War of Nerves

by Jonathan Tucker

  In August 1949, the Soviet Union successfully tested an atomic bomb on the remote steppes of Kazakhstan, achieving a nuclear weapons capability years earlier than Western intelligence services had predicted and radically shifting the global balance of power. Given the transformed strategic situation, U.S. policy makers were forced to reevaluate their heavy reliance on nuclear deterrence. If it was no longer possible for Washington to threaten the use of atomic weapons without inviting mutual destruction, what other weapons could help win a future global conflict?

  At this juncture, the Chemical Corps saw an opportunity to regain some of its lost influence. Stressing the need to reinforce nuclear deterrence by other means, the Corps called for an end to the “retaliation-only” chemical warfare doctrine that had been in effect since 1943, when President Roosevelt had stated that the United States would “under no circumstances resort to the use of such weapons unless they are first used by our enemies.”

  On the afternoon of February 1, 1950, President Truman met in the Cabinet Room of the White House with the members of his National Security Council to discuss U.S. chemical warfare policy. Secretary of Defense Louis Johnson gave an oral presentation and distributed a memorandum from General Omar N. Bradley, the chairman of the Joint Chiefs of Staff, arguing against a change in the “retaliation-only” posture. “While the United States must be at all times prepared for the initiation of gas warfare by our enemies—the time, place and purpose of initiation to be chosen by them—it is doubtful if the United States Government should adopt a policy of unrestricted gas warfare excepting in retaliation,” General Bradley wrote. Such a policy shift, he explained, would be unacceptable to the European NATO allies because of their “vulnerable and comparatively defenseless position” and would also be opposed by the American people, who remembered the horrors of gas warfare in World War I.

  Swayed by General Bradley’s arguments, President Truman reaffirmed the “retaliation-only” chemical doctrine in policy memorandum NSC-62 of February 17, 1950. This decision was a setback to the ambitions of the Chemical Corps, which had sought to remove any constraints on the acquisition and use of chemical weapons by the armed forces. Nevertheless, the U.S. Army moved forward with plans to acquire a stockpile of Sarin nerve agent. The Chemical Corps Technical Committee had decided in June 1949 that Tabun (GA) was inferior to Sarin and should not be mass-produced, although the Tabun-containing bombs and shells confiscated from Germany were still in good condition and would be retained as an emergency war reserve.

  The next step was to procure a stockpile of Sarin. Major General Anthony C. McAuliffe, the chief of the Chemical Corps, had become famous during World War II as the commander of U.S. forces at Bastogne who had said “Nuts” when the Germans told him to surrender. In a secret memorandum dated April 14, 1950, he estimated the U.S. stockpile requirement for Sarin at 48,000 tons. If the Army built a production facility with a capacity of 25 tons per day, it would have to operate continuously for five and a half years to produce that quantity of agent and could then be placed on standby status. In wartime, the estimated military requirement for Sarin would be 2,000 tons per month, requiring two additional production plants.

  In late April 1950, General McAuliffe gave a speech at a meeting in Detroit of the American Chemical Society in which he revealed publicly for the first time that the U.S. Army was working on a new generation of chemical warfare agents that attacked the nervous system. “Our use of them would be purely retaliatory,” he explained. “It is a well-known fact that many German scientific experts on toxic chemical warfare are being exploited by Soviet Russia. It must be assumed, therefore, that we are not the sole possessors of the offensive and defensive secrets of the new nerve gases.”

  After General McAuliffe’s disclosure, the nature and composition of the nerve agents became a topic of intense speculation among journalists and armchair military strategists. Because of the dense veil of secrecy surrounding the new weapons, misconceptions were widespread, such as the belief that nerve agents caused temporary incapacitation rather than death. According to a May 1950 article in Time magazine, “Presumably [nerve gas] would be sprayed over enemy cities by planes in the same way that whole areas are sprayed with mosquito-killing DDT, paralyzing the whole population. Then the attacking army, equipped with protective masks, would march in and take over.”

  ALTHOUGH THE U.S. nerve agent program was proceeding at a deliberate pace, world events suddenly transformed the situation overnight. At approximately 4:00 a.m. on June 25, 1950, the North Korean Army began to fire artillery and mortar shells at South Korean military positions south of the 38th Parallel, which marked the border between the two countries. Soon massive columns of North Korean tanks and infantry poured across the demarcation line at multiple points, and at 11:00 a.m., Pyongyang issued a formal declaration of war. Responding decisively to the Communist surprise attack, President Truman quickly organized a military intervention in Korea under United Nations auspices.

  On June 30, 1950, only five days after the North Korean invasion, a group of civilian advisers to the Pentagon called the Ad Hoc Committee on Chemical, Biological and Radiological Warfare submitted its report. The panel was chaired by Earl P. Stevenson, president of the consulting firm Arthur D. Little, Inc. Rejecting the notion that poison gas was uniquely immoral or inhumane, the Stevenson committee argued that chemical arms might be “exceedingly important as a supplement to weapons now in general use for holding back the advance of enemy ground forces.” Although the timing of the report was fortuitous—Secretary of Defense Louis A. Johnson had appointed the blue-ribbon panel back in December 1949—the sudden outbreak of war in Korea greatly enhanced its political impact.

  The Stevenson committee concluded that the United States lagged far behind the Soviet Union in chemical warfare capabilities. Whereas intelligence reports indicated that the Soviets had captured entire German factories for the manufacture of nerve agents, the United States possessed only limited stocks of mustard and phosgene and had not yet begun to produce nerve agents or suitable delivery systems. The committee blamed Roosevelt’s “retaliation-only” policy for putting the United States in a position of dangerous inferiority vis-à-vis the Soviet Union. “Such a policy has resulted in the assignment of low priorities to the research, development, and production of chemical weapons,” Stevenson wrote in his cover letter. “The security of the United States demands that the policy of ‘use in retaliation only’ be abandoned.”

  The Stevenson committee’s recommendations sparked an intense debate within the Joint Chiefs of Staff. Some senior military officials favored switching to a policy of chemical first use, both to deter the Soviet Union and to counter human-wave infantry attacks in Korea. They warned that the Chinese Communists led by Mao Zedong, who had seized power on the mainland in 1949, represented a new “Yellow peril” that might intervene on the North Korean side. Nerve agents, they argued, would be highly effective against the Chinese People’s Army, a technically backward force that lacked modern protective gear and whose chief asset was large reserves of manpower.

  Although both the U.S. Army and Air Force favored scrapping the retaliation-only policy, the Navy opposed an expanded role for chemical weapons and strongly defended the status quo. Aware that the Army wanted the other services to share the burden of delivering chemical weapons, Navy officials were concerned about the problem of storing nerve agent munitions on aircraft carriers and other warships. Because the cramped quarters of a ship at sea provided nowhere to run, a single leaking chemical bomb would be disastrous for the crew. For this reason, the Navy resisted a change in chemical warfare posture and ultimately prevailed in the internal Pentagon debate.

  On September 7, 1950, the Joint Chiefs informed the Office of the Secretary of Defense that they accepted all of the Stevenson committee’s recommendations except for the proposed change in U.S. chemical warfare doctrine. Not only did the United States lack the stockpiles and delivery systems needed to employ nerve agents o
n a large scale, but Great Britain supported the “retaliation-only” policy and a unilateral change in the U.S. posture would risk alienating America’s closest ally. General Bradley concluded that improved preparedness to conduct chemical warfare “can and must be achieved under a policy of retaliation-only” and suggested that any consideration of a change in doctrine be “deferred pending further developments.” On October 27, 1950, the new Secretary of Defense, General George C. Marshall, signed off on the JCS position.

  Despite this decision, the advocates of chemical warfare achieved most of their objectives. Without challenging the “retaliation-only” policy directly, Secretary of Defense Marshall directed the Pentagon to implement all the other recommendations in the Stevenson report, including actions needed “to make the United States capable of effectively employing toxic agents at the outset of a war.” During the fifteen months following the outbreak of the Korean conflict, the research-and-development budget of the Chemical Corps (which also included smoke and incendiary munitions) tripled in size, and the number of researchers grew from 2,100 to 3,700. Several private companies and universities also received government contracts to perform related R&D. To support this expanded effort, the Pentagon authorized the open-air testing of advanced chemical weapons and delivery systems, which required a vast amount of open space far from populated areas. During the summer of 1950, the Army reactivated and expanded Dugway Proving Ground, a chemical and biological testing site in the Utah desert that had been established in 1942 and placed on standby status after World War II.

  THE PRIMARY TASK facing Edgewood Arsenal was the design and construction of a full-scale Sarin manufacturing plant based on the German DMHP process and drawing on the technical details provided by Schieber and his team of Falkenhagen scientists. On October 31, 1950, Secretary of Defense Marshall, using contingency funds available to him, secretly authorized $50 million for the initial design, engineering, and construction of the Sarin plant under the code name “Gibbett.” The Chemical Corps established a task force to oversee all aspects of the project, including procurement, funding, security, and administration. Kellex Corporation (which later changed its name to Vitro Corporation) was selected as the prime contractor, and in November 1950 the design process began under the management of the Army Corps of Engineers.

  Earlier, the Chemical Corps had contracted with Monsanto Chemical Company to build and operate a small pilot plant to test the DMHP process, but corrosion had damaged the apparatus so badly that it could not be salvaged. It was therefore clear that the design, construction, and operation of the full-scale Sarin plant would demand a high level of engineering expertise. According to a memorandum by Major Stanley Levy, chairman of the Chemical Corps’s Industrial Mobilization Review Committee, “The chemistry of the G Agents embraces an entire new field and much time must be given to the instruction of any contractor selected for this work.”

  At the same time the Chemical Corps began designing the full-scale Sarin production plant, it was developing a specialized delivery system called the M34 cluster bomb. A metal cylinder with seventy-six Sarin-filled bomblets neatly packed inside, the bomb weighed a total of 1,000 pounds. Development of this weapon proceeded slowly until the outbreak of the Korean War, when the Pentagon authorized a “crash” acquisition program. In August 1950, even though the engineering and testing of the M34 were not yet complete, the Chemical Corps froze the design in order to move it rapidly into production. This telescoping of development and procurement gave rise to numerous technical problems later on.

  INITIALLY, the Army assumed that the entire Sarin manufacturing process would be carried out at a single location. As planning progressed, however, it was considered prudent to reduce the vulnerability of the production complex to attack or sabotage by dividing the five manufacturing steps between two separate facilities, designated Site A and Site B. Site A would perform the initial three-step process in which elemental phosphorus was converted into a chemical intermediate called methylphosphonic dichloride [CH3P(O)Cl2], known by the short name “dichlor.” The dichlor manufactured at Site A would then be shipped by rail in special tanker cars to Site B, where the final two-step conversion into Sarin would take place.

  The Chemical Corps decided to build Site A on forty-five acres of land purchased from the Tennessee Valley Authority near the town of Muscle Shoals, in the northwest corner of Alabama. This site, on the TVA’s Wilson Dam Reservation, was chosen because the land was government-owned and had an existing plant for converting phosphate ore into elemental phosphorus, ample electrical power and water, and a pool of trained operating personnel. The location selected for Site B was on ninety acres in the north-central portion of Rocky Mountain Arsenal near Denver, Colorado. Founded in 1942 to produce mustard agent and incendiary weapons for World War II, the arsenal sprawled over twenty-seven square miles of flat scrubland and cottonwoods along the foothills of the Rockies.

  The U.S. program to mass-produce Sarin took on new urgency in the light of an ominous assessment of the Soviet chemical warfare threat. On December 15, 1950, the Central Intelligence Agency issued a top secret National Intelligence Estimate (NIE) on the possibility of a Soviet chemical or biological attack against the United States. This report concluded that if the Soviets decided to launch a chemical attack, they would almost certainly use nerve agents. By 1952, the CIA predicted, the Soviets would have at their disposal “sufficient nerve gas for sustained extensive employment” on the battlefield, and by 1954 they would possess “new agents in sufficient quantity for limited mass lethal attacks on selected military or industrial targets in the US.”

  The NIE also discussed the possibility that Soviet covert operatives might smuggle nerve agents into the United States for sabotage attacks against key military installations. “Since the agents are odorless, colorless liquids, they can be transported in glass or suitably lined containers,” the report noted. “Hence, the agent could be shipped in any desired quantity disguised as innocuous liquids, such as champagne or perfume.” Because the unique characteristics of nerve agents would make it possible to identify the Soviet Union as the source of an attack, the CIA did not consider it likely that the Kremlin would resort to such weapons prior to the outbreak of general war. Nevertheless, the CIA’s alarmist assessment of the Soviet threat gave the Chemical Corps a powerful rationale to accelerate the production of nerve agents in order to deter a Soviet chemical attack.

  IN THE SPRING of 1951, Vitro Corporation began to build the Sarin production facilities at Site A (Muscle Shoals) and Site B (Rocky Mountain Arsenal). Given the urgent military demand for Sarin-filled weapons, every effort was made to expedite construction. To save time, the various production facilities were designed and built concurrently, without the usual exhaustive development and testing at the bench-scale and pilot-plant levels. Because much of the information needed to scale up dichlor production was lacking, Vitro engineers were forced to make numerous judgment calls, turning Site A into what was effectively a “pilot plant” of gigantic proportions. Although the Army had planned for Site A to be operational in November 1951 and Site B a month later, at a total cost of $30 million, these projections proved to be wildly optimistic.

  The Muscle Shoals complex began limited operation in June 1952. It consisted of a series of chemical plants, one for each step in the conversion of phosphorus to dichlor, plus a dedicated facility for the production of chlorine. Except for high fences and other security measures, Site A resembled an ordinary chemical factory. Mounted on steel superstructures in the open air were numerous corrosion-resistant reactor vessels interconnected with stainless-steel pipes, pumps, and valves. Local residents referred to the mysterious industrial facility at the TVA reservation as “The Thing” and speculated that the Atomic Energy Commission was using it to process uranium ore. To conceal its real purpose, the Army gave Site A the innocuous name of “Phosphate Development Works.”

  Because the Muscle Shoals facility had been a crash program based on inc
omplete development, Vitro Corporation and the Chemical Corps faced numerous technical problems in getting it fully operational. The highly corrosive chemicals used to make dichlor caused leaks at expansion joints and numerous failures in valves, lines, and other components, necessitating equipment changes, process modifications, and lengthy downtimes. In 1953, a runaway reaction in the phosphorus trichloride plant (Building 101) caused an explosion in which five workers died.

  The Phosphate Development Works at Muscle Shoals in northwestern Alabama. This facility manufactured dichlor, the main precursor chemical used to produce Sarin, from 1953 to 1957.

  A production plant for phosphorus trichloride—the main starting material for dichlor production—in the Phosphate Development Works at Muscle Shoals.

  A plant for the Step 1 process in dichlor production in the Phosphate Development Works at Muscle Shoals.

  Another problem arose during the third step of dichlor production, which generated an unwanted by-product called phosphorus oxychloride in quantities too large to be disposed of by sale on the commercial market. Even when the Muscle Shoals facility was running at only half capacity, it produced about 55 tons of phosphorus oxychloride per day. It was therefore necessary to build a separate reprocessing plant to convert the by-product back into phosphorus trichloride, which could then be reused as a chlorinating agent in steps 1 and 3 of the production process. The reprocessing plant cost about $9 million to build, was costly, difficult, and dangerous to operate, and never kept pace either with the volume of by-product or with the demand for raw material.