Whether the HEU from Apollo was used as weapons fuel or in Dimona booster fuel, either of these end uses would have justified the costs of establishing NUMEC, building the Apollo plant and paying the $2.2 Million in fines for the missing uranium. As things worked out, even some of those costs were avoided.
***
In response to pressure from the Joint Committee after its hearings in early 1966, AEC staff undertook interviews of NUMEC employees to gain further insight into what might have happened to the missing uranium.179 The interviews included 23 current and 14 former employees, ranging from company president Shapiro to plant workers, including two vault custodians and two security guards, plus two Westinghouse employees familiar with the Astronuclear contract.
Howard Brown asked Marcus Rowden, an assistant general counsel of AEC, to provide procedures to control the interviews. Rowden said the interviewers should tell the witnesses of their rights to refuse to testify and to have legal counsel of their choice present for their interviews. He “emphatically stated that under no circumstances should the Interview Team attempt to obtain written statements from any of the witnesses. He said this was a matter which FBI specifically objected to.”12 Other AEC staffers objected to Rowden’s instructions saying they would inhibit the witnesses and hinder the investigation. Rowden insisted on his way, going so far as to say, “if any information is uncovered which appears to be of pertinent value, the investigation should be discontinued until the facts can be related to HQ and, if necessary, the team might even have to come back to HQ to go over the situation.”180 Rowden kept the Interview Team under tight control. He certainly was being careful in protecting the AEC in this delicate matter.
The principal conclusions of the interviews were reported by AEC, as follows:181
1.All knowledgeable interviewees believed the losses were sustained during processing operations [AEC did not define a “knowledgeable” interviewee];
2.Terminated employees were nearly unanimous in saying NUMEC’s low wage scale led to a high rate of employee turnover;
3.All interviewees said they knew of no instance of enriched uranium being stolen or otherwise diverted from Apollo, but several interviewees said it would have been possible to do so without being detected, for small quantities;
4.NUMEC management personnel stated that to the best of their knowledge there were no instances of deliberate commingling of enriched uranium assigned to different contracts;
5.NUMEC had a line of credit with Mellon National Bank and Trust Company in an aggregate amount not to exceed $2.5 million. This line of credit had been used to make the first $500,000 payment to Westinghouse for the material missing from the Astronuclear contract, on or about December 23, 1965; and
6.NUMEC had sufficient internal controls on shipments, which, in the absence of deliberate collusion, should ensure that the quantities reported on transfer documents were those quantities shipped.
The AEC did not pursue the third conclusion that it was possible to divert U-235 from Apollo. The fourth conclusion contradicted what AEC told the Joint Committee about one of the root causes of the missing material, i.e., there was commingling of material between contracts. The fifth conclusion would interest investigators for another decade, as described below—the issue was not who made the loan but who paid it back.
The employees interviewed by AEC in 1966 were not as forthcoming as one employee would be 15 years later with the NRC and the FBI. That is, no one told AEC in 1966 about encountering armed strangers on the dock one night in early 1965 loading canisters of HEU onto a truck in racks that had not been seen before, heading to a ship bound for Israel. Nor did that NRC informant tell AEC in 1966 that a NUMEC manager threatened him to keep his mouth shut, or else.
***
In September 1965, NUMEC’s annual report to shareholders showed that the company was operating on credit, as follows:182
•A $2.5 million line of credit with Mellon National Bank,
•A $1.5 million term note from Mellon payable in sixteen quarterly payments commencing April 1, 1967,
•$1.8 million of convertible subordinated income notes issued to institutional investors due April 1,1978, and
•$509,500 borrowed from Pittsburgh National Bank under a participating loan agreement involving the Small Business Administration.
In its fiscal year 1965, NUMEC had annual revenues of about $7 million and net income of about $240,000. The company was more than $3.8 million in debt before it used its line of credit with Mellon Bank to pay the fine owed for the missing uranium on the Astronuclear contract. The company was highly leveraged.
***
Whatever caused the sloppy accounting of uranium at Apollo, the situation quickly changed. As soon as AEC began to more closely scrutinize operations at Apollo, its inspectors were told by the employees they interviewed, “there was marked improvement in the nuclear materials control posture of NUMEC [as reported by] nearly all of the [NUMEC employees interviewed by AEC in early 1966].”183
Material controls and accounting was not a difficult science in 1960—it never has been—and if NUMEC was quick to improve its material controls and accounting once AEC applied pressure to the situation, the question remains, why not sooner? However, the assertion that the company was quick to turn around its sloppy material control and accounting practices would soon be open to question.
***
In April 1966 Dr. Samuel McDowell authored a 23-page report, with 40 pages of attachments, describing AEC’s late 1965 and early 1966 investigation and independent inventory of Apollo and AEC’s subsequent interactions with NUMEC.184 McDowell was a leader among U.S. material controls experts. A few years later, FBI agents would rely on his expertise in dealing with another aspect of the NUMEC investigation. In 1966, however, the key points in his report on AEC’s 1965 and 1966 investigation of the missing uranium at Apollo were as follows:
•The total cumulative loss through October 31, 1965 was 178 kilograms of U-235 in the form of HEU;
•NUMEC had previously reported 149 kilograms as missing;
•The total cumulative loss as a percentage of throughput was larger than losses reported by other comparable facilities;
•It could not be stated with certainty that diversion did not take place, but no evidence was found to support this possibility;
•There were impediments to diversion, but there were weaknesses that would have facilitated diversion;
•The AEC investigation identified “all known loss mechanisms” and they totaled 84.2 kilograms of U-235;
•The loss of an additional 93.8 kilograms of U-235 owed to unknown mechanisms, but underestimates of known losses might account for 15 kilograms of U-235;
•NUMEC had not assigned the caliber of professional talent found by other companies to be necessary for adequate accounting of Special Nuclear Material;
•Foreign transfers of SNM placed reliance on the integrity of the shipper and the receiver for verification of quantities shipped; AEC provided no independent confirmation of the quantities transferred;
•Such controls, when properly implemented, should, in the absence of a deliberate collusion, ensure that quantities reported on transfer documents were indeed those shipped;
•Many of the inventory records requested by AEC were inadvertently destroyed by NUMEC supervisory personnel during a “clean up” campaign at the time of an employee strike that lasted from January 1 to February 25, 1964; and
•There was a “high degree of probability that WANL contract material was transferred to other contracts.”
Several things stand out among these conclusions. First, as AEC and FBI were to learn later, NUMEC prepared some large containers for shipment to Israel that could easily have hidden 93.8 kilograms of uranium, and NUMEC shipped them to Israel about the time that the uranium was discovered to be missing from the plant. If NUMEC diverted the HEU to Israel, then there was a prima facie case for collusion on both sides of the transaction. Second, in
ventory records were not supposed to disappear in cleanup operations; rather, they were required by AEC regulations to be protected for the life of the plant. Their destruction was a violation of AEC requirements for which AEC did not issue a citation. Third, McDowell did not accept Shapiro’s claim that there was no commingling of uranium among contracts.
Reading between the lines of the McDowell report, some of AEC’s influential commissioners and staff might have believed the missing uranium was held up in unrecoverable scrap or in cracks and crevices of the plant and its processing equipment, but they couldn’t be certain. Such a conclusion was unjustified because the amount held up in scrap was measured independently by the AEC in making the 1965 inventory. In addition, the burial pit was exhumed, the material found there was included in the final inventory and there were no other authorized waste disposal locations. Furthermore, if AEC believed more uranium was held up in different places than it had already identified, then it could have made an estimate of how much, just as it did for the amount it estimated to have been tracked out of the plant on the shoes of workers. Apparently, the AEC commissioners and staff had a collective mindset that did not permit recognition of the real possibility that someone had stolen the missing uranium. It is also possible that there was collusion within the AEC to protect the diversion of uranium to Israel. As the years went by, the CIA collected new evidence that the uranium found its way to Israel, and the attorney general ordered an investigation of possible collusion in government agencies, as described in the following chapters.
***
The AEC said the monetary penalty it imposed on NUMEC represented the cost of material of various enrichments. Contemporary documents confirm that most of the missing HEU exceeded 90 percent enrichment. For example, the 1965 audits attributed the losses of HEU to the Astronuclear contract, which was 93 percent enriched.
Similarly, in a July 27, 1965 letter to Charles Keller at Oak Ridge, Shapiro said that analysis of air filters at Apollo showed “The vast majority of the filters in storage were accumulated during a period when the predominant quantity of uranium processed at the plant was enriched to greater than 90%. . . . As mentioned previously, we were processing predominantly higher enriched uranium in the period during which the 905 drums were buried.”185 The fact that NUMEC began manufacturing 93 percent enriched uranium fuel for the Shippingport reactor in 1964 while it was processing HEU enriched to 97.7 percent for naval reactor fuel and 93 percent enriched uranium for the Astronuclear contract supports Shapiro’s statement that most of the uranium in the plant for the period in question was HEU.
On February 14, 1966, AEC’s general manager provided further confirmation of the HEU enrichment levels in the period of interest when he told the Joint Committee, “The quantity of 178 kg of U-235 represents enriched material ranging from slightly enriched to fully enriched at 93.15% U-235. Some SNM enriched greater than 93.15 has also been processed at NUMEC.”186 The more highly enriched SNM to which he referred would have been naval reactor fuel.
In June 1967, the GAO (General Accounting Office) reported to the Joint Committee on its assessment of material control at Apollo. The report quoted an Oak Ridge document of 1964 saying “the percent of MUF [inventory difference] disclosed by comparing the adjusted book inventory to the physical inventory was 8.62% loss of uranium above 75% U-235, 3.19% loss of uranium below 75% U-235 and 6.01% loss of leased nuclear materials.”187 That is, NUMEC was more than twice as capable in preserving U-235 in low enriched uranium than it was in preserving U-235 in highly enriched uranium.
The GAO also said, “During the period of our review, we found that additional losses had been disclosed and NUMEC’s records showed that cumulative losses of U-235 through December 31, 1966 totaled about 260 kilograms or about 1.2% of receipts.” GAO attributed its knowledge of the inventory in late 1966 to a letter from Zalman Shapiro commenting on a draft of the GAO report.188
Thus, 82 kilograms of U-235 in the form of HEU was lost at Apollo in the 14 months between the time the AEC audit ended on October 31, 1965 (178 kilograms) and Shapiro’s estimate for December 31, 1966 (260 kilograms). This was the highest rate of loss of HEU in the entire history of Apollo operations. Apparently this fact troubled no one at AEC or GAO, at least no one said so.
The losses at Apollo continued to accumulate. The AEC’s records of inventory differences, as documented by ERDA, for companies that held strategic quantities of special nuclear materials showed that for the eight years from 1960 through March 1968, the inventory difference of HEU at Apollo was 267.1 kilograms. For reasons described above, most of this missing uranium would be of enrichment levels in excess of 90 percent.189
In summary, the 1965 inventory by AEC included HEU of variable but generally high enrichments, including naval reactor fuel materials as high as 97.7 percent enriched. The 1967 GAO report and the 1977 ERDA report cited above confirm that the cumulative inventory difference for U-235 contained in HEU was 178 kilograms at the end of October 1965, grew to 260 kilograms by the end of December 1966 and was 267.1 kilograms by March 1968. That estimate of the total inventory difference was in close agreement with a 2001 Department of Energy report showing that the total inventory difference for U-235 in HEU at NUMEC from 1957 to 1968 was 269 kilograms.190 These figures were all in the form of inventory difference, i.e., they combined losses that could be attributed and estimated, after the fact, to routine causes (such as measurement error or estimates in lieu of measurements of radioactive effluents and waste disposals) and losses that could not be attributed to routine causes (such as theft). It is likely that by 1968 the total amount of HEU that could not be attributed to routine causes exceeded the 93.8 kilograms estimated by AEC in 1965.
***
While the AEC was struggling with NUMEC to deal with the missing uranium, there were developments on a related front. On July 28, 1966, Secretary of State Dean Rusk sent a telegram to the U.S. Embassy in Israel concerning U.S. policy on Israel’s nuclear program. Rusk said he told Israeli Ambassador Avraham Harman that the Soviets could get the Arabs to agree to denuclearization if Israel would agree to IAEA safeguards of its nuclear facilities such as Dimona. Israeli Envoy Gideon Rafael, who accompanied the Israeli Ambassador, replied to Rusk that Israel was not giving thought to that option because of the broader problem of conventional arms. Rusk responded that if Israel went nuclear it would lose U.S. support. “Secretary then told [the] Israelis that nothing would be more disastrous to GOI [Government of Israel] than [to] enter nuclear weapon field and urged them to agree to international safeguards. . . . He again noted that if Israel is holding open the nuclear option, it should forget U.S. support. We would not be with you, he said.”191
***
On September 18, 1966, an article authored by John W. Finney, dateline Washington, DC, appeared in the Paris edition of the New York Times. Finney was the newspaper’s lead correspondent on atomic energy matters. The article reported AEC’s discovery that one of its industrial contractors had lost more than 100 kilograms of enriched uranium, enough to make half a dozen atom bombs.192 This was the first public reporting of NUMEC’s missing U-235. Finney did not name his source, but someone must have been concerned with the coverup then underway by the AEC, the FBI and the Joint Committee.
In later years, there were various reports in the media of the amount of uranium missing from Apollo by the end of 1965.193 Those reports were based on what others recollected, not on the evidence summarized herein, that is, contemporary records of what AEC told Congress at the time and upon the underlying technical reports, such as the 1966 McDowell report, which supported the AEC’s conclusions.
In addition, well after the fact, there were various accounts of the fines that NUMEC paid for the missing U-235 on the Astronuclear project.194 The $2.2 million figure described herein, of which $764,000 was associated with uranium missing from the Astronuclear project, is based on records of what AEC told Congress at the time.
By the end of 1965, with the pric
e of highly enriched uranium ranging from $10 to $15 per gram, about ten times the price of gold at that time, despite AEC’s best efforts to find it, some 93.8 kilograms of U-235 in the form of HEU were missing from Apollo. No one, including the most qualified experts, could say for sure when or where it had gone. By 1968, even more HEU went missing from Apollo.
Because of NUMEC’s loss of so much HEU at Apollo, the AEC concluded that economic incentives based on a presumption of honesty were not sufficient to protect bomb-making materials in the commercial sector. The commission directed its staff to develop new systems for accounting and securing strategic quantities of special nuclear material.
***
Soon after concluding its investigation at Apollo, the AEC formed an Advisory Panel on Safeguarding Special Nuclear Material, headed by Ralph F. Lumb, director of the Western New York Nuclear Research Center in Buffalo. The panel visited sites around the country that handled special nuclear material, including Apollo, and discussed safeguards with the people in charge, including Shapiro.195 The panel recommended improvements of the safeguards provided for special nuclear materials, including physical security (fences, locks, vaults, guards, and other means to limit access to the material) and criteria for record keeping. The panel also recommended that AEC place resident inspectors at particular facilities, including Apollo.
Stealing the Atom Bomb: How Denial and Deception Armed Israel Page 11