Where Those Reactors and Centrifuges Came From

By Jeremy Bernstein, the author of the forthcoming “Plutonium: A History of the World’s Most Dangerous Element” (THE NEW YORK TIMES, 10/03/07):

THE six-party agreement signed with North Korea last month should certainly be applauded as a necessary first step in improving relations with the United States. While a good deal of the North Korean program is shrouded in mystery — just this week the United States again urged the North Koreans to disclose any uranium-enrichment activities — there are some things we do know, including the nature and status of the country’s reactors.

North Korea’s one functioning reactor, at Yongbyon, uses natural uranium for fuel and graphite as its moderator (the substance that slows the neutrons and enhances the fission reaction). These are the same ingredients used in the first reactor ever designed, which was tested by Enrico Fermi at the University of Chicago in 1942. The best estimate is that Yongbyon has produced about 100 pounds of plutonium since it went into full operation in 1990. This is enough for six to eight nuclear bombs, depending on their design. (The North Koreans might have used about six kilograms in their Oct. 9 test.) The construction of the larger reactors North Korea was building was apparently already suspended, for various technical reasons, before the agreement.

The North Koreans have been fairly transparent about their reactor program but almost totally opaque about their program to make natural uranium suitable for nuclear weapons by using centrifuges. We know that there is such a program, but we do not know where it is or how much, if any, uranium it has enriched. Centrifuges are much easier to hide than reactors.

The provenance of the North Korean centrifuge program is a useful lesson in nuclear proliferation. One can trace it back to the spring of 1945, when the Russians were overrunning Germany. Along with the army came a cadre of atomic and nuclear physicists who were looking for both German physicists and metallic uranium.

The latter had been made in large quantities — tons — by the Auer company, a subsidiary of the Degussa chemical company, in part by using slave labor from the concentration camps. The Soviets were able to take home about 300 tons of processed uranium.

Thanks to espionage, the Soviets knew where to look and whom to look for. (The United States had a similar program, called Alsos, that competed for many of the same people.) The Soviets collected a talented inventor of electronic devices named Manfred von Ardenne. He had made a great deal of money and had a large estate outside Berlin. On it he had a laboratory with a nuclear program financed by the German Post Office.

In May 1945 the Soviets shipped Dr. von Ardenne east with some of his colleagues and equipment from his laboratory. By June he had set up a laboratory, Institute A, in Sukhumi on the Black Sea in Georgia. Nearby, another laboratory, Institute G, had been set up by Gustav Hertz, a German physicist of Jewish ancestry who had shared the 1925 Nobel Prize in Physics. Dr. Hertz had been working out of sight at the Siemens company during the Nazi period.

The Sukhumi scientists were ordered to find methods of separating uranium isotopes. Dr. Hertz chose to study gaseous diffusion. Uranium hexafluoride gas is forced through tiny pores in a membrane to separate out the lighter isotope, uranium 235, which is needed for weapons. Dr. von Ardenne tried separation by using electromagnetic fields, a technique also used in the American uranium separation program at Oak Ridge, Tenn.

A third group, headed by a physicist named Max Steenbeck, investigated the centrifuge. Dr. Steenbeck, who had been arrested by the Soviets and put in a concentration camp in Poland, had previously been in charge of research for the division of Siemens that dealt with aircraft. While in captivity he wrote a letter to the Soviet secret police, the N.K.V.D., explaining his scientific background; he also ended up in Sukhumi. Dr. Steenbeck began with a small group and some antiquated Soviet centrifuges that certainly could not have been used to separate uranium isotopes.

In the summer of 1946 they were joined by an Austrian physicist named Gernot Zippe. Dr. Zippe had been in the Luftwaffe during the war and, after having been taken prisoner in the summer of 1946, he went from a prison camp to the relative luxury of Sukhumi, thanks to the initiative of Dr. von Ardenne. Neither Dr. Zippe nor Dr. Steenbeck had ever worked on centrifuges, but within two years they created the best centrifuge in the world — although at the time they did not know it. (To give some idea of its capacity, a typical laboratory centrifuge makes a few thousand rotations a minute. The Zippe centrifuge — this is the common name, although Dr. Zippe himself refers to it as the “Russian centrifuge” — can do 90,000 rotations a minute.)

In 1956, Dr. Zippe was allowed to return to Germany. Although he was not permitted to take any documents with him, he was able to reconstruct his work, and began consulting for various companies interested in centrifuges, including Degussa.

The private German companies, including the part of Degussa that was doing centrifuges, became nationalized in 1964. But in 1970 these national companies became part of an international consortium called Urenco. The Dutch had a branch in Almelo and, in 1972, a Pakistani metallurgist named Abdul Qadeer Khan joined it. Fluent in both Dutch and German, he was given the job of translating the German centrifuge plans into Dutch. He became familiar with both the German and Dutch versions of the Zippe centrifuge.

In 1974, India successfully tested a nuclear device, and Pakistan’s president, Zulfikar Ali Bhutto, put out a call to all the scientists in the Pakistani diaspora to return home and help make a bomb. Dr. Khan was one who answered and he brought with him the stolen plans for the Zippe centrifuge. This is not the place to go into the details of Dr. Khan’s activities, which in the end involved a variety of countries from Libya to China — to say nothing of Iran, whose centrifuges also have a Pakistani origin.

By the 1990s Dr. Khan was exchanging weapons information with the North Koreans for similar information about their long-range rockets. We know he gave them plans for the centrifuge and probably sample centrifuges. We do not know whether he gave them plans for a nuclear weapon, as he had done for the Libyans.

We also do not know to what extent the government of Pakistan was complicit in this. The army certainly was, and military aircraft were used to transport material. Pakistan has denied any involvement; Dr. Khan is under house arrest and no foreign intelligence representatives have been allowed to interview him.

The North Koreans have reluctantly admitted that they have a centrifuge program but have not let any foreign observers see it. Such a program, if limited, would have been allowed by the Nuclear Nonproliferation Treaty. But the North Koreans would have had to declare it to the International Atomic Energy Agency, which then would have had the right to inspect it.

This they did not do. Perhaps they enjoy the ambiguity. My own guess is that if they have an active program it is relatively small. And while so far the agreement we have made with them does not say anything about this program, clearly we must eventually insist on knowing its extent. The route that led from Soviet prisoners of war to the centrifuges in North Korea is so implausible that if one put it in a novel, no one would believe it.