128
Chemistry 1983
At the quantitative theoretical level, attempts were being made to account
for the barrier to reaction attending encounter and separation and that contri-
buted by charge trapping within the metal complex and by the surrounding
medium. The papers which most influenced the experimentalists, at least
during the formative period in question, are those by Marcus (33) and Hush
(34), dealing with adiabatic (35) electron transfer. Other theoretical ap-
proaches were being advanced during this period, and in some, attempts were
made to account for non-adiabaticity, and the various theories are compared
and evaluated in reference (33). (The current state of theory can be gathered
from a recent article by Sutin (36)). Because this very important aspect of
electron transfer reactions is not dealt with in this paper, it is essential to
mention that the processes as they occur at electrodes were not being over-
looked.
The correlation of the rates of cross reactions with the rates of the component
self-exchange reactions (5), has been widely applied, especially to outer-sphere
reactions. The limits of its validity were clearly set down by the author:
allowance must be made for the work of bringing the reaction partners together
and separating the products, electron delocalization must be great enough to
ensure adiabatic behavior, but not so great as materially to reduce the activa-
tion energy. (The last condition, it should be noted, does not necessarily limit
the applicability of the Marcus equation to outer sphere processes.) Hush’s
treatment (34) also leads to a correlation of rates of self-exchange and cross
reactions. It also takes into account the contribution by driving force, and in
fact the first calculation of the rate of a cross reaction, in this instance:
from those of the self exchange processes and the equilibrium constant appears
in reference 34.
Theory of another kind has profoundly influenced the development of the
field, even though it is qualitative. It responds to the question of how the choice
of mechanism, and relative rates, are to be understood in terms of the electronic
structures of the reactants. As is true also of rates of substitution, the observa-
tions are so sensitive to electronic structure that even qualitative ideas are
useful in correlating observations, and in pointing the way to new experiments.
Orgel (37) early applied qualitative ligand field theory in discussing the inner-
sphere mechanisms for the reduction of Cr(III) and low spin Co(III) complex-
es. When electron transfer takes place through a bridging group, it is important
to distinguish a chemical or “hopping” mechanism - here a low lying orbital of
the ligand is populated by the reductant, or a hole is generated by the oxidant
in an occupied orbital - from resonance transfer, that is, electron tunneling
through the barrier separating the two metal centers. This distinction was
drawn rather early by George and Griffith (38) who moreover proposed alter-
native mechanisms for resonance transfer. Shortly thereafter, Halpern and
Orgel (39) gave a more formal treatment of resonance transfer through bridg-
ing ligands. Concerns about the relation between electronic structure and the
observations on electron transfer strongly influenced my own work, but before
tracing this theme, I want to report on the progress made, mainly by others, in
extending the descriptive chemistry of electron transfer reactions.
H. Taube
129
The unambiguous demonstration of an inner sphere mechanism in a sense
introduced a second dimension to the field of electron transfer mechanisms.
That in certain systems reaction perforce took place by an outer sphere mecha-
nism had long been known, but until the experiments of references 16 and 17
were done, the inner sphere mechanism was only conjecture, and, as frequently
happens in research in chemistry, only after conjecture, however reasonable, is
upgraded by proof, is it accepted as a base for further development. That the
distinction between the two reaction classes is meaningful, not only in terms of
chemistry but also in rates, will be illustrated by a single comparison: the
specific rates of reaction of Cr
2+
(aq) with (NH
3
)
5
CoCl
2+
is ca. 10
8
greater than
it is with CO(NH
3
)
6
3 +
(40) (the latter can only react by an outer sphere
mechanism). The classification of reaction paths as inner sphere or outer
sphere, on the basis of rate comparisons, involving effects (such as those
exerted by non-bridging ligands) established with reactions of known mecha-
nism, became the focus of experiment and discussion when direct proof based
on product or intermediate identification was lacking.
Early in the 1960s new metal centers were added to the roster of those
proven to react by inner sphere mechanisms. For Co(CN)
5
3-
as reducing agent
(41), the demonstration of an inner sphere path again depended on the charac-
terization of product complexes by orthodox means. Sutin and co-workers have
been particularly resourceful in using flow techniques to provide direct proof of
mechanism for oxidizing centers ordinarily considered as labile to substitution:
thus
note
t h e p r o o f o f i n n e r
sphere
mechanism
f o r F e C l
2 +
( a q )
+
+
(43), even the much studied
exchange (44). It was early appreciated (17) that atom or group transfer is not
a necessary concomitant of an inner sphere process. Whether the bridging
group transfers to reductant, remains with the oxidant, or transfers from
reductant to oxidant depends on the substitution labilities of reactants and
products. Early qualitative observations (17) on the Cr
2+
(aq) +
system
had apparently exposed an example of reaction by an inner sphere mechanism,
but leading to no net transfer of the bridging atom. Here
is
formed as an intermediate, but this then aquates to
(later work (45) has shown the inner sphere path to be minor compared to the
outer sphere, and that
is the lesser product of the former path).
Experiments (46, 47) with
as reducing agent provided numerous
examples of systems in which substitution on the reducing complex is rate
determining for the net redox process (note that
because of its
electronic structure
is expected (18) to undergo substitution relatively
slowly). Unstable forms of linkage isomers were prepared by taking advantage
of the chemistry of the inner sphere mechanism:
by the
reaction
with
49)
by the reaction of
with
Oxygen atom transfer was shown to be com-
plete (51) in the reaction of
with
(The path involv-
ing direct attack on the aquo complex was later (52) shown to be unobservable
compared to attack on the hydroxo. Bridging by H
2
O has to date not been
demonstrated.) The inner sphere path was demonstrated (53) also for net 2e
-