10
THOMAS S. ELLENAX, JOHN 1 %'. REISIIUSAKD DONS. MARTIN, JR.
mmoles; in an entirely independent case, the corresponding values were 2.02 and 2.05 mmoles, respectively. With respect to the role of [Pt(en)(en-H)]O, the experiments involving this intermediate show that the decomposition of [Pt(en)z]O is more complex than originally expected. The added complexity, however, arises from the unavoidable presence of excess potassium in the reaction mixture; this results in concurrent reactions that are not involved in the over-all reaction of [Pt(en)z]I*with one equivalent of potassium since in this case the potassium is removed prior to the appearance of [Pt(en)z]". Although [Pt(en)(en-H) ] undoubtedly appears in the reduction of [Pt(en)~]Izwith two equivalents of potassium before the excess potassium is removed by washing, the intermediate must react only slightly since after 3 to 4 hr. the final decomposition product, [Pt(en-H)z]O, is present to the extent of only about 10% of the Pt used originally as [Pt(en)z]Iz. This interpretation is supported also by the persistence of the blue color a t the end of the addition of potassium solution. The unusual species derived from [Pt(en)~]Izand reported in this and the preceding communication2 may be summarized as (la) [Pt(en)JI (2a) [Pt(en)(en-H)]O (lb) [Pt(en)2]O (2b) K[Pt(en)(en-H)] (3a) [Pt(en) (en-H)]I (3b) [Pt(en-H)2l0 (3c) K[Pt(en-H) (en-2H)I
The complexes l a and l b contain platinum in the apparent 1+ and zero oxidation states. These products are unstable with respect to decomposition to elemental platinum and ethylenediamine a t room temperature; a t -33.5' they decompose with
L-01. SI
evolution of hydrogen and the formation of one or more of the basic complexes (3a, b, c). Although the dark red [Pt(en)n]I and the light pink [Pt(en)*]O are both insoluble in liquid ammonia, their stabilities are of such a low order that evidence for their existence could be obtained only by analysis of their decomposition products. The compounds 2a and 2b also contain platinum in their apparent I + and zero oxidation states, respectively, and their formation involves also the removal of a proton from a nitrogen atom in ethylenediamine coordinated with platinum. These are white ammonia-insoluble solids and evidence for their formation also was obtained through the study of their decomposition products. The complexes 3a, 3b and 3c appear not only as decomposition products of (la) and (lb) but also are formed by the interaction of [Pt(en)*]Iz with one, two and three equivalents of potassium amide. All three of these substances contain platinum in the 2+ oxidation state and although very hygroscopic, they are stable a t room temperature in a dry inert atmosphere. Accordingly, these products were isolated, purified and subjected to direct analysis. Aqueous solutions of these products are strongly basic; this of course indicates the displacement of a proton from water. Both [Pt(en)(en-H) ]I and [Pt(en-H)z]O are insoluble in liquid ammonia as well as in more than 20 common organic solvents. Further studies on the structure of these unusual species are either in progress or are anticipated. Certain evidences suggest that a t least some of these substances are polymeric; if so, the presence of platinum in 1+ and zero oxidation states may prove to be more apparent than real. AUSTIN,TEXAS
[CONTRIBUTION No. 603 FROM THE INSTITUTE FOR ATOMICRESEARCH AND DEPARTMENT OF CHEMISTRY, IOWA STATE COLLEGE. W O R K WAS P E R F O R M E D I N THE AMES LABORATORY O F THE u. s. A T O M I C ENERGY COMMISSIoX]
Trichloroamrnineplatinate (11) Ion. Hydrolysis, Isotopic Exchange of Chloride and the trans-Effect BY
THOMAS s. ELLEMAN, JOHN
w.REISHUS
AND
DONs. M A R T I N ,
JR.
RECEIVED APRIL 14, 1938 Rates of isotopic exchange between the chloride ligands in [Pt(I\;H3)C13]-and in the acid hydrolysis product, [Pt(SHs)Cl2(H20)],with C1- have been determined in aqueous solutions in the temperature range 0-30". For each eoinplex the exchange occurs by the observable acid hydrolysis and by additional chloride-independent processes. I t is proposed that the non-equivalent chlorides in each complex undergo separate acid hydrolyses. Although all such hydrolysis reactions have roughly the same rate constant and AH*, there is considerable variation in A?". In support of this proposal, the substitution of chloride by hydroxide in [Pt(NHa)C12(H20)]has been found by a tracer technique to be unsymmetric, indicatitig the presence of the cis-isomer. The mechanism of the acid hydrolysis reactions is discussed.
The authors have reported in an earlier publicaIntroduction The reaction of trichloroammineplatinate(I1) ion tion2 studies of the kinetics of the reversible acid with ammonia yields predominantly the cis- [Pt- hydrolysis of [Pt(NH3)CI3]- and [Pt(NHs)Cl*(NH3)2C1z]. This procedure for the preparation of (H20)] in dilute solutions. The system was dethe cis-isomer is one of the most frequently cited ex- scribed by the reactions amples of the trans-effect which illustrates that a k- 1 chloride ligand has a greater trans-directing influC1ence than ammonia. By this means the isomer is [Pt(NHs)C13]- + H20 f, [Pt(XH3)CI,(H20)] ki formed which is less stable in a thermodynamic (1) sense since i t is converted upon heating into the trans-isomer. (2) T.S. Elleman, J. W.Reishus and D S. Rlartin, Jr., TIIISJ f J t J R -
+
(11 A . V. Nikolaev, Compt. rend acad.
sci. U.R.S.S., 20, 571 (1938).
NAL,
8 0 , 536 (1958).
TRICHLOROAMMINEPLATINATE(~~) ION
Jan. 5, 1959 and HzO
+ [Pt(NHa)Clz(HnO)l
__ k-
11
which contains chloride trans to ammonia and other chloride trans to chloride.
z
Experimental
kz
[Pt(NHa)CI(HzO)z]
f C1-
(2)
Mateiials.-KIPt(NHa)Cla] was prepared by the method
described earlier.' [Pt(NHs)Clz(HnO)]. .was formed when solutions of [Pt(NHa)Cb] - were equilibrated with HzO by reaction 1. Passage of dilute equilibrated solutions of R set-1, A H I S = 18.9 kcal. kl = 3.6 X [Pt(NH3)Cla] through a column of anion-exchange resin, k-1 = 2.5 X IO-* !. mole-' sec-1, AHl+ = 16.8 kcal. Dowex 1, in the sulfate cycle, removed [Pt(NH&?l!]k z = 1 X 10-6sec.-1 and C1-. The remaining [Pt(NH,)Clz(HzO)] solution k-2 = 0.2 1. mole-' set.-' was stored in concentrations not greater than 0.005 M a t The equilibria in equations 1and 2 were re-established It was not possible by the earlier experiments to 0.0'. but the species [Pt(NHs)CL(H20)] comprised approxievaluate the stereochemical features of the reac- mately 85% of the total platinum. The remainder was tions. The trans-effect predicts that the reaction to mostly [Pt(NHa)Cl(HzO)2]+. Tetraphenylarsonium acetate, (As(C&)4) (CZH~OZ), was produce the cis-isomer occurs more rapidly, i.e., prepared by passing the (As(CeH&)Cl reagent, purchased k, kt for the reactions from the Hach Chemical Co., through a Dowex 1 anionexchange resin in the acetate cycle. The effluent solution LI I was standardized gravimetrically by the precipitation of (As(C+~)~)(C104). When the acetate reagent was added to equilibrated solutions of K[Pt(NHa)Cla], the compound (As(C&Q4) [Pt(NH3)C13] precipitated quickly. For the analyses of these precipitates, samples which had been dried at 95' were weighed and dissolved in hot water. An excess of LiClO, was added, and the. (As(CSH&)(ClO!) C1 I which precipitated was filtered and weighed. The platiHJG-Pt-OHZ C1num and NHs analyses of the filtrate followed the proI kt cedures used for the K[Pt(NH,)Cls] samples2 Results: c1 (As(CsH&)+ = 53.7, calcd. 54.5; P t = 27.2, calcd. 27.7; In the cis- [Pt(NH3)Ch(HzO)], of course, the two NH, = 2.5, calcd. 2.4. Chloride was not determined since C1-analyses of K[Pt(NHa)Cb] were generally of rather chlorides are non-equivalent, whereas in the trans- poor solution, precision? However, a (As(C&)JNOa compound the chlorides are equivalent. p H 4.65, was added to a solution of 0.01 M K[ Pt(NHa)C&], The present study was undertaken to determine pH 4.60, which had been aged for 20 hr. The p H after the was 4.65. Therethe rate a t which isotopically tagged chloride ion in precipitation of (As(C6Hs)r) [Pt(NH&!13] no appreciable amount of (As(C,&,)c) [P.t(NHa)Cl(OH)], the solution enters the various complexes. Reac- fore which would reduce the pH, could have precipitated. tions 1 and 2 offer a means by which such isotopic Chlorine-36, half-life 3.1 X 106 years, decaying by a single exchange with chloride can be accomplished. It is 0.72 MeV. ,%disintegration, was the tracer isotope. The of interest to find whether additional means exist to radioactive chlorine, carried in dilute HC1, was obtained the Isotopes Division, U.S.A.E.C., Oak Ridge, permit the tagged chloride to enter the complexes. from Tennessee. Under certain conditions the stereochemistry of Water for the preparation of all solutions was drawn from reaction 1 may be resolved. If, for example, k, the distilled water tap and was redistilled from alkaline k t , two chlorides may approach isotopic equilibrium permanganate. Platinum normally was recovered as the metal by a forin [Pt(NH3)C13]- much more rapidly than the mate reduction and reused in the experiments. A neutronthird and vice versa. However, reaction 2 and its activation analysis indicated a n iridium content of 20 reverse may permit the exchange of the chloride p.p.m. Other reagents which were used met A.C.S. specifications. which is not removed by the first hydrolysis (reacEquipment.-Reaction temperatures were maintained to tion 1). Also, a possibility exists that if the [Pt- 10.1' by a Sargent constant temperature bath. At O", (NH3)Cl(HzO)z]+ is exclusively the trans-complex, however, ice-baths were used. For other experiments the chloride trans to the NH3 will not exchange. below 20' the water cooling coil of the thermostat bath was The hydrolysis of all the other mononuclear replaced by a refrigeration coil. Radioactivity Measurements.-The rather conventional chloroamrnines of platinum(I1) has been studied. procedures for preparing, weighing, mounting and counting Grantham, Elleman and Martin3 found that [Pt- radioactive samples were described by Grantham, et al.3 CL]= undergoes acid hydrolysis but does not un- The practice of carrying several i3ter papers through the dergo additional isotopic exchange of chloride. washing, drying and weighing procedure was continued to corrections for humidity differences in the counting However, [PtC13(HzO) undergoes a chloride ex- provide sample tares. change which is independent of chloride and which For each sample the specific activity, S cts./min. mg. C1, may possibly proceed v i a acid hydrolysis. Ba- was always determined from the counting rate, corrected to nerjea, Basolo and Pearson* found that cis- [(Pt- standard geometry and self-scattering conditions, from the weight of the deposit and from its chloride content. Such (NH3)zC12] undergoes an acid hydrolysis. Although specific activities were proportional to the disintegration they detected no hydrolysis product for trans- [Pt- rates per mole of chloride. (NH3)zClz J nor [Pt(NH&Cl] +, the isotopic exExchange in Aged Solutions of K[Pt(NH3)Clo] and KC1.change with chloride occurred by a process which The proper quantities of K[Pt(NHa)Clsl, KC1 and NazS04 to give solutions with the desired concentrations and an was independent of chloride ion concentrations. ionic strength of 0.318molar were weighed into a volumetric Thus no exchange of chloride which is dependent flask, and water was added to the mark. The flasks usually on chloride concentration has yet been found for were wrapped with opaque tape to exclude light and any of the complexes. The present complex is of were thermostated a t the reaction temperature for a suflitime to approach closely equilibrium with respect to special interest for i t is the only one of the series cient the hydrolysis reactions 1 and 2. Aliquots were withdrawn (3) 'L. F. Grantham, T. S. Elleman a n d D. S. Martin, Jr., THIS a t intervals, and a fivefold excess of (As(C&)r) (C2H302) JOURNAL, 77, 2965 (1955). was added immediately to each aliquot to precipitate the ( I ) D. Ranerjea, F. Basolo and R. G. Pearson. i b i d , 79, 4055 [Pt(NHs)Cla]-. After 4 min. the aliquot was filtered. (1957) The precipitates were washed three times with H20 and +
where for 25'
>
,.
[
+
>>
1-
1-3
THOM.J,S s. ELLEMAN, JOHN b-. REISHUSAND
allowed to stand overnight. The (As(C6H&) [Pt(NH3)Cls] was deposited in an even, reproducible film on the filter paper, and its specific activity S was determined. To obtain S, for each reaction, a t least two aliquots were taken after a sufficient time so that no further change in the specific activity was detectable. These data provided a fraction of exchange for the [Pt(NH3)C13]- species defined as S/S,,. I n a few experiments with high chloride concentrations, the initial specific activity of the tracer-chloride was carefully determined, but this information was not generally required. In aged solutions, for which the chloride concentration was low (less than 0.02 Ad), [Pt(NH3)Cla]- and [Pt(KH&Cl~(H20)lwere present in comparable concentrations. The addition of AgZSOa t o the solutions quickly precipitated .lgC1 and Ag[Pt(NH3)Cls]. However, a drop in P H followed the addition of 4g2S04, which indicated the coprecipitation of significant amounts of insoluble Ag[Pt(NHs]Clz(OH)]. The precipitation of this compound was suppressed by the addition of HzSOa t o lower the pH to ca. 2. .it p H 2 the addition of Ag2S04, 20-30y0 in excess of the amount calculated to precipitate C1- aiid [Pt(NHa)C13] -, removed these anions effectively. The major portion of the [ P ~ ( S H , ) C ~ Z ( H S remained O)] in the filtrate after filtration through a glass sinter. The filtrate was treated with excess aqueous NH3 and heated for 4 lir. a t 80' to replace the The solution then was acidified chloride ligaiids by "3. with H2SOd and AgaSO, was added. The resulting AgCl precipitate was collected on a filter paper, dried, weighed and counted. iVith a careful control of the amount of -1g2S04 which was iriitially added, this silver-ioli scavenging procedure provided the specific activity of the [Pt(NH,)CIZ(H20)] species. Without such information analysis of the exchange would have been severely curtailed. Exchange in Fresh Solutions of K[Pt(NH3)C1,] and KC1.--For fresh solution exchange the proper quantities of KCI and Sa,SO, were dissolved in water of a predetermined volume and temperature. A weighed quantity of K [ P t (XHa)Cl3]then was dissolved in the solution and the chloride tracer was added immediately. At this time only negligible quantities of the hydrolyzed species were present. From this point the experiment was treated in the same fashion as the aged-solution experiments. Chloride Exchange with [Pt(NH3)Cla(H20j]. ---.Di!ute solutions of [Pt(NH3)C12(H20)II prepared by the ionexchange method, were transferred t o reaction flasks, and ?;azS04 was added to bring the ionic strength to 0.318 mole/. The flasks were stored for a period in the thermostat. Th,e following concentrations of a typical preparation a t 20 , after the hydrolysis equilibria 1 and 2 were reestablished, were calculated from the equilibrium constants: [ P t ( S H & M, [Pt(NH3)CI(H20)2+] = 3.4 Cl?(H20)] = 20.2 X M , and = X lo-, li,[Pt(NHJ)Cla-j = 0.4 X 2.5 X If. Enough solid KCI was added to increase the C1- concentration by a n increment of 29.0 X 211. The solution approached a n equilibrium state in which [Pt(NH~)CIz(HzO)]= 20.0 X lo-' M , [Pt(SH3)Cl(HpO)*+: = 0.4 X X,[Pt(NH,)Cla-] = 3.6 X lo-< M a n d (C1-) = 26.0 X -11. Initially, the [Pt(riH3)Cl(HsO)a] concentration decreased very rapidly, but the [Pt(NHa)Cld - formed slowly with a period of approximately 14 hr. a t PO . Approximately 1 hr. after the addition of the KC1, the chloride tracer was added. Solution aliquots were withdrawn after various time intervals. The specific activity of the chloride ligands then \vas determined by the d w r ion scavenger procedure described above. In some experiments the Cl36 tracer was included with the KCl which was added t u increase the chloride ion content of the equilibrated [Pt(NH3)Cl2(H2Oj! solutions. The specific activity of complexed chloride increased more rapidly in these experiments by virtue of the reverse of reaction 2 From the increase in the initial slope of S/S, the rate of t-eplacement of H2O i n [Pt(XHp)Cl(H20)2]-r by chloridc was estimated. .L' value 0.16 AI-' set.-' was indicated for k - 2 a t 20°, in satisfactory agreement with the earlier rough estimate of 0.2 for 25'. Substitution of NOz- and O H - in [Pt(NHJClz(HzO)] .- Experiments were performed to detect a possible nonequivalence of the chloride ligands in [Pt(SHs)Clz(HzO)]. [Pt(NH3)Cl~(HzO)]was allowed to exchange with tracer chloride for L: short period of time 111-2 hr.). The exchange I\-as then quc~icherihy the addition of .Ig?SO,. T h e aver:Lge
DOK
s. In\l.J,RTIN, J R .
Vol. 81
specific activity of the chloride ligands in a portion of the solution was obtained by the silver-ion scavenging procedure. Various amounts of either KN02 or KOH were added to other portions of solution. With KOH, AgzO formed quickly and was filtered off. The platinum complex was transformed to the anion [Pt(NH3)Clz(OH)]- and OHslowly replaced C1-. The reaction of hO2- with [Pt(NH3)C12(H20)] proceeded more rapidly. After a fraction of the chloride ligands had been replaced, the solution was acidified to pH 2 , and 4g2S04 was added to precipitate the free chloride. These precipitates were separated by filtration, and their specific activities were determined. The chloride ligands which had passed into the filtrate were recovered by the ammonia replacement method used in the silver-ion scavenging procedure.
Results
Exchange for [Pt(NH3)C1,]- in Aged Solutions.Preliminary experiments were intended to establish the number of chlorides in [Pt(NH3)Cl3]- which exchanged with chloride ion. The exchange solutions contained high chloride ion concentrations with a known specific activity. The specific activity of the [Pt(NH3)Cls]- fraction was determined after several days. If two chloride ligands had exchanged, the final specific activity, S, ( 2 ) will be .C7(2j = 2bS,G/j:i i h
+ 2n
( 3!
I
where = initial COIICII. of [Pt(;iH3jC1~]-, ~1.i h = initial concn. of C1-, X SO , = initial specific activity of C1-, cts./miii. lug. C1
(I
If all three chloride ligands had exchanged, the final specific activity will be .S,(3) = bS,O,(b
-t3
~ )
(4)
The results, shown in Table I, clearly indicate that all three ligands had undergone exchange. 'r.4BLE I EXCHASGEOF CHIXIRIDES IS [Pt(P\;H3)Cla]
-I1
Initial [Pt(XHa)Cia-] (a1 M
Initial chloride
0,0083
0.134
Vltjti
.(I75
);\
(a-)
a
Temp.
= 2U";
Ssd =
s, (3)
'Iimr,
.S,(2)
hr.
calcd.
calcd. cts./ min. mg. C1
65
9.3
S
obsd.
97.8 f.1 . 6 96.9 rt 1 . 6 94.2 rt 1 . 5 96.7 i 1 . 6 100
