Content uploaded by David L. Carroll
Author content
All content in this area was uploaded by David L. Carroll on Feb 16, 2018
Content may be subject to copyright.
DISTRIBUTION A: Approved for public release, distribution unlimited.
Work funded by AFRL/RQRS on contract FA9300-15-M-1003, Public Affairs Clearance Number 16504.
© Copyright by authors. Paper Tracking #4935
DEVELOPMENT OF
H2O2-BASED MONOPROPELLANT PROPULSION UNIT FOR CUBESATS (MPUC)
D. King, C. Woodruff, R. Burton, and D. Carroll
CU Aerospace, LLC
Champaign, IL
ABSTRACT
CU Aerospace recently completed an Air Force Phase I SBIR to test a proof-of-principle
Monopropellant Propulsion Unit for CubeSats (MPUC), consisting of a thrust chamber and demonstrating
complete catalyzed combustion of an H2O2-based propellant denoted as CMP-8. CMP-8 has zero
toxicity and no special measures are required for its long-term storage. The propellant was subjected to a
scaled UN Series 1 detonation test series and demonstrated no detonation propagation when confined
under a charge of high explosive. Potentiometric titrations with standardized sodium bisulfite
demonstrated no degradation in the CMP-8 over a four-month, room-temperature storage period. Thrust
stand tests achieved a thrust level of >100mN at Isp >183 s with an average input power of ~3 W, for hot
fire runs typically spanning >10 minutes. A single run of greater than one hour was also demonstrated. A
trade study was performed of CMP-8 and its nearest competitors. A number of operational metrics and
issues were examined, and while other propellants have a minor advantage in Isp, the MPUC propellant
has advantages in availability, cost, lower flame temperature (less thermal management and radiation
losses), lower pre-heat temperature, lower viscosity, and low thruster materials costs. MPUC designs
comprise a complete propulsion system technology for CubeSats and other small satellites, with a high
performance, nontoxic monopropellant that possesses benign storage characteristics. The conceptual
system also provides cold-gas attitude control and projects >1200 N-s/liter of volumetric impulse.
INTRODUCTION
Commercial interest in very small satellites continues to grow. In the 1-50 kg satellite sector,
launches have shifted from a fairly balanced distribution between civil, government, commercial, and
defense (2009-2015) to a distribution dominated by commercial interests [Doncaster, 2016]. Moving
forward, it is more important than ever that these satellites have access to propulsion systems to extend
their asset time on orbit. Such a system must occupy minimal bus volume and carry a high product of
propellant density times Isp. Avoiding the use of toxic propellants such as hydrazine that significantly
complicate the storage and handling of the propulsion system is also desirable [Hargus, 2010; Singleton,
2013]. While AF-M315E is a green monopropellant, it has a high flame temperature of 1800 ºC requiring
a thrust chamber constructed from refractory materials, which adds further expense and complication. A
survey of available propellants motivated the search for a less exotic and more easily sourced chemistry
to be utilized in these small satellite propulsion systems.
The authors sought to develop a new monopropellant from non-toxic, readily available reagents.
The candidate needed a flame temperature low enough to avoid refractory construction, while maintaining
high enough specific impulse and density to be competitive with legacy propellants. The chosen
propellant has heritage in bipropellant thrusters [Woschnak, 2013, Wieling, 2012] and as a US Navy Mark
16 torpedo propellant [Clark, 1972], but has not been developed for a thruster in the 100 mN class. A
variant has even been used to drive a turbine-powered Volkswagen Beetle [Delchev, 1987].
DISTRIBUTION A: Approved for public release, distribution unlimited.
2
Risk reduction (both in the laboratory and for the end product on the launch range) was
paramount in the propellant development program. Desirable monopropellant safety properties were
described by Hawkins, et al. [Hawkins, 2010] and include high thermal stability, low unconfined ignition
explosive response, low impact sensitivity, low friction sensitivity, low detonability, insensitive adiabatic
compression, low electrostatic discharge sensitivity, and low vapor toxicity. To that end, a considerable
effort was undertaken to ensure that the propellant conformed to detonability, ignition, and storability
behavior guidelines set to ensure user safety.
RESULTS AND DISCUSSION
A search of potential fuel formulations yielded a mixture of 50% (w/w) hydrogen peroxide and one
of several alcohols (tert-butyl alcohol (TBA), 2-propanol, ethanol, methanol and glycerol). Each of these
systems requires a catalytic element to enable combustion. Glycerol was favored early in the program for
its high product of density and maximum theoretical specific impulse. However, it soon became evident
that detonation concerns would seriously reduce its utility. Ternary detonability plots were gathered
(where available) for the candidate fuels, Figs. 1a-d [Shanley, 1947; Shreck, 2004].
a. Ethanol b. Tert butyl alcohol
c. 2-propanol d. Glycerol
Figure 1: Ternary plots for H2O2 / alcohol mixtures (the detonation region lies within the red shaded region).
Note the stoichiometric mixture denoted as a yellow line.
DISTRIBUTION A: Approved for public release, distribution unlimited.
3
Figure 2: Theoretical maximum specific
impulse of candidate MPUC fuels. All
fuels except ethanol were eliminated
because their stoichiometric mixtures
are detonable when mixed with 50%
H2O2, denoted here by diagonal striped
columns.
Adiabatic flame temperatures were computed for each of the candidate mixtures, assuming
complete stoichiometric combustion with all reactions going to water and carbon dioxide. Performance
estimates were then made for theoretical maximum thruster specific impulse Isp, Fig. 2. Safety concerns,
reagent availability, and performance levels led to the ultimate choice of CMP-8, a mixture of 10.2%
ethanol, 45% H2O2, and 44.8% H2O. With its flame temperature of just 1220 ºC, CMP-8 allows MPUC to
be constructed from common stainless steels.
A sodium bisulfite potentiometric titration [Gimeno, 2013] was used for determination of the peroxide
content in CMP-8 mixtures, and proved to be a robust technique for all of the candidate alcohols. Long-
term storage samples were prepared into
polyethylene storage vials and kept at room
temperature. The samples were titrated at
regular intervals throughout a 123 day study
period, maintaining their initial peroxide
concentration to within experimental error,
Table 1.
Detonation tests were performed to
confirm the detonability of the candidate
propellants. Due to a 400 kJ facility total energy limit in the 1800-liter blast chamber (University of Illinois
Energetic Materials Laboratory), we employed a ~1/6 scaled version of the UN Series 1(a) apparatus [TB
700.2, 2012]. This featured a 1.25” OD, 0.874” ID (0.188” wall), 6.62” long, drawn over mandrel steel
tubing as the pressure vessel and housed ~65 ml of propellant, topped by a 1” OD x ~0.5” tall booster
charge of PBXN9, initiated with an RP-81 detonator. Control runs were performed with water and a
known-detonable glycerol mixture before testing CMP-8. The tests resulted in a no-propagation finding for
CMP-8 via witness plate inspection, Figs. 3 and 4.
Figure 3: Detonation test setup: (left) before testing, (middle) after water control test, and (right) after
positive propagation glycerol test.
Table 1: Long term storage titration results
target H
2
O
2
% by initial mixture mass
43.5%
H
2
O
2
% by titration - 1hr
43.3%
H
2
O
2
% by titration - 36d
43.6%
H
2
O
2
% by titration - 75d
43.5%
H
2
O
2
% by titration - 123d
43.1%
DISTRIBUTION A: Approved for public release, distribution unlimited.
4
Water
(negative propagation)
Glycerol
(positive propagation)
Ethanol
(negative propagation)
Figure 4: Detonation test steel witness plates
Following the validation of the anticipated detonation limits, a series of ignition studies were
conducted. Each candidate fuel was tested with various ignition sources (butane flame, platinum glow
plug, Tesla coil, and piezoelectric spark) in a simple spot plate setup. It was observed that although CMP-
8 was not ignitable in air without the presence of a catalyst, it readily lit in the presence of several
common granular and metallic catalysts, Fig. 5.
Figure 5: Ignition tests with CMP-8 showing no ignition (left) without catalyst
and successful ignition (right) in the presence of catalyst.
A variety of combustion test fixtures were designed and manufactured. Common between all
fixtures were the following elements: inlet flow control orifice, catalyst bed (with resistive preheat), internal
ignition source (platinum hot wire or miniature spark plug), and nozzle. The geometry of each was varied
considerably, yielding a rapid increase in measured temperature after the first several modifications were
tested, Fig. 6.
DISTRIBUTION A: Approved for public release, distribution unlimited.
5
Figure 6: Chronological time
history of measured system
temperature as hardware was
modified.
Performance increases were realized mostly from geometric variation, thermal management, and
catalyst bed design. Once the “CTF-O” series hardware showed promise in the bench tests (a simple
vacuum chamber with electrical and fluid feedthroughs), the assembly was brought to the UIUC Electric
Propulsion Laboratory for thrust performance measurements. Propellant feed rate was monitored via a
spiral pressurized-tube feed system (with nitrogen pressurant) whose internal volume was calibrated with
a precision graduated cylinder. Propellant density was also carefully measured via the same graduated
cylinder and a precision laboratory balance. As seen in Fig. 7, the propellant was dyed green (<0.1%
w/w) for visibility. No degradation in performance was noted for dyed vs. non-dyed propellant.
Figure 7: CMP-8, dyed green for visibility: (left) in spiral feed system, and (right)
polyethylene storage cylinders.
The UIUC compact thrust stand [Wilson, 1997], Fig. 8, was used to make thrust measurements.
Test runs comprised a thrust stand calibration, a pre-heat of the catalyst bed, a propellant flow and CTF
burn of between 1 – 10 minutes, and finally another calibration run. Pre- and post-run calibrations are
necessary to account for the effects of changing thrust stand platform center of gravity and its resultant
influence on period and responsivity. Sample calibration and experimental measurements are shown
below in Fig. 9. During testing, the stainless steel CTF body glowed red hot, Fig. 10. The external body
temperature measured 690 ºC in this test.
DISTRIBUTION A: Approved for public release, distribution unlimited.
6
Figure 8: MPUC CTF mounted on top of UIUC compact thrust stand in the Electric Propulsion
Laboratory 1.5 m3 vacuum chamber. Gas flow is right to left.
Figure 9: Thrust calibration (left) and measurement (right) for a typical MPUC thrust test. This
test measured a specific impulse of 173 s.
DISTRIBUTION A: Approved for public release, distribution unlimited.
7
Figure 10: MPUC CTF glowing hot during combustion testing. Gas flow is left to right.
By varying the pressure of the feed system pressurant, propellant flow rate could be varied at any
time during the run, enabling throttling of the thruster to desired thrust levels. More than 2000 ml of
propellant was burned in various developmental test series. One long-duration test consumed a total of
315 ml (with ~20 s pauses to refill the spiral feed system).
Thruster performance was
repeatable and steady at ~180
seconds of specific impulse. Selected
data are presented below in Table 2,
which lists the conditions recorded
during a thrust stand test that
achieved one of the highest recorded
temperatures for the program. Note
that the exhaust gas temperature will exceed the measured external temperature, due to radial heat flow
in the combustion chamber.
Propellant viscosity was identified as a feed system risk factor for MPUC. The room temperature
kinematic viscosity of CMP-8 was measured to be 1.39 cSt with a Zahn Cup #1 efflux technique, Table 3.
Note that the Zahn Cup #1 method is less accurate for values below 5 cSt, however by measuring efflux
times for liquids of known viscosities (water and kerosene), we can interpolate a reasonable estimated
viscosity for CMP-8, which we believe to be accurate to within ±10%. Note that this kinematic viscosity of
1.39 cSt is similar to water and well below the kinematic viscosity of AF-M315E propellant at room
temperature. AF-M315E can have very small Re numbers in small flow channels which can create a
propellant feed problem, but this will not be an issue with CMP-8 propellant.
Table 3: Measurement of CMP-8 kinematic viscosity
Efflux Time [s]
Test #
Kerosene
Water
CMP-8
AF-M315E
1
30.25
25.87
26.50
2
27.98
25.55
26.62
3
27.54
25.73
26.57
4 30.10 25.68 26.57
5 32.75 25.70 26.65
Average
29.59
25.71
26.58
N/A
Kinematic
Viscosity [cSt] 2.71 1.0038 1.39 ≈ 25
Table 2: MPUC performance data.
Source Pressure [psia]
140
170
External Temperature [°C]
678
694
Mass Flow Rate [mg/s]
76.4
87.2
Thrust [mN]
137
160
Specific Impulse [s]
183
187
DISTRIBUTION A: Approved for public release, distribution unlimited.
8
CMP-8 is compared with AF-M315E and hydrazine in Table 4. Its advantages include low
viscosity, high availability, low cost, and low toxicity. Although the demonstrated specific impulse to date
is lower than AF-M315E and hydrazine, further development will close that gap. Compared to AF-
M315E, considerably lower thermal soak-back rates into the satellite are achievable with CMP-8 due to
the lower reaction temperature.
Table 4: Fuel comparison
Issue CUA CMP-08 AF-M315E Hydrazine
Detonability
not detonable by
blasting cap, ESD,
or impact, UN Class
1 classification antic.
UN 1.4C
classification high
Vapor Pressure very low very low high
Exhaust Product Toxicity non-toxic non-toxic Class 8 - highly toxic
PPE required Spill protection -
gloves / goggles Spill protection -
gloves / goggles Full SCAPE suit
Availability >2M metric tons
annually from COTS
reagents (low cost)
10000 lbs to date,
8000 lbs additional
capacity
>14000 metric tons
annually [1984]
Reaction Temperature [°C] 1200-1300 1700-1800 800
Necessary Catalyst Pre-heat Temp [°C] ≤200 400-450 25-250
Kinematic Viscosity [cSt at 25 °C] 1.4 25 0.9
Thruster materials necessitated stainless steels refractory metals stainless steels
Demonstrated Isp at 100 mN class 183 [CUA MPUC] 214 [Busek BGT-X1] 219 [Aerojet MPS-120]
Demonstrated Isp at 500 mN class n/a 220 [Busek BGT-X5] 227 [Airbus 0.5N]
Demonstrated Isp at 1 N class n/a 235 [Aerojet GR-1] 224 [Aerojet MR-103D]
FUTURE WORK
The MPUC combustor has been integrated into a complete CubeSat thruster design including 3-axis
ACS, using a self-pressurizing pressurant for both primary thruster fuel feed and ACS propellant. The
design philosophy of future MPUC systems is simple: maximize the H2O2/ethanol propellant load by
using as much of the available envelope as possible for propellant storage. By selecting pressurants with
a several-atmosphere vapor pressure over 0 to 60ºC (standard CubeSat temperature range requirement)
the system will self-pressurize, yet still retain a pressure low enough to allow a rectangular pressure
boundary. Conceptual models of a 1U MPUC with ACS are shown in Fig. 11. Notional internal
schematics have also been developed (not shown for brevity).
DISTRIBUTION A: Approved for public release, distribution unlimited.
9
Figure 11: MPUC 1U concept.
SUMMARY AND CONCLUSIONS
CU Aerospace (CUA) demonstrated a proof-of-principle Monopropellant Propulsion Unit for CubeSats
(MPUC) thruster during a recent Air Force Phase I SBIR. During this 6-month program, we took the
MPUC concept (TRL 2) to successful demonstration and performance testing as well as initial system
design (TRL 4). A specific impulse of 187 s was demonstrated during thrust stand testing. MPUC uses a
nontoxic propellant, with no special measures required for long-term storage. With its lower operational
temperature (and therefore, lower thermal soak back rate), the MPUC system provides longer duration
impulse capability for given thrust levels than higher-temperature propellants.
A principle program objective was to retire a significant amount of MPUC risk, which was achieved in
the areas of propellant toxicity, cost, detonability, viscosity, low flame temperature, storability and
combustion efficiency. Other primary technical objectives of the program were to prove stable, reliable
operation of a breadboard MPUC system and obtain preliminary thruster performance. Additional risk
reduction tasks that were accomplished included: (i) demonstration of reliable storage, (ii) detonation
limits, and (iii) experimental demonstration of specific impulse and thrust capabilities in a simulated space
environment. The conceptual system provides cold-gas attitude control and is estimated to have >1200
N-s/liter of volumetric impulse.
ACKNOWLEDGMENTS
This work was sponsored by USAF under SBIR contract FA9300-15-M-1003. The authors would
like to thank Dr. William Hargus (who served as our technical monitor), Professors Herman Krier and Nick
Glumac at the University of Illinois at Urbana-Champaign, as well as their students, for conducting the
detonation testing. The JANNAF Green Monopropellant Alternatives to Hydrazine Technical Interchange
Meeting was found to be a critically important resource for this MPUC project. Finally, special thanks go to
Dr. Mike Tsay of Busek Co., Inc., for constructive discussions.
DISTRIBUTION A: Approved for public release, distribution unlimited.
10
REFERENCES
Clark, J., Ignition!, Rutgers University Press (1972).
Delchev, N., US Patent 4,698,965 (1987).
Department of Defense Ammunition and Explosives Hazard Classification Procedures, TB 700-20,
NAVSEAINT 8020.8C, TO 11A-1-47 (July 2012).
Doncaster, B. and Shulman, J., 2016 Nano/Microsatellite Market Forecast, SEI website publication,
accessed on 9/28/2016. http://tinyurl.com/SEIreport2016
Gimeno, M P, M C C Mayoral, and J M M Andrés. A potentiometric titration for H2O2 determination in
the presence of organic compounds, Analytical Methods 5.6 (Jan 2013)
Hargus, W., Singleton, J., and Holmes, M., An Analysis of Cubesat Propulsion Options, 53rd JANNAF
Joint Propulsion Meeting, Colorado Springs, CO (May 2010).
Hawkins, T.W, Brand, A.J., McKay, M.B., and Tinirello, M., Reduced Toxicity, High Performance
Monopropellant at the U.S. Air Force Research Laboratory, AFRL Report AFRL-RZ-ED-TP-2010-
219 (2010).
King, S., Marx, P., and Taylor, D., The Aerospace Corporation Shell 405 Catalyst Evaluation
Program, AF Report #SAMSO-TR-69-271.VOL1 (August 1969).
Schreck, A, et al, Investigation of the explosive hazard of mixtures containing hydrogen peroxide
and different alcohols, Journal of Hazardous Materials, Volume 108, Issues 1–2, (April 2004).
Shanley, E.S., Prediction of the Explosive Behavior of Mixtures Containing Hydrogen Peroxide,
Journal of Jet Propulsion, Vol. 28, No. 6 (1958)
Singleton, J., Hargus, W., Carroll, D.L., Burton, R.L., Benavides, G.F., Hejmanowski, N., Laystrom-
Woodard, J., King, D.M., Day, C., Bhandari, R., Hageman, K., and Cardin, J.M., Propulsion for
Cubesats (PUC) – A Smart, Robust Propulsion System for CubeSats, JANNAF 2013-3243,
Colorado Springs, CO (May 2013).
Tsay, M. Feng, C., Paritsky, L., Zwahlen, J., Lafko, D., and Robin, M., Complete EM System
Development for Busek’s 1U CubeSat Green Propulsion Module, 52nd AIAA/SAE/ASEE Joint
Propulsion Conference, Propulsion and Energy Forum, (AIAA 2016-4905).
Vierea, R., Pham-Huu, C., Keller, N., and Ledoux M.J., New carbon nanofiber / graphite felt
composite for use as a catalyst support for hydrazine catalytic decomposition, Chemical
Communications (2002).
Wieling, Zandbergen, Mayer, and Schrijer, Development of a Hydrogen Peroxide Ethanol Thruster for
the Advanced Re-Entry Vehicle, Space Propulsion, Bordeaux, France (May 2012).
Wilson, M.J., Bushman, S.S., and Burton, R.L., A Compact Thrust Stand for Pulsed Plasma
Thrusters, IEPC 97-122 (August 1997).
Woschnak, et al, Development of a Green Bipropellant Hydrogen Peroxide Thruster for Attitude
Control on Satellites, Progress in Propulsion Physics, 4 (2013).