Full Text
Electroluminescent Light Amplifier
In a number of cases there arises a need to amplify a luminous flux, and it is often desirable that the amplification be accompanied by transformation of its frequency. As is known, such amplification is impossible by optical methods. It can, however, be achieved with the aid of combined electro-optical apparatus. Electron-optical image converters, sometimes using secondary-electron amplification, are usually employed as such apparatus. An entirely different principle underlies the device proposed for this purpose by the authors of the paper under review1.
The still little-studied phenomenon of electroluminescence was discovered comparatively recently; it consists in the fact that certain substances begin to glow when placed in an alternating electric field[^2]. The intensity of the glow, generally speaking, quite weak, increases within known limits as the field strength and its frequency are increased. Thus, if the intensity of the light beam being converted
FROM CURRENT LITERATURE
transform it into a voltage applied to an electroluminescent phosphor, the latter will begin to glow, and the intensity of its glow (i.e., the intensity of the converted light) will be a function (generally speaking, nonlinear) of the intensity of the irradiating beam. This can be achieved by means of the following simple device.
A transparent electrically conducting layer (electrode) is deposited on a glass plate and covered with a layer of electroluminescent material. Then a semiconductor layer is applied, forming a photoresistance with a large dark resistance. On top of the photoresistor a second transparent electrode is deposited, covered above by a glass plate. As a result, a rigid laminated system of the desired dimensions is obtained, convenient to handle and little subject to external influences.
An alternating voltage \(V \simeq 600 \div 800\) V with frequency
\[ \frac{\omega}{2\pi}=60 \div 3000 \text{ cycles} \]
is applied to the electrodes. Suppose that the semiconductor layer has electrical conductivity \(K\) (referred to unit surface area). Then the potential difference applied to the electroluminophor layer will be equal to
\[ V_p=\frac{KV}{(K^2+\omega^2 C^2)^{1/2}}, \]
where
\[ \frac{1}{\omega C} \]
is the capacitive resistance of the luminophor layer, referred to unit surface area (this expression is not entirely correct, since the electroluminophor also possesses active resistance, but the latter is usually much greater than the capacitive resistance). The semiconductor is chosen so that in the absence of light \(K \ll \omega C\) and, consequently, \(V_p \ll V\)—the luminophor remains dark. When the photoresistance is illuminated, \(K\) increases, and together with it \(V_p\) and the brightness of the luminophor glow increase.
If, in a first approximation, we assume that
\[ K=\gamma I^n, \]
where \(I\) is the intensity of the light irradiating the photoresistance, and \(\gamma\) and \(n\) are constants depending on the nature of the semiconductor, and that the brightness of the luminophor glow is
\[ B=\beta V_p^m, \]
where \(\beta\) and \(m\) are constants depending on the nature of the electroluminophor and on the frequency of the applied voltage, then, obviously, the brightness amplification coefficient is
\[ G_B=\frac{B}{I}= \frac{\beta V^m \gamma^{m} I^{(mn-1)}}{(\gamma^2 I^{2n}+\omega^2 C^2)^{m/2}}, \]
and the luminous-flux amplification coefficient is
\[ G_\Phi=G_B\frac{S_{\rm sv}}{S_{\rm osv}}, \]
where \(S_{\rm sv}\) and \(S_{\rm osv}\) are, respectively, the areas of the luminous and illuminated surfaces. For the device fabricated by the authors (CdS as the photoresistance), the constants had the following values:
\[ n=0.9,\qquad \gamma=1.01\cdot 10^{-4}, \]
\[ m=1.9,\qquad \beta=1.4\cdot 10^{-8}, \]
\[ C=75\ \text{pF}/\text{cm}^2; \]
Here \(I\) and \(B\) were expressed in lumens/cm\(^2\), \(V\) in volts, and \(K\) in megohms. As the authors indicate, the relation given satisfactorily describes the experimental results.
According to the measurements, the greatest amplification was attained at a frequency of 200 cps and \(I = 0.08\) millilumens/cm\(^2\). Under these conditions \(G_B \cong 24\), and \(G_\Phi \cong 480\) (\(S_{\mathrm{sv}} \cong 40\ \mathrm{mm}^2\) and \(S_{\mathrm{osv}} \cong 2\ \mathrm{mm}^2\)). In the case \(S_{\mathrm{sv}} = S_{\mathrm{osv}}\), obviously, \(G_\Phi = G_B\). Experiments showed that, for a device of dimensions \(7.5 \times 7.5\ \mathrm{cm}^2\), the smallest increase in brightness was about 2.
The authors note that if \(G_B > 1\), it is necessary to introduce a light-impermeable interlayer between the electroluminophore and the photoresistance layers, since otherwise optical feedback arises and the light intensity increases to saturation regardless of the magnitude of \(I\). It should be assumed that in some cases weak feedback may prove useful.
With regard to the resolving power of the device described, the authors give no data. There is every reason to think that it is small.
G. R.
Cited Literature
- R. K. Orthuber and L. R. Ullery, JOSA 44, No. 4, 297 (1954).
- G. Destrian, Phyl. Mag. 38, 700 (1947); Lehovec, Accardo and Jamgochian, Phys. Rev. 83, 603 (1951); S. Roberts, JOSA 42, 850 (1950).
-
B. Billings, W. J. Hitchehcock, and M. Zelikoff, J. Chem. Phys., 21, No. 10 (1953). ↩