Bertram N. Brockhouse
1918 - 2003
Nobel Prize in Physics, 1994
Bertram Neville Brockhouse was born July 15, 1918 in Lethbridge, Alberta, the son of Israel and Mable Brockhouse. At an early age he moved with his family to Vancouver. After graduating from high school in 1935, he worked as a laboratory assistant, and then as a self-employed radio repairman, both in Vancouver and in Chicago. He spent the war years in the Royal Canadian Navy Volunteer Reserve, mostly on the lower deck, partly as an electronics technician, and he then attended the University of British Columbia, from which he graduated in 1947 with first class honours in mathematics and physics. He entered the University of Toronto that same year, and the following year married Doris Miller, whom he had met in Ottawa in 1945. He obtained his Ph.D. in 1950, with a thesis entitled "The Effect of Stress and Temperature upon the Magnetic Properties of Ferromagnetic Materials".
In July 1950 Brockhouse joined the staff of the Atomic Energy Project of the National Research Council of Canada, later to become Atomic Energy of Canada Limited (AECL), at the Chalk River Nuclear Laboratories about 130 miles northwest of Ottawa. It appears that D.G. Hurst, head of the Neutron Spectrometer Section of the General physics Division, was looking for someone to do neutron scattering experiments, and that Bert Brockhouse was spotted at the University of Toronto and invited to apply for the position. One wonders whether the Brockhouses didn't have second thoughts about their move when the town of Deep River was overrun by a plague of caterpillars, only a few months after their arrival!
Brockhouse's first work at Chalk River, in collaboration with Hurst and M. Bloom, involved him in a series of studies of the resonant scattering of slow neutrons by strong absorbers such as cadmium and samarium. The measurements were made by placing the sample in a well shielded scattering chamber so that it was surrounded by an annular array of six Bismuth Fluoride detectors. The chamber was mounted on the arm of a single-axis spectrometer which had been built by Hurst and his group in the late nineteen- forties.
The idea of studying the inelastic scattering of slow neutrons occurred at a meeting attended by Hurst, Brockhouse, G.H. Goldschmidt and N.K. Pope, in December 1950: it was soon decided that such experiments were feasible at the Chalk River NRX (National Research Experimental) reactor which was at the time, and until 1953, the world's highest flux beam reactor. As a first step the resonant scattering apparatus, in slightly modified form, was used by Brockhouse and Hurst to study the energy distributions of initially monochromatic neutrons scattered by polycrystalline samples of aluminum, lead, graphite and diamond. The intensities of neutrons transmitted through various thicknesses of cadmium were compared with calculations based on ideal gas and Einstein models of the vibrational behaviour of the material.
In the early months of 1952 Brockhouse put together what he described as a "large aperture double spectrometer", in reality a triple-axis machine, hoping to be able "to measure the as yet unknown frequency distribution of normal modes" in a crystal. Much effort was put into trying to get the machine to work, including attempts to produce monochromator crystals with higher reflectivity, and improvements to the shielding, but by the end of the year the spectrometer was not producing results.
In 1953 Brockhouse took advantage of an unexpected shutdown of the NRX reactor to spend ten months as the first foreign guest scientist in the Reactor Department at Brookhaven National Laboratory. During this time he worked with L.M. Corliss and J.M. Hastings on a study of multiple scattering by flat specimens and magnetic scattering by zinc ferrite, on a powder magnetic diffraction study of copper oxide, on the development of improved monochromator crystals, on a study with G.H. Vineyard of the scattering by liquid aluminum, including "an (inconclusive) attempt to measure the energy distribution of the scattered neutrons by absorption methods", and on a measurement of the incoherent cross sections of copper and gold.
On his return to Chalk River, Brockhouse again set up his crude triple-axis spectrometer, using a fixed angle monochromator facility with an aluminum crystal monochromator, a makeshift sample table, and the old single-axis instrument acting as the analysing spectrometer. The scattering angle at the sample position was fixed for a give set of measurements but could be changed by turning the sample table and moving the analysing spectrometer on a set of rails. The machine was used successfully, thanks in large part to its improved monochromator, for studies of the phonon frequency distribution of vanadium and of the inelastic scattering by liquid lead and light and heavy water: these measurements were reported at the January 1955 meeting of the American Physical Society in New York City. The energy dependence of the paramagnetic scattering by materials such as the manganese oxides was also studied.
In the early months of 1955, "preliminary measurements (were made) of energy distributions scattered by an aluminum single crystal in several different orientations". This work, in collaboration with A.T. Stewart, led to the first successful determination of a phonon dispersion curve. It provided the first convincing demonstration of the power of the triple-axis method, at a time when groups at Saclay and at Brookhaven were concentrating their attention on a complementary time-of-flight technique.
Brockhouse next turned his attention to the possibility that neutrons might be used to investigate the "thermal disturbances of the magnetized arrays of...coupled magnetic moments (which) can be described by means of quantized wave excitations called spin waves". The ferrimagnetic material magnetite was chosen, for the very good reason that large single crystals were available. The measurements of scattered neutron energy were carried out for 12 different orientations of the crystal, using 1.52 Angstroms incident neutrons and a scattering angle of 18 degrees, and it was concluded that the observed excitations were not phonons, but indeed "in the spin system itself". This was the first experimental determination of a magnon dispersion curve.
These experiments on phonons and magnons were followed by extensive sets of measurements on a variety of crystals. It was not uncommon for Brockhouse and his collaborators to return to systems which had previously been studied, taking full advantage of improvements in experimental technique and instrument performance, in order to obtain more detailed and more extensive results.
The first work on a semiconductor, germanium, was reported by Brockhouse and his student P.K. Iyengar in 1957. In order to determine the frequencies of phonons along major symmetry directions, a method of successive approximations was adopted.
In the mid-fifties Brockhouse and Stewart had begun to put together a beryllium filter-chopper spectrometer, having previously concluded that this type of instrument was preferable to a double chopper machine. The instrument was similar to the slow chopper at Brookhaven except that a filter difference technique was used to improve the overall energy resolution. Nonetheless its use was discontinued in 1957 because of its limited resolution. Phonons in aluminum and beryllium were studied using the filter-chopper spectrometer, and an important series of measurements on water was completed.
A new type of high resolution time-of-flight instrument was devised by Brockhouse at about this time, following a conversation with D.G. Hurst. This was the rotating crystal spectrometer, first mentioned in an AECL Physics Division progress report in late 1957. The first version of this machine was installed at the NRX reactor, and an improved version, fitted with a cooled quartz filter and initially located at the NRX reactor, was later set up at the N5 hole of the new reactor NRU (National Research Universal). R.N. Sinclair, who had worked with Brockhouse as a postdoctorate fellow, built a similar machine at Harwell on his return from Chalk River.
The rotating crystal spectrometer was initially used for a detailed study of the quasi-elastic component of the scattering by water. The results were consistent with earlier Chalk River results, and strongly suggested that the supposed fine structure in the data of D.J. Hughes and his collaborators was spurious. The spectrometer was also used for an extensive series of experiments, by D.G. Henshaw, on liquid helium: the first inelastic work on helium at Chalk River had been undertaken using the filter-chopper spectrometer. Other early work using the rotating crystal spectrometer included phonon measurements on lead and sodium iodide.
The famous C5 triple-axis spectrometer, immortalised in Kittel's "Introduction to Solid State Physics", was installed at the NRU reactor in 1958. This machine remained in use for more than twenty years and was an important training ground for many present day triple-axis spectrometrists. It consisted of a monochromator section, designed by W. McAlpin, together with the double-axis spectrometer and associated electronics which had formerly been in use at the NRX reactor. (The analysing section had first been installed in 1956 at the NRX reactor, piggy-backed on the arm of the original single-axis spectrometer.) The first material to be studied using the C5 machine was a single crystal of silicon; lead was also studied, supplementing the earlier work on the rotating crystal spectrometer. Major improvements to the spectrometer were reported in a 1959 progress report, including the construction of monochromating crystal control units with which "it will be possible to change the wavelength of the incoming neutrons automatically and continuously over a wide range".
The report continues: "A modification to the positional spectrometer is under construction which will enable the specimen table to be automatically controlled. New control units have been designed and constructed (by E.A. Glasser) which will simultaneously control three of the four variables (incoming energy, outgoing energy, angle of scattering, and specimen angle) in a non-linear way according to prescriptions computed on the Datatron. These changes will make possible drastic improvements in experimental methods. For example, it will now be possible to measure energy distributions at a constant Q (wave vector transfer). This is highly desirable since the significant theoretical variables in a scattering experiment are the energy transfer and the wave vector transfer, and not the four experimental parameters listed above." The development of the constant Q technique paved the way for a very large number of sophisticated investigations of excitations in single crystals. It remains the most powerful triple-axis technique, and is in constant use at neutron scattering centres throughout the world.
With the capability to vary the incident neutron energy of the C5 spectrometer, a new method for the study of high energy excitations became possible. This was the beryllium filter detector method, which was first tried at Chalk River in early 1960. At about this time the N5 rotating crystal spectrometer was modified so that both the incident energy and the angle of scattering could be continuously varied.
In the late fifties and early sixties scientists from a number of countries visited Chalk River in order to observe and learn the new "Methods for Neutron Spectrometry". By all accounts it was an exciting time and Chalk River staff members who knew none of the details of what was happening in the field were aware of the importance of the neutron scattering experiments at the reactor.
Brockhouse became head of the newly formed Neutron Physics Branch in 1960. A.D.B. Woods, who collaborated with Brockhouse on many occasions, had joined the neutron scattering group two years previously, and G. Dolling arrived in 1961.
A number of scientists paid extended visits to Chalk River during this period, and participated in collaborative experiments. T. Arase and G. Caglioti participated in the extensive experimental studies of lead. W. Cochran spent an extremely successful year (1958-9) working with Brockhouse and Woods on experimental and theoretical aspects of the lattice dynamics of alkali halides: a detailed formulation of the "shell model" was developed and this in turn led to an important investigation of SrTiO_3 by R.A. Cowley, who first visited Chalk River as a pre-doctoral fellow in 1961. Others who spent significant periods of time with the group were M. Sakamoto, K.R. Rao, L.N. Becka, H. Watanabe, B.A. Dasannacharya, and J. Bergsma. Experiments performed during this period included detailed studies of phonons and magnons, the first observation (in 1961) of a Kohn anomaly, intensity measurements to determine phonon polarization vectors, observations of crystal field splittings, studies of moderators and hydrogenous materials, and measurements of the scattering from several liquids. Mention should also be made of Brockhouse's work on single crystal filters, and of his early studies of texture in metals.
Bert Brockhouse did not have very much spare time during his highly productive years at Chalk River, but he did find time to take part in a number of amateur dramatic presentations, including three Gilbert and Sullivan operettas, and a production of Shaw's "Arms and the Man". He enjoys music enormously, and many of his colleagues have told stories of his penchant for singing musical excerpts, whether from an opera or from a Broadway musical, while working at one of his neutron experiments.
In 1962 Brockhouse moved to McMaster University where he served as Professor of Physics until his retirement in 1984. He was chairman of the Department from 1967 to 1970. At McMaster he took an active part in teaching, and was able to communicate his enthusiasm for physics to undergraduate and graduate students alike. He was influential in building up the Department, and he and his graduate students, the first of whom were S.-H. Chen, J.M. Rowe, and E.C. Svensson, built new spectrometers at the McMaster Nuclear Reactor, and later at the Chalk River NRU reactor. Novel designs were employed, such as the double monochromator concept (independently conceived by R. Stedman), and the idea of placing analyser and detector in a common shield. At the same time many interesting experiments were performed, including studies of lattice excitations in dilute alloys such as Cu(Au), and in concentrated alloys such as the Bi-Pb-Tl system.
Professor Brockhouse received many honours over the years, including the Tory Medal of the Royal Society of Canada, the Buckley Prize of the American Physical Society, the Duddell Medal and Prize of the (British) Institute of Physics and Physical Society "for excellence in experimental physics", and the Centennial Medal of Canada. He was an Officer of the Order of Canada, a Fellow of the Royal Societies of Canada and London, and a Foreign member of the Royal Swedish Academy of Sciences. He received honorary D.Sc. degrees from the University of Waterloo and McMaster University. He was also a member of the Philosophy of Science Association.
We owe a tremendous debt of gratitude to Bert Brockhouse. He has inspired many people to accept the challenges of neutron inelastic scattering, and to work long and hard to improve methods, materials and equipment in order to be able to do experiments properly and convincingly. Throughout his career he has demonstrated an honesty, thoroughness and scientific passion which are an example to us all. The "absent-minded professor" stories are plentiful, and amusing, but the stories of his insistence on good experimental technique, and of his concern that time and money be efficiently used, are perhaps more to the point. His intuition, his dedication to research, and his kindness and concern for his fellow workers, are frequently mentioned by those who have had the pleasure to work with him.
Professor Brockhouse passed away October 13, 2003.
Footnote: Quotations in the text are taken from published papers and from progress reports of the National Research Council of Canada, Atomic Energy Project, and of Atomic Energy of Canada Limited, Chalk River Project. I am grateful to the many people who contributed information for this brief biography
Biography prepared by John R.D. Copley at the time of Dr. Brockhouse's retirement in 1984. It is taken from Physica 136B (1986) xxvii-xxxi (North-Holland, Amsterdam)
John R.D. Copley
McMaster Nuclear Reactor, McMaster University
Hamilton, Ontario, L8S 4M1, Canada