The FBI Laboratory: An Investigation into Laboratory Practices and Alleged Misconduct in Explosives-Related and Other Cases (April,1997) Office of the Inspector General, Department of Justice

In 1988, Yu Kikumura, a member of the Japanese Red Army (JRA) terrorist organization, was found with three homemade bombs in a car in a service area of the New Jersey Turnpike. Kikumura was later indicted on several counts of interstate transportation of explosive devices and passport violations. After a bench trial on stipulated facts, Kikumura was convicted on November 29, 1988. He is currently serving a sentence of 262 months.

Kikumura's first sentencing hearing was held on February 7, 1989. At that hearing, the government offered testimony by J. Thomas Thurman, who was then an examiner in the Explosives Unit (EU). As the principal examiner in this case, Thurman had prepared a Laboratory report, dated June 15, 1988, concerning the bombs and other evidence removed from Kikumura's car.

In a letter to the OIG dated February 17, 1996, Whitehurst alleges that Thurman lied on the stand about examinations done by the Laboratory, violated FBI procedures or protocol by testifying outside his expertise, misled the jury, and incorrectly suggested that the defendant intended to make a large and powerful bomb from ammonium nitrate, aluminum powder, and mercury fulminate.

We reviewed the Laboratory report prepared by Thurman and the related dictation and work papers of other examiners. We also reviewed transcripts from the hearings before the district court related to Kikumura's conviction and sentencing and the published decisions by the United States District Court and the United States Court of Appeals for the Third Circuit. Finally, we interviewed Thurman and Terry Rudolph concerning their work on the case.

We find no basis for the allegations that Thurman testified falsely or violated FBI policies in this case. In some areas, Thurman's testimony contains ambiguities or minor inaccuracies. We do not believe that his remarks in these areas reflect knowing and intentional misconduct. As with other cases we have reviewed, we think this case illustrates the desirability of clearer guidelines for, and effective monitoring of, examiner testimony. The case is also another example where the Laboratory would have benefitted from more rigorous policies for case work documentation, file review, and record retention.

On April 12, 1988, a New Jersey state trooper detained Kikumura after stopping him for a motor vehicle violation. The officer noticed several gunpowder containers and lead shot in a bag on the backseat of Kikumura's car. He also saw a cardboard box containing three red cylinders with black tape and wires on them. Kikumura invited the officer to examine these items. The officer concluded that they might be bombs and arrested Kikumura.

The three red cylinders and other evidence from Kikumura's car were sent to the FBI Laboratory for examination. As the principal examiner, Thurman prepared a report dated June 15, 1988. This report incorporated the work of many other examiners, including Lynn Lasswell in the Chemistry-Toxicology Unit (CTU) and Terry Rudolph in the Materials Analysis Unit (MAU). Among other things, Lasswell confirmed that powder from the red cylinders was a mix of six identifiable types of smokeless powders and one unidentified smokeless powder. Rudolph determined that three pea-sized objects found in a paper bag in the car were prills of ammonium nitrate.

The red cylinders were fire extinguishers that had been emptied and refilled with about three pounds of gunpowder, wadding, about three pounds of lead shot, and a flashbulb connected to some wires running out of the top. On one of the bombs, there was an assembled fusing system made from an electric timer, a toggle switch, some batteries, and jack connectors. This timer, Thurman concluded, would allow the bomb to be detonated up to an hour after it was set. The car also contained materials from which similar fusing systems for the other two bombs could be made.

On October 21, 1988, the district court denied Kikumura's motion to suppress the evidence discovered in his car. On the scheduled trial date of November 28, 1988, Kikumura proposed through his counsel that the parties stipulate to certain facts, waive trial by jury, and have a bench trial on the stipulated facts. For purposes of the bench trial, the parties stipulated that Kikumura had transported the explosives with knowledge that they could be used to damage or destroy property. Kikumura agreed that the government would be free to offer whatever evidence it deemed appropriate at a later sentencing hearing, including evidence of his intent to kill.

On November 29, 1988, the district court convicted Kikumura on all counts based on the stipulated facts. A sentencing hearing was held on February 7, 1989. The government offered two witnesses at this hearing: New Jersey Detective Joseph Fuentes and Thurman. Fuentes described how Kikumura had entered the United States on March 8, 1988, by using a stolen and altered passport. Over the next 30 days, Kikumura traveled some 7,000 miles through at least seventeen different states. During this journey, he bought at various places components of the bombs and other items found in his car. Fuentes described evidence suggesting that when Kikumura was arrested, he was en route to New York City, where he intended to detonate his bombs at a military recruiting office and then depart by plane on April 15, 1988. Fuentes also testified that in a search of Kikumura's car, two brown paper bags had been found that contained residues of what the FBI Laboratory later determined was ammonium nitrate.

Thurman testified about the bombs found in Kikumura's car. After describing their construction and stating that the completed timer was functional, Thurman discussed the destructive force of the bombs. He said that if the bombs were detonated in an occupied large room with a ten foot ceiling, there would be numerous casualties and significant property damage. He compared the bombs to Claymore mines and opined that, while they were capable of destroying property, they were meant as anti-personnel devices.

At the sentencing hearing, Thurman also testified about the significance of other chemicals found in Kikumura's car: aluminum powder, ammonium nitrate, and mercury. Kikumura possessed two pounds of fine aluminum powder. Thurman testified that if this powder were mixed with the right amount of ammonium nitrate, it could produce a 50-pound bomb capable of producing mass casualties and destruction in a room in which it was exploded. Only traces of ammonium nitrate were found in Kikumura's car, but Thurman testified that it is normally sold in 50-pound bags, and said it was logical to conclude from finding traces that a person likely once had a much larger quantity.

Kikumura also possessed mercury obtained by emptying thermometers. Thurman testified about two possible uses for this mercury. First, when combined with nitric acid and alcohol, it could be converted into mercury fulminate, a high explosive commonly used in blasting caps. Thurman admitted that Kikumura did not have all the chemicals needed to make mercury fulminate. Thurman also said that mercury might be used as a booby-trap to set off a bomb when it was moved.

Through affidavits, the government also introduced evidence at the sentencing hearing that Kikumura, as a member of the JRA, had received training in the manufacture of explosives at a terrorist camp in Lebanon and that he had been arrested in Holland in 1986 after attempting to smuggle over two pounds of TNT into Amsterdam. Kikumura did not call any witnesses or otherwise offer any evidence at his sentencing hearing.

After finding that Kikumura had possessed the bombs with an intent to kill or maim a large number of persons, the district court sentenced him to 360 months imprisonment. This sentence was reversed in 1990 because the Court of Appeals held that the district court had incorrectly applied the federal sentencing guidelines. On March 1, 1991, the district court resentenced Kikumura to 262 months imprisonment. This sentence was affirmed on October 15, 1991.

Whitehurst did not do any work on the Kikumura case. During an interview, he told us that he had not reviewed the underlying Laboratory reports or auxiliary examiner dictation, and that he also had not discussed the case with Thurman. Instead, he said that his criticisms were based solely on a review of the transcript of Thurman's testimony at the February 7, 1989, hearing.

The criticisms made by Whitehurst are summarized and discussed below.

Since this was a sentencing hearing, we find no basis for Whitehurst's assertions that Thurman by his testimony misled the jury or otherwise deprived Kikumura of a fair trial. Because Thurman testified at a sentencing hearing, the Federal Rules of Evidence did not strictly apply. More importantly, for the reasons presented in the following sections, we do not believe that Thurman's testimony at the sentencing hearing was materially misleading.

Whitehurst maintains that Thurman should have stated during his direct examination that his undergraduate degree was in political science and that he had no formal training in various areas on which he testified.

Thurman was asked on direct examination what formal education he had received that prepared him for his opinions in this case. He replied that he had received a master's degree in forensic science from George Washington University. He also was asked about his experience in the military and in the FBI Laboratory. Before the court accepted Thurman as qualified as an expert in the identification, construction, operation and the capabilities of explosive devices, defense counsel was allowed to conduct voir dire questioning.

Defense counsel did not ask Thurman any questions about his undergraduate education or his formal training beyond what Thurman described on direct examination. Given the questions asked during the direct examination and the voir dire, we do not think Thurman was obliged to volunteer additional information about his educational background.

Whitehurst complains that Thurman violated FBI protocols and procedures by testifying outside his expertise with regard to matters involving chemistry, electrical engineering, and the composition or manufacture of certain explosives.

As noted earlier in this Report, the FBI did not have any clearly stated policy concerning the permissible scope of testimony by a principal examiner about work done by auxiliary examiners. The unwritten practice described by many whom we interviewed was that, in their testimony, principal examiners could restate conclusions reached by others who contributed to the Laboratory report. Moreover, we note that a principal examiner may properly base his or her own opinion on analytical work done by other examiners. Similarly, expert opinions may sometimes properly rest on experience, as distinct from formal education or analytical tests specific to a particular case.

With that background, we do not think that Thurman improperly testified outside his expertise with regard to the identified matters. Thurman was asked if, after seeing the ingredients of the bombs, he later conduct[ed] a more thorough and scientific examination of the materials that were found in the defendant's car. After Thurman replied he did, he was asked if he had formed an opinion about the composition of the bombs taken from Kikumura's car. Thurman again answered affirmatively and said that the main charge was a combination of seven types of smokeless powder. Whitehurst asserts that Thurman simply lied on the stand because he did not conduct the examinations and has no training in chemistry.

We do not think Thurman lied by his use of the phrase I did when asked if he later conducted a more thorough and scientific examination. Thurman did in fact conduct additional examinations of the evidence. He also enlisted others to work on the case as auxiliary examiners. As for his opinion about the composition of the bombs, Thurman correctly restated the conclusions set forth in his Laboratory report, which in turn reflected analytical work done by chemist Lynn Lasswell. The report itself was introduced as an exhibit, and Thurman stated on cross-examination that he helped prepare the report. To avoid any possible misunderstanding about who did the work, Thurman should have said that the laboratory had conducted a more thorough examination of the items or that his conclusions reflected work done by others.

With respect to the discussion of smokeless powder, Whitehurst also asserts that the only Laboratory personnel who were deemed competent to render opinions about the chemical analysis of explosives were individuals in the MAU. We have discussed this issue previously with regard to the VANPAC case. When the work was done on the Kikumura case, the CTU also was analyzing materials to identify smokeless powders. We find no basis to conclude that Thurman violated FBI policies or procedures in having Lasswell analyze the powders found in the bombs or in restating the conclusions of the Laboratory report in testifying at the sentencing hearing.

Whitehurst also contends that Thurman is not an electronics expert and therefore should not have testified that he thought that the person who constructed the bombs, including the fusing system, had a very high level of expertise. This criticism fails to note that Thurman then explained, without any objection by defense counsel, the basis for his conclusion. Thurman explained that the fusing system had 14 soldered connections, including leads into the circuit board of the clock, insulated connections, and a safety switch. Thurman observed that the bomb maker would have to have considerable electrical experience and knowledge to effectively solder the leads to the circuit board. He also noted that a voltmeter was found in Kikumura's car, that a voltmeter is used to determine that there is a complete circuit, and that no instruction manuals were found in the car.

On cross-examination, defense counsel returned to the issue of Thurman's opinion concerning the bomb maker's expertise. Thurman agreed with the defense counsel that Kikumura was a very skilled bomb maker compared with the average citizen on the street. In response to further questioning, he also said that he thought Kikumura was more than a high intermediate when compared with people skilled in making bombs.

We do not think Thurman testified improperly in stating his opinion that Kikumura had a high degree of skill or expertise in bomb making. The construction of explosive devices is well within the expertise of an explosives examiner. The views stated by Thurman seem to have a reasonable basis in the evidence otherwise described in his testimony and Laboratory report. Thurman did not claim that his opinion was based on his being an electronics expert. After stating his opinion, Thurman discussed its basis during both his direct and cross-examination.

Whitehurst complains that Thurman was not qualified to testify about the identification of traces of ammonium nitrate found in Kikumura's car. Thurman testified, [t]he type of ammonium nitrate that we found here is commonly the type of ammonium nitrate agricultural-grade, that you would find in hardware stores, farm supply houses, but it's an agricultural-grade of ammonium nitrate. The Laboratory report stated that physical and instrumental analysis had determined white prills were agricultural-grade ammonium nitrate, which is a common ingredient in the clandestine manufacture of high explosives.

Thurman's testimony about the identification of the traces of ammonium nitrate was based on an auxiliary examination by Terry Rudolph. Dictation prepared by Rudolph stated that prills removed from a paper bag were identified as ammonium nitrate. The dictation also stated that, [t]hese prills were coated with diatomaceous earth and are probably of agricultural origin. The dictation further noted, as did the Laboratory report, that no residues or traces of explosives or ammonium nitrate were found on various other items. Rudolph's conclusions were reviewed and approved by Jerry Butler, who was then chief of the MAU.

Thurman further testified that agricultural grade ammonium nitrate could be found in farm supply houses and hardware stores. On cross-examination, defense counsel asked Thurman to confirm that the defendant had only three prills of ammonium nitrate and to explain how large a quantity that was. Thurman responded that ammonium nitrate normally comes in 50-pound bags and a prill is about the size of a pea. On re-direct, he again stated that ammonium nitrate in this form normally is sold in 50-pound quantities.

The statements by Thurman about where ammonium nitrate can be purchased and the quantity in which it usually is sold are not based on language in the Laboratory report. Thurman believes that he received this information from talking with Rudolph. The latter could not recall if he discussed these matters with Thurman. After reviewing Thurman's testimony, Rudolph also told us he thought it was accurate. Our own contacts with fertilizer manufacturers and distributors confirmed that agricultural grade ammonium nitrate is usually sold in 50-pound bags.

We do not find that Thurman testified improperly about the identification of ammonium nitrate, where it can be purchased, and the quantity in which it is usually sold. Thurman could properly testify to these matters based on the analytical work done by the MAU and information he obtained from talking with Rudolph or others. Moreover, for reasons previously discussed, we do not think Thurman violated any FBI protocols or policies by testifying on this subject.

Thurman in his report and testimony did not use the same language contained in the auxiliary examiner dictation concerning the agricultural origin of the ammonium nitrate. As noted above, Rudolph's dictation states that the ammonium nitrate was probably of agricultural origin, while Thurman's report and testimony state more positively that the prills were agricultural grade ammonium nitrate. Thurman told us he thought Rudolph approved the language used in the report. Rudolph cannot recall that conversation, but thinks it may have occurred. In these circumstances, we cannot conclude that Thurman intentionally overstated Rudolph's conclusions about the ammonium nitrate. This case, like others, illustrates the importance of principal examiners accurately reporting, whether in Laboratory reports or trial testimony, conclusions reached by other examiners. Moreover, if an auxiliary examiner agrees to a later modification or rewording of his or her conclusions, the file should reflect the basis for that action.

During his direct examination, Thurman testified that Kikumura had additional materials in his car that are commonly used in the manufacture of explosive devices. These materials were ammonium nitrate, aluminum powder, and mercury. He explained that with ammonium nitrate and the two pounds of aluminum powder found in the car one could make at least a 50-pound bomb. He also explained that mercury is an ingredient of mercury fulminate, a high explosive, and can also be used to make a booby trap for a bomb.

Whitehurst complains that Thurman left the jury with the incorrect impression that the defendant may have intended to construct a large and powerful bomb composed of ammonium nitrate and aluminum powder which was to be initiated with a mercury fulminate blasting cap. Based on Thurman's testimony, one could certainly conclude that Kikumura intended to make another bomb having a mercury fulminate detonator or a mercury switch and a charge made from a mixture of ammonium nitrate and aluminum powder. We do not agree, however, that this impression would be incorrect because Thurman's testimony was biased or otherwise flawed.

During his cross-examination, Thurman acknowledged that Kikumura did not have in his possession any bombs made from ammonium nitrate and aluminum powder. Thurman stated that only three prills of ammonium nitrate had been found in the car and agreed that a bomb could not be made with that amount. He further conceded that, based on the materials in the car, Kikumura could not detonate or even make a 50-pound bomb of aluminum and ammonium nitrate.

Defense counsel also asked Thurman if mercury fulminate was the same as mercury. Thurman replied, No sir. It starts with mercury and it is converted through a chemical process into mercury fulminate. The attorney asked him to explain the chemical process, and Thurman said two other ingredients, nitric acid and alcohol, are added. Thurman then explained, You add the nitric acid to the mercury fulminate. After that ceases its reaction and then you put in the alcohol. Thurman acknowledged that no nitric acid was found on Kikumura. Defense counsel then asked if there was any alcohol, and Thurman stated that common variety rubbing alcohol had been found. He also agreed that mercury mixed with alcohol would not produce mercury fulminate or a detonator.

Two aspects of Thurman's testimony about the manufacture of mercury fulminate merit comment. First, he misspoke in stating that nitric acid would be added to mercury fulminate as part of the chemical process. He apparently meant to say that nitric acid would be added to mercury, and after that reaction ceased, alcohol would be added to the mixture to produce mercury fulminate. Second, the isopropyl alcohol found in Kikumura's possession could not be used to make mercury fulminate. Whitehurst complains that by failing to note this point, Thurman misled the jury. That claim is unfounded: Thurman accurately answered the specific questions that defense counsel asked about the alcohol, which was an issue raised initially on cross-examination. In order to avoid the possibility that his testimony would be misunderstood, Thurman could have expressly stated that mercury fulminate could not be made from the rubbing alcohol found with Kikumura.

Whitehurst complains that Thurman failed to acknowledge that ammonium nitrate could be picked up from a lawn or an agricultural community and transported in Kikumura's shoes. The prills were found in the car inside a paper bag. We reject Whitehurst's suggestion that Thurman was obliged to volunteer that prills can be picked up in a person's shoes.

During his cross-examination, Thurman acknowledged that an explosive device could not be made with the three little peas of ammonium nitrate found in Kikumura's possession. On redirect, Thurman again said that ammonium nitrate prills are normally found in 50-pound quantities. The prosecutor then asked, And if you find traces of ammonium nitrate prills in someone's automobile, that's an indication that the person who is responsible for putting those prills there is also or likely had a much larger quantity of ammonium nitrate. Is that correct? After the court overruled an objection by the defense counsel, Thurman responded, That's the logical conclusion.

Whitehurst asserts that through this testimony, Thurman rendered an opinion that is too categorical and reached a logical conclusion that is obviously being presented to establish guilt. Whitehurst argues that prills can be picked up in a person's shoes from walking over a fertilized area and that ammonium nitrate can be purchased in small quantities. In making his criticisms, Whitehurst apparently did not recognize that the prills were found inside a paper bag, rather than on Kikumura's shoes or the floor of the car. Whitehurst also evidently misunderstood the nature of the sentencing hearing, which did not involve a jury or a finding of guilt, since guilt had already been determined. We do think Thurman should have been more careful in phrasing his response to the prosecutor's question. Thurman should have said that it was possible that whoever placed the prills in the automobile also had a much larger quantity of ammonium nitrate.

Thurman stated during his direct examination that he did not know the exact number of investigations in which he had been involved since joining the EU, but that it was in the hundreds. Whitehurst alleged that Thurman may have exaggerated the number of cases he had worked. Thurman joined the Laboratory in February 1981. In our investigation, we reviewed records from the FBI which indicated that Thurman worked on 129 cases between November 1984 and February 1989. We did not determine the number of cases on which he worked from the time he first joined the Laboratory in 1981 through October 1984. The FBI records indicate that Thurman accurately stated the number of examinations in which he had been involved.

Whitehurst contends that Thurman falsely testified that he had not attempted to force three pounds of powder back into one of the fire extinguisher cylinders because he did not want to take a chance of it exploding. Members of the EU, Whitehurst states, make pipe bombs by hand at the FBI bomb range at Quantico, Virginia, and thereby expose themselves to explosions. For this reason, Whitehurst says he does not believe Thurman had any concern about blast damage from the smokeless powder in the Kikumura case. In an interview with the OIG, Thurman again recalled that he did not attempt to force the powder into the cylinder because of the possibility of explosion. We found no basis to question the truthfulness of his testimony in this regard.

We did find minor inaccuracies or ambiguities in four aspects of Thurman's testimony that were not the subject of complaints by Whitehurst. First, in discussing the explosion that would result from the bombs found in Kikumura's car, Thurman described a fireball in somewhat ambiguous terms that might suggest the explosion would produce two fireballs when it likely would create only one.

Second, Thurman testified that mercury fulminate is commonly used in the manufacture of blasting caps. Although mercury fulminate was once so used, we question whether by early 1989 it was still commonly used in commercially manufactured blasting caps in North America. We think this is a relatively minor point, particularly because Thurman accurately stated that mercury fulminate can be used in a blasting cap for high explosives.

Thurman observed that the bombs found in Kikumura's car did not have as much directional capability as a Claymore mine. He explained that a Claymore mine would release steel balls in an arc of approximately 180 degrees, while in contrast the bombs would release lead shot in about a 360 degree arc. The bombs were described at the sentencing hearing as having lead shot at the top end of each cylinder. As described, they would release the shot in a manner similar to a shotgun shell. Thurman's testimony was incorrect or at least ambiguous insofar as it suggested that shot would be released in all directions from the bombs. If he intended to say that the bombs would release shot in a forward direction in a 360-degree circular pattern, he could have done so more clearly.

Finally, Thurman was technically incorrect in distinguishing high and low explosives based on their detonating velocities. Thurman correctly testified that high explosives have a detonating velocity of over 3200 feet per second, but he erred in stating that low explosives have a detonating velocity of less than 3200 feet per second. As discussed in Attachment C, infra, low explosives burn rather than detonate. By attempting to distinguish high and low explosives by detonating velocity, Thurman implied that both detonate, which is not normally the case.

We do not find that Thurman testified falsely in the Kikumura case. Nor do we find that he violated FBI policies or improperly gave biased or speculative testimony. To the extent we noted ambiguities or inaccuracies in his testimony, we think they illustrate the importance of having effective guidelines concerning the scope and monitoring of testimony by Laboratory examiners. The case also provides yet another example where the Laboratory should have improved its policies for the documentation of case work, file review, and record retention.

#####

The FBI Laboratory: An Investigation into Laboratory Practices and Alleged Misconduct in Explosives-Related and Other Cases (April,1997) Office of the Inspector General, Department of Justice