Towards rational design of solid acid catalysts

Article abstract of DOI:10.1524/zpch.1996.197.part_1_2.037

We discuss in this article methodologies and strategies for rational design of solid acid catalysts. In particular we will demonstrate how a new computational technique, de novo molecular design, is effective in suggesting new template molecules which can be utilised in the synthesis of target solids. A new microporous aluminophosphate-based catalyst, DAF-4, was prepared which is effective for converting methanol to light olefins. Other methodologies which may prove effective in designing new materials are also discussed. by R. Oldenbourg Verlag, 1996.

Full text of DOI:10.1524/zpch.1996.197.part_1_2.037

Zeitschrift für Physikalische Chemie, Bd. 197, S. 37-48 (1996)
by R. Oldenbourg Verlag, München 1996
©
Towards Rational
of Solid Acid
Design
Catalysts
By John Meurig Thomas and Dewi Wyn
Lewis*
Davy Faraday Research Laboratory, Royal Institution of Great Britain,
21
Albemarle St., London W1X 4BS, UK
Dedicated to Gerhard Erti on the occasion
of his
60th
birthday
(
Received June 3, 1996)
I
I
I Rational
Catalyst Design Computer-modelling
We discuss in this article methodologies and strategies for rational
of solid acid
design
new computational technique, de
molecular design, is effective in suggesting new template molecules which can be utilised
catalysts. In particular we will demonstrate how
a
novo
in the synthesis of target solids.
A
new microporous aluminophosphate-based catalyst,
DAF-4, was prepared which is effective for converting methanol to light olefins. Other
methodologies which may prove effective in designing new materials are also discussed.
Vor dem Hintergrund ökonomischer Notwendigkeiten und der intellektuellen Heraus-
neue
herzustellen, werden die Strategien zur rationalen
Festsàurekatalysatoren
forderung,
Planung der Synthèse eines speziellen Katalysators, DAF-4, der Methanol selektiv in
Ethen and Propen überführt, beschrieben. Die Konzepte beim Design geeigneter organi-
scher
zum Wachstum
auf der Basis chemi-
werden neben anderen Möglichkeiten zum
Templatmoleküle
mikroporöser Katalysatoren
scher Erfahrung und
Computermodellierung
Design neuer Katalysatoren, auch unter Einbeziehung der möglichen
Kombinatorik, erlautert.
Rolle chemischer
Introduction
and other
scientists can
such as artificial
Enzymologists
biological
nowadays successfully
and
new
that rival,
enzyme.
design
synthesize
catalysts
enzymes
natural
where
do not
the
of the
they
surpass,
performance
parent
How is it that there now exists this remarkable
of
and control,
degree
design
a
mutant
is superior to the wild form? Part of the reason
whereby
enzyme
of the
lies in the
of site-selective
that en-
discovery
technique
mutagenesis
ables the amino acid constituents of
*
proteins to
be
replaced by slightly
(or
a
Current address: Department of Materials Science and Metallurgy, University of
Cambridge, Cambridge,
CB2 3QZ, UK.
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

38
John Meurig Thomas and Dewi Wyn Lewis
different one in the
of the active site, which is
generally
radically)
composed
vicinity
of
a
lined
three or so
enviable
amino acids. Part of the
pocket
by
specific
reason for the
also arises because
enzymologists'
dexterity they
have, for over thirty years, been engaged in in
situ structural studies of their
into mechan-
ism. We shall return to the relevance of these facts later in this article.
have
catalysts. They
consequently gained revealing insights
But it is not
breed of
the
merchant
the new
measure of
only
enzymologist
biologically-oriented catalyst
who can
achieve
a
nowadays
large
success in
Those who harness genetically engineered
designing catalysts.
are in
a
to elucidate,
biocatalytic pathways.
in vitro reconstruction,
by
syntheses
strong position
the nature of
and
With the
biosynthetic
development
of monoclonal antibodies and
have been taken in the
strides
genetic engineering techniques, major
of
antibodies. The
for
development
catalytic
is to choose the right starter mol-
particular catalytic antibody
strategy
at
a
arriving
ecules, the so-called hapten, to generate an antibody with the appropriate
of
An
raised
against a
transition state
disposition
catalytic groups.
antibody
will bind it, and thus be
a
for the reaction that
passes through that
catalyst
state. The
are
the transition state analogues.
haptens
effectively
Those who work in the domain of
have also made
photocatalysis
signifi-
cant studies in recent
in
heterojunctions of binary inor-
years
the electronic structure of which has been
of solid-state
assembling
ganic solids,
engineered using
chemistry.
the
and solid-state
well-proven principles
physics
In this connection the work of Zamaraev and his collaborators is
pre-emi-
nent. These workers have
of metal sulphides suit-
produced microparticles
by harnessing
contaminated
H2S. And in the realm of
able for the
solar radiation
of
photocatalytic cleansing
inland seas that are
naturally
homogeneous catalysis also, although pre-ordained design
the
order
in the
by
is not
of
quite
advances have
high
accomplishment
biological
world,
major
Kaminsky
been made
that enable
metallocene precursors of the so-called
type
of
using
to be
with
a
stereospecific polymers
produced
greater degree
control than was hitherto achievable
Here,
using Ziegler-Natta catalysts.
however, there still remains
a
fair degree of 'black art' in that
a
good deal
of
exists
the nature and role of
like
uncertainty
concerning
co-catalysts
alumoxanes
fied
to render these sandwich cationic active centres
viable.
required
(typi-
by (Zr(cp)2)2+) catalytically
But what of the situation
solid
Are
concerning
heterogeneous catalysts?
we now in
has to be
a
position to design and synthesize these to order? The answer
out
So far as
and
which
spelled
carefully.
supported catalysts go
by
these one means
that are
and
any
all
multicomponent
multiphasic,
covers
and all those bifunctional
essentially
Fischer-Tropsch catalysts
used for
catalysts
that consist of
and accumulated
divided metals on
hydrotreating
inspiration
finely
oxides or
sulphides
still domi-
knowledge
nate. We are
a
from
able to construct an intellectual
quite long way
being
new
blueprint that
is then translatable into
a
But for
catalyst.
monophasic,
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

Towards Rational Design of Solid Acid Catalysts
the
39
so-called uniform
prospects are infinitely brighter,
as
catalysts [1, 2],
has been shown
on other occasions
There has, however,
[3, 4].
progressively
been
further
very
and it is upon these new
that we dwell upon
significant
progress
recently
with uniform
developments,
heterogeneous catalysts,
in the remainder of this article.
mostly
Uniform, heterogeneous catalysts, exemplified by zeolitic and other
and
solids, as well as
are,
microporous
definition, monophasic crystalline solids in which the active sites
uniform manner
their bulk. Because such
mesoporous
by pillared clays,
by
are dis-
tributed in
have the
a
spatially
throughout
fraction of the surface inside the bulk (be-
catalysts
overwhelming
ing microporous and crystalline),
the surface and the bulk of such solids are
effectively synonymous.
The need for new solid acid
catalysts
Acid
but the need to
mize the
have been used on an industrial scale for several decades,
catalysts
develop new
ones has
intensified in order to mini-
extensive use of corro-
recently
associated with the
disadvantages
present
sive and
such as concentrated
acids
environmentally dangerous liquids
phos-
and
acids.
reactions such as
of
phoric, sulphuric
hydrofluoric
Very strong
capable
of isobutene to
catalysing "demanding"
alkylations (e.g.
to
octane fuels or the isomerisation of C5-G, /i-alkanes
butane)
like the
yield
mentioned above, are often
on siliceous or
new solid acid cata-
and adverse
liquids
supported
clay
minerals. But there is an
need to
urgent
develop quite
that do not suffer from the environmental
lysts
disadvantages
corrosive
of such
There is also
a
need for new, intrin-
properties
liquids.
sically weaker,
acidic (and also
environmentally benign) catalysts
that
would be useful for the
of ethers and small
hydrocarbons by the
production
of alkanols under rather mild conditions.
dehydration
has been achieved
and admirable reviews exist [5—7]
adroit use of solid
Quite encouraging progress
by
the wide
heteropolyacids;
detailing
range of catalytic behaviour
that
exhibit. Moreover, some subtle
they
structural
the series of
and
variations have been introduced to
(KSbP20K, KSb2P08, K2SbPOft and
changes
compositional
phosphatoantimonates
Piffard et al. [8, 9] and others [10]. The
K3Sb3P20,4) by
advantages pos-
sessed
structures and
these two
families of materials is that
are extended
by
large
they
with our definition of uniform heterogeneous cata-
comply
Like zeolitic and other
amenable to structural
(see below)
they
are
lysts.
microporous catalysts
under in situ conditions
a
of
readily
study
by range
the
precise nature
of which has been described elsewhere [2,
techniques,
11].
Other solids,
Br0nsted acidity, have been studied in
for the isomerization of w-alkanes.
possessing strong
better
the context of
developing
catalysts
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

40
John
Meurig Thomas and
Dewi
Lewis
Wyn
these are
zirconia. For reasons that are not
Popular among
clear,
simply sulphated
sulphated
a
Zr02 surface treated with
acid
or
altogether
high-area
zirconium
sulphuric
but
one or
hydroxide
preferably containing
more additives as
are effective as
tures as low as 100°C [12-14],
(such as iron,
cobalt, nickel or
manganese,
promoters
in the isomerization of «-butane at
zinc),
catalysts
tempera-
In this communication we shall concentrate on the
of
a
development
new
of uniform
olefins. Demand for
not least because of their use in
for the conversion of meth-
family
heterogeneous catalysts
anol to
and
and
continues to soar
light
ethylene
propylene
packaging
piping.
Rational
of
a
of solid acid
design
family
catalysts
for the conversion of methanol to
olefins
light
It is well known that HF-ZSM-5 is an
exceptionally good, shape-selective
mixture of benzene, toluene
well known that A1PO
that converts methanol to
a
dehydration catalyst
and
(
[15]. It is also
xylene
equally
microporous
aluminium phosphate-based) catalysts, which possess smaller apertures,
channels and
ZSM-11,
than the
aluminosilicates such as ZSM-5 and
methanol to
olefins
to
cages
pentasil
shape-selectively dehydrate
light
up
pro-
attached
struc-
and butene. The locus of this
is the
pene
catalytic activity
loosely
proton on
tures.
the inner
of the channels of the
linings
microporous
Now Br0nsted
may be conferred upon an otherwise neutral solid
acidity
such as AlPOj in three ways:
of (a)
a
lower-valent metal
of Al(III) in to the framework, so as to form MeAlPOs with
by incorporation
in
place
Mn etc., or
a
tetravalent ion
in
Me=Zn, Mg, Co,
(b)
P(V) thereby creating SAPOs,
b), so as to yield MeASPOs [16-19]. Many distinct kinds of
structures are known in which substantial variation in the framework
(especially Si(IV))
of
or
a
combination of and
place
(c) by
(a)
A1PO
(
may
zeolites and
DAF-1 [22],
be effected [20, 21]. But even
though several "designed"
MeAlPOs have
been
reported
[19, 22-26]
recently
MeALPO-18 [19], SSZ-26 [23]. CIT-1 [24], UiO-7 [25] and STA-1
[
26]
there is still
a
need to "create" novel solid acid
compelling
(microporous)
if
because of the clamant demand for their
in
catalysts,
only
application
contexts.
petrochemical
In
the
two inter-related
be
meeting
challenge
approaches may
pursued,
based on
and we have
both of them: (i)
employed
design by experiment
chemical intuition and (ii)
Success in syn-
comes from an
from which the tem-
design by computer modelling.
acids
the first
new
thesising
microporous
using
approach
choice of constituents of the nutrient
appropriate
gel
treatment.
solid crstallizes
plated
during hydrothermal
Diisopropylamine,
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

Towards Rational Design of Solid Acid Catalysts
41
for
CoAlPO-18
[19, 27]
when cobalt acetate is
example, yields
added to
of choice
and so on. In
potential has
to investi-
mixtures of Al(OH), and
H,P04; cyclohexylamine
is the
template
to
and
new
ions for
produce CoAlPO-44;
decamethonium
of
DAF-1,
the second
a
and
approach
procedure,
growing power
evolved from the use of
the
minimisation methods
amines
energy
[28, 29]
gate
packing energies
of
organic
(such
as
triethylammonium
within distinct ALPO structures
to one
is used to build
in
for
ions)
[30]
which,
a
given
target solid,
an
appropriate algorithm
within the desired
up,
of the
cavity
target microporous catalysts, the template
molecule to
be used
in its chemical
synthesis.
We focus first on the
based on accumulated chemical intuition.
design
To make
our aim we are
out
CoALPO
to
"find"
the most
explicit
setting
appropriate
that will
yield a microporous
solid
template
(or SAPO)
such that,
after
the
active
driving away
template by calcination,
sites are formed of
the
that we have been
been able
type
tive terms as
diffraction and FTIR
independently
to describe in quantita-
a
result of in situ characterization
by X-ray adsorption, X-ray
This
Br0nsted
bound to the framwork
oxygen
acid site
that is
is, in effect,
a
detach-
[
31-33].
able
proton,
that has
adjacent to
a
Co(II) ion
isomorphously replaced Al(III) in the framework:
I
So in effect, our aim is to
both cobalt
and silicon
synthesize
con-
solid acids based on ALPO, in which the cavities are
taining
somewhat
different from those in Co-ALPO-18
and SAPO-18
[
19]
[34]. (In SAPO
a
the detachable
framework Si(IV) which has
is bound to an
which is
catalyst
proton
oxygen
next to the
isomorphically replaced P(V).)
We
it
necessary
to create cavities that were
in
judged
slightly smaller
a
one direction and
in another so as to achieve
larger
degree of shape-
that would
the methanol
selectivity
dehydration. Knowing
ALPO structure
prevent
yielding
1,4-diaminocyclohexane
the bulkier olefins
upon
rise
that
a
gives
to
template
a
we
that
solid with
porous layered
[35].
argued
microporous
1
,2-diaminocyclohex-
ane would favour the formation of
a
8-ring windows
(
i.e. eight oxygen atoms lining the window) as in the chabazitic materials
SAPO-34, CoAlPO-18 and CoAlPO-44, but
different in its
was found to be
significantly
cavity shape. But,
stable under the
because
1,2-diaminocyclohexane
un-
conditions of
synthesis, we
chose
instead the
hydrothermal
more stable
The
as the
2
-methylcyclohexylamine
template (Fig. 1).
methylcyclohexylamine template,
as well as
another, tropine (a
cycloheptanol having an N-methyl bridge
see Fig. 1), resulted in the
of the desired solids, which we have
DAF-4
production
designated
[
36]
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

42
John
Meurig Thomas and
Dewi
Lewis
Wyn
Fig. 1. Spacefilling representations of the templates used in the synthesis of DAF-4: the
unstable 1,2-diaminocyclohexane (left), 2-methylcyclohexylamine (centre) and tropine
(
right).
Fig. 2. The structure of DAF-4 with (left) the template in place and (right) after calci-
nation
Fig. 3.
the Br0nsted acid centres adjacent to the (blue) Co(II).
showing
overlay of the predicted position of the ZEBEDDE constructed amino-quinucli-
of the
A
dine analogue and the
[52]
N-methylquinuclidinium
template
experimental
The
is represented by the
positions
position
in the levyne
of the tetrahedral cations
cage.
cage
(
Al/P/Co) and the oxygens (which lie approximately midway between the vertices) are
ommitted for clarity.
Fig. 4. The growth processes involved in
in the levyne structure. Starting from
the formation of the
methane seed, the alky
1,2-dimethylcyclohexane
a
1
chain grows inside the
restricted void of the levyne structure. When an atom with
a
5th order neighbour within
a
distance of
3
A
is selected (along with the ring formation action), methyl-cyclohexane
is formed and further substitution of this
template
clarity.
then occurs. The
right, other images
of the
ring
predicted position
in the
unit cell is shown to the
have been omitted for
levyne
number
This
be
either as
a
CoALPO,
(
Davy Faraday
four).
may
prepared
with
to as many as one in five of the Al(III) framework ions
up
replaced
by Co(II), a
SAPO or as CoAPSO variants.
2
shows the
and
Fig.
forms of DAF-4 in its CoAlPO state.
of Co(II) into the framework and its retention
templated
detemplated
and
Incorporation
during
after calcination, was confirmed
our [32, 37,
38]
in situ combined X-
using
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

Towards Rational
of Solid Acid
and
43
Design
Catalysts
diffraction
made
ray absorption spectroscopy
X-ray
technique
possible
radiation. Even DAF-4
termal
rich in framework
by synchrotron
catalysts very
cobalt exhibited
at ca. 550°C. In the
high
stability
corresponding
SAPO
of DAF-4, FTIR
confirmed the occurrences
at
analogue
spectroscopy
of
the
Si-substitution,
bridging hydroxyl stretching frequency occurring
3
580 cm". Both the CoALPO and SAPO (DAF-4)
are active for
catalysts
the conversion of methanol to
olefins with selective
dimethyl
of
light
production
ethene and
some
ether, methane and traces of
propene. Although
butene are also formed, more than two thirds of the
product gases
pro-
1.29)
duced
consist of
at 350 °C with 77
and
conversion (WHSV
percent
typically
in the ratio of
three to one.
mineral
ethylene
propylene
roughly
DAF-4 is
akin to the
structurally
naturally occurring
levyne
(
idealized formula Cay(All8Si3ftO108] 50 H20) which is isotypic with some
other, previously described synthetic minerals notably ZK-20 [39], SAPO-
[20], and NU-3 [20].
35
for the
of
microporous solids
Designing templates
de novo
synthesis
using a
computer-modelling technique
In
a
Lewis et al.
would favour the
that
1,2-
production
companion computational study,
[30] predicted
disubstituted
of
a
cyclohexane templates
possessed by
structure that is
DAF-4
levyne
catalysts using computational
de novo molecular
methods.
design
De novo
have been
active materials
for some time in the search
ligand design
applied
for
see Refs. [40—46] for further details and
biologically
These
allow
a
molecule to be
so as to fit
computational
applications.
techniques
"grown"
into
a
The molecule is
favourable interactions between the substrate and the tar-
in
a
accurately
targeted space.
grown,
sense, to promote
But
to the
and
ill-defined nature of
are often made, such as
get cavity.
biological
defining
owing a
complex,
frequently
active sites, number of
approximations
the
target as property (CoMFA) map, [43, 47] restricting growth
a
to
units
and
dense
molecules
of atoms into
linking
[43—45]
annealing
arrangements
for the
suitable molecules [48].
of
Template
required
production
solids are
much
and smaller than
microporous inorganic
usually
simpler
those
in
contexts. Furthermore, the
over the
host (the
target
required
biological
microporous material)
has
many advantages
organic cavity. Firstly,
it is
in nature, and can therefore be
described atomis-
crystalline
readily
even
structures can be defined in this
tically;
hypothetical
way. Secondly,
the frameworks
reasonable
so that conformational
possess
rigidity
changes
in the
pore
sent in the
need not in
templating
be considered.
the interactions
pre-
general
Finally,
of
materials are dominated
van der
microporous
by
Waals interactions [40, 50]. Interactions
involving charge and hydrogen
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

44
John
Meurig Thomas and
Dewi
Lewis
Wyn
are less
are
in
in the
the final structure formed, al-
bonding
important
determining
gel chemistry that is an essential precursor
though they
significant
of the
crystalline product.
We have
recently developed [30] a
new
ZEBEDDE
code,
investigate
(ZEolites By
of de novo
De-novo
to
the
Evolutionary
DEsign),
application
to
for
materials. The
techniques
designing templates
microporous
target
is
space (the framework)
in which
with
a
molecule (the
template)
fragments
is to be
grown
from which the
the user,
a
of
supplied by
along
library
is to be constructed. The
selected actions until
tive fit of the molecule in the framework, is satisfied. We
then grows
a
number of
template
template
by
a
cost function, which measures the effec-
randomly
the
exploit
crystal-
line nature of the framework,
the
of the
to
by allowing
symmetry
template
be defined and for the
growth
to occur under
conditions.
periodic boundary
The actions available allow: (i) the addition of new atoms,
by selecting
molecular
plate; (ii)
from
a
and
them to the
tem-
fragments
library
bonding
multiple fragments
existing
rotation of
or
around the new bonds
single
formed; (iii) the displacement and rotation of the growing molecule within
the framework; (iv) the formation of
from chains; and (v) the
optimi-
rings
zation of the molecule
energy minimization, either in isolation or
using
within the reference framework.
these
the
is
func-
Throughout
actions,
growth
controlled
tions. Further details are
Our initial
the cost function and additional van der Waals
by
overlap
elsewhere [30].
given
of this new
have
to be
applications
methodology
proven
in
the effectiveness of the method for
fruitful, firstly demonstrating
generat-
identified
and also in
a
new template
ing previously
templates
predicting
which as mentioned earlier has been shown
to be an effective
independently
for the
framework [36]. We have considered
a
template
targeted levyne
number of frameworks, which exhibit different
features
topological
cages,
channels,
in our evaluation of the method and here
intersecting
pockets
we
a
number of these
of the
highlight
applications.
In our
we first
to
study
levyne structure,
attempted
"reproduce"
the
used
1-aminoadam-
experimentally,
zeolite NU-3 [51]
relatively complex cyclic templates
antane and
to form the
N-methylquinuclidium
which is isotypic with levyne [20]). This was accomplished in two ways.
First, by including adamantane and quinuclidine in the fragment library
we could test whether our could insert such
synthetic
(
algorithms
successfully
large
into
a
confined
we wished to see if such com-
fragments
plex fragments
verts chain molecules into
region. Secondly,
could be formed
our
which con-
by
ring-formation strategy
nth order
if
rings, by bonding
neighbours
they
are within
these
a
suitable distance (n
5 forms 6-membered
Each of
rings).
of both
successful and close
approaches proved
was
analogues
experimen-
tal
of and
3
shows the
location
templates
generated
Fig.
experimental
2-methylquinucline generated by
were
[52]
the
of
and
N-methylquinuclidium
code. In addition
a
number of other
cyclic templates
generated,
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

Towards Rational Design of Solid Acid Catalysts
45
which
was considered to be an excellent candidate
Other workers in our
1
,2-dimethylcyclohexane
its
given
high binding energy [28].
laboratory indepena-
to form
used an amino derivative,
dently
2-methylcyclohexylamine,
CoAlPO, designated DAF-4, which possesses the levyne
structure (see
4
illustrates how this
was
our code.
above). Fig.
template
generated by
To test further the trustworthiness of our methodology, we attempted to
for ZSM-5 that has
a
three-dimensional
successful and
channel
number of
in
grow templates
system. Again,
intersecting
the method
a
proved
generated
the
of
known to form this. Simulations of
variety
templates
templating
this structure allowed us to validate the
within the host.
of
importance
specifying template
such as
symmetry
Large templates,
tetrapropylammonium,
tend to be located at the channel
and
the
of
intersection,
specifying
growth
one
at each intersection encouraged the formation of
template
quaternary
amines and similar
if the
sym-
(
e.g. tripropylamine) templates. Similarly,
metry were specified
so as to favour
a
concentration, then
higher template
smaller, linear templates (such as n-diamines), located in the channels, are
formed. Further details of these simulations are
given elsewhere
[30].
We are
further our
this
number of
with results to date. In addition to
to test
en-
currently applying
computer modelling technique
in
a
and are
to
methodology
systems
a
particularly
couraged
program
generate templates
for new structures, we are also
on
a
number of structures where
focusing
new
will lead to the formation of new compositions and improved
templates
control.
A
notable case here is the chabazite structure, the
morphological
of which is known to
strict control of
concen-
synthesis
require
template
tration [29]. Here, we are
to grow an
which
attempting
optimal template,
not
also
a
lower concentration (i.e. one template per unit cell), but
only requires
a
lower
charge density.
In this way we can access MeAPO
possess
with lower metal content than
accessible.
highly localized charge, then we
compositions
if we can construct
be able to control the
currently
Similarly,
a
with
a
template
siting
of the substituent metal in the
may
framework.
resulting
Combinatorial and other
approaches
As hinted earlier in reference to the use of
for the
of
haptens
development
antibodies
science is
a
deal further forward than
catalytic
biological
good
the
In
science in
such
materials with desired catalytic properties.
physical
designing
are often identified by the synthesis and screening
species
biology,
of
collections, or libraries, of structures. If we reflect
the fact
upon
large
that the immune
can
and screen some 1012 antibody mol-
system
generate
ecules to
one that
and binds
a
identify
specifically recognized
foreign patho-
one realizes the immense potential that such
a
were it trans-
gen,
system,
ferred to other
situations, would have. One of the first applications
catalytic
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

4
6
John
Meurig Thomas and
Dewi
Lewis
anti-
Wyn
of molecular libraries to
bodies [53]. Soon, thereafter, there
was the
appeared a
of
of
chemistry
development
catalytic
number of methods for gen-
molecules in vitro for
and
for
on the
erating
binding,
pended
screening large populations
biological
or for both [54]. Great effort is
ex-
catalysis
currently being
of these combinatorial libraries to the
application
discovery
of new
(Combinatorial
is
a
novel
of
drugs.
chemistry
way
rapidly generating
a
number of related
The
so-called, solid
thirty years ago
large
compounds.
elegant phase
of
devised
Merrifield some
is often
synthesis
peptides
starting point
combinatorial
by
taken as
a
for the
of combinatorial chemical libra-
production
ries [55]).
A
to the
of new solid-state
recently been de-
approach
discovery
materials with novel
has
super-conducting properties
scribed
It is
in
et al. [56].
by Xiang
to declare whether such an
will
prove as
useful
premature
approach
as it
has in the realms of
antibodies. But ways must
of combinatorial
inorganic chemistry
undoubtedly
peptide, oligo-
nucleotide and
be
of har-
sought
catalytic
surely
nessing the general principles
as to allow
Until such
ways
a
chemistry
of the structure of new
priori prediction
inorganic catalysts.
are found, confluence of
and intuition as one stream and of
experience
as another, must be
to the fullest so as to
that have hitherto held
computer-modelling
exploited
eliminate much of the Edisonian
approaches
sway
in the
of many
discovery
inorganic catalysts.
Acknowledgement
We
financial
DWL).
from EPSRC (a grant to JMT and
rolling
acknowledge
support
a
ROPA award to
We are also
to Prof. C.R.A. Catlow and
grateful
Drs. G.
for many
Fig. 2.
Sankar, P. A. Wright,
J.
D. J. Willock and Mr. P. Barratt
Chen,
discussions. Dr. R. G. Bell is thanked for
stimulating
producing
References
1
.
J. M. Thomas,
Chem. Int. Ed. 27
1673.
Angew.
(1988)
2.
3.
4.
5.
J. M. Thomas, ibid. 33
913.
(1994)
J. M. Thomas, Phil. Trans.
Soc. A. 333 (1990) 173.
Roy.
J. M. Thomas and K. I. Zamaraev, Angew. Chem. Int. Ed. 33 (1994) 308.
Y. Ono, in Perspectives in
(ed. J. M. Thomas and K. I. Zamaraev),
Catalysis
Blackwells, Oxford (1992) p. 431.
M. Misono, in New Frontiers in Catalysis (ed. L. Guczi et al.) Elsevier, Amsterdam
6.
(
1993), p. 69.
7.
V. Kozhevnikov, Catal. Rev.-Sci. Eng. 37 (1995) 311.
8
9
.
.
Y. Piffard, A.
and M. Tournox, Rev. Chim. Miner. 22 (1985) 101.
Lachgar
Y
Piffard, A. Verbaere, S. Oyetola, S. Courant and M. Tournoux, ibid. 23 (1986)
766.
1
0. S.
and J. M. Thomas, Catal.
Discuss 100
12 (1992) 433.
Natarajan
Today
1
1. J. M.
c9.
Thomas, Faraday
(1995)
1
2. V. Adeeva, G. D. Lei and W. M. H. Sachtler. Appi. Catal. A, 118 (1994) LI 1.
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

Towards Rational
of Solid Acid
47
Design
Catalysts
1
3. M. A. Coelho. D. E. Resasco, E. C. Sikabwe and R. L. White. Catal. Lett. 32 (1995)
53.
2
1
1
1
4. F. C. Lange, T.-K. Cheung and B. C. Gates, Catal. Lett., in
press.
5. P. B. Weisz, Proc. 7th Int. Cong. Catalysis, Tokyo (1980), vol. 1, pi.
6. J. M. O. Lewis and J. B. Price, US Patent 4,873,390 (1989).
1
7. T. Inui, S. Phatanasri and H. Matsuda, J. Chem. Soc. Chem. Commun. 1990, 205.
1
8. J. M. Thomas, Y. Xu, C. R. A. Catlow and J. W. Couves, Chem. Mater.
3
(1991)
667.
1
9. J. Chen and J. M. Thomas, J. Chem. Soc. Chem. Commun. 1994, 603.
2
0. D. H. Olson and W. M. Meier. Atlas of Zeolite Structure Types, Butterworth-Heine-
mann, London 1992.
2
1. J. M. Thomas, C. R. A. Catlow and R. G. Bell, "A
Guide to the Structures
Catlaysts" in Handbook of Heterogeneous Catalysis (ed. G. Erti, H.
Synoptic
of
Microporous
Krozinger and J. Weitkamp), VCH, in press.
2
2. P. A.
R. H. Jones, S. Natarajan, R. G. Bell, J. S. Chen, M. B. Hursthouse
Wright,
and J. M. Thomas, J. Chem. Soc, Chem. Commun. 1993, 633.
2
3. S. I. Zones and D. S. Santilli, in 9th International Zeolites Conference (ed. von
Ballmoos. R., et al.), (Buuterwirj-Heinemann, Montreal 1992), 171.
4. R. F. Lobo and M. E. Davis, J. Am. Chem. Soc. 117 (1995) 3766.
5. D. E. Akporiaye, H. Fjellcag, E. N. Halvorsen, J. Husetveit, A. Karlsson and K. P.
LiUerud, J. Chem. Soc, Chem. Commun. 1996, 601.
2
2
2
2
6. P. A. Wright et al., in
preparation.
7. J. Chen, J. M. Thomas, R. P. Townsend and C. M. Lok. UK Patent
Application
8. D. W. Lewis, C. M. Freeman and C. R. A. Catlow, J. Phys. Chem. 99 (1995) 11194.
93186443 (1993).
2
2
3
9. D. W. Lewis, C. R. A. Catlow and J. M. Thomas, Chem. Mater.
8
(1996) 1112.
0. D. W. Lewis, D. J. Willock. G. J. Hutchings, C. R. A. Catlow and J. M. Thomas,
Nature 382 (1996) 604.
3
1. J. M. Thomas, G. N. Greaves, G. Sankar, P. A. Wright. J. S. Chen. A. J. Dent and
L. Marchese, Angew. Chem. Int. Ed. 33 (1994) 1871.
3
3
2. J. M. Thomas and G. N. Greaves, Science 265 (1994) 1675.
3. L. Marchese, J. S. Chen. J. M. Thomas, S. Coluccia and A. Zecchina. J. Phys. Chem.
98 (1994) 13350.
3
4. J. S. Chen, P. A. Wright, J. M. Thomas, S. Natarajan, L. Marchese, S. M.
Bradley,
G. Sankar, C. R. A. Catlow, P. L.
Phys. Chem. 98 (1994) 10216.
R. P. Townsend and C. M.
Lok,
J.
Gai-Boyes,
3
3
5. P. A. Barrett and R. H. Jones, J. Chem. Soc. Chem. Commun. 1994, 10216.
6. P. A. Barrett. R. H. Jones, J. M. Thomas, G. Sankar, I. J. Shannon and C. R. A.
Catlow, J. Chem. Soc. Chem. Commun., in press.
3
7. J. W. Couves, J. M. Thomas, D. Waller, R. H. Jones, A. J. Dent, G. E. Derbyshire
and G. N. Greaves, Nature 354 (1991) 465.
3
8. G. Sankar, P. A. Wright, S. Natarajan, J. M. Thomas, G. N. Greaves, A. J. Dent, B.
R. Dobson, C. A. Ramsdale and R. H. Jones, J. Phys. Chem. 97 (1993) 9550.
3
4
9. G. T. Kerr. U. S. Patent 3,49,676 (1969).
0. R. S. Bohacek and C. McMartin, in Computer-Aided Molecular Design. Applications
in
Materials and Pharmaceuticals (eds. C. H.
M. K. Hol-
Agrochemicals,
Reynolds,
loway and H. K. Cox), American Chemical Society, Washington D. C, 1995, pp. 82.
4
4
1. V. Gillet, A. P. Johnson, P. Mata, S. Sike and P. J. Williams, J. Mol.
Comput-Aided
7
(1993) 127.
Design
2. T. I. Oprea, C. M. W. Ho and G. R. Marshall, in Computer-Aided Molecular Design.
Applications in Agrochemicals, Materials and Pharmaceuticals (C. H. Reynolds, M.
K. Holloway and H. K. Cox), American Chemical Society, Washington D.C., 1995,
65.
pp.
3. R. S. Bohacek and C. J. McMartin. J. Am. Chem. Soc. 116 (1994) 5560.
4
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM

48
John
Meurig Thomas and
Dewi
Lewis
Wyn
4
4
4
4
4
4. H. J. Bohm, J. Comput-Aided Mol. Design
6 (1992) 61.
5. R. A. Lewis and P. M. Dean, Proc. R. Soc. Lond. (Biol.) 236 (1989) 125.
6. C. Verlinde, G. Rudenko and W. Hoi, J. Comput-Aided Mol Design
6 (1992) 131.
7. R. Cramer III, D. Patterson, and J. Bunce, J. Am. Chem. Soc. 110 (1988) 5959.
8. A. Miranker and M. Karplus, PROTEINS: Structure, Function, Genetics 23 (1995)
472.
4
5
5
5
5
5
5
5
9. H. Gies and B. Marier, Zeolites 12 (1992) 42.
0. H. Gies, Stud. Surf. Sci. Catal. 85 (1994) 295.
1. A. Stewart, Zeolites
9
(1989) 140.
2. L. B. McCusker, Mater. Sci. Forum 133-136 (1993) 423.
3. R. A. Lerner, S. J. Benkovic and P. G. Schultz, Science 252 (1991) 859.
4. P. G. Schultz and R. A. Lerner, Acc. Chem. Res. 26 (1993) 391.
5. G. Lowe, Chem. Soc. Rev. (1995) 309.
6. X. D. Xiang, X. D. Sun, G, Briceno, Y. L. Lou, K. A. Wang, H.
Y
Chang, W. G.
Wallace-Freedman, S. W. Chen and P.
G
Schultz, Science 268 (1995) 1738.
Brought to you by | University of California
Authenticated
Download Date | 12/25/15 8:05 AM