Stealing the Atom Bomb: How Denial and Deception Armed Israel

by Roger Mattson

  Fri did not tell the IG investigators anything about the third encounter that he said he had with CIA concerning NUMEC. It may have been the meeting with Brent Scowcroft in April 1976 that Giller recounted. Instead, Fri proceeded to recite the decade-old AEC explanation for the missing U-235, i.e., it could not be found because of sloppy accounting and was likely held up in the scrap processing line at Apollo where it adhered to pipes and filters.

  Eleven days after interviewing Fri, the two IG investigators turned their attention to General Alfred Starbird who had retired from ERDA.592 He told them that he accompanied Giller at the Rowden briefing in 1976 and joined Fri and Giller at the Shackley briefing in the summer of 1977. Starbird brought his personal attorney to the interview. He began by telling the investigators of the review he conducted relative to NUMEC in preparing for his high level review of the inventory difference report in the summer of 1977. His review included AEC documents describing the 1965 AEC investigation of NUMEC, interviews of ERDA staff that had been involved in that investigation, and examination of GAO and FBI investigations. He defended current safeguards as being superior to those in existence in the mid-1960s. He said that his review of these matters and his consideration of information presented by CIA led him to conclude, “that a significant amount of SNM had never been stolen from NUMEC.” He cautioned, however, “That any interpretation of the AEC/ERDA position on this question should not misconstrue his statements to mean he was ‘certain’ that no diversion had ever occurred.”

  Starbird described the briefing by Shackley and said Shackley showed him, Fri and Giller a talking paper [fifteen lines redacted].

  Starbird ended his IG interview with a long explanation of why previous written answers he provided to Dingell about his NUMEC-related testimony were nonresponsive, i.e., he was the victim of inept subordinates.

  The IG investigators conducted a second interview of Fri on March 8, 1979, ten months after his first interview.593 The DOE redacted much of the record of that interview before it was released. The unredacted information did not add much to the summary of Fri’s first interview. Fri said that he did not view Knoche’s briefing in 1977 as the official position of the CIA. One is left to wonder how much more official a CIA statement has to be than the words of the Deputy DCI in a closed meeting in the office of the National Security Advisor describing SECRET/SENSITIVE National Security Information.

  Based on the interviews by Anderson and Knauf, the Inspector General’s office of DOE on April 27, 1979 told DOE Undersecretary John Deutch that Fri “made a knowing misstatement of actual knowledge” to Dingell on August 8, 1977 by denying he knew of the views of a U.S. intelligence agency on the alleged diversion of SNM from Apollo in the mid-sixties. That same day the Department of Justice found there was insufficient evidence to justify prosecuting Fri for a “knowing misstatement” to Congress.594

  ***

  At least one person lost his job at DOE in the aftermath of the Fri investigation. Harvey E. Lyon, long-time head of ERDA and DOE’s Office of Security, was the person who originally refused to grant Jim Conran access to NUMEC files. Lyon also helped prepare Fri’s 1977 testimony to the Dingell and Udall committees. He resigned from DOE in January 1979 amid rumors he had been fired. His boss, Duane Sewell, DOE assistant secretary for defense programs, refused to comment on the reasons for Lyon’s departure, while Lyon said he was offered another position in an ongoing reorganization and “didn’t want it.”595

  ***

  Henry Myers recalled that in 1978 the Defense Science Board assigned John Deutch, then serving as undersecretary of energy, to perform an “all sources” review of the NUMEC affair. Myers noted, “In so doing he has seen materials not available to us.” Deutch promised to tell Congressman Udall the results of his review, but he never did.596

  ***

  On June 22, 1979, Michael McNulty warned Udall that Shapiro’s lawyers were “feeling the heat . . . with some sort of FBI investigation now breathing down their neck, the precariousness of Shapiro’s situation is causing them a lot of anxiety . . . now that they may be going to a grand jury; and so they’re going to jump on Henry [Myers].” McNulty urged Udall to be short and to the point with Shapiro’s lawyers in their efforts to stem the tide of damaging documents that were flowing to the press.597

  ***

  The press accounts of growing evidence of a diversion from Apollo apparently did not faze former AEC Chairman—he said he heard it all before when he corroborated the public statements of Fialka, Newsweek and other journalists concerning CIA’s 1968 discovery of HEU near Dimona. Seaborg said, in reference to the various newspaper articles about NUMEC that appeared in 1979, “these accounts as far as I have read them, revealed without gross inaccuracies much of what had occurred.”598

  Journalists Steve Weissman and Herbert Krosney, who relied mainly on interviews with government officials in the U.S. and the Middle East for their 1981 book on the Islamic bomb, also described CIA’s 1968 discovery of HEU in Israel. They went further than others in concluding that such material was used in Israel’s first atom bombs.599

  Duckett stated in his briefing that the CIA had varied evidence for its 1968 report. . . . The CIA had also followed up a disturbing rumor. American scientists returning from Israel were worried over signs that the Israelis appeared to be working with bomb-grade highly enriched uranium. The scientists informed the CIA, and the CIA put to work some highly sophisticated equipment to monitor air and soil samples from around Israel’s nuclear facility at Dimona. The “sniffers” and other instruments proved what the CIA most feared: Israel had highly enriched uranium, and the CIA concluded that Israel had enough of it to make several bombs. . . . When the CIA first reported that Israel had nuclear weapons in 1968, they apparently knew little or nothing of the French plutonium. What they did know was that Israel had the other bomb material, highly enriched uranium. Their “sniffers” and other fancy instruments had found unquestionable traces of it in waste taken from Dimona. . . . Agency officials were convinced that Shapiro and NUMEC had supplied Israel with more than 200 pounds of highly enriched uranium, or enough for several atomic bombs.

  ***

  Carl Duckett took a long time before he told anyone outside CIA that the Agency found traces of Portsmouth uranium in the environment near Dimona. Finally, in 1991, he discussed CIA’s 1968 discovery with investigative journalist Seymour Hersh. However, Hersh claimed that the uranium found by CIA near Dimona could not be associated with NUMEC. “Duckett and other government investigators into NUMEC acknowledged that there is no meaningful correlation between uranium processed in the NUMEC plant and the traces of enriched uranium picked up by the American agents outside Dimona.”600 If Duckett said this to Hersh, he was wrong.

  A statement of “no meaningful correlation” is not correct. The science of nuclear forensics had been around for twenty years by the time it was applied at Dimona. U.S. intelligence agencies, including AEC, started working with nuclear forensics immediately after World War II. They applied scientific methods at increasing levels of sophistication to diagnose nuclear operations and experiments by other nations, both friend and foe.601 In the case of highly enriched uranium near Dimona, the forensic analysis no doubt began with the fact that only five countries could enrich uranium in 1968, namely, France, England, Russia, China and the United States. Although the enrichment facilities in those countries were similar, they had unique design characteristics and unique enrichment capabilities. Therefore, the mixtures of uranium isotopes that they produced were also unique. However, because the operating cycles of enrichment plants frequently vary, it is difficult to distinguish a particular source of enrichment by gaseous diffusion. Nevertheless, attention would have focused quickly on the three U.S. gaseous diffusion plants. Then, the analysis would have centered on the Portsmouth plant that, beginning in 1964, was the sole source of 97.7 percent enriched uranium. Thus, any traces of that uranium found near Dimona had to have come from one of the companies that
processed it into naval reactor fuel. NUMEC’s Apollo plant was the leading candidate among the small number and size of such companies because of the relatively high volume of naval fuel it processed compared to the other plants, its historically high losses of HEU, and its close ties with Israel.

  ***

  Contrary to Hersh’s claim of “no meaningful correlation,” there is proof that ERDA staff (and its predecessor AEC staff) knew that spectrographic analysis could establish the signature of highly enriched uranium near Dimona. That proof is evidenced by ERDA’s reply to an inquiry from the FBI in 1976, not long after Duckett briefed NRC. Sam McDowell of the ERDA staff (the AEC official that signed the 1966 final report of AEC’s investigation of the missing HEU at Apollo) recorded his conversation with the FBI special agent as follows:602

  On May 20, 1976, the writer met with (redacted) in Rm A2 1304 [an office in the Germantown, MD headquarters of AEC]. (Redacted) asked what technique might be available to determine whether enriched uranium if such can be obtained from the Israelis could be traced to material from the Numec facility during the 1963-1965 time frame. The writer told (redacted) that two possibilities exist. One is comparison of the mass spectrometer data including minor isotopes of the Israeli material with reference to cascade mass spectrometer data on file at Oak Ridge for the same period. The second possibility would be to relocate any of the WANL fuel rods or elements containing the Numec material. There is a good chance that some of these may exist at Nevada, Idaho or Los Alamos. This is now being checked out. If located, the material could then be compared with the Israeli material, if such material can be obtained.

  Although McDowell focused his answer to the FBI on the Oak Ridge enrichment plant because the Astronuclear (he called it WANL) uranium was enriched there, it is clear from his description of the technology that it was equally applicable to uranium enriched at Portsmouth. No record has been found to show whether McDowell knew of the HEU with a Portsmouth signature found near Dimona. It is noteworthy that CIA apparently found that unique HEU eight years earlier than this FBI query of McDowell. However, CIA, by its own records cited above, admitted it did not tell FBI the details of what they found until the 1970s when Conran’s whistleblowing caused the attorney general to renew the FBI investigation. If so, then the FBI inquiry of McDowell might have been simply to confirm what CIA had concluded in 1976; i.e., it was possible to trace the origins of HEU with the unique signature of Portsmouth by spectrographic analysis.

  On May 28, the Special Agent in Charge of the Washington Field Office sent a Top Secret Teletype to the FBI Director giving the FBI’s version of the meeting with McDowell eight days before.603

  On May 27, 1976 (redacted, but probably McDowell) ERDA, Germantown, Maryland, advised that detailed information pertaining to analysis of material processed at NUMEC during the early 1960’s is on file at Oak Ridge and [Los Alamos], but cannot be removed from their libraries. Also actual material in the form of reactor rods probably can be retrieved from installations in the United States. (Nine lines redacted.)”

  Nothing has been found in later FBI memoranda and letters on NUMEC to indicate that FBI pursued this line of investigation. However, many FBI documents subsequent to this one were almost completely redacted before release.

  Modern day uses of “signature” technology include deployment of sensitive detection capabilities by the United States for monitoring materials and effluents from nuclear facilities around the globe. The IAEA also uses this technology.604 As reported by nuclear experts Thomas Reed and Danny Stillman, “Should a nuclear device detonate by surprise anywhere in the world, the forensic scientists of the nuclear weapons community will be able to identify the source of the fissile material within a few hours. The nuclear weapons states and the IAEA have built an excellent library of nuclear fingerprints that precludes anonymity.”605

  Duckett was the former chief scientific officer of CIA. He led the Agency’s tracking of Soviet nuclear strike capabilities for much of the 1960s and 1970s. He had to have known about nuclear forensics. Therefore, either Hersh misquoted or misunderstood Duckett or Duckett misled Hersh. Independent of Hersh’s rendition of the exchange between the two of them, by the late 1970s, Seaborg, Knauf, Anderson, Starbird and probably most of the other people that Shackley briefed in 1977, if not every one, knew of CIA’s 1968 discovery near Dimona of HEU with a Portsmouth signature.

  ***

  In addition to there being a number of sources for the fact that Portsmouth HEU was discovered by the CIA in the environment near Dimona, there were also a number of journalists that wrote about that discovery (Hersh, Fialka, the Cockburns, Newsweek, and Weissman and Krosney). In addition, no one has suggested in more than 45 years that there was any other reason for the 1968 CIA request that FBI renew its investigation of NUMEC or the much later 1976 FBI query of ERDA on how to interpret uranium enrichment signatures.

  Nevertheless, there is some uncertainty in the evidence supporting this fact. That is, no NRC commissioner or staffer has ever said that they were told by CIA of its 1968 discovery, near Dimona, of HEU with a Portsmouth signature. Such evidence, when viewed by technical experts at NRC, should have constituted strong evidence, if not proof, of a diversion from NUMEC to Israel. No one at NRC has ever claimed to have such proof. Instead, NRC has been content to say it had “no conclusive evidence,” as NRC Chairman Hendrie put it, of a diversion. The lack of an NRC source for CIA’s claim that it found Portsmouth HEU near Dimona is intriguing. It is clear that Duckett did not tell NRC of this evidence. In their interviews by McTiernan regarding the Duckett briefing in February 1976, Commissioners Mason and Gilinsky (the only commissioners with technical qualifications at that time) said there was no spectrographic information conveyed by Duckett. Mason went further:606

  When asked whether there was anything by way of a “fingerprint” (i.e., spectrographic analysis) linking Apollo material with anything that was in Israel, Dr. Mason emphatically answered that there was no such reference. That was precisely the sort of hard evidence that he was expecting to hear, and nothing of the kind was presented.

  Kenneth Chapman, director of NRC’s Office of Nuclear Material Safety and Safeguards and Builder’s boss, identified several items alleged in the press that were not contained in Duckett’s briefing. He said, “neither Mr. Helms nor the President were mentioned; there was no mention of the classification level of the briefing, other than it had ‘sensitivity’; details printed in later Fialka articles were not given, i.e., monitors, encoded phones, etc.; and there was no mention of the FBI.”607 Chapman said there were no “monitors.” Thus, as far as Chapman recalled, Duckett did not tell NRC anything about monitors.

  Later, Gilinsky recalled hearing that CIA radiation monitors found HEU near Dimona, but could not recall where, perhaps at the Department of State sometime after the Duckett briefing.608 Apparently Shackley did not tell the NRC people about Portsmouth HEU when he briefed them on February 2, 1978.

  ***

  On August 3, 1978, John Fialka reported that NRC investigator McTiernan had written a long memorandum to the NRC commissioners when he retired in July. McTiernan told the commissioners, “there is an aura or atmosphere about NUMEC/Apollo that leaves one with a strange feeling that there is some kind of story to be told, something that, for some reason, has not yet been said. I don’t know whether there is such a story, but, looking back, there are certain questions that now cross my mind.” McTiernan went on to say that Shapiro, “should be given an opportunity, if he chooses, to answer publicly the shadowy charges that have been leveled against him over these many years.”609

  * * *

  33 It is not clear from McTiernan’s summary whether Duckett referred to the Plumbat affair in 1968, which involved 200 tons of Belgian yellowcake via Germany, or Argentina’s 1963 direct sale of 80 to 100 tons of yellowcake to Israel.

  34 This person would have been James Lovett, an AEC materials expert, not the chief of AEC’s investigation. Shapiro hired Love
tt to improve material control and accounting at NUMEC shortly after the 1965 AEC investigation. From the summary of his remarks, it is unclear when Duckett referred to the MUF at NUMEC whether he meant the large inventory difference that accumulated by 1968, the inventory difference AEC found in 1965, or the unexplained portion of the 1965 inventory difference. Whichever it was, he thought it was “undue.”

  35 Edward Bennett Williams, who represented Shapiro in his quest for an AEC Sigma clearance in 1970, did not represent him in this instance. It may have been a matter of costs. Hadrian Katz of Arnold & Porter told Senator Arlen Specter (D-PA) in 2009 that Shapiro was “our long-time pro bono client and friend.”

  36 ISORAD continued in operation into the 21st Century, providing Israeli nuclear knowhow to foreign clients.

  37 Aliyah or aliya is enshrined in Israel’s Law of Return that accords any Jew the legal right to assisted immigration and settlement in Israel, as well as automatic Israeli citizenship.

  38 From 1955 to 1963, Scoville served as Deputy Director for Research at CIA. He helped develop technology required for independent verification of nuclear weapons capabilities without on-site inspection. He left CIA in 1963 to become assistant director for science and technology at the Arms Control and Disarmament Agency where he helped formulate U.S. positions for the Limited Test Ban Treaty and the Non-Proliferation Treaty.

  Chapter 14

  CIA Hydra (1979)

  The Central Intelligence Agency was associated with the NUMEC affair in various ways. However, it is difficult to know CIA’s roles in detail because it is a secret agency, the events of interest are decades old and the Agency is not telling everything it knows. The possibilities for CIA’s involvement range from simple observer of the facts in America, to espionage in Israel to covert operations at home and abroad.