<<

22.4 N&V 807 MH 16/4/04 5:01 pm Page 816

news and views

Obituary only a minor player (the dominant contributor being , with John A. Pople (1925–2004) whom Pople shared the ), he systematized the way such calculations Sir died on 15 March 2004 at were carried out and incorporated them the age of 78. He was a giant in his chosen into the program. This field, computational quantum , undoubtedly helped significantly in COOMER/AP

for which he was awarded the Nobel Prize accelerating the widespread acceptance B. in Chemistry in 1998. Pople revolutionized and use of density functional theory in the way chemistry is practised today by chemistry. making it possible to carry out research Computational is in the computer as a complement to the used today by myriad across a conventional chemistry of the laboratory. variety of fields. It has become a viable It was Pople who was most responsible adjunct to experiment, and is used in for making computational quantum solving fundamental problems. It is also chemistry available to the community being increasingly applied by industrial of chemists at large, and he dominated companies in more practical situations, the scene in this area during the past such as the design of new drugs and new five decades. materials. It is especially valuable in Pople was born in Burnham-on-Sea in studying substances that might be difficult south-west in 1925. His talents to examine experimentally, for instance in mathematics were apparent at an early because they have a very short lifetime or age. However, he describes in his Nobel are toxic or explosive. The computer is autobiography how he introduced oblivious to such hazards. deliberate errors into his mathematics Innovator in The success of computational exercises at Bristol Grammar School so as chemistry has of course been helped by not to appear too clever. It was not until a computational massive and continuing increases in new mathematics teacher arrived, and set chemistry computer power. The computer available a particularly challenging test, that he to Pople in 1970, when Gaussian 70 was succumbed to temptation and turned in a example, it should be unique, well defined, born, was a CDC 1604. A standard personal perfect paper, including multiple solutions unbiased, objective and widely applicable, computer today is roughly 100,000 times to several of the problems. Despite the as well as satisfying certain, more faster, has 5,000 times as much memory, remarkable achievements that were to technical, requirements (such as being and yet costs several hundred times less. follow, he never lost his modest manner. ‘size-consistent’). If the model performed Apart from being a brilliant researcher, Pople went on to study mathematics satisfactorily in systematic comparisons Pople was also a great communicator, as and at the University of with experimental data, it could then be anyone who heard him lecture will know. Cambridge. But eventually his interest in used to make predictions in cases where Some people have the knack of making pure mathematics began to wane and he experimental data are not available. simple things complicated. Pople kept decided to apply his mathematical skills Pople then proceeded to design a simple things simple, and had the gift of to chemistry. The germs of the ideas that series of theoretical procedures that could making complicated material appear were to provide the focus for his life’s serve as the basis for the model. In some simple as well. work — developing mathematical models cases, he was the inventor or co-inventor Virtually all of Pople’s prizewinning that could describe all of chemistry — of the new procedures — for example, the work was carried out while he was a were sown in 1952. There were some Pariser–Parr–Pople - theory professor at Carnegie–Mellon University digressions, including early seminal of the 1950s, or ‘complete neglect of in , Pennsylvania, from 1964 contributions and a classic text in the differential overlap’ (CNDO) and other to 1993, although it continued after he then-emerging field of nuclear magnetic ‘semi-empirical’ theories of the 1960s. moved to in resonance, work that was a forerunner to In others, he reformulated existing ideas, Evanston, . He also had close the magnetic resonance imaging of today. such as Møller–Plesset perturbation connections with the Australian National But the main game was always the creation theory, to make them practically and University in Canberra, which he described of models for studying chemistry. objectively applicable to a wide cross- as his second academic home, and which It was known long ago, and enunciated section of chemistry. he visited on nine occasions during the by Paul Dirac in 1929, that the laws of A key to implementing such procedures 1980s and 1990s. His death is a great loss could in principle was the construction of efficient computer not only for science but also for his many be used to predict all of chemistry. What programs, and this was one of Pople’s friends and colleagues. Pople did was convert this to a practical major contributions. His development and John Pople was married to Joy Bowers reality. His aim was to enable chemists to release of the Gaussian 70 program marked in 1952. They shared a close relationship straightforwardly predict the properties of a turning point in the field, making it for just under 50 years before Joy passed molecules, such as molecular structures possible to carry out broad theoretical away in 2002. He is survived by his four and the way molecules react with one studies on real chemical problems on a children, 11 grandchildren and one great- another, by using a computer rather than scale that had not previously been possible. granddaughter. by carrying out experiments. This was also the hallmark of Pople’s Leo Radom is in the School of Chemistry, He went about this task by first subsequent computational initiatives. Even University of Sydney, Sydney, formulating the essential characteristics in the more recently popularized density New South Wales 2006, Australia. of an acceptable theoretical model. For functional theory, where he was formally e-mail: [email protected]

816 NATURE | VOL 428 | 22 APRIL 2004 | www.nature.com/nature © 2004 Nature Publishing Group