Solid rocket propulsion technology for de-orbiting spacecraft

Adam OKNINSKI

Center of Space Technologies, Lukasiewicz Research Network - Institute of Aviation, Warsaw 02-256, Poland

KEYWORDS De-orbit;De-orbiting technology;End-of-life disposal;Satellite propulsion system development;Solid propellant;Solid rocket motor;Space debris mitigation;Space traffic management

Abstract This paper presents the topic of using solid rocket propulsion for de-orbiting spacecraft,in order to fulfil space debris mitigation requirements.The benefits and disadvantages of using such means are discussed.A dedicated system can be implemented in the satellite design phase and shall be a key subsystem of platforms inserted into orbit. Uncontrolled, semi-controlled and controlled de-orbit can be completed using solid rocket motors.Their impact on the space debris environment is discussed. Specific requirements for dedicated propellants and systems are provided. While the majority of presently developed rocket systems worldwide require high burn rates, several applications, including de-orbiting, benefit from solid propellants with decreased regression rates. This allows limiting spacecraft accelerations and loads during de-orbit manoeuvres. Moreover, the requirement of minimising solid particle generation is presented.Heritage technology from the Mercury and Gemini human spaceflight programmes, where de-orbit motors were used, is shown.Historical Soviet,American and Chinese film-return-capsule solid propellant retrorockets,enabling deorbit,are also presented.A detailed survey of current work worldwide on end-of-life disposal using solid propulsion is included. Challenges of developing dedicated systems are discussed. Finally, an outlook on solid rocket motor utilisation for space debris mitigation is provided.

1. Introduction

Abbreviations ADNAmmonium Dinitramide APAmmonium Perchlorate EoLEnd-of-Life ESAEuropean Space Agency HNFHydrazinium Nitroformate HTPB Hydroxyl-Terminated Polybutadiene ICBM Intercontinental Ballistic Missile IPOLInternal Policy ISSInternational Space Station LEOLow Earth Orbit GEOGeostationary Orbit GTOGeostationary Transfer Orbit RESIReduced Signature SDMSpace Debris Mitigation SPADES Solid Propellant Autonomous DE-orbit System SSOSun Synchronous Orbit SRMSolid Rocket Motor TRLTechnology Readiness Level TVCThrust Vector Control

While initial research on threats possible due to human-made space debris began in the United States during the 1970s,first recommendations were stated by Kessler and Cour-Palais in 1978 and early standards (titled ‘‘System safety requirements for ESA space systems”) were formulated in Europe by 1988.However at that point they were not yet directly devoted to Space Debris Mitigation.This was followed by dedicated standards in the United States, Japan and France in the last years of the 20th century.It became clear that the key method to limit new space debris generation is to avoid allowing new objects to remain in orbit after their end-of-life.Extensive work during the last two decades enabled raising the awareness of the space technology community. One of the forums allowing global discussion is the Inter-Agency Space Debris Coordination Committee(IADC).Since 2009 the problem of space debris became more evident due to the collision of Iridum-33 and Kosmos-2251 satellites and an earlier successful antisatellite weapon test by China in 2007. It became clear that post-mission disposal is crucial for limiting the speed of growth of the debris population.While Active Debris Removal is necessary to decrease the number of existing large debris,it is a very costly approach and the main solution is to make sure that all new objects delivered to orbit will have a system enabling decreasing their orbital lifetime after the end of their operations. This is crucial in order to secure existing orbital assets, such as the International Space Station.The problem becomes even more serious due to the growing concentration of satellites on relatively similar orbits, as shown in Fig. 1.The MASTER debris model is constantly updated and the significant change in debris spatial density can be analyzed.Moreover,numerous smaller spacecraft,unlike few large ones,are planned to be used in LEO mega constellations. Satellites are expected to be launched more frequently with limited lifetime due to the present business approach and technology becoming obsolete since the beginning of the given satellite’s payload development. Within the Space 4.0 trend numerous new players occur, some with limited experience and a multitude of small satellites with questionable reliability are close to getting to orbit. With over 1000 satellites being launched per year in a few-year horizon, a significant market for endof-life disposal can be seen. The establishment of the Clean Space initiative at ESA confirmed Europe being one of the leaders in Space Debris Mitigation.As of 2020 ESA has its dedicated Space Safety programme, including work on SDM.Meanwhile efforts worldwide continue to unify international standardsand space law.While a few years back spent satellites were to be de-orbited or re-orbited to higher cemetery orbits, recent guidelines propose not to use the concept of reorbiting for LEO, due to it not being forward-thinking in the age of intensive orbital infrastructural expansion.More details on the current status of SDM can be found in Refs..Discussions lead to the conclusion that apart from EoL disposal, Just-in-time-Collision Avoidance capabilityand a Space Traffic Management are needed in order to ensure sustainability of orbital operations.In case of Just-in-time Collision Avoidance recent concepts include use of SRM plume for debris deceleration.However this review shall focus solely on EoL propulsive disposal.

Fig. 1 Debris spatial density distribution on LEO: data from 2009 and 201812.

The key current requirement concerning spacecraft de-orbit is to conduct re-entry within 25 years from the satellite’s EoL.This means that all satellites on orbits above approximately 600 km(value depending on the exact orbit,solar activity, the mass-to-area ratio of the object and aerodynamic configuration) shall require a system enabling their deorbitation in order to fulfill the 25-year rule. The probability of successful EoL disposal is to be at least 90%.The manoeuvre shall ensure that the risk of casualties on ground is less than 10, what can be assessed analytically using dedicated tools for re-entry breakup analysis.The amount of generated debris while using propulsion, shall be limited. In particular debris over 1 mm of size are unacceptable.ESA Internal Policy(IPOL) on SDM excludes the release of particulate matter in orbit. Most recent recommendations are provided in Ref..However a lot is yet to be done,especially taking into account numerous planned commercial mega constellations.The European Cooperation on Space Standardization (ECSS) adopted ISO24113 in the space sustainability branch.Particular SDM standards adopted by specific countries can be found in Ref..

Fig. 2 Outcome of ESA Solid Propellant Autonomous DEOrbit System (SPADES) study comparison regarding use of different chemical propulsion systems for de-orbit30.

Methods for end-of-life de-orbiting of spacecraft include chemical rocket propulsion, electric propulsion, sails, tethers etc.Several de-orbiting strategies can be used: uncontrolled, semi-controlled and controlled de-orbit. In most cases only chemical propulsion enables obtaining relatively large thrust, which allows for controlled and semi-controlled deorbit.Chemical propulsion includes solid rocket motors,hybrid rocket motors, monopropellant thrusters, bipropellant engines and cold gas systems.While the last technology is usually discarded due to very low performance, the other four are commonly compared considering SDM.Monopropellants and bipropellants are often assumed as natural candidates due to their wide use on satellites and their significant flight heritage. Fig. 2presents results of a dedicated ESA study comparing liquid and solid propulsion for deorbit─relative costs, mass budget differences due to required spacecraft velocity increase (delta-V) and expected reliability have been computed for different spacecraft sizes. Bipropellants were not considered for satellites below 200 kg of mass due to system complexity. Reference missions considered SSOs.Solid propulsion was identified as very promising.More details on the study and its assumptions can be found in Ref..Several modern concepts consider hybrid propulsion as one of the best technical solutions for EoL disposal.Unlike for typical SRM, reignition and adapting the delivered total impulse is possible. However, hybrids lack in-orbit flight heritage, while most commonly used mono- and bipropellants include toxic compounds and are to be replaced in the longer term.The key benefit of solid propulsion is its high density performance, storability and autonomy. Several trade-offs show good potential for use of solid rocket propulsion for EoL disposaland this technology shall be discussed in this paper. Other analyses considering use of SRM can be found in Refs., but no survey of existing technical solutions and heritage is commonly available. From the technical point of view,each of the four mentioned chemical propulsion methods is feasible for use for de-orbiting. In today’s market the economic and operational factors shall be decisive.

Chemical propulsion allows de-orbitation for all spacecraft sizes as well as for launch vehicle upper-stages and jettisoned components (such as Sylda, Speltra).De-orbit manoeuvers are done for SDM, but also for re-entry and recovery of payloads. This relates to human spaceflight, biological payloads and recovery of reconnaissance film capsules. It can also be due to safety measures, as in one of the first concepts mentioned in 1966 - due to nuclear waste hazards of Systems for Nuclear Auxiliary Power (SNAP) satellites.Typical deorbit manoeuvres for objects on LEO require delta-Vs of up to a few hundred m/s, depending on the initial orbit, re-entry flight path,satellite configuration and type of disposal(uncontrolled or controlled de-orbit). Example data is presented in the following paragraphs.

2. Scenarios for modern application of SRM technology for SDM

Fig. 3 EoL spacecraft removal strategy decision tree by Airbus Defence & Space38.

Three typical approaches are discussed for EoL disposal using chemical propulsion. These include: uncontrolled, semicontrolled and controlled de-orbitation, presented in Fig. 3.The primary driver affecting the decision on the necessary re-entry scenario is the ground casualty risk. This decision must be taken already during the spacecraft initial development phase, since it is the key aspect influencing propulsion system sizing and autonomous SRM implementation in the platform.Typical components and subsystems which are decisive regarding spacecraft demisability include: propellant and pressurant tanks, reaction wheels, magneeters and solar array drive mechanisms. The first solution, uncontrolled de-orbit,requires transferring the given satellite to an orbit, which ensures that the 25 year disposal rule is fulfilled.This leads to a long de-orbit duration. One will not know exactly when and where the debris will re-enter the dense layers of Earth’s atmosphere. Therefore, large, not-fully-demisable objects should not be subject to this scenario. Only satellite disposal with ground casualty risk below 10can be allowed for this type of re-entry. The second approach – semi-controlled reentry – enables ensuring that the debris will fall within a few or preferably one orbit.This way it is possible to limit casualty risk,despite a large possible impact area,via targeting re-entry to occur over ocean areas.It is the most expensive and complex scenario and is inevitable for large satellites and launch vehicle upper-stages.Objects are targeted to enter the thicker layers of the atmosphere with an angle of approximately -1.5° and fall into the South Pacific Ocean Uninhabited Area(SPOUA),which is believed to be the safest place for re-entry in order to mitigate casualties:both in lives and property.Since 1971 over 260 satellites have been directed into this area.Considering a single burn and a Hohmann-like manoeuvre, the spacecraft will have its new perigee in the dense layers of the atmosphere, re-entering half an orbit after the burn delivery. Commonly a perigee altitude of 80 km or lower is assumed to allow immediate EoL disposal.Direct controlled de-orbitation limits the overall disposal time to a minimum.While a single burn scenario is considered in most cases, ESA data on the difference in delta-V for a two-SRMburn scenario for de-orbit is also available in Ref.. Delivering a few consecutive burns is possible using an SRM cluster.Clusters are anyhow expected in many satellite applications since long total burn durations are needed to limit maximum thrust and acceleration for spacecraft with fragile appendages.To ensure having the thrust vector aligned with the spacecraft’s centre of mass, firings of SRM pairs are typically proposed.In case of firing a single SRM, it is difficult to precisely control the impact location when initial high LEO is considered.Therefore a stepped scenario of controlled de-orbitation has been proposed by European Large Satellite Integrators. This way the first step would be performed by the satellite’s main propulsion system (potentially electric)and would lower the orbit from the nominal operational orbit of the given satellite to a circular orbit of approximately 300 km of altitude. The second manoeuvre, completed using the dedicated SRM system, would lower the orbital perigee altitude from 300 km to below 80 km,thus allowing controlled re-entry.For a conservative perigee altitude of 40 km,the theoretical delta-V is 77 m/s.This approach allows precise control of the re-entry footprint and makes efficient use of the SRM’s high thrust, while combining the SRM firing with an earlier delta-V delivered by a potentially higher specific impulse system. Another propellant-efficient method for controlled deorbitation assumes an active propulsive phase, followed by a passive phase when drag furtherly decreases the satellite’s orbit and a final firing is concluded after the passive phase.However, while limiting the total necessary delta-V, it requires lengthening the mission duration in comparison to direct controlled de-orbitation, thus increasing costs.

Fig. 4 Effect of initial circular orbit altitude and re-entry flight path angle on necessary delta-V to de-orbit34.

While as mentioned, for circular orbits below approximately 600 km no propulsion system firing is required to limit the orbital lifetime to 25 years, higher orbits require delta-V capabilities of spacecraft.For example for a typical 800 km circular SSO,lowering the orbit perigee to 80 km via a Hohmann transfer, thus allowing re-entry, is equivalent to a 200 m/s delta-V.Initial circular orbits were found to be the most demanding in terms of delta-V budget.There are cases when propulsion system firing is needed for very low orbits. This occurs in the case of controlled re-entry requirements. In general controlled re-entry leads to higher delta-V requirements for de-orbit propulsion systems for a given initial orbit(unless the de-orbit propulsion system is only used for the final burn and the majority of the orbital altitude change is done utilising the satellite’s main propulsion system). Different levels of thrust can be used.explains that a low-thrust continuous firing manoeuvre results in a spiral-type trajectory and leads to a higher delta-V than in case of impulsive manoeuvres. Fig. 4shows the necessary delta-V to enable a Hohmann transfer manoeuvre for several initial circular orbits in order to have reach a flight path angle in the range of -1.5° to -2.5° at 120 km of altitude.

3. Up to date use of solid rocket propulsion for space applications

Solid propellant motors are the first historical developments for both military and civil rocket propulsion applications.World War 2 advancements in the field of efficient castable composite solid propellants led to increased work on novel compositions and brought the introduction of Ammonium Perchlorate, which replaced Potassium Perchlorate as oxidizer in the late 1940’s.Further developments included work on new binder systems and in 1958 aluminium fuel was proposed by Rumbel and Henderson for performance improvement.Major relevant work in the Soviet Union took place in the 1960s, however significant theory of combustion was introduced in the preceding decades.

Space applications include use of solid propellants in i.e.boosters, main-core propulsion, ullage motors, separation motors, spin-control motors, launch abort systems, gas generators, landing, deceleration, descent and recovery systems.Further in-space use of SRMs has been performed numerously through utilisation of SRM kick stages.The first successful use of an SRM in orbit, apart from ullage SRMs etc., was in the third stage of the Vanguard vehicle in March 1958. The Grand Central 33KS2800 SRM used a polysulfide and AP composite propellant, designated GCR-201C.Other historical solid propellant upper stages were present in vehicles including: Thor-Delta (since 1960), Delta (1962), Scout(1964), Thor-Burner (1965), Lambda 4 (1970), Mu vehicle family (1971), Long March 1 (1970) and Black Arrow(1971).Up till today several active launchers use SRMs in their upper-stages, although most new architectures under development utilise liquid upper-stages for enhanced mission flexibility for a wide range of payloads. Historical in-orbit use of SRMs also includes propulsion for reaching GEO from GTO and to perform interplanetary trajectories. Example kick motors include STAR systems from ATK and European MAGE motors.One of the SRM which was the beginning of the STAR motor family is the retrorocket of the Surveyor unmanned lunar lander, the STAR-37.The B-4 Surveyor retrorocket used over 600 kg of propellant, having a burn duration of approximately 40 s and a thrust level of 44 500 N.While it was the main contributor to the spacecraft deceleration, it was ejected before final touchdown.Similarly, the Europa lander mission currently planned by the Jet Propulsion Laboratory, considers an SRM for final vehicle deceleration.An off-loaded STAR-48B SRM is proposed for the de-orbit burn.The topic of this paper, however,focuses on applications of dedicated de-orbit motors, used in Earth orbit, thus solid propulsion for exploration descent elements is only mentioned.Several other applications are shown,in case of similar propellant or system requirements.

4. Aluminised propellant impact on space debris environment

Current state-of-the-art solid propellants usually include aluminium powder. For typical SRMs more than 30%of the initial propellant mass is expelled as AlO.However, most particles rapidly re-enter the thicker layers of the atmosphereand less then 5% of aluminium oxide particles survive over 1 year in orbit.This is naturally highly dependent on the orbit considered and SRM firing conditions.Particle size,area and surface shape affect drag and particle orbital lifetime,while J2 gravitational potential zonal harmonics, solar radiation pressure and luni-solar attraction also play a role.Dust particles have exhaust velocities up to 3000 m/s. While metalized formulations generate small particles during nominal operation,tail-off burning at lower combustion chamber pressures leads to significant oxide slag. This is presented in Fig. 5, where nominal and tail-off SRM performance in high-altitude conditions are shown for the first stage of the Pegasus launch vehicle and in Fig. 6with the STAR-37 SRM during vacuum ground testing. While most debris from SRM firings are below 10 μm of diameter,research shows that bits of AlOexceeding 10 mm in diameter are common for SRM burn tail-offs. Some sources mention particles up to 50 mm of size, with velocity emissions in the order of tens of metres per second.The mass ejected during tail-off is approximated as in the range of 0.04%-0.65%of the initial propellant mass. Since during this transient period low ejection velocities occur,low relative velocities are present and lead to slag orbits being close to the orbits of the SRM themselves.Ref.discusses various non-numerical methods of studying the effect of SRM solid ejecta. Examples from radar observations, particle impacts on space vehicle surfaces and data from ground tests of SRMs are available. The relation between expelled particle size and SRM nozzle throat diameter is discussed in Ref..However, the dependance of AlOsize on SRM dimensions was already suggested in Refs..

Fig. 5 Pegasus launch vehicle first stage SRM firing: caption before burn tail-off (left) and caption taken 15.5 s after burn tail-off(right)62.

Fig.6 STAR-37 SRM tests at a vacuum facility:steady-state burning with chamber pressure above 3.1 MPa(left)and burn tail-off with chamber pressure below 7 kPa (right) 62.

Ref.presents the key sources of large objects in orbits up to 2000 km in altitude. One can see the large impact of SRM slag, which is larger than of sodium–potassium alloy (NaK)droplets released from orbital nuclear reactors before the end of the 1980s, and significantly larger than that of LMRO(Launch and Mission Related Objects). The main source of large debris are fragments of vehicles. Ref.also shows the spatial density of debris on LEO. It can be seen that a relatively constant spatial density is present in case of SRM slag,decreasing for altitudes below 400 km,where orbital residence time is low.In Ref.,results of slag spatial density modelling for LEO, GTO and GEO SRM firings are shown. Fig. 7 and Fig. 8, taken from Ref., provide insight into debris source population fluxes as of 2014, based on the ESA MASTER-2009 model(Meteoroid and Space Debris Terrestrial Environment Reference). Data for the International Space Station(ISS) orbit, a typical SSO (used by Defense Meteorological Satellites), as well as GTO and GEO is presented. The highest cumulative debris flux originating from SRMs among the four orbits considered is visible for the SSO.Fig.7 and Fig.8 show the distinction between SRM dust and slag.The range of debris between 1 and 10 mm is marked and referenced as the critical size range. This is due to its particular importance for human and robotic space programmes since such debris are usually not tracked, while being large enough to pose threat to missions.Fig. 9presents the expected growth of debris spatial density in case of continuing the use of aluminised propellants in-orbit. It can be seen that, apart from orbits below 1000 km of altitude, peaks of SRM-related debris occur for approximately 20 000 km and 36 000 km altitudes. These represent firings dedicated to the circularization of orbits of the GPS constellation and transfer orbits into GEO.Ref.shows that the slag population growth will continue,even taking into account presently limited in-space use of SRMs and a recommendation to forbid orbital use of SRMs is given. The same reference provides information that the largest amount of particles are approximately 100 μm in diameter and the most slag residue comes from apogee kick motors.Ref.recommends limiting orbital SRM use, in particular for Medium Earth Orbits and GEO regions, where particle residence times are long.

Fig. 7 MASTER-2009 debris source population fluxes for ISS orbit (left) and example SSO (right) 67.

Fig. 8 MASTER-2009 debris source population fluxes for GTO (left) and GEO (right) 67.

Fig. 9 Prediction of debris spatial density growth assuming current SRM rate of utilisation 63.

More data can be found in ESA MASTER,which allows to assess the distribution of SRM ejecta on Earth orbits. Over 1100 in-orbit SRM firings have been registered since 1958.More information is provided in Refs..The ESA Clean Space initiative has been looking into the topic and nonaluminised propellant research was initiated in 2013 to ensure the possibility to use solid propulsion for EoL spacecraft disposal. Limiting solid combustion products is critical for modern SDM propellant compositions. This ensures compliance with SDM requirements and safety of solar arrays and other critical satellite components, as well as human spaceflight missions.

5. American de-orbit SRMs

Use of SRM for de-orbiting applications began at the very beginning of the space age. However, re-entry was attempted not for the need of SDM, but to recover valuable hardware with data or to recover human spaceflight capsules. The Corona spacecraft family used capsules to recover photoreconnaissance films.Each satellite had one or two capsules,each with a STAR-12 SRM attached.The first launch of a Corona spacecraft took place in 1959. Fig. 10presents Corona J-1,which had two recoverable capsules, which was the spacecraft’s version used most extensively (52 satellites with 94 film recoveries during 1963–1969).The STAR-12 SRM had a gross mass of 28 kg, propellant mass of 18.28 kg, vacuum thrust of 4450 N and burn duration of 10.4 s,providing a vacuum specific impulse of 256 s. It had an external diameter of 305 mm and a compact envelope due to the use of an end-burning grain configuration. Use of small film re-entry capsules was continued through a few decades via operations of the Corona spacecraft, as well as KH-5 Argon, KH-6 Lanyard,KH-7 Gambit,KH-8 Gambit 3 and the much larger KH-9 HEXAGON, which included five return capsules using one SRM each. It is shown in Fig. 11.More recent American photoreconnaissance satellites used electro-optical and synthetic aperture radar payloads,thus not requiring retromotors for film re-entry.

Fig. 10 Corona spacecraft draft with de-orbit SRMs 71.

Fig. 11 HEXAGON spacecraft draft with de-orbit SRMs71.

As for human spaceflight,SRMs also developed by Thiokol Chemical Corporation were used for de-orbiting the Mercury spacecraft. A design very similar to TE-M-236 (STAR-12)from the Corona programme was used. The Mercury capsule was fitted with 3 retrograde motors also called retrorockets(term not to be confused with retrorockets used for launch vehicle stage separation). Each SRM had a burn duration of 10 s. The motors were fired with 5 s intervals between the end of one firing and the initiation of the firing of the next one. Successful de-orbitation was possible even if one of the SRMs failed, due to the corrections of the automatic control system. Each of the SRMs had a mass of 30.1 kg and 5100 N of thrust.Their position in the vehicle is shown in Fig. 12.SRMs were protected from contamination and micrometeor damage due to the use of a metal housing. The SRM had a length of 368 mm and diameter of 305 mm. The de-orbit subsystem was jettisoned after performing the burn.The Mercury de-orbit motor is shown in Fig. 13(next to the Gemini SRM).Key SRM parameters are provided in Table 1.

Fig. 12 Bottom view of Mercury capsule (left) and Mercury retrograde rocket system (right)73.

Fig. 13 Mercury TE-316 (left)75 and Gemini TE-385 (right)77 used for de-orbiting manned capsules.

Table 1 Key data of Mercury and Gemini SRMs for capsule de-orbiting.

Similar SRMs of Thiokol Chemical Corporation were used in the Gemini spaceflight programme. The TE-385 was a derivative of the STAR-13 TE-M-485 orbit insertion motor.Four TE-385 motors were used for de-orbiting the capsule and fired in ripple fashion.Apart from this, the SRM could be used during a high altitude launch abort, separating the capsule from the launch vehicle. This however never took place.SRMs were used for mission de-orbit in ten manned Gemini flights in 1965 and 1966.Details on the Gemini TE-385 design can be found in Ref..Key motor parameters and its comparison to the Mercury TE-316 can be seen in Table 1.The motor had a star grain configuration and used a polysulfide-based propellant. A spherical titanium alloy combustion chamber was used. The SRM had a submerged nozzle configuration,the expansion cone was made from molded vitreous silica phenolic resin and the throat insert was made from high density Graph-I-Tite G-90 graphite.

While Mercury and Gemini capsules used SRMs for deorbiting manned spacecraft, future American designs did not include SRMs for de-orbiting manned capsules. Neither has this approach been utilised in Russian or Chinese vehicles(Soyuz and Shenzhou). Use of SRM for de-orbiting manned capsules is expected to remain only a historical solution.

As for newer designs, a wide range of SRMs is provided in the ATK SRM catalogue.The document features the STAR-3A motor being advertised as useful for de-orbitation missions.This motor thrust peaks at 800 N but burns only for 0.5 s,providing an effective specific impulse of 241.2 s. Such a short boost either generates too high accelerations for small satellites or requires clustering for bigger ones since its total impulse is equal to 284 N∙s. The STAR-4A on the other hand has an end-burning configuration,giving a burn duration of 10.3 s and a total impulse close to 2650 N∙s, while having a propellant mass fraction of 0.65 and delivered specific impulse of 269.4 s. The ATK catalogue also provides another SRM for minimum acceleration and extended burn duration delta-V impulse, which is the STAR-5A,which has a burn duration of 24.7 s and a total impulse of 6000 N∙s,providing an effective specific impulse of 250.8 s.This SRM has 2.3 kg of propellant and a propellant mass fraction close to 0.50.

Another example of using an American SRM for de-orbit is the Japanese concept of the Lunar-A penetrator,which was to use ATK Thiokol’s STAR-30BP motor to provide 84%of the required total deceleration delta-V.Fig. 14 presents the penetrator with the attached SRM for de-orbit.However, the Lunar-A project was never finalised.

6. Soviet and Russian de-orbit SRMs

Numerous Soviet and Russian photoreconnaissance missions utilised SRM for de-orbiting capsules, as well as satellites.The primary objective was to return exposed film to Earth.Three different systems have been used extensively: Zenit,Yantar, and Orlets. Their sketches are shown in Ref..

Fig.15presents the timeline of major technology utilisation.The maximum of de-orbits was achieved in the mid-1980’s at southern latitudes between 60° and 80°. The contribution of the Zenit retro-firings is dominant. Almost all firings took place between 150 and 400 km of altitude.Key SRM data are presented in the following paragraphs.Fig.16shows parts of Zenit,Yantar,and Orlets,which were recovered,using SRMs.

Fig. 15 Mass of released orbital dust and slag due to in-orbit firings of Soviet and Russian SRMs84.

Fig. 16 Zenit with the TTDU SRM (top), Yantar-2 K’s SpK with the 11D864 SRM (bottom left), and Orlets-1 capsule with TOR (17D712) (bottom right) (drawing of Orlets-1 film return system is an artist view) 84.

The de-orbit propulsion system of the Vostok used a liquid propellant engine of OKB-1,but the NII-125 design bureau in parallel developed an SRM alternative, the Vostok-Meteor kick stage. It was implemented in the Zenit satellite series(and flew on Zenit-4 M,4MT and 4MK).The SRM was used to de-orbit the whole 6000 kg satellite. The SRM was named TTDU and had a designation 11D82M. It is shown in Fig. 17. Ref.states that 508 in-orbit firings were completed. 11D82M had a thrust level of 30.7 kN, a propellant mass of 246 kg and a burn duration of 23 s.

Fig. 14 Lunar-A penetrator with its SRM for deceleration82.

Fig. 17 Zenit spacecraft’s TTDU SRM assembly88.

The fourth generation reconnaissance satellite Yantar-2K used 11D864 MKB Iskra SRMs for de-orbiting its two small capsules named SpK (Spuskayemaya Kapsula). Initial inspace testing of the capsules with the dedicated SRM took place on the Gektor spacecraft.The SRM is can be seen in Fig.18and its location on-board the Yantar spacecraft is visible in the left photograph. A toroidal gas reaction engine(8005D-0)was present around the SRM.It was used for stabilizing the capsule after braking and for angular velocity control.In total 250 in-orbit firings are known to have taken place.The SRM had a mass of 52 kg and a burn duration of 13 s at a nominal chamber pressure of 8 MPa. Being 450 mm long, it had a diameter of 380 mm. The de-orbited mass was 274 kg, including 29 kg of payload. The 11D864 had a thrust of 5.9 kN.

Fig. 19 TOR (17D712) propulsion module85.

Fig. 20 Impact of Soviet and Russian SRM firings on space debris84.

The third major SRM for de-orbit in Soviet programmes was used in Orlets capsules. 122 SRM retro-burns took place in the Orlets programme till 2009.Sixth generation photoreconnaissance satellites with return film capsules (Orlets-1) had 8 capsules, while the seventh generation of the spacecraft(Orlets-2) has 22 capsules.Each of the capsules uses a dedicated TOR (17D712) SRM for de-orbit, which has a toroidal shaped combustion chamber and utilises four nozzles.It is shown in Fig.19.The SRM length is 190 mm,outer diameter 350 mm and inner diameter 205 mm.Its mass is 17 kg,out of which 8 kg is the mass of the propellant.Ref.mentions that the 17D712,developed at the Federal State Unitary Enterprise Design Bureau ‘‘Arsenal”, could be regulated and deliver a precise impulse.

Fig. 18 Yantar-2 K spacecraft (left) and its descent capsule with the 11D864 SRM (right) (photos obtained with permission from Mr.Jakob Terweij)89.

Another example of SRM de-orbit technology application is the Mars 96 de-orbit retrorocket, which was to provide a de-orbit burn with a delta-V of 30 m/s to land the Mars soil penetrator.However, this has not been demonstrated due to a failure of the launch vehicle.

In summary, a large amount of the known in-space SRM firings are due to Soviet and Russian activity. Fig. 20, however, shows that these firings are not the major contributor to the orbital SRM dust. Russian retro-burns pose no ongoing long-term impact on the debris environment. This is mainly due to the fact that very low altitude firings were done,mainly in order to recover photoreconnaissance capsules. As for slag,the exhaust velocity was very low and the objects were expelled during the end of the burn only,when the orbit of the capsule was already significantly lowered. This occurred in near-tocircular orbits for slag released from the TTDU and 11D864 motors and resulted in immediate re-entry. Generated particle clouds are considered to have had orbital lifetimes of up to half a year.

7. Chinese de-orbit SRMs

China became the third nation to successfully recover its payload from space.This was achieved by the Fanhui Shi Weixing(FSW)photoreconnaissance spacecraft.Film capsules were de-orbited in a blunt re-entry vehicle of 1.5 m length,propelled by one SRM. Several versions of the FSW satellite were used in orbit between 1975 and 2005, while the first launch in 1974 was not successful due to the launch vehicle’s failure.Five series of FSW spacecraft were developed: FSW-0, FSW-1,FSW-2,FSW-3 and FSW-4.They used SRMs produced by China Hexi Chemical & Mechanical Company. This included: FG-14, FG-23 (also designated as SpaB/68 shown in Fig. 21) and FG-23 SRM variants.Several differences in the SRM designs included using different combustion chamber technology (steel or titanium) and different propellants(ranging from Carboxyl-Terminated-Polybutadiene (CTPB)-based to HTPB-based)and propellant mass.All motors utilise internal burning grain with a total impulse exceeding 520 kN∙s and later versions had increased performance. More on the FSW and its SRMs can be found in Refs.. In particular,details on FSW missions and their outcomes are discussed in Ref.. Typical perigee altitude was between 170 and 200 km and the apogee altitude between 350 and 400 km.The orbit inclination was equal to 63°-65°.Importantly,the re-entry system was fired very steeply downward from orbit.Ref.mentions that the SRM injected the capsule into a 179 km × 3031 km orbit with an inclination of 56°. This required a very large delta-V, significantly exceeding delta-Vs of similar Soviet and American missions, which are estimated as about 175 m/s.Ref.mentions a delta-V of 650 m/s–such a manoeuvre was used and ensured that the re-entry command could be given to the spacecraft while it was over China and the direct descent trajectory provided near certainty that the capsule would be recovered on Chinese territory. Since the FSW-2 vehicle configuration, the SRM was jettisoned after its firing in order to limit the mass necessary to be recovered later by parachute.Several FSW missions and the later evolution of the FSW spacecraft allowed its use as a microgravity platform. The version without film capsules (after the FSW programme) was called the SJ (Shi Jian). Experimental payload re-entry capability was maintained.

Fig. 21 FG-23 (SpaB/68) SRM96.

8. Revisiting idea of using SRMs for de-orbiting spacecraft

While multiple sources discuss the use of SRMs for SDM, the goal of this review is to collect current and historical developments worldwide in the field of SRMs for de-orbit and associated technologies in one paper. Solid propulsion for end-of-life disposal is considered, assuming use of a dedicated SRM system or an SRM cluster,implemented in a given spacecraft during its design and integration.While Active Debris Removal is also possible to be carried out by using solid propellant space-tugs,it is not considered widely for such applications due to the need of precise attitude and orbital control, while SRM provide a given total impulse. The topic of rendezvous with a non-cooperative,un-prepared,tumbling debris is a difficult field itself and is not discussed herein.Table 2presents a summary of advantages and disadvantages of using solid propulsion for EoL disposal of spacecraft. This analysis shows that dedicated SRMs for EoL disposal may be a good solution for SDM.

9. Requirements for SRMs for modern end-of-life de-orbitation

Key modern requirements for effective and safe use of SRM technology for de-orbitation include:

(1)Limiting the inert mass of the system and minimising the necessary SRM size.

Effect: use of high performance propellant and a high propellant mass fraction SRMs (efficient material selection - in particular low-regression-rate ablative insulation).

(2)Limiting the thrust level to limit spacecraft accelerations during SRM firing – maximum accelerations depend on the design of the spacecraft, however, for typical large platforms of European Large Satellite Integrators the limit is 0.04 g if large appendages are used– this leads to relatively low thrust of solid rocket motors and very demanding long firing durations. However, if the SRM is used just for the final burn during direct controlled de-orbitation,some break-up is possible at altitudes below 600 km, where debris residence time is below 25 years. As for smaller spacecraft, without delicate appendages, relatively large thrust may be possible – accelerations above 1 g are acceptable for typical CubeSats and for many other small satellites without foldable solar panels and antennas.

Table 2 Advantages and disadvantages of SRMs for EoL disposal.30,105–108

Effect: very low-burn-rate propellant and an end-burning configuration of the SRM (especially for spacecraft with appendages), use of clusters of SRMs in case of high total impulse requirements.

(3)Limiting generation of debris occurring due to the presence of solid particles in the SRM exhaust (both combustion products, as well as bits of eroded insulation and other parts of the system).

Effect: use of non-metalised propellants, SRM design shall utilise a conditioning and ignition system that does not generate debris during firing initiation, while internal thermal insulation technology shall prevent injection of particles into the exhaust, especially during burn tail-off.

(4) Ensuring storability and reliability of the motor.

Effect:well-known propellant type with good aging properties, immune to radiation and with a wide operating temperature range shall be used.

(5) Ensuring passivation of the spacecraft by limiting energy reservoirs and ensuring their depletion at end-of-life.

Effect: limited impact on the SRM due to its single-burn capability (however at system-level battery discharge and inhibitingpyrotechnicaldevicesisneededand de-pressurization of the Thrust Vector Control is required if a pressurized medium is utilised).

(6) Design for demise.

More on this topic can be found in publications showing results from recent ESA projects.

10. Propellant selection

One of the keys issues is to define the most promising propellant for SRMs for SDM. It is commonly assumed that it should be a composite propellant due to in-space heritage of such systems.Double base propellants are less attractive due to their lower specific impulse and larger challenges considering including them on-board spacecraft platforms,due to their higher sensitivity (Class 1.1. transportation hazard, which means that they can transfer from deflagration to detonation, unlike typical Class 1.3. composite propellants).Despite their shortcomings, use of double base propellants for EoL disposal is discussed in Ref.. However, composite propellants allow obtaining significantly higher specific impulse (and density specific impulse) and a very wide range of burn rates (from over 50 mm/s to single mm/s). Therefore composite propellants shall be discussed within this paper.

10.1. Oxidizer and energetic material selection

The preferred oxidizer for dedicated composite propellants is ammonium perchlorate,due to its heritage,good performance and storability.ADN propellants deliver high specific impulse,but have very high burn rates,what eliminates their use in most SDM applications,despite of utilisation of ADN-based propellants in some upper stage SRMs of Soviet and Russian vehicles. Alternative oxidizers, such as HNF are not characterised well and they have little heritage,even regarding on-ground applications.HNF is considered hazardous and most research involving its use has been stopped. Several other propellants using novel oxidizers are at the laboratory stage of development.Some propellants are based on CL-20(Hexanitrohexaazaisowurtzitane,HNIW),which is a missile-proven technology,but its costs are high since most commercially manufactured batches are small.Other explosives as 1,3,5,7-tetranitro-1,3,5,7-tetraazacyclooc tane(High Melting Explosive,HMX)and 1,3,5-trinitrohexahy dro-sym-triazine(Research Department Explosive,RDX)may also be used.Although,while contributing up to 10%of the propellant mass they do not change the propellant safety classification, use of explosives is avoided for in-space propulsion. Other well-known oxidizers have limited use due to lack of performance. For example ammonium nitrate could be an interesting oxidizer allowing to limit burn rates,but its performance and issues with storability make it less attractive,although its thermal stabilisation is possible.No information in open literature was found regarding use of non-AP propellants for SDM apart from preliminary concepts.AP remains the baseline oxidizer due to its prevalence and novel oxidizers lack industrial-size facilities,what is linked to the evaluation of the economic impact regarding the potential necessary initial investment required for use of several novel compositions.

In order to obtain high performance in terms of specific impulse, despite using no energetic additives and no metal powder or hydrates, very high AP loadings must be used.The size of AP particles are one of the key parameters effecting overall propellant burn rate.Compositions with bimodaland trimodalAP are advantageous for propellant packing and obtaining better combustion stability. However,very fine AP fractions must be limited not to increase burn rate. It is notable that well-tuned bimodal compositions can give lower burn rates than unimodal ones.There is, however,an upper limit on AP content - due to increasing the viscosity of the propellant slurry during casting, bubble contamination of grains occurs and the burn rate cannot be controlled. In practice the maximum AP mass fraction for bimodal AP is about 86%.For relatively coarse AP, which allows limiting the burn rate, the maximum AP mass fraction is even lower and it furtherly decreases in case of the necessity to include burn rate augmentation compounds.

10.2. Fuel and binder system

Most importantly, to limit condensable combustion products, metal powders shall not be used in modern SRM for SDM. This is discussed in the following paragraphs of this paper. Similarly, several other compounds commonly used for instability suppression and damping, due to their solid phase, should be limited. As for binder systems, they have impact on propellant storability and mechanical properties.Expected mechanical properties should be optimized for end-burning grains, since this geometry shall be dominant for most spacecraft. Typical propellants based on HTPB binders are promisingand storability data is available regarding AP/HTPB propellant shelf life.However new developments do not have to be limited to this binder system. Several alternative ones may actually enable a decrease of the propellant burn rate.Typically hydrocarbon burn rates decrease for compounds with higher numbers of as in a molecule.Higher performance binders, such as Glycidyl Azide Polymer (GAP), may also be of interest for future developments,as well as lower burn rate ones,including 3,3-bis(azidomethyl)oxetane (BAMO).This also considers use of advanced binders with explosives.More on modern energetic binders can be found in Refs.. However, in-space use of explosive materials is expected not to be widely considered by satellite integrators,even in case of specific impulse gains.

10.3. Burn rate suppressants

While most rocket application require high propellant burn rates, several sources describe various additives for standard AP-based composite propellants, enabling obtaining lowburn-rates, which are required for efficient deorbiting.Compounds enabling decreasing propellant burn rates are called suppressants, moderators or inhibitors. A detailed review of such compounds for SDM has been conducted in the recent project ‘‘Pre-Qualification of Aluminium-Free Solid Propellant” of ESA, where over 60 burn rate suppressants have been listed.Among the most promising burn rate suppressant is oxamide (CHNO),1which is one of the most commonly used burn rate suppressants in missiles. Oxamide is also used as an additive to low-regression liners and thermal insulation materials.Other effective propellant additives for burn rate moderation include lithium fluoride,titanium oxide,melamineand several ammonium salts.Particular impact of the selected compound depends on the moderator concentration,surface area and aggregation,as in case of catalysts.Different mechanism of decreasing burn rates are described in Refs..These include cooling the combustion process and propellant heating,as well as directly affecting chemical kinetics of AP decomposition. More on AP decomposition can be found in Refs.. However, some suppressants influence not only AP thermal decomposition, but also binder system decomposition.More burn rate suppressants can be found in Ref., where 117 compounds, with reported research concerning achieving low burn rates,are listed.Other references include Refs..

11. Modern developments of SRMs for end-of-life spacecraft disposal

In Europe several studies and developments concerning the use of SRM for de-orbitation have been on-going for a number of years. The SPADES (Solid Propellant Autonomous DE-orbit System) concept developed at ESA is the workhorse of development in this area in Europe. Previous research was published since the beginning of the 1990s by representatives of Schonenborg Space Engineering B.V. and Scho¨yer Consultancy B.V. Their concept is based on the idea of including SRMs on board of S/C at their design and integration phase,ensuring an autonomous, highly reliable, storable propulsion system providing the source for the delta-V needed at S/C end of life.

In 2013 Dutch entities: Innovative Solutions in Space B.V,Aerospace Propulsion Products B.V. and Netherlands Organisation for Applied Scientific Research (TNO) proposed an SRM for NanoSat de-orbitation. A Nanosatellite Kick Stage(NKS) was planned to de-orbit 3-unit CubeSat class satellites in a predictable manner from a 1000 km orbit.With a total impulse of 600 N∙s and approximately 180 N of thrust,a 3.9 kg CubeSat would find itself under 5g acceleration. Ref.presents pictures from the prototype ground firing. The project never advanced to higher Technology Readiness Levels(TRLs).

With these words he continued his way, and very soon met the third knave13, who said to him, God bless you, sir; are you by any chance coming from the market? Yes, I am, replied Simon

Outside Europe, also in 2013, the American Ad Astra Rocket Company proposed a Low Earth Orbit Large Debris Removal system, which would enable de-orbitation of inter alia spent Zenit rocket upper stages. An active space tug propelled by advanced VASIMR and solar electric propulsion systems would enable the attachment of SRMs to large objects being already in orbit. Ad Astra’s idea assumes igniting an SRM at an altitude of 400 km,enabling controlled re-entry.

In 2016 the Los Alamos National Laboratory published research on an innovative propulsion system that could serve as a kick-stage for small satellites, as well as it could be used for de-orbiting application.The original SRM configuration assumed using a separate fuel grain and a separate oxidizer grain.

Fig.22 Pacific Scientific’s SRM and an example SRM cluster for in-space applications181.

Fig. 23 Pacific Scientific’s PACSCISAT and its MAPS module demonstrator183.

A newer concept,presented in 2017 by the Pacific Scientific Energetic Materials Company,is a configuration dedicated to small satellites utilising dozens of small SRM in a cluster,which would be implemented as part of the S/C. It is shown in Fig.22.The system allows numerous in-orbit manoeuvres including final de-orbitation. While the presented version of the system consists of over 170 SRMs, the manufacturer underlines ability to maintain very high overall reliability and as described in Pacific Scientific Energetic Materials Company’s patent,up to 1000 SRMs can be effectively integrated in one system.MAPS heritage includes technology development for the Exo Kill Vehicle where extremely precise sequencing of SRMs was achieved.The company declares the use of a clean-burning propellant with a specific impulse of 210 s in vacuum, what is slightly below state-of-the-art hydrazine monopropellant thrusters.The use of MAPS has been demonstrated via orbit raising of the 3-unit PACSCISAT satellite in 2017, which included four small SRMs - two firings of two different motor pairs was carried out.The satellite’s mechanical configuration and the SRM assembly are shown in Fig.23.The company declares that the system is compatible with many other small satellites and has a lifetime of over 10 years with three independent inhibits against unwanted motor firing. MAPS is capable of delivering a delta-V of over 50 m/s for a 1-unit satellite.Further information on the propulsion system can be found in Refs..

Fig. 24 D-SAT SRM108.

Another example of use of SRMs for SDM is provided by D-Orbit from Italy. The company declares the will to sell autonomous de-orbit systems. However, it is not directly involved in developing the SRMs themselves.The company cooperates with Bayern Chemie,from Germany,regarding the propellant and SRM to be used. Initial work of D-Orbit and Bayern Chemie on the SRM assumed a propellant with SiOinclusions for burn stability. The cooperation led to the DSAT 3 Unit CubeSat.The solid rocket motor had a burn time of 3.2 s and was to deliver 375 N of thrust and 836 N∙s of total impulse.The SRM total mass was 0.9 kg with a 0.33 propellant mass fraction and its envelope slightly exceeded the 1-unit standard. It is shown in Fig. 24. While a case-bonded propellant grain would allow obtaining a better propellant mass fraction, a cartridge grain enabled assembling the SRM at the launch complex.The maximum expected operating pressure was 7 MPa, while the nominal value was 5 MPa and a specific impulse of 266 s was to be achieved. A nozzle area ratio of 20 is quoted.The 2018 in-orbit demonstration was carried out, however D-Orbit elevated the orbit instead of decreasing it.Early SRM system versions shown to the public did not meet ESA maximum acceleration requirements,however the final product is to be also useful for satellites with appendages.The Bayern Chemie propellant utilised is from the RESI (REduced SIgnature composite propellant) family. The company states that one of the RESI propellants was aged for 10.5 years at 62 °C, what is equivalent to over 100 years of storage in room temperature. The only property degradation issue concerned strain-capability at low temperatures.However, an official shelf life between 12.5 and 16 years was stated. As for the specific RESI propellant used in D-SAT, it has a burn rate of 12.5 mm/s at 10 MPa and a pressure exponent of 0.43.Its operational temperature range was-30°C to+71°C(while qualification took place at-34°C).Irradiation tests were done on a propellant simulant with no oxidiser.More data on RESI propellants is provided in Ref..

In early 2016 the European Commission began the TeSeR(Technology for Self-Removal of Spacecraft)project with Airbus Defence and Space being the leader. The goal of the project was to develop on-ground prototypes of spacecraft EoL disposal systems. Among three different removal subsystems one was based on solid propulsion.However only an inert(excluding real propellant) configuration was builtand the project finished in 2019. Work allowed designing the system including: the SRM with a safe-and-arm device and ignition control electronics. Basing on the outcome D-Orbit included various versions of the D3 (De-orbit Decommissioning Device)subsystem in its products line.A newer product of D-Orbit under development is the Fenix small propulsion module for CubeSats (20 mm of diameter and under 100 mm of one SRM). The baseline configuration uses four motors of this type,situated along the vertical edges of a 1-unit CubeSat frame.

Other developments in Europe, following the increasing impact of the ESA Clean Space,were stimulated by small contracts awarded by ESA in 2016 within the‘‘CleanSat:Technology assessment and Concurrent engineering in support of LEO platform evolutions” project.The multidisciplinary study was supported by European Large Satellite Integrators.Numerous SDM technology building blocks were assessed using the ESA concurrent engineering environment,including use of solid propulsion for de-orbit. Two concepts were dedicated to SRMs(by Institute of Aviation from Poland and Nammo from Norway)and two were dedicated to Autonomous De-orbit Systems, which may use SRMs (by D-Orbit from Italy and GMV from Spain).Fig. 25,presents the concept of Nammo. Meanwhile the initial phase of ESAsupervised TVC development for SRM for SDM was ongoing and was done by a Swiss-Italian consortium combining Almatech, D-Orbit and Alta Space. A small study regarding use of SRMs for nanosats was also carried out in Norway, funded by the Norwegian Space Agency.

Fig. 25 Nammo’s SRM for small-sat de-orbiting (left) and example small-sat visualization (right)196.

Fig. 26 Polish SRM for de-orbit (left) and subscale SRM atmospheric firing (right)105.

Fig. 27 ESA-supported SRM design as of 2020197.

After the ‘‘CleanSat: Technology assessment and Concurrent engineering in support of LEO platform evolutions”ESA initiated a more detailed engineering project ‘‘Pre-Qualification of Aluminium-Free Solid Propellant”, which was completed in Poland and allowed obtaining unprecedented results including confirming radiation-resistance and storability of high performance low-burn-rate solid propellants.The propellant based on AP, HTPB and oxamide completed pre-qualification and obtained an ESA TRL of 6.The continuation is ongoing with further development, via a consecutive contract in ESA General Support Technology Programme (GSTP), titled ”Solid Propellant Engineering Model Development”, where the SRM technology is going to be increased to TRL = 5. Further SRM qualification and flight model development is planned. An in-orbit-demonstration would allow increasing the community’s awareness of the technology’s capabilities and prove technology readiness.Fig. 26,presents the Polish SRM design and successful ground firing of prototype using the newly developed dedicated propellant and Fig. 27,presents the design optimised as of 2020 for 1.5 t satellites, All efforts are done in-line with ESA and European Large Satellite Intergrators’ requirements in mind.

12. Thrust vector control of SRMs for EoL de-orbit

D-Orbit’s issue during its in-orbit demonstration of its SRM prototype was due to an attitude control problem and lack of TVC.Thrust vector control has been identified by European primes as the key system-level challenge to be solved before introducing dedicated SRMs for EoL disposal to a wider range of spacecraft.While an alternative is to use spin stabilisation of the satellite during the final burn, this may require spin-up thrusters and is not possible for each spacecraft due to i.e. appendages. In case of Iridium satellites,mentions that a maximum allowable angular acceleration of 0.42 rad/sand an angular velocity of 0.7 rad/s is expected to ensure spin-stabilisation of the satellite. Requisite spin-up may be achieved by using small SRMs if no thrusters are onboard or if early passivation of the main liquid propulsion system is needed. Spin-stabilisation can be also achieved by adequate geometry of the exhaust nozzle. This approach may be attractive for nanosatellites,where minimal system complexity is needed. However, caution is needed regarding nutation instability, also called the PAM-D coning anomaly (named due to its discovery during flights of the American Payload Assist Module using a STAR-48 SRM).This instability may occur after motor ignition and is visible via a lateral angular wobble, occurring despite gyroscopic effects, as shown in Fig. 28. This wobble may remain post the SRM firing in case of spin stabilisation.Naturally its occurrence is dependent on the mass distribution and mass change during propellant burn out. A recent studyprovides a major outcome showing that use of end-burning solid rocket motors allows gradual vehicle stabilization, unlike in case of radial-burning propellant grains.Therefore,in case of typically envisaged SRMs for de-orbit, spin stabilisation shall be considered and may not require additional nutation control subsystems.In case spin-stabilisation is not possible and the existing attitude control system is not able to provide adequate torques to counteract aerodynamic effects,which grow with decreasing altitudes, TVC shall be necessary. The ESA SPADES study mentions two technical solutions, shown in Fig. 29. In case of Pacific Scientific Energetic Materials Company’s MAPS,the array of small SRMs allows for firings of adequate charges in order to conduct planned manoeuvres. A combination of using reaction wheels and SRM firings is utilised. Ref.states that this way all necessary orbital adjustments can be achieved, including effective attitude control and higher delta-V manoeuvres.

Fig. 28 Nutation instability of spin-stabilised vehicles using SRM systems198.

Later ESA projects led to a larger number of concepts from industry including Almatech,Nammo and Warsaw Institute of Aviation.Fig.30presents two SRM TVC configurations presented during the ESA‘‘CleanSat:Technology assessment and Concurrent engineering in support of LEO platform evolutions” project. Gimballing of the nozzle downstream of the throat (supersonic split-line technology), gimballing of the whole nozzle, as well as gimballing of the whole SRM were considered.

Moreover,a trade-off of a number of further solutions was conducted at Lukasiewicz Research Network - Institute of Aviation (Fig. 31). Some concepts included components which could also be used for closing the SRM nozzle outlet,thus enabling SRM conditioning with a neutral gas without the need of introducing a nozzle bladder(which is problematic in terms of opening without releasing additional debris).In the trade-off mentioned two TVC solutions were selected for further analyses – the gimballed one from Fig. 30 (left) and one using external nozzle flaps from Fig. 31 (bottom right). This topic has been also covered in the ESA ‘‘Pre-Qualification of Aluminium-Free Solid Propellant”, where initial system-level aspects have been analysed. The problem of TVC and spacecraft dynamic behavior has been studied in Ref.. This also includes analysis of TVC use in case of firing two SRMs at one time,while due to uneven satellite heating one motor may have a higher propellant burn rate.

An ESA activity concerning dedicated TVC has been completed by a consortium led by Almatech.A wide range of technical solutions for TVC has been considered.Design activities led to a new patent application considering a frictionless flex-gimbal solution.As in one of the Polish concepts, the whole SRM is to be gimballed, what is an unprecedented approach for SRMs. Almatech’s solution allows for a nonlubricated design and has been shortlisted by its consortium.The design is shown in Fig. 32. Importantly, earlier work assumed developing TVC with nozzle exit vanes, as successfully used in several missiles worldwide. However, issues with testing were acknowledged and no sufficiently representative cold-flow test method was identified.This is due to a very high envisaged expansion ratio of the SRM nozzle. Use of gas cold-flows would lead to having condensation at the nozzle exit and high-mass flow tests lead to very demanding vacuum facility requirements. With vanes located at the SRM nozzle exit, it is very difficult to provide representative testing conditions.Hot firings of SRMs with vanes could be carried out in a high-performance vacuum facility, however vacuum testing of SRM is very rare and only several adequate facilities exist worldwide.

Fig. 29 ESA SPADES concepts for thrust vector control systems: with deflecting nozzles (left) and vanes in nozzle exhaust (right)30.

Fig. 30 ESA CleanSat study TVC concepts using gimballing27.

Fig. 31 TVC concepts for preliminary trade-off at Warsaw Institute of Aviation41.

Fig. 32 Frictionless Flex-Gimbal design202.

Recent ESA advancements with other building blocks of the ultimate de-orbit system, such as the dedicated propellant and long-burn-duration SRM technology, show that there is need for rapid TRL increase of TVC technology for SRMs.Naturally, the main driver is commercial interest of satellite integrators due to legal requirements concerning SDM.In late 2020, a new ESA activity titled ‘‘Solid Propellant Rocket Motor Thrust Deflection System”has been initiated in Poland.This contract, being the forth ESA activity in Poland within this topic is to deliver TVC technology to TRL = 5+. Latest Polish, Norwegian, Italian and Swiss experience and feedback from European primes show that a robust,low-cost TVC solution is needed, which would allow minimisation of spacecraft architecture modifications. While gimballing the whole SRM may lead to increased system envelopes, use of components in the SRM outflow, such as flaps or vanes, is also difficult due to challenges with representative vacuum testing conditions. Different providers may decide to settle upon different TVC technologies. Regardless of final technical trade-off results, in-orbit demonstrations shall be beneficial for the development, including development costs, and may be required for some of the solutions due to the mentioned issues with full on-ground validation.

13. Challenges in development of modern SRM systems for spacecraft de-orbit

Typical design procedures for SRMs can be found in Refs.. However, some SDM requirements lead to specific technical difficulties. One of the first trade-offs is the propellant type. Combining performance without using metal additives and low-burn rate is a major challenge itself. On the other hand, propellants with no aluminum do not exhibitloss of performance due to aging (no AlOgeneration occurs during aging). Lack of AlOin the combustion products decreases mechanical erosion of system components such as:the nozzle, including the throat region, and the TVC subsystem. However, high-AP-loading leads to intense oxidation and chemical erosion within the SRM. A non-eroding nozzle insert is needed for high performance, while refractory and other non-demisable materials should have limited use in motors used for spacecraft EoL disposal.Long burn durations cause increased heat losses, which decrease overall system performance. Long burn durations also require using endburning configurations with significant mass of thermal insulation, what decreases overall system delta-V capabilities. Very favourable (in comparison to liquid propulsion) system dry mass estimations from Ref.may be slightly optimistic if long burn durations are required. High system-level performance requires maximising the propellant mass fraction and advanced ablative insulations must be used,possibly with variable thickness along the SRM axis. Historical developments show the possibility of grain coning throughout the burn and that difficulties in reaching the desired thrust characteristic may occur.Low chamber pressure is expected to be beneficial,as in most in-orbit propulsion systems,in terms of maintaining low structural mass of the SRM and in order to limit insulation regression and the propellant burn rate. Very thin combustion chamber wall thickness may not be possible due to manufacturing challenges for high elongation motors and structural oversizing may need to be done,also in order to deal with machining,integration and handling loads.Low chamber pressure may also be a disadvantage since each propellant combination has its own low pressure limit of combustion.For some formulations low pressure ignition and steady combustion may be a challenge.Lack of solid particles in the combustion products may lead to issues with SRM burn stability.It is however possible to have stable combustion using nonmetalised propellants.While negligible amounts of solid and condensable combustion products lead to acceptable ultra-fine particles in the exhaust,attention must also be given to other motor components. A dedicated, fully combustible,thermal insulation shall be used and the nozzle burst-disc must not generate additional debris.Similar work must be also performed on optimisation of igniter subsystems in order to minimise solid particle generation.The ESA IPOL requirement of debris below 1 mm of size is valid(however combustion products are to be two-orders of magnitude smaller).Another challenge is connected to SRM in-orbit storability. Long-duration radiation exposure must not pose a threat not only to the propellant itself,but also to SRM components such as sealings and pyrotechnics.Recent data on the impact of space radiation on solid propellants has been presented by Nowakowski in Ref.and Caffrey et al. in Ref.. While, the results show that radiation shall not be an issue for most cases,the combustion chamber wall may be treated as potential shielding if needed. Thermal cycling, especially significant in terms of LEO applications, must also be considered. A challenge itself is SRM testing and qualification.A limited number of facilities worldwide enable vacuum testing of SRMs. While qualification does not have to require vacuum testing, some TVC systems may need to be tested in vacuum chambers,depending on the physical phenomena utilised. Attitude control of spacecraft is a complex task and in case of introducing a solid propulsion system, which has relatively high thrust in comparison to other technical solutions, applicable Attitude Control System (ACS) margins or dedicated TVC are needed.Naturally, for many spacecraft spin stabilization is also possible, as has been historically done.

Table 3 Technical maturity of key technologies.

Fig. 33 The 15D161 SRM213.

While several challenges occurring during the development of dedicated SRM systems for de-orbit were shown herein,available modern SRM technologies allow meeting requirements. Example solutions were discussed by Schonenborg and Schoyerand Nowakowski et al.and several on-going research and development activities worldwide address key issues regarding effective use of SRMs for EoL disposal.

14. Technical maturity

In order to summarize the status of SRM development for propulsive EoL spacecraft disposal,a brief review on the technical status of key technologies has been done. It is briefly encapsulated in Table 3.While SRM technology is well established, new requirements concerning system performance and SDM issues lead to necessary further development work in this field. SRMs of relatively similar design (end burning grain,very long firing duration) are present in several sustainer rocket motors used in i.e. air-to-air missiles. Larger SRMs of this type were utilised in ICBMs for propelling nuclear warhead reentry vehicles.Systems of Yuzhnoye used nonmetalised propellants in case-bonded grain configurations with SRM burn durations up to 500 s.Fig. 33,presents Yuzhnoye’s 15D161 rocket motor used in the 15A14 (S-18) ICBM.Another historical example is the American Propelled Decoy motor called the Five-minute Rocket Motor,which had a burn duration exceeding 300 s (although the low burn rate baseline propellant used aluminum powder).More recent developments include end burning SRMs with low burn rate AP/HTPB propellants for drone application.

Numerous development activities could be proposed regarding even further technology development than listed in Table 3. Next generation systems could include use of non-AP/HTPB solid propellants. Moreover, some of the largest drawbacks of conventional SRMs, thus no restart-ability and no thrust termination capability, could be tackled. Appropriate technologies for small and medium size SRMs existand are implemented into military designs.In case of dedicated SRMs for de-orbit, at this stage of development the goal is to have them demise during spacecraft re-entry. If payload ground recovery would be the goal, SRM recovery and reusability after refurbishing the thermal insulation and repeated propellant casting could be the long term goal. This however, is not considered within this review due to multiple challenges to be faced before this future phase.Other examples of further work include developing de-orbit kits with full autonomy, by including power supply, control, a simple onboard computer etc. in de-orbit kits, regardless of the systems already present on the satellite platform. This would allow nearly full spacecraft passivation, with only the de-orbit kit remaining active during the de-orbit manoeuvre.

15. Conclusions

Use of solid rocket motors for SDM is considered one of the most propitious solutions among high-thrust propulsive SDM methods.While the first use of solid propellants in space and for de-orbit applications goes back to the very beginning of the space-age, their utilisation was continued throughout the decades till the 21st century, mainly on-board American,Soviet and Russian spacecraft.Most applications since the late 1950s were dedicated to recover payloads (manned and unmanned capsules) and the peak use of SRM for de-orbit occurred in the mid-1980’s. In the 1990’s Schonenborg and Scho¨yer identified the potential of SRMs for Space Debris Mitigation and this was one of the catalysts for the development of the concept - after 2012 global interest in commercial applications in this field has been identified.The following conclusions regarding use of SRM technology for EoL propulsive disposal of spacecraft can be drawn:

(1) Initial applications of SRMs for EoL disposal occurred due to re-entry needs, not due to SDM awareness.

(2) Despite the fact that Soviet and Russian applications were a major part of registered in-space uses of SRMs,their impact on the current space debris environment is minimal due to their use on very low orbits.More significant impact on the number of orbital debris is due to use of SRMs as kick-stages and upper-stage propulsion for launch vehicles.

(3) Despite significant heritage in SRM retro-firings, SRMs for de-orbit have been in their nascent stages until the early years of the 21st century, if modern SDM requirements are considered.This is due to the extensive,up-todate, use of solid propellants with aluminium powder,generating significant amounts of AlOand other SRM-related debris. Typical small SRMs do not ensure long burn durations and limiting acceleration levels for spacecraft with appendages. The use of small SRMs with burn durations in the order of minutes has been historically very limited. The only relevant applications known are Soviet SRMs used for propelling nuclear warheads and several experimental designs, hardly used operationally. Naturally, relevant technologies are also present in sustainer SRMs used in missiles. However,burn durations above 60 s are rare and combining clusters of such motors is unprecedented.As for propellants,dedicated solid propellants for de-orbit have not been available until recently. Modern advances led to the development of low burn rate propellants with high performance.Most technically relevant solutions are unmetalised composite propellants used for missile sustainer propulsion. Both case-bonded and free-standing grain technologies can be utilised and technically justified.Supporting cases, proving that the technical maturity is either present or can be achieved in the next few years has been shown.

(4) While up-to-date progress on TVC for SRMs for deorbiting spacecraft is not mature yet, major advances are expected in the 2020s,leading to operational systems.Heritage is also largely-based on missile technologies.

(5) Since heritage of SRMs for de-orbit is relevant for very short-duration reconnaissance and human spaceflight missions, thus little data on long-term in-space storability is available in open literature. However, recent advances show good resistance to radiation and no show stoppers regarding use of AP/HTPB propellants in deorbit SRMs have been identified.In fact,dedicated nonmetalised AP/HTPB propellants are considered also safe regarding potential compliance with possible future international regulations (both in terms of SDM and limiting use of toxic propellants).

(6) In the past few years several European companies initiated developments of solid rocket motors for de-orbit with the goal of introducing them to satellite platforms.However up to date no existing solutions meet requirements concerning acceleration limits (0.04g for spacecraft with deployable appendages) given by ESA and European Large Satellite Integrators.It can be seen that a reliable, high performance and low-acceleration deorbitation motor has yet to be developed. The creation of the ESA Clean Space initiative led to work of several entities in this field.

(7) While numerous advantages of SRM systems have been identified, challenges connected with the development and use of solid propulsion for SDM have been listed.Key challenges include attitude control during the deorbit manoeuvre and reaching high propellant mass fractions for very long SRM burn durations.

(8) While low-acceleration SRM systems,and in some cases SRM clusters, are needed for spacecraft with large appendages, small satellites, especially nanosatellites,may enable conducting de-orbit manoeuvres with accelerations even exceeding 1g, thus significantly reducing SRM development challenges. Such systems will have considerable higher propellant mass fractions than systems of the same size, which have long burn durations.Therefore, it may be expected that small satellite applications of SRM de-orbit technology will serve as an important direction of de-orbit technology growth, next to developments for larger platforms within ESA and other international programmes.

(9) Currently three countries worldwide are leading major SRM developments for SDM.This includes commercial initiatives (system-level work in Italy and in the United States) and ESA projects (SRM development projects in Poland).Consecutive in-orbit demonstrations of relevant technologies are expected between 2022 and 2027.The outlook on further SRM developments and utilisation for SDM looks promising. No showstoppers have been identified.

(10) The key step regarding wide SRM technology application for de-orbitt is matching satellite integrators needs and industry flexibility is essential to adapt current spacecraft platforms’ mechanical configurations to integrate SRM within them. Use of SRM for uncontrolled,semi-controlled and controlled de-orbit is possible.They can be used as fully autonomous modules or in combination with other on-board propulsion systems. The NewSpace market is still shaping, thus the future of SRM for de-orbit will depend on the final commercial appeal of the technology.

(11) Follow-up work should explore the impact of recent and expected advances in liquid and hybrid propulsion on the de-orbit propulsion system trade-off. This should also include an updated cost vs. reward analysis. The range of missions where high thrust manoeuvres are essential and will not be replaced by electric propulsion firings should be defined to allow market size estimation.

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.

This work is based on results of activities led by the Lukasiewicz ResearchNetwork-Instituteof Aviation.Itwas prepared within the ‘‘Space Research Development” project. The project is financed by the Polish National Agency for Academic Exchange(No. PPI/APM/2018/1/00032/U/001). Work presented from projects of ESA includes work done during preliminary phases of ASPro (Pre-Qualification of Aluminium-Free Solid Propellant) and SPRODEM (Solid Propellant De-orbit Motor Engineering Model Development) projects funded within the General Support Technology Programme(GSTP).The author would like to thank all of the members of the Team who committed to the development of SDM technology and would also like to show his deepest gratitude to ESA Clean Space and all ESA technical and contractual Officers involved.Moreover,cooperation with various entities working in the field of Space Debris Mitigation, as well as individuals collecting space-related data is highly appreciated. Fruitful cooperation allowed sharing unique figures and data for the purpose of this review.