IN PARALLEL WITH the construction of the troubled dichlor plant at Muscle Shoals, Vitro Corporation, supported by hundreds of subcontractors and suppliers, built the North Plants complex at Rocky Mountain Arsenal to perform the final two steps in the production of Sarin. The manufacturing process was complex and extremely dangerous. First, dichlor was reacted with hydrogen fluoride (HF) gas to yield a roughly 50:50 mixture of dichlor and difluor [CH3P(O)F2]. This so-called di-di mixture was then combined with isopropyl alcohol to produce Sarin, giving off hydrochloric acid (HCl) as a gaseous by-product. Sarin was purified by passing it through a distillation column in which the temperature was precisely controlled; it was then stabilized with tributylamine or triethanolamine and loaded into munitions.
Because the Sarin manufacturing process involved two highly corrosive chemicals that could erode stainless steel (hydrogen fluoride as a reactant and hydrochloric acid as a by-product), the design, construction, and operation of the Rocky Mountain plant posed unique engineering challenges. To resist corrosion and prevent the gradual destruction of the process equipment, Vitro used parts made of a high-nickel-steel alloy called Hastalloy. The company also developed new fabrication and welding techniques to make optimal use of this special material. Construction of the Sarin production facility ultimately consumed 150 tons of nickel, fifteen tons of Hastalloy, thirty-five tons of carbon steel, thirty-five tons of copper, and five tons of aluminum.
The Sarin production equipment was installed in three bays inside Building 1501, a windowless, five-story blockhouse that was sealed to contain the lethal fumes. One of the largest poured-concrete structures in the United States at the time, the blockhouse was designed to withstand a major earthquake and hurricane-force winds of 100 miles per hour. The Rocky Mountain complex also included a munitions filling line, an administration building, a hospital, quality control laboratories, utilities, and a waste treatment plant. In view of the extreme hazards posed by nerve agents—breathing air containing only one part per million of Sarin vapor for ten minutes could be fatal—several U.S. government agencies and industrial safety experts analyzed every conceivable risk from the production facility to the surrounding communities. Vitro engineers developed methods to prevent leaks and ensure the safe operation of the plant that went far beyond all previous safety requirements.
By late 1952, Site B at Rocky Mountain Arsenal, which the Army gave the deceptive cover name of “Incendiary Oil Plant,” was ready to begin operation. Production was delayed, however, by the ongoing technical and management problems at Site A, which had become a major thorn in the side of the Chemical Corps. Although various parts of the Muscle Shoals complex were operational, the facility as a whole could not achieve the planned sustained rate of dichlor production. In addition to persistent technical problems, inadequate supervision and poor personnel management caused further delays.
The Air Force was becoming increasingly impatient over the lack of Sarin to fill the M34 cluster bombs. Although the weapon had been scheduled to enter production in May 1951, the start date had been delayed five times. In order to work around the chronic problems at Muscle Shoals, the Chemical Corps hired the Shell Chemical Company to manufacture dichlor at Rocky Mountain Arsenal, using an alternative to the DMHP method called the aluminum phosphorus complex (APC) process. Dichlor produced by the APC process was ultimately used to produce about a third of the Sarin in the U.S. stockpile. Although the APC method worked fairly well, it had several liabilities: the process was complex, created an explosive hazard, and generated a large volume of hazardous waste for each pound of product.
IN 1952, Otto Ambros was released from prison after serving only two years of his eight-year sentence at Nuremberg. The sentence had been reduced at the request of the West German government and John J. McCloy, the U.S. High Commissioner for Germany. After Ambros’s release, EUCOM offered him a job advising the Army Chemical Corps. Dr. Wilhelm Hirschkind, the Dow Chemical scientist who had interviewed Ambros in July 1945, also renewed contact and arranged several meetings for him with Dow executives. As a result, Ambros soon became a successful consultant to both the German and American chemical industries.
Meanwhile, the Korean War was still raging. In March 1953, the U.S. Army Chief of Staff recommended shipping a stockpile of chemical weapons to Okinawa, an island off the coast of Japan that had been controlled by the U.S. military since the end of World War II. Deploying chemical munitions on Okinawa would enable the Army’s Far East Command to retaliate if the North Koreans or their Red Chinese allies resorted to chemical attacks. According to a memo to the Joint Chiefs from the Army Chief of Staff, the deployment of chemical arms to Okinawa required “the utmost secrecy . . . in order to forestall disclosure as long as possible. However, it should be noted that disclosure becomes an increasing possibility after the shipments of munitions are set in motion from United States depots.”
The Korean War ended on July 27, 1953, without any use of chemical weapons. Several years later, however, Brigadier General Jack Rothschild revealed that U.S. Army field commanders in Korea had requested permission to use poison gas to break the military deadlock but had been turned down.
Shortly after the end of the war, the Chemical Corps launched a public relations campaign to educate ordinary Americans about chemical warfare and civil defense. Army officials invited well-known journalists and authors to visit the Muscle Shoals and Rocky Mountain production plants and receive briefings on the Soviet chemical warfare threat. One of the more famous writers to accept this invitation was Cornelius Ryan, a popular military historian and the author of the best-selling World War II epic A Bridge Too Far. Ryan’s article, titled “G-Gas: A New Weapon of Chilling Terror,” was the cover story in the November 1953 issue of Collier’s Magazine.
Citing Chemical Corps experts, Ryan reported that a single Soviet TU-4 bomber could drop seven tons of Sarin-filled bombs on a major American city, potentially inflicting a death toll in the millions. A map accompanying the article compared the effects of a nerve agent attack on Washington, D.C., with the detonation of a small atomic bomb. Whereas the A-bomb would kill everyone within a three-mile radius, the Sarin cloud would drift up to fifty miles downwind, blanketing an area of a hundred square miles with lethal vapor. After painting this grim picture, Ryan quoted Major General E. F. Bullene, the chief of the Chemical Corps, who argued that the best way to deter a Soviet chemical attack was for the United States to possess the means to retaliate. “At this time,” General Bullene warned, “the only safe course is to be prepared to defend ourselves and ready to use gas in overpowering quantities.”
However improbable the scenario that Moscow would launch a massive chemical first strike against U.S. cities, the idea took root in the fertile soil of Cold War paranoia. The federal Civil Defense Administration prepared a thirty-minute color film showing how enemy aircraft might spray clouds of lethal nerve gas to kill and demoralize the U.S. civilian population, and how ordinary citizens could protect themselves. Because nerve agent vapors tended to hug the ground, the narrator intoned, the best way to survive a chemical attack on one’s home was to close the windows on the lower floors and shelter in an upstairs room.
By December 1953, Site A was two years behind its original projected date for full-scale operation and millions of dollars over budget. It was not until 1954 that the Muscle Shoals facility finally began producing dichlor at full capacity. Because the chemical was a solid at room temperature and highly corrosive, it had to be transported to Rocky Mountain Arsenal in special railroad tank cars lined with nickel. When the trains arrived at their destination, heating coils inside the tank cars melted the dichlor into liquid form, which was then pumped into glass-lined tanks for storage.
Site B operated twenty-four hours a day, seven days a week, converting dichlor into thousands of tons of Sarin per year. The interior of the blockhouse was divided into three operating bays that contained the chemical-processing units; in an emergency, each bay could be automatically seale
d off from the other parts of the building. In addition to these physical barriers, powerful ventilation fans kept the interior of the blockhouse at a negative atmospheric pressure so that in case of a breach in containment, the deadly fumes would be retained inside.
An aerial view of the North Plants complex at Rocky Mountain Arsenal, which produced Sarin from 1953 to 1957.
Rocky Mountain Arsenal blockhouse, which contained the final steps in the Sarin production process. The building was hermetically sealed to prevent the escape of deadly gases.
Soldiers guard Sarin-filled ton containers at Rocky Mountain Arsenal.
Because of the extremely hazardous nature of the Sarin production process, technicians rarely entered the manufacturing areas while the plant was in operation. Instead, they manipulated valves with long-handled levers from outside the sealed enclosures, while monitoring gauges that provided a continuous readout of the reaction temperatures and pressures. Between runs, however, occasional visits to the manufacturing bays were necessary to perform adjustments, repairs, and maintenance. Whenever workers entered the “hot zone,” they wore full-body rubber suits and masks, carried syrettes filled with atropine, and worked in pairs so that they could assist one another in an emergency. Over the years, the plant workers developed a healthy respect for Sarin’s insidious power. They had their cholinesterase levels checked once a week at the base dispensary; if the value fell below a specified threshold, they stopped working in the hot zone until the test returned to normal.
Despite these elaborate safety precautions, numerous mishaps occurred. In 1954, more than seventy technicians at Site B received low-level exposures to Sarin and had to be treated with atropine at the base hospital. A few individuals were hospitalized for several days with “small eye” (pinpoint pupils and blurred vision), cramps, chest pain, shortness of breath, and nausea. Some of them experienced wild dreams, extreme anxiety, and an inability to make decisions. These mental symptoms suggested that exposure to even extremely low doses of nerve agents could cause psychological disturbances and distort the judgment of commanders and troops in combat.
At the end of the production process, the distilled Sarin was pumped from the blockhouse into the munitions-loading plant (Building 1601), a narrow, windowless, bunkerlike structure about 600 feet long. Airtight, this building contained filling lines for the various types of munition, including artillery shells (155 mm, 8-inch, 105 mm), aerial bombs, and submunitions for cluster weapons. Each filling line had four or five stations enclosed inside sealed metal cabinets to prevent the escape of toxic fumes. At the first station, the machine loaded a projectile or bomb with a preset amount of liquid Sarin pumped from an underground storage tank. A conveyor belt carried the munition to the next station, where an overlay of helium gas was injected into the space remaining inside. The filling aperture was then capped, welded with a double seal, and vacuum-tested for leaks using a helium detector. Because helium is an extremely small molecule that can penetrate the slightest leak, the presence of the inert gas in the air was a reliable indicator of defective welds.
Next the filled shells or bombs moved along the conveyor belt to another automated station that decontaminated the outside surfaces. Finally, the finished munitions were transported to an open packing area, where workers weighed, painted, stenciled, assembled, and crated them for storage. Because of a shortage of empty bomb casings and projectiles, thousands of gallons of bulk Sarin were stored temporarily in steel ton containers, two feet wide by eight feet long. Hundreds of these containers, painted silver, were lined up in rows in the storage yard.
Elaborate safety measures were designed to protect the workers at the Rocky Mountain Arsenal facility. Thirty M5 Automatic G-Agent Fixed Installation Alarms, each seven feet high and weighing 725 pounds, were installed throughout the blockhouse and the filling plant to monitor the concentration of Sarin vapor in the air. The detectors contained a solution that reacted with Sarin to yield a fluorescent compound; a photometer measured the fluorescence and triggered an alarm in about ten seconds. Human operators then shut down production and sealed off the affected unit. To back up the electronic alarms, cages containing canaries and white rabbits were placed at strategic points around the facility. Because these animals were more sensitive than humans to nerve agent exposure, they would provide a few minutes’ warning of an accidental leak; the rabbits’ large pink eyes made it easy to see when their pupils were constricted. The Sarin plant also incorporated an advanced pollution abatement system in which contaminated effluent air from the blockhouse and the filling plant passed through a series of four caustic Venturi scrubbers before being exhausted out a 200-foot stack.
THE AMOUNT OF TIME, money, and effort invested in the Sarin production program turned out to be vastly greater than anticipated. Whereas the original cost estimate for construction of the two facilities had been $30 million, the actual total was well over $100 million. Nevertheless, there was no outcry from Congress or the public over the huge cost overruns because the entire project was shrouded in secrecy. It was not until July 1954 that Major General William M. Creasy, Bullene’s successor as chief of the Chemical Corps, disclosed to the public that the real purpose of the Phosphate Development Works at Muscle Shoals was to produce an intermediate chemical used in the manufacture of nerve gas.
Even as the mass production of Sarin was under way, chemists at Edgewood Arsenal continued to assess a variety of novel compounds as potential nerve agents. In 1953, the Advisory Committee on New Agents requested the toxicity screening of 157 candidate chemicals, selected from a list of about 400. Edgewood scientists also studied agent cocktails such as Sarin and mustard, and experimented with thickeners and other additives that could modify the droplet size and physical properties of Sarin, increasing its persistence or ability to penetrate clothing or skin.
The Chemical Corps issued numerous research contracts for studies of Sarin production, stabilization, detection, and decontamination to outside entities such as the National Bureau of Standards, the Standard Oil Development Company, the University of Kansas, and Louisiana State University. The Air Force also commissioned Project Big Ben, a study group of statisticians, mathematicians, and engineers at the University of Pennsylvania, to analyze optimal ways of dispersing Sarin over area targets such as a military formation.
The M34 aircraft-delivered cluster bomb was filled with 76 Sarin-filled bomblets that were dispersed over the target area.
In 1954, the Air Force and the Army Chemical Corps conducted a series of ten field tests of the M34 cluster bomb at Dugway Proving Ground. During these trials, which were performed at night, two B-47 bombers flew from Eglin Air Force Base in Florida to Dugway and each dropped two M34 bombs from an altitude of 35,000 feet. Fire pots laid out on the desert floor delineated the target area: a 6,000-foot square superimposed on a circle 8,000 feet in diameter and crosshatched with grid lines. As each cylindrical bomb fell to earth, it deployed a parachute to slow its descent. The metal casing then burst open and scattered its cargo of seventy-six bomblets, which detonated on impact with the ground and discharged their content of Sarin (2.6 pounds). Distributed over the test grid were sampling devices designed to collect Sarin vapor and droplets, and cages containing test pigeons. The target area also included two family-style model homes and a slit trench to measure the ability of the toxic cloud to penetrate these structures. In addition to the M34 cluster bomb, the Chemical Corps standardized two artillery projectiles in 1954: a light 105 mm shell that held 1.6 pounds of Sarin, and a heavy 155 mm shell that held 6.5 pounds.
Sarin-filled munitions were stockpiled at several Army depots on U.S. soil, and some were secretly deployed overseas. Although the first chemical weapons transferred to Okinawa in 1953 had been mustard-filled, in 1954 the stockpile was augmented with munitions containing Sarin. Outside the Chemical Corps, however, Army field commanders viewed chemical weapons as more trouble than they were worth, contaminating the battlefield and forcing troops to wear clumsy protecti
ve gear that degraded their fighting efficiency. This antipathy led to a strong resistance to integrating chemical arms into the Army’s force structure and war-fighting doctrine.
This 155 mm artillery shell was loaded with Sarin (GB) and marked with three rings, the symbol for chemical nerve-agent munitions.
As a result, apart from the small stockpile deployed on Okinawa, the vast majority of chemical munitions were stored within the continental United States, mostly at or near the original production locations and far from coastal ports where they could be readily deployed to Europe or the Pacific. Moreover, a large fraction of the nerve agent stockpile was not loaded into munitions at all, but remained in bulk storage tanks.
Throughout the 1950s, the Pentagon continued to justify the mass production of nerve agents with the specter of a large-scale Soviet chemical attack. The Joint Strategic Plans Committee of the JCS predicted, “The Soviets have been producing at least one of the ‘G’ agents since 1949 and hence, by 1956, will probably be capable of extensive employment of nerve gases.” Kremlin leaders, for their part, saw the United States moving forward aggressively with Sarin production and were determined not to be left behind in the chemical arms race between the superpowers.
CHAPTER EIGHT
CHEMICAL ARMS RACE
War of Nerves Page 17
