Berni Alder (1925–2020)

Obituary Berni Alder (1925–2020)

Theoretical physicist who pioneered the computer modelling of matter.

erni Alder pioneered computer sim- undergo a transition from liquid to solid. Since LLNL ulation, in particular of the dynamics hard spheres do not have attractive interac- of atoms and molecules in condensed tions, freezing maximizes their entropy rather matter. To answer fundamental ques- than minimizing their energy; the regular tions, he encouraged the view that arrangement of spheres in a crystal allows more Bcomputer simulation was a new way of doing space for them to move than does a liquid. science, one that could connect theory with A second advance concerned how non- experiment. Alder’s vision transformed the equilibrium fluids approach equilibrium: field of statistical mechanics and many other Albert Einstein, for example, assumed that areas of applied science. fluctuations in their properties would quickly Alder, who died on 7 September aged 95, decay. In 1970, Alder and Wainwright dis- was born in Duisburg, Germany. In 1933, as covered that this intuitive assumption was the Nazis came to power, his family moved to incorrect. If a sphere is given an initial push, Zurich, Switzerland, and in 1941 to the United its average velocity is found to decay much States. After wartime service in the US Navy, more slowly. This caused a re-examination of Alder obtained undergraduate and master’s the microscopic basis for hydrodynamics. degrees in chemistry from the University of Alder extended the reach of simulation. California, Berkeley. While working for a PhD In molecular dynamics, the forces between at the California Institute of Technology in molecules arise from the electronic density; Pasadena, under the physical chemist John these forces can be described only by quan- Kirkwood, he began to use mechanical com- tum mechanics. Without solving the quantum puters to explore how molecules in solids and problem, molecular dynamics would not be liquids moved in relation to each other. of spheres and followed the state of the sys- able to make precise predictions. The devel- The question he set himself, which occupied tem over time. They considered hard spheres opment of very accurate simulation methods him for the following two decades, was: “How because the dynamics could be exactly deter- for extended quantum systems is an unsolved does a system of hard spheres [representing mined, silencing criticism that the results problem to this day, although excellent results molecules] behave under various conditions?” were the product of inaccurate computer have been obtained for some situations. The During his PhD, he and the computer scientist arithmetic. The advantage of this technique simulations of a uniform system of electrons Stan Frankel developed an early Monte Carlo over Monte Carlo methods was, as its name that I performed with Alder in 1980 represent algorithm — one in which the spheres are given suggests, that it could address the dynamics one such situation. Our results for the corre- random displacements — to calculate the prop- of many-particle systems as well as their equi- lation energy of electrons underlie the theory erties of the hard-sphere fluid. The advance librium properties. used in most studies of the microscopic prop- was scooped by Nicholas Metropolis and his erties of condensed materials. group at the Los Alamos National Laboratory “He was the impresario In 1963, Alder, with Teller and others, helped in New Mexico. to set up the Department of Applied Science at After completing his PhD in 1951, Alder of younger colleagues, the University of California, Davis, to establish returned to Berkeley to teach chemistry. In pushing them to work on a graduate-training programme associated 1953, he began working as a consultant at the difficult problems.” with the Radiation Laboratory at Livermore. University of California Radiation Laboratory Alder was also one of the founders and the at Livermore (later the Lawrence Livermore editor of the Journal of Computational Physics. Laboratory), newly founded by nuclear phys- The invention of molecular dynamics is Berni Alder’s personality was unique. He icists Edward Teller and Ernest Lawrence, and Alder’s greatest legacy, and has led to appli- preferred intuitive understanding over math- in 1955 he joined the staff. The laboratory was cations in materials science, biochemistry and ematical derivations, always focusing on what well funded as part of the US effort to promote biophysics, as well as physics and chemistry. he regarded as the fundamental scientific innovation during the cold war. Alder and his Two of his fundamental contributions to sta- problems, not on short-term progress. He was group used the spare capacity of the increas- tistical mechanics stand out, and convinced never the programmer, but the impresario of ingly powerful electronic computers that the the scientific community of the technique’s younger colleagues, pushing them to work on lab deployed in the design of nuclear weapons. utility. Until he published his findings, solids difficult problems through his curiosity and Alder returned to his interest in the proper- were thought to exist as a result of attractive intense interest. ties of systems of spheres. In the mid-1950s, in interactions between molecules: the regular collaboration with his Livermore colleague arrangement of atoms in a crystal lattice is the David Ceperley is founder professor of Thomas Wainwright, he developed algo- configuration that minimizes their energy. physics at the University of Illinois at Urbana- rithms to simulate many-body systems. The Alder and Wainwright (and others using Champaign. He worked with Berni Alder at technique they used, molecular dynamics, Monte Carlo methods) showed in 1957 that as Livermore from 1978 to 1987. modelled a sequence of collisions in a system systems of hard spheres are compressed, they e-mail: [email protected]