APPARATUS FOR STUDYING THE POLARIZATION OF RECOIL PROTONS IN MESON PHOTOPRODUCTION REACTIONS

P. S. Ananyin, I. V. Glavanakov, V. I. Kryshkin, V. N. Stibunov,
A. G. Sterligov, Yu. P. Usov

(Presented by the Scientific and Technical Seminar of the High-Energy Laboratory of the Nuclear Physics Research Institute)

In experiments on the study of meson photoproduction, apparatus is used that consists of a proton telescope and a Cherenkov spectrometer. The layout of the experimental setup is shown in Fig. 1.

Fig. 1. Arrangement of the apparatus in the hall. SP-58 is the cleaning magnet.

1. Proton Telescope

The proton telescope consists of three scintillation counters \((C_1, C_2, C_3)\) and four spark chambers. The solid angle is determined by counter \(C_3\) and is equal to 4.05 millisteradians in the laboratory system. The dimensions, positions, and characteristics of the counters of the proton telescope are given in Table 1.

In counters \(C_1\) and \(C_2\), the Plexiglas light guides have the form shown in Fig. 2. The light guide of counter \(C_3\) is shown in Fig. 3 (the design of the light guide for the round scintillation counter \(C_4\) of the photon telescope is analogous to the design shown in Fig. 2, but the upper ...

Fig. 2. Design of the light guides of counters \(C_1\) and \(C_2\):
\(l_1=d,\ l_2=5\ \mathrm{cm}\).

the surface for coupling to the scintillator was made cylindrical). The scintillators were glued to the light guides with OK-72 optical cement, which provides optical contact and mechanical strength. The scintillators and light guides were wrapped in Whatman paper; an aluminized-Mylar reflector was placed on the detector face farthest from the light guide. Light shielding of the counters is provided by means of an opaque polyethylene film.

Fig. 3. Design of the light guide and scintillator of the counter:
1 — light guide, 2 — scintillator.

The uniformity of light collection was studied using a collimated \(Cs^{137}\) source. The counters \(C_4\) and \(C_3\) (for each of the two photomultipliers separately) had a light-collection uniformity of 30–40%. Figure 4 presents the results of measuring the light-collection uniformity of counter \(C_3\) in two mutually perpendicular directions when summing the signals from photomultipliers A and B. To monitor the operation of the photomultipliers, GaP pulsed light sources were glued to the light guides.

The photomultipliers were selected for maximum sensitivity; for each of them, the supply mode was chosen using a light diode powered by a G5-11 generator. All divider resistors were shunted by capacitances of \(1 \div 2 \cdot 10^3\ \mathrm{pF}\). In counter \(C_1\), the last 9 dynodes of the photomultipliers were shunted by capacitances of \(\sim 20\ \mu\mathrm{F}\). In addition, to eliminate voltage redistribution due to high loads \((1—2 \cdot 10^5\ \mathrm{pulses/cycle})\), the last dynodes were powered from a separate high-current rectifier of the UIP-1 type.

Fig. 4. Dependence of the light-collection coefficient on the point at which the particle passes. \(J\) is the photomultiplier current, \(r\) is the distance from the center of the scintillator. 1, 2—measurements along the axis passing through photomultipliers A and B and in the perpendicular direction, respectively.

Table 1

The design of the counter housings provided satisfactory shielding against electromagnetic pickup generated by the high-voltage equipment.

Calibration of the Counters

The time resolution of the scintillation counters was measured with relativistic electrons from the pair magnetic spectrometer by means of a time-to-amplitude converter. Counter \(C_2\) was used as the “start” counter, while the “stop” counter was varied. The time resolution—the full width of the time-distribution peak at half its height—is presented for different counter combinations in Table 2.

To measure the time resolution of the combination \(C_3^{A} + C_3^{B}\), two time-to-amplitude converters were used; their signals were summed and fed to a multichannel analyzer. The improvement in the time resolution for this combination, compared with operation of each photomultiplier separately, is connected with compensation of the time spread due to the geometrical dimensions of the scintillator.

Table 2

Calibration of counter \(C_2\), which measures the ionization losses of particles \((dE/dx)\) and is used to suppress the electron background, was carried out on a pair magnetic spectrometer. Electrons with a momentum of 300 MeV/\(c\) passed through counter \(S_0\) \((10\times10\times1\ \mathrm{cm}^3)\) and counter \(C_2\), located at a distance of 1 m from one another. The block diagram of the calibration is shown in Fig. 5. The spectrometric signal from the 9th dynode of the photomultiplier tube of counter \(C_2\) was split on a linear fan-out (LR). One of these signals, through a linear “gate” opened by the coincidence circuit CC 1 for 40 ns, was fed to an integrator (I) and then analyzed by a multichannel analyzer (AA). To avoid overloading the spectrometric amplifier, control of the linear gates was performed from the dead-time former (MV) \((f_{\mathrm{oper}}=10\ \mathrm{kHz})\). The circuits of the spectrometric channel are described in more detail in [1].

Fig. 5. Block diagram of the calibration of the \(dE/dx\) counter. K — lead collimator.

By varying the angle \(\alpha\) between the detector plane and the electron beam from \(90^\circ\) to the limiting angle determined by the scintillator width and the diameter of the electron beam, the path \(l\) traversed by the electron in the scintillator changes, \(l=d/\sin\alpha\). The dependence of the amplitude distribution on the effective thickness of the scintillator was measured, i.e., on the energy \(E_e\) lost by the electron in the scintillator,

\[ E_e=\left(\frac{dE}{dx}\right)_e l, \]

where \(dE/dx\) is the specific ionization loss of the electron.

Using the known data on the ionization losses of protons in the scintillator and taking into account the nonlinearity of the scintillator light output as a function of the type of particles and their specific ionization [2], one can convert from the energy lost by electrons to the energies of the incident protons. Figure 6 shows the dependences of the energy resolution and of the counter pulse amplitude on the absorbed energy.

For calibrating the integral discriminator, control of the linear gates was performed from the CC-2 coincidence circuit \((2\tau=25\ \mathrm{ns})\)

Fig. 6. Dependence of the pulse amplitude and the energy resolution \(dE/dx\) of the counter on the energy lost by particles in the scintillator.

Fig. 7. Amplitude distributions of \(dE/dx\) of the counter for energies lost in the scintillator of 1.86 and 3.6 MeV. The dashed curves are amplitude distributions for different discrimination levels at an energy loss of 1.86 MeV.

(Fig. 5). The threshold level was chosen from the condition of efficient registration (\(\sim 100\%\)) of protons with energies up to 300 MeV. A suppression of the electron background by a factor of 5 was obtained in this case. Figure 7 shows the amplitude distributions of the counter pulses for different energy losses and the amplitude distributions of the counter as a function of the discrimination level. From these distributions the uncertainty of the discrimination level was found to be \(\pm 0.24\) MeV.

2. Cherenkov \(\gamma\)-Spectrometer

The radiator of the spectrometer consists of two blocks of TF-1 glass: a cylindrical one 340 mm in diameter and 100 mm high, and a conical one with base diameters of 340 and 300 mm and a height of 200 mm, bonded with OK-72 glue. The side surface of the radiator is covered with an aluminum reflector. One spectrometric photomultiplier of type FEU-49 was used. The signal for amplitude analysis was taken

from the last dynode of the photomultiplier, and the “time” signal from the anode of the multiplier. The spectrometer was calibrated with a pair magnetic spectrometer. The analytical dependences of the energy resolution and of the amplitude of the spectrometer pulse on the electron energy \(E\) and on the displacement of the point at which the electron strikes the radiator from the center \(r\) have the form:

\[ \delta = \frac{0.22}{\sqrt{E(\text{GeV})}}, \qquad A \simeq A_0 \cos(6^\circ r). \]

The time resolution of the spectrometer, measured with a time-to-amplitude converter, is 13 ns (full width at half maximum) and is practically independent of the energy in the range 100–600 MeV.

3. Spark chambers and high-voltage power supply

The design and principal dimensions of the chamber are shown in Fig. 8. The working volume of the chamber is formed by two electrodes 1 and 2 and glass walls 3. Electrode 2 and the brass frame 4 are glued to walls 3 with ED-5 resin. Electrode 1 is removable; it is fastened to frame 4 by means of

Fig. 8. Design and principal dimensions of the spark chamber.

screws around the perimeter of the frame; sealing is accomplished by means of vacuum grease. Ten polyethylene plates 5, each measuring \(350 \times 320 \times 10\ \text{mm}^3\), are secured by means of polyethylene spacers located at the corners of the chamber. Electrodes 1 and 2 are made of duralumin; their thickness in the working region of the chamber is 0.1 mm. The chamber is filled with neon.

The capacitance of one gas gap of the chamber is 100 pF, and the capacitance of a polyethylene plate is 220 pF. Therefore, for an electric-field strength in the gas gaps of 20 kV/cm, the required amplitude of the high-voltage pulse is 280 kV.

The layout of the spark chambers and the connection of the pulse generators to them are shown in Fig. 9. In the first version, the pulse generators were made according to the Arkad’ev–Marx circuit with two-sided charging [3]. The pulse generator is triggered when the trigatron spark gap in the first stage of the generator is fired by a pulse with an amplitude of 8–10 kV. Adjustment of the pulse duration at the output of the pulse generator was carried out by means of a cutting spark gap of the trigatron type. The pulse generator is filled with gaseous nitrogen to a pres-

Fig. 9. Layout of the spark chambers and connection of the PFLs to them. 1 — Plexiglas plate, 2 — contact system, 3 — feedthrough insulator, R — discharge resistors.

pressure of 5–6 atm. The maximum pulse amplitude at the PFL output was 360 kV, the pulse rise time when the spark chambers were switched on was 10 ns, and the PFL triggering time was 70 ns. As the electrodes of the arresters wore out \((10^4\text{–}10^5\) operations), this time increased to 150–200 ns.

Operation of the spark chambers revealed a number of shortcomings of the initially selected version of the high-voltage power supply:

1) the unreliability of K-15-4 ceramic capacitors with an operating voltage of 30 kV in the short-circuit-discharge mode;

2) instability of the cutting arresters of the PFLs, appearing after \(10^3\) operations and leading to a change in the duration of the high-voltage pulse.

Therefore, in the second version, K-15-4 capacitors with a capacitance of 1000 pF and an operating voltage of 40 kV were used in the PFLs, with 5 capacitors in each stage of the PFL. A circuit with one-sided charging of the 10th stage of the PFL was used. In addition, in the second version of the high-voltage power supply the pulse of fixed duration is formed not by a cutting arrester, but by a passive circuit—a discharge resistance \(R = 80\ \Omega\), Fig. 9. Each resistance \(R\) is formed by five TBO-20 type resistors connected in parallel.

4. System for Recording Information from the Analyzers

Simultaneously with photographing the track in the spark chamber, the following were recorded: the particle time of flight, the energy of the decay \(\gamma\)-quantum, and the dose of radiation per event.

The chains of pulses from two BADs were fed to PP-9-1 counting instruments (p.p.). For more effective use of the p.p., one more input was added and the indication was divided into two groups of three decades each. For this purpose, the manufactured BU-3 amplifier-discriminator unit was inserted in place of the BP-17 unit. To increase the repetition rate of the input pulses at the second input, one of the BP-18 units was modified according to the circuit of the BP-20 unit. The program-control unit (BPU) was started by the pulse of the first BU-13 and simultaneously controlled both amplifier-discriminators and the entire indication scale [4].

To eliminate the influence of electromagnetic pickup, the PP-9s were placed in a shield made of sheet iron.

5. Spark-Chamber Trigger System

The block diagram of the electronics is shown in Fig. 10. Counters \(C_1\) and \(C_3\) measure the particle time of flight over a 3 m base. Counter \(C_2\) is located at a distance of 50 cm from \(C_1\), and its role is to reduce the number of accidental coincidences between counters \(C_1\) and \(C_3\). Coincidences between the pulses

Fig. 10. Block diagram of the electronics. \(\Phi\)—shaper, \(D\)—integral discriminator, \(V\)—“gate,” \(LV\)—linear “gate,” \(CC\)—coincidence circuit, \(\Phi MV\)—dead-time shaper, \(EP\)—emitter follower.

from the photomultipliers viewing scintillator \(C_3\) eliminate triggering by noise pulses (the noise-pulse level was rather high, since to improve the time resolution it was necessary to raise the voltage on the photomultiplier) and by pulses from Cherenkov radiation in the light guides. The resolution of circuit \(CC2\), equal to 4 ns, was determined by the method of [5] and corresponded to the time spread in the scintillator and photomultiplier. The resolving time of coincidence circuit \(CC3\) was chosen so that it encompassed the flight times of protons with the required kinetic energy. Since this coincidence circuit controlled linear gates through which the signals from counters \(C_1\) and \(C_3\) passed, the time-to-amplitude converter recorded signals with a time interval no greater than the resolving time of coincidence circuit \(CC3\). The linear range of the converter is 60 ns [6]; the coincidence circuit cuts out a segment of about 30 ns, and delay line \(L3\) shifts this range so that particles with \(\beta \sim 1\) are not registered. In addition, the use of the “gate” sharply reduces the load on the TAC, which has a large “dead” time. The signal from counter \(C_1\) passes through the linear gate to coincidence circuits \(CC4\) and \(CC5\). The signal from circuit \(CC5\) controlled the linear gates through which

the signal of the Cherenkov spectrometer passed to the analyzer ADC. The signals of the CC5 coincidence circuit switched on an indicator lamp. The impulse arcs from the AI-256 ADCs were fed to PP-9-1 scaling instruments operating in the pulse-illumination mode. A pulse of the Cherenkov spectrometer coinciding with a pulse from the proton telescope triggers the high-voltage supply. The “width” of the gate, equal to 0.6 μs, was sufficiently small to keep the number of random coincidences below 5%, since the loads at the “control” and “input” gates were very small. A polyethylene absorber 6 cm thick in front of counter \(C_4\) reduces the background of low-energy particles and the neutron background.

Signals from the CC4 coincidence circuit were sent to a PP-9-1; the count from this circuit was calibrated by the quantometer, the reproducibility accuracy of the count was no worse than 2%, and the number on the scaling instrument indicated the dose of \(\gamma\)-quantum radiation corresponding to triggering of the I.C. Simultaneously, signals from the ADCs and from the CC4 coincidence circuit were sent to the “Print” scaling device, from which they were read out to a digital printing device.

7. Optical system for photographing tracks in the spark chambers

The system for photographing the spark chambers is shown in Fig. 11. Two projections of the spark chambers were photographed (Fig. 12) through two Plexiglas

Fig. 11. Optical system of the spark chambers.

lenses \(87 \times 50 \times 10\ \text{cm}^3\) with focal length \(f = 4\) m. The mirrors are glass. Mirrors 4 and 5 serve to introduce into the frame information from two PP-9-1 scaling instruments. Photography was performed with an RFK-5 camera with a “Gelios-40” lens, \(f = 5\) cm, on type 15 film, 1300 GOST units. To calculate corrections for the geometrical distortion of the track coordinates by the optical photographing system, eight crosses were applied in four planes parallel to the walls of the spark chambers. The illumination of the crosses is pulsed and synchronized with the RFK-5 trigger pulse. The crosses were applied on a type 2A450 coordinate ruling machine, with an accuracy of \(\pm 10\ \mu\text{m}\). The accuracy of positioning in the vertical plane is \(\pm 100\ \mu\text{m}\). Figure 12 shows one of the stereophotographs of the track of a proton scattered in the plates of the spark chamber.

8. Calculation of the Registration Efficiency

The registration efficiency of the process

\[ \gamma + p \to \pi^\circ + p \]

is determined mainly by the efficiency of registering the decay photons with the Cherenkov spectrometer, whose aperture is smaller than the opening angle of the decay photons. Consequently, the experimental setup will register the indicated process with less than 100-percent efficiency. To calculate it, the method described in

Fig. 12. Stereo photograph of the track of a proton scattered in a chamber plate.

Fig. 13. Registration efficiency of \(\pi^\circ\)-mesons from decay photons by the Cherenkov spectrometer as a function of the bremsstrahlung-photon energy, calculated by the Monte Carlo method.

work [7]. The calculation was performed by the Monte Carlo method on a BESM-4 computer. Figure 13 gives the obtained value of the registration efficiency of \(\pi^\circ\)-mesons from the \(\gamma\)-quanta of the decay as a function of the energy of the bremsstrahlung photons. The accuracy of the calculation was \(\sim 3\%\).

References

1. V. A. Aulchenko, Yu. V. Korshunov. Preprint, Institute of Nuclear Physics, Siberian Branch, Academy of Sciences of the USSR, 14, Novosibirsk, 1970.

2. F. D. Brooks. Progr. Nucl. Phys., 5, (1956), 252; T. J. Gooding and H. G. Pug. Nucl. Instrum. and Meth., 7, (1960), 189.

3. P. S. Ananyin, V. I. Kryshkin, A. G. Sterligov, Yu. P. Usov. Proceedings of NIIYaF, issue 2, 1972, 92.

4. P. S. Ananyin. This collection.

5. V. I. Kryshkin, V. N. Stibunov, N. P. Fedorov. PTE, 4, 1971, 99.

6. V. I. Kryshkin, V. N. Stibunov, N. P. Fedorov. PTE, 6, 1970, 64.

7. C. Bacci, C. Mencuccini, G. Penso et al. Accademia Nazionale dei lincei, 49, (1965), 452.