Physicists' Response to the Challenge of X-rays

HISTORY oF SciENCE SociETY 1995 ANNUAL MEETING 26-29 October 1995 Minneapolis, Minnesota Radisson Hotel Metrodome CoNTENTS HSS Officers/Program Chairs 3 Acknowledgements 3 General Information 4 Maps 5 Twin Cities Points of Interest 7 Selected Restaruant and Club List 15 Overview Schedule of Events 18 PROGRAM SCHEDULE 22 HSS Distinguished Lecture 43 Index of Participants 48 Book Exhibito"' 51 Advertisers 52 Friday. 27 October Richard W: Burkhardt, Jr. (University of Illinois, Urbana-Champaign): ''Adaptive Radiation at Oxford: Niko Tinbergen and the Reformation of the Aims and Methods of Ethology" Joe Cain (UniversityofMinnesota): "Going Public: Post-Synthesis Popular Writings of George Gaylord Simpson" 24. Centennial of Roentgen's Discovery ofX-Rays BAKKEN LIBRARY CHAIR: *David J. Rhees (The Balrken Library and Museum) Spencer Weart (American Institute of Physics): "Roentgen Before the Roentgen Rays" Nahum Kipnis (The Balrken Library and Museum): "Physicists' Response to the Challenge ofX-Rays, 1895-1912" Joel Howell (University of Michigan): "Early Diagnostic Radiology: Machines, Pictures, and Power, 1895-1925" CoMMENTATOR: Nancy Knight (American College of Radiology) NOTE: This session will be held at the Balrken Library and Museum. Transportation to and &om The Balrken will be provided. "Perfecting Tradition and Re-thinking Revolution: Discarded Images and New Visions of the Scientific Revolution" Special Session in Honor of David C. Lindberg's 60th Birthday NOLTE ROOM Chair: David C. Lindberg (University ofWisconsim, Madison) A. Mark Smith (University of Missouri-Columbia): "Through a Glass Darkly: The Problem of Image-formation in Medieval and Renaissance Optics" William B. Ashworth (University of Missouri-Kansas City): "Visual Perceptions: Images, Optics, and the Scientific Revolution" ｒｯ｢ｾｲｴ＠ Harch (University ofFlorida): "A&er Images: The Retina, the Witness, the Private Eye" Commentator: Robert S. Westman (University of CaliforniaSan Diego) l'1S S Minneupocs l0 , t7. 5i Physicistsr Response to the Challenge of X rays Nahum Kipnis Bakken Library and Museum 3537 Zenith Ave S, Minneapolis, MN 55416 USA 612-927 -6508 ; [email protected] The quest for establishing the nature of X rays ended in 1922 with Arthur Compton's explanation of the wavelength change in X-ray scattering, and its last decade was discussed in detail by Roger Stuewer. Here I am going to talk about the earlier period ending in 1912 with the discovery of diffraction of X rays by Laue, Friedrich and Knipping. My main focus will be on the first two years after Rdntgen's discovery. That was the "heroic" period, when the nature of X rays was of a primary concern. It lasted long enough for all competing theories to emerge and for physicists to realize that the direct attack on the problem had failed, and that the solution is not at all close. I. The Discovery 1. Riintgen In the summer of 1895, Wilhelm Konrad Rcintgen (1845-1923) decided to switch to a new research area: the cathode rays. He began with repeating some experiments of his predecessors, including those of Philip Lenard. On November 8, 1895, while checking the dark cover of his cathode-rays apparatus, he noticed that a fluorescent screen laying on the table shone in the dark. Rrintgen conducted an intensive investigation and communicated its results to the President of Wiirzburg PhysicalMedical Society on December 28,1895. The new phenomenon possessed properties that it was difficult to reconcile. On the one hand, the new agent behaved like light, producing sharp images on a photographic plate, and for this reason Rdntgen called it the "X rays." On the other hand, these rays differed from light: they easily penetrated many opaque bodies; they could not be regularly reflected or refracted; and they showed no sign of either polarization or interference. Thus, Rontgen concluded that, despite their similarity to ultaviolet light in their ability to produce fluorescence and chemical reactions and discharge electrified plates , X rays could not be transverse waves. Nor could they be cathode rays exiting the tube, because X rays had much greater penetrative @ 1995 by Nahum Kipnis .. 2 power than cathode rays, and they did not deviate even in a strong magnetic field. Rontgen decided that the only choice open was to assume that X rays were longitudinal waves in the ether. To understand his arguments, we need to review the contemporary views on cathode rays and on longitudinal waves. 2. Background: Cathode Rays By 1895, an electrical discharge in a low pressure gas has been a subject of intensive studies. A glass tube with two metal electrodes when connected to a source of high voltage displayed a number of fascinating phenomena, beginning with light inside it, and a fluorescent spot on the glass wall that could be moved by a magnet. It appeared that something moved away from the cathode producing shadows, heating an obstacle, and exerting pressure on it. Two beams produced by adjacent cathodes appeared to repel one another. William Crookes and other English physicists explained these phenomena by a movement of negatively charged particles repelled by the cathode. Crookes in particular thought they were molecules of the residual gas. However, in the next 15 years German physicists found a number of objections to his theory and favored a wave explanation of cathode rays. In particular, E. Wiedemann found that closing one of the two beams did not affect the trajectory of the other. Heinrich Hertz discovered a fluorescent spot within a shadow of a thin metal screen, which he interpreted as the result of the cathode rays passing through the screen. Besides, he could not obtain any declination of the cathode rays by an electric field. Finally, Philip Lenard managed to let the cathode rays out of the discharge tube through a very thin aluminum window. In the vicinity of the window, he observed fluorescence and obtained a photograph of an object placed in an aluminum box. By attaching to the window another evacuated tube Lenard could observe how cathode rays behave there at different pressure. He found two types of rays, which considerably differed in the degree of deviation by a magnet. Lenard thought that these phenomena spoke against particles and for some role of ether waves with very small wavelengths. The corpuscular theory recovered to some extent at the end of 1895 when Jean Perren definitely demonstrated that cathode rays carry a negative charge and deviate in an electric field. However, a number of questions remained to be answered, and the lack of those answers strongly affected the understanding not only of cathode rays but also of their offspring, X rays. 3 3. Background: Longitudinal waves We see that Rontgen had reasons to dismiss the corpuscular theory of X rays. But did he opt for such an extravagant solution as longitudinal waves? Actually, in 1895 such an idea was not at all extravagant. Since 1830, no theory of light whether it assumed the ether to be a solid body or an electromagnetic medium, could explain the transmission of light by a transparent body without assuming a partnership of t transverse and longitudinal waves in this. Almost all theoreticians used longitudinal waves in their theories of optical phenomena in the hope that sooner or later experimenters will find a way to detect them. That is why Rontgen's discovery appeared to a number of physicists to be the long awaited manifestation of the existence of longitudinal waves. II. Physicists take the challenge The first articles on X rays began to appear late in January 1896, about 3 weeks after Rontgen's discovery became known. While most of them dealt with repeating and modifying his experiments, there was no shortage in attempts to explain the phenomenon. Despite Rontgen's reputation of a thorough and a careful investigator, all his experiments and theoretical conclusions were re-examined. The work on the subject was so intensive that new findings were reported almost on a weekly basis, and, as it is only natural for such a rush, not all of them were correct. Several theories sprang out almost simultaneously, and each of them had a big name attached to it. 1. Longitudinal waves? Let's begin with Rontgen's hypothesis. It enjoyed a comfortable support, in particular from such renown scientists as Boltzmann and Lord Kelvin. On February 13, 1896, Kelvin presented a paper to the Royal Society where he suggested generating longitudinal waves by means of an electrical discharge. A circular metal plate was placed inside a metal box, insulated from it, and charged. Then an electrical discharge was produced between the bottom of the plate and the box. In Kelvin's view, the electromagnetic waves created above the plate were primarily longitudinal. They could affect a photographic plate placed in the upper part of the box or even outside it. Kelvin's idea was based on an experiment by Lord Blythwood that allegedly proved that X rays can be created without a vacuum tube. In this experiment, a photographic plate in a wooden holder was wrapped into a black cloth and placed close to a working powerful electrostatic generator. The • 4 interpretation of this experiment was erroneous, of course, but this error was quite typical for the time. It must be noted, however, that Kelvin had a suspicion that Blythwood's result could have been a direct effect of electricity rather than of X rays. Since no one ever observed longitudinal waves and their exact properties were unknown, it was difficult to conceive experiments revealing their effects. As the result, evidence in their favor was gathered from experiment consistent with the idea of waves but contradicting their transverse character. One was the absence of polarization, another was the fact that, unlike ultra-violet rays, X rays discharged equally well positive and negative charges. Still another was J. Precht's interference experiment, from which he found wavelengths for X rays within the range of visible light. Since these rays penetrated black paper, which transverse visible rays don't do, he interpreted them as longitudinal waves. 2. Transverse waves? Surprisingly, but transverse waves, dismissed by Rontgen, was the first theory to challenge his own. The major problem was how to deal with the absence of refraction. As early as January 23, 1896, Arthur Schuster suggested that if the properties of the ether remain constant within the sphere of action of a molecule, and if the wavelength of X rays were comparable with this distance, such waves could travel in the substance with the same speed as in vacuum, which means no refraction. As to the origin of necessary very high frequency of waves, he pointed to electrical vibrations within the molecule. Soon D. Godhammer noted that in contemporary theories of dispersion the index of refraction for extremely short waves (A.zO} must be very close to 1, which means no refraction. Another problem was the absence of diffraction. Gaston Sagnac pointed out that in Fresnel's theory of diffraction, for extremely small wavelengths diffraction effects become unnoticeable. Polarization was an even more difficult obstacle. Since X rays did not reflect, refract, or diffract, the only hope for observing polarization was by different absorption in a pair of double-refracting crystals at different orientation of their optical axes . However, the results of most experiments were negative. Another way of determining polarization of X rays would be by changing the color of fluorescence of a crystal at a certain orientation. That was known for ultra-violet light , but nothing of the sort was observed for X rays. It took some time for physicists to realize that the technical means in their possession were inadequate to observe the desired effects. In particular, one way of 5 testing the transverse wave theory was by proving that the wavelength of X rays is much smaller than those of visual light, which could be done in experiments on interference or diffraction. Experimenters found, however, great difficulties in measuring very small diffraction. First, the photographic technique could not provide a necessary precision. Secondly, sometimes it was difficult to distinguish the true effects from the false ones. A few positive observations of an enlargement of the shadow or fringes (Bungenziano, L. Fomm, L. Calmette and G. Lhuillier) were explained by critics as due to a finite size of the source of X rays or some optical illusions. The only positive result in experiments on polarization (Golitsyn and Karnojitsky) was explained by a lack of homogeneity in crystals of tourmaline. Lack of homogeneity of x rays was another cause suggested as masking the effect of their interference. For these reasons the wavelengths determined were only estimates, in many cases deduced from negative results. And since they varied from 10 A to 8300 A, it was not good enough to support any wave theory of X rays. And here particles came back on the stage 3. Cathode rays/particles? From the very beginning, a number of Continental physicists viewed the common features that X rays had with the cathode rays and with the Lenard rays to be too important to dismiss their connection, as Rontgen had initially done. They argued that each of these radiations was very heterogeneous as to both their absorbability and magnetic deviation, and the difference in this respect between X rays and the cathode rays could be relative rather than absolute. In other words, X rays, with their minimal absorbability and zero of magnetic declination, were supposed to be one extreme of the continuum of the cathode rays. Later in 1896, the interest in the analogy between X rays and the cathode rays declined, but in 1897, it came back when Lenard and Rontgen became its chief proponents. Some physicists thought that one of the causes of the heterogeneity of cathode rays was that they consisted of both charged and neutral particles, with X rays representing the neutral kind. In their view, the dependence of the absorption of X rays on the density of substances could not be explained by their wave nature. A. Battelli, A. Garbasso, and Salvioni viewed X rays are a part of the cathode rays inside the tube that were filtered out when exiting it. According to Oliver Lodge, X rays consist of those particles in the cathode-ray beam that lost their charge when colliding with the glass wall or the anti-cathode. The neutrality of these particles explains the absence of deviation of X rays in a magnetic field and their high 6 penetrability. On the other hand, A. Vosmaer and F. Ortt believed that the X rays consist of a mixture of positive, negative, and neutral particles in various proportions. As the evidence, they cited an experiment by Lafay, in which X rays acquired the property to deviate in a magnetic field after passing an electrified metal screen. T. Porter repeated the experiment with the negative result. B. Walter, however, deemed Porter's experiments inconclusive, since the electrostatic repulsion was notorious for being difficult to show even for charged cathode rays. 4. Ether vortices? Some theories of X rays were quite exotic. One of them was Albert Michelson's "ether-vortex" theory. These vortices were supposed to originate at the surface of the cathode and be projected out from there. It was known from mechanics that vortices are capable of rectilinear propagation, but not of reflection, refraction, polarization, or interference. According to Michelson, an ether-vortex can pass through a substance in the same way as a smoke-ring passes through a wire gauze screen. Michelson never tried to mathematize his hypothesis, and it didn't excite much interest. 5. Sound waves? Another curiosity was Thomas Edison's theory of X rays as sound waves of very small wavelengths (which means longitudinal waves much longer than those of light). In his view, the shadows shown in photographs taken in X rays resembled those produced by sound. To support this theory, he described an experiment where X rays penetrated deep inside the shadow cast by a steel plate. A short debate around this theory ended when it was shown that, unlike sound, X rays can travel in vacuum. 6. Pulses? Apparently, the first idea of pulses was conceived independently by Wiechert and Stokes in 1896, and Stokes' theory was published early in 1897. To him, absence of interference meant that the radiation was not periodicaL He supposed that X rays consisted of a long succession of transverse pulses, produced when the charged molecules emitted by the cathode strike the glass or another target. Since such pulses are independent, they cannot produce coherent vibrations at any point of a screen, thus there is no interference, and no diffraction can be observed. According to Stokes, the absence of refraction was due to a peculiar mechanism of 7 energy transfer. When a light wave falls upon a body, it excites vibrations of the ether inside it which after some time are transferred to the molecules of the body. When the wave is periodical and lasts long enough, this does not pose any problem. As the result of join vibrations a part of the original kinetic energy of the ether is absorbed by the body's molecules, which reduces the energy of the ether vibrations inside the body, which means a reduction in the speed of propagation of light, or refraction. The situation is different, however, for X rays. Since their pulses strike a body independently of one another, they cannot establish any continuous vibrations inside it, there is no resonance, and thus there is no refraction. Early in 1898, J. J. Thomson reinterpreted Stokes's idea of pulses in terms of electromagnetic theory of light. He argued that X rays are thin pulses of electric and magnetic disturbances created when small negatively charged particles moving with a very high velocity are stopped by an obstacle. The pulse's thickness equals the particle's diameter. If the pulses are so thin that the time they take to pass over a molecule of a substance is much shorter than its period of vibrations, there will be neither refraction nor diffraction. 7. Changing views What was the fate of all these theories? Since the late 1896, transverse waves became the favorite theory, with neutral particles, and pulses being major contenders. During the first period, the popularity of transverse wave grew despite continuing failures of diffraction and polarization experiments. Apparently, physicists favored this theory because it was well developed and definitely demonstrated in another area (optics), and because it did not contradict most experiments with X rays (assuming very small wavelength for them). The appearance of the new theory of pulses (1897) did not diminish the interest in the other two, on the contrary, their adherents increased in numbers, in particular, Rontgen switched to particle theory. The ups and downs of these theories were primarily due to erroneous experiments rather than an improvement of experimental techniques. The pulse theory was brought to life not by new experiments but rather as an alternative explanation of the old negative evidence against transversal waves (Stokes) or as a theory of the cathode rays G.J. Thomson). During the second period (1898-1912) X- rays research changes its character. There was a sharp decline of interest in the nature of the rays. Very few physicists continued to experiment with diffraction of X rays. For instance, throughout this 8 period, Haga & Wind on one side and, B. Walter & R. Pohl on the other argued whether an enlargement of the image of a narrow slit was due to diffraction or other reasons. In 1912, Sommerfeld tried to summarize their results using new microphotometrical measurements of their photogramms by P. Koch. However, like his predecessors he only established the upper limit of the wavelength, 0.4 A, which corresponded to his theory. Most physicists however abandoned this apparently hopeless line of research and started studying what was within their reach, which happened to be absorption of X rays. It was a purely empirical research that did not promise to shed any new light on the nature of X rays. However, it was this research that eventually brought about, as a by-product, the results wanted in the beginning. Charles Barkla demonstrated in 1905 a partial polarization of primary X rays and a total polarization of secondary X rays. Subsequently, he discovered that the X ray spectra of heavy metals were linear rather than continuous. This concept of monochromatic X rays inspired Laue in 1912 to try measuring their wavelength using a new type of a diffraction grating, a crystal, which was accomplished by Friedrich and Knipping. While that was an important event in the history of physics, the contemporaries saw its importance elsewhere but not in demonstrating that X rays were periodical transverse waves. To those who wanted to believe in it, it was old news, while the reasons inspiring their opponents still preserved their power: the theory of pulses was very successful in absorption, and the idea of neutral particles retained its appeal. Only later was it realized that each of these theories represented one aspect of the same entity, which became known as the "particle-wave duality." Thus the quest for the nature of X rays exhibits several unusual features. One was the ease with which physicists initially chose theories out of several options despite the absence of positive evidence. Another was their satisfaction with this situation, which resulted in abandoning the quest for the true theory in favor of empirical research. Finally, the view on X rays that eventually evolved consisted of parts of three competing theories. However unusual these features appear, it can be of interest to check their presence in the reception of other theories as well. I