Where is North Korea’s Nuclear Program Heading?

Siegfried S. Hecker

[This article is an edited version of a report prepared by Dr. Hecker soon after his return from North Korea’s Yongbyon Nuclear Complex in November, 2010; we are grateful for his permission to run it. A related article appeared as a “Back Page” in APS News (March, 2011). Dr. Hecker’s full article and related reports can be found at his website. Dr. Hecker is a Professor (Research) of Management Science and Engineering, a Senior Fellow at the Freeman Spogli Institute for International Studies, and Co-Director of the Center for International Security and Cooperation, all at Stanford University. Trained as a metallurgist, he is regarded as one of the leading experts in the world on the properties of plutonium, and served as Director of the Los Alamos National Laboratory from 1986-1997 - Ed.]

On November 12, 2010, John W. Lewis, Robert Carlin and I visited North Korea’s Yongbyon Nuclear Complex. There we were shown a 25 to 30 megawatt-electric (MWe) experimental light-water reactor (LWR) in the early stages of construction, along with a modern uranium enrichment facility. This reactor is North Korea’s first attempt at LWR technology. These facilities appear to be designed primarily for civilian nuclear power as opposed to boosting North Korea’s military nuclear capability.

This visit allowed us to answer some questions about Pyongyang’s nuclear directions, but it also raised many more. In this article I describe our visit and offer some comments on how the response of the United States and its partners to these developments may help to shape whether Pyongyang will rely more on bomb development for diplomatic leverage or begin a shift toward nuclear electricity, which it desires both for economic and symbolic reasons.

Yongbyon Nuclear Scientific Research Center

This trip was my seventh to North Korea and my fourth to the Yongbyon complex, which is located about 90 km north of Pyongyang. During my first visit in January 2004 I was shown a sample of plutonium metal that had been reprocessed from spent fuel rods that had been stored since 1994 as part of the Agreed Framework, and which was subsequently used as bomb fuel for North Korea’s first nuclear test of 2006 [1]. During all of my previous trips to North Korea, government officials and technical specialists denied the existence of any uranium enrichment activities. Following their 2009 rocket launch and second nuclear test, Pyongyang expelled the U.S. technical team and international inspectors and declared that it would build its own light-water reactor (LWR) and produce its own fuel. For this visit, I asked to see the key nuclear sites in order to judge their current status and to see if the enrichment technology that they announced at that time was successful [2]. Our visit was supported by a number of foundations, including the Ploughshares Foundation, the Carnegie Corporation, and the MacArthur Foundation.

We were met by a small technical team and representatives of the General Bureau of Atomic Energy. The senior technical official gave the following introduction: “In the 1980s and 1990s, we agreed to give up our reactors for LWRs, 2,000 Megawatt-electric (MWe) by 2003. In the early 1990s we built 50 and 200 MWe reactors (of gas-cooled, graphitemoderated design). Now they have become ruined concrete structures and iron scrap. We have not been able to contribute to the national demand for electricity. So, we decided to make a new start. For us to survive, we decided to build our own LWR. On April 15, 2009, the Foreign Ministry stated that we will proceed with our own LWR fuel cycle … Our nuclear program has not proceeded as expected, we have not delivered electricity and that has impacted the economic condition of our country. We will use our economic resources to solve the electricity problem. We are willing to proceed with the Six-Party Talks and the September 19, 2005 agreement, but we cannot wait for a positive agreement. We are trying our best to solve our own problems. We will convert our center (Yongbyon) to a LWR and pilot enrichment facility. It is a high priority to develop uranium enrichment. We will have some difficulties with this, but we are proceeding with the LWR fuel cycle. We have designated a site for the LWR and also for uranium enrichment – it is the first stage, so it is first priority. The construction is completed and the facility is operational. You will be the first to see this facility.” [3]. Unlike during my previous visits to Yongbyon, the technical team clearly had instructions to show us only the basics at two facilities and answer a minimum of questions. We were hurried along at every stage but eventually spent 3.5 hours at the site before lunch. The chief engineer of the 5 MWe reactor showed us the site and answered questions, but only when pressed.

Experimental 25 to 30 MWe LWR construction

At the 5 MWe reactor site we were taken to a construction site that had been identified previously from overhead imagery [4]. A large excavated pit roughly 40 meters by 50 meters by 7 meters deep containing a concrete foundation 28 meters square with round concrete preforms for the reactor containment vessel was visible. The containment vessel is designed for a power level of 100 MW (thermal) and was about one meter high at the time we saw it. We were told it will be 22 meters diameter, 0.9 meters thick and 40 meters high. We were not told the electrical power but were informed that the conversion efficiency is typically 30 percent, so I estimate the electrical power to be 25 to 30 MWe.

This power level is much smaller than the two 1,000 MWe LWRs that were to have been constructed as part of the KEDO project at the Kumho (North Korea) site [5]. They explained that the LWR design is different from their experience base of gas-graphite reactors and so they are building a small prototype first. Once they have mastered this technology they will build a bigger LWR. However, even with the 25 to 30 MWe reactor they will build two electrical generators that will supply electricity to the local communities and be hooked into the national grid. The chief engineer said that construction was started on July 31, 2010, and that the target date for operations is 2012. This seemed to us unreasonably optimistic, but coincides with the centenary of Kim Il-sung’s birth. There were nearly 50 workers on the floor. We enquired about reactor safety analysis and practices; they claimed to have excavated down to bedrock and to have performed seismic analysis of the site.

The pressure vessel will be fabricated out of high-strength steel. The chief engineer said that they will be able to manufacture it and all other reactor components domestically. I asked if they have a nuclear regulatory agency; the response was that the National Nuclear Safety Commission has oversight and has inspectors on-site.

The reactor will be fueled with uranium dioxide fuel enriched to 3.5%, typical of LWR fuel but very different from the metallic uranium-alloy fuel rods used in the gas graphite reactor. A full load of fuel will comprise four metric tons of uranium. We were told that North Korea has ample domestic uranium ore resources. They were not certain what cladding material would be used, stating that they are still working on many of the details. The reactor design team is a young group without reactor design experience, but they are being mentored by the experienced gas-graphite reactor designers. The new designers are in their 40s, graduated from North Korean universities, and have spent their careers at Yongbyon. They have not brought any of the North Korean KEDO LWR team members to Yongbyon at this time but may do so for the operational phase. (The KEDO team did not necessarily have design experience because the reactors to be provided were to come from South Korea.) I had expected the old design team from the gas-graphite reactors to be involved. My hosts said that they specifically tasked a fresh design team because this was a new technology.

Uranium enrichment facility

At the fuel fabrication plant we entered a building that appeared to be a new but which we identified later as the former metal fuel rod fabrication building which I had visited in February 2008 to verify disablement actions. The view through the windows of the second-floor observation deck into two long high-bay areas was stunning. Instead of seeing the few small cascades of centrifuges which I believed to exist in North Korea, we saw a modern, clean plant of more than a thousand centrifuges all neatly aligned and plumbed. We were told that construction had begun in April 2009 and was completed a few days before our visit. Overhead imagery now shows a building with a blue roof about 120 meters long (see http://www.globalsecurity.org/wmd/world/dprk/yongbyonffp- imagery-02.htm).

We estimated the centrifuges to be about 8 inches in diameter and approximately 6 feet high. They looked like smooth aluminum casings (no cooling coils visible) with three small stainless steel tubes emanating from the top to central plumbing that ran the length of the facility. In response to persistent questioning, the chief process engineer told us that the facility contained 2,000 centrifuges in six cascades [6]. He would not provide us with the physical dimensions, stating that the United States would also not release such proprietary information. When asked if they were Pakistani P-1 centrifuges, he said no [7]. When pressed, he said the rotors were made of alloys containing iron, which likely makes them P-2 models [8]. He claimed all components were manufactured domestically, but modeled after the centrifuges at Almelo (a URENCO facility in the Netherlands) and Rokkasho-mura (a Japan Nuclear Fuels facility). We were able to extract the most important detail, the enrichment capacity, which he said was 8,000 kg SWU/year [9]. With this capacity North Korea could produce up to 2 tonnes of LEU per year, or, if the cascades were reconfigured, up to 40 kg HEU.

The control room was astonishingly modern. Unlike the reprocessing facility and control room for the gas-graphite reactor, which looked like 1950s U.S. or 1980s Soviet instrumentation, this control room would fit into any modern American processing facility. Five large panels had numerous LED displays of operating parameters. Computers linked to flat-screen monitors displayed flow diagrams and numbers, but we were ushered past too quickly to tell what they signified.

I expressed surprise that they were apparently able to get cascades of 2,000 centrifuges working so quickly, and asked again if the facility is actually operating now. We were given an emphatic “yes”. We were not able to independently verify this, although it was not inconsistent with what we saw. We attempted to probe more deeply into their claims of indigenous fabrication but received no concrete answers. The technical official claimed that they produce uranium hexafluoride as feed material for gas centrifuges, material which they had never admitted having produced in the past [10]. I also asked again about the fuel – will it be UO2 and how will they make it? He said that the process for learning how to make UO2 was difficult but had begun, and confirmed that they are currently enriching uranium in the facility. When I pointed out that the outside world will be concerned about their ability to convert the facility to make HEU, he stated that anyone can tell by looking at the monitors in the control room that the cascades are configured for LEU. Besides, he said, they can think what they want.

Status of existing plutonium production facilities

We were not taken to the plutonium production facilities, but the 5 MWe reactor, which is adjacent to the LWR construction site, appeared dormant. We were told that it is in stand-by status with regular maintenance. We were reminded that the cooling tower had been destroyed (June 2008), but the chief engineer was confident that they could restart the reactor should they decide to do so; I estimate that it would require approximately six months to do so. We were told that fresh fuel which could be used to refuel the reactor was still stored in the same warehouse in which I last saw it in 2008. The 50 MWe reactor, which was near completion in the mid- 1990s but abandoned during the Agreed Framework, was being dismantled with large cranes. No activity was apparent at the reprocessing facility as we drove past it.

To summarize: The 5 MWe reactor has not been restarted since it was shut down in July 2007. The spent fuel rods were reprocessed following North Korea’s termination of the Six- Party talks in April 2009. No new fuel has been produced and the fresh fuel produced prior to 1994 (sufficient for one more reactor core) is still in storage. Pyongyang has apparently decided not to make more plutonium or plutonium bombs for now. My assessment is that they could resume plutonium operations within approximately six months and make one bomb’s worth of plutonium per year for some time to come.


The findings from this trip answer many questions about the direction of North Korea’s nuclear program, but they also raise at least as many. I will give a preliminary analysis here.

The plutonium program associated with the now shutdown 5 MWe graphite reactor remains frozen, and has perhaps even taken another step backward. They have converted the metal fuel rod fabrication facility into the centrifuge cascade halls, thereby making it more difficult to make fuel for the plutonium production reactor. The LWR will produce plutonium, but it is much less suitable for bombs than that from the 5 MWe reactor. In addition, the reprocessing facility operations would have to be reconfigured to reprocess the LWR fuel. My previous estimate of the North Korean plutonium inventory from its 5 MWe reactor of from 24 to 42 kilograms (sufficient for four to eight primitive nuclear weapons) still stands [11].

A North Korean uranium enrichment program has long been suspected. I believe that they started early, perhaps in the 1970s or 1980s, but did not accelerate the effort until their dealings with A.Q. Khan in the 1990s [Editor’s note: Khan is regarded as the father of the Pakistani nuclear-weapons program. He was involved with proliferating smuggled URENCO centrifuge technology, and has been accused of proliferating weapons technology to Iraq, Libya, Iran, and North Korea.] However, the 2,000-centrifuge capability significantly exceeds my estimates and that of most other analysts [12]. We were not able to confirm that the facility is fully operational. It typically requires much time to bring cascades of this size into full operation [13]. Nevertheless, they have either done so or are most likely capable of doing so shortly. The LEU capacity is consistent with the requirements of the LWR under construction. It would have to be expanded significantly if North Korea builds a larger LWR in the future. Whether LEU or HEU is produced in the facility is easy to monitor with on-site presence or on-site instrumentation. However, the greatest concern is that a facility of equal or greater capacity configured to produce HEU exists somewhere else. Such a facility would be difficult to detect as demonstrated by the fact that this facility was undetected in the middle of the Yongbyon fuel fabrication site. The only factors that would limit North Korea’s ability to build more are the procurement or production of specialty materials and pieces of equipment such as maraging steel, high-strength aluminum alloys, ring magnets, frequency converters, special bearings, vacuum equipment and flow meters. We have little knowledge of the North’s indigenous fabrication capabilities. If North Korea claims its uranium program is strictly peaceful, then the burden of proof is on it, especially since they continued to deny it during the Six-Party negotiations. [Editor’s note: maraging steels are ultra-high-strength steels that derive their strength from precipitation of intermetallic compounds containing Ni, Co, Mo, and Yi. They are very machinable and weldable, and suffer little dimensional change after being heat-treated. They are used in products as diverse as rockets, fencing blades, golf clubs, and, centrifuges.]

One of the most puzzling issues is how they got this far? Albright and Brannan recently presented a detailed analysis of the status of North Korea’s uranium enrichment program [4]. Their work demonstrates a clear pattern of cooperation and exchange with Pakistan, including crucial elements such as training of North Korean technical specialists at the Khan Research Laboratory. They also show a troubling procurement scheme, particularly with commercial entities in China. I have previously stated my concern about potential cooperation and exchanges in uranium technologies between North Korea and Iran. However, a detailed analysis and reevaluation taking into account the findings from this trip is now in order. A better understanding is important because it will help us better judge the capacity of current and planned enrichment capacity.

It is an understatement to say that trying to discern North Korea’s motivations is difficult. In an essay published elsewhere I argued how an initially security-driven motivation for the bomb took on important domestic and international dimensions [11]. Pyongyang has clearly stated that it will retain its nuclear weapons as a deterrent so long as current U.S. policies persist. North Korean officials with whom we met on this trip made it abundantly clear that there will be no denuclearization without a fundamental change in U.S. – North Korean relations. Pyongyang has seriously pursued nuclear electricity; it has both practical and symbolic importance [14]. It views LWRs as the modern path to nuclear power, and was prepared several times in the past to trade its bomb-fuel producing reactors for LWRs. This time we were told, “We have given up; we will do it on our own.” They can claim with some justification that the uranium enrichment program is an integral step toward an LWR and nuclear electricity.

I believe that although this peaceful program can be diverted to military ends, the current revelations do not fundamentally change the present security calculus of the United States or its allies. Pyongyang has gained significant political leverage already from the few plutonium bombs they have. Building more sophisticated bombs that can be mounted on a missile is better done with plutonium than HEU. However, the production of large quantities of HEU and additional nuclear tests would allow them to increase the size of their arsenal. Even more troubling would be the potential of export of fissile materials or the means of producing them, means which now include centrifuge technologies. For these reasons, the United States should not sit idly by.

Where do we go from here?

Is Pyongyang really pursuing a modern nuclear electricity program? If so, what are its chances of success without outside help? Have they decided to abandon their plutonium production complex or at least keep it dormant? Do they have additional centrifuge facilities that could be dedicated to producing HEU bomb fuel? How did they acquire centrifuge technology at such a level of sophistication and when? Why did Pyongyang decide to show us the facilities now and how does this fit into their broader strategy of how to deal with their domestic and international challenges?

Much more work will have to be done by many more analysts to address these questions. One thing, however, is certain: these revelations will cause a political firestorm. Some will use them to prove that Pyongyang cannot be trusted. Some will use them to justify the October 2002 U.S. decision to confront Pyongyang about uranium enrichment, a confrontation which resulted in termination of the Agreed Framework. Some, most likely China and Russia, will claim that North Korea is within its sovereign rights to develop nuclear energy. The issue is complicated by the inherently dual-use nature of nuclear technology. It is possible that Pyongyang’s latest moves are directed at generating much-needed electricity. Yet, the military potential of uranium enrichment technology is serious. Waiting for Pyongyang to return to the Six-Party talks on terms acceptable to the United States and its allies will only exacerbate the problem. A military attack is out of the question. Tightening sanctions further is likewise a dead end, particularly given the advances made in their program and the economic improvements we saw in general in Pyongyang. The only hope appears to be engagement. The United States and its partners should respond to these latest developments so as to encourage Pyongyang to finally pursue nuclear electricity in lieu of the bomb. That will require addressing North Korea’s underlying insecurity. A high-level North Korean government official told us that the October 2000 Joint Communiqué, which brought Secretary Madeleine Albright to Pyongyang, is a good place to start [15].

Notes and References

[1] The “Agreed Framework” referred to in Dr. Hecker’s article was signed in October 1994 between North Korea and the United States. The agreement provided for replacing North Korea’s already-operating graphite-moderated 5 MWe reactor and associated plutoniumprocessing plant and 50 MWe and 200 MWe reactors then under construction with two 1000 MWe Light-Water Reactors (LWRs) by 2003, and for normalization of relations between the two countries. The agreement was to be implemented by the Korean Peninsula Energy Development Organization (KEDO), a consortium of several nations including the United States. Construction of the first LWR did not begin until August, 2002. However, in October, 2002, a U.S. delegation confronted North Korea with a U.S. assessment that the North Koreans had a uranium enrichment program. In January 2003, North Korea announced its withdrawal from the Nuclear Non-Proliferation Treaty. In December, 2003, KEDO suspended work on the LWR project, and terminated the project altogether in May, 2006. North Korea tested nuclear weapons in October, 2006, and May, 2009. However, they did not restart work on the two reactors that were frozen under the agreement, and the plutonium-producing 5 MWe reactor was shut down in July, 2007. A detailed description of the Agreed Framework can be found at the website of the Arms Control Association.

Their site also hosts a detailed chronology of North Korean Nuclear developments from 1985 to the present. The Joint Communiqué referred to at the end of the article is also described at this website. Secretary of State Madeleine Albright visited North Korea in October, 2000, and the communiqué was issued on October 12 of that year. The relevant portion reads: “The two sides stated that neither government would have hostile intent toward the other and confirmed the commitment of both governments to make every effort in the future to build a new relationship free from past enmity.” A January 7, 2003 joint statement from the United States, Japan, and South Korea reaffirmed this commitment in writing, stating that the United States “has no intention of invading North Korea.”- Ed.

[2] In a September 4, 2009 letter to the President of the UN Security Council, the North Korean permanent representative to the United Nations stated that North Korea’s “experimental uranium enrichment has successfully been conducted to enter into completion phase.” (Korean Central News Agency – KCNA).

[3] This quote is based on our notes of the interpreter’s version of the technical official’s comments. The September 19, 2005 date refers to the Joint Statement of the Fourth Round of the Six-Party Talks.

[4] David Albright and Paul Brannan, “Taking Stock: North Korea’s Uranium Enrichment Program

[5] The Korean Peninsula Energy Development Organization (KEDO) project was to have the United States and other parties build two 1,000 MWe LWRs for North Korea at the Kumho site. It was established in March 1995 as part of the Agreed Framework. The LWR project was terminated in 2006.

[6] The chief process engineer told us at the outset that they did not want to show us this facility, but their superiors told them to do so. Consequently, they showed us as little as possible, did not volunteer any information, and hurried us along as much as possible.

[7] The P-1 designation refers to the Pakistani design copied from the least advanced URENCO centrifuges. These contain high-strength aluminum alloy rotors and high-strength aluminum alloy casings.

[8] P-2 centrifuges were based on the German G-2 design developed as part of the URENCO consortium. These typically have high-strength steel rotors that can by spun much faster than the aluminum rotors, thereby increasing the throughput.

[9] The kg SWU is an acronym for kg of separative work units. It refers to the amount of isotope separation achieved (separating the fissile U-235 isotope from the non-fissile U-238 isotope).

[10] In spite of prior North Korean denial, the nuclear materials recovered in Libya in 2003 when Col. Gadaffi relinquished his nuclear weapons program were reputed to include a shipment of uranium hexafluoride from North Korea.

[11] Siegfried S. Hecker, “Lessons learned from the North Korean nuclear crises,” Daedalus, Winter 2010, pp.44-56. [12] For example, see Hui Zhang, “Assessing North Korea’s uranium enrichment capabilities,” Bulletin of the Atomic Scientists, 18 June 2009. See also Ref. [4].

[13] It took Iran about 20 years to procure, build and operate cascades of this size. Iran has roughly 8,000 centrifuges installed with 4,000 P-1s working for a total capacity of ~ 4,000 kg SWU.

[14] Siegfried S. Hecker, Sean C. Lee and Chaim Braun, “North Korea’s Choice: Bombs over Electricity, “The Bridge, Vol. 40, No. 10, Summer 2010, pp. 5-12.

[15] For some further analysis of the situation, see http://www.foreignaffairs.com/articles/67023/siegfried-s-hecker/whati- found-in-north-korea and http://www.thebulletin.org/web-edition/features/redefiningdenuclearization- north-korea-0

Siegfried S. Hecker
Center for International Security and Cooperation
Stanford University

These contributions have not been peer-refereed. They represent solely the view(s) of the author(s) and not necessarily the view of APS.