STUDY OF SOME PROPERTIES OF SCINTILLATION COUNTERS

A scintillation counter consists of a combination of a phosphor, in which fast particles produce scintillation flashes, and a photomultiplier, by means of which these flashes are recorded.

The use of such a combination was proposed independently and almost simultaneously by Kalman and by Coltman and Marshall[^2]. Coltman and Marshall proposed using inorganic phosphors of the zinc chloride type to increase the efficiency of the multiplier; Kalman established that organic phosphors, such as naphthalene, transparent to their own radiation, are better suited for this purpose. A review of the first works on the study of the properties of scintillation counters is contained in[^3] (section 5).

Detailed studies of the properties of various phosphors with respect to their efficiency and light output were carried out by Munn[^4] (inorganic phosphors) and by Kalman[^5] (organic and inorganic).

At present, scintillation counters have found wide application in various experimental works in nuclear physics, which we shall not discuss here.

It was established that solutions of various substances (“liquid counters”)[^6] can also be used as particle detectors, which offers certain advantages, since the need to prepare crystals is eliminated and, in addition, solutions prove to be more transparent to radiation than large crystals.

Already in the first works on scintillation counters it was also noted that the decay time of organic phosphors is significantly less than the decay time of inorganic phosphors, which opens attractive prospects for measuring short time intervals (in measuring the lifetimes of short-lived elements and in other experiments).

The decay time of some phosphors was measured in[^7]; the measurements were made with a micro-oscillograph[^8] and, for the three phosphors tested, gave the following decay-time values at the temperature of liquid nitrogen:

\[ \begin{aligned} \text{naphthalene} &\ldots\ldots\ldots (5.7 \pm 0.5)\cdot 10^{-8}\ \text{s},\\ \text{anthracene} &\ldots\ldots\ldots (1.3 \pm 0.2)\cdot 10^{-8}\ \text{s},\\ \text{phenanthrene} &\ldots\ldots\ldots (0.9 \pm 0.2)\cdot 10^{-8}\ \text{s}, \end{aligned} \]

whereas the decay time for inorganic phosphors is of the order of \(10^{-7}\)–\(10^{-6}\) s.

Recently a number of notes[^9][^12–^14][^20–^23] have been published in which results are given on the study of the decay time of various organic phosphors. These works are of interest not only because of the results obtained in them, but also because of the new technique used in the investigations.

In[^9] the dependence of the decay time of anthracene on temperature was measured; in addition, the dependence of the magnitude and height of the pulses on temperature was investigated.

The pulses were fed to an oscilloscope with a distributed-gain amplifier[^10] (see also[^11]), with a rise-time constant of \(6\cdot 10^{-9}\) s, and were photographed. The decay time was determined graphically and averaged over 20 pulses at each temperature. At two points, to control the processing method, all 20 pulses were plotted graphically and the decay time was determined from the resultant graph;

the discrepancy between the results obtained by these two methods in both cases did not exceed 2%. The dependence of the decay time of anthracene on the temperature of the specimen is given in Fig. 1.

In the same work, qualitative results were obtained concerning the spectral composition of the radiation from anthracene, which show that the scintillation spectrum includes regions with \(\lambda < 3400\) Å (see also \(^{24}\)).

Fig. 1. Decay of scintillation flashes in anthracene. The small vertical strokes on the graph indicate the interval into which 95% of the measured values fall.

The dependence of the light output on the temperature of the specimen for anthracene, naphthalene, and stilbene was also studied in \(^{19, 19a}\); for anthracene and naphthalene the dependence on temperature is small (the light output decreases by two to three times when the temperature changes from 120 to 310°K), while for stilbene no change of light output with temperature was noticed at all.

The results of work \(^{9}\) regarding the decay time agree with the data published in \(^{20}\), where the following values were obtained for the decay time of anthracene in the temperature interval from 4 to 290°K:

If the straight line in Fig. 1 is extrapolated to 0°K, we obtain \(\tau_{\varphi K} \approx 0\). An attempt to verify this circumstance over a wide temperature interval for anthracene and stilbene was undertaken in \(^{21}\).

The measurements in this work were carried out with the aid of the following apparatus: the phosphor was mounted at the end of a quartz rod about 20 cm long and placed in a cooling bath. At the other end of the rod there was a 931-A photomultiplier. Measurements of the decay time were made in the temperature interval from 4 to 298°K for anthracene and from 78 to 298°K for stilbene, using apparatus with a delayed line (details concerning the construction of the measuring circuit were not published); irradiation was produced by a radium source in 10 \(\mu\)C.

The results obtained in this work differ from the data obtained in \(^{9}\) and \(^{20}\). Thus, for anthracene, as the temperature was lowered, the decay time tended toward a limiting value (at 4°K \(\tau \approx 16 \cdot 10^{-9}\) sec.); for stilbene it remained almost constant, changing from \(9 \cdot 10^{-9}\) sec. at 78°K to \(10 \cdot 10^{-9}\) sec. at 195°K and \(12 \cdot 10^{-9}\) sec. at 298°K. It is possible that the observed tendency toward a limit is caused by the imperfection of the measuring apparatus, all the more so because the value of the decay time obtained in this work for stilbene is substantially greater than the values obtained in subsequent works.

In \(^{12–13}\), the following method was used to investigate the decay time: voltage pulses up to

5 kilovolts, with a duration of up to 2.5 microseconds; in a number of multiplier specimens no breakdowns were observed. The pulses at the output of the multiplier from a stilbene irradiated with γ-rays from Ag¹¹⁰ reached 80 volts. Through a coaxial cable loaded with a wave resistance of 100 ohms, the pulses were fed directly to the deflection plates of a high-voltage cathode-ray tube of type 5RP11A. To damp free oscillations at the cable output, damping resistances of 100 ohms were connected between the cable and the tube inputs. The shape of the curves was analyzed both graphically (for a single pulse) and by cutting out pulses with a short-circuited line. The authors believe that the pulse rise time observed on the tube screen may have been appreciably increased owing to the inductance of the plate leads inside the tube.

Fig. 2.

The width of the pulses from single electrons emitted from the photocathode (background pulses) was approximately \(5\cdot 10^{-10}\) sec, in agreement with Sard’s calculations¹⁵.

Using the described method, very pure crystals of trans-stilbene (1,2-diphenylethylene) were investigated and, for comparison with previous works⁷ and ⁹, an anthracene crystal of unknown purity. The results obtained are given in the following table (in \(10^{-9}\) sec.):

The results for anthracene at the temperature of liquid nitrogen are in agreement with the results obtained by Birks⁹. However, at room temperature in this work the decay time for anthracene proved to be substantially smaller than in⁹ (Fig. 2).

The magnitude of the decay time was determined by comparing smoothed photographs of the pulses with plotted exponentials, moreover

the authors note that at room temperature the decay of the pulses from anthracene turns out to be very nonuniform, with noticeable jumps, so that the exponent has to be fitted very conventionally; it is possible that this is the reason for the discrepancies with \(^{9}\); at the same time the pulses from stilbene are sufficiently smooth at all temperatures (it is possible, of course, that they are smoothed by the circuit itself to a greater extent than the pulses from anthracene, because of the greater steepness of the decay).

In \(^{28}\) (by the same method as in \(^{12-13}\)) the decay time (and efficiency) of a saturated solution of terphenyl (\(p\)-diphenylbenzene) in toluene was investigated; for \(\tau\) the value \(\tau \simeq 2.2 \cdot 10^{-9}\) sec was obtained; in processing the data the pulse rise time was not taken into account; the observed rise time of the pulses from 10 to 90% in the various cases lay in the range from 0.6 to \(1.0 \cdot 10^{-9}\) sec. Similar results were also obtained for a solution of terphenyl in xylene (a mixture of the ortho-, para-, and meta-modifications).

In \(^{35}\) measurements of the steepness of the leading edge of multiplier pulses and the burn-out time of phosphors were made as follows: photomultipliers 1P21 operated at a voltage of 1500–1800 volts; in this case the pulses from the radium source at the output of the cathode follower (lamp 6AG7), connected directly to the multiplier, were 25 volts. To the output of the follower there was connected a shorted coaxial cable with a characteristic impedance of 100 ohms, the length of which could be varied. The pulses then went to a discriminator on germanium diodes and to a counting circuit; the length of the cable and the bias on the discriminator were adjusted until a constant counting rate was obtained for the given source.

For the pulse widths the following values were obtained (in \(10^{-9}\) sec):

(Stilbene and anthracene—in the form of colorless single crystals, the remaining substances—in the form of microcrystals); the temperature of the samples at which the measurements were made is not indicated. The excessive width of the background pulses (one order of magnitude greater than in \(^{13}\), and several times greater than in \(^{14}\), where the same 1P21 multipliers, operating in normal mode, were used—see below) attracts attention.

In work \(^{14}\) the burn-out time of several new organic phosphors was measured by means of the following device: the crystals under study were placed between two 1P21 multipliers and were irradiated with \(\gamma\)-rays from a radioactive preparation. The pulses were taken from the 9th dynode of the multipliers and, through coaxial cables with a characteristic impedance of 100 ohms, used to introduce delay, were fed to amplifiers with distributed amplification (see \(^{10}\) and \(^{3}\)) with a rise time of \(2.5 \cdot 10^{-9}\) sec. After amplification the pulses were fed to a coincidence circuit with germanium diodes. The burn-out curves could be obtained by changing the length of the coaxial cable in one of the channels. By changing the magnitude of the bias on the coincidence discriminator, it was possible to select cases corresponding to the incidence on the photocathode of the multiplier (in the channel without delay) simultaneously of one, two, three, etc. photons.

For background investigations, the limiting width of the pulses at the input of the coincidence circuit (meaning the width determined by the multiplier and ampli-

liters), the number of registered coincidences is determined by the expression

\[ N=\sum A_n \exp(-n t_3/\tau), \]

where \(t_3\) is the delay time of the pulses in one of the channels relative to the other, \(\tau\) is the decay time of the phosphor, \(n\) is the number of photons simultaneously incident on the photocathode of the multiplier in the channel without delay, and \(A_n\) is a constant independent of \(t_3\).

In the experiments described, only those coincidences were selected which corresponded to the simultaneous incidence (in the channel without delay) of no fewer than three photons. In this way it was possible to separate true coincidences more reliably from false ones with large single pulses; in addition, this method gives better resolution in the sense of determining the decay time. By changing the bias at the discriminator, coincidences caused by the simultaneous incidence of two and more, and of four and more, photons in the channel without delay were also measured. Figure 3 presents curves corresponding to the term \(\exp(-3t_3/\tau)\).

Crystals and multipliers operated without cooling at room temperature (\(\sim 24^\circ\) C).

In order to estimate the width of the pulses from a single electron emitted from the photocathode, the self-coincidences from the thermal-noise pulses of one multiplier were also measured (curve \(F\) in Fig. 3); in another experiment, for the same purpose, a strong source of \(\gamma\)-rays from \(Co^{60}\) was placed between the multipliers.

Fig. 3. Decay curves \(\exp(-3t_3;\tau)\) for the following substances:
\(A\)—1,4-diphenylbutadiene; \(B\)—\(p\)-terphenyl; \(C\)—\(m\)-stilbene; \(D\)—anthracene; \(E\)—naphthalene; \(F\)—noise pulses. The ordinates of curve \(E\) are multiplied by \(10^3\); this curve also passes through the point with coordinates: 55 feet (\(\simeq 7\cdot 10^8\) sec.), \(0.004 \pm 0.001\) coincidences per second.

The mean decay time of the excited crystals is three times greater than the value corresponding to the slope on the graph in Fig. 3 (see the formula), and is given in the following table:

1,4-diphenylbutadiene . . . . \(4.2\cdot 10^{-9}\) sec.
\(p\)-terphenyl . . . . . . . . . . \(4.2\cdot 10^{-9}\) sec.
\(m\)-stilbene . . . . . . . . . . \(5.7\cdot 10^{-9}\) sec.
Anthracene . . . . . . . . . . . \(24\cdot 10^{-9}\) sec.
Naphthalene . . . . . . . . . . \(60\cdot 10^{-9}\) sec.

The decay time for stilbene and anthracene agrees with the values obtained in \(^{12}\).

The width of the pulses from single electrons, determined on the basis of curve \(F\) (Fig. 3), is due to the rise time of the pulse—

... in amplifiers; this explanation is also confirmed by experiments without an amplifier in front of the coincidence circuit; in this case the pulse width was less than \(10^{-9}\) sec., which is generally consistent with the calculated value^15 and with the experimental value obtained in^13.

The bending of the curves in Fig. 3 near the ordinate axis is caused by saturation in the multipliers and by the superposition of pulses from several photons.

Crystals with a decay time of \((3—4)\cdot 10^{-9}\) sec. may make it possible to measure small time intervals with an accuracy up to \((5—10)\cdot 10^{-10}\) sec. This permits their use not only for measuring the lifetime of short-lived elements, but also for the direct measurement of the velocity of fast particles: at \(\beta = 0.7\) and a path length of \(2\) m the flight time will be \(10^{-8}\) sec. and can be measured with an accuracy of \(5—10\%\). Direct measurements of particle velocity would make it possible to determine directly—from velocity and momentum—the mass of particles without having to resort to various theoretical formulas relating the particle range to its energy or the ionizing power of the particle to its velocity. In addition, such measurements would make it possible to subject these theoretical formulas to experimental verification for fast particles other than electrons.

A study of certain other properties of organic phosphors of the aromatic series can be found in the following works:^16 (various substances, including stilbene and phenanthrene),^17 [1,2-diphenylethane (dibenzyl), 1,2-diphenylethylene (stilbene), and diphenylacetate],^18 [anthracene, stilbene, \(p\)-diphenylbenzene (terphenyl), and diphenylethane (dibenzyl).]

V. Kh.

CITED LITERATURE

1. H. Kalman, Nature und Technick, July (1947).
2. J. W. Koltman and F. H. Marshall, Phys. Rev. 72, 528 (1947).
3. V. Kharitonov, UFN 39, 402 (1949).
4. R. T. Moon, Phys. Rev. 73, 210 (1948).
5. H. Kalman, Phys. Rev. 75, 623 (1949).
6. G. T. Reynolds, F. B. Harrison a. G. Salvini, Phys. Rev. 78, 488 (1950) (tests in irradiations by \(\gamma\)-rays of OT C\(_0^{60}\): naphthalene in benzene; naphthalene + anthracene in benzene; terphenyl in benzene and xylene; liquid dibenzyl with positive results).
   H. Kalman, Phys. Rev. 78, 621 (1950) (anthracene, phenanthrene, and other substances in various solvents). M. Ageno, M. Chiozzotto and R. Querzoli, Phys. Rev. 79, 720 (1950) (detailed data on a naphthalene solution in xylene and discussion).
7. G. B. Collins, Phys. Rev. 74, 1543 (1948).
8. G. M. Lee, Proc. I. R. E. 34, 121 (1946).
9. G. G. Kelley and M. Goodrich, Phys. Rev. 77, 138 (1950).
10. G. G. Kelley, Rev. Sc. Instr. 21, 71 (1950).
11. Ginztoun, Hewlett, Tasberg and Noe, Proc. I. R. E. 36, No. 8 (1948).
12. R. F. Post and N. S. Shiren, Phys. Rev. 78, 80 (1950).
13. R. F. Post and N. S. Shiren, Phys. Rev. 78, 81 (1950).
14. A. Lanby, Phys. Rev. 80, 477 (1950).
15. R. D. Sard, J. Appl. Phys. 17, 768 (1946).
16. H. T. Gittings, R. F. Taschek, A. R. Ronzio, E. Jones and W. J. Mazilum, Phys. Rev. 75, 205 (1949).
17. W. S. Koski and. C. O. Thomas, Phys. Rev., 78, 308 (1949).
18. R. Hofstadter, S. H. Liebson and J. O. Elliot, Phys. Rev. 78, 81 (1950).

1. S. H. Liebson and J. W. Keller, Phys. Rev. 78, 305 (1950).

19a. S. H. Liebson and R. T. Farrar, Phys. Rev. 79, 733 (1950) (relative light output and temperature dependence of the light output for naphthalene and anthracene, and for naphthalene with various anthracene contents).

1. S. H. Liebson and J. O. Elliot, Phys. Rev. 78, 65 (1950).

2. J. O. Elliot, S. H. Liebson and C. F. Ravilious, Phys. Rev. 79, 393 (1950).

3. O. Martinson, P. Isaak, H. Brown and Rudernau, Phys. Rev. 79, 178 (1950).

4. R. F. Post, Phys. Rev. 79, 735 (1950).

5. F. B. Harrison and G. T. Reynolds, Phys. Rev. 79, 732 (1950) (study of the spectral composition of the radiation of terphenyl and anthracene).