BRIDGED MECHANISM FOR THE FLATINUM(IH) CATALYSIS OF CEILBWE EXCHANGE IN CHLOROAMMINE PLATIIWJM(I[V) COMPLEXES 1 BY FRED BASOLO, MELVIN L. MORRIS AND RALPH G. PEARSON Dept. of Chemistry, Northwestern University, Illinois Received 1 1 th January, 1960 A kinetic study was made of tho platinum(I1)-catalyzed chloride-ion exchange of various chloroamineplatinum(1V) complexes). The exchange was observed to follow the rate law R = k[Pt(IV)][Pt(II)][CI-] and to be essentially the same as the rate of platinum exchange. The rate of chloride exchange with trans-Pt(NH,),Cli + is about 2000 times faster than with the cis isomer and about 10,000 times faster than with Pt(NH3)5CP+. There is no chloride exchange in the system trans-Pt (tetrameen),Cli ++ Pt(tetrarneen)$++*Cl-. All of these observations are explained on the basis of a two- clectron-change redox reaction involving a bridged activated complex.The mechanisms of redox reactions have been recently reviewed by Taube.2 In general, these reactions are found to occur either by way of (i) an outer-sphere activated complex or (ii) a bridged activated complex.* The classification as a reaction proceeding by an outer-sphere activated complex is given to systems where there is no evidence for the formation of a bridged intermediate and where it appears that, because of the nature of the species involved, such an intermediate might not be formed, e.g., Fe(CN)%-+ Fe(CN)z-, Cobhen):+ + Co(yhen)?+. In some of the cases that have been studied there is direct evidence that a bridged activated complex is involved.For example, it was shown 3 that the reduction of Co(NH3)5X2+ by Cr(H20)2+ proceeds via the activated complex 3 (NH3)5CO- - X- - Cr(H20)<+ and that the exchange of chromium in the system Cr(H2O)sX2++ *Cr(H20);+ involves the bridged complex 4 (H20)sCr- -X- -*Cr(H20);+. Most of the studies that support a bridged activated complex for the redox reaction have been made on systems such as those mentioned above and which involve a one-electron change. However, some two-electron-change processes are also known to proceed by such a reaction path. For instance, oxygen-18 experi- ments indicate 5 that the oxidation of SO$- by Cloy occurs through the intermediate 02S--O--Cl02. Similarly, the exchange of chloride and of platinum in the system trans-Pt(en)ZCl:+ +Pt(en)$+ +Cl-, also an example of a two-electron change, is believed to involve the bridged intermediate Cl(en)#t- -C1- -Pt(en)2C13+.Making use of this mechanism as a guide, it was possible to prepare several com- pounds 6 of the type trans-[Pt(en)zX2]X2 starting with trans-[Pt(en)2C12]Cl2. In order to test further the generality of this bridged mechanism, the platinum(I1) catalysis of chloride exchange with various chloroammineplatinum(IV) cations has been investigated and is reported here. EXPERIMENTAL PREPARATION OF COMPOUNDS Except for [Pt(tetrameen)2]Cl~ . 2H20 and trans-[Pt(tetrameen)2Cl&l2 . Ha0 the compounds used in this investigation are known compounds and they were prepared by * Tn the earlier literature on this subject, category (i) was referred to as electron transfer and (ii) was designated as atom transfer.Since experiments and theory to date do not permit such a detailed assignment of mechanism, the classification into outer-sphere or bridged activated complexes as suggested by Taube is preferable. 80F . BASOLO, M. L. MORRIS AND R . G . PEARSON 81 methods described in the literature.8 These compounds were purified by recrystallization and characterized by means of analyses (table 1). The tetramethylethylenediamine complexes were prepared as follows. A reaction mixture containing 5 ml of 1.6 M tetrameen, 3-3 g of K2PtCb. and 25 ml of water was heated on a steam bath until the solution became a pale yellow and no more solid appeared to separate. At this point the mixture was cooled in an ice bath after which the crystalline Pt(tetrameen)C12 was collected on a suction filter and washed with a small amount of water.This complex was then suspended in 100 ml of water and 6 ml of 1-6 M tetrameen was added. The mixture was then refluxed until complete solution had taken place, concentrated until crystals began to separate, and cooled in an ice+salt bath; crystals were collected on a suction filter and finally washed with ethanol and ether. Further crops of the white crystals were obtained from the mother liquors. The product was air-dried at room temperature and found to weigh 4.0 g (95 "/o yield) and have an analysis corresponding to [Pt(tetrameen)2]C12,2H20 (table 1). Chlorine was slowly bubbled through a solution containing 0.5 g of IPt(tetrameen)2]- C12.2H20 in 30 ml of water for a period of 5 min.During this time the solution turned yellow, then a deep brownish yellow. A vigo- rous stream of air was then passed through the solution to remove the excess chlorine. The mixture which was slightly cloudy was passed through a filter and the clear filtrate was concen- trated to the point of crystallization on the steam bath. After cooling in an ice+salt bath the straw-yellow crystals that separated were col- lected on a filter and washed with acetone. Additional product was obtained by further concentration of the mother liquor. The pro- duct was air-dried at room temperature and found to weigh 0.5 g (91 % yield) and have an analysis corresponding to [Pt(tetrameen)2Cla] C12. H20 (table 1). CHLORIDE EXCHANGE All of the experiments were carried out in 10 ml volumetric flasks which were covered with aluminium foil to exclude any light. The re- action solutions were prepared either by adding known volumes of freshly prepared stock solutions of the platinum complexes or weighed amounts of the solid to the flask.This solu- tion was then allowed to reach equilibrium temperature in a constant (f0.1 "C) temperature bath and a known amount of radioactive chloride ion was added in the form of a stock solution of H36Cl. At various times 0-2 ml of the sample was withdrawn and added to a small centrifuge tube containing 0.1 ml of a solution of in 0.2 N KN03. The amount of AgN03 was always approximately 40 % in excess of the chloride ion. This mixture was then centrifuged82 BRIDGED MECHANlSM OF CHLORIDE EXCHANtiE for 15 min after which time 0.2 ml of the clear solution was carefully removed with a 200 A pipette and placed on a piece of filter paper contained on an aluminium planchet. One drop of concentrated NH40H was added and the sample was dried by means of an infra-red lamp.The radiocounting was done with a Nuclear Chicago model 181A scaler. Each sample was counted for a period of 1 h and the reproducibility of successive sample counts was within 2 %. The rate expression used to calculate the rate of exchange of n-chlorides where there was no net chemical change was Rt = ii[Pt(IV)][CI-]/n[Pt(IV)] -f- [Cl-] x 2-303 log [(CW - C,)/(C, - C,)], (1) where CW is the count at infinite time, CO is the count at zero time and Ct the count at any time t .The velocity constant k was calculated from the expression R = ic[Pt(IV)][Pt(II)][CI- J. (2) In soiiie cases the chloride exchange is accompanied by a net chemical change, e.g., Pt(NH,),C13+ + Pt(NH,)i' +H+ + C1-+Pt(NHJ)z' + trans-Pt(NH,),Cl;' + NH;. (3) In this case the measured radioactivity was converted into concentrations and the apparent rate constant kapp. was calculated using the equation (a-b)kapp.t = 2.303 log [b(a-x)/a(b-x)], (4) where (a-x) is the concentration of the reactant C1- and (b-x) is the concentration of the original Pt(IV) complex at any time t. The true velocity constant k was then obtained from the relationship, k = kapp./[Pt(II)I* ( 5 ) The experimental iiifinity counts always agreed with the calculated infinity Colin t to within 5%.PLATINUM EXCHANGE The platinum exchange was examined in the ethylenediamine system only, These exchange studies were done in three different ways and in three different laboratories. Polarimetric studies using laevo-propylenediamine complexes were made by Dr. R. G. Wilkins in this laboratory. The optical rotation of trans-Pt(1-pn),C$ + is approximately three times that of Pt(l-pn);+ ; the propylenediamine and ethylenediamine complexes are of comparable stabilities and all of the complexes are substitution inert. Therefore it was possible to determine the rate of platinum exchange polarimetrically by taking ad- vantage of the equilibrium c1- Pt(en)i++ trans-Pt(l-pn),CIi+ + trans-Pt(en),CI;+ + Pt(1-pn) :+. (6) In a typical experiment a solution was prepared which was 6.0 x 10-3 M in [Pt(en)2](C104)2 and in trans-[Pt(l-pn)2Clil(N03)2 and had an optical rotation of 1-25" at the sodium-D line.The optical rotation of this solution changed only very slowly but upon the addition of sufficient solid KCl to give a concentration of 3-4 x 10-3 M KC1 there is a rapid measur- able equilibration to an optical rotation of 0.75". The rate of equilibration of (6) was followed polarimettcially at room temperature (approx. 25°C) in a dark room with the reaction solution being exposed only to the light source of the sodium vapour lamp. Essentially the same experiment was done starting with trans-[Pt(en)2C12]C104 and [Pt(l-pn)2](ClO& and the results by these two methods were in good agreement. Making use of carbon-14 labelled ethylenediamine, Dr.R, G. Wilkins at Sheffield University has followed the rate of platinum exchange in the reaction c1- Pt(en*)z + + trans-Pt(en),CIj4 + + trans-Pt(en*),Cl; I- -I- Pt(en)2, + . ('7)F . BASOLO, M. L. MORRIS AND R. G . PEARSON 83 A reaction mixture which was 1.44 x 10-3 M [Pt(en*)2]Cl2, 2.72 x 10-3 M trans- [Pt(en)2C12](C104)2 and 9.43 x 10-3 M HC1 was placed in a volumetric flask covered with aluminium foil and thermostatted (rtO.l°C) at 25°C.. Aliquots were removed at various times and added to a centrifuge tube containing an excess of aqueous K2PtC14 to pre- cipitate [Pt(en)2] [PtC14]. This precipitate, after quick centrifuging, was washed, dried and counted as the solid. Using platinum-195 as a tracer, the exchange of platinum in the system trans- Ft(en),Cli +-I-Pt(en)z ++ C1- was investigated by Prof.D. S. Martin 9 at Iowa State College. Details of these studies are to be published later by Martin. CHLORIDE CONSUMPTION The rate of decrease of chloride ion concentration in the reaction mixture of Pt(NH3)&13++Pt(NH3)$++ 61- according to eqn. (3) was followed by means of titrations at various time intervals using the Volhard method. The calculated and experimental value of the chloride ion con- centration at zero time agreed to within 3 %. These solutions were prepared and handled in the same way as described above for the chloride exchange studies. The product of this reaction (3) was isolated and shown by analysis and infra-red spectrum to be trans- [Pt(NH&CIz]C12. A similar observation had been reported earlier.10 The chloride ion concentration was also observed to decrease in the system ci~-Pt(NH3)&1$++Pt(NH3):++ C1-.However, since in this case it is likely that several products are formed, no quantitative study of the rate of disappearance of chloride ion was made. For the same reason, the fact that chemical reaction was occurring was ignored in calculating the rate of exchange by radiochemical methods. The reaction was slow and was followed to no more than 25 % of completion with a fair excess of chloride ion present. Since linear first-order plots were obtained, changes in composition were not great enough to affect the rates. Initially there was a very fast exchange corresponding to essentially a 6-7 % zero time exchange when plotted on a time scale convenient for the remainder of the reaction.It was assumed that this was due to some 6-7 % of the trans isomer contaminating the cis. HYDROLYSIS Conductivity measurements and chloride ion determination on solutions of the chloro- ammineplatinuni(1V) complexes studied showed that there was no appreciable hydrolysis for any of these during the time and experimental conditions for chloride exchange. How- ever, this was not true for cis- and trans-Pt(NH&C14. Solutions of both of these com- pounds gave evidence of hydrolysis even when kept in the dark. Because of the com- plications due to hydrolysis no attempt was made to investigate the exchange of chloride in these systems. RESULTS The uncatalyzed exchange of radiochloride ion with the chloroammineplatinum- (IV) complexes studied at these experimental conditions was extremely slow.In all cases the exchange was catalyzed by platinum(I1) and kinetic studies show a first-order dependence on each of the three reactants so that the rate law is given by eqn. (2). Linear plots of the kinetic data, log[(C, - Co)/(C, - C,)] against t, were obtained. Results of experiments on the system trans-Pt(en)2ClZ++ Pt(eii)$++*Cl- are given in table 2. The activation energy for exchange is ap- proximately 11.5 kcal. Addition of hydroquinone or of barium diphenylamine sulphonate does not detectably alter the rate of exchange, whilst the presence of aniline only slightly decreases the exchange ratc. Qualitative observations on the rates of platinum exchange using complexes of laevo-propylenediamine are summarized in table 3.Quantitative data using carbon-14 and platinum-195 also are given. The first of these gives excellent agreement and the second moderate agreement with the rates of chloride exchange. In the case of carbon-14, it was assumed that Pt(en)2 exchanged as a unit.84 BRIDGED MECHANISM OF CHLORIDE EXCHANGE TABLE 2.-&TE OF CHLORIDE EXCHANGE IN THE SYSTEM trans-Pt(en),Cl; ++€%(en); ++ *C1- AT 25°C IN THE DARK k, 1.2 mole-2 min-1 run [trans-~t(en)2~122+1 [~t(en),+l [HCU 1 2 30 4b 5 6 7c 8 9d 1 Od 1 Id 12d 0.001 M 0.001 0.001 00005 0.001 0.002 0.002 0.001 0.001 0.001 0.001 0.002 0-00005 M 0~00010 0~00020 0*00020 0-00020 0.00023 0*00023 040025 0~00012 0.00025 0*00012 - 0.01 M 0.01 0.01 0.01 0.02 0.015 0.01 5 0.01 0.01 5 0.01 5 04085 0.017 9.2 x 102 9.2 x 102 9.2 X 102 8.8 X 102 8-8 X 102 9-2X 102 7.4x 102 no exchange in 2 days 2 3 x 102 2.4 x 102 1.6 x 102 1 .6 ~ 102 (a) Making the reaction mixture 0002 M in hydroquinone after one half-life did not nieasurably alter the exchange rate. This same reaction at 37.5"C has a value of k = 2.0 x 103 1.2 mole-2 min-1 allowing an estimate of EA = 11.5 kcal/mole. (6) Making soIution 0.001 M in barium diphenylamine sulphonate after one half-life did not measurably alter the exchange rate. (c) Reaction mixture was 0004 M in aniline. ( d ) For runs 9-12 the Pt(I1) catalyst is Pt(NH,)i+. 10 20 30 4 0 5 0 time, h FIG. l.-The rate of exchange of chloride ion in the system trans-Pt(NH3),C14, +Pt(NH,)$+ +*c1- at 25°C in the dark. Calculated on the basis of two exchangeable chlorides, 0 ; The experimental observations on all other systems are tabulated in table 4.me rate data reported for trans-Pt(NH3)3CI$ are based on the assumption of two replaceable chlorides only. When on the basis of three exchangeable chlorides, A. The justification for this is shown in fig. 1.TABLE 3.-POLARIMETRIC STUDIES OF PLATINUM EXCHANGE IN trans-Pt(AA),Xz ++Pt(AA)g ++x- SYSTEMS b run 13 14 16 17 18d 19e 20 21 22 1 9 concentration Pt(1V) Pt(IV) WII) 0.006 M 0.006 0.006 0.012 0.01 3 0.010 0.004 0.006 0.006 0.005 Pt(I1) 0006 M 0.006 0.006 0.01 3 0.01 3 0-012 0-004 0,006 0.004 0.005 optical rotation initial 040" 038 1.26 0.37 1.26 0.40 0.80 0.45 0.78 0.37 final 0.40" 0.75 1-14 0.76 0.76 0.82 0.80 0.86 0.47 0.37 remarks F no change in 15 min.Add 0.015 h.I KCI td and get equilibrium value of 0.75 in 20 min 0 r after 1 h, 0.49" ; complete overnight after 1 h, add 0.0034 M KCI and t4-20 min ,O equilibrium reached in 20 min, &,-0-85 % same as 16 above equilibrium reached in 30 min r % no change in 24 h 0 equilibrium reached in 10 min, KEq.-'1 TI same as 20 above M no change in 24 h ; no effect of added KCl 9 Z U (a) Reaction mixtures were at room temperature (approx. 25°C) and in only the light path of the sodium vapour lamp. w (b) Using 1"-en, platinum exchange in the dark at 25°C for the reaction mixture 0.00144 M [TPt(en*)2]C12, 0.00272 M trans-[Pt(en)2C12l(ClO& p and 0.0094 M HC1 has a value of k = 9.7 x 102 1.2 mole-2 min-1. Preliminary experiments with 195Pt give a value of k = 6*6+ 1 x 102 1.2 mole-2 cd min-1.In both of these studies the rate law (2) was confirmed. M P (c) Chloride exchange at these conditions has t+ = 17 min. (d) The P t O complex is supposedly the cis isomer, a sample of which was obtained from G. Johnson. (e) In more concentrated solutions of trans-Pt(en),Cl$ ++ Pt(NM,)$ f+ Cl- precipitation with PtCl$- gave no indication of the formation of Pt(en);+. It would seem that K E ~ . for this reaction is small.TABLE 4.-&TE OF PLATINUMQI) CATALYZED CHLORIDE EXCHANGE OF CJ3LOROAMMINEPLATINUM@v) COMPLEXES AT 25°C IN THE DARK run 23 24 25 26 27 28 29a 30b 3 1 C 32 33 34a 35 3 6d 37e 38a 39f 40a 41b 42 43 44g 45 platinum complexes W V ) WlI) t rans-Pt (NH 3)4C1: + Pt(NH 3 >t + 3) Y Y ¶ Y YY Y Y - $9 [O-OOS h4 Ce(IV)] Y Y ]Pt(en)$ + cis-Pt(NH,),CI$ * Pt@?H,)i+ Y Y Pt(NH3) Ibl3 * Y Y - Pt(NH,): + Y, Y f Y Y Y Y Y Y - trans-Pt(NH,), Cll Pt(NH3 ): + - Y Y Y Y [0.004 M C e O ] Y Y Pt(NH,),CI+ trans-Pt(tetrameen),Cl: * Pt(tetrameen)z + Y Y Y, WIV) 0.0010 M 0*0010 00010 o*m91 0.00091 0.00036 0.0010 Q*OOlO 0.0008 1 0-0184 0.0258 0.0279 0.01 15 0.0101 00125 0.0122 00010 0~0010 0.0010 0.0010 0.00069 0-00178 0.001 87 Pt(1I) 0.00025 M 000025 000012 0.00025 0.00025 0.00025 I - 0*00012 0.01 64 0.0336 0.0152 0.01 52 0.0037 0*00025 - - - - 0*00014 0*00009 0.0022 0*0001 concentration k, 12.mole-2 min-1 HCl 0.0085 M 3-8 X 102 3.8 X 102 0.01 5 4.8 X 102 0.015 0.01 5 3.3 x 102 0.0088 4-3 x 102 0.014 3.7 x 102 0.01 5 t4-44 h 0.015 no exchange in 2 days 0.010 initial rate very fast 0.050 1.7 >( 10-1 0064 1.4 X 10-1 0.896 0.047 3.9 x 10-2 0.043 3.2 X 10-1 0.050 4.0 0.049 0.014 tg-66 h 0.014 0-014 0.0084 04184 0.0122 no exchange in 14 days no exchange in 14 days 0.014 1.3 x 103 1.8 x 103 no exchange in 3 days no exchange in 14 days no exchange in 2 days no exchange in 1 day (a) In these runs no Pt(1I) was added.However, this does not exclude the possibility of catalytic amounts of Pt(I1) being present because in (b) The solutions of Pt(1V) complex containing Ce(IV) were allowed to stand overnight in the dark at room temperature prior to the addition (c) This exchange is complicated by the accompanying chemical reaction yielding Pt(NH,);+ and trans-Pt(en),CI$ + (see run 19, table 3). ( d ) Reaction temperature was 50°C. Under the same conditions the rate of consumption of chloride ion due to the formation of trans- (e) Reaction temperature was 81°C.The activation energy EA is 16.7 kcal/mole. (f) The value of k is calculated on the basis of two exchangeable chlorides (see fig. 2). (g Reaction temperature 50°C. most cases the Pt(IV) complex is prepared by the chlorination of a Pt(1I) compound. of HCP. Pt(NH3)4Cl$+ (see eqn. (3)) has a value of k = 2 8 x 10-1.F. BASOLO, M. L . MORRIS AND R. G . PEARSON 87 the data are plotted for three exchangeable chlorides, a linear plot is not obtained. However, theuse of a calculated C, based on only two exchangeable chlorides gives reasonable linearity. The experimental C, slowly approaches that for three replaceable groups. The activation energy for the chloride exchange in Pt(NH3)&13++ Pt(NH3):'f *Cl- is found to be 16.7 kcd.This reaction mixture yields the products shown in eqn. (3) and the rate of loss of chloride ion concentration is equal to the rate of chloride exchange if it is assumed that the product trans-Pt(NH3)&1$+ contains two completely exchanged chlorides. DISCUSSION The direct exchange of chloride with the chloroammineplatinum(1V) complexes investigated is either extremely slow or does not occur under the conditions of these experiments (see runs 8, 29, 34, 38 and 40). Runs 30 and 41 show that even the very slow exchange is almost completely inhibited by prior treatment of the platinum(1V) complex with cerium(1V). I t is believed that this is due to the oxidation of catalytic amounts of platinum(I1) in the platinum(1V) complexes which were prepared by the chlorination of the platinum(I1) compound.The kinetic data in tables 2 and 4 show that the rate law for exchange is given by eqn. (2). The mechanism proposed 1 to explain these results is illustrated by the following equations : fast €%(en)$ + + C1- + Pt(en),CI+ (8) en en slow en en slow en en en en ci--Pt-c12 + + pt-ci + + cl--Pt - - - c i - - - lpt-ci3 + + en + en en en Cl-Pt 3. CL-Pt-C12 + (9) There is now considerable evidence 11 in support of the addition of other groups to square planar complexes as suggested in (8). Similarly the bridged inter- mediate in (9) with the structure shown in fig. 2is analogous toIcomplexes originally CI 6' I FIG. 2.-Bridged intermediate proposed to explain the two electron change redox reaction in certain platinum(1V)-platinum(I1) systems.believed to contain platinum(II1) but later shown to be platinum(I1)-platinum(IV) bridged compounds.12 There is also some evidence for the existence of this particular bridged complex, [Pt2(en)&13]C13 in the solid state. It was observed that upon evaporation of a colourless aqueous solution containing equivalent amounts of trans-[Pt(en)2C12]C12 and [Pt(en)2]C12 an orange solid is obtained. The individual complexes are white. Thus the orange colour of the resulting solid88 BRIDGED MECHANISM OF CHLORIDE EXCHANGE is indicative of a platinum(I1)-platinum(1V) bridged complex, since such systems are known usually to be highly coloured. No spectral evidence for this species in solution could be found. This bridged mechanism for chloride exchange requires that platinum exchanges at the same rate by means of a two-electron-change process.The data in table 3 show that there is an exchange of platinum in these systems. Footnotes (b) and (c) of this table give quantitative data on the rates of platinum exchange which are in agreement with the rate of chloride exchange. It should also be noted that chloride ion is required for rapid platinum exchange. Perchlorate and nitrate ions are not nearly so effective as is shown by runs 13, 14 and 15. Cl Rapld c hlori dc and Pt platinum cxc hanqe similar c 1- equations Pt + NH 3 NH4 FIG. 3.-Mechanism for the chloride exchange and reaction of Pt(NH3)Qt in aqueous hydrochloric acid solution containing Pt(NH,)$+.The only chloroammineplatinum(IV) complex investigated that does not exchange chloride ion in the presence of platinum(I1) is trans-Pt(tetrameen),Cl:+. This is a significant observation as it is exactly what the bridged mechanism would predict. Because of the bulkiness of the C-methyl groups on the chelate rings, the Pt(tetrameen)$+ cannot get close enough to form a bridged complex through the chloro group. Therefore the bridged mechanism is not available to this system and no chloride exchange is expected. Similarly the less bulky Pt(en)f+ does not catalyze chloride exchange in Pt(tetrarneen),Clf-'- presumably because there is still steric resistance to the formation of the bridged complex (see runs 43, 44 and 45). A comparison of runs 23-28 and 32 and 33 show that the rate of chloride ex- change of trans-Pt(NH&ClZf is approximately 2000 times faster than that for the cis isomer.This is believed to be a direct consequence of the stronger Pt-N bond compared to the Pt-CI bond, for the trans isomer reduction of the plat- inum(1V) complex to platinum (11) by the bridged mechanism requires the rupture of the Pt-C1 bond opposite the chloro bridge as represented in fig. 2. However, in the cis isomer ammonia is opposite the chloro bridge and reduction necessitates cleavage of a Pt-N bond as shown in fig. 3. This difference in the rate of chloride exchange for the geometric isomers of platinuw(1V) complexes is in agreement with the observations that in such systems the cis isomer is more difficult to reduce than is the trans form.10 This observation is also consistent with Orgel's 13 suggestion that the bridge axis be designated as the z axis and then the electrons enter the dZ2 orbital.Since ammonia has a stronger crystal field than does chloride ion, the dZ2 orbital will be at a higher energy value for the cis isomer where ammonia is opposite the bridging group and consequently the transfer of electrons would require more energy than for the trans isomer. However, the fact that trans-Pt(en),(OH)$+ does not exchange with Pt(l-pn)$f even in the presence of chloride ion (table 3, run 22) indicates that the strength of the bond to be broken, Pt-OH, is more important than crystal field effects. Hydroxide ion and chloride ion are close to each other in the crystal field series.F.BASOLO, M. L . MORRIS AND R. G . PEARSON 89 The rate of chloride exchange with Pt(NH3),C13f (run 35) is approximately 10,00Q times slower than it is for tran~-Pt(NH~)~Cl;+. The activation energy for exchange in Pt(NH3)5C13+ is 16-7 kcal compared to 11.5 kcal for trans-Pt(en),Cl$f. Presumably this slower rate and higher energy of activation is again due to the stronger Pt-N bond opposite the chloro bridge as explained above and as shown in fig. 3. Since the Pt-Cl bond is weaker than Pt-N, it follows that step (1) is rapid compared to step (2). However, (1) results in no change, neither chloride exchange nor chemical reaction. For exchange of chloride step (2) is required which also necessitates a net chemical reaction. The tran~-Pt(NH~)~Cl$+ product then undergoes relatively rapid exchange by step (3) in the usual manner without further chemical change.Two observations were made on this system that afford excellent support to the reaction scheme shown in fig. 3. One is that the rate of decrease of chloride ion concentration is approximately the same as the rate of chloride-ion exchange (see footnote ( d ) in table 4) calculated on the basis of two chlorides exchanged per chloride ion removed from the solution. The other is that trans-[Pt(NH3)&12]C12 was isolated from the reaction mixture. This con- firms and explains the interesting observation of Rubinstein 10 that catalytic mounts of Pt(NH3);-+ will convert Pt(NH3)5C13+ in high yield to trans- Pt(NH3)4C1$f. Only catalytic amounts are required because Pt(NW3)2+ is re- generated as in (3).el I ' / I \ ' CI I F' @ I 1 68 "4, QAO (Q 1 FIG. 4.-Bridged intermediates proposed for the Pt(NH3)3Clf catalyzed chloride exchange of trans-Pt(NH&Cl;. Kinetic data collected on the exchange of chloride with trans-Pt(NH3)3C1$ show that two chlorides exchange much more rapidly than does the third (fig. 1). This can be explained by comparing structures (A) and (B) of fig. 4. The rate of exchange of the two chloro groups trans to each other via (A) requires the cleavage of a Pt-C1 bond and is fast, whereas the exchange of the chloro group trans to ammonia via (B) involves a Pt-N bond rupture and is slow. On the basis of the results mentioned above a difference of at least 103 in the two rates is expected. It is of some interest to note the effect of the charge of the platinum(1V) complex on the rate of chloride exchange.A comparison of the exchange rates show that trans-Pt(NH,),CI$ >tran~-Pt(N€€~)~Cl;+ (runs 23 and 39) and c~s-P~(NH~)~CI;+ >Pt(NH3)5C13+ (runs 32 and 35) by factors of approximately three and four re- spectively. This might be the result of a smaller tendency to form the bridged complex because of the greater repulsive interaction of the more highly charged cations. The charge on the platinum(I1) complex appears to have only a slight effect on the rate of chloride exchange. Runs 39 and 42 show that Pt(NH3)3ClC90 BRIDGED MECHANISM OF CHLORIDE EXCHANGE is only a slightly better catalyst than is Pt(NH,)z+. This may be due to corn- pensating opposing effects ; the smaller the cationic charge, the greater the tendency to form the bridged complex but the smaller the tendency to form the “five- co-ordinated ’’ species (see eqn.(8)). It should also be pointed out that the rate of exchange of tran~-Pt(en)~Cl;+ is about three times that of tran~-Pt(NH~)~Cl$+. One possible explanation for this difference is that the ammonia complex is more highly solvated and thus offers greater resistance to the formation of the bridged complex. The polarimetric studies reported in table 3 indicate that the efficiency of the bridging groups in facilitating electron transfer between platinum(1V) and plat- inum(I1) is Br->CI->OH- (compare runs 16, 20 and 22). This same order was observed for the system chromium(III)+ chromium(II).4 For platinrrm(IV)+ platinum(I1) the non-bridging group is also of considerable importance. Runs 14, 15 and 16 serve to show that the chloride ion is much more efficient than is either nitrate or perchlorate ion.This work was recently extended to show that in the system tran~-Pt(en)~@l$+Pt(en)~ +Xu, when X- is Br-, C1-, C N , CNS-, CNO- and NO; there is rapid exchange of platinum whilst when X- is OH-, SO;-, ClS:, NO,, C2M30; and F- the exchange is very slow.6 Catalysis of chloride exchange in PtC1;- by PtC1;- has been observed by Rich and Taube 14 and explained on the basis of a chain mechanism involving plat- inum(II1). Such a mechanism is probably not operating in the cases reported here because of the rate law and because the reaction is not affected by the in- hibitors which Rich and Taube found effective.Similarly, the suggestion 15 that the mechanism may involve the equilibrium trans-Pt(en),Clg + +Pt(en)i + + Clz accompanied by rapid exchange between C12 and *Cl-, seems unlikely because run 7 shows that aniline, an efficient chlorine trap, has little effect on the rate of exchange. Also Pt(tetrameen)z+ and Pt(tetrameen)2Clz+, which do not exchange, are relatively easy to halogenate and dehalogenate respectively. One observation which does not fit into the general scheme outlined above for the bridged mechanism of exchange is that ci~-Pt(en)~Clzf seems to exchange at a rate similar to that of the trans isomer (run 18, table 3). This is quite unexpected in terms of the bridge mechanism which predicts a very slow rate of exchange as found for cis-Pt(NH,),Clz+ and Pt(NH3)5C13f.A detailed study 16 is being made of the Pt(en)g+ catalyzed chloride exchange of cis-Pt(en),Cl;+. The authors wish particularly to thank Dr. R. G. Wilkins for doing the polari- metric and 04-ethylenediamine experiments and for his very helpful suggestions during the early stages of this investigation. We also thank Dr. M. J. G. Williams who did the tetrameen experiments and Prof. D. S. Martin, Jr., for making available to us his Pt195 results prior to publication. This investigation was supported by a grant from the U.S. Atomic Energy Commission under contract AT(11-1)-89, project no. 2 and in part by a grant from the U.S. Air Force Office of Scientific Research under Contract No. AF 49(638)-3 15. f for a previous communication see Basolo, Wilks, Pearson and Wilkins, J. Inovg. 2 Taube, Advances on Inorganic Chemistry and Radiochemistry, Emeleus and Sharpe 3 Taube, J. Amer. Chem. Soc., 1955,77, 4481. 4 Ball and King, J. Amer. Chem. SOC., 1958, 80, 1091. 5 Halperin and Taube, J. Amer. Chem. SOC., 1952, 74, 375. For a general discussion 6 Johnson and Basolo, J. Inorg. Nucl. Chem., in press. Nucl. Chem., 1958, 6, 161. (editors) (Academic Press Inc., New York, 1959), vol. I, pp. 1-53. see Edwards, Chem. Rev., 1952, 50,455.F. BASOLO, M . L. MORRIS AND R . G . PEARSON 91 7 Symbols used are en = NH2CH2CH2WH2, tetrameen = NH2G(CH3)2-C(CH3)2NH2 8 Gmdin’s Handbuclz der Anovganischen Chemie B, 1930, 58. 9 D. S. Martin, private communication. 10 Rubinstein, U.R.S.S. Compt. rend., 1940, 28, 55/58 ; Izvest. Plat., 1947, 20, 53/83 ; 11 Tschugaeff, Compt. rend., 1915, 161, 563. Harris and Stephenson, Chem. and 12 for references and discussion of apparent platinum(ZI1) compounds, see Watt and 13 Qrgel, Rept. Xe Conseil Inst. Int. Chim. Solvay, 1956, p. 289. 14 Rich and Taube, J. Amer. Chem. Soc., 1954,76,2608. 15 J. Halpern, private communication. 26 J. C . Bailar and G. Johnson, private communication. and pn = NH2CH(CH3)CH2NH2. 1947, 56. Ind., 1957, 14, 426. Harris, Livingstone and Reece, J. Chem. Soc., 1959, 1505. McCarley, J. Amer. Chem. Soc., 1957, 79, 4585.