Toxicologic Assessment of the Army's Zinc Cadmium Sulfide Dispersion Tests (1997)

BECAUSE OF CONCERN over possible use of biologic warfare (BW) by a foreign power against the United States and its allies, U.S. Secretary of War Henry Stimson in 1941 asked the National Research Council/National Academy of Sciences to investigate all phases of BW. In response, the Research Council appointed a committee of 9 prominent scientists, known as the War Bureau of Consultants (WBC) Committee, which conducted its work in the utmost secrecy.

The committee concluded in a classified report (NAS 1942) that BW was distinctly feasible and urged that appropriate steps be taken for defense against its use. (The report was declassified in 1988.) The report stated in part:

Secretary Stimson conveyed the committee's recommendations to President Roosevelt, who in May of 1942 authorized the secretary to create an

organization within the Federal Security Agency to conduct the U.S. Biological Warfare Program so as to avoid public concern over America's vulnerability. The exchange of information on this subject with the United Kingdom and Canada, which had been inaugurated some months before, was continued, and provision was made for the interchange of biologic-warfare personnel among the 3 countries (U.S. Army 1977). The research and development work from these programs extended knowledge of the military use of pathogenic microorganisms, as well as identifying ways to defend against them. It was recognized that an effective program involving BW agents, weapons systems, and production could not be achieved without large-scale developmental operations. In 1942, the U.S. Army Chemical Warfare Service assumed the responsibility for a large-scale research and development program. Fort Detrick in Frederick, MD, was selected to carry out the biologic-warfare task.

The policy of the United States concerning BW between 1941 and 1973, when the entire stockpile of offensive BW agents was destroyed as a consequence of a directive issued by President Nixon in 1969, was to deter its use against the United States and its allies and to retaliate if deterrence failed. Fundamental to the development of a deterrent strategy was the need for a thorough study and analysis of our vulnerability to overt and covert attack. The policy also required examination of the full range of retaliatory options and development of a retaliatory capability that used pathogenic agents. In short, the policy required extensive research and development to determine precisely our vulnerability, the efficacy of our protective measures, and the tactical and strategic capability of various BW agents and delivery systems.

In the beginning and continuing throughout the BW program, there was a paucity of scientific and engineering knowledge and principles related to the vulnerability of the United States and its allies to BW attacks. Vulnerability testing was conducted to provide information on the agents likely to be used, means of disseminating them, sizes of areas that could be attacked, environmental effects on agents, dispersion of agents, the obstruc-

tive effects of buildings and other structures and terrain on agents, the detection and identification of agents, U.S. areas and forces most likely to be attacked, the extent of damage possible, and physical and mathematical models that could be used as substitutes for live, open-air testing.

The pathogenic organisms considered most suitable as weapons were those of anthrax, brucellosis, tularemia, Q fever, and psittacosis. The tactical use of BW agents required the development of tables of munitions requirements (the quantity of material required to achieve a particular military objective) for the strategic use of BW agents against target cities. The Army recognized two major challenges. In the case of tactical use of BW agents, one of the greatest difficulties was in selecting conditions that would lead to successful results when logistically feasible amounts of material were used on comparatively small areas. As the potency of the newly developed agents increased, the objective became the distribution of small amounts of them over comparatively large areas. Thus, to use phosgene (a highly toxic chemical-warfare agent) effectively, it was desirable to identify and use conditions that would enable the effective use of several hundred pounds of phosgene per 100 yd², whereas some of the BW agents could be used in quantities of a few pounds per square mile.

The second challenge arose from the impossibility of field testing BW agents in test areas that are typical of the probable targets, such as selected cities of the former Soviet Union. If pathogens or extremely toxic chemicals were to be tested, they had to be handled in remote, isolated areas, but such areas generally are atypical of habitable regions. The Army wanted to know the effects of buildings, terrain, meteorologic conditions, and so on, on the dispersion of BW agents. To obtain some estimate of the amount of BW material required to meet particular objectives in given cities or industrial areas, an indirect approach would be needed.

Three approaches to the problem of estimating munitions requirements for the strategic use of BW agents against target cities were available. The first, which was investigated by the British, used wind-tunnel studies on city models and a small number of field tests within a city itself (Aanensen 1951). Important results were obtained, but it had not been established that the necessary range of meteorologic conditions could be reproduced within a wind tunnel, particularly the unique mesometeorologic conditions induced by the city itself. The second and third

approaches were based on full-scale field experiments. The second approach was to construct a test target typical of known targets in an area resembling as nearly as possible typical terrain conditions and having typical meteorologic conditions. Because cities themselves modify meteorologic conditions to some extent, the target had to be large enough to induce such modifications. The third approach was to simulate the BW agent, rather than the target, and to run tests with the simulant in suitable areas. This approach was chosen after the first 2 were rejected—the first because it was inadequate and the second because it was too expensive. That left researchers with two more choices: choosing the simulant and selecting test cities or other appropriate test locations.

Every effort expended in open-air testing was first directed toward the use of biologic and nonbiologic simulants to obtain the necessary data for evaluation (U.S. Army 1977). Biologic simulants are defined as living microorganisms that are not normally capable of causing infection, that represent the physical and biologic characteristics of potential microbiologic agents, and that are considered medically safe to operating personnel and surrounding communities. Nonbiologic simulants are nonliving inert (usually inorganic) materials. They are chosen to resemble the size of BW agents for penetration in the respiratory tract; they are not themselves BW agents.

The Army used both biologic and nonbiologic simulants in dispersion tests. The biologic simulants used included Serratia marcescens, Bacillus globigii, Bacillus subtilis, and Aspergillus fumigatus; all 3 are considered to be generally nonpathogenic or of low virulence to normal populations. The nonbiologic simulants studied included zinc cadmium sulfide, ZnCdS; sulfur dioxide, SO₂; and soap bubbles.

Setting a munitions requirement (the quantity of material required to achieve a particular military objective) for a BW agent required researchers to consider, among other things, the toxicity and dose of an agent. Determination of the dose involves 2 variables: toxicity depends on the size of the agent, which determines where in the respiratory system the

organisms or particles would lodge (Stanford University 1952, pp. 16-17); and deposition of organisms or particles in the alveoli is much more effective than deposition in the upper respiratory tract.

ZnCdS, a fluorescent pigment, was chosen as a simulant not just because of its detectable glow, but also because of its particle size. Its particle-size range, 0.5-3 µm, approximates that considered most effective in penetrating into the lungs (Stanford University 1952, p. 55). Other properties that made it desirable as a simulant were its economic feasibility; its lack of toxicity to humans, animals, and plants; its stability in the atmosphere; and its dispersibility. ZnCdS was considered to be ''a simulant having no viability'' or an "inert material" (Stanford University 1952, p. 22).

To accomplish the Army's goal of estimating munitions requirements for the strategic use of BW agents against cities, the researchers considered as test areas North American metropolitan areas that most closely matched the meteorologic, terrain, population, and physical characteristics of the Soviet cities of interest, such as Moscow and Leningrad (Stanford University 1952, pp. 32-37). For winter conditions, it became apparent that the upper Mississippi Valley and adjacent areas presented the best possibility of matching some climatic characteristics and bracketing others. In this general geographic region, the following cities were considered: Oklahoma City, Kansas City, Omaha, Cincinnati, St. Louis, Chicago, Minneapolis, and Winnipeg. Of those, St. Louis and Minneapolis appeared to bracket the range of climatic values of interest to the best advantage (Stanford University 1952). The broad topographic requirement to be met was that of flat to rolling country at elevations generally below 1,000 ft above sea level. St. Louis and Minneapolis qualified in those respects. They are on one or more rivers—another desired feature.

In summer, that general area has considerably more precipitation than the Soviet area of interest. However, areas in North America with substantially the same precipitation as that in the Soviet area of interest were

disqualified from further consideration if their terrain was mountainous, they were too high above sea level, or their rainfall was of a peculiar coastal nature (Stanford University 1952). To qualify as a site for summer tests, a city had to have cloudy and clear days and rainy and dry days within specified ranges of temperature and windiness. St. Louis and Winnipeg met the desired summer temperature range, and in other respects they also qualified as summer-test cities (Stanford University 1952).

Population was an additional consideration used in the decision to test in Minneapolis, St. Louis, and Winnipeg. The Stanford University (1952) study noted that "of the 82 Russian cities whose populations are greater than 100,000, only two have populations exceeding 1,000,000" (Stanford University 1952, p. 34). Minneapolis, St. Louis, and Winnipeg fit within that range, with populations of 500,000, 800,000, and 250,000, respectively (Stanford University 1952, p. 34). Population density is also an important factor, but this information was not available on the Soviet cities.

The structural characteristics of cities of the former Soviet Union could not be completely matched in American cities, but the essential features were found in various degrees. St. Louis, in particular, has structure heights in general not exceeding 3 stories. Industrial instillations occur both in built-up areas and in isolated areas.

The presence of universities in Minneapolis, St. Louis, and Winnipeg was yet another advantage in that the universities provided an "ample pool of qualified personnel" to assist with field testing and data reduction (Stanford University 1952, pp. 40, 73). In Minneapolis, researchers also believed that the University of Minnesota laboratories might prove useful if the need arose (Stanford University 1952, p. 73); a substantial number of part-time personnel from the University of Minnesota were employed (Stanford University 1952, p. 34).

The cooperation of local officials and the local staffs of the U.S. Weather Bureau was important (Stanford University 1952, p. 40). Cooperation of the police departments and air-pollution control officers was enlisted to avoid problems with local officials in connection with the operations. To avoid disclosing the exact nature and purpose of the operations, a cover story was devised: city officials were told that the work was to obtain data pertinent to smoke screening of cities to prevent aerial observation (Stanford University 1952, p. 76).

Other test locations were selected to simulate other Soviet cities (such as San Francisco, CA and Panama City, FL), forests (such as Chippewa National Forest, MN), flatlands (such as Fort Wayne, IN, and Corpus Christi, TX), deserts, and unpopulated areas (Dugway Proving Ground, UT). Appendix B provides the reasons for selecting other locations.

In tests with the biologic and nonbiologic simulants, public safety was stated to be the foremost consideration (Stanford University 1952). Organisms and materials that were considered by the scientific community to be safe were selected (U.S. Army 1977). A total of 160 tests using various simulants were conducted at 66 locations (both military and civilian targets) in the United States (including Alaska and Hawaii) and Canada. The specific dates, locations, and substances used are in an Army report (U.S. Army 1977). In the conduct of BW testing, specialized sampling and analysis techniques were used to determine the various parameters of the test requirements and the downwind travel distances. These were supplemented by rather complete meteorologic data gathering systems to define meteorologic conditions. Meteorologic conditions constituted an absolute controlling factor in whether a test was permitted to start or continue.

The three most commonly used biologic simulants were Serratia marcescens, Bacillus globigii, and Aspergillus fumigatus. The Army released S. marcescens in eight tests to determine the vulnerability to enemy attacks (U.S. Army 1977). In addition, the Army conducted field testing with B. globigii and Aspergillus. S. marcescens and Bacillus subtilis were also released in San Francisco and New York subway systems, respectively, to study their dispersion.

Dispersion tests were carried out in many cities and rural locations with particles of the nonbiologic simulant ZnCdS. A total of 34 tests involving the dispersion of ZnCdS particles were conducted in various U.S. and Canadian locations. Some of those tests involved simultaneous releases of ZnCdS and Serratia marcescens or Bacillus globigii.

During the 1950s and 1960s, Stanford University and the Ralph Parsons Company (both contractors for the U.S. Army Chemical Corps) conducted atmospheric-dispersion tests with ZnCdS particles in Minneapolis, MN; Corpus Christi, TX; St. Louis, MO; Fort Wayne, IN; and 29 other urban and rural locations in the United States and Canada. The tests were purportedly used to develop and verify meteorologic models for estimating the dispersion of aerosols in various environments. However, the real purpose was to obtain information that would be useful for estimating the potential dispersion of BW agents and determining munitions requirements for the strategic use of BW agents against selected cities of the former Soviet Union. The tests were designed to provide information on the dispersion of BW agents over a short distance (for example, within a city, such as Minneapolis), over many miles (Fort Wayne, IN), and over several thousand square miles (Large Area Coverage test).

Operation LAC, which took its name from "Large Area Coverage," was the largest test ever undertaken by the Chemical Corps. The test area covered the United States from the Rockies to the Atlantic, and from Canada to the Gulf of Mexico. The tests proved the feasibility of covering large areas (thousands of square miles) of a country with BW agents. Many scientists and officers had believed this possible, but LAC provided the first proof.

Aanensen, C.J.M. 1951. Wind Tunnel Experiments of Diffusion in a Built-up Area. Porton Technical Paper 257. The Travel of Gas in a Built-up Area. Porton/TU 1206/340/51.

NAS (National Academy of Sciences). 1942. Report of the WBC Committee. (Declassified in 1988)

Stanford University. 1952. Behavior of Aerosol Clouds within Cities. Joint Quarterly Report Number One. July-September. Stanford University and Ralph Parsons Company. 78 pp.

U.S. Army. 1977. U.S. Army Activity in the U.S. Biological Warfare Programs, 1962-1977, Vol. II .