -ouo soyouesy [Ins puv savy Am conoyd v kBz0U0 yeo]TouR om PULL “ST \g yo susan uy (2L1°2) “bay Aq 20 $ yo smmz0y wy (L1°2) “Pa Aq oad 81 uonooje pasnees undue) w Jo ABi0ue oHeUry om AUT MOUS “ZT ‘ASW 9840 PUB ‘987"0 ‘26000 “UY “A2W OF PUB ‘AW T1S‘O ‘APT OT JO 80 ~1o1e uoroqd-yueprour 20} wojoyd pazowsnos amp Jo AFz000 op PUL %06 Jo oy8ue ue ySnoutp woDeps we kq poxeyiees NrdmoD s woroNd Y “TE tor honx lt . ole uy ~-€¥ 060°0 30 inSuapexem w anny suinoud pozortess uorduion ibm spBaw sey ry” “wogseo IOyE atp uty “g M9 IMOGD FY 2oOz"9 seUy -sKeu-x ur poyerper somod oqeur -Rroudde ath avewpsa {poonposd ates oxp Jo sa t9uapscen.sroye Om SETOUAL “ANT 09 JO Souaromr Tenuanod wae yur sy SuiaIs fjddns sodod op v tna poreiodo sy 7a81m unnsum v maya oqmy Ker Uy “p oUStY ST uy MMT pBUDPPABAL-q20Ns 9x 78 suOIS yd 207 o}BuB Su;DUO;T Sm Pwo out sip Jo" HABuopava guy puLy “.9"ST 9q OF puMog st aut] PY NO 2m 20} pwisAso TOBA ¥ MpL oue TuUNS oyT, “Ay 7 Jo soUd “RUFP nURIOd w ae poywsado a} 72820) raddoo vinta aqm kx Uy g a neh fy 9203 sup e9ps0 4825 aT (Y 162°0) UonIPe MY OM Joy oue Fupumy om ey rey eanG0 uUNoDYOE ikea wrpeye a worempemsy ogg202 Inlroduclion to Modern Phyice . 21, Verify that a freely moving electron cannot spontaneously emit or absorb a photon. 22, When a high-energy photon undergoes Compton scattering at an angle of 30°, the recoil electron is projected with ¢ = 30°. Find the energy of the incident photon. Ans: 2.18 MeV 23. (a). The average intensity of a beam of plane-polarized electro magnetic radiation is jcc, where Ep is the amplitude of the electric vector. Calculate the classical amplitude of vibration of a free electron under the influence of a beam of x-rays with wavelength 0.7 A and an intensity of 20 W/m?. (8) What power is reradiated as scattered x-radiation by the electron? (©), Find the intensity required to give the electron an amplitude of 0.05 A, 24. A spectrometer to determine the energy of incident rays operates as follows. Collimated + rays strike a thin target, and Compton recoil electrons proceeding in the direction of the original y's are analyzed magnetically. (Note that the Compton scattering angle is then 180°.) If ris the radius of the electron’s path and B is the magnetic induction, derive a relativistically correct equation for the y-ray energy in terms ofr, B, and fundamental constants. 25. In the inverse Compton effect a photon gains energy in an elastic collision with a high-energy electron. If in the laboratory reference frame, a 10-eV photon collided head on with a.500-MeV electron, find thé energy of the Compton scattered photon in the laboratory system. 26, One means of producing monoergic photons with high energy (and nearly 100 percent polarization) is through the inyerse Compton elfect, (see Prob. 25). If an intense laser beam of 6942-A photons is made to je head on with a beam of 20-GeV electrons, find (a) the energy of the incident photons in a reference frame moving with the electrons, and (6) the energy of the back-scattered photons in the laboratory reference frame, Some of the high-energy photons observed in cosmic rays may have gained energy through the inverse Compton process. chapter eight (Radioactivity and the Nuclear Atont The discovery of radioactivity by Henri Becquerel in 1896 marks the ‘tarling point of nuclear physics. As a result of his work and the investi- galions by the Curies, Rutherford and his collaborators, and many olhers, il became evident thal this phenomenon involves the operation of forces of a different order of magnitude from any familiar lo chemists or physicisis of that time and thal one was dealing with the aclual transmutation of the elements, for which men had searched in vain sinée the lime of the Egyptians. ‘As a direct oulgrowth of these studies came the identification of a host of malerials possessing lhe remarkable property of spontaneously transforming rom one into another, dnd with the unraveling of these successive (ransforma- tions came the first clue lo the inlimale relation belween the elements, Exploi- talion of the enormously enérgelic radiations which these substances emilled led to the discovery of the, nucleus itself, to the nuclear model of the atom, and finally'to the arlifically induced transmutation of the slable elements. 203chap. 9. 0€ BROGLEE'S POSTULATE 84 a respect to A oF Ap i pat fb Note that casialy the minimum energy would be sero) [A-TY tube manufatorer i attempting t improve the picture resolution, while keeping eats down by desizing an ceton gu that products a electron bea which will make the smal! posible sp onthe face ofthe tube usig oly an ectron emitting cathode fellowes by # system oft well spuced apertures, (a) Show that there isan optimus tiameter for the second apertre (9) Using reasonable TV tube parameters. estimate the trina posable spot = MLA hoy on top ofa ladder of height I i¢ dropping marbles of mass m tothe Noor and trying to ita ctack inthe Noor. To si, he using equipment ofthe highest posible preckion. a) Show thatthe marbles ill mis she crack By am average distance ofthe ter oth a, wheres the acelration due to rant. (hi Using reasonable Salus of ff andi, evaluate this disince 38, Show that in onder to be able o determine through which slit ofa doublet sytem each photon pastes without destoying the éovbleat diaction pattem, the eoadition IRyAp, sh must be satis, Since this contin volts the uncertainty princi. i EISBELG - RESNAC QLANTUN PHYSICS “ 4 BOHR’S MODEL OF THE ATOM properties of mov parties mug scaterng Geiger. Marsden experi men afore of model RUTHERFORO’S MODEL, nce e-partle teaetores: impact poramcer and distance of closest ap proach: Rutherford’ calculation: comparzon with Geiger Marsden expe ‘ment nuclear ead: definition of eiferental roe section; sll ange Ru ‘ered settering ers section ‘THE STABILITY OF THE NUCLEAR ATOM radiation by an acecerate cassia charged dy ‘Tome sPEoTRA line spect: hydrogen series: Balmer Formula: Ryders constant alkali serie! absorption spect ‘BOHR'S POSTULATES statement of pestultes orbital angular momeatum quantistion: apps BonR's MoDeL Boh calculation: orbit ai onecetron atom energy quantization: com arson with Palmer formu singly ionized lm ‘CORRECTION FOR FINITE NUCLEAR MASS reduced mass: Ryders consnt evaluation: postemivm: dewterim: mv ATOMIC ENERGY STATES Feanek-Herty experiment: ionization energy omtinaum sates INTERPRETATION OF THE QUANTIZATION RULES Wikon Sommerelé quantization rules: phase space and phase diagrams simple harmonic asiator ane-eleton atoms and de Broglie’ interpret "im pacicle in one-dimsnsonal to ‘SOMMERFELD'S MODEL ‘quantization of epic obits: principal and acimathl quantum numbers epeneracy eet of relativity. Rydrogen ie structure, Unesteetare coe 1085 ‘07JEL OF THE ATOM 88 ‘chap. 4 BOKR'S MOD: 4:11 THE CORRESPONDENCE PRINCIPLE ” statsment of princi jvstifation: charged simple harmonic oxeilatr: hydrogen stom 4:12 A CRINQUE OF THE OLD QUANTUM THEORY 18 rocaptulston flores of the old quantum theory: earch ors replacement ‘vestions 19 PROBLEMS. 120 41 THOMSON'S MODEL By 1910 experimental evidence had been accumulated which showed that stoms contain electrons (eg. scattering of xrays by atoms, photoelectric effect, etc}. These {experimenis also provided an estimate of Z, the number of electrons in an atom. They found it to be roughly equal to 4/2 where is the chemical atomic weight of the atom in question. Since atoms are normally neutral they must also contain Positive charge equal in magnitude to the negative charge carried by their normal complement of electrons. Thus a neutral atom has a negative charge ~ Ze. where ~e isthe electron charge. and also positive charge of the same magnitude. That he mass of an electron is very small compared t0 the mass of even the lightest atom implies that mos ofthe mass ofthe ator must be associated with the postive charge These considerations naturally Ted 10 the question ofthe distribution ofthe positive and negative charges within the tom. J. Thomeon proposed tentative description, ‘oF meds), of an atom according t0 which the negatively charged electrons were located within a continuous distribution of positive charge. The postive charge dis: Iribution was assumed to be spherical in shape with a radius of the known order ff magnitude of the radius of an atom, 10°1""m. (This value can be obtained from the density of a typical solid, ite wtomic weight, and Avogadro's number) Owing to their matual repulsion, the electrons would be uniformly distributed through the sphere of positive charge. Figure 41 illustrates this “plum pudding” model of the ‘tom. In an atom in is lowes! possible energy state the electrons would he Mixed ‘at theieequilibriam positions, In excited atoms (eg, atoms in a material at high Temperature), the elecirons would vibrate about their equilibrium pesitions. Since Classica electromagnetic theory predicts that an accelerated charged body, such as a vibrating electron, emits leciromagnetic radiation, it was possible to understand {atitaively the emission of such radiation By excited atoms on the basis of Thom= Son's modsl. Quantitative agreement with experimentally observed spectra was lack ing. however, Example 4-1. (9) Assume that thare is one lett of change —e inside a sphrial region ‘Funfor postive charge density Thomson hydrogen atom). Show tha i motion. i fos kinetic energy can be sl harmonic oxilaion about the center of the sphere Ci Docc eas of postive charge Ler the eletzon be dispoed to a distance fom the center, with aes chan the rads ofthe sphere, From Guuse's low, we kag that we can calelae the frce-on it by wsing Coulomb tw Cea ee where (43a isthe net postive charge in a sphere of rade a Hence, we can write F= “ha. whete the constant k= peg Wf the ekcton a a eed with no inital velocity. this force wil produce simple harmonic motion alg & dameter ofthe sphere since Ii says ddrected towards the center and has a srength which is proportional tothe Ginplacement From the centr < (©) Let ihe total postive charge hive the magnitude of one elton charge (so that the azom fst na net charge) and let be dributd vera sphere of rads P= HO 10"? me. Find the force constant kad the Frequency ofthe motion o the elim PWelhave 90 10" nme oul 9 (Lh 10. coal (1d 1 a “The fequency of the simple harmonic motion i then ved [ELA [23% 0? mum ie i 1 hg Since tin analogy to tadition emited hy electrons osilaing in a antennal the radiation ited hy the ator wil have ris sme frequency. will eerespond faa wavelength 225 ¢ 10" # nV a7 as sc10 ee 12" 187m In the far ultraviolet portion of the electromagnetic spectrum, It is etry to show that an tletron moving ins sable circle orbit of any rae side the Thomson atm Yevnes st this same Frequency. and sot would rate thi egusney so, Orcourse. ferent asumed radi of the sphere of pov charge Would give ile Irequeney. But the fat that x Thomson hydrogen atom has only one charac emision frequeey cnties wih he very large number a diferent quencies observed inthe spect of hydrogen < 00 A Conclusive proof ofthe inadequacy of Thomson's model was obtained in 1911 by Emest Rutherford, a former student of Thomson's, om the anslyss of experiments ‘om the scattering of x particles by atoms, Rutherford’ analysis showed tha, instead ‘of being spread throughout the atom, the postive charge is concentrated in a Very small region, or nufeus. at the eoater of the atom. This was one of the most in portant developments in atomic physice and was the foundition of the subject of helene physics Rutherord had already boon awarded the Nobel Prize in 1908 fo his “investigations ia regard to the decay of elements and. the chemistry of radioactive subvtnces,” He was talented, hard-working phi wih enormous drive ead selconfence after writen Tae in ie the hen Ltd Rherord wre “Tve jus een reading some of my erly papers And. you know, when Ta finshed I said to mysll. "Rutherford, my bo, you wed (0 Bet Game clever ellow:” Though peace at winning « Nobel Prize he was not happy that ‘wat chemistry pris rather than one in physics (Any research inthe elements was them v8 28 10001 S.NOSHOHLchap. 4 BOHR'S MODEL OF THE ATOM 68 seine oO ofan s-pate' scattering experiment, The region traversed by considered chemisty) ln bis spcch accepting the prize be noted that he had observed many tromslormations in his work with racioativiy hu never had een one as tapi as his own From physics! to chem Rutherford already knew x particles to be doubly ionized helium atoms (ic. He atoms with two electrons removed), emitted sponianeonsly from several radioactive ‘materials at high speed, In Figure 4-2 we show a typical arrangement that fe and his Colleagues used to study the scattring of » particles on passing through thin foils of Sarious substances, The radioactive source emits particles which are collimated into ‘narrow parallel beam by & pair of diaphragms, The parallel beam is incident upon {foil of some substance, usually a metal, The fol isso thin that the particles pass Completely through with only a small deerease in speed. In traversing the foil, how= fever, cach x particle experiences many small deflections due to the Coulomb force fcting between sts charge and the positive and negative charges of the atoms ofthe foil. Since the deflection of an 2 particle in passing through a single atom depends ‘on the details of its irajectory through the atom, the net defletion in passing through the entire foil wll be different for different: particles in the beam. As a result, the beam emerges from the fil not as a parallel beam but asa divergent beam. A quanti (ative measure ofits civergence is found by measuring the number of « particles Scaltered into each angular range © to @ + 4®., The 2 particle detector consisted of ‘layer ofthe erysalline compound ZnS and a microscope. The crystal ZnS has the Useful property of producing 2 small flash oflght when struck by an a particle. If ‘observed with a microscope, the flash due (othe incidence of single a particle can te distinguished, In the experiment an observer counts the number of light ashes produced per unit time #8 fonction ofthe angular positon of the detector. [Let 1 represent the number of atoms that defect an x particle im its passage through the foil IT represents the angle of deflection in passing through one atom, asin Figure 4-3, and @ isthe net deflection in passing through all the atoms in its Figure 4-3. An x particle passing through a Thomsen model atom. The angle (species “aaecory through the fi, then statistical theory shows that Byer wee en Here ©)" is the root mean square net defection, or sattering. angle and (FY is the Foot mean square scattering angle in 8 defection from a single atom The factor VF comes fom the randomness ofthe deflection: al delstions wer in the same Airetion.cearly we woul bain. instead of v4 More generals statistical heory ves the following angular distribution ofthe settered particles’ (ond = UE 6 40 2 a ‘where N(O)d isthe number ofa seatered within the angular range © to @ + 40. and Fis the number of pasting through the el Because electrons have aver} small mass compared to thea particle they can in any ease produce only smal particle defections and because the positive charge = Aisiributed over ll the volume of the ~ 10"! m radivs Thomson atom cannot provide Coulomb repulsion intense enough to proce a large deletion ofthe 2 peril. Indeed, using Thomson's model we find thatthe deletion esused by one tom is 10. *rad. This result and (-1)and (2 comprise the apart etering predictions of the Thomson modo of the atom. Rutherford and his group tested these predictions Example 42. (a) In 9 typical experiment (Geiger aid Marden, 1900), particles were seatered by gol 10m thick The averape scattering ale was found ta Be(@) = % 10-* rad, Callate 2° >The numberof atoms wavered by the x partis & approximately equal tothe thickness fhe fi ide by the diameter othe atom Hence ‘The average defection ange in tiversng a singe atom then from (4-1. aya ON 20? a, Ge Te ade 10-trad no in disagreement with he Thomson stom estimate & 10 rad « {by More than 99, of the » urils were seater at angles lee than 3, The measure ments, using 1 for (7), were in agreement with 42) for M(@)dO for angles © in ths ‘ange: bat the angle dirbution of the small normher of particles setter a larger angles wearin marked disagreement with (42. was found, for example, tht te faction of 3 scattered at amles peter than 90, NG > 9 Vw about 0 What doe 2 predic? P We have mo>907_ a ' ' 4 strikingly diferent result than the eapsriment vale of 10°* general the eumber of scttered particles was observed to be very much Lager than the peeled number for sll scattering anges greater than 9 few degrees ‘< we 90 8 yg 8880 The existence of a small, but non7ero probability for sattoring at large aneles could not be explained at all in terms of Thomson's model of the atom. which basically involves small angle scattering from mans atoms. To scientists accustomed to thinking in terms ofthis mode it came as a preat surprise that some 2 particles were deflected through very large angles, up to 180" In Rutherford's words! “It nas quite the most incredible event thal ever happened to me in my life. I was as incredible as if you fired a 15-inch shell at a piece of tissue paper and it came back and hit you.” by 208 68 7a00W S.NOSWONLlonap. 4 BOHR'S MODEL OF THE ATOM 90 Experiments using foils of various thicknesses showed that the number of large angle scatterings was proportional {0 -1, the number of atoms traversed by the particle, This #8 just the dependence on "+ that would arse if there were a small probability that an 2 particle could be scattered through a large angle in traversing 2 single atom, That cannot happen in Thomson's model of the atom, and this led Rutherford in 1911 to propose a new mode 4-2. RUTHERFORD'S MODEL In Rutherford’s model of the structure of the atom, all the positive charge of the Atom, and consequently essentially al iis mass ae assumed :0 be concentrated in a small region in the center called the nucleus I the dimensions of the nucleus ate small enowgh, an a particle passing very near itcan be scattered by a strong Coulomb repulsion through ‘large angle in the traversal of a single atom. If instead of using F'~ 10" m forthe radius ofthe postive charge distribution ofthe Thomson ‘atom, which leads to a maximum deflection angle 0 = 10°* rad, we ask what the racias of a nicleus should be to obtain ~ 1 rad, say, we find y= 10." m, This, as we shall se, turns out Co be a good estimate of the radius of the atomic nucleus, Rutherford made a detailed calculation ofthe angular distribution to be expected for the scattering of« particles from atoms of the type proposed in his model. The calculation was concerned only with seattering at angles greater than several degrees Hence, scattering due to atomic electrons can be ignored. The scattering is then due to the repulsive Coulomb force acting between the positively charged x particle and the positively charged nucleus. Furthermore, the calelation considered only the scattering from heavy atoms, (0 permit the assumption that the mass ofthe nucleus is so large compared to that of the « particle thatthe nucleus does not recoil appreciably (remains fixed in space) during the scattering process. It was also assumed that the particle doesnot actually penetrate the nuclear region, so that the particle and the nucleus (both assumed to be spherical) act like point charges as far as the ‘Coulomb fore is concerned. We shal se later that all these assumptions are quite valid except for the scattering of 2 particles from the lighter nucle, and we can correct for the nite nuclear mass in sich cases, The calculation, finaly, uses nonrelativistic mechanies, since we = 1/20 Figure 44 illustrates the scattering of an » particle, of charge +2e and mass M, in passing near a nucleus of charge +Ze. The nucleus is fixed atthe origin of the coordinate system. When the particle i very far from the nucieus, the Coulomb force fom itis negligible so that the particle approaches the nucleus along a straight line ‘vith constant speed v. After the scattering, the particle will move off finaly slong. a straight line again with constant speed ©. The position of the particle relative to the nucleus is specified by the radial coordinate r and the polar angle g, with the latter measured from an axis drawn parallel tothe initial trajectory line. The perpendicular distance from that axis tothe lie of initial motion is called the impact peramerer, speciied by b. The scattering angle is just the angle between the axis find a line drawn through the origin parallel to the line of final motion: the perpendi cular distance between these wo lines is Example #3. Shov that o~ 6 and b= The force acting onthe parti, heing 4 Coulomb force, i always in the radi! direction. Hence, the angular momentum of the particle about the origin has x constant value, Specially then, the inal angular Momentum equal tothe final angular momentum, oF Mob = Me's = OF course, the kinetic energy ofthe potcle doesnot remain constant during the scattering, ht the ini Kinetic energy ast Be eaual to the Binal inet energy since the nucleus | | Figure 4-4. The hyperbole Runerors trajectory. showing the polar coordinates, @ and the parameters b 0 These two parameters completely determine ie rajctoy. im ari lar the scattering angle and the dstance of closest approach The near BOI charge Zelion at focus ofthe branch of he hyperbola sumed to rena satonay. Tas Saebane? See ede eee re eee By a straightforward calculation of clasical mechanics, using the repuive Cow. omb force (eg(2Ze"/r?. we can obtain the following equation forthe trajctory ofthe 2 partite (see Appendix E for a derivation) t > jain @ + pros #0) 4a) the equation of a hyperbola in polar coordinates. Here D is a constant, defined by trey MER CD Ttisa convenient parameter equal to the distance of closest approach tothe nucleus in a head-on collision (b= 0), since D i the distance at which the potential energy (deegn2Ze"/D) is equal tothe inital Kinetic energy Mu#/2 (simply equate the two and solve for D), At this point the particle would come to a stop and then reverse its direction of motion. The scattering angle 8 follows from (4-3) hy finding the valve of gas op and selting = x. In thie way’ we find 0m 45) Bxample 4-4. FraluteR, the distance of closet approach ofthe partie othe center ofthe rucieus tbe origin in Figure 4-4 >>The radial eooedinate wll equal R when the polar angle it @ = tr ~ 2. Frauatng (43) meee 2 (ts) 6 TROON ScyosBKINY 27-298CChap. 4 BOHR'S MODEL OF THE ATOM 92 Now, from (481 we ean put 3D anual] 9 ‘head-on collision. R -+D, the distance of closest approach. Ako, as -+ 0, corresponding to ro dflction a il hath Band go to int, ar ould he expected “ From (4-5) we sce that, in the scattering of an x particle by a single nucleus. if the impact parameter i in the range h to h + dh then the sattering angle i in the range f! to H+ df. where the relation between b and # is given by the equation, This is illustrated in Figure 4-5. The problem of calculating the number N(@}d ‘of particles scatered into the angular range © to © + 4@ im traversing the entice foil i therefore equivalent to the problem of calculating the number which are incident, with impact parameter from b to b + dh upon the nucle in the fil. AS we ‘Show in the following example the result ie (aa) (ne) “nto al Nive where 1 is the number of a particles incident on a foil of thickness r cm containing ‘nck per cubic centimeter Example 5. Verify (7. > Consiern sepment of the fi with w rosesetional are of 1 an?, ax shown in Figure 46 AN ring. inner rae band outer radius b+ dk ie dre around an incident axis psing Thvouh each auclus the area of each ring being 2b The number of rach rings the ‘segment ofthe foi st. The probability that an 2 porte will pase through one of thee rings, Pla. oq to the tal sect obacorod bythe ings, as seen by the icient Partick, divided by she total area of the segment. We assume the fit be thin enough that the cin ignore vertaing of ring rom ferent nis. The prosese insoles singe scatering snd the pehblity for appreciate seatiering Hy more than ane motes very low. Hence bsdh = road Figure 45 Tho relation between the impact parameter 6 and the scattering angle 2 [At b ineroaseslo6s close nuclear approach) the angle (decreases (smaller tearing angio}. The 2 paricles with impect parametors belwoen b and b + ab are scatered info the angular range between ? and +d ao Figure ¢§ A beam ot parties incent on fings, which are purely geometrical contre land not anyiting physical ara canerea on fuel. Actually here ate enormously many 4 more ings than shown and he rings avery ‘much araller than sow, bu b= W/2h cot 2) so that ee ae sa and napa Demonte oF sae Thos Pind = To? ino ® sin ‘But ~Pthidh is equa co the probability thatthe incident particles wil e ssttered into the angular range 010 +l. The minus sign arses from the fact that & dcerstse 1h 2h, corteponds to increas in angie vn passing through the eae fi Nene a). Using our elie aotation © forthe sate dh <2 pp? 88.080 = Pra = Fo? OS Filly. with D = (1 Sreph-2e2 M022, we obtain 7. < If we compare the Rutherford atom result, (47, to the Thomson atom result (4-2) we see that although the angular factor dcereases rapidly with increasing angle in hoth, the deerease is very much les rapid for Rutherfords prediction. Large angle ‘stalling is very much more probable in single seattering from a nuclear atom than in multiple small angle scatering from a plum pudding atom. Detailed experimental tests of 4-7) mere performed within a few months of ils derivation by Geiger and Marsden, withthe fllowing results 1. The angular dependence was tested, using foils of Ag and Au. over the angular range 5° to 130), Although N(@)dO varies By a factor of about 10° over thie ane, the experimental data remained proportional to the theoretical angular distribution 1o within a few percent 2. The quantity N1@)dO was found indeed to be proportional tothe thickness ¢ ‘of the foil for a range of about 10 in thickness forall the elements investigated. 3. Equation (4-7) predicts thatthe namber of scatered 2s will be inversely proportional to the square oftheir Kinetic energy, Mr). This was tested by using 2 particles from several eiferent radioactive sources and the predicted energy depen dence was confirmed experimentally over an available energy variation of about & factor of 3 4. Finally, the equation prediets N(@)d® to be proportional to Ze, the square ‘of the nuclear charge, AC the time 7 was not known for the various tome. Assuming % 29 295Cchap. 4 BOHR'S MODEL OF THE ATOM 94 (4-7) to be valid, the experiment was used 0 determine Z and it was found that Z. ‘was equal (o the chemical atomic member of the target atoms. This implied that the first stom, H, in the periodic table contains one electron, the second atom, He, ‘contains two elecirons, the thd atom, Li, contains three, ef sinee Z is also the umber of eleetrons in the neutral atom, This result was soon independently confirmed by soray techniques that will be discussed in Chapter 9, Rutherford, his model now confirmed, was able to put limits on the sie of the ‘nucleus. The distance of closest approach, D. is the smallest value that R takes on, ‘which is Rat © = 180, Hence ' Roa = = Fee, Mol The nucleus radius must be no larger than D because the results are based on the sssumption thatthe force acting on the a particle is always strictly a Coulomb force between two point charges, This assumption would not be true ifthe particle pene= trated the nuclear region tits distance of closest approach. The previous equation shows that Ry ap decreases us Z decreases. The question arises: How much can Ryyg Aecrease before Rigo is less than the nuclear radius? Departures from the predicted Rutherford scattering were actually observed from the very light (ow Z) nuclei, Part ‘of this was due fo a violation, for the very light nucle, of the assumption that the hnuclear mass is large compared to the alpha particle mass; however, deviations c= mained even after the finite nuclear mass Was taken into account in the theory. This Suggests that penetration of the nucleus occurs in hese cases thereby altering the predicted scattering. Hence, the nuclear radius can be defined as the value of Rat the limiting scattering ange, o limiting incident energy, at which deviations from Rutherford seattering set in, In Figure 47, for example, we show data from Ruther ford’s group for the scattering of x particles, of various energies. at a fixed large angle from tn Al foi. The ordinate isthe ratio of the observed numberof sattered particles fo the number predicted by the Rutherford theory (corrected forthe inte hclear mass). The abscissa isthe distance of closest approach caleulated from (+6) These data imply thatthe radius of the AI nucleus is about 10 '*m = 10F. (The unit of distance used in nuclear physis isthe fermi, which equals 10. '* m. Note that F = 10 * A, whore A, the angstrom, isthe unit used in atomic physics) ‘The Rutherford scattering formula, (4-7, is usually expressed in terms ofa differential cross section defdS. This quantity is defined so that the number UN of = particles sattered into solid angle dO at seattering angle © is de an © nda 48) } eo Figure 47. Sonte data obtained in th seatering of» particles ftom s radioactive source by sluminlum. The toaciea = he distance o! closest approach tothe nucesr Cena. neni ersued ~~ Figure 48. lustrating the definition of he diferetal cross section dof If the target is thin enough fran incident partici to have negigibe chenco ot ntracing with more than fone nucleus while passing Through the target, then N= (aia ot IF particles are incident om a target fol containing » nuclei per square centimeter. The definition is analogous to the definition ofa cross section 2 ia (2-18) N=aln iis itlastrated in Figure 48. The solid angle dO, which is essentially a wo. dimensional angular range i measured numerically by the area which the angular range includes on a sphere of anit radius centered where the scatterings occur, For Rutherfard scattering, which is symmetric about the axis of the incident beam, we ‘re intrestd in the solid angle corresponding to all events in which the seattering faze ies in the range d® al @. As is shown in the figure = 22 sin 40 Using this in (4-7, writing N(@)J6 in that equation as dN and also writing the term pt appearing there asm, we immediately obtain 0 -(Ge) (ite) aoa Comparison with the deiton of 4) thn shows that the Ruther ctor “ (.) (Ze) act iin (ate) (Gite) se 49 43. THE STABILITY OF THE NUCLEAR ATOM The detailed experimental verification of the predictions of Rutherford’s nuctear ‘model ofthe atom lef litle room for doubt concerning the validity of the model [At the center of the atom isa nucleus whose mass is approximately that ofthe entire tiom and whose charge is equal fo the atomic number Z times e: around this rncleus there exist Z electrons, neutralizing the atom as a whole, But serious ques tions emerge about the stability of such an atom. If we assume. for example, thatthe clectrons in the atom are stationary. there exists no stable arrangement of the elec tuons which stould prevent the electrons from falling into the mucleus under the influence of te Coulomb attraction. We cannot allow the atom t0 collapse (back °0 1 nuclear-sized plum pudding) because then jis radius would be of the order of a BMLJOALNIWIS IML F998 96‘chap. 4 BOHR'S MODEL OF THE ATOM 98 nuclear radius, whichis four orders of magnitude smaller than diverse experiments show the radius of the stony to be At first glance it seems thet we can simply allow the electrons to circulate about the nucleus in orbits similar to the orbits of the planets circulating about the sun, Such a system can be stable mechanically, asi the sola system, A serious diiulty arises, however, in trying to carry over thi idea from the planetary system to the atomic system, The problem is that the charged clectrons would be constantly accelerating in their motion around the nucleus and, according to classical eleteo- ‘magnetic theory. all accelerating charged bodies radiate energy inthe form of elec tromagnetic eadiation (see Appendix B), The energy would be emitted atthe expense (of the mechanical energy of the electron, and the electron would spiral into the nucleus. Again we have an atom which would rapidly collapse to nucle dimensions. (For an atom of diameter 10” !m the time of collapse can be computed to be ‘10% scl) Furthermore, the continuous spectrum ofthe radiation that would be ‘mitted in this procest is not in agreement with the discrete spectrum which is known to be emitted by atoms. “This dificult problem of the stability of atoms actully led to a simple model of atomic structure. A key feature ofthis very successful model, proposed by Niels Boh in 1913, was the prediction of the spectrim of radiation emitted by certain atoms nce, it is appropriate at this point 10 desribe some of the principal features of such spectra 44 ATOMIC SPECTRA Atypical apparatus used inthe measurement of atomic specea is indicated in Figure 4.9. The source consists ofan eleciicdischarge passing through a region containing ‘8 monatomic gis. Owing to collisions with electrons, and with each other. some of ‘the atoms inthe discharge are pt into state in which their total enerpy is greater ‘than i isin @ normal atom. In returning to their normal energy state, the atoms give up their excess energy by emitting electromagnetic radiation. The radiation is coli- mated by the slit and then it passes through prism (or difraction grating for better resolution) where itis broken up into its wavelength spectrum which is re cenrded on the photographie plate. “The noture ofthe observed spectra is indicated on the photographic plate. In contrast to the continuous spectrum of electromagnetic radiation emitted, fr instance, from the surfce of solids at high temperature, he electromagnetic radiation emited Pte late | by free atoms is concentrated ata numer of discrete wavelengths. Fach ofthese wave> length components called a line because ofthe line (image of the slit) which it pro {duces on the photographic plate. Investigation of the spectra emitced from diferent kinds of atoms shows that cach kind of atoms has its own characteristic spectrum, ie. a characteristic set of wavelengths at which the lines of the spectrum are found, “This feature is of greatest practical importance because it makes spectroscopy avery useful addition to the usual techniques of chemical analysis. Chiefly for this reason much effort was devated to the accurate measurement of atomic spectra, and. infact, uch effort was needed because the spectra consist of many hundreds of lines and in general are very complicated. However the spectrum of hydrogen is relatively simple. This is perhaps not surprising since hydrogen, which contains just one electron, i tse the simplest atom Most of the universe consists of isolated hydrogen atoms so thatthe hydrogen spc trum is of considerable practical interest. There are historical and theoretical reasons as well for studying it as wil become apparent later, Figure 4:10 shows that part of the atomic hydrogen spectrum which falls approximately within the wavlength range ‘of visible light, We see thatthe spacing, in wavelengths, beween adjacent fines of the spectrum continuously decreases with decreasing wavelength ofthe lines, so thatthe {ries of lines converges tothe so-called series limit at 3645.6 A. The short wavelength lines, including the series limit, are hard to observe experimentally because of theit close spacing and because they are in the ultraviolet. “The obvious regularity of the H spectrum tempted several people to look for an empirical formula which would represent the wavelength ofthe lines. Such a formal was discovered in 1885 by Balmer, He found thatthe simple equation 3646 sin units) 7 where n= 3 for Hy, =4 for Hy. = 5 for Hy tc, was uble to prediet the wave: Tength ofthe fist nine fines of the series, which Were all that were known atthe time, to better than one part in {000, This discovery initiated a search fr similar empirical formulas that would apply to series of lines which can sometimes be identified in the ‘complicated distribution of lines that constitute the spectra of other elements. Most ofthis work was done around 1890 by Rydberg, who found it convenient to deal with the reciprocal af the wavelength of the lines, instead oftheir wavelength fv terms of feciprocal wavelength the Balmer formula can be written RYU? — Uh?) 9 =RAS... (410) where Ry isthe so-ealled Ryder constant for hydrogen. From recent spectroscopic Ket “| ili 6 vyanoaes oMOLY PY 298chap. 4 BOKR'S MODEL OF THE ATOM 98 ‘able The Hyarogen Series Names___Wavelength Ranges Formuls Lyman Uteavioit werse Balmer Netti and nana Paschen — tniared ne 5.6 racket Iniared n= 562 Phund fred na 628 ata, its value is known to be Ry = 10967757.6 4 1.2m! “This indicates the accuracy possible in spectroscopic measurements. Formulas of this type were found for a number of series. For instance, we now know of the existence of five series of Tnes in the hydrogen spectrum, a8 shown in Table 4 For alkali clement atoms (Li, Na, K,...) te series formulas are of the same general structure Thats \ 1 ae *Lin a tn 7] Gay where R isthe Rydberg constant forthe particular element, a and b are constants for the particular series, i an integer which is txed for the articular series, and » isa variable integer. To within about 0.05% the Rydberg constant has the same value for all elements, although it does show a very slight systematic increase with increasing ‘atomic weight ‘We have been discussing the emission spectrum of an atom. A closely related prop erty is the absorption spectrum, This may be measured with apparatus similar to that shown in Figure 4.9 except tha a source emitting a continuous spectrum s used and a glass-walled cel, containing the monatomic gas to be investi ‘somewhere between the rource and the prism, After exposure and development, the ‘photographic plate is found to be darkened everywhere except for a number of un exposed lines. These lines represent set of discrete wavelength components which ‘were missing from the otherwise continuous spoctrum incident upon the prism. and Which must have been sbsorbed by the atoms in the gas cll is observed that for {every line in the absorption spectrum of an element there isa corresponding same ‘wavelength line in its emission spectrum; however, the reverse is not true. Only ceritin emission lines show up in the absorption spectrum. For hydrogen gas, nor- ‘mally only ines corresponding to the Lyman series appear in the absorption spectrum: but, when the gas is at very high temperatures, at the surface ofa sta, nes corresponding to the Balmer series are found 4.5. BOHR’S POSTULATES All these features of atomic spectra. and many more which we have not discussed, ‘must be explained by any suevessel model of atomic structure. Furthermore, the very reat precision of spectroscopic measurements imposes severe requirements on the sccuracy with which such a mode! must be able to predict the quantitative features of the spectra Nevertheless, in 1913 Niels Bobr developed a model which was in accurate quant {ative agreement with certain of the spectroscopic data (eg the hydrogen spectrum It had the additonal attraction that the mathematics involved was very easy t0 understand. Although the student has probably seen something of Boht’s model in studying elementary physics, or chemisiry, we shall consider i in detail here in order to obtain various results that we shall want fo make comparisons with elsewhere in thie book, and also in arde to take a careful look atthe rather confusing postulates fon which the model is based. These postulates are 1. Am electron in an atom moves in a circular orbit about the nucleus under the in ‘uence of the Coulomb attraction between the electron and the nucleus. obeying the laws of classical mechanics. 2 Instead ofthe infinity of orbits which would he possible in classical mechanics i Is only possible for an electron ta mace in an obi for which its orbital angular momen- ton Lis an integral multiple of 8, Planck's constant divided by 2x 3. Despite the fact that itis constantly accelerating, an electron moving in such an allowed orbit does not radiate electromagnetic energy. Thus, ts total eneras E remains 4. Electromagnetic radiation is emitted If an electron, initally moving in an orbit of total energy E,- discontinuousy changes is motion so that t mares an orbit of total ‘energy Ey. The frequency ofthe emived radiation is equal tothe quantity (E, ~ Ey) dlicded by Planck's constant ‘The first postulate bases Bahr’s model on the existence of the atomic nucleus. The second postulate introduces quantization, Note the difleronce, however, between Bohr’ quantization of the orbital angular momentum of an atomic elecizon moving under the influence of an inerse square (Coulomb) force Lam 2 Re GID and Planck's quantization of the energy of a particle, such as un electron, executing simple harmonic motion under the influence of a harmonic restoring force: B = rhs n=0, 1, 2,..-. We shall sce inthe next section thatthe quantization of the orbital Angular momentum ofthe atomic electron does lead to the quantization of is total energy, but with an energy quantization equation which is diferent from Planck's ‘equation. The third postulate removes the problem of the stability of an electron moving in a circular obit due to the emission of the electromagnetic radiation ‘required ofthe electcon by classical theory. by simply postulating that this particular feature of the classical theory is not valid for the case of an atomic electron, The pos: tulate was based on the fact that atoms are abserced by experiment 0 be stable even thauigh this not predicted by the classical theory. The fourth postulate Fk, 7 eh (et is really just Einstein's postulate thatthe fequency of a photon of electromagnetic radiation is equal to the energy carried by the photon divided hy Planck's constant, ‘These postulates do a thorough job of mining classical and nonclassical physics. The eeetron roving na rela obits assumed tn hey clase mechanics and yet the pela idea Sf quantization of orbital angulse momentum is inladed. The eecton is assumed 10 obey fone feature of lass electromagnetic theory (Coulomb's awh sd ye Ha to obey aaier Feature (emison of radiation by a accelerated charged edly) However, we sould nat be surprised f the lm of lasial physics, which ate hated om out expriene with macroscopic systems, are aot completly valid when dealing with microscope systems such atthe af SBLYTISOd SuHoR $9 205. 66100 (chap. 4 BOHR'S MODEL OF THE ATOM 4.6 BOHR’S MODEL “The justification of Boh's postulates, or of any set of postulates, can be found only bby comparing the predictions that can be derived from the postulates withthe results, ‘of experiment. In this section we derive some ofthese predictions and compare them ‘withthe data of Section 4-4 ‘Consider an stom consisting of a nucleus of charge + Ze and mass M, and a single clectron of charge —e and mass m. For a neutral hydrogen atom Z = 1, for a singly fonized helium stom Z = 2.for a doubly ionized lithium atom Z = 3,ete. We assume thatthe electron revolves in a circular orbit about the nucleus. Initially we suppose the mass of the electron to be completely negligible compared to the mass of the ‘nucleus, and consequently assume that the nucleus remains fixed in space. The con- ition of mechanical stability ofthe electron is 1 zee mee ae where the spood of the electron in is orbit, and ris the rads ofthe orbit. The left side ofthis equation i the Coulomb force acting onthe electron, and the right side i ma, where ai the centripetal acceleration Keeping the electron in it ireular orbit. Now, the orbital angular momentum of the electron, L= mer, must be a constant, because the foree acting on the electron is entirely in the radial direction. Applying the quantization condition, (12) to L, we have or = wh Raha 15) Solving for » and substituting into (4-14) we obtain whe re beg 23. oe 16, and ze g-2 meL2Re IN “The application of the angular momentom quantization condition has ested the possible cirola orts to thos of radi sive by (4101 Note that these radi are proportional tothe Squsce ofthe quantum manor nf we evaluate the radius ofthe smallest orbit (n= 1) for 3 fydrogen atom (2 1) by inserting the known values of km and «, we obtain Tie 10." m= 05 A. We sll show later tha the cloton hs its minim toa ners) ‘ahen in the orbit corresponding to n= f- Consequently we may interpret the radius ofthis Orbit asa metre ofthe radius of hydrogen atom in its poral sat Is in god wareement Sth the estimate mentioned previous. thatthe order of magnide of an atomic eas i TTA. Hence, Bors postulates predict a reasonable ize forthe atom. Evaluating the orbital Nelecity of an cecton in the stale orbit of @ hydrogen atom fom (8-17, we Bind r= 22 = 10" masse It apparent from the equation that thi i the Largest velocity posible fora hydrogen ato cectrom, The fc thal this velocity ess tha 1% ofthe velocity of ight isthe sic fr using lsc! mechanics instead of relaivitie mechanics inthe Bohr ‘model On the other hand. (217) shows that for large values of Z the electron velocity Fecomes relativistic the model ould not be applied in such cases That equation also makes itapparent why Bohr could not allow the quunturs number ever to asame the value m= 0, 1s Ir may in Planck's quantization equation. [Next we calculate the total energy of an atomic electron moving in one of the allowed orbits. Let us define the potential energy to be 2er0 when the electron is infinitely distant from the nucleus. Then the potential energy V at any finite distance ‘rean be obtained by integrating the work that would be done by the Coulomb force ‘acting from r 10 22, Thus rf The potential energy is negative because the Coulomb force is attractive: it takes work to move the electron from r to infinity against this force. The kinetic energy ofthe electron, K, can be evaluated, with the aid of (4-14 0 be Ze * rel The (otal energy ofthe electron, Fis then ze -K+ ane Using (4-10) for + in the preceding equation, we have mort rego ot We se that rhe quantization ofthe orbital angular momentune ofthe electron Ieads to «quantization ofits total eneray ‘The information contained in (418) is presented as an energy-tevel diagram in Figure 4-11. The energy of each level, as evaluated from (418), is shovin on the let, in terms of joules and electron volts and the quantum numberof the level is shown fm the right. The diagram isso constructed that the distance from any level to the level of zero energy is proportional to the energy ofthat level. Note thatthe lowest {rmost negative) allowed value of toal energy aocurs forthe smallest quantum number n= I. As n increases, the total energy of the quantum state becomes less negative, with E approaching zero as n approaches infinity. Sinee the state of lowest total energy is ofcourse, the most stable state for the electron, we see that the normal State of the electron ina one-clecteon atom isthe state for which n =I. {411 An eneeay-ovel diagram for he hydrogen atom to THO0N SuHOR oF 29602 .ap.4 BOHR'S MODEL OF THE ATOM [Example 46. Calulte the binding eneray of the hydrogen atom the enray binding the tleston to the nucle) fom (19), Soveponding on tine Ths a wth Ze &= (Ga) (20 107 tom2joo 911 10°51 kg (140 10" cout Which wpress very well withthe experimentally abserved binding energy for hydeopen. [Next we calculate the frequency y of the electromagnetic radiation emitted when the electron makes transition from the quantum stale n, tothe quantum state ry, ‘that is, when an electron initially moving in an orbit characterized hy the quantura number 1 discontinuously changes its motion so that it moves in an orbit charac terized by quantum number ny Using Rohr's fourth postulate (4-13) and (4-18), we have Bote (Lame (t re GE (i) Tn (op tn terms of the reipracsl wavelength 1 “Ge cone (hh) oben a(S am ig ane) dre i, this is tind where n, and ny are integers “The essential predictions ofthe Bohr model are contained in (4-18) and (2-19) Let us first discuss the emission of electromagnetic radiation by a one-electron Bohr ator in terms of these equations 1. The normal state of the atom will be the state in which the electron has the lowest energie the sate = I, Tis iscalled the ground stave, (Ground state means fundamental slate, the term originating from the German word grund, meaning fandamental) 2. In an electric discharge, or in some other process, the atom receives energy due to collisions. ete. This means that the electron must make a transition to 4 state of higher energy, or excited state, in which > 1 1. Obeying the common tendency ofall physical systems, the atom will emit its ‘excess energy and return to the ground state, Ths i= accomplished by a teries of transitions in which the eleccon drops to excited states of successively lower energy. finally reaching the ground state. In cach transition electromagnetic radiation is emitted with a wavelength which depends on the energy lost by the electron, ie. on the inital an final quantum numbers. I atypical case, the electron might be excited ito state n= 7 and drop suocessvely through the states n= 4 and n=? 10 the round state n= I. Thice lines of the atomic spectrum are emitted with reciprocal wavelengths given hy (19) form = 7 and ny = 4. = 4and my = 2,and n;= 2 and nat 4. To the very large number of excitation and deexcitation processes which take plsce during a measurement of an atomic spectrum, all possible transitions occur nd the complete spectrum is emited. The reciprocal wavelengths, or wavelengths, (ofthe set of lines which constitute the spectrum are given by (4-19), where we allow ‘mand n; to take on all possible integral values subject only tothe restriction that For hydrogen (Z = 1) let us consider the subset of specteal lines which arises from transitions in which a, = 2. According to (4-19) the reciprocal wavelengths of these lines are given by k= RyCUnF — Wa) ny =2and n,m, = RUD — Ye) n=RAS6, This is identical with the series formula forthe Balmer series of the hydrogen spe trum (4-10) if R, is equal to Ry. According to the Bohr Model (aa) sare R Although the numerical valves of some ofthe quantities entering into this equation ‘were not very accurately known at the time, Bohr evaluated R., in terms of these {quantities and found thatthe resulting value was in quite good agreement with the experimental valve of Ry Inthe next section we shall make a detailed comparison. using recent data, between the experimental value of Ry and Bohr’ prediction, and wwe shall show that the two agree almost pete. ‘According to the Bohr model. each ofthe five known series ofthe hydrogen spec eum arts fom a subset of transitions in which the electron goes to certain final quantum stale n. For the Lyman series nj = I: forthe Balmer j= 2:for the Paschen nip = 3: for the Brackett n, = 4: and for the Pfund ny =. The fist three of these tries are conveniently illustrated in terms ofthe energy-Ivel diagram of Figuce 4-12. ‘The transition giving ris to a particular fin of series is indicated in this diagram by an arrow going from the initial quantum state m0 the final quantum state , ‘Only the arrows corresponding tothe fist Felines of each series and to the series limit are shown, Since the distance between any two enerey levels in sucha diagram is proportional to the diference between the energy ofthe two levels, and since (4-13) states thatthe Frequency + (or reciprocal wavelength) is proportional to the energy dlilference, the length of any arrow is proportional to the frequency (or reciprocal wavelength) forthe corresponding spectral line. The wavelengths of the lines ofall these series are ited very accurately by (419), by using the appropriate value of ny. This was a great aiumph for Bohr's mod! The success of the modi! was particularly impressive beeause the Lyman, Brackett, and Pfund series had not been discovered atthe time the madel wis developed by Bohr. The cristence of these series was predicted, and the series were soon found experimentally by the persons after whom they are named. “The mode! worked equally well when appli to the case of one-electron atoms ‘with Z=2. ie, sinly ionized helium atome He’. Such atoms can be produces 69 passing a particularly violent electri discharge (a spark) through normal helium gs, ‘They make their presence apparent by emitting a simpler spectrum than that emitted by normal helium atoms fa fact. the atomic spectrum of He is exactly the sme #8 the hydrogen spectrum except thatthe reciprocal wavelengths ofall the fines are almost exactly four times as great, This is explained very easily. in terms of the Bair model, by seting 2° = 4 in (4-19) The properties ofthe absorption spectrum of one-electron atoms are also easy to Understand in terms of the Bohr model, Since the atomic electron must have a Cota ‘energy exactly equal to the energy of ane of the allowed energy stats, the atom can nly absorb diserete amounts of energy from the incident sletromagneticcadiatio, This fact leads to the idea that we consider the incident radiation to be a beam of hotons, and that only those photons can be absorbed whose frequency is given by TRON SuHoR G4 298 SOLBOHR'S MODEL OF THE ATOM 108 conap.¢ zoo aia soy noo zolnd ad Teo 10 Bo ao "rao He For 12 Top th nea vl dara hyrogen wih he gaat sun or imei is crowded. tn betwoon the levels mariad: n't and nee” Sonor Toe Cosrespaniog spats lnas torn thee series inated inn each erase cpctal tng teow a ragaar pater, approaching tne aves ta to shortwave and ores ‘s_Aa drawn nee, neter the wavelength nor requoney scale nea. boing chosen ‘tney are moray tr clarty of vation ine waveleogh este would mare nea, feprecet the actual appearance o the potogapne piste Sane rom spectroscope. ‘hw riche an Plo cee, wich aver shown, tem he Ta tata pat ne = Iv, whete Eis one ofthe discrete amounts of energy which can be absorbed by the atom. The process of absorbing electromagnetic radiation is then jest the inverse of the normal emission process. and the lines of the absorption spectrum will have ‘exactly the sume wavelengths asthe lines of the emission spectrum. Normally the atom is alvays initially inthe ground state = I, that enly absorption processes from m= to n> 1 can occur, Thus, only the absorption lines which correspond {For hydrogen) to the Lyman series will normally be observed. However, if the gas containing the absorbing atoms is at very high temperate, then, owing to colisions, ome of the atoms will initially be in the fist excited state n = 2, and absorption Tines coresponding to the Balmer series will he observed Example 4-7. Fstimate rhe temperature ofa gas containing hydrogen atoms at which the Falmer sere inet wl be obwervd in the absorption spectrum. The Boltzmann probability distribution (ce Appendix) shows that the ratio ofthe rum horny of ators inthe it ctited state tothe number my of atoms in the proud st, lege simple thermal eum a temperature Tit ‘where k is Boltzmann's constant, k = 1.38 x 10 79 joule’ K = R62 x 10 *eV/'K. For we 411: Ey = —136€V, y= —33%V. Hence heefce sgnifieant ration of the hydrogen atoms wil aly bein the fs exci nly when Ts not foo much smaller than 0° Keand only when they abso fi State ean they praduce absorption lines of the Balmer series The stuation complicated by the fact tht the n= level ot ar ave the => 2 ee “his proximity makes the probability tha hydrogen stom wil intl be vn with ere temperature about ae rapidly a ths probity thatthe sions il nal Ten thf ested state But no absorption nes tales Be produced by val toni hydrogen toms, Dtiedcalelotions pri! that the maxim mount of Blot aPsory= tion should be observed when th temperature is aboat 10" Balmer atworpton lines are actualy bsersed i the hydrogen gas of some star stm sphere Thin gies ust may af estimating te emperatre ofthe sive fa stse ot 4-7 CORRECTION FOR FINITE NUCLEAR MASS Im the previous sestion we assumed the mass of the atomic nucleus (0 be infinitely large compared tothe mass of the atomic electron so that the nleus remains tved in space. This isa good approximation eve for hydrogen, which contains the lightes ‘nucleus, since the mass ofthat nucleus is wbout 2000 times larger than the electron mass. However the spectroscopic data are s0 very accurate that before we make a detailed numerical comparison of these data with the Bahr mode we musi fake into fccount the fet that the nuclear mass is actualy finite. In such a ease the eleetron and the nucleus move about theit common center wf mass. However. itis ot diffu foshow that in such a planetarylike system the electron moses relative 1 the mucleus as though the nuclus were fixed and the mas m ofthe elotron were slightly redaced forthe value the reduced mass ofthe system, The equations of mation ofthe wstert ic the same as those we have considered f we simply substitute form, where Sine is fess than m by factor 1/11 + m/M). Here SIs the mass ofthe nucleus To handle this situation Bohr modified his second postulate ta require that the total orbital angular momentum of heaton, Lisa inte Planks com stant divided by 2x. This 8 hieved by generalizing 120) perm oh Using 1 instead of oy in this equation itkes inte aevount the angular momentum of the nucleus as well as that of the electron. Making similar modifications to the Fst fof Bohrs derivation forthe cise of finite nuclear mass, we find that many of the {equations are identical with those derived before. except that he electron miss mis replaced by the reduced mass In particular, the formu forthe ree Tengths ofthe spectes ines becomes 1 ty w " “The quantity Ry is the Rodhery constant for « mucteus of mass M. AC SEIn—2 431 isappureni that Ry» the Rydberg constant for an flinitely heawy nucle whieh appears in (419). tn gencal the Rydberg constant Ry less than R., by the fair Tit + mAh. For the most extceme case of hydrogen, Mior = F836 and Ry ts les than R, by about one purt in 200. we evaluate Ry rom (422) using the currently accepted valves of th im, M-e.e-and A, we find Ry = 0968100 m_!. Comparing this with the evperimen {al value of Ry given in Section 4-4, we see that the Boh move carecte for finite nuclear muss, agrees with the spectroscopic daa to within Uhee parts in HBF so. pr 398 ‘Seyi HV3TONN auiNld Hos NOLO3HUOS106 BOHA'S MODEL OF THE ATOM ‘chap. 4 Example 48. In Chair 2 spo of he pain om” esr "stom const of poson ad "eSion evn shout oon eo hh ly Bey eMac str ees al so How wl eso pect ome ta cof the hydrogen stom? m= aa Pinner” mas Ataf the positon. wich quam he mas fhe clectron, Hence. the reduced mass (4-20) is “ - aM ont mee The eoresponding Reerg somtant Ry is according 10 4-22) The enetay state ofthe positonium stom then would e given bs RybeZ? Rh Eien = sand the reir wavelengths ofthe emit spel lines by Balt) The lrquencis ofthe emit ines would then be hall andthe wavelengths double that of a hudrogen atom twit infinitely heavy auckush. 7 being equal to one for postsonivm and Tengen Fl {hy What would he the electron separator nthe ground Ne obit Dinca we merely teplace m by j= m2 and we find tate the distance ofthe elton fom the “nucle ewie ss grat im the positronium stom atin the Bydragen atom with infnl heavy nucleus) Example 48. 4 miami atom contains a nucleus of charge Ze and u negative mum fig abot i The an eect pte wih charger ana man at 0 Times ts Iepeas ah lest mass {a9 Caleta the muon-nucleus separation, D, ofthe fit Bohr orbit of muonic stom wth 7 f Therese mass ofthe system, with my = 200 and A HES fo 4-20) 20m, 1836, 2m, + 1830, hen, ha (AION ih w= Z = sre 18n2 186 a ess 210 ma 2k FA ‘hewn he emu lg te ayn ace ta the dso ns do stor. IC 6 this Fate Which makes sch monic atoms intresting. information about hae propertion hing reveal Pom ther tay < {hy Casula the Ning energy ef a muonic atom with Z = £ DP FromiitMh with Z= hone Land m= n= IMne we have spin wING N60 = —25%0eV as the ground sate energy: Hence, the binding energy i 2530 eV, < what he maven the et nn the Laman ees or sucha tom? Prom (222, wth 2 = he we have wl iwimaR,, = I86R, . Hence For the fist Lyin line, n= 2 and ny =. In this a8 Ry tae (1-2)= 8 With Ry. = 109737 om! we obtain as6s8 so thatthe Lyman tines ein the say part ofthe spectrum. Xray techies ate necessity. terefore. to sady the spoctrum of moni atoms < Example 4-10, Ondinary hydrogen contains about one par in 600 of deverium. or easy hydrogen This sa hydronen stom whose mleus contains a proven and a neutron, How docs the doubled nuclear mae ast the atom spectrum? he spocteum would be hentia! it were nol for the correction fr Brite nuclear mass. Fora normal yogen atom Ry _ 106797 om : wT Cea) For an atom of heavy hydrogen or deuterium eR. teoritem soso ays (Ey ren : Hence, Ro iit large than Ry 80 tha the pete lines ofthe deterum aor ae shied tovsphly shorter wavelengths compares to hydrogen Indeed. deuterium was disoverad i 1932 by H.C Urey following the observation of hese shifted spectral ins. By increasing the coneentation af the heayy Kotope above is normal “aloe in hydrogen Uissharge tue, we now ean enfance the intensity a the deuterium fines Urhich ordinary, are dil fo detet. We then readily observe pis of yop lines the Shorter wevelnsih members of the pair sorepond exw to those prodited {fom Ry finer The resolution need cai ebvained, the Hele pir being separated by about T'S Afr example, sever thousand ties greater than the inimom resoahle separation 4 48 ATOMIC ENERGY STATES “The Bohr mode! predicts that the total enerey ofan atomic electron is quantized. For ‘example, (18) gives the allowed energy values for the electron in a one-cectron ftom. Although ve have not attempted to derive similar expressions fr the electrons ina multilectron atom, itis clear that according to the model the total eneray of cach ofthe electrons wil ako be quantized and, consequently, that the same must he Itve of the atoms total energy content, The Planck theory of blackbody radiation hnad also predicted that in the process of emission and absorption of radiation, the ‘toms inthe cavity wall behaved as though they had quantized energy states. Hence, faccording to the olf quantum theory every atom ean have only certain discretely separated energy states. Direct confirmation that the internal energy states of an atom are quantized came from a simple experiment performed by Franck and Hertz in 1914. The type of apparatus used by these investigators is indicated in Figure 412. Fleetrons are emitted thermally at low energy ftom the heated eathode C. They are accelerated to the anode A by a potential 1 applied between the two electrodes. Some of the eleetrons pass through holes in and travel (o plate P, providing their Kinetic enerey lupon leaving A is enough t0 overcame small etarding potential V, applied betseen P and A. The entire tube is filled at a low pressure with a gas’ or vapor of the toms to be investigated. The experiment involves measuring the clectron current SBLVAS AOUENE INGLY GF 298 LOETHE atom 108 hap. 4 BOHR'S MODEL OF eee Figure 4.13. Schematic of the apparatus used by Franck and Herz fo prove that atomic fnorgy sates are quantzod ceaching tna bythe ctrent owing through the meer as function of ‘he accelerating wltage The fist experiment wa performed with he ube containing He vapor, The nature of the results are indicated in Figure 4-4. At low sceserting voltage the curent {i observed to increase wth incressing vllage V- When ¥ reaches 49° ecurent sirup drs. This was interred as indian that some inaction between the cline and the Hg atoms soddely hens when the ckctrony aun a kinetic energy of 49. Apparent» icant acon telecon see et the He atoms and in xo doing entirely lose ht kinetic enra. It i only sgt more hun 49 Ve excanpross mast ocr i inn Fe aoe A and afer the proess he eactrone cannot gain enough Kneis energy in ling toward “overcome the retarding potent Hand each plate P At somenhat Ive, Vth electrons can ain enough Kncic ery afer excitation proces fo ove come Hand reich P The sharpness ofthe beak inthe curve idan tht cle trons of enoray les than 49 eV are notable to trarfr thei ney to an Hg atom Tris interpretation is convtem withthe existence of discrete energy sites for the Ha atom. ASsoming the fis exited sate of He tobe 49 6V biher in energy than the round sae an He atom woul simpy not beable to acep energy frm the bombarding eleians unless these electrons had a ast 49 eV Ho Figure 4-14 The vollage dependence of the current measured in the Franck Mertz Now, ifthe separation between the ground state and the first excited state is ac tually 49 eV, there should be a fine in the Hg emission spectrum corresponding to the atoms [oss of 49 eV in undergoing @ transition from the ist excited state to the round state. Franck and Hertz found that when the energy of the bombarding Becirons isles than 4 eV no spectral lines at all are emitted from the Hg vapor in the tube, and when the energy is not more than a few electron vols greater than this value only a single line is seen in the spectrum. This line is of wavelength 2536 A, which corresponds exacly 10 a photon energy of 49 eV “The Franck-Hertz experiment provided striking evidence for the quantization of the energy of atoms, It also provided a method for the direct measurement of the nergy diferences between the quantum states of an atom—the answers appear on the dial of a volimeter! When the curve of / versus V is extended to higher volt- ages, additional breaks are found. Some are due to electrons exciting the firs ex- {ted state of the atoms on several separate occasions in their trip from C 10 A: but Some are due to excitation of the higher excited states and. from the postion of these breaks, the energy differences between the higher excited states and the ground state can be directly measured "Another experimental method of determining the separations between the energy states of an atom # to measure its atomic specieum and then empirically to construct Sct of energy states which would lead to such a spectrum. tn practice this is often {uite dlfcult to do since the set of ines constituting the spectrum. a8 well asthe set Of energy states, is often very complicated: however in common with all spectro- Seopic techniques, itis avery accurate method, In all cases ia which determinations ofthe separations between the energy states ofa certain atom have been made. using both this technique and the Franck-Hertz technique, the results have been found to be in excelent agreement In order to illustrate the preceding discussion, we show in Figure 4-15 a con: siderably simplified representation of the energy states of Hg in terms of an energy~ level diagram. The seperations between the ground state and the first and second excited states are known, from the Franck-Hertz experiment, to be 49 eV and 6.7 eV. These numbers ean be confirmed, and in Fact determined with much higher accuracy by measuring the wavelengths ofthe two specteal lines corresponding to transitions ‘olan electron in the Hg atom fram these (o slates tothe ground state. The enerEy = —104eV. of the ground state wative to a state of 7ero total energy. #8 not ‘determined by the Franck-Hertz experiment. However, it can be found by measuring the wavelength of the line corresponding to @ transition of an atomic electron from eT ol eseet seo) Jott Figure 4.35 considerably simpli energytevl diagram for mercury. Lying above the figs! discrete energy level al € = 0 fea continu of levels, eo. SaLVIS ADHEN3 OLY 94 296state of zero total energy 10 the ground state, This isthe series limit of the series terminating on the ground state. The energy & can also be measured by measuring the energy which must be supplied to an Hg atom in order to send one of is clectrons from the ground state to state of ra total energy, Since an electron of zero total energy is no longer bound to the atom, A is the energy required 10 ionize the atom and is therefore called the jonizarion energy: Lying above the highest discrete state at F = 0 are the energy states ofthe system ‘consisting of an unbound electron plus an fonized Hg stom. The total energy of an unbound electro (aie electron with E> 0} is ot quantized, Thus any enerey E> 0 is possible fr the electron, and the energy states form a continuum. The electron can ba JinE and om VET be excited from its ground state to a cantinowm state the He atom receives an en a exay greater than 1014 eV. Conversely iti possible for an jonized Ng atom to capture 4 free electron into one of the quantized energy states ofthe neutral atom. In this, process, radiation of frequency greater than the series limit corresponding to that uated fr te eatin einen ‘The quantization intra fps is mont easy esas for Tho thats imposed th qunion ve considers poms terpctation. The ‘Galion team p and «i the equation ofan lips, Any instantaneous ste a ‘notion af th ovate epresented by sme plat aa pot of this equation on two-dimensional space having coorlnac pad x. We il sch a spas (he fo i plane) phase space. and the plot is a phase diagram of the linear oscillator, shown D Figure’ During one oe of entation the print preening he pston and ‘Damertum tepals one around the cIgee.Thesomise and hehe Site rates ret are een, by comparison ath our equation. er 205 [Now the area ofan elipse is rah. Furthermore, the value ofthe integral § 7d is just equal to that area (To see this note thatthe integral over a complete osilation hap 4 BOMR'S MODELS OF THE ATOM ‘equals an integral in which the representative point travels from <= —a to.x = +a State will be emitted. The exact value of the frequency depends on the intial energy ‘over the upper half of the lips plus an integral in which the point travels hack Co E of the fee electron. Since # can have any value, the spectrum of He should have Se a over the lower half. In the frst integral both p, and dv are postive und its 4 continuum extending beyond every series limit in the direction of increasing fre- ‘quency. This ean actually be seen experimentally, although with some difficulty. These comments concerning the continuum of energy states for > 0, and its consequences, hhave been made in reference to the He ator, but ‘S37 NOLLVZILNYNO 3H 40 NOILYA 4-9 INTERPRETATION OF THE QUANTIZATION RULES | tm our case The suocess of the Bohr model, as measured by its agreement with experiment, was i certainly very striking: but it only accentuated the mysterious nature of the postulates i fn which the model was based, One ofthe biggest mysteries was the question ofthe | Felation between Bohr’ quantization of the angular momentum of an electron mov- | | ing ina circular orbit and Planck's quantization of the foal energy of an catty, such as an electron, executing simple harmonic motion, n 1916 some light was shed ton this by Wilson and Sommerfeld, who enunciated a set of rules for the quan: | lization of any physical system for which the coordinates ute periodic functions of time. These rules included both the Planck and the Boht quantization as special cases. They were aso of considerable use in broadening the range of applicability of the quantum theory. These rules can be stated as follows For anv physical system in which the cvordinates are periodic functions of time. there exists a quamtum condizon for each coordinate, These quatum conditions are hose where is ne of the coordinates isthe momentum associated with that coordinate 2ngisa quantum number sshich takes om integral values aad § means tht the integration taken orer one periad of the coordinate q The meaning of these rules ean best be iMustrated in terms of some specific ex amples. Consider a onesdimensional simple harmonic oscillator. Its fatal energy can he written, in terms of position and momentum, 3 A (423) iagram of tho motion ofthe representative pot fora ‘eoresered by lipes tose reas phase space ae given by mh The sce between ‘tlocen!slipaes or example he shacod area) hasan area Figure 416 Top: A phase spaz but iim = anv where vi the Frequency of the osilation, so that §nar= Bm we now use (4:23), the Wilson-Sommerfeld quantization rule, we have Pred Et nh nh Em nhy which i identical with Planck's quantization law Not thatthe allowed states of osiltion are represented by a series of eligses in phase space, the area enclosed between sueesive elipses always Being (ace Figure #16) Again We Find thatthe easieal sitontion corresponds ta h-~ 0 al vals of and hence all pe being allowed if that were tr. The quantity § pf sometimes called a phase intcgran ‘asia physics ic he integral ofthe dynamical quantity caled the ation over ane oil tion ofthe motion. Hence the Planck energy quantiraion i equivalent tothe quam cation of We can also deduce the Bohr quantization of angular momentum from the Wilson Sommerfeld rule. (4-23). An electron moving in a citeular orbit of radius r as an angular momentum, mer = 1, whichis constant. The angular coordinate is 0, which 's periodic function ofthe time. That is versus #8 a s4-tooth function, increasing linearly rom zero to 2x rad in one period and repeating this pattern in each sueczeding period. The quantization rule Grae Bom hocomes in this ease so that L=nhian enh which is identical with Bob's quantization law ‘A-more physical interpretation of the Babr quantization rule was given in 1924 by be Broglie. The Bohe quantization of angular momentum can be writen asin (415) er = pr = nhs man where pis the linear momentum ofan electron in an allowed orbit of radius». If we Substitute into this equation the expression for p in terms ofthe corresponding de Wotie wavelength ‘he Rohr equation becomes Lake. 28) ‘us te allowed orbits are those in which the circunference ofthe orit can contain slam integral numberof de Broglie wacelengths Timagine the electron to be moving ina ereular orbit at constant speed, with the sssncated wave following the electron, The wave, of vavelenath 4 1s thea wrapped ‘epeately around the eteular orbit. The exultant wave that is produced will have ‘era intensity at any point unless the wave at each traversal is exactly in phase a that pint withthe wave in other traversal. If the waves fn each traversal are exactly in fphase. they join on perfectly in orbits that accommodate integral numbers of de Foplic wavelengths as illusirated in Figure 4-17. But the condition that this happens ‘s just the condition that (4-24) he satisfied. If this equation were violated, then in a laste mumber of traversals the waves would interfere with each other in such a way that ther average intensity would be zero. Since the average intensity ofthe waves. “V5 is supposed to be a measure of where the particle is located. we interpret this as ‘mening that an electron eannot be found in such an orbit, This wave pictare gives no suggestion of progressive motion, Rather, it suggests standing waves, asin a stretched string ofa given length, Ina stretched string only Certain wavelengths, or Frequencies of vibration, are permitted. Once such modes are feted, the vibration goes on indefinitely if there is no damping. To get standing lvaves, however, we need oppositely directed traveling waves of equal amplitude. For the atom this requirement Is presomably satisfied by the fact that the electron can leaverse an orbit in ether dieection and still have the magnitude of angular momentum required by Bohr. The de Broglie standing wave interpretation, ilustrated in Figure 417, therefore provides a satisfying basis for Bohr’s quantization rule and, Tor this case, ofthe more general Wilson-Sommerfeld le ‘Thee is another example of a system in which th origin of the Wilson-Sommerfeld “quantization rule ean be understood ia terms of the requirement thatthe de Brodie Figure 417_ilustrating standing de Grooie waves set up Inthe fist ties Bot omits The locations of he nodes ean. of course, be found anywhere on each ait provided that thei spacings are as shown HL 4ONOMWLSHAMBLNI OF 29S CLL zing NowwzuNyno4 1p. 4 BONR'S MODEL OF THE ATOM ares associated with «pa Je underqoing perindic mation form a set of standing wares. Consiter a particle which moves fcely along the « axis from x= —a/2 to = 4/2. ut which dacs not penetrate into the regions outside these limits This system can be thought of as representing approsimately the mation of 3 conduction electron in a one-dimensional piece of metal that extends from —a/? to 4u'2, The particle hounces ack and forth between the ends of the region with momentum hat changes sign at each ounce, bat maintains eanstant magnitude p. So the Wilkon-Sommerfeld equation reads $ret haw 428) ut hip i just the de Brose wavelength of the particle, so we have Thasan integral number of de Broglie wavelengths just ts into the distance covered hy the particle in ane traversal of the region, and this allows the waves associated With successive traversal to hein phase and 50 st up a standing wave We shall se in the following chapters that the properties of standing waves ane equally important in the qusatization conditions of Schroedinger's quantum me stiomies. And the timesindependent Features of the standing wave associated with am eleteon in the ground state ofan atom will make it possible to understand in a simple way the fendamental guestion of why the eleceon does ot emit electromag rete radiation and spiral into the muck, 410 SOMMERFELD'S MODEL ‘One ofthe important applications ofthe Wilson-Sommerfeld quantization rule isto the case of a hydrogen atom in which it was assumed that the electron could move in eliprcat orbits. This was done by Sommerfeld in an attempt to explain the fine stra ture of the hyalrogen spectrum, The fine structure i a splitting ofthe spectral lines, inca several distinct components, which is fou in all atomic spectra H can he ah served only hy using equipment of very high resolution since the separation. in terms of reciprocal wavelength, hetweee adjacent components ofa single spectral line is of the order of 10 + times the separation betwen adjacent lines. According to the one model, this must mean that what we had thought was a Single enerey state of the hydrogen ator actually consists of several states which are very elene together Sommerfeld fst evaluated the size and shape of the allowed elliptical orbits, as wll asthe total energy of an cletron ming in such an orbit using the formulas of classical mechanies, Deseribing the motion in term ofthe polar coordinates and he applied the swe guar extions fri fos ire he fist coniton yields the samme Yesrition om the ofbital angular momentum t 1233, {hat it does forthe eireulr arhit theory. The second condition (which was not appl Ne in the Tong eave of purely circular webs) leads to the following elation ud o between Land a/b, the rato ofthe semimajor axis to the semiminor axis ofthe elipse Lai — y= mb n= 0.12.3% By applying the condition of mechanical stability analogous to (4-14). third equ tign i obtained. From these equations Sonmer‘eld evaluated the semimajor and ‘semiminor axes 4 and b which give the size and shape of the eliptial orbits. and also the total energy £ of an electron in such an orbit. The results are eon? ian ma ale 14-266) 8a) SF 1369 Where jis the reduoed mass of the eleetron. and where the quantum number 1 is defined by Since ny 1.2.3,. nd y= 0,42, Ao o5 can EaRE on the values ne 1234, Fora given vue of nn cn assume only he valves “The neg nis lle the pric quanta mmber, nd Melle the cima ‘uation (20) shows thatthe shape ofthe ori the ratio the semimajor tthe semimino sus dtermined by the ato yon For y= 1 the obits are ces Frain Note that te eguation giving in terms 9 Meni with (19) the Squitionaivng the rads ofthe ere Bot orbits, Remember that 46) wil have telaced hy if proper ncoant fs ken ofthe Brite mosear mass) Fire {Ptashne th psu oi rego te tre vale fhe inal quantum numer Corresponding teach tale of the principal quantum unter nike are tet allowed ete One of ths. he cela ost thr oret decribed bythe oviinl Bohr mode. The offers are elipieal. But despite {hesery diferent path flloned by an let moving the diferent possible orbits thea given tc) ells stat te ttl energy ofthe lesron ithe sme The ta Sere ofthe ccctron deponts only on m The several vfs chaaceized hy 0 Sermon sale of ate silt be deynerate The energie dire! sates ofmt00 the sme ttl ene. “degenerae” Figure 4-18 Some elipical Boh-Som the common focus of the olpses, ind 200m ScTRSHINNOS OL 29S GLE116 Conap. 4 BOHA'S MODEL OF THE ATOM ‘This degeneracy inthe total energy ofan electran, following the orbits of very di ferent shape but common ns the result of a very delicate bulance between potential land kinetic energy, which is characteristic of treating the inverse square Coulomb force by the methods of classical mechanics, Exell the same phenomenon is found in planetary o satellite motion, which is governed by he inverse squre gravitational force. For instance. a satelite may be launched into any one of whole family of elliptical orbits. all of whieh correspond to the sime lotal energy and have the same semimajor axis. OF course there is eflectvely no quantization of the ot in these macroscopic eases, but as fas degen nalogous to the cas of hydrogen atom Sommerfeld “removed the degeneracy” in the hydrogen atom by next trating the problem relaeistcully: In the discussion fllowing (417) we showed that. for an electron ins hydrogen atom, re = 10. * or les. Thus we would expect the eltivisic cerrections to the total energy. de to the relativistic variation of the electron mass which wil be of the order of ty 0 be only of the erder of 10“: however, this is just the order of magnitude ofthe splitting in the energy states of hydrogen that would he needed to explain the fine structure ofthe hydragcn spectrum. The actual size of the correction depends on the average velocity of th electron which, in (urn. depends fon the elipiity of the orbit Alter a ealeulation which is much too tedious te [produce hete, Sommerfeld showed that the total energy of an cletron in an seb jhtracerized by the quantum aumbers 9 and iy egal (O ae a eats ota aa EIT Oa parameters xy is concerned they are completely 1 (27h In Figure 419 we represent the frst few energy states of the hydrogen atom in terms of an energyctevel diagram. The separation hetwoen the several level with common value of has Be ee vrata for the sike of cavity, Arrows in dicate transitions between the various energy states which produce the lines of the alomie spectrum. Lines cortesponding to the transitions represented by the solid arrows are olwerved in the hydrogen spectrum. The wavelengths ofthese lines are in very good agreement with the predictions derived from (4-279, However. the lines corresponding to the transitions represented by dashed arrows in Figure 419 are not found inthe speteum. The transitions conesrned Jo not take place. Inspection of the figure will demonstrate that transitions only accu if iy tay = £1 (28) no hand 2 Figure 4-19 The Fne-siructre sphiting of some eneray levels of he hysragen str The Slings reat exaggerated Traneions mien pranecesbseres los of he hyarogon Spectrum ae inciateg by sod arrows a “This is called selection rule. I selects from all the transitions those that actually 4:11. THE CORRESPONDENCE PRINCIPLE justification of selection rules could sometimes be found with the aid of an auxiliary postulate known as the correspinuence principle. This principle, enunciated by Bohr fn 1923, consists of two parts 1 The predictions of dhe quan theors forthe Behavior of any physical system must conrespand to the prediction of classical physies the fit in whieh the quanturn tumivers specif the state ofthe sestem become very large A sletion rule hols true eer he entire range ofthe quanta naraber concerned. “Thus any selection rides whic are necessar to abiain the required correspondence i the classical lint (large mals apply inthe quasum lini (mall ‘Concerning the first part. it fs obvious that the quantum theory must correspond to the classical theory in the limit in whieh the system behaves classically. The ony ‘question is Where is the classical limit? Bohs assumption is that the classical limit Js alvays to be found in the imi of arge quantam nombers. In making ths asump- tion he was guided by certain evidence available atthe time. For instance, th classical Rayleigh-feans theory of the blackbody spectrum agrees with experiment in the limit ‘of sinall x. Since Planck's quantum theory agrees with experiment everywhere, We ‘ee thal correspondence hetween the quantuay and classical theories is Found, in this casein te limit of small, But it is easy to se thal as ¥ bocomes small the average Value fof the quantum number specifying the energy state of blackbody eleciro- imagnetic waves of fequcney twill come large. (Since = why. we have = ay But as v0, KT, sin in this limit r= AT. whieh isa constant. Thus 6 35 ‘= inthe classiea! mit, Note also that if we fe inthe relation aly = AT" const and take f= 0 9s we freqently have in considering the classical mit, we again find Ji» / in that limit) The ssoond part of the correspondence principle was purely ‘in assumption, but certainly a reasonable one. ‘Let us lustrate the correspondence principle by applying it to a simple harmonic osellator seh as a pendula oscillating a fequeney «, One prediction of quantum theory for this system is that the allowed energy states are given by E = nv. In the discussion in Chapter 1 we sw than the limit of large this prediction isnot in disagreement with what we actually know about the energy states ofa elascal pen- {dulum. Ia this case of simple harmonic osilater. the quantum and classical theories do correspon! for n+ 1 insofar as the energy slates are concerned. Next assume that fe pendulum hob carries a eletrc charge so that we ean compare the predic~ tions ofthe to theoris concerning the emission and absorption of electromagnetic radittion by such & sytem, Classically the system would emit radiation de to the ‘overated motion of the charge, and the frequency of the emited radiation would be evactly +. According to the quantum physies, radiation is emitted as a result of the sjstem making a transition from quantum state w, 0 quantum state. The ‘energy emitted in suc a transition is equal to E, — = n ~ i. This energy is farsi away by a photon of frequency (Ey ~ E,W — myles Thus, in order to ‘obtain correspondence helwccn the classical ané quantum predictions of the fre ‘quency of the emited radiation, we must require that the selection rule m, — y= fe vali in the classes imi of large n. A siniar aegument concerning the absorption ‘of radiation by the charged pendulum shows that in th classical lint there ix also the possibilty of 4 transition in which m, ~ uy © ~L. The validity of these selection rules in the quantum lini of small mc he lest by investigating the spectrum of radiation emitted by « vibrating diatomic molecule. The vibrational energy states for such a system are just those of a simple harmonic oseilator since the force which ie 295 uth “sYatoNIs FONBENOASIHHOD AHL8 Conap, 4 BONR'S MODEL OF THE ATOM Table 42_ The Correspondence Principle for Hysrogen ” ‘e , * Bilerence 5 52x10 ax 10 1 esr 7x 00 asm 10" gar? x 108 1s Lom 68719 5 1065878 x 108 ats Jom 6577910" -gTH9 «10? nis leads tothe equilibrium separation of the two atoms has the same form as a harmonic «tring fore From he ibyational spectra can be semi tha the ston rule m)— ny +1 actully is in oporation in the mit of small quan in agreement withthe cond pat ofthe corespondenee princi. -aditumber of other section rales wee covered empirical nthe analyst of ‘omic and molecular spectra. Sometimes, but not always, it was possible to under stand these selection rues in terms of a correspondence principle argument im numbers, Example 11. Appin ieconerondene rn toy ogn lmao thet >The tency flat y of an cto ina Bare flo fo 16 an 17 (gy ma According to cassal pies he fegvency othe Hh emi in sch va, the frequency of revolution. aD “Quantum phys petits that the requeney ofthe emit ight i, from 4-19) es ive-(e) Sel ce) tarlazin (oe) Sole Css) tn Lie it a? _appecuches 217 so that ¥ > ¥p as m+ " He sau bashes above i rpndence or tage n « Is nrucive fo mot tha although both altho both prs ofthe corespondence principle gtr where rh mmo utr es rt apecs with experiment inthe hydrogen atom consiered inthe preceding ee frie For cperiner shows athena en I hh wa er) to sai) the fst prt ofthe pnp for Lrg m dose Hor apply to the hydopen atmo oa Trantion ae or tsb ate ow ch ihe aunt non ir na yeh one Ti at hat i i en ye bemade to gre with experiment, however 4-12 A CRITIQUE OF THE OLD QUANTUM THEORY Inte pt or hes a ve diced ne he elon wih 1 mote gtum mn Ts pment tow fel Yo 5 the ol nt thor. nan espete titer) ever sce cen ore tha rns be apparent to the seen feeuse me hae not entoned numberof sce ful applications of the old quantum theory to phenomena. such asthe heat capacity Df sols at tow temperature, which were nexpliable in terms of the classical theo: fies However, the old quantum theory certainly was not free of erties. To com pete our discussion of this theory we must indicate some of is undesirable aspects: T The theory only tells us how to treat systems which are periodic. by using the ‘Wilson-Sommerfld quantization rules, but there are many systems of physical interest which re nol periodic. And the number of periodic systems for which & Fiysicl basis of these rules ean he found in the de Broglie relation i very small y atthough the theory does tll us how to calculate the energies of the allowed states of certain systems, and the frequency ofthe photons emitted or absorbed when Siyetemn makes a transition between allowed states, it docs not tell us how to Sattalte the rare at which such transitions take place. For example, it does not tol fo how to caleuate the intensities of spectral lines. And we fave seen that the theory cannot always fell us even shih transitions actwally are observed to occur and whieh are not shen applied to atoms, the theory i really only successful for one-electron atoms The alkali elements (Li, Na, K, Rb, Cs! can be treated approximately, but Sly because they are in many respects similar to & one-electron atom, The theory fl badly even when applied fo the neutral He atom, which contains only two clecteons. +t pinally we might mention the subjective eritcism that the entire theory seems somehow to lack coberence~to be intelecally unsatisfing “That some af these objections are really of very fundamental nature was realized by everyone concerned. and much effort was expended in attempts to develop Guuntuin theory which would be free ofthese and other objections. The effort was Gin rewarded, In 1925 Erwin Schroedinger developed his theory of quantum me cranks Although itis 4 generalization ofthe de Broglie postulate, the Sehroedinger theory isin some respects very diferent from the old quantum theory, For instance the prture of atomic structure provided by quantum mechanics i the antithesis of the picture, used in the ld. quantum theory. of electrons moving in welefined eae Nesertheles, the old quantum theory i sil frequently employed as a first crpreximation to the more accurate description of quantum phenemena provided sy quantum mechanics. The reasons are thatthe ol quantum theory i often capable Er Sting mumerealy correct reslis with mathematical procedures which are con Cideebiy tes complicated than those used in quantum mechanics, and that the old seetture theory item helpful in visualizing processes which ae dificult to visualize ii terme af the rather absteact language of quantum mechanics QUESTIONS, na colon between un x particle and an escteon, what genera considerations tiit th moment renter? Doe theft thatthe Force bs Colm aya role im ts 22 How does the Thomson atom die from & andom dsthation of protons and ecto in aahencal ego? List nbjetions to the Thomson mode of the atom, 3k Why do me epcy tha the il be thin in exerimans intend to check the Rutherford scattering formals” |The scattering a x partes at very small anges digress with the Rutherford format for such ames. Eapain 6. ow does the destction of 431, which giver the taccory of «partite moving under THe etree apulive ivese square Couloe force. ifr from the dedtion of swoussno ttPART 2 STATISTICAL PHYSICS 10 Classical Statistical Mechanics ~ 11 Thermodynamics 12 Thermal Properties of Gases 13 Quantum Statistics432 In the preceding chapters of this text we have studied the fundamental constit- uusats of mistter: particles, nuclei, atoms, and molecules. But we cannot see or feel individual atoms or moleoules st work. Rather we observe the result of a large number of them acting in a more or less organized manner; ie., macroscopic processes. ‘The properties of matter in bull (called macroceopie properties), as we ordinarily observe them, are the result of these collective eetions. ‘The collective behavior of large numbers of atoms and moleoules is basically » result of thelr sleetromagnetic interaction, since gravitational interaction plays only a minor role aud the strong and weak interactions affect only nuclear processes. Familiar Processes, euch as melting and vaporizatior, diffusion, thermal and electrical conduction, emission of electrons by hot metals, and many others, as well as such concepts as temperature, heat capacity, heat of change of phase, eto, fall ia this category of colleetive properties. It is the understanding and control of theoe collective phenomena that primarily interest the applied physicist, the engineer, ‘he chemist, and other ecientists in their endesvor to use the forces of nature for the welfare of rhan. To deseribe processes involving a very large number of partieles, special methods must be devised. These methods are, by necessity, of a statistical nature, since ‘we cannot take the detailed motion of individual particles into account (or is it necessary to make such an effort to obtein results of practical value). For this reason the discussion of macroscopic provsasee from the molecular point of view constitutes what is called satiatical physics. The word “statitical” must be inter preted as applying to a technique for describing processes involving large numbers of particles whose individual interactions are known, without considering the individual behavior of each particle, In this respect, statistical methods have been. applied to the discussion of many-loctron atoms and many-nucleon nuclei, In this part of the text we shall first develop the statistical method known as classical slatistcal mechanics, discuss its general applications to what is called thermodynamics, and apply the method to analyze the properties of gases, We shall conclude this part with a brief introduetion to quantum statistical mechanice, illustrating some of its applications, 10 CLASSICAL STATISTICAL MECHANICS 10.1 Introduction 10.2 Statistical Equilibrium 10.3 The Maxwell-Boltzmann Distribution Law 10.4 Temperature 10.5. Thermal Equilibrium 10.6 Application: to the Ideal Gas‘=P ~Lubie centimeter of a gas at STP there are about 3 x 10! molecules, It (436) Clasical statistical mechanic (102 (44) Gls lista mechanics 10.1 Introduction ‘Mechanics is founded on certain general principles, such ss the conservation of energy and momentum, that are applicable to the motion of interacting particles. In this chapter we shall extend the principles of mechanios to systems of meny. particles, emphasizing the methods used to obtain collective or masroscopie prop- 102) Statistical equilibrium 435 ‘This expression for the total energy of the system implicitly assumes that_the particles are noninteracting (or that they interact only slightly), so that to each pow bute an energy depending only on the chondinates of the Wa teractions, we must add to Hq. (10.2) terms of the form ade particle we can, ile. If we oor ‘erties of the system, without considering the detailed motion of each particle. This technique is called slatzetieal me "We use “particle” here in a broad sense, ‘meaning a fundamental particle, such a an electron, or an aggregate of funda. ‘mental particles, such as an atom or molecule. Thus a “particle” will be each of the welldefined and stable units composing a given physical system, ‘The fect that we need a statistical approach when we are dealing with the ‘macroscopic properties of matter is easily recognized when we note that in one ‘not aly practically impossible, but also unnecessary to take into aeeount the motions of each of these molecules in detail to determine the bulk properties of the gas, ‘such as its pressure or temperature, On the other hand, to make a statistical analysis of @ many-particle system, we have to make some reasonable estimate about the dynam i. particle based.on. the general_properties of the. artisles. ‘We make this estimate by introducing the concept of the probability of distribution of the particles among the different dynamical states in which they ‘may be found. When we introduce the idea of probability, this does not imply ‘that we assume that the particles move randomly or in a chaotic way, without obeying any well-defined laws. The concept of probability arises from our method of estimating the dynamical states of the particles in a system, not from the mechanism by which, as a result of their interactions, the particles of a system are distributed in nature among the possible dynamical states. Henee the validity of the statistical analysis of a many-particle system is directly related to the validity of our assumptions concerning the probability distribution of the particles. 10.2) Statistical Equilibriam Consider an isolated system composed of a large number N’ of particles, ia which ‘ach partiloHas avaliable tot eoverl states With soon By Bac Be ‘energy states may be quantized (as i ahtat molecule) oF they may form a practically continuous spectrum (as for the trans- “ional Wnelic energy of the moleoules in a gas). At a particular time the particles are distributed ainong the diferent states, so thst n, particles have engrxy Ex:na particles have energy Bs; and so on. ‘The total number of particles is =Sme | io.) Eyis + Boia +--+ Eyag 7 cofresponding to the pot ‘etion between pairs of paiticlas, Each term includes the coordinates of both interacting particles. In such a ease we cannot speak of the energy of each particle, but only of the system. ‘Tt may seem at first sight that our discussion is therefore unrealistic, since all particles that make up physical systems are interacting. However, under special conditions we can use a technique called the self-consistint field, in which each article is considered subject. to the averaze interaction of the others, with an average potential energy which depends only on its eoordinates, so that we ean still write U as in Eq. (10.2), but now E; = Eys-+ By ave. For eases in which the interactions among the particles must be considered explicitly, other techniques inust be used. We shall disouss these in Chapter 12 in connection with real gases. If the system is isolated, the total energy U must be constant, However, as it of their mutual interaati isk but les among the available energy states may be changing. For example, in a gas a fast molecule may collide with a slow one; after the collision the fast molecule may. have slowed down and the slow one may have sped up. Or an excited atom may collide inelastically with another atom, with a transfer of its excitation energy into kinetic energy of both atoms, Hence, in both examples, the particles after the collision ae in diferent states. In other words, the numbers tay. sigh {ive the partition (or distribution) of the N particles among the available energy States, may be changlog. Tt Ts reasonable to ascume that for each macrsenpie state of a system of particles there is a partition which is more favored than any ther. In other word, we may say Chat ginen alata conditions of the sytem articles (that is, the number of particles, the total energy, and the structure SESE diatamsaebae nein, Whee Oi setiees ects the svstem is said to be i ala rs ‘Aavetem in statistical equilibrium will not depast from the most probable partition (except for statistical fluctuations) unless it is disturbed by_an extemal action. By this we mean that the partition numbers m,, na, m3, ... may fluctuate. around the values corresponding to the most probable partition without no- ticeabe (or observable) macroscopic effects, For example, suppoge that ve have & gas in statistical equibrium, in which a molecule of energy By collides with molecule of energy E;; after the collisions their energies are E, and E,. We may assume that within a short time another pair of molecules is removed from energy states E, and E, and the same or another pair of molecules is moved into energy states By and Ey, so that, statistically, the partition has not changed, ‘The key problem of statistical mechanics is to find the most probable partition PQ: (10.2) (or distribution law) of an isolated system, given its composition. Once the most — probable partition has been found, the next problem is to devise methods for de- a (U = miBy + mgBo + maBa tes = Do midi. |436 Clasical statistical mechanies wos riving the macroscopically observed properties from it, To obtain the distribution lav, certain assumptions are required. One may try several plausible assumptions, mt a distribution law in accordance with experimental results is obtained. "Theo distribution las or statistis are presently used. One is called the Mazwell ‘Goleman disirbuion lav, which i the basis of dlasbical slatintice, We shall study 5 i sa pee gE ORS RGIRRGES OED nEEOUOEEERSOaSEEOOSEO . Giassioal statistical mechanios was developed in the last part of the nineteenth, century and the beginning of the twentieth century as the result of the work of Tndwig Boltzmann (1844-1906), James C. Maxwell (1881-1879), and Josish W. Gibbs (1899-1003), Classical statistical mechanics has a very broad applicability, especially to the discussion of many properties of gases. ‘The remaining two dis tribution laws, called Permi-Dirac and Bloee-Binslein belong to quantum saliales seat ie catered ee Chasior13--Chasial stata mechanics e436 CO5- Sidered as a limiting value of the two quantum statistics (see Section 13.9). f0.3\The Maxwell-Boltzmann Distribution Late AC considera system composed ofa lange numberof identica! and di le “particles, By identical particles, we mean that the particlesyhave the same struc- “are and composition. By distinguishable particles, we mean that we can distinguish, or tell the difference, betwen one particle and another identical partiole. 'At frst sight it seems that there is a contradiction between identical and distin- fuichable, end indeed ths isthe case! Later on we shall reconsider this spparent Tack of loge; however, the results we shall obtain now are suficiently simple to justify a preliminary discussion of such a system. &y—_—_— Bg 0 tt kt Fig, 10-1. Distsibution of particles TF among diferent encrey states, hn a z a3 rs 4 us representa particular partition ny, na, 75... by the geometreal arrange ment shown in Fig. 10-1, Each line represents e particular slate of enorgy By and {he number of dots indicates the number n, of particles in each energy state. In our example n; = 3, na = 0,3 = 2, na = 1, ete ‘Our frst ascumption is thet all energy states are equally accesible to the par. ticles of the svstem: Le, that all eneray slates have the same probability of being 403) ‘The Mazuell-Bollzmann distribution low 437 _sccupiedThen we shall assume that different waye in which the particles can be distributed among the asail. cable energy states to produce the partition. Let us then se, as an example, the number of different ways in which the partition of Fig. 10-1 canbe obtained. By diferent we enn that the numbers my nya. tue fied, but the particles in each state are different. ‘That i, a partition in whickC, particle a isin atate £, and particle p is in state B is considered different from partition in which particle ais in state By and particle pis in state By. This is 0 consequence of assuming thst the component particles are distinguishable. I they” were indistinguishable, we would have to assume that the two partitions were the same. ‘Tostart filing the state Ry us may select the frst particle from any one of th ‘pasties avalbble, ‘Then ther ae distinguishable wave of selating he et particle. The second particle canbe selected in N —1 diferent ways, since only N ~ 1 particles are left available. And the third particle can be selected in, — 2 different ways, So the totel number of different distinguishable ways in which we may select the frst three particles to place in state Ey is NW — NW - 2) (10.3) Let us designate the three chosen particles by a, b, and o, We may pick them in any of the 3! = 6 different orders or permutations abe, boa, ab, bac, acb, cba But these 31 diferent orders of placing particles a, 2, and c in Ey give the same action, sine hry ll corespond to parts andi state By pation is determined only by the number and labeling of particles in each state and not by the order in which they were placed there, which is immaterial once the partition is arranged. Therefore, to obtain the total number of distinguishable different, ways in which we may select the first three particles in state E1, we must divide Eq, (10.3) by 3, resulting in ais: aw 3 ‘The general expression for the total number of distinguishable different ways of plscing’ny particles in state 2 is then es wt aaa | a0.) ‘This is just thé number of permutations of N objects, taken ny ata time. When we pass to the second state B, only NV — n, particles are available. So if we want to Place np particles in state E, we must use Eq. (10.4) with N replaced by N — m438 Classica statistical mechanies (wos and n; replaced by nz, resulting in (10.5) For the third state we have only N’ — n; — ms particles left, and we are placing ‘ng particles in it, so that instead of Eq. (10.4) we get mal ‘The process can be continued until all energy states have been considered. ‘The total number of distinguishable different ways of obtaining the partition m, na ns, cT7is objained by multiplying expressions (10.4), (10.8), and the successive ex- pressfons for all remaining energy states. Tn doing this the student may note that number of factors, such as (N — n,)!, (V —m — nz)! ete., cancel out, resulting in the simplified expression ___ | wm; \r matmalnal (10.6) which gives‘the number of different distinguishable ways in which the partition ‘hy, Ra, May .-. can be obtained. We shall now assume that the probability of obtaining thst partition is proportional to P. For example, for the partition of Fig. 10-1, the probability P is ‘We must take O! equal to 1, since n; = 0 can be chosen in only one way. ‘The numbers nj, ng, ng, ..- in Eq. (10.6) must be chosen in such a way that Eqs, (10.1) aoe ee rave assumed that all aval slates have the same probability of Pokal ia being oooupied. ILmay happen that ss have different intrinsic probabilities usmntcces 9, For example, a certain energy state may be compatible with more different Ce ee ee anor ates tnd reste Wns Maly to bo Hee a ate pasabiley taen ito sous te pkey en eee SUNY atent em Be C0, ObruR ee aa Pa cet Ue orc coeng Ba pot of id tivo particles in that level is 9, Xo; = g2, and for n, particles it is 9; X gv X ge X See TET Tor Sat of a riven pration, inead of beg given by Eq. (10.6), is given by } yRighigha pp i (0.7) "We thall now illustrate an application of 10.3) The Mazwel-Bollemann distribution law 439 Finally let us remove the condition of distinguishability. If all particles are sdgazal su RAvGRPUUTDIG Se canoe Tesogaize tho diterence if, for example, in the partition in Fig. TO-I particles « and’p exchange places. Therefore all N! permutations among the pasticles ovcupving the different sistes give the same partition, That means that we have to divide Eq. (10.7) by WI, eating in’ PBX Ou 4 distur oF Maxwell Bo pcsets (108) ‘This is the expression for the probability of a disibution in MacsaliBattemaan si ‘We ean obtain the equilibrium state, corresponding to the most probable pa tifion, by finding the maximum of P (given by Eq. 10.8) compatible with condi. tions (10.1) and (10.2), with V and U as constants. The method is s straight forward mathematical technique, and is explained in detail ik Example 10.J. The result is that the partition having maximum probability is given byw yunt. gi ng = ge PR / aos) where a and 8 are two parameters which, as will be shown, are related to the physioal properties of the system, reseed in terms of. the total number of particles, N, Using Eq, (10.9), we have that ny tna tig + gre PFs gee Pa gye"PF + gye-BP2 + gye-PP2 + ~~») Hee) Fa sper] Prrciene PARTIOH & cory ‘The quantity Z, ealled the @arliffon Functiom is a very important expression which appears quite often in many ealoulations. We may thus write e“* = N/Z, and Ex, (10.8) becomes a LE6GE OI ISR. a Goan =F oe ze Expression (10.11) constitutes the Maxwell Boliemann, Sie a] ‘The quantity 8 is related to the energy of the system, or more precisely, 0 the average energy of partisle, as will be explained in Section 10.4. 1 (10.11). From the definition of average value, we have that the average value (for a given partition) of a physical400 Chessell mechanics wos property F(E) that can be expressed as a function of the particle energy His given Zz oT [ Fave = LD narce), = fre=ale~re | do = F and for the most probable oF equilbrium partitign we have, using Eq, (10.11), Fave = FD OEE. 10.12) X¢ _ For example, if the particles of a system can be in only two states of energy, B, = +e and Ey = ~¢, both with the same probability (g1 = g2 = 1), then i mts gt I Soak pe 7 ~€ tanh Be. {The total energy is EXAMPLE 10.1._De éetuions To won probaba ox equliium partion comeponds (by dfn) to ena P Ps maxims in tary crarponds tothe sitution fn which the caren acai sco GP sO for taal hangs das nn dyin ont care ae a ears nad of etsning the masinim ef yi caren er i's bain the maximum of InP, wiih conesponts tothe sie Sina dP Now, tom Bq, (00), wo hove mP tion of the most probable or equilibrium partition In gi + na ln ga-+ ns In gat --+ = In mi! — In ng! — In nal — Using Stiring’s formula for the logarithm of the factorial of a very large number (see Appendix V), Inzlw zine — 2, ‘and sesuming that m1, na, ns, «+ are large numbers (since physieal systems are in gen- feral composed of a grest many particles), we have InP =m Ing molnga+ ns lngs+- = (a lam — m1) — (azn na — a) — (ng ln ns — 1s) — + = (and In ni/ge — Dns dln m/ed =D lero nndes— Dnt = =D (en) nlp — Dan. (10.14) But from Eq, (101), singe dN’ = 0 because Nis constant, we have Den (0s) ‘Therefore Eq. (10.14) reduces to a(n P) =D Gam/pd anu # 0. «aoe We write zero on the right side of Bg, (10.16) because we are looking for the equilibrium Sao DEORE Emin Gras O-and therfore dd P) = Poa? = 0. Teall the changes dny, dnz, dns, ... were arbitrary, we could eailsy Ti (1OT6) EF making Inmi/or = 0, In ma/g2 = 0, In ns/oa = 0, «++ OF my = gu na = However, the changes dns ate not entirely arbitrary because of condition (10.18), which wens darted Rom the Constanoy in he number of patil, sua ear conaioon, Deu= ao.17) resulting from the constancy of the internal energy and obtained by dif Eq. GO) end setting dU = 0. “To compensate for the two conditions (10.18) and (10.17), we introduce two arbitrary parameters @ and B, called undelermined multipliers, according to a mathematical tech- igus suggaried By-Lagrmage Gee Appendix VID. Malipiying Bq. (015) by «, Bq. (10.17) by 8 and adding these to Ea, (10.16), we obtain D> (ia n/gst a+ BEd) dns = 0. ‘This eds two new arbitrary coeficients « and 8 to compensate for the two restrictive conditions (10.18) and (10.17). The equilibrium distribution is then obtained if Inndobat BE: = 0 nm os Which is the result proviously stated in Eq. (10.9)W2 — Clasizal statistical mechanics (wos {X BKAMPLE 102, Aeyeam iacompond of 00 partes which may bein oe of thee cee rdtsor bec Sey, honors nee nd Zeal ving tee {Releio petty Gee Fe 10-2) Thi cium, for etopin eld ena of same wine Ene ental paced i 0 mags El Compa te inive'pbeblies of the partion whch thera 200 parses lover ee, T1000 he mide eel and 00 nto appr orl withthe psn euing rom he tate oon parle om theme vel wo the ver ovel and ante oe upper Ieee proces sompeile wth he cnsrraton of ener Byte ns Brae my Fig. 10-2. System with three energy levels. eee ms Solution: According to Eq, (10.8), the probabilities for the first and second partitions are ge gto Pu seooimoomoa’ «=? ™ aooTNeRRTOTT Instead of computing the values of Py and Pa (which we could do by using Eq. 10.13), ve shall simply find their ratios: Pa | 1700 1699 _ 2,888,900 _ 4 5 Pi go0r ani ~ ooaan ~ ** ‘Thus the mere transfer of two particles out of 4000 to other levels changes the probebility by a factor of 4.8. ‘This means that partitions Py and Pp are both far from being the librium partition; this situation is due to an ex i ‘Therefore the system Will try te evolve to's state in which the middle level is less populated, Tt is suggested thet the student repeat the problem, considering ather possible distributions of particles, all compatible with the same total energy. hift two more particles from the middle level or move one partie from the upper level and ancther from the lower level into the middle level and recompute the relative probabilities.) | EXAMPLE 10.3. Determine the most probable or equilibrium partition for she system of Example 10. Solution: The aystem is composed of N = 4000 particles, and according to the data given in Example 10.2, its total energy (ooe Fig. 10-2) is 2000 X O-+ 1700 X «-+ 300 x (24) = 2800. Using Bq, (10.9) for the most probable partition, we must have, according to the notation of Fig. 10-2, and with 9: = 92 = 99 = 0) ng = gett ay = oe" a = ge or, if we designate e-# by z, we have nz = iz and ng = m2, ‘Thus conditions (10.1) ‘and (10.2), which give the total number of particles and the total energy, respectively, Decome mit met mz = 4000, (ms)e-+ (mia) (20) = 28006, 40.4) Temperature — 448 ‘Canceling the common factor ¢ in the second relation, we have m(l-+ 2+ 2%) = 4000, ma(e-+ 22%) = 2300. Eliminating mz, we obtain an equation for 2: 4122+ 17s ~ 8 = 0, oz = 0.5084. (Only the positive root is used. Why?) Therefore m = 2277 (the figure hhas been sounded). Accordingly, na = 1146 and ng = 877. The corresponding partition probability is. ce ‘TST P Lot us now compute the change in P when two particles are removed from the intermediate level and transferred to the upper and lower levels. The new partition probability is BRA TaT P ‘The ratio of the two probabi Pe sex 114s _ 1,812,170 P ” aare x oT ~ 1,316,684 9068, ‘Therefore the two probabilities are essentially the same; this is as it should be, since, if P is a maximum, AP must be zero or very small for a small change in the distributi numbers of an equilibrium partition. 10.4. Temperature "The parameter 8 of Eqs. (1010) and (10.1) is directly related to physical quan iity, the temperature, which was originally introduced more to describe a sen- ‘sorial experience than a statistical property of an aggregate of particles. We first note that, if we are to have dimensional consisteney in Eqs. (10.10) and (10.11), ‘we must express 6 in reciprocal energy units: that is Io}, eV—! ete. Using Eq. (10.11), we have that the total energy of « system of particles in statistical equilibrium is [= mB, trae + mabe + iN Zz 2 (Cane) coup (GiB yet + galize*¥* + gsBse*™ +++)44s Classical statistical mechanics 104 Using the definition (10.10) of the partition function, we may write U_in the alternative form ° yd i —N Fn), | (00.19) which is an important relation between the total energy and the partition funetion of a system in statistical equilibrium. ‘The average energy of a particle is Sax ‘Note that, given.a physical system described hy the g/’s.and the K/s, the partition funtion Z (and hence the totale wwerage.encrgy-of a par- a {and alzo other parameters which determine the ‘macroscopic state, such as volume and pressure). That is, we may use the parameter 8 to characterize the internal energy of the system. However, it has been ne ‘more convenient luce a new physical quantity instead of 6. This led the absolute temperature of the system; it is designated by T, and (10.20) SATVE jetined according to the ration HaxwelLiouoy ico 12 FOUR erat. Mine aie SE at . Bi tre Bg: SP et ‘The quantity kT must obviously be expressed in units of such as joules orelton ws, Th cnn) cle te earn all ran tensed atler the nite for rar chum, eaten tbe thown that the ‘temperature we have defined here coincides with the temperature as measured by, say, a gas thermometer. We shall do this later on (Section 12.2). At this time it is enough to say that when 7 is expressed in a unit. called desrece keluin, designated by °K (which was introduced before statistical mechanics was developed), the value of the Boltzmann eonstant is, o21) _ = 1.3805 x 1078 J°K-! = 8.6178 x 10-5 eV °K, Of course, we could make k= 1 and measure the temperature directly in energy units, such a8 joules or electron volts. ‘This is perfectly possible. However, the tradition of expressing temperature in degrees, introduced long before the relation between temperature and the molecular structure of a system was recognized, is s0 well entrenched in physics that it is almost impossible (and perhaps also undesirable) to do away with the use of degrees. ‘The quantity 8 is positive (except in some special cases), as will be shown later; Renee the abso erature Ti positive. ‘The lowest temperature is zero, which corresponds to what is galled solu 2070, Te must be clearly kept in mind that the statistical definition of temperature, os given by Eq. (10.21), is valid only for-a system of particles in statistical ecu ‘umand hence does not apply toa single particle or to a system which is not in 104) Temperature 445 yuilibrium, ‘The reason for this is that, Iris swith the cal of ition of aaystem, which corresponds, Y definition, to the equilibrium state, If the system is not in equilibrium, we ‘may still speak of an “effective” temperature of each small portion of the system, assuming that each small portion is almost in equilibrium. ‘When we introduce the definition (10.21) of temperature in Eq. (10.10), we may write the equilibrium partition fanction as (10.22) (10.28) which gives the Maxwell-Boltsmann distribution law in terms of the temperature. Since @ = 1/kT, then. d8 = —dT/ET?, and we can write Eq. (10.19) as Eun = Moe $a. 2). (10.25) “The exact relation obviously depends on the microscopic structure ofthe system, which is exprested in the partition function Z, and ia diferent for an ideal gas, & agg real gas, a liquid, or a solid. Hence we may say thet the temperature of «system in statition!equtlibrium is a quantity related (othe cnerage enerou rer parte of the VE aystem, the relation depending on the structure of the system. Tn the general ease, the average value of any particle property, F(Z), which ‘was defined in Eq, (10.12), now becomes 7 wFEde", Fave (10.26) and is thus a function of the temperature. For example, for the system composed of partioles of energies +e and —e, considered at the end of Section 10.3, the average energy of a particle at the temperature 7 is Eye = —etanh (/ET), Since the exponential e~#s'*7 in Eq. (10.23) is a decreasing function of E,/k7, the larger the ratio Z;/kT, the smaller the value of the occupation number ny Therefore, at a given temperature, the larger the energy Bi, the smaller the value446 Classical statistical mechanics 04 FO rrr stray cere = a an re smaller values of E;/kT for given energy) the relative population of the higher vergy levels increases (again as shown in Fig. 10-3) by transfer of particles from ee armsigtaed saemanalite nant ee ee eee econ or a SAS n nals [ae | 0.27) where AE is the energy difference be- ‘tween the two levels. ‘Thus nzand nj are comparable only if A is much smaller than ET eymBoiet _ Os sein, - ‘Very low temperature Fig. 108. Maxwell-Boltzmann distribue tion at three different temperatures. ‘asin. 104. Given tengrctas of 10% 20" om tee, a 1000°K,, determine for each temperature the ratio between the occupation numbers for too levels corresponding to AZ equal to: (a) 10~* eV, which isa value equivalent to the pacing of rotational levels for ms 5 (b) 8X 10-# eV, which corresponds to molecular vibrational levels, and (c)3.00 eV, which i of the order of magnitude of eles. “Boni exaltation of atoms and molecules. Assume g = 1 ( Solution: Remembering that k = 8.6178 X 108 eV °K-1, we have that kT = 8.6178 20-®7 eV, and AB/ET = 1.1608 X 10*AB/T, where AE is expressed in electron volts. ‘Therefore, for the indieated values of AB, the values of ny/n; at the three temperatures are as given in the following table. ‘AE,ev | 100% | 300°K 10+ J osses | 0.9962 0.9988 sxio-? | 3x 10-8 | 1.45 x 10-1 | 5.60 x 10-* 3.00 3X 1o-t | 8x 10-# | BX 1078 ‘Therefore, for AE = 10-+eV, the two levels are practically equally populated at all GapSSUES aRTRTSN Te expla, Tor Sama Wigs Silos rotational a 104) Temperature 447 scxption apectrum of molecules is observed at normal temperature, a illustrated in WESas tor HCLATAH = 85CTD-*eV, Whe population of the upper level is alread re als of soon numa 91h means ar many lcs re an excl! wppesee Lntate at Toom temperatures, However, for AE = 3eV, the ratio m/ny isso Tull that ie plausible to consider the upper level as essentially empty stall tempers {Ges tonidered. Thus most stoms and molesules (at room temperature) are in thir ‘ound electronics Sac eatemely high temperatures, as in very Wot sla, are Fern nt and moloues in excited electron states in any appreciable smounty Tn the laboratory, elestronie excitations are produced by means of inelastic collisions of gas molecules with fast eleetrons in sn electric discharge. EXAMPLE 10.5. Consider a system of polar molecules which are placed in uniform tioeea field, but which are otherwise isolated from any other exterel action. Compute Solution: Let us designate the electric dipole moment of each molecule by po. We shall assume that the effect of the appliod electric field is only to modify the random brientation of the molecules without disturbing their translational or internal (rotational or vibrational) motions, and thet the molecules are freo to follow the orienting action of the electri field. ‘The energy of a molecule whose dipole moment makes an angle @ with the electric field 6 is £(@) = —po8 cos 6. Since the angle @ can have any value between O and, the energies of the available states are not diserote but vary continuously. ‘Thus, instead of looking for the number of molecules oriented at an angle we must find the number of molecules with their dipoles oriented within the eolid angle d@, defined as being between cones of angles @ and @-+ d9 (Fig. 10-4). ‘The value of this solid angle is aQ = 2rsin @ dB, and using Eq. (10.22) with g; replaced by df and the summation replaced by an integral, we may write the partition function as Ze ii epeteatiM 2g sin 80 — 4(kT/po8) sinh (Po8/KT). ‘The orientation of the molecules must be symmetric relative to the direction of the eleo- trie field and the average value of their electric dipole moment is parallel to 8. Thus, to find Pare we must evaluate (Po 005 Bae since po 008 6 is the component of the electric ipole moment in the direction of 8. Using Eq. (10.26), with the summation replaced “ by an integral, we have 4 pow Ef con nereoae sin bab m(oate—t) as h \ result whieh is cated Langevin formate) As. very large & or very small temperature (hat is, for po5.2> RI, we have that coth (petETT—=T and FIT 0B — 0, <0 that 2 electri fold, or at temperatures ‘oriented paraliel to the electric field. For small & Bese = po. This means that for a ve ‘te zero, practically ll molaeul448 Classical statistical mechanics 05 or largo T (that ic, po8/EI” <1), we use the asvmptotio expansion cothe = et e/8t-, Pan = pib/3H If there are n molecules per unit volume, the polarization of the substance is ‘and thus © = mPa = (nph/BkT)8. ‘This expression is used in the caloulation of the electric permittivity of a medium com ‘posed of polar molecules. ‘The magnetization of a substance placed in an external magnetic field when the molecules have a permanent magnetic moment is obtained by & imilar calculation, ‘This isthe orientation effect which gives rise to paramagnetism, In this ease, hqwever, « complication arses. The magnetic moment is related to the angular momentum, which is restricted in orientation, as indicated in Section 8.4. Thus instead of performing an integration, we must perform s summation, The final result ie till very similar to that obtained for an electric dipole moment (see Problem 10.8). (40.5) Thermal Equilibrium Consider « system eomposed of two different groups of particles, We shall sy that each group of particles constitutes » subsystem, Our two subsystems may, for example, consist of a liquid with a slid immersed init, a mixture of tro gases of liquids, or to solids in contact. By means of collisions and other interactions, may be exchanged between the tw ‘buf the total energy of the whole svatem is assumed to be Ssed—Let us designate The total numbers of particles in each aubayatem by N and NV’ and the corresponding energy levels available to the particles by E, Bs, Bs, ... and Ei, Bs, By, If there are no reactions ticles in the two 8 he tot number of particles in each sul ; also the total energy of the system is constant. But the energy of each subsystem is not eonserved because, due ‘ thelr interactions, energy may be exchanged between them. Therefore the following conditions for the two subsystems must be fulfilled by the occupation num bers nj, na, ng, ..., and nj, m,n}, ... in @ given partition: N= Sone = const, 0.28) Wr = Son} = const, (0.29) | U = Dine Dee ‘const. (20.20) ‘The probability of a given partition or distribution is given by the product of two expressions similar to Eq. (10.8), one for each subsystem: (20.81) 105) Thermal equilibrium — 49 fe can obtain the equilibrium of the composite system by requiting that P bea TEEN compatible with Bqs. (10:28) (10.29), and (10.30). The resale (oe “Fample 10.6) is that Fa fi n= Zoe Me 8H aMe where Z and Z” are the respective partition functions of the two subsystems. We Hy (20.82) Tn view of our definition of temperature, Bq. (10.21), we conclude that LEG : a ap tam, 0 Hum must have ie same termerature, Ee a statement called the zeroth law of armodynamics, It is then said that the subsystems are in thermal equilibrium. Hence, when we replace 8 by 1/RT, instead of £4. Bq, (10.32) we may write : No p-Po! ya Bezier, ne ge, nm Ze gse TNT, where 7's the common absolute temperature of the two subsystems. The expressions for ns and nj show that at thermal equilibrium each subsyatemn attains the same ae, ist piatlesaon nite Soule and ote rove wolated and et the same tempore T— ‘These relations therefore express the fact that, ina statistical sense, after thermal equilibrium is sttained the energy of each subsystem remains constant. This means thet, although both systems may exchange energy at the microscopic level, the exchange takes place in both directions and, on the average, no net energy is exchanged in either direction. Therefore statistically, the energy of each subsystem remains constant. It then follows that when two bodies at different, ‘temperatures are placed in contact they will exchange energy until both reach ‘thermal equilibrium at the same temperature. At this point no further net exchange of energy takes place. ‘The zeroth law is the basis of the measurement of the temperature of a body; this temperature is determined by placing it in contact with a properly calibrated standard body (or thermometer). The zeroth law is also in agreement with our common understanding of the concept of temperature by which, if a “eold” and 4 “hot” body are placed in contact, the eold warms up and the hot cools off un both are “felt” at the same temperature, Therefore, although the concept of temperature was frst introduced to correspond to a sensorial experience which is, of course, escocisted with a physical condition of the human body, it ean now be defined precisely by means ofthe statistical relation (10-21. “ Solution: When P is given by Eq. (10.31), if we follow the same procedure used to obtain. ‘Eq, (10.16), we have a(n P) = Sy ta ni/gi dns + Dla nj/oh dns = 0. 0.33)450 Classical slatistcal mechanice 0s Th restricting conditions (10.28), (10.29), and (10.90) now give Yaw =o, (1036) Lay (20.35) 7 DL kant Sy Bans = 0. (10.36) Multiplying Eq, (10:30 by a, Bq, (10.5) by a’, and Eq. (103) by 8, and introducing ‘hese in Eq, (10.33), we get Sinaia at BB) dnt TS (anh/o a + BB dn = 0, ‘which requires (for the same reason as in the ease of only one kind of partic) that Inndoctat BBs = 0, Inni/gi-+al+ BEY = 0. Accordingly we get n= ge OH, ny = ge or, using relations (10.28) and (10.28) to eliminate e-* and e~=" Ny gO. tw My 8 ma keet, he Fale Note hut we wea and a because the numberof partis ofeach subeystem is conserved separately, bit we use only one 8 eeuuse its the total energy which fs conserved. 10.6 Application to the Ideal Gas Qur next step is to determine whether there is apy system of particles in nature i fetions of the Maxwell-Bolizmann _ distribution law. Up to this point we have presented this distribution only as a more or less reasonable theoretical construction. Experience confirms the fact that bed according to Maxvell-Boltamann statities over a wide range of temperature. For simplicity we shall now consider ideal gases composed of monatomic molecules, In this way we do not have to include the potential ennay di F rach ‘intemal rotational or vibrational motions_of the molegles, and all moleeular energy is translational kinetic; that is, 2; in Eq. (10.11) is c= hot = tet aos7) But we note, bs explained in Example 2.6, that the kinetio energy of an ideal gas Jing a rather large volume may be considered as not bel i having a eoaiinuonsspesirom. ‘Therefore we must revrite Bq, (10.22), replacing E ects? _ fe oe I GE = ew 108) Application fo the ideal gas 451 ‘the sum by an integral in the form a= oman ae 20:38) ‘where (F) dB. replaces g:.and tensesents the number of molecular states inthe energy tange between Band E-.d#, This number arises from the different ‘rientations of the momentum p for a given energy. Trom Bq, (2.49) of Exam- ple 24h we have that (BE) dE (10.39) hare Vi th volume ssupied by the gas, Therefore (, [orca fret OS. ‘The integral appearing on the right can be found (sée Appendix IV) to have the value $V 7(E2)S. Therefore (10.40) which gives th partition jc gas as a function of temperature and . Then, by taking the natural logarithm of ‘ach side, we have InZ =C+3inkT, where C includes all the remaining constant quantities. Substitut value of InZ in Bq, (10.25), we obtain the average energy of the molecules as (10.41) ‘Therefore the avazage Kinetic energy of the moleoules of an ideal gas in statistical ‘equilibrium is proportional to the absolute temperature of the gas. Historically, Bq, (10.41) was introduced in the nineteenth century, long before the development of statistical mechanies, in connection with the kinetic theory of gases. Tt was this relation that suggested the identification of 6 with 1/k7. ‘The total energy ‘of & gas composed of N molecules is then U = NEw = HNT. We conclude then that the internal energy of an ides! monatomis only on its temperature, ‘This is # direct result of our definitions of an ideal ges ahd of temperature. We do not expect the same relations to hold for real gases or other substances, since their internal energy is partly potential and thus depends con the separation of the molecules; that is, on the volume of the substance. . ae = 2%482 Glassicalstatstical mechanict (os . 100° Fig. 10-5, Molecular energy distribu tion at two temperatures (100° and 00°K). Note, from Eq. (10.44), that the cengy distribution is independent of the os 353040 8080 BO (0? eV) TLx is the number of moles* of the gus, and N4 = 6.0225 x 10" mole? is Avogadro's constant, which is the number of molecules,in one male of the substance and which is the same for all substances, we have w= N/N a. )Thus Peder = ber, (10.42) (Sey 8.3143 Jmole"! 2K=) 1.9860 calories mole“! *K-! 1894 X 10"? eV mole“ K~* where (10.43) is called the gas constant. Expression (10.28) forthe distribution of molecules among the different enersis, with ge replaced by 9(B) dB as given by Bq. (10.38), is now eV (omi)!!? | an = B80 g(e) ap = ¥ BVO preven al where . By E he value of Z 9s ei Eq. (10.40), we have Svbuw Oa: lin 22 pie, zr ‘ote ste B= Gera ) ow which is Mexwell’s formula for the. in ap ide joE. NOW gaa This eootiates one of the ears aplitatins of sane metinds In omen Se phystes. It was originally derived by James C. Maxwell around 1857, using a different logic. A plot of Bq. (10.44) for two different temperatures is represented a in Fig. 10-5. * sub {te molecular mass expressed in amu. is an emai 1 substance, in grams, equal 106) Applicaton tothe ideal as 453 ae Fig. 10-6, Molecular velocity distribution for oxygen at two Vemperatures(80°K and 800°K), Note, from Eq. (10.45), ‘that the velocity distribution depends on the molecular mass. oT 60m =) ‘Sometimes we require the velocity distribution rather than the energy distribution. Noting that Z = 4mv%, we then have that dn dB dn @B ay aE fash gun, coe veto: 25_It gives the number of moleoules moving with s velocity between v and ‘0 + dy, irrespective of the direction of motion. ‘The velocity distribution for ‘oxygen at two different temperatures is represented in Fig, 10-6. ‘A crucial teat of the applicability of Maxwell-Boltzmann statistics to ideal gnses is to see whether the energy and velocity distributions illustrated in Figs. 10-5 and 10-6 actuslly occur. One indirect way of doing this is to analyze the depen dence of the rate of chemical reactions on the temperature. Suppose that a particular reaction occurs only if the colliding molecules have a certain energy equal to or larger than Eg, ‘The rate of reaction at a given temperature then depends on the number of molecules having an energy equal to or larger than E., ‘These numbers, for two different temperatures, are given by the shaded areas under the low temperature and high-temperature curves of Fig. 10-7. We note that there are more molecules available st high than at low temperatures. By proper caleula- tion, we can predict the effect of these additional molecules, and the theoretical prediction oan be compared with the experimental data. Experimental results are in excellent agreement with Eq, (10.43); this confirms the applicability of Maxwell- Boltzmann statities to gases, Towle a7 Rexwee —10s Enerey Fig. 107. Number of molcules with energy larger than Eat two diferent tempers- tures. ‘The number in each eas is indiated by the shaded area ‘A more direct verification ui “count” of the number of molecules i z Several experimental arrangements bbave been used for this purpose. One method, using a velocity selector or “chopper,” isillustrated in Fig. 10-8. ‘The two slotted diss, D and D’, rotate with an angular velocity w and their slots are displaced by an angle #. When gas molecules eseape from an oven at a certain temperature (a process called fusion), they pass through both slots and are received at the detector only if their velocity is » = au/@. Actu- ally, since the slots have a finite width, the transmitted molecules have velocities in a certain range Av about the given value. When cither w or @ is changed, the velocity of the molecules received at the detectors ean be changed. If one makes several observations for different velocities », the velocity and energy distribu: tions are obtained. ‘The experimental results confirm the prediction of Maxwell- Boltzmann statistics, as expressed by Big, (10-44) and (10.45) As explained in Section 8.6, neutrons produced in the fission process in a nuclear reactor are slowed down by means ofa suitable moderator until they are in thermal Estuing Fig. 10-8. Molecular velocity selector. 408) ‘Application lo Ue ideal gos 455 equilibrium at the temperature of the reactor. Neutrons in a thermal nuclear ‘ad de Sac AEDS agra WT Hiatisties, This Tact is essential in nuclear reactor design. If the neutrons emerging. [from a porthole in reactor are made to pass through a chopper similar to that of Fig. 10-8, a monoenergetie beam of neutrons is obtained. \\ EXAMPLE 10.7, Obtain the most probsble energy and velocity of the gas molecules j/st a given temperature; these values correspond to the maxima of dn/aB and dn/ds, respectively. Solution: To obtain the maximum of dn/dB, given by Eq, (10.44), ata certain tempera ‘ure, it e necessary to compute only the maximum of y = BY?¢-# Pa oe B- (ee — 4, from whieh we have B welave Bay ~ 00120, ‘Similarly, to obtain the maximum of dn/d, given by Bq, (1048), we must computa the musinim ofp = oe Then JET. Thus at room temperature, for which RT ~ 0.025 eV, EXAMPLE 10.8. Average velocity (2) and root-mean-square velocity (Ha) of the ‘molecules in an ideal gas Solution: The average velocity is defined by ey weed [oad [te Inteedusing the value of dn/do given by Ea. (10.45), we have ‘We first transform the integral by setting w = o?, 20 that du = 2edo, resulting in afro, Next we integrate by parts, resulting in 4(2k7'/m)?, so thati a e Fig, 10-9. Most probable, average, tnd root-mean-square velocities of a as. o Ty Pe rae 7 ‘We define the root-mean-square velocity by ime = (ear and we obtain the average of o from the expression “ wed [een Noting that »? = 26/m, we may also write 2ift 2 me Zi |, Bee 2 Baw since the average energy is defined by Ea. = (1/N)fE dn, But for an idesl gas, according to Bq. (OAD), Buy = JOT. This the = an = Zager) = BE v= (2) ‘The three velocities, tmp, Samy and tw are indicated in Fig. (10-8). and References 1. *What is Probability?” R. Carnap, Sei. Am., September 1953, page 128 2, “Thermodynamies, Statistis, and Information,” L. Brillouin, Am. J. Phys. 29, 318 asst) 8. “The Work of T. H. Berlin in Statistical Mechanics,” M. Kac, Physics Today, Octo- ber 1964, page 40 4. "New Approach to Statistical Mechani (1967) 5, Slatistical Mechanice, G. Rushbrooke, London: Oxford University Press, 1949 8. Slatietical Phyeice, C. Wannier. Now York: John Wiley, 1968, Chaptors 1, 2, and 3 5." R. Woinstock, Am. J. Phys. 35, 710 t Problems 457 7. Pundamentale of Statistical Thermodynamics, R. Sonntag and G. Van Wylen. New ‘York: John Wiley, 1966, Chapters 1 and 2 8, Statistical Thermodynamics, B. Schr6dinger. Cambridge: Cambridge University Press, 1064, Chapters 1 and 2 9, The Feynman Lectures on Physics, Volume I, R. Feynman, R. Leighton, and M. Sands. Reading, Mass.: Addison-Wesley, 1968, Chapter 40 Problems 10,1 Referring to Example 10.2, find the change in the probsbility of the distribution if two particles, one from the upper fand one from the lower level, are transferred to the intermediate level. Repeat for the most probable partition found in Example 103. 10.2 In Example 10.2, compute the ratio Po/Pa, using Eq, (10.13) for la P. 10.3 Determine the temperature of the system of Example 10.8 when it is in statistical equilibrium, Assume ¢ = 0.02 40.4 Find the ratio of the number of electrons having their spins parallel and anti- parallel to a magnetic feld as a function of ‘the temperature. Evaluate the number for 8 temperature T equal to 10°, 300%, and 1000°K. [Hint: Recall thet the spin mag netic moment of the electrons is given by Mz = —2unS, Bq. (8.33),) 105 (@) Show that the partition function of an electron gas placed in a magnetic field @ is Z = 2eosh (un/E7), where us fis a Bohr magneton. (b) Compute the magnetic energy of an electron gas in the magnetic field; then show that the paramagnetism of free electrons corresponds to 8 magnetization M = nus tanh (unG/KD}, ‘where n isthe number of electrons per unit volume, (c) Find the limiting values of the partition function and the magnetization at very low and very high ‘temperatures. 10.6 ‘The possible particle energies of @ system of particles are 0, ¢, 26 «+ M4 (@) Show that the partition function of the system (with gs = 1) ia Bane, (©) Compute the average energy of the particles. (6) Find the limiting value of the average energy when ¢ is much smaller than FF. 107 Considering a system of particles Ihaving energies 0, ¢,2e..+ m4 Plots raph showing the occupation numbers for {@) 100°K, () 200°K, (6) S00°K, given that the value of the energy eis (a) 10~¥ eV, (@) 0. 6V. [Hints Use the result of Prob! Jem 10.6for the partition function} 103 The magnetic moment of atoms (or rolecles) having an angular momentum Fis given by Bq. (G40): Mo —nng. {@ Obtain an expression giving the number of atoms having 6 value of J, equal to smi, if the stoms are placed in « magnetic field @ oriented parallel to the Z-axis. (b) Show that the partition function of the system of atoms is z= Sab IG + Done /ke} sink (HngG/ET) (6) Verify that for j = } the partition function reduces to that obtained for electrons in Problem 10. 10.9 Obtain the average magnetic energy ‘of the atoms considered in the proceding problem. Verify from this result that the average magnetic moment of the atoms ia 1 direction parallel to the magnetic field is Mave = hogl(j + 4) coth (7+ $)2 — booth $2]i 5 458 Classzalsatistical mechanics where z= sapB/EP. Find the limiting values of Mure for # very small and very large compared with 1 10.10 At what temperature is the number ‘of molecules of an ideal ges per unit energy range at 2X 10-FeV, one-fourth of the number corresponding to 1X 10-# eV? Find wthe ratio of the nunber of ‘molecules of an ideal gas per unit energy range at energies of 0.2eV and 0.02eV, given thatthe gas temperature i (a) 100°K, (b) 800°K, (@) 600°. 10.12 Compute enough points to eon- struct graphs of the molecular energy distribution funetion in an ides! gas for one Kilomole at 200°K and 600°K. 10.18 («) Compute the root-mean-square, average, and most probable velocities of oxygen molecules at s temperature of 300°K. (b) Compute the most probable velocity of oxygan molecules at the following temperatures: 100°K, 300°K, 1000°K, 10,000°K. 10.14 (4) Compute the mean translational Kinetic energy of an ideal gas molecule at 300°K. (b) Compute the root-mean-square velocity if the gas is hydrogen (Ha), oxy- ‘gen (Oa), or meroury vapor (Hg). Compare Your results for hydrogen and oxygen with ‘the velocity of sound in those gases. 10.15 Compute the root-mean-square velocity of (a) helium atoms at 2°K, (b) nitrogen molecules at 27°C, (0) mercury ‘toms at 100°C. 10.16 At what temperature is the mesn ‘translational kinetic energy of « molecule of an ideal gas equal to thet of singly charged ion of the same mass thet has Deen secelerated from rest through potential difference of (a) 1 volt, (b) 1000 volts, (@) 1,000,000 volts? Neglect relativistic effects, 10.7 Determine the fractional number of ‘molecules of an ideal gas with velocities Tpetween thye and 1.2 from Eq. (10.45) ‘aking (a) v = tyre and do = 0.20 (0) P= Lltaye and do = O20. 10.18 Show thet the number of molecules of an ideal gas moving with velocity having ‘components in the range vy and %_ + des, ay and vy + doy, og and eg + dy, ia given by an = Mom/2ekT)AM I ye dy do Note that, in Bq. (10.45), for this ease we ‘must replace the integration element eo? do by deedey de... Justify this replacement) 10.19 Show that the number of molecules of an ideal gas that have an X- component of velocity between t and a+ dts irespective of the values of the ny and v, components is an (ifint: Integrate the expression obtsined in Problem 10:18 over t, and vy. ‘The limits ‘of integration in each ease must be from setts) 10.20 Use the result of Problem 10.19 to ‘obtain the average value of (a) v, (b) 02, (©) [eal for an ideat gt 1021 The error function exf (2) is defined by 1 etek On Fle In Table 10-1 we give some values of this function. Using the result of Problem 10.19, show that the number of molecules of sn ideal gas having an X-component of velocity between 0 and og is N@,».) = Wert @), where 2 = (m/2¢7)"%0,, and the number of molecules having an’ X-component of velocity larger then vis N(va ©) = NU — ext (I 10.22 Whst fraction of the molecules of an ‘desl gas have positive X-components of velocity greater than 2m? ‘TABLE 10-1 Values of orf (2)* et @) 0.4882 0.4948 04977 04991 0.4998 0.4999 0.5000 |_2.5000 "For larger values of 2, the values of xf (2) remain constant st 0.5000, up to four significant figures. 10.28 Compute the fraction of molecules of an ideal gas that have a velocity with a component along tho X-axis (a) smaller than tap, (b) larger than vm, (c) smaller than sw, (2) larger than tye (Hint: See Problem 10.21] 10.24 What fraction of the molecules of ‘an ideal gas have X-components of veloc ity between —tap and-Hias? 10.25 Show that the number of molecules ‘of an ideal gas that have a velocity between zero and » is given by on =f 1 ve where 2? = mo#/2ET. Also find the number of molecules that have a velocity larger than ». [Hint: Use Bq. (1045) and integrate by parts] 10.28 Using the result of Problem 10.25, compute the fraction of molecules of an ideal gas that have a velocity (a) smaller than tap, (b) larger than tap, (c) smaller than tury (d) larger than tre. 10.27 What percentage of oxygen mole- ‘cules have velocities greater than 10% m s™* ats temperature of (a) 100°K, (b) 1,000°K,, (©) 10,000°K? Illustrate graphically in ‘terms of the distribution function. Probleme 459 10.28 Using the result of Problem 10.25, find the percentage of molecules of an ideal ss that have an energy larger than 0.5 eV, siven that the temperature is (a) 300°K, (b) 600°K, (6) 1,000°K. (Hint: Note that in the expression derived in Problem 10.25, = E/eT] 10.29 Show that the number of gas molecules with velocity between v and o-+ do colliding with the wall of « container, per unit area and per unit time, is to(dn/ds) av. ‘Then show that the total number of mol= cules colliding with the wall per unit time is tay, Where m in both eases refers to the number of molecules per unit volume, Note that the molecules with velosity » colliding per unit ares and per unit time tand moving in a direction making an angle @ with the normal to tho wall aro those within ® cylindrical volume of height equal to v cos 4, Also the fraction of molecules moving within the solid angle d0 is a2 and df = Drsing a8, 06 Assume vost a small leis made in of an oven containing 4 gas at temperate 7. Show that dhe nimber af molecule with velocity between? and ot do excaping ‘per unit aren and Per where NV gives the total number of molecules per unit volume in the oven. Find ‘the total number of molecules escaping per unit area and per unit time. Also square molecular velocities in the outgoing beam are a wR tema = VERE Fm. andALONSO (2413 12 THERMAL PROPERTIES OF GASES 12.4 Introduction 122 The Equation of State of an Ideal Gas 12.3. Equation of State for Real Gases 12.4 Heat Capacity of an Ideal Monatomic Gas 12.5 Heal Capacities of an Ideal Polyatomic Gas 12.6 The Principle of Equipartition of Energy494 Thermal properties of gases «ea 42.1 Introduction Ia the two preceding chapters we have Inid the groundwork for « discussion of the general properties of inatter in bulk: the mechanical, elecrial, thermal, and chemical properties of mater or large aggrogates of atoms ot molecules, "We ean snprosch the dscusioa of these properties from tro opposite directions. The ‘perimental approach calls for extensive laboretory measurement ad tabulation of properties euch as density, specific heat, thermal and electrical conductivities, Yiseadly, clastic moduli, suface tension, thermal expansion, chemical reaction Inies, ete, and-the empitieal dependence of these quantities on extraal factors tuch as presure and temperature, applied eleire and magneti file, el. This has been the traditional procedure for those interested primarily in a knowledge ofthese properties for specise applieatios. ‘The alterative approach is heretical; {hie approach consists in an evaluation of the bulk properes of matter ia terms ofthe atomic and molecular structure and in terms of th interactions between ttoms and meleculs. Although thepe teractions are esentaly lestromauueticn come simplifying assimptions of « largely shenomenologeal nature are made (such as a reasonable intermolecular potential energy). Beesuse of the large number of particles involved, the theoretical approach requires the use of siatistial methods this chapter we sallustrate the theoretical proosdure by working out some representative problems related to the thermal properties of gues. The extension ‘ofthe methods of this chapter to other states of matter otto special topics belongs to eagioeering,physieal chenisiry, and applied physica, and wil not bo davonod in this texts 12.2 The Equation of State of an Ideal Gas ‘The simplest of all systems of particles is an ideal gas, and in this section we shall, obtain its equation of state. In Example 11.6 we derived Hq, (11.30), that is, az _aw vu ww =F + Rar. a2) ‘This equation relates the change in the partition function to the work done by system and its-change in temperature. For the case of a gas, in which the only work is expansion work, we have dW = pV. Also d(ln Z) = dZ/Z. ‘Thus we may rewrite Eq, (12.1) in the form pa , Van. rtp kN a(inzZ) = If the temperature is constant, dT’ = 0, and solving for p, we have wr [220], (12.2) 4 | 122) ‘The equation of sate ofan ideal gas 495 where the subscript T' indicates that the temperature is constant. Equation (12.2) rolates the pressure of a system to its temperature 7, its volume V, and the internal structure of the system, as expressed by Z, ‘Therefore it provides a relation of the form fip, V,T) = 0. Hence we may call Eq. (12.2) the equation of state of the yster. For an ideal gas, the partition function is given by Eq. (10.40), Vermin)? | z a za) which, when substituted in Bq. (12.2), gives (24) Fauation (12.4) is the equation of sate of an ideal gas. Tis sometimes written as pv = ENT, (25) cr, since B= kN, = KN/w, where N= N/N« is the number of moles of the ‘gas and V4 is Avogadro's number, pV = sRT. aze) ‘We can derive this equation in other, perhaps more direct, ways. In one case, we can compute the pressure of the gas by analyzing the change in momentum of ‘the gas molecules when they hit the walls of the container. In another method, we can use the virial theorem to show that Eq. (12.5) gives the pressure of # gas when the intermolecular foros are neglected.* ‘The fact that we arrive at the same regult in all derivations shows the consistency of our methods. =X werenlowsl os Note on the measurement of temperature., In Section 10.4 we associated the temper- sture of a system of particles with tho average energy of a particle. In Eq. (10.41), whichis Ei. = $27, we wore more sped about the relation between the temperature fan ideal gutind the average Kinets energy of the gas molecules. However, we rust how eotaider two important aspects: First, in the dann equation (10.28), we intro- ‘duced two new quantities, 7 (the sbsolute temperature) and & (Boltsmann’s constant), land we must decide how they can be measured independently. Second, all human beings Ihave an intuitive coneept of tamperature based on sensorial experience, a3 reacted by our fealngs-of hot and call, ‘We ae all scoustomed to meacuring temperature in terms ‘of e number given by a devie called a termomater. ‘Therefore we must correlate our ‘atistical detitin of temperature with this intuitive notion, Let us consider a mass Mf of «gas containing 8 molecules. If we nogeot the effect of ‘he intermolecular forces, the equation of states given by Bq. (12.8); that is, 7 = ENT. *See, for example, Fundamental University Physics, Volume I, Section 9.13 and Exam- ple 9.6.496 ‘Thermal propertie of gases 22 Suppose that we bring the gas into thermal equilibrium with some other physical system, ‘which we sesuime may be kept at « fixed temperature. This system may be an equilib: rium mixture of water and ioe at the standard pressure of 1 stm. ‘hiss called the normal freesing point of water, We measure the pressure and the volume of the gas a this fixed, temperature, and obtain the values po and Vo, respectively. Next we decide to assign a convenient (but arbitrary) value To to the Sxed temperature, which is also the tempera- tre of the gas. ‘Therefore we may write poVo = kNTo. ‘This automatically fxes the vvalue of the Boltamann constant, k = poVo/NTo, since we can obtain WV if we know the mass of each molecule, “To determine the temperature of the gas when Its pressure is p and ite volume is V, so that pV = ENT, we simply eliminate the factor kN, using the standard values, and obtsin 1 = To(p¥/p0¥o, which gives 7 in terms of our standard referenoe temperature Tp and other messursble (quantities. In this way our mace of gas has become a gae thermomeer. Tf the volume of ‘the gus ie maintained constant and equal to Vo, we have T = To(p/po), resulting io a conatant-olume gas thermomler. We tay use other substances as thermometers instead of enses, such a3 liquide or metals whose dimensions (volume or length) change with the temperature, Other thermometers use electric conductors (such as platinum wires) whose resistance varies with the temporature. Since the equation of state of these substances is more complicated, in practice we calibrate these thermometers sgainst a ges thermometer, Th this ease the thermomter agrees with the gas thermometer only at the calibration points, Since the property chosen may not vary linearly with the gas temperature, there may be slight diserepancias at intermediate temperatures. ‘We may choose the valu of To on the basis of several points of view. For example, we ‘may choose another provess which conceivably occurs at a fixed temperature, such 0s ‘water boiling at the standard pressure of 1 atm, which is called the normal boiling point Of water. ‘Thon we may decide that the temperature of this second reference point is $00 units, or degrees, above Zo, chosen as the normal freezing point of water. Given that ‘pi and V are the pressure and volume of the gas at this new temperature, we have that piVi = EN(To-+ 100), Solving for kN from the equation poVo = kNTo and substi {toting this value in the above equation, we find that To = 100poVo/(p.¥1 — poVo), from which we can obtain a numerioal value for 7 in this arbitrarily chosen sesle. The value obtained for 7o as a reqlt of this type of experiment (and many other experiments using different techniques) ie To = 273.15. Each of the units is called a depres Kelvin, designated by °K. Nowadays it is preferred to simply assign, by definition, the value To = 213.16 °K to the temperature of the normal freezing point of water. The value of ‘the Boltamsnn constant then becomes k = 1.3805 X 10-# J °K, “Wis important to realize that the technique we have explained for measuring tempere- ture fs bused on the ideal gas approximation. If wo use different gases, the results obtained will not be the same because the effet ofthe intermolecular forees, as it appears ia Eq, (127) of the next section, is diferent for each ges. Usually bydrogen or helium is used. Tt is most desirable to have a tompersture scale independent of the substance being used ase measuring medium. We osn accomplish this by using s reversible thermal engine operating in a Carnot oyele (see Example 11.7). Then the eficiency of the engine 123) Eequalion of stale for real gases 487 ia independant of the substance used in the engine, and is given by Eq. (11.47), B = (Ty — 72)/Ts. Given that 9 is our standard temperature J and Ty is the tempera ‘ture T to be measured, we have _fof, F Hence if we measure B, we obtain 7. Te has been found experimentally that for a thersaal engine operating between the normal boiling and freesing pointa of water, E = 100/373, If we choose T’— To = 100 degrees for those two temperatures, we again have that To = IS". ‘The temperature obtained using a reversible engig is called the thermodynamic temperature. "This method was proposed by Kalvin. Both Kelvin snd Joule rade careful experiments comperiag the temperature measured by ® constantvolume hydrogen thermometer with the thermodyaamio temperatare. 12.3 Equation of State for Real Gases ‘When we are dealing with real gases, we must take into account the in {goes yod the fn dimen of the nol "The inlermoleular frees ae of ith the distance between moleoules. There- fore the pressure of a real gas will be closer to the ideal gas value, Eq. (12.4), the larger the volume per molecule; ie. the larger V/x. This suggests that we express ‘the pressure of a real gas in terms of the series pa BE MA MB ae. wan in negative powers of V/x. We can consider Eq. (12.7) as the equation of slate of a real gas. A, B, C, ... are quantities characteristic of each gas, called the second, third, ete,, virial coefficients. They are functions of the on the strength of the intermoledilat forves. By in ‘tures and volumes, we can obtain the virial coefficients A(7), B(T), . tally. However, to find a clue to the correlation between the virial coefficients and the intermolecular forces, we must obtain certain theoretial relations. One possibility is to use the virial theorem, which is derived in-mechanics texts.* ” This ‘theorem, when applied to a gas, becomes 2 Ey 3, a varea(E Fv) ; * aaa where Fy is the force on molecule i due to molecule j, rz is the position vector of molecule 7 relative to molecule j, and the summation extends over all pairs of molecules, When we compare Hq. (128) with Eq. (12.7), we find that, ue Furs) (Aree), 2.9) ‘See, for example, Fundamental University Physics, Volume'l, Section 9.12.498 Thermal properties of gases «23 which in principle allows us to correlate the virial cosficients A, B,C, ... with the intermolecular forces. The methods of statistical mechanics allow a more straight= forward ealeulation of the virial coefficients, as we shall presently see. However, Decause the statistioal mechanics of systems composed of interacting particles is slightly more complex than that of systems of noninteracting perticles, we shall have to omit some derivations. Tet us introduce the quantity Z = Z”/NI, called the grand partition Junction of a system of noninteracting particles. Then, noting that nz = In @*/N}) = NinZ — aN}, ‘we may also write the equation of state (12.2) in the form Sonn), (02.10) For an ideal gas, using Ba. (1238), we have 1 [Ve@rmery ng = CONEY, am Win er couig ea cade dng vi ting eras aetna aie eee een eet Bet Soe a cc eeiias aden ta Fe et pee asin ase za fOr) ff. ferent arsara-arn, 0212 whee the 1 velume intel eorespond toons foreach melee, This ppt Shy foidableenrenn i fas, vry singe exten of the ariion ‘Rtn for an ego, esas fort el ga we ust make By =O nd hen 72, — 1. Thus the multiple integral becomes Jfp fart aren 00-7 Aimotecalen =v, (a23) since each moleeule may be found throughout the entire available volume. Then Eq, (12.12) becomes 1 wi ray 3 [aman Fy nt Zyat=miL [eormeny. in agreement with Eq. (12.11). ‘Now, retuming to Eq. (12.12), we must write the potential energy as Ey= Y Boy a 12.3) Equation of slate for real gases 499 ‘and use a reasonable expression for the intermolecular potential energy. Once we know Ep, we may in principle evaluate Z. By applying Eq. (12.10), we can obtain ‘the pressure p and expand the result in inverse powers of the volume V. In this ‘way we can establish # relation between the intermolecular potential By and the virial coefficients 4, B, .... Since we can determine the virial coefficients A, B, ... experimentally, this provides a very useful guide for investigating the form of the intermolecular potential Ey.j on a sort of trial-and-error basis until we can calculate the correct viral coefficients, We shall illustrate some of these techniques, in the remainder of this section Let us consider the multiple integral of Fr) Bq. (12.12); that is, Ta ffeoof PM avy ava a¥in svhere the summation B= D Boy x ‘contains N (NV — 1) terms, which is the total number of different pairs of molecules. Noting thet gett ‘we have that a = Fig. 12-1. Intermolecular potential energy. where the symbol IT means the product of all terms appesting, Also when Ei; is, very small, the exponential factor is practically 1. But (except when the molecules are rather close) Ei, is very small, due to the short range of the intermolecular forces (o0e Fig, 12-1). ‘Thus, using the expansion &~* = 1—2-+ fa? —---, we can write teh me — Batt 4 3 (Batty nt fey 21H and the quantity fy is very small except when the to nilesules are-very cose ‘Therefore Pt TT tha) = 1+ D fst ay a ‘The terms that have been omitted involve products of 2, 3, .. fg’s, and we shall noglect them, although, in « more detailed theory, they must be considered, With500 ‘Thermal properties of gases wes ‘this approximation we may write the integral J as Tm fff (AE fat) arate ae ‘Te term having te factor 1 obvioly gives after iteration in agrement Ti ie povows rout for an ent gua in Bq. (2218). Suopedig tera then TREES Sitton of tho intermolecaar Coens Tho AQ — 1) terns ofthe Stout a allt, becouse fy hs hese form forall pairs of moleues. ‘Thus we may wets their ontsibation in the form BVO — DYNO ff fina¥a dV, a8) where we have chosen the pair of molecules 1 and 2 and the factor V¥~* results from the volume integral for the remaining N — 2 molecules. In evaluating the double integral, we may first choose our origin of coordinates, at molecule 1 to perform the integral over dV (Fig. 12-2). Designating the distance between 1 and 2 by 1, we may then write f [feo ari are = J {[pfistrter? a} ar, where we have used dV, = 4rr* dr for the ‘volume element, because of the spherical symmetry of the problem. ‘The integral oe 4" = [falar ar 02.18) Figure 12-2 is independent of the position of molecule 1 (60 long as it ia not close to the walls of the container), and therefore J, [fat avrara = J aavs = afar = oY. Replacing 4N(N — 1) by 4N, which is «valid approximation when V is very Cae Clave Ba (1210) in the form 4N?7¥—%9, and we may write the integral J as ay x2), Ta Vt NAD tp = 7 (14 EP However, if successive terms are taken into account in the expansion of ¢” the result one obtains for I is (1 NB) meth ayer, ee inst 123) Equation of sate for real gases S04 ‘The previous result is just the first two terms in the binomial expansion of the above expression. The grand partition function of the real gas, Eq. (12.12), is now tiven by the expression 1 [V@rmery!® a\” z=af anmbt) T O48 (237) up to the frst order of approximation. ‘This must be compared with Eq. (12.11) fora real gas. The last factor in Bq, (12.17) isthe contribution ofthe intermolecular {oroes to the grand partition function. : To obtain the equation of state ofa real gas, we now use Eq. (12.10). From the ‘expression (1217) for Z, we have ize wing + in (t+ BP +P), where we have included in F(Z) the remaining terms that are either constant or depend on the temperature and do not affect the derivative in Eq. (12.10). We now approximate the second term, using In (1+) ~ 2 under the assumption that 2 (=N6/2V) is small compared with 1 (the quantity N'8/2V is of the order of 10~* at STP). ‘Thus nz=NnV ‘Therefore which, substituted in Eq. (12.10), gives v But remembering that N = NV, and R= kNa,-we may write NRT _ WRT 8 ee (2.8) which is the resulting equation of state of the real gas expressed in virial form, to ‘the first order of approximation. Wenote that in Eq. (12.18) we have obtained only ‘two terms; this is a result of the type of the approximation which we have made in the evaluation of Z. When we take these terms into account, the complete virial expansion results. Comparing Eq, (12.18) with Eq, (12.7), we see that the second virial coefficient in our approximation is A(T) = —4RTN a8, (12.19)502 Thermal properties of gases 23 which directly relates the intermolecular interaction (identified by 6) and the “Virial coefficient A(T) (determined experimentally). This is the connection between the intermolecular interaction and the virial coefficients that we indiosted before. ‘We shall not pursue our disoussion any further; what we have said is enough to indicate the method of attacking the problem of the equation of state of s real gas, and in gonerel of any eystem composed of interacting molecules. The case for liquids and solids is more complex, and we shall not discuss it in this text. EXAMPLE 12.1. Evaluate the second vitial coofcient for the case of & gus composed of noninteracting hard spheres of radius ro Solution: Tis is « rather “unrealistio” real gas because no intermelecular foroes are fssumed until the centers of the molecules are 2 distance 2ro apart, at which time ‘Snong repulsion sels in. ‘Thus the intermolecular potential energy is Byia ~ Oforr > 2ro fand Eyis = © for < 2rp, resulting in fiz = O for r > Bro and fis = —I for < 2ro. ‘This “hard-core” potential, os itis usually called, is represented in Fig, 12-3. We have ‘Called this an unrealistic potential energy because it cannot produce condensation, since $e baa no minimum and therefore no stable separation between the molecules. However, He does give us a simple model with which to check our physical ideas without undue Jatheroatical complications, Introducing the values of fiz in Eq. (1216), we obtain 30er) oe B= f (-1)4er dr = —$x(2r0)* = “When we substitute this value in Hq. (12.10), the second virial coefficient becomes : ied (2) = a, where b = Na(16nr#/2) is four times the volume of the molecules in one mole. Hence The equation of state of e gas composed of hard apheres, to the first-order approximation, 4 RTD EXAMPLE 12.2, Extend the results of the previous example to the case in which the se niewulac forces are attractive but very weak, except for r < 2ro, where very strong repulsion sets in, making the molecules almost ike impenetrable hard spheres Solution: As a logical extension of the calculation performed in the previous exemple, Srimay introduce rather weak but attractive intermolecular forces for r > 2roy as ills: Teted ia Fig. 12-4 Then we may continue with the assumption that fy = —t for Peas But from Bq, (12-14) if Zpy/KT i small compared with unity, we may write fia = -Bpia/BP for r > 20. Therefore Bq. (12.16) gives ord | a 7 te - 2 Be f (tar? dr ja (—Byua/kT Ma? dr = — 129) equation of slate for real gases 503 Bp Bane Epa e ® Fig. 12:3. Hard-core intermelecular po- Fig. ‘2-4. Intermolecular | potential fesiial energy. No attraction at any energy with ahard core and weakly attac- distance, tive larger distances. where am f° (Byte ae is postive quantity since Eqis Ss negative (ee Fig. 12-4 fot & Bra: Substituting in BE Ebioy, Neng Wis proves deGutdon of? sad elting e = Na, we obtain 4 = Tb —o for tho second virial coefficient, The equation of state, to the firt order of approximation, is then, using Eq. (12.7), 12.20) ‘This equation is setisGed with fainly good socursey by many real gases, especially for large values of V/s. ‘The ooeficienta a and b are called oan der Waals constants. They are given in Table 12-1 for several real gases. TABLE 12-1 Van der Waals Constants : *, 2 Substance | yy m¢ kg~* mole? | m® kg? mole“ aliom Bad6 x 105 0.02570 ‘Hydrogen 24,68 0.02661, Neos 2128 o.o1709 Nitrogen sao 0.03013, Oxygen ana 0.03183, Ammonis aaa onsi07 Carbon dioxide | 3628 0.08267 Sullurdioride | 6731, 0.05636, Water sia /5s¥ | 09900/55,5505 Thermal properties of gases. (24 RA Heat Capacity of an Ideat Monatomie Gas In Section 11.7 we defined the hest capacities of a substance at constant volume ‘and at eonstint pressure as where H = U-+p¥ is the enthalpy of the substance. In addition to their importance as coefficients in several practical calculations, the theoretical éalculation of the heat capacities of a substance affords a means of verifying the correctness, of the model chosen to desoribe the substance. In this section we shall ealeulate the hest eepacities of an ideal gas, and by comparing these with the observed values for real gases, we shall be able to conclude to what extent the ideal-gas ‘model is a good approximation. é ler an ideal monetomio gas. ‘The internal energy of such a gas is purely translational kinetieenergy, ond is given by U = §wRT. Using Eq, (12.21), we obtain Cy=4R 2.4715 J mole? °K-* = 2.9807 cal mole“! *O-, (4222) ‘Taking into account Eq, (12.6) (PY = SRT), we have that the enthalpy of an ideal gas is H=U-+pV = gk. ‘Thus Eq, (12.21) gives Cy = §R = 20.7858 J mole! °K 4.9678 eal molew 12.28) ‘Therefore all idesl monatomic gases have the same heat capacities, independent of the structure of the atoms. We may note, from the above results, that G—Cy=R, 42.24) so that C, is larger than Cy by the amount R. The reason for this is that Cy is related only to the change in internal energy, while C, includes, in addition, the expansion Work of the gas when its temperature increases 1 dogreo at constant. pressure, It is simple to verify that this work is exactly equal'to R. When the pressure of the gas is constant, p dV = NZd7, and if the increase of temperature ia one degree, the work done is, Wy [oar = [wear = wr. Hence the work’per mole done by the gas is F, From this proof wo see that {s valid for alfideal/gses, either monatomic or otherwise. ‘Eq. (12.24) is valid for allfdeal gases, cither monatomic or ot 124) Heat capacity of an ideal monatomic gas $05 Another relation among the heat capacities of an idesl monatomic gas is 1 Op/Cy = § = 1.667. (12.25) ‘This relation is followed rather closely by most fronstomie my wo in ‘Table 12-3 (at the end of this chapter). - a EXAMPLE 12.3. The equation of state of an ideal gas in terms of pressure, volume, ‘and entropy. Solutions Sinn entzopy ia» pope ofthe site of «gust can be used a 8 vial sate of Te aa, oh cr eapection. Fen Inp+ nV = laxR+InT. Coar . Dieretaion yids a oe, at (12.26) or ihe cso ofan ial en in which th intemal eierey depends sly on the tamperature, we have that dU = Cy dT. Therefore the first law of thermodynamic Eq, (11.87), dU = TaS — pa¥, gives ee xGy dT = Ta8— pav, from which (dividing by 7 and using the equation of state) we obtain a a nov = as — xR. 227) liinating 47/7 between Eq, (1226) and (1227) and ung the relation C — Cr = 2, eee pty NG" whete 7 = C,/Cy. Integrating, we obtsin -5. inp +1 = 55+ In (oot) PVT = (const)esr, (12.28) ‘which fs the equation of state of an ideal gas in terms of p, V, and S. If transforma. tion is adiabatic and reversible (Le, if it is isentropio), the equation reduces to PVT = const 2.29) ‘This equation finds many uses in processes involving gases.2 t = | conn ; histone Ba HD, = EE Te tite 506 Tharmal properties of gases 2s 12.5 Heat Capacities of an Ideal Polyatomic Gas = When the ideal ens is. nat mansio, 6 mustek the structure of the molecules into account when we ealeulate the Heat capacities. The energy of # polyatomic molecule is composed of three terms: translational energy, rotational energy, and. vibrational energy; that is,* : Bnotecste = Bes + Brot + Evi 12.20) ‘We shall ignore the electronic energy of the molecules becauie it seldom partici- ‘ates in the thermal excitation of the gas. Electronic excitation requires an energy Of the order of 1 eV at least; euch energy is about 40 times greater than the average thermal kinetic energy at room temperature (298°K), and therefore » temperature, of the order of 10¢ °K is required to produce a substantial number of molecules in Greed elovirgnie states (ove Example 10.4). At these temperatures, of eourse, mont of the gaa molecules are disociated by collisions ‘On the other hand, the rotational kinetic energy of polyatomic molecules is of “the onder 10724, and therefore molecules exn easly be carried to excited Tols- ‘onal levels, even at temperatures that are low compared with oom temperature, vi ‘nergies are nthe range of 10~* eV to 10-1 eV, and therefore at room enperature moleaules may be found in a few low-lying excited vibrational states. “To compute the contribution of the internsl motions to the heat capscity of & polyatomic gas, we must first (using statistical methods) find the distribution of the gas molecules among the rotational ond vibrational states. i elves to the simplest case of ¢ diatomio gas. ‘The rotational Iinetio energy of iatomie molecules, according to Eq, (6.12), is a whore Tis the moment of inertia of the molecule relative to a perpendicular axis, ‘passing through the center of mass and 1 determines the angular momentum of {he molecule relative to the center of mass. As explained in Section 5.7, the angu- Jer momentum may have 21-+ 1 diferent orientations, all with the same energy, to that the ge factor used in Eq, (10.28) is 21+ 1. ‘The equilibrium distribution of the molecules among the available rotational states, when we use Maxwell-Boltz- ‘mann statistics, is then am Beare nemennns = F are neem, a2.) where @, = A?/2Tk is called the characteristic temperature of rotation. Values of 8, ave given in Table 12-2 for a few diatomic gases. Looking at this table, we can ‘We shall assume, for simplicity, that the three energies are additive. However, strictly “peaking, there are some cross terme among the rotational and vibrational energies. 125) Heal eapacilze of an ideal polyatomic gas 507 1/04 a iareees eaneas SneeaR 1/88 1 7 77 1 7 Fig. 12-5, Occupation of rotational loves in © diatomic gas for three values of /@,. ‘ee thet even for hydrogen these values are very small compared with room temperature, The values of nro are illustrated in Tig. 12-5 for three values of 1/®,. ‘As T increases, the number of molecules in excited rotational states also increases, ‘The rotational partition function Z.4, according to the general definition in Bq. (10.22), is Bros = Sy Ql Yer VOT, (12.32) ‘Once we:compute the rotational partition function Z,91, we ean obtain the rota tional energy Uses of the gas by using Eq, (10.24) Ug = WT? aad. 2s508 Thermal properties of gases 25 TABLE 12-2 Characteristic Temperatures for Ro- tation and Vibration of Diatomic Molecules ; ‘Bubstance Hydrogen Carbon monoxide Oxygen Chlorine Bromine Sodium Potassium ‘The total internal energy of the gas is U = Ut Uno = BRT + Urey 12.34) since Ue = §NRT. To obtain the heat capacity at constant volume, Cy, we next substitute Eq, (12.34) in Eq, (12.21). ‘The heat oapacity incresses gradually with the temperature, as indicated in Fig. 12-6, because onergy is required not only to increase the translational energy of the molecules, but also to increase their rota~ tional energy by bringing more molecules to excited rotational levels. Figure 12-6 indiéates that, at a temperature high relative to @,, Cy levels off at about §2- o Fig. 12-6, Variation of heat capacity at #4) ‘constant volume of « diatomic gas due to ‘excitation of rotational levels. 2 o yon eine te At pests Si etn ein ir: wees mi ee 0 is Gy yo sl, etl ‘energy, and we may compute Zrot by replacing the summation in Eq. (12.32) by ‘an integration. We also substitute 2l for 21+ 1. and 2? for (@-+ 1). Then Bea = [Be OHP al = 7/8 225) Heat capacities of an idea polyatomic gas 509 ‘Thus In Zyot = In T — In @, and Eq, (12.88) gives Uso, = KANT = NRT, (12.35) sat = uP Tan tov woe D seven (20 ay 2 Cy =4R. (42.37) ‘Thus, at temperatures high compared with @,, stional kineti Se orec faites ee ae oe of, polyatomic molecules, tatoos’ toabeg & Ts mee one age ae ent foe on ot mee acael agree ‘according to Eq. (5.16), are given by Byip = (v-+ 4)hu,” Therefore, when we again use Maxwell-Boltsmann statistics and g; = 1, which corresponds to this wa Nett al tie = (12.38) where @, = hus/k is called the characteristic temperature for vibration. Its value {for a few diotomic molecules is given in Table 12-2. Note that in all cases @, is much larger than @,. Also in most cases @, is greater than room temperature. ‘Figure 12-7 shows the values of yi» for three values of 7/®,. As T increases, ‘the number of molecules in excited. vibrational levels also increases; but because ©, > @, the excited low vibrational levels begin to be appreciably populated at temperatures at which many rotational levels are already occupied. Tn some ceases the molecules dissociate at energies lower than those at which the higher vibrational levels contribute appreciably to the internal energy. ‘The vibrational partition function Zyiy is, Lay = eH ws MT (SD oF), ‘Using the expression for the sum’ of an infinite decressing geometric progression, De- and noting that in our case funetion (12.39) tetetet-e-= rs re fore <1, 5-67, we have for the vibrational partition eer Zan (12.40) oy

Dispense Di Fisica Moderna

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Dispense Di Fisica Moderna

Caricato da

Copyright:

Formati disponibili

Potrebbero piacerti anche