Oct. 5, 1961
NITRITESUBSTITUTION IN CHLOROAMMIKEPLATINUM(IV) COMPLEXES
Pauli-Peierls equationlg
where m* is the effective mass, I.CO is the Bohr magneton, n is the electron density, m is the electron mass and h is the Planck constant. Without the subtractive term in the parentheses, the magnetic susceptibility is the Pauli susceptibility arising from the unpaired electrons a t the top of the Fermi distribution. The subtractive term with m = m” is the Landau diamagnetism which amounts to onethird of the Pauli moment. Introduction of m* after Peierls corrects for departure from a perfectly free electron gas. Except for the value for Na 78W03,the effective masses show a rise to a maximum a t M.30w03and then a fall-off. (It might be noted here that we have restricted ourselves to cubic bronzes only, so problems of anisotropy have been avoided.) The existence of this maximum in m* suggests that there is a perturbation of the band structure which diminishes as x in M,W03 exceeds 0.3. The decay in m* a t large .ZI is consistent with the increasing mobility of the electrons in the alkali-richer bronzes. The source of the initial rise in m* is not clear. It may only be an apparent effect coming from the assumption that all the electrons are free. As discussed above, there might be localization of the electrons in the dilute bronzes which would reduce the number of free carriers. However, in Ago.01iV0320where trapping would be more likely because of the higher ionization potential of silver, the conductivity data seem to imply complete freeing of the carriers a t room temperature and above. (19) See, for example, A. H Wilson, “ T h e Theory of Metals,” 2nd Ed., Cambridge University Press, Cambridge, 1954, p. 155. (20) M. J. Sienko and B. R. Mazumder, J . Am. Chem. Soc., 82, 3508 (1960).
[CONTRIBUTION FROM THE CHEMICAL
3943
Alternatively, the initial rise in m* might be attributable to the fact that a t low values of x in M,W03, the concentration of M + ions introduced may not seriously perturb a 5d, band, if the latter indeed is the principal conducting band. As more M+ is introduced into the lattice, the concentration of scattering centers increases, each M -tappreciably lowering the energy of the eight tungsten neighbors that comprise the unit cell. States are thus effectively removed from the band, since banding can occur only with approximately equal energy states. On the other hand, when sufficiently high concentrations of M + have been added, a major portion of the tungsten atoms may be so perturbed and hence their states could mix to give band formation. The higher the concentration of M+, the more valid is the approximation of a regularly periodic potential. In summary, the model that begins to develop for the tungsten bronzes is a 5d, conduction band with local traps arising from tungsten atoms that have M+ions in their vicinity. There should be a finite excitation energy from these traps, but a fairly high mobility and low effective mass for any electrons that have been ejected from the traps. As the concentration of b9+ increases, the number of trapping centers increases, leading to reduced mobility and higher effective mass. Eventually the traps begin to overlap, the activation energy vanishes and metallic conduction sets in. From there on, increasing the concentration of M + reduces the aperiodicity in the band and leads to increasing mobility and decreasing eff eetive mass. A major test of these ideas would come from experiments a t lower temperatures. Acknowledgments.-We acknowledge with thanks the stimulating discussions we have had with Professors Raymond Bowers and James Krumhansl of the Cornel1 Physics Department.
LABORATORY O F NORTHWESTERN UNIVERSITY, EVANSTON, ILLINOIS]
Nitrite Substitution in Chloroammineplatinum(1V)Complexes BY HERBERTR. ELL IS ON,^ FRED BASOLO AND RALPHG. PEARSON RECEIVED OCTOBER 31, 1960 A kinetic study was made of the nitrite substitution reaction in several chloroammineplatinum(1V) complexes. I n the absence of platinum(I1) the reaction was found to involve an induction period of from 1 to 2 hr. and variable dependence upon platinum(1V) concentration. With the addition of platinurn(I1) the induction period disappeared and the reaction was found to obey the rate law R = k[Pt(IV)][ P t ( I I ) ][NOz-] and to be roughly the same as the rate of chloride exchange. All of these observations are explained on the basis of a slow reduction of platinum(1V) to platinum(I1) by nitrite ion and then a second two electron oxidation-reduction reaction involving a bridged intermediate. trans- [Pt(tetrameen)zCl~] *+.* reacts with nitrite ion only to be reduced t o [Pt(tetrameen)2l2+and no substitution occurs.
Much of the information concerning reaction mechanisms of hexacoordinated complexes has been gained from kinetic studies with the many complexes of cobalt(II1), 3 Preliminary investigations by Wilks4 of the reaction between trans(1) Department of Chemistry, Wheaton College, Norton, Massachusetts. (2) Symbols of ligands are en = KHzCHzCHzNHz and tetrameen = NHzC(CHa)zC(CHa)zNHz. (3) F. Basolo and R. G. Pearson, “Mechanisms of Inorganic Reactions,” John Wiley and Sons, Inc., New York, N Y . , 1958, Ch. 3. (4) P. H. Wilks, Masters Thesis, Northwestern University (1957).
[Pt(en)zClz]2 + and nitrite ion indicated that the reaction proceeded by the direct replacement of one chloride ion to yield [Pt(en)~N02CI]~+. Recently, Muskets observed that the reaction was much more complex than it previously had been believed to be.6 For example, Musket found t h a t the reaction has an induction period lasting from 1 t o 3 hr. and that the reaction is catalyzed by the ( 5 ) S.F. Musket, Masters Thesis, Northwestern University (1960). (6) F. Basolo, A. F. Messing, P. H . Wilks, R. G. Pearson and R . G . Wilkins, J. Ilzorg. Nuclear Chem., 8 , 203 (1958).
3944
HERBERT R. ELLISON, FREDBASOLOAND RALPHG. PEARSON
Vol. 83
TABLE I
Iv) COMPLEXESa AT 50" AND WITH N O ADDED Pt(I1) trans-[Pt(en)L!l,] a + [Pt(IV)], M [NaXOi], M [XaC104], M [HClOaJ, M k'(M-lmin.-L)O Ind. per., min. kl(M-:min.-9 0.0012 0.232 ... ... 0.832 x 10-1 165 1.42 x 10-4 ,0023 .227 ... ... 0.840 x lo-* 85 1.38 x 10-4 .200 0.026 ... 1.64 x 10-2 90 1.68 x 10-4 .003 1 .OO49b ,220 ... ... 1.83 x 10-2 130 1.33 x 10-4 .0050 .I50 ,055 0.015 1.57 X 85 1.35 X lo-.! .0060" ,200 ... .020 1.72 x 10-2 70 1.40 x 10-4 .O05Od .200 ... .020 1 . 7 5 x IO-* 95 1.37 x 10-4 .OO5lb .200 ... .020 1.58 x lo-* 90 1.31 x 10-4 .zoo .020 ... 1.72 x 90 1.50 x 10-4 ,0051 .0055 ,200 ... .020 1.61 x lo-* 100 1.14 x 10-4 .005 ... 2.56 X lo-* 70 1.43 x 10-4 ,0096 .200 .0098 ,200 .005 ... 2.59 x 50 1.57 x 10-4 ,0099' .lX0 ... ,085 2.48 x lo-* 85 1.34 x 10-4 .0140 .190 ... ... 2.82 X lo-* 60 1.21 x 10-4 .0190 .17,5 ... ... 3.79 x 10-2 65 1.39 X IO-' .oow ,200 ,020 ... 5.80 x 40 ......... .0050' ,200 .020 ... 1.94 X 85 ......... cs-Pt( NH2)4C12] . 0025h ,200 ,027 ... 1.58 x 10-4 1600 1.5 X ~ ~ C Z N S[ Pt( NH#)rCL]s+ ... 1.00 x lo-' 10 6 . 4 x 10-4 .00 10h ,200 .032 .0012" ,200 ... ,026 1.01 x 10-2 60 5.9 x 10-4 .0024' ... 1.73 x 40 6.2 x 10-4 ,200 ,027 , OO-i5*.~ ,026 1.62 x 30 5 . 4 x 10-4 ,200 ... .00,jt~ ,200 .020 ... 1.84 X 10-2 30 5 . 9 x 10-4 a Used as the perchlorate salt unless otherwise noted, all runs made at g = 0.235. Used as the nitrate salt. Dissolved complex treated with KMnOd before reaction. (NH4)*Ce(NO& added to reaction mixture, [Ce(IV)]o = 2.5 X M. CuS04 added t o reaction mixture, [Cu(II)] = 10-6 M. f Fes(S04)a added t o reaction mixture, [Fe(III)] Pseudo-first order constants divided by nitrite ion concentration. lo-* M .
RATECONSTANTS FOR
NITRITE SUBSTITUTION I N CHLOROAMWNEPLATINUM(
'+
additiou of [Pt(en)z]2+. The entire system has now been studied in some detail and the results of this investigation are reported here. Experimental
Results Since the nitrite ion was present in large excess it was possible to treat the experimental data by plotting log ( V , - V ) vs. time, where V is thc Reagents and Equipment.-All of tlie platinum corn- volume of silver nitrate used a t time t and 1'pounds used in this investigation are known compounds and is the amount a t infinite time. After an induction they were prepared by methods described in the literature.7~8 period, these pseudo first order plots usually were These compounds were purified by recrystallization and characterized by means of analyses. All of the reagents used fairly linear without too much scattering; an exin this study were of the best grade available. A constant aniple is shown in Fig. 1. The second order rate temperature bath capable of maintaining the temperature constants, k', that were calculated from the slopes within 1 0 . 1 " :!t 60" for a long period of time was used. are shown in Tables I and 11. Spectrophotometric measurements were made on a Cary hlodel 11 recording spectrophotometer. X microamDiscussion meter, silver electrodes and a microburet were used t o Inspection of the data from the truns-[Pt(en)zriiake arnperometric titrations of chloride ion. Procedure.-Reaction mixtures were usually about 0.005 Cl2l2+ runs in Table I reveals several interesting A I in the platinum(1V) complex and about 0.1 t o 0.2 M features. The most important is that the obin nitrite ion. All runs were carried out in volumetric flasks served second order rate constants are a function of coated with black Apiezon wax to prevent photocatalysis. All reactants were thermally equilibrated before mixing in the initial concentration of the platinum(1V) comorder t o assure that the induction period was not caused by plex when there is no platinum(I1) present. To temperature differences. The extent of reaction was fol- check whether this was a consequence of small lowed by amperometric titrations of the chloride ion that amounts of platinum(I1) being present as impuriwas released. Titrations with this method are somewhat faster than with the Volhard method and are reproducible ties (the platinum(1V) salts were prepared by to about 1 0 . 0 2 m'i. The procedure followed in these titra- chlorination of the analogous platinum(I1) comtions was to place small aliquots of the reaction mixturL,in plexes), several runs were made after treating about 100 ml. of ice cold acetone t o which had beeen added the platinum(1V) complex with an oxidizing agent.g 20 ml. of concentrated nitric acid per liter of acetone. Then a few drops of wool violet (1 g./l.) were added and the snlu- In every case there was no change in the observed tion titrated with 0.01 k?silver nitrate solution. The etid- rate constants. Also of interest was the appoint was takcti as tlie minirniim o f R plot OF niamp. us. pearance of an induction period of from 1 to 2.5 hr nil. of titrant. which had previously been observed by Musket.b ( 7 ) "Gmelins Handbuch der Anorganische Chemie." Vol. 08, 1)~). This too was unaffected by treatment with per470-010, Verlap Chemie (1957). manganate or ceric ion. Most of the runs were 18) F.Basolo. M . L.hforris and R. G. Pearson, Discussions Faraday SOC.,29, 80 (1960).
(9) Upon adding nitrite ion the excess oxidizing agent is destroyed.
NITRITE SUBSTITUTION IN CHLOROAMMINEPLATINUM(IV) COMPLEXES
Oct. 5, 1961
3945
made in the absence of acid t o avoid complications due to the decomposition of nitrous acid. With the addition of small amounts of platinum(11) t o the reaction mixture, the results changed dramatically. The pseudo h s t order plots were quite linear and showed no induction period. The rate increased with increasing platinum(I1) concentration. As the concentration of the reagents were varied i t became clear that the catalyzed reaction was first order in platinum(II), platinum(IV) and nitrite ion. Similar results were obtained with cis- and trans- [Pt(NH3)4C12]2+ and trans-[Pt(en)2Clz] 2+, the latter ion being most extensively studied. The catalysis by platinum(I1) and the form of the rate law is identical with the behavior shown by a large number of other substitution reactions of chloro complexes of platinum(1V) .*310811 The mechanism for the catalytic reaction then can be written in the same way as for the examples studied earlier. Using trans- [Pt(en)zCL] 2+ catalyzed by [Pt(er&] 2 + as an example, it would be fast Pt(en)z2+ NO^Pt(en)ZNOp+ en en slow C1-Pt-Cla+ Pt-NOo+ en en fast en en CI-Pt ..,C1 ...Pt-NO,'+ en en en en fast Cl-Pt ...Cl ...Pt-N0p3+ en en slow en+ en Cl-Pt Cl-Pt-N02'+ en en fast Pt(en)2C1+J_ Pt(en)22+ CI-
+
(1)
+
(2)
+
(3) (4)
That two chlorides are not released and a dinitro complex formed, may be due t o an unusual stability of the chloronitro complex compared t o the dinitro. A more likely explanation is that a nitro bridged intermediate, l 2 which is needed by this mechanism t o form a dinitro product, has a much smaller tendency t o form. The product of reaction between trans- [Pt(en)zCIS]^+ and nitrite ion in the absence of platinum(11) is also the chloronitro complex. This suggests that the same catalytic mechanism is operating since a simple substitution reaction would be expected t o give a dinitro product eventually. In all runs, with or without platinum(II), even long standing failed t o release more than one mole of chloride ion per mole of dichloro complex. I n the case of samples not containing added platinum(11), a few per cent. excess chloride ion was found. This, however, was accounted for by the appearance of 5-10% of platinum(I1) after times corresponding t o infinity on the scale for which kinetic studies were made. a (10) F.Basolo, P. H. Wilks, R. 0.Pearson, R. G Wilkins, J . Inorg. Yudcar Chcm., 6, 161 (1958). (11) F. Basolo and R. Johnson, ibid., 18, 36 (1960). (12) We have called the bridged species shown in equations 2 and 3 an intermediate, for reasons discussed in ref. 8. It could equally well be an activated complex. (13) Platinum(I1) was determined by permanganate titration after destroying the nitrite ion with sulfamic acid. Platinum(I1) in small
order plot; [Pt(en)~Cl~]'+ = 0.005 M, [NOo-] = 0.200 M.
The oxidation-reduction potential for the over-all " 0 3 HnO
Pt(I\')
I -
+
Minutes Fig. 1.-Pseudo-first
+
+
Pt(I1)
+ Nos- + 3H"
(5)
reaction is 0.4 to 0.8 volt for Eo,the value depending on the groups coordinated t o p 1 a t i n ~ m . l ~This indicates that nitrite ion readily can reduce platinum(1V) to platinum(I1). It is also known4 that excess nitrous acid can oxidize platinum(I1) complexes t o a dinitro complex of platinum(1V). Tables I and I1 show that adding small amounts of acid (loyoof the nitrite ion) has little effect on the rate when no platinum(I1) is added or when only small amounts are added. However, if larger amounts of platinum(I1) are added, nitrous acid does cause a large drop in rate, presumably by oxidizing the divalent platinum. I n view of all of the above observations, it seems logical t o assume that the reaction between a chloro complex of platinum(IV) and nitrite ion to produce a nitro conlplex is a two stage process in
+
[ P t ( e n ) ~ C l ~ ] ~NOZ+ --+ [Pt(en)2ClX02]2+ C1-
+
(6)
which the fist step is reduction to platinuni(I1)
+
+
ki
-+
[ P t ( e n ) p C l ~ ] ~ + NOpH20 Pt(en)2+ 4- NO'-
4- 2C1-
+ 2H+
(7)
of some of the original complex. This platinum(11) then causes a catalyzed reaction of the platinum(1V) with nitrite ion according t o equations 1 to 4, and with an over-all rate constant kp. That this scheme can explain the observed results can be shown by deriving the rate law. By comamounts always was found during the kinetic runs as well, but the analyses were not reliable. (14) W. M. Latimer, "The Oxidation States of the Elements and Their Potentials in Aqueous Solutions." 2nd Ed., PrentictHall, New York, N. Y.,1952.
Vol. s3
HERBERT R. ELLISOX, FREDBASOLOAND RALPHG. PEARSON
3946
TABLE I1 RATECONSTANTS F O R
XITRITE SUBSTITUTION I N cHLOROAMXINEPLATINUM(
Iv) COMPLEXESa AT 50" AND WITH i l D D E D Pt( 11)
I r a n ~ - [ P t ( e n ) ~ C 1 ~6 ] 2 + [ P t ( I I ) ] ,.If
[ P t ( I V ) I ,A i
,0047 ,0048 ,0094 ,0144 ,0047
x x 0.55 x 1.00x 1.00 x 2.00 x 4.00 x 5.00x 6.00 x 1.00 x 1.00 x 4.00 x
,0025 ,002*5
2.5 2.5
0.1 0.1
0,0048 ,0046" ,0050
,0050
.0050d.' ,0050 ,0047
x x
[NaKOz], M
[NaClOa], M
0.200 0.020 ,200 ... ,200 .020 ,200 .020 ,200 ... ,200 ,020 ,180 ,040 ,200 ,020 .070 ,150 ,200 ,005 ,190 ... . 100 ,120 [ Pt( XHy)jC1] + ,200 ,020 ,200 .012
10-4
10-4 10-4 10-4 10-4 10-1 10-4 10-4 10-4 10-4
10-4
10-4 10-3
k' (iW-lrnin.-l)
2.12 x 1.72 X 3.39 x 6.03 X 2.27 X 12.0 x 23.5 X 30.4 X 37.8 X 6.10 X 6.05 X 23.6 X
kzc(M-zmin.-I)
.........
10-2
lo-*
......... 6.18 X lo2 6.03 X lo2
10-2
lo-*
......... 10-2
lo-*
lo-?
6.00 X 5.86 X 6.08 x 6.30 X 6.10 X 6.05 x 5.89 x
10? 10' 10' lo2 10' 102 102
S o C1- in 2 weeks t = SO", no CI-in 3 days, -207c rp-
lease in 4 weeks cis-[Pt(NH3)1Cl~]~+',' 2.43 x 10-4 1.22 x IO-' ,200 ,020 ,0025 0.0020 3.23 X IO-* 1.29 X 10-I ,200 .020 .0025 .0030 tmns- [Pt(NH;),C12] ,200 ,027 3.93 x 10-2 1.96 X lo2 .0025 2 . 0 x 10-4 3.82 X 1.91 x 102 ,200 ,026 .003 2 . 0 x 10-4 ,200 .026 3.92 X 1.96 X l o 2 2 . 0 x 10-4 ,003 7.83 X 1.96 X lo2 ,180 .046 4 . 0 x 10-4 .003 10.32 X 2.06 X lo2 ,150 ,076 5.0x 10-4 .003 11.00 x 10-2 1.83 x 102 ,150 ,076 6 . 0 x 10-4 .003 Pt(I1) added as [Pt(en)?]All platinum salts are perchlorates unless otherwise noted, all runs made a t p = 0.235. Perchloric acid added; [HC104] = 0.020 M . e Potassium permanganate added to dissolve (C104)2. kz = k ' / [ P t ( I I ) ] . complexes before reaction. f Pt(I1) added as [PtiKH3)4](C104)2. Pt(1V) used as the nitrate salt.
'+
(1
bining equations 3 , 4 and 5 and using abreviations one obtains the following schematic mechanism.
+ N -+
ki
A
B
k2
+B +N+B
+ 2C D
(9)
where A is [Pt(en)zC12]2+,B is [Pt(en)2I2+,N is the nitrite and C is the chloride ion. Making note of the fact that the nitrite ion was present in large excess, it is possible to write differential equations using the relationships kl' = kiNand kz' = k&. dA/dt -ki'A - ks'AB dB)dt ki'A
(10) (11)
Dividing (10) by (11) and integrating (assuming no B a t zero time) results in A
Ao - B
-
'/2ky'/ki'B2
(12)
Substituting (12) into (11) and again integrating vields where Y
= (2Aokz'/klf
+ 1)'/2
and K
=
k2'/kl'
(13)
Expanding and rearranging (13) and making use of the fact that r>>l and hence r2 = 2Aokz'/kl' (since experimentally kz' is much larger than kl') gives From the stoichiometry, A . can be shown that
=
A
(16)
(8)
+C+
-4
C = '/'kz'/ki'B2
Substituting (16) and (14) into (15) and rearranging extensively finally results in
+ B + C, it
Equation 17 states that a plot of the left hand side vs. time should be a straight line with zero intercept. Such plots are indeed found; two examples are shown in Fig. 2 . Deviations occur only a t S0-9070 completion. The slopes of these lines are given by rk1'/2.303 which can be rearranged t o give kl
=
(2.303 X slope)2 2AokzN2
(181
The value of kz to be used in calculating k l can be taken from runs made in which the iirst slow step of the mechanism is by-passed by adding platinuni(11). These values are found in Table 11. The last column of Table I shows that over a wide range of concentrations of platinum(1V) complex, nitrite ion and acid the values of kl calculated from equation 18 are fairly constant. T h a t this mechanism and the resulting equations can reproduce the experimental data over most of the reaction is shown by Fig. 3. The only chloroammineplatinum(1V) complex investigated that did not appear to fit this mechanism was trans- [Pt(tetrameen)2C12)2+.With this complex 1.5-2.0 chloride ions per ion of complex were released. Pseudo k s t order plots were linear with no induction period but there was more
Oct. 5, 1961
NITRITES U B S T I T U T I O N
1.6
I N CNLOROhMrdI1;EPL.4TINUM
(
3947
COMPLEXES
ZUI
0
1
1.2
I
I 0
200
I
400
I 600
Minutes. Fig. 2.-Plot of left-hand side of equation (15) os. time; A, [Pt(en)2C12]2C = 0.0050 M , [NOz-] = 0.200 M ; 0, [Pt(en)~Clz]Z+= 0.019 M , [NO,-] = 0.175 M .
scatter t o the data. Variation of the initial amount of platinum(1V) appeared t o have little effect on the observed second order rate constants while the addition of platinum(I1) in the form of [Pt(tetrameen)a] 2 + or [Pt(en)~] 2 + had no catalytic effect. Runs with varying amounts of acid resulted in similar rate constants, although somewhat lower than runs made with no acid present. Most significant, however, were the results of titrations with permanganate. These titrations showed that most, if not all, of the release of chloride ion was due to the reduction of platinum(1V) t o platinum(11). Thus it appears that there is essentially no substitution with this particular complex. Such a result is completely consistent with the postulated mechanism. Because of the bulkiness of the C-methyl groups on the chelate rings, the [Pt(tetrameenj2]2 + cannot get close enough to form a bridged complex through the chloro group. Therefore the bridged mechanism is not available t o this system and only reduction can take place. The mechanism for this particular process however is not clear from this investigation and needs further study. That substitution does not take place is also consistent with the observation that there is no chloride exchange with this complex.8 From the first order plots for [Pt(tetrameen)zCl,] 2+ pseudo-first order rate constants were obtained, which on division by the nitrite ion concentration gave a value of kl equal t o about 6 X M-l min.-'. If this can be identified with the rate of reduction as in equation 7, then it is of interest that this is some ten times faster than kl for trans- [Pt(en)&!lx] 2+. Steric strain in the alkylated complex evidently gives rise to a tendency to go to a state of reduced coordination number.
Minutes. Fig. 3.-Calculated and experimental [Cl-1 os. time: , calculated from equations 14 and 13; A, experimental points, [ P t ( e n ) ~ C l ~ = ]~+ 0.005 M, [NOz-] = 0.200 M ; O, experimental points, [Pt(en)2Cl2lz+ = 0.019 M , [NOz-] = 0.175 M.
Further comparison of the chloride exchange data reported by Basolo, Morris and Pearsons and the present nitrite substitution results indicates a close similarity between the two reactions. Although there is a 25' difference in the ternperatures of the measurements inspection of the results listed in Table I11 shows a striking correlation between the two sets of constants. Indeed, a loglog plot of these constants is linear with a slope of almost unity. The fact that the chloride exchange is somewhat faster may be due t o the inclusion of the formation constant of the five-coordinate intermediate of reaction 3 in kz. However, it would be expected that this intermediate would be more stable for nitrite ion than for chloride ion since nitrite ion is a good nucleophilic reagent for platinum(II).'5 A more likely explanation is t h a t reTABLE I11 RATE CONSTANTS FOR
THE
PLATINUM(II)CATALYZED RE-
ACTION
Temp. Platinum complex
trans- [Pt(en)zCL]+2 trans- [Pt(en)~Clz]+2
+'
Nucleophile
NOzb c1* NorC Ci *
("C.) k ( M - 2 min.-l)a
50
6.07 X lo2
25
9.1
x
102
1.26 X lo-' &-[Pt(NHs)4ClzJ 50 1 . 6 X lo-' cis-[Pt( NHs)aClz] +' 25 1.95 X lo2 trans- [Pt( NHs).iCl%]+2 50 trans- [Pt(NHl)rClz]+2 c1* 25 2 . 0 x 102 Nore [Pt( "8)jCll +$ 50 Very slowe c1* 25 3 . 9 x 1 0 - 2 [Pt(NH06Cll +3 trans- [Pt(tetrameen)zCIZ] +a NO2 50 No substitution trans- [Pt(tetrameen)z25 c1* No exchange ClP] +a a Chloride exchange results are from ref. 8. 1.39 b k1 f 0.09 x 10-4 M-1 min.-l. c k 1 = 1.5 X 10-4. kt = 6.0 f 0.3 X lo-'. No chloride released in two weeks., a t 80" there was some after one week. f k1 = approximately60 X M-'min.-'. (15) R. G. Pearson, H. B. Gray and F. Basolo, J. A m . Chcm. Soc., 89, 787 (1960).
39 kS
HERBERT R.Er,r,rsox. FREDBASOLOAND RALPHe. PEARSON
Vol. 83
the Pt-N bond is stronger than tlic 1%-C1 bond, it follows that the rate with the cis isomer ud1 be slower than with the trans isomer. This is cornpletely in agreement with the observation that in such systems the cis isomer is more diiiicrrlt to reduce than is the trans forrn.ls Although a chloronitroamniineplatinuin (1V) complex was not isolated and completely characterized during this study there is sufficient cvidence in support of such products. In all of the runs, except with trans- [Pt(tetrameen)zClz]2t which underwent reduction, only one chloride ion was ever observed to be released if the reactions were carried out in the dark. This had previously been ( a! (B) reported by Wilks4 and M ~ s k e t . Such ~ an observation argues strongly for a chloronitro complex, Fig. 4.--Bridged iiitermediates prc)posed to explain nitrite on the basis of the bridge mechanism it is substitutioti in 1ruus- (a) and c ~ . ~ - [ P t ( ~ H : ~ )(b). . ~ ~ l ~ ] since z+ impossible to form a dinitro complex as mentjoned action 2 is readily reversible. Thus the ratio of previously. From the preparative point of view tlie reverse rate constant of (2) to i.he forward salts of such coniplexes as trans-[Pt en(NHy)* constant of ( 3 ) may be rather large. I n the iso- C1NO.I *+, trans- [Pt(NH3)3C1ClN@2]+, trans-(Pttope exchange reaction this ratio is necessarily (NH enCl)ClN02) +, trans- [Pt(NH-,enNOz)-CIKunity. O,] + and trans-[Pt(a en b)CIN02] + where a is Re.asoning from the chloride exchange resultss methylamine or ethylaniine and b is C1 or NO, leads one to predict that the nitrite reaction with have been i ~ o l a t e d . ~In every case the chloro[Pt(NHJ5C1I3+ would hare a ka of about 3 X nitro complex was prepared by treating nitrite ion lo-* 1.2 mole-2 min.-l, Le., about l/20,000 as fast with the corresponding trans dichloro complex, as the nitrite reaction with tr~ns-[Pt(en)~Cl2] 2f. often at high temperatures and for long periods However, on the basis of the postulated mech- of time. These results can be completely acanism with bridge formation taking place through counted for on the basis of the bridge mechanism. the chloro group, one would predict that trunsRecently Chernyaev, Nazarova and MiranovaL7 [Pt(NH,?),lUO,Cl]2 + would be formed and that reported their results of a study of the reaction of ainrrionia would be released and not chloride ion. nitrite ion with chloropl3tinate:IV) ion All that can be stated from our results is that the reaction is certainly very slow as predicted by the [PtC16]2 + ?i No,- + [Pt(S02)zCl,z]*z C1- (18) mechanism. An attempt to make a run on a preparative scale to isolate Irans- [Pt(NHs)dNO2Cl] + failed because of extensive reduction (about They were able to prepare chloronitro cornpleies 30’%) and because both dichloro and chloronitro with x from 1 to 5 but were not able to isolate the products seemed to be formed. The observation hexanitro compound. To explain their results that [Pt(NHJaC12]?+ is forrred during tlie reaction they proposed that platinum(1V) was reduced has been noted before by Basdo, Morris and Pear- to platinurn(I1) and direct substitution of from 1 son6 and by Rubinstein16 who reported that, to 4 chlorides by nitrite took place. These comin the presence of chloride ion, catalytic amounts plexes then were oxidized to platinum(1V) uith of [Pt(N€13)4] 2 + will convert [Pt(NI-Ia),Cl]3 + in two chloride ions taking up trnnb positions in the high yield to trans- [Pt(NH,) *+. Because of coniplexes. They then proposed that nitrite ion this and tlie reduction by nitrite the reaction is ieplnces one of the chlorides in the Cl-Pt-CI axis certainly more complex t h m those of the other to give the final products. The authors eniphasized that their experimental results shmved chloroarnn~inepl~tinurll(IV) coiiqAexes studied. Examination of Tahle 111 shows that the rate that chlorine was not replaced directly by the of nitrite substitution ( K 2 ) of tram- [Pt(NHJ*- nitro groups in chloroplatiti?tejlv). This tonClp]2 + is about 1,500 times faster than that for the clusion is precisely the same as the one arrived :It cis isomer. This difference is roughly the same as in this study, namely, that nitritc substitution that found in the chloride eschnrige nieasurenients in chloroammineplatinuni(1V) coniplexes 1rroceeds (see Table 111). X proliable reason for this may through a complex reduction-oxidation nTechanism involving a bridge between the reduced and be seen by considering the bridged iiiternicdiates oxidized forms of platinum. Thus the pro1 osed for these two compleses. For the traiis isomer in Fig. 4.4 reduction of the platinuIn(1T.’) complex bridge rrechanism can be used to explain completely why only one chloride ion is replaced in the last t o platinum(I1) by the bridged mechanism requires the rupture of the Pt-Cl honrl opposite the step of the chloroplatinateiIl’)-nitrite ion reaction. Acknowledgment.-We wish to achnowledpe the chloro ,group, as represented by tlie asterisk. However in the cis isomer this process requires the support of the U. s Xtomic Energy ~ o ~ ~ i x n i s s i o ~ i cleavage of a Pt-N bond (asterisk in Fig. 4B), Contract At (11-1) F,!)-Project No. 2 and the since ammonia is opposite the chloro bridge. Since Xational Science Found.ition grant NSF-GT,lOi
+
(16) A. hf. Rubinstein, U.R.S.S. Coinpf. r e n d . , 28, 5.5/58 (1940); Izwrt. PIaf , 2 0 , 5 3 / 9 3 . 50 ( I W T ) .
(17) I I Chrrnyacv 1- 4 N a z i r o r a Inorg Cheni 4, 310 (1950)
11111
2 h \Iirani,va
Kiiss J
