Atomic Energy for Military Purposes (The Smyth Report)

The Official Report on the Development of the Atomic Bomb Under the Auspices of the United States Government


In Chapter I we pointed out the importance of ionization in the study of radioactivity and mentioned the electroscope. In this appendix we shall mention one method of historical importance comparable with the electroscope but no longer used, and then we shall review the various methods now in use for observing alpha particles, beta particles (or positrons), gamma rays, and neutrons, or their effects.


The closest approach that can be made to "seeing" an atom is to see the bright flash of light that an alpha particle or high-speed proton makes when it strikes a fluorescent screen. All that is required is a piece of glass covered with zinc sulphide, a low- power microscope, a dark room, a well-rested eye, and a source of alpha particles. Most of Rutherford's famous experiments, including that mentioned in paragraph 1.17, involved "counting" scintiliations but the method is tedious and, as far as the author knows, has been entirely superseded by electrical methods.


When a high-speed charged particle like an alpha particle or a high-speed electron passes through matter, it disrupts the molecules that it strikes by reason of the electrical forces between the charged particle and the electrons in the molecule. If the material is gaseous, the resultant fragments or ions may move apart and, if there is an electric field present, the electrons knocked out of the molecules move in one direction and the residual positive ions in another direction. A beta particle with a million electron volts energy will produce some 18,000 ionized atoms before it is stopped completely since on the average it uses up about 60 volts energy in each ionizing collision. Since each ionization process gives both a positive and a negative ion, there is a total of 36,000 charges set free by one high-speed electron, but since each charge is only 1.6 × 10-19 coulomb, the total is only about 6 × 10-15 coulomb and is still very minute. The best galvanometer can be made to measure a charge of about 10-10 coulomb. It is possible to push the sensitivity of an electrometer to about 10-16 coulomb but the electrometer is a very inconvenient instrument to use.

An alpha particle produces amounts of ionization comparable with the beta particle. It is stopped more rapidly, but it produces more ions per unit of path. A gamma ray is much less efficient as an ionizer since the process is quite different. It does occasionally set free an electron from a molecule by Compton scattering or the photoelectric effect, and this secondary electron has enough energy to produce ionization. A neutron, as we have already mentioned in the text, produces ionization only indirectly by giving high velocity to a nucleus by elastic collision, or by disrupting a nucleus with resultant ionization by the fragments.

If we are to detect the ionizing effects of these particles, we must evidently use the resultant effect of a great many particles or have very sensitive means of measuring electric currents.


Essentially the electroscope determines to what degree the air immediately around it has become conducting as the result of the ions produced in it.

The simplest form of electroscope is a strip of gold leaf a few centimeters long, suspended by a hinge from a vertical insulated rod. If the rod is charged, the gold leaf also takes up the same charge and stands out at an angle as a result of the repulsion of like charges. As the charge leaks away, the leaf gradually swings down against the rod, and the rate at which it moves is a measure of the conductivity of the air surrounding it.

A more rugged form of electroscope was devised by C. C Lauritsen, who substituted a quartz fiber for the gold leaf and used the elasticity of the fiber as the restoring force instead of gravity. The fiber is made conducting by a thin coating of metal. Again the instrument is charged, and the fiber, after initial deflection, gradually comes back to its uncharged position. The position of the fiber is read in a low-power microscope. These instruments can be made portable and rugged and fairly sensitive. They are the standard field instrument for testing the level of gamma radiation, particularly as a safeguard against dangerous exposure.


An ionization chamber measures the total number of ions produced directly in it. It usually consists of two plane electrodes between which there is a strong enough electric field to draw all the ions to the electrodes before they recombine but not strong enough to produce secondary ions as in the instruments we shall describe presently.

By careful design and the use of sensitive amplifiers an ionization chamber can measure a number of ions as low as that produced by a single alpha particle, or it can be used much like an electroscope to measure the total amount of ionizing radiation present instantaneously, or it can be arranged to give the total amount of ionization that has occurred over a period of time.


While ionization chambers can be made which will respond to single alpha particles, it is far more convenient to use a self amplifying device, that is, to make the ions originally produced make other ions in the same region so that the amplifier circuits need not be so sensitive.

In a proportional counter one of the electrodes is a fine wire along the axis of the second electrode, which is a hollow cylinder. The effect of the wire is to give strong electric field strengths close to it even for relatively small potential differences between it and the other electrode. This strong field quickly accelerates the primary ions formed by the alpha or beta particle or photon, and these accelerated primary ions (particularly the electrons) in turn form secondary ions in the gas with which the counter is filled so that the total pulse of current is much increased.

It is possible to design and operate such counters in such a way that the total number of ions formed is proportional to the number of primary ions formed. Thus after amplification a current pulse can be seen on an oscilloscope, the height of which will indicate how effective an ionizer the initial particle was. It is quite easy to distinguish in this way between alpha particles and beta particles and photons, and the circuits can be arranged to count only the pulses of greater than a chosen magnitude. Thus a proportional counter can count alpha particles against a background of betas or can even count only the alpha particles having more than a certain energy.


If the voltage on a proportional counter is raised, there comes a point when the primary ions from a single alpha particle, beta particle, or photon will set off a discharge through the whole counter, not merely multiply the number of primary ions in the region where they are produced. This is a trigger action and the current is independent of the number of ions produced; furthermore, the current would continue indefinitely if no steps were taken to quench it. Quenching can be achieved entirely by arranging the external circuits so that the voltage drops as soon as current passes or by using a mixture of gases in the counter which "poison" the electrode surface as soon as the discharge passes and temporarily prevent the further emission of electrons, or by combining both methods.

The Geiger-Muller counter was developed before the proportional counter and remains the most sensitive instrument for detecting ionizing radiation, but all it does is "count" any ionizing radiation that passes through it whether it be an alpha particle, proton, electron, or photon.


It is one thing to describe the principles of various ionization chambers, counters, and the like; quite another to construct and operate them successfully.

First of all, the walls of the counter chamber must allow the particles to enter the counter. For gamma rays this is a minor problem, but for relatively low-speed electrons or positrons or for alpha particles the walls of the counter must be very thin or there must be thin windows.

Then there are great variations in the details of the counter itself, spacing and size of electrodes, nature of the gas filling the chamber, its pressure, and so on.

Finally, the interpretation of the resultant data is a tricky business. The absorption of the counter walls and of any external absorbers must be taken into account; the geometry of the counter with relation to the source must be estimated to translate observed counts into actual number of nuclear events; last but not always least, statistical fluctuations must be considered since all nuclear reactions are governed by probability laws.


There is one method of observing nuclear particles that depends directly on ionization but is not an electrical method. It uses the fact that supersaturated vapor will condense more readily on ions than on neutral molecules. If air saturated with water vapor is cooled by expansion just after an alpha particle has passed through it, tiny drops of water condense on the ions formed by the alpha particle and will reflect a bright light strongly enough to be seen or photographed so that the actual path of the alpha particle is recorded.

This method developed by C. T. R. Wilson in Cambridge, England, about 1912 has been enormously useful in studying the behavior of individual particles, alphas, protons, electrons, positrons, mesons, photons, and the fast atoms caused by collisions with alphas protons, or neutrons. Unlike the scintillation method, its companion tool for many years, it has not been superseded and is still used extensively, particularly to study details of collisions between nuclear particles and atoms.


The tracks of individual particles passing through matter can also be observed in photographic emulsions, but the lengths of path are so small that they must be observed under a microscope, where they appear as a series of developed grains marking the passage of the particle. This method of observation requires practically no equipment but is tedious and of limited usefulness. It is possible, however, to use the general blackening of a photographic film as a measure of total exposure to radiation, a procedure that has been used to supplement or to replace electroscopes for safety control in many parts of the project.


None of the methods we have described is directly applicable to neutrons, but all of them are indirectly applicable since neutrons produce ions indirectly. This happens in two ways - by elastic collision and by nuclear reaction. As we have already described, a fast neutron in passing through matter occasionally approaches an atomic nucleus so closely as to impart to it a large amount of momentum and energy according to the laws of elastic collision. The nucleus thereby becomes a high-speed charged particle which will produce ionization in an ionization chamber, counter, or cloud chamber. But if the neutron has low speed, e.g., thermal, the struck nucleus will not get enough energy to cause ionization. If, on the other hand, the neutron is absorbed and the resultant nucleus breaks up with the release of energy, ionization will be produced. Thus, for the detection of high-speed neutrons one has a choice between elastic collisions and nuclear reaction, but for thermal speeds only nuclear reaction will serve.

The reaction most commonly used is the B10(n, α )Li7 reaction which releases about 2.5 Mev energy shared between the resultant alpha particle and Li7 nucleus. This is ample to produce ionization. This reaction is used by filling an ionization chamber or proportional counter with boron trifluoride gas so that the reaction occurs in the region where ionization is wanted; as an alternative the interior of the chamber or counter is lined with boron. The ionization chamber then serves as an instrument to measure overall neutron flux while the proportional counter records numbers of individual neutrons.

One of the most valuable methods of measuring neutron densities by nuclear reactions depends on the production of artificial radioactive nuclei. A foil known to be made radioactive by neutron bombardment is inserted at a point where the neutron intensity is wanted. After a given time it is removed and its activity measured by an electroscope or counter. The degree of activity that has been built up is then a measure of the number of neutrons that have been absorbed. This method has the obvious disadvantage that it does not give an instantaneous response as do the ionization chamber and counter.

One of the most interesting methods developed on the project is to use the fission of uranium as the nuclear reaction for neutron detection. Furthermore, by separating the isotopes, fast and slow neutrons can be differentiated.

Since the probability of a neutron reaction occurring is different for every reaction and for every neutron speed, difficulties of translating counts or current measurements into numbers and speeds of neutrons present are even greater than for other nuclear particles. No one need be surprised if two able investigators give different numbers for supposedly the same nuclear constant. It is only by an intricate series of interlocking experiments carefully compared and interpreted that the fundamental facts can be untangled from experimental and instrumental variables.