Filtration of simulated Martian atmosphere for in-situ oxygen production
Introduction
In-Situ Resource Utilisation (ISRU) has the potential to dramatically reduce the mass and cost of planetary missions. For example, producing 1 kg of consumable in situ on Mars reduces the payload mass required to be launched from Earth by approximately 10 kg (Sanders, 2018). To increase the technology readiness level of ISRU, the Mars 2020 rover Perseverance will carry the Mars Oxygen ISRU Experiment (MOXIE). MOXIE is a small-scale ISRU technology demonstrator that will produce oxygen from Mars’ atmospheric carbon dioxide using solid oxide electrolysis (Hecht and Hoffman, 2015).
Mars’ atmosphere contains a large amount of dust that, if ingested by MOXIE, may contaminate the electrodes of the solid oxide electrolysis subsystem. MOXIE is protected from dust by a High-Efficiency Particulate Air (HEPA) filter. However, as dust accumulates on the filter, the pressure drop across it will increase. Should the pressure drop become sufficiently large, there is a risk that MOXIE’s compressor will not be able to deliver the required mass flow rate of carbon dioxide, limiting the rate of oxygen production.
The dust loading rate and pressure drop versus dust loading relationship of fibrous filters have been extensively investigated in terrestrial conditions, where the continuum and slip flow regimes apply (see, for example, the textbooks of Brown, 1993; Thomas et al., 2017). However, fewer studies have examined low-pressure conditions where the transition and free molecular flow regimes apply (Li et al., 2014), and even fewer have explored Martian conditions, where both a low-pressure carbon dioxide atmosphere and Mars dust simulant are required. Within the context of filtration for ISRU, previous work has focused on cabin air filtration, instead of atmospheric filtration (for example, the Scroll Filter System (SFS), Agui and Perry, 2017; Agui et al., 2017); lunar dust, instead of Martian dust (Agui, 2008); flat filters, instead of pleated filters (Agui, 2016); or electrostatic removal of dust, instead of removal by fibrous filters (Calle et al., 2011).
In addition to understanding the level of risk presented by filter pressure drop increase, experimental characterisation of the MOXIE HEPA filter is important for several reasons. First, it is difficult to predict filter performance analytically without accurate knowledge of the dust Particle Size Distribution (PSD), particularly at the HEPA filter Most Penetrating Particle Size (MPPS) of approximately 0.3 . Although models of the dust PSD on Mars are in general agreement above 1 , they vary over several orders of magnitude below 1 (Esposito et al., 2011). This uncertainty hampers efforts to predict the dust loading rate and the pressure drop versus dust loading relationship. Second, it will not be possible to monitor the dust loading on the MOXIE HEPA filter on Mars by visual inspection, as the intake is covered by a baffle. Third, the risk presented by dust increases in severity for potential full-scale ISRU plants after Mars 2020. Compared to MOXIE, a full-scale ISRU plant will operate for longer, at higher flow rates, and continuously instead of intermittently, introducing the possibility of operation in a wide range of dust particle number densities. Depending upon filter design, this may result in a higher dust loading rate and pressure drop.
This paper addresses two main objectives and seven sub-objectives that relate to the performance of HEPA filters in simulated Martian conditions. The two main objectives are to determine the dust loading rate, , and pressure drop as a function of dust loading and filtration velocity , for the MOXIE HEPA filter. Dust loading, m, is defined as the mass of dust captured by the filter per unit filter media area. Only suspended dust is considered. Saltated (wind-blown) dust is excluded as it is assumed that MOXIE’s baffle will deflect all saltated particles. The seven sub-objectives are to characterise how the dust loading rate is affected by particle size, filter geometry (pleated or flat), whether or not carbon dioxide is being actively drawn through the filter (active or passive dust loading), whether or not the filter is fitted with a baffle, and filter orientation to the wind; and how the pressure drop is affected by inlet face velocity and wind tunnel pressure.
It was expected that the dust loading rate would be reduced for larger particles and pleated, passive, and baffled filters aligned at an angle of 90∘ to the wind, and that the pressure drop would increase with dust loading, inlet face velocity, and atmospheric pressure. These hypotheses, and the underlying theory supporting them, are discussed in Sect. 2. The dust loading rate was determined by exposing various filters to simulated Martian conditions and measuring their mass before and after exposure. The pressure drop as a function of dust loading was determined by loading flat filters to various dust loadings, measuring the pressure drop versus inlet face velocity for each one, and measuring their mass before and after loading. Materials and methods are discussed in Sect. 3. Results are reported and compared to theoretical predictions in Sect. 4, before Sect. 5 concludes with implications for MOXIE, Mars 2020, and future atmospheric ISRU on Mars.
Section snippets
Dust loading rate
Dust loading, m, is the mass of dust captured by the filter per unit filter media area. The dust loading rate, , can be estimated as:where is the dust particle density, and are the dust particle number density and diameter, is the inlet face velocity, and and are the areas of the inlet and filter media. Eq. (1) assumes that the particle density, number density, and size are uniform. Therefore, when applied in previous work, typical values for these
Equipment
The Mars Simulation Laboratory at Aarhus University, Denmark, has several wind tunnels to study dust transport. The Aarhus Wind Tunnel Simulator II (AWTSII, the “large wind tunnel”) is a 2.1 m diameter, 10 m long recirculating wind tunnel (Holstein-Rathlou et al., 2014) that was used to address the first objective by exposing various filters to simulated Martian conditions for a set amount of time. The Aarhus Wind Tunnel Simulator I (AWTSI, the “small wind tunnel”) is a 0.8 m diameter, 3 m
Dust loading rate
Table 3 shows the dust loading rates for all of the filter configurations and dust simulants tested in the large wind tunnel. Fig. 11 shows the increase in the mass of the flight-representative filter versus time.
The dust loading rate was found by taking the slope of the best-fit linear trend, resulting in (0.19 0.02) mg m−2 h−1, which is equivalent to (4.6 0.4) mg m-2sol-1. This dust loading rate was found using an inlet face velocity of 7.1 cm s−1. Dust loading rates for other inlet
Conclusion
In-situ resource utilisation is a powerful way to reduce the mass and cost of planetary missions. MOXIE will demonstrate ISRU of atmospheric carbon dioxide on Mars 2020, however dust ingestion is a risk. An experimental investigation was carried out to determine the dust loading rate and the pressure drop as a function of dust loading for the MOXIE HEPA filter.
The dust loading rate was found to be (0.19 0.02) mg m−2 h−1 using a flight-representative filter (Sect. 3.1.1). The dust loading rate
Declaration of competing interest
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: A. Wiegmann is the co-founder and CEO of Math2Market GmbH. M. Azimian is an employee of Math2Market GmbH.
Acknowledgements
This study was funded by Europlanet, the Danish Council for Independent Research grant no. 4002-00292B, the UK Space Agency, and the Val O’Donoghue scholarship at Imperial College London. Europlanet 2020 RI has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 654208. We thank Morten Bo Madsen for initial funding, Robin Vinther Nielsen and Simon Jacobsen for valuable assistance during the testing, Jeff Mellstrom and Adrian Ponce
References (27)
- et al.
An environmental simulation wind tunnel for studying Aeolian transport on Mars
Planet. Space Sci.
(2008) - et al.
Salten Skov I: a Martian magnetic dust analogue
Planet. Space Sci.
(2009) - et al.
Clogging of HEPA fibrous filters by solid and liquid aerosol particles: an experimental study
Filtrat. Separ.
(1999) Pressure characteristics of fibrous aerosol filters
J. Colloid Interface Sci.
(1971)- et al.
Airborne nanoparticles filtration performance of fibrous media: a review
Sci. Technol. Built Environ.
(2018) Flow and particulate transport modeling of a lunar dust filtration testing facility
(2008)Filter media tests under simulated martian atmospheric conditions
- et al.
Life Support Filtration System Trade Study for Deep Space Missions
(2017) - et al.
Exploration mission particulate matter filtration technology performance in a simulated spacecraft cabin ventilation System
Air Filtration: an Integrated Approach to the Theory and Applications of Fibrous Filters
(1993)
Dust removal technology for a Mars in situ resource utilization System
Les fontaines publiques de la ville de Dijon. Exposition et application des principes à suivre et des formules à employer dans les questions de distributions d’eau
Victor Dalmont
Air Filtration
Cited by (0)
- 1
Now at MIT Haystack Observatory, 99 Millstone Rd, Westford, MA, 01886, United States ([email protected])