key: cord-0779413-887fr4xv
authors: Evans, Michael E.; Graham, Lee D.
title: A Flexible Lunar Architecture for Exploration (FLARE) supporting NASA's Artemis Program
date: 2020-07-28
journal: Acta Astronaut
DOI: 10.1016/j.actaastro.2020.07.032
sha: f4ff69647c494f05db4cce8c9fb12e51161e6d7b
doc_id: 779413
cord_uid: 887fr4xv

The Flexible Lunar Architecture for Exploration (FLARE) is a concept to deliver four crew to the lunar surface for a minimum of seven days and then return them safely to Earth. FLARE can be implemented whenever the component vehicles are operational. FLARE was developed as an alternative to NASA's Human Landing System (HLS) reference architecture from the Design Analysis Cycle (DAC) #2 created in 2019. The DAC2 guidelines required utilization of the Gateway vehicle in a Near- Rectilinear Halo Orbit (NRHO). Instead, FLARE chooses a Low Lunar Frozen Polar Orbit (LLFPO) for lunar rendezvous of components, and an optional Gateway vehicle. The LLFPO provides a stable orbit that overflies the south pole every 2 h, ensuring easy access to the lunar surface for surface aborts with a much lower propellant requirement than NRHO. The minimum FLARE concept uses one Space Launch System (SLS) launch, one Orion, one European Service Module (ESM), and one human lander (launched on commercial vehicle(s)). FLARE adds the SpaceTug, based upon the mature and successful ULA “Common” Centaur Upper Stage vehicle, with modifications to create an Earth-Moon transfer vehicle. In the FLARE baseline mission, the SpaceTug provides propulsion needed to return the Orion + ESM from LLFPO to Earth. The SpaceTug also provides propulsion to deliver the separate human lander components – the Descent Element (DE) and the Ascent Element (AE) - from Low Earth Orbit (LEO) to LLFPO. The SLS Block 1 then launches the Orion + ESM and completes a rendezvous with the mated DE + AE components in LLFPO. FLARE offers optional phases beyond the baseline mission. The SpaceTug can deliver components of the planned Gateway, including the Power and Propulsion Element (PPE) and the Habitable and Logistics Outpost (HALO), to LLFPO. FLARE provides an option to deliver precursor equipment to the lunar surface to enhance and extend the human mission. With these components, including an inflatable habitation module and airlock, individual crew mobility vehicle(s), an In-Situ Resource Utilization (ISRU) demonstration, and science and technology experiments, the crew can explore and conduct science on the lunar surface for up to 14 days.

The Flexible Lunar Architecture for Exploration (FLARE) is a practical methodology to deliver four crew to the lunar surface for 7-14 days then safely return them to Earth. FLARE can be implemented whenever the component vehicles are ready for launch. FLARE was developed as an alternative to NASA's Human Landing System (HLS) reference architecture from the Design Analysis Cycle (DAC) #2 [1]. Their goal was to develop an architecture delivering two crew to the lunar surface by 2024 using only a commercial human lander, the SLS Block 1, the Orion, the European Service Module (ESM), and Gateway in a Near Rectilinear Halo Orbit (NRHO). The DAC2 team selected NRHO for Orion due to documented propulsion limitations of the ESM [2] . The Orion+ESM alone has insufficient propellant to deliver the crew to any circular Low Lunar Orbit (LLO) and safely return them to Earth. Although outside of their documented constraints, DAC2 discussed the requirement for a "Transfer Element" (TE) as an additional vehicle to move assets from NRHO to LLO and back. The HLS team was unsuccessful at closing an architecture within their constraints due to the large mass needed (both propellant and human logistics supplies) for transfer between the lunar surface and NRHO [1] . The FLARE expands upon the TE concept by modifying a mature, upper-stage vehicle into a "SpaceTug" to transfer mass between Low Earth Orbit (LEO) and LLO. FLARE is supported with a technical analysis of multiple factors, including mass and change in velocity (∆V) calculations including crew, cargo, and propulsion systems. FLARE develops a plan for launch and mating of necessary components in Earth and lunar orbit. FLARE provides a reference design for the SpaceTug and a human lander, including both the pressurized Ascent Element (AE) and a "common" Descent Element (DE) capable of delivering either crew or cargo to the lunar surface. Payload volumetric evaluations are considered within existing launch vehicle fairings, and also for crew logistics on the lunar surface (within both the lander and in an optional inflatable habitation module). A lunar surface concept of operations is presented for Extra-Vehicular Activity (EVA) traverses and crewed exploration activities with a 7-14 day campaign. FLARE also provides an optional reference concept for an individual crew mobility device, called the "Lunar ATV" (LATV), to support extended surface traverses for early human surface missions. Similar to the SpaceTug function for orbital transfers, the LATV provides lunar surface transfers for science and crew.

The FLARE utilizes launches on various mature, Commercial Launch Vehicles (CLV) to lift uncrewed components to LEO. The FLARE launch schedule requires a nine-week period that integrates ULA, SpaceX, and NASA launch pad availability with predicted boil-off rates for vehicle cryogenic propellants. The key to FLARE is the SpaceTug (based on an existing upper-stage vehicle) that is launched on a CLV. FLARE utilizes a Low Lunar Frozen Polar Orbit (LLFPO) with inclination of 86.5° at an altitude of 100 km over the lunar surface. SpaceTugs transfer assets, such as the human lander, between Low Earth Orbit (LEO) and LLFPO. A single SLS Block 1 lifts crew in the Orion+ESM to Trans-Lunar Injection (TLI), and the ESM provides the propulsion to insert the Orion+ESM in LLFPO. A dedicated "Return SpaceTug" (RST), delivered to LLFPO before the crew launches, provides the necessary propulsion for crew return to Earth. [4] ; however, these elements are available as optional phases in FLARE. FLARE also provides another optional phase for a precursor cargo mission, launched on a SpaceX Falcon Heavy (FH). The precursor mission lands directly on the lunar surface with an inflatable habitation module, LATV(s), a science trailer, and technology demonstrations and experiments.

3.0 3.0 3.0 3.0 FLARE FLARE FLARE FLARE P P P PHASES HASES HASES HASES FLARE sequence naming convention is XXY, where the XX identifies the vehicle element (see Appendix A Acronyms), and Y is an incrementing counter for those vehicle elements. For SpaceTugs, the naming convention is repeated, e.g. Tug1DE1 (which identifies the first SpaceTug that pushes the first Descent Element from LEO to LLFPO). Generic vehicle discussion uses their acronym only without a counter (Y). When discussing the launch vehicle for each element (see Table 2 and Table 3 ), the designated CLV is listed first, e.g. A5_AE1 (identifies the ULA Atlas V that lifts the AE1 to LEO).

The FLARE sequence is divided into five Phases A-E (with each subdivided into subphases a-b): A. Deliver equipment to create the Gateway in LLFPO (Optional) B. Deliver lunar surface precursor equipment (Optional) C. Deliver vehicles to LLFPO for crewed mission support (Required) D. Deliver crew to LLFPO, then lunar surface, then to LLFPO (Required) E. Return crew to Earth (Required)

A summary of the FLARE mass budget and number of vehicles is provided in Table 1 The minimum sequence of steps for the FLARE crewed mission (to support four crew on the lunar surface for 7 days living in the AE) is shown in Table 2 , with a graphical description of required phases provided in Figure 1 The sequence of steps for Optional FLARE Phases is shown in Table 3 , with a graphical description of optional phases provided in Figure 6 thru Figure 9 . These maneuver burn and margin calculations are provided in Appendix B Tables B.1 Optional Phase A assembles Gateway in LLFPO. NASA has existing contracts for commercial launch of the PPE and HALO, former known as "Mini-Hab" (MH), elements in NRHO [3, 5] . NASA then decided on a single launch of these two integrated elements in 2023 [6] . With FLARE moving the Gateway elements from NRHO to LLFPO, vehicle thermal and power systems may need to be modified (based on the assumed stack attitude timeline in the lower lunar orbit). The PPE could provide a valuable communications relay for lunar surface operations, and HALO could provide unpressurized docking adaptors for SpaceTugs. Without Phase A, Orion+ESM must dock directly to the AE1 of the human lander (composed of AE1+DE2). This is exactly the Apollo program approach, and requires compatible docking adaptors on each vehicle (which are different from the requirements for each vehicle docking to Gateway). Within FLARE, each Gateway element is launched separately to LEO on an Atlas V (A5). Each SpaceTug is launched to LEO on a SpaceX Falcon (F9). Thus, four additional CLV launches are required to construct Gateway in LLFPO. Each SpaceTug conducts an autonomous Rendezvous, Proximity Operations, and Docking (RPOD) with the Gateway element in LEO and then transfers it to LLFPO.

Optional Phase B provides precursor hardware to the lunar surface, including an inflatable surface habitat and inflatable airlock (to support four crew living on the lunar surface up to 14 days living in the habitat). The precursor hardware also includes human logistics and EVA support, individual human mobility vehicle(s), science and technology experiments and deployed communication/navigation satellite(s). Phase B is necessary for developing early lunar surface sustainability as a human "field station". Phase B requires one Falcon Heavy (FH) launch to deliver a "common" Descent Element (named DE1) and 4.5 mt cargo payload of precursor hardware to TLI. The DE1 then deploys communications & navigation satellite(s), performs the lunar descent, and delivers the payload directly to the lunar surface. Despite the coronavirus global pandemic in 2020 closing NASA centers [7] , NASA continues to press towards a human lunar mission in 2024, even developing a framework of principles for lunar development with the "Artemis Accords" [8] . There is concern, however, that the Congressionally required budget to support this launch date will not be provided to NASA [9] . FLARE supports the Artemis Program with a launch sequence applicable whenever the vehicles are operational. The minimum crewed mission (FLARE Phases C, D, E) to the lunar surface requires the launch of five Falcon 9 (F9) for SpaceTugs, two Atlas V (A5) for human lander components (AE and DE), and one SLS for crew inside Orion+ESM. All these launches occur within a 9-week period (see Figure 10 ). The launch timing of these phases is critical since the SpaceTug and lander DE vehicles use LOX/LH 2 as propellant that boils away in orbit. FLARE calculates the required timing and sequence to provide necessary margin in ∆V calculations, including the loss of LOX/LH 2 from "boil-off" in orbit (see Appendix B Tables B.4-B.10). The SLS is the final launch in this 9-week period, and its launch window must include considerations for lunar polar plane rotation in LLFPO to minimize ∆V for TEI return, and also lunar surface lighting during the crew lunar surface campaign. In the minimum mission, all four crew spend seven days on the lunar surface living in the AE1 (see Figure 5 ). Adding optional Phase B, all four crew would live for 14 days in the inflatable habitat provided by the precursor cargo mission (see Figure 9 ). Phase B requires one additional SpaceX Falcon Heavy (FH) launch to carry a "common" DE and payload to TLI. The DE1 vehicle then lands the 4.5 mt payload on the lunar surface. Note that all optional phases (A, B) occur before the 9-week sequence leading to the launch of the crew (see Figure 10 ).

The SLS launches Orion+ESM to TLI, and the ESM then provides the required 1.0 km/s ∆V for LLO insertion and RPOD to the AE1 waiting in LLFPO (or to Gateway if Option A is implemented). This consumes most of the ESM propellant. The FLARE provides a SpaceTug in LLFPO, named the Return SpaceTug (RST), with sufficient propellant to return the Orion+ESM and crew back to Earth (including boil-off margin). The RST is launched 1.5 months before SLS. After delivery to LLFPO, the RST experiences 56 days of propellant boil-off yet retains sufficient ∆V to return the crew to Earth (including a maximum 14-day surface duration with Optional Phase B). All of the necessary elements for lunar descent, ascent, and return are therefore in place before the crew launches in Orion.

Recognizing the risk of reliance on a SpaceTug in LLFPO for crew return to Earth, FLARE considered a possible alternative of this crewed sequence using a SpaceTug to push the Orion+ESM to LLFPO. This allows the Orion+ESM in LLFPO to provide the ∆V for crew return to Earth (no RST required). Preliminary analysis suggests this approach is possible but requires assembly of a large, multistage vehicle in LEO composed of three SpaceTugs mated to the Orion+ESM to conduct TLI and LOI (see Appendix B Tables B.12 The HLS DAC2 evaluation guidelines required Gateway in a NRHO [1]. The NRHO is highly elliptical lunar orbit with a period of 6 to 8 days. It has been well studied as a potential staging orbit for deep space exploration using the Earth-Moon-Sun L1 and L2 Lagrange Points [10] . Altitudes above the lunar surface can vary from 2,000 to 75,000 km during each NRHO period [2] . A specific NRHO with a 9:2 lunar synodic resonant, chosen for the NASA HLS DAC2, places apolune over the lunar south pole. This orbit is favored for its low orbital maintenance maneuver requirement and infrequent eclipse by the Earth and Moon [11] .

In a representative 6.5-day period, southern NRHO, a spacecraft spends the bulk of every week at the far end of the orbit (relative to the Moon) with only 1 to 2 days near the lunar surface. For a brief time the orbiting vehicle is difficult to reach from the lunar surface (passing at high velocity nearly 2000 km above the surface) and for the majority of the week the orbiting vehicle is impossible to access (>30,000 km away). NRHO thus forces lunar human mission designers into one of two difficult choices. First, a short-duration "grab and go" mission to descend to the lunar surface from Orion+ESM in NRHO, consisting of a very brief surface exploration campaign (< 12 hours), and then ascend to the Orion+ESM. This short mission provides little surface science opportunity or surface infrastructure development, and requires an extremely long crew wake period (>24 hours to prepare in orbit for descent, descend, explore, ascend, and dock with Orion). The second option is a weeklong mission on the lunar surface; however, the crew has no ability to rapidly abort to Orion+ESM once they are on the surface for more than 4 hours (since Orion+ESM are too far away and moving rapidly further from the Moon). A few interim abort-to-orbit opportunities exist, but the crew must survive in the ascent vehicle for 2 to 4 days while conducting a rendezvous with the distant Orion+ESM. For the nominal mission, the Orion+ESM overflies the landing site 6.5 days after landing, and the crew can ascend and rendezvous in NRHO. Adding a surface habitation module and pre-positioned infrastructure components (e.g. crew mobility devices, EVA tools, and contingency consumables) increases safety and reduces risk for the longer surface missions while Orion+ESM is too distant for rendezvous.

Science opportunities from the NRHO include Earth observations (outside the Earth's magnetosphere), heliophysics, fundamental physics, and microgravity or radiation studies of biological and physical systems [12, 13] . The first instruments selected for Gateway observe space weather and monitor the Sun's radiation environment [14] . With an orbiting spacecraft in NRHO, however, the vehicle is so distant from the lunar surface that telescope observations from Earth exceed resolutions possible from likely equipment available viewing from a window, or externally attached to HALO. Note the current Gateway HALO concept provides little volume for internal or active, external science instrumentation and experiments. HALO has no science airlock or robotic arm to mount and remove experiments for return to Earth, although perhaps the addition of the ESA ESPRIT module would add these features after the initial Gateway vehicle arrives in orbit [12] . NASA has announced that Gateway is no longer a required component of the Artemis architecture [15] , so early science may be limited to capabilities in Orion or the human lander only. The SLS B1 vehicle with the ICPS can deliver the 26 mt Orion+ESM to TLI [16] , but the 8.6 mt of propellant and oxidizer in the ESM [17] delivers a maximum total ∆V of only 1.25 km/s [2] . This ESM therefore provides insufficient delta-V (∆V) to both insert the Orion+ESM into a 100 km circular Low Lunar Orbit (LLO), requiring a ∆V = 0.952 km/s, and return it to Earth from LLO, requiring a ∆V = 1.256 km/s [18] . It might be possible for Orion+ESM to return from LLO to Earth alone (with little excess margin); however the Orion+ESM must then be delivered to LLO by another vehicle. A summary of the relevant ∆V requirements is provided in Table 4 .

To Table 4 :

The NRHO is an elegant mathematical solution to the ∆V limitations of Orion+ESM and the SLS B1. The NRHO significantly reduces the total ∆V cost for a near-lunar orbiting spacecraft, requiring a total Orion+ESM ∆V=0.850 km/s for insertion and exit in a 21-day mission [2] . This reduction in ∆V for access to NRHO from TLI, however, forces any lunar lander to increase their ∆V lunar ascent propellant budget by 0.85 km/s to achieve the higher orbit from lunar surface [1]. This increased lunar ascent propellant mass ripples through every possible architecture with impacts on lunar descent propellant, lander dry mass, and ultimately launch mass for the components.

The FLARE human lunar lander reference design AE upsizes the two-crew Apollo Lunar Module (LM) ascent stage from 2.4 mt dry mass [19] to 4.0 mt to accommodate four crew for seven days. For this 4.0 mt AE, the LLO orbit requires only 60% of the ascent and descent propellant compared to NRHO. The NRHO increased propellant requirement drives larger AE and DE vehicles (see Appendix B Table B .11 for a comparison study of lander mass for NRHO and LLO where the SpaceTug, not the DE, performs the lunar orbit insertion maneuver). The NRHO integrated human lander (AE and DE) for this 4.0 mt AE would have a launch mass of 29.6 mt. One SLS Block 1, capable of lifting 26 mt to TLI [20] , could not lift this integrated NRHO lunar lander. To reduce the lander weight on NASA's currently planned Artemis-3 mission, the HLS plans to only send two of the four astronauts from Orion to the lunar surface and back to NRHO. Two other crew will remain aboard Gateway in NRHO [21] . With FLARE, however, an integrated LLO human lander has a launch mass of 20.4 mt (including 0.5 mt of AE payload). This lander could be launched on a dedicated SLS Block 1 to TLI; however, FLARE assumes only one SLS launch/year dedicated to the Orion+ESM (limited by NASA budget and SLS production rate). The FLARE LLO human lander components (AE1 and DE2) are thus launched separately on CLVs (see Figure 1 and Figure 3 ). SpaceTugs then transfer the lander components to lunar orbit and provide the RPOD for mating. Future CLVs may lift an integrated human lander, or perhaps the Orion+ESM, directly to TLI -but the lander or ESM must then perform the required lunar orbit insertion burn and RPOD.

During the spring of 2020, HLS continued to study possible orbits that Orion+ESM can achieve using the SLS Block 1 launch vehicle. The annual NASA management presentations to the NASA Advisory Council (NAC) provide limited details on this orbit. The ECPO is an elliptical orbit that varies from 4500km to 6500 km at apolune to a 100 km perilune, with an approximate 9 hour orbital period. The orbit requires more maneuvers to enter and exit lunar orbit than a NRHO, and it greatly delays the Earth return window for a missed TEI maneuver [22] . This orbit likely requires less propellant mass for transfers to/from the lunar surface than NRHO, but extensive lander design analysis of this orbit is not available yet. Lessons learned from Apollo teach that vehicles in LLOs do not all have the same propellant budget to maintain their orbit. For example, the Apollo 16 mission released an orbital scientific satellite to study charged particles and magnetics fields around the Moon. The vehicle crashed into the moon after only 35 days due to unknown subsurface gravity mass concentrations ("mascons") that altered the satellite orbit with each revolution, thus causing it to deorbit much sooner than planned. Subsequent lunar missions have mapped the locations of these mascons and identified their gravity impact on lunar orbits [23] . A few special LLOs are less affected by these mascons. These "frozen orbits" have "constant mean eccentricity, mean inclination and mean argument of perigee" [24] , and provide multi-year stability requiring no corrective maneuvers [2] . Frozen orbits ensure a constant altitude while minimizing the station keeping propellant budget, and are thus favored for orbiting reconnaissance spacecraft [25] . To support a lunar south polar landing site, FLARE selects a specific Low Lunar Frozen Polar Orbit (LLFPO) with I=86.5° and e=0.153 [25] .

With a period of approximately 2 hours, this orbit provides frequent overflight of the near-polar landing site by the orbital vehicle. The entire lunar surface is available for observation and mission access at some point during a lunar month. From this inclination, a small 2.5° plane change during descent provides access to likely landing sites on flat areas near the Persistently Illuminated Regions (PIRs) of the south pole [26, 27] . The ∆V cost for this 2.5° plane change maneuver is calculated, based upon orbital velocity for a 100 km lunar altitude, to be 71 m/s (see Appendix B Table B.19 ). This is additional to the descent propellant requirement for 2.180 km/s [18] . During ascent (when the vehicle velocity is low) the 2.5° plane change is not budgeted with additional propellant to the required 1.968 km/s [18] for ascent to the 100 km altitude. A graphical comparison of NRHO and LLFPO and associated ∆V requirements is shown in Figure 11 . 

The average Earth visibility from possible landing sites at the lunar south pole vary from 30%-70% during a typical month, and no likely site has 100% coverage [26] . This limitation can easily be included as a constraint in launch window development for Moon rendezvous, with the short mission of 7 to 14 days planned to occur when communications with Earth is in direct line-of-site. For a sustained, long-term human presence at the Lunar South Pole, the Earth visibility becomes problematic. To ensure continuous Earth communications, additional equipment needs to be placed either on lunar surface topographic features (e.g. atop a tall nearby mountain such as the rim of Malapert crater) or in lunar orbit. The surface solution could be implemented using a CLPS lander to deploy a communications tower. FLARE, however, chooses to place a satellite in orbit for continuous communications between Earth and the lunar landing site. Co-manifested with various elements of FLARE (possibly with PPE, HALO, DE1, AE, or DE2), the satellite(s) could be deployed after the payload stack achieves sufficient ∆V for TLI. The FLARE SpaceTug is based upon a successful, mature flight-proven upper-stage developed by the United Launch Alliance (ULA). The "Common" Centaur (evolved from the Centaur-III) uses a standard RL-10 engine powered by Liquid Oxygen (LOX) and Liquid Hydrogen (LH 2 ) to deliver payloads to LEO atop an Atlas launch vehicle [28] . The FLARE goal is to create a SpaceTug vehicle that is capable of autonomous RPOD in Earth and Lunar orbit. It can then transfer payloads, reboost platforms, and potentially store and transfer propellants to other vehicles.

The SpaceTug requires the ULA Integrated Vehicle Fluids (IVF) technology, developed for the new ULA Advanced Cryogenic Evolved Stage (ACES) [29] , to re-pressurize the system and provide power to the vehicle [30] . This also removes the need for hydrazine or helium as tank pressurizers. Additional electrical power is provided to the SpaceTug with solar arrays affixed on each side, which cover Multi-Layer-Insulation (MLI) blankets to reduce solar heating into the propellant tanks. A new Tug Adaptor (TA), replacing the Common Centaur Payload Adaptor atop the LH 2 tank, provides the electronics, batteries, and re-pressurization system components for the SpaceTug. The TA also houses the retractable docking struts and umbilical connections for mating the SpaceTug to other vehicles (including other SpaceTugs). SpaceTug configurations are demonstrated in Figure 12 .

The dry mass for the SpaceTug is 2.75 mt (adding 0.5 mt for the above modifications to the 2.25 mt "Common" Centaur dry mass) with a propellant load of 20.05 mt (slightly reduced from the Centaur-III to keep the total mass within the expected SpaceX F9 28.5° LEO capability) [28] . The SpaceTug can be stacked together to become a 2-stage vehicle for pushing heavy payloads from LEO to TLI, NRHO, or LLO (see Figure 12 ). A single SpaceTug has the ability to deliver 15.5 mt from LEO to TLI, which is nearly as much as SpaceX Falcon Heavy. A double SpaceTug can deliver 32.9 mt from LEO to TLI, which is significantly more than a SLS B1 with ICPS. The double SpaceTug can deliver 21.2 mt from LEO to LLO (see Appendix B Tables B.12-B.17 for SpaceTug transfer calculations).

Each SpaceTug is launched on a SpaceX Falcon 9 (F9) CLV. The F9 Block 5 delivers 22.8 mt to a 28.5° LEO orbit [31] for an unspecified altitude. Prior SpaceX documentation reveals approximately a 5% reduction in payload delivery with a F9 Block 2 between a circular 200 km orbit (delivery = 10.454 mt) and a circular 400 km (delivery = 9.953 mt) orbit [32] . The F9 Block 5 is thus expected to incur the same Performance Fraction (PF) for similar LEO altitudes. It is unknown how much reserve SpaceX maintains for the F9 Block 5 capacity. FLARE thus assumes the F9 Block 5 can deliver 22.8 mt to a 400km circular orbit of 28.5° inclination. The 2019 Falcon 9 fairing is 5.2m in outer diameter and 13.2m high overall and it can accommodate payloads of 4.6m diameter and 11m tall (barrel volume). A longer fairing is needed for the SpaceX Falcon 9 (F9) to hold the 12.7m tall SpaceTug derived from the ULA Common Centaur. There are longer fairings available, developed by RUAG for the ULA Atlas V, that have previously been discussed to support Department of Defense (DoD) payloads [33] . The SpaceX launch pad also needs modification to allow cryogenic refill of the SpaceTug immediately prior to launch.

The boil-off of cryogenic LOX/LH 2 must be minimized in vehicles on-orbit. This cryogenic propellant is chosen for both the SpaceTug and the human lander reference design. The original Titan/Centaur was designed to support an eight-hour mission with a boil-off of 2%/day [34] . ULA has developed and patented numerous concepts to store propellant on-orbit [35, 36] . A new design, using the Centaur upper-stage as a secondary tank, called the CRYogenic Orbital Test (CRYOTE) concept, conceives of up to 1 year of storage of cryogenic propellants on-orbit [37, 38] . Building on CRYOTE tests, the next ULA concept uses a "Drop Tank" which waits in LEO for "days, weeks, or even months" to refill a Centaur upper-stage launched on a subsequent mission [34] . The Drop Tank remains attached to its depleted upper-stage and spins slowly (1°/sec) to provide centrifugal acceleration that settles the cryogenic fluids. The Drop Tank design includes features to minimize boil-off with insulated blankets, lightweight materials, a vacuum insulated common bulkhead and low conductivity struts. This expected boil-off is under 0.1%/day of the total propellant load. Studies with "Zero-Boil-Off" systems show that spacecraft with proper insulation, and/or small cryo-cooling systems, could provide for months of liquid hydrogen storage without evaporation [39, 40] . FLARE assumes the LOX/LH 2 boil-off rate of 0.5%/day (based on a full cryogenic propellant tank) for both the SpaceTug and the reference Descent Element (DE) of the human lander.

Although FLARE does not require on-orbit fluid transfer of LOX/LH 2 , it provides an opportunity to demonstrate this capability in LLFPO (see Sequence #18 in Table 2 ). NASA has investigated cryogenic fuel transfer and fuel depot concepts using the Centaur, with a goal of reducing boil-off to 0.1%/day [41] . A SpaceTug in LLFPO could transfer residual propellant to the human lander DE or another SpaceTug. NASA considered the transfer of LOX between vehicles in Earth Orbit Rendezvous (EOR) as one of four architectures for lunar exploration. The Apollo plan required transfer of oxidizer from a tanker S-IVB upper-stage to a Trans-Lunar Injection Stage containing the CM, SM, Lunar Touchdown Module and Lunar Braking Module in order to achieve TLI [42] . A challenge of cryogenic fluid transfer on-orbit is evolved gas release during the fill process. "No-vent-fill" designs, tested by NASA in the 1990s [43, 44] , use cold liquid thermodynamic properties to condense the vapor in the tank [34] . In 2019, a company built an experiment "Furphy" on the ISS that successfully demonstrated the transfer of water on-orbit [45] . NASA and Yetispace have conducted tests on Earth demonstrating successful liquid nitrogen transfer under flight-like conditions [46] . The Shuttle program demonstrated an astronautcontrolled remote transfer of hydrazine between two tanks mounted in the payload bay of STS-41G [47] . Originally planned as a water transfer, Astronaut Dave Leestma convinced NASA management to allow the transfer of toxic hydrazine in order to better simulate refueling of spacecraft on-orbit [48] . Similarly, the Apollo 14 crew demonstrated liquid transfer of an inert fluorochemical, perfluorotributylamine, from one container to another using a hand pump operated by an astronaut [49] . The propellant transfer technology gap must be closed to support long-term harvesting of planetary resources to fuel space vehicles. SpaceTug on-orbit transfer of LOX/LH 2 enables future Mars exploration and lunar commercialization.

The FLARE minimum required crew phases requires five SpaceTugs, of which three (Tug1AE1, Tug1RST, Tug2DE2) ultimately crash on the lunar surface. Adding an auxiliary Electric Propulsion (EP) system to the SpaceTugs would allow the depleted LOX/LH 2 vehicles to remain on-orbit and be repurposed as communications, navigation, or science experiment satellites. NASA has employed EP on three science missions: Deep Space 1, Dawn, and Space Technology 7 [50] . Current NASA Solar Electric Propulsion (SEP) plans employ a 100 kw system with a 13.3 kw Hall thruster system for the PPE [51] . Development of a low power SEP, perhaps using the SpaceTug solar arrays (anticipated power in the 100s watt range), has previously been studied [52] .

The SpaceTug could provide both the storage system and the transportation engine to deliver In-Situ Resource Utilization (ISRU) harvested LOX/LH 2 propellants from the lunar surface to LLO (or beyond). Numerous lunar fuel depot studies have been published detailing the necessary technology and infrastructure requirements for this capability [53] [54] [55] [56] . ULA has developed a concept to transform an ACES upper stage into a horizontal lander. Called XEUS, it provides a novel design to deliver crew and cargo to the lunar surface [30] . For FLARE, a vertical "lander" SpaceTug is estimated to require an additional 1.25 mt (dry mass = 4.0 mt) of hardware modifications for structural components (legs, pads, tanks) and surface refilling equipment. FLARE calculations predict that a modified SpaceTug could achieve LLFPO with ~11.25 mt of propellant remaining from a full (~20 mt) propellant load on the lunar surface (see Appendix B Table B .18). Orbiting SpaceTugs could then refill other orbiting vehicles.

An additional future concept for future SpaceTug design is aerocapture. Rather than have a propellant depleted vehicle de-orbit to Earth after pushing components from LEO towards TLI, necessary decelleration could be performed with an inflatable shield that would slow the SpaceTug again to LEO velocities for RPOD with ISS [55] . The SpaceTug could then be refilled and reused from LEO.

6.0 6.0 6.0 6.0 LUNAR LANDER LUNAR LANDER LUNAR LANDER LUNAR LANDERS S S S FLARE provides a reference 2-stage human lander design concept, consistent in mass and volume with numerous previous lunar lander designs [57] . The components are a "Common" Descent Element (DE) for either cargo or crew delivery to the lunar surface from LLFPO, and a pressurized Ascent Element (AE) for crew transfer to and from the lunar surface. The AE also provides a lunar surface residence for four crew up to seven days. The FLARE concept can accommodate any commercial lander for humans or cargo that falls within the mass, diameter, and height constraints of available CLVs launching elements to LEO. The sequence would follow a similar configuration for the FLARE reference human lander components, such as AE1 (See Figure 1 ) using one SpaceTug for transfer from LEO to LLFPO, or DE2 (See Figure 3 ) using two SpaceTugs for the transfer to LLFPO. In April 2020, NASA announced three commercial contracts for development of a human lunar lander supporting the Artemis Program [58], although these landers are not required to follow HLS's DAC2 requirement for rendezvous with Gateway in a NRHO [15] . NASA had previously selected nine companies in 2018 to provide unmanned landers for lunar exploration with the Commercial Lunar Payload Services (CLPS) program [59] . An additional 12 NASA payloads and experiments were selected in early 2019 [60] , with another 12 selected in summer 2019 [61] . Beginning in 2021 with payloads of at least 10 kg, the CLPS landers are expected to grow to support future payloads of up to 500 kg or larger [59] . The CLPS landers can then provide preliminary science or human infrastructure equipment for human surface missions. Selecting the AE mass is the key design driver for FLARE. The AE supports the crew as they descend from LLFPO, land, live on the lunar surface for 7 days, then return to Orion (or Gateway) in lunar orbit. The AE has a dry mass of 4.0 mt (upscaled from the 2-man Apollo vehicle), and thus requires ascent propellant of 4.4 mt to achieve LLFPO (see Appendix B Table B .9). The AE volume is 8.4 m 3 (or approximately 2 m 3 per person) and has a pentagon-shaped outer mold line (see Figure 13 ). The AE is designed such that the crewmembers stand during descent and ascent. The AE supports crew use of either full xEMU spacesuits or Orion Launch and Entry Suits (LES). The crew has sufficient volume to don/doff their xEMU and/or LES two-at-a-time, and also supports crew sleep periods (by use of hammocks) inside the pressurized volume. The AE is designed to carry 100 kg of lunar surface and crew biological samples (allocated as payload). The AE dry mass includes 20 kg of additional science supporting equipment such as containment boxes within the pressurized volume. The AE also has four Reaction Control System (RCS) thruster quads fed from the same propellant and pressurant tanks as the OME.

The main pressure wall material of the AE habitable volume uses the common 0.040" aluminum 6061-T6 material (slightly thicker than the Apollo LEM 0.012" thick pressure wall). Similar to the DE, it has external MLI interleaved with additional sheets of inexpensive fiberglass fabric for micro-meteoroid protection [62] , and polyethylene sheets for radiation protection [63] The entire AE module itself is structurally attached to the DE-based support structure using (4) pyrotechnic bolts with ZipNuts. These pyrotechnically-modified Snap-On © ZipNuts [64] mate the 2 elements (AE and DE) on-orbit and only require a straight push to fully engage. They therefore do not require any turning of the nut to ensure proper torqueing. These also have the advantage that the greater the tension the better the split nut threads grip the bolt shaft. The bolts themselves have been modified to fragment the bolt head upon the separation firing command. The AE will fire the four pyrotechnic bolts to separate from the DE immediately prior to liftoff from the lunar surface.

The FLARE AE is designed for an atmospheric pressure of 8.2 psi to 14.7 psi. The AE is able to support multiple depress/repress cycles for EVA, including the optional Phase B extended surface mission. With the optional surface habitation module in Phase B, the crew depress the AE to vacuum, power down the AE, traverse to the inflatable habitation module, complete the surface mission, return to the AE, repress the AE, power up the AE, liftoff and ascend to the Orion+ESM. The AE has a single large Ingress/egress hatch for individual crew access in and out of the module while on the lunar surface. This allows a single crewmember to quickly traverse across the sill while only stooping slightly. A second hatch exists for crew entrance/exit on-orbit when docked to either the Orion or Gateway. The AE has two windows, each of which is a triple pane system to protect the pressure pane from inadvertent impact from the crew or impact from micro-meteoroids in lunar orbit or on the surface. After the AE1 vehicle returns to Orion in LLFPO, it could be repurposed as an additional pressurized volume or as an airlock for Gateway (if present). 6.3 "Common" 6.3 "Common" 6.3 "Common"

The FLARE "common" DE concept is shown in Figure 14 , representing the vehicle as a lunar descent stage for either a cargo platform (4.5 mt payload) or for the crewed AE (8.4 mt vehicle).

The DE uses two flight control computers for command and control when an AE is not attached. When an AE is attached, the AE flight computers control the integrated stack. By implementing the optional Phase B, the "common" DE is fully flight tested and demonstrated prior to its required service for landing the crew. The DE lightweight composite truss structure has a dry mass of 2.5 mt, which requires a total mass of 11.5 mt consisting of: a) two LOX and two LH 2 tanks carrying a total of 9.0 mt propellant (7.714 mt LOX, 1.286 mt LH 2 ) b) four helium pressurant tanks at pressure of 2.0684E7 Pascals (3000psi) c) four small self-contained monopropellant RCS thruster pods each containing 40 kg hydrazine d) a single gimbaled RL-10A-4-2 engine (specific impulse of 450.5 s) e) four landing legs that are launched folded up (to fit inside the FH CLV fairing), then deploy and lock into place after the TLI burn is completed f) an overhead composite support structure for a maximum 8.4 mt payload mass (crewed AE) g) a base composite support structure for the Main Propulsion System (MPS) tanks, leg attachment and main engine h) two composite thrust deflection ramps to redirect the AE module ascent plume away from the DE propellant and pressurant tanks during AE ascent.

The required ∆Vs (shown in Table 4 ) for lunar descent and landing from LLFPO are 2.180 km/s ∆V for altitude change [18] , and 0.071 km/s for the plane change (see Appendix B Table B . 19 ). Note that FLARE Sequence #19 (see Table 2 ) assumes the Descent Element (DE2) for crew delivery to the lunar surface provides this entire ∆V (see Appendix B Table B.8 ). An option exists using a SpaceTug for part of this ∆V. Initially, Tug2DE2 pushes the DE2 from TLI to LLFPO and has residual propellant available when the crew arrives to LLFPO. Although not included in these calculations, the Tug2DE2 could conduct a perigee adjust maneuver for the crewed lander (AE1+DE2) to decrease the deorbit propellant needed by DE2 for the lunar landing.

The DE MPS consists of two cylindrical, dome-capped 2.75m long LH 2 fuel tanks (1.286 mt at a density of 70.8 kg/m 3 ) and two cylindrical, dome capped 1.9m long LOX oxidizer tanks (7.714 mt at a density of 1141 kg/m 3 ). This provides a total of 9.0 mt of propellant. Ullage volume was assumed to be 10% for both the LOX and LH 2 tanks. The DE employs a single gimbaled RL-10A-4-2 deep throttle-able engine mounted on the base thrust structure burning an oxidizer-to-fuel ratio of 6:1. As stated previously, FLARE assumes a cryogenic LOX/LH 2 boil-off rate of 0.5%/day (for both SpaceTugs and DE).

The DE has four landing legs which are launched stowed to fit within the 4.6m dynamic envelop of the FH fairing. The legs are deployed immediately following the completion of the TLI burn. Each leg is deployed using small electrical motors and the legs lock in place when fully extended. Internal to each leg is a crushable aluminum honeycomb structure which absorbs a portion of the touchdown loads. One leg has the crew surface access ladder attached to it. The ladder is in segments to "fold" when the landing leg compresses from the touchdown loads. This prevents the buckled ladder from blocking the crew access to the surface. Approximately 2 m above the landing pad is a small open-grated "porch" which also functions as a station for dust removal from the xEMU, as well as a storage location for tools and other equipment.

DE attitude control is maintained by a combination of reaction wheels, RCS thrusters and main engine gimbaling. The DE uses a reaction wheel module mounted on the support structure that torques the DE when slower rate movements are required. When faster rates are required, the four 40 kg RCS thruster pods are also utilized. These are mounted on the four landing legs of the DE and are self-contained monopropellant (hydrazine) systems. The "up" firing thruster on the RCS thruster pod on the leg with the ladder is normally inhibited from firing, but can be commanded to fire through the opengrated porch platform. The opposite leg, opposite thruster is normally used to accomplish this "pitch down" motion. When faster rates are required, a combination of reaction wheel operation, RCS thruster firings, and main engine gimbaling are used to accomplish the movement.

The AE support structure on the DE is a composite frame attached to the base thrust structure and is intended to support the loads induced from a fully loaded AE during lunar orbit maneuvering and landing on the lunar surface. It is not flown during any precursor supply mission. The top of the AE structural support structure on the DE is located 5.0 meters above the lunar surface and consists of an open web platform with the four attach points for the AE. Also attached to this structure are two ascent engine exhaust "chutes". These are placed to redirect the AE OME liftoff exhaust plume such that it doesn't impinge on the DE LOX/LH 2 tanks and potentially cause an explosion. These main LH 2 /LOX propulsion system components are mounted inside this support structure and, with the chutes in place, are able to withstand the brief pressure and temperature spikes seen during AE liftoff.

The DE does not have a "hard" shell around its moldline. Surrounding DE propulsion components are "soft side" sheets of MLI interleaved with additional sheets of inexpensive fiberglass fabric. This provides a viable short-term Micro-Meteoroid and Orbital Debris (MMOD) shield similar to what is used to protect the SpaceX Dragon commercial cargo vehicles [62] .

A FLARE stretch goal is to have the DE capable of on-orbit refueling from a SpaceTug (see Sequence Step #18 in Table 1 ). Mounted on the underside of the base thrust structure are four propellant transfer/structural attach points for the SpaceTug extendable docking struts. These ports provide pressurized fluid transfer as well as structural attach points for the SpaceTug. In addition, there are also two ports on the underside for power/data umbilical attachments. These redundant ports provide communication and power transfer to/from other SpaceTugs when they are attached. The FLARE "common" DE design provides a delivery vehicle for crew or cargo to the lunar surface with mission DE1 (see Table 3 , Sequence #10 and Figure 8 ). With the additional launch of one SpaceX Falcon Heavy to TLI, the DE1 can land 4.5 mt of precursor cargo payload mass directly to the lunar surface. This option could be executed once, or multiple times, to build a sustainable lunar surface infrastructure. The first precursor payload (landed with DE1) includes an inflatable habitation module (2.0 mt), consumables supporting four crew on a 14-day surface mission (0.35 mt), two additional xEMU space suits (0.38mt), individual mobility vehicle(s) (0.92mt), science experiments on a portable trailer (0.30mt), and an In-situ Resource Utilization (ISRU) demonstration (0.40 mt) selected from NASA contractors investigating space resource collection [65] . Addition smallsats (0.15 mt) carried onboard the DE1 are deployed prior to descent to support lunar surface communications and navigation. 

The primary component of the precursor mission is an inflatable habitation module. Based upon the proven design of the ISS Bigelow Expandable Activity Module (BEAM), the 3.6m 3 packed volume expands to 16m 3 when inflated [66] , which exceeds the reference AE volume of 8.4 m 3 . The ISS BEAM dry mass of 1.4 mt is increased to 2.0 mt for FLARE to include surface components for crew sleep stations, toilet, galley, and an inflatable airlock. The assumed internal pressure is 14.7 psi, although detailed design may lower the desired pressure to 8.2 psi to accommodate EVA preparations. Since the inflatable hab and airlock are delivered directly to the lunar surface, they do not require any orbital docking with other vehicles. Thus, the hatch diameter can be increased from the Orion hatch size.

An additional mass of 0.35 mt is also included in the DE1 payload for crew logistics to support a 14-day surface mission for 4 crew (water/air/food), which is 6.25 kg/person/day. Similar to the AE, the inflatable habitation module has external MLI interleaved with additional sheets of inexpensive fiberglass fabric for micro-meteoroid protection, and polyethylene sheets for radiation protection [63] . Vertical solar arrays atop the inflated habitat (located at least 8m above the surface) provide nearly continuous power generation for the field station, despite the low polar sun angle [26] . The FLARE surface mobility concept is modeled after the successful Antarctic Search for Meteorites (ANSMET) program. The ANSMET program, sponsored by the National Science Foundation (NSF), has recovered more than 22,000 meteorites from the ice of Antarctica since 1976 [67] . A small team of scientists is deployed each year to a scientific "field station" in the remote Antarctic mountains for 4 to 8 weeks of research and sample recovery. The ANSMET team conducts exploration traverses by walking and using individual mobility devices, the Ski-Doo snowmobile. The FLARE optional Phase B precursor mission has limited mass and volume that likely precludes an unpressurized 2-person rover, but could accommodate smaller 1-person vehicle(s) that fits (partially disassembled) within the deflated habitation module. NASA is currently seeking commercial designs for an unpressurized 4-wheel, 2person "buggy"-style Lunar Terrain Vehicle (LTV) for Artemis missions [68] . Following are considerations for selecting a one-person Lunar All-Terrain Vehicle (LATV), rather than LTV for early mission crew mobility on the lunar surface.

Modification of an existing commercial design is likely much faster and cheaper than developing a new vehicle, so the LATV might be available before a LRV. Many vendors provide "hardened" All-Terrain Vehicle (ATV) designs for off-road, sand, snow, or ice conditions with electric propulsion systems using removable batteries (which allows for rapid replacement of depleted batteries). The LATV must survive the Moon environment, including extremely cold conditions and dusty regolith. The LATV must provide batteries that can be easily recharged. The LATV must be compatible with the lunar xEMU suit design so that astronauts can sit and ride on the vehicle comfortably. Commercial companies may contribute "in-kind" development resources for a LATV to gain publicity and market attraction for collaboration with NASA. A reference design by Doohan [69] provides an example of a commerciallyavailable, 3-wheeled, electric vehicle with an Earth weight of ~160 kg.

A typical ATV weighs from 100-300 kg on Earth, which is 50 kg (approximately 110 pounds) on the Moon with 1/6 th Earth's gravity. A single astronaut can partially or completely lift, rotate, or push this mass on the Moon. If the LATV is stuck in soft regolith, the crew could lift the wheel or pull the vehicle towards a more solid surface.

The LATV can be designed to easily assemble/disassemble into large pieces, so that they could perhaps be initially unloaded from the hatch in the deflated surface habitation module (FLARE optional Phase B with DE1) or from a CLPS nearby lander. The crew then rapidly assembles the components into a functional vehicle on the lunar surface. Reversing the process allows long term storage of the vehicles in the habitat, or in a lunar surface shed that protect them from solar wind and micro-meteoroid exposure while providing warmth and power to survive lunar nights. ANSMET teams in Antarctica have often cannibalized parts from one snowmobile to keep other vehicles functional [70] . Having multiple identical vehicles on the lunar surface similarly provides an inventory of spare parts in the event of unforeseen failures.

The mobility system needs solar array(s) with recharging capability for the LATV batteries. As previously discussed, solar arrays atop the inflatable habitat could be used for recharging the LATV batteries, or a separate ground station could be installed. Multiple excess batteries should also be charged and ready for rapid replacement once an LATV has depleted the energy in its current battery. The recharging of each battery could either be done inductively with the battery remaining on the vehicle (necessary for telerobotic operations), or the battery can be removed by the crew and attached to the recharging terminal (either outside or inside a pressurized volume).

Telerobotics on a LATV adds multiple capabilities to the entire mission. NASA and terrestrial industries have developed the capability to robotically drive a vehicle using a remote operator. For remote lunar operations, the vehicle must be designed to recharge the electrical battery directly with easy connectors to a charging station or through inductive field charging.

Telerobotics allows exploration of surface regions that are precluded from humans (due to slopes, temperatures, or surface roughness/softness). The astronaut could observe the LATV directly and remotely drive the vehicle in conducting operations in these more difficult terrains. The LATV could also be telerobotically operated as a rescue vehicle to recover a stranded astronaut at a distant location.

The LATV can support lunar surface traverses up to 10km from the landing site. Using a pair of LATVs provides redundancy and rescue capability. The pair of astronauts could work collaboratively or independently on tasks. This increases the efficiency of astronaut EVA time and enhances the flexibility in executing traverses. Each LATV is capable of carrying two xEMU-suited astronauts. A rescue LATV could be driven to a stranded astronaut by one crew, or an empty LATV could be telerobotically driven to the disabled crew location.

Similar to the SpaceTug concept for orbital transfers, the LATV provides surface transfers for science and crew. The LATV provides a simple, flexible vehicle derived from existing commercial designs at minimum cost. To minimize development cost and schedule risk, FLARE separates the surface science function on a dedicated trailer. Fixed and deployable instruments, including a robotic arm, are mounted on a chassis that provides power, communications, and computer resources. The trailer is pulled to a desired location by the LATV (driven by crew or telerobotically), left to collect data (but commanded remotely), and then later retrieved by the LATV. One science trailer (allocated mass 0.3 mt) is included in the optional precursor FLARE Phase B mission, but it could be delivered separately on a CLPS lander. After a successful launch and transit to the Moon, the Orion+ESM docks with assets in LLFPO, which is, at a minimum, the human lander (AE1 + DE2) and the RST. The PPE and HALO may also be present (as Gateway) but are not required. On Crew Flight Day 8 (see Figure 10 ) all four Orion crew board the AE1 and seal the hatch. They then descend to the lunar surface and land at the designated site. After safely landing on the lunar surface, the operations concept depends upon the available surface assets. If Phase B is NOT implemented, the four crew will conduct limited walking EVAs alternating pair teams using the 2 xEMU suits in the AE. The surface duration may be very brief (hours) or extend up to one week before the crew returns to Orion.

The FLARE Option B extends to a 14-day surface mission with longer science traverses using the prepositioned assets. All four Orion crew descend to the lunar surface inside the pressurized AE1. Two of the crew are dressed in xEMU suits, and two are dressed in the Orion LES. Once the DE2 has landed safely on the Moon, the AE1 is depressurized and the two crew dressed in xEMU suits depart. The two crew remaining in the AE1 -dressed in the LES that can keep the astronauts alive for up to 6 days [71]then repressurize the AE1 and wait for the return of their crewmates. The two crew in xEMU suits walk to the deflated habitation module and unload the crew mobility vehicles and two xEMU suits from inside. They then inflate the habitation module and attached inflatable airlock, then return to the AE1. The AE1 is again depressurized to allow the xEMU crew to enter, then the AE1 is pressurized and the two crew left inside doff the LES and don the xEMU. The AE1 is again depressurized and all 4 crew depart to the DE1 landing site. The crew ingress/egress the habitation module via the inflatable airlock. Science, ISRU, EVA tools, and solar array mobility equipment are unloaded and deployed. The crew sleeps in the habitat for the next 14 nights. Each day, one team conducts traverses using the LATVs, and one team remains at the habitat landing site (either inside the habitat or conducting walking EVAs). Each traverse pair includes one crewmember on a LATV, and the other one on an LATV pulling the science trailer. A third LATV (if available) can be pulled behind the non-science trailer pulling the LATV, driven telerobotically, or left behind at the habitat site (for rescue). Each team of two crew will alternate days as an LATV traverse team or local habitat team. Upon completion of the 2-week campaign, the crew will reconfigure the site for remote science operations by enabling telerobotic equipment and stowing equipment and configuring the habitat module for human absence. The crew then reverse the sequence of walking between the AE1 site and habitat site returning two xEMU suits to the habitat and bringing two xEMU suits into the AE. On Flight Day 22 (see Figure 10 ), all four crew in the AE1 return from the lunar surface to Orion in LLFPO. Table 5 [20, 31, 32, [72] [73] [74] [75] [76] [77] . Mass fractions are used to equalize delivery mass to LEO and TLI across vendors. The 0.94 mass fraction for performance reduction from LEO 28.5° to LEO 51.6° is based upon published values for the SpaceX F9 Block 2 [32] . The 0.27 mass fraction for performance reduction from LEO 28.5° to TLI is based upon published values for SLS [75, 78] . Note the Space Shuttle (retired) is an entirely different architecture from CLVs that does not follow the mass fraction for 51.6°, and it was not capable of delivering payloads to TLI. Also included in Table 5 is the predicted performance of a ULA SpaceTug (FLARE concept) in either a single (See Figure 1 ) or double-stacked (See Figure 3) configuration.

The FLARE LEO assumption is a 400km circular orbit at 28.5° inclination. The chosen CLV for human lander (AE1 and DE2), or optional Gateway (PPE and HALO), components is the ULA Atlas V (A5) 551 rocket, which is capable of lifting 18.5 mt to LEO [79] . The FLARE reference CLV for the SpaceTug (including the RST) delivery to LEO is the SpaceX Falcon 9 (F9) Block 5 rocket, which is capable of lifting 22.8 mt to an unspecified altitude [31] . Prior SpaceX documentation reveals approximately a 5% reduction in payload delivery with a F9 Block 2 between a circular 200 km orbit (delivery = 10.454 mt) and a circular 400 km (delivery = 9.953 mt) orbit [32] The Federal Aviation Administration (FAA) publishes an annual report on the estimated launch costs for commercial vendors [72] , and frequent website announcements provide general industry details on price modifications. FLARE assumes a launch cost of each SpaceX F9 is $62M, each SpaceX FH is $90M, each ULA A5 is $153M, and each SLS B1 is $2000M [80] . In general the SLS cost overwhelms the cost of all other components. Using only one SLS launch for the crewed Orion+ESM minimizes this expense, but forces the human lander elements (AE1 and DE2) to be launched separately on CLVs to LEO and then pushed to LLFPO by SpaceTugs for autonomous docking.

The unit cost (after development) of a SpaceTug is expected to be similar to the existing ULA Common Centaur. The SpaceTug assumed build cost is $80M each (~1/2 the cost of an Atlas V 551 launch). The development costs of the human lander components is not included. The optional precursor mission (Phase B) costs are also not estimated, but FLARE has chosen existing components which could be modified for less cost than new development.

In summary, the transportation costs for the minimum required phases (C,D,E) is estimated at $3,016M (5 SpaceTugs, 5 F9, 2 A5, and 1 SLS B1 The FLARE provides a reasonable, practical sequence to deliver four Americans to the Moon and then return them safely to Earth. The FLARE supports the Artemis Program using components currently being developed by NASA. The underlying FLARE concept is to maximize available commercial technology for the mission, and limit development of entirely new systems or vehicles. As new technology or vehicles become available, FLARE provides multiple growth opportunities for their integration. Transportation costs are minimized using existing CLVs for delivery of components to LEO, and the SpaceTug (which is a modification of mature, successful technology combined with newer, proven innovations) provides the necessary propulsion for transfer between LEO and LLFPO (which is the optimal lunar orbit for sustained lunar surface operations). A new human lander is required, but commercial contracts provided by NASA are underway for its further definition and development. FLARE provides a human lander reference design that can be replaced with any of the selected commercial landers. FLARE allows for inclusion of Gateway elements (PPE and HALO) and lunar surface precursor equipment to extend and enhance human surface operations. Crew lunar surface rovers being investigated by NASA can be added, although FLARE provides a reference concept for an individual vehicle, called the Lunar-ATV (LATV), for early human surface missions. Advanced technology demonstrations for on-orbit fluid transfer, ISRU propellant resupply, and deep space communications satellites are included in FLARE to enable enhanced exploration of cislunar space, then on to Mars. 

Human Landing System (HLS) Architecture Analyses of Alternatives, Design Analysis Cycle (DAC) 2, Attachment A-12

Options for staging orbits in cislunar space

The Gateway Power and Propulsion Element: Setting the Foundation for Exploration and Commerce

ISS Partners endorse modified Gateway plans

NASA to sole source Gateway habitation module to Northop Grumman

NASA planning to launch an integrated Lunar Gateway in 2023

NASA Administrator Marh 17 Statement on Agency Response to Coronavirus

NASA's new Artemis Accords govern how we cooperatively and safely explore the moon and Mars

NASA's 2024 Moon mission is in serious trouble

Strategic considerations for cislunar space infrastructure

Targeting cislunar near rectilinear halo orbits for human space exploration

NASA's Gateway: An Update on Progress and Plans for Extending Human Presence to Cislunar Space

The fall, recovery, orbit, and composition of the Tagish Lake meteorite: A new type of carbonaceous chondrite

Selects First Science Instruments to Send to Gateway

The Lunar Gateway is No Longer a Required Part of the Artemis Mission to Return to the Moon by 2024

An overview of NASA's exploration mission 2 (EM-2)

The multi-purpose crew vehicle European service module: a European contribution to human Exploration

NASA human spaceflight architecture team: Lunar surface exploration strategies

Apollo by the numbers: a statistical reference

Space Launch System: SLS Payload Delivery Capability Overview

NASA's Space Launch System: Artemis 1 Vehicle Nears Completion

Advanced Exporation Systems Status to NASA Advisory Committee

Recent gravity models as a result of the Lunar Prospector mission

Lunar frozen orbits revisited

Frozen orbits about the moon

Illumination conditions of the lunar polar regions using LOLA topography

Illumination conditions at the lunar south pole using high resolution digital terrain models from LOLA

The Centaur Upper Stage Vehicle. 4th IAC on Launcher Technology

SNC Selects ULA for Dram Chaser Spacecraft Launches

The ACES Stage Concept: Higher Performance, New Capabilities, at Lower Recurring Cost

Falcon 9 Launch Vehicle Payload User's Guide Rev 1

SpaceX may have signed a fairing agreement with ULA supplier RUAG (Update: no agreement)

Distributed Launch-Enabling Beyond LEO Missions

Cryogenic propellant depot and integral sunshield

Simple, robust cryogenic propellant depot for near term applications

Cryogenic Orbital Testbed (CRYOTE) Development Status

Passive zero-boil-off storage of liquid hydrogen for longtime space missions

Zero boil-off system testing

NASA interest in an interplanetary highway supported by Propellant Depots

Apollo lunar orbit rendezvous architecture decision revisited

Comparing the results of an analytical model of the no-vent fill process with novent fill test results for a 4.96 cu m (175 cu ft) tank

Cryogenic Fluid Management Technologies for the Space Launch Initiative

Orbit Fab demonstrates satellite refueling technology on ISS

Liquid Nitrogen Testing of ISRU Liquefaction Methods in Unsteady Applications

35 Years Ago: STS-41G -A Flight of Many Firsts

Bold They Rise: The Space Shuttle Early Years

Liquid transfer demonstration on board Apollo 14 during transearth coast

Electric propulsion research and development at nasa

Development of High Power Hall Thruster Systems to Enable the NASA Exploration Vision

Scaling of Ion Thrusters to Low Power. 1998. 53. NASA. The Second Conference on Lunar Bases and Space Activities of the 21st Century

The Lunar Base Handbook, An Introduction to Lunar Base Design. Development, and Operations

Lunar Propellant Infrastructure for Flexible Path Exploration and Space Commerce

The Value of the Moon: How to Explore, Live, and Prosper in Space Using the Moon's Resources

NASA Lunar Lander Concepts Beyond Apollo, in NTRS. 2019, NASA. 58. NASA. NASA Selects Blue Origin, Dynetics, SpaceX for Artemis Human Landers

NASA to pay private space companies for moon rides

Selects Experiments for Possible Lunar Flights in 2019

NASA Selects 12 New Lunar Science

High-Performance MMOD Shielding

Polyethylene/boron nitride composites for space radiation shielding

Zip Technology: ZipNut and ZipCONNECTOR

Selects US Companies to Advance Space Resource Collection

Bigelow Expandable Activity Module (BEAM) ISS Year-One. 2017. 67. Institution, S. ANSMET, the Antarctic Search for Meteorites

Send Ideas for Lunar Rovers

Quick Status Report and a Short Essay of the Evils of Snowmobiles

Suit Equipped to Expect the Unexpected on Artemis Missions

The Annual Compendium of Commercial Space Transportation

Space shuttle: The history of developing the national space transportation system

ULA. Atlas V Launch Services User's Guide

The NASA SLS Upper Stage Development and Mission Opportunities

Atlas V Launch Services User's Guide

The White House Puts a Price on the SLS Rocket

The authors acknowledge and thank our coworkers and management in the Astromaterials Research and Exploration Science (ARES) division at NASA's Johnson Space Center (JSC) for their support and editorial comments, especially Paul Abel, John Gruener, and also Blake Demesnil with Jacobs Technology, Inc. for his graphics skills.

11.0 11.0 11.0 11.0 REFERENCES REFERENCES REFERENCES 

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Both authors, as employees of NASA, created this research while supporting the Artemis Program. The authors are not participating in any selection board for contracts regarding commercial lunar landers or surface mobility rovers. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: