ALL is not optical glass that glisters, but on the other hand not all that is not optical glass is contemptible. In a paper published in the Journal of the Optical Society of America, Vol. 28, No. 1, Dr. W. B. Rayton, of the scientific staff (lens designer) of the Bausch and Lomb Optical Co., makes some pertinent observations about this question:

"A Committee of this society spent many hours of discussion and carried out an extended correspondence in an effort to formulate a definition of optical glass without conspicuous success. Presumably such a definition should permit one to determine by inspection whether a sample piece of glass is optical glass or not. A definition in that sense is impossible. Certain specimens might be broken out of electric insulators that might, if judged only on a quality basis be classed as the finest optical glass while other specimens of glass, very difflcult to produce at all because of the extraordinary optical properties prescribed, would be immediately classed as ordinary glass of poor grade.

"In the consumer's mind, bubbles in glass are very offensive although from the optical standpoint they are the most harmless thing in the world. Bubbles are due to the volatilization of some of the materials in the batch in the melting process and because of greater viscosity they are held by some glasses much more tenaciously than by others. The manufacturer would like to take advantage of the more desirable optical properties of these glasses many times when he is compelled to use a less desirable glass because it is free of bubbles. Because of the refusal of the consuming public to accept bubbles in glass, he pays more for optical instruments than would otherwise be necessary and sometimes has to accept inferior performance.

"The best glass today absorbs not more than four to six tenths of a percent per centimeter except in the dense barium crowns and the densest flint glasses.

"Regarding the development of a sort of metallic luster generally known as tarnish: H. Dennis Taylor discovered to his surprise years ago that tarnished surfaces had a greater light transmission and a lower Fresnel reflection than clean, freshly-polished surfaces. Except for the appearance, then, which creates uneasiness in the mind of the owner or sales resistance in the prospective purchaser, this effect is not serious."

IN the same number (an optical glass number) of the journal named, George W. Morey, of the Geophysical Laboratory, Carnegie Institution of Washington, an optical glass expert, makes this statement: "Ordinary window glass today is of better quality than some pre-war optical glass, chiefly because of the reduction in iron content of the sand."

Incidentally, in the same article he mentions an interesting list: "Besides the eight disks supplementary to the 200-inch, all of which are of 'ribbed' structure, and one of which, the 120" flat, is larger than any disk previously made, there have been made in Corning seven solid type disks for reflecting telescopes. These are: a 24" for Cornell University; a 30" and a 36" for the Foundation for Astrophysical Research; a 60" for Harvard University, a 76" for the University of Toronto; an 81" for McDonald Observatory, Texas; and a 98" for the University of Michigan."

CONTINUING our gleanings on glass from recent technical papers, we take the following from an article on the X-ray determination of the structure of liquids and glass, by Dr. B. E. Warren, a Massachusetts Institute of Technology physicist who specializes in the X-ray study of the arrangement of the atoms and molecules of matter, published in the Journal of Applied Physics, Vol. 8, No. 10:

"Glass is usually called an under cooled liquid, the name suggesting that, although it differs has many of the mechanical properties of a true solid, it differs from the crystalline form of matter by not having passed through a sharp or definite transition in solidifying from the melt. From the X-ray studies we shall conclude that glass and liquids are; similar in that both are amorphous forms of matter. In one respect, however, their structures differ; in a glass each atom has permanent neighbors at a fairly definite distance, while in a liquid the neighbors about any atom are continually changing."

HOW the atoms are arranged in matter has been fairly well worked out within recent years by X-ray analysis. This is not to be confused with ordinary X-raying but consists of using X-rays as "feelers" for the atoms. X-rays are light having about 1/10,000th the wavelength of the light which our eves perceive. Thus they get down into the realm of actual atom size. If light of this short wavelength is shot into matter, a part of it will be reflected from or diffracted by individual atoms and the emerging rays can be photographed. (The rays which pass on through in the commonplace X-ray manner are ignored.) If the atoms are arranged in a pattern, as in crystals, the photograph taken proves also to exhibit a systematic pattern. By means of this research tool physicists during the past few years have been able to ascertain at least as much concerning the atomic arrangement within matter as a blindfolded man could determine about the arrangement of objects in a box by feeling around within it-almost as much as if it were directly visible. In fact, the extent to which this technique has been developed and the complication of existing atomic arrangements revealed by it are remarkable. For most practical purposes, then, we can now "see" the atoms in matter as satisfactorily as we can see the rows and cross rows of trees in an orchard (though we cannot see individual atoms). And it has turned out that most common things are crystalline: wood, for example-even rubber!

But glass is an exception-it is amorphous. Before the X-ray technique was devised we were partly sure that glass was amorphous but could not prove it so directly as now. And in the Technology Review (Vol. 39, No. 6), edited at the Massachusetts Institute of Technology, Philip M. Morse shows what glass is like. Largely, he also points out, it has been the same Professor Warren, quoted some distance above, who has done research on the X- ray patterns and atomic arrangement of amorphous materials including glass. In these the regularity and symmetry of atoms existing in crystals are absent. However, the atoms remain about uniform distance apart, and Figure 1, reproduced from the review named, gives an idea of the difference in atomic arrangement of the same substance (B is the example chosen), first in its crystalline form (above) and then when turned into borate glass (below). "A glass," Prof. Morse says, "is a clumsy caricature of a crystal of the same material, distorted and with parts left out here and there."

COGNATE with all this is the fact, recently discovered by the same X-ray diffraction method, that even liquids, including water, have some orderliness of atomic arrangement. Debye, the German chemist, states as a result of his researches that liquids are much more closely related to solids than they are to gases.

JUST what happens when glass is polished ? This is discussed in "ATM," pages 326-331, but since that note was written considerably more experiment has been performed and the subject has waxed in interest among physicists-particularly because we now, have the X-ray method described above.

Let us first summarize the several competing theories of the nature of polishing. First the theory of Newton and the younger Herschel: Polishing is nothing more than grinding or submicroscopically scratching down the protuberances with smaller and smaller abrasives until as Newton put it the visible "scratches and frettings of the surface become too small to be visible." Thus there is no essential difference between grinding and polishing. A theory of Elihu Thomson's described in "ATM " page 328 is a relative of this one: the rouge particles embed themselves in the pitch their cutting edges coming automatically to a common level and make submicroscopic scratches. (Perhaps this explanation makes more appeal to the common sense than any other but more recent evidence indicates that The Theory 3 below is closer to actual fact.)

Second, Rayleigh's theory that the operation is a molecular one. No pits are formed as in grinding with hard surface against hard by the breaking out of fragments but the material is worn away at first on the eminences almost molecularly. The microscope shows that as soon as the polished local areas can be observed at all they appear absolutely structureless. In its subsequent action the polishing tool extends the boundaries of these parts but does not enhance their quality (paraphrased from Rayleigh Trans. Opt. Soc. Oct. 1917).

Third the "butter" theory of Beilby whose experiments threw an entirely fresh light on the nature of polish. He demonstrated smearing or flowing of the surface layer. Polishing at right angles to scratches caused a flowing that filled up and hid the scratches. Etched with hydrofluoric acid the polished surface again revealed these scratches.

"The rouge particles hardly penetrate below the surface," Beilby states, "but coming into almost molecular contact with the sheet of molecules on the surface drag it off like a skin. The fresh molecular layer left by the removal of the skin retains its mobility for an instant and, before solidification, is smoothed over by the action of surface tension thus producing the liquid-like surface which is the necessary condition of a perfect polish."

Commenting on this Selby says: "In many respects glass is a liquid of extremely high viscosity-not a solid. Energy expended in polishing is manifested by heat which is sufficient to lower the viscosity of the glass near the surface to such a degree that this hyper- thin film-'beta layer'-can be made to flow." While heat is not a factor in the theories of Beilby or of French it is in those of Macaulay of the Royal Technical College at Glasgow also of Bowden and Hughes of Cambridge University. The latter two made experiments using two different metals as a thermo-couple and showed that in their sliding contact the surface temperature may be very high; in glass it would be still higher.

Letter from A. W. Everest: "I am beginning to lean toward plastic flow in polishing. I read an excellent paper pointing out the high temperatures generated in polishing- high enough in a thin layer of glass at the surface actually to melt it. If this is true then of course plastic flow occurs. However I still feel that most of the glass is removed.")

H. H. Selby, author of the chapter on flat making in "ATMA" and a chemist next describes his own experiments:

"Glass and light have one thing in common-dual personality. Only by assuming that light is both corpuscular and undulatory can some optical phenomena be explained So, in a way, must glass be considered to be in some respects solid, in others, liquid.

"For years, the several theories of polish have had their advocates, who have been engaged among themselves in acrimonious polemics and contradictory experiment. A year or so ago, the Lowers and the writer did all but take to poniard and rapier over the matter. The Lowers discounted the beta film (butter) theory of Beilby and of French and the abrasion theory of Rayleigh, but held to the planing, or imbedded particle, idea. The writer clung piteously to a combination of planing and surface flow as best describing the polishing of glass. As practical evidence, the Lowers offered the observation that the polishing liquor became less red and more white as polishing proceeded, claiming that the color change was due to removal of glass from the surface and suspension in the liquor.

"In an attempt to test this assumption, an f/l sphere was polished face up for 14 hours with a pitch lap and all the rouge liquor was saved for chemical analysis. If the Lowers were right, the solids suspended in the liquor should be quite high in silica (SiO₂), since the glass used was a mixture of the oxides of sodium, calcium and silicon. If, however, the writer's contention that the color change was due to emulsification of pitch constituents was correct, very little SiO₂ would be found.

"884 ml. (approximately 1 qt.) of liquor was evaporated to constant weight at 105 degrees C. Thus, the water, turpentine, and other volatiles were removed. The residue weighed 47.408 gm. This was ignited to constant weight at 850 degrees C. to remove the gums, resinates, etc., of pitch. This second residue was 44.971 gm.-a loss of 2.437 gm. A 5-gm. aliquot was then treated with hydrochloric acid, as in the usual SiO₂ determinations, and the treatments continued until the washings were free from iron. (The washings removed were examined nephelometrically and found free from colloidal SiO₂) The residue of silica, ignited, weighed but 0.0174 gm. over and above the SiO₂ found in pure pitch (.001 gm.), acid (.0003 gm.), rouge (.0066 gm. and water .0000 gm.), run as controls.

"The above results indicated that the rouge liquor was not whitened by glass, for an equivalent amount of SiO₂, as glass dissolved in alkali and precipitated by acid, had no effect, nor did ten times that amount, as silica gel, silex, diatomite or tripoli. However, this experiment helped very little in elucidating the question of the theory of polish, since it merely indicated, nay, proved, that some glass was removed-it did not show that no flow occurred.

"Another experiment was undertaken therefore to shed some light on the flow question, for the writer has reason to consider glass as a supersaturated solution of normally crystalline silicates which cannot crystallize due to the very high viscosity of the solution. (On a scale in which water=l, some glasses, even melting, would run to a viscosity of l0,000,000.)

"A 15 cm. flat, 23 mm. Thick, was very finely marked with radial grooves 4 cm. long, using a tungsten carbide pencil (Figure 2). The grooves were made in a lathe chuck, in triplicate, by drawing the spring-loaded pencil outward between guides, using kerosene as a lubricant. After many trials, smooth grooves were cut, the depth of which could be measured along the circle A, using the fine adjustment of the microscope, which could be read to 0.001 mm. Also, the thickness of the disk could be measured at B, using a micrometer calipers graduated to 0.01 mm. and readable to 0.001 mm.

"It was assumed that surface flow would be proved if a series of grooves of a given depth could be obliterated before the disk thickness had been decreased by a similar amount. Such proved to be the case.

"A typical pair of cases follows: (Each measurement was repeated ten times, 24 hours after last polishing period, at a constant temperature l degree C and the average value reported.)

A. Hard pitch, 5r% rouge suspension. 2

meters per min. Pressure, 0.3 Kg/sq.cm. Each wet dried. 3 grooves, av. depth 0.023 mm. obliterated in 22 wets. Thickness change 0.008 mm.

B. Hard pitch, 15% rouge, 2 M./min., 0.02 Kg/sq.cm., no wets dried. 0.011 mm. grooves obliterated in 60 wets. Thickness change 0.007 mm.

"The above trials, among many others, were repeated three times, with results of the same order of magnitude.

"The writer draws the following conclusions:

1. Surface flow does occur during pitch-polishing.

2. Glass is, at the same time, planed away.

3. Under the usual conditions of figuring (low pressure and thick rouge mixture) planing predominates, almost to the exclusion of flow.

4. Under rough and rapid polishing conditions, surface flow is marked and performs the major part of the polishing."

ON reading much of the literature about polish one conclusion becomes evident an oversimplified answer will never do.

The Society for Amateur Scientists (SAS) is a nonprofit research and educational organization dedicated to helping people enrich their lives by following their passion to take part in scientific adventures of all kinds.