at:
il
xfo
for
8 S
mo
wer
ly
in the nature of their active sites for attacking the substrate, but here we stress that their differences often rest in the value of their frame-
works. The frameworks contribute to activity through binding of substrate, creating selectivity, or even by directly aiding the catalytic act
sometimes by opening up a pathway different from that of
and enzymes, heterogeneous solids, non-conducting and
conducting), Table 1. Here we wish to compare all these
Now the catalysis may differ in that the rate controlling
step may involve either very short-term contact, collision,
or binding between catalyst and substrate. The framework
may have little consequence if reaction with the active
*
Tel.: +44 (0) 1865 272621; fax: +44 (0) 1865 272690.
E-mail address: bob.williams@chem.ox.ac.uk
Available online at www.sciencedirect.com
Journal of Inorganic Biochemist
JOURNAL OF1. Introduction
The basis of catalysis is the enhancement of rate of a
reaction by a substance without overall change in that sub-
stance, the catalyst. It is not an over simplification to state
that the activity of the catalyst arises because it assists the
re-arranging or moving of atoms or electrons of the bound
reactants from one centre to another while forming prod-
ucts more quickly than would happen in its absence. In
other words, it reduces the free energy barriers to reaction,
divisions concentrating upon their similarities in that they
all have centres for attack on substrates which have often
been called active sites. We shall be referring to catalysts
which mostly contain one metal ion as the attacking group
although many enzymes do not use these ions. We shall
show that as the framework holding these centres of activ-
ity is increased the cause of activity often becomes more
widespread in the catalyst, when it is preferable to describe
activity in terms of active regions. It is in the part played by
the framework that enzymes have the strongest advantages.of transforming the substrate to the product, when there is an active region rather than a site. It may also provide limited directed motion
aiding effective progress of the active groups themselves through a cycle of activity. The article highlights the difficulties in the use of
other kinds of catalysts as aids to the understanding of enzymes. Part A is a general description and Part B is a set of examples of
the catalysts.
 2007 Elsevier Inc. All rights reserved.
Keywords: Catalysts; Enzymes; Catalysts: solid state; Catalysts: molecular; Catalyst frameworks; Active sites
Part A
General considerations of catalysis
the easiest un-catalysed reaction. Many kinds of catalyst
have been used in similar organism, laboratory and indus-
trial processes but the basis of functional activity has often
been obscured by the manner of study of each of the four
principal divisions of catalysts (homogeneous moleculesAbstract
The purpose of this review is to compare four kinds of catalyst:
conducting solids, both heterogeneous, in order to show the full po
to describing catalysts containing single metal atom or ion units, onReview
A comparison of types of catalys
R.J.P. W
Inorganic Chemistry Laboratory, University of O
Received 28 March 2007; received in revised
Available online0162-0134/$ - see front matter  2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jinorgbio.2007.08.012rticle
The quality of metallo-enzymes
liams
*
rd, South Parks Road, Oxford OX1 3QR, UK
m 7 August 2007; accepted 17 August 2007
eptember 2007
lecular and enzymic, both homogeneous, and non-conducting and
of metallo-enzymes. For ease of comparison we restrict ourselves
briefly mentioning more complex units. Their common ground lies
www.elsevier.com/locate/jinorgbio
ry 102 (2008) 1–25
Inorganic
Biochemistry

olut
le-li
n w
(Heterogenized variants) Membrane catalysts including enzymes
sul
ed
ined
ganattacking groups follows collision. In this case reaction is
conventionally divided between a collision rate Z, a prob-
ability factor, P, and a free energy barrier DG
Rate constant; k ¼ ZPe
DG=RT
ð1Þ
see for example, Ref. [1] for detailed analysis.
Major interest lies in the ability of a catalyst to reduce
the transformation energy, DG, from that of the uncataly-
sed reaction. This is achieved in a unimolecular reaction on
collision of catalyst and reactant provided the catalyst
increases the fraction of molecules which have sufficient
energy in that part of the reactant molecules which is to
undergo change. Collision alone is ineffective. There may
not be time for energy equilibration, although the collision
could be thought of as being sticky, as any binding is tran-
sient and does not affect the rate. DG is not then simply
related to the energy partition function [1]. We shall also
be interested in P, which is the probability of correct orien-
tation and can be biased by a framework. However, for
many catalysed unimolecular reactions there will be a bind-
ing step and this becomes almost a necessity for bimolecu-
lar reactions which, when catalysed, require a termolecular
interaction. Termolecular collisions are very infrequent. In
order to avoid complications, which unfortunately are
often observed we shall generalise the analysis of them
using saturation or Michaelis–Menten kinetics. Assuming
3. Heterogeneous solid systems Continuous solids such as oxides and
Nanoporous solids Open frameworks including immobilis
Dendrimers
Metallic or semiconductors Metals, alloys, conductors
Note. The active site atoms in all the cases except the first may be constraTable 1
Principal types of catalysts
Name or classification Properties
Homogeneous
1. Molecular systems Single atom/ion site (not constrained)
(Heterogenized variants) (a) Soluble small molecule active in s
(b) Small molecule attached by flexib
2. Enzymes Soluble large proteins active usually i
2 R.J.P. Williams / Journal of Inorthat the binding comes close to equilibrium, see Appendix
at the end, we can express rate, free of concentration terms,
as a product of an equilibrium binding constant, K and a
subsequent rate step, k
0
Rate constant ¼ K  k
0
ð2Þ
Here K involves complications in that the substrate may
well be very precisely oriented in the binding pocket, com-
pare P in Eq. (1), and the binding itself may assist reaction
rate not only by increasing the effective concentration
product of a bimolecular reaction in the active region,
but by straining a substrate and hence increasing the over-
all rate while not being the rate limiting step. Note the
energy of binding, which is temperature dependent, is
described by an equilibrium free energy change, DG
0
.Itisclear that here the framework, immobile or mobile, can
play a large part in the stereochemistry and the overall rate
of reaction. The rate constant itself takes the form
k
0
¼ P
0
e
DG
00
=RT
ð3Þ
where P
0
is a further orientation factor if the substrate
and/or catalyst adjust after initial binding. Once again
DG
00
is not simply related to the partition function of
energies of the bound complex [1]. In fact there may be
many steps in the bound state before product release
and they require small adjustments of molecular structure
of both the intermediates and the catalyst. The progres-
sion of steps is often described by transition state theory
[1] where the rate of each step is limited by the binding
energy difference DG
b
between the energy of the interme-
diate and the top of the barrier to the next intermediate,
treated as at equilibrium with it. Energy distribution is
assumed to be rapid in the bound states. The rate step
is then the vibronic frequency which causes reaction.
Now often, as a consequence of many steps, Eq. (2)
has to be replaced by a much more complex form. Here
unfortunately this only serves to obscure comparisons. In
view of these difficulties of comparable representation of
catalysis by mathematical expressions, and hence our
ability to determine related parameters, we are only able
to compare major structural features of catalysts includ-
ion
nker to a surface
ater including proteins such as antibodies modified for catalytic purposes
fides
atoms, complexes or nanoparticles of metals
.
ic Biochemistry 102 (2008) 1–25ing the structure and functions of attacking groups, and
those of the frameworks which hold them. At the very
elementary level to which we have reduced the equations
of rate, see footnote at end, it is the structure which con-
trols the binding and access to the site expressed by K but
it is the attacking groups which control k
0
. It is one inten-
tion of this article to indicate how this separation is of
very different security for different types of catalyst. Prob-
lems of comparison are increased by differences in small
motions after binding. We shall not be interested, how-
ever, in translational motion of substrates on, off, or in-
side the catalyst and we shall not concern ourselves
with the complications of major reorganisation of the cat-
alyst itself until late in this article. Now the difficulties of
comparison are increased by the way the different cata-
lysts have been studied.