Films & processing

All photographic films - colour negative and transparency, black-and-white, infrared, process, etc. - are sensitive to ultraviolet radiation (because silver halide crystals are sensitive to ultraviolet). There is no significant advantage to using colour film for reflected ultraviolet work because only the blue layer of the tri-colour pack will be exposed by the ultraviolet (the integral yellow filter effectively preventing exposure of the red and green layers). Some workers however have designed systems that use source/filter combinations and colour film to give false colour renditions; these are dealt with in reflected ultraviolet photography with colour. For the pure recording of reflected ultraviolet radiation therefore black-and-white negative emulsions are invariably used.

The recording of ultraviolet radiation involves a number of problems even though silver halide has an inherent sensitivity to all radiation of shorter wavelength than the visible. Firstly, the emulsion grains themselves, and more importantly the gelatin support structure of the film, absorb ultraviolet significantly. This results in the ultraviolet image lying near the surface of the emulsion because the radiation is unable to penetrate very far. Modern films with flat, tabular-grained silver halide crystals near the surface of the emulsion, have a natural advantage therefore in recording ultraviolet. Unfortunately, though, in order to remove the generally unwanted complication of ultraviolet sensitivity, most films now have an ultraviolet absorbing filter built into the overcoating of the emulsion - this has the effect of reducing the relative sensitivity of these films to ultraviolet. Silver halide crystals are inherently sensitive to short wavelength radiation but the progressive absorption by the gelatin from 300nm downward results in an effective cut-off at 250 nm (90% transmission). The spectral sensitivity curve for a generalized panchromatic emulsion demonstrates that it has a relatively high sensitivity between 350nm and 400nm (Figure 42).

Figure 42 (above). The spectral sensitivity curve for a generalized panchromatic emulsion demonstrates that it has a relatively high sensitivity between 350nm and 400nm. Silver halide itself has an inherent sensitivity to all radiation of shorter wavelength than the visible but unfortunately the gelatin support of photographic film absorbs UV strongly from 300nm downwards resulting in an effective cut-off at 250 nm.

In order to determine the best film for reflected ultraviolet photography of patients extensive tests were carried out by the authors. A "high speed" film was indicated, as five or six stops more exposure were typically required over the visible equivalent, and a relatively small aperture was required to achieve adequate depth-of-field in patient photography. A high sensitivity to ultraviolet was therefore required. ISO film speeds are not necessarily a good indicator of ultraviolet sensitivity; previous workers had reported that slow "non-colour" sensitive films, such as Ilford's N531 (0.1 ISO), were just as sensitive to ultraviolet as panchromatic films of 400 ISO (Arnold et al., 1971).

Unfortunately, many of the special emulsions, such as Kodak's Spectroscopic Film Type 103-0, have either disappeared from the market, or are special bulk order only items. Also, many of the traditionally useful films now have the ultraviolet absorbing overcoats mentioned above. A range of commonly available films including the following: Kodak Tech Pan, Ilford Pan F, Ilford FP4, Kodak T-Max 100, Kodak T-Max 400, Kodak Tri-X, Ilford HP5 Plus and Kodak T-Max 3200 were tested by recording a resolution chart, 18% grey card and step wedge (Figure 43). Each film was exposed to visible radiation (no filter) and ultraviolet radiation (18A filter) in a strictly controlled manner. The films were developed as recommended by the manufacturers in ID-11, T-Max developer, or HP5 Plus developer, to a gamma of 0.55 or 0.60 (as determined from the density step wedge). Accurate densitometric and resolution data were then obtained from the negatives using a transmission densitometer and a travelling microscope. The various films responded differently in the ultraviolet spectrum in terms of sensitivity-ISO speeds were not particularly indicative - and some relatively fast films performed less well than medium speed equivalents. These tests measured relative sensitivity to ultraviolet, and resolution, in a practical working situation.

Figure 43 (left). An example of a test result from the extensive film testing undertaken by the authors. Each standardized photograph of skin was recorded using reflected ultraviolet radiation and incorporated a resolution chart, 18% grey card and step wedge.

The results from these tests showed that typically fifty times more exposure was required in the ultraviolet than in the visible to achieve a correctly exposed negative. The Tri-X turned out to be about one and a half times the speed of T-Max 400 in the ultraviolet spectrum, although nominally they are both the same ISO rating. However, the Tri-X had poorer resolution and was lower in contrast than the T-Max. It was interesting to note that the Ilford film, HP5 Plus, again 400 ISO, was in fact over four times the speed of T-Max 400 in the ultraviolet region and that FP4 was approximately 50% faster than the T-Max 400 in the same region!

It was clear that the HP5 Plus was the fastest of all the films, ie. more than three times faster than Tri-X, its next nearest rival, in relative sensitivity to ultraviolet. Initially, therefore, it was concluded that HP5 Plus was the best film due to its speed. Even though the HP5 Plus appeared the film of choice from these initial tests, its relatively poor sensitivity meant that an aperture of only f/8 could be employed, clearly not sufficient for patient photography.

The real problem remained that of sensitivity to ultraviolet, and it was obvious that "push" processing was required to increase film speed still further. From publications of previous workers (Frair et al., 1989) it was known that T-Max films responded very well to "push" processing. Further tests, therefore, involved processing the two selected films with extended development to determine which performed the best; HP5 Plus to 1600 ISO, and T-Max 400 to 1600 ISO. Figure 44 shows the family of D Log E curves for T-Max with progressively increased development. The results were compared to T-Max 3200 processed normally. It was found that HP5 Plus performed poorly, with marked increase in granularity that reduced resolution considerably. Although the T-Max 3200 was found to be faster than the T-Max 400 rated at 1600 ISO, the resolution was not as good. T-Max 400 can be "pushed" quite satisfactorily to 3200 ISO and it was quite interesting to note that even "pushed" to this speed the image quality obtained was actually better than the T-Max 3200 film itself (Figure 45)! Figure 46 shows the film's spectral sensitivity curve.

Figure 44 (above). The effects of increasing development with T-Max 400 film as shown by a family of DlogE curves.

Figure 45 (left). A reflected ultraviolet photograph of a resolution test chart taken with Kodak T-Max 400 rated and 'push' processed to ISO 3200; the whole image (top) and a much enlarged section from the centre of the negative (below) demonstrates the remarkable speed/resolution of this combination.

Figure 46 (above). The spectral sensitivity curve for Kodak T-Max 400 film.

Reflected ultraviolet photography suffers from the gamma/lambda effect - which is a reduction in contrast with decrease in wavelength . Figure 47 shows the reduction in contrast between a photograph taken with visible radiation and that imaged with reflected ultraviolet. As the wavelength decreases through the blue region to the ultraviolet, the gamma, or contrast, of the negative gradually falls (Figure 48). This is well known in the graphic arts industry when making colour separations; traditionally the blue separation has always been given more processing to raise its contrast, and similarly increased development is required in reflected ultraviolet photography. Processing must be increased by 25-30% to achieve a normal contrast index. Where "push" processing is being used to increase film speed it will automatically raise the contrast of the finished negatives. In order to obtain the required speed index of 3200 ISO, a processing time of 9.5 minutes must be used. This fortuitously raises the contrast index to approximately 25% higher than normal.

Figure 47 (above). DlogE curves for a test chart image recorded with visible frequencies (upper curve) and with reflected ultraviolet radiation (lower curve) show the reduction in contrast at the shorter wavelengths.

Figure 48 (above). A graph to show the effect of reducing contrast (Gamma) as the wavelength of the imaging source falls.

The film/processing combination to be recommended for reflected ultraviolet photography of patients is therefore T-Max 400 processed in T-Max developer, 1:4 for 9.5 minutes, at 24ºC, to give a relative speed of 3200 ISO.

References

• Arnold, C., Rolls, P. and Stewart, J., 1971, Applied Photography Focal Press, London.
• Frair, J. West, M. and Davies, J., 1989, "A new film for ultraviolet photography," J. For. Sci. 34 (1):234-238.

< Optical considerations

Exposure >