FBI photos of the material in the letter sent to Senator Pat Leahy (left) and to the New York Post (right), from the report.

A report from McClatchy provides important new evidence and analysis in the FBI’s Amerithrax investigation of the 2001 anthrax attacks. The report shows that the FBI ignored as potentially erroneous a measurement of silicon in one anthrax sample and then hid this information from Congressman Jerrold Nadler (D-NY). Even more importantly, the high silicon measurements in at least two samples also were coupled with high tin measurements, opening up the possibility that silicon was added to the attack material in a form that is not mentioned in any of the FBI documents. Significantly, it is virtually impossible that Bruce Ivins, whom the FBI has concluded acted on his own to carry out the attacks, would have been able to perform the necessary chemical manipulations involved in this treatment of the spores.  Ivins likely also would not have had access to the necessary laboratory equipment to perform this treatment.

The presence of silicon and how it may have gotten into the anthrax material has been a point of great controversy throughout the entire investigation. This question is important because the chemical nature of the silicon and the level at which it is present is presumed to be an indicator of whether the anthrax spores have been “weaponized” to make them suspend more readily in air so that they are more effective in getting into the small passageways of the lungs of the intended targets of the attack. Early in the investigation, Brian Ross published “leaked” information that the spores had been weaponized through addition of bentonite and that Iraq had a weaponization program that used bentonite. This report turned out to be false, as no evidence for bentonite has been found. A more sophisticated type of weaponizing would rely on mixing the spores with nanoparticles of silica (silica is the common name for the compound silicon dioxide) to make them disperse more easily.

The FBI carried out a special form electron microscopy that could identify the location of the silicon in the spores from the attack material. They found that the silicon was in a structure called the the spore coat, which is inside the most outer covering of the spore called the exosporium. If silica nanoparticles had been used to disperse the spores, these would have been found on the outside of the exosporuim (see this diary for a discussion of this point and quotes from the scientific literature) because they are too large to penetrate it.  No silicon signature was seen on the outside edge of the exosporium.  What is significant about the type of silicon treatment suggested in the McClatchy piece is that both high silicon and high tin measurements were found in several samples and that there is an alternative silicon treatment that would involve a tin-catalyzed polymerization of silicon-containing precursor molecules. McClatchy interviewed scientists who work with this process and they confirmed that the ratio of silicon to tin found by the FBI is in the range one would expect if such a polymerization process had been used.

What McClatchy doesn’t mention in their report is that it would seem for a polymerization process of this sort, the silicon-containing precursor molecules would be small enough to penetrate the exosporium before being polymerized, or linked together into much larger molecules, once they reached the spore coat. This would mimic the location of silicon incorporated “naturally” into spores.

I had previously suggested that the amount of silicon and its location in the spore coat in the attack material most likely indicated that the spores had been cultured in the presence of an antifoam agent more commonly used in large bacterial fermenters and that this tended to exclude Ivins as a suspect since he did not have access to a fermenter.  This analysis, however, was carried out in the absence of the key finding that the New York Post anthrax powder contained up to 10% silicon.  This level approaches a non-biological level, as the highest silicon content I have seen in the literature is 6.3% for the anthrax relative Bacillus cereus.

The National Academy of Sciences panel reviewing the FBI data did have the 10% number.  It appears in this table:

As McClatchy reports, the FBI hid this high silicon reading from Congressman Nadler when he asked about it in 2008:

The FBI guarded its laboratory’s finding of 10.8 percent silicon in the Post letter for years. New York Democratic Rep. Jerrold Nadler asked FBI Director Robert Mueller how much silicon was in the Post and Leahy letters at a hearing before the House Judiciary Committee in September 2008. The Justice Department responded seven months later that silicon made up 1.4 percent of the Leahy powder (without disclosing the 1.8 percent reading) and that “a reliable quantitative measurement was not possible” for the Post letter.

McClatchy also points out the significance of the finding of silicon and tin together:

The lab data, contained in more than 9,000 pages of files that emerged a year after the Justice Department closed its inquiry and condemned the late Army microbiologist Bruce Ivins as the perpetrator, shows unusual levels of silicon and tin in anthrax powder from two of the five letters.


The existence of the silicon-tin chemical signature offered investigators the possibility of tracing purchases of the more than 100 such chemical products available before the attacks, which might have produced hard evidence against Ivins or led the agency to the real culprit.

Finally, chemiists who work with the tin-catalyzed silicon polymers find it unlikely a biologist such as Ivins would be able to carry out this chemistry:

Several scientists and former colleagues of Ivins argue that he was a career biologist who probably lacked the chemistry knowledge and skills to concoct a silicon-based additive.

“There’s no way that an individual scientist can invent a new way of making anthrax using silicon and tin,” said Stuart Jacobsen, a Texas-based analytical chemist for an electronics company who’s closely studied the FBI lab results. “It requires an institutional effort to do this, such as at a military lab.”

As for the FBI, McClatchy reports that their opinion on these unresolved questions is that “it’s all a moot point, because they’re positive they got the right man in Ivins.” Such is the state of justice in the United States in 2011.

Clearly, more scientific investigation is warranted on the silicon question now that the possibility of tin-catalyzed silicon polymerization (see this pdf for some of the chemistry involved here) has entered the picture. Some of the currently conflicting observations possibly could be resolved if it turns out that the attack material was weaponized with this process. At any rate, the coupling of tin with silicon eliminates the most likely sources of silicon that could have been exogenously added by Ivins, a common treatment of glassware to prevent biological material sticking to it or even antifoam being added to shake flasks (see this diary where the analysis of a typical antifoam shows less than five parts per million of heavy metals like tin).