close

Вход

Забыли?

вход по аккаунту

?

Book Review Has Quantum Chemistry Grown Up A Handbook of Computational Chemistry. By T. Clark

код для вставкиСкачать
BOOK R E V I E W S
Has Quantum Chemistry Grown Up?
This question is prompted by the appearance of the following two new books:
A Handbook of Computational Chemistry. By T. Clark.
Wiley, Chichester 1985. 332 pp., bound, X 35,80.-ISBN
0-471-8821 1-9.
Semi-empirical Methods of Quantum Chemistry. By J .
Sadlej. Ellis Horwood, Chichester 1985. 386 pp., bound,
X 42.50.-1SBN 0-85312-167-2
Instead of giving a conventional review of these books, I
shall try here to examine the development of quantum
chemistry since its emergence, and to consider its present
state and its continually changing relationship to “established” chemistry.
Quantum chemistry is the application of quantum mechanics to chemistry, even though the activities in which
quantum chemists are engaged can vary quite widely. Does
a theoretical statement become a useful contribution to our
knowledge only when it is expressed as a mathematical
formula, or d o numbers that are arrived at by less explicit
methods, and yet are real in terms of our observations, also
have a knowledge value? This question can be put more
succinctly as follows: d o we want quantum chemistry to
provide generalized statements, o r are its most useful results those that are more specific to particular substances?
At present quantum chemistry is well on the way to adopting a more sceptical attitude towards broad generalizations. Thus, for example, benzene now seems to be the
only molecule which is truly aromatic, although some definitions of aromaticity are quite unrelated to material
properties, but are of a topological nature, and are therefore shown to be inadequate. Again, the Hiickel4n 2 and
4n rules now show more exceptions than regular cases, and
elements which are said to be homologous are often radically different. Quantum chemistry today ought not to impose limitations on the scope of its applications, e.g. by
avoiding the study of compounds not yet prepared synthetically, but instead should test familiar general statements
more widely and, where necessary, re-express them in a
more restricted form. Theory is thus approaching closer to
chemical reality, as it concerns itself more with statements
related to particular substances, and rejects over-generalized statements when required. These observations allow
us to formulate an up-to-date definition of applied quantum chemistry as the application of computational methods, with the help of a computer, to give data o n structures
and energies.
We can now see that the question posed in the title can
hardly be answered by a simple yes or no. What criteria
can be used to judge the development progress of this
scientific discipline?
+
0 In the preface to Clark’s book Paul von Rague Schleyer
writes: “The chemist now has incredibly powerful mathematical tools at his disposal. These tools are simple to use.
. .. The future directions of chemical research are already
programmed!” The physicist Dirac was moved to express a
rather similar hymn of praise in 1929 when he first encountered the Schrodinger Equation. Predictions of this sort
158
were also made during the sixties, when quantum chemistry had passed through its critical first phase of development and become established as a chemical discipline. The
essential difference now, though, is that the statements
come from chemists, showing that one of the aims of quantum chemistry, namely to become accepted as an integral
part of chemistry, has been achieved.
0 The first book published on quantum chemistry-and
in fact the expression itself-came in 1937 from Hans Hellmann (H. Hellmann: Einfiihrung in die Quantenchernie,
Deuticke, Vienna 1937). If one could ask the author for his
view of quantum chemistry today, one might expect a reply
something like the following: quantum chemistry is now in
its adolescence-much grown in size, enormously powerful, but intellectually immature. Thus, for example, a
chemist using a computer program for configurational interaction (CI) calculations, which relies on Brillouin’s theorem, does not need to have this theorem in mind each
time he does so, nor even ever to know that he is using the
theorem. In its application quantum chemistry today is for
that very reason a matter of volition rather than of knowledge and ability. Is this a situation that is appropriate to
the academic ideal?
0 This criticism of the process of growing u p can, however, be seen in a different way. In the book by Clark an
attempt is made for the first time to separate science from
technology. From being originally the preserve of a few
physicists with a mathematical background, the application of quantum chemistry now no longer calls for an expensive mathematical training, and it is therefore accessible to the whole of the chemical community. This situation
reminds one of Coulson, who several decades ago was bold
enough to venture into chemistry, and successfully replaced the mathematical formalism of quantum mechanics
by “primitive patterns of understanding” (C. A. Coulson:
VaZence, Clarendon Press, Oxford 1952).
Clark mainly sticks to his stated intention of not writing
a book for theoreticians, which is fortunate, since those
brief attempts at theoretical explanations that have been
inserted into the text have a confusing effect. Not much
can be made of statements like “the Walsh orbitals in cyclopropane are responsible for much of the unusual chemistry of three-membered ring systems”, “the Born-Oppenheimer approximation is a very good one, except in the
case of extremely flat potential surfaces”, or “in the C I
treatment there is mixing of electronic states”. The section
on electron correlation, though, is unforgivable, where it is
stated, and illustrated by a diagram, that in the CI treatment the wave function of a singly excited singlet state is
mixed with that of the S C F ground-state! What has become of Brillouin’s theorem in this case? Inevitably at this
point one has misgivings that the knowledge of technology
alone might open the way to its misuse.
How d o the facilities available for applying quantum
chemistry now look? Essentially three programs are described, for which the manner of feeding in the input and
the interpretation of the output are described with numerous examples, namely:
1) classical force field calculations
(molecular mechanics);
Angew. Chem. In[. Ed. Engl. 26 (19871
No. 2
2) semi-empirical MO procedures
( M I N D 0 and MNDO);
3) ab initio procedures
(GAUSSIAN 82).
The fact that such a wide range of possibilities exists
may be taken as an indication of the increasing maturity of
quantum chemistry, for only a few years ago it would not
have been possible to bring together in a single book computational procedures based on such differing philosophies. The book will be of interest to all chemists who are
interested in molecular structure and energetics.
Semi-empirical methods of calculation have made a contribution to the development of quantum chemistry whose
importance can scarcely be overestimated. The great heyday of semi-empirical methods occurred in the sixties, and
therefore it is hardly to be noticed that the second of the
books under review, by Joanna Sadlej, should describe the
state of the topic in 1978. It has apparently taken seven
years for the book to be translated from Polish into English. During that time the following developments have occurred :
1) the C N D O method, and the procedures derived from it,
2)
3)
4)
5)
apart from MNDO, have practically disappeared from
the journals;
studies of molecular structures and energies of reactions
have become much more common than the 40 pages
here would suggest;
semi-empirical calculations of spectroscopic quantities
have declined to such an extent that the 95 pages occupied in the book now seem far too many;
C N D O calculations on hydrogen bonding are now recognized to be basically incorrect, so that the chapter of
38 pages is now superfluous;
the size of compounds for which semi-empirical calculations are indispensable is now considerably larger.
Unhappily the one method which might have been kept
unaltered from 1952, and still retains its standing today,
namely the extended Hiickel theory (EHT), is given only
one page in the appendix.
Despite these deficiencies, however, Joanna Sadlej
shows in her book in a very impressive way the importance
of semi-empirical methods in the development of quantum
chemistry, particularly through the many examples and
their associated literature references. Whereas many of the
early books on semi-empirical methods reflect mainly the
personal views of their authors, Sadlej has tried-successfully for the most part-to remain objective, so as to produce as complete a work as can be achieved by a single
author. It is unfortunate that this care is to some extent
wasted, since all that this book contains; and a great deal
more besides, is contained in volumes 7 and 8, both edited
by G . A . Segal, in the series Modern 7heoretical Chemistry
(Semiempirical Methods of Electronic Structure Calculations, Plenum Press, New York 1977).
The relationship of quantum chemistry to physics and
chemistry, and that between the alternative approaches of
ab initio and semi-empirical methods, can best be discussed by considering Table 1, which lists the annual totals
of papers on quantum chemistry in several journals; the
figures in parentheses give the number of papers on a b initio methods included in the total.
In the years after 1960 the numbers of quantum chemistry papers in physics journals increased sharply as a result
of computer developments. They reached a maximum in
Anyen,.
Chem. Inr. Ed. Engl. 26 (1987) No. 2
Table I
Year
J . Chem. Phys.
I960
1961
1962
1963
1964
1965
1966
1967
88
1
136
217
244
292
288
209
252
0
1985
60
J. Am. Chem. SOC.
1
14
I
19 (1)
23
18 (4)
145 (121)
Angew. Chem.
0
I
0
1
2
3
2
2
6
the middle of the sixties, and at the same time papers on
these topics began to appear in chemical journals, even in
those published in countries where a strong traditionalism
in chemistry is said to exist. This breakthrough was mainly
due to the work of Dewar and Hoffmann. Quantum chemistry was at that time divided into two camps, the physical
(ab initio methods) and the chemically oriented (semi-empirical methods). Papers giving results obtained by a b initio methods were accepted only hesitantly in the chemical
literature from the mid-sixties. Now, twenty years later,
quantum chemistry work is carried out to a considerably
greater extent in chemical than in physical research
groups, and semi-empirical methods play only a secondary
role, the only appreciable contributions in 1985 for the
three journals of,Table 1 being 17 on M N D O and 7 on
EHT methods. Two of the arguments put forward in the
past for using semi-empirical methods, now becoming less
valid, are that “qualitative results are often sufficient”, and
that “calculations on large molecules call for drastic approximations”. Meanwhile a b initio calculations have
been successfully applied to the spherical C60 molecule
and to a D N A fragment with three stacked base pairs. The
trend in quantum chemistry is plainly towards obtaining
numerically reliable results from a b initio calculations.
The development of quantum chemistry is undoubtedly
closely linked to that of computers. However, to avoid the
possibility of any misunderstanding with regard to quantum chemistry reaching its maturity, this attempt at a study
of its historical development will conclude with a quotation from Coulson: “Our understanding ... must be in
terms of simple physical models which can readily be visualized, and not in terms of numbers gushing from a computer. If our concepts and patterns of understanding are
sound they will be supported by rigorous calculation; if
not they must be rejected.”
RudolfJanoschek [NB 775 IE]
Institut fur Theoretische Chemie
der Universitat Graz (Austria)
Medicinal Chemistry Research in India. By H. Singh. A . S.
Chawla, and V . K . Kapoor. National Information Centre
for Drugs and Pharmaceuticals, Lucknow (India) 1985.
184 pp., bound, Rs. 150.00/$35.00.-1SBN 81-85042-004
Medicinal chemistry is a line of research which has been
pursued more intensively in India since independence.
This book gives a survey of Indian research in this field, in
twenty concise chapters which cover published work u p to
1985.
An introductory chapter describes the achievements of
Indian pharmaceutical research, six of its drugs having
come on to the market. The following sixteen chapters are
159
Документ
Категория
Без категории
Просмотров
0
Размер файла
268 Кб
Теги
chemistry, clarke, book, quantum, handbook, review, computational, grow
1/--страниц
Пожаловаться на содержимое документа