Materials Research Activities

Interview with Professor Peter Hirsch

Professor Sir Peter Hirsch, Materials Department, Oxford University

Bernadette Bensaude-Vincent, December 12, 2002

Bernadette Bensaude-Vincent (BBV):Could you tell me about how metallurgy developed in Oxford and how it grew into Materials Science?

Peter Hirsch (PH). Let me first say a few words about the way I see the early developments of Materials Science in the UK. There were various trends that came together after the War. First of all there was Professor Nevill Mott in Bristol, developing Solid State Physics. His interests included inter alia defects and how they influence properties of materials. He had important activities on properties related to point defects and on the photographic effect. In the course of these studies they also developed techniques for visualizing dislocations. The group included Charles Frank, a physical chemist by training who was a material scientist par excellence. He developed a theory of crystal growth. Then there was Nabarro who worked on dislocations. So in Bristol they worked on defects, dislocations, crystal growth, the photographic effect, and also optical properties of crystals.

BBV Where were they based?

PH: In the Physics department at Bristol University. It was Solid State Physics. The electron theory of metals was developed there too. And Jacques Friedel was there for some time. In my opinion Solid State Physics opened the way to Materials Science. In parallel with this there was the group in the Metallurgy Department in Birmingham, lead by Alan Cottrell. He is much more of a materials scientist than I am. He was subsequently instrumental in developing Materials Science in Cambridge. His activity in Birmingham was very important because his group attempted to explain mechanical properties in terms of disclocation theory. And the third trend was what happened here in Oxford through Hume-Rothery's classic work on electron phases of alloys. He stimulated work on electron theory of metals and alloys in Physics Departments elsewhere. He was trained as a chemist.

BBV: So it came out of these three trends?

PH: These are the main three. I may be unfair to other groups but these were the most influential groups.

There was some Metal Physics going on in the Cavendish Laboratory after the war. Bragg, who was Head of Department was of course an X-Ray crystallographer. In order to understand the intensities of diffraction spots on an X-Ray diffraction photograph, it had to be assumed that the crystals were not perfect, i.e. that crystals consisted of mosaic blocks. This was based on a theory by Darwin (1914) of the intensities of diffracted X Rays. Bragg published a note in the 1940s on the relationship between strength and particle size (mosaic blocks) in crystals. There were two groups in the Cavendish in the Metal Physics field when I came in 1946. One was Bragg's little group which developed the bubble-model, typical of Bragg's simple but brilliant ideas. Another group was Orowan's Metal Physics group which studied plasticity, fracture, creep, those sorts of topics

BBV It is in the Cavendish Laboratory that you started your career, isn't?

PH: It was 1946 when I joined and there were several activities. My own work started from Bragg's interest in work hardening(when you deform a metal it becomes stronger). I actually went into the Crystallography Department of the Cavendish to work on a PhD problem that Bragg gave me. It was conceptually very simple. If you take a metal and work it, does it break up into smaller blocks? Bragg always had very simple and brilliant ideas. If the crystal breaks up into smaller blocks (subgrains) the diffraction pattern consists of individual spots from each little subgrain. If you make the beam small enough you illuminate such a small number that you will get a few spots on the diffraction ring, whereas, with  too large a beam diameter you get a continuous ring due to overlapping spots. You can count the spots (on discontinuous rings) and deduce the size of the particles. That was the project. It did actually work for heavily cold worked aluminium.We derived a particle size of 2 microns. By the time we managed to do all this by X ray diffraction, Bragg had lost interest in it. He did not actually supervise me. Bragg was interested in proteins at that point.  My formal supervisor was W.H. Taylor who was head of the Crystallography Department.  His interest was the structure of minerals.

BBV: When you worked in Cambridge did you consider yourself as a crystallographer?

PH: I was a physicist working in a department of crystallography. In those days most conventional crystallographers determined crystal structures. I was one of the relatively few people not doing that.

By the time we found that cold worked aluminium breaks into subgrains, Heidenreich at the Bell Laboratories published the first pictures of metals by Transmission Electron Microscopy (TEM). He observed directly the little subgrains in heavily beaten aluminium foil. That depressed us very much because we needed exposures of many hours for our x-ray diffraction photographs, while  he had a ten second exposure with his electron microscope. So we went into this field of TEM and finally we saw individual dislocations.This had a big impact because there were many metallurgists who did not believe in dislocations, who considered them as figments of the imagination of solid state physicists working out theories in tremendous detail without much supporting experimental evidence. With our technique you could see dislocations directly and see them move. And we made movies. I remember showing a movie at MIT to Bert Warren who was a well-known X Ray crystallographer.  His comment was symptomatic of many metallurgists. Seing is believing. We converted people.

BBV: In terms of institutions could you describe the shift from Metallurgy to Materials departments?

PH: There were many Metallurgy departments in this country:  Sheffield, Birmingham, Imperial College, to mention a few. The activities of the groups that I mentioned gave an impetus. But an important impetus came from the United States. Bill Baker and Herbert Holloman in particular had this vision of multidisciplinary activities leading to the development of better materials, whiskers, ceramics etc.. An enormous amount of money was spent at General Electric, and Bell Telephone and the Ford Motor Company. When Alan Cottrell became Professor in Cambridge that was the real beginning of an institutional effort  to develop Materials Science in this country. In my opinion that was the defining moment. Alan Cottrell left Birmingham in 1955 and went to work at the Atomic Energy Authority at Harwell where he worked on uranium and materials for nuclear reactors. Then he became Goldsmith Professor of Metallurgy in Cambridge in 1958. He started projects on ceramics and composite materials. Tony Kelly joined him to work in these areas. Alan Cottrell ‘s interest in composite materials probably stemmed partly from activities in the US but mainly from his own views on strong materials. He also supported work on superconductors. His initiative to work on different types of materials was probably the beginning of the shift from metallurgy to materials science in Universities in this country.

BBV: Could you tell something about the implementation of Materials Science here in Oxford?

PH:  The department of Metallurgy and an Honour School in Metallurgy were started by the University in 1957. Jack Christian wrote a paper on the early history (Materials World, April1997) when the department celebrated its 40th anniversary. You see from this that Hume-Rothery started as a chemist and worked in the Inorganic Chemistry Department. But he was interested in metallic phases. The idea of a separate department arose gradually. Jack Christian was appointed as demonstrator in 1951 and initially lectured on metallurgical topics to chemists. Metallography was a supplementary subject in the Chemistry Honour School. Then the Pressed Steel Company at Oxford established a readership in metallurgy named after George Kelley. Hume-Rothery became the first George Kelley reader in 1954. Jack Christian was appointed lecturer in 1955, then John Martin came from Cambridge and joined in 1957, followed by Angus Hellawell. There was a move to increase the Engineering School and to develop a Department of Metallurgy. Francis Simon who was professor of physics here was keen to establish something on the lines of the Laboratory for the Study of Metals in Chicago, which was famous in those days.  Chicago was a research institute, not a teaching department, just a research institute for postgraduates. This is what Simon had in mind for Oxford. However, the industrial advisors expressed the view that we needed to  educate metallurgists to go into industry. British industry wanted a teaching department. Monty Finniston, who was at the time the head of the metallurgy division at AERE Harwell and later became chairman of British Steel, made approaches to the Wolfson Foundation for financial support. The result was the establishement of the Isaac Wolfson Chair in Metallurgy which Hume-Rothery held until his retirement in 1966.

At the same time the condition that the Wolfson foundation made for giving the money for the Chair was that the University Grants Committee, an organization distributing the money from the Department of Education to the Universities, should provide the funds for the building. Funds were provided by the University Grants Committee and the building was opened in 1959. Initially the Honour School was a Joint Honour School in Chemistry and Metallurgy,  which subsequently became Metallurgy. There is an interesting quote in Jack Christian's paper which indicates the University's unease with technology at that time. The Honour School should teach «no more technology than is involved in the degree courses of chemistry and physics» and it stated that «the man who has studied pure science at a university can take up technology on entering industry much more easily than one who has studied technology can later take up the pure science which may be required for his work». It was still quite difficult to really expand engineering and to get a proper engineering metallurgy course at that time. The research that was going on here was on alloy phases (Hume-Rothery's research); John Martin worked on mechanical properties of alloys; Angus Hellawell worked on solidification studies; Jack Christian worked on phase transformations - the martensitic transformations - of metals and alloys, and also on plastic deformation of body centred cubic metals. So that was the development of metallurgy in Oxford before 1966. Clearly this department was a metallurgy department rather than a materials department. But Hume-Rothery's work was influential in encouraging solid state physicists to become interested in metals. He stimulated the understanding of the structures of metals and alloys on the basis of the fundamental electron theory of metals. He empirically developed some structure rules. He hoped that electron theory of metals would enable the prediction of the structures of metals and alloys and appointed a theoretical chemist, Simon Altmann, to work in this area.

BBV: And then you came in this Metallurgy department. Why did you move from Cambridge to Oxford?

PH: I came in 1966. I was a physicist who had worked on electron microscopy of defects in materials. I had the opportunity to take a Cambridge chair or an Oxford chair. I decided to go to Oxford for two reasons. 1) I had been in Cambridge quite a long time (23 years) and a change opens new vistas. 2) The challenge was much greater here and my own predilection was always for building things up. The Cambridge Metallurgy department was a large and successful department built up by Alan Cottrell. In Oxford there was really a nucleus of a department (a distinguished one) and in principle one could do a job to build it up.

I had already built up a large research group in the Cavendish in Cambridge.  When Bragg switched his interests to proteins (Perutz, Kendrew et al) Orowan left and went to MIT. I don't know the details but my impression is that Bragg did not work hard enough to get Orowan a senior permanent job. So the Metal Physics group in Cambridge folded up. Bragg's own little group folded up too. We made a significant effort to set up a new Metal Physics group at the Cavendish. The vision behind this was: We now had a technique to enable us to actually see with the TEM what is going on inside metals and one could see the defect distribution after deformation or irradiation of the material. One could then determine in principle what the properties of the materials were. That was the Holy Grail. There were three steps: 1) to try to understand the properties of the defects and how they interacted; 2) to try understand how they control the macroscopic properties; 3) if one understood the basic relation between defects and properties then one could eventually go further and predict what processing should be done to optimise properties. That was the vision but while we were successful in the first step, there was only limited success in the second step, and we never got as far as the third step.

When I came to Oxford, my aim was to get this technique of TEM - to see what goes on in the materials - transferred to a metallurgy/materials department. I wanted to get it closer to applications to «real» materials. Whereas physicists work on models, - on pure copper for instance , a metallurgy department should be looking at materials of interest  technologically, such as alloys that are much closer to practical needs. My aim was to apply TEM to technologically interesting materials, real materials rather than the model materials that we looked at in Cambridge.

When I got here I did attempt to build up Materials Science. Right from the beginning my aim was to shift from metallurgy to materials science which should cover all kinds of materials and applications.

BBV Where did this project come from?

PH: There was no model course. We had to build this up incrementally. When I first came the University provided three new permanent posts, very generously. One was for Professor Whelan, to establish the electron microscopy group. One went to John Hunt, a solidification expert who had worked atHarwell and Bell Laboratories; and another one was for Geoffrey Groves who worked on ceramics. Gradually we built up a Materials Department by securing more appointments. You have to realize that when I came in 1966 the number of staff was very small (4 faculty plus the Professor).The course developed gradually and changed from metallurgy to materials science.  In the early years a considerable part of the teaching was carried out by research assistants or fellows supported on research grants or fellowships.  The research activities were built up first, and the research groups then helped in the teaching of the Honour School.

BBV When did you officially change the name of the department?

PH: That was much later, 1990. The name of the Honour School was changed to Metallurgy and Science of Materials in1969 (The University would not accept "Materials Science" because Materials was not an adjective!).

BBV Did you appoint chemistsas well?

PH: We didn't appoint chemists to teaching posts. We appointed metallurgists and physicists. But we did appoint a chemist to a research post in the department, in high resolution electron microscopy. When I retired in 1992 the department  consisted mainly of metallurgists or materials scientists and physicists or ex-physicists. We appointed inter alia people who had expertise in semiconductors, superconductors, magnetic materials, materials processing, corrosion. Gradually we extended the scope of the courses in the department. The one sticking point was polymers. For a very long time we could not get a good polymer scientist. It has changed now. We now have two polymer scientists.

BBV Did you have contacts with John Goodenough who came to the chemistry departmentin 1973?

PH: We did have contacts but we probably did not make the best of the opportunity. In this department we were more interested in the effect of microstructure on materials, metals, ceramics, semiconductors, superconductors, magnetic materials etc., rather than e.g. in the intrinsic magnetic properties of perovskites and other materials.Our interests were not sufficiently close. We became a Materials Department in that we were concerned with the effects of microstructure on properties of a wide range of materials of technological interest and with the effects of defects on devices. The work on superconductors was  initially on low temperature superconductors, but later on high temperature superconductors, and the more recent studies focussed on processing.  There was some interaction with the Inorganic Chemistry Laboratory on electron microscopy of catalysts and superconductors.

BBV: Did you keep a close link between courses and research in the department?

PH: Yes. We started as a metallurgy department. We built up a number of research groups, and the expertise in some of these, e.g. semiconductor and magnetic materials, enabled us to teach courses on these topics and broaden the curriculum to Materials Science. At some point we had a group working on cements, and this too led to a course in the Honour School. And there were always undergraduate and postgraduate courses on materials characterisation, where we had particular strengths.  In the 1980s, I became more interested in the output end, in getting closer to the engineers.

BBV: Did you train students in engineering?

PH: The engineers had their own faculty to teach materials to Engineers.  But we did some teaching for them, and over some years they taught polymers for us.  In the mid 1980s we started a joint course with the Engineers on Electronic and Structural Materials Engineering, changed to Engineering and Materials in 1992.  This led to closer collaboration with the Engineers in teaching and research.

BBV:Did you also have contacts with the nuclear physics department in Oxford?

PH: The only contact we had with the Nuclear Physics Laboratory in Oxford was on their proton microprobe, a materials characterisation tool, and fairly recently this activity was transferred to our department.

We had a lot of contacts with Harwell. From 1973 onwards I got very much involved with the study of the integrity of pressurized water reactors through the Marshall Committee. From 1982 to 1984 I became part-time chairman of the Atomic Energy Authority. This period had a strong influence on me. I felt the need to produce materials engineers, not only materials scientists. I learned during my period with the Atomic Energy Authority that the education of engineers in materials tended to be relatively poor. And this could result in inappropriate component design or manufacture. These problems sometimes led to costly mistakes.  In order to develop a joint activity between engineers and metallurgists we got some support from the engineering department. To cut a long and difficult story short, we took an opportunity which presented itself around that time in the form of additional funding for engineering courses from the Government to start a course on Electronic and Structural Materials Engineering.

BBV: Was it in the 1980s?

PH: Yes.

BBV: Did you then feel the need to add «engineering» in the name of the department on the lines of the departments in US universities that are called Materials Science and Engineering with an emphasis on the E of Engineering?

PH: What happened here is that we developed a multiplicity of courses to provide more options for the undergraduates to increase our intake.  The courses were Metallurgy and the Science of Materials, then Metallurgy, Economics and Management (1979), and finally Engineering and Materials. There is close collaboration between us and engineering through the teaching and contacts in research. This depends on individual contacts of course. In the 1980s we set up with the Engineers the Oxford Centre for Advanced Materials and Composites (OCAMAC) to foster collaborative research and contacts with industry. There is now also strong collaboration with people in chemistry and physics on a number of research programmes. We have become much more interdisciplinary, if you like. We have always considered ourselves as a bridge between the Science and Engineering Departments, with contacts with both.  The fact that Physical Science and Engineering are all part of the same faculty in Oxford is advantageous to us.  We did consider changing the name of the Department to Materials Science and Engineering - but that was unacceptable to the Engineers who considered Materials Engineering to be their responsibility.  So we decided to change the name in 1990 to "Department of Materials".

BBV: It seems that your story is quite different fromthat of  the US materials departments.

PH: I think that it is different ; you are quite right. Here the initiative to have Materials Departments came from academics, whereas in the US the development was led by Industry, and industrialists promoted Materials Science in Universities.  (This does not mean that no materials science was going on in industry in the UK - e.g. the carbon fibre work at Farnborough.) I think it started in Cambridge first. We came along in the late 1960s. By that time there were materials science activities going on in many departments, at Imperial College and Birmingham, for instance. We could not claim to have inspired Materials science in the UK. We came along and did our own brand. There were many departments, but they were on a small scale.  Concerning the definition of Materials Science I am with Merton Flemings. Structure, properties, processing, performance/application.  I think it is the engineering applications that are fundamental. The philosophy of the department here was to get theoretical and practical people together. People able to develop models with people who really know what applications are important. And industrial links were fostered.  It is still the philosophy today, even more so. In the last few years after my retirement the department has gone from strength to strength in that direction. There is now another site in Begbroke, five miles away from Oxford centre. It provides opportunities for collaborative research between the Department and Industry.

BBV: Where did the money come from?

PH: The funding for work in the department came partly from the University, the Research Council and Industry.  Some from charitable foundations, e.g. for building expansions. We had a large budget from the Research Council. That is another important aspect, somewhat  similar to the US picture.

It is difficult to attract undergraduates to read Materials Science and Metallurgy. Enormous efforts have been spent on attracting more undergraduates, most of them not very successful.  (But a recent appointment of a Schools Liaison Officer seems to be effective.)  Materials Science is not a school subject, although elements are introduced into some of the school examination papers e.g. the physics curriculum.  The number of courses and options that we offer in Oxford  has helped a bit but it remains a problem. Some small departments in the UK were able to survive because they had a large research activity (supported on non-University funds)compared to the undergraduate activities.  But in the UK as a whole some Materials Departments have closed or have been amalgamated with Engineering Departments. The change of name of department from Metallurgy to Materials does not only reflect the change in content of the courses.  It is also pragmatic because metallurgy has an old-fashion ring about it and materials science is a much broader subject likely to attract more students. The image of materials for computers, aeroplanes and cars is more exciting than that of dirty blast furnaces in the steel industry.

BBV: What would you say about multidisciplinarity in the development of this department ? More specifically could you compare the situation here in Oxford with this diagram published by the National Acdemy of Science in 1969 with a hard core in mathematics physics and chemistry and applications around?

PH: I would agree with this: mathematics, physics and chemistry are basic inputs in Materials Science with applications in ceramics, polymers, and so on. But in these days I would include materials for medical applications. There is somebody in this department working on biomedical materials for implants. There is also a large group in Cambridge. There is a big scope for materials in medicine, particularly for prostheses.

BBV: Finally would you consider yourself today more as a physicist or as a material scientist?

PH: I think I am physicist who «saw the light». True physicists would no longer consider me as a physicist. I consider myself as a materials scientist because my interest is in the effect of microstructure on the properties of materials. I am interested in quite complex materials, with potential applications e.g. high temperature intermetallics, and in modelling their complex mechanical properties. I ended up as a materials scientist. But there are materials scientists who would consider me to be a rather theoretical materials scientist. In the later years of my conversion I supported and promoted materials processing in the department  although it took me rather a long time to get to this view, to appreciate the importance of this field, and to realise the need and potential for modelling.

BBV: What is your concept of materials science?

PH: I am close to Merton Flemings's concept. To me materials science is an enabling science. We study material composition, structure, properties and processing for applications in engineering. There is now a strong group on processing here. Not only casting but various kinds of processing like spray forming, coating, making magnetic and superconductor devices etc. There are also two lecturers working on polymers now.

BBV: Does polymer synt-esis now belong to Materials Science?

PH: Polymer processing and modelling properties belong to materials science, but synthesis of new kinds of polymers - I doubt if this is a proper activity for a materials department, although it would be quite appropriate as a joint research activity with Chemistry. That would  be my view for what it is worth. But composition, structure, properties, performance, Merton Fleming's picture, defines Materials Science.

BBV: What about the recent addition of end-users to this picture?

PH: Yes end-users are important but I would consider that this links in with performance. The interaction with industry is important. Quite apart from the problem of funding it is vital for materials science, as an enabling science.

This page was last updated on 5 February 2003 by Arne Hessenbruch