J Solid State Electrochem (2002) 6: 359±360 DOI 10.1007/s10008-001-0256-1
B OO K R E V I EW
Walter Langel
T. Helgaker, P. Jorgensen, J. Olsen, T. Helgaker: Molecular electronic-structure theory Wiley, New York, 2000. 938 pp (ISBN 0-471-96755-6) US$ 300.00 Published online: 17 November 2001 Ó Springer-Verlag 2001
Computers change our approach to chemistry. The classical chemist worked hard in the laboratory and left theory to others, since he could not see much relevance in it. On the other hand, physicists thought that they had understood all fundamentals of chemistry, but that trying to address its practical problems was not very fruitful. This attitude is summarized in the famous Dirac statement: ``The underlying physics necessary for the mathematical theory of a large part of physics and the whole of chemistry are thus completely known, and the diculty is only that the exact application of these laws leads to equations much too complicated to be soluble''. Nowadays a major percentage of experimental chemical papers refers to calculations supporting the proposed structures and reaction mechanisms. This is due to the development of ecient computing methods as well as of the performance of computers. Thus the demand for supercomputers is now supported to a major part by chemists. However, computers would not be of much help without the availability of a whole range of theoretical approaches consisting of dierent compromises between precision of the result on the one side and size of the system and scope of the evaluated properties on the other. This spectrum starts at highest precision relativistic calculations for metal dimers, passes via systems of a few hundred atoms, which still can be calculated ab initio in reasonable approximation, and extends to large force-®eld calculations for biomolecules which nowadays obtain their essential parameters from quantum chemistry. Its sophisticated algorithms are not yet covered by many textbooks. Classical physical chemistry usually treats the SchroÈdinger equation for small systems such as the hydrogen molecule, but not the approaches to
W. Langel E.-M.-Arndt UniversitaÈt, Institut fuÈr Chemie und Biochemie, Soldmannstrasse 23, 17489 Greifswald, Germany E-mail:
[email protected]
multielectron systems. Theoretical physicists view the subject from a more general point of view and do not go into the details of chemical calculations, whereas texts on molecular modelling often avoid a very strict mathematical formulation. One still has to learn a lot from the handbooks and web pages for programs, but not everybody wants to use the packages as ``black boxes''. The intention of the present book is to strictly derive current computational chemistry from the principles of quantum mechanics. The presentation of the material is very logically structured, starting with Fock matrices and the second quantization (chapters 1 and 2). Chapters 3 and 4 describe the handling of wavefunctions in general (orbital transformations, expectation values and Hellman Feynman forces). A major part of the book (chapters 5±9) is devoted to the development of speci®c basis sets for chemical applications and their integration. At this stage the book goes far beyond typical theoretical physics. In chapters 10 and 14 the Hartree-Fock approximation and its most common extension, the Mùller-Plesset perturbation calculation, are introduced. Here most texts on molecular modelling would stop, since the Hartree Fock method is still a sort of workhorse for the computational chemist. Chapters 11±13 show, however, that this method only yields a ®rst trial function for more serious multicon®guration approaches such as con®guration interaction and coupled cluster theories. A ®nal chapter (15) deals with the calibration of the results. According to its title, this part is written from the perspective of the theorist who wants to demonstrate the capabilities of his methods, and does not focus as much on the interpretation of measured data as could be desired by the experimentalist. The book aords a profound knowledge of basic quantum mechanics as is acquired in theoretical physics courses. It will primarily be of interest for scientists who want to develop or implement new methods, but not so much for the common user of Gaussian or another package. The mathematical overhead is well explained, and even some elementary mathematical subjects (Taylor series of sine and cosine) are treated in broad
360
and very instructive exercises. The text addresses those who want to understand the methods in depth, but this will be prohibitive for readers who just want to be informed on the state of the art or even for graduate chemistry students, unless they are planning to join a theory group. As also the price is fairly high, this is a book for the library and for the theoretical chemistry department.
Many people in the community would appreciate a volume II, treating density functional theory (DFT) with similar explicitness. In combination with plane wave basis sets, DFT now goes much further into the description of laboratory experiments than CI and even HF, but textbooks closing the gap between simple program manuals and pure theory texts are even more seldom.
