Experimental Study of the Effect of Intensifying Screens on the Sensitivity and Contrast of X-ray Films

V. I. Gorbunov

(Presented by the scientific seminar of the Faculty of Physics and Technology)

Radiation having a maximum quantum energy in the spectrum of 25–30 MeV is the most penetrating in the case of radiographic inspection of a steel object. When the inspected object is very thick, the exposure time becomes very considerable. This is explained primarily by the special nature of the action of hard X-radiation on the photographic emulsion of X-ray film.

It is known [1, 2] that, at high hardness of the rays, the latent image is produced mainly as a result of the action of electrons that are formed in the surrounding medium, since the absorption of hard X-rays in the photosensitive layer itself is negligibly small.

The curve of the specific sensitivity of photographic film as a function of the hardness of the X-radiation (Fig. 1) clearly illustrates this situation. The maximum of the curve at a ray hardness on the order of 80–100 kV is explained by the fact that at low ray hardness their absorption by the photosensitive layer is very considerable; in other words, the latent image in the photoemulsion is formed mainly owing to electrons liberated in the photosensitive layer itself.

Fig. 1

The constancy of the specific sensitivity of photographic film to the action of harder radiation is explained by the small absorption of hard X-rays by the photographic layer itself.

In X-ray and γ-defectoscopy, intensifying screens have found wide application; they increase the sensitivity of the film to radiation. Two types of intensifying screens are usually used: metallic foils and fluorescent screens.

Metallic Intensifying Screens

It follows from the foregoing that, in the case of hard X-radiation, the latent image in the film emulsion is formed as a result of the action of electrons liberated by X-rays in the medium surrounding the film. Consequently, by placing a substance with a high atomic number next to the film, one can thereby considerably increase the number of electrons reaching the film per unit time; in other words, increase the blackening of the film.

It is known that the intensifying action of metallic foils depends on the material of the foil, its thickness, and the wavelength of the radiation.

The literature contains extensive information on the choice of metallic intensifying screens for radiography of metal articles with radiation of hardness up to 2 MeV [1, 2, 3, 4, 5, 6].

For harder radiation (20–30 MeV), there are no definite recommendations for the selection of metallic intensifying screens, although a number of articles have been published in which this question is considered [7, 8].

On the basis of the theory of the interaction of radiation with matter, we carried out a series of experiments on the selection of metallic intensifying screens for work with X-radiation having a maximum energy of 25 MeV. The experiments were conducted in two directions: toward selecting the optimum thickness of intensifying screens, taking into account their maximum intensifying effect, and from the standpoint of the greatest detectability of defects.

In addition, we considered the question of selecting the material of the intensifying screens. Let us first dwell on this question.

As already noted, to increase the sensitivity of the film it is necessary to use screens made of heavy metals, since the number of secondary electrons and, consequently, the degree of intensification increases with increasing atomic number of the screen material.

In practice, lead intensifying screens are used; they have a considerable specific gravity.

However, when selecting a screen it must be taken into account that not only the intensification effect but also the detectability of defects depends on the material of the intensifying screen, since, owing to scattering of secondary electrons in the screen itself, the contours of the image are blurred. The dimensions of the blurred image increase in proportion to the mean angle of electron scattering, and also with increasing screen thickness.

It is known [9] that the scattering angle increases with an increase in the atomic number of the screen material, whereas the required screen thickness decreases with increasing specific gravity.

From this one can draw the important practical conclusion that, in order to obtain a high-quality image with a large intensification coefficient, it is necessary to choose a screen material with a low atomic number and, at the same time, with as high a specific gravity as possible.

Consequently, lead is not the best material for screens, since tantalum, gold, and platinum have almost twice as high a density.

with a lower atomic number. This means that, if these metals are used as materials for screens, we can obtain, along with a high intensification factor, a high image quality.

Unfortunately, in our experiments we did not have the above-mentioned materials at our disposal; therefore all subsequent experiments were carried out with the aid of lead screens. The use of tin screens is less effective, since lead screens have a considerable advantage over them owing to the release of a larger number of electrons.

In all experiments to select the optimum thickness of lead screens, double screens were used: a front and a rear one. The front screen must perform two tasks. The first of these is to produce the greatest photographic effect on the emulsion owing to the intensifying action of the screen. The second, no less important task is to absorb the scattered electrons formed in the specimen under test so that they cannot reach the film and darken the image.

The rear screen is also used to increase the photographic effect by back scattering and, in addition, it improves the image quality to a slight degree. On the basis of all the foregoing, we carried out a series of experiments to select the most effective combinations of lead intensifying screens.

Two lead screens of complex shape were made (Figs. 2, 3). The front screen (Fig. 2) has the form of a square measuring \(130 \times 130\) mm, inside which there are 8 sectors with different thicknesses from 0 to 5.8 mm.

Fig. 2

On top of the lead sectors, using BF-2 glue, a set of similar sectors made of Plexiglas was glued; as a result, a composite sectorial screen was obtained, making it possible in a single exposure to obtain 8 blackening values as a function of screen thickness.

In addition, the sectorial form of the screen makes it possible to eliminate completely the influence of the intensity distribution in the beam, which considerably increases the accuracy of the experiment.

The rear lead screen (Fig. 3) also has the form of a square plate (\(130 \times 130\) mm), inside which there are 4 concentric [[unclear: word cut off at bottom of page]].

with a height of 1.1, 2, 2.9, and 4.5 mm. As in the case of the front screen, Plexiglas disks were glued on top.

The rear screen made it possible to obtain simultaneously four values of blackening depending on the thickness of the concentric rings. The distribution of intensity in the beam was taken into account by means of the exposure of an X-ray film, which was then photometered at points separated from the center of the beam by distances equal to the mean radii of the concentric lead rings. The lead surface of both the front and the rear screens was carefully treated in order to ensure the best contact with the surface of the X-ray film.

Fig. 3

The study of the intensifying action of the lead screens was carried out for two cases—without an absorber and with an absorbing steel body. Exposures without an absorber were made with the distance from the target to the film equal to 250 cm. Several types of X-ray film were exposed: “Roentgen XX,” “Roentgen X,” and “Agfa-Duro.”

In the course of the study it was established that, among the tested types of films, there is no fundamental difference from the point of view of the action of lead intensifying screens on them.

In studying the intensifying action of lead screens in the absence of an absorber, all types of X-ray film show increasing blackening with increasing screen thickness. The strongest blackening was found in fields with a front screen 1.5–2.5 mm thick and a rear screen 2.5–3 mm thick.

In the case of tests of lead screens behind a steel absorber, a different picture was observed: a thinner screen (0.5 mm thick) gives a more distinct intensification in comparison with fields without front screens.

Figs. 4 and 5 present the results of measurements \(D=f(d)\) for three exposures of sets of lead screens: Fig. 4 without an absorber and Fig. 5 with a steel absorber 100 mm thick. Each curve shows the course of blackening with increasing thickness of the front screen for the corresponding thickness of the rear screen. Comparison of the curves gives the value of the intensifying action of the various rear screens.

From Fig. 4 it is evident that, with a 2.5 mm lead screen, radiation equilibrium is practically maintained, i.e., maximum intensification is achieved. A further increase in the thickness of the front screen does not lead to an increase in blackening, but, owing to absorption of the primary radiation, leads to its steady decrease.

Fig. 4

Fig. 5

The blackening curves in Fig. 5 show a noticeable initial rise, which can be explained by a disturbance of radiation equilibrium due to filtration of the secondary electrons released in the body under study by the front aluminum wall of the cassette, which intercepts more electrons than it releases. In this case a lead screen 0.3–0.5 mm thick is sufficient to restore radiation equilibrium and correspondingly intensify the blackening. The subsequent, flatter course of the blackening indicates that, for front-screen thicknesses greater than 0.5 mm, absorption and release of secondary electrons are approximately equal.

The observed decrease in blackening at front-screen thicknesses from 2 to 3.3 mm of lead (Fig. 5) can be explained by the filtration of Compton electrons released in the test body and possessing higher energy.

After studying the effect of the thickness of the lead screen on the degree of blackening of the film, experiments were carried out to determine the dependence of image quality on screen thickness.

Consequently, with a smaller screen thickness, the sharpness of the image of defects increases. However, the screen thickness must not be too small, since otherwise the scattered electrons formed in the specimen under test and in the low-atomic-number walls of the cassette will be able to reach the film and darken the image. From this point of view, the thickness of the front screen should be approximately equal to the mean range of the electrons.

However, as already indicated above, increasing the thickness of the lead screen beyond a certain value will have an adverse effect on the quality of the radiograph. This is clearly confirmed by the schematic graph of the change in the intensity of hard X-rays and secondary electrons formed in the absorber and in the lead screens (Fig. 6).

Fig. 6

It follows from the graph that the density of secondary electrons formed in the front screen increases sharply, above all in the first third of the mean electron range, whereas the rise in the last third of the range can practically be neglected (curve 3). The increase in image sharpness due to absorption in the last third of the screen, equal to the mean range of the secondary electrons formed in the test body and in the cassette walls, is so small (curve 2) that it is compensated by the stronger scattering of the screen electrons.

Thus, the image quality as a function of the thickness of the front screen should improve only up to \(2/3\) of the electron range; thereafter this dependence will be insignificant. For the rear screen, the indicated dependence is preserved.

The experiments we carried out to determine the dependence of image quality on the thickness of the intensifying screen showed good agreement with theoretical calculations.

All experiments were carried out with a steel absorber 100 mm thick. The distance from the focus to the film was 1.5 m; the film, in an aluminum metal cassette, was placed directly behind the absorber. As the front screen, the sectorial lead screen already described above was used. The back screen was changed each time before the next exposure; in all, 7 back screens were used, with thicknesses from 0 to 4 mm in steps of 0.5 mm. On the surface of the steel absorber located closer to the focus, steel wires 1 mm in diameter were placed in the form of radial lines, arranged so that opposite each sector of the front screen there was one wire. All experiments were carried out with “Rentgen XX” film.

The results of the study are shown in Fig. 7. Fig. 7 shows the relative detectability of the defect as a function of the thickness of the front and back screens. It follows from the graph that the best detectability is obtained for a front lead screen 3.0 mm thick and for a back screen—2.5 mm thick.

Fig. 7

Comparing the results of measurements for determining the dependence of blackening on the thickness of the lead screen with the results obtained in determining the detectability of the defect as a function of screen thickness, one can select the most effective thickness of the lead screens.

Such thicknesses, in the case of X-rays with an energy of 25 MeV, are: a front lead screen of 2 mm and a back lead screen of 3 mm.

Subsequently, in all experiments where it was necessary to use lead intensifying screens, lead screens of the indicated thickness were used.

Fluorescent intensifying screens

As further experiments on the radiography of steel objects showed, exposures for steel more than 50 mm thick with the use of lead intensifying screens become so large that it was necessary to employ—

... less effective intensifying screens. Such screens are fluorescent intensifying screens, which, as the experiments have shown, can be used with comparative success in radiography with hard X-rays.

Modern fluorescent screens produced by domestic industry consist of cardboard with an active layer deposited on it. The active layer is crystalline phosphors, which, under the action of X-rays or γ-rays, give the greatest output of luminescent light. The action on the film of the additional light from the fluorescent screen increases the overall blackening of the film.

The excitation of fluorescent screens by X-rays occurs not by the X-rays themselves, but by photo-, Compton, and secondary electrons formed in the screen material as a result of the absorption and scattering of X-rays.

Moreover, for a given luminescent substance the excitability by X-rays is the greater, the higher its absorption coefficient for X-rays. This high absorption coefficient is characteristic of substances containing heavy atoms (1, 2). From this point of view, the most acceptable phosphors may be considered the alkaline-earth tungstates, for example \((\mathrm{CaWO}_4)\), as well as uranium salts and double salts of the type barium platinocyanide \([\mathrm{BaPt}(\mathrm{CN})_4 4\mathrm{H}_2\mathrm{O}]\).

It is known that tungstate screens have low visual brightness of luminescence; however, their spectral composition is very suitable for the photographic emulsion of X-ray films [1, 10].

Figure 8 shows the dependence of the energy, in relative units, emitted by calcium tungstate on the wavelength of the luminescent light [1]. It follows from the graph that calcium tungstate emits energy predominantly in the ultraviolet region of the spectrum, partly including the violet and blue regions of visible light.

Fig. 8

The relative sensitivity of an X-ray film is shown by curve 2 (Fig. 8), from which it follows that the silver bromide salt of the emulsion has the greatest sensitivity to wavelengths located predominantly in the ultraviolet region of the spectrum.

Depending on the loading of the luminous layer (the weight of phosphor per unit surface area), the types of tungstate screens indicated in Table 1 are currently manufactured [1].

Table 1

We had at our disposal tungstate screens Nos. 1, 3, and 4. The characteristics of tungstate screens for industrial radiography with radiation of hardness 25 MeV were investigated using a steel absorber of various thicknesses.

First of all, experiments were carried out to compare the performance of intensifying screens with different coating weights of the luminescent layer (40, 120, 160 mg/cm²). The tests were performed for the domestic film “Roentgen X,” with the distance from the target to the film equal to 150 cm. Screens with different coating weights of the luminescent layer were cut in the form of sectors, pasted onto black paper, and then placed in a cassette.

After radiographing steel plates 50 and 100 mm thick with different exposure times, all films were developed under identical conditions. Subsequently the films were photometered with an MF-4 microphotometer. The results of the investigation, in the form of curves, are given in Fig. 9. It follows from Fig. 9 that, with an increase in the coating weight of the luminescent layer, the intensification factor of the screens increases not in proportion to the increase in the coating weight of the luminescent layer. This can be explained by absorption and scattering of the fluorescence radiation on its path from the place where it arises to the photographic emulsion.

Fig. 9

In addition, consideration of the curves also shows that, as the absorber increases, the difference in the intensification factor for screens with different

with the loading decreases. From this one may conclude that, with a greater thickness of the steel absorber (i.e., precisely in the region of application of fluorescent screens), screens with a loading of 120, 160 mg/cm² and higher give approximately the same amplification.

Subsequently, in all our experimental studies we used screens with an active layer of calcium tungstate with a loading of 120 mg/cm² (front and rear screens).

When radiography is performed with fluorescent screens, along with the reduction in exposure, the sharpness of the image of defects is also reduced to a considerable degree. As the experiments showed, for large defects such a decrease in sharpness has a very slight effect. However, the detectability of small defects (cracks, lack of fusion) drops sharply in comparison with radiographs obtained using lead intensifying screens.

This shortcoming of fluorescent screens is explained by the scattering of fluorescence light by the grains of the active layer, by the scattering of electrons released by X-rays on their path to the film within the volume of the fluorescent screen and the object being radiographed, and also by the imperfect contact of the fluorescent screen with the film.

Fig. 10

All this compelled us to seek the possibility of increasing the sharpness of defect images when radiography is performed with fluorescent screens. As a result of the study it was found that the use of composite intensifying screens (lead and fluorescent together) makes it possible

significantly increase the sharpness of the image of defects than in the case with fluorescent screens alone.

Figure 10 shows the dependence of defect detection on the thickness of the steel plates being radiographed for three cases: lead screen 2 mm—film—rear lead screen 3 mm (2P3); fluorescent screen with a coating weight of 120 mg/cm²—film—rear fluorescent screen with a coating weight of 120 mg/cm² (FPF); lead 2 mm—fluorescent screen 120 mg/cm²—film—fluorescent screen 120 mg/cm²—lead 3 mm (2FPF3). As follows from the graph, the detectability of defects in the case of the composite screen (2FPF3) is much higher than in the case of the screens (FPF). All experiments were carried out with “Agfa Duro” film, which in each case was placed at a distance of 1500 mm from the target.

To compare the intensifying action of various combinations of intensifying screens, several “Agfa Duro” and “Rentgen XX” films were exposed behind a steel absorber 200 mm thick at a focal distance of 1500 mm. As a result of photometry it was found that the use of lead intensifying screens (2P3) reduces the exposure, as compared with the exposure of film without screens, by a factor of 2–2.5; the use of fluorescent screens (2FPF3) reduces the exposure by a factor of 7–11; and, finally, the use of composite screens (FPF) reduces the exposure by a factor of 10–15.

Conclusions

1. The best lead screens, both in terms of defect detectability and in terms of maximum sensitivity, are screens 2 mm thick (front screen) and 3 mm thick (rear screen).

2. Fluorescent screens with an active layer of calcium tungstate and a coating weight of 120 and 160 mg/cm² differ almost not at all in their action when the thickness of the steel absorber being radiographed is 200 mm and above.

3. Although fluorescent screens considerably reduce the exposure when radiographing steel (by 10–15 times), the detectability of defects when they are used is considerably lower than in the case of lead intensifying screens.

4. The use of composite intensifying screens (2FPF3) makes it possible to increase the sharpness of the image of defects while increasing the exposure as compared with FPF screens (by a factor of 1.5).

Literature

1. Rumyantsev S. V., Grigorovich Yu. A. Control of the Quality of Metals by γ-Rays. Metallurgizdat, 1954.

2. Tatochenko L. K., Medvedev S. V. Industrial γ-Defectoscopy. Metallurgizdat, 1955.

3. Trapeznikov A. K. X-ray Defectoscopy. Mashgiz, 1948.

4. St. John E., Isenburger G. Industrial Radiography. Mashgiz, 1947.

5. Malyshev M. V. Radiography of Metals. Metallurgizdat, 1952.

6. Umanskii Ya. S. X-rayography. Mashgiz, 1951.

7. Schittenhelm R. ATM. Januar 1956, S. 1–8.

8. Möller H. Archiv für das Eisenhüttenwesen, May–June, 1954.

9. Shpolskii E. V. Atomic Physics, Vol. I, II. State Publishing House of Technical-Theoretical Literature, 1950.

10. Riehl N. Luminescence. Foreign Literature Publishing House, 1932.