Reminiscences of the Early Days of
Nuclear Magnetic Resonance at Stanford University
by Warren G. Proctor, Research Laboratory, Varian AG, written around 1967.
 
Historic document, transcribed by Stanislav Sýkora, Extra Byte, Via R.Sanzio 22C, Castano Primo, Italy 20022
in Stan's Library, Ed.S.Sykora, Vol.I. First release September 30, 2005
Permalink via DOI:  10.3247/SL1Nmr05.002
Stan's LIBRARY | SCIENCE Links | NMR Links | Early NMR References. Stan's NMR Blog | Stan's HUB

Forword by the Editor

The title page of the Warren G.Proctor's document published on this web page bears an autographed signature of Lars Olaf Andersson which marks it as his personal copy. Lars has also added a pencilled note which says:

I have found this in my old Zürich lab file and showed it to Martin Packard.
He has added some comments on the last page.

I no longer remember whether the message was originally addressed directly to me. However, it is almost certainly the case since the notes added by Martin Packard are dated June 1984, a year during which I had been in frequent contact with Lars at the Varian offices in Zürich (Switzerland). I have obtained the document directly from Lars who knew about my interest in these things.

The document is typewritten on eight sheets of relatively low-quality paper (A4-format). The count does not include the title page and the seven notes by Martin Packard which are on a separate (page 9). It seems to be an original rather than a high-quality xerox copy but, due to the document's age, I can't be completely sure of it. In any case, I keep the original in careful custody.

I have transcribed the document and taken the Editor's liberty of commenting it in a slightly humourous way. While the transcriptions are in framed panels with a white background and use size 11-pt font, my comments are in simple paragraphs with a tree-bark background and size 10-pt font and are preceded by my initials. To facilitate the reading, the notes by M.Packard are inserted at the indicated locations in the text, rather than collected at the end the way it is done in the original. They are in blue color, enclosed in square brackets and preceded by the initials MP. The transcript is verbatim, except for five quite obvious single-letter mistypes. To enhance the pulse of history which permeates the lines of this documents, I have typeset in bold the names of all persons it mentions - there are no such typographic enhancements in the original.

I have no way to know whether, by publishing the document, I am transgressing somebody's legal rights. Sincerely, I do not think so and I believe that the importance of the document to the NMR community is such that it should be made public no matter what. I certainly have no intention to profit on it, not even by means of ads which I normally use to pay for the maintenance of this site.

Stan Sykora, October 2005, Castano Primo, Italy
 

Transcript and comments
 

I began my graduate studies during the first days of January of 1946, the beginning of the winter trimester at Stanford University. There were altogether about 15 other beginning graduate students in physics, including M.E. Packard and E.H. Rogers (who are now respectively Director of Research and Manager of the Instrument Division at Varian Associates). The faculty at that time was quite small, but it is important to mention that Professor Felix Bloch, Professor W. Hansen were among the half dozen faculty members.

A memorable thing happened to me almost immediately -- it was related to the Physics Department's colloquia, meeting each Thursday afternoon. The first talk I heard was given by Professor Bloch, who described with pride the success of Professor Hansen, a graduate student named Martin Packard, and himself in observing nuclear magnetic resonance, which at Stanford at that time was called nuclear induction. I must admit that I failed to grasp the significance of the discovery. I was fascinated by the instruments which had been carried up from the Physics Department basement to the lecture platform, where young Packard, while Professor Bloch was speaking, unsuccessfully attempted for an hour to show us a proton signal. As I said, this was my first physics colloquium. I naturally thought tha all colloquia would be similar - each professor in turn describing to us the various interesting things that he had observed in recent weeks.

Professor Bloch's ideas on magnetic resonance (he told me later) had been quietly worked out at home during the war years that he had spent at Harvard University. This made it possible to perform the experiment within a few months after his return to Stanford after the end of the war. The brilliance of his concept was clouded by only one thing: he had neglected to anticipate the narrowing of lines in liquids due to their thermal motions, as explained by Bloombergen, Purcell and Pound in their classic paper a year or so later. Until news of Purcell's work had reached Stanford, each time nuclear magnetic resonance was seen, the broad lines that had been observed were always thought to have been their natural width, in keeping with Professor Bloch's simple static dipole-dipole line braodening mechanism. They were, in truth, due to inhomogeneities of the magnets available at the Stanford Physics Department in those years.

SS: Apart from the disarming ingenuity of young students of yester times, one can not fail to notice that
- even then, NMR was mostly done in damp basements,
- the equipment of F.Bloch was portable, anticipating the times by about 60 years,
- public demos used to fail then just as they do today, contradicting the Science's dogma of reproducibility, but confirming the Murphy's law.
- top physics is best done at home, reclining on a sofa and sipping wine,
- even Nobel prize winners are prone to overlook quite a few things,
- whatever you observe always fits well your theories and makes you happy. Long live experiments!

However, I am slightly ahead of my story. Although Martin Packard was unable to demonstrate resonance during that colloquium, the research group of Bloch had in fact certainly seen a magnetic resonance a few weeks before, and I believe it was the day after Christmas, 1945 [MP: A few days after Christmas]. Packard had registered as a graduate student during the fall trimester, thus he must have been one of only a very few students, since the semester started so close to the end of the war. He was immediately pressed, without the benefit of course work, into this research program, and what would normally be a reasonable systematic search for a nuclear magnetic resonance [SS: of proton] was started, using the spectroscopic value for its magnetic moment. However, as the weeks went by and no signals were observed, less and less time was spent in physics basement by Professor Bloch and Hansen, leaving poor Packard alone in the basement to look for the proposed phenomenon [MP: We were all in the Cyclotron Lab that night]. Packard could not find it anywhere and someone suggested the idea of turning up the field very high and then turning it off. Certainly as the field fell over the resonance value something would be seen. This was tried and it worked. Unfortunately I am unable to tell you whose idea this was for sure, or who was present when the first signal appeared on the oscilloscope. Since our student days are always so pleasant to remember, I would like to pretend that the first resonance signal ever to be seen was seen by a lonely and desperate [MP: Not too lonely and desperate] graduate student, Martin E.Packard, in the basement of the physics building of Stanford University on the day after Christmas 1945. The incredibly important role that this phenomenon subsequently was to play in the chemistry and physics should now serve to comfort Dr. Packard for all those miserable and lonely hours he spent looking for that first signal. Of course, had he, due to a bad selection of magnet, sample material or some other variable, failed to observe the signal, the invention would not have been lost, for at aproximately the same time, at least within a matter of days [MP: Actually a few weeks earlier], a similar experiment was carried out by Purcell and Pound, using microwave techniques, and there the nuclei observed was also protons, but from paraffin with which they have filled up a cavity. The phenomenon was announced in the letters of the Physical Review in the same issue, January 1946.

SS: Lessons learned:
- when everything else fails, quench the magnet and/or smash the equipment, but keep observing,
- there is always somebody else trying to achieve the same thing for which you have sacrificed a Christmas ...
- ... and he sure as Hell has a better detector.

This was followed, for me, by a period of two years which was totally unconnected with magnetic resonance. One heard more and more about it at Stanford University, of course, and I remember two further colloquia which exposed me to the subject. One was given by Packard himself, an enjoyable talk from an experimentalist's point of view. Another talk was given somewhat later by Professor Bloch, who had taken Packard with him for a stay of several months in Los Alamos where an apparatus had been assembled to detect the resonance from the compressed gas tritium, perhaps the only tritium in existence at that time. The object was, of course, to measure its magnetic moment, for the emphasis about this time was on the measurement of nuclear magnetic moments, and no doubt it was hoped that a precise measurement of these moments would provide a key to understanding the structure of the nucleus.

SS: Today we understand the structure of nuclides pretty well, yet we still find ab-initio computation of their moments nearly impossible.

The way that contact was taken up, as far as I am concerned, with nuclear magnetic resonance again was the following: I was earning a little money by correcting homework and examination papers for Professor Hansen, who was giving a course on microwaves. Professor Hansen came into the room where I was working one day and said that one of his graduate students had become more interested in another one of his ideas and he asked me to take up the work which his student (Dr.) R.F. Post (now at the University of California's Livermore Laboratories, an international authority on fusion methods) would like to drop. Professor Hansen had given the thesis problem of building an automatic instrument that was to search for the magnetic resonance from all stable isotopes.

It was a very shiny and complex apparatus with many chasses, which had been conceived, including a method of phase detection (which is of course now used in high resolution techniques) for being able to select U and V modes without having to tune for them at the probe. The reason for this was that the probe was simply unaccessible. The magnet had been designed by a weird-looking fellow named Russel Varian. I still hold many impressions of Russel Varian during those days, almost 20 years ago, but one or two of them will suffice for the moment.

It was lunch-time and, as students do, we walked out to the street to pile into an old automobile to drive to our midday meals home or elsewhere. There in the sunshine, in an open car, sat Russel Varian alone and thoughtfully eating his lunch. His lunch was simply a bunch of carrots which were scattered on the seat behind him. His head was leaning back restfully on the car seat so that the carrot was vertical, and as we watched, one carrot slowly was ground away only to be replaced by another one.

At one point I was bold enough to ask Professor Hansen why this fellow, who seemed to do little but wander around, was hired by the Physics Department. I still remember Hansen's reply: "You don't think he's doing anything, my boy, but the wheels are going around, the weels are going around."

It was this thinker who had designed the magnet [MP: The magnet was a follow-on of the design by Bitter of MIT]. He had realized that some magnetic moments would be very small, therefore the highest possible fields would be necessary. Consequently Russel Varian designed a magnet which was totally enclosed, that is to say, the magnet had complete cylindrical symmetry about a line running through the center of its poles. When looking at it, one was simply faced with a large cylinder of steel and there was obviously no access to the gap, which, of course, there must be. Instead of opening a window at one side, as one would presumably do today, Varian felt it to be more sensible to have long holes drilled through the pole pieces and the pole caps into the gap region since, NMR lines being broad, homogeneity was not a question but field strength was!

SS: I suppose that at the roots of every successful Company there is a person who knows or does something special. Like keeping carrots vertical while eating them. Seriously, though, did you notice the use of the symbols "U" and "V" for the two orthogonal signal modes? They are still used and I often wondered where that habit came from. It also looks like NMR quadrature detection was born for the mere necessity to overcome mechanical shortcomings of a magnet ... Pity that it was later abandoned for nearly a couple of decades and then re-discovered again.

It was with this instrument then that my practical introduction to Nuclear Magnetic Resonance began. It did not take me long to learn that the remaining part of this magnificient electronic apparatus was, for reasons I have forgotten, equally unsuited for the job it was built for. So I started off alone and built the apparatus which was described in the paper Phys.Rev. 79, 35 (1950). The work described in this paper became my dissertation, and once again, the point was to build an apparatus to locate resonances and measure the magnetic moments and spins of as many stable isotopes as possible. I had found about eight or ten.

Prof. Bloch persuaded me to stay on for another year, since this business was proving so highly successful. It was arranged that I should be joined by Dr. F.C. Yu who had just received his degree from Ohio State, so that, as a team, we might in another year possibly measure the majority of the yet unmeasured moments. This year with Dr. Yu was extremely successful and our work is described in the paper Phys.Rev. 81, 20, 1951. It was during this year with Dr. Yu that we discovered both the chemical shift and spin-spin interaction.

With respect to the former, we found that we were using nitrogen as a reference nucleus and wished to measure its magnetic moment more carefully (I believe the resonance of N14 had been seen and reported by Prof. R. Pound at Harvard a few months before). I remember going into the chemistry stockroom and searching for a compound which would give us the strongest possible resonance. I fell upon the bottle marked NE4NO3 because it was obviously very soluble in water and secondly there were two nitrogens per molecule. Downstairs we made up a concentrated solution and set it in the instrument. Dr. Yu and I were facing each other, leaning on the magnet, anticipating the resonance any moment as a small clock motor gradually changed the frequency of the spectrometer. As expected, a lovely resonance appeared on the recorder tape and, as we watched it, we were asounded to see the first resonance followed by a second one of equal amplitude. We repeated this measurement several times to make sure there was no possible instrumental error and then took the traces upstairs into Prof. Bloch's office. He became very excited and we discussed the reasons for it. In our discussions we decided there could be two reasons - one far more interesting than the other, namely that there were possibly stable nuclear isomers with slightly different properties! We were hoping that we had made such a discovery. The second reason could be, of course, that it was some nasty chemical phenomenon, most probably a difference in a diamagnetic shielding due to different electron densities at the nitrogen nucleus. It was at this conference that Prof. Bloch explained to me for the first time the next perturbation, the Van Vleck paramagnetism, which we later decided must contribute.

Our next job was to decide which of the two alternatives it was. Prof. Bloch telephoned to Prof. Segrè at the University of California in order to borrow a little N15. If the two resonances were not present in ammonium nitrate made up with N15, then obviously the explanation would be nuclear. Waiting for the preparation of this sample took about three weeks, which in our excitement was very agonizing. The sample arrived and within hours we had seen again two resonances with the same proportional shift between them, and hence we were led to the conclusion that this was an annoying chemical affect which could terribly impede our progress in trying to measure the magnitude of nuclear magnetic moments. Our work was described in a letter, Phys.Rev. 77, 717 (1950). By coincidence, the same issue carried news of the chemical shift of F19 which had been observed by W.C. Dickenson at M.I.T.

SS: These two paragraphs are so delightful to make one lough and have cold shivers all at the same time! Even the mistyped formula of ammonium nitrate (it is like that in the original) underlines a physicist's Freudian lack of trust when it comes to nasty chemistry. The two guys would not even think about preparing the solution in the chemistry lab upstairs. They would carry the bottle down (hiding it under a lab coat?) and do it there ... probably stirring the solution with a used coffee spoon. On the other hand, the investigation which pinned down the origin of the phenomenon is spotless.
     It is interesting that the discovery of chemical shifts, just like the discovery of NMR itself, happened nearly simultaneously and independently at two different places.

With respect to the spin-spin interaction, we were attempting to measure the magnet moments of the antimony isotopes, and had chosen the material NaSbF6 for its solubility and symmetry. This time we were astounded to see not two lines but a multiplet of five lines. (This was later shown to be seven). This was of course manifestly impossible unless the SbF6- ions were impeded in their rotation. I remember sitting down and working out the pattern of magnetic fields one would expect to find at the antimony nucleus if one had static distribution of six fluorine nuclei equally distributed over all angles of the magnetic field, taking into account the complexities imposed by the spin distribution. The magnitude of the splitting would be equal to the magnetic moment of the splitting would be equal to the magnetic moment of the fluorine divided by the cube of the interatomic spacing. The incredible thing was that our theory was a fairly satisfactory description of the multiplet we had observed, but most convincing was the correct magnitude of the distance between the lines. I remember being fairly happy happy with the explanation, but I remember equally well being totally unable to convince more experienced physicists or chemists of it. This did not, however, prevent Dr. Yu and myself from describing this unusual phenomenon in print. We were supported in our views when Prof E.R. Andrews suggested a mechanism which could explain the impeded motion: a long chain of alternating hydrogens and SbF6- ions. However it was about a year after this that Slichter at Illinois and E.Hahn, then an instructor at Stanford University, noticed these interactions in hydrogen spectra and simultaneously proposed the now well-known scalar interaction. Dharmatti and Weaver then went back to re-examine SbF6- spectrum to make sure that the new interaction was indeed correct.

SS: This is an excellent example of how a theory need not be correct for the mere reason that it fits the data. I feel great respect for those physicists and chemists who, on a hunch, did not trust the proposed theory even though it seemed to work. Anyway, who really discovered the J's as we know them today ? Certainly Proctor and Yu, in Felix Bloch's lab, were the ones to observe them first, but it was C.P. Slichter and E.L. Hahn to interpret them correctly (in the original, the two names are mistyped as Schlichter and Eahn). To be fair, we should also add the names of H.S. Gutowsky, D.W. McCall, E.B. McNeil and D.E. Maxwell (see Phys.Rev. 84, 1951, pages 589, 1245 and 1246).

The discovery that one could actually see chemical shifts in hydrogen spectra was made in 1951 at Stanford University by Packard, Arnold, Dharmatti. This was unfortunately a year that I was absent from Stanford, being at the University of Basle. The story, as it was explained to me, was the following. Arnold's dissertation was to measure again the magnetic moment of tritium [MP: possibly He3]. To do this he obviously needed a very homogeneous magnetic field, for the amount of tritium was limited and one wished to have lines which were as narrow as possible. Dharmatti, who was visiting from the Tata Institute, was working with him, and Packard, who by that time had received his degree, had stayed on as an instructor. They were working together to shim the magnet using a small probe to plot the magnetic field. For a reason I cannot explain, Dr. Dharmatti suggested that one use ethyl alcohol instead of water as a sample material [MP: To see if we could see the protein chemical shift]. Fortunately the job of magnet shimming had been done well enough that one was able to see on the oscilloscope the famous triplet of lines with the intensity ratio 1:2:3, corresponding to the proton concentration in the three chemical groups. The publication of this paper caught the eye of Dr. J. Shoolery who was just completing his work (in microvawe spectroscopy) at the California Institute of Technology, and he applied to Varian Associates for a job.

SS: The first hydrogen spectrum (today we would say proton spectrum) of ethanol is often reproduced in textbooks and heralded as being the very first spectrum showing chemical shifts. Which, as we have seen, is not true. What, to me, is even more interesting is the fact that ethanol was chosen thinking that, with its CH3, CH2 and OH groups, it might be a good model compound for proteins !!! Naturally, the guys thought in terms of a single chemical shift for water, another single one for protein, another for sugar, etc. Until they tried in practice ...

Shoolery was convinced that this phenomena was to play an important role in the future of chemical spectroscopy, and he is to be credited for this foresight. At the end of the term Packard moved to Varian, and two years later Arnold joined as well. Many other students of Bloch's group subsequently joined Varian.

I end my story by remarking that Varian Associates was started in 1948 by a group of about six men, who left good jobs and sold their homes to start the company. The Varian brothers, who had discovered the Klystron with Prof. Hansen, were obliged to take a license from Stanford University and from Sperry Gryoscope Company to start its manufacture. The second "Product Line" of the new company was to revolve about nuclear magnetic resonance. Russell Varian was convinced of its great value, but no one was quite sure what it was good for.

SS: Varian Associates have gone a long way. During my stay at University of Illinois (1971-1974) I have seen one of the very first NMR spectrometers build around a superconducting magnet (200 MHz) and it was marked Varian. Later there were wild ups and downs in the Corporation's attitude towards NMR (some of which most puzzling) but today Varian is again one of the few top-class manufacturers of NMR spectrometers.
 
I find it fitting to complement W.G.Proctor's exposition by the starting and closing paragraphs of the 1952 Nobel Prize lecture by Edward M.Purcell:
Professor Bloch has told you how one can detect the precession of the magnetic nuclei in a drop of water. Commonplace as such experiments have become in our laboratories, I have not yet lost a feeling of wonder, and of delight, that this delicate motion should reside in all the ordinary things around us, revealing itself only to him who looks for it. I remember, in the winter of our first experiments, just seven years ago, looking on snow with new eyes. There the snow lay around my doorstep - great heaps of protons quietly precessing in the earth's magnetic field. To see the world for a moment as something rich and strange is the private reward of many a discovery.
-------
This has been a long story and a complicated one. We are dealing not merely with a new tool but with a new subject, a subject I have called simply nuclear magnetism. If you will think of the history of ordinary magnetism - the electronic kind - you will remember that it has been rich in difficult and provocative problems, and full of surprises. Nuclear magnetism, so far as we have gone, is like that too.
 
E.M.Purcell, 1952. The full lecture can be downloaded in pdf format from Purcell's Nobel Prize web page.

 

TOP | Stan's LIBRARY | SCIENCE Links | NMR Links | Early NMR References. Stan's NMR Blog | Stan's HUB
Copyright ©2005: Stanislav Sýkora    DOI: 10.3247/SL1Nmr05.002 Design by Stan Sýkora