Silvering Glass
Application to Telescope Mirrors

      Today we are aware of a number of different procedures for reducing silver onto the surface of polished glass. Originally, these procedures only were intended to form a sort of silvering on the reverse side, like on a looking-glass. No one bothered about the uniformity of the thickness of the deposited layer, nor about whether it adhered well to the glass, nor about the extent to which it was polished on its back side. No one worried about speeding up the reaction by raising the temperature; the only concern was cost.

.

When applying this process to optics, the cost of silvering is pretty much insignificant. We have all the latitude in the world to satisfy conditions that take on a major importance at the moment that the metallic layer, deposited by chemical means, is called on to reflect light by its exterior surface, to form images, and to reproduce precisely the underlying glass surface. The Drayton process, which uses very pure alcohols as solvents and uses as reducing agents very expensive balsamic essences, is the one which we employed at the time of our first experiments. After three years of experience, it still seems to us to be the best method. It acts at ordinary temperatures, and the layer of silver which it forms on the glass is already reflective when it leaves the bath. It presents a uniform thickness and has shown itself to be sufficiently adherent to support a prolonged rubbing by a piece of leather reddened with iron oxide. Polished, it reflects about 75 percent of the incident light.

.

We have not changed anything in the procedure itself; but since we needed to make a novel and very delicate application of it,, we have been led to regularize some of the details of the manipulation, to change slightly the proportions of the elements that enter into the formula, and above all to study the empirical influence of each ingredient by either adding a bit more or a bit less. This was the only way to go to arrive, in all circumstances, at the best mixture of the variable products which one can find in commerce.

.

There are three operations which must be executed in turn on a glass mirror in order to give it the lively metallic brilliance of silver: the preliminary preparation or cleaning of the surface, the formation of the silver deposit, and the polishing of this same metallic surface.

.

The preparation of the glass surface which is to receive the silver layer exercises a great influence on the manner in which the reduction will operate. The silver-bearing solution, which has the special property of reducing itself when in contact with solid and polished surfaces, acts faster and forms a layer that sticks better and in a more homogeneous manner when the surface is more free from foreign bodies on its surface layer. For a glass surface to present this degree of chemical purity, it is not enough that is appears to the eye to be perfectly clear and shiny. When cleaning it, we must use measures whose efficacity have been demonstrated to need no other verification than that of the silvering operation itself.

.

Whether the surface has already been silvered or not, we begin by wetting it with a few drops of pure nitric acid, which we then spread around quickly with a wad of cotton. Then we rinse the surface with water, and wipe it with a dry cloth. In this state, the surface only retains on it that which comes from the water itself and the cloth we used to wipe it off. To purify the surface, if not completely, then at least to fix it uniformity, we powder it with pulverized Spanish chalk, add enough distilled water to form a paste, and spread it all over using another wad of cotton. We leave the glass to dry on its back long enough for the water to evaporate; the soluble substances will fix themselves in the chalk, which will serve as a carrier. Now it is the turn for this chalk to disappear. We take some carded cotton [in the 1850’s, people were still carding, combing, and spinning cotton and wool by hand – gfb.], without squeezing it, and with light rubbing we attack the layer of white chalk. The chalk will separate from the surface while still leaving it covered with a uniform veil-like haze. When this haze is removed, the glass will be left in the best state for being silvered. We form a new wad of cotton by superimposing regular layers of cotton taken from a card, and we rub lightly all the parts of the surface, taking care to remove the surface layer of cotton as soon as it is full of chalk. In this way, the haze which covered the glass slowly dissipates without any discontinuities or lines of demarcation being visible. Now we feel the cotton sliding on a clean surface. This is the moment to take a firmer cotton wad and to work it energetically on the glass while paying special attention to the area near the edges. After a while, when we feel that the surface cannot improve, we brush away with the cotton the dust that has a tendency to attach itself to the glass because of the static electricity produced by the rubbing. Then we lay the glass down while waiting to immerse it in the silvering bath. But before describing the latter operation, it is wise to give the formula to follow for preparing the solution.

.

The composition of the silver bath is fairly complex. Its primary ingredients are water, alcohol, silver nitrate, ammonium nitrate, ammonia, galbanum gum (see this page for what it is – Bob), and essence of cloves. Before entering the final bath, these elements are united in temporary solutions, whose compositions follow herewith:

 

1. Diluted ammonia. We take pure, commercial ammonia and we dilute it with distilled water until it reaches 13 degrees on the Cartier densimeter. (The Cartier densimeter is a predecessor to the Baumé Hydrometer –see this page for further info – Bob)
2. Ammonium nitrate solution, with ammonia. Into 200 grams of water, we dissolve 100 grams of dry ammonium nitrate, and we add 100 cc of the previous diluted ammonia solution; we then have a solution whose composition follows:
   • Dry ammonium nitrate                 100 grams
   • Distilled water                               200 grams
   • Ammonia, diluted to 13o Cartier   100 cc
3. Tincture of galbanum. One can find, in commerce, under the name of galbanum gum, a gum resin that is a bit soft, blond in color, and with a strong unpleasant smell. We reject any that crumbles, or is hard, or is odorless, or is greenish and mixed with a sort of inert chapeture (basically contaminants – Bob). We take about 20 grams of this substance with 80 cc of alcohol that is rectified to 36 degrees Cartier, and we mash it all together in a porcelain mortar heated to 40 or 50 degrees Celsius. We then obtain a solution of the resinous portion, still accompanied by an insoluble gum. We decant this into a flask and let it rest. We filter the liquid portion, throw out the opaque part, and by adding alcohol we bring this solution up to 29 degrees on the Cartier densimeter.
4. Tincture of cloves. This is a solution which results from the mixture of alcohol and the essence of cloves in the following proportions:
   • Essence of cloves                25 cc
   • Alcohol at 36o Cartier          75 cc

From all the products previously mentioned we then form a mixture composed as follows:

• Melted (??) silver nitrate                        50 grams
• Distilled water                                        100 cc
• Ammonium nitrate with ammonia (#2)   7 cc
• Diluted ammonia                                  24 cc
• Alcohol rectified to 36o Cartier             450 cc
• Tincture of galbanum (#3)                   110 cc

First we dissolve the silver nitrate in water, then we add the ammonium nitrate, which acts to prevent the solution from precipitating when we add the free ammonia. Then comes the alcohol, and lastly the tincture of galbanum. In other words, the substances should be incorporated one with the others, following the same order one sees in the formula.

.

The resulting solution promptly turns brown and forms a precipitate that deposits itself in a couple of days. We decant the clear portion and store it in darkness, where we keep it for use labeled “standard solution.” This solution, inactive by itself, has nonetheless a great tendency to reduce itself on contact with glass as soon as one adds three percent tincture of cloves (solution #4).

.

Meanwhile, the deposit which forms rapidly at 15 or 20 degrees Celsius, despite its good appearance, does not have all the consistency needed to resist a final polishing. The addition of four or five percent pure water, which slows the reaction, also gives the silver deposit greater solidity. If we add too much water, the solution will become too slow-acting, and the barely-formed layer of silver would stop in its development at such a degree of thinness that it could never acquire its normal coefficient of reflection. Thus, it takes careful observation and experience to help us decide precisely what quantity of water one needs to add to the standard solution to obtain the best deposit of silver.

.

The same concept holds with the ammonia, which, entering the mixture in very small amounts, is almost never added in precise enough quantities on the first try. If there is not enough ammonia, the solution works slowly, and then one has to decide whether this slowness comes from an excess of water or from a lack of alkalinity.. When it’s the ammonia that is lacking, the deposit of silver when taken out of the bath presents a very pronounced violet color and seems to be covered with a whitish veil. If on the contrary there is too much alkali, then under the influence of the cloves, the solution reduces itself en masse and to the detriment of the elective action of the walls (???), and the deposit on leaving the bath seems tarnished and covered with a sort of dark gray crumbly layer. The correct proportion of ammonia is that which gives the deposit a rich golden color tending towards rose, with the formation of a light ash-gray veil.

.
But while the addition of water happens by hundredths, the adjustments of pure and concentrated ammonia must never be more than by the thousandths. If by mistake one has added too much ammonia, the solution is not completely lost, because it is easy to repair it with nitric acid. The only result would be a slight increase of ammonium nitrate, which does not exercise a harmful influence on the deposit. To sum it up, it is by the water and the ammonia that one adjusts the solutions correctly. To avoid losses of time, one would do well to prepare in advance large quantities of standard solution, to pour them together in a single flask, to treat them en masse for final adjustments, and to store them hermetically sealed under the label of “tested solution.”

.

One should never try to silver an important piece unless one has an already old and previously-tested solution on hand. The operations takes place for large pieces in copper basins that have been silvered on the inside by electrolysis, so that will not be attacked on contact with silver nitrate. The diameter of the basin must be about 3 to 5 cm larger than the piece being silvered, and it must be deep enough as well. For mirrors of small dimensions, one can simply use porcelain platters that one can find in commerce.

.

It is necessary to finish off the back of the mirrors with a polished surface, and to leave this surface free of any obstacle that would impede the access of light and would prevent one from watching the progress of silvering. If a mirror is large enough that one cannot handle it securely by holding it only on the sides, then it is necessary to carve a groove along the sides where one can fasten two handles made out of cord, attached solidly by several turns of string. One must also prepare three small pieces of wood, or better yet whalebone, narrow and beveled that one will slide under the edge of the mirror immediately after its immersion in the bath, to prevent it from touching the bottom of the basin. This will also provide some space for the circulation of the liquid. Finally, when one works on mirrors that are fairly heavy, we make the basin sit on a plank of wood that has curves formed in it to act as sort of a cradle. In any case, the operation should take place in broad daylight and in a location that has been brought to a temperature between 15 and 20 degrees Celsius, because light and heat exercise an absolutely necessary influence on the reduction of silver.

.

Even if the surface to be silvered has undergone a perfect cleaning, if the immersion in the bath is not done with all the required precautions, various sorts of blemishes can occur in the silvering, or else unevenness or halting of the process. The basin having been cleaned with Spanish chalk, one prepares, to pour the solution into, a large cone of glued paper that one puts into a funnel like a paper filter. We cut a hole about 2 or 3 millimeters in diameter in the point of the paper cone for the liquid to flow out of. This hole is maintained 3 or 4 centimeters above the bottom of the basin. At the very moment we begin operating, we mix, while agitating them in the same vessel, the tested solution and the 3 percent of tincture of cloves that determine the reaction. Of this we pour a small amount into the basin and we spread it around with a wad of cotton; then, we pour the rest into the funnel, which flows out the hole and renewing its surface, and when flowing it only encounters walls that have been already wetted. We then take the mirror by its handles, hold it obliquely in order to make it firstly to rest at an angle of its principal face, and we lower it with a uniform movement which determines the progressive invasion of the layer of liquid. We slide the three holders into three equidistant points so that they will keep the mirror from touching the bottom, and we place the basin on its cradle while exposing it freely to the full light of the sun. From this moment on, one simply must stir the liquid gently while leaning the apparatus from one side to side, and also must turn the basin around by half-turns.

.

In the first few moments before the reaction begins, the surface submerged in a liquid that is less refractive than glass gives to exterior objects a visible image through the thickness of the disk. But soon, under the influence of the first deposits, this image weakens, takes on a brownish tint, and almost completely disappears. Then it suddenly reappears with a metallic shine, whereby one deems that the reflection has changed its nature. The elapsed time between immersion of the mirror and the reappearance of the reflected image is important to note, because it serves as a guide for the total amount of time for the reaction, which generally requires only about five to six times longer to produce a complete silvering. In normal conditions of temperature and light, the reappearance of the image occurs five minutes after immersion, and after a further 20 to 25 minutes in the bath, the layer of silver acquires all the thickness needed.

.

When one deems that the deposit has thickened sufficiently, one should take the mirror out of the bath, let the liquid drip off until it threatens to dry out, and then put it into a second basin containing ordinary alcohol diluted by water to a level of 67 degrees on the alcohol meter of Gay-Lussac or 25 degrees on the densimeter of Cartier. We stir the liquid until the drops coming off are no longer colored, and we then transport it into a third basin containing ordinary filtered water. A certain amount of stirring, without letting the surface emerge into the air, will hasten the dissolution of the alcohol into the water, but it is always prudent to prolong this washing beyond the six to eight minutes that are strictly necessary.

.

The mirror is finally put into distilled water, and from there placed on its edge in an almost vertical position in contact with a cloth, where we let it dry. When the operation has been conducted properly, we see the water level pull back and leave uncovered a surface with a golden-yellow color, tending towards rose, and covered with a light ash- gray veil. When examined by looking through the layer of silver, one only sees objects that are brightly lit, and they appear with a strong blue color.

.

Now we need to remove this light veil which colors the silver and reduces its reflectivity. Experience has taught us that it is necessary to start by rubbing this surface with a chamois skin that is placed over a soft wad of carded cotton stuffing. One must refrain from putting any polishing powder on the chamois leather, understanding that this preliminary rubbing is mostly intended to press down upon the silver deposit, to crush its inherent velvety structure, and to impart to it a solidity which will permit it to withstand a full polishing.

.

There is a singular phenomenon which never fails to happen and which seems to show that under the soft pressure exercises by the skin, the layer of silver modifies it constitution. The transparency which it displayed to a small degree when leaving the bath, diminishes noticeably during the rubbing; the transmitted blue color becomes darker as if the very small interstices capable of transmitting white light had just been obliterated because of the crushing of the gaps. After having been polished, the layer of silver, which has actually lost rather than gained material, in fact transmits less light than previously.

.

When the untreated chamois skin has produced its effect, one takes a second one set up the same way, but impregnated with fine English rouge that has been washed with the utmost care. We move it with circular strokes all over the surface, paying particular attention to the edges, which always have a tendency to remain behind. Bit by bit, the silver recovers it whiteness and gains a polish which reproduces that of the surface it rests on. This is the polish of the glass in its perfection, increased by the intensity of the metallic reflection. During an hour or two, depending on the extent of the surface to be polished, the mirror’s reflective qualities increase steadily. But later, once the reflection of objects that are in the shade gives a handsome black image, one must stop prolonging this treatment which otherwise would alter the texture of the thin layer of silver.

.

These are all the details of the process that we have followed for regularly silvering glass mirrors, without the surface of them displaying the slightest visible change under the different (optical) examination procedures.

.

We do not claim that all these precautions are strictly indispensable for succeeding in producing a silvering good enough for use; but having many times observed that only rarely does one resign oneself to accept even the slightest defects that impair the uniformity of a good surface, we have understood that we should indicate all of the methods, whatever they may be, to obtain mirrors without blemishes.