From American Cinematographer, Non-Glare Coating Makes Lenses One Stop Faster by William Stull, A.S.C (March 1940)
For many years photographers have accepted the fact that a considerable proportion of the light that enters a lens never reaches the film. The greater part of this loss is occasioned by reflections from the various polished glass surfaces- inside and out- that make up the lens, and it increases with the increase in the complexity of the lens' design.
Thus the loss is considerably greater in many of the modern, complex high-speed lenses than in the simpler, older objectives. But since lenses depend for their action upon the use of many highly polished- and therefore reflective- glass surfaces, it has seemed that nothing could be done to avoid these losses.
Almost exactly a year ago, two Eastern research groups, working quite independently of each other, separately announced the development of methods of treating glass to eliminate surface reflections. That this created a sensation in photographic circles would be to put it mildly.
Here was the development so radically advantageous that it would practically revolutionize photography! But the excitement subsided when it was learned that in both cases the project was still in the experimental stage: that while lenses could be treated, the treatment was extremely delicate and not at all lasting.
Today the subject takes on renewed interest when it is learned that the stage of commercial practicability that the Paramount Studio has been using a set of treated lenses on actual production, with such excellent results that treatment of all the studio's lenses is being contemplated.
The process used is that developed by Dr. John Strong of the California Institute of Technology. Essentially, this consists of depositing upon each of the glass surfaces treated an ultra-microscopically thin chemical film. This film measures but four millionths of an inch (0.000004 inches) in thickness- exactly one-quarter the length of a light-wave.
As light falls upon this film, rays are reflected from both its upper and lower surfaces. Since the coating is exactly one-quarter wave-length in thickness, these reflections from the upper and lower surfaces are equal in intensity, and opposite in phase. What follows is similar to the well known interference effect in that these oppositely phased reflections cancel each other out. However, classic scientific theory does not account for the fact that with this treatment there is an increase in overall light transmission, which would occur if these reflections merely offset each other.
Some of the processes announced a year ago, produced this coating by immersing the glass surface to be treated in a tank of liquid on the surface of which was an infinitely think film (one molecule thick) of an insoluble soap.
Forms Permanent Coating
Repeated immersions or dippings, approximately 22- built up the desired quarter-wave-length coating. It was still, however, a delicate, soapy film, with all the inevitable drawbacks of a soap-film.
Dr. Strong's method, however, is different. In this, the desired surface coating is deposited by evaporation in a vacuum. Instead of a soap-film the coating is a metallic fluoride. While the details of the Strong process cannot, of course, be given as yet, it may be stated that the present treatment produces a coating sufficiently durable to withstand any normal handling.
Some of the lenses used at Paramount have been washed to remove accidental fingerprints, without in the least disturbing the non-glare coating.
In addition, exhaustive tests by the optical experts of the studio's engineering staff indicate that the treatment is in no way harmful to the normal optical qualities of the lens. It is non-corrosive, non-tarnishing, and does not pit or scratch the glass surface. It is further stated that the expense of treating lenses is not prohibitive.
The practical advantage gained from using treated lenses are many. Probably the most startling is the increase in effective speed. Normally there is an average light-loss of 5.22 percent from reflection for each air-to-glass surface inside or outside the lens: thus with a typical high-speed motion picture anastigmat like the Astro "PanTachar," which has eight such glass-air surfaces, the loss of light from reflections is in excess of 41 percent.
If such a lens is treated on all its external and internal glass-air surfaces, this loss is reduced to negligible proportions.
Increased Shadow Detail
But this accounts for only part of the actual gain in speed. Much of this reflected light finds its way back to the film as scattered, fog-producing light which tends to veil the shadowed areas. With this scattered light eliminated, it is possible with a given exposure to record a great deal of shadow detail which is normally lost.
Conversely, it is possible to obtain a given effect, as measured by shadow detail, with considerably less light than would be needed to give the same effect with an untreated lens.
These two gains are cumulative, and add up to a practical increase in speed of virtually one full stop, or between five and six printer-light settings in the Paramount laboratory. Thus a normal f/2.3 lens, when treated is the equivalent in speed of an f/1.6 objective, but still retaining the depth of field, definition, and optical quality of the f/2.3 design!
The elimination of the internal reflections gives a marked increase in the apparent definition of scenes photographed with treated lenses. The effect may be compared to that seen when using a fast lens with and without an adequate sunshade. The picture as a whole is visibly more crisp, and details not previously evident are suddenly revealed.
In the same way, depth of field is apparently considerably increased by the treated lens. It is quite possible that the circle of confusion is affected, since the resolving power is known to be increased.
Shooting Into Lights
Every photographer is familiar with the lens-flare which ordinarily results from shooting directly into strong sources of light, such as the sun or a studio lamp. The reflections from the several glass-air surfaces of the lens produce multiple, distorted images of the light-source or the iris diaphragm, usually with strong secondary halation streaks.
Comparative scenes filmed though treated lenses show an almost complete absence of these effects. Instead, a surprisingly clear image of the scene and the light source is obtained: such halation as is present is obviously photographic rather than optical, and attributable to photographic overexposure and to reflections from the film base itself.
The practical advantages which this treatment offers to cinematography can be well imagined. The increased speed can be of tremendous value in simplifying lighting. The increases in depth and definition should be of almost equal benefit under modern conditions, especially in the case of "follow focus" and dolly shots.
It may be mentioned, too, that insofar as can be determined as yet, the use of treated lenses should be equally feasible in natural color cinematography, in Technicolor or an other process. While a treated lens, if examined in the hand, by reflected light, appears to have an iridescent magenta sheen, the coating does not appear to have the slightest effect upon the actual color transmission of the lens, nor upon its color correction.
The visual image viewed on the focusing screen of a camera shows no trace of color alteration, and monochrome tests of standard color charts made under identical conditions with treated and untreated lenses show no difference in color rendition. It may therefore be assumed that the treated lenses may be used equally well in color photography; and due to the inherent limitations in speed and definition of all color processes, they could be used to even greater advantage in color than in monochrome.
Naturally , this non-glare treatment need not be confined solely to motion picture camera lenses. It will be equally beneficial when applied to the lenses of still cameras, optical printers, projection lenses, and the like, while the advantages to be gained from applying the treatment to the optical systems used in recording and reproducing sound, where speed and extreme resolving power are so necessary, should be equally revolutionary.