Vol. 8, No.

10, October 1969 OXIDATIONS

OF TRIS-OXALATO
COMPLEXES220 1

CONTRIBUTION
FROM THE DEPARTMENT
OF CHEMISTRY,
BOSTON
UNIVERSITY,
BOSTON,MASSACHUSETTS
02215

The Cerium(1V) Oxidations of Tris(oxalato)chromate(III) and

Tris(oxalato)rhodate(III) Ions in Aqueous Sulfuric Acid1I2
BY MARTHA W. HSU, HARRIET G. KRUSZYNA, AND RONALD M. MILBURN

Received February 17, 1969

and Rh(C204)~~- react with Ce(IV) in aqueous sulfuric acid with the stepwise oxidation of coordinated oxalate,
the initial reactions proceeding to produce bis-oxalato complexes. For the Cr(C20&3--Ce(IV) system evaluation of rate
constants for the initial direct redox reaction requires consideration of the competing aquation of Cr(Cz0&3-. In 1 M
sulfuric acid a t 25” the second-order rate constant for the direct one-electron oxidation of Cr(Ct04)33- is 4.0 (f0.5)X 10-2
M-1 sec-1, while for the corresponding direct one-electron oxidation of Rh(C20&3- the rate constant is 6.1 ( & l . O ) X 1 0 - 4
M-1 sec-1. Rate behavior for several aqueous-acidic sulfate solvents, differing in concentrations of hydrogen ion and sulfate
ion, suggests that activated complexes for the initial direct redox reactions of Cr(C~04)~3- and R h ( C ~ 0 4 ) ~
are
~ -of generally
similar but not identical compositions. The measured AH* values for both systems exhibit slight temperature dependences
which are probably caused by small changes with temperature in the average compositions of activated complexes. A t
corresponding temperatures the AH* value for the rhodium system is significantly higher than that for the chromium system.
The initial direct redox step for each complex is interpreted as involving oxidation of a coordinated oxalate to an oxalate
radical anion. Factors which are likely to contribute to differences in the redox reactivities of Rh(C204)33-and Cr(C20&3-
are discussed.

The area concerned with reactions of coordinated reactions have much in common and are conveniently
ligands in metal complexes has received increased described together. Some different features appear for
attention over the past several years3 A matter of the reactions of Ce(1V) with C0(C204)3~- and with
interest to us has been the reactions of inert metal Ir(C204)33-,and these systems will be described sep-
complexes containing oxidizable ligands toward ex- arately.’
ternal oxidizing agents, and in this connection partic-
Experimental Section
ular attention has been centered on the oxidation of
Materials.-Preparative methods for potassium tris(oxa1ato)-
metal-oxalato complexes by cerium(1V). An earlier
chromate(II1) and for solutions containing cis-bis(oxa1ato)di-
study reported on the oxidation of oxalato complexes aquochromate(II1) ion have been described.j Solutions of these
of chromium(III), and while major emphasis of this complexes used for kinetic studies were standardized by analysis
work was on the C~~-C~(H~O)Z(C~O~)~--C~(IV) reaction, for chromium as before.6 Potassium tris(oxalato)rhodate(III )
a preliminary account of the Cr(Cz04)2--Ce(IV) re- was prepared by Barton’s modification of Werner’s method.8
The salt was dried under vacuum, allowed to equilibrate with the
action was also given.6 Aqueous acidic-sulfate media
atmosphere, and finally stored in a calcium chloride desiccator.
were used to avoid polynuclear species of cerium(IV)6 After the salt had reached constant weight, it was ana1yzed.Q
and to keep reaction rates in a conveniently measurable Anal. Calcd for K3Rh(C204)3.1.5H20: C, 14.10; H, 0.59;
range. As part of an inquiry into the role of the central K, 23.15. Found: C, 14.1; H, 0.48; K, 23.2. For kinetic
metal of the complex in reactions of this type, the experiments solutions of potassium tris(oxalato)rhodate(III)
were prepared from the weighed salt. The method of Ayres and
Cr (C204)33--Ce(IV) reaction has now been examined
Forresterlo was used to prepare a sample of [Rh(H20)~](C104)3.
in further detail, and the Rh(C204)33--Ce(IV) reaction The preparation of standard solutions of cerium(1V) and
has been studied under similar conditions. The two cerium(II1) and analytical methods for cerium(1V) and cerium-
(111) were as before.6 The solvent used for the majority of
(1) Supported by the National Science Foundation.
(2) Presented at the Tenth International Conference on Coordination experiments was aqueous 1.00 il4 sulfuric acid, but to investigate
Chemistry, Tokyo and Nikko, Japan, Sept 1967; see Proceedings, p 302. the role of hydrogen ion and sulfate ion some additional aqueous
(3) See, for example: (a) “Reactions of Coordinated Ligands and Homo- acidic-sulfate solvents consisting of mixtures of sulfuric acid and
geneous Catalysis,” Advances in Chemistry Series, No. 37, American Chemi-
sodium sulfate were used.
cal Society, Washington, D. c.,1963; (b) M. Anbar in “Mechanisms of In-
organic Reactions,” Advances in Chemistry Series, No. 49, American Chemi- Stoichiometries.-Our earlier report6 gave support for the
cal Society, Washington, D. c.,1965; (c) J. P. Collman, et al., J . A m . Chem. view that Cr(C~04)3~- is oxidized by Ce(IV) in a stepwise manner,
Soc., 89, 1082 (1967), and earlier papers; (d) M. A. Bennett, et al., J. Chem. with initial reactions proceeding according to the stoichiometries
Soc., A , 2301 (1967); (e) M. T. Beck and L. Dozsa, Inovg. Chim. Acta, 1,
134 (1967); (f) A. D. Allen, et al., J . A m . Chem. Soc., 89, 5595 (1967); (9) +
Cr(C204)3S- 2Ce(IV) 2H20 ---f+
H. A. Scheidegger, J. N . Amor, and H. Taube, ibid., 90, 3264 (1968); (h)
S. M. Caldwell and A. R. Norris, Inorg. Chem., 7, 1667 (1968); (i) K. Schug,
+
ci~-Cr(H2O)~(C~04)2-2Ce(III) + 2C02 (1)
et ol., ibid, 7 , 1669 (1968); (j) E. Blinn and D. H. Busch, J . A m . Chem. cis-Cr(H20)2(C20.&- + 2Ce(IV) f 2H20 e
SOC.,90, 4280 (1968), and earlier papers; (k) J. E. Hix, Jr., and M. M. Jones, Cr(H20)4C204f+ 2Ce(III) + 2C02 (2)
ibid., 90, 1723 (I968), and earlier papers; (1) J. P. Candlin, K. A. Taylor,
and D. T. Thompson, “Reactions of Transition Metal Complexes,“ Elsevier The subsequent reaction of Cr(H2O)rCzO,+ with Ce(1V) proceeds
Publishing Co., Amsterdam, 1968, Chapter 3. significantly more slowly than the cis-Cr(HpO)z(CzO&-Ce(IV)
(4) In certain cases it may not be clear whether a reaction of a complex
with an external redox reagent is best described as a reaction of a ligand or reaction, and the process following Ce(IV) addition to cis-Cr-
ligands, a reaction of the central metal, or a reaction of the complex as a (HZO)Z(C~O~)Z- therefore adheres closely to the stoichiometry
whole. In the cases of present interest the ligand has, at least by the end
of the reaction sequence, undergone an obvious chemical change. (7) T o be submitted for publication.
(5) J. E. Teggins, M. T . Wang, and R. M. Milburn, ref 38, pp 226-242. (8) D. Barton and G.M. Harris, Inovg. Chem., 1, 251 (1962).
(6) See P. R. Danesi, Acta Chem. Scand., 21, 143 (1967), and references (9) Analysis by Schwarzkopf Microanalytical Laboratory.
therein. (IO) G . H. Ayres and J. S . Forrester, J . Inorg. Nucl. Chem., 3, 365 (1957).
2202 A m R.M. MILBURN
M. M:. Hsu, H. G . KRUSZYNA, Inorganic Chemistry

represented by reaction 2. On the other hand, reactions 1 and 2 For the Cr(CzO4)a3-system, the Ce(1V) concentration in reac-
proceed a t comparable specific rates, and a stoichiometry ap- tion solutions was measured as a function of time by three sepa-
proximating reaction 1 can be expected only for the early stages rate methods. The first of these involved quenching aliquots
of reaction between Cr( C204)33-and Ce( IV). of the reaction mixture a t various times by addition to excess
The Rh(Cz04)33--Ce(IV)system presents a similar situation, iron(I1) sulfate solution. The resulting Fe(II1) concentrations,
with redox processes continuing while both Ce(1V) and oxalate which were a measure of the Ce(1V) concentrations immediately
remain. Kinetic observations on Ce(1V) consumption, moni- before quenching, were determined from the absorbance of the
tored as described below, are consistent with the initial stoichi- thiocyanate complex a t 478 mp, using solutions which were 0.20
ometries M in potassium thiocyanate, ilf in sulfuric acid, and about
+
R ~ I ( C ~ O ~ )'Ce(1Y)
~~- 6 X A!! in Fe(II1). Because the iron(I1I) is sloivly reduced
Rh(C204)2- + 2Ce(III) + 2COr
by the thiocyanate, the absorbances were measured as a function
(3)
of time and extrapolated back to the time of quenching. The
Rh(Cz04)s- + 2Ce(IV) +RhC204++ 2Ce(III) + 2COs (4) extrapolated values were used to calculate [Fe(III)] by compari-
with reactions 3 and 4 proceeding a t quite similar specific son with a calibration curve obtained under similar conditions.
rates.l1-'3 Thus, when the initial molar ratio of Ce(1V) to Rh- In the second method, which we shall term direct spectropho-
(cso4)33-is in excess of 2 : 1, redox reaction continues a t a steady tometry, absorbances of reacting solutions were measured as a
rate even after 2 mol of Ce(1V) has been consumed per mole of function of time a t each of two wavelengths, namely, 480 and
Rh(C204)33-. Also, one can aquate Rh(Cz04)33-in aqueous acid 560 mp. A t 560 mp there is no significant absorption by the
to the bis-oxalato complex and free oxalate.8~12~'a When Ce(1V) Ce(IV), while a t 480 niM the Ce(1V) and the various chromium-
in 1 M sulfuric acid is added to solutions containing the bis- (111) species both contribute to the absorbance. From the
oxalato complex and free oxalate, prepared by aquation in either stoichionietries of reactions 1 and 2 and the absorbancy
1 M sulfuric acid or 2 A4 perchloric acid, and where the Ce(I1') coefficients of Cr ( C Z O -,~ ) ~cis-Cr(H20
~ ) 2 ( CSO~)Z-,Cr (H 2 0 14-

to rhodium ratio is in excess of 2 : 1, the initial rapid consumption Cz04+, and Ce(IV) a t each of the two wavelengths, i t is possible t o
of Ce(1V) by the free oxalate is followed by a continued steady calculate values of [Ce(lTr)]. -4 complication is experienced here
Ce(1V) consumption by the bis(oxalato)rhodate(III). I t is in that measured absorbances a t 480 nip, extrapolated t o zero
apparent here also that a stoichiometry approximating reaction time, exceed values calculated from the measured absorbances
3 can be expected only for the early stages of the Rh(C~04)3~-- of the individual components by up to about 8%. On the other
Ce(IV) reaction. hand, measured absorbances a t 560 mp, extrapolated to zero
Measurements of Rates of Ce(IV) Consumption.-Reactions time, are in very close agreement with values calculated from
were initiated by mixing, a t zero time, thermostated solutions of the absorbances of the components. The deviations a t 480 mp
K3Cr(Cz04)3or K3Rh(C$04)3with Ce(IV) solutions. The sul- are attributed to rapid complex formation between Ce(IV) and
furic acid was either included in the Ce(1V) solution, or added Cr(C204)s3-prior to the initial redox step. It may be noted here
to the solution of complex immediately before initiating reaction. that similar small deviations between calculated and observed
Where appropriate, sodium sulfate was also included in the absorbances in the vicinity of 480 mw were observed also for the
reaction mixtures. The rates of the reactions were followed by Rh(C204)33--Ce(IV) reaction. For the purpose of calculating
measuring the Ce(IV) concentration as a function of time. Ce(1V) concentrations a t various times for the Cr(C~0,)3~-
For the R h ( C ~ 0 4 ) &system
~- the time dependence of the Ce- system, an effective absorbancy coefficient for Ce(1V) a t 480 mp
(IV) concentration was measured by two independent methods. was obtained for each experimental run from the absorbance
In the first of these aliquots of the reaction solution were quenched extrapolated to zero time. Any uncertainties which could be
a t recorded times by addition of appropriate amounts of standard introduced by this procedure were reduced below the level of
Fe(I1) solution, and the excess Fe(I1) was titrated potentiometri- experimental significanceby concentrating attention on the initial
cally with standard dichromate solution. In the second method, stages of reaction 1, and this was necessary anyway because
aliquots of the reaction solution were at recorded times diluted reaction 2 and additional reactions become inconveniently im-
(100-fold to 1000-fold) by addition to sufficient 1 A l sulfuric acid portant a t larger reaction times. Confidence in the procedure is
to bring the Ce(1V) concentration into the range (2-10) X provided by the observation that Ce(1V) concentrations calcu-
iM, and the absorbance of each solution was measured immedi- lated by this direct spectrophotometric approach checked satis-
ately in a 1-cm cell a t 320 mp. The dilution had the effect of factorily with values obtained by the quenching method described
markedly slowing down the reaction (initial reaction being second above. The direct spectrophotometric method was more con-
order), and no significant Ce(1V) consumption took place be- venient than the quenching method and was therefore used for
tween the time of dilution and the time of absorbance measure- the majority of kinetic runs. At a later time, after the kinetics
ments. In 1 M sulfuric acid Ce(1V) obeys Beer's law a t 320 of the Rh(C204)33--Ce(IV) system had been examined by the dilu-
mp, and a t this wavelength it is the dominant absorbing species tion technique in which the absorbance of Ce(1V) was examined
with E 5520. For the other reactant, Rh(Cp04)33-, e is 605, and directly a t 320 mfi, this procedure was also applied to the Cr-
for the presumed initial rhodium product, Rh(Cz04)2(HzO)$-,e (C20~)33--Ce(IV) reaction. Results for this third and more
is ~ 2 5 0 , 1 4while for Ce(III), e is <50. The Ce(1V) concentra- direct procedure also agreed satisfactorily with the two other
tions were computed for the initial stages of reaction on the methods.
basis that the decreases in absorbances at 320 mp could be at- Evaluation of Rate Constants.-Kinetic studies of the Rh-
tributed to the Ce(1V) consumption together with the conver- (C204)33--Ce(IV)reaction were carried out in 1 iM sulfuric acid a t
sion of R h ( C z 0 , ) P t o Rh(G04)z(H20)~-. The rates of Ce(1V) 25.0' with initial Rh(C~04)3~- concentrations varied through the
consumption as measured by the two experimental methods were range 0.01-0.05 M while the initial Ce(1V) concentrations we'e
in good agreement. varied from 0.01-0.07 M . The initial redox reaction is first order
in each of the two reactants, as shown by satisfactory constancy
of second-order rate constants obtained from the initial slopes
(11) T h e Rh(Cz0a)z- ion formed by reaction 3 is probably predominantly
of t h e cis form, as i t is for the bis-oxalato complex formed b y acid-catalyzed for plots of the integrated second-order rate expression for the
aquation of Rh(C204)aS- ion,s,la313with residual coordination positions about varying initial reactant concentrations (Table IA). At inter-
the rhodium occupied by water and/or sulfate. mediate reaction times deviations from linearity of the second-
(12) K. V. Krishnamurty, 1nor.g. Chem., 1, 422 (1962).
(13) R. D. Gillard and G. Wilkinson, J . C h e m . Soc., 2092 (1963).
order plots become apparent, with Ce(1V) consumption now
(14) T h e value 250 was obtained when Rh(Cz0a)aa- was aquated in 2 M proceeding more rapidly than would be anticipated from the
perchloric acid for 15 days a t 45"; when t h e aquation was carried out in 1 ILI initial second-order rates (see Figure 1). The apparent accelera-
sulfuric acid, t h e observed absorbancy coefficient was 270. Use of the second
absorbancy coefficient would influence t h e calculated changes in the Ce(1V)
tion is attributed to the consecutive reaction of Ce(1V) with the
Concentration b y only -0.2%. bis(oxalato)rhodate(III) product of reaction 3 (see above). On
Vol. 8, No. 10, October 1969 OXIDATIONS O F %IS-OXALATO COMPLEXES 2203

+
cis-Cr(H20)z(C204)2- 2Ce(IV) 2H20-+ +
Cr(HzOhCz04+ ZCe(II1) + 2C02 (7) +
+
2Ce(IV) +2CO2 +
o,061
C202- ZCe(II1) (8)
Whereas reactions 1 and 2 were written to imply stoichiometry
alone, reactions 5-8 are written to imply kinetically distinct
reaction paths-although they do not represent elementary
steps in a mechanism.16 The kinetics of reactions 6 and 7 have
been previously examined in other media.6JsJ7 For the purposes
of the present study the rates of these two processes were re-
0.04 - examined in 1 M sulfuric acid and in some additional aqueous
acidic sulfate solvents as described below. For reaction 6 in 1 M
sulfuric acid studies of the absorbance changes at 410 mp gave
0.03 - the following fist-order rate constants: 15.0", k = 1.1 x 10-6
sec-l; 25.0°, k = 4.6 X 10-6sec-1; 35.0°, k = 1.51 X 10-4sec-1.
For reaction 7, studies according to the procedures previously
described gave the following second-order rate constants for 1 M
0.02 -
sulfuric acid, expressed in terms of Ce(1V) consumption: 25.0°,

TABLE
I
0.01 .
SECOND-ORDERRATECONSTANTSFOR REACTION
3 in 1 M SULFURIC
ACID^

0.00 A. S t 2 5 . 0 '
0 40 80 120 No. of 102[Rh- 102[Ce- lOak,
Time, min. expt (CgOda8-10 (IV) Io M-1 sec-1

1 2.048 4.34 1.05

Figure 1.-Second-order rate plot for the Rh(C~0~)3~---ce(IV) 2 2,048 6.52 0.99
reaction in 1 Msulfuricacid at25.0° where [Ce(IV)]o = 0.0434 N 3 2,066 6.52 1.15
and [Rh(CzOe)a3-]0 = 0.0827 N . Note that concentrations are 4 1.033 4.34 1.41
here expressed in normalities, computed on the basis of reaction 3. 5 1.033 6.52 1.16
Deviation from linearity is attributed to increasing importance of 6 2.066 2.172 1.29
reaction 4. 7 1.034 2.1i2 1.28
8 1,036 2.172 1.47
9 1.036 4.34 1.34
the other hand, the Rh(CzO&3--Ce(IV) reaction shows evidence 10 2.071 4.34 1.26
of being slightly inhibited by Ce(III), as indicated by separate 11 4.14 4.34 1.09
experiments in which varying amounts of Ce(II1) were added to 12 4.39 4.34 1.09
the initial reaction mixtures (Table I D ) . Because of the con- 13 4.71 2.172 1.24
secutive reaction of the bis-oxalato complex and the apparent Ce- Av 1.22 (ztO.20)
(111)inhibition and in order to avoid difficulties associated with
B. At35.0'
possible involvement of sulfate in the coordination sphere of the
rhodium, rate constants for the initial redox reaction between No. of
Rh(CzO&s- and Ce(1V) were evaluated from the integrated expt 102[l<h(CnOa)3~-lo 10ZICe(IV)10 lo%, .Lf-1 sec-1
second-order rate law plots considering the first -10% of reac- 1 1.975 2,136 7.56
tion. The rate constants obtained in this way are estimated to 2 0,988 2.136 7.70
be reliable to within about 15%. The data in Table I were ob- Av 7.63
tained on this basis. Note that the rate constants in Table I
C. A t 4 5 . 0 '
are expressed in terms of consumption of Ce(IV) and refer to t h e
No. 01
over-all initial two-electron redox reaction 3. (The rate constants expt lu
l O ~ ~ R h ( C ~ 0 d r J - ~ l10glICe(IV)
11 IOV<, M-1 sec-1
for one-electron oxidations of the rhodium and chromium com-
1 1,127 1,086 5.56
plexes given in the Discussion are equal to half of the rate con-
2 0.266 1,084 4.44
stants for the over-all two-electron oxidations.) In addition to
3 0.32% 4.34 5.80
the data for 25.0°, results are given for 35.0 and 45.0' (Table IB
4 0.543 1.086 5.36
and C).
Av 5.29
From preliminary kinetic analysis of the Cr(C204)s3--Ce(IV)
reaction it became apparent that the direct redox reaction could D. A t 25.0", with Ce(II1) Added
be satisfactorily described by a second-order process, with the No. of
reaction rate being proportional to the concentration of each expt 102[12h(C20a)aa-]o 102[Ce(I\')lo 102[Ce(III)lo 104k, M-1 sec-1
of the two reactants. However, the acid-catalyzed aquation of 1 2.059 4.65 0.57 9.70
Cr(Cz04)33- to cis-Cr(HtO)~(C~04)~- proceeds for the chosen 2 2.059 4.46 1.19 9.30
conditions sufficiently rapidly that it needs to be taken into 3 2,059 3.96 2.12 7.56
account in calculating the rate of the direct redox reaction. 4 2.059 4.12 2.09 7.50
A41so, as previously described,E the bis-oxalato complex reacts a Concentrations are in moles per liter, and rate constants
with Ce(IV) at similar rates to the tris-oxalato complex. Evalua- refer to the consumption of Ce(1V).
tion of rate constants for the direct reaction of Ce(1V) with Cr-
(C20&3- therefore requires consideration of the competing and
consecutive processes
+ +
Cr(C904)~~- 2Ce(IV) 2H20 ---f
(15) See Discussion.
(16) K. V. Krishnamurty and G. M. Harris, J . Phys. Chem., 64, 346
+
ci~-Cr(H~O)~(C204)2-2Ce(III) + 2C02 (5) (1960).
(17) C. A. Bunton, J. H. Carter, D. R . Llewellyn, A. L. Odell, and S. Y.
Cr(C204)3a- $- 2H20 -+ +OC~ )~Z0-4 ~ -(6)
C~S-C~(HZO)~(CZ Yih, J . Chem. Soc., 4622 (1964).
2204 M. W. Hsu, H. G. KRUSZYNA,
AND R. 141. MILBURN Inorganic Chemistry

k = 0.20 A C ' sec-'; 35.0°, k = 0.32 A P sec-l. In compari- of hydrogen ion and sulfate ion on the reactions. The com-
son to reactions 5-7, reaction 8 is rapid and may be considered positions of these additional media, which were computed
to proceed to completion without delay.'8 For the reactions from the data of B a e ~ , are
~,~ ~ in Table 111. The rate
given
which ensue upon addition of Ce(1V) to Cr(C204)33-we then constants obtained for these media a t 25.0' are summarized
write for the Cr(C204)a3-system and for the initial R ~ ( C Z O ~ ) ~ ~ - - C ~ ( I V )

-dC/dt = kjAC + k7BC + 2k6A (9)

reaction in Table I V .

where A , B , and C are the concentrations of Cr(C204)33-, cis-

Cr(H20)2(Cn04)2-, and Ce(IV), respectively, and the subscript TABLE 111
on each rate constant is the number of the reaction t o which it COWPOSITIOSS OF THREEAkQGEOUS
refers. Also we have SULFURIC
ACID-SODIUM SULFATE
MEDIA
-dA/dt = '/zkjAC + keA (10)
Desig-
nation Stoichiometric Equilibrium Ionic

dB/dt = ksA + '/ZkjAC - '/?k7BC (11)

of --concn, JI--- -concn,
media [HzS041 [NaS041 LHSOa-I LH+I
.W--.
[SO421
strength,
M

In addition it is necessary to consider the stoichiometric rela-

I 0.950 ... 0,675 1.225 0.276 1.50
I1 0.588 0.238 0.518 0.592 0.276 1.34
tionships which are summarized by the equations I11 0.641 0.020 0,492 0,790 0.169 1.00
IV 0.691 0.207 0.597 0.785 0.301 I..X
A =Ao-w-u (12)
B =Bo - V + (w+u) (13)
IV
TABLE
C = C0-2(w+u+v) (14) R.4rE CONSTANTS" FOR T'ARIOGS ?VfEDIA AT '5.0"
where subscript zero denotes zero time, w is the concentration of
Cr(C204)~~- consumed by reaction 5, u is the concentration of
Reaction
n0.C ---.-------.---~,I~diumb-------.----
I I1 I11 IV
-
Cr(C204)33-consumed by reaction 6, and v is the concentration of 6d 3.5 X 8.4 X 1.6 X 1.7 X
cis-Cr(H20)2(C204)2- consumed by reaction 7. Exact mathe- 'ie 0.19 0.13 0.21 0.14
matical solutions to the set of differential eq 9-11 cannot be 5e 8 . 0 x 10-2 4 . 3 x 10-2 9 . 0 x 10-2 4 . 9 x 10-n
~ b t a i n e d . ' ~Plots of Ce(1V) concentrations against time were 3s 1 . 0x 10-3 0.20x 1 0 - 3 0.75 x 10-3 0 . 2 0 x 10-3
therefore used to estimate values of - d C/dt for successive closely
spaced times (0.5-min intervals were in general used). These
a T'alues based on duplicate measurements. * For composi-
tions see Table 111. Reference is to number designated to
values together with the known values of kg and ki and equa- reaction iii text. First-order rate constant in sec-'. e Second-
tions 9, 10, 12-14 were then used in an iterative procedure which
order rate constant in sec-l.
allowed values of k j to be calculated for each of the time intervals.
The Cr(C204)a3--Ce(IV) reaction was followed in this way for
about the first 20% or the reaction. Reasonable constancy of
the k, values was obtained over this portion of the reaction, and Discussion
the values also appeared to be independent of the initial con- The R ~ ( C Z O ~ ) ~ ~ - - C ~
redox
( I V )reaction and the
centrations of the reactants. Results for two typical reaction
runs are summarized in Table 11. The average value of k j for
second-order path for the Cr(C204)a3--Ce(IV) redox
four reaction runszoa t 25.0" in 1 M sulfuric acid is 8.0 (i.1.0) X reaction may be interpreted most simply as involving
sec-I. With coinparable studies a t other tempera- rate-determing one-electron oxidations of the tris-
tures in 1 &'sulfuric acid we find at l5.0", ki = 3.2 X A-1 oxalato complexes, followed by rapid one-electron oxi-
sec-I and a t 35.0", k j = 2 . 5 X lo-' ~ l 4 - lsec-I. dation of the metal-oxalato intermediates by additional
Ce(1V). This type of mechanism has been previously
TABLE I1 outlined and supported5in relation to the Ce(1V) oxida-
SECOSD-ORDER RATECONSTANTS
FOR tions of Cr(C204)33-,Cr(H20)4(C204) +, and especially
REACTION5 IS 1 M SULFURIC czs-Cr(H20)z(C204)2-, and the general arguments pre-
-Expt 1, [Cr(CzOa)83-]0= 4.62 X -Expt 2 , [Cr(Cz0da3-1o = 2.31 X sented for these chromium complexes apply also to the
10-8, [Ce(IV)]o = 2.172 X 10-2- 10-3, [Ce(IV)]. = 2.172 X 10-2-
Time, sec 1 0 2 k s , hf - 1 sec-1 Time, sec 102ks,A - 1 sec-1 inert Rh(Cz04)s3- ion. The essentials of the suggested
30 8.3 30 6.7 common mechanism, as applied to the tris-oxalato
60 8,s 60 7.3 complexes, can be specified without reference to resid-
90 7.5 90 7.2 ual coordination positions in the bis-oxalato products
120 7.0 120 7.8 or to details of the coordination sphere about the
150 7.7 150 7.7
180 7.5 180 6.0
cerium. For the rhodium system slight reversibility
Av 7 . 8 Av 7 . 2 of rcaction 15 is indicated by the data, implying that
Q Concentrations are in
refer to the consumption of
moles per liter, and rate constants
Ce(1V); temperature is 25.0". &f1r1(c204)33- CeIv + (z) +
slow
?VI(C>O~)~~- CeIII (15)
ht(C20a)32- + CeIV +1MXI1(C2O~)~-
+ CeIII + 2C02 (16)

Studies in Other Media.-Several reactioii media in addition the intermediate is efficient as both an oxidizing and a
to 1 174 sulfuric acid were used to obtain information on the roles reducing agent. Cerium(II1) inhibition of this type
was clearly evident; in the Ce(1V) oxidation of cis-
(18) Y.El-Tantawy and G. A . Rechnitz, A w o l . Che717., 36, 1774 (1964). Cr(H20)2(C20&-. For Cr(Cz04),33 - the kinetic analy-
(10) See, for example, S. W. Renson, "The Foundations of Chemical
Kinetics," McGraw-Hill Book Co., Inc., New York, N. Y . , 1960, p 42. sis is more difiicult. and a clear conclusion on the
(20) Additional experiments in a medium of 1.83 41 sulfuric acid reillforce
t h e view t h a t t h e initial direct reaction is first order in each of Cr(C;On)&
and Ce(1V). (21) C . F. Baes,J. A m . C h i n . Soc., 79, 5611 (1957)
Vol. 8, No. 10, October 1969 OXIDATIONS COMPLEXES2205
OF TRIS-OXALATO

reversibility of reaction 15 for this complex has not a t evidence for such similarly bridging oxalate in the solid
this time been drawn. phase has been noted for [ { Ni(en)z)2C204] (NO3)? on
Before comparing the redox reactivities of Rh- the basis of ir and X-ray studies.29 On the other hand,
(CZO4)s3-and Cr(C~04)3~- in further detail, it may be evidence has accumulated which suggests that in
instructive to refer to the redox reactivity of the latter aqueous acidic solution both Cr (C204)33- and cis-Cr-
complex in relation to observations which have been (HzO)z(Cz04)z- are in rapid equilibrium with forms in
made on C ~ ~ - C ~ ( H ~ O ) Z ( C Cr(Hz0)4Cz04+,
ZO~)~-, and which an oxalate is coordinated to the chromium by a
free oxalate. Observed rate constants, in 1 M sulfuric single o ~ y g e n . ~It
~ Jis~therefore not unreasonable to
acid a t 25" for the Ce(1V) one-electron oxidations, are: suggest alternatively that binuclear Cr-Ce complexes
Cr(Cz04)33-, 4.0 X M-l sec-l (k15, this work); could be rapidly formed in which oxalate is coordinated
cis-Cr(HzO)z(C~04)~-,1.0 X 10-1 M-l sec-1;22 Cr- to one or both metal centers by a single oxygen.
(Hz0)4Cz04+,1.5 X M-l sec-1;22 and HzCz04, Likely structures are
28 M-l sec-l.18 For free oxalic acid and the three chro-
mium complexes it is apparent that one cannot correlate
the rates in any simple way with the charges carried by
the principal oxalato species.2ar24The relatively high
reactivity of free oxalic acid is most likely related to the
facility with which Ce(1V) and oxalic acid combine
where the sixth coordination position about the chro-
rapidly to form cerium(1V)-oxalato complexes. Evi-
mium could possibly accommodate a water molecule.
dence has been presented18that the predominant tran-
I n the chosen medium one would not anticipate that
sient complex formed in sulfuric acid mediaZ6tz6 is Ce-
the concentrations of any of the above binuclear types
(S04)z(Cz04)2-, and this has the same composition as the would be large, because of the high concentrations of
transition state for the initial and rate-determining
sulfate and its considerable affinity for cerium(1V)26
redox process. The apparent facilitation of the oxida-
and because an already coordinated oxalate would not
tion by direct coordination of oxalate to the cerium is
be expected to have a particularly high affinity for a
similar to the situation which has been long recognized
second metal center. The significantly greater re-
for the Mn(II1) oxidation of oxalic
activity of free oxalate compared to the several chro-
The Ce(1V) oxidations of the chromium(II1)-
mium complexes may be a t least partially attributable
oxalato complexes also almost certainly proceed via the
to higher concentrations of rapidly formed complexes
initial formation of Ce(1V) complexes. For the Cr-
in which the cerium(1V) and oxalate are brought into
(Cz04)33--Ce(IV) system direct evidence for a rapidly close association. Rationalization of the relative oxi-
established preequilibrium reaction is provided by the
dation rates for the several different chromium com-
observed differences between measured and calculated
plexes is a more difficult proposition. It cannot be
absorbances for zero redox reaction time. These
assumed, for example, that the presumed reactive bi-
spectral differences are not large, and we have not a t
nuclear intermediates (and likewise transition states
this point been able to make a detailed investigation of for the initial redox step) will necessarily have identical
the preequilibrium reactions. Such rapid associations
compositions in soh2-and H+. Unraveling the precise
between any of these oxalato complexes and Ce(1V)
role of s04'- and H+ is an intricate matter and as a
could be achieved, without chromium-oxygen bond
minimum necessitates examination of various media
cleavage, if oxalate functions as a double bidentate
together with consideration of the way that the equilib-
ligand rium quotient for bisulfate dissociation can depend on
the specific composition of the medium.5 Some studies
of this type have been carried out in relation to the
Ce(1V) oxidation of each of Cr(H~0)4Cz04+ionIz2
cis-Cr(HzO)z(Cz04)2-ionJ6 and C r ( C ~ 0 ~ ) 3ion ~ - (this
study). The reactions for these three complexes show
Such a bridging function for oxalate was proposed a different specificities toward different media, implying
number of years ago as a means to interpreting the transition states of differing compositions in relation to
effectiveness of metal ions in catalyzing the racemiza- the components of the solvent electrolyte. It might
tion of the Cr(Cz04)33-ionIz8and more recently direct be anticipated that complexities of the above type
would be minimized, however, as one compares the
redox reactivities of the identically charged and other-
(22) M. W. Hsu and R. M. Milburn, unpublished results. wise similar complexes, Cr(C204)33- and Rh(Cz0~)3~-.
(23) I n t h e 1 114 sulfuric acid medium the principal oxalato species will
correspond t o the formulas as written above. For a discussion of the basicity For 25' and 1 M sulfuric acid we obtain the following
of Cr(CiOa)sa-, see ref 24. values for the rate constants for the forward reaction,
(24) H. Kelm and G. M. Harris,Inovg. Chem., 6, 1743 (1967).
(25) I n 1 111 sulfuric acid the predoinitianl cerium(lV) species is, presurn- eq 15: with M = Cr, k15 = 4.0 (A0.5) X 1W-l
ably,*o Ce(S04)3*-. sec-l; with M = Iih, k l h = G . 1 ( = t l . O ) X l W 4 M-'
(26) T. J. Hardwick and E. Robinson, Can. J . Chem., 29, 828 (1951).
(27) H. Taube, J . A m . Chem. Soc., TO, 1216 (1948). (29) N. F. Curtis, J . Chem. Soc., A , 1584 (1968).
(28) J. H. Carter, Doctoral Thesis, University College, London, England, (30) J Aggett, I. Mawston, A. L. Odell, and B. E. Smith, ibzd., A , 1313
1956. (1968); also references therein.
2206 M. W. Hsu, H. G.KRUSZYNA,
AND K. M. MILBURN Inorganic Chemistry

sec-I. Thus, for the given conditions there is an ap- tion enhance the reaction rate, and on this basis it
proximately 70-fold difference in the rates observed appears reasonable to ascribe specific roles to Hf and
for the two metal complexes. in the reaction^.^^ The quantitative effects of
I n accord with the above discussion we suggest that the medium changes are, however, rather different
reaction 15, rather than being a concerted process, is for C r ( C ~ 0 ? ) 3 ~and - Rh(Cz04)s3-. This observation
likely to involve equilibrium complexation reactions illustrates the important point that the activated com-
which precede the rate-determining redox step. In plexes for the initial and rate-determining redox re-
simple terms a likely set of initial reactions is actions (reaction 19) are not identical.
If one accepts the view of Hardwick and Robinson2G
that the principal species of Ce(1V) in these media is
Ce(S04)32-,the inverse dependences on sulfate ion are
consistent with the view that cerium is coordinated to
oxalate in the transition state for the rate-determining
step. Possibly the greater sensitivity which the Rh-
(C204)33-reaction exhibits toward sulfate implies a
transition state in which bridging oxalate is still chel-
ated to the rhodium center, while for the Cr(Cz04)33-
Reaction 18 is written to represent the rapid equilib- reaction the transition state could involve bridging
rium formation of a binuclear complex, as discussed oxalate coordinated to the chromium through a single
above for Cr(C204)33-,in which the cerium is immedi- oxygen, as suggested above. It would be expected that
ately adjacent to oxalate which remains coordinated to efficient coordination of cerium t o an already "fully
M. Reaction 19 is the rate-determining electron-trans- coordinated" (i.e,, chelated) oxalate would be signifi-
fer process. Both M(C204)32-.Ce111and 1cI(C~04)3~-, cantly aided by more extensive loss of sulfate from the
formed by rapid equilibrium dissociation of this species, cerium(1V). This possible difference in the nature of
would be expected to react rapidly with additional the transition states for the Cr(C204)s3- and Rh-
Ce(1V) to consummate the redox process. (C2Od)a3- reactions is in accord with the relative labil-
It cannot be assumed automatically that the greater ities of the Cr-0 and Rh-0 bonds in the two com-
reactivity of Cr(Cz04)33-compared to Rh(C204)33- a t p l e ~ e s . ~ , The ~ ~ role , ~ of~ hydrogen
~ ~ ~ ~ion ~ is~ even
- ~ ~
25' is a reflection of the greater ability of the chromium more difficult to ascribe with assurance, but it seems
center to facilitate electron transfer in reactions of type reasonable that protonation of sulfate, coordinated to
19. A simple comparison between chromium and the cerium, would facilitate electron transfer to the
rhodium in these terms would be justified only if the cerium (IV)center.
compositions, stabilities, and structures of the binuclear From the temperature dependences of the Rh-
complexes (formed according to reactions of type 18) (C204)33--Ce(IV) and direct Cr(Cz04)33--Ce(IV) redox
and the compositions and structures of the transition reactions in 1 M sulfuric acid we calculate the following
states (for reactions of type 19) are identical or very apparent enthalpies of activation: rhodium, AH" =
similar. Although from a simple electrostatic point of 35.0 (f1.5)kcal mo1-I (at 35"); chromium, A H * =
view close similarities for chromium and rhodium would 17.6 ( h l . 5 ) kcal mol-1 (25'). For each reaction the
seem reasonable, since the Cr(Cz0&3- and Rh(Ca04)33- A H * values exhibit temperature dependences which
ions must be of quite similar size, this simplicity is not are likely to be an indication of changing compositions
completely borne out by the results of our studies in in transition states with temperature. For comparison
the several media. Thus, as one moves from medium purposes we calculate for the one-electron oxidations
11 to I, which involves doubling the hydrogen ion con- a t 30": chromium, AH* = 20 kcal mol-', AS* = +B
centration while [SO4+] remains constant (at the same eu; rhodium, AH* = 32.5 kcal molw1,AS" = +35 eu.
time [HS04-] increases by -30% and the ionic The measured AH" and AS" values will include con-
strength increases by -l2yO), the observed rate con- tributions from preequilibrium reactions, and these
stant for the direct Ce(1V) oxidation of Cr(C~04)3~- contributions will presumably differ for the Cr ( C Z O ~ ) ~ ~ -
( k b ) increases by a factor of 1.85; a similar medium and Rh(C204)33- systems. Because of the nonidentity
change for Rh(C20&a- results in an increase in the ob- in compositions of transition states, detailed interpreta-
served rate constant (&) by a factor of 5. Also, as one
moves from medium IV to 111, which involves essen- (31) It might be argued t h a t t h e differentkinetic behaviors f o r t h e s e v c i i ~ l
tially halving the sulfate ion concentration while [H+] media may be due princiually t o differences in ionic strength. This view i i
less appealing than ascribing a specific role t o H ' and SOax- in view of t h c
remains constant (at the same time [HSOd-] de- types oi observations made in extended studies of medium effects for the
creases by -18% and the ionic strength decreases by cerium(IVj-czs-Cr(H2O)r(C~O~jx- systems and because changing the medium
lends t o significant quantitative differences in the kinetic liehador exhibited
-33%) k b for Cr(C204)33-again increases by a factor hy C'r(GOr)+-and Kh(CpOd37 .-.
of 1.85; whereas a similar medium change for Rh- (?Z) I,. Damrauer and I<. nl. hlilburn, J . Ain. i ' h e i i i . Soc., 90, :3884 (1968)
(33) A. L. Odell, R. m7. Olliff,and F. B. Seaton, J . Chein. Sx.,2280 (1965).
( C Z O ~ )results
~ ~ - in an increase in the observed rate (34) T h e formulation of K3Iih!C20dj.4.5H20 proposed by Poi-te, el a l , Y
constant k3 by a factor of 3.7. For both complexes, would suggest, however, t h a t hali of t h e rhodiums would initially contain il
monodentate oxalate.
medium changes which involve increases in hydrogen ( 3 6 ) A. L. Porte, H. S. Gutowsky, and G. M. Harris, J . Chew. Phys., 34, 66
ion concentration or decreases in sulfate ion concentra- (1961).
VoL 8, No. 10, October 2965, 2207
~~U~~-~IS(TRIET~Y~P~O~PHINE)CHL6RO(MESITYL)PLA TI~UM(~~)

tion of these values is scarcely possible. However, the tailed consideration of electrode potentials for Cr-
values are consistent with the suggestions that AH* for (1V)-Cr(II1) and Rh(1V)-Rh(II1) couple^,^^-^^ while
the rhodium system may include a significant contribu- not leading to unambiguous conclusions, suggests that
tion from energy terms of a type common to substitu- the one-electron oxidations of the Cr3+and Rh3+ metal
tion a t the metal center, such as rhodium-oxygen bond centers are unlikely to be particularly favorable from a
stretching, and that the difference in the AS* values thermodynamic standpoint. Free oxalate is efficiently
for the two reactions could in part be a reflection of a oxidized in sulfuric acid by both Ce(1V) and Mn(III),
lower charged transition state for rhodium, with con- however, and it seems to us more likely that the initial
sequent greater freedom for solvent, and in part an one-electron oxidations of Cr(Cz04)33-and R h ( C ~ 0 ~ ) 3 ~ -
indication for chromium of rather specific geometric will involve oxidation of a coordinated oxalate to a
requirements for the transition state. If it were not radical ion, Undoubtedly coordination to the chro-
for the relatively large positive value of AS* for the mium and rhodium centers would polarize the oxalate
Rh(C20Js3--Ce(IV) system, this reaction would pro- electrons and make i t more difficult, in both a ther-
ceed many orders of magnitude more slowly than the modynamic and a kinetic sense, for an external oxidant
corresponding Cr(C~04)3~- reaction. to be effective. For identical coordination such polar-
It is relevant to inquire if the components ization should be greater in the case of rhodium where
M(Cz04)32-,produced on electron transfer to the Ce- metal-ligand bonds are likely to have more covalent
(IV), are most appropriately described as complexes in character. The difference in polarization would cer-
which M remains in oxidation state + 3 (;.e., d3 or de tainly be greater still if bridging oxalate is chelated to
complexes, each containing a coordinated oxalate the rhodium and coordinated to the chromium by only
radical ion) or complexes where M is in the oxidation a single oxygen. Also, for an electron transfer from an
state +4 (Le., d2 or d6 complexes). These distinctions oxalate, as for transfer from an inert metal center, some
are chemically meaningful, although, for the transient prior metal-oxygen bond adjustment may be required
species under consideration, not easily made. 36 De- which would involve larger energy requirements in the
case of the rhodium. In view of these likely contribu-
tions the observed order of reactivity Cr(C204)33- >
( 3 6 ) It is recognized t h a t in some cases the oxidation s t a t e of a central Rh(C204)33-is not surprising.
atom in a complex may be difficult t o specify in other t h a n a n arhitrary way.
This situation arises especially where electrons may be considered t o be
delocalized over a large p a r t of the complex. F o r most isolable complexes of (37) F. P. Dwyer and H. N. Schafer, J . Puoc. Roy. Soc. N . S. Wales, 81,
chromium and rhodium there is little ambiguity in the oxidation s t a t e of t h e 294 (1948).
metal, b u t t h e situation is less clear for highly reactive intermediates, such (38) J. Y. P. Tong and E. L. King, J . A m . Chem. Soc., 82, 3805 (1960).
as Cr(C*Oa)az-, etc. In certain cases direct distinctions between extreme (39) E. A. WI.Wetton and W. C. E. Higginson, J . Chem. Soc., 4890 (1965).
possibilities are, a t least in principle, feasible. Thus t h e number of unpaired (40) J. Feldman, R. S. Nyholm, and E. Watton, ibid., 4724 (1965).
electron spins for Cr(CiO6)aZ- would depend on whether it is a complex of (41) C . S. G. Phillips and €2. J. P. Williams, “Inorganic Chemistry,” Vol.
Cr(IV) or of Cr(II1) associated with a n oxalate radical ion. 11, Oxford University Press, London, 1966, Chapter 23.

CHIMICA FISICA,
CONTRIBUTION FROM ISTITUTO D I
DI MESSINA,MESSIXA98100, ITALY
UNIVERSITA

Substitution Reactions of the Sterically Hindered

trans-Bis (triethy1phosphine)chloro(mesityl)platinum(I I)
BY G. FARAONE, V. RICEVUTO, R. ROMEO, AND M. TROZZI

Received January 20, 1969

Kinetic data for substitution reactions of trans-Pt(P(CzH5)8)z(ms)C1
(ms = mesityl) with various nucleophiles in methanol and
dimethyl sulfoxide are reported. The results are discussed in terms of a balance between steric factors and electronic struc-
ture in determining the reaction mechanism. Depending on the solvent and nucleophile, the kinetic behavior is similar
either to that found for “pseudooctahedral” complexes or to the associative mechanism typical of square-planar substitution
reactions.

Introduction energy, available for a u bond in forming the transition

In substitution reactions of low-spin ds systems, state.
both steric factors and electronic structure favor an It has recently been pointed out that in sterically