USE OF THE CYCLOTRON IN STUDIES ON THE RADIATION CHEMISTRY OF SOLIDS

A. N. OBLIVANTSEV, V. M. LYKHIN, L. P. EREMIN

(Presented by the scientific seminar of the sector of radiation-chemical analysis, Research Institute of Nuclear Physics)

Studies in the radiation chemistry of ionic crystals are receiving ever greater development. However, until recently these studies were carried out mainly with the use of lightly ionizing radiation, and the basic regularities of the radiation decomposition of ionic crystals were obtained using X-ray and gamma radiation. It is obvious that these regularities, without experimental verification, cannot be fully transferred to radiation decomposition under the action of heavy charged particles, for which specific features may appear that are associated with their low penetrating power and high ionization density.

The small number of experimental works known to us addresses these questions using external [1, 2] or internal isotopic sources of charged particles [3]. The use of particle accelerators for these purposes—for example, cyclotrons—appears more interesting. The possibility of accelerating in them various particles, such as protons, deuterons, helions, and other multiply charged ions, makes it possible to compare the action of radiation with different ionization densities.

Over the last 5 years, on the cyclotron of the Research Institute of Nuclear Physics at TPI, we have carried out work on studying the radiation-chemical stability of ionic crystals under the action of an accelerated-proton beam extracted into the atmosphere. The objects of study were azides, nitrates, perchlorates, and permanganates of alkali metals.

By varying the thickness of the foil for extracting the beam, the distance between the irradiated sample and the extraction window, or by placing an additional metal foil of the required thickness in the path of the beam to the sample, one can reduce the energy of the protons incident on the sample to any desired value.

Uniformity of the particle flux over the cross section of the beam, with dimensions of \(26 \times 100\ \text{mm}^2\), was achieved by installing in the ion guide a scattering aluminum foil \(5.25\ \text{mg}/\text{cm}^2\) thick.

The cyclotron also makes it possible to vary the flux density of charged particles within considerable limits. The high intensity of cyclotron radiation makes it possible, in a short time, to accumulate large absorbed doses and to study radiation-chemical decomposition up to large percentages of conversion.

1. Irradiation of Samples

Since protons have low penetrating power, the irradiation cell must ensure uniform dosing of the sample at a very small sample thickness.

For irradiation we used a cell consisting of an aluminum base measuring \(40 \times 90\ \mathrm{mm}^2\), onto which aluminum foil of thickness \(10\ \mathrm{mg/cm^2}\), with four holes of diameter \(14.5\ \mathrm{mm}\), was glued. Into the resulting recesses a weighed amount of the ground substance to be irradiated was placed in a thin layer. The recesses were covered with aluminum foil of thickness \(2\ \mathrm{mg/cm^2}\), which prevented the substance from spilling out in the vertical working position. Irradiation was carried out directly adjacent to the exit window of the cyclotron.

For irradiation of samples at elevated temperatures, the cell, together with a thermocouple located in a special pocket near the substance, was fastened to the flat surface of an electric furnace. The manufacture of the cell from aluminum provides good heat removal at room temperature and rapid heating of it in the case of irradiation at elevated temperatures; moreover, upon its activation a rapidly decaying \(\beta\)-activity is formed.

Fig. 1. Increase in sample temperature during irradiation with 4.7 MeV protons at current density: 1—\(0.35 \cdot 10^{11}\); 2—\(0.7 \cdot 10^{11}\); 3—\(1.4 \cdot 10^{11}\) protons/\(\mathrm{cm^2 \cdot sec}\).

By regulating the change in the density of the proton current, it is possible to study the influence of dose rate on radiation-chemical processes. In this case, however, the maximum attainable dose-rate values are limited not only by the technical characteristics of the accelerator (in our case, with good accuracy it was possible to obtain proton flux densities in the interval of average current values \(0.001\)—\(0.1\ \mu\mathrm{A/cm^2}\)), but also by heating of the sample at high current densities, caused by the fact that the rate of transfer of radiation energy, dissipated in the sample in the form of heat, substantially exceeds the rate of heat removal.

Heating of the sample during irradiation was checked experimentally for potassium permanganate. The dependence shown in Fig. 1 of the excess of the sample temperature over room temperature on irradiation time and current density (the current was increased after a constant temperature had been established at its lower value) shows that macroscopic heating of the sample can be neglected at the proton-current densities used. Similar conclusions were reached by the authors of work [4], who investigated the temperature distribution over the depth of alkali-halide crystals irradiated with protons on the same cyclotron. It has been shown [4, 5] that the change in thermometry of a thermocouple when operating in radiation fields is small and corresponds to an error of \(1^\circ\mathrm{C}\).

2. Determination of the Absorbed Dose

In studying the radiation stability of compounds, the correctness of the estimate of the dose absorbed by the sample is of great importance. If the proton flux and the energy losses of each particle in passing through the irradiated sample are known, then the measured current, integrated

over time, makes it possible to calculate the absolute amount of energy transferred by the beam to the sample.

Measurements of the ion current were carried out with an M-95 galvanometer. The current was taken from rectangular copper sensors, each with an area of \(1\ \mathrm{cm}^2\), which were installed in the ion guide in front of the exit copper foil at various points in the beam cross section and, without interfering with irradiation of the samples, make it possible to record both the total current from \(6\ \mathrm{cm}^2\) and the current from each sensor. The latter circumstance is used when adjusting the uniformity of the beam.

When the current is measured in this way, the emission of secondary electrons during bombardment of the sensors by protons, and the associated overestimation of the current value, are not taken into account. Measurements of the true proton current carried out by D. I. Vaisburd using a Faraday cylinder gave a correction coefficient for conversion from the value measured by the galvanometer to the true current equal to 1.5, which was taken into account in the calculations.

Since the thickness of the layer of substance in our experiments was less than the range in the samples, the absorbed dose was calculated from the formula:

\[ D=\frac{I}{q}\cdot\frac{s_0}{m}\cdot\Delta E\cdot\tau\ \mathrm{eV}/\mathrm{g}, \]

where
\(I\) is the true current density in CGSE units;
\(q\) is the proton charge in CGSE units;
\(\Delta E\) is the energy loss of the proton in the sample, eV;
\(\tau\) is the irradiation time, sec;

\[ \frac{m}{s_0} \]

is the thickness of the sample, \(\mathrm{g}/\mathrm{cm}^2\).

The energy of the protons in the ion guide before the scattering foil is \(6.8\ \mathrm{MeV}\). To determine the energy of the protons incident on the sample, it is necessary to determine the energy losses in the scattering aluminum foil (\(5.25\ \mathrm{mg}/\mathrm{cm}^2\)), the copper foil of the exit window (\(42.8\ \mathrm{mg}/\mathrm{cm}^2\)), and the window of the aluminum cell (\(2.0\ \mathrm{mg}/\mathrm{cm}^2\)). Taking these losses into account, as determined from the known dependences of proton range on their energy in copper and aluminum [6], the energy of the protons incident on the samples is \(4.7\ \mathrm{MeV}\).

3. Calculation of energy loss in the samples

The energy losses \((\Delta E)\) can be found from the dependence of range on proton energy in the irradiated samples. Literature data for determining the energy losses of protons as they pass through complex substances are lacking; therefore, we calculated the stopping powers and obtained the dependences of range on proton energy for azides, nitrates, perchlorates, and permanganates of alkali metals.

The stopping powers

\[ \left(-\frac{dE}{dx}\right) \]

were determined from the Bethe–Bloch formula for heavy charged particles [6], using the effective excitation potential and the effective atomic number determined according to Bragg’s additivity rule. Figure 2 gives the theoretical curves of stopping power (linear energy transfer) in \(\mathrm{eV}/\text{\AA}\) for alkali-metal nitrates in the proton energy interval \(0.5\)—\(5\ \mathrm{MeV}\).

The proton range in the irradiated substances was determined by approximate numerical integration of this stopping power. Figure 3 gives theoretical curves of range—energy loss for protons with an initial energy of \(5\ \mathrm{MeV}\) in nitrates. Similar curves were calculated for azides, perchlorates, and permanganates of alkali metals.

Fig. 2. Linear energy transfer for protons in alkali-metal nitrates: 1—KNO\(_3\); 2—NaNO\(_3\); 3—RbNO\(_3\); 4—CsNO\(_3\).

Fig. 3. Theoretical range–energy loss curves for protons with initial energy 5 MeV in alkali-metal nitrates: 1—NaNO\(_3\); 2—KNO\(_3\); 3—RbNO\(_3\); 4—CsNO\(_3\).

metals. Taking the sample thickness into account, the average energy losses of protons in the samples were determined from these curves (the data are summarized in Table 1), and the absorbed dose was calculated.

Table 1

Energy losses of protons during irradiation of samples

At a particle flux density of $0.02\ \mu\text{A}/\text{cm}^2$, the calculated average absorbed-dose rate was on the order of $10^{19}$ eV/g sec.

The method we have described for calculating the absorbed dose under proton irradiation proved to be sufficiently accurate. It was checked on potassium nitrate. For this purpose a sample of thickness $31.9\ \text{mg}/\text{cm}^2$, corresponding, according to our calculations, to proton energy losses of 2.8 MeV, was irradiated, as was a sample of thickness $95.7\ \text{mg}/\text{cm}^2$, greater than the proton range in potassium nitrate. In the second case all the incident energy is absorbed, and the energy-loss calculations given above can be avoided. By analyzing the entire irradiated sample, the amount of nitrite formed in it was determined and the radiation-chemical yield was calculated. The discrepancy between the yields determined with complete absorption of the proton energy and with partial absorption, determined by calculation, was 2%, which lies within the error of the analysis.

4. Studies carried out on the cyclotron

The irradiation and dosimetry methods we applied on the cyclotron made it possible to carry out a number of interesting studies in the radiation chemistry of solids.

For azides, nitrates, and perchlorates of alkali metals it was possible to show that the dependence of radiation stability on the free volume of the unit cell of the crystal, observed under electromagnetic irradiation, is preserved under proton irradiation, but becomes less sharply expressed.

Over a tenfold interval, no substantial effect of dose rate on the radiolysis of alkali-metal nitrates and perchlorates was detected.

The influence of homophase impurities on the radiation-chemical stability and post-radiation thermal annealing was studied for alkali-metal nitrates irradiated with protons. In particular, for potassium nitrate it was established that impurities of Sr$^{++}$, Pb$^{++}$, and Ag$^+$ ions lower its radiation-chemical stability; that the presence of Sr$^{++}$ and Pb$^{++}$ significantly accelerates the process of post-radiation thermal annealing and the fraction of this annealing. An effect of irradiation temperature on the radiation-chemical yield of nitrates was found.

Using the example of alkali-metal permanganates, the change in the thermal stability of solid salts after proton irradiation has been studied.

For the results obtained, the applicability of the hypothesis concerning the role of thermal effects in the track in the mechanism of radiolysis has been considered.

These results are presented in greater detail in papers [7—10].

REFERENCES

1. C. J. Hochanadel. Rad. Res., 16, 286 (1962).

2. C. J. Hochanadel. J. Phys. Chem., 67, 2229 (1963).

3. G. E. Boyd, T. G. Ward. J. Phys. Chem., 68, 3809 (1964).

4. I. G. Berezina, L. S. Sokolov. Izvestiya vysshikh uchebnykh zavedenii. Fizika, 3, 123—126 (1963).

5. Hubert P. Jockey. Phys. Rev., 101, 4, 1426 (1956).

6. Experimental Nuclear Physics, ed. by E. Segrè, vol. 1, IL, Moscow (1955).

7. A. N. Oblivantsev. Candidate’s dissertation, Tomsk, 1966.

8. A. N. Oblivantsev, V. V. Boldyrev, L. P. Eremin, V. M. Lykhin. Kinetika i kataliz, 7, 1015 (1966).

9. L. P. Eremin, V. M. Lykhin, A. N. Oblivantsev, V. V. Boldyrev. Zhurnal fizicheskoi khimii, 41, 1180 (1967).

10. A. N. Oblivantsev, V. M. Lykhin, V. V. Boldyrev, L. P. Eremin, V. F. Komarov. Present collection, p. 122.