See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/20236686

The new International Temperature Scale of 1990 (ITS-90)

Article  in  Clinical Chemistry · April 1989

Source: PubMed

CITATIONS READS
3 428

1 author:

Billy W Mangum
National Institute of Standards and Technology
60 PUBLICATIONS   1,239 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Billy W Mangum on 27 August 2015.

The user has requested enhancement of the downloaded file.

CLIN. CHEM. 35/3, 503-505 (1989)

The New InternationalTemperature Scale of 1990 (ITS-90)

B. W. Mangum1

A new international temperature scale, the ITS-90, will re-
place the International Practical Temperature Scale of 1968
(amended edition of 1975), IPTS-68(75), on 1 January 1990.
Temperatures on the ITS-90 will agree more closely with
thermodynamic temperatures; therefore, the ITS-90 repre-
sents a substantial improvement over the IPTS-68(75). For-
tunately for the clinical laboratory community, the change in
the scale will be at most only 0.05 #{176}C
or less in the range from
0 to 60 #{176}C,
but corrections in primary calibrations should be
made so that the calibrations are based on the ITS-gO.
In a previous publication (1), the most recent internation-
ally adopted temperature scale, the International Practical
Temperature Scale of 1968 (amended edition of 1975),
designated IPTS-68(75), was discussed in some detail and a
brief description was given of its evolution. A new tempera-
ture scale will be implemented on 1 January 1990, the same
date that changes in some electrical quantities will be made
(2-6). The new scale (7),the International Temperature
Scale of 1990 (1TS-90), will supersede the IPTS-68(75) and
also a provisional low-temperature scale, the Echelle Provi-
tare values listed in Table 1 are estimates based on recently
published data (8-15). The dashes after the temperature
values indicate that the final values have not yet been
firmly and officially assigned, so the digits that will eventu-
ally be in the positions of the dashes are not yet given.
Standard Instruments and Temperature Determinations
The standard instruments for interpolating between the
defining fixed points of the ITS-90 will be:
1. The constant volume gas thermometer, for the region
from 3 K to the triple-point temperature of neon at about
24.6 K, and
2. The platinum resistance thermometer, for the region
from the triple-point temperature of equilibrium hydrogen,
at about 13.8 K, to the freezing-point temperature of silver,
at about 1235 K (962 #{176}C).
The kelvin, K, is the unit of Kelvin thermodynamic
temperature, T; it is defined as the fraction 1/273.16 of the
Kelvin thermodynamic temperature of the triple point of
water. Celsius thermodynamic temperature, t, with the unit
degree Celsius, #{176}C,
is defined as t = T 273.15 K.
-

soire de Temperature de 1976 entre 0.5 K and 30 K (EPT-

The iTS-gO will use both International Kelvin Tempera-
tures, T, and International Celsius Temperatures, t,
76).
The major changes of the ITS-90 from the 1PTS-68(75) with units kelvin, K, and degree Celsius, #{176}C, respectively.
These temperatures are related thusly:
will be:
#{149}
extending the temperature scale downward from 13.81 K t90 = T - 273.15 K.
to about 0.65 K
#{149}
eliminating the Type S thermocouple as a standard
Table 1. Defining Fixed Points of the ITS-90
instrument of the scale
Temp.raturec
#{149}
extending the range of the platinum resistance thermom- EquIlibrium
eter upward to the freezing point of silver Materiai stateb T, K tC
changing the reference point for radiation thermometry
#{149} 4He VP 4.2-- -268.9--
from the freezing point of gold to the freezing point of silver e-H2 TP 13.8-- -259.3--
#{149}
assigning temperatures on the scale that agree closely e-H2 VP or CVGT 17.0-- -256.1--
with Kelvin thermodynamic temperatures. e-H2 VP or CVGT 20.2-- -252.8--
Ned TP 24.5-- -248.5--
Defining Temperature Fixed Points 02 TP 54.3-- -218.7--
Ar TP 83.8-- -189.3--
Ten of the 13 defining (and alternative) fixed points of the Hgd TP 234.3-- -38.8--
IPTS-68(75) plus an additional seven new points will be H20 TP 273.16 0.01
used as the defining fixed points of the ITS-90. For all 17 Gad TP (or MP) 302.9-- 29.7--
fixed points of the ITS-90, new values of temperature, in md FP 429.7-- 156.5--
Sn FP 505.0-- 231.9--
close agreement with thermodynamic values, will be as- Zn FP 692.6-- 419.5--
signed. The 17 fixed points will be defined as equilibrium AId FP 933.4-- 660.3--
states of pure materials as given in Table 1; 14 of these Ag FP 1234.9-- 961.7--
states will be triple points or freezing points. The tempera- Au FP 1337.3-- 1064.1--
CUd FP 1357.6-- 108-4.5--
‘U.S. delegateto the ComiteConsultatifde Thermom#{233}trie
(C(Th Indicates equilibriumhydrogen,that is, hydrogen
#{149}eH2 withtheequilibrium
of the Comit#{233}
International des Poids et Mesures (CIPM); Group distributionof its ortho and pare states.
Leader, Thermometry Research and Calibration Services, Tem- bvp, vaporpressure point;CVGT, constant-volume gasthermometerpoint;
perature and Pressure Division, National Institute of Standards
TP, triple point; FP, freezingpoint;MP, melting point.
and Technology (NIST) [formerly-i.e., prior to August 23, 1988- cThe dashesaftermosttemperaturesindicate thatthe finaldigits for these
the National Bureau of Standards (NBS)], Gaithersburg, MD
20899.
valueshavenotyet been officially assigned.
dprevjouslythese weresecondary fixedpoints.
ReceivedNovember29, 1988;acceptedDecember 7, 1988.

CLINICAL CHEMISTRY. Vol. 35, No.3, 1989 503

 cate for any given point may differ slightly, but the differ-
Temperatures in the range from about 0.65 K to 5.0 K will
be determined from specified vapor pressure-temperature ences, for all practical purposes, will be negligible.
relations of 3He and (or) 4He. Those in the range from 3 K
Differences between t and t
to about 24.6 K will be determined by means of a constant-
volume gas thermometer calibrated at three points, namely, Estimates of the magnitude of the changes to be expected
the triple point of neon, the triple point of equilibrium between temperatures on the ITS-90 and those on the JPTS-
hydrogen, and at some temperature in the region from 3 K 68(75) in the region of temperature of interest to the clinical
to 5.0 K, the value of the selected point being determined laboratory community are given in Table 2. The tabulated
from the vapor pressure-temperature relation of helium. values are approximate, subject to change, and are more or
Those in the range from the triple-point temperature of less the means of the published values (8-15). Figure 1
equilibrium hydrogen to the freezing-point temperature of shows an approximation of the differences over the range of
silver will be determined with platinum resistance ther- the fixed points of the ITS-90.
mometers calibrated at the defining fixed points in the Although clinical laboratories use liquid nitrogen (77 K)
range of these thermometers. Temperatures in the range for preserving some things for long times, the region of
above the freezing-point temperature of silver will be deter- precision thermometry in such laboratories usually doss not
mined by means of radiation thermometry, referenced to the extend below -223 K (-50 #{176}C). Consequently, the low
freezing-point temperature of silver (or, if one chooses,the temperature end of the 1TS-90 is of little concern to these
freezing-point temperature of either gold or copper), and laboratories; the platinum resistance thermometer is the
Planck’s Law of Radiation. In all cases, temperatures of the standard instrument in the temperature range of interest to
1TS-9O will agree closely with thermodynamic tempera- them. Fortunately, the changes in the temperature values
tures. in the region of precision thermometry used by the clinical
In the temperature range from about 13.8 K to 24.6 K, the laboratory community are less than 0.05 #{176}Cand, thus, will
1TS-90 will have both the constant-volume gas thermometer have negligible effects for most applications. Nevertheless,
and the platinum resistance thermometer specified as stan- the changes are measurable with good-quality clinical labo-
dard instruments. Either or both of these instruments may ratory thermometers and the new temperature values on
be used, being of equal status. In the region of overlap, the 1TS-90 should be used.
however, the temperatures the two instruments may indi-
Secondary Fixed Points
The Consultative Committee on Thermometry of the
Table 2. Approximate Differences between International Committee of Weights and Measures (CIPM)
Temperatures (tso) on the ITS-90 and Temperatures (tes) has in the past published (16) accepted temperature values
on the IPTS-68(75) in the RegIon of Interest to Clinical
Laboratories
T,K C -,Cx1O3
50 -223 -6.5
58 -215 0.0
73 -200 8.0
100 -173 9.5
123 -150 12.0
150 -123 13.8
173 -100 13.6
200 -73 12.0
223 -50 9.2
250 -23 4.5
273 0 0.0
300 27 -5.8
323 50 -11.5
350 77 -19.0
373 100 -26.0 t(c)
423 150 -32.0 Fig. 1. Differences between Celsius thermodynamic temperature
473 200 -37.0
vaues, and those on the IPTS-68(75), t, as a function of

Table 3. Major National Standards Laboratories That Will Calibrate Thermometers on the ITS-90
Country city Laboratory
Australia Lindfield CSIRO NationalMeasurementLaboratory(NML)
Canada Ottawa NationalResearchCouncil(NRC)
F.R.G. Braunschweig, Berlin Physikalisch-Technische Bundesanstalt(PTB)
France Pans Institut Nationalde Metrologie(INM)
G.D.R. Berlin Amt f#{252}r
Standardisierung,Messweseni.mdWarenprufung(ASMW); DeutschesAmt f#{252}r
Mass und Gewicht (DAMG)
Italy Turin Istituto di Metrologia “G. Colonnetti” (IMGC)
Japan Ibaraki National Research Laboratory of Metrology (NRLM)
P.R.C. Beijing NationalInstituteof Metrology(NIM)
U.S.S.R. Moscow Physico-Technical and Radio-TechnicalMeasurementsInstitute(PRMI)
Leningrad 0. I. Mendeleev Instituteof Metrology(VNIIM)
U.K. Teddington National Physical Laboratory (NPL)
United States Gaithersburg,MD NationalInstituteof Standardsand Technology(NIST)

504 CLINICAL CHEMISTRY, Vol. 35, No. 3, 1989

 of secondary fixed points, some of which were listed in Table
2 of reference 1. As can be seen by comparing that table with
Table 1 of this article, some of the points listed before will be
defining points of the ITS-90. Some new fixed points will be
listed in the next publication of secondary fixed points, and
some of these will be of considerable interest to the clinical
laboratory community. Included will be materials such as
the following:
#{149}
succinonitrile, with a triple-point temperature at
58.080 #{176}C
#{149}
n-docosane, with a triple-point temperature at 43.891 #{176}Cthe electrical units. IEEE Trans Instrum Mess 1987;JM-36:672--5.
#{149}
1,3-dioxolan-2-one (ethylene carbonate) with a triple-
point temperature at 36.324 #{176}C
These values are on the IPTS-68(75) and will be converted to
the appropriate values on the ITS-90.
Cells for realizing the defining fixed-point temperature of
gallium and triple-point cells of succinonitrile are available
from the Office of Standard Reference Materials of the
National Institute of Standards and Technology as Standard
Reference Materials 1968 and 1970, respectively; triple-
point cells of ethylene carbonate and n-docosane will be
available as Standard Reference Materials 1972 and 1973,
respectively, in 1989. Table 3 lists the major national
standards laboratories that will calibrate thermometers on
the 1TS-90.
Some of the work involved in developing the n-docosanefixed
point wassupported by an interagencyagreementfrom the Nation-
al Cancer Institute, Agreement No. Y01-CM-70140. The Officeof
Standard Reference Materials provided somesupport for the work
involved in developing new Standard Reference Materials 1972and
1973 (ethylene carbonate and n-docosane, respectively).
References
1. Manguin BW. The gallium melting-point standard: its role in
our temperature measurement system. Clin Chem 1977;23:7 11-S.
2. Taylor BN. History of the present value of 2e/h commonly used
View publication stats
for defining national units of voltage and possible changes in
national units of voltage and resistance. IEEE Trans Instrwn Meas
19871M-36:659-64.
3. KoseV. Changes in electrical units? Contributions from national
standards laboratories in Europe. IEEE Trans Instrum Mess
19871M-36:665-7.
4. Shida K. Circumstances in Japan forthe proposed changesin the
electrical units. IEEE Trans Instrum Meas 1987;IM-36:2668-9.
5. Skinner AD. Changes in the electric units and the effect on
industry in the U. K. IEEE Trans Instrum Mess 1987;IM-36:670-1.
6. Jaeger KB, Taylor BN. U.S. perspective on possiblechanges in
7. International Temperature Scale of 1990. BIPM Corn Cons
Thermom#{233}trie
1987;16:Document CCT/87-22A.
8. Martin JE, Quinn TJ, Chu B. Further measurements of thermo-
dynamic temperature using a total radiation thermometer: the
range -130#{176}C
to +60 #{176}C.
Metrologia 1988;25:107-12.
9. Edsinger RE, Schooley JF. Gas thermometric determinations of
t(KTTS) vs t(IP’1’S-68)in the range 230#{176}C
to 660 #{176}C.
10. Guildner LA, Edsinger RE. Deviation ofinternational practical
temperatures from thermodynamic temperaturesin the tempera-
ture range from 273.16 K to 730 K. J Res NatI Bur Stand
1976;80A:703-38.
11. Astrov DN, BelyanskyLB, Dedikov YA, Polunin SP, Zacharov
AA. A thermodynamic temperature measurementin the 20 K to
300 K range. BIPM Corn Cons Thermometrie 1987;16:Document
CCT/87-62.
12. Kemp RC, Kemp WRG, BesleyLM. A determinationofthermo-
dynamic ternrratures and measurements of the secondvirial
coefficientof He between 13.81 K and 287 K using a constant-
volumegas thermometer.Metrologia 1986/87;23:61-86.
13. Steur PPM, Durieux M. Constant-volumegas therrnometry
between 4 K and 100 K. Metrologia 1986;23:1-18.
14. WeberF. Gasthermometermeasurementsat 27K, 54K and90
K. BIPM Corn ConsThermom#{233}trie 1984;15:DocunientCCT/84-6.
15. MoldoverMR, Trusler JPM. Accurateacousticthermometry I:
the triple point of gallium. Metrologia 1988;25:165-87.
16. The International PracticalTemperatureScaleof 1968,amend-
ed edition of 1975. Metrologia 1976;12:7-17.
CLINICAL CHEMISTRY, Vol. 35, No. 3, 1989 505