香京julia种子在线播放

The WRS can be subdivided into two assemblies operating in concert with each other. The Water Processor Assembly (WPA) collects condensate from the cabin atmosphere originating from crew perspiration and respiration, water from the OGS produced during the Sabatier reaction, and distillate from the second module, the UPA. The UPA’s position in the ECLSS is shown in Figure 1. In Figure 2, a schematic overview of the UPA module is provided (Bagdigian et al., 2015; Volpin et al., 2020).

In total, an average of 1.80l of urine (4–9gN/L d⁻¹) and urine flush water per crew member per day is collected in the Wastewater Storage Tank Assembly (WSTA). Here, the urine load is kept sterile and is chemically stabilized to avoid scaling (precipitation of solid minerals in urine collection systems) by mixing it with an H₃PO₄ and Cr⁶⁺ solution. Originally, H₂SO₄ played the role of H₃PO₄ in the system, acidifying urine and converting the volatile ammonia (NH₃) originating from the hydrolysis of urea to non-volatile ammonium (NH₄⁺). However, astronaut’s urine contains higher amounts of Ca²⁺, which can precipitate as CaSO₄ and cause scaling and pipeline clogging which eventually leads to system failure. This issue was initially addressed by reducing UPA water recovery from 85 to 75%, which prevented the CaSO₄ concentration from reaching its solubility limit in water. The introduction of H₃PO₄ to acidify stored urine significantly reduced SO42-content and hence scaling potential. This allowed the increase of water recovery efficiency up to 85% again. Cr⁶⁺ acts as an oxidizing agent, preventing urea hydrolysis and thus avoiding NH₄⁺ formation (Muirhead et al., 2018; Volpin et al., 2020). Chemically stabilized urine is pumped from the WSTA to the distillation assembly (DA) after enough urine has been collected in the storage tank (Bagdigian et al., 2015; Anderson et al., 2018; Volpin et al., 2020).

In the DA, water is separated from waste products through evaporation, leaving a concentrated waste mixture behind. This waste stream passes through the Advanced Recycle Filter Tank Assembly (ARFTA) where concentrated brine is collected with a series of membrane filters (Link et al., 2010; Bagdigian et al., 2015; Volpin et al., 2020). A downstream brine filter removes leftover solid precipitates formed during the process (Link et al., 2010). After every cycle, the ARFTA becomes saturated and is emptied and the removed concentrated brine is stored until eventual disposal. The resulting filtered waste stream is then returned to the DA for a new distillation cycle. Finally, the produced water vapor from the distillation process is separated from purge gases before passing to the WPA (Bagdigian et al., 2015; Volpin et al., 2020). In the WPA, water undergoes the final purification steps. Here, contaminants are removed in a four-step process: (1) removal of inorganics and non-volatile organic compounds in an ion-exchange resin multifiltration bed, (2) oxidation of low molecular organics with oxygen at 130°C, (3) liquid–gas separation of oxygen and other leftover gaseous by-products, and (4) removal of carbonate and bicarbonate ions through an ion-exchange system (Volpin et al., 2020).

Although the current ECLSS is suitable for space missions in Low Earth Orbit (LEO), there are several limitations that need to be addressed to enable its use in long-term space missions. In order to make long-distance space exploration feasible, a life support system should recover more than 98% of water and nutrients from waste streams and produce food to meet the human metabolic needs, with minimal use of additional resources (Pickett et al., 2020). The UPA operating efficiency of only 85% results in a significant loss of available H₂O. In addition, not all water vapor that can be used for oxygen production is condensed out of the CH₄ waste stream in the OGS, thereby venting valuable water to space. The combination of the H₂ lost under the form of CH₄ (as H-atoms) and the loss of uncondensed H₂O produced in the Sabatier reaction results in only 50% of O₂ recovery produced with the Sabatier reaction from metabolic CO₂ (Greenwood et al., 2018). Consequently, though designed for high reliability, the current ECLSS does not meet the necessary requirements for long-distance space travel. Water resupply from Earth is necessary to enable system operation. Moreover, 0.21kg of disposable hardware (saturated filters, malfunctioning hardware, etc.) is consumed per 1l of potable water produced, which need to be replaced during resupply (Volpin et al., 2020). In conclusion, while payload size and storage requirements on the ISS have been vastly reduced by the use of the ECLSS, they are still considerable and should be minimized further. Moreover, energy requirements to employ ECLSS modules are substantial (Jones et al., 2014). Hence, technological improvements (e.g., higher water recovery efficiencies, nutrient recovery, and higher system reliability) or a switch to new innovations, such as a BLSS, is required (Jones, 2016). The latter would allow tapping on waste streams rich in valuable compounds (N, P, and K), such as brine produced from urine during water purification and solid wastes, both of which are currently stored to be incinerated upon re-entry into Earth’s atmosphere. These compounds could be alternatively used for the production of fertilizer for food production (Volpin et al., 2020). It would help to (partially) solve the inability of the current system to produce food and reduce the need for frequent supply missions to space.

Russia spearheaded the development of BLSSs with the development of the Hybrid Biosphere System (BIOS) program, which was a large scale confined analogue environment manned by humans. It used microalgae and higher plants to meet a person’s oxygen demand (BIOS-1 and BIOS-2) and later expanded to also include plant-based food production (BIOS-3). Here, unprocessed human urine (containing urea) was mixed directly with nutrient solutions to successfully cultivate higher plants hydroponically (Salisbury et al., 1997; Guo et al., 2017). Other similar projects are the Biosphere 2 project of NASA, the Closed Ecological Experiment Facility (CEEF) of the Japan Aerospace Exploration Agency (JAXA), and the Lunar Palace 1 program of the National Natural Science Foundation of China. These facilities were built with a focus on in situ cultivation of higher plants and/or animals for future Lunar or Martian bases (Ashida and Nitta, 1995; Guo et al., 2017; Xie et al., 2017). The Biosphere 2 consists of a 1.27ha structure of different biomes (marshland, rainforest, savannah, etc.). In the 1990s, it was used to study artificial closed ecology systems with a human crew. Human urine was deposited in the marshland biotope, where it was digested through natural processes and bioavailable nitrogen could be redirected toward the remaining biomes (Nelson et al., 1999; Guo et al., 2017). In the CEEF, JAXA chose to incinerate human fecal material, urine, and inedible crop waste in an effort to recover CO₂ for crop production. No nitrogen recovery system was put in place (Tako et al., 2017). More recently, China completed a 106-day terrestrial manned mission in the Lunar Palace 1 facility. A BLSS was successfully used to support a crew of four with minimal external input of food, water, and oxygen for the duration of the trial (Xie et al., 2017; Cheng et al., 2019). During the experiment, nitrogen was recovered from urine with an efficiency of 20.5% through reduced-pressure distillation, where ammonia was distilled together with the water vapor. The distillate from this process was combined with kitchen and sanitary wastewater and processed in a membrane bioreactor followed by activated carbon treatment. The former is populated by bacteria (including nitrifying strains) originating from the Lunar Palace environment and wastewater stream (Xie et al., 2017). It is worth mentioning that this system design was mainly focused on pollutant removal rather than optimization of nutrient recovery. In the second phase of Lunar Palace, an anaerobic and aerobic membrane bioreactor will replace the current setup. In preliminary experiments preceding a new manned mission in Lunar Palace, successful ureolysis has been observed in the anaerobic reactor, while the ammonium removal efficiency in the aerobic reactor sits between 80 and 99% during stable operation (Cheng et al., 2019).

The German Aerospace Center (DLR) initiated two smaller scale BLSS research projects. In the 1990s and the 2000s, the Closed Equilibrated Biological Aquatic System (CEBAS) combined aquatic vertebrates and invertebrates with microalgae, aquatic plants and nitrifying bacteria in a closed-loop ecosystem. An ammonia oxidizing biofilter, which contains bacteria from the Nitrosomonas and Nitrobacter genera, processed the excreted ammonium from organisms in the aquatic animal tank (Blüm et al., 2003). The second, more recent Combined Regenerative Organic-food Production (C.R.O.P) project utilizes a biofiltration process with a natural bacterial community (including Nitrosomonas and Nitrobacter) that populates a trickle filter. In this setup, nitrogen can be recovered from synthetic human urine that carries ammonium as N-source. The microalgae species Euglena gracilis provides oxygen necessary for nitrification activity in this setup. Tomato plants are cultivated using the resulting nutrient stream from the trickle filter. Depending on the dilution of a synthetic urine matrix, a 66–87% nitrogen recovery efficiency was realized. Here, an undiluted feed negatively impacted the nitrate production rates, while 60 and 80% diluted urine seemed to be within the optimal range for nitrogen recovery (Bornemann et al., 2015, 2018; Hauslage et al., 2018).

European Space Agency initiated the ambitious MELiSSA program in the late 1980s. It is currently the longest-running BLSS program to date. Based on a lake ecosystem, MELiSSA aims to develop a closed-loop bioregenerative system using an engineering-based approach. Five interdependent compartments utilize a variety of specific microorganisms or higher plants to produce oxygen, water, and food from human metabolic waste (Hendrickx et al., 2006; Lasseur et al., 2010). In contrast to the previous described projects and facilities that focused on the development of a large analogue human test facility, the MELiSSA Pilot Plant, a testbed facility of the MELiSSA loop, is designed to sustain the respiration needs of a single person while also providing 20–40% of necessary nutrition (Alemany et al., 2019). Hence, it is not designed to be a closed full-scale testing facility that can fully support a complete crew. The plant has been operative since 2009. The MELiSSA program is the oldest and most advanced circular BLSS currently in development. At the same time, it serves as a pioneer for the terrestrial circular economy as it provides state-of-the-art technology that can also be applied on Earth (Paladini et al., 2021).

In the MELiSSA context, astronauts act as consumers, microorganisms are the recyclers, and plants and cyanobacteria close the loop as the producers (Figure 3; Lasseur et al., 1996; Hendrickx et al., 2006). In compartment I (CI), a consortium of bacteria digest/liquefy fecal and solid wastes and urine mainly into CO₂, volatile fatty acids (VFAs), and minerals, including free NH₄⁺. Bioreactor conditions are thermophilic and anaerobic to prevent possible pathogen proliferation and methanogenesis, respectively. Methane is a potentially hazardous and flammable biogas and hence it should be avoided on space vessels. Moreover, its potential use as a biogas fuel results in loss of carbon from the loop while the biological recovery of carbon from methane implies the introduction of two more bioreactors in the already highly complicated loop (Hendrickx et al., 2006; Hendrickx and Mergeay, 2007). Furthermore, carbon sources will also be lost as methane and recovering carbon from methane implicates two additional compartments, overcomplicating the MELiSSA design (Hendrickx et al., 2006). VFAs and minerals are further processed in Compartment II (CII) by the photoheterotrophic bacterium Rhodospirillum rubrum to produce biomass and additional NH₄⁺ (Hendrickx et al., 2006; Mastroleo et al., 2009). Produced biomass could potentially serve as an additional nutrition source, while NH₄⁺ is diverted to the third, nitrifying compartment (CIII; Hendrickx et al., 2006; Mastroleo et al., 2013). Moreover, removal of VFAs from the waste stream is necessary to ensure proper functioning of CIII, since its presence negatively affects nitrite oxidation in the dedicated compartment (Oguz et al., 2006; Mastroleo et al., 2013). Nitrogen availability is crucial for cyanobacteria (Limnospira indica) and higher plant growth in compartment CIVa and CIVb, respectively (Paradiso et al., 2014; Clauwaert et al., 2017; Alemany et al., 2019; Poughon et al., 2020). As such, to ensure optimal food production, an efficient nitrogen recovery system is indispensable. Although nitrogen fixation from urea by plants is possible, studies have shown that it is ineffective as plant fertilizer (Ilgrande et al., 2019a). Plants prefer to assimilate nitrogen from inorganic NH₄⁺ and NO3-molecules while L. indica prefers NO3-over NH₄⁺ and urea (Clauwaert et al., 2017; Ilgrande et al., 2019a). Therefore, in MELiSSA, urea is biologically converted to NH₄⁺ and NO₃⁻. In the context of a BLSS for space applications, NO3-is especially preferred due to the fact that NH₄⁺ can easily convert to the volatile NH₃, which is toxic at high concentrations in atmosphere. It can be harmful to the astronaut’s health if leaked into a closed environment, such as the ISS or any other spacecraft (Ilgrande et al., 2019a). Finally, NH₄⁺ is also detrimental for plant growth at high concentrations (Paradiso et al., 2014; Ilgrande et al., 2019a). In CIII, NH₄⁺ is oxidized to nitrate in a two-step process using O₂ as electron acceptor. Nitrosomonas europaea is an ammonium-oxidizing bacterium that converts ammonium to nitrite, which is subsequently consumed by the nitrite oxidizing bacterium (NOB) Nitrobacter winogradskyi. Here, nitrogen is effectively converted to the form preferred by plants or cyanobacteria. It serves as a fertilizer in the subsequent biomass-and oxygen-producing and CO₂-consuming compartment IV (Farges et al., 2012; Clauwaert et al., 2017; Ilgrande et al., 2019b). The final compartment V is the crew, who is consuming produced biomass and O₂ on the one hand, and producing CO₂, urine and fecal waste on the other hand. These waste products enter CI, closing the MELiSSA-loop.

In the MPP, bioreactors are operating and being characterized both individually and in an interconnected setup (Albiol et al., 2000; Gòdia et al., 2004). An important objective is to design associated technology that facilitates a continuous process that is robust and that runs as optimally as possible, as well as to develop mathematical models based on experimental data obtained from real operating conditions (Albiol et al., 2000).

The metabolic pathway of nitrification in N. europaea and N. winogradskyi is shown in Figure 4. Full nitrification is achieved by the nitrifiers in the following reactions (Cruvellier et al., 2016; Caranto and Lancaster, 2018):

The formation of nitrite from ammonium in N. europaea is called nitritation (1), and it is achieved in several enzymatic reactions. In the first step, ammonia is converted into hydroxylamine (NH₂OH) by the membrane-bound ammonia monooxygenase (AMO). The next step involves hydroxylamine oxidoreductase (HAO), which was originally hypothesized to catalyze the oxidation of NH₂OH to NO₂- with O₂ acting as the electron acceptor (Chain et al., 2003). Recently, however, nitric oxide (NO) has been demonstrated to be the enzymatic product of HAO instead of NO₂⁻. Meanwhile, the production of NO₂-from NO is probably catalyzed by a third, unidentified enzyme in the nitritation (Coleman and Lancaster, 2020). The presence of such an enzyme is necessary to outcompete the side reactions that can produce by-products, such as NO3-and/or N₂O, but also to prevent the spontaneous formation of NO₂⁻. The latter is required since non-enzymatic oxidation implies a loss of electrons, which are captured during the enzymatic reaction (Caranto and Lancaster, 2018). Finally, NO₂-is oxidized to NO3-by the membrane-bound nitrite oxidoreductase (NXR) of N. winogradskyi during the so-called nitratation (2), also using O₂ as an electron acceptor (Starkenburg et al., 2006; Cruvellier et al., 2016; Caranto and Lancaster, 2018; Coleman and Lancaster, 2020).

These nitritation and nitratation processes only take place in aerobic conditions since they are highly dependent on the oxidative power of O₂. An anaerobic environment imposes a lack of oxygen on the cells for the nitrification reaction. In those situations, NO3-can be reduced to NO₂-by N. winogradskyi using the NXR enzyme, which now acts as a reductase instead of an oxygenase under these circumstances (Freitag et al., 1987). N. europaea can further convert NO₂-into gaseous N₂O, NO, and N₂ (Schmidt et al., 2004). This so-called nitrifier denitrification leads to nitrogen loss and production of undesired gasses. Following the requirements of an efficient and safe BLSS, denitrification should be strongly avoided.

The nitrification reactor in the MPP is a fixed-bed bioreactor of 8.1l that houses a nitrifying co-culture of N. europaea and N. winogradskyi. It operates at a pH of 8.1 at 28±0.1°C and is magnetically stirred at the inlet to enable mixing of the bioreactor content (Montras et al., 2008). Due to the autotrophs’ characteristic low biomass growth and high conversion yield, the co-culture is grown as a biofilm immobilized on 4mm diameter polystyrene beads to avoid washout (Zeghal et al., 1994; Perez et al., 2004; Montras et al., 2008). Growth on biofilms allows decoupling of the liquid residence time in the reactor from the biomass retention time (critical for slow-growing organisms, such as nitrifiers), although substrate diffusion through biofilms can result in lower maximal conversion rates (Perez et al., 2004).

During an extended period of continuous operation (4.8years), nitrogen conversions in the order of 95–100% efficiency were reported in the MPP CIII, at ammonium loading rates of up to 1.35kgNm⁻³ d⁻¹ (Gòdia et al., 2002; Perez et al., 2004; Montras et al., 2008). Despite the observed long-term stable operation of the bioreactor, it remains important to characterize the effects of certain perturbations on the process stability in the reactor. As mentioned, a critical parameter that influences nitrification efficiency is the dissolved oxygen (DO) availability in the reactor. Depending on how extensive oxygen limitation is, different steps in the nitrification process are affected (Gòdia et al., 2002). Hence, experiments were performed to quantify the bulk DO level needed in the system to ensure full nitrification (Zeghal et al., 1994; Gòdia et al., 2002; Perez et al., 2004). In a two-step decrease of DO concentration from 80 to 40% and 40 to 20%, ammonium was not completely converted to nitrite but all nitrite was oxidized to nitrate in the first step DO concentration decrease (80 to 40%). This indicates that DO concentration in this range is limiting for N. europaea, but not for N. winogradskyi (Gòdia et al., 2002). In the second DO decrease step (40 to 20%), however, ammonium was fully oxidized after an adaption period but nitrite conversion was incomplete. Here, N. europaea regained the upper hand in competition for oxygen with its co-culture counterpart and dominated the co-culture. These results showed that at 80% DO concentration, full nitrification was achieved while lower DO concentrations caused partial nitrification. It is thus essential to monitor DO in the bioreactor in order to control the nitrification process both in steady state and in dynamic situations (Gòdia et al., 2002).

Aside from high nitrification efficiency, another requirement for MELiSSA is stringent controllability of the compartments. MELISSA considers three control levels to ensure controllability (and thus stability) of the complex biological system: (1) real-time monitoring and control of predetermined parameters in each compartment (such as pH, temperature, and dissolved oxygen (DO) in the CIII), (2) implementation of a control law for each bioreactor chamber, and (3) an intercompartment coordination system for the loop (Gòdia et al., 2004). Specifically for the CIII compartment, several mathematical models have been constructed and validated to simulate and control the MPP CIII bioreactor in real-time. For instance, a hydrodynamic model able to predict the N-species response to ammonium load disturbances was developed, hence enabling the user to keep nitrite accumulation to a minimum (Perez et al., 2005). This is required due to the latter compound’s toxic properties for both plants and humans. Using the model, one can simulate the N-species (NH₄⁺, NO₂⁻, and NO₃⁻) concentrations and biomass concentration in the fixed-bed bioreactor. The model serves as a valuable tool to maintain stable operation in CIII by identifying boundary conditions and operating the bioreactor within these boundaries (Perez et al., 2005). Beyond that, the heterogeneous distribution of the co-culture species was characterized along the length of the bioreactor after 4.8years of operation and this information was used to expand on the previous model to predict population dynamics and nitrification efficiency, while also considering diffusion of nutrients into the biofilm (Montras et al., 2008). Finally, Cruvellier et al. was able to determine that online measurements of base addition to counteracting nitritation-driven acidification (1) and online measurements of oxygen consumption can be used as predictive variables for partial nitrification (Cruvellier et al., 2016).

Nitrogen recovery from ammonium oxidation has been well characterized in ammonium-containing medium in the MELISSA CIII compartment at the MPP (Gòdia et al., 2004, 2004; Perez et al., 2015; Cruvellier et al., 2016, 2017). However, CIII cannot be used to directly process urine at present. Urine currently enters CI together with other waste products (Ilgrande et al., 2019b). Urea can be converted to ammonium by ureolytic bacteria through the action of the hydrolytic urease enzyme (Defoirdt et al., 2017; Ilgrande et al., 2019a). Urease catalyzes the conversion of urea into ammonium and carbamic acid. The latter then spontaneously hydrolyzes into a second ammonium molecule and carbonic acid, resulting in two ammonium molecules from one urea molecule and an increase in pH due to bicarbonate (HCO₃⁻) formation (Defoirdt et al., 2017; De Paepe et al., 2020b).

Full urine hydrolysis can take more than a month when relying on the spontaneous action of urease-positive bacteria populating urine pipelines and storage equipment. Storing urine for these extended periods of time is impractical for space applications (Defoirdt et al., 2017). By including an ureolytic heterotrophic bacterium directly in the nitrifying community in CIII, urea can be converted into ammonium (Defoirdt et al., 2017; Christiaens et al., 2019a; Ilgrande et al., 2019a). Low concentration organic compounds (~10g COD (Chemical Oxygen Demand) L⁻¹) present in urine should also be removed from urine by the heterotroph to provide a fertilizer free of CODs, reducing the risk of biofouling of the nutrient solution for CIV (Christiaens et al., 2019a; Ilgrande et al., 2019a; De Paepe et al., 2020a).

Currently, research on adding one (or multiple) ureolytic strain(s) to the nitrifying co-culture in the CIII compartment is ongoing. The combination of a heterotroph with nitrifying bacteria in a defined community has not been demonstrated before, as opposed to wastewater treatment with mixed microbial communities. In the same study that demonstrated halotolerance of the nitrifying co-culture by Christiaens and co-workers, the co-culture was combined with three heterotrophic ureolytic strains (Pseudomonas fluorescens DSMZ 50090, Acidovorax delafieldii DSMZ 64, and Delftia acidovorans DSMZ 14801). Complete ureolysis and nitrification were achieved at similar efficiencies as a nitrifying co-culture (107±8% and 94±8%, respectively) when treating a 5% fresh real urine feed. When using a 10% urine influent, ureolysis and nitrification efficiencies significantly dropped to 66±9% and 35±3%, respectively. In another study by Ilgrande and coworkers, five ureolytic heterotrophic strains (A. delafieldii, Comamonas testosteroni I2, Cupriavidus necator DSMZ 13513, D. acidovorans, P. fluorescens, Vibrio campbellii LMG 22895) were also combined individually with the nitrifying co-culture. C. necator and V. campbellii containing communities did not show active nitrification of NH₃. Meanwhile, ammonia oxidation rates in consortia with P. fluorescens and C. testosteroni were twice as high as those observed in A. delafieldii or D. acidovorans communities, indicating a beneficial interaction between the heterotroph and N. europaea (Ilgrande et al., 2019a). Moreover, in axenic conditions, C. testosteroni, V. campbellii, and P. fluorescens were proven to hydrolyze urea present in low concentrations, while the other strains (A. delafieldii, C. necator, and D. acidovorans) did not exhibit ureolytic activity here (Ilgrande et al., 2019a). Hence, the three former strains are favored over the latter three strains for possible inclusion in the nitrifying community since urea consumption will be more complete. Halotolerance of those ureolytic heterotrophs was also assessed and no ureolytic activity was observed for the selected species in high salinity (> 30 mS cm⁻¹) conditions in an artificial urine matrix. However, ureolytic activity at lower salinities was not inhibited, implying the need for dilution of urine before feeding it to the bioreactor (Ilgrande et al., 2019a). Finally, organics removal was limited with all influent urine concentrations and has to be improved before space applications are possible (Christiaens et al., 2019a). Nonetheless, the inclusion of an ureolytic step in a nitrifying co-culture already shows successful ureolysis and nitrification, which is encouraging to the implementation of direct urine treatment in CIII.

Due to storage limitations in space, urine accumulation in a tank for extended periods of time is discouraged. However, storage cannot be avoided because urine is produced discontinuously over the course of a day. Unfortunately, untreated urine is highly unstable. As show in eq. 4, urea hydrolysis by urease-positive bacteria in the urine collection and processing system results in ammonium formation and a pH increase, causing ammonia (NH₃) volatilization. This impacts nitrogen recovery and can have detrimental effects on crew members health. Moreover, salt precipitation due to high pH levels causing scale formation and clogging of the urine processing system (Coppens et al., 2016; De Paepe et al., 2020a,b). The toxic compounds (H₃PO₄ and Cr⁶⁺) that are used currently in the UPA system in ISS to stabilize stored urine can be replaced by a safer, electrochemical method (De Paepe et al., 2020b). Recirculation over a cathode in a three-compartment electrochemical cell is a potential alternative. Here, hydroxyl (OH⁻) ions are created from H₂O and the pH is increased to 10–11 (due to hydroxyl-ion production), inhibiting enzymatic urease activity in a non-toxic manner (De Paepe et al., 2020b). Meanwhile, calcium and magnesium concentrations are reduced due to precipitation caused by supersaturation. This mineral removal process in the electrochemical cell also minimizes downstream clogging of the system (De Paepe et al., 2020b).

A key hurdle for nitrification in space applications is the diffusion of oxygen into the nitrifying culture due to reduced gravity conditions on Mars and microgravity conditions in LEO and deep space. Fluid dynamics are restricted to diffusion due to the absence of convection forces in these scenarios, causing limited oxygen distribution (Monti, 2001; Ilgrande et al., 2019a; De Paepe et al., 2020a; Acres et al., 2021). One possible way to circumvent this problem is the use of a membrane-aerated biofilm reactor (MABR) in which aeration takes place via gas-permeable membrane tubes populated by a biofilm of nitrifiers on the outer surface of the membranes. Since microorganisms directly take up the oxygen that diffuses through the membrane, no air bubbles are formed. However, bacteria closest to the gas-permeable membrane will consume most oxygen. This could be an issue if fast-growing ureolytic heterotrophs are present in the system, potentially limiting oxygen availability for nitrifiers on the outer layers of the biofilm. Additionally, anoxic areas may lead to denitrification by the autotrophic nitrifiers or by heterotrophic denitrifiers (using available COD) if an open community is used in the MABR. These denitrification reactions result in the loss of nitrogen as N₂. COD removal with an ureolytic heterotroph in the reactor to minimize denitrification is too low. One option to tackle that issue while also preventing oxygen limitation for autotrophic nitrifiers is the addition of an upstream microbial electrolysis cell (MEC) populated by a ureolytic strain. The bacteria in the MEC remove organics efficiently using an anode as electron acceptor (Figure 5). Here, energy from COD is also partially converted to hydrogen gas. Ultimately, a low COD, nitrate-rich effluent suitable for CIV cyanobacteria and plant growth is obtained after treatment in a MEC-MABR setup (De Paepe et al., 2020a). The combination of physicochemical and biological treatment of urine optimizes nitrogen recovery at minimal energy input and high conversion efficiency while avoiding the use of toxic compounds for urine storage (De Paepe et al., 2020a,b).

Potential application of an electrochemical urine pretreatment module and a MEC-MABR module in the MELiSSA CIII shows promises, but still require modifications for operation in LEO. For example, hydrogen gas formed in the liquid in the MEC in space cannot escape due to the microgravity conditions. A possible solution is the use of a gas diffusion air cathode that prevents H₂ formation in the MEC (De Paepe et al., 2020a).

While the former technologies are promising for improving the overall process of urine nitrification, the use of an anaerobic bioreactor followed by an aerobic bioreactor as will be applied in the Lunar Palace also show promise (Cheng et al., 2019). Thus, in recent years, exciting concepts have been demonstrated for the combination of physicochemical and biological elements for urine processing in space.

Several terrestrial applications based on space research for nitrogen treatment have been developed or are currently in development. For example, Biostyr™ is an innovative process that can be implemented to remove, among others, nitrogenous (NH₄⁺, NO₃⁻) compounds using both nitrification and denitrification in a compact structure, thereby presenting a low environmental footprint. Another technology based on a self-sustaining groundwater filtration system that removes nitrates with denitrification from the ground water to produce potable water has been installed and is currently in used in Morocco (MELiSSA Foundation, 2021). Finally, the Antarctica research facility Concordia has been fitted with a grey water processing system that includes biological components to remove nitrogen (MELiSSA Foundation, 2021).

As mentioned, untreated urine causes scaling and biofouling in downstream processes during urine recycling in CIII. This also holds true for pipelines in sewer and sewage plants. Stabilization of urine by controlled anaerobic ureolysis at the source rather than in a centralized manner can reduce divalent cation salt precipitation (Ca²⁺ and Mg²⁺) and release NH₃ in a controlled manner. The latter is important to avoid nitrogen loss and thus improve downstream nutrient recovery, yielding environmental benefits as a consequence (Christiaens et al., 2019b). In a second potential system, an electrochemical cell drives precipitation of the divalent cation minerals at the toilet and, unlike the aforementioned technology, inhibits urea hydrolysis through alkalization of the urine. Source-separated urine can be stored and, when needed, discharged to a wastewater treatment plant that can recover nitrogen, reducing nitrogen loss (De Paepe et al., 2020b).

Hydrogen-oxidizing Bacteria (HOB) have been under the spotlight in recent years due to their high growth rate, low resource requirements, and a high protein content (> 70% of their cell dry weight). Hence, they have great potential to be used as microbial protein (MP) source (Christiaens et al., 2017; Yang et al., 2021). In recent studies, the growth of HOB on recycled nitrogen has been investigated (Christiaens et al., 2017; Yang et al., 2021). HOBs typically rely on ammonium as their nitrogen source and use H₂ and O₂ as electron donor and an electron acceptor, respectively, to fixate CO₂ as carbon source (Yang et al., 2021). However, conventional ammonia removal and production with the Haber-Bosch process (an artificial industrial process to fixate nitrogen by converting atmospheric N₂ to NH₃ with hydrogen under high pressure and temperature) has a high energy cost. As a potential alternative, a proof-of-concept study validated the use of an electrochemical cell to extract NH₄⁺ from hydrolyzed urine produced in, for example, a MEC. In the system, the urine is alkalified in a cathode compartment, which enables gas stripping of H₂ and NH₃. Then, the effluent is redirected to an anode compartment where NH₄⁺ is pulled to the cathode compartment over an ion-exchange membrane. The O₂ and H₂ needed are generated at the anode and cathode, respectively, whereas the CO₂ used as carbon source is a by-product of ureolysis. The use of this technology greatly reduces energy costs compared to NH₄⁺ production with Haber-Bosch. A study on HOB species C. necator 335 confirmed its ability to grow at high NH₄⁺ loads. Hence, it is possible to grow MP from HOB via NH₄⁺ recovered from high NH₄⁺ load waste streams such as urine (Yang et al., 2021). By supplementing human food or feed for cattle with MP from HOB produced from recovered nitrogen, nitrogen losses, energy costs, and surface area per unit of protein mass produced can be significantly reduced. This offers a strategy to tackle to challenges of future food production for an ever-growing human population (Christiaens et al., 2017; Yang et al., 2021).

Depending on the location in spaceflight, microgravity decreases from 10⁻³ to 10⁻⁶ times the terrestrial gravity (Huang et al., 2018). These conditions indirectly impact bacterial life through the alteration of fluid mechanics. Buoyancy due to density differences, convection and hydrostatic forces are eliminated, causing a low-shear fluid environment for bacteria, in which nutrients and metabolites only spread by diffusion along a concentration gradient. This results in the formation of nutrient-depleted and metabolite-enriched zones around bacteria (Monti, 2001; Huang et al., 2018; Senatore et al., 2018; Acres et al., 2021). Microbial behavior in real microgravity conditions has been analyzed on the orbital stations Mir and ISS. Most experiments, however, have been conducted on Earth, using different variations of devices called clinostats, which simulate microgravity. The rotating wall vessel (RWV) is a type of 2D clinostat that rotates perpendicular to the gravity vector, causing low-shear fluid conditions and preventing the microbe from adapting to a specific gravitational orientation. These factors result in a low shear-modelled microgravity (LSMMG) environment (Orsini et al., 2017; Huang et al., 2018; Senatore et al., 2018; Acres et al., 2021). Alternatively, the random positioning machine (RPM), a type of 3D clinostat, changes the position of an experiment randomly at arbitrary speeds and directions in the 3D space. In doing so, the object in suspension experiences a net gravity vector close to zero over a certain time interval. For microorganisms, the RWV is the most commonly used device for experiments in simulated microgravity, while the RPM is generally applied for experiments with higher organisms, but several applications for bacterial experiments are also known (Mastroleo et al., 2009, 2013; Huang et al., 2018; Senatore et al., 2018, 2020; Acres et al., 2021). The effects of an LSMMG environment have been assessed for a wide variety of microbial species and have been observed to impact cell growth, cell morphology, cell metabolism, cell–cell communication, cell pigmentation, biofilm formation, stress response, gene expression, virulence, genetic transfer, and even cause genomic changes (Mastroleo et al., 2009, 2013; Orsini et al., 2017; Tirumalai et al., 2017; Senatore et al., 2020; Acres et al., 2021).

Very limited work has been performed on ureolytic and nitrifying bacteria or microbial consortia in simulated microgravity. Ilgrande and co-workers exposed a synthetic community of C. testosteroni, N. europaea, and N. winogradskyi to LSMMG in a 2D clinorotation experiment, showing similar ureolytic and nitritation activity in regular cultures compared to those subjected to microgravity (Ilgrande et al., 2019a). However, nitrite accumulated in the LSMMG-exposed samples, indicating inactive nitratation by N. winogradskyi. This observation has been attributed to the experiment setup, which was performed in a fluid processing apparatus used in former spaceflight experiments (Ilgrande et al., 2019a). It is hypothesized that this device could limit oxygen availability. The resulting competition for oxygen probably inhibited N. winogradskyi in the nitrifying community, likely causing nitrite accumulation in this scenario (Laanbroek and Gerards, 1993; Ilgrande et al., 2019a). Further work should ensure that the setup allows adequate oxygen availability for all members of the synthetic community. Moreover, an omics approach will aid in unraveling the underlying molecular mechanisms and constructing a thorough understanding of the behavior of nitrifying consortium in simulated microgravity conditions.

A second major difference in the space environment compared to the terrestrial environment is the chronic low-dose ionizing radiation exposure. In LEO, radiation dose rates (400–600 μGy d⁻¹) can be 150–200 times higher than those experienced on Earth (2–4 μGy d⁻¹; Vanhavere et al., 2008; Mastroleo et al., 2009; Dachev et al., 2015). The average ionizing radiation exposure in lunar orbit was measured at 200–300 μGy d⁻¹ and recent lunar surface measurements determined an average absorbed dose rate of 316.8 μGy d⁻¹ (Reitz et al., 2012; Zhang et al., 2020). Finally, radiation doses in Mars transit are estimated to average 460 μGy d⁻¹ while the Mars surface dose rate is approximated at an average of 210 μGy d⁻¹ (Durante, 2014). Biological effects of ionizing radiation result from both direct and indirect damage to biomolecules. First, molecular bonds in biomolecules can be broken directly by radiation, which interacts with molecules by excitation and ionization. High energy, high density (HZE) particles are the main contributors to these effects. However, ionizing radiation mainly causes indirect damage by interacting with H₂O molecules, creating reactive oxygen species that attack biomolecules and cause oxidative stress (Dartnell et al., 2008; Moeller et al., 2010; Hassler et al., 2014; Siasou et al., 2017).

There have been many studies on the effects of ionizing radiation on bacterial species. Cell survival to acute high-dose exposure is suggested to not rely on just a single mechanism, such as the ability to repair DNA, but rather a complex set of cellular responses. A review on studies of the effects on bacterial populations exposed to long-term chronic low doses in the Chernobyl restricted zone describes that radioresistance seems to improve (Mollert and Mousseau, 2016). This implies adaption of bacteria to long-term chronic low-dose radiation exposure. However, more rigorous experimentation is required to solidify this claim (Mollert and Mousseau, 2016). In experiments with E. coli, selective pressure caused by high-dose acute ionizing radiation (1,000Gy) led to highly radioresistant phenotypes (Byrne et al., 2014). The findings in the discussed experiments do imply the presence of selective pressure when bacteria are exposed to either chronic low-dose irradiation or acute high-dose irradiation.

Effects of space relevant doses of ionizing radiation mimicking LEO conditions have also been studied. It is impossible to precisely imitate the complex radiation environment on board the ISS due to the wide array of radiation types stemming from secondary particles generated from interactions of primary particles with the spacecraft’s hull. HZE particles are the main protagonists in radiobiological damage in space. Hence, irradiation with these types of particles in terrestrial simulation can give a close approximation of the biological effects of cosmic radiation (Moeller et al., 2010). Moreover, to more closely mimic the radiation environment in space, one could irradiate biological samples with both low-and high-energy particles. An experiment on R. rubrum (a MELiSSA CII strain) comparing the impact of simulated LEO ionizing radiation and a spaceflight experiment in ISS, showed a pronounced transcriptomic response to the radiation exposure. R. rubrum is thus able to respond and cope with conditions linked to spaceflight in LEO (Mastroleo et al., 2009).

To our knowledge, no research has been done on the effects of ionizing radiation on ureolytic and nitrifying strains or consortia. Only one study indicates that members of the ammonium-oxidizing Nitrosomonadacaea family, of which N. europaea is a member, are more sensitive to γ-irradiation than other soil microorganisms, and post-exposure recovery proceeds slowly (Shah et al., 2013). To provide an idea of the performance of the axenic ureolytic heterotrophs and nitrifying autotrophs in space, it is necessary to evaluate the biological effects of chronic low-dose ionizing radiation exposure on these strains. Moreover, valuable information can be gathered on radioresistance of these bacteria with an acute high-dose irradiation survival experiment.

Several studies have already provided researchers with insight into the effects of the LEO space environment on bacteria. Since it is hard to mimic the exact conditions of real space ionizing radiation and microgravity conditions and the combination of both, space experiments provide very valuable information of bacterial response to real space conditions. Such effects include variations in cellular metabolism, microbial proliferation rate, cell division, biofilm formation, cell morphology, cell motility, and genetic transfer between cells. Limited information is available on molecular responses of bacteria in space (Senatore et al., 2018; Milojevic and Weckwerth, 2020).

Nitrification in space has already been demonstrated in biofilters of the CEBAS experiment during the SPS-89 mission of the Columbia space shuttle (Blüm et al., 2003). A second biofilter experiment called C.R.O.P. failed due to hardware malfunctioning during spaceflight (Hauslage et al., 2018). Furthermore, two spaceflight experiments (NITRIMEL and BISTRO) have been conducted to specifically study the effects of space conditions on nitrifying bacteria and consortia. Both focused on the storage and survival of inactivated cultures of nitrifiers, rather than assessing the effects of space conditions on active cultures and the active nitrogen conversion process (Lindeboom et al., 2018; Ilgrande et al., 2019b). The impact of space conditions on reactivation ability of the nitrogen cycle cultures is important in case of system failure during spaceflight (Lindeboom et al., 2018). During the NITRIMEL experiment, reactivation of pure cultures of ureolytic heterotroph Cupriavidus pinatubonensis, N. europaea, and N. winogradskyi, a bipartite culture of N. europaea and N. winogradskyi, a synthetic consortium of all 3 strains, and a nitrifying bioreactor culture were assessed after a 44-day mission in LEO on board the FOTON-M4 research satellite. Samples were exposed to 547–827 μGy d⁻¹ and microgravity (10^{–3 _} 10⁻⁴g) conditions. No impact on ureolysis, nitritation or nitratation rates was observed compared to control cultures stored on the ground (Lindeboom et al., 2018). These results were substantiated by the BISTRO fight experiment where C. pinatubonensis, autotrophic nitrifying Nitrosomonas ureae, N. europaea, N. winogradskyi, the bipartite community of N. europaea and N. winogradskyi, and a synthetic consortium of the aforementioned strains with or without N. ureae added to the community, were sent to the ISS for 7days. Here, the radiation exposure was 400 μGy d⁻¹. A general biomass decay was observed for all ISS cultures as opposed to control samples, but reactivation of nitrifying processes was successful in all populations (Ilgrande et al., 2019b). The observations in both studies conclude that ureolytic and nitrifying strains can cope with exposure to space conditions for a duration of missions that can take us to at least the Moon and encourage further experimentation toward their application in a BLSS for space travel (Lindeboom et al., 2018; Ilgrande et al., 2019b).

In the current URINIS (Urine Nitrification in Space) project, the impact of spaceflight conditions on actively growing nitrification cultures in space will be assessed for the first time. A batch experiment followed by a bioreactor experiment will be flown to ISS to assess the in situ effect of space conditions on the nitrifying activity of pure strains and the synthetic consortium of C. testosteroni, N. europaea, and N. winogradskyi. Activity tests, viability tests, transcriptome, metabolome, and proteome analysis, phenotypical characterization and biofilm analyses will aid in understanding microbial responses to that environment and provide more answers for the feasibility of CIII in space travel.