1939: Physics

Nuclear Fission.

It is always perilous to venture what future generations will accept as the weightiest discovery of any particular year, but for physics of 1939 there seems less risk than usual in making a guess: it is, the discovery of 'nuclear fission.' Several hundreds of transmutations had been produced and understood in the twenty previous years, but in no one of them did a massive atomic nucleus divide itself into two nearly equal parts. We now are acquainted with at least four examples of this novel type, for which the name was borrowed from biology where it is used to refer to the division of cells. In these four cases, the 'fissuring' nuclei are formed by the entry and absorption of neutrons into nuclei of the last three elements of the Periodic Table. viz., thorium (atomic number 90, symbol ₉₀Th), Protactinium (₉₁Pa) and uranium (₉₂U). With thorium and protactinium neutrons of high energy are required, and it is thought that the commonest isotope of each element is responsible for the whole effect. With uranium the fission is caused both by fast and by very slow neutrons, and it is thought that two different isotopes are affected, making up the total of four cases.

What was first observed was the formation of many radioactive bodies, which through their chemical properties have been identified as belonging to elements in the middle range of the Periodic Table: among them are isotopes of bromine (₃₅Br), krypton (₃₆r), rubidium (₃₇Rb), strontium (₃₈Sr), yttrium (₃₉Y), tellurium (₅₂Te), iodine (₅₃I), xenon (₅₄Xe), caesium (₅₅Cs), barium (₅₆Ba) and lanthanum (₅₇La). Some of these originate from others of the list, and it is still not clear which pair is directly formed in the initial partition, nor even whether it is always the same pair which is initially formed. Whichever the initial pair, the atomic numbers of its members must add up to the value characteristic of the neutron-bombarded element, e.g., 92 for uranium; for atomic number is proportional to nuclear charge, and the charge of the exploding nucleus must be divided between the fragments. We may as well suppose that uranium splits into barium (56) and krypton (36); for two important consequences can be derived from this assumption which would be equally true for any other fragment-pair, consisting of two elements lying near the middle of the Periodic Table, and having atomic numbers adding up to 92.

The first of these is, that the particles of the pair must jump apart with energies of tremendous motion (on the nuclear scale). This is because the masses of nuclei at rest depend upon atomic number in such a way that the sum of the masses of the uranium nucleus and the absorbed neutron is definitely and markedly greater than the sum of the masses of our hypothetical fragment-pair at rest. This superfluity of mass would according to older ideas be obliged to vanish; but according to the doctrine of relativity, the mass of a body increases with its speed, and we can assume that the fragments are moving just rapidly enough to raise their masses to the degree needed to take care of that apparent superfluity. The requisite speeds correspond to a kinetic energy now estimated at about 180 Mev for the two particles jointly. (One Mev = one million electron-volts = 1.5910^-6 erg.) Particles of such energy can make unprecedented amounts of sudden ionization in an ionization-chamber, and by virtue of this property they were observed in January 1939. They can also leap clear out of the superficial layers of a neutron-bombarded layer of uranium or thorium, and be collected by themselves on a suitably-located plate, which then shows the radioactivity of themselves and their descendants. Estimates of the kinetic energy of the particles can be made from their ionizing power, or from the thickness of air which they are able to traverse, or by thermal measurements upon the bombarded layer; they agree well with the theoretical value.

The second consequence is altogether different and yet flows also from the weights of the hypothetical fragments. Consider the fission of the uranium isotope of mass nearly 238. (Unit of mass: one-sixteenth of the mass of the commonest type of oxygen atom.) The explosive nucleus is formed by the entry into this of a neutron of mass nearly 1, and accordingly has a mass of nearly 239. The fragment-masses must then add up to a value nearer to 239 than to any other integer. The radioactive barium which is observed in the experiments is almost certainly the barium isotope of mass nearly 139, or Ba¹³⁹. If this is one of the fragments the other must be Kr¹⁰⁰. But the heaviest stable isotope of krypton is Kr⁸⁶, and to find an element having a stable isotope of mass nearly 100 we have to go along the Periodic Table all the way from krypton to molybdenum, which with its atomic number of 42 is six elements beyond krypton.

To put this result in general terms: whatever the initial fragment-pair, one at least (and probably each) of its members is bound to be a nucleus 'much too heavy for its atomic number.' But remembering that atomic number is a measure of nuclear positive charge, one sees that by ejecting a sufficient number of negative electrons (six, in the case of Kr¹⁰⁰) the nucleus could elevate its charge to such an extent that the actual mass would be stable for the new value of atomic number. The electrons might come off seriatim with intervals between; in such a case there would be a chain of radioactive substances, each an emitter of negative electrons. Again, the fragment might conceivably shed neutrons until its weight was reduced to a value compatible with its charge; or it might shed neutrons and electrons alternately. Consider also that different fissions may result in different initial fragment-pairs — and one sees that fission may be expected to result in an almost countless multitude of radioactive bodies emitting negative electrons, and also in the release of quantities of neutrons. Both of these expectations are verified by the facts.

The releasing of neutrons might conceivably have a dramatic, even a terrific consequence — the new-made neutrons might produce fresh fissions of nearby nuclei, these in turn fresh neutrons and so onward by cumulation, so that a block of thorium or uranium would be a high explosive liable to be detonated not only by any deliberate bombardment but even by any wandering neutron. If for this it were a sufficient condition that more than one fresh neutron should be produced at every fission, such an explosion would already have happened; for that condition is almost certainly realized (a recent estimate is that 3 to 4 neutrons are released in consequence of every fission). Luckily it is not a sufficient condition; partly because the likelihood of fission depends very much on the energy of the neutrons, and those released in these operations do not initially have the suitable energy; mainly because free neutrons can be captured and made harmless in ways which do not entail anything so violent as fission. It is even possible (and frequent) for a neutron to be captured by a nucleus of uranium 238 in such a way that the resulting nucleus of uranium 239 does not undergo fission at all, but remains in existence for hours or days, then emits a negative electron, and thenceforward is apparently stable for good and all. If the target consists of a mixture of other elements with the uranium, the possibilities of harmless capture are much increased by the others. It is still debatable whether the explosion might occur in an ideally pure block of uranium, or perhaps a block consisting of only one uranium isotope. The consensus of opinion at present is that it would not.*

* The names of some of those important in these discoveries are: Hahn, Strassmann (Berlin); Bohr, Frisch, Meitner (Copenhagen); Joliot, Curie, Savitch (Paris); Dunning, Tuve, McMillan, Abelson (U. S. A.). Some of the radioactive bodies resulting from fission were discovered years ago, but the elements of which they are isotopes were not properly identified and the mode of their formation was quite differently (and quite wrongly, as it now appears) conceived. The discovery of the initial fragments through their ionizing-power was made independently in many laboratories, first at Copenhagen; the isolation of the radioactive bodies by use of the ability of the fragments to leap out of the target was first achieved by Joliot.

Atom Nuclei.

While the phenomena of fission have diverted attention considerably from the more familiar types of transmutation, work on these continues apace with the aid of the ever-growing number of electrostatic generators and cyclotrons. So great a proportion of these is in America, that one may reasonably hope that nuclear physics is not destined to suffer a serious set-back from the war. The latest and greatest of cyclotrons, the new 60-inch built by Lawrence, and his colleagues at Berkeley, has proved itself competent of imparting energies of 25,000,000 electron-volts to deuterons and 50,000,000 to alpha particles. The scattering of beams of fast-moving protons and neutrons by hydrogen — which is to say, by assemblages of stationary protons — has been intensively studied for the information which it is capable of giving as to the short-range forces of attraction between protons and protons on the one hand, neutrons and protons on the other. These are the forces which hold together the constituents of atom-nuclei, and so make possible the existence of atoms.

Many atom-nuclei display the connected properties of magnetic moment (M) and angular momentum (P). There are several methods of ascertaining P, and two of determining M, but none is applicable to all nuclei and there are some nuclei which resist them all, so that knowledge and ignorance in this field have been strangely and unpleasantly mixed. Lately Rabi, with his colleagues Millman and Kusch, has developed a remarkable new method of measuring M/P, applicable to the nuclei of all atoms which can be built into chemical compounds satisfying certain fairly easy requirements as to vapor-pressure and stability. It depends on two facts, first that a nucleus traversing a steady magnetic field of strength H precesses about the direction of the field with a frequency (M/P)H and second that an oscillating field oriented at right angles to the steady field will have a pronounced effect on the nuclei when and only when its frequency of oscillation n is very close to the precession-frequency. The effect in question is a reorientation of the nuclei, and the experiment is so set up that the reoriented nuclei quit the beam in which they were traveling so that a detector set up in the prolongation of that beam reports a sharp falling-off in intensity when n passes through the value MH/P or H passes through the value of Pn/M. Very exact values of M/P have been obtained in this way for both isotopes of hydrogen, lithium, boron, potassium, chlorine and rubidium and for the principal isotope of nitrogen, fluorine, sodium, aluminium and caesium. In many of these cases P was already known, so that a good value of M could at once be computed.

Liquid Helium II.

Recent progress in the study of 'liquid helium II' makes 1939 a good year for describing the qualities of this remarkable fluid. (Another and a sadder reason is, that the description stands a chance of remaining up-to-date for years to come, since the laboratories where nearly all of the work has been done are mostly diverted by now to the uses of war.) When gaseous helium liquefies at 4.2° K. or thereabouts the liquid at first presents no very singular properties, but when it is further cooled it undergoes at 2.19° K. a sudden transformation and becomes a substance which, though dense and mobile as a liquid, behaves as though it were less viscous than any gas and yet a better conductor of heat than solid copper itself. One experimenter (Kapitza) planned to separate two optically-flat surfaces by a narrow gap and let the liquid flow slowly between them, but found that even when the surfaces were actually resting one upon the other the helium slid between them at so rapid a rate that no measurement could be made.

When descending through narrow capillaries (from a few tenths to a few thousandths of a mm. in diameter) under the pull of gravity, this fluid violates all of the laws made familiar by a century of study of ordinary liquids and ordinary gases. For a tube of constant length, the velocity of the fluid increases less rapidly than the pressure-head; for a constant pressure-head the velocity decreases less rapidly than the reciprocal of the length; when tubes of different bores are compared, the velocity passes through a minimum for a certain value of diameter. For the narrowest available tubes (which are really slits between wires squeezed tightly together) the velocity is almost independent of the pressure and the gradient thereof.

Measurements of the transport of heat are also made with columns of liquid helium II contained in narrow capillaries. It is found that the rate of flow of heat along the column is not proportional to the temperature-difference between the end thereof; the ratio of these two quantities is highest when the two have their lowest measurable values, so that people speak of helium II as having a thermal conductivity which decreases with increasing heat-flow. Curious phenomena involving motion occur in these experiments; the liquid helium tends to creep bodily along the capillary in the opposite direction to the heat-flow, and if instead of being in a tube the liquid is perfusing the crevices between the particles of a tightly-packed powder, it may be made to spring in a jet several inches high by transfer of heat to the powder (by shining the light of an ordinary flashlight against it).

The strange facility of motion suggested by some of these observations is displayed very clearly in experiments on beakers or jars with open tops. If such a beaker, empty, is lowered partway into a bowl of liquid helium II, the liquid promptly fills it up until the menisci within and without stand at just the same height; if the partly-filled beaker is now lifted out, the liquid comes out of it rapidly though invisibly, collects in droplets on the bottom and drips back into the bowl. The influx or efflux is at a rate strictly proportional to the periphery of the beaker; it is therefore in a flowing surface-film that the liquid travels. To evaluate the 'speed' of the film in cc. of drained-off liquid per second per cm. of periphery is relatively easy; to evaluate the literal speed in cm/sec requires however a knowledge of the surface-density of the film, and this is obtained by allowing a broad expanse of coiled metal to cover itself with the film and then (when heated) to shed the film into a very narrow calibrated capillary closed at the bottom, in which the liquid derived from the film can be seen and its amount can be estimated. The figure given for the speed is 20 cm/sec at 1.1° K.; it is not dependent on the difference in height between the menisci within and without. It seems quite probable that the curiosities of the passage of the liquid through narrow channels and the passage of heat through narrow columns of the liquid are partly at least attributable to this remarkable type of motion.* See also NAVAL SCIENCE, AMERICAN.

* Some of the outstanding names in this field of research are Onnes, Keesom, Miss Keesom (Leyden); Mendelssohn, Daunt, Simon (Oxford); Allen, Jones, Misener (Cambridge); Burton (Toronto).