ELECTRON TRANSFER BETWEEN METAL
COMPLEXES - RETROSPECTIVE
Nobel lecture, 8 December, 1983
by
HENRY TAUBE
Department of Chemistry, Stanford University,
Stanford, CA 94305
This will be an account in historical perspective of the development of part of
the field of chemistry that I have been active in for most of my professional life,
the field that is loosely described by the phrase “electron transfer in chemical
reactions”. In the short time available to me for the preparation of this paper, I
can’t hope to provide anything significant in the way of original thought. But I
can add some detail to the historical record, especially on just how some of the
contributions which my co-workers and I have made came about. This kind of
information may have some human interest and may even have scientific
interest of a kind which cannot easily be gathered from the scientific journals.
For publication there, the course of discovery as it actually took place may be
rewritten to invest it with a logic that it did not fully acquire until after the
event.
Simple electron transfer is realized only in systems such as Ne + Ne
+
. The
physics already becomes more complicated when we move to N
2
+ N
2
+
for
example, and with the metal ion complexes which I shall deal with, where a
typical reagent is Ru(NH
3
)
6
2+
, and where charge trapping by the solvent, as
well as within the molecule, must be taken into account, the complexity is
much greater. Still, a great deal of progress has been made by a productive
interplay of experiment, qualitative ideas, and more sophisticated theory,
involving many workers. Because of space limitations, I will be unable to trace
all the ramifications of the field today, and will emphasize the earlier history of
the subject, when some of the ideas basic to the field were being formulated.
This choice of emphasis is justified because, by an accident of history, I was a
graduate student at the University of California, Berkeley, about the time the
first natal stirrings of the subject of this article occurred, and at a place where
these stirrings were most active. As a result, I may be in a unique position to
deal knowledgeably and fairly with the early history of the subject. The
emphasis on the early history is all the more justified because most of the topics
touched on in this article, and also closely related topics, are brought up to date
in a very recent volume of the series, Progress in Inorganic Chemistry (1).
Chemical reactions are commonly classified into two categories: substitution
or oxidation-reduction. The latter can always be viewed as involving electron
H. Taube
121
transfer, though it is agreed that when we consider the mechanisms in solution,
electron transfer is not as simple as it is in the Ne + Ne
+
case. Rearrangement
of atoms always attend the changes in electron count at each center, and these
must be allowed for. I will, however, simplify the subject by considering only
processes of simple chemistry: those in which electron transfer leaves each of
the reaction partners in a stable oxidation state. While substitution reactions
can be discussed without concern for oxidation-reduction reactions, the reverse
is not true. The changes that take place at each center when the electron count
is changed is an essential part of the “electron transfer” process, and may be
the dominating influence in fixing the rate of the reaction. Moreover, most of
the early definitive experiments have depended on exploiting the substitution
characteristics of the reactants, and of the products. Thus, the attention which
will be devoted to the substitution properties of the metal ions is not a
digression but is an integral part of the subject.
An appropriate place to begin this account is with the advent of artificial
radioactivity. This enormously increased the scope of isotopic tracer methods
applied to chemistry, and made it possible to measure the rates of a large
number of oxidation reduction reactions such as:
Hevesy and coworkers (2) who used naturally occurring isotopes to follow
Pb(IV)/Pb(II) exchange in acetic acid.) Because chemists were there involved
in the discovery of many of the new isotopes (3) an early interest in this kind
of possibility developed in the chemistry community at the University of
California, Berkeley, and was already evident when I was a graduate student there
(1937-40). Mention is made in a review article by Seaborg (3) devoted to
artificial radioactivity, of an attempt (4) to measure the rate of the Fe
3+/2+
exchange in aqueous chloride media, the result of this early attempt being that
the exchange was found to be complete by the time the separation of Fe(III)
from Fe (II) was made. It was appreciated by many that the separation procedure,
in this case extraction of the Fe(III)-Cl
-
complex into ether, might have induced
exchange. It was also appreciated that Cl
-
might have affected the reaction rate,
possibly increasing it, and that quite different results might be obtained were the
experiment done with Cl
-
being replaced by an indifferent anion.
There were several reasons for the interest, among many physical-inorganic
chemists, in a reaction such as (1). That the interest in chemical applications of
the new isotopes was keen in Berkeley may be traced in part to the involvement
of much of the research body in teaching in the introductory chemistry course.
We all had a background of qualitative observations on oxidation-reduction
reactions of simple chemistry - as an example, on the reaction of Ce(IV)(aq)
with Fe
2
+
(aq), - from experience in qualitative or quantitative analysis. Still, to
my knowledge, at the time I was a graduate student, not a single measurement
had been made of the rate of this kind of reaction. That a field of research,
which has since grown enormously, was started by studying “self-exchange
reactions” (5) such as (1), rather than net chemical changes (descriptor,
“cross-reaction”) (5), may reflect the intervention of a human factor. Measur-