ACA Logo

Avoiding Enrichment: Using Financial Tools To Prevent Another Khan Network

Geoffrey Forden

The existence of Abdul Qadeer Khan’s nuclear black market, which supplied entire uranium-enrichment plants to the Libyan nuclear weapons programs, has shocked the world. It should also cause a fundamental rethinking about the frightening speed, nature, and ways further proliferation might occur in the 21st century. Khan’s network, far from being the exception, is likely to prove closer to the rule for how states will try to acquire nuclear, chemical, or biological weapons and related delivery systems in coming years. Rather than slowly developing their own technology or getting assistance from other states, developing countries seeking such weapons can save time, money, and diplomatic capital by buying complete production facilities as well as the know-how to use them from all too many proliferation profiteers.[1]

Yet, although such acquisition pathways afford a state intent on gaining dangerous technology many advantages, they also present the world community with new opportunities for detecting and stopping state proliferation. Such proliferation schemes make use of the same financial instruments as legitimate international trade and leave a tell-tale trail that can be used to detect, stop, or inhibit the efficiency of this proliferation path. Current international nonproliferation regimes could be significantly strengthened if they were buttressed by a dedicated international auditing authority that would track down and investigate suspicious international deals.

Today we rely on customs agents in individual countries to spot the transfer of components for WMD production, an impossible task for individual countries. An international auditing authority, however, could use access to bills of lading, letters of credit, customs reports, and other international trading documentation to unearth the correlation of shipments among purchasers, manufacturers, and trading companies and reconstruct a complete picture of proliferation deals. Private banks wanting to do business with major industrialized countries could be required to grant auditing privileges to the international authority and to do business only with participating banks. Suspicious trade activity could be tagged and examined in detail, including underlying contracts, yielding further insight into suspicious shipments.

Nor do we need to start from scratch in establishing such an authority: the United Nations Monitoring, Verification and Inspection Commission (UNMOVIC) developed experience and expertise in these matters when monitoring Iraq’s commercial transactions regarding missile, chemical, and biological technologies.[2] Extending UNMOVIC’s mandate to include auditing suspicious letters of credit worldwide could appreciably improve the current nonproliferation regimes.[3]

If international investigators had these tools years ago, they might have fleshed out Khan’s activities far before his public confession in February 2004 or the October 2003 intercept of a cargo ship carrying centrifuge components to Libya for use in enriching uranium. That intercept gave U.S. intelligence officials the tangible proof they needed to show that Khan had been secretly selling Tripoli centrifuge technology that could help make essential fissile material for nuclear weapons, although Libya and nearly all of the half-dozen countries involved in the Khan network (with the exception of Pakistan) had acceded to the nuclear Nonproliferation Treaty (NPT).

How States Try to Acquire the Bomb: The Libyan Case
A state such as Libya, seeking to acquire a militarily significant stockpile of highly dangerous weapons, faces a different task than a terrorist group. In the past, terrorist groups have been satisfied with small quantities of either chemical or biological agents and have not had the capacity to build nuclear weapons.[4] States, on the other hand, are trying to acquire these weapons on an industrial scale and must, therefore, master efficient production techniques. Saddam Hussein’s Iraq, for instance, produced 3,000 tons of mustard gas[5] and 8,445 liters of concentrated anthrax.[6]

Hiring a consulting engineering firm to build a complete production facility or licensing the know-how from a proliferation profiteer has clear advantages for a country eager to acquire such weapons. First, it allows a proliferator to acquire production capability quickly and without research and development costs. Second, it is more reliable than other methods such as copying or reverse engineering imported items. Third, international proliferation profiteers can use their business contacts and networks—contacts not associated with a suspect state—to shield their actions from export controls. Finally, developing countries are often familiar with such deals from acquiring civilian technologies.[7]

Libya was apparently aware of these benefits in 1997 when it began its most serious bid to acquire nuclear weapons. That year, Tripoli’s representatives met with Khan and his associate, Buhary Syed Abu Tahir, about acquiring a gas centrifuge plant capable of producing weapons-grade enriched uranium.[8] (Tahir also allegedly facilitated Khan’s sale of used centrifuge units to Iran in 1994 and 1995 by arranging for their transport.)

It appears that Khan functioned less as a single supplier than the broker for a number of interrelated transactions: Khan supplied designs and specifications, but most of the hardware came from his contacts in other countries. For example, to help put together the gas centrifuges, 300 tons of aluminum tubes and bars were first purchased from a company in Singapore. These semi-finished pieces were then transported to Scomi Precision Engineering (SCOPE) in Malaysia. SCOPE produced several thousand roughly three-feet-long “stationary tubes” and other components. These components were on their way to Libya via a Dubai trading company when they were intercepted aboard the freighter BBC China in Italy. Other suppliers, operating from Germany, Switzerland, Turkey, and the United Kingdom, also provided electrical components, building designs, and advanced machinery to Libya.

It is clear that Libya was acquiring an entire enrichment facility and a machine shop capable of repairing broken units. Reports by the International Atomic Energy Agency state that Libya placed an order for 10,000 additional second-generation units, sufficient to enrich enough uranium for nearly 15 bombs per year. They also indicate that Khan’s associates were not only involved in the training of Libyan workers in the use of the advanced machinery but were likely to have given instructions in the workings of a centrifuge plant.[9]

Why were so many components allowed to be made and shipped? Why were customs officials not able to stop vital centrifuge components from leaving their countries? According to Malaysia’s inspector general, the Khan network was able to disguise shipment destinations and split the manufacture of critical components among different companies and even different countries. This obscured the magnitude of the operation. To quote the Malaysian report, “[W]ithout knowing the full or a significant portion of the total subassembly, no definitive [assessment] of the possible device may be made.”[10]

Applying the Auditing Authority to the Khan Network
To see how an international auditing authority might have fared against the Khan nuclear network, we need to make assumptions about some aspects of the network’s operations that investigators have so far been unable to pin down. First, an as-yet undisclosed precision engineering company other than SCOPE was producing the rotors for the Libyan centrifuges out of high-strength aluminum. This is a reasonable assumption since the aluminum that SCOPE imported from Singapore for its parts was a very common, low-strength variety and, therefore, not ideally suited for centrifuge rotors.[11] It is also consistent with the conjecture that the Khan network had components fabricated by different manufacturers to conceal the overall designs and hence their ultimate use.

To further obscure the trail, assume that the rotors or the requisite aluminum were not being procured by the Dubaibased SMB Group—the front company that was dealing with SCOPE—but by an import/export company in another country. Also assume that the rotors are shipped to an intermediate point of call, much like the SCOPE tubes were transshipped in Dubai.

In today’s world, such a scheme, with so many different companies and front companies, can be hidden from national customs agents. This is likely to still prove the case even if all states fully implement UN Security Council Resolution 1540. The measure, approved last year, calls on all states to take steps to improve their export control regimes in order to prevent terrorists from gaining access to nuclear, chemical, or biological weapons and related materials.[12] An international auditing agency, with a global view of trades, could more easily spot the significant connections between high-strength tubes and components purchased from other companies in other countries. It could do this by comparing both shipments’ bills of lading, letters of credit, and customs reports because they can track them to their ultimate destination and trace them back to their original purchasers.

In the case of Khan’s transshipment of SCOPE centrifuge tubes in Dubai, such an agency might have quickly detected the ultimate destination by noting that no customs reports were filed when they arrived at their first port. This is just one example of using a port’s material balance—adding up all the items coming into and leaving the port—to spot the diversion of goods. The international auditing agency could use such procedures to detect other tricks the proliferator might try, such as altering the tubes’ bill of lading during transshipment in Dubai to disguise their high-strength aluminum properties.

If these actions can be carried out in real time, the auditing agency can alert the relevant customs agencies and, if those agencies request expert assistance, could inspect the suspect shipments. If it proves impossible to correlate all the relevant pieces of documentation fast enough for such an intervention, the auditing agency can still perform a valuable service. It can alert the world to a potential proliferation problem allowing concerned states to initiate sanctions or other nonproliferation actions. The case for such sanctions will be much stronger and the debate more focused on policy rather than disputed facts because of the international auditing agency’s efforts.


Problems Still to Be Overcome

To be sure, there are practical and political problems associated with the auditing authority proposed here. On the political side, some might find a comprehensive international regime objectionable no matter what safeguards are taken to preserve confidentiality. Of course, this point of view ignores the fact that numerous individuals in international banking and insurance institutions already see this information. Nonetheless, some are sure to object to creating any international agency with broad oversight authority. If that view prevails, there will be nothing but voluntary associations of “supplier” countries to stem the tide of proliferation. Although these supplier regimes play an important role, they do not have a global view to detect components for complete facilities manufactured in different parts of the world.


If a truly functioning authority is to be created, some practical issues will also need to be addressed. In particular, auditing the world’s trade of manufactured goods could imply reviewing 20 million trades per year, an apparently monumental task.[13] Yet, modern information technology makes this feasible. Once the documents have been entered into the computer, an automatic pattern recognition program can flag questionable deals for attention by human analysts. Furthermore, the intelligence services from supporting governments can also call the auditing authority’s attention to specific trades, greatly simplifying the recognition problem. It should be remembered that UNMOVIC reviewed Iraq’s imports during the period of the Goods Review List.[14]

In fact, the largest problem is not analyzing all these records but obtaining them in electronic format. Currently, only a small but growing percentage of these trades use electronically formatted bills of lading and letters of credit. Therefore, the principal bottleneck will be scanning in the documents and converting them to computer-readable forms.

A final objection might be raised by analysts concerned that proliferators will simply bypass the international trade instruments that have been carefully worked out over the years to protect both trading partners. It is true that proliferators could bypass this nonproliferation regime in such a fashion. In fact, government-to-government proliferators can clearly avoid this regime altogether. Such proliferation is presumably handled exclusively through national banks that do not have the same financial incentives that private banks have to submit copies of letters of credit and other documentation to the auditing authority. Many of the recent examples of proliferation, however, have occurred with governments purchasing packaged technologies from private entities, which do not have their own banks and would hesitate to deal in cash. After all, both parties give up a considerable amount of security when they bypass international institutions, which makes procurement efforts that much more difficult. Many companies that would otherwise be willing to contribute to the production of weapons of mass destruction might not if they lacked the standard guarantees.

Iraq, even during the height of sanctions, did not completely give up standard trade mechanisms. Instead, the Iraq Survey Group has found that Iraq would either pay cash to intermediaries who signed letters of credit or used nomineenamed letters of credit to obscure that the goods were destined for Iraq.[15] Thus, a new proliferation auditing authority stands a good chance of either detecting and preventing proliferation or inhibiting proliferation by forcing proliferators to work outside all the established mechanisms for international trade.

Jump-Starting Proliferation: Examples of Purchasing Packaged Weapons Plants

When it comes to seeking weapons of mass destruction and related delivery systems, developing countries prefer to acquire packaged factories from foreign suppliers. Slower alternatives tend to be used only when that path is closed off.

Libya’s Uranium-Enrichment Program
Purchasing a gas centrifuge plant from Abdul Qadeer Khan was not Libya’s first attempt to buy a uranium processing plant. In the 1980s, Libya initiated an effort to procure a plant to produce uranium hexafluoride, the feedstock for enriched uranium. Uranium, when enriched to high enough levels, can be used as the explosive material in nuclear weapons.[1]

After two rounds of bids for a complete factory designed and constructed by foreign suppliers, during which time Libya must have examined detailed proposals, including engineering details, the regime of Moammar Gaddafi designed its own plant and then contracted out its production to a country in the Far East. Yet, Libya did not get the training for shop-floor technicians, which is often included in such deals. As a consequence, Tripoli never used the plant to produce the feedstock for its enrichment plants.

India’s Ballistic Missile Development Program
India’s missile development program was profoundly influenced by its purchases of packaged production technology from foreign sources. It started in the 1960s when India first sent scientists and engineers to NASA’s Wallops Flight Facility, where they learned the procedures for preparing, assembling, and launching solid-propellant sounding rockets for scientific experiments in the upper atmosphere.[2] Next, with substantial assistance from France and the United States, India established its own sounding rocket facility where imported sounding rockets were flown for many years. Soon, India licensed the Centaure 2B sounding rocket production technology from Sud Aviation in France.[3] As part of this deal, India received the complete designs and specifications for the rocket and the process for its production. At roughly the same time, Indian scientists and engineers worked in a French solid-propellant production facility, learning production details.

India was doing more than just using imported technology. It was assimilating that technology by experimenting with, for instance, new formulations for solid propellants. Nothing makes this assimilation clearer than India’s development of thrust vector control—a technology needed to progress from unguided sounding rockets to militarily useful guided missiles. To do this, however, India purchased from foreign suppliers a complete facility for static-testing its motors, including a key piece of equipment: a six-degree-of-freedom test stand. Only with such a test stand could India test its ideas and designs for steering the missile’s thrust during powered flight.[4] With such assistance, India was able to build a space-launch vehicle, the SLV-3, on which it based its first medium-range ballistic missile, the Agni.

Iraq’s Chemical and Biological Weapons Program
Iraq would have liked to purchase complete facilities for the production of chemical and biological weapons from abroad. Fortunately for the world, Hussein could not find a commercial dealer for such unprofitable specialized products. He did, however, do the next best thing: he purchased complete chemical and biological pilot plants and then tried to develop the recipes for his deadly brews.[5] Nevertheless, even with training in chemical plant design from the Karl Kolb Co. of Germany, Iraq was unable before the Persian Gulf War to convert a laboratory-scale process for making stable VX (the so-called Route D method) into the required industrial-scale process needed for militarily significant amounts of this deadly nerve agent.[6]

Iraq followed a similar acquisition path for its biological weapons. In 1987, Iraq began its first serious attempt to produce large quantities of anthrax and botulinum toxin when it transferred an imported pilot-scale fermenter and its controls from the Single Cell Protein plant—a project using bacteria to convert natural gas to foodstuffs for animals—at Al Taji to its biological weapons program.[7] Virulent strains of anthrax were imported from the American Type Culture Collection, although Iraq still conducted research on storage, growth factors, and virulence.[8] Then, in 1988, Iraq’s biological weapons program tried to purchase an even larger fermentation facility, this one with a 5 cubic-meter capacity, from Chemap of Switzerland, but the purchase was blocked because Chemap could not get an export license.[9] Instead, Iraq used a collection of smaller imported fermenters, the largest having a 450-liter capacity, to produce its biological agents.


1. International Atomic Energy Agency, “Implementation of the NPT Safeguards Agreement of the Socialist People’s Libyan Arab Jamahiriya,” GOV/2004/33, May 28, 2004.

2. Gopal Raj, Reach for the Stars: The Evolution of India’s Rocket Programme (India: Viking by Penguin Books, 2000).

3. Universite de Perpignan, “From Belier to Eridan,” available at

4. NASA, “Captive-Fired Testing of Solid Rocket Motors,” SP-8041 (March 1971).

5. “Iraq’s Chemical Warfare Program,” Comprehensive Report of the Special Advisor to the DCI on Iraq’s WMD, vol. 3, September 30, 2004, p. 6, available at

6. Ibid., p. 21.

7. “Biological Warfare,” Comprehensive Report of the Special Advisor to the DCI on Iraq’s WMD, vol. 3, September 30, 2004, p. 38.

8. Ibid., pp. 6, 21.

9. Ibid., p. 9.

Trading Documents for Goods

How does a company such as the Malaysian engineering firm SCOPE, which unwittingly produced parts destined for Libya’s gas centrifuge program, ensure payment? Why did the firm not demand advance payment? If it had, would the buyer, either the ultimate buyer, Libya, or Abdul Qadeer Khan’s middleman, Buhary Syed Abu Tahir, have been willing to advance SCOPE $3.5 million[1] and trust that the quality of the work would be as specified?

Of course, Tahir had established a business relationship with the company and the end user that helped smooth the way, but what if SCOPE, a fairly small company, could not make its payroll unless it was paid in a timely fashion? The answer to all these questions—the exchange of payments for documents and documents for goods and services—has been worked out over millions of legitimate international trades.[2]

When a buyer and a seller reach an agreement, the contract they sign will specify how and when the seller will be paid. A standard method that shares the risk equally between the buyer and seller involves guaranteed letters of credit and bills of lading that form the basis for much international trade.

First, the buyer arranges with his bank (the Libyan government, for instance, might use its national bank) for a letter of credit for the specified amount to be issued to the seller. The seller then presents the letter of credit to his local bank, which verifies that the buyer’s bank will pay the amount specified when a bill of lading—a document the shipper issues that describes the goods being shipped—is forwarded to the buyer’s bank. The letter of credit must specify the product or services sufficiently accurately so that the buyer feels comfortable with the seller’s bank committing the buyer’s money in exchange for a bill of lading that matches the description of goods in the letter of credit.

When the product is completed, the seller arranges for a shipper to pick it up and receives the bill of lading. Shippers confirm the contents of the bill of lading to at least a minimal extent to avoid cancellation of their insurance; insurance companies charge premiums for hazardous cargoes that might endanger a vessel.

The seller—SCOPE, in our example—then presents the bill of lading to his bank. If the description of the goods matches the letter of credit, the seller’s local bank pays the amount specified, recording the payment in its books as a loan to the buyer. The bill of lading is then exchanged with the buyer’s bank for payment. At that point, the buyer can pay his bank the amount he owes in exchange for the bill of lading. When the cargo arrives, the buyer can then use the bill of lading to claim his goods.

Documents describing the goods provided pass through multiple hands as the product and payments are exchanged. Other objective observers, such as insurance and customs agents, check the validity of these documents. Services, such as training of technicians or engineering design, also generate a similar trail, although the service equivalent of the bill of lading, sometimes referred to as certificates of completion, is often less descriptive.


1. Polis Diraja Malaysia, “Press Release by Inspector General of Police in Relation to Investigation on the Alleged Production of Components for Libya’s Uranium Enrichment Programme.”

2. Details of international trade, letters of credit, and bills of lading come from Edward G. Hinkelman, A Short Course in International Trade Documentation: The Documents of Exporting, Importing, Shipping and Banking (Novato, CA: World Trade Press, 2002); Edward G. Hinkelman, A Short Course in International Payments: Letters of Credit, Documentary Collections and Cyberpayments in International Transactions, 2d ed. (Novato, CA: World Trade Press, 2003).

Geoffrey Forden is a research associate with the Science, Technology, and Society Program at Massachusetts Institute of Technology. He served as chief of the multidisciplinary analysis section of the UN Monitoring, Verification and Inspection Commission (UNMOVIC).


1. Know-how is generally understood to mean “all inventions, trade secrets, technical information, data, shop practices, plans, drawings, blueprints, specifications and methods.” J. L. Enos and W. H. Park, The Adoption and Diffusion of Imported Technology: The Case of Korea (London: Croom Helm, 1988), p. 63.

2. Extending the mandates of the Organization for the Prohibition of Chemical Weapons and the International Atomic Energy Agency (IAEA), which already have mandates to inspect chemical and nuclear production facilities, to include auditing financial documents could also improve current nonproliferation regimes. Of course, there are significant practical advantages to having a single agency handle all such auditing. A single such agency avoids unnecessary replication of information technology infrastructure as well as reduces the burden on private industry in submitting duplicate paperwork. There are also more substantive reasons for having a single auditing authority: understanding a shipment’s end use will necessarily require a multidisciplinary approach. We can expect that the proliferation profiteer would do everything possible to conceal the shipment’s true usage. For instance, the Malaysian engineering firm that manufactured components for Libya’s uranium centrifuges believed their products were to be used for petroleum processing. Determining their ultimate usage would benefit from having a team of both chemical and nuclear analysts; a multidisciplinary approach that the UN Monitoring, Verification and Inspection Commission (UNMOVIC) is already familiar with.

3. See Paul Kerr, “New Reports Cite Looting at Iraqi Sites; UNMOVIC Future Discussed,” Arms Control Today, April 2005, pp. 36-37] for a discussion of the status of the debate surrounding UNMOVIC’s future.

4. Aum Shinrikyo used a total of six liters of impure sarin during its attack on the Tokyo subway on March 20, 1995. See Defence R&D Canada Suffield, “Aum Shinrikyo: Chemical and Biological Terrorism,” available at

5. UNMOVIC, “Unresolved Disarmament Issues: Iraq’s Proscribed Weapons Programmes” March 6, 2003, pp. 75-76.

6. Ibid., p. 95.

7. Linsu Kim, Imitation to Innovation: The Dynamics of Korea’s Technological Learning (Boston: Harvard Business School Press, 1997).

8. Polis Diraja Malaysia, “Press Release by Inspector General of Police in Relation to Investigation on the Alleged Production of Components for Libya’s Uranium Enrichment Programme.”

9. Among the documents and other evidence that the Libyan government presented to the IAEA was a hard drive containing a promotional film from Abdul Qadeer Khan showing the operations of a gas-centrifuge uranium-enrichment plant, indicating a willingness to train technicians for such a plants operation. IAEA, “Implementation of the NPT Safeguards Agreement of the Socialist People’s Libyan Arab Jamahiriya,” GOV/2004/33, May 28, 2004, p. 6.

10. Polis Diraja Malaysia, “Investigation on the Alleged Production of Components,” p. 9.

11. Marvin Miller, communication with author, Boston, Mass., March 17, 2005. Aluminum grade 7075, for instance, is almost 70 percent stronger than the more common 6061 aluminum that SCOPE used in producing its parts. Making the rotor out of this higher strength aluminum would have increased each unit’s separative power by up to a factor of five.

12. For more details, see Wade Boese, “U.S. Disappointed with Worldwide Response to WMD Resolution,” Arms Control Today, December 2004, p. 38.

13. The International Trade Administration, in the U.S. Department of Commerce, lists the imports of manufactured goods into the United States at approximately $1 trillion for 2002. Assuming a threshold value of $500,000 for each trade reported to the auditing authority implies 2 million imports into the United States each year. Assuming that U.S. imports make up about 10 percent of global trade implies 20 million trades per year worldwide. See

14. UN Security Council Resolution 1409 (2002).

15. Charles Duelfer, “Regime Finance and Procurement,” Comprehensive Report of the Special Adviser to the DCI on Iraq’s WMD, vol. 1, September 30, 2004, p. 48.