On Enhancing the Contrast of an Electron Image by Cutting Off “Slow” Electrons

V. I. Gorbunov, D. I. Sviryakin, K. B. Umbliya, V. V. Evstigneev

As a natural result of the demands of modern technological progress, in recent years a new direction in science and technology has arisen and begun to develop in our country—introscopy.

The main aim of introscopy is the creation of methods and means that, with the aid of penetrating radiation and fields, make it possible to study the internal properties of materials that are opaque to ordinary visible light.

The physical basis of introscopy is the interaction of penetrating radiation or fields with the substance in which they propagate. This interaction may be expressed both in changes and variations of the absorption coefficient of the radiation along its path of propagation inside the body under study, and in changes in characteristics associated respectively with reflection, scattering, dispersion, polarization, diffraction, the Compton effect, the Mössbauer effect, nuclear reactions, magnetic resonance, and other phenomena.

The information capability of introscopy means is determined by the nature of the penetrating radiation and by the mechanism of its interaction with the substance of the object being inspected; by the intensity, spectrum, and geometry of the distributed flux of penetrating radiation; by the material, structure, and dimensions of the object under investigation; by the rate of obtaining and processing multielement information; by the method of recording images in invisible fluxes of penetrating radiation; by the fraction of effectively absorbed energy in the receiving element; by the type and parameters of the converter and image amplifier; and by the method of recognizing images of introscopic information.

As an agent capable of carrying multielement information about the internal structure of opaque bodies, many types of optically formed or spatially distributed fluxes of penetrating radiation may be used: from high-energy gamma quanta to radio waves of the millimeter and submillimeter ranges, as well as certain types of corpuscular radiation.

Recently, interest has grown sharply in methods of radiography using a beam of high-energy electrons. When radiographing specimens of low specific density with X-ray radiation, it is not possible to obtain sufficiently high contrast sensitivity, whereas electron beams, owing to the peculiarities of their interaction with matter, provide the sensitivity required for introscopy. The first experiments on radiographing products with a beam of fast electrons showed that this method is very promising [1], but in order to make full use of its possibilities, further—

development of detectors of spatially modulated electron fluxes, as well as systems for selecting an electron flux by energies and scattering angles, i.e., systems for changing the scattering function. A schematic of the simplest electron introscope is shown in Fig. 1.

Fig. 1. Schematic of the simplest electron introscope: 1 — object being transilluminated; 2 — phosphor layer on a transparent substrate; 3 — objective; 4 — electron-optical brightness amplifier; 5 — mirror; 6 — image analyzer.

The relative sensitivity of such an introscope is determined by the relation

\[ \frac{\delta}{d}=\frac{\Delta S}{S}\cdot\frac{B}{\mu d}, \tag{1} \]

where \(\mu\) is the coefficient of linear attenuation;
\(d\) is the thickness of the object at the point of transillumination;

\[ \frac{\Delta S}{S} \]

is the threshold contrast of the receiver;

\(B\) is the accumulation factor of radiation scattered in the material of the object being inspected;

\(\delta\) is the minimum size of a defect detected by the introscope.

The contrast sensitivity of the method depends to a significant degree on the threshold contrast of the optical-image receiver, in this case the eye, and improving the contrast sensitivity of the entire system requires the use of methods for contrasting images at the individual stages of acquisition and transmission of video information. To increase contrast sensitivity, it is expedient to proceed by enhancing contrast when converting the electron image into a light image. In the general case the optical contrast \(K_{\text{opt}}\) is not equal to the electron contrast \(K_{\text{e}}\); the relationship between them is determined by the relation [2]

\[ K_{\text{opt}}=\gamma K_{\text{e}}, \tag{2} \]

where \(\gamma\) is the contrast coefficient, which characterizes the enhancement of image contrast introduced by the converter. The larger the contrast coefficient, the smaller the electron contrast, or the smaller the thickness difference of the part being inspected, that can be distinguished under the same observation conditions.

The most widely used converters of electron fluxes into visible light—fluoroscopic screens and scintillation crystals—convert electron contrast into optical contrast almost without distortion. For these converters, practically \(\gamma=1\) and \(K_{\text{opt}}=K_{\text{e}}\). Additional enhancement of the contrast of the image of the internal structure of the object being transilluminated can be obtained by applying selection of electrons in a spatially modulated beam by energy. The design of an electrostatic selection system is schematically shown in Fig. 2. The retarding electric field between the grids \(C_1\) and \(C_2\), con-

Fig. 2. Diagram of the electrostatic selection system.

Fig. 3. Illustration of the operating principle of the electron selector:
a—curves of the electron energy distribution in defective (1) and defect-free (2) regions;
b—electron energy distribution after passage through the retarding field and cutoff of electrons with lower energy;
c—electron energy distribution after additional acceleration by an electric field.

determined by the potential difference \(U_1 - U_2\), does not transmit electrons with energies less than \((U_1 - U_2)\).

Fig. 3 illustrates the operating principle of the electron selector. In Fig. 3a, approximate curves of the energy distribution of electrons upon leaving the specimen are shown for a defective region (1) and a defect-free region (2).

The integral of the energy density of the outgoing electron flux in the defective region \(W_{\text{d}}\) will be determined by the area enclosed between curve (1) and the \(E\) axis (Fig. 3a). In the remaining part of the specimen, the integral of the energy density of the outgoing flux \(W\) will be determined by the area between curve (2) and the \(E\) axis. The contrast of the electron image is determined by the relation

\[ K_{\text{e}} = \frac{W_{\text{d}} - W}{W}. \]

Analysis of the electron energy-distribution curves shows that, when slow electrons are cut off, the difference \(W_{\text{d}} - W\) decreases, though only slightly, whereas the decrease in \(W\) is very substantial. By bringing the cutoff level to \(E_2\), one can obtain an image of the defect with 100% contrast; however, the number of electrons incident on the screen may then prove insufficient to form a sharp image with the brightness required for transmission. By varying the voltage \(U_2\), the optimum cutoff level can be selected.

With this filtration method, the flux density and energy of the electrons incident on the luminescent screen decrease sharply (Fig. 3b), resulting in a reduction in screen brightness. To increase the brightness, an accelerating electric field is applied between the second \(C_2\) and third \(C_3\) grids; its magnitude is controlled by selecting the voltage \(U_3\) (Fig. 3c).

This principle for increasing the contrast of images obtained with electron betatron introscopes [3] is sufficiently effective and simple to implement.

REFERENCES

1. V. V. Evstigneev, D. I. Sviryakin, V. I. Boiko, Yu. N. Adishchev. Resolving power of the scintillation method of introscopy with fast electrons. Defektoskopiya, 1973, No. 1.

2. V. A. Dobromyslov, S. V. Rumyantsev. Radiation Introscopy. Moscow, Atomizdat, 1972.

3. V. I. Boiko, V. I. Gorbunov, V. V. Evstigneev, D. I. Sviryakin. Betatron electron introscope. Defektoskopiya, 1973, No. 6.