APPENDIX

(Published Papers)
 Jcnr. Indian Cheni. Soc., VoL 33, No, 7, 1956]

COMPLEX COMPOUNDS OF BI-POSITIVK SILVER WITH PYRIDINE

CARBOXYLIC ACIDS. PART I. ARGENTIC
NICOTINATK AND tsoNICOTINATK

By Bireswar Banerjee and Priyadaranjan Ray

Complex compounds of bi-positive silver with nicotinic and fionicotinic acids have been prepared
in the form of cinnamon-red crystals by the action of potassium persu’phate on silver nitrate and
the organic ligand in aqueous medium. The compounds are paramagnetic with ,u„ = 173 Bohr
(approx.). A study of the X-ray diffraction of the crystals of Ag11 - and-Cu11 • /sonicotinates furnishes
evidence in support of their isomorphous character.
It is suggested that the configuratirn. of the complexes is best'represented bv assuming that their
molecules are either dimeric, or polymeric with the formation of an unending two dimensional sheets
or layers. They are - to he regarded as inner-ruetallic penetration complexes with planar dsp1
hybrid bonds. ... .

Bi-positive silver is known to form a very stable chelate complex of the inner-
metallic type with pyridine-2-carboxylic acid (picolinic acid). This was first studied
by Barbieri (Atti Accad. Lincei, 1933, 17, 1078) and was fcund by Cox, Wardlaw
and Webester (/. Chem. Soc;, 1936, 775) to be isomorphous with copper picolinate.
Burada (Ann. Sci. Univ. Jassy, 1935, 20, 71) also reported the preparation of a red
crystalline bi-positive silver complex of quinolinic acid (pyridine-2:3-dicarboxylic
acid). In both these acids, the functional groups —N and —COOH are in very
favourable positions to give rise to'stable five-membered chelate rings by co-ordination
with the silver (II) atom. With a view to studying whether these functional groups at
other positions of the pyridine ring can at all lead to the formation of any inner-
metallic complexes with bi-positive silver, a systematic investigation with all the
available mono-, di- and tri-carboxylic acids of pyridine was considered desirable. A
preliminary qualitative examination with nicotinic acid (pyridine-3-carboxylic acid),
wonicotinic acid {pyridiue-4-carboxylic- acid) and cinchomeronic acid (pyridine-3:4.
dicarboxylic acid) gave very promising results. In the present communication
argentic complexes of pyridine monocarboxylic' acids, nicotinic'and fsonicotinic acids
have been described. Both nicotinic and fsonicotinic acids, on being treated with
silver nitrate and potassium persulphate solution, furnished insoluble cinnamon-red
crystalline complexes with a paramagnetic moment (ca. 1.73 ys), corresponding to
the presence of one unpaired electron, characteristic of bi-positive silver. Of these,
argentic isonicotinate has been found to be isomorphous with the corresponding cuprjc
compound from an X-ray study of their powdered crystals. A similar examination
of the nicotinate complexes (Chackraburtty and Banerjee, Indian J. Phys., 1055, 29,
357) also indicates a general resemblance between the argentic and cupric compounds,
though no definite statement can be made about their isomorphous character,
8—1914P—7
 N

504 B. BANERJEE AND P. RAY

The structures of these argentic complexes present a rather intriguing problem.

Their properties indicate that they should be viewed as inner-metallic penetration
complexes with planar dsp% hybrid bonds. Both show the same composition
Ag(C6H40,N)2:
N__ • __ '
Ag«^ ^COO)* and Ag(N<( ^>—COO),

Agu -liicotinate. Agn ■ fsoDkotixiate.

On the basis of these simple structures, 1:3 or 111 eta positions in the nicotinate
and 1:4 or para positions in the Eonicotinate complex are to be bridged by the silver atom
with formation of 6- and 7-membered chelate rings, which are rather improbable and
are unknown in the 'case of organic compounds. But the possibility of the meta
positions of nicotinic acid being bridged by the silver atom cannot be altogether
excluded in vew of a very large single-bond covalent radius (1.53A) of the silver atom,
.which is almost twice that of carbon (0.77&). But such an interpretation cannot be
•seriously entertained in the case of isonicotinic acid, where the para positions have
to be .bridged. Plausible configurations for these complexes can, however, be
represented by assuming that their molecules are either dimeric as in (I), or polymeric
as in (II) with the formation of an unending two-dimensional sheets or layers. The
silver atoms in these molecules occupy the centre of a square, being co-ordinated with
N- and O- atoms of the acids in one plane.
Fig.

Fig. i

oc- -w* -w-

>Ag-< -NC H
5 4
NC H.
f 1 coV-'P N- 0

f
\ Ml 1
1 / i CO
— -1
CO
'A: tyl,
N°
'0 — co —W .0
1
-HCN—>Ao-*- NC R—
i
(I) H 5 j* 5 4 co-o-Aff-
r
e*H
5n4
(II)
It can therefore' be stated that all the three pyridine monocarboxylic acids (a, A
and y) afford bi-positive chelate silver complexes of the inner-metallic type with similar
properties.
Experimental

Argentic Nicotinate.—Nicotinic acid (3 g.) was dissolved in 50 c.c. of water (if

necessary by heating) and mixed at the room temperature with a solution of silver
 COMPLEX COMp6uPd8 OF BI-POSITIVE SILVER &0£>

nitrate (1.36 g. in xo c.e.) when silver (I) nicotinate separated as a precipitate. A

solution of potassium persulphate (6 g. in the minimum quantity of water) was after­
wards added dropwise to this mixture with constant stirring. The stirring was continued
for about three hours by means of a magnetic stirrer in the cold. A chocolate-red crystal­
line powder formed the final product. This was filtered through a sintered-glass bed,
washed with cold water till free from sulphate, and then dried in a vacuum desiccator
over CaClj ; yield 3.5 g.
The substance forms brick-red, fine, needle-shaped prismatic crystals, insoluble
in water, but readily decomposes in dilute mineral acids or alkalies. With cold
dilute acetic acid also it suffers decomposition on warming. The substance can be
heated to 1150 without any appreciable decomposition. It remains unchanged, when
kept dry, for months. <u? = 1.74. Prom an acid solution of potassium iodide it
immediately liberates iodine. [Found : Ag (oxidimetric), 30.2 ; Ag (gravimetric as
AgCl), 30.4 ; N, 8.r. Ag(C6HjN02)3 requires Ag, 30.68; N, 7.95 per cent].
Tables I and II summarise the results of measurement on the X-ray pictures of
copper and silver complexes of fsonicotinic acid.

Table I

Copper isonicotinale.

$. Visual. d. e. Visual d. e. Visual d.

intensity. intensity. intensity.

5*27' v.s. 8.10A 13*54' w. 3.21k ° f

20 12 w. 2.23ft

7°io' w. 6ni7 14*37' w. 3-05 m. 2.11

9*45' m. 4-55 15*11' w. 2-S4 22a30' w. 2.01

11*03' s. 4.03 16*30' w. 2.74 23*56' w. 1.90

n*54' ni. • 3-74 16*54' w. 2.65 24*30' w. 1.86

13*53' w. 3.45 18*55' w. 3-37 26*39' w. 1.72

Table II
Silver^ iso nicotinate.
9. Visual d. 9. Visual d. 9. Visual d.
intensity. intensity. intensity.
>*
M
00

5*27' v.s. s.ioA 14*02' w, 17*20' w, 2.58ft

7*10' w. 6.17 15*03' \v. 2.97 18*46' w. 3-39

9*53' in. 4.49 16*02' w. 3.79 20*55' m. 2.16

io?53.' s. 4.0S 16*37' w, 3.69 23*13' in. 2.04

12*02' s. 3-69 23*47' w. 1.91

Argentic isoNicotinate.— isoNicotinic acid (1 •87 g.) was dissolved in 100 c.c. of
boiling water and treated with a solution of silver nitrate (0.87 g. in 5 c.c.). On
cooling, silver (II wonicodnate separated as a fine precipitate. The mixture was then
506 B. BAFERJBB AND P. RAY

treated dropvvise with constant stirring with a solution of sodium persulphate (3 g.)
in the least quantity of water. The stirring was continued with the help of a magnetic
stirrer for about 4 hours in the cold, when the white silver (I) tsonicotinate was completely
converted into cinnamon-red crystals of the bi-positive silver complex. The crystals
were filtered, washed and dried as iu the previous case ; yield 1.5 g.
Argentic isonicotinate resembles the corresponding uicotinate complex more or less
closely in all its properties, = 1.60. [Found: Ag, 30.6 (oxidimetric), 30.8
(gravimetric) : C, 41.52. Ag (C8HjNOj)3 requires Ag, 30.7 ; C, 40.9 per cent].
Tim powder X-ray photograph of the substance as well as of the corresponding
copper isonicotinate. CulCuHjNOJj, were taken in a cylindrical camera of radius
'5.0 cm. by using copper radiation from a Haddiug tube run at a voltage of 40. kv.
with'a tube current of 15' 111a, [cf. Fig. 3 (a. and b)]. Fig. 4 (a and b) shows-the
powder photograph of copper and silver complexes of nicotinic acid.

Fig. 3a
Cu-Iso
uicotinate

,‘Fig. 3b Agn .

Cu-nicotinate
Fig. 4a

AgH-

Fig . 4b

It is evident from the tables that the d values and also the intensity of the
corresponding powder lines of the two wonicotinate complexes are almost identical,
indicating their structural similarity.
The' authors desire to express here their best thanks to Dr. D. R. Dasgupta of
the Department of Magnetism of this Association for the powder photographs
and to Shri S. Dutta Ray of the same department for magnetic measurements of the
substances.

Department of Inorganic Chemistry,

Indian Association for the Cultivation' . -. ' Received February 15, 1956.
of Science, Calcutta-32.
[Jour. Indian Cbem. Soc , Vol. 34, No. 3, 1957]

COMPLEX COMPOUNDS OF BIPOS1T1VE SILVER WITB PYRIDINE

CARBOXYLIC ACIDS. PART II. ARGENTIC COMPOUNDS OF
PYRIDINE DICARBOXYLIC ACIDS

By Bireswar Banerjeb and Priyadaranjan Ray

In continuation of our previous work on bipositive silver complexes of pyridine monocarboxylic

acids, we have now been able to prepare the argentic complexes of all the dicarboxylic acids cf
pyridine with the exception >f that of dinicctinic acid, which is not at hand for the present.
Argentic complexes of quinolinic, cinchomeronic, fsocinchomeronic, lub'dinic and dipicolinic
acids are formed by the oxidation of silver nitrate in presence of the corresponding acid by alkali persul­
phate in aqueous medium. They form red to dark red, finely divided crystals, more or less sparingly
soluble in water. The argentic complex of dipicolinic acid has, however, been obtained in two
different colours-dark red and green, the last one being produced at a lower temperature. The com­
position of all these complexes is given by the general formula Ag'XH'j (where XH3 = one molecule of
the pyridine dicarboxylic acid). With-the exception of cinchomeronic acid (3 : 4), all the complexes
may be represented by a monomeric structure like that of the argentic picolinate. The structure of
the argentic cinchomeronate resembles possibly that of nicotinic or fsonicotinic acid, being either
dimeric, or polymeric with the formation of unending two-dimensional sheets and layers. The
complexes are all decomposed by acids and alkalies.
All these argentic complexes are paramagnetic and give moment values of approximately x.73 mb .
X-ray powder photographs show that the green and chocolate modifications of the argentic complex
of dipicolinic acid differ in their crystal structure. Cupric and argentic quinolinates crystallise with
different .water molecules and hence show, as might be expected, different X-ray patterns.

In a previous paper (this Journal, 1956, 33, 503) an account of the preparation,
properties and constitution 0: the argentic complexes of pyridine monocarboxylic acids,
like nicotinic and isonicotinic acids, has been given, which, with the argentic
picolinate previously described by . Barbieri (Atti Accad. Lined, 1933, 17, 1078),
show that all the three pyridine monocarboxylic acids (a, ft and y) are capable of
forming chelate inner-metallic complexes with bipositive silver. It has now been
observed that pyridine dicarboxylic acids' also behave in a similar manner. Of the
six different pyridine dicarboxylic acids, argentic complexes with five of these, viz.,
quinolinic (2:3), cinchomeronic (3:4), fsocinchomeronic (2:5), lulidinic (2:4) and
dipicolinic (2:61 acids, have been prepared for the present. It has, however, not yet
been possible to obtain the remaining 3 :5-dicarboxylic acid of pyridine (dinicotinic
acid) in sufficient quantities to investigate the formation of its argentic complex.
The argentic complexes of all the above pyridine dicarboxylic acids are formed by
the oxidation of silver nitrate in the presence of the acid concerned by means of sodium
or potassium persulphate in aqueous medium in the cold. 'All the complexes excepting
that of 2:6-acid (dipicolinic acid) form red to dark red crystalline products ; while
the latter gives rise to two different coloured forms, green and dark chocolate. Their
composition is represented by the general formula, Agn(XH)a (where XHa = one
molecule of the dicarboxylic acid). One of the -COOH groups of each ligand molecule
2U8 B. BANERJEE AND P. BAY

remains obviously free in the complex. All the complexes, with the exeeptioE"of
that of cinchomeronic acid (3:4), are possibly represented by a monomeric structure
like that of argentic picolinate. The structure of the argentic complex of cincho­
meronic acid is presumably related to that of nicotinic or tsonicotinic acid, being
either dimeric, or polymeric with the formation of unending two-dimensional sheets
or layers, (cf. Banerjee and R&y, loc. cit.}. All the argentic complexes, as expected,
are paramagnetic with a moment value of approximately 1.73 fiB.
Diibsky and Ok&e (Coll. Czech. Chem. Com., 1931, 3,465), Eukes and Jurecek (ibid.,
1948, 13, 131), as well as others have' described Cu11 salts of many of these pyridine
dicarboxylic acids. They have, hdwever, found that in most cases the composition
of the copper compound is given by one atom of copper combining with one of the
acid molecule. But in the case of lutidinic (2:4), quinolinic {2:3) and dipicolinic
(2 :6) acids, copper compounds with two molecules of the acid per atom of copper have
also been described. The compositions of these latter differ from their corresponding
argent,c compounds, as described here, in their water of hydration, which presumably
accounts for their different X-ray pictures obtained for the compounds of quinolinic
acid (2:3). Burada (An. Sci. Univ. Jassey, 1935, 20, 71) described a silver" quinoli-
nate with one molecule of water, resembling the corresponding copper compound.
We have, however, failed to verify this, as the silver complex prepared by us has
always been found to. contain two molecules of water of hydration.

ExPERIME N'T A L

All the complex argentic compounds of pyridine dicarboxylic acids were prepared
by adding the finely pov;dered acid to the solution of sodium or potassium persulphate
almost simultaneously with silver nitrate solution, while the mixture was constantly
stirred by means of a magnetic stirrer. The stirring was continued for about 3 hours
for complete oxidation. It should, however, be noted that silver nitrate and the acid
should not be allowed to come together in the absence of persulphate, as in that case
highly insoluble univalent silver salt of the pyridine carboxylic acid is formed, which
it is very difficult to oxidise.
1. Argentic Quindinate.—It was obtained by -Burada (loc. cit.) by oxidising
silver' quinoline nitrate, Ag(C9HrN)2N03, with an excess of ammonium persulphate
in aqueous solution. The composition of his compound is given by AgI! (XH)2.H30.
In the present case, quinolinic acid was prepared by the oxidation of quinoline
with alkaline potassium permanganate solution on the water-bath (Koenigs, Ber.,
1879, 12, 983). For the preparation of the argentic quinolinate, quinolinic acid
(0.6 g.) in a - finely divided state was added to sodium persulphate (3 g.), dissolved
in the minimum quantity of cold water, followed immediately by a solution of silver
nitrate (0.3 g.) in a little water. The mixture was stirred continuously for 3 hours.
The colour of the mixture gradually turned orange-red. When no further change
had occurred, the product was filtered through sintered glass-bed filter, washed with cold
water till free of sulphate, and then dried in a vacuum desiccator, placed in a refrigerator.
Yield.0.7-0.8 g- .
 COMPLEX COMPOUNDS OF B±POSITIVE SILVEg, $09

The orange-red crystalline product is very sparingly soluble in water. It is

quite stable in dry air in the cold and remains unchanged even after 3 months. The
substance is readily decomposed by acids and alkalies ; it liberates iodine from acid
solutions of potassium iodide. [Pound : Ag(oxidimetric), 22.41; Ag(gravimetric), 22.83!
C, 35.81. Ag(CTH4NOjl3.2HsO requires Ag, 22.69 ; €,35.30%]. The substance, was
found to be paramagnetic. Xm =2.236 x io-6 ; xif (corr. 1 = 1247 x io~6 ; /xE =1.743.
2. Argentic Cinchomeronate.-—Cinchomeronic acid was prepared by the oxidation
of quinine in a boiling mixture of concentrated and fuming nitric acid for about 150
hours (Kass, Monatsh., 1902, 23, 252).
The argentic complex was prepared exactly in the same way as the argentic
quinolinate described above, yield 0.8 g. The substance forms dark red crystals and is
quite stable in the dry air for months. It behaves like the previous compound. It does
not decompose even when heated to 70° for one hour and' is unaffected by light.
[Found : Ag(oxidimetric), 24.41 ; Ag(gravimetric), 24.70; C, 38.98. Ag(CTH4N0.i)i
requires Ag, 24.54 ; C, 38.18%]. Xm - 2.434 x io~‘ > xm (corr..) = 1232 x io~6 ; ^ = 1.742.
3. Argentic isoCinchomeronaie — isoCinchomeronic and lutidinic acids were pre­
pared by the oxidation of a mixture of 2:57 and 2 :4-dimethylpyridine (commercially
available) with potassium permanganate solution on a water-bath with mechanical
stirring (cf. Garkow, Bet., 1904, 37, 2063 ; Tadenburg, Ber., 1885, 18, 915). The di-
methylpyridine mixture (10 g.) was oxidised with' KMnO* (72.g.) in aqueous; solution.
After complete oxidation, when the colour of the permanganate, disappeared, the mixture
was filtered from the precipitated oxides of manganese. The. clear colorless filtrate
was cooled to 10° and acidified with HCI (indicated, by Congo-red), when a-heavy
white precipitate of the mixture of 2 .-4-and 2:5-pyridine diearboxylic acids separated
out. These were then separated from each other by repeated extraction with hot
water. The 2:4-acid (lutidinic acid), soluble in hot water, was freed from the
admixture of 2:5-acid (tsocinchomeronic acid), the latter is difficultly soluble even in
boiling water. The purity of the products was verified by determination, of their
melting points.
For the preparation of argentic complex of tsocinchomeronic acid, 0.413 g. of
the acid was employed for 0.3 g. of AgN03 and 3 g. of sodium persulphate ; yield'
0.6 g„ The preparation was carried out at 30° and' the stirring was continued for
about 5 hours. As the acid is highly insoluble, any excess should be carefully,
avoided.
The substance forms brick-red powder, insoluble in water; it is quite stable in air
and remains unchanged for months. Its properties are similar to those of the previous--
compounds. [Found: Ag (oxidimetric), 24,21; Ag-(gravimetric), 24.36;. C 38.12-
Ag(C7H4NOi)2 requires Ag, 24.54 > C, 38.18%]. x® =2.346 x io-6; xu (corr.) = 1194 x io-6;
/j-b —1.712.
4. Argentic Lutidinate.—The complex argentic compound of lutidinic acid (2-.4)
was prepared like the previous compound, using the same quantity of materials.
The colour and-other properties are closely allied to those of argentic t'socinchomeronate.
 210 B. Bakebjbb And p. ray

[Found: Ag (oxidimetric), 24.18 ; Ag (gravimetric), 24.66; C, 38.08. Ag(CjH4N04)3

requires Ag, 24.54 ; C. 38.18%]. Xm = 2.713 x io-6 ; xm (corr.) = 7091 x io~s; ;uB= 1.644.
5. Argentic Dipicolinale.—'a) Green. This was prepared like the previous
compound, but the reaction temperature was maintained near about 15°. The mixture
turned dark green with the separation of fine green crystals. Yield is almost the same
as that of the argentic quinolinate.
• The substance is moderately soluble in water, giving a green solution. In the
solid state it is rather unstable except in the cold and dry state. When kept at the
room temperature it gradually turns dark grey. [Found : Ag (oxidimetric), 21.75 ;
Ag (gravimetric), 20.99 ; C, 33.02. Ag(CrH4N04)2.4H50 requires Ag, 21.09;
0,3281%]. x= = 2.2 x io~6 ; xu (cor*.) = 1252 x 10; /<» = 7.747.
(b) Chocolate.—The preparation of this variety is similar to that of the green
form, with this difference that the temperature is maintained at about 250. From the
green solution the product separated as a dark chocolate crystalline powder.
The substance, however, dissolves in water forming a green solution. It is
somewhat more stable than the green form. In all other respects they behave alike.
[Found: Ag (oxidimetric), 22.55; Ag (gravimetric), 22.83; C, 36.20; H, 2.65.
Ag(CrH4N04)3.2H20 requires Ag, 22.69; C, 35.30; H, 252%]- Xm = 2.441 x io"6 ;
Xm (corr.) = 1345 x 10"8 ; = 1.807.
The difference in colour of these two modifications is possibly caused by their
difference in hydration, and the consequent difference in their crystal structure as
revealed by their X-ray powder photographs (vide infra). There, is no evidence,
however, to prove that they might be related as cis-lrans isomers of one and the same
compound.
Table I
Argentic quinolinate, 2H20 (Fig. 1).
e. d. Intensity. e. d. Intensity. e. d. Intensity.
4*42 9.4004 8q 14*8 3-1546 2 21*26 2.1064 5

6°i8 i4°53 21 “43 2.0817

7.0192 25 3.0021 5 7

7*20 6.0354 40 i5°ix 2.9222 22*14 2.0356 25

7

8*5 5.4780 S5 16*7 2.7749 25 23°3 1.9673 ■ 5 •

9*3 4.8968 X i6°5* 2-6547 85 24*3° 1.8574 20

9°49 4-5178 30 77*25 2-5735 2 25*54 17634 75

10*27 4 2468 IO i8°I2 2.4661 26*21 3 ■
5 7-7354

n°30 3.863; 100 i8°54 2.3779 7 26*59 1.6971 2

12*15 3.6302 25 79°56 2.2529 25 S7°24 1-6737 I

12*21 3-6013 85 20*1 2.2502 27*52 1 6479 X

5
I3°45 3.2407 E 20°30 2.1994 28*22 1.6313
25
BANERJEE AND RAY

Fig. l

Fig. 2

fig. 3

Fig. 4
 COMPLEX COMPOUNDS OF BIPOSITIVE SILVER 21:

Table II

Copper quinolinate, h2o (Fig;, 2)..

e. d. Intensity. e. d. Intensity. 0. d. Intensity,

s‘52 7.5362 95 15*21 2.9098 25 21*49 2.0726 25
6*18 7-0192 16*21 2.7363 10 22*48 1.9877
95 7
6*31 6.7871 15 16*58 2.6396 23*12 2
5 1-9552
6*55 6.3963 20 17*24 2.5758 10 24*l6 1.8742 2

II°2I 3 9197 100 17*56 2.5016 10 24*45 1.8398 . 5

11*58 3 7154 18*25 2.4381 25*12 1.8091 5

15 5
12*52 19*25 2.3170 25*58 1-7592 7
3-4591 35 15

iJ°i5 3-3607 19*58 1 27*10 1.6871 20

5 2.2557

i4°x ' 3-1803 7 20*6 2.2412 10 23*3 1.6380 12

M'58 2-9825 2 21*10 2.1332 29*9 1-5813 18

5

Table III

Argentic dipicolinate, 4HS0 (green) (Fig. 3).

$, d. Intensity. e. d. Intensity. e. d. Intensity.

4*21 10.1570 n”53 3.7406 2 20*36 2.1892 10

9°
6*9 12*17 3.6204 1 22*3 2.0518 1
7-I903 95
7*12 6.1456 13*23 100 22*39 2.0002 1
15 33277

5-4556 14*29 3.0798 5 23*10 1.9578 10

8*7 i

2.8975 1 24*37 1.8492 3

9*5 4.8789 25 15*25
10*19 16*28 2-7173 7 25*13 I.8080 1
4-30H 25

2.4948 26*10 I.7466 10

] 1*18 3-931° 20 i7°59 10

19*25 2.3170 JO

Table IV

Argentic dipicolinate, 2H30 (chocolate) (Fig. 4).

0. d. Intensity. 6. d. Intensity. e. d. Intensity.

4* 18 85 11*48 3.7666 95 20*9 2.2359 40

10.273

5*36 7-8933 95 12*41 3 5081 1 31*35 2.0939 40

6*4 7.2886 13*36 22*33 2.0086 5

1 3-2757 95
60 14*44 3.0287 23*19 1.9460 15
7*33 5-8625 3

8*g
5-4335 60 15*27 2.8914 I 23*54 1.9012 •5
9*1 4.9147 1 16*22 I 24*Si 1.8329 5°
2-7335
100 17*8 2.6147 40 25*31 1.7881 3
9°44 4-5S6i
10*43 4.1486 3° ■i8*6 2-4793 40 26*9 1-7477 30
18*48 2.3901 10
8—i94?P—3
212 B. BANERJEE AND ,P. BAY

X-ray measurements.—X-ray powder. . photographs of argentic quin

Ag(C7H4N04)3.2H20, and of the corresponding copper compound, Cu(CrHjN04'
as well as. of the green' and chocolate modifications of argentic dipicolinate wer
in a-cylindrical 9 cm Unicam camera with Cu.Ka-radiation from Raymax demo
rotating anode tube at a voltage of 30 kV with a tube current of 50 mA. T
of exposure in each case was 3 hours.
It is evident from the measurement given above that cupric quinolina
argentic quinolinate differ in their crystalline structure, as might be expected fr<
difference in their water of hydration. The two differently coloured modificat
argentic dipicolinate also show different structures, as already stated.
The authors desire to express their best thanks to Mr. B. Bishui ol
and Ceramic Research Institute for the X-ray photographs, and to Dr. Atmara
Director of the Institute, for necessary permission.
Department of Tnoxganic Chemistry, .
Indian Association for the Received Septemba
Cultivation op Science,
Jadavpur, Calcutta-33.
 SONDERDRUCK AUS
DIE
NATURWISSENSCHAFTEN
SPRINGER-VERLAG / BERLIN ■ GOTTINGEN ■ HEIDELBERG
1955 HEFT 14, S. 416/17 42. JAIIRGANG

lodoxlne as a New Analytical Reagent.

Iodoxine (5,7-Diiodo-S-hydroxyquinoline)1) like Chloroxine
(5,7-dichloro-8-bydroxyquinoline) *) and Bromoxine (5,7-ci-
bromo-8-hydroxyquinoline)3) has been found to react with
different metallic ions in solution giving characteristic colours
and precipitates. Iodoxine possesses the same reactive
grouping as S-Hydroxyquinoline (Oxine)4) and exhibits many
of the analytical characteristics of the latter. Consequently a
large number of metallic ions have been found to be estimated
or detected by the use of this new reagent. The precipitates
and colour reactions of this new reagent with various metallic
ions have been fully investigated. The reagent (m.p. 208°),
Iodoxine, is insoluble in water and only sparingly soluble in
alcohol, ether, acetone, and chloroform. The iodoxine, howe­
ver, is fairly soluble in pyridine and dioxane. Dioxane solution
of the reagent (l —2 per cent) was therefore used in this course
of investigation. A number' of metallic ions obtainable in
aqueous1dioxane solution were examined and the results are
summarised in the following. Pyridine solution of the reagent
behaves similarly.
The colours or precipitates for the different ions are:
Cu'H' greenish-yellow ppt., Ni++ red-brown ppt., Co++ chocc-
late-brown ppt., Fe+++ greenish-grey ppt., UOJ+ orange-red
ppt., Pd** and Th++++ orange-yellow ppt., Zr++++ cream-
yellow ppt., Ti++++ pink-yellow ppt., VO^* light yellow ppt.,
Pb++ yellow ppt.
The reagent has been found to react quantitatively with
copper, nickel, iron, zirconium, thorium, uranium, vanadium
and titanium at different pE values. A precipitate is ob­
tained with a solution containing 0*4 mg. of ferric iron in
200 ml. and with solution of zirconium, thorium, uranium and
copper containing only 0*2 mg. of the ions in 200 ml. The
precipitates of copper could be weighed directly after drying
to constant weight at 110—120°. The reagent has also been
found to react with very small amounts of iron and uranium
to give characteristic colours, and is thus suitable for the
colorimetric determination of iron and uranium. An aqueous
solution of the reagent can not be used for the colorimetric
determination of iron or uranium due to the insolubility'
of the reagent in water. An intense green colour is devel­
oped by the addition of a dioxane or pyridine solution of the
iodoxine to a very dilute ferric salt solution, followed by the
addition of a few ml. of dioxane or pyridine. The maximum
absorption lies below 400 mp; but as the reagent also shows
an absorption below 360 mp, measurements were made ad
380 mp. Uranium also gives a strong orange-red colour
 - 2
this reagent in presence of dioxane or pyridine. The maximum
absorption of the uranyl complex in dioxane also lies below
400 m p,; but with pyridine solution of the reagent, the maximum
absorption is at 400 mp. This reagent may be used advanta­
geously in the colorimetric determination of iron and uranium,
since the ferric and uranyl complexes are soluble in chloro­
form, carbon tetrachloride, benzene and carbon disulfide.
Thus, small quantities of iron and uranium may be extracted
from aqueous solution and the sensitivity of the reaction in­
creased. The depth of colours in these reactions has been
found to be proportional to the concentration of the metallic
ions and consequently is well suited for the colorimetric
determination of these ions. The colours are found to be
stable and remain unchanged even on long standing. But
the intensity of the colours is variable with the hydrogen
ion concentration of the solutions and only suitable for
measurements over a definite pH range.
Further investigations, including the gravimetric estimat­
ions of copper, nickel, iron, uranium, thorium, zirconium,
titanium and vanadium and their separation from each other
and the colorimetry with solvent extraction are in progress.
Details of the methods will be published elsewhere in due
course.
Our sincere thanks are due to Professor P. Ray of this
department for his helpful suggestion and interest.
Department of Inorganic Chemistry, Indian Association
for the Cultivation of Science, Calcutta-32, India.
Ajit Kumar Mtjkherjee and Bireswar Banbrjee.
Eingegangen am 3. Juni 1955.
*) Papesh, V., and R. Burtner: J. Amer. Chem. Soc, 58, 1314
(1936).
*) Fresentos, N.: Z. analyt. Chem. 96, 433 (1934). — Berg, R.:
Z. anorg. allg. Chem. 204, 208 (1932).
*) Berg, R., and H. KOstenmacher : Z. anorg. allg. Chem,
204, 215 (1932). — Ishimaru, S.: J. Chem. Soc. Japan 58, 201
(1934).
4) Hahn, F.: Z. angew. Chem. 39, 1198 (1926). — Berg, R.:
J. prakt Chem. 115, 178 (1927). — Fleck, H. R.: Analyst 58, 388
(1933).

Dmck der Universltatsdruckerei H. Stiirtr AG., Wtlriburg

 [Reprinted from Scihnob and Ctmrima
VoL 20, pp. 229—231, November, 1954.]

AGE OF NAGARJUNA
BIRESWAR BANERJEE
H3DIA1T ASSOCIATION FOB THB OtTLTIVATIOK' OP SOIBirOT,

OAXiCTOTTA.

■THE age of the great Indian alchemist
^ Nagarjuna has been a subject of great
controversy among the scholars.. Different
writers have placed him in different times from
which it may be assumed that possibly several
Nagarjunas flourished in India at one or the
other time during 200 B.G. to 900 A.D. It
has, however, been generally accepted now that
there were two Nagarjunas, one the philosopher
who preached the doctrine of MahSyanism,
and the other the alchemist, the inventor of
the processes of distillation, sublimation, etc.
The chronology of Nag&ijuna presents a problem
for solution, for authorities like Hiouen Thsang,
Tar&nath and others have given a rather con­
fused account, assuming that the philosopher
and the alchemist is one and the same person.
It is not safe without positive evidence to
place much faith in these early chronicles.
Nagarjuna is indeed a magical figure in the
history of the rise of Mahayanism, as also in
that of Indian alchemy. According to Lassen1.
Nagarjuna flourished at about 23 A.D. during
the reign of Kanishka, the famous Kushana
king. It is said that he joined the Third Bud­
dhist Council as a prominent leader of Maha­
yanism. This Council was held under the patro­
nage of king Kanishka. Kalyana Misra (11th
century A.D.), the author of Rajatarangini,
placed Nagarjuna 150 years after •Sakyasimha,
from which it may be stated that he lived nearly
by the first quarter of the third century B.C.
But Hiouen Thsang, the Chinese traveller,
who visited India in 629 A.D., writes relying
upon the local legends that the Buddhist al­
chemist Nagarjuna was a friend of king Sata-
vahana; this was supported by the contemporary
author, poet Vana, in his work—The Life of
King Harsha. The friendship between Nagar­
juna and king Satavahana was further support­
ed by the Chinese and Tibetan sources, where
we get the reference of king Udayana as a friend
 t * 3
of the sage. Again Rasaratnakara, one of the establishing the age of the alchemist NSgarjuna
works of Nagarjuna, contains a dialogue we can not depend upon these sources.
between the author and the king Salivahana.
All these names, Udayana, Salivahana, possibly Now, the other convincing evidences that the
belong to the same character S&tavahana who alchemist Nagaijuna can not be placed in the
is a prominent figure in the history of South first century A.D. with his philosopher name­
India. He was one of the Satavahana kings, sake as an identical person, has been compiled
who ruled over the Deccan from circa 73 B.G. from Kakshaputatantram2, one of the works
to circa 218 A.D. All these facts place the ascribed to that reputed author of the Indian
philosopher Nagaijuna, the founder and syste- alchemical school. The present writer made
matiser of Mahayanism, near about the begin­ a through search in the Kakshaputatantram
ning of the Christian era. We can, therefore, and has come to the conclusion from, the
accept the views of chose Western scholars, who evidences found in it that the age of the author
placed the age of Nagaijuna, the philosopher, of Kakshaputatantram must be placed on or
in the first century A.D. after the 6th century A.D. From these we can
definitely conclude that the alchemist Nagar­
According to some authors there were several juna was a different person from the philosopher
Nagarjunas, or there was a school of alchemy Nagaijuna, who lived in the early Christian
whose leader assumed the title of Nagarjuna. era. The evidences on which the writer makes
The editor of Kakshaputatantram* in its preface his conclusion are given here.
writes that there were four Nagarjunas in the
early days. But these assumptions have little Kakshaputatantram provides us with two
authentic evidence in their support. It is important clues regarding this matter. The
quite probable that there were only two Nagar­ name of this work is very significant. Tantras8
junas—the second one being the celebrated are books enbodying some secret and mystic
alchemist. The dialogue between the alchemist knowledge. The Tantras are divided into two
Nagarjuna and the king Salivahana1 in Rasa­ groups the Agamas and the Nigamas and are
ratnakara seems to place him in the early presented in the form‘of conversation between
Christian era. But the figure of Salivahana Shiva and P&rvati. The Tantric cult became
might be viewed as an allegorical one; for, no predominant one in the post-Gupta period and
reference of the improved chemical knowledge there is not a single Tantra which can be consi­
described in Rasaratnakara is to be found, in dered to be of earlier origin than the Gupta age.
the treatises of Gharaka, Susruta or even in Thus, considered from this point of view, the
Vagbhat. Kakshaputatantram must have been composed
Taranath, relying upon the Tibetan sources, later than the age of the Guptas, which ter­
writes of Nagarjuna as an adept in magic, minates at about 550 A.D.
conjuration and alchemy, and places him in The next point is the reference of Dinara in
the early 7th century A.D. The Tibetan and Kakshaputatantram. There are several slokas
Chinese..sources also furnish us with some in­ with reference to Dinara in it, out of which
formation about this celebrated Indian alche­ only one is mentioned below :
mist. These hold the view that both the philo­
sopher and the alchemist are represented by
one and the same person, for which there is spgjrt *1% 1
ho definite evidence. Mr. M. Walleser3, while
assessing the value of the Tibetan and Chinese
sources, writes,—“the name of a man of whom
we can not positively say that he really existed, “After midnight the goddess appears and bestows
—still less that he is the author of the works on the worshipper tweaty-flve dinaras (gold ooins)
ascribed to him : this name is Nagarjuna.” daily”.
Walleser again comments in another place
“the personality of Nagarjuna himself became The Roman coin dinarius6 first came to.
surrounded in an increasing manner by a mystic India during the time of the Kushana king
veil so that an unlimited number of legends Kadphises II, who ruled, according to Rapson,
arose about him, that an unlimited lifetime Thomas and others, just before 78 A.D. But
brought a close by himself, and the acquire­ in the Sanskrit literature the use of the term
ments of the rarest magic powers, were ascri­ diriara is, however, a late one. During the age
bed to him.” Mr. S. Pathak4 also in a recent of Guptas we find the first reference of the use
article in the Indian Historical Quarterly com­ of the term in the Sanskrit language. The
ments that the Tibetans have mixed up die .two mention, of it in Kashaputatantram proves
Nagarjunas as an identical person. So, for conclusively that the currency had become
 t 3 3
popular one, and as such we may presume that
the author lived in the Gupta age or later, when
there was a great Sanskrit revival in India.
The theory of the two Nagfirjunas is further
supported by Vidhushekhara Bhattacharya7
and Giuseppe Tucci8, Pandit Vidhushekara
placed the philosopher in the 2nd century A.D.;
for, according to him the 2nd Nagarjuna
flourished in 7th century A.D. His argument
to establish the age of Nagarjuna bears a resem-
blence with that of Taranath, because both
of them depended upon the Tibetan sources
mainly. Giuseppe Tucci wrote in his works,
—“we find so many treatises ascribed to NSgar-
juna which are certainly works of a later date
and written by another Nagarjuna '(Siddha
Nagarjuna). From these we can come to the
conclusion that alchemist Nagarjuna or Siddha.
Nagarjuna, the reputed author of Rasaratnakara
or Kakshaputatantram, flourished near about
the 7th century A.D. However, to determine
his exact age we shall have to consider some
more records given by the other authors.
The frequent reference of Nagarjuna as an
authority in alchemy, is found in the works of
Vrinda1 and Chakrapani,1 which are impor­
tant sources for establishing the age of
this great alchemist. Dalvana1 also associates
this venerable name with the redactor of the
Susruta. The works of Nagarjuna contain
more advanced knowledge- of chemistry than
those of Gharaka, Susruta and Vagbhat. So
the age of the alchemist Nagaijuna should
lie between those of Vagbhat and Vrinda—.
Chakrapani. The age of Vagbhat may be
assigned to oirca 800 A.D.8, whereas the age
of Vrinda may be placed near about 1000
A.D. So we can place the age of Nagarjuna
some where in the 9th century A.D. Alberuni1
says, “A famous representative of this art
(alchemy) was Nagaijuna, a native of the
fort Daihak, near Somnath. He excelled in
it and composed a book which contains the
substance of the whole literature on this subject
and is very rare. He lived nearly a hundred
years before our time”. We do not know how
far we can believe the records given by this
great and cultured Arab who visited India in
the 11th century A.D. At that time the Hindu
Schools of science and learning were unapproach­
Published by Hemendra Nath Saha, Assistant Secretary, on behalf of the Indian Soienoe News Association, 92,
Upper Circular Boad, Calcutta, and printed by Halipada Mukherjee at F,ka Press, 204/1, B. T. Boad, Oalcatta-35.
able for Mohammedans; so Alberuni had to
depend upon the local traditions of those places
of India which he visited. This information
does not shift the age. of Nagarjuna far from
the 9th century A.D.
One of the greatest authorities of the history
of science, George Sarton10 believes that Nag&r-
juna flourished in the 8th century A.D. Parting­
ton11, the well-known chemist, places the al­
chemist Nagaijuna in the 9th century A.D.,
though according to him very little is known
about this historic character. However, con­
sidering all these facts we can accept the 9th
century A.D. as the correct estimate for the age
of the great Indian alchemist Siddha NSgarjuna.
But still this problem requires further investiga­
tion, and only a close study of the internal evi­
dences of the works ascribed to this great author
will enable us to come to a satisfactory conclu­
sion.
Several works are known, the authorship of
which is ascribed to the alchemist Nagarjuna,.
but these are not available now. Any infor­
mation regarding these works will be gratefully
appreciated by the writer, as that would facili­
tate his investigation.
The writer hereby, expresses his sincerest
thanks to Prof. P. R&y of the Indian Associa­
tion for the Cultivation of Science for his kind
suggestions and guidance. Thanks are also due
to Prof. J. N. Banerjee of Calcutta University
and Prof. Sudhakar Chatterjee of Viswabharati
University for their helpful suggestions.
BsaiBBEaroiss
1 P. C. Bay, History of Hindu Chemistry, Vol. I and
VoL n.
2 K. M. Chatterjee, Siddha Nagarjuna Kak^haputam.
5 M. Walleser. Life of Nagarjuna from Tibetan and
Chinese sources.
4 3. Pathak. Life of Nagarjuna published in Indian
Historical Quarterly March 1964.
6 Bagchi—Studies in the Tantras.
® Brown, Coins of India.
7 Vidhushekhara Bhattacharya, Mahayanavims aka.
8 Giuseppe Tucci. Dinnaga Buddhist Texts on logio
from Chinese sources.
0 D. C. Bhattacharya, New light on Vaidyaka Litera­
ture, Indian Historical Quarterly, Vol. 23, 1947.
10 G. Sarton, Introduction to the History of Science,
Vol. I.
11 Partington. History of Chemistry.
 [Reprinted from Sctbnob and Cultubw
Vol. 21, pp. 26—28, July, 1956.]

RASAYANASASTRODDHRIT1 FROM THE TIBETAN

XYLOGRAPH
SUNTTY KUMAR PATHAK
‘ VISVABHABATI UNrSTIBSITY, SANTINTlCElTAN,
AND
BIRJBSWAR BANERJEE
INDIAN ASSOCIATION FOB THK CULTIVATION OF 80IEN0B, CALCUTTA

CTHE Tibetan text of the Rasayanasastrod- The equivalent Sanskrit terms of the Tibetan
A dhriti (gser.’ gyur. gyi. bslan. bcos. works current in alchemy have been taken from
bsdus. pa) is found in .the Bstan-’ gyur collec­ the Amarkosa and Mahavyuipatti. The language
tion (mdo go pp. 3b : '2-54 : 5) of the Snar- of the text is obscure in some places and this
Than edition. For studying this text the might not unreasonably be expected consider­
Xylographs belonging to the Asiatic Society of ing the fact that the Tibetan translator could
Bengal,- the Calcutta University and the not express the actual details of the. processes
Visvabharati University Library have been of Indian alchemy in his translation, as it was.
made use of. Among these, the Xylographs a tradition among the alchemists' to make a
belonging to the Calcutta University Library secret of their efficiency. However an attempt
are comparatively clearer and more legible. has been made to cull the ‘scientific5 knowledge
The comparatively illegible portions have been from this Tibetan text which with its complete
deciphered and translated by making use of English translation by S. Pathak will be pub­
the probable readings suggested by the Rev: lished shordy in the volume of History of Chemistry
G. R. Lama, Head of the Indo-Tibetan studies, in Ancient and Mediaeval India, by the Indian
Visvabharati University. The text, as is evi­ Chemical Society, edited by Prof. P. Ray.
dent from its name is a collection of a few
verses from the works on Rasayana of ancient As is usual with all alchemical texts, Rasa-
India, and was possibly translated from the yanasastrouddhriti, also begins with a homage
original Sanskrit into the Tibetan language to the Almighty Lord, Iswara, who bestows
sometime during the early Mediaeval Ages. peace and good upon all beings, and with a
A relevant point of great importance lies in salutation to Maheswari, the consort of Lord
the fact that there is no mention of the original Siva. The whole text comprises mainly of
Sanskrit text. discussions on several metals, their properties,
Two other texts from Tibetan sources purifications and also on preparations for polish­
styled Sarvesoararasayana and Dhatuvada were ing them. Let us first consider the verses (3)
translated into English by Pandit Vidhu- to (9), the English translation of which is given
shekhara Bhattacharya, which was published in below :—
Acharya Ray Commemoration volume in 1932. (3) Mercury Products : After compounding
Rasayamsastroddhriti is an allied work dealing with various .materials many useful products
mainly with alchemical ideas, which has an can be prepared from pure and solid mercury
entirely original stamp of its own regarded from (dnul. chu.).
the standpoint of the history of science. Pandit
Bhattacharya has translated the Tibetan name (4) Merits of Mercury : Mercury has three­
of the text as Dhatuvadasastra or Dhatuvadasastro- fold merits (yon. tan) :
ddkriti. No mention of its author and translator
is found in the original Tibetan Xylograph, (1) Burnt mercury (rgyal. ba.-being con­
but according to the Tibetan Catalogue Pandit quered) cures diseases; (2) heated mercury
Bhattacharya referred this text to be a part (smin. pa) removes poverty (of die person
of Vyadipada’s work. In Catalogue Du Ponds engaged in experiments); (3) confined (by
Tibetain by P. Cordier we also found, that this mystic mantras of the Tantric school) mercury
text may be ascribed to Bhali-pa or Vyadi- (bcins. pa) goes into the sky.
pada. But curiously enough in the Japanese
Tibetan Catalogue published by Tohuku Imperial (5) Merits of Mica : Heated mica (chan,
University, the name Ratnasri has been men­ tsher) has eightfold merits. Heated with sul­
tioned as its translator, without any reference to phur, it has six-fold merits.-' Like mercury,
its author. However from these limited facts it confined mica with gold facilitates flight in the
is not desirable to draw any conclusion regard­ sky (lit. how can you check diem going into
ing the matter of authorship of this text. the sky).
 [ 2
(6) Copper Polish : Take zinc or tin (gsa’.
tse=tin) and mercury dust isediment) of equal
proportion and powder them. Add (zla. ba)
juice (of a kind of white coloured flower), dou­
ble in proportion, with half of (the above men­
tioned) powder according to the method spoken
by Chandrapakasa. This compound can polish
many sheets of copper.
(7) This compound without mercury is red­
dish. It enriches (the compounder). The heat­
ed compound in a little viscid form is turned
into sky-blue colour. Then it will last long.
(8) Method of Melting : The process of
melting : Boil mercury properly and add
silver or gold (proportionately). They will be
transformed into one body and will not be
decomposed any more.
(9) Method of Vandha (confining) : The
method of confining is : add (burnt) mercury
(dban.— lit, subdued) with heated mica. It
enriches (the experimenter) and leads him
to the door of religion.
In these verses primarily, the three-fold
merits of mercury have been mentioned.
Burnt mercury cures diseases, possibly indicat­
ing the useful medicinal properties of its oxide.
The yellow mercuric oxide has a good many
medicinal uses in ' modem pharmacy. The
mercuric oxide can be prepared by heating
mercury in air at about 300°G. in an open
container, hence it seems to be identical with
the burnt mercury of that period. But how
the heated mercury can enrich the experi­
menter is not discussed. Thirdly, we see that
the confined mercury goes into the sky. Verses
in the same strain are alsc found in different
Sanskrit alchemical texts; as for example in
the Kakshaputatantram by Nagaijuna, there is a
sloka with the sense that if some mercury can
be kept closed in a seed, then by applying heat
it can be taken to the sky, provided always
the mercury is kept within it. If some mercury
is confined in a closed vessel and is strongly
heated, then the gaseous metal without getting
an outlet would move the vessel by its pressure.
Probably from this observation the alchemists
took it for granted that confined mercury tends
to fly to the sky. It was a traditional belief
also, that an expert can fly by holding confined
mercury.
Then in the verse (6) an amalgam of
mercury and tin has been mentioned as being
used in preparing a copper polish. We can
not follow what kind of reddish compound can
be prepared from tin and a kind of fruit
juice only, as mentioned in verse (7). In the
next verse (8) an amalgam of mercury has
]
been described, but in verse (9) the statement
in the text,—confining mercury with heated
mica would enrich the experimenter, is un­
intelligible.
In the Tibetan text verses (10) to (24)
mainly deal with the preparations of metal
polishes, which describe some fantastic ideas
typical of the alchemical works. There are
two processes for gold polish, one from milk of
lioness and head of the chataka bird, and the
other from a mixture of iron, mercury, copper,
tin, svarna (a kind of flower) and a kind of
white stone used for medicine; the preparations
of both of these are described in rather mystic
words'. The verse (13) also mentions a process
for making gold from mercury, sulphur, hari-
tala, maksika (pyrites), and kiintika sikhi;
we do not know whether this mixture would
ever produce a gold-like colour even. In the
remaining verses the preparation of several
arsenic compounds has been described in the
same mystic manner. In one, mention is made
of arsenic being mixed with yellow colour,
white metals, mica, etc., in equal proportions
with kakusatha and melted to form a com­
pound. This compound after distillation yields
a glazy fluid for polishing the metals. The
preparation of these polishes from several
materials can furnish no clear idea of the nature
of the product formed. Even the properties
ascribed to the gold polish suggest no sensible
idea. Mention is made of the fact that these
polishes prepared from mercury are usefully
applied on pure copper, steel and lead.
The verses (17) and (22) describe mainly
the classification of copper and the qualities of
gold. The English translation of them are
given below :—
(17) Classification of Copper : Copper is red
and hard, and is of three categories : (1) of the
best quality, copper is imported from Nepal,
(2) of the medium quality, copper is as a
flower leaf or petal (’dab. ma), (3) of the in­
ferior quality, copper has no glaze but spots.
(22) Qualities of Gold : Gold has got six qua­
lities. It is (a) separable by cutting (not by any
other means), (b) yellow coloured when it is
rubbed, (e) of saffron colour, being burnt in
fire, (d) bright (e) heavy and (f) glittering like
kunda flower or a shining star.
Obviously the above two verses are the
inferences from some observations. The diff­
erent types of copper obtained from various
sources have been classified according to the
current knowledge, and the quality of gold has
been judged from its properties. The method
 [ 3
of powdering metals described in verses (25)
and (26) seem rather fantastic.
The verses (27) to (39) described the puri­
fication of different metals. The English trans­
lation of some of them is given below :—
(27) Copper and steel (glazy iron) will
be purified by acid wash.
(28) Another Method of Purifying Copper :
Add elephant (glan-chen) bone dust (rus. phye)
with glass jewel powder, and sprinkle for
seven times the urine of goat upon these two
powders.
(29) Learned men know the use of copper
and the process of its purification. Purified
copper will be used in many ways.
(30) Method of Purifying Brass: The method
of purifying brass is to be learnt properly.
The learned knows the method of making
brass dust by rubbing it with salt and urine
for several days, and afterwards putting (brass
dust) in three earthenware pots again and
again (for washing). Then brass will be
purified.
(31) Method of Purifying Bell Metal : The
method of purifying bell metal (’khar.pa) is
similar to that of brass. (You) should learn
that well. Fine dust from it will be made.
Then, add salt-mixed urine with those bell
metal dusts, and wash the material again and
again for seven times.
(32) Method of Purifying Lead : Put lead
into fire until it gets copper-like colour. When
it will have copper-like colour, add some brass
again on it.
(34) By arsenic powder which is (as strong)
as burning, and srikhanda’s juice it will be
purified by washing for a day (again and again).
(35) Lead will be well purified by burning
in fire, and by (arsenic compound) powder.
Published by Hemendra Nath Saha, Assistant Secretary, on behalf of the Indian Science News Association, 92,
Upper Circular Road, Calcutta, and printed by Kalipada Mukherjee at Eka Press, 204/1, B. T. Road, Caloutto-85.
]
Within seven days, lead will be made pure by
washing with the juice of three kinds of fruits.
(36) Method of Purifying fine : The process
of purifying zinc is like that of lead, with the
seed of kunali and the foamy content of wine.
(38) General Method of Purification.
The method of purification is well des­
cribed; for, many people carry out this process
wrongly. Scholars know how to burn lead in
fire to make it pure by (removing alloy) adding
malt.
From the above translation, we find that
the general procedure for purification was
carried, either by washing with acid or alkali
(urine). Help of some salts ( ? ) was also
taken. By these methods only the surface
could be cleared. Attempts to purify lead by
arsenic and zinc, by washing with castor oil
and karanja oil appear somehow peculiar.
In the remaining portions the causes of
impurity in gold and the power of Siddhas
have been described.
Throughout the text it would be found that
a mystic veil has been drawn over the prepara­
tions and procedures, with the assertion that
the learned ones know the methods well. This
is a favourite announcement of all the alchemi­
cal texts, the learned ones easily attaining
accomplishment of changing iron and other
metals into gold, some times even not by
applying any experimental method. The state­
ment is made perhaps to influence the readers,
so as to commend their respect for the mystic
powers involved. Finally, we are of opinion,
that this text displays very little systematic
chemical bias*.
*We express our sincere thanks to Prof. P. RSy of
the Indian Association for the Cultivation of Soience,
for guidance and encouragement, and to Rev. C. R.
Lama of Visvabharati University for taking kind inter­
est in the progress of this work. Thanks are also due
to Sree Benoy Chandra Mukherjee for helpful sugges­
tions.
 [Reprinted from Science and CuI/TOBS
Vol. 21, pp. 195—197, October, 1955.]
SIDDHA NAGARJUNA KAKSHAPUTAM OR
KAKSHAPUTATANTRAM
BIRESWAR BANERJEE
INDIAN ASSOCIATION FOB THB OTIiTIVATION OP SOIBNOB, OAIODTTA
THAMES of several alchemical texts, besides
the well-known treatise Rasaratnakara, the
authorship of which is ascribed to Siddha
NagSijuna or the alchemist Nagarjuna, are
known, but most of these are not now available.
Copies of two of these, Jogosara and Joga4ataka,
lie in the Nepal Durbar Library; many others,
such as Arogyomanjari, Kautuhalachintamani,
etc., have perhaps been lost. For, no infor­
mation regarding them is available. Two other
works of Siddha NagArjuna, which one finds
in the archives of National Library, Alipore,
are Kokaiastra and Kakijhaputatantram, edited
by VolAnAtha Vidyanidhi and Satish Chandra
Mukheijee respectively. Though the author­
ship of these works is ascribed to the reputed
leader of the Indian alchemical school, still
neither of the editors tried to sift out the
chemical information contained therein. The
reason probably is that the contents of the
texts are given in mystic and fantastic verses
from which it is not easy to extract intelligible
information.
Koka&istra is a manual on sexology. It is
said to contain an account of the first fruit
of knowledge, acquired by man. Siddha
Nagarjuna in this work appears as a commenta­
tor of this fundamental knowledge that came
from Lord Siva. This work is divided into
eleven Padas and starts with an introduction
of the author himself (Siddha Nagaijuna) who
was accomplished in the art of all sorts of
esoteric practices and lived by the side of
river Narmada. This work mainly contains
symbolic interpretation of sexual sciences but
no chemistry. However, it is doubtful whether
it really belongs to that renowned alchemist
NAgaijuna or was written by some one else in
a later age, under the garb of a great name.
Kakshaputatantram is however, a composi­
tion of greater importance and interest from
the stand-point of history of ancient Indian
chemistry. In fact, it is one of the prominent
works of the great sage, Siddha Nagarjuna.
It deals mainly with the esoteric doctrines of
magic, sorcery, alchemy,' incantations' and
allied subjects. It is stated that Nagaijuna
always kept this text under his armpit,, just to
protect it from theft, for which this work was
given the name Kakshaputatantram,' i.e the
tantra which remains under the armpit. The
legend goes, that for this habit Nagaijuna
lost the use of his left hand after some ' years.
Kakshaputatantram is divided into 31 Pa talas,
describing different types of supernatural power
like subjugation, charms, spells, making one
invisible, producing diseases, invocation o: evil
spirits, etc. The original text was divided
into twenty Patalas which are sub-divided by
the editor. Though this work is mainly
devoted to supernatural objectives, still there
are a few verses in it which reveal some rudi­
mentary knowledge of experimental sciences.
It is the intention of the present writer to
make a critical study of the chemical informa­
tion found in this text.
As usual in all tantras, the Kakshaputatan­
tram begins with worship of the deity, the
supreme Lord of this world. The verses, which
deal with chemical information, are quoted
below along with their translations.
 [ 2 ]
The fumes from this incense kill mosquitoes,
lice, bugs, snakes, rats and scorpions.
:mr# I«fi5wrfir ■w u
XIV
TTSSTcT 3?H.I
If powdered sulphur be spread upon the
surface of water in a tank and this water be
sprayed upon vegetables and groves of plants,
the latter are thereby destroyed.
qg'wfer ^ ppfptt jpit gig tori: ii
<J#?T ftl# pfesn* I xvn
enrararc ^rrf^er gftrai *fw n When mutha (Cyprus rotunda), white mustard
XVII (Brassica alba), marking nuts (Semecarpus anacar-
dium), elephant apple, treacle, dust of fruits
If a small idol be made out of paste con­ of sun plant (swallow wort) and resin, are
sisting of orpiment and whey, and this be mixed together and burnt inside the house,
placed inside the house, then all flies, smelling bugs, mosquitoes, snakes, rats and other poison­
the paste, leave the house. There is no doubt ous insects flee from the house.
about it.
II
gsn =*r XIX
3P&$g jj# ’ll II Fumes from burning sulphur change the
XVII colour of flowers.

When treacle, milk of arka (Calotropis g § <11^1

gigantca), gunja (Abrus precatcrius) and powdered
sesame ate mixed and spread over the leaves of ftfcftfVlgfllsFg sRJff 3 TO#rfcT; I
arka, and the latter are then placed in a house,
all the rats in the house are killed thereby.
XIX
ft«R I A hole is bored in a Mandabi fruit (?) and
a fine drop of mercury, is poured into it, the
^3 ftw# *fbn§# ^Tcjif i . mouth of the hole is then closed by means of a
wick, which is then ignited. This is then
«rfe«r ^ #1 i thrown into air, taking care that the mercury
remains always closed inside the hole; it is
*Rprra.q«Tt ?nfar gsorraf gf#r snm n found that the seed rises up in the air,
XVII
A paste is made of the powdered stramonium sppg #1TOU
(Datura fastuosa), white arsenic, and ground #T55 3T wfaoT ft##l
sesame; this is then thoroughly ground with fterc; ftftftf#,i
VishapdSSna (?) in fish oil. Pills are then made
of this mixture and kept in the house with all
the water pots covered safely with lids. Hats
on testing these pills suffer immediate death FPTO , f#R55Tft aI
out of thirst. V I
ftsr - g s# gsTTOin# stes#
ft# fwngern t erfewsftw umt u
®wfwIs4 wm ^ jphstri XXV
■’SfreT^Iff W^i#5TOI II The substances lying .inside an earthen,
XVII wooden or iron vessel, or associated with algae
or soil, have got to be purified. A ground
Prepare incense by grinding together treacle, mixture of salt and :sand is thrown oyer the
frankincense, marking nuts (Semecarpus anacar- •substances. The whole is then heated by a
■dium), biranga (Embelia ribes), three myrobalans, mild fire till the salt is completely dried up.
lac, and leaves of akanda (Calotropis gigantea). By this means gold and all other gems are
 [ 3 1
purified. In another process twigs of aijuna of a flower, indicates, a knowledge of the bleach­
(Terminalia aijuna), the beleric myrobalan and ing action of sulphur dioxide. Another verse
of this Patala deals with the traditional belief
the chitraka (.Plumbago zeylonica) tree are crushed
together and mixed with an equal amount of of going into the sky with the help of mercury.
salt; this is then agitated with fermented rice. It is a common description found in many
The gems (noble metals) are then painted with alchemical texts for defying gravitation or
this paste and heated in fire, when they are acquiring levity. The experiment described
purified. simply shows that the vapourisation of mercury
by heat inside the closed space of a body forces
the latter high up into the air. These are only
w errs 'ftw i the records of some observations without any
TTgT yicMI II explanation, because of ignorance. In Patala
XXVI- XXV, two processes for the purification of
gold and gems have been described. The
A mixture of ten parts of gold, twelve parts first method of purification is to digest them
of copper and sixteen parts of silver is known with salts and sand, but the second one may
as ‘trilauha’. This proportion in ‘trilauha’ is be said to be a process of reduction. Here
to be found under all circumstances. the twigs would provide carbon particles for
the generation of reducing gases. The coating
Only one verse has been taken from XIV of oxides on the surface of noble metals are
Patala which describes the poisonous effect of thereby removed. In Patala XXVI, an alloy
sulphur on plants and vegetables. This shows, of gold, copper and silver has been described.
that sulphur can be used as weed-killer
was known to the author. The verses quoted In conclusion, I would like to mention
from XVII Patala deal mainly with recipes here that Jogasara and Jogoiataka, the other
for overcoming household troubles from rats, two works of Siddha Nagarjuna are still lying
lice, flies, insects, etc. Orpiment, which con­ in the Nepal Durbar Library, perhaps without
tains arsenic, is a well-known poison. It is any attempt- at estimation of their values, so
used as a insect-killer and was thus known far as chemical informations present in them
at the author’s time. Madder-gum, gunja, are concerned; .These rare works may con­
stramonium are also well known poisons, and tain some more valuable data of Indian
have been correctly described as poison for Alchemy. Prof. P. Ray tried to bring copies
rats and rodents. The last two verses of this of these manuscripts from Nepal, through
Patala describe the preparation of poisonous the National Library, but unfortunately met
incense, by which snakes, rats, mosquitoes, with little success. Any information regarding
bugs etc., would leave the place at once. the other works of the alchemist Nagaijuna
This might be due to the production of poison­ would be much appreciated by the writer of
ous fumes from organic compounds, used in the present article.
the preparation of the incense.
My best thanks are due to Prof. P. Ray
The XIX Patala mainly describes some of Indian Association for the Cultivation of
experimental observations. That the smoke Science for guidance and taking keen interest
from burning sulphur can change the colour in the progress of this work.

Published by Hemendra Nath Saha, Assistant Secretary, on behalf of thb Indian Soienoe News Association, 92,
Upper Circular Road, Calcutta, and printed by Kalipada "MukhcrjeC at Eka Pfess, 204/1, B. T. Road, Calcutta-30.