Quantitative Measurements of Hazardous Gas Effluents from the Combustion of Crew Waste Simulant in Microgravity

Current planning for crewed space exploration is being developed within the ‘Moon-to-Mars’ framework (NASA, 2023). Although the most immediate and visible activity within this framework is exemplified by the Artemis missions, planning for crewed missions to Mars is underway. Proposed Mars mission durations range from 700 to 1,200 days; thus, intensive consideration of mission logistics is required for such a mission to be feasible (Chai et al., 2021). NASA’s Logistics Reduction program is currently evaluating numerous facets of mission logistics, including supply organization via RFID technology (Fink et al., 2017), crew clothing and laundry (Ewert and Jeng 2015; Broyan, Chu, and Ewert 2014), handling of metabolic waste (McKinley, Borrego, and Broyan, 2022), re-purposing waste items (Shull, Polit Casillas, and Howe 2012), and disposal of crew trash (Anthony and Hintze 2014; J. Olson et al. 2021; Wetzel et al. 2018; Mesa et al. 2021; Sepka et al. 2022). Among these, the authors have performed several studies on the disposal of crew trash via the so-called Trash-to-Gas (TtG) approach (Pitts et al. 2022; Meier et al. 2021; 2020; Shah, Meier, and Toro Medina 2019). In this paradigm, a thermal process is used to convert solid crew trash into a relatively inert gas that can be safely vented into space, thus reducing the total mass of a vehicle and, correspondingly, reducing the propellant required for mission maneuvers (e.g., orbit insertions and transit injections) (J. A. Olson et al. 2021). Various thermal processes have been investigated by NASA and via public-private SBIR partnerships, including combustion/incineration, steam reforming, pyrolysis, torrefaction, and plasma processes (combustion and pyrolysis) (J. Olson et al., 2021). It is projected that the thermal processing of crew trash could result in substantial mass savings of >6,000 kg for a Mars transit vehicle, representing more than 13% of the payload mass (J. A. Olson et al., 2021). Such mass savings are highly desirable since they would allow for additional valuable (e.g., scientific) payloads. TtG systems may also be applicable to a sustained lunar presence, which is expected beyond the three currently planned Artemis missions (U.S. Government, 2020). A critical consideration for a TtG system is the possibility of contaminating the habitable volume with volatile organic compounds (VOCs). The purpose of this study is to compare the combustion VOCs produced under microgravity vs. analogous terrestrial laboratory tests to evaluate potential differences in performance.

Researchers at NASA’s Kennedy Space Center (Florida, USA) have undertaken several TtG test campaigns wherein the combustion of simulated crew trash was performed under microgravity conditions to determine the feasibility of such an operation and to compare trash combustion in a terrestrial venue versus microgravity (Pitts et al. 2022; Meier et al. 2021; 2020; Shah, Meier, and Toro Medina 2019). The system for these studies is the Orbital Syngas Commodity Augmentation Reactor (OSCAR) (Toro Medina et al. 2020). As described below, OSCAR is a flight-capable subscale TtG system designed to perform tests in microgravity as well as under terrestrial laboratory conditions to provide a direct experimental comparison of performance. OSCAR utilized combustion during the microgravity demonstration described here due to time constraints inherent in sub-orbital flight, but it can perform various thermal degradation processes. The OSCAR microgravity test activities have included two drop-tower campaigns (at Glenn Research Center, Cleveland, OH, USA) and two flights on board Blue Origin’s New Shepard spacecraft (from Blue Origin’s Launch Site One, in Van Horn, TX, USA) (Meier et al. 2020; Shah, Meier, and Toro Medina 2019; Meier et al. 2021). While the drop-tower campaigns provided microgravity durations of seconds, the New Shepard vehicle provided over three minutes of microgravity, wherein a substantial amount of simulated waste material (~10 g) was combusted. The results presented here are from the second flight test, after which gas samples were collected for trace gas analyses using a single combustion test. Also note that, due to test differences and anomalies (especially in the first OSCAR flight test), it is difficult to draw direct comparisons between the two flight tests.

NASA has pursued analyses of combustion in space on board the International Space Station (ISS) and associated vehicles. While the primary motivation for these studies is the evaluation of spacecraft fire hazards, they are also of interest for their contribution toward understanding fundamental aspects of the behavior of flames. Combustion testing on board the ISS is complicated by the significant crew hazard associated with it. Thus, much previous work was with small amounts of fuel, and under very strictly controlled self-contained atmospheres (Irace et al. 2021; Sandra L. Olson and Ferkul 2017; D. L. Dietrich et al. 2017; Ma et al. 2015; Daniel L. Dietrich et al. 2014; D.L. Dietrich et al. 2000; Nayagam et al. 1998). More recently, as part of NASA’s Saffire Project, larger amounts of fuel (56 g of cotton) were combusted in microgravity in the uncrewed Cygnus logistics module after separation from the ISS (Sandra L. Olson et al. 2023; Thomsen et al. 2022; Li and Liao 2021). These studies revealed important differences between combustion under conditions of terrestrial gravity versus microgravity. Under microgravity conditions, a flame cannot generate the convective flow resulting from the natural buoyancy of the heated effluent gases; thus, quiescent microgravity combustion results in a diffusion-limited condition. To compensate, several microgravity studies (such as Saffire) utilized forced flow to compare with terrestrial buoyant convection (Sandra L. Olson and Ferkul 2017; Thomsen et al. 2022; Li and Liao 2021; Urban et al. 2019). While these studies have provided excellent insight into combustion under microgravity and have informed the designs of more practical combustion devices, note that they used controlled experimental arrangements utilizing 1- or 2-dimensional single-component fuels with (typically) imposed flow parallel to the fuel surfaces. Though very informative, it is not clear whether these results are applicable to the combustion of mixed materials, such as a comprehensive crew waste stream. Additionally, while forced flow can be designed into a trash reactor, it will not have controlled gas flow over all fuel surfaces. The combustion of trash in microgravity represents a more complex process and provides the motivation for the OSCAR test campaigns.

For crew safety, NASA pays careful attention to the constituents in spacecraft cabin air and maintains a list of potentially hazardous and ubiquitous compounds of greatest concern called the Spacecraft Maximum Allowable Concentrations for Airborne Contaminants (SMAC) list (Ryder 2022). The SMAC list includes exposure limits that have been determined from careful appraisal of the physiological effects of the contaminants. The SMAC list includes 56 entries (see Supporting Information) encompassing 70 individual chemical species, including 64 VOCs. Exposure limits are organized according to durations from 1 hour to 1000 days. SMAC exposure limits are not equivalent to other similar measures, such as OSHA’s Permissible Exposure Limit (PEL). In general, the SMAC limits decrease (or in some cases remain the same) with increased exposure duration, though some species are not given exposure limits for all durations.

The SMAC list is not exhaustive regarding potential gaseous hazards on a spacecraft. Indeed, analyses of cabin air on the ISS have observed additional contaminants, including 1-propanol, ethyl acetate, R134a, Freon 218, tetramethylsilane, acrylonitrile, and isobutane (Lewis 2015). However, the SMAC list forms the basis for airborne crew hazards and is emphasized here. The analyses described here included other non-SMAC contaminants that are routinely measured on the ISS.

As mentioned, the OSCAR system was developed for both flight and laboratory operations; it flew as a payload aboard the Blue Origin New Shepard vehicle during the NS-17 mission on August 26, 2021. The flight provided >3 min of microgravity during which it combusted ~10 g of simulated crew trash (Pitts et al. 2022). While pertinent aspects of the OSCAR experiments are described here, the reader is directed to the cited work for a more detailed description of the flight experiment. The OSCAR system is shown in Figure 1(A). Figure 1(B) shows a schematic of the pertinent systems of OSCAR, including the reactor, oxygen supply, trash injection tubes, condenser/filter, and gas collection tanks. For the NS-17 flight, ~8 g of trash simulant was placed into the reactor, and ~1 g of trash simulant was placed into each of two injection tubes, for a total of 10 g (see Supporting Information for a complete description of the waste simulant). Upon achieving microgravity conditions, the reactor was heated, thus initiating the combustion of the trash. The trash from the injection tubes was pneumatically injected into the reactor at set times to sustain the combustion under microgravity. Toward the end of the microgravity phase, solenoid valves opened sequentially to collect the product gases (after filtering via the heat condenser/filter) into collection tanks. Thus, the combustion product gas collection was performed only under microgravity.

Figure 1.

(A) A photograph of the OSCAR system mounted in the flight enclosure prepared for mission NS-17 at the Blue Origin Launch Site One near Van Allen, TX. (B) Schematic of the pertinent systems of OSCAR. The blue arrows indicate the direction of gas flow. Note that the gases in the collection tanks were filtered, whereas the gases from the reactor were not.

The OSCAR reactor/gas system is sealed from the external atmosphere. Feedstock O₂ from a 103 L tank at 1000 psig (6205 kPa) regulated down to 100 psig (589 kPa) supplied oxygen at a constant flow rate of ~3 SLPM owing to a 0.0094” orifice located upstream of the reactor (890 cm³), and a back pressure regulator set at 30 psig located downstream of the reactor. This higher pressure of pure oxygen was required to complete combustion in the time allowed during the microgravity phase of flight. The rate of flow and reactor volume resulted in a residence time of 139 s. Pure O₂ was used to prime the reactor before ignition and maintain the reaction until all the fuel was consumed. Oxygen entered the reactor from heated feedlines at 650°C at two inlets that were angled tangentially to the reactor circumference to create a gas vortex flow to maximize mixing and compensate for the lack of buoyant convection in microgravity. The reactor reached a maximum temperature of ~800°C.

The combustion gas effluent was passed through a multistage heat exchanger/condenser/filter upon exiting the reactor. The first filter was located inside the reactor and consisted of a small metallic 100-mesh screen that prevented unburnt fuel particles from exiting, especially during micro-gravity. The second filter was provided by glass wool at the entrance and exit of the condenser/filter. The condenser/filter was used to filter, condense, and dry the combustion products. Condensation was achieved by passive convection heat transfer via the finned dissipative condenser. Moisture was collected via a 3:1 mass ratio of silica gel and 4A mol sieve 4–8 mesh beads between the glass wool plugs. Lastly, the cooled gas was filtered through a 40 micron-rated sintered stainless-steel filter prior to the collection tanks.

The product gases were sequentially discharged into the three collection tanks (T1-3). T1 was collected before the initiation of combustion and represents a control sample. T2 was collected after combustion was initiated and included the first trash injection tube. T3 was collected after T2 and included the second trash injection tube. Upon flight completion, each collection tank and the reactor remained sealed via the closed solenoid valves.

Within ~2 hours post-flight, hand valves were closed to isolate the reactor and the collection tanks that were then retrieved. Samples were then transferred from each collection tank into grab sample containers (GSCs, for trace gas analyses) by connecting directly to each collection tank. A control sample of room air was also collected. The following day, the reactor was retrieved from the New Shepard capsule, whereupon an additional GSC was used to collect gases directly from the reactor. The OSCAR gas system remained under positive pressure after the flight, thus eliminating the possibility of contamination from the atmosphere.

Triplicate analogous terrestrial laboratory tests were completed following the flight operations. These tests were performed with identical timings as for the flight experiment (i.e., actuation of pertinent solenoid valves, duration of heater cartridge actuation, timing of trash injections, etc.). For the lab tests, corresponding gas samples were collected for trace gas analyses. The only procedural difference between laboratory and flight tests was that reactor gases were collected on the same day for the laboratory experiments versus the next day for the flight experiment. However, temperature data indicated that the reactor temperatures were below 100°C for the flight experiment during the time for which laboratory gas samples were collected. Thus, no additional combustion occurred before the collection of the reactor flight gas samples. Gas samples were analyzed by the Environmental Chemistry Laboratory at NASA’s Johnson Space Center, Houston, TX.

The preparation of the GSCs included leak checking, cleaning, proofing (verifying cleanliness), evacuating, and dosing with a surrogate. The GSCs were leak-checked to ensure that valves and fittings were not damaged and were able to maintain a sufficient vacuum before sampling. See Supporting Information for more details on the GSC preparation procedure.

Following the return of the gas samples to the Environmental Chemistry Laboratory, each canister’s sample pressure was measured. This initial pressure and the dosing information provided the calibration for each canister. Subsequent GC/MS analyses quantitated the recovery of each of the compounds. The VOC analyses were based on EPA Method TO-15A, which was modified to accommodate constraints related to spaceflight operations. See Supporting Information for a complete description of the trace gas analysis method.

The analyses generated a large dataset whose raw results (reported in mg/m³) are provided in the Supporting Information. A total of 93 compounds were evaluated. Of these, 28 compounds were not observed in any of the samples (i.e., below the reporting limit) and are listed in Table 1 (all halogens). No fluorinated or brominated VOCs were detected. Additionally, few chlorinated compounds were observed beyond mono-substituted species.

Of the polychlorides, only 1,2-dichloroethane and 1,1,2,2-tetrachloroethane were observed in any of the samples, with the latter observed in the reactor gases of only one of the post-flight triplicates (0.2 mg/m³), and the former in all three of the post-flight reactor samples (0.44, 0.28, and 0.41 mg/m³). Chlorinated compounds that were observed included chloromethane, vinyl chloride, chloroethane, and 3-chloropropene. Therefore, it appears that the combustion process/waste simulant used here does not produce a significant amount of polychlorides. This is not surprising since chlorine represents only 0.11% by mass of the waste simulant. For easier comparison between the lab and flight tests, the 63 observed compounds were categorized into groups: carbonyls, other carbohydrates (i.e., CHO), aromatic hydrocarbons, aromatic halogens, furans and dioxins, aliphatic hydrocarbons, nitrogen-containing (CHN) compounds, aliphatic halogens, thiols and sulfides, and siloxanes. To compare between tests, the total mass of each VOC was calculated from the pressure and volume of the pertinent collection tank/reactor and then normalized by the amount of trash combusted for each test (calculated from the differential dry mass of the trash before and after combustion). The resulting values are then reported in mg VOC/kg trash combusted and are shown in Table 2. Here, the lab average values include 95% confidence intervals (CIs). Where no intervals are provided, only a single lab trial was measured above the reporting limits; for these instances, the individual result is given. Cells shaded in blue denote chemical categories where the flight value falls outside of the 95% CI for the lab measurements.

The differences between the flight and laboratory tests are apparent. The flight results fall outside of the 95% CI for most of the VOC categories (where replicate measurements were observed), as well as for the total VOCs measured. Thus, VOC production from combustion in microgravity is statistically differentiable from combustion in a terrestrial laboratory. The compositions of the VOCs also vary, as shown in Figure 2.

Figure 2.

Comparison of VOC % mass compositions by category.

(A) The composition of the VOCs observed in the flight reactor.

(B) The average composition of the VOCs from the terrestrial laboratory analog tests.

Figure 2 shows important differences between terrestrial and microgravity combustion. Specifically, the microgravity VOC composition shown in Figure 2(A) favors the production of both aliphatic and aromatic hydrocarbons relative to the VOC composition from laboratory tests shown in Figure 2(B) (total of 60.25% for flight/microgravity versus 33.57% for lab tests). Furthermore, the microgravity combustion produced a lower fraction of carbonyls (26.50% for flight/microgravity versus 55.87% for the lab tests); indeed, carbonyls constitute the majority of the VOCs for lab tests.

These results suggest that laboratory combustion was more efficient than for microgravity. The statistically significant larger mass of VOCs for the flight/microgravity test indicates that microgravity combustion was less complete. This agrees with the previously reported lower solid-to-gas conversion observed for the flight/microgravity test than for the laboratory tests (Pitts et al. 2022). Specifically, aliphatic and aromatic compounds represent fuels that combust to form CO₂ and water in an efficient system. Here, however, it was observed that the flight/microgravity test left a larger fraction of these hydrocarbon compounds unreacted than the lab tests. Additionally, the higher composition of carbonyls (a combustion intermediate) for the lab tests shows a greater degree of oxidation than for microgravity. This corroborates reported comparisons between these tests, indicating a higher combustion efficiency (based on the CO₂:CO ratio) of the laboratory tests than the microgravity test (Meier et al. 2021). This also agrees with other reports of decreased combustion efficiency in microgravity due to the lack of buoyant convection (Cao et al. 2015; S. L. Olson 1991; Urban et al. 2019).

Of the 93 analyzed compounds, 37 of them appear on NASA’s SMAC list (described above). Of those 37 SMAC compounds, 31 were observed in the OSCAR reactor either in the flight/microgravity test and/or in one or more of the laboratory tests. Table 3 lists the observed SMAC compounds and their reactor concentrations (in ppm), along with the SMAC concentrations for their different exposure durations.

The concentrations shaded in blue indicate SMAC compounds with microgravity concentrations outside of the 95% CI for the laboratory tests; in all but one of these cases (acrolein), the microgravity concentration was higher than the CI upper limit. This is consistent with the higher total VOC production observed for the flight/microgravity test. The lone exception is acrolein, whose flight/microgravity concentration was lower than the lower limit of the 95% CI for the lab tests (this is further discussed below). Figure 3 graphically represents these concentrations, with each plot displaying a successively higher concentration range.

Figure 3.

Plots of the flight/microgravity SMAC concentrations (blue squares) with the laboratory test average concentrations (black line) and the upper/lower boundaries of the 95% CI (gray dashed lines).

Again, significant differences can be observed between the flight/microgravity test and the laboratory tests. Ten SMAC compounds for the flight/microgravity test fell outside of the 95% CI for the laboratory tests. These differences suggest that the VOC production for combustion in microgravity is significantly different than for terrestrial gravity.

Regarding the SMAC limits, concentrations in the reactor are critical quantities since they are higher than those in the collection tanks. A TtG system must be considered within the context of a transit vehicle. The current Mars mission architecture proposes a vehicle habitable volume of 116 m³ (Ewert, Chen, and Powell, Camilah D. 2022). The authors’ design considerations for a TtG reactor have a maximum volume of ~0.0165 m³. Thus, for a worst-case critical failure scenario allowing the reactor gases to escape into the cabin, the volume ratio is 7,030:1. Table 4 shows the flight/microgravity SMAC concentrations as a percent of each SMAC limit before and after this dilution.

Both the undiluted (UD) reactor concentrations and the diluted (D) concentrations are included in the table. Where the result equals or exceeds the SMAC limit, the cell is shaded red. None of the diluted concentrations exceed the SMAC limits. The highest is furan at 110,000%, or ~1000-fold excess of its 180-day SMAC limit. But even for furan, dilution into the habitable volume results in 15.6% of the 180-day SMAC limit. Interestingly, acrolein was higher in concentration for laboratory tests (51.61±8.88 ppm) than for the flight test (1.19 ppm). Acrolein is a carbonyl that has been reported to be a product of the oxidation of aliphatic species as well as complex materials (Sun et al. 2023; Hartikainen et al. 2018; Shields et al. 1995; Terrill, Montgomery, and Reinhardt 1978; Einhorn 1975). The SMAC limits for acrolein are shown in Table 3. For the lab tests, undiluted percentages for acrolein are 68,800%, 147,000%, 344,000%, 344,000%, and 645,000%, and diluted percentages are 9.8%, 20.9%, 48.9%, 48.9%, and 91.7%, for the respective durations. These are the highest values for all of the SMAC compounds observed. Thus, acrolein represents an important hazard, especially as microgravity TtG reactors are developed with higher efficiencies.

The primary conclusion here is that the difference in VOC production between combustion in microgravity versus terrestrial gravity is significant. There was a statistically significant higher VOC production for microgravity than for the laboratory tests. Indeed, for total VOCs, the flight values fell outside the 95% CI by as much as 32 times the range of the CI from the lab test averages. While this corroborates the previously reported observation that the flight test showed a decreased ratio of CO₂:CO versus the lab tests, the statistical treatment of the CO₂:CO ratio was not definitive. (Pitts et al. 2022) In that case, the values for the microgravity test fell just within the 95% CI for the three laboratory results; thus, the null hypothesis (that the measurements are not differentiable) could not be rejected. In the case of VOCs, however, the rejection of the null hypothesis was quite definitive, indicating a statistically significant difference between the combustion of trash in microgravity versus terrestrial gravity.

The reason for this difference is less apparent. While the flight test was very successful, there was one important anomaly. Namely, there was a coarse steel wire mesh in the reactor meant to hold the trash near the bottom (as for laboratory tests) that moved during flight due to the vibrations associated with the launch. As a result, under microgravity, the trash migrated into a larger volume within the reactor. Laboratory tests with the mesh in different positions revealed a strong dependence of the combustion efficiency (via the CO₂:CO ratio) on the mesh position. For laboratory tests with the mesh partially raised to the height observed during the flight (measured from the flight reactor video), a larger CO₂:CO ratio was observed (26.1:1 for the lab test versus 12.0:1 for flight) (Pitts et al. 2022). Furthermore, lab tests with the mesh fully raised (to the top of the reactor) showed a ratio of 77.5:1, corresponding to more efficient combustion and suggesting that the mesh is a flame and heat barrier. However, trace gas analyses were performed only for the ‘fully raised mesh’ laboratory tests (at the time understood to be the position of the mesh during the flight) corresponding to more efficient combustion. Note that correlations with combustion efficiencies and CO₂:CO ratios have been reported for numerous fuel types (Matthes and Hartmann 2017; Gopan et al. 2017; Chien, Escofet-Martin, and Dunn-Rankin 2016; Svoboda et al. 2009; Wallington, Kaiser, and Farrell 2006; Lee et al. 2005; Johansson et al. 2004; Cliffe and Patumsawad 2001; Hasselriis, Floyd 1992). In any case, it is apparent that combustion in microgravity is less efficient than in an equivalent terrestrial arrangement. Furthermore, the raised mesh/larger volume of the flight test is an important result since it more closely approximates a likely reactor scenario for a full-scale system.

The SMAC list is not exhaustive regarding harmful compounds, especially considering the myriad VOCs present in a combustion effluent. If thermal processes are used to process trash on spacecraft, a more comprehensive appraisal of the potential hazards in the effluent gases will be required. Additionally, the SMAC list includes hazardous species that were not analyzed here and should be evaluated in future work.

While a final reactor design should protect against leaks into a spacecraft’s habitable volume, this study suggests that the resulting dilution should provide some protection even in the event of a catastrophic failure; however, it must be noted that VOC concentrations in a full-scale reactor might be higher than for the sub-scale OSCAR reactor. Additionally, hazardous VOCs could be reduced by effective filtering/sequestering and/or other abatement methods, such as reducing VOCs in a plasma system. It is likely that a final TtG system will include multiple abatement systems. Indeed, abatement systems (such as plasma) might be used to produce commodities as feedstocks for other systems. These results provide a foundational basis from which to evaluate the performance of such VOC abatement systems.