A significant number of scientists and biological warfare experts are expressing skepticism about the FBI's view that a single disgruntled American scientist prepared the spores and mailed the deadly anthrax letters that killed five people last year.
These sources say that making a weaponized aerosol of such sophistication and virulence would require scientific knowledge, technical competence, access to expensive equipment and safety know-how that are probably beyond the capabilities of a lone individual.
As a result, a consensus has emerged in recent months among experts familiar with the technology needed to turn anthrax spores into the deadly aerosol that was sent to Sens. Thomas A. Daschle (D-S.D.) and Patrick J. Leahy (D-Vt.) that some of the fundamental assumptions driving the FBI's investigation may be flawed.
"In my opinion, there are maybe four or five people in the whole country who might be able to make this stuff, and I'm one of them," said Richard O. Spertzel, chief biological inspector for the U.N. Special Commission from 1994 to 1998. "And even with a good lab and staff to help run it, it might take me a year to come up with a product as good."
Instead, suggested Spertzel and more than a dozen experts interviewed by The Washington Post in recent weeks, investigators might want to reexamine the possibility of state-sponsored terrorism, or try to determine whether weaponized spores may have been stolen by the attacker from an existing, but secret, biodefense program or perhaps given to the attacker by an accomplice.
The Defense Department and FBI refused repeated requests from The Post to discuss recent developments in the anthrax investigation. But in some important respects, the official version of events -- developed in part during the early, frantic days of the probe -- is at odds with the available evidence, the experts say.
A profile of the attacker issued by the FBI last November described an angry, "lone individual" with "some" science background who could weaponize the anthrax spores in a basement laboratory for as little as $2,500. The FBI acknowledged that the sender may not have been a native English speaker but emphasized that there was no "direct or clear" link between the attacks and foreign terrorism.
More recently, investigators appear to have abandoned the idea of an amateur attacker, but they continue to focus on a lone, domestic scientist, probably an insider. Attention has centered on medical doctor and virologist Steven J. Hatfill, a former U.S. Army scientist identified by the Justice Department as a "person of interest" in the investigation. Hatfill vigorously denies any involvement.
Scientists suggested that the loner theory appeared flawed even in the opening days of the investigation. The profile was issued three weeks after U.S. Army scientists had examined the Daschle spores and found them to be 1.5 to 3 microns in size and processed to a grade of 1 trillion spores per gram -- 50 times finer than anything produced by the now-defunct U.S. bioweapons program and 10 times finer than the finest known grade of Soviet anthrax spores. A micron is a millionth of a meter.
"Just collecting this stuff is a trick," said Steven A. Lancos, executive vice president of Niro Inc., one of the leading manufacturers of spray dryers, viewed by several sources as the likeliest tool needed to weaponize the anthrax bacteria. "Even on a small scale, you still need containment. If you're going to do it right, it could cost millions of dollars."
Possible Foreign Source
Also early in the case, U.S. authorities dismissed the possibility that Iraq could have sponsored the attacks because investigators determined that the spores had been coated with silica to make them disperse quickly, rather than the mineral bentonite, regarded by the U.S. Army Medical Research and Materiel Command as Iraq's additive of choice.
However, Iraq's alleged preference for bentonite appears to be based on a single sample of a common pesticide collected by U.N. authorities from Iraq's Al Hakam biological weapons facility in the mid-1990s. By contrast, the U.S. Defense Intelligence Agency warned in declassified documents as early as 1989 that Iraq was acquiring silica to use as a chemical weapons additive.
In 1998, Iraq reported to the United Nations that it had conducted an artillery test of a live biological agent that used silica as a dispersant. And U.N. and U.S. intelligence documents reviewed by The Post show that Iraq had bought all the essential equipment and ingredients needed to weaponize anthrax bacteria with silica to a grade consistent with the Daschle and Leahy letters.
Daschle, Leahy and a few other senators and representatives have received periodic FBI briefings on the investigation, and Leahy said last week that the agency "has not foreclosed the possibility of a foreign source of this attack." However, the FBI's continued focus on Hatfill shows the agency's preoccupation with a domestic loner.
Bush administration officials have acknowledged that the anthrax attacks were an important motivator in the U.S. decision to confront Iraq, and several senior administration officials say today that they still strongly suspect a foreign source -- perhaps Iraq -- even though no one has publicly said so.
That Iraq had the wherewithal to make the anthrax letters does not mean it is the guilty party. Still, the FBI's early dismissal of the possibility may have prematurely closed a legitimate line of inquiry.
"Iraq almost certainly had their anthrax spores in a powdered form," Spertzel said. "They had used silica gel to aid in dispersibility of [wheat] smut spores, and also indicated they were looking at it as a carrier for aflatoxin," a carcinogen.
Outer Limits of Technology
Since the attacks one year ago, scientists have been able to identify the anthrax bacteria used in the Daschle and Leahy letters as the "Ames strain," a virulent anthrax used in U.S. biodefense programs.
Analysts are examining lab variants of the Ames strain to find possible sources for the original spores, but scientists and biowarfare experts say the additive used to disperse the spores may be as instructive as the spores themselves.
Even the sparse evidence made public by the investigation -- the uniformly tiny particle size and the trillion-spore-per-gram concentration -- has been enough to show many researchers that whoever weaponized the spores was operating at the outer limits of known aerosol technology. The mailer was brutally efficient in making a very special product for a very special mission.
The anthrax mailer needed a powder that could negotiate the U.S. postal system without absorbing so much moisture that it would cake up. At the end of the trip, the coated spores had to be light and supple enough to fly into the air with no delivery system beyond the rip of a letter opener through an envelope.
Finally, the spores had to be small enough for potential victims to inhale them deep into their lungs so that only a tiny number of spores would be needed to kill -- far fewer than the dosages anticipated by the U.S. government for the cruder aerosols of the past.
The answer was silica -- the same silicon dioxide that comprises substances ranging from beach sand to window glass. The attacker needed a special kind of silica, however, because the aerosol that delivered the spores was as sophisticated as any on the market.
"You need to get a drug into the bloodstream as an alternative to injecting it," said pharmaceutical scientist Richard Dalby of the University of Maryland's Aerosol Lab. "You need the drug to get much deeper into the lung, where the membranes are thinner, and to do that, you need smaller particles."
The pharmaceutical industry is the leader in this technology, Dalby added, but "there's only been an interest in generating tiny particles for that purpose for about the last 10 years."
Several sources agreed that the most likely way to build the coated spores would be to use the fine glass particles, known generically as "fumed silica" or "solid smoke," and mix them with the spores in a spray dryer. "I know of no other technique that might give you that finished product," Spertzel said.
According to William C. Patrick III, the former chief of product development for the U.S. Army's now-defunct bioweapons program, U.S. government scientists made biological agents using spray dryers, but did not spray dry anthrax.
Fumed silica grains are between 0.012 and 0.300 of a micron in size, and will readily adhere to the surface of any larger particle, such as an anthrax spore. Coated particles will easily disperse, because the grains act as tiny "ball bearings," enabling the larger bits to skid past one another.
Under an electron microscope, fumed silica would look like cotton balls strung together into strands that branch out in every direction. Their extremely small size gives them an aerodynamic quality, and their high surface area allows them to readily trap moisture, acting as a natural dessicant.
"If you packaged this stuff in a container, it would float out, and it's highly dispersible and messy to deal with," said C. Jeffrey Brinker, a University of New Mexico chemical engineer and a senior scientist at the Sandia National Laboratories.
Moreover, Brinker added, simply by shaking the particles in a jar, they acquire an electric charge, which causes them to repel one another and not clump together. A few passes through a mail-sorting machine would create the same effect. The particles would float, but they would remain separated.
"This concept of using something that would serve as a dessicant and a carrier at the same time is new," said Harvard University chemical engineer David Edwards. "It's a diabolically brilliant idea."
Fumed silica has myriad uses, mostly as a thickening agent in products including ceramics, house paint, toothpaste and cosmetics. It is not widely known as an aerosol additive.
"If you're going to put it into the lung, there has to be a mechanism to clear it, otherwise you just fill up somebody's lung with silica after repeated dosings," said Dalby, of the Aerosol Lab. The anthrax mailer, he noted, obviously wasn't worried about giving his victims silicosis.
Some fumed silicas are extremely difficult to make, but at least two -- Aerosil and Cab-O-Sil -- are readily available and sold commercially in bulk. Either product, in theory, could be used to coat anthrax spores. Aerosil is based in Germany and Cab-O-Sil, in Boston. Both firms have offices around the world.
Ken Alibek, a former deputy director of the Soviet bioweapons program now running an Alexandria biotechnology firm, said the Soviets used Aerosil in agent powders, and a classified Defense Department memo in 1991 said Iraq had "imported approximately 100 MT [metric tons] of Aerosil during the last 8-9 years." Spertzel said the United Nations reported in the 1990s that Iraq had 10 metric tons of Cab-O-Sil, probably destined for its chemical weapons program.
The United Nations also documented the presence of three Niro Inc. spray dryers in Iraq in the 1990s. Spertzel said two were destroyed, and the third was scoured and sterilized before inspectors could examine it.
In spray drying, a technician mixes fumed silica and spores with water, then sprays the mist through a nozzle directly into a stream of superheated air shooting from a second nozzle into an enclosed chamber. The water evaporates instantly, leaving spores and additive floating in space.
"Surface tension will pull those little [silica] particles together onto the big one," said California Institute of Technology chemical engineer Richard Flagan. "You will end up with some degree of coating."
Whoever made such an aerosol would "need some experience" with aerosols and "would have to have a lot of anthrax, so you could practice," Edwards said. "You'd have to do a lot of trial and error to get the particles you wanted." It would also help to have an electron microscope to examine the results.
This would mean at least several hundred thousand dollars worth of equipment, several experts said. Niro's cheapest spray dryer sells for about $50,000. Electron microscopes cost hundreds of thousands of dollars.
In all, said Niro's Lancos, "you would need [a] chemist who is familiar with colloidal [fumed] silica, and a material science person to put it all together, and then some mechanical engineers to make this work . . . probably some containment people, if you don't want to kill anybody. You need half a dozen, I think, really smart people."
One way to assemble such a team would be with "the knowing complicity of the government of the state in which it [the agent] is made," Spertzel said. Another way to acquire the agent, several sources acknowledged, would be to steal it from a biodefense program that uses live biological agents for research or training purposes.
The Biological and Toxin Weapons Convention of 1972 bans offensive biowarfare research, but it clearly allows signatory nations to undertake biodefense programs using small quantities of live agents.
The Daschle and Leahy letters each contained 1.5 grams of anthrax powder or less, well within the boundaries of what researchers describe as "laboratory quantities" of agent. It is impossible to account publicly for all the anthrax powder that may exist in the United States, because most of the defense projects that use it are classified.
The Post asked the Defense Department whether the U.S. armed forces have made any anthrax powder comparable to that which was mailed to the Senate. The department declined to comment, citing the ongoing anthrax investigation.
There is, however, no public evidence that the Army has used spray-dried agents in recent biodefense projects, choosing instead to test small amounts of irradiated -- and therefore nonlethal -- anthrax bacteria that had been dried with older technologies.
In a written response to questions about the U.S. interpretation of the weapons convention, the Defense Department said its personnel may use live biological agents in a number of research settings: for vaccines and treatment; protective clothing and containment; alarms and detection; and decontamination.
The department "does not set quantitative thresholds for the agents or toxins in its possession," but "these quantities are generally small," the response said. "DOD continues to evaluate its procedures to ensure dangerous materials are safely stored and properly disposed of when no longer required."