960
JOSEPH
A. SCHUFLE AND H. MORRIS EILAND
Vol. 76
[DEPARTMENT OF CHEMISTRY, NEW MEXICOINSTITUTE OF MININGAND TECHNOLOGY J
Indium Halide Complexes Studied By Ion-exchange Methods BY JOSEPH A. SCHUFLE AND H. MORRIS EILAND RECEIVEDJULY 27, 1963 The aqueous complex systems of indium with fluoride, chloride, bromide and iodide have been studied from their equilibria in the presence of a n ion-exchange resin, Amberlite IR-120: The solutions were kept a t a constant ionic strength of one with sodium perchlorate, and at a constant pH of 3.8. The distribution of indium between the solution and resin phase was followed by using 50-day In114as the tracer. T h e data were analyzed by the method of S u r e Fronaeus. This method makes allowances for the formation of complexes higher than the first complex, and for the adsorption of the first complex as well as the metal cation itself on the resin. The results indicate the existence of three complexes in the bromide system for bromide concentrations up t o 0 . 2 mole/l.; in the chloride system for chloride concentrations up t o 0.12 mole/l.; and in the fluoride system for fluoride concentrations up t o 0.004 mole/l. Formation constants calculated are: In-Cl system: PI = 26; PZ = 170; p3 = 1680. In-Br system: PI = 16; BZ = 60; P 3 = 300. In-I system: PI = 2. In-Fsystem: P I = 1000; 6 2 = 6 X lo5; ( 3 3 = 4 X lo*. The concentrations at which the first two complexes in the chloride and bromide systems predominate are determined
The aqueous indium halide system is interesting in that a regular series of complexes seems to be formed as predicted by Bjerrum.' Earlier work a t high chloride concentrations2 showed the possibility of InClz+ and InC14- complexes existing. Work by Schubert3 and F r ~ n a e u s among ,~ others, has shown that the ion-exchange method may be applied to the study of complex systems. Schubert's treatment is especially applicable where only the first complex is formed in appreciable amounts and it assumes that the adsorption of the complex cation on the resin is negligible.5 Neither of these requirements is strictly met in the indium halide systems, since complexes higher than the rnonocomplex can be demonstrated and there is an appreciable adsorption of the first complex on the resin. Therefore the data were analyzed in the more general terms of Fronaeus.
CM = total metal concn. in soln. in all forms ( M R ) , ( M A R ) = equil. concn. of metal ion and complex ion in the resin (A) = equilibrium concn. of halide ion in s o h . = distribution coefficient = (MR) (MAR) (M) (MA) (MA21 (MAX) yo activity in resin vol. s o h . % activity in soln. mass resin
+
+
+
+
+
P,,
(MA,) = formation constant of the nth complex = ___ (M)(A)"
a,
(MA ) = degree of formation of the nth complex = 2
Z
= av. no. of ligands attached to M = n
1
I
CM
+ ddP(A) log a, ~
Only the terms (MR) and (MAR) are shown in the numerator of #J as defined above since it has been shown that the adsorption of the MA2+ complex in such a tri-univalent system as this is probably negligible.6 If the following equation for #J is deduced
i!
p"'"
/
L
0 -
IO
20
30
J
(AI.
Fig. 1.
The following notation, largely that of Fronaeus and Bjerrum is used (M), (MA), (MA?), . . , ( M A N ) = equil. concn. of metal ion and complex ions in s o h . (1) J. Bjerrum, "Metal Ammine Formation in Aqueous Solution,' P. Haase and Son, Copenhagen, 1941. (2) J . A Schufle, M . F. Stuhbs and R. E . Witman, THIS JOORNAL, 73, 1013 (1951). ( 3 ) J . Schuhert, E. R. Russell and L. S. Rlyers, J r . , J . Bioi. Chr?n.. 1 8 5 , 387 (1950). (1) S. Fronaeus, A c t a Chenz. Scand., 6 , 8.59 ( l n j l ) . ( 5 ) J . Schubert and A . 1,indenbaum. THTS~ O l J R W ¶ I . , 7 4 , %2!J
(1552).
Fronaeus' has shown that equation 1 may be ten in the alternate form f = 6141 - xz where XZ = Pz &(A) Pi(A)' ...... Also from equation 1 and the definition of may show that the lim 41 = (PI - I ) .
+
+
+
writ(2) (3)
one
(A)+O
By a graph of l/#Jvs. (A) one obtains l / l ~by extrapolation to zero concentration of (A) (Fig. 1). By a graph of #JI vs. (A) the quantity (PI - I ) is obtained by extrapolation. The function f may now be calculated for various values of (A), and X 2evaluated from equation 2 for various values of (-4). ( 6 ) S. Fronaeus, S w t v y c k U I Saensk Keinisk T i d c k ; , i f l , 66, 19 (1958) ( 7 ) S. Fronaeus, i b i d . , 64, 317-324 (1952).
Feb. 20, 1954
!)(i I
INDIUM HA4LIDECOMPLEXES STUDIED BY ION-EXCHANGE LIETHODS
TABLE I Sample no.
(A), mole/l.
... 1, 2 3, 4 5, 6 7, 8 9, 10 11, 12
0 0.06 .10 .15 .20 .25 .30
...
0 0.02 .05
1, 2 3, 4 5, 6 7, 8 9, 10 11,12
... 1, 2 3, 4 5, 6 7, 8 9, 10 11,12 13, 14
... 1, 2 3, 4 5, 6 7, 8 9, 10
In A soln , 7" ml./g.
.. 50.7 51.4 53.4 53.8 55.0 56.0
0 0.02 .n4 .06
.08 .09 .10 .12
0 0.88 X 1.76 2.20 2.64 4.40
Indium Iodide System ( 1 .2) 1.6 1.3 1.5 1.2 1.2 1.2
100 175 242 356 310
.. 57.4 64.4 71.4 77.0 79.0 79.2 84.5
(110) 76.0 55.3 40.1 29.9 26.5 26.3 18.4
Indium Chloride System (0.009) (21.0) (360) ,0132 23.1 380 425 ,0181 25.2 ,0250 29.7 480 ,0335 34.0 552 ,0377 35.0 580 ,0381 32.3 567 ,0545 42.1 709
1000 1625 2000 2400 2442 2070 2910
61.4 76.8 81.0 86.0 92.4
(110) 60.8 30.2 23.5 15.6 7.9
Indium Fluoride System (0.009) (550) (0.5 105) 0 . 7 X 105 ,0166 940 ,0331 1520 2.9 ,0426 1700 4.1 ,064 2300 6.0 .13 3000 10.8
At low values of (A) Xz
(0,009) ,0103 ,0106 ,0118 ,0116 .0122 ,0127
f,
(I./mole)2
93.5 65.5 50.4 37.8 26.9 23.1
51.7 60.4 66.5 72 6 78 8 81.2
.16 .20
97.0 94.6 87.3 86.0 81.8 78.6
g./ml.
41, I./mole
Indium Bromide System (0.009) (12.3) (137) ,0117 ..... .. ,0153 13.6 142 ,0199 14.8 151 ,0265 15.9 166 ,0372 19.5 194 ,0432 18.8 199
..
.08 .12
(1 10)
I/+,
Pz
(110)
+ PdA)
x
(4)
and from equations 2 and 4 we obtain A.f = PIA$JI - P,A(A)
(5)
Thus, from a graph of Af/A(A) v s . A+l,'A(A), PI is obtained from the slope of the line and Paas the intercept. From equation 4, a graph of X z vs. (A) should give a straight line of slope equal to /33 and intercept equal to Pz. Experimental Part In114,half-life 50 days, in solution as the perchlorate, was used. The indium concentration in the final equilibrium mixtures was 2.6-5.2 X mole/l. Probably because of the low concentration used, + was not found to vary with metal ion concentration as Fronaeus found in his work. Indium showed appreciable amounts of adsorption on the walls of the flask at this concentration, and it was necessary t o run blanks (no resin) with each system t o correct for this adsorption. Distribution ratios (6) were determined by radiochemical assay of the solutions after equilibrium was reached. One-milliliter portions of solution were evaporated in metal planchets one inch in diameter. Samples were counted with a n end-window G M counter a t uniform (approximately 5%) geometry. Sample counts of the order of 1000 c . p . m . were obtained, with a background count of 15-20 c . p . m . The per cent. indium o n the resin was obtained from the difference in activity between the original solution and the solution in equilibrium with the resin. Solution and resin were brought to equilibrium by mechanical shaking for one hour, followed by standing for 24 hours in
..
26.0 31.3 30.0 43.8 32.5
70 80 90 110 100
...
(180) 210 220 250 320 320 260 370
105 105 145 163 156 113 176 ...
. .
0.3 X 10* 1.4 1.6 2.1 2.3
4 . 4 X lo5 5.5 5.2 6.7 5.5
(6 X 106) 8 . 6 X 105 12.4 12.8 17.2 18.9
a constant temperature bath a t 25'. Studies on the time necessary for equilibrium to be reached showed no appreciable change in the system after 2 hours. The resin used was Amberlite IR-120 A . G . (Rohm and Haas Co., Philadelphia) converted into the sodium form a t p H 3.8, air-dried, screen size 28-60 meshes per inch, moisture content 12.68%. All values of were calculated on the basis of oven-dry resin (0'% moisture). The resin had a n exchange capacity of 3.8 meq. per gram of oven dry resin. The P H of the solutions was maintained at 3.8, since In(OH), precipitates a t pH values slightly above this. A Beckman P H meter and glass electrode were used t o measure pH. Distribution ratios (+) were found to be constantwhen solution to resin ratio was changed, indicating absence of colloidal indium. The ionic strength of all solutions was kept a t one usingsodium perchlorate as the added electrolyte. Additional work has been done using carrier-free In116 aq the tracer for radiochemical assay, and general confirmation of the work with In114was obtained. However, the adsorption of In115on the container walls at the carrier-free concentrations used (approximately mole/l.) was a considerably greater factor than for the concentrations of 10-5 mole/l. used in the work with In114. Therefore these results were considered less reliable and are not reported here. The data obtained are given in Table I. The data for the indium chloride system gave the most satisfactory values for PI, P? and Pa. Values of Aj/(A) were plotted against values of A+l/(A) according to equation 5. From the slope of the line, PI = 26, and from the intercept, pi = 1500. From a graph of X? v s . (A) (equation 41, a more reliable value of 03 = 1680 was obtained from the slope of the line, and 01 = 170 as the intercept. An attempt to determine p4 from equation 3 resulted in a negative value for &, indicating the data are insufficiently accurate for determination of p4a t concentrations of chloride ion up t o 0.12 mole/l.,
+
962
JOSEPH
-4. SCHUFLE
4ND
and t h a t the amount of InC14- formed a t these concentrations is probably very small. (pi - 1) obtained by extrapolation of $1 to (A) = 0 was 21.0. Thus 1 = 5, 11 = 0.021 l./g. = ( M A R ) / ( M A ) . The value of ( K a R ) / ( N a ) is (0.0038 mole/g.)/(l mole/l.) = 0.0038 l./g. Thus the ratio (MAR)/(MA) is about five times the ratio ( X a R ) / ( S a ) shoying an appreciable exchange of h-a+ for MA++ on the resm. The data for the indium bromide system, treated in the vime way as the data for indlum chloride, gave formation constants as: p1 = 16; p. = 60; p 3 = 300. (PI - I ) was found to be 12, and thus 1 = 4 and ZI = 0.028 l./g.
The data for the indium iodide system did not justify estimation of formation constants higher than pl. Appreciable formation of complexes higher than the first seems to be lacking in iodide concentrations up t o 0.3 mole/l. In the iodide system, (PI - 1) = 1.2. The average value of 11 from the chloride and bromide systems is 0.02 1. g. If this value is used for the iodide system: (PI - I ) = PI - (Zi/lo)pi = - (0.02)/(O.llO)p1. Thus PI =
H.
VOl. 76
PITORRIS EILAND
Values of Pz = 6 X lo5and 6 3 = 4 x lo8were found and are probably significant only as to order of magnitude. A consideration of the effect of the hydrolysis of indium on the formation of the halide-complexes, as suggested by Hepler and H u g ~ sindicates ,~ that the value of p1might be increased by about a factor of two by the presence of InOH++ ion in solution, a t p H 3.8. When an additional factor of approximately five is included to correct for differences in ionic strength, the total factor of ten brings the results of this work into reasonable agreement with the work of Hepler and Hugus. The degree of formation of the individual complexes, a,, may be calculated, once the formation constants are known, from the equation an =
B,,(A)"/S
(7)
Then from the relation given by Bjerrum10
3 -.
In studying the indium fluoride system i t was necessary to correct the concentration of added fluoride for the formation of HF. The fluoride concentration was calculated from the equation
if log a, is plotted against p(A), G will be equal to n where the slope of the curve is zero. Thus the data for the indium chloride system given in Table 11, and Fig. 2, show the areas of chloride concentra(6) (F-)= ( ( H ) / K d 1)!A) tion in which the first two chloride complexes of where (A) is the concentration of added sodium fluo- indium predominate. InCl+ is the predominant ride. KHF is 6.9 X for low ionic strength.8 species a t chloride concentrations of 0.05 mole/l., The value of this constant would be approximately whereas InClz+is the predominant complex a t 0.16 doubled for an ionic strength of one. Therefore a mole/l. A similar determination from the data was used for KHF,and a t PH for the indium bromide system shows that the Invalue of 1.4 x 3.8 equation 6 becomes (F-) = 0.88(A). The data B r + + and InBrz+ ions predominate a t bromide ion for the indium fluoride system gave = 1000. concentrations of 0.08 and 0.3 mole/l., respectively Summary.-The aqueous indium halide systems have been studied from their equilibria in the presence of an ion-exchange resin and the data analyzed by the method of Sture Fronaeus. For mation constants obtained are In-C1 system: 81 = 36; p 2 = 170; p3 = 1680 In-Br system: 6, = 16; p? = 60; ps = 300
+
In-I system: p1 = 2 In-Fsystem: p1 = 1000; p2 = 6 X 105; p8 = 4 X lo8
Halide concentrations a t which the first two complexes in the chloride and bromide systems predominate are determined. TABLE I1
I
.5
'a
1.5
1.0 p (A).
Fig. 2. (8) "International Critical Tables," Vol. 6, p. 260.
2.0
I
(A), mole/l.
$(A)
a1
-log a,
ar
-log a,
0.01 .03 .06 .10 .15 .20 .25 .30
2.000 1.523 1.222 1.000 0.324 ,699 ,602 .523
0.201 ,391
0.697 ,408
0.014
1.854 1.109 0.758
,438
,358
.372 .268 .194 ,145 .lll
,430
.572 ,712 ,840 .959
.078 .175 .245 .266 ,256 .240 .221
,611 .575 .592 ,620 .656
SOCORRO, KETVMEXICO (9) L. G. Hepler and 2. 2. Hugus, Jr., THISJOURNAL, 74, 6115 (1952). (IO) J. Bjerrum, ref. 1, p. 22.
