The NUMEC / Apollo Affair

Sometime in the months before April 1968, CIA officers in Israel collected environmental samples from around the reactor at Dimona. The results came back showing a rare type of uranium used in American naval fuel, enriched to 97.7 percent. Israel was not believed capable of producing that material itself at the time.

The obvious question was where it had come from, and there was an obvious place to look. A plant in Apollo, Pennsylvania, operating under U.S. government contract, could not account for a very large quantity of exactly that kind of uranium.

Nobody was ever charged. Nothing was ever proven. Sixty years on, key files remain classified, and the United States government has never said either that the uranium was stolen or that it was not.

The plant

The Nuclear Materials and Equipment Corporation was founded in 1957 by Zalman Shapiro, a respected chemist who had contributed to early American nuclear work. It turned government-owned enriched uranium into fuel for naval reactors and commercial use, which meant it handled large quantities of weapons-usable highly enriched uranium under contract, with strict accounting obligations.

Shapiro was a committed Zionist with extensive contacts in Israel. That is not an inference about him, it is a matter of documented association. He was in touch with the head of Israeli military intelligence and with the head of its nuclear weapons program, and he later acknowledged knowing Binyamin Blumberg, who ran Israel’s bureau of scientific liaison, the economic espionage outfit better known as LAKAM. LAKAM appears elsewhere in this archive, running Jonathan Pollard and handling Arnon Milchan.

The hole in the books

The core anomaly was an accounting one. Department of Energy records later showed NUMEC with the largest highly enriched uranium inventory loss of any commercial site in the United States: 269 kilograms unaccounted for before 1968, and another 76 after. Estimates of the total have ranged from roughly two hundred to six hundred pounds, with the FBI investigation focused on about 206.

There is an innocent explanation and it has never been disproved. Processing loses material. It plates out inside pipes, collects in filters, ends up in waste. Every plant of this kind loses some. NUMEC lost more than any other, by a wide margin, and that is what kept federal investigators coming back for fifteen years.

What the government did, and did not do

The Atomic Energy Commission looked into it in 1966. Victor Gilinsky, later a commissioner of the Nuclear Regulatory Commission, went through the declassified files decades afterward and described that effort as “anemic at best.” Officials visited the plant, took no formal statements, and did not pursue the leads they had. Senior figures at the Commission, notably its chairman Glenn Seaborg, pressed to keep the matter in-house and the FBI at arm’s length. Seaborg wrote that while diversion was possible, it was “unlikely.”

The CIA’s samples changed the temperature. In April 1968 the Director of Central Intelligence, Richard Helms, wrote to Attorney General Ramsey Clark saying that highly enriched uranium processed at Apollo might have ended up at Dimona, and asked that the FBI restart its investigation, the Agency being barred from domestic work.

Helms also told the President. According to the declassified record, Lyndon Johnson’s response was: “Don’t tell anyone else, even [Secretary of State] Dean Rusk and [Defense Secretary] Robert McNamara.”

Clark put Shapiro under physical and electronic surveillance. It turned up meetings between Shapiro and known Israeli intelligence figures.

The visitor

One detail in the file has a long shadow.

FBI and AEC documents record that in September 1968 four Israeli citizens known to be connected to Israeli intelligence visited the NUMEC plant at Apollo. One of them was Rafi Eitan.

Nineteen years later, Eitan was exposed as the head of LAKAM and the man who ran the Pollard operation. The future chief of Israel’s scientific espionage bureau walked through the plant then under federal investigation for losing bomb-grade uranium to Israel, during the period of maximum suspicion. It is among the most pointed facts in the record. It is also, like the rest of the case, circumstantial.

What the people who read the files concluded

The strongest statements about NUMEC come not from partisans but from nuclear regulators who went through the documents.

In 2010, Victor Gilinsky and Roger Mattson, a former NRC physicist who had investigated NUMEC, published a reassessment in the Bulletin of the Atomic Scientists. Their conclusion: “When the known facts are put together, it is difficult to escape the conclusion that Israel probably did steal highly enriched uranium from the United States.” Mattson later published a book-length study through the National Security Archive.

The word is “probably,” not “certainly,” and that distinction is load-bearing. These are the people best placed to judge the record, and what they reached was a probabilistic judgment, not a finding.

The non-conclusion

No charges were ever brought against Shapiro or anyone else across fifteen years of investigation. He denied diverting anything, consistently, and died in 2016 at ninety-six with nothing ever alleged against him in court.

The official position is a study in refusing to decide. In 2009, prompted by Shapiro’s lawyer, Senator Arlen Specter asked the Nuclear Regulatory Commission to clear him. The Commission would not. It said it had “found no documents that provided specific evidence that the diversion of nuclear materials occurred,” and in the same breath that it did “not have information that would allow it to unequivocally conclude that nuclear material was not diverted from the site.”

The government would neither accuse the man nor exonerate him.

A declassified Government Accountability Office review of the investigations put it more sharply, concluding that “a timely, concerted effort on the part of these three agencies would have greatly aided and possibly solved the NUMEC diversion questions, if they desired.” That last clause is doing a great deal of work.

The plant left one unambiguous legacy. The Apollo and Parks Township sites stayed contaminated and required an expensive federal cleanup by the Army Corps of Engineers decades later.

Why it matters

If the diversion happened, the fissile material at the heart of Israel’s first weapons was stolen from the United States, out of an American plant, under an American contract. And the American government, holding the CIA’s samples from Dimona and its own surveillance of Shapiro, chose not to prosecute and not to conclude, with a President instructing his intelligence chief to keep it from the Secretaries of State and Defense.

The case has to be stated with care, because it was never proved and the innocent explanation was never ruled out. The honest verdict is the Commission’s own: it cannot confirm the theft and will not rule it out. What the record supports is narrower than a finding of guilt and heavier than a rumor. Credentialed nuclear regulators who read the files concluded that a theft of American bomb material by an allied government was the probable explanation, and the response of the United States, then and since, has been to leave the question open for sixty years with the files still sealed.

Sources

• National Security Archive (George Washington University), “The NUMEC Affair: Did Highly Enriched Uranium from the U.S. Aid Israel’s Nuclear Weapons Program?” edited by Roger Mattson, Avner Cohen and William Burr — declassified documents including the Helms-Clark correspondence and the record of the Rafi Eitan visit
• Victor Gilinsky and Roger Mattson, “Revisiting the NUMEC Affair,” Bulletin of the Atomic Scientists, 2010; and “Did Israel Steal Bomb-Grade Uranium from the United States?” 2014
• Roger Mattson, Stealing the Atom Bomb: How Denial and Deception Armed Israel (2016)
• Foreign Policy, “What Lies Beneath,” March 23, 2015 — the CIA environmental sampling at Dimona and the Johnson instruction
• Nuclear Regulatory Commission statement declining to clear Zalman Shapiro, 2009 (prompted by Senator Arlen Specter)
• U.S. Government Accountability Office review of the NUMEC investigations (declassified May 2010)
• U.S. Department of Energy inventory-loss records for NUMEC