MOLDING SUCCESSFUL MIRROR PROJECT : ARIZONA CAST TO FASHION 40,000 POUNDS OF GLASS TO USE IN TELESCOPE.

Byline: Arthur H. Rotstein Associated Press 

Beneath the University of Arizona's football stadium, a massive, spinning furnace is being tested, empty except for a giant mold inside.

The furnace is being heated nightly for the next few weeks to 1,180 degrees Celsius - more than 2,150 degrees Fahrenheit. Later this month, when cooled, it will be opened and the mold checked to make certain there are no cracks.

Workers will load it with nearly 20 tons of glass, and sometime in January it will be fired up for real.

The 40,000 pounds of glass will be cooked to make the first of twin 8.4-meter mirrors about 3-1/2 inches thick for the Large Binocular Telescope.

``It's the routine pre-fire act that we go through before every casting,'' says Roger Angel, head of the Steward Observatory Mirror Laboratory.

``Right now we're heating the mold empty . . . before we commit $700,000 of glass,'' adds John Hill, director of the $60 million telescope project and Steward's mirror-casting director.

The single-piece slabs of glass will be the key ingredients in the Large Binocular Telescope. Each will be polished to form a mirror more than 27 feet in diameter. Together, they will provide the telescope with viewing power to make it one of the world's most powerful optical instruments.

The project on 10,700-foot Mount Graham near Safford, 125 miles northeast of Tucson, finally is on track to completion after a decadelong court fight with an environmental coalition. Opponents said they feared building the telescope would lead to extinction of an endangered subspecies of red squirrel found nowhere else.

Angel came up with the idea about 15 years ago to revolutionize mirror-making by creating huge, relatively inexpensive, single-piece mirrors that are lightweight yet rigid.

The key was to melt borosilicate glass in a spinning furnace to create a glass blank with hollow backing and a steep focal ratio. In other words, part of the structure of the mirror is built the same way as a bee's honeycomb - partly hollow but still strong and lightweight. The focal ratio is considered critical because it holds expenses down by reducing the height of the building needed to house the telescope.

No one has cast as large a blank of glass conventionally, Hill said. One of comparable thickness and focal ratio would weigh five times more - 100 tons.

When the furnace is turned on for real, he said, it will be kept at its maximum temperature for about 11 hours.

``That's the time for the glass to melt and fill the mold,'' Angel said. The furnace then will spin for three days to let the glass solidify, followed by up to four months of slow cooling so that it tempers and hardens without becoming brittle. That process is called annealing.

``It may be as slow as 5 degrees a day,'' said Angel, who received a five-year, $330,000 grant last spring from the John D. and Catherine T. MacArthur Foundation in recognition of his creativity.

Two periods in the mirror production will be critical. The first is when the temperature is cranked all the way up, he said.

``The period of maximum danger, I would guess, is while the glass is liquid, because it exerts strong buoyant forces on the mold,'' Angel said.

``It's trying to lift up, and there are significant forces trying to tear the mold apart. We think we've made it strong enough. That's the acid test.''

The other will be at temperatures of about 500 degrees Celsius to about 450 degrees, during the annealing, Angel said.

The mirror lab has produced glass blanks 3.5 meters in diameter and two of 6.5 meters, and each time the size has increased, new things have been learned, Angel said.

Angel hopes there won't be too many new lessons this time.

``We try to fix everything we can anticipate,'' he said. ``So the new lesson will be things we didn't anticipate.''

After the glass is cast, it will take about two years to polish the first mirror by computer - once smaller ones already in queue are finished.

The cost for each big mirror, when cast and polished, will be about $12 million, Hill said.

``This probably won't go to the mountain until the fall of 2000 or something like that,'' Hill said.

CAPTION(S):

Photo

Photo: University of Arizona's Dean Ketelsen checks out a finished mirror before grinding and finishing in the mirror laboratory in Tucson, Ariz.

Associated Press