Grantville Gazette - Volume V (35 page)

Where can we find lead oxide? As it turns out, lead oxide has, for thousands of years, been a byproduct of the "cupellation process" of producing silver. There is a small amount of silver in galena, the principal lead ore. The main component of galena is lead sulfide, and it is readily oxidized, when roasted in a wood fire, to form lead oxide. The lead oxide ("litharge") can then be separated out by absorbing it with bone ash.

Encyclopedia Americana
lists Germany as a leading lead-mining country, although well behind the United States and Australia. It also says that there are major deposits of galena in Germany. Down-time miners should be well aware of it, as it has a very distinctive appearance.

Borosilicate glass contains, as you might expect, silica and boron oxide. It is used primarily in chemical glassware and in ovenware.

Modern borosilicate glass was developed in 1912. Curiously, by 1225, the
Chinese
were already aware of the use of borax in
Arab
glassmaking. Zhao Rukuo noted that "borax is added so that the glass endures the most severe thermal extremes and will not crack" (Smith).

Seventeenth-century European glassworkers were extremely secretive about their craft. It is conceivable that prior to 1632, there was some European use of borax in glassmaking. However, the first definite reference dates back only to 1679, when Johann Kunckel (1630-1703) mentioned borax in a recipe for an artificial gem (Smith).

The
Encyclopedia Britannica
gives us a starting point for formulating these glasses: for chemical glassware, use 81% silica, 12% boron oxide, 4.5% (sic) sodium oxide, and 2% aluminum oxide. The same table also gives a formula for an optical borosilicate glass, "crown" glass: 68.9% silica, 10.1% boron oxide, 8.8% sodium oxide, 8.4% potassium oxide, 2.8% barium oxide, and 1.0% zinc oxide. Two more borosilicate glass formulas can be gleaned from
Collier's
; these leave out the barium and zinc oxide, but do contain a little aluminum oxide. And a fifth recipe appears in the
World Book Encyclopedia
. So take your pick. Collectively, the indicated range in boron oxide content is roughly 10 to 25%.

Boron oxide is readily obtained from borax (hydrated sodium borate), other borate salts, or boric acid. In 1632, borax was imported to Europe from Tibet under the name of "tincal." We shouldn't have any difficult getting our hands on tincal from our Venetian trading partners. The real question is price. Borax from Tibet was a luxury item in Renaissance Europe, used primarily by goldsmiths and assayers. If we want to make more than just small quantities of borosilicate laboratory glassware, we will want to exploit nearer sources. Fortunately, the encyclopedias provide some clues.

Boric acid can be obtained from the lagoons in the "Maremma" of Tuscany (this source was not known in our time line until 1777). The 1911 EB entry for "boric acid" describes in some detail how the boric acid is recovered.

The encyclopedias also reveal that "pandermite," a hydrous calcium borate, can be obtained from Panderma (Panormus) on the Sea of Marmora: "it occurs as large nodules, up to a ton in weight, beneath a thick bed of gypsum." Panderma (Panormus) also is said to have a trade in "boracite."

Boracite, a mineral containing magnesium borate, can even be found in Germany. "Small crystals bounded on all sides by sharply defined faces are found in considerable numbers embedded in gypsum and anhydrite in the salt deposits at Lüneburg in Hanover, where it was first observed in 1787. . . . [A] massive variety, known as stassfurtite, occurs as nodules in the salt deposits at Stassfurt in Prussia." (1911 EB, "boracite"). One of my field guides to minerals actually has a photograph of a boracite crystal found in Bernburg, in Thuringia, Germany. (Hochleitner, 206).

Down-time glassworkers will need to adapt to the special properties of borosilicate glass. It has a softening point of 820 deg. C and a working point of 1,245 deg. C. In contrast, the values for the familiar soda lime glasses are typically about 750 and 1,000, respectively. (Just to complete the picture, the values for lead-alkali glass are 677 and 985, respectively. All these numbers are in the
Encyclopedia Britannica
.)

There are four other major types of modern glass: aluminosilicate; fused quartz; fused silica; and 96% silica. These glasses are all more resistant to high temperature, heat shock, and corrosive agents than borosilicate glass, but also more difficult to make and work. The other USE industries have not yet advanced to the point where they are needed.

Down-time glass makers already have many colorants and decolorants. However, they don't yet know how to make the famous ruby glass of Bohemia, because its inventor, Johann Kunckel (1630-1703), is still in diapers. The secret to reproducing this glass is the use of microscopic particles of gold chloride.

Glass is
cast
by pouring it, while liquid, into a mold. Because of its viscosity, glass does not fill a complicated mold shape without assistance. A gob of molten glass can be forced, by means of a plunger, to spread throughout the cavity. This is called
pressing
.

Glass may also be
blown
. A bubble of molten glass is placed inside a mold and more air is forced into it, causing it to expand into contact with the mold.

In
drawing
, a tool called a
bait
is lowered into the molten glass and then raised. The glass adheres to the bait, and depending on the shape of the bait, a thread, rod or sheet of glass is drawn up. Glass may also be
extruded
through holes, as a result of centrifugal force, or a blast of air.

Molten glass can also be squeezed between rollers to produce a flat glass.
Rolled
glass may subsequently be
floated
on a bath of molten metal, so it smooths out.

Almost all of the processes mentioned above were initially carried out by hand, and later by machine. As a nation of "master mechanics," Grantville will certainly attempt to find ways of automating the seventeenth-century glasshouse. However, it would be a grave mistake for them to attempt to duplicate, with machinery, a process which they have not carried out manually. All sorts of little things go wrong when you try to automate a complex process, and, if you don't have a deep understanding of the handicraft, then you aren't sure whether the problem is with the machinery, the raw materials, or whatever.

Once you have an understanding of the subtleties of hand blowing, casting, drawing, pressing, rolling, grinding and polishing glass, you are ready to consider whether any of the steps can be automated. The
World Book Encyclopedia
sketches out methods of blowing glass bottles, pressing glass dishes, and drawing glass tubing (the Danner process). The 1911
Encyclopedia Britannica
describes mechanical methods of pressing, blowing and drawing glass. It even discusses the famous 1904 Owens "suck-and-blow" bottling machine. The modern
Encyclopedia Britannia
divulges two more methods of making glass tubing (the downdraw and Vello processes), the workings of the 1926 "ribbon machine" (which makes 30 light bulb shells per second), and the basics of blow-and-blow bottle making on the "Individual Section" machine. Both modern reference works describe the jewel of the modern glass industry crown, the Pilkington float process (discussed in a later section) for making plate glass without polishing steps.

In general, the encyclopedias describe the basic operations that the modern machinery performs, but not the specific mechanisms which accomplish them. For example, the
World Book Encyclopedia
(WBE) describes a mechanical method of blowing glass bottles: drop a gob of glass into a bottle-shaped mold (neck end down); blow air in to force the glass into the neck; flip the mold; and then blow air in again to force the glass down to form the walls of the main compartment (WBE 214). But the neophyte mechanical engineers of Grantville will need to figure out the exact gadgetry that will ladle out or suck up that gob, manipulate the mold, and blow on command. They will also need to design the appropriate conveyors, guides, controls, safety rails, etc.

The same reference also explains the 1917 Danner process for mechanically drawing tubing: let a ribbon of glass entwine itself around a slowly rotating mandrel, and then pull it off. The mandrel itself is hollow, and air is blown through it so the walls don't collapse upon the hollow center. The tubing is fed by rollers to a cutting device (WBE 214). Someone still needs to figure out how the operating temperature, the rate at which the glass is supplied, the material of which the mandrel should be made, the rate of rotation and angle of inclination of the mandrel, and the rate of air flow through it.

Improved Manufacturing Methods: Plate Glass 

To make good windows, you need to be able to make plate glass. Plate glass is not merely any old flat glass. It is glass which is clear, and has flat, parallel surfaces. Otherwise, it will afford only a distorted view. Some techniques produce plate glass directly, others produce a rough-surfaced sheets which must be ground and polished to convert them into plate glass.

There were two down-time methods of making of large panes of glass. In the Crown glass ("Normandy") method, a bubble of glass was blown, cut open, and spun about. The spinning resulted in a circular pane. The glass was frequently reheated during this process, giving it a high polish (fire polish). The glass was cooled, and the workers first cut out the center (bulls' eye), and then cut out straight pieces.

In the Broadsheet ("Lorrainer") method, the glass was blown, then swung to form a long cylinder, a "sausage." The craftsmen cut off the ends, opened the cylinder lengthwise ("muffing"), placed it in a flattening oven, and polished it (Gros-Galliner, 32-33).

The first logical improvement would be the introduction, half a century earlier than in our time line, of the French method of making plate glass by grinding and polishing "table cast" glass. In 1687, Bernard Perrot "invented a method, hitherto unknown, of casting glass into panels, the way one does with metal . . ." He poured molten glass onto an iron plate covered with sand, and then rolled the glass flat. After it cooled, it was ground and polished with iron disks, abrasive sands, and felt disks. The Perrot method, as practiced in the great French manufactory of Saint Gobain, allowed production of plate glass (and flat glass mirrors) on a theretofore unheard-of scale. Saint Gobain produced the 306 mirror panes of Versailles' Hall of Mirrors, which effectively served as a permanent advertising kiosk for the French plate glass industry.

In 1903, John H. Lubbers partially automated the medieval Lorrainer method. A circular bait, at the end of a blowpipe, was dipped into a draw pot, and a cylinder of glass was drawn up. The cylinder still had to be manually cut and flattened. Nonetheless, the Lubbers technique allowed the fabrication of larger sheets of glass by less skilled workers. This method is mentioned in the 1911 EB, although that reference mysteriously remarks that during the drawing operation, the cylinder is "kept in shape by means of special devices."

By 1905, Emile Fourcault succeeded in vertically drawing a continuous sheet of glass directly from the glass furnace. The Fourcault process is also described in the 1911 EB. However, it naturally does not reveal the improved process (featuring a device called a debiteuse) which Fourcault developed in 1913, so the glass did not narrow at the base as the leading edge was drawn up (Douglas, 155). The drawn glass was still marred by rollers, and needed to be ground and polished to be suitable for optical use.

What truly revolutionized the plate glass industry was Alastair Pilkington's float glass process (1952), in which the glass spreads out over a layer of molten tin. Because the surface of the molten tin is flat, the glass also becomes flat, settling to a thickness of six millimeters.

The
Encyclopedia Britannica
favors us with a schematic diagram of the Pilkington float process, and with a few process parameters. It is certainly worth trying to duplicate once we have mastered the earlier plate glass production methods. However, it is important to recognize that it took seven years, and seven million pounds, to reduce the idea to practice, using 1950s' technology. We do have the advantage of having a description of the perfected process, but there is no doubt in my mind that the explanation leaves out important details. For example, the operator needs to control the viscosity gradient by appropriate settings of the water coolers along the process line. And some of the details it does give, such as the need for a controlled hydrogen-nitrogen atmosphere to prevent oxidation of the tin, are daunting.

The down-time state of the mirror-making art was the technique developed by Venetians Andrea and Domenico de'Anzolo del Gallo in 1507. They realized that the Venetian
cristallo
could be given a highly reflective surface by hammering tin into thin sheets, amalgamating it with mercury, and then laying the sheets of
cristallo
onto the amalgam.

We can greatly improve upon this century-old technique, dispensing with the poisonous mercury, and also obtaining a more uniform coating of controllable thickness. In 1835, Justus von Liebig discovered that silver could be deposited in a thin film on glass. There are many variations on the Liebig process, but in all of them, a solution of a silver salt is used as a source of silver ions. A reducing agent reduces the silver ions to neutral silver atoms, and the latter are deposited on the glass. In Liebig's original work, an ammoniacal solution of silver nitrate was heated with and reduced by an aldehyde (e.g., formaldehyde) to elemental silver. The process was commercialized in the 1840s, and true silvering replaced foiling.