Phys. perspect. 1 (1999) 123–135 1422–6944/99/020123–13 $ 1.50+0.20/0

© Birkha¨user Verlag, Basel, 1999

Early History of Magnetic Resonance Norman F. Ramsey*

The early history of magnetic resonance to around 1950 is discussed from the point of view of a participant in it. I. I. Rabi’s theory of space quantization in a gyrating magnetic field and his molecular beam experiments in the 1930s laid the foundation of the magnetic resonance method, which he and his associates subsequently pursued and developed further at Columbia University, leading eventually to the development of NMR after World War II and the invention of the separated oscillatory fields method in 1950.

Key words: Molecular beams; nuclear magnetic resonance; magnetic resonance imaging; microwave spectroscopy; radiofrequency spectroscopy; oscillatory field method; nuclear magnetic moments; paramagnetic resonance; optical pumping.

Introduction Any history written by a participant has the advantage of providing added details and insights to the events described. But these add length to some descriptions and thereby under-emphasize events in which the writer did not participate. Alternatively, if the author describes only the events in which he participated, perspective is lost and unbalance is even worse. I have chosen the first of these alternatives, and I have written in the first person for events in which I actively participated. I end this ‘‘early’’ history around 1950.

Precursors to Magnetic Resonance During the two decades preceding the first successful magnetic resonance experiments there were several experiments and theoretical speculations related to, but different from, magnetic resonance. I became interested in this early period while preparing for my Ph.D. final exam. Since mine was the first Ph.D. thesis to use magnetic resonance,1 I anticipated that my examining committee might ask searching questions about the origins of magnetic resonance. Although I was not asked such questions, I was fascinated by how the ideas developed.2 * Norman F. Ramsey is Higgins Professor of Physics, Emeritus, at Harvard University. He received the Nobel Prize in Physics in 1989 for the invention of the separated oscillatory fields method and its use in the hydrogen maser and other atomic clocks. 123

124

N. F. Ramsey

Phys. perspect.

The earliest reported search for a dependence of magnetic susceptibility on frequency was that of M. H. Belz in 1922 with different solutions of paramagnetic salts.3 No frequency dependence was found. Acting on suggestions of H. F. E. Lenz and P. Ehrenfest, G. Breit four years later also searched for such a frequency dependence but found none.4 Perhaps this disappointment contributed to Breit’s decision to concentrate in theory, where he had a highly productive career. Theoretical ideas related to magnetic resonance go back to early quantum theory speculations and experiments on the changes in space quantization when the direction of a magnetic field is changed. The problem was first posed and partially solved in 1927 by C. G. Darwin,5 and his analysis was subsequently improved by P. Gu¨ttinger,6 by E. Majorana,7 and by L. Motz and M. E. Rose.8 In the period 1931 – 33, T. E. Phipps and O. Stern, and O. R. Frisch and E. Segre` in Stern’s Hamburg laboratory observed the changes in the space quantization when the direction of the magnetic field was changed,9 but their observations agreed only partially with theory. In 1936, I. I. Rabi pointed out that the discrepancy between theory and experiment was due to the neglect of nuclear spins in previous theories.10 These theories had neglected nuclear effects because the magnetic moment of the nucleus is about 2000 times smaller than that of the electron. But the spin angular momenta of nuclei and of the electron are comparable in size so the nucleus produces observable effects when the electron and nuclear angular momenta are tightly coupled together at low external fields. Although the techniques of microwave spectroscopy are quite different from those of magnetic resonance, there is sufficient similarity to note that in 1933 C. E. Cleeton and N. H. Williams, using magnetron generated microwave electric fields, observed a single broad spectral line corresponding to the tunneling transition of the NH3 molecule at a wave length of 1.1 cm.11 There were, however, no subsequent microwave spectroscopy experiments for the next thirteen years.12 Another different, but related, experiment was that of B. G. Lasarew and L. W. Schubnikow, which at low temperature showed that the nuclear magnetic moments in solid hydrogen contribute significantly to the observed static magnetic susceptibility of solid hydrogen.13 In 1936, using a calorimetric detection technique, C. J. Gorter successfully observed a frequency dependence of the paramagnetic relaxation of a number of alums.14 He found that the observed effects depend on the frequency n as n x, where x is a number, usually between 1 and 2; no resonance effects were observed. Utilizing the same calorimetric technique, Gorter attempted to observe resonant heating of a substance in a strong static magnetic field when subjected to a weak field oscillating at the precession frequency of a constituent nucleus, but found no resonances.15 In 1942, subsequent to the success of the molecular beam nuclear magnetic resonance experiments, Gorter and L. J. F. Broer reported unsuccessful attempts to observe nuclear magnetic resonance in powders of LiCl and KF.16 They tried to detect the resonance as a frequency pulling of a self-excited oscillator as the applied magnetic field was moved through the resonant value. The reasons for the failures remain a mystery, but they are probably due to detection inefficiencies since N. Bloembergen many years later was able to obtain a resonance with one of the same

Vol. 1 (1999)

Early History of Magnetic Resonance

125

crystals used by Gorter. The first published use of the name ‘‘nuclear magnetic resonance’’ is in Gorter’s 1942 paper where he attributes the coining of the phrase to I. I. Rabi. While Gorter was pursuing his unsuccessful experiments, I. I. Rabi and J. Schwinger were developing theories for the transitions induced when atoms or molecules in a molecular beam traverse successive regions of space with different directions of magnetic fields.17 In his brilliant 1937 paper, ‘‘Space Quantization in a Gyrating Magnetic Field,’’ Rabi assumed for simplicity that the fields were oscillatory in time even though the application was to molecules moving through fields that varied in space rather than time. Consequently, the formulae in that paper are applicable to resonances with oscillatory fields, and the paper, without alteration, provides the fundamental theory for all subsequent magnetic resonance experiments. However, the theoretical transition probability was written in a form that obscured its resonant nature while the subsequent averaging over molecular velocities eliminated sharp resonances, so Rabi did not immediately recognize the advantages of fields oscillating in time rather than space.

Molecular Beam Magnetic Resonance Methods After writing his paper on the gyrating field, Rabi discussed with some of his colleagues (Figure 1) the possibility of using oscillatory rather than space varying magnetic fields, but his laboratory had a full program of important experiments so

Fig. 1. Gathering of former students and friends for Rabi’s retirement from Columbia University in 1968. Left to right: Norman Ramsey, Jerrold Zacharias, Charles Townes, Rabi, V. W. Hughes, Julian Schwinger, Edward Purcell, William Nierenberg, Gregory Breit. Courtesy of Helen Rabi.

126

N. F. Ramsey

Phys. perspect.

no experiments utilizing oscillatory fields were seriously considered during the first five months following the publication of his paper on the gyrating magnetic field. In September 1937, Gorter visited Rabi’s laboratory at Columbia University* and described his brilliantly conceived but experimentally unsuccessful efforts to observe resonant heating of a substance in a strong static magnetic field when also subjected to a weak field oscillating at the precession frequency of the nucleus in the static field.18 Rabi then fully appreciated the advantage of using an oscillatory field and promptly invented the molecular beam magnetic resonance method. Two successful molecular beam magnetic resonance apparatuses were soon constructed by Rabi, J. R. Zacharias, S. Millman, and P. Kusch,19 and by J. M. B. Kellogg, Rabi, Ramsey and Zacharias.20 A schematic view of the method is shown in Figure 2. In these experiments the atoms or molecules are deflected by a first inhomogeneous magnetic field and refocused by a second one. When a resonance transition is induced in the region between the two inhomogeneous fields, the final beam intensity is reduced by the failure of refocusing. As a result, when the oscillator angular frequency v is equal to the Larmor angular frequency v0 of a nucleus in magnetic field B0, a sharp resonance is obtained for v =v0 = gIB0, where v0 is the angular precession frequency of a classical magnetized top with the same ratio gI of magnetic moment to angular momentum. As expected, the first magnetic resonance was observed by Rabi, Zacharias, Millman and Kusch (ref. 19), who were studying the easily detected LiCl molecule. Figure 3 shows the first reported nuclear magnetic resonance curve. Kellogg, Rabi, Ramsey, and Zacharias soon extended the method to the molecule H2 (ref. 20). When we started these experiments, we thought the only measurable interactions would be those of the magnetic moments with the external magnetic field so there should be a narrow single resonance such as Figure 2 for each magnetic moment. Instead we were disappointed to observe a broad irregular resonance region with small overlapping peaks that were hard to distinguish from noise. We then tried D2 and HD and found dominant single resonances from which we could accurately measure the proton and deuteron magnetic moments, which had been our primary objective. Columbia University than had a rule that each Ph.D. thesis should have a single author; so I was assigned the investigation of the backgrounds and peculiar H2 resonance patterns, which we thought were interesting, but not so fundamental that all of us should share the authorship. While the others in our group planned apparatus modifications to improve our resolution, I continued measurements with the old apparatus and soon found that we had been using much too strong oscillatory fields. Through reductions in the oscillatory fields, I found that our confusing broad resonance pattern for H2 became six separate well-resolved resonances. It then became clear that the multiple resonances in H2 were a radiofrequency spectrum of the molecule with protons interacting magnetically, not only with the external magnetic field but also with each other and with the magnetic field * Rabi refers to the visit of Gorter in a footnote to the first paper experimentally demonstrating a successful molecular-beam magnetic resonance (ref. 19), and Gorter twenty-nine years later published an article giving his own somewhat different recollections of the same visit (ref. 18).

Vol. 1 (1999) Early History of Magnetic Resonance 127

Fig. 2. Schematic diagram showing the principle of the first molecular beam magnetic resonance experiments. The two solid curves indicate two paths of molecules having different orientations that are not changed during passage through the apparatus. The two dashed curves in the region of the B magnet indicate two paths of molecules whose orientation has been changed in the C region so the refocusing is lost due to the change in the component along the direction of the magnetic field. Source: Rabi, et al., ‘‘Molecular Beam Resonance Method’’ (ref. 19), p. 527.

128

N. F. Ramsey

Phys. perspect.

Fig. 3. First observed nuclear magnetic resonance. The refocused beam intensity is plotted for various values of the magnetic field. One ampere corresponds to about 1.84 × 10 − 4 T. The frequency of the oscillating field was held constant at 3.518×106 Hz. Source: Rabi, et al., ‘‘New Method’’ (ref. 19), p. 318.

generated by the rotation of the molecule. The resonance would then occur when the angular frequency v of the oscillatory field was equal to the Bohr frequency v0 given by "v0 =Ei −Ef. Also, with weaker oscillatory fields, the ill-defined background fields we had observed in D2 and HD were suggestive of additional weak resonance peaks. The frequency spread of these peaks in D2 was even greater than in H2, which was surprising, since the nuclear moments were less and the molecule rotated more slowly, so the magnetic interactions should have been less. This suggested that there must be an additional nonmagnetic interaction in the molecule, such as that from an unexpected deuteron quadrupole moment. The four of us agreed that the structure of the multiple resonances and the deuteron electric dipole moment were too important for a single-authored thesis, since all of us had built the apparatus and discussed the interpretations. So I changed my thesis to the study of the resonances I had found associated with the

Vol. 1 (1999)

Early History of Magnetic Resonance

129

rotational magnetic moments of the molecules (ref. 1). The four of us, with the improved apparatus, went on to measure accurately the spin-spin and spin-rotational magnetic interactions and to confirm the existence of the deuteron quadrupole interactions (ref. 20). The observed spectrum for H2 is shown in Figure 4. These resonances were the first instances of multiple-line spectroscopy with coherent electromagnetic radiation, and our reports for the first time used the phrase ‘‘radiofrequency spectroscopy.’’ Rabi was a great thesis supervisor; he had fascinating original ideas on fundamental physics and other subjects. He encouraged creativity and gave his students and collaborators a great deal of freedom. He had high standards that the research be of high quality and fundamental interest. His critical remarks about uninteresting research annoyed many scientists, and he was sometimes wrong, but usually right; Rabi’s critical remarks often stimulated his friends and associates to even higher standards. The first molecular beam magnetic resonance experiments were with 1S molecules, but in 1940 Kusch, Millman and Rabi extended the method to paramagnetic atoms and in particular to DF = 9 1 transitions of atoms.21 For such transitions, the relative orientations of the nuclear and electronic magnetic moments are changed, so the resonance frequencies are primarily determined by internal properties of the atom. The same magnetic resonance technique was utilized by L. W. Alvarez and F. Bloch to measure the magnetic moment of the neutron with a neutron beam.22 The first publication on the neutron magnetic resonance came about two years after the first molecular beam magnetic resonance papers, so it is usually considered that the

Fig. 4. Radiofrequency spectrum of H2 in the vicinity of the proton resonance frequency. The resonance frequencies are primarily determined by the interaction of the proton magnetic moment with the external magnetic field, but the states of different mI and mJ are displaced relative to each other by the different values of the nuclear spin-spin and spin-rotational magnetic interaction energies. Source: Ramsey, Molecular Beams (ref. 2, 1956, 1985), p. 6. Reproduced by permission of Oxford University Press.

130

N. F. Ramsey

Phys. perspect.

neutron studies of Alvarez and Bloch were adaptations of the resonance methods developed earlier by Rabi and his associates. However, Alvarez has said that Bloch thought of doing the neutron beam resonance experiment before either of them had heard of the earlier molecular beam resonance experiments. It must have been a bitter disappointment to Bloch and Alvarez to learn that their clever idea for magnetic resonance had been anticipated. It is to their credit that they did not let this disappointment blight their research careers; instead each went on independently to win a Nobel Prize for other research. Molecular and neutron beam resonance experiments were discontinued during World War II. But shortly after the end of the war J. E. Nafe, E. B. Nelson and Rabi used the molecular beam magnetic resonance method to show that the hyperfine separation in the hydrogen atom was different from that expected theoretically.23 G. Breit pointed out that this could be caused by an anomalous magnetic moment of the electron,24 so Kusch and H. M. Foley started an experiment to measure the electron magnetic moment directly.25 As the anomaly in the atomic hydrogen hyperfine separation was being measured, W. E. Lamb and R. C. Retherford, with a different form of atomic beam apparatus, measured the atomic hydrogen fine structure.26 They showed that the separation of the 22S1/2 and 23P1/2 levels were 1057.77 Mhz different from the prediction of the Dirac theory. This large and accurately measured anomaly, called the Lamb shift, was quickly interpreted by H. A. Kramers, H. A. Bethe and others in terms of nonrelativistic quantum electrodynamics (QED). Soon thereafter, J. Schwinger developed relativistic QED,27 which accounted for the anomalous hyperfine separation and predicted an anomalous magnetic moment for the electron in agreement with the value measured by Kusch and Foley. During the first two decades of radiofrequency spectroscopy, little explicit use was made of the coherency of the radiation. In 1950 I invented the methods of separated or successive oscillatory fields in which the oscillatory field inducing the transitions is coherently applied in two short regions at the beginning and end of the uniform field region or as two short coherent pulses.28 These methods have a number of advantages including a narrower resonance, reduced first-order Doppler effects, and applicability at much higher frequencies. It is used in most accurate atomic clocks. I also studied theoretically the effectiveness of more than two pulses, of a variety of phase shifts between the coherent pulses and of different pulse shapes.29 Soon after my paper on the separated oscillatory fields method, E. L. Hahn invented the ingenious NMR spin-echo technique which uses several coherent pulses and is especially effective in measuring nuclear spin relaxations.30 Since then a variety of pulse shapes and phases have been used in NMR, in Fourier transform spectroscopy and in magnetic resonance imaging (MRI) by Hahn, C. Slichter, R. Ernst, W. Anderson, A. Pines, J. Waugh, W. Warren, P. Lauterbuhr and many others.

Electron Paramagnetic Resonance Experiments in Condensed Matter Despite his unsuccessful efforts to observe nuclear magnetic resonance, Gorter successfully observed paramagnetic relaxation in condensed matter but his attempts

Vol. 1 (1999)

Early History of Magnetic Resonance

131

to observe paramagnetic resonances failed (ref. 14). The first successful paramagnetic resonance experiments in condensed matter were those of E. Zavoisky, who in his Ph.D. thesis in 1944 observed paramagnetic resonance with CrCl3.31 Shortly after Zavoisky’s pioneering work, other observations of electron paramagnetic resonances were made by R. L. Cummerow and D. Halliday32 and many others.

NMR Following World War II, two groups independently developed nuclear magnetic resonance methods for liquids, solids and dense gases with the resonances being detected by their effects on the electromagnetic field. E. M. Purcell, H. C. Torrey and R. V. Pound at Harvard University observed the nuclear magnetic resonance absorption of the applied electromagnetic radiation,33 while F. Bloch, W. W. Hansen and M. Packard at Stanford University observed the resonance by the nuclear induction signal.34 The two experiments were independently conceived with the Harvard letter being received by the Physical Re6iew editor in December of 1945, just one month before that from Stanford. Many variations of these experiments soon followed,35 often with modified forms of the apparatus. Experiments in this field were given the collective name NMR (an acronym for nuclear magnetic resonance, with the three letters used to distinguish these experiments from the molecular beam experiments in which the term nuclear magnetic resonance earlier had also been applied). When gas or liquid samples are used in NMR there are fewer but narrower resonance lines than in crystals or in the molecular beam resonance method. Bloembergen, Purcell and Pound attributed this to the phenomenon of ‘‘collision narrowing’’ in which the molecular state is changed so frequently by collisions that its effective field at the nucleus is just its average value. Collision narrowing has the advantage of giving very narrow resonances, but the disadvantage that internal molecular interactions that can be studied with molecular beams can not be seen in NMR. For measuring nuclear magnetic moments, NMR and molecular beam resonance experiments have been complementary. The molecular beam method is usually best for finding an unknown magnetic moment since the search can be simplified by using a low magnetic field to reduce the region to be searched and by using a strong oscillatory field to broaden the resonance. By contrast, once the resonance is found, the narrow resonance in NMR permits a high precision measurement of its frequency. The relaxation times that can be measured in NMR can also provide useful information and identification as in MRI. The relaxation time T1 measures how long it takes test samples to return to their natural alignment after the magnetic field is removed, and T2 measures the duration of the signal from the sample. In determining nuclear magnetic moments from the observed resonance frequencies, a correction has to be made for the magnetic shielding of the nucleus by the circulation of the electrons induced by the static magnetic field. For a single atom

132

N. F. Ramsey

Phys. perspect.

the induced circulation is in simple circles and requires only a simple calculation, first provided by W. Lamb.36 In a polyatomic molecule the calculations are much more complicated since there is no single center of force and the induced motions are not circular. I derived and published a general expression for the magnetic shielding of polyatomic molecules,37 which was later extended and numerically applied by a number of theorists. One implication of magnetic shielding is that for the same atom in the same external field the resonance frequency will be slightly different in different molecules and in different locations in the same molecule. This chemical shift was observed by Bloembergen,38 W. D. Knight,39 W. C. Dickinson,40 W. G. Procter and F. C. Yu,41 and many others; it is the basis of the important use of NMR for chemical analysis. A different frequency shift in NMR was discovered experimentally by H. S. Gutowsky, D. W. McCall, C. P. Slichter and E. B. McNeil,42 and by E. L. Hahn and D. E. Maxwell.43 They found a shift proportional to I1 · I2 which they could not account for by any previously known interaction. Ramsey and Purcell pointed out that this could be attributed to an electron coupled nuclear spin-spin interaction.44 Such a newly considered interaction could arise from each nucleus interacting magnetically with the electron spin of its own atom. Since the electrons are tightly coupled together by the exchange coupling that leads to zero resultant electron spin for most molecules, the nuclear spins are effectively coupled together.

Optical Pumping In 1949 A. Kastler and J. Brossel developed the optical pumping technique for detecting magnetic resonances by exciting the atom with polarized light followed by spontaneous emission.45 Since the selection rules are different for excitation by polarized light and for de-excitation through spontaneous emission, the magnetic substrates of the ground state become unequally populated. These populations are changed when transitions are induced by a resonant radiofrequency field so the existence of the resonance can be detected from the change in the optical absorption. In subsequent years this method has been extended and extensively applied, but most of these developments were after the time period covered in this history.

Conclusions Although the history in this article terminates in 1950, progress did not stop then. The field of magnetic resonance has continued to advance rapidly up to the present, with the invention of revolutionary new devices and methods, such as the Overhauser effect, masers, lasers, Fourier transform spectroscopy, magnetic resonance imaging (MRI) and laser cooling. There is every reason to expect such rapid progress to continue for many years into the future.

Vol. 1 (1999)

Early History of Magnetic Resonance

133

References 1. N. F. Ramsey, Jr., ‘‘The Rotational Magnetic Moments of H2, D2, and HD Molecules. The Rotational Radiofrequency Spectra of H2, D2, and HD in Magnetic Fields,’’ Phys. Re6., 58 (1940), 226–236. 2. Norman F. Ramsey, Molecular Beams (Oxford: Clarendon Press, 1956; second ed. 1985); ‘‘History of Atomic Clocks,’’ Journal of Research of the National Bureau of Standards, 88 (1983), 301 – 320; ‘‘Early History of Magnetic Resonance,’’ Bulletin of Magnetic Resonance, 7 (1985), 94 – 99; ‘‘Origins of Magnetic Resonance,’’ in David M. Grant and Robin K. Harris, eds., Encyclopedia of Nuclear Magnetic Resonance. Vol. 1. Historical Perspecti6es (New York: Wiley, 1996), pp. 565 – 569. 3. Maurice H. Belz, ‘‘The Heterodyne Beat Method and some Applications to Physical Measurements,’’ Phil. Mag., 44 (1922), 479– 501. 4. G. Breit and H. Kamerlingh Onnes, ‘‘Magnetic Researches. XXVI. Measurements of magnetic permeabilities of chromium chloride and gadolinium sulphate at the boiling point of liquid hydrogen in alternating fluids of frequency 369,000 per second,’’ Communications from the Physical Laboratory of the Uni6ersity of Leiden, No. 168c (1926), 21 – 31. 5. C. G. Darwin, ‘‘Free Motion in the Wave Mechanics,’’ Proc. Roy. Soc., [A] 117 (1927), 258 – 293. 6. P. Gu¨ttinger, ‘‘Das Verhalten von Atomen im magnetischen Drehfeld,’’ Zeit. f. Phys., 73 (1931), 169–184. 7. Ettore Majorana, ‘‘Atomi Orientati in Campo Magnetico Variabile,’’ Nuo6o Cimento, 9 (1932), 43–50. 8. Lloyd Motz and M. E. Rose, ‘‘On Space Quantization in Time Varying Magnetic Fields,’’ Phys. Re6., 50 (1936), 348–355. 9. T. E. Phipps and O. Stern, ‘‘U8 ber die Einstellung der Richtungsquantelung,’’ Zeit. f. Phys., 73 (1931), 185–191; R. Frisch and E. Segre`, ‘‘U8 ber die Einstellung der Richtungsquantelung. II,’’ ibid., 80 (1933), 610–616. 10. I. I. Rabi, ‘‘On the Process of Space Quantization,’’ Phys. Re6., 49 (1936), 324 – 328. 11. C. E. Cleeton and N. H. Williams, ‘‘Electromagnetic Waves of 1.1 cm Wave-Length and the Absorption Spectrum of Ammonia,’’ ibid., 45 (1934), 234 – 237. 12. Robert Beringer, ‘‘The Absorption of One-Half Centimeter Electromagnetic Waves in Oxygen,’’ ibid., 70 (1946), 53–57; George E. Becker and Stanley H. Autler, ‘‘Water Vapor Absorption of Electromagnetic Radiation in the Centimeter Wave-Length Range,’’ ibid., pp. 300 – 307. Robert H. Dicke, Robert Beringer, Robert L. Khyl, and A. B. Vane, ‘‘Atmospheric Absorption Measurements with a Microwave Radiometer,’’ ibid., pp. 340 – 348. 13. B. G. Lasarew and L. W. Schubnikow, ‘‘Des Magnetische Moment des Protons,’’ Physikalische Zeitschrift der Sowjetunion, 11 (1937), 445 – 457. 14. C. J. Gorter, ‘‘Parametric Relaxation,’’ Physica, 3 (1936), 503 – 514; ‘‘Negative Result of an Attempt to Detect Nuclear Magnetic Spins,’’ ibid., pp. 995 – 998. 15. C. J. Gorter, ‘‘Paramagnetic Relaxation in a Transversal Magnetic Field,’’ ibid., pp. 1006 – 1008. 16. C. J. Gorter and L. J. F. Broer, ‘‘Negative Result of an Attempt to Observe Nuclear Magnetic Resonance in Solids,’’ ibid., 9 (1942), 591 – 596. 17. I. I. Rabi, ‘‘Space Quantization in a Gyrating Magnetic Field,’’ Phys. Re6., 51 (1937), 652 – 654; Julian Schwinger, ‘‘On Nonadiabatic Processes in Inhomogeneous Fields,’’ ibid., pp. 648 – 651. 18. C. J. Gorter, ‘‘Bad Luck in Attempts To Make Scientific Discoveries,’’ Physics Today, 20 (January 1967), 76–81. 19. I. I. Rabi, J. R. Zacharias, S. Millman and P. Kusch, ‘‘A New Method of Measuring Nuclear Magnetic Moment,’’ Phys. Re6., 53 (1938), 318; ‘‘The Molecular Beam Resonance Method for Measuring Nuclear Magnetic Moments. The Magnetic Moments of 3Li6, 3Li7 and 9F19,’’ ibid., 55 (1939), 526–535. 20. J. M. B. Kellogg, I. I. Rabi, N. F. Ramsey, Jr. and J. R. Zacharias, ‘‘An Electrical Quadrupole Moment of the Deuteron,’’ ibid., 55 (1939), 318 – 319; ‘‘Magnetic Moments of the Proton and the Deuteron,’’ ibid., p. 595; ‘‘An Electrical Quadrupole Moment of the Deuteron. The Radiofrequency Spectra of HD and D2 Molecules in a Magnetic Field,’’ ibid., 57 (1940), 677 – 695.

134

N. F. Ramsey

Phys. perspect.

21. P. Kusch, S. Millman and I. I. Rabi, ‘‘The Radiofrequency Spectra of Atoms. Hyperfine Structure and Zeeman Effect in the Ground State of Li6, Li7, K39 and K44,’’ ibid., pp. 765 – 780; S. Millman and P. Kusch, ‘‘On the Radiofrequency Spectra of Sodium, Rubidium and Cesium,’’ ibid., 58 (1940), 438–445. 22. Luis W. Alvarez and F. Bloch, ‘‘A Quantitative Determination of the Neutron Moment in Absolute Nuclear Magnetons,’’ ibid., 57 (1940), 111 – 122. 23. J. E. Nafe, E. B. Nelson, and I. I. Rabi, ‘‘The Hyperfine Structure of Atomic Hydrogen and Deuterium,’’ ibid., 71 (1947), 914–915; John E. Nafe and Edward B. Nelson, ‘‘The Hyperfine Structure of Hydrogen and Deuterium,’’ ibid., 73 (1948), 718 – 728. 24. G. Breit, ‘‘Does the Electron Have an Intrinsic Magnetic Moment?,’’ ibid., 72 (1947), 984. 25. P. Kusch and H. M. Foley, ‘‘Precision Measurement of the Ratio of the Atomic ‘g Values’ in the 2 P3/2 and 2P1/2 States of Gallium,’’ ibid., pp. 1256 – 1257. 26. Willis E. Lamb, Jr. and Robert C. Retherford, ‘‘Fine Structure of the Hydrogen Atom by a Microwave Method,’’ ibid., pp. 241 – 243. 27. Julian Schwinger, ‘‘On Quantum-Electrodynamics and the Magnetic Moment of the Electron,’’ ibid., 73 (1948), 416–418; ‘‘Quantum Electrodynamics. III. The Electromagnetic Properties of the Electron – Radiative Corrections to Scattering,’’ ibid., 76 (1949), 790 – 817. 28. Norman F. Ramsey, ‘‘A New Molecular Beam Magnetic Resonance Method,’’ ibid., 75 (1949), 1326; ‘‘A New Molecular Beam Resonance Method,’’ ibid., 76 (1949), 996; ‘‘A Molecular Beam Resonance Method with Separated Oscillating Fields,’’ ibid., 78 (1950), 695 – 699. 29. Norman F. Ramsey, ‘‘Resonance Experiments in Successive Oscillatory Fields,’’ Re6. Sci. Instruments, 28 (1957), 57–58; ‘‘Molecular Beam Resonances in Oscillatory Fields of Nonuniform Amplitudes and Phases,’’ Phys. Re6., 109 (1958), 822 – 825. 30. E. L. Hahn, ‘‘Spin Echoes,’’ ibid., 80 (1950), 580 – 594. 31. E. Zavoisky, ‘‘Paramagnetic Relaxation of Liquid Solutions for Perpendicular Fields,’’ Journal of Physics, Academy of Sciences of the USSR, 9 (1945), 211 – 216; ‘‘Spin Magnetic Resonance in the Decimetre-Wave Region,’’ ibid., 10 (1946), 197 – 198. 32. Robert L. Cummerow and David Halliday, ‘‘Paramagnetic Losses in Two Manganous Salts,’’ Phys. Re6., 70 (1946), 433. 33. E. M. Purcell, H. C. Torrey and R. V. Pound, ‘‘Resonance Absorption by Nuclear Magnetic Moments in a Solid,’’ ibid., 69 (1946), 37 – 38. 34. F. Bloch, W. W. Hansen, and Martin Packard, ‘‘Nuclear Induction,’’ ibid., p. 127. 35. References to more publications of the early scientists can be found in early textbooks on magnetic resonance such as Norman F. Ramsey, Nuclear Moments (New York: Wiley, 1953) and C. P. Slichter, Principles of Magnetic Resonance (Berlin and New York: Springer Verlag, 1978). 36. W. E. Lamb, Jr., ‘‘Internal Diamagnetic Fields,’’ Phys. Re6., 60 (1941), 817 – 819. 37. Norman F. Ramsey, ‘‘The Internal Diamagnetic Field Correction in Measurements of the Proton Magnetic Moment,’’ ibid., 77 (1950), 567; ‘‘Magnetic Shielding of Nuclei in Molecules,’’ ibid., 78 (1950), 699–703; ‘‘Dependence of Magnetic Shielding of Nuclei upon Molecular Orientation,’’ ibid., 83 (1951), 540–541; ‘‘Chemical Effects in Nuclear Magnetic Resonance and in Diamagnetic Susceptibility,’’ ibid., 86 (1952), 243 – 246. 38. N. Bloembergen, ‘‘Fine Structure of the Proton Magnetic Resonance in CuSO4 · 5H2O,’’ ibid., 75 (1949), 1326. 39. W. D. Knight, ‘‘Nuclear Magnetic Resonance Shift in Metals,’’ ibid., 76 (1949), 1259 – 1260. 40. W. C. Dickinson, ‘‘Dependence of the F19 Nuclear Resonance Position on Chemical Compound,’’ ibid., 77 (1950), 736–737; ‘‘Factors Influencing the Positions of Nuclear Magnetic Resonances,’’ ibid., 78 (1950), 339. 41. W. G. Proctor and F. C. Yu, ‘‘The Dependence of a Nuclear Magnetic Resonance Frequency upon Chemical Compound,’’ ibid., 77 (1950), 717. 42. H. S. Gutowsky and D. W. McCall, ‘‘Nuclear Magnetic Resonance Fine Structure in Liquids,’’ ibid., 82 (1951), 748–749; H. S. Gutowsky, D. W. McCall, and C. P. Slichter, ‘‘Coupling among Nuclear Magnetic Dipoles in Molecules,’’ ibid., 84 (1951), 589 – 590; E. B. McNeil, C. P. Slichter, and H. S. Gutowsky, ‘‘‘Slow Beats’ in F19 Nuclear Spin Echoes,’’ ibid., pp. 1245 – 1246. 43. E. L. Hahn and D. E. Maxwell, ‘‘Chemical Shift and Field Independent Frequency Modulation of the Spin Echo Envelope,’’ ibid., pp. 1246 – 1247; ‘‘Spin-Echo Measurements of Nuclear Spin-Spin

Vol. 1 (1999)

Early History of Magnetic Resonance

135

Coupling,’’ ibid., 85 (1952), 762; ‘‘Spin Echo Measurements of Nuclear Spin Coupling in Molecules,’’ ibid., 88 (1952), 1070–1084. 44. N. F. Ramsey and E. M. Purcell, ‘‘Interactions between Nuclear Spins in Molecules,’’ ibid., 85 (1952), 143–144; Norman F. Ramsey, ‘‘Electron Coupled Interactions between Nuclear Spins in Molecules,’’ ibid., 91 (1953), 303–307. 45. Jean Brossel and Alfred Kastler, ‘‘La de´tection de la re´sonance magne´tique des niveaux excite´s: l’effet de de´polarisation des radiations de re´sonance optique et de fluorescence,’’ Comptes Rendus, 229 (1949), 1213–1215; Alfred Kastler, ‘‘Quelques suggestions concernant la production optique et la de´tection optique d’une ine´galite´ de population des niveaux de quantification spatiale des atomes. Application a l’expe´rience de Stern et Gerlach et a la re´sonance magne´tique,’’ Journal de Physique et de Radium, 11 (1950), 255 – 265. Lyman Laboratory of Physics Harvard University Cambridge, MA 02138, USA e-mail: [email protected]

A Scientist’s Responsibility I. I. Rabi had this to say about a scientist’s responsibility: Our great need is for a better understanding on the part of the scientist that he has a real responsibility for science. This responsibility extends well beyond just doing good honest scientific work. He must try to understand the consequences of scientific discovery, and to communicate this understanding to the public. I do not mean that he must make propaganda for science, but rather that he must disseminate knowledge about science, what its place in our society is now and what it is to be in the future. Whether he thinks of himself in this way or not, the scientist is the custodian of the immense inherited wealth of discovery and advance in science and technology of all the ages. He alone has the key and therefore the access to this treasure. The scientist should take the responsibility of this position in our world as seriously as the physician should take his Hippocratic oath. I. I. Rabi, Science: The Center of Culture (New York and Cleveland: World Publishing, 1970), p. 114.

.