Hawley’s Condensed Chemical Dictionary says that diethyl zinc is a “high-energy aircraft and missile fuel”07—but missiles were just the beginning. U.S. troops have used DEZ to create firewalls, according to an Army scientist at Fort Belvoir, Maryland, by the name of Divyakant Patel. “During the war,”08 Patel told me (he didn’t make clear which war) “when they want to separate out one section from another section, they can throw this vapor in the air, just like creating a screen, a big screen of fire.” Patel has himself recently experimented with diethyl zinc: using a hypodermic syringe, he filled bullets with the chemical and shot them at land mines. A few drops of DEZ, diluted in toluene, is more than enough to set TNT on fire, and the bullet hole releases the combustion pressure so that the mine doesn’t explode—or at least not quite so violently.
Dr. Allen Tulis, current chairman of the International Pyrotechnics Seminar (where investigators in the fields of combustion, explosion, and flame propagation gather every year to share their research), has worked with diethyl zinc on and off for decades and knows its behavior as well as anyone. In the sixties, as a researcher at the Illinois Institute of Technology Research Institute, a not-for-profit contractor that does weapons testing and development for the Department of Defense, Tulis found diethyl zinc’s properties useful in his “encapsulated flamethrower,” which was later weaponized, to use the military’s verb, for the U.S. Army. Previous to Tulis’s work, soldiers operating flamethrowers didn’t survive very long in battle, because the plumes of fire revealed their whereabouts and they were quickly shot. An encapsulated flamethrower,09 so I gather, launched frangible capsules containing diethyl zinc. When they hit their target and broke, the capsules created the sort of deflagrational mayhem that was produced by a standard flamethrower, but without betraying the thrower’s location.
Diethyl zinc is one of a class of tricky organometallic compounds called metal alkyls; it grabs any available oxygen atoms, including those in cellulose and in human tissue, and uses them to create fire. Tulis recalls one time when some droplets of diethyl zinc blew out of a prototype flamethrower capsule and splattered onto his lab notebook, scorching the pages. “It’s very dangerous,” Dr. Tulis says, “because if you spill a few drops on your body somewhere, it will eat right into the flesh, and you can’t really stop it. It continues to react with moisture and your flesh as it eats into your body.” Chemists have on occasion been badly burned—as have enemy troops, presumably.
In the seventies, bomb designers also saw possibilities in DEZ, as they searched for a suitable “initiator” to employ in a fancy new kind of fuel-air weapon. The old fashioned fuel-air bombs, used in Vietnam and much later in Iraq, blew out a cloud of fuel into the air above a target and then, using a second, delayed charge (or a cluster of charges), lit the cloud. The result was a doughnut-shaped firestorm accompanied by a high-pressure shockwave that could be substantially more destructive than that created by a conventional bomb—an “overpressure” closer, in fact, to that produced by a small nuclear detonation. Fuel-air bombs were used to clear landing zones for helicopters (they could blow a stand of trees flat), to clear minefields by triggering all the mines, and simply to kill or stun people in quantity. Even for those who escaped the worst effects of the blast, the shock wave’s implosive undertow could rupture eardums,10 collapse lungs, and cause a bubbling in the blood similar to a deep-sea diver’s case of the bends.
The military wasn’t completely satisfied with these fuel-air bombs, however. In 1979, the Air Force’s Office of Scientific Research funded a workshop at McGill University in Montreal in which scientists from Atlantic Research Corporation and elsewhere discussed possible ingredients for a “FAE III” (that is, a third-generation Fuel-Air Explosive bomb)—a weapon whose fuel cloud would be lit as it expanded not by small secondary explosives, but by its own voraciously combustive chemistry.11 Dr. John Lee,13 a scientist at McGill, experimented with a number of compounds, including triethyl aluminum and diethyl zinc, as potential initiators and shock-wave amplifiers for this new bomb. Lee recalls an unsettling incident in his laboratory. A holding tank was supposedly empty, but in fact it still contained a tiny amount of diethyl zinc. “We thought it was completely, completely gone,” Lee told me; but when air was allowed to enter the test chamber where the tank sat, there was a “small explosion.” Diethyl zinc is, Lee said, “really wicked stuff.” If he heard that someone was playing around with hundreds of pounds of it, “it would scare the hell out of me.”
CHAPTER 13
* * *
Getting the Champagne out of the Bottle
The Library of Congress was playing around with hundreds of pounds of it. Their aim was to build a processing facility that would deacidify a million books per year: to do that, they would need about forty-five thousand pounds of their fractious agent—three percent of the weight of the books, assuming that an average book weighs a pound and a half. Every five-thousand-book run would require between two and three hundred pounds of vaporized diethyl zinc: the word “sobering” comes to mind. (Compare this to the manufacture of plastic or rubber, where to make, say, twenty tons of your final product you might need a single pound of diethyl zinc intermingled with other catalysts and a non-pyrophoric solvent such as hexane.) The scientists’ task was made more complicated by the fact that DEZ, as it reacts with water, produces quantities of ethane, which is flammable. I described the deacidification process to Allen Tulis. “Ethane is a very good fuel,” Tulis said. “Now the problem I see there is: How do you discharge this material [the diethyl zinc and ethane] and properly neutralize it? Because you’ve got the makings of a fuel-air in there. If the vacuum would be breached and air would blow into the system, you’d have a humongous explosion.” The library had in effect designed a large fuel-air bomb that happened to contain books.
The strangeness of the idea seems not to have troubled the scientists’ sleep, however. There were two principal developers of the DEZ process: Dr. John Williams, the library’s head of the Research and Testing Department, who had designed water filters and had made an improved inflatable rubber bag during the war (a bag used for gluing helmet liners into Army helmets); and a quiet, deliberate man named George Kelly, Jr., whom Williams hired in 1971, with the help of a grant from the Council01 on Library Resources.
Kelly had a B.A. in chemistry from the University of Maryland; during World War II, he taught recruits at the Army’s Chemical Corps School how to fire 4.2-inch chemical mortars. He did some research on floor tiles and had an unhappy time at a pesticide company,02 and then moved to Westvaco, where he tested the methods of refining trona to make bicarbonate of soda (trona is a kind of rock mined deep below Wyoming) and was part of a team that made a couple of hundred thousand pounds of a rocket fuel called hydrazine. In the sixties, Kelly moved to Union Carbide, where he experimented with water-soluble polymers as paper coatings. “I’m pretty much a problem solver,” Kelly said. After Union Carbide canceled his project, Kelly wrote John Williams and asked if there was a job for him at the Library of Congress.
Williams liked Kelly’s work on paper coatings, and assigned him the job of testing potential agents for deacidification. Kelly was willing—in fact, he did so much work with a pungent group of compounds called amines that his sense of smell was destroyed. After a few years, he and Williams together narrowed their search down to diethyl zinc, which they ordered in quart bottles from Texas Alkyls. With a hypodermic syringe, Kelly sucked some DEZ from the bottle and quickly plunged the needle of the syringe into a cork; to start a test run, he stabbed the needle through a rubber gasket in the crypto-pressure-cooker that held five or six test books and gave them a shot. The library’s safety officer was not happy with these experiments, according to Kelly. “You’re going to burn our library down!” the officer would say. “No, I’m not,” Kelly would answer. Recalling an early visit from Kelly, Scott Eidt, the chemist from Texas Alkyls, told me: “We made a joke of it, because of what he was going to do. We’d call it ‘book bu
rning.’ ”
Williams and Kelly, both intelligent applied scientists with no practical knowledge of traditional book conservation, got their first diethyl-zinc patent in 1976. Kelly convinced a somewhat doubtful manager at General Electric’s aerospace unit to let them use a vacuum chamber in Valley Forge, Pennsylvania, for some early tests. Each test, performed on lots of four hundred books, consumed between thirty and seventy pounds of liquid DEZ.03 The testing proceeded without any scares, except once when Texas Alkyls shipped the wrong colorless liquid: scientists unknowingly filled the vacuum chamber with it, leaving the books “thoroughly acidified.”04
There were problems with the process, though—book covers decorated with rainbow-colored auras of zinc-oxide residues, binding adhesives weakened, papers darkened, scorched, or bad smelling—and the penetration of the acid-neutralizing vapor was incomplete in many cases. Treated books also proved to be more sensitive to light damage than untreated papers. General Electric was lukewarm05 about the project, and it and the library parted ways in the late seventies—which was just as well, since the GE chamber had “small air leaks”06 anyway. But by 1980, an optimistic Kelly announced that the process was “now ready for commercial use.” He warned, however, that diethyl zinc was “extremely hazardous” and must be used only in an industrial setting by trained personnel. But, he said, “we have demonstrated07 it can be used successfully and safely to treat books.” Neither Kelly nor anyone else at the library alluded to the missile-propulsive and incendiary incarnations of their chosen chemical: it was described as a polymer catalyst, tout court. “We deliberately stayed away from so-called military uses,” one former Library of Congress scientist told me.
Kelly and Williams retired from the library—Kelly went back to consult for his old company, FMC (which had bought Westvaco), and Williams took a job working on artillery propellants for Armtec, a defense contractor in California. Ideally, that would have been the end of diethyl-zinc deacidification forever. But the process found a new and ardent friend in Peter Sparks, the aptly named head of the library’s Preservation Directorate, who arrived in 1981.
Sparks took up the organometallic cause, began feeding quoteworthy quarter-truths to the press, and succeeded in extracting millions more in deacidificational funding from Congress. He and the librarian of Congress, Daniel Boorstin, liked big, round numbers; Sparks told a writer for the Associated Press that a diethyl-zinc treatment would add “400 to 600 years”08 to the life of a book. He showed off a guillotined copy of Thomas Carlyle’s Sartor Resartus to reporters and claimed that seventy-seven thousand books a year go brittle at the Library of Congress. (The Library of Congress had fun with Carlyle, microfilming eight separate old editions of Sartor Resartus, three editions of his Life of Schiller, Lecky’s edition of his French Revolution, and a shelfload more.) The New York Times believed Sparks in 1984 (why shouldn’t they?): “The pages of at least five million volumes09 now crumble at the touch of a finger, simply because daily exposure to the elements has turned them brittle. Every year 77,000 more volumes are affected.” Sparks was “a very good marketing person,” Chandru Shahani told me. “He was, I think, too good a salesperson, so he sold this with a lot of promises when the process wasn’t ready. He hadn’t really anticipated the problems.” Daniel Boorstin authoritatively backed up Sparks’s salesmanship, too: the same New York Times article quoted Boorstin as saying that the diethyl-zinc process was “one of the most important steps toward the preservation of knowledge around the world. . . . It would be hard to overstate its significance.”
As for safety, not to worry, the “handling of diethyl zinc10 by trained operators has been reduced to a routine matter,” according to the library’s Information Bulletin. Daniel Boorstin wrote a letter to Congress assuring them that there were “no known safety risks11 to personnel or books with the Library of Congress’ process.”
Once funding was secured, an idle vacuum chamber at NASA’s Goddard Space Flight Center, conveniently located in suburban Washington, D.C., became the center of testing operations. In 1982, the library (through NASA) hired Northrup Services to draw up plans and operating procedures. Several Northrup engineers went down to Texas Alkyls for an eyebrow-raising diethyl-zinc briefing. One of the engineers began to feel that the Library of Congress had not been forthcoming in its description of the hazards of the process. “Basically what you had was a huge combustion chamber,” the engineer told me. “It became patently clear to me that we were heading for trouble, because we were modifying a system that was not designed for chemical processes.” NASA’s vacuum chambers were built to simulate conditions in outer space, after all, and the engineers at Northrup Services were thermal-vacuum scientists, with no experience in the design of prototype chemical plants.
Northrup needed the work, according to the engineer; they therefore played down some of the difficulties and hazards in their pro forma presentations to NASA. “What we were told to say to [NASA’s] safety committee versus what was really going on were a little divergent,” the engineer says. And there were too many managers. The project was paid for by the Library of Congress, constructed in NASA’s buildings, employing Northrup: nobody was really in charge, and nobody really knew what they were doing. The detailed chemistry of diethyl zinc’s reactions with most of the hundreds of filler substances in old paper were undocumented. The library had never published detailed descriptions, in peer-reviewed journals, of the evolving permutations of their invasive treatment and its often discouraging experimental results. The Library of Congress planned to do something radical and irreversible to millions of its books, and yet, rather than inviting outside comment and help, they were behaving like weapons procurers12 at the Department of Defense.
The first NASA run, in November 1982, was the most ambitious one—after months of preparation, the testers loaded five thousand ex-library books into the chamber, the number they planned to treat each time when they were at full production volume. The process wasn’t particularly kind to the books, as it took almost two weeks and involved extremes of heat and cold. First, they heated the books to 113 degrees13 Fahrenheit for a few days in order to drive off most of their moisture; next, they pulled a vacuum, slowly, which caused the temperature to drop. When they were sure the air was gone, they began fogging the chamber with the necessary two to three hundred pounds of their chemical. As the reactions progressed, the chamber grew first warm, then quite hot, although it was supposedly kept “well below” 212 degrees. (It wouldn’t be at all good if the temperature rose above 250 degrees because diethyl zinc begins to decompose above that point; at pot-roast temperatures, the decomposition becomes “self-sustaining and uncontrollable,”14 and, depending on conditions and quantities, you could have a thermal explosion.)
Four thousand of the books were discards, a thousand were from the library’s shelves. For six days, in this iron lung, they underwent their hot chemotherapy. Then the technicians created another vacuum, and the temperature dropped again—but slowly: it would be bad if the books froze, because the DEZ vapor might condense on them and remain to cause trouble upon repressurization. In an effort to prolong these books’ lives, the scientists were subjecting them to conditions that mimicked a rigorous accelerated-aging experiment, minus the air—no wonder the paper sometimes tested weaker immediately after treatment.
The five-thousand-book run was not a success. (The results were “mixed,”15 according to Congress’s Office of Technology Assessment.) Analysis determined that a cloud of ethane gas had taken up position at the top of the chamber, while a cloud of heavier diethyl zinc pooled at the bottom; because the vapor was in a vacuum, there was little convectional mixing. Thus many of the stacked books16 got no deacidificational benefit, while some got too much. There were “tide marks,” darkened paper, and distasteful odors. (“Cause of odor a mystery17 since known chemistry cannot explain it,” noted the still-baffled scientists years later.) Also the library couldn’t possibly meet its stated goal of deacidifying a million book
s a year if each five-thousand-book run had a cycle time of two weeks—things had to move along much quicker than that. Hoping to work out the problems, the scientists made a number of much smaller tests in another NASA vacuum chamber, and then, in the spring of 1985, with funds running low and many uncertainties dangling, the library decided to abandon the prototype plant altogether and build still another, smaller pilot facility, in a different NASA building, with a rectilinear chamber this time instead of a round one, so that the DEZ gas wouldn’t idle in the waste space, and with faster pumps, so that the gas cloud would swirl turbulently around the books, reacting with all of them. There was an air of desperate haste at this stage—a NASA electrician later reported that in November 1985, “a Library of Congress representative18 tried to talk [name whited out] into running DEZ into the chamber when there were system leaks and no procedures.” The electrician stopped them by “explaining the consequences of those size leaks, but did not feel he should have been the one to do this.”
The hurry had to do with money: the library had spent several million on the NASA project so far (and the NASA project followed the GE project, which hadn’t been cheap), and Peter Sparks and William Welsh (the deputy librarian of Congress) needed to get beyond feasibility studies and tap into the $11.5 million that Congress had budgeted for a bona fide high-volume treatment plant, to be built at Fort Detrick, which would possibly handle books for other libraries as well. Later, Welsh admitted19 that the library had diverted $1.7 million from the library’s personnel accounts into the diethyl-zinc fund without congressional permission. (“The time drivers20 on this project [were] the limited funding, and the Library of Congress pushing to get the job done,” said one anonymous respondent in an official NASA report. “There were overruns, and budget problems in October 1985 that caused the Library of Congress to pressure Northrup on this project.”)