A.  Start-Up

In addition to taking a background sample, all Fox crew training emphasized a standard operating procedure both before the start of the vehicle’s mission and after the MM-1 alerted to the possible presence of a chemical warfare agent.^[14] The start-up procedure consisted of function tests, confidence checks, and calibration tests. Function tests check various MM-1 components^[15] which must each pass for the MM-1 to operate properly; the function tests should be repeated until all components pass. Confidence checks use chemical simulants that a Fox crewmember holds to the sampling probe. These checks further assure the vehicle’s sampling and detection equipment transports chemical vapor samples through the system properly.^[16] Calibration tests ensure the MM-1 can take samples continuously and allow the MM-1 to properly gain access to chemical warfare agent information stored in the library.^[17] After successfully performing function, confidence, and calibration tests, the operator runs a temperature program to destroy any remaining test substances in the sampler and the MM-1 system and eliminate any unwanted residual ion activity displayed on the operator’s video screen. If the temperature program is not successful, chemicals in the sampler unit can emit residual ions, which can cause false alarms when combined with sampled air.^[18]

The MM-1 collects samples in three ways: the MM-1’s sampling probe draws vapors from the surrounding air; the MM-1 operator uses silicone sampling wheels to raise liquid samples from the ground to the retractable sampling probe; or the MM-1’s sampling probe draws directly from the source of the contamination. Spare sampling wheels are stored outside the vehicle and can be changed from inside the vehicle using the protective glove. The MM-1 operator also may use the protective rubber glove to manually collect physical samples. A small window allows the operator to see outside while working (Figure 3).^[19]

Figure 3.  Air/Surface Sampler Unit

The MM-1 can operate in two modes: Air Monitor or Surface Monitor. Each mode has several sampling methods. The Air Monitor mode normally is used for continuous sampling while the Surface Monitor mode is used for periodic sampling and monitoring.^[20] MM-1 operators were trained to use only the Air Monitor mode; therefore this paper addresses only that mode.^[21] During Desert Storm, Fox crews normally used three methods—Air/Hi, Wheel/Hi, and Surface/Lo—in the Air Monitor Mode to search for or detect chemical warfare agents. Table 1 shows the method, the sampler unit temperature, whether the sampling wheels are used, and sampling probe position for each method.

The Air/Hi method is used to define edges or borders of previously identified contaminated areas by detecting chemical warfare agent vapors in the air. The sampling probe, approximately one meter from the ground, draws in vapors in its immediate surrounding area. The sampling wheels are not used. The sampling probe operates at the high temperature of 180� Celsius (C) (356� Fahrenheit). Consequently, the molecules of the sample move more quickly to the MM-1. At this temperature, many chemicals could enter the mass spectrometer at the same time, making it more difficult for the MM-1 to assign an ion to a particular chemical warfare agent, rather than another substance.

The Wheel/Hi method searches for and alerts the crew to the possible presence of liquid chemical warfare agents on the ground. The temperature of the probe is again 180� C and as in the Air/Hi method, the probe position is approximately one meter from the ground. However, in this method, the wheels directly touch the ground and pick up possible liquid chemical warfare agents rather than drawing in vapors from the surrounding air. The wheels rise off the ground and deliver the liquid chemical warfare agent to the sampling probe’s head, which then vaporizes the sample and delivers a highly concentrated agent vapor to the MM-1 for analysis. The ability to directly capture liquid chemical warfare agents makes this method more sensitive than the Air/Hi method.

In the Surface/Lo method, the sampling probe is at 120� C (248� Fahrenheit)—cooler than the Air/Hi or Wheel/Hi methods. The lower sampler temperature allows maximum separation of the multiple chemical compounds in a sample. After the MM-1 initially alerts from the Air/Hi or Wheel/Hi methods, the MM-1 operator allows the probe to cool. The operator then lowers the probe to within three to five centimeters of the suspected contamination. The lower temperature and proximity to the source of the contamination of the Surface/Lo method allows the MM-1 to acquire a well-prepared sample and deliver possible liquid chemical warfare agents for analysis.

The MM-1 continuously monitors air passing through it, checking for the presence of chemical warfare agents identified on a pre-selected, initial target list of up to 22 chemical compounds, derived from the full library of 60 compounds. During Desert Storm, Fox vehicle operators normally used a target list of 10 compounds. Table 2 shows the 10 chemical warfare agents most commonly monitored in an initial search during Desert Storm.^[22] This target list consists of a four-ion combination for each chemical compound. During the initial identification step, the MM-1 fragments each sample into a pattern of ions and then compares each four-ion combination on the target list against the sample, searching for a match.^[23]

If the MM-1 makes an initial match, and the ion intensities exceed a minimum detectable threshold (unique for each agent), an alarm sounds to alert the operator. This alert also displays on the MM-1 operator’s screen and can be printed on a paper tape.^[24] This initial alert indicates the possible presence of a chemical warfare agent. However, since many chemical compounds have some of the same or similar ion masses as those on the chemical warfare agent target list, it is also possible that another substance with similar ions was detected.

To better determine what chemical warfare agent is most likely present and increase confidence the MM-1 actually detected a chemical warfare agent, the MM-1 must analyze a spectrum of the sample, preferably in the Surface/Lo method. A spectrum analysis involves optimizing the MM-1 by reducing the temperature of the sample line from 180�  C to 120�  C for better molecule separation, cleaning the sample probe to remove residual ion activity (contamination), and lowering the probe to within three to five centimeters of the contamination, re-acquiring a sample, and terminating the air flow. This allows the MM-1 to work with a better-prepared sample. The MM-1 then searches its entire 60-compound chemical library of four-ion combinations, compares them against this spectrum from the improved sample, and attempts to find a match.^[25] The MM-1 operator should print a tape, which saves the details of the spectrum as a hard-copy historical record.^[26] It takes several minutes to collect a sample and obtain a good spectrum analysis, but this process properly evaluates the sample for any suspected chemical warfare agents and minimizes the possibility initial MM-1 indications were affected by contaminants (e.g., exhaust, oil, etc.). Although an MM-1 operator can produce a spectrum in other ways, this is the proper, most accurate method.^[27] Should the properly performed spectrum procedure identify a chemical warfare agent, the MM-1 operator and Fox commander can be confident, though not assured, the agent is present. Conversely, if the spectrum analysis does not identify one of the chemical warfare agents in the MM-1 library, the MM-1 operator and Fox commander can be confident the chemical warfare agent displayed during the initial alarm is not present. A mass spectrometry expert’s further analysis of the spectrum tape printout can further increase or decrease the degree of confidence in the detection. Additionally, MM-1 operators are taught to collect a specimen of the contamination (e.g., a soil sample) to further aid confirmation of the substance through laboratory analysis. If the MM-1 cannot match a chemical warfare agent in its library and "unknown" is displayed during the spectrum sample, the area is considered to be contaminated by an unknown substance,^[28] but not a substance in the MM-1’s library. A reading of "fats, oils, wax" refers to petroleum-based hydrocarbons (e.g., fuel, vehicle exhaust, or burning oil well fire smoke) that act as contaminants. This is considered a false alarm.

It is important to note that when performing the spectrum analysis, the MM-1 analyzes only the substance in the initial alarm with the highest relative intensity even if several substances may be present. The MM-1 has the capability to evaluate the substances in a sample with reduced intensities. This procedure, known as a series spectrum, involves taking a spectrum analysis on each substance identified in an initial alarm. This paper, however, does not describe the series spectrum because during the Gulf War US operators were not trained on this procedure.^[29]

Although sufficient to alert for the possible presence of chemical warfare agents on the battlefield, the MM-1 does not confirm chemical warfare agent presence to the standards required for recognition by the United States national command authority or neutral third parties. Therefore, an MM-1 detection would lack legal standing in an international court of law, for instance, in a dispute over first use of chemical weapons. International protocols have been established for final confirmation of chemical warfare agent use. Evidence must be sent to a more accurate laboratory spectrometer certified and recognized by the international community.^[30] A certified laboratory must analyze the records of the sample site, physical evidence (e.g., soil, weapons fragments, or tissue remains), a chain of custody record of the physical sample, and the spectrometer’s analysis results on printed tapes. Results are sent to neutral third parties, such as the United Nations, before confirming the presence of chemical warfare agents.^[31] After alerting to the possible presence of a chemical warfare agent, only by following these protocols will the confirmation of the presence of chemical warfare agents be legally proven to the international community’s satisfaction.

A false positive is an alert for a chemical warfare agent that is not present. A variety of contaminants can cause false positives in the MM-1, including diesel fuel, vehicle exhaust, and oil well fire smoke. False positives can occur because the ions used to detect chemical warfare agents in the MM-1 library also are prevalent in petroleum-based hydrocarbons (e.g., oil, fuel and benzene derivatives) found on the battlefield.^[32] Table 3 lists just a few examples of contaminants in which ion patterns under certain conditions are known to resemble those of certain chemical warfare agents.

The four-ion combination used to initially alert for the nerve agent sarin and the riot control agent CS are similar. For example, sarin has an ion reading at 125.0 molecular weight and a relative intensity of 25.0%, while the riot agent CS has an ion reading at 126.0 molecular weight and a relative intensity of 18.7%. Because these readings are so similar, the MM-1 may initially alert for sarin when CS is actually present.^[33] Conducting a spectrum analysis in this instance would eliminate false positives because the MM-1‘s full spectrum resolves this ambiguity by examining other ion readings and their relative intensities.

Tests of the Fox conducted after the Gulf War determined the silicone material in the Fox sampling wheels and silicone lubricants used on the wheels emit certain chemical vapors that act as contaminants when the wheels are raised to the heated sampling probe.^[34] Ions from the silicone caused an initial alert for the chemical warfare agent lewisite. A spectrum analysis would indicate such an alarm was in fact a false positive by not identifying lewisite. Although this defect has since been corrected, it is relevant to the Fox vehicle configuration during the war and helps to explain several of the unconfirmed alerts to the chemical warfare agent lewisite, alerts that in all probability were due to the silicone contaminant rather than lewisite.^[35]

Similarly, the MM-1 did not routinely monitor benzyl bromide (a tear-producer and skin irritant), which was in the Fox chemical library and could be identified by spectrum analysis. The ions used to identify benzyl bromide also are in toluene (a common solvent).^[36] All these examples demonstrate that battlefield contaminants could have affected the MM-1 during the Gulf War to cause false alarms.

Since not alerting to a potential chemical warfare agent hazard jeopardizes unprotected soldiers’ safety, the MM-1’s design specifically works to ensure an alert occurs if there is a chance a chemical warfare agent is present to produce maximum warning time and safety even at the expense of generating potential false positives. However, to prevent the Fox from continuously alerting to a variety of substances, the MM-1 mobile mass spectrometer contains certain variables and design considerations that help to minimize the false positives. The MM-1 uses three compound-specific parameter values to help separate genuine alarms from alarm errors. These values are the interference, reliability, and impossible ion parameters uniquely set for each compound in the library. Table 4 shows examples of these values.^[37]

In general, the function of the interference parameter is to suppress false alarms due to contaminants, yet still alarm for a chemical warfare agent when large amounts of contaminants are present. Higher values mean a larger amount of a contaminant is required to suppress an alarm. The interference parameter scale is logarithmic because of the need to express rapidly increasing values. For example, compared to a chemical warfare agent with an interference parameter set at 1.0, to suppress an alarm for a chemical warfare agent with an interference parameter of 2.0 would require not twice but 10 times the amount of a contaminant be present in the vapor sample. Compared to a chemical warfare agent with an interference parameter of 1.0, the alarm for a chemical warfare agent with an interference parameter of 8.0 (e.g., sarin) would be suppressed only if the presence of the contaminant were 100,000,000 (10⁸) greater.^[38] By setting the interference parameter properly, the MM-1 will alarm for the presence of a small amount of chemical warfare agent in the presence of a large amount of other compounds.

The reliability parameter compares the accuracy of the values in the programmed library used to identify a chemical warfare agent against each of a sample’s four ion masses and their relative intensities. Determining whether to use a higher or lower setting for this parameter is a trade-off between detecting an actual chemical warfare agent and generating a false positive. The higher the setting, the more likely an MM-1 will identify a chemical warfare agent as well as signal a false positive.^[39] The lower the setting, the more likely the MM-1 will miss a positive detection while also decreasing the chances of a false positive.^[40]

An impossible ion is one not present in the mass spectrum of a dangerous compound but present in another compound with similar ion and mass readings. If the MM-1 detects the impossible ion, it can distinguish between the dangerous compound and the other, similar compound. For example, the MM-1 monitors four ions for phosgene: 65.0, 63.0, 98.0, and 109.0. The mass 109.0 is an impossible ion for phosgene and is set with a relative intensity of 0.0%. Detecting the mass 109.0 at any relative intensity other than 0.0% would enable the MM-1 to assume the sample could not be phosgene, so the MM-1 would not alert to phosgene.^[41]

Considering the process by which the MM-1 initially searches for chemical warfare agents and the parameter settings used to minimize alarm errors, the MM-1 can initially alert when a chemical warfare agent is not actually present. In a multi-chemical environment (e.g., a smoke- filled battlefield), the MM-1 must compare the ions encountered in the outside air with the ion patterns of the chemical warfare agents on the target list. Accepting initial false positives is a design compromise between MM-1 sensitivity and selectivity. The MM-1 is sensitive enough to rapidly detect chemical warfare agents to provide maximum warning and selective enough to rule out most false positives with subsequent testing.^[42] In other words, if there is any question about the identity of the suspected substance, safety considerations require the MM-1 to issue an initial alarm for the dangerous chemical and try to determine the validity of the initial alarm later with a spectrum analysis.

An additional tool to help minimize errors during the initial search for chemical warfare agents is the temperature program, which cleans the sampling probe to ensure the most accurate spectrum possible. When the operator’s screen shows high ion activity above a certain threshold, the operator should run the temperature program to eliminate any residual contamination in the vapor sampling system, or else the spectrum taken may contain some contamination.^[43]

Every time the MM-1 performs certain functions, the operator can record them on a paper tape that looks similar to a grocery receipt. The printed tape records information such as calibration tests, warnings, alarms, sampling method changes, and spectrum analyses results. If the MM-1 operator activates the autoprint button, the MM-1 prints data to a tape automatically; otherwise, the MM-1 operator must press the print button.

Following are four examples of MM-1 tapes in different scenarios (Figures 4-7).^[44] These illustrations are not actual Gulf War MM-1 detections. Listed first on all four tapes is the word "background," referring to a background air sample taken at the start of each mission. All tapes list the sampling method in use, for example, Wheel/Hi or Air/Hi. All tapes list Air Monitor, the only mode of operation taught to US MM-1 operators before the Gulf War. The Vehicle Orientation System provides location information of a possible detection based on distance and direction traveled since a known starting point was put into the system. In each scenario, a compound that initially alerts the MM-1 is listed along with the time, detection number, and relative intensity that appears above the minimum detectable amount. The letter "A" preceding the relative intensity shows the sampling wheels are not in the expected position relative to the sampling port to take a wheel sample while "C" shows the wheels are in the proper position.^[45] The tape also shows results of the spectrum analysis and are indicated by the letter "S." The spectrum analysis can increase the operator’s confidence an initial alarm is correct or determine the substance identified is unknown because it is not in the MM-1 library.

| First Page | Prev Page | Next Page |