THE KINETICS OF THE OXIDATION OF
SUBSTITUTED 2,2' -BIPYRIDINE AND
1,10 - PHENANTHROLINE COMPLEXES 0F
IRONGI) WITH PEROXYDISULFATE IONS

ThesIs for the Degree of Phil).
MICHIGAN STATE UNIVERSITY
SAROJA N. RAMAI'I
1968

 

 

 

 

,rhc.’\3

 

This is to certify that the
thesis entitled

THE KINETICS OF THE OXIDATION OF SUBSTITUTED

2 , 2 ' -BIPYRIDINE AND 1 , 10-PHENANTHROLINE COMPLEXES

OF IRON(II) WITH PERQXEDISULFATE ION
presente g

Saroja N. Raman
has been accepted towards fulfillment

of the requirements for

PhoDo degree in Chemistry

(fox—QB, MR ,

Major professor

Date July 19, 1968

0-169

. “‘ BUIIK BINDER‘I’ INC.

UIRARY BINDEHS

 

ABSTRACT
THE KINETICS OF THE OXIDATION OF SUBSTITUTED

2.2'-BIPYRIDINE AND 1,10-PHENANTHROLINE COMPLEXES
OF IRON(II) WITH PEROXYDISULFATE IONS

by Saroja N. Raman

The kinetics of oxidation of 2,2'—bipyridine, 4,4'-
dimethyl-2,2'—bipyridine, 1,10-phenanthroline, S-chloro-,
5-nitro- and 5-methyl—1,10-phenanthroline complexes of
iron(II) with peroxydisulfate have been studied. The

complexes are oxidized according to the mechanism

K _ k
e > [Fe(II)°82082 ] 1

 

 

2.—
Fe(II) + 5208 > [FGIIIII +

k_1

2— -0
so4 1+ so4
k2

2—
fast> Fe(III) + so4 .

Fe(II) + so4-°

The reverse step is formulated as a bimolecular step
involving the Fe(III)-SO42_ ion pair and sulfate radical
ion. The forward rate constants, the equilibrium constants
and the activation parameters are reported.

The formation of sulfate radical ion in the rate de—
termining step of the reaction has been proved from the
study of the effect of alcohol on the rate of oxidation of
the t£i§f(2,2'-bipyridine)-iron(II) complex. At high
concentrations of sulfate ions, the complex cations form
ion pairs, which seem to be the principal species under-

going oxidation under these conditions.

Saroja N. Raman

There is a reasonable linear relationship between the
free energy of activation and the standard free energy of
the rate determining step, for these complexes. Such a
relationship is also found between the pKa and the Hammett
substituent constants of the ligandsand the logarithm of

the forward rate constants.

THE KINETICS OF THE OXIDATION OF SUBSTITUTED
2,2'-BIPYRIDINE AND 1,10-PHENANTHROLINE COMPLEXES

OF IRON(II) WITH PEROXYDISULFATE IONS

BY

Saroja N. Raman

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Chemistry

1968

ACKNOWLEDGMENTS

The author is very grateful to Professor Carl H.
Brubaker, for his continued assistance and encouragement
throughout the work, which made this thesis possible.

No words are adequate to express her deep gratitude to her
husband K. V. Raman and to her daughter Kalpana, for their
patience and understanding.

The financial support from the Atomic Energy Commis—

sion is gratefully acknowledged.

ii

II.

III.

IV.

V.

VI.

TABLE OF CONTENTS

INTRODUCTION . . . . . . . . . . . . . .
HISTORICAL . . . . . . . . . . . . . . .
EXPERIMENTAL . . . . . . . . . . . . . .
A. Preparation of Reagents . . . . . .
B. Analytical Methods . . . . . . . . .
C. Procedure . . . . . . . . . . . . .
D. Errors . . . . . . . . . . . . . . .
THEORETICAL . . . . . . . . . . . . . .
RESULTS . . . . . . . . . . . . . . .

Determination of the Rate Constants . .
A. Effect of Alcohol on the Reaction .
B. Effect of Sulfate Ions . . . . . . .

C. Activation Parameters . . . . . . .
D. Relationship Between AG* and AGO. .

E. Relationship Between the Rate Constants

and the pK s of the Monoprotonated Ligands

and the Hammett Substituent Constants

F. Errors . . . . . . . . . . . . . . .
DISCUSSION . . . . . . . . . . . . . . .
A. Mechanism . . . . . . . . . . . . .
B. Role of Alcohol . . . . . . . . . .
C. Mode of Electron Transfer . . . . .
D. Activation Parameters . . . . . . .
E. Effect of Substituents . . . . . . .
LITERATURE CITED . . . . . . . . . . . .

APPENDIX . . . . . . . . . . . . . . . .

iii

Page

52
52

56
56
64
66
67
68
7O

76

TABLE

II.

III.

IV.

V.

VI.

VII.

VIII.

LIST OF TABLES

Molar absorptivities of the substituted
1,10-phenanthroline and 2,2'—bipyridine

complexes of iron(II) and iron(III) . . . . .

Dependence of the second order rate constants
on the concentration of peroxydisulfate

Rate constants for the complex perchlorates

Forward rate constants and equilibrium

constants . . . . . .
Values of the ratio

Rates of reaction,

1:11.
R2

in the presence of excess

tris—(2,2'-bipyridine)-iron(II) sulfate . . .

Effect of alcohol on the rate of the reaction

Effect of sulfate ion concentration on the

rates . . . . . . . .

Activation parameters

iv

Page

20

32

34

39

42

42

44

45

50

FIGURE

1.

10.

11.

LIST OF FIGURES

Graph of 1/kOb versus peroxydisulfate con—
centration for tris-(2.2'-Bipyridine)—iron(II)
sulfate . . . . . . . . . . . . . . . . . .
Graph of l/kOb versus peroxydisulfate con-
centration for tris-(4,4'-Dimethyl-2,2'-bi-
pyridine)-iron(III sulfate . . . . . . . .
Graph of l/kOb versus peroxydisulfate con-
centration for tris-(l,10-Phenanthroline)-
iron(II) sulfate . . . . . . . . . . . . .
Graph of 1/kOb versus peroxydisulfate con-
centration for tris—(5-Methyl-1,10-phenan—
throline)-iron(III sulfate . . . . . . . .
Graph of 1/kOb versus the peroxydisulfate
concentration for tris-(5-Nitro-1,10—phen-
anthroline)—iron(III sulfate . . . . . .
Graph of 1/kOb versus the-peroxydisulfate
concentration for tris-(5-Chloro-1,10-phen—
anthroline)-iron(III sulfate . . . . . .
Probable occurrence of the reverse reaction.
Graph of -{[Fe(II)]t - [Fe(II)]eq] versus
time for tris-(l,10-phenanthroline)-iron(II)

sulfate . . . . . . . . . . . . . . . . . .

Effect of sulfate ion concentration on the
rate for tris-(2,2'-Bipyridine)-iron(II)
sulfate . . . . . . . . . . . . . . . .

Arrhenius plot. Graph of log k versus l/T

f

Relation between the free energy of activa-
tion and the standard free energy change of
the oxidation-reduction step . . . . . . .

Relation between the forward rate constants
and the pKa's of the ligands . . . . . . .

V

Page

36

36

37

38

38

38

41

47

48

51

53

LIST OF FIGURES (Cont.)
FIGURE Page

12. Relation between the forward rate constants
and the Hammett substituent constants (C)
of the ligands . . . . . . . . . . ... . . 54

vi

INTRODUCTION

There has been extensive study during the past fifteen
years in the field of kinetics and mechanisms of inorganic
reactions in solution. The reactions studied can be grouped
into two major classes: (1) Ligand substitution reactions12
and (2) electron transfer reactions which include both
oxidation-reduction and simple electron exchange processes,
which do not involve a net chemical reaction.3'22'42'52'94'95

A great majority of the early work was concerned only
with the evaluation of stoichiometry and thermodynamics and
it is only in the last fifteen years that the area of the
mechanisms of reactions has received serious attention. The
mechanisms are evaluated from an experimental knowledge of
the form of the rate law for the reaction in question and
also from the knowledge of other factors such as the influence
of added substances, dielectric constant and ionic strength
of the medium on the rate of the reaction.

But the more fundamental question of just how the elec-
trons are transferred from reducing to oxidizing agents, in
reactions which take place in solutions is yet to be answered
fully.22 Much of the current work in the field of oxida-
tion-reduction is concerned with an attempt to answer this
question.

A large body of data from the rates of reactions, thermo-
dynamic parameters and other important evidence concerning
the participation or nonparticipation of certain species in

1

2
the reaction, has lead to the postulation of three possible
pathways for effecting electron transfer in reactions in
solution.3:22:42:52r94l95
Effective transfer of electrons may result from an
atom or group transfer such as is found in one path for the

exchange reaction35

Pc13 + P*c15 -——> P*c15 + P*c13 (1)

via P*c15 -—> P*c13 + c12 (2)

A sodium atom transfer has been proved by E.S.R. measurement
in the case of electron transfer to benzophenone from its

sodium ketyl.1 Similarly an H atom transfer is sup-

posed to be involved in the exchange between Fe2+ and
Fe3+ in water.57
(H20)5FeII-OH----O-FeIII(H20)5 (3)
H H

There seems to be no doubt that atom, radical or molecule
transfer does occur in some systems and accounts for the
change in oxidation states of the participating species.
But it is not clear how general such a mechanism may be for
solution reactions.

More prevalent are the other two models for electron
transfer designated broadly as the "outer sphere“ and "inner
sphere" mechanisms. In reactions which are supposed to pro-
ceed by an outer sphere mechansim31'89'93'106 electron
transfer occurs through an extended activated complex or

outer sphere complex, in which the first coordination shell

'3
of each metal ion is presumably intact. The evidence for
this type of mechanism is usually the rate law correspond-
ing to an activated complex containing all the ligands in
the first coordination shells of both central atoms [e.g.
The rate law for the electron exchange between Co(II) and
Co(III) complexes of ethylenediamine is of the form: Rate
of exchange = k [CoII(C2H8N2)3][CoIIHC2H8N2)3]} and/or
the demonstration that electron transfer is faster than sub-
stitution into the coordination shell of either of the re-
actants. For these systems, orbital overlap for the two ions
is small, so that the frequency of electronic transition is
small and there is no substantial binding between the two
exchanging centers. The reactions in this class cover a

wide range of rates from k = 10"8 M-1 sec-1 for the elec-

III(

tron exchange between CoII(NH3)6 and Co NH5)6 to k =

_ —1 .
105 M 1 sec for systems like Fe

II(Ph)3 and‘ FeIII(Ph)3§7:93
In general,metal ions surrounded by unsaturated or large
polarizable ligands such as orthophenanthroline, bipyridine,
cyanide etc. exchange electrons faster.

In the case of inner sphere reactions, electron trans-
fer is preceded by substitution into the coordination shell
of one of the ions with the formation of a bridged species
in the transition state or even an intermediate51 in which
the two metal ions are linked by a common ligand. In such
cases, there is a strong interaction of the orbitals and

interpenetration of the coordination spheres. A significant

body of work has been done by Taube and coworkers Mdth

4

IIIX plus reducing agents such as Cr(II) in aqueous

(NH3I5C0
solutions, where X is H20, OH-, Cl_, OAc-, (H-succinate),
(H-phthalate), (Me—fumarate), etc. Cr(III) and Co(III)
are inert and in many cases they have been able to prove

the formation of a bridged intermediate by the presence of

the X group in the coordination Sphere of the Cr(III)

formed during the reaction.

III III

+ .
(NH3)5CO x + or“ + 5H ——> Cr x + Co2+ + 5NH4+ (4)

The rates of inner sphere reactions are very sensitive to the
nature of the bridging group, reflecting the essential role
of the latter in the electron transfer process. Thus there
is a 106 fold variation in the rate of oxidation of Cr(II)
by Co(NH3)5X, when the nature of bridging ligand is varied.
In the outer sphere mechanism, transfer of electrons
proceeds through a quantum mechanical barrier penetration.
Such systems lend themselves to an exhaustive theoretical
analysis and Marcus73.74.75.77 has successfully derived a
quantitative expression for the a priori prediction of the
rate constants in such reactions. Other theories include
one by R. J. Marcus, Zwolinski and Eyring',’8 in which an ex-
perimental system is necessary for the calculation of param-
eters. Hush58 has derived an adiabatic theory which gives
essentially the same result as that of Marcus. Review of
the various theories has been made by Marcus?6
The theoretical rate constant according to Marcus' theory

can be obtained as follows:74

5

-AG*/RT
k = Ze (5)

where k is the rate constant, AG* is the free energy of
activation, Z is the collision frequency of (hypothetical)
uncharged species in solution. AG* is comprised of three
different energy terms: Standard free energy of the oxida-
tion—reduction step, the coulombic interaction of the ionic
charges of the reactants and products, and the energy for
the rearrangement of the inner and outer spheres of the

charged reactants. (AG*==AG*-2.81(cal)

AG* = w + mzx (6)

where

AGO + wp - w

m = -[ + )\ ] (7)

NH‘

where w is the work needed to bring the two reactants
together and wp the work needed for separating the products.
AG0 is the standard free energy of the elementary electron
transfer step in the electrolytic medium under considera-
tion. A is the reorganization term and includes the effect
of changes in bond distances and bond angles in the inner
coordination sphere of each reactant and the solvent re-

organization around the charged species.

 

l = 7\outer + 7\inner
A depends on the size and shape of the reactants.
outer
_ 1 1 .l 1 _ 1 2
x. - <——2a1 + 2% r><—n, —D We) (8)

k.kP o 2
A = Z ‘41—]— (Aq .) (9)
j kj + k? 3

where a1 and a2 are the radii of the spherical particles
undergoing reaction, r is the mean distance between the
centers of the activated complex (a1 + a2). n and DS

are the refractive index and static dielectric constant
respectively, (Ae) is the number of electrons transferred,

k., k? are the force constants of the jth vibrational

J
coordinate for a species as reactant and product respective-
ly, Aq:_ is the change in the bond distance in the inner
coordination sphere of each reactant. To determine hi .
the summation is over the jth normal modes of each react—
ant and over all bonds involved in a particular mode.

If AGO + wp - w * w + wp A
A

o
is small, AG z—T—+—+§§__

4 2 (10)

This is the key expression around which interest has
centered for the past ten years. If for a series of reac-
tions the coulombic and reorganization energy are constant
and the only parameter varied is AGO, then a plot of AG*
versus AGO should be a straight line with a slope of 0.5.

Sutin and coworkers have made this type of verification
of Marcus' theory. They have studied the oxidation of
Fe(II) phenanthroline complexes with Ce(IV)25:36, and Mn(III)34
and the reduction of Fe(III) complexes with Fe(II) and have
obtained a slope41 of 0.5 for the plot of AG* versus AGO
which has proved as an important experimental verification

of Marcus' theory.

7

The exact nature of electron transfer in the inner
sphere mechanism is not completely settled. The bridging
process may occur with two different results; merely to
hold the reactants together in similar environments and
in proximity until some other process such as electron
tunnelling can occur72, or to serve as a path for electron
conduction53. In either case, the Franck-Condon principle
could be operative and the transfer of electrons would oc-
cur after the environments of the two bridged species has
readjusted to such an extent that coordination number and
bond distances would be intermediate between the initial and
final positions for each species.

Libby"2 favors the idea that bridge mechanisms operate
simply to connect the reactants, allow them to satisfy
Franck-Condon principle and permit tunnelling transfer to
take place. He has rationalized most of the results of
Taube's reactions on the basis of direct transfer, but there
are instances where a facilitation of conduction of elec-
trons may be the consequence of bridge formation. This
mechanism seems obvious at least in the Taube reaction when
X is nicotinamide or isonicotinamidesz. The formation of

III-nicotinamide complex has been proved

an amide-bound Cr
by Nordmeyer and Taube82 which shows that .Cr2+' attadks the
Co(III) complex at a remote amide oxygen. Since the cobalt
atoms are coordinated to the pyridinyl nitrogen, in both

cases the ring electron must have been transmitted through

the pyridine ring. Wang's work97 on the oxidation of

8
cytochrome g_also demonstrates electron conduction through
4,4'-bipyridine bridges. According to Brubaker22 both
mechanisms should be possible depending on the nature of
the reaction and it is unlikely that either mechanism will
be found completely general.

The trig complexes of phenanthroline, bipyridine and
their variously substituted derivatives with transition
metals like osmium, ruthenium and iron form a convenient and
interesting group of compounds from a standpoint of kinetic
studies. They form stable nonlabile complexes which dis-
sociate only slowly in acidic or alkaline mediumE1v17v23r24t33
These complexes have very strong absorption bands in the
visible region of the spectrum, and are ideally suited for
spectrophotometric work. The ligands exhibit a wide range
of pKa values and so the complexes differ in their stabil-
ities and the metal complexes cover a wide range of oxida-
tion-reduction potentials.

The nonlabile nature of the complexes guarantees that
the inner coordination sphere is held intact during oxida-
tion-reduction reactions and so may be conducive to an elec-
tron transfer through the outer sphere mechanism. Further
there is no change in geometry and only a small change in
bond lengths in the two oxidation states, which should lower
the reorganization energy. Kinetic studies with extensively
v bonded systems like phenanthroline and bipyridine tend
to indicate that a large degree of delocalization of the

electrons in these systems may be favorable to a tunnelling

9
mechanism.52 Hence, the idea of working with the Fe(II)
system seemed very promising and interesting to test
Marcus' outer sphere theory.
The wide range of pKa values of the ligands enables
one to seek a correlation between the rate constants and

o, the substituent constant of the Hammett equation.62

ks
log(i;J = Op (11)

where kS and k0 are the equilibrium constants or the

rate constants for a series of reactions involving the un-
substituted and substituted (meta or para) derivatives of

the benzene ring. 0 is the substituent constant and depends
solely on the nature and position of the substituent. o is
obtained directly by measuring the effect of that substituent

on the ionization constant of benzoic acid in water at 25°.

 

KX-C H COOH
C6H5COOH
d ' ' ' -
where KX-C3H4COOH an KC6H5COOH are the ionization con

stants for the substituted and unsubstituted benzoic acids.
p is a reaction constant and depends on the nature of the
reaction. The Hammett equation provides a quantitative re-
lation between the nature of the substituent and the reac-
tivity of a side chain.

A large variation in the standard oxidation-reduction

potential is ideal in any investigation of Marcus' theory

10
since this is a direct measure of the standard free energy
change of the system.

—AG° - RTan = nFE° (13)

In a series of oxidation-reduction reactions, if the
coulombic and the reorganization energy are held constant,
the energy of activation is directly related to AGO, for
an outer sphere electron transfer. The Fe(II) system
fulfills this condition as evidenced from the work of Sutin
and coworkersgs'34'36'41

In the present investigation, Tl(III) was meant to be
used as an oxidizing agent. Tl(III) is especially inter-
esting because its reduction involves a two-electron trans-
fer and it is not yet established conclusively whether the
chemical reduction takes place gig two, one-electron steps
or by a concerted two-electron transfer, although the elec-
trochemical reduction has been proved to be two one-electron
steps.74 Much chemical evidence, also points to two,
one-electron transfer steps in chemical reactions.7054
Theoretical calculation of the exchange rate of Tl(I) -
Tl(III) from the Marcus theory,seems to agree with this.74
A similar calculation for the Fe(II) —— Tl(III) system,
however, gives a low value for the rate as compared to the
experimental value. It was hoped that the reaction of Tl(III)
with the Fe(II) complexes might lead to some categorical
proof of either mechanism. George and Irvine“8 have re-
ported the rate constant for the oxidation of trigjphenan-

throline)iron(II) with Tl(III) in a 2 M,HC104 medium at a

11
(Pl(III) concentration of 3.5 x 10-2 M, In the present
iJrvestigation, it was found that under conditions of high
arzidity, which is essential for preventing the hydrolysis
cxf Tl(III), the acid dissociation of the Fe(II) complexes
“Has fast enough to compete with the oxidation92. Irvine59
later reported that the oxidation takes place irreversibly
xvhich probably suggests that considerable hydrolysis of the
<:ompound had occurred in his studies. Extensive hydrolysis
\Mas found to occur in the oxidation of E£i§_bipyridine
:Lron(II) complex with Tl3+ in 2 M_HC104.87 Hence, the
Situdy was abandoned and peroxydisulfate was chosen as the
cu<idant since it showed a reasonable rate of oxidation of
tlie iron(II) complexes.

With peroxydisulfate, the oxidation involves breaking
up) of the peroxide bridge and probably no categorical postu-
liation as regards the mode of electron transfer Can be made.
Tflae simplified assumption that (AGo + wp — w)/A is small.
iJi the derivation of AG* = w + wp/Z + A/4 + AGO/2, may
(Dr may not be true. Even if it is, the reactions may not
Eill have similar magnitudes for the A term (see the dis-
Cnassion in the final chapter). Hence, only a trend can be
lJdeed for in the change of AG* versus AGO. Further
<20mplications with the peroxydisulfate system, which are de-
Escribed in the later chapters have added to the difficulty;
Ihowever, a reasonable straight line with AG* versus AG0

has been obtained.

HISTORICAL

Peroxydisulfate ion is one of the strongest oxidizing
agents known in aqueous solution. The standard oxidation-

reduction potential for the reaction:

> 5203'2(aq) + 2e" (14)

 

2804-2(aq)

is estimated to be -2.01 v.2:71 Although it is a two-
electron oxidizing agent, in a majority of reactions $2032—
ion undergoes either a one—electron reduction with the forma-
tion of one sulfate radical ion or a fission of the 0-0
bond to form two sulfate radical ions. The highly reactive
sulfate radical ion may further undergo reactions with the
solvent or with various substrates present in the solution.
In some cases, the one-electron oxidation product of the
substrate may be a reactive intermediate, which may further
react with the solvent or other substrates present in the
solution, or the peroxydisulfate ion itself. The best ex-
ample for this is the reaction of $2082” with Fe2+ ions
in the presence of ethyl alcohol.68

One of the most extensively investigated areas in the
chemistry of peroxydisulfate ion is its use in organic poly-
merization reactions. The sulfate radical ion produced by
reacting $2082- with a metal ion such as Fe(II) is able
to induce oxidations and polymerizations in organic sub-
stances and its presence has been confirmed by the incor-
poration of 35S from labelled peroxydisulfate into polymer

12

13
chains.4°"53'7°I31'91 The polymerization processes find
wide applications. The iron-peroxydisulfate induced poly-
merization is quicker and more efficient than the early use
of peroxydisulfate alone. A very large part of the work
done in the field of peroxydisulfate concerns its use as a
radical generator, often in conjunction with a metal ion to
suit the type of polymerization.

There has been a comparatively meager amount of investi-
gation in the field of oxidations of metal and metal complex
ions by peroxydisulfate in spite of its intense oxidizing
power. One of the reasons is that for many of these reducing
agents, oxidation by peroxydisulfate does not proceed at a
convenient rate around 25° unless a catalyst is present,
even though the free energy change is so favorable. At
higher temperatures, peroxydisulfate itself undergoes de-
composition and introduces serious complications in the sys-
tem.69 Further, the peroxydisulfate oxidations almost always
involve the production of the highly reactive sulfate radical
ion (exception50), which brings about undesirable and un-
resolvable obstacles to the elucidation of the mechanism of
many systems. Kinetics is highly susceptible to influence
by traces of foreign matter in the water and to oxygen and
so extraordinarily painstaking efforts are needed in some
cases to get reproducible and meaningful results. A large
number of early published works on this subject reflect such

anomalies.29'44

14

The peroxydisulfate oxidations are very similar to
the oxidations by hydrogen peroxide, which involve the OH.
radical as the intermediate in many of its reactionsq9:I°°r1°1
In both systems, the peroxide bond is symmetrically cleaved
and all the complexities that arise from the generation of
highly reactive inorganic radicals are common to both of
them.

The unfavorable rates of oxidation of many metals and
complexes with peroxydisulfate necessitate the use of a
catalyst to make them proceed at reasonable rates. One of
the most widely used catalysts has been Ag(I) ion.56:102
Other metal ions such as Cu(II). Fe(II) have also been used
to some extent. The work in this area has been fairly ex-
tensive and has lead to the observation of some simplified
and common trends in the diverse systems.56.1°2

The Ag(I) catalysis of $2082- ion oxidations was first
observed by Marshall?9 Various kinetic investigations show
that for a majority of the reducing agent316I29'32'44'49I64v
65:56:103I10‘:1°5 studied, the rate law is given by

'dlszosg-I _
dt ““= k2[82082 IIA9+I (15)

 

This indicates that the rate of the reaction is proportional
to $2082— and Ag(I) ion concentrations, and is independent
of the nature and concentration of the reducing agent. The
nature of the rate law proves that the rate determining
reaction between peroxydisulfate and Ag(I) ion yields one

or more intermediates capable of rapidly oxidizing the

15
reducing agents in these reactions. Further, when positive
ions are the reducing agents, the second order rate constants

have similar magnitudes and can reasonably be represented as

log k2 = -.193 - .38u1/2 56 ' (16)
The values for the rate constants for uncharged molecules
and negatively charged ions are much larger and do not con-
form to any apparent simple form.

Two different mechanisms have been proposed to explain
the Ag(I) ion catalysis of peroxydisulfate oxidations.
Yost103 proposed the step involving the formation of Ag(III)
ion

2— +

+
+ Ag 3

 

5203 > 25042’ + Ag (17)

followed by the rapid oxidation by Ag(III). Bawn and Mar-
gerison16 proposed that the rate determining step involves

a one-electron oxidation—reduction reaction.

— ++

2' + > 504'“ + 5042 + Ag (18)

8208 + Ag

 

There have been several publications favoring either of them.
According to Wilmarth and Haim102 the chemical prOper-

ties of the higher oxidation states of silver are compatible

with either of these. In highly acidic media, Ag(II) may

be readily prepared and magnetic susceptibility and oxida-

tion potential measurementss4v85 indicate that the position

of the equilibrium in

++ > Ag+- _+ Ag3+ (19)

 

2Ag

 

16
is far to the left in HNO3 and HClO4 at concentrations
greater than 4 M, There is evidence that basic Ag(III)
salts are produced when Ag(II) solutions are diluted.63
None of the reactions studied identifies the species formed
in the rate determining step of the peroxydisulfate and
Ag(I) ions, or the reaction intermediates responsible for
the oxidation of the reducing agents. Moreover, the reac-
tion intermediates may not be the same in all the cases and
in some cases,more than one intermediate may be involved in
the oxidation.

Another anomaly in the proposed mechanism is that the
entropies of activation of many Ag(I) catalyzed reactions are
highly negative, which is in conflict with the value of ap-
proximately +20 caloK-lmol-1 expected from the proposed rate
determining step.56 In this connection however, the alternate
inital step suggested by Bekier and Kijowski18 and

Chaltykyan and Beilerian27 may be cited.

+ 2- + 2-
Ag + 5208 {22> [A9 ’5203 I (20)
+ _ .-
[Ag 620.2 1 —> Ag"+ + 250.2 (21)
+ _ _ ...o
or (Ag '52032 > ‘_R A92+ + $042 + 504 (22)

The overall negative entrOpy can be explained because step
21 or 22 will involve a large negative entrOpy.45 .In the
present investigation, a similar step has been proposed for
the Fe(II) complexes and peroxydisulfate. Although the

overall entropy is negative, a positive entropy of about

17

15 (2.2110161

mol-1 has been calculated for the corresponding
step 1.

Chain reactions represent another widely investigated
area in the chemistry of peroxydisulfate ion. These are
mostly in organic systems which are capable of partici-
pating in chain initiation and termination steps. This
subject will not be dealt with any further in this work.

Very few cases of direct oxidation of metal ions and
their complexes by peroxydisulfate have been studied. These
include the oxidation of Fe2+,4°:68 some t£i§_complexescfi‘
Fe(II) and Os(II) with bipyridine and phenathroline deriva-
tive825:5°:61 and ferrocyanide.23v55 The rate determining
step was postulated as

2+ 3+

_ 2- _.
M + $2082 -—> M + so4 + so4 (slow) (23)

804-. + M2+ -—> M3+ + SO4_- (fast) ‘ (24)

A slightly different step has been suggested in this investi-

gation as explained in later chapters.

EXPERIMENTAL
A. Preparation of Reagents:

Reagent grade materials were used throughout. 2,2'—Bi-
pyridine, 1,10-phenanthroline and their derivatives were
obtained from K. and K. Laboratories Inc. ‘The water used
to prepare the solutions was doubly distilled over alkaline
potassium permanganate in a Pyrex vessel and stored in stop-
pered Pyrex flasks. This method was preferred to demineral-
ization by passing through organic resins as the latter
might introduce some organic matter as an impurity. The
presence of even traces of organic matter has been shown to
affect the course of the peroxydisulfate reactions very
seriously and leads to irreproducible and erroneous results.

One such illustration is the presence of alcohol, which
is discussed later in this work.

The glassware used was thoroughly washed with chromic
acid solutions, rinsed with distilled water, conductivity
water and then was air dried.

Standard solutions of peroxydisulfate were prepared
by weighing exact quantities of the reagent and dissolving
them in appropriate quantities of doubly distilled water:
the solutions were prepared fresh daily and were never kept
for more than 10-12 hours. This precaution prevented the
possibility of radical initiation or any other.comp1ication
that might arise from the decomposition of peroxydisulfate.

18

19
The rate of decomposition of the peroxydisulfate under the
conditions of the experiment, however, was very slow and
the results were reproducible within the limits of experi-
mental errors.

2,2'—Bipyridine, 1,10-phenanthroline and their deriva-
tives were used without further purification. The sulfates
of these complexes were prepared by dissolving three equi-
valents of the organic base in water and adding a freshly
prepared aqueous solution containing one equivalent of fer-
rous sulfate. The complexes were formed instantaneously.
The absorptivities were checked spectrophotometrically for
the complete formation of the complex.

The complex perchlorates were precipitated from the
sulfate solutions with sodium perchlorate, washed thoroughly
with water and redissolved. Solutions of sodium sulfate and
sodiumperchlorate were prepared by dissolving the salts in

distilled water.
B. Analytical Methods:

Spectrophotometry was used to determine the initial
concentration of the Fe(II) complexes and to follow the
course of the reactions. The wavelength of the maximum ab—
sorption and the molar absorptivities of the trig Fe(II)
and Fe(III) complexes were checked and were found to agree
with the values reported earlier.41 The values of molar

absorptivities are given in Table I.

20

 

osv.w can mam Ham mcfleflusmgn.m.muassumsanu.e.e

 

 

 

omm.w own man has mcfieflummamu.m.m
oos.HH cos «Hm mam maaaounocmcmamuoa.HuouoHnoum
oom.HH onm can can mcHHoHnucmcmnmuoH.Huouuazum
oom.fia one can smm magaougpamcmnmuofi.Huamnpmzun
ooa.HH osw can mom mcflaounuamamnmnofl.fi
“HchouH AHHHpcouH AHchouH AHHchouH enemas
AanHoE «Eov Ajfiv soaumnomnm
mufl>flumH0QO Hmaoz Edeflxmfi mo sumcoao>m3

 

 

.AHHHVGOHH 0cm AHHVGOHH mo moxoamfioo OGHGAHmQHQI.N.N
paw ocHHonnuamcm£QIOHJ wmusuflquSm ms» mo mofluw>flumnomnm Hmaoz .H magma

21

Solutions of Fe(II) complexes were found to obey Beer's
law in the range of concentrations used. The absorptivi-
ties of the Fe(III) complexes are very low compared to the
Fe(II) complexes and so no correction for their absorption
was used except in the later parts of some reactions where
it was significant.

In the experiments to check the stoichiometry, Spectro-
photometry was used to determine the Fe(II) complex concen-
trations. The peroxydisulfate was determined according to
the following method:67

Potassium iodide was added to a known volume of peroxy-
disulfate and the mixture was allowed to stand in the dark
for 15-20 min. The iodine liberated was determined by titra-
tion with a standard solution of sodium thiosulfate, by use
of a freshly prepared solution of starch as indicator. Very
good endpoints were obtained.

The same procedure was used to check whether any loss
of K28208 had occurred during the reaction. The Fe(II)
complex and K28208 were allowed to react and after the re-
action was complete, KI was added. In this case, however,
the Fe(III) complex also oxidized the iodide and the total
quantity of iodine liberated was determined as before. The
total amount of $2032- before and after the titration were

compared.

22

C. Procedure

Exact volumes of the standard potassium peroxydisulfate
(usually 10-2 M) and Fe(II) complex solution were placed in
the two separate arms of an H-shaped veSsel joined together
at the top. The amount of Fe(II) was adjusted so that it
would be in the concentration range within which Beer's
law is obeyed. The volumes of the samples were adjusted
to be 10 ml so that the concentration of each of the react-
ants after mixing was easily calculated. The tubes were
stoppered with Pyrex caps or polythene stoppers and were
allowed to warm up in a constant temperature water bath to
the required temperature. The temperature of the bath was
measured with a thermometer graduated in 0.10 degree and
the temperature of the bath could be adjusted to within
i 0.050. Water from the bath was circulated by a centrifugal
pump through the water jacket of the spectrophotometric cell
compartment.

The spectrophotometer cells were washed with conductiv-
ity water and after air drying were placed in the thermo-
stated cell compartment. When the cell had attained the
bath temperature, the reactants were mixed by inverting the
tubes a few times while still being immersed in the bath,
and were then quickly transferred to the cells. The time
of mixing was taken as the zero time of the reaction. Read-
ings on the spectrophotometer could be started in most cases

within 30 seconds of mixing.

23

To monitor the course of the reactions, a Beckman
Model DU Spectrophotometer (later an Unicam Model S.P. BOO-B
spectrophotometer) was used. For most reactions, cells with
a 1 cm pathlength were used: however, in.a few reactions
where concentrated Fe(II) solutions were involved cells
with 0.1 cm path length were used. The temperature of the
solutions after the reaction was checked and found to be
unchanged. Duplicate experiments with the Beckman Model DU
and the Unicam spectrophotometers agreed very well. The
extent of the reaction was determined from the value of the
absorbance at various times.

As the reactions were mostly carried out under pseudo-
first order conditions by keeping the $2082- in large excess
with respect to the Fe(II) complexes the pseudo first order
rate constants were obtained from the slope of linear por-
tion of the plot of -ln (absorbance) versus time. The
second order rate constants were obtained by dividing the
first order constants by twice the concentration of 82082..

For each set of concentrations, a total of at least
four experiments were performed and the average rate constants
were used. The reproducibility was excellent for all the
complexes except in the case of 5-N02- and 5—Cl—phenauhroline
complexes. In those cases, the reaction deviated from the
first order plot so early in the runs, that reproducibility

was difficult and a large number of runs were performed for

each concentration.

24

D. Errors

There are several sources of error in this investiga-
tion. Spectrophotometry is limited to a maximum accuracy of
about 1% or more depending upon the magnitude of the absor-
bance reading. Fortunately in this investigation, the ab-
solute value of the initial concentration of the Fe(II)
complexes was not needed accurately for most cases, since
the reaction was followed under pseudo first order condi-
tions by keeping $2082- in excess. Hence an error in the
absorbance readings should introduce a comparatively smaller
error in the rate constants determined from the slope of
the first order plot.

The most serious source of errors in this investigation
is caused by the complicated side reactions discussed in the
DISCUSSION section. This is especially true for the oxida-
tion of 5-nitro— and 5-chloro-phenanthroline complexes.
Although the measurement of absorbances was started in less
than 30-40 seconds after mixing, and the rate constants
calculated from the linear portion of the first order plot,
the rate constants may be subject to unestimable and perhaps
large errors, in view of the deviation from the regular
first order plot very early in the reaction.

In the other cases (bipyridine, phenanthroline, di-
methylbipyridine, 5-methyl phenanthroline) however, the
deviation occurred only after considerable time interval

and the reproducibility of the results over a period of about

25
two years using different reagents, different spectrophotom-
eters and water baths, gives one confidence in the results.
The temperature of the water bath could be controlled
to within i 0.050 and the errors caused by the variations

of temperature could be considered negligible.

THEORETICAL

The determination of the rate law for simple reactions
involves the use of some direct graphical or mathematical
procedures available in many of the standard text bookséll9I45
If however, the system under study is more complex and in-
volves opposing, concurrent, consequent reactions or forma-
tion of stable intermediates prior to reaction etc., elemen-
tary methods of determining the rate law may be impractical
or impossible. In such cases, the data are fitted to equa-
tions for the various possible rate laws to determine which
one is obeyed. Additional information about the reaction in
question, such as attainment of equilibrium, concentrations
at equilibrium, nature of influence of some added substances
(inert or participating in the reaction) etc., can be of
great help in narrowing the search for the correct rate law.

In the present investigation, studies were conducted
under pseudo first order conditions by keeping a large ex-
cess of $2082. with respect to the Fe(II) complex. The
concentration of 82082. was varied from 1 x 10-3 Mgto
5 x 10'.3 M_ and that of the complexes was maintained at

5-7 x 10"5

M. for most of the experiments. A few experiments
with higher concentrations of Fe(II) and 82082. were per-
formed. The data was analyzed as first order rates:

- 933911 = 2kztszosz'1 [Fe(II)] <25)

26

27
Since [32082—] is essentially constant in each experi-

ment, it can be written as a pseudo first order reaction

.. .illgglll = kéIFe(II)I (where k; = 2k:[82032-I) (26)

and

Fe II __ I
1“ ‘Irzifiiio ‘ W (27)

k; is obtained as the slope of the graph of

Fe II .
1n [Fe II Io versus time.

* k2
k. = ~--————-- (28)
2—
2[3203 ]

In this study, -log of the absorptivities of the solu-
tion at various times were plotted against time, because
the concentration of the Fe(II) complex is proportional to
the absorbance.

[Fe(II)] = absorbance/Molar absorptivity.

slope x 2.303
2tszosz'1

 

*
k2 was evaluated from

The second order rate constants so obtained showed an
inverse dependence on the $2082_ concentration and the
values could be fitted to a slightly modified form of the
second order reaction, in which the formation of a loose
complex or ion pair between the Fe(II) and 82082- ions
is assumed to form before the oxidation. The proposed

mechanism is:

28

2- Ke 2- k1
Fe(II) + $208 ———> [Fe(II) - $203 ]

<———

 

> Fe(III) +

<_

-1

SO42— + 304" (29)

k2
fast

 

_. 2..
so4 + Fe(II) > Fe(III) + so4 (30)

The reverse step with rate constant k_1 is proposed to
explain the deviation from the second order conditions, as
eXplained in the DISCUSSION section.

If k2 >> k1 and excluding the reverse step for ini-
tial stages of the reaction, the reaction rate is

_ Sigélll = 2k1[Fe(II)'52032-] = 2k1KeIFe(II)IISzoez-I (31)

[52082-1 is constant under experimental conditions. Inte-

gration gives

 

 

 

lee
ké*==kob = - 2_ [Ref. 43] (32)
1 + Kelszos I
2—
-—l— = 1 + Ké[8208 ] (33)
kob k1Ke lee

lee is the forward rate constant of the reaction. A plot

of 1/kOb versus [82082-1 gives the intercept = 1/kf =
1/k1Ke and slope = Kekf from which Ke is calculated.

If the reverse step involving the reaction of Fe(III) -

804 ion pair and 804-. is incorporated, the rate law is

29

- Sigélll = kIKeIFeIIIIIISzoez-I' k-1[Fe(III)°SO42_][SO4-o]
+ k2[Fe(II)][SO4-'] (34)

If a steady state approximation is assumed,

k1KeIFe(II)I[52082-I

 

 

" = 35
[804 155 k2[Fe(II)] + k-1[Fe(III)] (
If k_1[Fe(III)] << k2[Fe(II)I'
[so."lss = kiKelszosz-I/kz (36)
Substituting in (34)
k“'1
§§§é531-= ZkiKeIFe(II)I[52082-I‘ §;—'(kiKelsaosz-IIFeIIIIII
= ZKIFe(II)Io - ZIE + k:1][Fe(III)] (37)
where 2_
_. k1Ke[8208 I
K = _
1+Ke[32082 ]
and
I k 2-
k_1 = 1/2 Ezl'(k1Ke)[5208 1. [Fe(II)]0 = initial con—
2
centration of Fe(II) complex.
At equilibrium.
RIFe(II)]o = (R'+ k:1)[Fe(III)] (38)
Writing
[Fe(III)]
eq = f (39)

 

[Fe(IIIIo

where [Fe(III)]e is the concentration of Fe(III) complex

q
at equilibrium,

 

kll = KI}: - 1) (40)
or
- 2 . .1. _
k-1/k2 1+Ke[82082_] (f 1) (41)

Hence if the initial and equilibriUm concentrations of
Fe(II) are known, k_1/k2 can be calculated.

kll can be graphically determined as follows, if
the reaction is considered as a simple first order revers-
ible reaction (since the concentration of 82082- is

essentially constant)

k

Fe(II) (ri:> Fe(III) (42)
k
r
where
k-
kf=2kK [502'] andk =——i(k1<)[soz']

The rate law for this reaction, if the initial concentra-

tion of Fe(III) is zero is:

- 1n {[Fe(II)]-[Fe(II)]fi}
[[FeIIIIIO-[FeIII)quI

 

= (kf +kr)t [Ref. 4] (43)

where [Fe(II)], [Fe(II)]o, [Fe(II)]eq are concentrations
of the Fe(II) complex at time .E, zero time and at equili-

brium, respectively. Since kf is known, kr can be

calculated.
k_1 2k
r;- = r;— (44)

The ratio k_1/k2 obtained by the two methods can be com-

pared.

RESULTS
Determination of the Rate Constants

Preliminary studies to check the order of the reaction
showed that the rate of the reaction was first order with
respect to Fe(II) and the rate constant was invariant with
the change in concentration of Fe(II). But when the concen-
tration of Fe(II) complex was kept constant, and the concen-
tration of peroxydisulfate varied, it was found that the
second order rate constants showed a decrease with increas-
ing concentrations of peroxydisulfate. See Table II. The
pseudo first order constants seemed to level off at high
concentrations of peroxydisulfate.

In a few of the initial experiments, the complex per—
chlorates were used. Later, due to their limited solubility,
the perchlorates were replaced by sulfates. The rate con-

stants for the perchlorates studied are given in Table III.
They compare well with the values for sulfates, for the

same concentrations of peroxydisulfate.

The experimental data could be successfully fitted to

the rate law corresponding to the mechanism:

K _ k1
ZEZ> [Fe(II)-$2082 I

 

Fe(II) + 32082" > Fe(III) +

<—

< k_1

SO42- + so4'°

_. k2 2-
so4 + Fe(II) EEEE" Fe(III) + so4

As eXplained in the THEORETICAL section, the values of the

forward rate constants and the equilibrium constants have

31

Table II.

32

Dependence of the second order rate constants
on the concentration of peroxydisulfate.

 

 

 

 

 

 

 

Second Order Rate Constants (M-lsec-l) x 101
[$2082'1x103g; 25° 30° _ 35° 40°
(a) Tris-(2,2'—Bipyridine)-rion(II) sulfate:

1.0 3.35 4.52 5.49 7.1
1.5 2.90 -- 4.78 6.34
2.0 2.69 3.46 4.40 5.5
2.5 2.48 3.34 4.15 5.2
3.0 2.33 3.10 3.74 4.62
3.5 -- 3.06 -- --
4.0 2.14 2.75 3.49 4.06
5.0 1.85 2.60 -- --
(b) Tris-(4,4'-Dimethy1-2,2'-bipyridine)-iron(II) sulfafieloo
0.50 4.67 6.02 6.93 8.10
1.0 3.55 5.15 5.56 6.50
1.5 3.07 3.84 4.52 4.95
2.0 2.75 3.23 4.00 --
2.5 -- 3.10 -- --
(c) Tris-(1,10-Phenanthroline)-iron(II) sulfate x 101
1.0 1.70 2.30 3.36 4.20
1.5 -- -- 3.04 3.36
2.0 1.30 1.92 2.60 3.08
2.5 -- -- 2.40 --
3.0 1.23 1.73 2.24 2.61
3.5 -- 1.60 -- --
4.0 1.01 1.54 1.92 2.28

 

33

Table II. (Cont.)

 

 

Second Order Rate Constants (M—lsec-l) x 102

 

2-
[$203 11:10.31; 25° 30° 35° 40°

 

(d) Tris-(S-Methyl-l,10-phenanthroline)-iron(II) sulfate

1.0 6.83 11.0 12.9 20.0
1.5 -- 9.40 11.6 17.5
2.0 5.80 8.40 11.1 13.3
2.5 -- 7.40 10.0 12.0
3.0 5.00 . 6.78 9.45 11.0
4.0 4.30 6.00 8.40 10.5

 

(e) Tris-(5-chloro-1,10-phenanthroline)-iron(II) sulfate 2

 

 

 

x 10
1.0 9.50 22.1 31.0 51.5
2.0 4.93 9.15 19.5 30.0
3.0 3.60 6.80 12.5 21.1
4.0 2.86 5.10 9.5 17.9
(f) Tris-(5-Nitro-1,10-phenanthroline)-iron(II) sulfate 101
x
1.0 1.72 3.98 5.42 9.00
2.0 0.844 2.03 3.12 5.57
3.0 0.603 1.53 2.45 3.76

4.0 0.456 1.10 1.66 3.03

 

34

Table III. Rate constants for the complex perchlorates.

 

 

2_ Second order
Complex [3208 ] x 103g_ rate constants

(M-lsec-l) x 101

Tris-(2.2'-Bipyridine)- 0.5 3.55
iron(II) sulfate 1.0 3.34
2.0 2.71
Tris-(1,10-Phenanthroline)—
iron(II) sulfate 1.0 1.86
2.0 1.54
Tris-(5-Nitro-1,10-phen- 1.0 1.67
anthroline)-iron(II) 2.0 0.91
sulfate
Tris-(5-Methy1-1,10-phen- 1.0 0.710
anthroline)-iron(II) 2.0 0.575

sulfate

 

35
been calculated graphically by plotting the reciprocals of
the observed second order rate constants against the con-
centration of peroxydisulfate. (Figures 1,2,3,4,5,6)
Least square analysis was done on all graphs and the best
values for slopes and intercepts along with the standard
deviations are given in Table IV.

Another important observation in these reactions is
the deviation from the first order plots after different
intervals of time for the different complexes. The devia-
tion is large in 5-nitro and 5-chlorophenantmxfline complexes,
starting in a few minutes after the reaction was started.
It is less for 5-methyl phemmmmmoline complex, starts only
after about 50% of the reaction for phenanthroline complex
and after 80 to 85% of the reaction for the bipyridine com-
plex. There was very little deviation from the rate equa-
tion for the dimethyl bipyridine complex. This deviation
was also observed by Irvine51.

In the present work, it is tentatively suggested that
a probable reverse step, with rate constant k_1, involving
the reaction of 804-. radical and the Fe(III) - SO42—
ion pair, may at least partly explain the deviation from
second order rate law. A detailed discussion of other fac-
tors that may be important is given in the DISCUSSION sec-
tion.

An attempt has been made in the case of phenanthroline
and 5-methylphenanthroline complexes to estimate the magni-

tude of k_1/k2 by calculation from Ke and concentrations

Figure 1.

Figure 2.

Graph of l/k versus peroxydisulfate

ob.

concentration for tris-(2,2'-Bipyridine)-

iron (II) sulfate.

Graph of l/k versus peroxydisulfate

ob.

concentration for tris-(4,4'-Dimethy1-

2,2'—Bipyridine)-iron(II) sulfate.

36

 

 

92.33 CH. o...

-GzaEEa _~.~ - 425234;..- 25

HH

 

 

 

32:5 :5 e.
-Ez_o_5a_an.~.~.-m_E

Figure 3.

Figure 4.

Graph of l/k versus peroxydisulfate

ob.
concentration for tris-(l,lO-Phenan-

throline) — iron (II) sulfate.

Graph of l/kob. versus peroxydisulfate
concentration for tris—(S-Methy1-1,10-

Phenanthroline )- iron (II) sulfate.

 

 

./

74

 

 

 

as 53
a. 2453 :5 a. ”:53 :H 3..
13238536....-o.._-:z;z-2..m_E 121.81.543.35. .72:

H , H

Figure 5.

Figure 6.

Graph of l/kob.

concentration for

Phenanthroline)

Graph of l/kob.

concentration for

Phenanthroline)

versus peroxydisulfate

tris-(S-Nitro—l,10-

iron (II) sulfate.

versus peroxydisulfate

tris-(S-Chloro-l,10-

iron (II) sulfate.

.3 5’

 

“23:39 .1.
13238535.... $310-2 - 2,:

9H

J

 

om

 

mm

 

n v m N _. o
\0\" IN
\e\° o\o 1?
qu% \\
\° C .
3mm \ -m
g
o 1N.
. .2
LON

 

a.

00

3.533 AH: a...

w
_.l:
O
.8

-Gzzoeizazmzaé...22-2-25

H

39

 

 

on. a mm.m o.e a o.mma oe
em. a m.» o.» a m.mm an
we. a e.» o.m H «.mm on mumeHSm AHHveoua
on. a me.m o.m a m.en mm -Aweaeausean..m.muamsumeanu.e.evumaee
H.m a s.om o.v a o.mm ov
m.m a n.3m o.m a s.om mm
H.m a m.ma H.m a o.nH om mumUHSm AHHveoua
m.H a o.ae H.m a m.mH mm uxmeaaounuemeoseuoa.HIOHSSZISVImaHe
o.m a o.ea o.m a m.ma ow
m.m a H.em ov. a me.s mm
H.m H «.mm an. a sm.e om mommaem AHHveoua
m.H a m.mm om. a ms.m mm -Ameaaoueeememaenoa.HuouoHeoumvumaue
we. a mw.m mm. a Hm.m es
en. a mm.m me. a mm.H mm
mm. a as.m am. a Hm.a om mummHSm AHHveoua
«H. a me.m ma. 0 mmm.o mm -Ameaaoueuememneioa.Huasnumzumvnmaue
mm. a we.m ma. 5 mm.m ov
me. a ma.m SH. a mm.v am
am. a mm.a «H. a as.m om 00mmH5m
we. a mm.m m. a oH.m mm AHHveOHHIAmeaaoueuememeeuoaeavnmaee
oH. a mm.m ea. 5 Hm.m ow
mm. a em.m m. a we.o mm
an. a mm.m no. « mm.m om «Sewage
«a. a mm.m SH. a mm.m mm AHHveeufluxmeaeauseam-.m.mvnmmmw
o m .moo xoamfiou
«IOH x AHISQ M HOH x Auloomulzv M Ema

 

 

mucmumcoo ESHHQHHHSUU can mucmumcoo oumu UHMBHOM.

.>H magma

40

of reagents, and graphically as explained in the THEORETICAL
section. The equilibrium concentration of the iron(II)
complex has been estimated by trial and error to give a good

linear graph of
-ln {[Fe(II)]-[Fe(II)]eq]/([Fe(II)]o-[Fe(II)eq} versus t,

(see Figure 7). The linearity of such graphs is satisfactory
and there is a fair agreement between the directly calculated
and graphically determined values of the ratio k_1/k2.
(Table V).

A few experiments were performed under conditions of
excess t£i§(2,2'-bipyridine)-iron(II) sulfate. These reac-
tions seemed to reach an apparent equilibrium, with less
than two equivalents of the Fe(II) complex oxidized for each
mole of 82082- present. The second order forward rate
constants, calculated from the early stages of these reac-

tions are given in Table VI.
A. Effect of Alcohol on the Reaction

In the earlier stages of this investigation, difficul-
ties arose on account of the insufficient solubilities of
the perchlorates of 5-mdflufl; 5,6-dimethylphenanthroline
complexes etc. in water. Hence the use of an ethyl alcohol-
water mixture of uniform composition was contemplated, as
in the work of Sutin and Fordsmith‘l. It was soon realized
that alcohol was not just providing a suitable medium in

terms of better solubility. A serious retardation by alcohol

Figure 7. Probable occurrence of the Reverse reaction.
Graph of - {~[1:‘e(II)]t - [Fe(II)] eq}

versus time for tris-(l,10-Phenanthroline)-

iron (II) sulfate.

41

_- ._. C
( (4(3) ‘ C'IJ " {(1.1) (LII)
(‘1 O ‘3) KO
0 O O .
H H O (D

 

 

p‘. r" 3’_ufl‘."“ "—' n ‘; .‘a..$.u .w- s'i. V'M It.

I I!

~-'&- “~-r‘li .11" 10'?! I“ «In! A...“ ‘4‘ . r.'.4'\.

l\- to m

[(3:1) €(1{.1)1?.3:JI301'

 

 

‘oy'Uh‘I—VB -a'K—8l‘\’:.l,\. LAI- Rhasfi- ‘J

W“

0 O

C)

KO
r4

(\J

MINUTES

42

Table V. Values of the ratio k-1/k2.

 

 

(k_1/k2) x 101

 

2- .
Temp. [$308 ] Direct Graph-
Complex 0C 3 Calcula- ical
x 10 M .
- tion
Tris-(1,10-Phenanthroline)- 25 1.0 3.2 4.0
iron(II) sulfate 1.5 2.0 1.6
2.5 1.2 2.5
35 1.0 5.2 5.1
1.5 4.0 4.4
2.0 3.3 3.0
Tris—(5-Methyl-1,10—phen- 35 2.5 6.2 5.1
anthroline)-iron(II) 3.0 5.2 4.9
sulfate . 3.5 5.0 4.8

 

Table VI. Rates of reaction, in the presence of excess

tris-(2,2'-bipyridine)-iron(II) sulfate. (25°)

 

 

Initial Concentrations (M x 104)

Second Order Rate Con-
stant (M lsec'1 x 101)

 

tszoaz'l [Fe(II)]
2.0 10.25 2.2
4.0 10.1 2.3

 

43

even when present in amounts comparable to [Fe(II)] and
[$2082-] suggested its chemical interaction in the course
of the reaction (see Table VII). A systematic study of

the reaction by use of a wide range of alcohol concentra-
tions not only clarified the nature of interactions of
alcohol but also provided conclusive evidence for the sug—
gested rate determining step of the reaction which had only

been a conjecture in earlier work.61
B. Effect of Sulfate Ions

The effect of sulfate ions in the region u = 0.1 to
0.4 M; was studied for all the complexes. The sulfate ion
concentration was varied by varying the amount of sodium
sulfate added. The concentration of the complex was main-
tained at 5-7 x 10-5M_and that of the peroxydisulfate ion
at 5 x 10-3M, As seen from Table VIII. the rate constants
showed very small decrease in this region contrary to the
result at lower ionic strength. This phenomenon of little
dependence on ionic strength when sulfate ions are in large
excess can be explained on the basis of the possible forma-
tion of ion pairs between the Fe(II) complex and sulfate
ions, which may be the principal species undergoing oxida—
tion at high concentrations of sulfate ions. Such species
should react more slowly than free ions and show no dependence
on the ionic strength.

An approximate value for the association constant of

tris(bipyridine)-iron (II) and sulfate ion is determined by

44

Table VII. Effect of alcohol concentration on the rate of
the reaction. (Temp. 25°)

Initial Concentrations (M).

 

 

 

[Fe(DiPy)3$O4] [$2082-] [c2350H] k2(gf1sec'1)
x 105 x 103 x 102
6.9 5.0 11.9 1.3
8.1 5.0 8.5 1.2
7.9 5.0 3.4 1.1
6.9 5.0 1.7 1.2‘
7.5 5.0 0.65 1.2
6.9 .0 0.17 2.3
7.9 5.0 0.017 11.5
7.9 5.0 0.0017 19.2
7.7 5 0 0.0 21.0

 

45

Table VIII. Effect of sulfate ion concentrations on the rates.

 

 

 

Initial Concentrations Secgnd
(M) or er
2_ '- 2_ Rate const-
Complex [3203 ] [804 ] 0x101 _and
x 103 102 (M 1390-1)
x x 101
Tris-(2,2'-Bipyridine)- 1.0 0.1 0.06 2.78
' iron(II) sulfate 0.5 0.45 0.15 2.20
1.0 0.4 0.15 2.19
2.0 0.3 0.15 2.07
4.0 0.1 0.15 1.85
5.0 0.1 0.18 1.85
5.0 0.2 0.21. 1.75
5.0 0.5 0.30 1.58
5.0 1.0 0.45 1.43
5.0 2.0 0.75 1.20
5.0 2.8 1.0 0.84
5.0 6.2 2.0 0.84
5.0 9.5 3.0 0.81
5.0 12.0 3.75 0.88
Tris-(1,10-Phenanthro- 5.0 2.8 1.0 0.23
line)-iron(II) sulfate 5.0 6.2 2.0 0.22
5.0 9.5 3.0 0.20
5.0 12.8 4.0 0.20
Tris-(5-Chloro-1,10- 5.0 2.8 1.0 0.19
phenanthroline)- 5.0 6.2 2.0 0.19
iron(II) sulfate 5.0 9.5 3.0 0.22
5.0 12.8 4.0 0.18
Tris-(4,4'-Dimethy1-2,2'- 5.0 2.8 1.0 16.4
bipyridine)-iron(II) 5.0 6.2 2.0 14.0
sulfate 5.0 9.5 3.0 12.0
5.0 12.8 4.0 11.6

 

46
plotting -- kobs versus the concentration of sulfate ions
(Figure 8). The concentration of 52082. was maintained at
5 x 10-3M and that of SO42- varied from 0-1 x 10-2M, The

slope of this graph is kaKa , from the relationship

2..
kob = k0 + kaKa[SO4 1 [Ref. 99] (45)

where kOb is the observed rate constant, kc and ka
are the rate constants for the free and associated Fe(II)

species and Ka = association constant. Assuming k3 =

8.4 x 10-2 Mlsec-1 which is the average limiting value

in the presence of excess of sulfate ions, the value of

Ka = 6.1 x 101 M71. The intercept agrees well with the ex-

perimentally obtained value for k0 at $2082- concentra-

tion equal to 5 x 10—3M,

C. Activation Parameters

 

The activation energy for the forward rate constants

was obtained in the usual way from the relationship

E

ln kr = - §%-+ constant (see Figure 9). (46)

From this the enthalpy of activation, entropy of activation
and the free energy of activation were calculated from the

transition state equation:

(AS*/R - AH*/RT)

—AG*/RT e (47)

.111:
h

Figure 8. Effect of sulfate ion concentration
on the rate for tris—(2,2'—Bipyridine)-

iron (II) sulfate.

47

 

 

OI XIIO’I

Figure 9. Arrhenius Plot.

Graph of log k versus l/T.

f
l. Tris-(4,4'-Dimethy1~2,2'Bipyridine)-

iron (II) sulfate. 2. Tris—(S—Nitro-l,10—
Phenanthroline) - iron(II) sulfate.

3. Tris- (2,2'-Bipyridine)-iron (II) sulfate.
4. Tris-(S-Chloro-l,10-Phenanthroline)—

iron (II) sulfate. 5. Tris—(1,10-Phenan-
throline) - iron (II) sulfate. 6. Tris-

(S-Methyl-l,lO-Phenanthroline) - iron (II)

sulfate.

48

 

E: 8.5sz

 

Tn n.m mm _.m o.n
e / .
III IN —l
:8!
III 10._I
h. I. In!) .
V I/./. IILuI iml
M. / [of X.) .
/. lo; a, -e-
. /QI/ / 6 [III ¢.
0 II I l
/A I ,0 II .II
D D .l. 1N.I 0
/. .D
N </ l./ 1 «I
l/OI /.W/ o
/0’l /@ // IN.
/m/ / lee
\ / -8.
/0 [mo
/ L .

49

k = Boltzmann's constant,

8* = entropy of activation per mole,

h = Planck's constant,

T = absolute temperature

H* = enthalpy of activation per mole,

R = gas constant,

H = Ea - RT (for reactions in solution),
G* = H* - TS*.

The results are shown in Table IX.
D. Relationship between AG* and AG°

The plot of AG* versus AG° of the oxidation-
reduction step is shown in Figure 10. AG° is computed by

combining the oxidation potentials of the couples:

$042. + so4'°

 

> 82032. + e' [-1.6 v (Ref. 102)]

<—

and

> Fe(III) + e . (48)

 

Fe(II) 4___
(The oxidation potentials of the corresponding complexes
are used.)

A fairly good linearity is observed for all except
5-nitro- and 5-chlorophenanthroline complexes. The failure
of some of these points to lie on the line may be due to
experimental errors, as the reproducibility of the rate
constants for these complexes was poor, or it may be due to

some other factors arising from the negative substituents.

The slope of this line is 1.4. This value is in agreement

50

 

 

 

Table IX. Activation parameters.
AS*
AH* AG*

Complex kcal/mole cal°K 1mol"1 kcal/mole
Tris- 2.2'-Bipyridine)10.0 i 1.6 -26.7 18.0 i 1.6
-iron II) sulfate
Tris-(4,4'-Dimethyle 7.64 i 1.74 -29.2 16.3 i 1.74
2,2'—bi yridine)—
iron(II sulfate
Tris-(1,10-Phenan- 12.1 1 1.67 {21.5 20.0 1 1.67
throline)-iron(II)
Tris-(5-Methyl-1.10- 12.2 i 1.66 -24.7 19.6 i 1.66
phenanthroline)-
iron(II) sulfate
Tris—(5-Chloro-1,10- 17.3 i 2.10 - 7.4 19.5 i 2.1
phenanthroline)-
iron(II) sulfate
Tris-(5-Nitro-1,10- 9.10 1 2.1 -27.6 17.3 1 2.1

phenanthroline)-
iron(II) sulfate

 

Figure 10.

Relation between the free energy of

activation and the standard free energy

change of the oxidation-reduction step.

1. Tris-(4,4'-Dimethy1—2,2'aBipyridine)—

iron (II) sulfate. 2. Tris-(2,2'-Bipyridine)-
iron (II) sulfate. 3. Tris-(5-Methy1-l,10-
Phenanthroline) - iron (II) sulfate.

4. Tris-(l,lO-Phenanthroline) - iron (II)

sulfate.

 

. 51

25-

 

 

2320. 04'
_: \o
.7 \
L9
<1

:5-

'00 I0

-AG° Ik.ca|.)

52
with the slope of Irvine's graph61 of log k versus the
overall oxidation potential change. As explained in the
DISCUSSION section, not much importance can be attached to

the value of this slope.

E. Relationship between the Rate Constants and the pKas of

 

the Monoprotonated Ligands and the Hammett Substituent

 

Constants o

 

The graph of log k versus pKa of the ligands41 is
given in Figure 11. A reasonably linear relationship is
obtained at all temperatures studied. Similar straight line
relationships are also obtained for the graph of log k
versus 0, the substituent constants80 of the Hammett equa-
tion62, suggesting that Substituents do influence the rates
of oxidation of the Fe(II) complexes (Figure 12). Such

effects have also been observed by other workers.21,41
F. Errors

The results of the present investigation are subject to
the usual experimental errors arising from the limitations
of the various methods used. Apart from these, the reac-
tions studied are potentially susceptible to a large number
of other complicating factors resulting from the sensitivity
of the complexes and the organic ligands to, hydrolysis and
oxidation respectively, in the slow reactions. These are
discussed in the DISCUSSION section. The formation of the

highly reactive 804-. radicals is another source of

Figure 11. Relation between the forward rate
constants and the pKa s of the

ligands.

53

 

 

0000
0000
Once:
Q’MMN
u<>0<l
DOO<I
<1
CI<>oo
//D/OM
1
g l
o o 0 <2:
«3' N —

z+1601

pka

Figure 12. Relation between the forward rate
constants and the Hammett substituent

constants ( ) of the ligands.

54

 

    

o
Q IO "Adv—Myoc—

ho._n_ 1.5.2.2":

 

2+) 50I

55
complication expecially in the presence of organic ligands.
Added to these is the probable reverse reaction. In view
of these difficulties, it is difficult to assess the extent
of accuracy, although the reproducibility of the results

was good for all except 5-nitro- and 5-chloro-complexes.

DISCUSSION
A. Mechanism

In addition to the reasonably good fit to the rate law of
the proposed mechanism for the reaction of the E£i§_com-
plexes of Fe(II) with the substituted 2,2'-bipyridines and
1,10—phenanthrolines, the factors which could be respons4
ible for the decrease in the rate constant with increasing
peroxydisulfate concentrations, were considered as follows:
the decomposition of the peroxydisulfate itself under the
conditions of the experiment is slow69 and could not be re-
sponsible for the lowering of the rate constants. No oxygen
evolution was observed. Moreover, the cxmermration of the
peroxydisulfate ion in all these experiments is much greater
than the iron concentration and so a slight undetectable
decomposition of peroxydisulfate ion should not alter the
rate constants greatly. Another possibility is that the
sulfate radical formed in the rate determining step could
be involved in the oxidation of the ligand or impurities.
Such side reactions have been anticipated or reported by
some of the earlier workers§°'61'1° Irvine°° reported that
in the oxidation of trig(bipyridine)osmium(II) with peroxy—
disulfate ion, only about 1.5 moles of the Os(II) complex
were oxidized for each mole of peroxydisulfate ion. He
ascribed it to the competition of the metal chelate ions
and reducing impurities for the sulfate radical ion formed

56

57
in the rate determining step. He could, however, recover
all the Os(II)-bipyridine complex by reduction, thus proving
the absence of oxidation of the ligand.

Barb and Baxendale1° reported a considerable oxidation
of bipyridine if it is present during catalytic decomposi-
tion of H202 by Fe(III) ion. The conditions of their ex-
periments and the conentrations were much more drastic than
in the present investigations.

In the present investigation, absence of extensive
ligand oxidation and the oxidation of nearly two moles of
the Fe(II) complex for each mole of 82032., were verified
as follows for the trig(bipyridine)- and tris(dimethyl
bipyridine)-iron(II) complexes: the iron(II) complexes were
recovered after the reaction by reduction. Solutions of
the trig(bipyridine)-iron(ll) complex, were oxidized with a
known excess of peroxydisulfate and the total peroxydi-
sulfate plus iron(III) was determined iodometrically after
the reaction.67 The quantities of sodium thiosulfate needed
after the reaction were nearly the same as those required
for the blanks, containing the same concentrations of’
82082-, used initially in the reactions. This suggested
the oxidation of two moles of Fe(II) complex for each mole
of 82082-, because iodine equivalent to the original peroxy-
disulfate could be formed after the reactions, only if the
peroxydisulfate was-used up in the oxidation of Fe(II) to
Fe(III). Fe(III) formed could liberate iodine, correspond-

ing to the amount of $2082- used in the oxidation. There

58

was, however, considerable difficulty in detecting the end

point in these titrations when excess Fe(II) complex was
used, due to the intense color of the complex.

Irvine61 calculated the value of k0 by assuming an
ionic strength dependence for k2 by plotting log k2
versus 'fh (in the region u - 2.5 x 10—3 to 10"2 M)
and extrapolating it to zero ionic strength. He obtained
slopes ranging from -2.5 in the case of trig(2,2'—bipyri-
dine)-ruthenium(II) to -4.4 for tris(dimethylbipyridine)-
iron(II) complex. In the present investigation, slopes of

—3 to -6 were obtained in this ionic strength region when
K28208 was used. When the ionic strength was varied by
varying the sulfate ion concentration, the plot of log k2
versus Jhd gave a lepe of about -2 at lower concentrations
of sulfate ion and practically a zero slope from u = 0.1
to 0.4 M, Further the rate constants showed a dependence
on the peroxydisulfate concentration and could be explained
better on the basis of the formation of ion pairs between
Fe(II) and sulfate ions. There is justification for assum-
ing the existence of ion pairs because earlier workers have
also postulated similar effects. Wells and Salam99'1°° ob-
served ion pairs of Fe2+ and SO42— in the oxidation with
H202 and Irvine5° in the oxidation of tris(bipyridine)-
osmium(II) with peroxydisulfate. Other complexes such as
hexammine cobalt(III) and tris(ethylenediamine)-cobalt(III)
are also shown to form ion pairs with sulfate ion.47 Ion

. . . _28
pairs are shown to be important in the ox1dation of Fe(CN)64

and I- by $2082-.86

59

The formation of an intermediate between the iron(II)
complex and $2082- ion is supported by the evidence which
follows: Intermediates such as [Mn+-82082-] have been con-
firmed in the catalytic oxidation of iodide by peroxydisul~
fate in the presence of metal ions.4° Such an intermediate
has been postulated for the oxidation of copper(I) by per-
oxydisulfate and in the catalytic oxidation of some compounds
by peroxydisulfate in the presence of silver ions.17:27 A
stable diamagnetic intermediate has been obServed in the
iron( II) iodide-peroxydisulfate system.2°

The deviation from the second order rate law after dif-
ferent time intervals for the different complexes was at-
tributed by Irvine61 to the possible hydrolysis of Fe(II)
and Fe(III) complexes. Another possibility is the partici-
pation of species other than the complexes of the Fe(II)
species in the oxidation. Fordham and Williams4° concluded
from the bipyridine concentration dependence of the rates
in the reaction between peroxydisulfate and an equilibrium
mixture of Fe(II) and bipyridine that the reacting species
is the bi§(bipyridine)-iron(II) complex. They studied the
reaction at pH 3.9, a value sufficiently low for the equi-

librium between the mono, bis, and tris species to be rapidly

 

established.

In the present investigation, oxidation of the ligand
and deviation from the expected stoichiometry seem to be
small. Hydrolysis is slow under the present conditions,

and further according to Fordham and Williams'4° conclusions,

60

the rate of the reaction for the oxidation of the his
species is very much greater than for the t£i§_species.
Hence the hydrolysis of the t£i§_complexes to form a big
species should show an increase in the rate of the reaction
rather than the observed decrease. Moreover, the rates of
hydrolysis of the tris(bipyridine) and tris(dimethylbipyri-
dine)-iron(II) complexes are greater than that of the phen—
anthroline complexes, and yet these two fit the rate law
better than some others.

The tentative suggestion of a possible reverse step to
explain, at least partly, the deviation from the second
order rate law was based on Uri's work on the ferrous ion-
hydrogen peroxide system.96 In an extensive review on in—
organic free radicals in solution he gives a critical discus—
sion on the Haber—Weiss mechanism of the catalytic decomposi-
tion of H202 in the presence of Fe3+ and its revisions by a

number of authorss'9'90'98 to give a rate law of the form

‘dIH202I

_ 2+
dt — 2k[Fe IsslHZOZI

. 2 .
2+]ss = steady state concentration of Fe +. Uri

where [Fe
suggests an additional important step not taken into account

by earlier workers, namely the reaction:

+ - . 2+
Fe3 -OH + OH —> Fe + H202

This is the back reaction of the primary step of the Haber-

Weiss mechanism. In the original Haber-Weiss mechanism, the

61

radical initiation step was represented as:

+

3 _
2+ > Fe + OH-- + OH“.

Fe + H202

 

according to which, a reverse step would be highly unlikely
due to the necessity of termolecular collisions. Uri con-
siders the ion pair formation necessary and hence a re-

verse step to the radical initiation step is possible and
important. This step is also exothermic to the extent of

5 - 10 kcals.96 The rate equation obtained for the catalytic
decomposition of H202 by Fe3+ is revised to include the re-

verse step as

‘dIHzozI 2
_ + 3+. r -
T — 2mm 15501202] -2k-,tFe OHIIOHISS
where [Fe2+]ss, and [OI-1.]SS denote steady state concen-

trations. He could thus successfully fit Anderson's data6
to the above rate law. Anderson had been unable to explain
his data adequately.

Similar steps have also been prOposed for the photo-
oxidation of water by Ce(IV) ion by Evans and Uri§sr39

k-l 3+

4+ > Ce + H202

 

[Ce ~0H'] + OH‘

k2 4+ _

e3+ > Ce + OH

 

C + OH'

and the ratio of k2 to k_1 is calculated to be ~v10.
In the reaction of Co(III) ions with water, the rate

of the reaction is second order with respect to Co(III) at

high Co(III) concentrations and tends to a first order at

low concentrations of the latter1.3I14'15 Uri96 has proposed

62

 

the steps

3 —

CO +°OH ———>v Co2+ + OH'
3 ..

or C0 + + OH > Co2+ + OH'

k
3 - . -1 2
Co +‘OH + OH -———>. Co I + H202

to explain the change in order, with change in Co(III) con-

k2 +

> C03 + OH-

 

. . 2 + .
centrations and conSiders Co + OH

. . . . . 2
could become Significant only at high concentrations of Co +

ions.
+
n+ . k2 > M(n+1)

The rate of the two reactions M + OH +OH-

(n+1) k“1 n+

> M + H202, in all these

> M(n+1)+.

 

and [M +°OH-] + OH‘

 

cases depends on the oxidation potential of Mn+

2
Thus, for Fe +, ka/k_1 is very high and becomes important

only at high concentrations of Fe3+. For Ce3+, kz/k_1
is approximately 10 and for Cos+, the reverse step is pre-
dominant except at high concentrations of Co(II).

In view of the similarities of the bimolecular oxida-
tions with $2032- ions and H202, it may be appropriate to
suggest a Similar reverse step, gig.

k
2- _. —1
[Fe3+-so4 ] + so4

 

2 2—
> Fe + + $208

The formation of the Fe(III) - sulfate ion pair eliminates
the highly improbable termolecular step. The main point of
interest here would be the comparison of k_1/k2. From Uri's
arguments, it seems that this ratio would probably be greater

than that for Fe2+ and less than that for Ce3+, taking

63
into‘amxnnt only the oxidation potentials of the systems.
The nature of the radicals and factors such as the activa—
tion energies and entropies would affect profoundly the
ratio of k_1/k2 in peroxydisulfate system. Nothing is
known about these factors in these systems and so it is hard
to speculate.

There is no a priori evidence for the reverse step but
the observation that the deviation from the scond order
becomes more significant as the oxidation potential of the
system decreases seems to encourage this suggestion. Further
in the case of iron(II) complexes with bases such as phen-
anthroline, which are highly stable, Amphlett5 has shown
that under suitable conditions Fe(III) is quantitatively
reduced as a result of they radiolysis of water, whereas
with simple Fe3+, no reduction occurs. In the present in-
vestigation, the relative stabilities of the Fe(II) and
Fe(III) complexes may also be a factor in determining the
relative magnitudes of k_1, and k2.

Although the deviation from the pseudo first order
rate law suggests a probable reverse step, equilibrium was
not observed due to the following difficulties; the reac-
tions were carried out under conditions of large excesses
of peroxydisulfate and on long term storage, the solutions
became colorless due to hydrolysis or oxidation of the
ligand by excess peroxydisulfate. However, when excess of
Fe(II) complex was reacted with peroxydisulfate, the apparent

stoichiometry measured from the decrease in absorptivity of

64
the Fe(II) complex solutions, showed that less than two
equivalents of iron were being oxidized for each $2082-
present. But in many cases, the iodometric titrations did
not indicate any appreciable loss of peroxydisulfate to
other reactions. It is tempting to infer the establishment
of an equilibrium. However, the mixture could not be kept

too long due to deterioration from other complications.
B. Role of Alcohol

The results of the effect alcohol on the rate of oxi-
dation of Erisjbipyridine)-iron(II) indicated that the second
step of oxidation is indeed due to the 804-. radicals
formed in the rate determining step and eliminated the pos-
sibility that OH' radicals formed by the rapid equilibra-

tion so," + H20

 

> H804- + OH' may be the second

<—

stage oxidant. All these conclusions were reached from a

direct comparison with a study of Kolthoff, Medalia and

2+

Raaen68 on the peroxydisulfate - Fe systems and the work

of Merz and Waters81 as revised by Kolthoff, Medalia and

68

Raaen, on the relative rates of reaction of sulfate and

+ and ethyl alcohol.

2

hydroxyl radicals with Fe
The role of alcohol in the tris(bipyridine)-iron(II)-
sulfate-alcohol system has been explained on the basis of

the following scheme of reactions:

k _ _.
1 > Fe(III) + so,2 + so4 (1)

 

Fe(II) + 82082-
k2

 

804-. + Fe(II) > Fe(III) + 8042- (2)

65

 

> 'C2H4OH + HSO4— (3)
R4

504" + C2H50H

 

°C2H4OH + Fe(III) > Fe(II) + CH3CHO + H+ (4)

k5

 

'C2H40H + 32082" > CH3CHO + HSO4- + so4"(5)

Reactions (2), (3) and (4) are fast.‘ The extent to which
(2) and (3) occur depends upon the concentration of alcohol
and Fe(II) and on the rate constants. k3/k2,according to
the calculations of Merz and Waters?1 is equal to 0.0062

for the peroxydisulfate-Fe2+ system. This value was later
revised by Kolthoff, et. ai.68 to;0:015. This ratio for

the corresponding steps of the hydrogen peroxide-Fe2+ system
namely

k2

 

0H" + Fe(II) > Fe(III) + 0H" (2')

 

OH' + C2H50H > °C2H4OH + H20 (3')

is 2.05. Assuming similar magnitudes for the tris(bi-
pyridine)-iron(II)-peroxydisulfate system, there should be
practically no reaction if the alcohol concentration is
very much greater than Fe(II) concentration, on account of
steps (3) and (4). Table IV gives the values for the rate
constants at different concentrations of alcohol. It is
evident from the data thatat [alcohol]/[Fe(II)] ratio
greater than 2.4 x 103(up to 1.9 x 105) there is very
little overall oxidation of the Fe(II) complex, indicating
that steps (3) and (4) predominate. At [alcohol]/[Fe]
ratios less than 2.0 x 102 the rate constant gradually in-

creases to reach almost the value at zero alcohol concentration

66

when this ratio is about 21, indicating that step (2) dom-
inates. If the OH‘radical is the second stage oxidant, then
at this ratio for [alcohol]/[Fe(II) complex], the correspond-
ing step (3') should have been predominant by virtue of

the much higher value for ké/ké. These results are very
much in parallel to the work of Kolthoff, Medalia and Raaen68
on the Fe(II)-peroxydisulfate system, and suggest that the

sulfate radical is the reactive intermediate in these oxi-

dations.
C. Mode of Electron Transfer

For a complicated system such as the one in the present
investigation involving breaking of bonds in 82082. and
where orbital overlaps may be important, it is hard to
speculate whether an outer sphere or an inner sphere elec-
tron transfer occurs. It seems reasonable that a close con-
tact between the Fe(II) complex and the peroxydisulfate ion
may not be easy although it is sterically possible to ac-
comodate the peroxydisulfate anion between the ligand mole-
cules.25 At the same time, the extensive delocalization of
electrons in the w system of the ligands does not seem to
play a very favorable role, if one compares the rates of
oxidation of Fe2+ and the Fe(II) complexes with peroxydisul-
fate“!o According to Marcus theory74 for a series of reac-
tions in which the enthalpy of activation is effectively
constant and the coulombic interaction of the charges of

the reactants are constant there should be a linear

67
relationship between the free energy of activation and the
standard free energy of the oxidation-reduction step, with
a lepe of 0.5. The values of entropy in the present in-
vestigation differ quite a bit and hence no quantitative
relationship could be expected. The linearity of the plot
of AG* versus AGO does, however, indicate that the free
energy change of the oxidation-reduction step plays an im—
portant role in the rate of the reaction. In view of the
complexities involved in the present investigation and also
due to the standard deviation of about 2 kcal in the acti-
vation energies, much importance cannot be attached to the
slope of this line. The value of the slope is, however, in
agreement with Irvine's plot of log k versus the total change

in the standard CXidation-reduction potential.
D. Activation Parameters

The reactions are all accompanied by extremely large
negative values of entropies of activation, which is con-
trary to the value expected for reactions between posi—
tively and negatively charged species. These are much
smaller than the -8 calo K.1 mol-1 for the corresponding
oxidation of Fe2+ by peroxydisulfate ion, and was not ex-
plained in the earlier work. Wilmarth and Haim102 have sug-
gested that the small value of entropy of activation may be
attributed to the long separation between the two species in

the activated complex and thus making transfer of electrons

by tunnelling over a considerable distance, a relatively

68
inefficient process. By contrast, the ferrous ion may come
in closer contact with the peroxydisulfate ion in the acti-
vated complex which would contribute favorably to the entrOpy
of activation in two ways, by reducing the electron trans-
fer distance and by the loss of water from the solvation
sphere of the Fe(II) ion.

The activation energies for these reactions is consider-
ably lower than the value of 33.5 kcal required for the sym-
metrical fission of the peroxydisulfate ion“.9 Qualitatively,
the lowering of the activation energy below the value of
33.5 kcal may be viewed as stabilization of the activated
complexes caused by the partial transfer of electrons to
the nascent sulfate radical ion.102 The increase in the
free energy of activation with a decrease in the oxidation
potential of the complex is compatible with this. The forma-
tion of a stable Fe(II)-$203 intermediate proposed in this
investigation amounts to a delocalization of the electrons.
The formation of this entity also qualitatively explains the
low values of the entropy of activation. Separating these
two steps, the calculated value for the ion pair formation

-1

. . '1 . .
step is obtained as about + 15 cal0 K mol which is

reasonable for such a process.
B. Effect of Substituents

The slope of the Hammett plot and the plot of log k
versus pKa indicates that the overall rate of the reaction

is enhanced by withdrawal of electrons from the iron atom,

69
and is unexpected. However, a similar effect has been re—
ported recently by Rund and Claus88 for the catalytic decar-
boxylation of B—keto acids by the Mn(II) complexes of sub-
stituted phenanthrolines. The mechanism suggested is a
similar one involving an equilibration and a breaking up of

the intermediate complex.

 

10.

11.

12.

13.

14.

15.

16.

17.

LITERATURE CITED
Adam, F. c., and S. I. Weissman, J. Am. Chem. Soc. 89,
1518 (1958).
Afanasev, P. v., J. Phys. Chem.-U.S.S.R. 2, 568 (1937).

Albery, W. J., Ann. Rep. Prog. Chem. (Chem. Soc.) ‘62,
40 (1963).

Amdur, I., and G. G. Hammes, "Chemical Kinetics",
McGraw-Hill Book Company, New York. 1966.

Amphlett, c. 3., Nature 165, 977 (1950).
Anderson, 5., Acta Chem. Scand. .2, 1 (1948).

Ashurst, K. G., and W. C. E. Higginson, J. Chem. Soc.
3044 (1953).

Barb, W. G., J. H. Baxendale, P. George and K. R.
Hargrave, Trans. Faraday Soc. 41” 462 (1951).

Barb, W. G., J. H. Baxendale, P. George, and K. R.
Hargrave, Trans. Faraday Soc. 41, 591 (1951).

Barb, W. G., J. H. Baxendale, P. George, and K. R.
Hargrave, Trans. Faraday Soc. .41, 935 (1955).

Basolo, F., J. C. Hayes, and H. M. Newmann, J. Am. Chem.
Soc. ‘16, 3807 (1954).

Basolo, F. and R. G. Pearson, "Mechanisms of Inorganic
Reactions", Second Edition. John Wiley and Sons
Inc., 1967.

Bawn, C. E. H. and A. G. White, J. Chem. Soc. 331 (1951).

Bawn, c. E. H. and A. G. White, J. Chem. Soc. 339 (1951).

Bawn, C. E. H. and A. G. White, J. Chem. Soc. 343 (1951).

Bawn, C. E. H. and D. Margerison, Trans. Faraday Soc.
'51, 925 (1955).

Baxendale, J. H. and P. George, Trans. Faraday Soc.
29. 736 (1950).

70

18.

19.

20.

21.

22.

23.
24.
25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

71

Bekier, E. and S. W. Kijowski, Roczniki Chem.
136 (1935). Chem. Abstracts 39, 3306 (19367?

Benson, 8. W., "The Foundations of Chemical Kinetics"
McGraw-Hill Book Co., New York. 1960.

Bhatnagar, S. S., B. Prakash, and J. Singh, J. Indian
Chem. Soc. 17,133 (1940)

Brandt, W. W. and D. K. Gullstrom, J. Am. Chem. Soc.
2:3,. 3532 (1952) .

Brubaker, C. H. (Jr.), Record of Chemical Progress
2:1. 181 (1963) .

Burgess, J. and R. H. Prince, J. Chem. Soc. 5752 (1963).
Burgess, J. and R. H. Prince. J. Chem. Soc. 4697 (1965).
Burgess, J. and R. H. Prince, J. Chem. Soc. 1772 (1966).

Campion, R. J., N. Purdie, and N. Sutin, Inorg. Chem.
3, 1091 (1964)

Chaltykyan, O. A. and N. M. Beilerian, Izvest. Akad.
Nauk Armian S .S .R , Khim. Nauki. 11,13 (1958).
Chem. Abstracts 53,1902 (1959).

Chlebek, R. W. and M. W. Lister, Can. J. Chem. ‘33,
2411 (1967).

Cone, w. H., J. Am. Chem. Soc. 31, 78 (1945).

Csayni, L. J. and F. Solymosi, Acta Univ. Szegediensis,
Acta Phys. et Chem. 3, 34 (1959). Chem. Abstracts
£3, 13924 (1960)

Deck, C. F., and A. C. Wahl, J. Am. Chem. Soc. .13,
4054 (1954).

Dekker, A. 0., H. A. Levy, and D. M. Yost, J. Am. Chem.
Soc. 92. 2129 (1937).

Dickens, J. E., F. Basolo, H. M. Neumann, J. Am. Chem.
Soc. .19, 1286 (1957).

Diebler, H. and N. Sutin, J. Phys. Chem. 93, 174 (1964).

Downs, J. and R. E. Johnson, J. Chem. Phys. 22, 143
(1954).

Dulz, G. and N. Sutin, Inorg. Chem. .3, 917 (1963).

37.

38.
39.

40.

41.

42.

43.

44.

45.

46.

47.
48.

49.

50.

51.

52.

53.

54.

72

Eichler, E. and A. C. Wahl, J. Am. Chem. Soc. ‘39,
4145 (1958).

Evans, M. G. and N. Uri, Nature 166, 602 (1950).

Evans, M. G. and N. Uri, "Symposium No. 5, Carbon
dioxide Fixation and Photo—synthesis, Society of
Experimental Biology, Sheffield", 1950, p. 130,
Cambridge.

Fordham, J. W. L. and H. L. Williams, J. Am. Chem. Soc.
.13, 1634, 4855 (1951).

Ford-Smith, M. H. and N. Sutin, J. Am. Chem. Soc.
33, 1830 (1961).

Frazer, R. T. M., Rev. Pure Appl. Chem. (Australia)
11. 64 (1961) .

French, T. C. and T. C. Bruice, Biochemistry 3, 1589
(1964).

Fronaeus, S. and C. O. Ostman, Acta Chem. Scand. ‘19,
320 (1956).

Frost, A. A. and R. G. Pearson, "Kinetics and Mechanisms"
Second Edition, John Wiley and Sons, New York.
1961.

Galiba, H., L. J. Csanyi, and Z. G. Szalo, Z. Anorg.
Allgem. Chem., 287, 169 (1956).

Garrick, F. J., Trans. Faraday Soc. 33, 486 (1937).
George, P. and D. H. Irvine, J. Chem. Soc. 587 (1954).

Gupta, Y. K. and S. Ghosh, J. Inorg. Nuclear Chem. 9,
178 (1959).

Haim, A. and H. Taube, J. Am. Chem. Soc. ‘33, 3108 (1963).

Haim, A. and W. K. Wilmarth, J. Am. Chem. Soc. .33,
509 (1961).

Halpern, J., Quart. Rev. '33, 207 (1961).

Halpern, J. and L. Orgel, Discussions Faraday Soc.
39, 32 (1960).

Harkness, A. C. and J. Halpern, J. Am. Chem. Soc.
_33, 3526 (1959).

55.

56.

57.

58.
59.
60.
61.
62.
63.

64.
65.
66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

73

Holluta, J. and W. Hermann, Z. Physik. Chem. 166A,
453 (1933).

House, D. A., Chem. Rev. 31, 185 (1962).

Hudis, J. and R. W. Dodson, J. Am. Chem. Soc. 13,
911 (1956).

Hush, N. 8., Trans. Faraday Soc. 31, 557 (1961).
Irvine, D. H., J. Chem. Soc. 1841 (1957).
Irvine, D. H., J. Chem. Soc. 2166 (1958).
Irvien, D. H., J. Chem. Soc. 2977 (1959).
Jaffé, H. H., Chem. Rev. 33, 191 (1953).

Jirsa, F., Z. Anorg. Allgem. Chem. 148, 130 (1925).
and 158, 33 61 (1926).

King, C. v., J. Am. Chem. Soc. 33, 2689 (1927).
King, C. V., J. Am. Chem. Soc. '39, 2080, 2089 (1928).

King, C. v., and F. L. Griswold, J. Am. Chem. Soc.
9.2.. 1493 (1930) .

Kolthoff, I. M. and R. Belcher, "Volumetric Analysis
(III)", Interscience Publishers, New York, 1957,
p. 288.

Kolthoff, I. M., A. I. Medalia and H. P. Raaen, J. Am.
Chem. Soc. .13, 1733 (1951).

Kolthoff, I. M. and I. K. Miller, J. Am. Chem. Soc.
13. 3055 (1951).

Kolthoff, I. M., P. R. O'Connor, J. L. Hanson,
J. Polymer Sci. 13, 459 (1955).

Latimer, W. M., "The Oxidation States of the Elements,
and their Potentials in Aqueous Solution", Pren-
tice Hall, New York, N.Y., 1952, p. 78.

Libby, W. F., J. Chem. Phys. 33, 420 (1963).

Marcus, R. A., J. Chem. Phys. 33, 966 (1956); 33,
867 (1957); 29, 872 (1957).

Marcus, R. A., J. Phys. Chem. 31, 853 (1963).

Marcus, R. A., Can. J. Chem. 31, 155 (1964).

76.
77.

78.

79.

80.

81.

82.

83.

84.

85.

86.

87.
88.

89.

90.

91.

92.

93.
94.

95.

96.

74
Marcus, R. A., Ann. Rev. Phys. Chem. .13, 155 (1964).
Marcus, R. A., J. Chem. Phys. '33, 679 (1965).

Marcus, R. J., B. Zwolinski, and H. Eyring, J. Phys.
Chem., 33, 432 (1954).

Marshall, H.,Proc. Roy. Soc. (edinburgh) 33, 163 (1900).

McDaniel, D. H. and H. C. Brown, J. Org. Chem. 33,
42 (1958).

Merz, J. H. and W. A. Waters, Discussions Faraday Soc.
3, 179 (1947).

Nordmeyer, F. and H. Taube, J. Am. Chem. Soc. '33,
1162 (1968).

Noyes, A. A., J. L. Hoard and K. S. Pitzer, J. Am.
Chem. Soc. 31, 1221 (1935).

Noyes, A. A. and A. Kossiakoff, J. Am. Chem. Soc.
31, 1238 (1935).

Noyes, A. A., D. De Vault, C. D. Coryell, and T. J.
Deahl, J. Am. Chem. Soc. '33, 1326 (1937).

Perlmutter-Hayman, B., and G. Stein, J. Chem. Phys.
4_0. 848 (1964).

Raman, S. (Unpublished results) ‘
Rund, J. V., and K. G. Claus, Inorg. Chem. 1, 860 (1968).

Simon, A., W. Haufe, Th. Reetz, and R. Preisler,
Z. Anorg. Chem. 230, 129 (1936).

Sloth, E. N. and C. S. Garner, J. Am. Chem. Soc. 11,
1440 (1955).

Smith, w. v., J. Am. Chem. Soc. 11, 4077 (1949).

Stoner, P.,. M.S. Thesis, Michigan State University,
1966.

Stranks, D. R., Discussions Faraday Soc. ‘33, 73 (1960).
Sutin, N., Ann. Rev. Nucl. Sci. 13, 285 (1962).

Taube, H., Progress in Inorg. and Nuclear Chem. .1,
1(1959).

Uri, N., Chem. Rev. 33, 375 (1951).

97.

98.
99.

100.

101.

102.

103.

104.

105.

106.

75

Wang, J. H., and W. S. Brinigar, Proc. Natl. Acad. Sci.,

fig, 958 (1960).
Weiss, J., Experientia 1, 135 (1951).
Wells, C. F. and M. A. Salam, Nature 203, 751 (1964).

Wells, C. F. and M. A. Salam, Trans. Faraday Soc.
fig, 620 (1967).

Wells, C. F. and M. A. Salam, J. Chem. Soc. (A), 308
(1968).

Wilmarth, W. K. and A. Haim, "Peroxide Reaction
‘Mechanisms", Interscience Publishers, London
(1961) p. 175.

Yost, D. 14., J. Am. Chem. Soc. 33, 152 (1926).

Yost, D. M., J. Am. Chem. Soc. 33, 374 (1926).

Yost, D. M., and W. H. Claussen, J. Am. Chem. Soc.
2;, 3349 (1931).

ZwiCkel, A. M. and H. Taube, Discussions Faraday Soc.
22, 42 (1960).

APPENDIX

The rate constants calculated from the different ex-
periments are summarized in the following tables. The
averages of these rate constants were used for further calcu-
lations of the forward rate constants, graphically. In some
cases, the readings, which appeared obviously out of line
with the others, were not considered in computing the aver-
age. Most of these were in the case of 5-nitro- and 5-chloro-
phenanthroline complexes, where a small delay in starting
the runs, caused a serious difference and a few were traced

back to faulty temperature control.

76

77

 

 

Temp. [32032.] x 103(3) Second Order Rate Constants
0C (M-lseC-l) x 101

 

Tris-(2,2'-Bipyridine)-iron(II) sulfate

25 1.0 . 3.34, 3.47, 3.37, 3.22
1.5 2.88, 2.93
2.0 2.69, 2.69, 2.72
2.5 2.48, 2.50, 2.46
3.0 2.33, 2.35, 2.34, 2.29
4.0 2.14, 2.13, 2.14, 2.15
5.0 1.85, 1.92
30 1.0 4.54, 4.69, 4.32
2. 3.46, 3.46, 3.48, 3.48
2.5 3.34, 3.84
3.0 3.12, 3.12, 3.08, 2.98, 2.98
3.5 3.06, 3.06
4.0 2.71, 2.80, 2.43, 2.96, 2,96
5.0 2.6 '
35 1.0 5.48, 5.45, 5.53, 5.54
. 4.78
2.0 4.50, 4,43, 4,31, 4.51
2.5 4.18, 4.12
3.0 3.83, 3.75, 3.63
4.0 3.53, 3.53, 3.41
40 1.0 7.11, 7.15, 7.06, 7.2, 7.1
1.5 6.34, 6.41
2.0 5.6, 5.4, 5.25, 5.49, 5.65, 5.75
2.5 5.17, 5.40, 5.01
3.0 4.62, 4.62, 4.71, 5.1, 4.31

4.0 4.08, 4.03. 4.06

78

 

 

Temp. [52082—] x 103(3) Second Order Rate Constants
_1 _
OC 4 (M sec 1) x 101

 

Tris-(1,10—Phenanthroline)-iron(II).sulfate

25 1.0 1.69, 1.70, 1.81, 1,80, 1.61, 1.62
2.0 1.31, 1.29, 1.39, 1.21, 1.30
3.0 1.23, 1.26, 1.31, 1.41
4.0 1.01, 1.01, 1.02, .98, 1.13
30 1.0 2.33, 2.27, 2.3, 2.76
2.0 1.98, 1.92, 1.92, 1.88
3.0 1.73, 1.71, 1.73, 1.81
3.5 1.60, 1.71, 1.60
4.0 1.54, 1.54, 1.56, 1.51, 1.66
4.5 1.44, 1.43
5.0 1.17, 1.18
35 1.0 3.47, 3.41, 3.29, 3.49, 3.38, 3.22
1.5 3.1, 2.98
2.0 2.63, 2.57, 2.63, 2.73, 2.93, 2.32
2.5 2.33, 2.28
3.0 2.25, 2.19, 2.27, 2.35, 2.14, 2.34
4.0 1.97, 1.97, 2.01, 1.81
4.5 1.91
40 1.0 4.16, 4.21, 4.54, 3.95, 4.28
1.5 3.34, 3.36
2.0 3.05, 3.03, 3.17, 3.11, 3.21
3.0 2.61, 2.61, 2.63, 2.91, 2.31
4.0 2.26, 2.30, 2.31, 2.16

79

 

 

Temp. [82082-] x 103(3) Second Order Rate Constants
0C (M-lsec-l) x 100

Tris-(4,4'-Dimethyl—2,2'-bipyridine)-iron(II) sulfate

 

25 0.5 4.59, 4.75, 4.76, 4.58
1.0 3.55, 3,56, 3,54
1.5 3.07, 3.10, 3.06
2.0 2.67, 2.81
30 0.5 6.02, 6.26, 5.82
1.0 5.24, 5.07
1.5 3.84, 3.84
2.0 3.09, 3.36
2.5 3.22, 2.97
35 0.5 6.96, 6.91
1.0 5.43, 5.49, 5.81
1.5 4.74, 4.62, 4.21
2. 4.32, 3.76
40 0.5 8.15, 8.05
1.0 6.40, 6.62, 5.86, 7.12
4.84, 5.05

-1

 

(M sec-1) x 102
1333-(5éMethyl-1,10—phenanthroline)-iron(II) sulfate
25 1.0 6.97, 6.72, 7.51, 6.12
2.0 6.25, 5.83, 5.62, 5.08, 5.92, 6.31
3.0 5.41, 5.50, 5.52, 5.4, 4.65,
4.26, 4.51

4.0 4.3, 4.26, 4.35, 4.61, 4.12

80

 

 

 

 

 

Temp. [82082-] x 103(3) Second Order Rate Constants
0C (M-lsec-l) x 102
30 . 11.8, 10.5, 12.4, 10.1, 12.1
1.5 9.4, 9.4
2.0 8.5, 8.3, 7.95, 8.82, 7.62,
9.31
2.5 7.11, 7.32, 7.60
3.0 6.67, 6.91, 6.83
4.0 6.01, 6.01, 5.97
35 13.6, 12.4, 12.5, 13.0, 13.4
1. 11.4, 12.1
2.0 11.1, 10.8, 10.1, 12.5
2.5 10.0
3.0 9.46, 9.48, 10.3, 10.4, 8.64,
8.95
4.0 8.53, 8.29, 8.36, 8.86, 8.11
(M-lseC—l) x 101
40 1.0 1.95, 2.00, 2.08, 2.22, 1.75
1.5 1.7, 1.77, 1.84
2.0 1.34, 1.26, 1.41, 1.50, 1.3
2.5 1.21
3.0 1.11, 1.05, 1.09, 1.11, 1.08,
1.21
4.0 1.02, 1.06, 1.12, .993
(M-lsec-l) x 102
$3137(5-Chloro—1,10-phenanthroline)-iron(II) sulfate
25 1.0 9.45, 9.33, 9.69
2.0 4.78, 4.94, 5.54, 5.02, 4.51,
4.5, 5.35
3.0 3.29, 3.62, 3.76, 3.69, 3.71,
3.82, 3.39, 3.29
4.0 2.79, 2.78, 2.95, 2.86, 2.91,

2.72

81

 

 

 

 

Temp. [52082-] x 103(3) Second Order Rate Constants
OC (M-lsec-l) x 102
30 1.0 22.4, 23.4, 20.7, 22.2
2. 9.15, 9.10, 9.0, 10.1, 8.2,
1001’ 803
3.0 6.77, 6.97, 6.90, 6.71
4.0 4.21, 5.03, 4.32, 6.51, 4.93, FT
5.83 )
(M-lsec-l) x 101 L
35 1.0 3.45, 2.3, 3.99, 2.7, 4.1, 2.21 4
2.0 1.91, 2.0, 1.81, 2.21, 1.72,
2.31
3.0 1.23, 1.34, 1.17, 1.02, 1.38,
1.21
4.0 .96, .95, .98, .92, .91, 1.06
40 1.0 5.21, 5.13, 5.26, 5.17
2.0 3.43, 3.31, 2.62, 2.91, 2.61
3.0 1.81, 2.20, 2.39, 2.06, 1.92,
I 1.90, 2.21
4.0 1.75, 1.83, 1.93, 1.72
Tris-(5—Nitro-1,10-phenanthroline)-iron(II) sulfate
25 1.0 1.55, 1.78, 1.62, 1.72, 1.85,
1.79
1.5 1.13, 1.31
2.0 .82, .88, .92, .85, .78, .73,
.91 “
2.5 .69, .71, .71
3.0 .61, .55, .65, .62, .64, .54
4.0 .45, .46, .43, .49, .42, .48
30 1.0 3.97, 3.5, 3.61. 4.01. 4.71
2.0 2.06, 2.06, 1.97, 1.72, 2.31
3.0 1.59, 1.47, 1.21, 1.84
4.0 1.04, 1.1, .96, .92, 1.2, 1.01

82

 

 

 

Temp. [82082-] x 103(3) Second Order Rate Constants

0C (M-lsec-l) x 101

35 1.0 5.41, 5.43, 5.52, 5.34
2.0 3.15, 3.09, 3.21, 3.05
3.0 2.49, 2.17, 2.34, 2.71, 2.61
4.0 1.51, 1.62, 1.74, 1.74

40 1.0 7.81, 9.5, 8.1, 8.0, 7.5, 10.8
2.0 5.55, 5.53, 5.13, 5.25, 5.91
3.0 3.83, 3.75, 3.74, 3.52, 3.96

4.0 3.01, 3.04, 3.03, 3.12, 2.98