A Concept for an On-Orbit Propellant Depot

Update 10-25-2022:  a version of this was presented at the 2022 Mars Society convention at ASU in Tempe,  AZ.  It was well-received.

---------------------------

This article describes a concept for an on-orbit propellant depot capable of refilling visiting craft,  which presumes the visiting craft have rendezvous and docking capabilities.  This is only a concept,  which has not had any design analysis. 

Such a depot needs orbit adjustment,  reboost,  and debris avoidance propulsion capability.  There is the option to propel the depot sufficiently that it might “go where the job is”,  instead of having the visiting craft come to it,  which could assist with some aspects of orbital debris removal.  But because the facility is so large,  a separate space tug is a far better option for such missions.

Classes of Propellants

There are fundamentally two completely different classes of propellants to handle.  These are the more-or-less room-temperature storable propellants,  now most commonly used in thruster and attitude-control systems,  and there are the cryogenic “main stage” propellants,  which do in fact include the storable material kerosene.  The thruster storables need to be supplied in relatively small quantities,  while the main stage cryogens (and the kerosene) need to be handled in very large quantities.

Zero-gravity propellant ullage problems have been thoroughly discussed in Ref. 1.  The solution for the storable propellants (and the storable kerosene) is the bladdered tank.  This could be a free bladder,  a side-everted bladder,  or an axially-everted bladder.  The exact bladder geometry to be selected is not the real issue here,  in a concept design description.  Those geometries are discussed in Ref. 1.

The solution for the cryogenic propellants has been rather difficult to find.  As discussed in Ref. 1,  the historical solution for various rocket stages requiring free-fall restart has been application of ullage thrust.  The same is said in Ref. 2.  There have been some ideas tested and flown that take advantage of surface tension effects,  as discussed in Ref. 2.  Those same references point out that application of ullage thrust can be expensive in terms of ullage motor propellant quantities,  if you have to do it many times.  The solution selected for this quandary is the spinning tank technology described in Ref. 1.

The other piece of this puzzle is the facility occupancy by a crew.  There is no need to continuously-man this facility!  You only need to send up a crew when there is a propellant transfer to be made.  Once done,  the crew returns to Earth.  There are already standard docking ports of the necessary types (plural) on the International Space Station.  You just equip this facility’s crew module with those same docking ports,  plus some room for a few future new designs.

What Propellants to Have In Stock

The small-quantity thruster propellants under consideration here are the four common hydrazine variants,  and hypergolic nitrogen tetroxide (NTO) as the oxidizer for all of them.  Those hydrazine variants are plain hydrazine (C₂H₄),  monomethyl hydrazine (MMH),  unsymmetrical dimethyl hydrazine (UDMH),  and the brand-name product Aerozine-50,  (a 50-50 blend of UDMH and plain hydrazine).  These are commonly stored at significant pressure,  for pressure-fed thruster systems.

The mostly-cryogenic combinations are (storable) rocket-grade kerosene (RP-1),  cryogenic liquid methane (LCH₄,  which is not the same as liquified natural gas),  and cryogenic liquid hydrogen (LH₂),  with cryogenic liquid oxygen (LOX) as the oxidizer for all of them.  These are commonly stored at rather low pressures,  only such that boiloff is slowed or prevented.  Such engines are usually fed by turbopumps.  List follows:

Fuel                       oxidizer                application

C2H₄                     NTO                      small-quantity thrusters

MMH                    NTO                      small-quantity thrusters

UDMH                  NTO                      small-quantity thrusters

Aerozine-50          NTO                      small-quantity thrusters

RP1                       LOX                       main stage propellant

LCH₄                     LOX                       main stage propellant

LH₂                        LOX                       main stage propellant

There are two other storable oxidizers,  although neither has been used in spacecraft for some years now.  Those are inhibited red fuming nitric acid (IRFNA) and high-test hydrogen peroxide (H₂O₂).  If such should ever be demanded again,  you just add extra storable oxidizer modules to the facility for them.

Properly-Applying Spinning-Tank Technology to Solve Cryogen Ullage Problems

The “trick” is not to spin the cryogenic tanks while they are attached to the depot!  That introduces severe sealing problems and the resulting logistical problems. 

Instead,  the craft to be refueled docks to a tank,  detaches it from the depot,  and moves off some distance,  then spins up with the tank docked to the craft,  to effect the transfer tank-to-spacecraft with pumps.  Then it de-spins and returns the tank to the depot. You do this twice:  once for the fuel,  then again for oxidizer.  See Figure 1.  Vapor exchange is required along with liquid transfer,  per Ref. 1. 

Figure 1 – Spinning-Tank Technology Applied to Cryogenic Refill On-Orbit

Transfers to the depot from tanker craft work exactly the same way:  detach the empty tank to be filled,  move off and spin up the docked tank-and-spacecraft,  transfer the liquid (and vapor),  then de-spin and return the filled tank to the depot.  See again Figure 1. 

Spin-up/spin-down thruster propellants are minimized by the use of “rifle-bullet spin”,  and also very much by being limited to only the inertia of spacecraft plus tank,  not the entire depot or its entire assemblage of tanks!  It is end-to-end docking plus that “rifle-bullet spin”,  of only the tank and visiting craft,  so the moment of inertia,  and total impulse required of the spin thrusters,  is minimized.  This was well-discussed in Ref. 1.  At about 1 rpm and near 2 m radii,  the spin induces about 0.0022 gee (corrected value 2-4-22). 

That's for the cryogenic materials only.  The storables are in the same sort of bladdered tanks as the spacecraft that use them.  You just dock with the depot module and pump,  from bladdered tank to bladdered tank.  Pressurant gases are involved to push on the bladders.  There is no spin needed.

The only design requirements imposed on the craft transferring-out cryogenic liquids,  are the need for propellant suction points out on the tank periphery,  as well as on the aft domes,  plus the presence of perforated radial baffles in the tank that speed the spin-up of the liquid globules inside.  See again Ref. 1.  Craft that only receive liquids do not need periphery drains or baffles,  spin or not.

Depot Station Layout

Excepting kerosene,  the bladdered storables are primarily for attitude/maneuvering thrusters.  Those quantities are far lower,  no matter which combination you desire.  So there are hydrazine tanks for around a 1 cubic meter each of hydrazine,  MMH,  UDMH,  and Aerozine-50,  plus tanks for about 5 cubic meters of NTO.   Those are 1-ton-class quantities of each fuel material,  and a 5-ton-class quantity of NTO to support any-and-all of them. 

The storable kerosene,  and the cryogenics LOX,  LCH₄,  and LH₂ are needed in far,  far larger quantities,  being main stage propellants.  The kerosene transfers just like the other storables,  except the tanks are far larger,  around 300 cubic meters for a nominal 250 ton quantity.  The LCH₄ is near twice that volume,  and the LH₂ around 11-12 times that volume.  This assumes very large receiving craft,  and roughly-equal frequency of demand for each fuel type,  which may or may not be actually true.  The shared LOX volume to support each of these is simply huge (at around 2300 cubic meters),  that being the largest volume stored on the depot station.

See Figure 2 for some crudely-estimated quantities.  These were done in an Excel spreadsheet.

Figure 2 – Crudely-Estimated Quantities for Storables and for Cryogens

Reboost and attitude control could be a combination of the storable propellants plus electric propulsion powered by some solar panels,  all located on (or in) a “power and propulsion module”.  This power and propulsion module would be at one end of the depot station.  The crew compartment would be at the other end.  You want the NTO tanks near one end,  as physically far from the 4 types of hydrazine tanks as possible,  to avoid any possibilities of explosions-upon-contact,  if there are ever any leaks.  These materials are hypergolic,  even in vacuum.

The large kerosene bladdered tanks should probably be at the same end as the hydrazine tanks,  just closer than the hydrazines to the center of mass.  The largest and heaviest item is the LOX tankage,  which should be near center of mass.  The LH₂ tanks can be at the same end as the NTO storage,  just closer to the center of mass than the NTO.  I show the LCH₄ tanks forward of the LOX,  just to better center the LOX on the center of gravity.  Because of the toxicity of NTO residues from any spills,  the crew compartment should be far away from the NTO storage.  That puts the propulsion and power module at the NTO end of the facility.  See Figures 3,  4,  and 5.

Figure 3 – The Bladdered-Tank Storable Modules

Those modules associated with bladdered tanks need an adjacent docking adapter for the visiting spacecraft,  with the tanks situated such that there is clearance to come in and dock.  Once docked,  you just hook up the transfer lines,  and do the fluid transfer.  This would also apply to tankers refilling the facility.  There is no spin,  and the pressurant gas supplies plus the plumbing and electrical,  are within the depot module truss.  The visiting craft undocks and then redocks elsewhere,  to switch from fuel to oxidizer.  There are manipulator arms on the modules to assist with the docking operations.

Figure 4 – The Cryogenic Tank Modules

Figure 5 – Recommended Facility Layout

The cryogenics are a different situation.  These large tanks must be perpendicular to the axis of the depot facility,  and must have appropriate docking gear on both ends of the tanks.  The visiting craft docks directly to the end of the tank,  and makes the fluid and vapor connections.  Then it detaches the tank from the depot,  and moves off a short distance (for safety).  The docked vehicle-and-tank gets spun up for the fluid transfer,  then de-spun once completed.  It redocks the tank with the station,  undoes the connections,  and finally undocks from the re-docked tank. 

If both fuel and oxidizer are cryogenic,  then one repeats this procedure with the appropriate other tanks.  There are a few relocatable manipulator arms that attach to the cryogenic tank docking structures,  which assist with all the docking operations.  Not every tank needs a manipulator arm.

Preventing Incorrect Connections

If the visiting vehicle is to be refilled from the facility,  its tanks need the plumbing connections to make the indicated hook-ups.  This is pretty simple for the storables,  including the kerosene.  For the cryogens,  the visiting vehicle does not necessarily need the perforated spin baffles or periphery fluid drains that are needed in the depot tanks.  But,  any tankers attempting to refill facility tanks will need these extra plumbing features,  as was indicated in Figure 1.

Since the oxidizer quantities are nearly always larger than the fuel quantities,  I recommend making the hose/plumbing connections and fittings for oxidizers about 1.5 to 2 times physically larger than the fuel fittings.  That prevents incorrect hook-ups,  something particularly important with the hypergolic hydrazines and NTO. 

However,  there also needs to be some sort of keying feature on all of the fuel connections,  so that only the correct fittings can be coupled,  even though they are of equal size.  You do not want to mix species among the hydrazines,  or get kerosene when you wanted a hydrazine,  or vice versa. 

You also need to key the oxidizer fittings by species,  just in case IRFNA or high-test H₂O₂ ever get added to the facility.  It would be harder,  but not impossible,  to mix up NTO and LOX,  just because of the difference in bladdered versus spinning-tank technologies.  But keying the fittings absolutely stops that.

These fitting sizes and keying features are a critical safety requirement,  and should be standardized “up front”,  so that everyone is using the same equipment!

Power and Propulsion Module

This item (also shown in Figure 5) has the thrusters and the solar panels required to power the entire facility,  and to move it as needed.  Those thrust applications would include reboost for orbital decay,  deorbit at end-of-life,  debris avoidance maneuvers,  and general attitude control.  I’d recommend something similar to the Space Shuttle “OMS” and attitude-thruster systems,  which were a hydrazine variant-NTO thruster design,  plus some sort of electric thruster,  for its far-higher specific impulse,  for the reboost operations and any other orbit changes. 

My own preference for the electric thruster would be an iodine thruster,  for the easy fuel storage and handling,  even though those are not yet very common,  or even in wide use.  The bigger the electric system,  the more solar panels would be needed on this module,  and there are geometric limits to that.

Crew Habitat and Docking Module

The layout in Figure 5 also shows a crew habitation module at the other end from the power and propulsion module.  There is attached to it a multi-port docking module for crew transfer vehicles.  Crews would be required during any refilling operations for visiting vessels,  or for tankers re-supplying the facility.  Otherwise,  there is no need for crew aboard,  not even during transits to other orbits.  Crews would be aboard relatively short-term for the necessary fluid transfer operations.  Long term occupancy (like the International Space Station) is not an issue here!

Options For Spinning-Up (and De-Spinning) Cryo-Tank and Vessel

The concepts selected here do minimize the thruster propellant quantities required for spin-up and de-spin,  assuming thrusters are used for this purpose.  That is the historically-proven way to do it. 

However,  adding a flywheel for electric torque spin-up and de-spin is another viable (if undemonstrated) option.  This flywheel should probably be within the docking adapter on the end of the tank adjacent to the vessel,  so as to be nearest the docked center of gravity. 

The flywheel will need to be rather massive,  and will spin quite fast,  so that its angular momentum magnitude equals the angular momentum magnitude of the docked cryo tank and vessel.

Adding a “Go-To-The-Job” Capability Is Unattractive

If you want this depot facility to take on the role of scavenging abandoned spent stages for their propellants,  before de-orbiting them,  then you need to add very considerable extra propulsive capability to the depot station.  It would have to go and rendezvous with every target spent stage. 

You do that with very much larger fuel tanks for the electric propulsion and for the storable-powered thruster,  both located in the power and propulsion module.  That option is also shown in Figure 5,  as extra tanks to be added surrounding the power and propulsion module.  The orbital changes to do this are made unmanned,  controlled from the ground.  You send up crews temporarily,  only when propellant transfers are to be made.

Otherwise,  there is just no need for a facility propulsive capacity that large!  That is why the option takes the form of added tanks.  Everything else is the same.  Although,  choosing instead to create a specialized vehicle to operate as a “space tug” to go get the spent stage,  and transport it to the facility,  is probably the better concept,  and by far.  Such a space tug is just not that massive,  while this propellant depot facility is simply enormous.

Thrust Sizing

But in any event,  the thrust-induced accelerations need to be quite low,  in order to avoid large bending loads in the cryogen tank attachments.  Accelerations on the order of 0.01 gee or less should be fine,  with the storable thruster.  It is quite likely the electric propulsion would provide accelerations more in the ballpark of 0.0001 gee.  While very low indeed,  that’s all right:  the electric propulsion would only be used while unmanned,  to make significant orbital changes that are not time critical.  More time-critical adjustments (such as debris avoidance),  would use the storable-propellant thruster.

References

#1. “Propellant Ullage Problem and Solutions”,  Gary W. Johnson,  posted 18 August 2021 to http://exrocketman.blogspot.com. 

#2. “A Detailed Historical Review of Propellant Management Devices for Low Gravity Propellant Acquisition”,  Jason W. Hartwig,  NASA Glenn Research Center, Cleveland, OH, 44135, USA; undated AIAA paper but ca. 2015;  (pretty much the same information was also in the Hartwig Ph.D. dissertation).

For articles posted on the “exrocketman” site,  use the navigation utility left side of page.  Click on the year,  then on the month,  then if need be on the title.