PROCEEDINGS
OF THE TOMSK ORDER OF THE RED BANNER OF LABOR POLYTECHNIC
INSTITUTE named after S. M. KIROV

Volume 184 1970

RADICAL-RECOMBINATION LUMINESCENCE OF SOLIDS. I. AN APPARATUS FOR INVESTIGATING RADICAL-RECOMBINATION LUMINESCENCE

V. V. STYROV

(Presented by Prof. V. A. Sokolov)

In recent years, a comparatively little-studied type of nonequilibrium radiation—candle luminescence—has received experimental and theoretical investigation [1, 2]. It has been established that the appearance of luminescence of solids in flames is caused by recombination processes on the surface of the solid involving free radicals and atoms present in the flame. However, a detailed study of the mechanism by which such luminescence arises is hindered by the complexity of the physicochemical nature of the flame. It is therefore expedient, apart from the flame, to study luminescence under the action of free radicals and atoms produced artificially, for example by means of an electric discharge (radical-recombination luminescence—RRL). The study of radical-recombination luminescence, in addition to being of purely scientific interest (the catalytic properties of the surface, migration of recombination energy to luminescence centers, the participation of surface levels in luminescence), also has practical significance. Thus, in 1960 [3] a new principle was proposed for increasing the luminous efficiency of incandescent lamps by adding to the radiation of tungsten the emission of a phosphor on the lamp walls, excited as a result of recombination of atomic hydrogen on its surface. In this case, to obtain atomic hydrogen it is proposed to use its dissociation on the surface of heated tungsten. However, with typical phosphors it has not yet been possible to increase the luminous efficiency because of the low quantum yield of the process [4]. There are very few works on radical-recombination luminescence, and the method for studying it has not been developed. Further study of radical-recombination luminescence and the determination of its maximum quantum yield are of interest. The apparatus described here serves these aims.

The vacuum apparatus (Fig. 1) is intended for exciting and recording the spectra of radical-recombination luminescence of powders, films, and single crystals over a wide temperature range (77–1273°K) and in an atmosphere of atoms of various gases. The apparatus also makes it possible to study the dependence of luminescence on an admixture of another gas, on the concentration of atoms in the gas phase, etc. Free atoms of simple gases (H, N, O) are obtained in an electrodeless high-frequency discharge. To excite the discharge, any HF generator with a power of ~200 W is applicable (we used a 20 MHz, 200 W generator from the workshops of the Siberian Physicotechnical Institute, as well as an LDI-1 generator of the Taganrog plant with regu-

Fig. 1. Schematic of the vacuum apparatus for investigating radical-recombination luminescence.

Visible labels in the diagram: VIT-1, EPP, FEU, VG-22, IM-2, GVCh.

controlled output power up to 4 kW, 40 MHz). The use of an electrodeless discharge has the advantage, compared with a low-frequency electrode discharge (the so-called Wood tube), that, first, the yield of atoms is considerably increased owing to the reduction of their loss on metal electrodes and, second, the discharge products are not contaminated by emissions from the electrodes. The quartz discharge tube 1, 32 mm in diameter, is attached to water-cooled ground joints 2 (cooling prevents the vacuum grease from softening when the tube is heated). High-frequency power is supplied to the discharge tube as shown in Fig. 1. Quartz is used because of its low loss angle at RF frequencies and its high melting temperature. The gas pressure in the discharge tube is measured with a shortened McLeod manometer 3 and an oil U-shaped manometer 4 (high-vacuum silicone oil). Free atoms are obtained in a flow system. Pumping is carried out by a VN-461M5 fore-vacuum pump and an N1-S2 diffusion pump 6. The system makes it possible to obtain a vacuum down to \(10^{-4}\) Torr, measured with vacuum gauge 26.

The gas-flow rate and the pressure in the discharge tube are regulated by a system of capillaries 7 connected in parallel through small vacuum valves. The method of making and calibrating the capillaries is described in 5. The capillary diameters lie within the range 0.05–0.2 mm, and their throughput, at a pressure difference of one atmosphere, can vary from 3 to 90 cm/min. To avoid pumping the system through the capillaries, bypass valves 9, 10 are included. The apparatus makes it possible to obtain hydrogen and oxygen on site by an electrolytic method and to purify them of impurities. Hydrogen is purified from oxygen impurity by passing it through a purification column 11 with a solution of pyrogallol A, and is dried from electrolyte vapors by passage through a trap 12 cooled with liquid nitrogen. Pure hydrogen is collected in a reservoir 13 of volume 10 l. Oxygen is dried by passage through a trap 14 cooled with liquid nitrogen and is collected in a reservoir 15 of volume 5 l. The hydrogen and oxygen pressures are measured with mercury U-shaped manometers 16, 17. The separation trap 18 serves to prevent moisture and oil vapors from entering the vacuum system. The phosphor 19 is placed at the end of the quartz stopper of ground joint 20 in tube 21. The tube is provided with a quartz window 22 and a Wood horn 23, which absorbs scattered light from the discharge. The phosphor is deposited on a glass or quartz substrate and is heated through the quartz wall by a miniature electric furnace 24. In place of the furnace, a refrigerant may be poured in for cooling. Near the phosphor, an atom-concentration meter of the Wrede–Harteck type is sealed into the tube wall (Fig. 2). Its main part is a narrow slit 1 (\(\varnothing\) 0.15 mm), whose diameter is considerably smaller than the length of the free path. Behind the slit there is a piece of aluminum foil 2, on which the atoms passing through the slit are converted into mole-

Fig. 2. Wrede–Harteck-type atom-concentration meter:
1 — slit, 2 — aluminum foil,
3 — tungsten filament, 4 — platinum leads

... molecules as a result of recombination. Thus, in front of the slit there are atoms and molecules, while behind the slit there are only molecules. As a result, a pressure difference \(\Delta P\) is established, recorded by two identical Pirani micromanometers [6] connected differentially in the two arms of a Wheatstone bridge. The unbalance of the bridge when the discharge is switched on serves as a measure of \(\Delta P\). The dependence of the resistance of the Pirani manometer on pressure at a given heating current is found by preliminary calibration against a McLeod manometer. The calibration graph is shown in Fig. 3. The degree of dissociation \(\alpha\) is calculated from the formula [6]

\[ \alpha = 3.41 \frac{\Delta P}{P}. \]

The pressure is measured with a McLeod manometer. We were able to register \(\Delta P = 10^{-4}\) torr at \(P = 10^{-1}\) torr. The working filament of the Pirani manometer was a tungsten coil from a 220 V, 15 W lamp, sealed in through platinum leads 4.

Fig. 3. Calibration graph of the Pirani manometer

The luminescence spectra in our apparatus are recorded automatically. Condenser 25 collects the light onto the entrance slit of a UM-2 monochromator, at the output of which there is an FEU-29 or FEU-22 photomultiplier. The signal from the photomultiplier passes through a balanced cathode follower to an automatic potentiometer EPP 09 (10 mV full scale). The spectrum is scanned by means of a Warren motor (2 rev/min).

To study the effect of an impurity of one gas or another on the radical-recombination luminescence, below the discharge a flask filled with the foreign gas is soldered on. To determine the quantum yield of radical-recombination luminescence, the absolute radiation flux is measured by one method or another (for example, with a calibrated photomultiplier). The number of recombinations per second is calculated from the known concentration of atoms in the gas phase and the recombination coefficient on the given surface. Some results obtained with this apparatus are presented in the following article.

REFERENCES

1. V. A. Sokolov. Optics and Spectroscopy, 4, 409 (1958).
2. F. F. Volkenshtein, A. N. Trofimov, V. A. Sokolov. In the collection Surface Properties of Semiconductors, Academy of Sciences of the USSR, Moscow, 1962, p. 179.
3. U.S. Patent, 1960, class 2, 920, 222.
4. G. N. Rokhlin, Z. N. Kobina, E. B. Volkova. Svetotekhnika, No. 1, 8, 1965.
5. A. N. Zaidel, A. A. Petrov, G. V. Vainberg. Spectral-Isotopic Method for Determining Hydrogen in Metals. Leningrad State University Publishing House, 1957.
6. I. C. Greeves and I. W. Linnett: Trans. Far. Soc., 55, 1338 (1059).