Characterization of Ethylenediamine Bisboraneas a Hypergolic Hybrid Rocket Fuel Additive

Mark A. Pfeil,∗ Jacob D. Dennis,† Steven F. Son,‡ Stephen D. Heister,§

Timothee L. Pourpoint,¶ and P. Veeraraghavan Ramachandran**

Purdue University, West Lafayette, Indiana 47907

DOI: 10.2514/1.B35386

A boron-based catalytic and energetic fuel additive is explored as a mechanism to enhance performance of hybrid

motors using nitric-acid-based oxidizers. Ethylenediamine bisborane is highly hypergolic with nitric acid, thereby

eliminating the need for a separate ignition system and providing the potential for restart. Ignition delays of

2.9� 0.3ms for ethylenediamine bisborane powder and 31.7� 19.6 ms for ethylenediamine bisborane/hydroxyl-

terminated polybutadiene combinations have been measured using white fuming nitric acid as the oxidizer. Theo-

retical specific impulse values higher than any other nitric-acid-based hypergolic hybrid combination tested are

obtained; in addition, ethylenediamine bisborane is air stable and not toxic. Also presented is a method of

manufacturing large quantities of high-purity ethylenediamine bisborane.

I. Introduction

H YPERGOLICITY is a highly desired characteristic for hybridmotors, as it removes the need for a separate ignition system

and provides for restart capabilities and energy management.Conventional hybrid motors require substantial energy to initiate,particularly for large motors that require high oxidizer flows. Suchcharacteristics make hypergolic hybrid rockets attractive for militaryuse, as these propellants can be used for various operational missionsfrom dynamic attitude control systems to motors that require deepthrottling capabilities. A collateral potential benefit of hypergolicityis an increase in regression rate, which is a desirable characteristic forhybrid motors, as compared to conventional fuels [1–10]. Priorresearch in this field has uncovered fuel additives that providehypergolicity with both hydrogen peroxide [1,4,7,11] and nitric-acid-based [7,8,12–16] oxidizers, but not all their characteristics areacceptable.Both metal hydrides and amine-based materials have been shown

to exhibit short ignition delays in a loose powder form ranging from 1to 10 ms for metal hydrides reacting with hydrogen peroxide [4,11]and from 27 to 4000 ms for amines reacting with white fuming nitricacid (WFNA), red fuming nitric acid (RFNA), or dinitrogen tetroxide(N2O4) [12–14]. These loose materials must be incorporated intosolid fuel matrices for practical application in a rocket system, whichtypically results in longer ignition delays. For example, ignitiondelays increased ranging from 0.2 ms to not hypergolic [1,11,15] forfuel binders incorporating metal hydrides when reacted with 70%nitric acid or 90% hydrogen peroxide and from 110 ms to nothypergolic for fuel binders incorporating amine-based materialswhen reacted with WFNA or RFNA [7,16]. Notably, ignition delays

ranging from 2 to 59ms were achieved when reacting metal hydridestogether with amine-based materials in a solid fuel with 96%hydrogen peroxide or WFNA [4,7,8].Although the ignition delays of some of these additive/binder sys-

tems are promising, a major drawback of several of the amines andmany of the metal hydride additives is that they are typically notstable at atmospheric (especially, humid) conditions. Such stabilityissues make them difficult to manufacture in large quantities orimplement in hybrid rocket systems [17] and have led to poor per-formance, presumably due to fuel grain degradation over time [18].Additionally, some metal hydrides are pyrophoric and toxic, whichfurther complicates fuel grain production and use. One method ofaddressing these concerns is to encapsulate the additives in protectivefuel binders such as dicylcopentadiene dihydrate [19]. However, theencapsulation of nondegraded additives must be completed in aninert environment, further complicating the grain manufacturingprocess.The stability limitations of the various hypergolic additives have

resulted in limited efforts to develop hypergolic hybrid systems.Although none of these systems has reached operational continualuse, a few propellant combinations have been used in flight demon-strations. Combinations ofWFNAandTagaform (a fuel consisting ofan aromatic amine and an aldehyde [8]), WFNA and p-toluidine/p-aminophenol, or RFNA and metatoluene diamine/nylon are most, ifnot all, of the propellants included in this category [20]. Hydrogen-peroxide-based systems have not been used in flight except asignition systems [21] that is likely due to the high freezing point andstability limitations for long-term storage of hydrogen peroxide.These issues are problematic, as hypergolic hybrids are well suitedfor missions that experience low temperatures or require long-termstorage. Other hypergolic oxidizer and fuel combinations (Table 1[1,3–8,20,22–24]) have been experimentally tested in rocket com-bustors. Sagaform A is an improved version of Tagaform. Most ofthese combinations have drawbacks, such as low melting temper-atures (ideally, melting temperatures will be above 344K for militaryapplications), air stability issues, and toxic components.Amine boranes are a class of hydrogen-dense amine materials

proposed for propulsion applications [16,25–31] that exhibit aspectsof having high hydrogen content and being an amine-based material.They are very similar to metal hydrides, in that the borane containsa metalloid, boron, and hydrogen; whereas metal hydrides containa metal and hydrogen. They are also related to a boron-containingmixture known as triethylaluminum-triethylborane that is commonlyused to pyrophorically ignite rocket engines, suggesting that highreactivity is possible [32]. They tend to be hypergolic with shortignition delays of 4 to 80 ms for powder samples [16,25] and 64 to1264 ms for powders cast in fuel binders when reacted with RFNA/WFNA [16]. Limited hybrid rocket testing resulted in a modified

Received 19 March 2014; revision received 19 August 2014; accepted forpublication 31 August 2014; published online 12 December 2014. Copyright© 2014 by the authors. Published by the American Institute of AeronauticsandAstronautics, Inc., with permission. Copies of this paper may be made forpersonal or internal use, on condition that the copier pay the $10.00 per-copyfee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers,MA 01923; include the code 1533-3876/14 and $10.00 in correspondencewith the CCC.

*Doctoral Student, School of Aeronautics and Astronautics, 500 AllisonRd. Student Member AIAA.

†Doctoral Candidate, School of Aeronautics and Astronautics, 500 AllisonRd. Student Member AIAA.

‡Professor, School of Mechanical Engineering, 500 Allison Rd. SeniorMember AIAA.

§RaisbeckEngineeringDistinguished Professor, School ofAeronautics andAstronautics, 701 W. Stadium Ave. Fellow AIAA.

¶Associate Professor, School of Aeronautics and Astronautics, 500 AllisonRd. Senior Member AIAA.

**Professor, Department of Chemistry, 560 Oval Drive.

365

JOURNAL OF PROPULSION AND POWER

Vol. 31, No. 1, January–February 2015

Dow

nloa

ded

by P

EN

NSY

LV

AN

IA S

TA

TE

UN

IVE

RSI

TY

on

May

30,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.

2514

/1.B

3538

6

regression rate (sensitive to configuration and flux level) [27,31,33]and increased C� efficiency of the rocket motor [27].Ethylenediamine bisborane (EDBB) (C2H14B2N2) is an air-stable

amine borane consisting of 16.3 wt% hydrogen that decomposesrapidly at temperatures above 363 K [34,35]. Historically, the mainuse of EDBB has been as a hydrogen storage material [34–38] with9.2–11.4wt%yields of hydrogen [35,38]. The high hydrogen contentof EDBBwould suggest that it could be a high-performance additivedue to the low molecular weight of hydrogen. However, no previousstudies have been made on the use of EDBB for propulsion appli-cations, likely due to prohibitive production costs, resulting fromsmall-batch synthesis methods used previously for EDBBproduction. Recent efforts by Ramachandran et al. [38] demonstrateda synthesis method that produces EDBB in large quantities, reducingcost and making it possible to consider this amine borane in a rocketsystem. Still, the current synthesis could be further improved.In this work, the objectives are to describe a new synthesis method

of producing large high-purity quantities of EDBB that are air stableand then characterize this powder and the powder made via theprevious method [38]. Characterization of physical features, theo-retical rocket performance, and hypergolic ignition delay in powderform and in a solid fuel matrix are presented.

II. Experimental Methods

A. Fuel Synthesis

Two procedures were used for the preparation of EDBB. Theformer involved the nucleophilic displacement of dimethyl sulfidefrom commercial borane-methyl sulfide (BMS) in solvent. Theamine borane produced by this protocol was coarse powder and willhereafter be referred to as EDBBA powder. The latter procedure in-volved the nucleophilic displacement of ammonia from a tetrahydro-furan (THF) solution of ammonia borane synthesized according tothe procedure described by Ramachandran and Gagare [39]. TheEDBB prepared by this protocol appeared more crystalline and fineand will hereafter be referred to as EDBB B powder. The latterprocedurewas applied for large-scale production of EDBB due to theair and moisture stability of ammonia borane compared to theextreme care necessary in handling BMS, as well as the low humanolfactory threshold of stenchy dimethyl sulfide.In a 12 liter three-neck round-bottom flask equipped with a

mechanical stirrer, a mixture of ammonia borane (524 g, 16.91 mol)and ethylenediamine (507 g, 8.45mol) in THF (8.5 liters) was heatedat 338 K with stirring for 12 h. The solvent was then removed undervacuum, and the remaining solid mass was stirred with a 2 liter 1∶1mixture of water and isopropanol for 12 h. After that, the solidmaterial was filtered using a Büchner funnel, the residuewas washedwith 1∶1 H2O-PrOH (0.5 liters) and i-PrOH (1 liter), respectively,and dried to yield EDBB.

B. Powder Characterization

Particle size distribution of the EDBB powder was measured bylaser diffraction using a Malvern Mastersizer Hydro 2000 μP andhexanes as a dispersant. To perform thesemeasurements, the index ofrefraction of EDBB was first obtained following the proceduresoutlined by Saveyn et al. [40] and Malvern Instruments, Ltd. [41],and using a Fisher Scientific (model 334620) refractometer with

anhydrous methanol as the solvent. A Hirox KH-8700 optical micro-scope was used to take high-resolution images of both EDBB pow-ders. These images provided details about surface features andrelative particle sizing to compare with the results of the laserdiffraction particle sizing.Electrostatic discharge (ESD) sensitivity experiments were

performed with 0.003–0.004 g of confined EDBB B powder in theair permilitary standardMIL-STD-1751 [42]. The standard thresholdof 250 mJ was discharged onto the prepared sample. As very littleoxidizer (air in this case) was in the confined space, it was not likelythat the EDBB powder would react; therefore, the experiments wererepeated with unconfined powder in the conductive cup, allowing airto surround the sample. A reaction or “Go” condition was charac-terized by any visible or audible reaction.The heat of combustion of the EDBBBpowderwas determined by

using a Parr 1281 bomb calorimeter. Fuel samples, ranging from 0.1to 0.3 g,were firstmade by pressing powder into pellets. Fuel sampleswere then placed in the calorimeter and burned in a pure-oxygenenvironment. The density of EDBBwasmeasured using an AccuPycII 1340 pycnometer using helium gas pressurized to 24 psi followingASTM International standard B923-10 [43].

C. Fuel Sample Preparation

The EDBB powder was incorporated into solid fuel matrices forfurther characterization. The fuel matrices consisted of 100% EDBBpressed into pellets and mixtures of EDBB and hydroxyl-terminatedpolybutadiene (HTPB) cured into a pellet. The pressed pellets wereproduced by applying 109.2 MPa to 100% EDBB powder in a 10-mm-diam stainless-steel dye using a Carver press. The powder waspressed for 10 min and then removed, producing cylindrical fuelpellets from 1 to 5 mm long.Powdered EDBB andHTPBweremixed for∼10 minwhile under

vacuum on a LabRam Resodyn resonant mixer with R-45polybutadiene resin, dioctyl adipate, and isophorone diisocyanate asthe components of the binder system. The fuel mixture was thentransferred to cylindrical molds with a 12 mm diameter and placed inan oven at 333 K for five days. Other R-45 curatives wereimplemented including Isonate™ 143L, Desmodur® N 3200, andDesmodur E 744. Pellets containing up to 58% EDBB were made.Although 58% EDBBmay be considered too high of a concentrationto be referred to as an additive in most fields, it is common in thehybrid community to refer to solid powders as additives in a fuelbinder despite the concentration.

D. Analysis of Fuel Samples

The ignition behavior and ignition delay time of both the powderand solid fuel sampleswere studied using a droplet ignition apparatus(Fig. 1). A droplet of 99% WFNA, purchased from Sigma Aldrich,was dropped from a syringe at a height of 127mmonto either a pile ofloose powder or cylindrical sections of the solid fuel matrices. Thesubsequent ignition delay was measured as the interval betweeninitial droplet contact with the fuel surface to first visible lightemission. The droplets of WFNA in this experiment had an averagediameter of 2.91� 0.02 mm. Various fuel surfaces were used, in-cluding pressed surfaces for pressed EDBB pellets, surfaces cut witha razor blade for EDBB/HTPB pellets, and surfaces sandedwith 100-grit sandpaper for both types of pellets. The rectangular Teflon® base

Table 1 Hypergolic hybrid rocket oxidizer and fuel combinations that have been used in experimental rocket combustors

Fuel type Oxidizer Comments

Tagaform [7,8] WFNA Softening point: 346 K/unknown toxicitySagaform A [22] RFNA Melting point: 408 K/unknown toxicityp-toluidine/p-aminophenol [7,20] WFNA p-toluidine melting point: 317 K/toxicMetatoluene diamine/nylon [4,5,20] RFNA/hydrogen peroxide Metatoluene diamine melting point: 371 K/toxicLithium aluminum hydride/polyethylene [3] Hydrogen peroxide Unstable in air/toxicManganese dioxide/sodium borohydride/polyethylene [1,23] Hydrogen peroxide Unstable in air/toxicDifurfurylidene cyclohexanone/polyisoprene [24] RFNA Difurfurylidene cyclohexanone melting point:

418 K/unknown toxicityAniline formaldehyde/magnesium [6] 99% RFNA/1% ammonium vanadate Aniline formaldehyde melting point: 423 K/toxic

366 PFEIL ETAL.

Dow

nloa

ded

by P

EN

NSY

LV

AN

IA S

TA

TE

UN

IVE

RSI

TY

on

May

30,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.

2514

/1.B

3538

6

on which the fuel sample was placed was cleaned between eachexperiment to prevent cross contamination between samples. Ignitiondelays and images were recorded through the use of a high-speedcolor Phantom v7.3 camera and a Nikon 28–105 mm lens. A 3 × 4array of Cree XLamp XP-G Star light-emitting diode (LED) lightswere used for illumination.

III. Results and Discussion

A. Fuel Characterization

The new fuel synthesis, discussed in Sec. II.A, resulted in 560 g(75%) of awhite powder, withmelting point of 413K. The boron andproton nuclear magnetic resonance spectrum of the powder in THF

solvent revealed that it was devoid of any impurities. Hydride anal-ysis revealed that the powderwas greater than 99%pure. This powderwas stored at atmospheric conditions for over two years before use inthese experiments, demonstrating long-term stability of EDBB.The density of EDBB B powder was measured 10 times using an

AccuPyc II 1340 pycnometer and found to have a value of 0.8317�0.0004 g∕cm3. This is very similar to the value reported byGroshensand Hollins of 0.82 g∕cm3 [35].Particle size distributions of the two EDBBA and B powders were

measured based on a volume percent basis using a Malvern Master-sizer Hydro 2000 μP; see Fig. 2. The index of refraction for thesemeasurements was found to be 1.868. Both EDBBA and B powdersexhibited a relatively single modal distribution. The majority of theparticles for theEDBBApowderwere between 100 to 2000 μm, withthe highest volume percent around 475 μm. The size of particles forthe EDBB B powder was generally smaller, falling between 10 and500 μm, with the highest concentration around 40 μm. Both powdersshowed a small percentage of particle sizes below these ranges; theEDBB B powder had some nanosized particles.Images of the EDBB powders indicate relatively clear, crystalline

particles, as indicated in Fig. 3. The angled facets and protrudingsurfaces of the EDBB A particles provide a clear indication of thecrystalline structure of these particles. Crystalline structures are alsoapparent in the EDBB B powder. The EDBB A particles generallyappear to be somewhat rectangular or elongated in one direction,whereas the EDBB B particles are irregular. These images also pro-vide general sizing of the particles, indicating that the EDBB Apowder size is on the order of 100 μm, whereas the EDBBB powderis an order of magnitude smaller: sizes that confirm the results madeby laser diffraction measurements.The standardmethod for testing ESD sensitivity [42] resulted in no

reactions for 10 separate experiments. Seven of those experimentswere repeated in an unconfined configuration. All of the unconfinedexperiments resulted in a reaction consisting of a green flash followedby smoke generation. This suggests that EDBBpowder is susceptibleto electrical energy; however, reaction leading to ignition will notoccur unless an adequate oxidizing environment is available. It is thussuggested that EDBB powder be handled appropriately to reduceaccidental ignition.The heat of combustion of the EDBB B powder was measured six

times using a bomb calorimeter, resulting in an average heat ofcombustion of 3651.9� 9.5 kJ∕mol. The heat of formation ofEDBBwas calculated from this measurement by assuming completereaction of EDBB with oxygen in the following manner:

C2H14B2N2�solid� � 7O2�gas� → 2CO2�gas�

� 6H2O�liquid� � N2�gas� � 2HBO2�solid� (1)

Equation (1) is used in conjunction with the following equation:

ΔHc � Hreactants −Hproducts � ΔnRT (2)

Fig. 1 Droplet ignition experiment for measuring hypergolic ignitiondelay and observing ignition behavior.

10-1

100

101

102

103

104

0

2

4

6

8

Particle Size, µm

Vol

ume

Per

cent

EDBB A Powder

EDBB B Powder

Fig. 2 EDBB particle size volume distribution for both EDBB A and Bpowders.

Fig. 3 Optical microscopy images of a) EDBB A and b) EDBB B powders.

PFEIL ETAL. 367

Dow

nloa

ded

by P

EN

NSY

LV

AN

IA S

TA

TE

UN

IVE

RSI

TY

on

May

30,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.

2514

/1.B

3538

6

whereΔHc is the heat of combustion of EDBB,Hreactants is the sumofthe heats of formation of the reactants,Hproducts is the sum of the heatsof formation of the products, Δn is the change in moles of gas fromreactants to products, R is the gas constant, and T is the initial/finaltemperature in the bomb calorimeter. This results in a heat offormation of−445.6� 1.2 kJ∕mol for EDBB.HBO2 solid productswere assumed, as this is the product calculated by both Cheetah 6.0[44] and NASA CEA [45] in equilibrium calculations for themeasured temperature within the bomb calorimeter. Solid precipitatewas formed in the bomb calorimeter sample holder after performanceof the experiment, consistent with the assumption of solid HBO2

products. However, the presence of solid precipitate in the sampleholder suggests some of the heat from the combustion process is notbeing transferred to the cooling jacket, and thus not measured. Themeasured heat of combustion and calculated heat of formation thenbecome lower limits of what the actual value may be.Samples of EDBB and HTPB composites resulted in gas genera-

tion during the curing process, producing voids in the cured sample.Such behavior was observedwith all four curatives used in this study.Cured fuel samples exhibited variations of structural characteristicsover time from firm and flexible to soft and deformable composites.Once soft, no further changes in the composite were observed. Thesecharacteristics suggest that EDBB is not compatible with the isocya-nate curatives used to crosslink the R-45 resin. It is therefore recom-mended that further efforts be made to identify polymer/curativecombinations that do not react with EDBB.

B. Theoretical Performance

The theoretical performance of EDBB compared to other nitric-acid-based hypergolic combinations is provided in Table 2 [46].Theoretical performance was obtained using the program Cheetah6.0 [44], whereas specific parameters for each propellant were ob-tained from the National Institute of Standards and TechnologyChemistryWebBook [47], Cheetah 6.0 [44], and calculating the heat

of formation for Tagaform following the procedures discussed inSec. III.Awhile using the heat of combustion and chemical formulaprovided by Magnusson [8]. The heats of formation were notavailable for some fuels. The performance of HTPB is provided, as itis a potential binder for EDBB powder. For the known nitric-acid-based hypergolic hybrid propellants, the theoretical Isp of EDBB isthe highest, whereas ρIsp is 26 s ⋅ �g∕cm3� below the highest value.Theoretical combustion product species in the combustion cham-

ber and the exhaust are provided in Fig. 4. Products are all in the gasphase for oxygen-to-fuel ratio (O∕F) values greater than 2.1. This isencouraging for internal flow in the motor, as there should not be anytwo-phase flow losses (or slag produced) in contrast with other high-energy additives, such as aluminum. Of particular interest are theexhaust products, as condensed phase productswill result in smoke inthe exhaust. Smoke is not preferred for military applications.Between anO∕F of 3.8 to 4.8, the exhaust products of EDBB are allin gaseous phase, providing an acceptable range of O∕F values forminimum smoke applications; however, it is difficult to determine ifcondensed phase products will not be formed behind the rocketwithout performing flight tests. The characteristic green light emis-sion caused by boron oxidation could also be a drawback if this fuel isused for minimum smoke applications. In the end, the strength of theboron emission will depend on the overall mass fraction in theexhaust.

C. Hypergolic Ignition

The ignition delay for a hypergolic propellant combination is animportant factorwhen designing a rocket system, and thus needs to becharacterized. In this work, EDBB was found to be hypergolic withWFNA for powders, pressed pellets, and EDBB cast in HTPB; theresults of which are tabulated in Table 3. The EDBB powders had theshortest ignition delay, with an average of 5.7� 1.0 ms for theEDBBA powder and 2.9� 0.3 ms for the EDBB B powder: a resultprobably due to particle size and resulting total surface area. This isthe shortest reported ignition delay time for amine or amine-boranepowders igniting with WFNA, RFNA, or N2O4 oxidizers, and it ismore on par with metal hydrides and hydrogen peroxide combi-nations. These results also support the trend that the combination ofamine and hydride materials result in short ignition delays.An example of a hypergolic ignition event between WFNA and

EDBB B powder is depicted in Fig. 5. A faint light green emission,indicative of ignition, is first observed 2.8 ms after the oxidizerdroplet touches the EDBB powder. A light green flame envelops thepowder and propagates outward, whereas an intense green flame isproduced at the location of where the EDBB andWFNA first came incontact. The green flame is an indication of boron combustion andsuggests that the borane groups on the EDBBmolecule are the initialparticipants in the hypergolic reaction. As the reaction proceeds, yel-low flames, characteristic of carbon combustion and soot formation,begin to appear near the edge of the green flame zone. The delayedappearance of the yellow luminosity may suggest that carbon does

Table 2 Maximum theoretical performance of various hypergolicoxidizer/fuel combinations using IRFNA IIIA1a [46] as the oxidizer(chamber pressure of 6.89 MPa, perfectly expanded to atmospheric

conditions, and using shifting equilibrium)

Fuel type O∕F Isp, s ρIsp, s ⋅ �g∕cm3�C�,m∕sEDBB 3.9 263.1 346.7 1532HTPB 4.6 264.1 363.4 1515Tagaform 3.8 259.9 373.0 1488Sagaform A — —— — — — — —

50% p-toluidine/50% p-aminophenol 3.6 257.1 360.7 1476Metatoluene diamine/nylon — —— — — — — —

Difurfurylidene cyclohexanone/polyisoprene

— —— — — — — —

Aniline formaldehyde/magnesium — —— — — — — —

aIRFNA IIIA (13–15% NO2, 1.5–2.5% H2O, 81.6–84.8% HNO3, 0.6–0.8% HF, and

≤0.1% solid nitrates).

0 2 4 6 8 10

a) b)

0

0.1

0.2

0.3

0.4

0.5

0.6

O/F

Mol

e F

ract

ion,

mol

/mol

0 0.5 10

0.05

0.1H2

H2O

N2

HO

CH4

BN(s)

COB2O3(L)

BHO2 O2CO2

C(S)

0 2 4 6 8 100

0.1

0.2

0.3

0.4

0.5

0.6

O/F

Mol

e F

ract

ion,

mol

/mol

H2

CO

B2O3(L)

BHO2

O2 B2O3(L) CO2

CH4

BN(s)C(S)

N2

H2O

Fig. 4 Theoretical product species vs O∕F ratio in a) the combustion chamber and b) the exhaust for EDBB and IRFNA IIIA combusted at a chamberpressure of 6.89 MPa, perfectly expanded to atmospheric conditions, and using shifting equilibrium.

368 PFEIL ETAL.

Dow

nloa

ded

by P

EN

NSY

LV

AN

IA S

TA

TE

UN

IVE

RSI

TY

on

May

30,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.

2514

/1.B

3538

6

not directly participate in the ignition process, but it may just indicatethat soot is produced later. As the reaction continues, the green flamebegins to fade, whereas the yellow flame intensifies. The resultingflame ball continues to expand until it has expanded to occupy avolume nearly 40 times greater than its initial volume, at which pointthe reaction ends. The EDBB A powder exhibits similar ignitionbehavior.Pressed EDBB fuel pellets exhibit fast, violent reactions when

exposed to a WFNA droplet. The droplet first impacts the fuel pelletand spreads out across the surface of the pellet (Fig. 6). A reactionbetween WFNA and the EDBB pellet appears to commence uponcontact, resulting in bubbling of the oxidizer. Eventually, a point isreached when the reaction rate increases notably, producing largequantities of gaseous products between the liquid oxidizer and solidfuel. This results in expulsion of the liquid WFNA from the pelletsurface and quenching of the reaction beginning 2.9� 0.3 ms afterthe WFNA droplet encountered the fuel pellet. It is likely that thispressure rise is a result of ignition between the solid fuel and liquid

oxidizer layers, as this is the ignition delay timeof the powder samples.The rapid depressurization caused by the expulsion of the oxidizerfrom the fuel surface would then cause the reaction to quench. Anysubsequent ignition of the fuel pellet was caused by the oxidizer fallingback onto the fuel pellet. These ignition dynamics will probably resultin short ignition delays in an actual rocket motor, as oxidizer will becontinuously coming in contact with the fuel surface.Experiments were also performed with pressed EDBB pellets that

were sanded to create a rough surface. These fuel pellets exhibitsimilar ignition behavior to that of the unsanded pellets, except thegas production leading to expulsion of the oxidizer from the fuelsurface appears to occur almost immediately upon contact. Localizedignition events also occur in the gases surrounding the fuel pellet;presumably decomposed EDBB and possibly EDBB fuel particlesare expelled from the fuel pellet that continue to react with the oxi-dizer. Again, more reaction leading to ignition occurs if any of theexpelled oxidizer falls back onto the fuel pellet.The EDBB and HTPB composite is hypergolic with WFNA.

Various concentrations of EDBB/HTPB pellets weremade; however,most fuel pellets had voids resulting in high surface areas, makingcomparison invalid. The ratio of 58%EDBB/42%HTPBdid producepellets that lacked voids, and were thus studied and reported. Experi-ments were performedwith both cut and sanded surfaces, resulting inignition delays of 65.5� 13.6 ms and 31.7� 19.6 ms, respectively.The ignition delay for the cut surface was as fast as any reportedamine-based material in a solid fuel matrix, whereas the sanded-surface ignition delay was the fastest reported in open liter-ature. These results are promising, despite the apparent incompat-ibility of EDBB and R-45 isocyanate curatives. Upon contact, the

Table 3 Ignition delay of various forms of EDBB

Fuel typeSurfacetype

Ignitiondelay, ms

Violent reactiondelay, ms

EDBB A powder — — 5.7� 1.0 — —

EDBB B powder — — 2.9� 0.3 — —

Pressed pellet, EDBB B powder Pressed — — 2.9� 0.3Pressed pellet, EDBB B powder Sanded — — 058% EDBBB powder/42% HTPB Cut 65.5� 13.6 — —

58% EDBBB powder/42% HTPB Sanded 31.7� 19.6 — —

0.0 ms 10 mm

Droplet

2.8 ms 10 mm

Ignition

10 mm4.2 ms 10 mm7.8 ms

10 mm13.2 ms 10 mm21.2 ms 10 mm44.6 ms 10 mm59.8 ms

Fig. 5 Hypergolic ignition of a droplet of WFNA with EDBB B powder.

0.0 ms 10 mm 1.2 ms 10 mm 2.2 ms 10 mm

4.0 ms 10 mm 6.2 ms 10 mm 11.2 ms 10 mm

Fig. 6 Heterogeneous reactionbetweenWFNAandpressedEDBB fuel pellets, resulting in shattering of the droplet and expulsion of the oxidizer from thefuel pellet.

PFEIL ETAL. 369

Dow

nloa

ded

by P

EN

NSY

LV

AN

IA S

TA

TE

UN

IVE

RSI

TY

on

May

30,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.

2514

/1.B

3538

6

WFNA droplet spreads over the fuel pellet surface and begins toreact; see Fig. 7. The reaction results in brown gas being producedbetween the liquid and fuel pellet that diffuses or bubbles through theliquid oxidizer and away from the fuel. Ignition is eventuallyachieved near the fuel pellet in the gas phase, resulting in a brightgreen flame that remains until all of theWFNA has reacted. Once theWFNA is consumed, the fuel pellet continues to burn with thesurrounding air until quenched. This behavior is similar for both cutand sanded surfaces, but it occurs on different timescales, resulting indifferent ignition delays.The ignition delay of the EDBB/HTPB fuel is compared against

other hypergolic combinations in Table 4 [7,8,22,24,48–51]. TheEDBB/HTPB fuel has a faster ignition delay than all of the motor-tested combinations, except for Sagaform A. The chemical makeupof SagaformAhas not been disclosed, but it is an improved version ofTagaform. It is thus likely that the fast ignition delay of Sagaform isbrought about by lithiumborohydride, which is the same additive thatreduced the ignition delay of Tagaform to 2 ms but made the fuelunstable in air [8,52]. The ignition delay of metatoluene diamine isnot available. This system used a phosphorus tri-N-methylimidecoating for hypergolic ignition [17], signifying that it is likely thatmetatoluene diamine has a long ignition delay. Therefore, for air-stable nitric-acid-based hypergolic propellant combinations, EDBB/HTPB has the shortest ignition delay, and thus probably the fastestresponse time.

IV. Conclusions

Ethylendiamine bisborane is a promising additive for implemen-tation in hypergolic hybrid rockets. Ethylendiamine bisboraneexhibits very short ignition delays of 2.9� 0.3 ms in powder formand violent reactions as fast as 2.9� 0.3 ms before the oxidizer isexpelled and the reaction quenched in pressed pellets. Despite theapparent incompatibility of EDBB and HTPB, these fuel pelletsproduced ignition delays as fast as 31.7� 19.6 ms. These delaytimes make EDBB one of the fastest reacting hypergolic hybridrocket fuels, suggesting that high fuel regression rates may beobtainable. The EDBB hypergolic combinations have high theoret-ical performance values, putting them above the other tested nitric-acid-based hypergolic combinations for which performance valueswere obtainable. The combination of short ignition delays and highperformance indicates that EDBB is an attractive additive for hyper-golic hybrid rockets.Ethylendiamine bisborane is themost practical fuel component for

producing hypergolicity and increasing performance of hybrid rock-ets of the fuels discussed in this work. Most of the additives used toproduce hypergolicity are difficult to manufacture, sensitive toatmospheric conditions, degrade over time, have storage temperaturelimitations, and tend to be classified as toxic materials. On the otherhand, EDBB is air stable, can be stored under awide range of temper-atures, and is classified as an irritant, making it easier to handleand implement. Additionally, the synthesis method outlined hereprovides a straightforward pathway to produce large quantities ofhigh-purity EDBB. The combination of these aspects along withperformance and ignition delay makes EDBB an attractive fuel forall hybrid rocket applications, from launch vehicles to militaryapplications.In the future, effortswill need to bemade to identify fuel binders that

are fully compatible with EDBB. Such binder/EDBB combinationswill likely result in shorter ignition delays than those reported here.Changes inparticle sizewill also likely influence ignition delay, both asa powder and in a fuel binder, and will need to be investigated.Additionally, other amine boranes will be investigated as potentialhypergolic fuels.

Acknowledgments

The authors would like to thank the Science, Mathematics, andResearch for Transformation scholarship, the National Defense

0.0 ms 10 mm

Droplet

10 mm8.0 ms 10 mm18.8 ms

10 mm51.8 ms 10 mm61.4 ms

Ignition

10 mm129.4 ms

10 mm202.6 ms

Fig. 7 Hypergolic ignition and subsequent combustion of 58% EDBB/42% HTPB with WFNA.

Table 4 Ignition delays of various hypergolic oxidizer/fuelcombinations

Fuel type OxidizerIgnitiondelay, ms

58% EDBB B powder/42% HTPB WFNA 31.7Tagaform [7,8] WFNA 150Sagaform A [22] IRFNA 5p-toluidine/p-aminophenol [7] WFNA 110Metatoluene diamine/nylon RFNA ––

Difurfurylidene cyclohexanone[24,48,49]

RFNA 45, 60–70,255

Aniline formaldehyde [50] 99% RFNA/1%ammonium vanadate

4800

50% aniline formaldehyde/50%magnesium [51]

RFNA 1134

370 PFEIL ETAL.

Dow

nloa

ded

by P

EN

NSY

LV

AN

IA S

TA

TE

UN

IVE

RSI

TY

on

May

30,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.

2514

/1.B

3538

6

Science and Engineering Graduate fellowship program, and theHerbert C. Brown Center for Borane Research. They would also liketo thank Bidyut Biswas, Brandon Terry, and Raghav Ramachandranfor their support in performing experiments.

References

[1] Austin, B. L., Heister, S. D., Dambach, E. M., Meyer, S. E., andWernimont, E. J., “Variable Thrust, Multiple Start Hybrid MotorSolutions for Missile and Space Applications,” 46th AIAA/ASME/SAE/

ASEE Joint Propulsion Conference and Exhibit, AIAA Paper 2010-7121, July 2010.

[2] Lo, R., “Empirical Laws for Hybrid Combustion of Lithium Hydridewith Flourine in Small Rocket Engines,” DLR German AerospaceCenter, Research Rept. 67-57, Sept. 1967; also NASA, TechnicalTranslation F-12,336, 1969.

[3] Osmon, R. V., “An Experimental Investigation of a Lithium AluminumHydride—Hydrogen Peroxide Hybrid Rocket,” Aerospace Chemical

Engineering, Vol. 62, No. 61, 1966, pp. 92–102.[4] Pugibet, M., and Moutet, H., “Utilisation dans les Systèmes Hybrids de

l’eau Oxygénée Comme Comburant,” La Recherche Aerospatiale,No. 132, 1969, pp. 15–31.

[5] Nadaud, L., andBaisini, J., “TheAblation of Solid FuelsUsed inHybridEngines,” La Recherche Aerospatiale, No. 115, 1966, pp. 45–55.

[6] Munjal, N. L., and Parvatiyar, M. G., “Regression Rate Studies ofMetalized Aniline Formaldehyde Hybrid Fuel,” Journal of Spacecraft,Vol. 13, No. 9, Sept. 1976, pp. 572–573.doi:10.2514/3.27928

[7] Schmucker, R., Hybridraketenantriebe: eine Einführung in Theoreti-

sche und Technische Probleme, Goldmann, Munich, 1972, pp. 184–187.

[8] Magnusson, U., “A Fuel for Hybrid Rockets,” World Aerospace

Systems, Vol. 2, 1996, pp. 50–52.[9] Smoot, L. D., and Price, C. F., “Pressure Dependence of Hybrid Fuel

Regression Rates,” AIAA Journal, Vol. 5, No. 1, 1967, pp. 102–106.doi:10.2514/3.3914

[10] Smoot, L. D., Price, C. F., and Mihlfeith, C. M., “The PressureDependence of Hybrid Fuel Regression Rates,” 3rd Aerospace SciencesMeeting, AIAA Paper 1966-0113, Jan. 1966.

[11] Sippel, T. R., Shark, S. C., Hinkelman, M. C., Pourpoint, T. L., Son, S.F., and Heister, S. D., “Hypergolic Ignition of Metal Hydride-BasedFuels with Hydrogen Peroxide,” 7th US National Combustion Meeting,Paper H24, Eastern States Section of the Combustion Institute, Atlanta,March 2011.

[12] Gune, S. G., Kulkarni, S. G., and Panda, S. P., “Synergistic HypergolicIgnition of Solid Substituted Anilines Mixed with Magnesium Powderand Red Fuming Nitric Acid,” Combustion and Flame, Vol. 61, No. 2,1985, pp. 189–193.doi:10.1016/0010-2180(85)90164-6

[13] Jain, S. R., Vittal, J. J., and Mimani, T., “Some Mechanistic Aspects ofHypergolic Ignition—Reaction of Dinitrogen Tetroxide with SolidAmines and Their Mixtures with Magnesium,” Defense Science

Journal, Vol. 42, No. 1, Jan. 1992, pp. 5–12.[14] Bernard,M. L., Cointot, A., Auzanneau,M., and Sztal, B., “The Role of

Surface Reactions in Hypergolic Ignition of Liquid-Solid Systems,”Combustion and Flame, Vol. 22, No. 1, 1974, pp. 1–7.doi:10.1016/0010-2180(74)90002-9

[15] DeSain, J. D., Curtiss, T. J., Metzler, K. M., and Brady, B. B., “TestingHypergolic Ignition of Paraffin Wax/LiAlH4 Mixtures,” 46th AIAA/

ASME/SAE/ASEE Joint Propulsion Conference and Exhibit, AIAAPaper 2010-6636, July 2010.

[16] Jain, S. R.,Murthy,K.N., andThanoo,B. C., “IgnitionDelay Studies onHypergolic Fuel Grains,” Defense Science Journal, Vol. 38, No. 3,July 1988, pp. 273–286.

[17] Jain, S. R., “Self-Igniting Fuel-Oxidizer Systems and Hybrid Rockets,”Journal of Scientific and Industrial Research, Vol. 62, April 2003,pp. 293–310.

[18] Corpening, J. H., Palmer, R. K., Heister, S. D., and Rusek, J. J.,“Combustion of Advanced Non-Toxic Hybrid Propellants,” Interna-

tional Journal of Alternative Propulsion, Vol. 1, Nos. 2–3, 2007,pp. 154–173.doi:10.1504/IJAP.2007.013018

[19] Shark, S. C., Pourpoint, T. L., Son, S. F., and Heister, S. D., “Perfor-mance of Dicyclopentadiene/H2O2-Based Hybrid Rocket Motors withMetal Hydride Additives,” Journal of Propulsion and Power, Vol. 29,No. 5, 2013, pp. 1122–1129.doi:10.2514/1.B34867

[20] Chiaverini, M. J., and Kuo, K. K., “Fundamentals of Hybrid RocketCombustion and Propulsion,” Progress in Astronautics and Aeronau-

tics, AIAA, Reston, VA, 2007, pp. 12–30.[21] Tsohas, J., Appel, B., Rettenmaier, A., Walker, M., and Heister, S. D.,

“Development and Launch of the Purdue Hybrid Rocket TechnologyDemonstrator,” 45th AIAA/ASME/SAE/ASEE Joint Propulsion Confer-

ence and Exhibit, AIAA Paper 2009-4842, Aug. 2009.[22] Ankarsward, B., “The SR-1 Propulsion System,” British Interplanetary

Society Symposium on Sounding Rockets, Cardiff, U.K., April 1970.[23] Wernimont, E. J., Meyer, S. E., and Ventura, M. C., U.S. Patent and

Trademark Office Patent No. 5,727,368 for “HybridMotor SystemwithConsumable Catalytic Bed a Composition of the Catalytic Bed and aMethod of Using,” March 1998.

[24] Paul, P. J.,Mukunda,H. S., Narahari, H.K., Venkataraman, R., and Jain,V. K., “Regression Rate Studies in Hypergolic System,” Combustion

Science and Technology, Vol. 26, Nos. 1–2, 1981, pp. 17–24.doi:10.1080/00102208108946942

[25] Gao, H., and Shreeve, J. M., “Ionic Liquid Solubilized Boranes asHypergolic Fluids,” Journal of Materials Chemistry, No. 22, 2012,pp. 11,022–11,024.doi:10.1039/c2jm31627g

[26] Chen, W. M., Murfree, J. A., Martignoni, P., Nappier, H. A., and Ayers,O.E., U.S. Patent No. 4,157,927 for “Amine-Boranes as HydrogenGenerating Propellants,” 1979.

[27] Weismiller, M. R., Connell, T. L., Risha, G. A., and Yetter, R. A.,“Characterization of Ammonia Borane (NH3BH3) Enhancement to aParaffin Fueled Hybrid Rocket System,” 46th AIAA/ASME/SAE/ASEE

Joint Propulsion Conference and Exhibit, AIAA Paper 2010-6639,July 2010.

[28] Risha, G. A., Connell, T. L., Weismiller, M., Yetter, R. A., Sundaram,D. S., Yang, V.,Wood, T. D., Pfeil, M. A., Pourpoint, T. L., Tsohas, J., andSon, S. F., “Novel Energetic Materials for Space Propulsion,” U.S. AirForce Office of Scientific Research, Rept. A818645, Arlington, VA,April 2011.

[29] Pfeil, M. A., Groven, L. J., Lucht, R. P., and Son, S. F., “Effects of Ammo-nia Borane on the Combustion of an Ethanol Droplet at AtmosphericPressure,” Combustion and Flame, Vol. 160, No. 10, 2013, pp. 2194–2203.doi:10.1016/j.combustflame.2013.04.014

[30] Pfeil, M. A., Son, S. F., and Anderson, W. E., “Influence of AmmoniaBorane on the Stability of a Liquid Rocket Combustor,” Journal of

Propulsion and Power, Vol. 30, No. 2, 2014, pp. 290–298.doi:10.2514/1.B34950

[31] Lee, J. G., Weismiller, M. R., Connell, T. L., Risha, G. A., Yetter, R. A.,Gilbert, P. D., and Son, S. F., “Ammonia Borane Based Propellants,”44th AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit,AIAA Paper 2008-5037, July 2008.

[32] Sutton, G. P., and Biblarz, O., “Hybrid Propellant Rockets,” Rocket

Propulsion Elements, 7th ed., Wiley, New York, 2001, p. 583.[33] Zaseck, C. R., Shark, S. C., Son, S. F., and Pourpoint, T. L., “Paraffin

Fuel and Additive Combustion in an Opposed Flow Burner Configu-ration,” 48th AIAA/ASME/SAE/ASEE Joint Propulsion Conference and

Exhibit, AIAA Paper 2012-3963, 2012.[34] Parry, R. W., “Some Significant Compounds of Boron,” Inorganic

Syntheses, Vol. 12, McGraw–Hill, New York, 1970, pp. 109–113.[35] Groshens, T. J., and Hollins, R. A., “New Chemical Hydrogen Storage

Materials Exploiting the Self-Sustaining Thermal Decompositionof Guanidinium Borohydride,” Chemistry Communications, No. 21,2009, pp. 3089–3091.

[36] Sahler, S., and Prechtl,M.H. G., “Advancement inMolecularHydrogenStorage Systems,” ChemCatChem, Vol. 3, No. 8, 2011, pp. 1257–1259.doi:10.1002/cctc.201100109

[37] Kelly, H. C., “Electronic Effects of theNH2BH2 Group. The Hydrolysisof Diamine Bisboranes,” Inorganic Chemistry, Vol. 5, No. 12, 1966,pp. 2173–2176.doi:10.1021/ic50046a020

[38] Ramachandran, P. V., Gagare, P. D., Mistry, H., and Biswas, B., WorldIntellectual PropertyOrg. Patent No.WO/2012/006347 for “Proceduresfor the Synthesis of EthylenediamineBisborane andAmmoniaBorane,”2012.

[39] Ramachandran, P. V., and Gagare, P. D., “Preparation of AmmoniaBorane in High Yield and Purity, Methanolysis, and Regeneration,”Inorganic Chemistry, Vol. 46, No. 19, 2007, pp. 7810–7817.doi:10.1021/ic700772a

[40] Saveyn, H., Mermuys, D., Thas, O., and Meeren, P., “Determination ofthe Refractive Index of Water-Dispersible Granules for Use in LaserDiffraction Experiments,” Particle and Particle Systems Characteri-

zation, Vol. 19, No. 6, 2002, pp. 426–432.doi:10.1002/ppsc.v19:6

PFEIL ETAL. 371

Dow

nloa

ded

by P

EN

NSY

LV

AN

IA S

TA

TE

UN

IVE

RSI

TY

on

May

30,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.

2514

/1.B

3538

6

[41] “Rapid Refractive Index Determination for Pharmaceutical Actives,”Malvern Instruments, Ltd. TN-MRK529-03, Worcestershire, England,U.K., 2003.

[42] “Military Standard Safety and Performance Tests for Qualificationof Explosives,” U.S. Air Force Dept. of Defense, MIL-STD-1751,Washington, D.C., Aug. 1982.

[43] “Standard Test Method for Metal Powder Skeletal Density by Heliumof Nitrogen Pycnometry,” ASTM International, STD-B923-10, WestConshohocken, PA, 2010.doi:10.1520/B0923-10

[44] Bastea, S., Fried, L. E., Glaesemann, K. R., Howard, W.M., Kuo, I. W.,Souers, P. C., and Vitello, P. A., Cheetah 6.0, Software Package,Lawrence Livermore National Lab., Livermore, CA, 2010.

[45] NASA CEA, Chemical Equilibrium with Applications, SoftwarePackage, Ver. 2, NASA Lewis Research Center, Cleveland, OH, 2002.

[46] Wright, A. C., “USAF Propellants Handbook Nitric Acid/NitrogenTetroxideOxidizers,”MartinMariettaCorp., AFRPL-TR-76-76,Vol. 2,Denver, CO, Feb. 1977.

[47] Afeefy, H.Y., Liebman, J. F., and Stein, S. E., “Neutral ThermochemicalData,”NIST Standard Reference Database Number 69, NIST ChemistryWebBook, edited by Mallard, W. G., and Linstrom, P. J., NationalInstitute of Standards and Technology, Gaithersburg, MD, 2011.

[48] Kulkarni, S. G., and Panda, S. P., “Role of Thermal Degradation ofHybrid Rocket Fuels in Hypergolic Ignition and Burning with RFNA as

Oxidizer,” Combustion and Flame, Vol. 39, No. 2, 1980, pp. 123–132.doi:10.1016/0010-2180(80)90012-7

[49] Panda, S. P., and Kulkarni, S. G., “Furfurylidene Ketones–ANewClassof Hypergolic Rocket Fuels with Red Fuming Nitric Acid (RFNA) asOxidizer,” Combustion and Flame, Vol. 28, 1977, pp. 25–31.doi:10.1016/0010-2180(77)90005-0

[50] Munjal, N. L., and Parvatiyar, M. G., “Ignition of Hybrid Rocket Fuelswith Fuming Nitric Acid as Oxidant,” Journal of Spacecraft, Vol. 11,No. 6, June 1974, pp. 428–430.doi:10.2514/3.62093

[51] Panda, S. P., Kulkarni, S. G., Kakade, S. D., and Rahalkar, A. B.,“Synergistic Hypergolic Ignition of Amino End Group in Monomersand Polymers,”Defense Science Journal, Vol. 36, No. 4, 1986, pp. 429–437.

[52] Mukunda, H. S., Jain, V. K., and Paul, P. J., “A Review of HybridRockets: Present Status and Future Potential,”Proceedings of the IndianAcademy of Sciences Section C: Engineering Sciences, Vol. 2, No. 2,1979, pp. 215–242.doi:10.1007/BF02845033

L. MauriceAssociate Editor

372 PFEIL ETAL.

Dow

nloa

ded

by P

EN

NSY

LV

AN

IA S

TA

TE

UN

IVE

RSI

TY

on

May

30,

201

5 | h

ttp://

arc.

aiaa

.org

| D

OI:

10.

2514

/1.B

3538

6