LVAYNEW. DUNKING AND DONS. MARTIN, JR.
5566
monium 6-molybdochromate(III) as well as his refined directions for preparation of ammonium 12molybdocerate(1V). \Ye are also indebted to Mr. William A. Light for assistance in obtaining some of
[COSTRIBUTION Xo. 717
FROM THE
Vol.
s1
the preliminary basicity data and Mr. Donald Broadhurst for obtaining some of the coagulation data. POTSDAM,
SEW
YORK
INSTITUTE FOR ATOMICRESEARCH ASD DEPARTMEST OF CHEMISTRY, IOWA STATE COLLEGE]
Mixed Tetra-(ch1orobromo)-platinates(I1). Equilibrium Constants for Formation in Aqueous Solution' BY i $ r i a ~ n -W. ~ DUNNING AND DONS. ~UXRTIP';, JR. RECEIVED APRIL I, 1959 A radiochemical procedure, combined with a least squares treatment, has been used to obtain the four equilibrium constants for the stepwise replacement by bromide of t h e ligands in [PtClr]'. The equilibrium constants for t h e single ligand replacement reactions a t 25' are 14.5, 8.3, 2.0 and 1.65. The AHo for each step is estimated t o be -1.8 + 0.3 kcal. The results fail t o show that the higher stability of the bromide ligand bond results primarily from x-bonding. Upper limits for the acid hydrolysis equilibrium constants in the system were also obtained.
Introduction
a-bonding. Electron pairs of the platinum ion, normally considered as non-bonding, are utilized evaluate the set of four equilibrium constants for in the a-bonds which therefore give multiple bond the successive substitution reactions indicated by character for the complexes. Chatt and Wilkinsg have found that AHo for the isomerization of ciseq. 1 to trans-dichlorobis-(triethy1phosphine)-platinum[PtC14-nBrn]- B r - % (11) is nearly +2.5 kcal./mole. They suggested [PtC13--nBrn+1]-4- C1-, K , (1) that the stability of the cis-isomer was possibly where n = 0 , I , 2 , 3. Platinum(I1) belongs to the due to the a-bonding, since the two n-bonds to rather limited group of ions for which the order phosphorus could utilize both the d,, and d,, of stability for halide ligand bonds is reversed from orbitals of platinum, whereas in the trans-isomer the normal, z.e., Br- is bonded more strongly than only one of these orbitals is suitable for forming C1-.2 L a t h e r 3 has estimated that for the over- a-bonds to the phosphorus atoms. Therefore, a n-bonding ligand is expected to detract from the all reaction n-bond trans to itself. [PtC14]4Br- --+ [PtBrr]' + 4C1(2) It appeared worthwhile to attempt an evaluation L I F O is -6.7 kcal. In addition, Leden and Chatt4 of the successive bromide substitution equilibrium found an equilibrium constant of 3.4 (AFO = -0.72 constants for [PtC14]-. If the stability of the kcal.) for the replacement by bromide of labile chlo- platinum-bromide bond does result froiri its ride trans to ethylene in trichloro-(ethylene)-plati- n-bond character, i t is expected that [PtCl,Br]= nate(I1). and cis- [PtC12Br2]', which have no bromides trans The higher stability of the bromide bond, com- to other bromides, will be especially stable. If the pared tochloride, may result from additional a- equilibrium constants could be deterniinetl with bonding for the bromide. Bromide has a stronger sufficient accuracy, it was expected that high values trans-directing effect than chloride, and Chatt, for KOand possibly K 1 would lend support to the et al.,5 and OrgeP have attributed the trans- n-bonding hypothesis. directing property of a ligand to its a-bonding Platinum(I1) forms the well-known square conicharacter. plexes which are inert. Consequently, halide In addition, bromide ligands in platinum(11) complexes with coordination number greater than complexes are more labile than chloride toward 4 have not been considered. Ample time must he exchange with the free ion.' Banerjea, Basolo provided to attain equilibrium in the systems. and Pearsons have presented evidence that the At moderate halide concentration, the acid hysubstitutions for ligands in platinum(I1) coni- drolysis (or aquation) can yield the trihalide complexes are rapid for entering groups which are plexes according to reaction 3 generally considered to provide a high degree of [PtCla- nBrn]- + H,O f, (1) Work was performed in the Ames Laboratory of t h e TI.S. Atomic [PtCl3-,,Br,(H2O)]C1-, K,,, ( 3 ) Energy Commission. wheren = 0, 1, 2, 3. (2) R. G. F. Carleson and H. Irving, J . Chem. Soc., 4390 (1954). (3) W. R l . L a t i m e r , "Oxidation Potentials," 2nd E d . , Prentice-Hall, Grantham, Elleman and ?\Iartin!O reported Inc., New York, h-.Y., 1952, p. 54, 59, 205. 0.018 mole/l. a t 25' for Ka0, the equilibrium con(4) I. Leden a n d J. C h a t t , J . Chem. Soc., 2936 (1955). stant for the nquation of [PtC14]=. An evaluation ( 5 ) J. C h a t t , L. A. Duncanson a n d I,, AI. Venazi, Chemist73 3 I n d u s Lvy, 749 (1935). of the acid hydrolysis equilibria of the mixed ( 0 ) L. E. Orgel, J . Inorg. Nuclear C h e v i , , 2 , 137 (1956) complexes is interdependent upon the dctermi-
X radiochemical procedure has been used to
+
+
+
(7) A. A . Grinberg and L. E . Nikolskaga, Z h u v . Puzklad. Khiwz,, 22, 5.42 (1949). (8) D. Banerjea, I'. Basolo a n d R. G . Pearson, TIXIS J O U R Z A L , 79, 4 0 j 5 (1957).
(9) J. C h a t t and R. G. Wilkins, -7. C h e i i i . Soc., 4300 (19Z2). (10) L . F. G r a n t h a m , T. S. Elleman and D. S. I I a r t i n , J r , J O L J R N A L . 77, 296.5 (1955).
7'111s
Nov. 5 , 1959
MIXEDTETRA-(CHLOROBROMO) -PLATINATES(II)
nation of the bromide substitution constants, the
K,'s. The Radiochemical Procedure.-To our knowledge, the experimental method for the determination of the equilibrium constants has not been utilized previously, and it deserves some comment. A weighed quantity of Kz[PtC14] and measured volumes of standard KC1 and NaC104, to adjust the ionic strength, were mixed in a volumetric flask and a small portion of C136tracer solution was added. Water was added to bring the solution to the mark. The solution was allowed to stand for a sufficient time so isotopic equilibrium between all the exchangeable chemical forms of chlorine was assured. The chlorine of perchlorate does not exchange. A measured volume of a standard NaBr or KBr solution was added to one portion of this solution. These bromide-containing solutions remained in a thermostat for a t least 40 hr. so equilibrium was again established. Aliquots were withdrawn from both the bromide-free and the bromide-equilibrated solutions. To each aliquot was added tetraammineplatinum(I1) nitrate solution. [Pt(NH3)4]++precipitates the very insoluble [Pt(NH3)4][PtCld] and [Pt(NH3)4][PtBrd]. Grantham, et aI.,1° found that it also precipitated I-, apparently without fractionation [PtC13(H20) from [PtC14]=. It was therefore assumed that it precipitated effectively all the mixed anion species indicated in equations 1 and 3 . The average specific activities (cts./min. g. ppt.) of the p_recipitates from the bromide-containing splution, Si, and from the bromide-fyee-solution, SF, were determined. The ratio, &/Sio, is a measure of the extent to which complexed chloride has been replaced by bromide; and this ratio was determined for a series of bromide-chloride ratios. Since the C1- contents of [Pt(NH&] [PtCL] and [Pt(NH&] [PtCl3(HzO)Izare virtually identical (6.662 and 6.647 mmoles/g., respectively), the aquation does not significantly influence the determination of specific activity per g. atom of chlorine in the bromidefree precipitates. However, acid hydrolysis in the bromide-containing systems must be considered. Therefore, the extent of hydrolysis was measured a t four different bromide-chloride ratios. This method possesses the advantage that necessarily laborious and difficult analyses for the individual halides in the mixtures are not required. The specific activity of the precipitates can be readily obtained by conventional radiochemical procedures, which must, however, be carefully executed. A complete recovery of precipitates need not be effected so long as a fractionation of the various complexed forms of chloride does not occur during the precipitation. The precipitates must be washed free of the solution and dried sufficiently to give representative deposits of all the chlorocomplexes. With the consideration of acid hydrolysis in the system, a satisfying evaluation of the equilibrium constants from the data would not have been feasible without a capable electronic digital computer. An International Business Machines 650 computer was used for the calculations. With the computer i t was possible to employ a least squares
5567
treatment which fully utilized the experimental data. Experimental Materials and Equipment.-The KZ[PtCla] was similar to the material used by Grantham, et al.1° An activation analysis of the platinum indicated an iridium impurity of ca. 20 Gg. Ir/g. Pt. [Pt(NH3)4]C12 was prepared by the method of Keller." A solution of [Pt( N H B ) I I ( K O ~was ) ~ prepared by adding t o a solution of the chloride salt. The nitrate was used as the precipitating agent for the halide complexes, so t h a t large quantities of inactive chloride which might disturb the isotopic equilibrium would not be added. However, a small amount of chloride was allowed t o remain in the solution since silver ions also precipitate the anions. Standard solutions of KBr and h-aBr were prepared by weighing the dried reagent grade salt into a volumetric flask and adding water to the mark. Standard solutions of Sac104 were prepared by titrating NaOH with reagent grade HC104 t o a pH of 7. The C136 tracer was obtained from the Isotopes Division of the U S . A.E.C., Oak Ridge, Tennessee, and it was carried in dilute HCl. Distilled water, redistilled from alkaline KMn04, was used for the preparation of the solutions. The counting equipment and the procedures for preparing the precipitates, for weighing and counting the samples have been described in ref. 10. Data.-Sixteen equilibrium solutions with the tracer were prepared a t 2,5". All solutions contained a quantity of the K2/PtC14] to give a platinum concentration of 0.0166 molar. The minimum halide concentration was 0.134 molar to suppress the acid hydrolysis of the complex; however, for a chloride system the aquation amounts to 12.57'. The ionic strength was maintained a t 0.318 M (based on zero acid hydrolysis) by the addition of the calculated amount of the standard NaC104 solution. For each specific activity determination, five to seven samples were withdrawn and precipitated separately. The experimental conditions and results of these determinations are given in Table I . Each of the indicated uncertainties rspresents the standard deviation for the average, S, or S,O of the N determinations, calculated by the expression usi =
(Si
- Si)Z/N(N - 1)
(4)
l i i
The ratio of the specific activities for each solution, RI-obad. I t s standard deviation, UR,,was calculated by the expression = .$,/SF, is included in Table I.
+
u 2 R I= d $ , / ( S ? ) * ~g,Os,~/(s?)' (5) Five solutions were prepared which contained 0.0166 mole &[PtClJ/l. and different concentrations of KBr. Enough A-aClOa had been added to give an ionic strength of 0.318 M . After equilibration with respect to aquation, the acid content was determined by rapid titration with AaOH, employing a glass electrode. A pH of 9 was taken as the end-point for the titrations. The equivalents of acid in the solution were assumed to equal the moles of HzO in the trihaloaquoplatinate(I1) species. Halide concentrations were purposely held to a low level so a high degree of acid hydrolysis would be provided for the experiments. The conditions and the results of these determinations appear in Table 11.
Computation of Equilibrium Constants In all of the solutions, the ionic strength was substantially constant so the activity coefficients were therefore assumed to be constant also. It is expected, therefore, that the equilibrium concentration ratios, defined in equations 6 and 7 will not vary, and they have been treated as equilibrium constants.
-
n -
[PtCls-,Br,+~'l [PtC14-%Br,']A
(11) R. M. Keller, "Inorganic Syntheses," Vol. 11, W. C. Fernelius. Ed., McGraw-Hill Book Co., Inc , New York, K.Y . , 1946, p . 250.
WAYNEW. DUNNING AND DONS. MARTIN, JIZ.
5568
Vol. 81
TABLE I [pt( S & ) 4 ]
E X P E R I M E N T A L C O S D I T I O X S AS11 l < E S U L T S FOR TlIE P R E C I P I T A T I O X 01' h h E D BROMIDE-CHLORIDE
[I'tSa]-- [Pt(Nk13)4] [l'tSs(IirO)]z
FROM
SOLUrIOSS
Temp. = 2 5 O , K2[PtCI4] added = 0.0166 mole/l., and ionic strength = 0.315 tuole/l. 1 3 ~ -added,
CI added,' molesll.
Specific activity of ppt., ct./min. mg. Bromide Chlori4e only
[Brt,,t]. moles/l.
sI 47.37 f 0 . 2 1 0.0083 ,0166 62.90 i . 1 1 51.37 i , 2 5 .I1249 ,0382 45.26 i .26 ,0415 32.82 i . 2 4 ,0498 3 1 , 0 5 d= .30 ,0581 26.43 i .30 ,0004 23.56 i .16 ,134 ,0830 21.94 i . i l 4 ,134 ,0996 22.61 i . 12 ,134 , 1162 18.30 i . 0 6 13.1 1328 16.51 i .05 ,1494 13.94 f . l I ,1188 ,1092 ,1660 13.04 rir: .07 0856 ,1826 13.33 f . 0 4 ,0690 .1992 10.97 =k .04 I n addition t o 0.0664 triule Cl-/l. in
0.134 ,134 ,131 ,134 . 134 ,134 ,134 .134
O4
Si0
['l'iterohid. 1. erluiv. acid/.
0 0,0090 .0170 .0408 .0827
0.0103
0 .0104
. 00791
007XG ,005DO
00588 002x" . 00104
00'85 CIO 1iJ 3
Calcd. ratio R , - calcd.
0.891f 0.006 53.06 i 0 . 3 0 77.81 i .55 ,808i ,000 ,737 f .006 6Sl.67 i . 4 1 68.41 i . 1 i ,662 + ,004 55.44 f ,83 ,592 rt ,006 57.55 i ,36 .540 i ,005 53.37 f .30 ,495 i ,005 49.35 rt .l-4 ,457 zt ,003 54.56 + . 6 1 ,402 rt .0')5 66.1% f .23 ,342 f ,003 63.28 i , 18 ,291 i .001 6-2.6s i . 4 0 ,255 i ,002 60.30 i. .77 ,231 i ,004 68.39 i . 4 7 ,191 rt ,002 83.80 i .56 .147 i ,001 91.95 f .50 .119 i .001 the 0.0166 iriole L[PtCl,]/l.
T.4BLE 11 E~~~~~~~~~~~c~~~~~~~~~ AsD R~~~~~~ FOR PH T I T R A T I O N S O F BROMIDE-CONTAINING SOLUTIONS OF K*[PtC14] Temp. = 2Z0, Km[PtCI4] = 0.0166 mulc/l., ionic stretigtli = 0.318 ixiole 11.
I < r - adrled [Rriotl, nioles/l.
Obsd. ratio Ri - obsd.
=
A calcd. [13rr]/
0.1410 .1457 ,1500 . 1539 .1574 ,1605 ,1634 ,1660 ,1704 . 1740 . 1770 ,1796 .lo75 .l539 .1398 ,1254
0.0216 ,0443 ,0684 ,0944 ,1214 ,1500 ,1812 ,2127 ,2776 ,3465 ,4159 ,4581 ,6030 ,7453 ,9202 1.1395
ic1-1
6i
$0.006
0.897 ,806 ,720 . tj57 ,598 .54G
- ,003 - ,011 - ,005 ,006 .00i
+ + + +
,499 ,459
,004 .002 - ,009 - ,002 ,007 .007 -,.011 - ,009 ,001 ,001
,393 ,340 ,293 ,363
+
+
. ""0 Y-
,181 ,148 ,120
+ +
neither A nor [el-]is obtained directly from the data. The calculations always started with the assumption of values for the two sets of constants, k , and k a x . For each experiment in Table I [Cl-] and A were found by a trial and error reiteration in which the concentration of each species was coniputed by equations S and 9. From a set of calculated concentrations, the complexed bromide and coniplexed chloride concentrations, [ETcomp 1 and [Clcomp],were calculated by equatioiis 10 and 11 4
[BrcnlOp] =
ti [PtCL11
where n tions
CI calcd , moles/l.
,J3rr8=]f
=0
3
0, 1, 2 , 3 ancl brackets denote concentra-
~ [ P ~ c I ~ - , , I ~ ~ , , ( H (10) ~o)--I 7z=O
A
=
[Br-]/[CI-]
4
[c~,~,,,,I
The constants for reaction 3 are
+
(4 - ~ ) [ P ~ c I ~ - ~ u ~ ~ ~ I
= 1,
=0
( 3 - n)[PtClS-.Br,,(€I,O)
where ?z = 0, 1, -3, 3. I t can be easily showi froiii equations ti arid '7 that the concentration of each tetralialoplatinate(I1) specics is given by the general expression IPtC14-,,Br,,"]
= [Ptt,,t]
(2
kn..l .Af*
n =0
where [Pt,,,]
+
=0
)
(8)
total platinum coiiceIitration (o.OI(jti 31.)- k-l = 1; KO = K O ;ki = Kok'l; k2 = K o K I K s ; and ha = KOKlK2K3. The concentrations of the trichloroaquoplatinates(I1) are given by =
[PtCl,. ,zBrrL(l120)-] = [Ptt,t]k,- I . ~ ' ' K ~ ~ / [ C X~ - ] 1
+ iCl-1
(12)
+ [Brol,mp]i 0.001 JI
(13)
lClt"t1 = [CL"mlJl
k , , . LA'LK:,n/[Cl-l 72
\
'I'he total chloride coiicentration [Cl,,,] and tlie total bromide concentration [BrtOt]arc known for each of the 16 experiments in Table I. However,
(11)
[Cl-] aiid A were varied for the cxperiineiit until the calculated [Clconr,,]and [Brcuiiil,] satisfied the relatioiis 12 and 13
k,, . I L ~ " / 3
]
n=O
arid [Brt,t] = A[C1-]
X calculated value of X i - c a l c d . was obtained for the experinient by the expression given as equation (141, where AfW = molecular weight of the corresponding [Pt(NH&++] - salt, 0 refers to the solution which has no added bromide. The quantity 6i = Rl-calcd. - X i w o b a d . was next calculated. 'I'heii Z 8 i 2 / R 2 i - o ~ s ~ ~was . evaluated for the entire set of experiments in Table I. 'I'he cyuilibriuiri constant ko was then changed by soiiie illcremetit a i i t l with such a change in ko, a new value of z 6 i Z / R 2 i - o ( , a d . was calculated. The process was repeated until the value of ko, within a 25& change, was fouiid
MIXEDTETRA-(CHLOROBROMO)-PLATINATES(~~) 5569
Nov. 5, 1959
+ [PtC13(H,O)-lD MW/2) 3
the evaluation parameter. However, a change [Cloomp]O [ P t C L ,,Brn'1 IMW [PtCla- .BrdHtO)-I M W D ) in AFnoof 0.25 kcal. will \n=O n=O / also double 86i2/R2i-,,bsd. which gave the minimum value of z6i2/R2i-,,bsd. Therefore, the uncertainties in the AFo0 for the reNext in turn, k l , kz and k3 were each individually actions in equation 1 have been arbitrarily and varied to give the minimum value in z6:8i2/RZi-obsd. rather conservatively, it is believed, set a t + 0.25 The cycle was repeated until no changes in the kcal. an's were obtained. Originally, a set of k,'s was TABLE IV evaluated with all K,,,'s assumed equal to zero, DEPENDENCE OF THE EVALUATIOX PARAMETER Z6,2/R2i-ot,.d. ie., with the assumption of no acid hydrolysis. This initial set of equilibrium constants is given in UPON THE BROMIDESUBSTITUTION CONSTANTS, Kn'S WITH THE FINAL SET OF K,%'s FROM TABLE I11 Table 111. Z6i2/
Ri-cs~od.
[Clcomp] ( [PtC11'l0 MW -
=
(
4
+
TABLE 111 CALCULATED EQUILIBRIUM CONSTASTS Kan
fin 71
-(Initial
0 1 2
15.7 127 252 418
3
(14)
set)-
0
0 0 0
kn
Kil.7,
final set)---
14.5 120 239 395
0.018 .0141 .0055 ,0039
This initial set of k,'s from Table I11 was combined with a set of assumed Kan's. The consistent values of [Cl-] and A were then evaluated for each experiment mentioned in Table 11. The acid content [Titercalod.]for an experiment was calculated by the expression 3
[Titercalod.]=
[PtCla. .Br,(H,O)-] n=O
(15)
KO Firial s e t 14.5 16.7 18.9 9.7 14.5 14.5 14.5 14.5 18.9 14.5 9.7 18.7
KI
Ka
8.30 8.3 8.3 8.3 7.2 10.8 8.3 8.3 6.4 8.3 8.3 6.4
1.99 1.99 1.99 1.99 1.99 1.99 1.73 1,99 1.99 2.59 1.99 2.58
Ks
1.65 1.65 1.65 1.65 1.65 1.65 1.65 1.90 1.65 2.15 2.48 1.27
R2i
-
obsd.
0,0069 0.007 1 ,0082 ,0137 ,109 .l54 .0201 .0216 ,0136 .18U7 ,1210 .01410
Discussion For the bromide substitution reactions 1, the charges of product ions are the same as the charges of the reactant ions. Since in all the experiments the ionic strength was substantially constant, the calculated equilibrium constants, k , or K,,, given in Table 111 and IV, are expected to be a very good measure of the standard free energy changes of the reactions according to the familiar expression AFn' = -RT loge Kn (16) Values of the indicated AFO for the four successive bromide substitution reactions are given in Table V. Latimer and Jolly12 have proposed a
The bromide-free experiment, of course, fixes K,o. The value of K,1 was changed (by increments of 59.;) until the best fit for the acid content of the second and third points (0.0090 M and 0.0179 M total bromide) was obtained. With this new value of Kal, K,2 was then changed until a fit with the fourth experiment was obtained. With the new values of Ka1 and Ka2, Ka3 was changed until a fit with the fifth experiment was obtained. The cycle for the calculation of the three K,, constants was repeated until there was no further change in them. With the new sets of constants, kn's and Kan's, TABLE V which were then a t hand, the two program cycles THERMODYNAMIC QUANTITIES AT 25' FOR THE REACTIOKS of computation were alternately repeated until [PtCln-nBr],' 4-Br- & [PtCI3-,,Brr, + J' + CIthere were no further changes in any constant. Statistical factpr A H d (Estimated) kcal. The final sets of equilibrium constants are also CIS: CIS.' Pd, cis trans t r a m = cis lvans trans= given in Table 111. I t is to be noted that the final n Akcal. only only 2:1 only only 2:l values of the kn's were not greatly changed by the 4 0 -1.58 4 4 -1.74 -1.74 -1.74 treatment which included the acid hydrolysis. -1.99 -2.23 -2.64 1 1/2 3/2 The values of ~8i2/R2i-ohs~. for a number of 1 -1.25 1 -0.98 -1.63 2/3 -1.39 2 trial sets of equilibrium constants, K,,have been 2 -0.41 -2.10 -2.10 -2.10 listed in Table IV. The minimum value of 26i2/ 3 -0.30 1/4 1/4 1/4 - _ _ R2i-&sd. for the final set of equilibrium constants -3.54 -7.46 -7.46 -7.4G was 0.9 x 10-3. This value is somewhat larger than Ea2~i/R2i-obsd. means to estimate A S for the replacement of H 2 0 Such a result suggests in a metal ion complex by a ligand. An extension which was 1.3 X that other than purely statistical errors may be of their method gives A 9 = - 3 . 3 e.u. for the recontained in the Ri's. However, the number of placement of chloride by bromide on a metal ion individual determinations for each condition was complex in the +2 oxidation state, ;.e., for each iiot sufficiently large for conipletely satisfactory of the reactions in equation 1. The entropy statistics. In addition, the indicated individual changes, however, must include contributions of values of 6i are not alarmingly large. From Table statistical factors arising from the symmetry of the IV it is noted that K Ohas not been established as species in the reactions. The importance of such accurately as the other K,'s. For example, a statistical factors for acid-base equilibria was change of 30'z in K1, Kz or K 3 more than doubles (12) W. M. Latimer and W. L. Jolly, THISJOURNAL, 76, 1548 (1953).
55'70
LYAYNE
w.D U N N I N G AiXil) D O N s. hI,\RTIN,
emphasized in the review article by Br@nsted13 and the topic has recently been discussed by Benson.14 The present experiments do not distinguish between the possible cis- and transisomers of dibromodichloroplatinate(I1) ; and in the equilibrium expressions [PtClzBrz=]corresponds to the sum of the two isomers' concentrations. Therefore, unambiguous statistical factors for the contribution to ASlo and ASzocannot be given. The statistical factors have been included in Table V for the three cases: (1) cis-isomer is dominant; ( 2 ) trans-isomer is dominant; (3) cis-isomer : trans-isomer = 2:l. The latter case corresponds to the distribution of isomers expected on the basis of statistics alone, ;.e., if AIIO for isomerization equals zero. The values of total entropy changes were therefore calculated from the expression AS,? = -3.3
+ R log,
(statistical factor) (17)
+
The estimated values of 4H,2 == AF,? TAS$ also included in Table V. The high value for AHzo together with the low value of AH2 for the trans-isomer only do not appear reasonable. The alternation in values of AHO , for the cis-isomer only is somewhat larger than the expected uncertainty of 0.3 kcal., and no obvious cause for such alternation is apparent. In particular, the estimated values of AHoo and AH? are approximately equal, and there appears to be no exceptional stability indicated for the addition of the first bromide. It is believed likely, therefore, that the distribution of isomers is given approximately by the statistical ratio of cis:trans = 2:l. In this case, the estimated AH,? for all the substitutions differ by no more than the uncertainties, and there is no indicated additional stability for placing a bromide ligand trans to a chloride rather than trans to a bromide. Poulsen and Bjerrum'j have pointed out that the assumptions of Latinier and Jolly, l 2 which predict constant AS0 values for successive substitution reactions, do not give ideal agreement with mme metal ion-ammonia systems when the assumption of constant AIIO for successive reactions is applied. Hence, variations in the actual AS0 values provide additional uncertainties to the estimates for the AH0 values in Table V. I t is interesting to note that with an entropy cliaiigc of - 3 . 3 e.u., the AFO of -0.2 kcal. for the replace-
ment of the trans-chloride in trichloro-(ethylene)platinate (11)3 corresponds to a 4H0 = -1.72 kcal., in very good agreement with the estimated values for the reactions of equations 1. I n that case, the trans ligand is believed to be strongly a-bonded. With the substituted phosphines in the complexes which Chatt and Wilkinsg studied, the a-bonding to phosphorus is expected to be much stronger than to bromide ligands. Also, a portion of the 2.4 kcal. which they obtained for Ai3 for the cis to trans isomerization of dichlorobis-(triethylphosphine)-platinum(I1) may result from the interaction of the dipole moment of the cis-isomer with the solvent. With the uncertainties in the values of the Ii,i's and the low values for the AH,O, it did not seem profitable to attempt the determination of the temperature dependence in the equilibrium constants. Some recent experiments with Kz[PtCl,] solutions in this Laboratory have indicated that in dilute solutions with low halide concentration a second acid hydrolysis process occurs to a detectable degree according to equation 18 [PtCla(HpO)]-
+ Hi0
[PtClz(HiO)2]
+ C1-
(18)
Such a second acid hydrolysis would be significant in the acid-content titration for the determination of K,, in Table 11; but i t would be negligible in the high halide solutions for the broniide-substitution experiments of Table I. Therefore, the values of K,,, obtained must be considered upper limits to the actual first acid hydrolysis constants. In the treatment of the bromide-substitution experinleiits of Table I, the acid hydrolysis has therefore probably been overestimated. It is interesting to note that the four acid hydrolysis constants K,o: K,1: I&: Ka3, with the exception of K,?, followed rather closely the ratios predicted by the statistical factors of 1:3,'4:1/'2:1/4 for the case of cis:tians = 2 : l . Since the K,, values did not change substantially when the acid hydrolysis effect was introduced, the final set of eyuilibriuni constants for reactions 1 appear to be valid. .I further investigation of the second acid hydrolysis processes of [PtC14]=is underway. Acknowledgment.-V'e wish to acknowledge the assistance of the Statistical Laboratory of lowa State College in providing the facilities for the calculations with the IBSI 650 computer.
Reus., 5 , 3 2 2 (1928). (14) S. 1%'. Benson, 1'HIS J O I J R S A I . , 80, 5151 (1958). IO\VA l. (16) I. Poiilsen a n d J. Bjerrum, . I c I , i ('Izeni. 5 ' ~ a n d, 9 , 1107 ( l ! ~ s ~ ~ . ~ AhZES, (13) J . pi. Bronsted, Cheix.
Vol. s1
JR.
