omputin J Froese Fischer I OUG1AS AYNER AKTREE His Life in Science and Computing This page is intentionally left blank OUG1AS AYNER AKTREE His Life in Science and Computin; Froese Fischer Vanderbilt University, USA \HJP World Scientific NEW JERSEY • LONDON • SINGAPORE • SHANGHAI • HONGKONG • TAIPEI • BANGALORE Published by World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, Singapore 596224 USA office: Suite 202, 1060 Main Street, River Edge, NJ 07661 UK office: 57 Shelton Street, Covent Garden, London WC2H 9HE British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library. DOUGLAS RAYNER HARTREE: HIS LIFE IN SCIENCE AND COMPUTING Copyright © 2003 by World Scientific Publishing Co. Pte. Ltd. All rights reserved. This book, or parts thereof, may not be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage and retrieval system now known or to be invented, without written permission from the Publisher. For photocopying of material in this volume, please pay a copying fee through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA. In this case permission to photocopy is not required from the publisher. ISBN 981-238-577-0 Printed in Singapore by World Scientific Printers (S) Pte Ltd Preface Anyone familiar with quantum mechanics will have heard of Douglas R Hartree and his self-consistent field method. Though Fock pre-empted Hartree's work on equations with exchange, Hartree later extended the equations to include configuration interaction. In doing so, he played an important role in laying the foundation not only of modern day atomic physics but also of quantum chemistry. Not nearly as well known are his contributions in other areas. In fact, it has been said that Hartree only had this one idea, churning out atom after atom. This was hardly the case. Early in his career he became interested in radio wave propagation and derived the equation for the index of refraction that in atmospheric physics today is called the "Appleton-Hartree equation." During World War II, he applied the self-consistent field idea to the study of the cavity magnetron then part of radar development. Computationally, this was a far more challenging problem than atoms because the geometry of cavities and slots precluded the assumption of radial symmetry, requiring the solution of two-dimensional partial differential equations. From these computations, he made a number of discoveries, one being the "Hartree condition" for the operation of a magnetron. There were others. The common denominator of Hartree's research was differential equa tions, either ordinary or partial. He combined his wide range of interests with a talent for problem solving. Given a problem, he would express the phenomena investigated in the form of one or more differential equations, devise a scheme for solving the equations in their simplest possible form, and perform the calculations. In the 1930's this led him to the differential analyzer and its application to a variety of problems, the most important being in control theory and fluid dynamics. Hartree knew the accuracy of the differential analyzer was insufficient for most atomic structure work, VI Douglas Rayner Hartree but he believed there were many industrial applications where great accu racy was not needed and the differential analyzer would be a very welcome tool. In this he was correct. During the war, he dealt with problems that were partial differential equations, and in later years, Hartree was greatly interested in equations arising in hydrodynamics. Because he believed in solving a problem with any tool available - slide-rule, desk calculator, or differential analyzer - he acquired great computational skills and knowledge of what was later called numerical analysis. Hartree did not hide his "love of arithmetic." This was not universally greeted with admiration or acclaim. Slater, in his autobiography, classified scientists as the "hand-waving magical type" or the "matter-of-fact type." He placed Hartree in the latter category, saying that the hand-waving- magical type considered him a "mere computer," yet he made a greater contribution than most of them. So why would someone under those circumstances take such an uncon ventional stand and pursue "arithmetic?" I believe the answer lies in the unusual family background, and the Bedales school Hartree attended prior to University. Both nurtured the development of his unique talents and at Bedales he learned to consider his role in society. For this reason, the present biography starts with the Hartree family. Because the extent of Hartree's contributions is not well known, the rest of the book attempts to chronicle his science as well as his life. Many of his publications are not readily accessible today, so a summary is appropriate. Hartree did not consider himself important and did not keep correspon dence, neither with scientists nor with family members. Fortunately, some archive collections have kept Hartree's letters - Edward V Appleton, Niels Bohr, Vannevar Bush, Robert B Lindsay, Fritz London, Rudolf Peierls, John C Slater, Ivar Waller - but particularly before World War II, letters were written by hand without carbon copies so the "conversations" in many letters are somewhat one-sided. As a document of his activities, it has been necessary to rely on Hartree's own publications supplemented by letters, as available, and documents from the Public Record Office, London, relating to war time activities. Hartree spent the 1920's in Cambridge, first as an undergraduate, then as a graduate student and research Fellow at the Cavendish laboratory. During this time, Ernest Rutherford was Professor of Physics and Ralph H Fowler, a lecturer in Mathematics, supervised those more theoretically in clined. Rutherford attracted many physics students from around the world to the Cavendish. The students of these two men were the "community" Preface VII with which Hartree identified. The science world at that time was much smaller than today so the same "cast of characters" will appear in different chapters of this book, ranging from World War I, quantum theory, World War II, and the emerging computer era. In placing Hartree's early work in context, it has been necessary to outline some of the developments in quantum theory, but I do not con sider this to be a history of quantum mechanics. John Slater, who was more intimately involved in some of the theoretical ideas, has written an autobiography, Solid State and Molecular Theory: A scientific biography, which, in spite of its title, presents an account of historical developments, and could be an excellent companion for readers desiring a more detailed account still at a simple level. In discussing Hartree's science, I made the decision that, when neces sary, differential equations would be included, but often they only show the nature of the problem in a concise way. The detailed meaning is of lesser importance. His papers are not cited here, but a complete list of his publications is included at the end of the book. Finally, let me say that, as Hartree's student, I unfortunately did not get to know him well. I never attended any lectures that he presented. Shortly after my arrival at Cambridge, Hartree was getting ready for his year at Princeton. When he got back, he understandably was exceedingly busy and soon thereafter, I was getting ready to return to Canada. The subsequent year we corresponded and Hartree arranged for my dissertation defense at the University of Toronto. Six months later he died unexpectedly. Thus not much of this book is from personal recollections. This biography could have been a lot different had it been started sooner. I did contact Bertha Swirles Jeffreys and Jack Howlett in 1999, shortly before they died, getting some valuable information. I had one let ter from David Myers who had been Dean of Engineering at the University of British Columbia during my time there, but we were just establishing contact when he too passed away. I was not aware how intimately he had been associated with Hartree and his work on the differential analyzer until starting this biography. However, I was fortunate in other respects. I got to meet Hartree's first PhD student, Arthur Porter, who alerted me to the importance of their work on control theory. In the last few years, a great deal of in formation has become available on the Internet that was a great asset. The "Nobel e-Museum" has a wonderful selection of biographies and lec tures (http://www.nobel.se). Anyone not familiar with computing de- vni Douglas Rayner Hartree vices can visit the museum of HP calculators (http://www.hpmuseum.org) and get more information about slide-rules, calculating machines, and cal culators than they might wish to know. And there are many others. In some instances, the web sites were a source of information, such as http://www.alanturing.net which has on-line copies of the minutes of NPL meetings in which Hartree participated, but in others, such as the one for calculators, they illustrated the fact that readers can readily find addi tional information of a general nature. Unfortunately, web sites come and go, therefore they have not always been included here as references. The Internet also made it possible for me to contact the Hartree family. I was desperately searching for pictures of Hartree that are few in num ber. I knew Hartree had collaborated with Robert Bruce Lindsay at Brown University. While searching the Brown University site, I came across the address of his son-in-law who provided me with the address of Hartree's daughter, Margaret Hartree Booth. As soon as we made contact, the Hartree family (Margaret, Oliver, and Richard) graciously shared with me the family pictures included in this book. Several factors motivated me in undertaking this project. I noticed that the Emile Segre Visual Archives had 36 (it now has 53) pictures of Dirac and only one of Hartree, and that is a group photo with Einstein. This indicated to me that pictures of Hartree were badly needed for an accurate historical "picture gallery" of the scientists of his time. Then in searching the McGraw-Hill online encyclopedia, I found Hartree's name associated not only with atomic physics, quantum chemistry, band-theory of solids, the nuclear molecule, and nucleon-nucleus scattering, all of which I was aware of to various degrees, but there also was the Hartree equation in electronics! This was totally new to me. I later found it to be related to the operation of a magnetron. At the same time, there was no mention of his work with the differential analyzer though the biography of Vannevar Bush mentioned that a large differential analyzer had been completed in 1935 at Manchester University and was now an exhibit in the Science Museum, London, without mentioning that the person to do so was Hartree! All this suggested to me that a biography was needed to describe the many facets of his work. In this biography, I have attempted to gather available information about Hartree and his work as well as give a sense of the times in which he lived. Charlotte Froese Fischer Acknowledgments This biography has relied on many valuable sources of information. The Hartree family - Margaret, Oliver, and Richard - provided me with the pictures included in this biography, unless indicated otherwise. Christ's College, Cambridge, was most helpful in getting me started, sending me Margaret Hartree Booth's memoirs, and obtaining letters from Hartree to Edward V Appleton from the Special Collections, Edinburgh University Li brary. Dr Michel Godefroid of the Free University of Brussels, shared with me his collection of articles on the Meccano differential analyzer. Dr Jon Agar, of the Centre for the History of Science, Technology and Medicine, University of Manchester, provided me with some difficult to find Hartree publications and an exceptional picture of Hartree demonstrating his differ ential analyzer. Dr James Peters, Manchester University Archivist, clarified Hartree's involvement in the development of the Faculty of Music. The Niels Bohr Library, Center for History of Physics, American Insti tute of Physics, College Park, MD through their oral history interviews, provided notes of a conversation with Elaine Hartree, transcripts of inter views with Sir Nevill Mott and John C Slater as well as the Robert Bruce Lindsay autobiography. Through the Archives for the History of Quan tum Physics program, I received letters from Hartree to Samuel Goudsmit, Robert B Lindsay, and Niels Bohr. The Niels Bohr Archive in Copenhagen provided letters to Niels Bohr, Oskar Klein, and Leon Rosenfeld. The Center for History of Science, of the Royal Swedish Academy of Science, Stockholm, kindly furnished copies of correspondence between Hartree and Ivar Waller. Letters between Hartree and Fritz London are part of the Fritz W London Papers in the Duke University Archives whereas the correspon dence with John C Slater is part of a collection found in the library of the American Philosophical Society, Philadelphia. I am grateful to all these ix