INVESTIGATION OF THE NEUTRON BACKGROUND OF A BETATRON AT 25 MeV USING NUCLEAR PHOTOEMULSIONS

A. K. Berzin, B. M. Yakovlev, A. A. Yatis

Introduction

When a betatron with a maximum radiation energy above 10 MeV is operated, as a result of the interaction of bremsstrahlung $\gamma$-radiation with the structural parts of the accelerator, neutrons are produced whose energy and angular distribution are characteristic of each accelerator installation individually. The intensity of the fluxes and the energy of the neutrons vary within considerable limits and depend mainly on the yield and energy of the betatron bremsstrahlung $\gamma$-radiation.

The main role as neutron sources in an accelerator installation is played by the target and filter, as well as by the accelerating chamber, the diaphragm system, and the protective screens.

In carrying out physical experiments on the study of photonuclear reactions—for example, in investigating the angular and energy distributions of photoneutrons from various elements—the contribution of neutrons from the betatron to the experimental results may substantially alter the true picture and introduce certain errors into the measurements. When betatrons are used in medicine for therapeutic purposes, and also in biological investigations, neutron fluxes in the irradiation field of the object create a certain additional dose which, apparently, must be taken into account, especially for long irradiation times, bearing in mind the high biological effectiveness of neutrons.

On the questions indicated above, there are relatively few data in the literature on measurements of the yield and the energy and angular distribution of neutrons from electron accelerators of the betatron or synchrotron type. For quantitative measurements and angular distributions of neutron fluxes, the method of induced activity was used, by activating foils of various materials: rhodium, indium, gold, or by activating samples of aluminum, manganese, or phosphorus. BF$_3$ counters have found wide application in measurements of neutron fluxes [1, 2]. To determine the neutron yield from a betatron, Ilford C—2 and Agfa photoemulsions were also used [3]. The author made an approximate determination of the neutron flux in the beam of bremsstrahlung $\gamma$-radiation of a betatron at 15 MeV, equal to $3.8 \cdot 10^5$ neutrons/cm$^2 \cdot$ sec, with the mean energy of fast neutrons taken as ...

equal to 1.33 MeV; it is not indicated at what distance from the accelerator target the measurements were carried out. An approximate estimate was made of the magnitude of the correction coefficient (somewhat greater than 100) for determining the absolute flux of fast neutrons. The energy distribution of fast neutrons is also given without any corrections for background.

Measurements of the fluxes of thermal and fast neutrons from several 15- and 25-MeV betatrons, carried out in 1957–59 [4, 5, 6], showed that the measurement technique and experimental apparatus we used—BF₃ counters, an “all-wave” neutron counter, and the induced-activity method—do not make it possible to carry out a complete estimate of the neutron yield from the betatron, their energy and spatial distribution. In connection with this, in 1958–60 we carried out experiments using domestic nuclear photoemulsions in order to estimate the magnitude of the neutron background, as well as to determine the spectrum of fast neutrons in the field of the betatron bremsstrahlung γ-radiation, the spectrum and distribution of neutrons in a phantom, and to determine the additional doses due to neutrons in the irradiated object.

The present work gives the results of measurements of the neutron background in a betatron laboratory by means of nuclear photoemulsions.

Experimental arrangement

For measuring neutron fluxes, nuclear photoemulsions of the NIKFI T-3 and Ya-2 types were used. The essence of the nuclear-photoemulsion method is that fast neutrons, passing through the emulsion, as a result of $(n, p)$ scattering on hydrogen, knock out recoil protons, which leave in the emulsion clearly distinguishable and precisely measurable tracks, from the magnitude of which one can judge the energy of the particles.

As the optimum thickness of the emulsion layer for our experiments, thicknesses of 100 and 200 microns proved most suitable, since at such thicknesses a sufficient number of tracks was found in the emulsion and, moreover, the methods of developing such plates, although rather lengthy in time, are not too complicated.

When plates were irradiated in the bremsstrahlung γ-radiation beam of a 25-MeV betatron or in fields of scattered γ-radiation, it was first of all necessary to determine the optimum γ-radiation dose for these plates. In doing so it was necessary to take into account, on the one hand, the circumstance that the magnitude of the γ-radiation dose should be large, so that it would be possible to form a sufficient number of recoil protons in the emulsion as a result of nuclear reactions under the action of neutrons; but at the same time the dose, on the other hand, should be small, so that there would not be too strong a blackening of the plates, which would hinder viewing of the particle tracks, especially those which pass inside the emulsion at large angles.

Preliminary irradiations were carried out of a series of T-3 plates, with emulsion thickness 200 microns, and then of several series of Ya-2 plates, with emulsion thickness 100 microns, in a bremsstrahlung γ-radiation beam with energy 24 MeV; the plates were placed at a distance of 1 m from the target at an angle of 10° to the beam axis. The optimum do-

for $\gamma$-radiation proved to be $4 \div 6$ r. At an irradiation dose of $8 \div 10$ r, the blackening of the plates was so strong that it made identification of particle tracks difficult, and a large fraction of the tracks whose length was small fell out of the field of view. At a $\gamma$-radiation dose of 15 r, the plates were completely black, and it was impossible to discern any tracks on them.

Before carrying out the experiment, it was necessary to clarify how the packing material and its thickness affect the number of tracks in the emulsion. The packing material (the black paper in which the plates are wrapped in order to prevent their premature exposure) contains the elements hydrogen $\mathrm{H}^1$, carbon $\mathrm{C}^{12}$, nitrogen $\mathrm{N}^{14}$, and oxygen $\mathrm{O}^{16}$; and under the action of bremsstrahlung $\gamma$-radiation with energy up to 24 MeV, photonuclear reactions of the $(\gamma, n)$ and $(\gamma, p)$ types can occur on these elements. As a result, in addition to recoil protons formed by fast neutrons caused by radiation from the accelerator installation itself, the emulsion will also register protons owing their origin to $(\gamma, p)$ reactions on various elements, and recoil protons produced as a result of $(\gamma, n)$ reactions. The photographic plates were wrapped in several layers of paper and were then irradiated in the beam of bremsstrahlung $\gamma$-radiation from the betatron, the $\gamma$-radiation dose being kept constant in each case. Examination of these plates showed that in plates with one layer of packing paper about 140 tracks/cm$^2 \cdot$ sec were recorded, whereas in plates with 5 layers of paper the number of tracks was 2.5 times greater. Therefore, in subsequent experiments, in order to avoid possible errors due to protons caused by the presence of the packing material, irradiation of the plates was carried out without it, in a darkened room.

For examining the plates, an MBI-3 microscope with oil-immersion magnification was used.

When the plates were examined under the microscope at a magnification of $900 \times$, at different depths in the emulsion there were observed all kinds of tracks, forks, and stars. Recoil protons formed as a result of the interaction of fast neutrons with the nuclei of various elements constituting the emulsion have characteristic tracks with a barely noticeable beginning and a somewhat bent, thickened end. Protons produced as a result of $(n, p)$ scattering in the glass base of the emulsion, whose tracks enter the emulsion from the side of the base, were not taken into account. Tracks leaving the emulsion into the base and emerging from the emulsion were counted in the examination of the plates, and their range was estimated approximately from the mean probable range of the main number of recoil-proton tracks having both beginning and end in the emulsion. Their number amounted to about $5 \div 10\%$ of the total number of tracks. Protons scattered by protons were also recorded in the emulsion, and the proton energy in this case was determined from the law of conservation of energy and momentum in the center-of-inertia system. Protons formed as a result of $(\gamma, p)$ or $(n, p)$ reactions on the nuclei of elements contained in the air, as well as in the collimator and protective shields, were registered by the emulsion; however, it should be noted that the cross section of these reactions is small and the reaction yield is insignificant. The relationship between the proton energy and its range in air gives grounds for assuming that the distance from the nearest points of the surface of the collimator and shielding to the plates, $20 \div 25$ cm, will constitute an obstacle to the registration in the emulsion of protons with energies up to 4.0 MeV. Nevertheless,

a significant number of proton tracks was recorded, beginning at the surface of the emulsion not only in these plates, but also in plates placed at a distance of 70–80 cm from the surface of the collimator and the protective shields. Consequently, recoil protons arise not only inside the emulsion, but also on its surface. Moreover, upon careful study of the tracks of protons entering the emulsion from outside and of the tracks of recoil protons formed on the surface of the emulsion, differences can be observed in the place where the track is formed in the emulsion: the track of a recoil proton has a poorly defined, diffuse beginning, whereas the beginning of a proton track is very sharp.

The total number of recoil-proton tracks is also contributed to by protons and neutrons formed as a result of reactions of the type: \((\gamma,n)\), \((\gamma,p)\), \((\gamma,np)\), and also by the reaction \(N^{14}(n,p)C^{14}\) on the nuclei of the isotopes of oxygen, carbon, and nitrogen that make up the emulsion. The tracks of these events can be identified unambiguously from the reaction products and the geometry of the disintegration. A reaction of the \((\gamma,n)\) type is characterized by the presence of only one track—the track of the recoil nucleus—which distinguishes it from other reactions. Reactions of the \((\gamma,p)\) and \((\gamma,np)\) types are characterized by the presence of two tracks: that of the proton and that of the recoil nucleus. In order to identify these two reactions, it must be borne in mind that in a \((\gamma,p)\) reaction in the emulsion the angle between the tracks of the proton and the recoil nucleus is close to \(180^\circ\), and the total momentum lies in the same plane with them and coincides with the direction of the \(\gamma\)-ray beam. In a \((\gamma,np)\) reaction, the emitted particles may have momenta differing in magnitude and direction and any angles with respect to the direction of the \(\gamma\)-radiation beam. The tracks formed as a result of the reaction \(N^{14}(n,p)C^{14}\) on slow neutrons can be identified from the total range of the proton and the recoil nucleus (about 6 microns). As an example of the tracks of particles in the emulsion of the plates, Fig. 1 shows a photograph of a recoil-proton track.

To calculate the range of recoil protons with the aid of the ocular scale of the microscope, the horizontal projection of the track and the vertical component (“\(n\)-immersion”) were measured; a correction was introduced for the shrinkage coefficient of the emulsion, the value of which for different plates varied within the limits 1.83–2.2. The accuracy of the measurements in determining the vertical projection of the track was limited by the division value of the micrometer screw (up to 1 micron).

Fig. 1

To calculate the energies of recoil protons from their ranges in the emulsion, we used the range–energy nomograms given in [7]. These nomograms were compiled for Ilford C-2 plates, but they can also be used for domestic plates, since the difference in the composition of these emulsions is small and the errors in determining the energy of recoil protons will be insignificant.

Measurement Results

The arrangement scheme of type Ya-2 photographic plates, 100 microns thick, in the premises of the betatron laboratory is shown in Fig. 2. The main ...

particular attention was paid to measurements of the neutron flux in the bremsstrahlung \(\gamma\)-ray beam, and also in the control-room area near the periscope (11) and the control console (12), where the experimenters and operating personnel are located during operation of the accelerator.

As was already mentioned above, the optimum dose of \(\gamma\)-radiation for plates placed in the beam at a distance of \(1\ \text{m}\) from the target was \(5\ \text{r}\). Constancy of the \(\gamma\)-radiation intensity was monitored by means of two ionization chambers, one of which

Fig. 2

was located in the bremsstrahlung \(\gamma\)-ray beam at a distance of \(1\ \text{m}\) from the target, and the other in the scattered radiation of the accelerator. In studying neutron fluxes in the control room, the dose of \(\gamma\)-radiation in the beam at a distance of \(1\ \text{m}\) from the target was \(100\ \text{r}\). Irradiation of the plates in the experimental hall was carried out without packaging in a darkened room; the plates located in the control room were placed in single-layer envelopes of thick black paper. The results of the measurements are given in the table.

As is evident from the measurement results, the largest value of the fast-neutron flux is \(6.38 \cdot 10^{4}\ \text{neutrons}/\text{cm}^{2}\) per \(1\ \text{r}\) of bremsstrahlung \(\gamma\)-radiation. From the data given in the table for the fluxes of fast neutrons measured at points 2, 3, 4, and 5, located symmetrically with respect to the axis of the bremsstrahlung \(\gamma\)-ray beam, it is seen that at point 2, located in the horizontal plane at a distance of \(20\ \text{cm}\) to the right of the beam, the flux of fast neutrons is greater than at the other points. This can apparently be explained by an additional contribution from neutrons reflected from the wall of the room. A decrease in the flux of fast neutrons in the bremsstrahlung \(\gamma\)-ray beam is observed as the distance from the accelerator target increases. The increase in the neutron-flux value at the point removed from the target by a distance of \(4.2\ \text{m}\), in the “absolutely black body,” is explained by backscattering of neutrons from the walls of the “black body.” These measurements

Table

Note. At points 11 and 12 the measurements of fast neutrons are referred to a \(\gamma\)-radiation dose of 100 roentgens in the beam at a distance of 1 m from the target.

once again confirm the results of neutron-flux measurements carried out earlier by other methods [8].

The data presented in this work on measurements of fast-neutron fluxes by means of photographic plates are the first results of investigations of neutron fluxes, their energy and spatial distribution, using nuclear photoemulsions, carried out on betatrons with a bremsstrahlung energy of 25 MeV.

REFERENCES

1. J. S. Laughlin. — Nucleonics, 8, No. 4, 5, 1951.
2. H. Waffler. Das Brown Boveri Betatron, p. 17, 1958.
3. D. Hoffmann. Strahlentherapie, 97, 239, 1955.
4. V. L. Korolev, G. A. Cherevatenko, B. M. Yakovlev. Scientific report No. 139, TPI, 1958; Scientific report No. 65, NII TPI, 1958.
5. A. L. Gryaznov, R. I. Meshcheryakov, B. M. Yakovlev. Scientific report No. 085771, NII TPI, 1959.
6. B. M. Yakovlev, R. I. Meshcheryakov, A. L. Gryaznov. Report at the III All-Union Interuniversity Conference on Electron Accelerators, Tomsk, 1959.
7. P. Demers. Ionographie, p. 252, 1958.
8. R. I. Meshcheryakov, B. M. Yakovlev, A. L. Gryaznov, G. A. Nyba. Report at the III All-Union Interuniversity Conference on Electron Accelerators, Tomsk, 1959.