Method for predicting hypergolic mixture flammability limits: Application for non-ionic liquid based systems

Combustion and Flame 176 (2017) 547–553

Contents lists available at ScienceDirect

Combustion and Flame journal homepage: www.elsevier.com/locate/combustflame

Method for predicting hypergolic mixture flammability limits: Application for non-ionic liquid based systems Jérémy Sabard a, Laurent Catoire a,∗, Steven D. Chambreau b, Ghanshyam L. Vaghjiani c a

Department of Chemical Engineering, UCP, ENSTA ParisTech, Université Paris-Saclay, 828 Boulevard des Maréchaux, 91762 Palaiseau Cedex, France ERC Incorporated, AFRL/RQRP, 10 E. Saturn Blvd., Edwards AFB, CA 93524, USA c Air Force Research Laboratory, Propellants Branch, AFRL/RQRP, 10 E. Saturn Blvd., Edwards AFB, CA 93524, USA b

a r t i c l e

i n f o

Article history: Received 1 June 2016 Revised 13 July 2016 Accepted 21 November 2016

Keywords: Limits Ignition Flammability Hypergolicity

a b s t r a c t A numerical method is demonstrated in which a simple flame temperature criterion of 2700 K is used to map out flammability diagrams as a function of total mixture pressure and equivalence ratio in the hypergolic system, MMH/NTO/He. The computed results are in good agreement with experimentally determined ignition diagrams for MMH/NTO/He. The method is used to predict the lower and upper hypergolicity limits of other mixtures known to be hypergolic at 298 K and 1 atm. Comparisons between available experimental data (mixing ratios) and calculated limits lead to the conclusion that the present numerical method may allow the screening of other fuel/oxidant systems potentially of interest for spacecraft propulsion as well as for the determination of the range of mixing ratios able to ensure autoignitibility (hypergolicity) in combustion devices. In the safety field, the same method allows for the rapid assessment of hazards in terms of hypergolicity. © 2016 Published by Elsevier Inc. on behalf of The Combustion Institute.

1. Introduction The concept of limits was first formulated over 200 years ago by Humboldt and Gay-Lussac. Mixtures of combustible materials (such as gaseous or vaporized fuels and also particles) and air will only burn if the fuel concentration lies within the well-defined bounds of lower limit and upper limit, which are determined experimentally, and often referred to as flammability or explosion or ignition limits. LFL (lower flammability limit) is the lowest concentration of a gas or a vapor in air capable of sustaining a flame in the presence of an ignition source (such as an arc or flame). Concentrations lower than LFL are too lean (in fuel) to burn. UFL (upper flammability limit) is the highest concentration of a gas or a vapor in air capable of sustaining a flame in the presence of the ignition source. Concentrations higher than UFL are too rich (in fuel) to burn. These limits are generally expressed in terms of volume percentage at a given condition of temperature and pressure. It is also possible to ignite a mixture, without the presence of an ignition source, by heating the mixture up to the auto ignition temperature (AIT). In that case it is also possible to define and measure lower and upper limits. In this study, we are interested in hyper-

∗

Corresponding author. Fax: +33 181872024. E-mail address: [email protected] (L. Catoire).

http://dx.doi.org/10.1016/j.combustflame.2016.11.020 0010-2180/© 2016 Published by Elsevier Inc. on behalf of The Combustion Institute.

golic ignition, which is another type of ignition that requires neither an ignition source nor heating. In most cases, liquids or gases once mixed do not react violently at room temperature. The resulting mixtures often are comprised of no reactions at all or of ones with slow-rate reactions. However, some combinations, termed hypergolic, are known to react rapidly and under appropriate mixing ratios can lead to violent explosions. In that case, lower and upper limits can also be defined, which we will refer hereafter as LHL (lower hypergolicity limit) and UHL (upper hypergolicity limit). Quantitative knowledge of these hypergolicity limits is relevant to the optimization of fuel oxidation in a combustion engine. One such combination is that of monomethylhydrazine (MMH) and nitrogen tetroxide (NTO), which for many years has been the preferred choice for creating propulsive thrust in numerous spacecraft applications [1]. This energetic mixture affords a great advantage for the propulsion ignition system since the mixture can be auto-igniting at low temperatures and low pressures, thereby circumventing the need for an external ignition source, but its components also possess drawbacks. MMH and NTO are toxic and carcinogenic compounds, and pure MMH vapor is flammable (it can act as a monopropellant). One of the alternatives proposed to replace MMH is ionic liquids. Ionic liquids exhibit physico-chemical properties (of low vapor pressure, thermal stability, etc.) which make them ideal for replacing MMH and some of these ionic liquids are also hypergolic with common oxidants (like nitric acid and

548

J. Sabard et al. / Combustion and Flame 176 (2017) 547–553

NTO) [2,3]. However, due to the vast number of possible ionic liquids (1018 ) [4]. it is crucial to be able to a priori predict the autoignition behavior that may arise from the mixing of two or more liquids. Such information would be valuable in the development of new chemical rocket propulsion technologies or for the purpose of mitigating the explosion risks [5] associated with the transportation and management of these rocket fuels. An extensive search in the open literature has shown that there are no computational chemical kinetics methods or thermochemical methods currently available to address these tasks. The objective of this work, in the propulsion field, is to present a numerical procedure to predict mixture hypergolicity limits (LHL and UHL) to model ignition conditions. Here we restrict our analysis to the MMH/NTO system since the procedure presented herein can be validated against previous experimental data [6]. We then make predictions for other hypergolic systems for which limited or no experimental flammability limit data is available. Another objective of this study is relevant to the safety field: it is to allow the screening of numerous systems to determine if these systems are potentially hypergolically hazardous.

2. Flame temperature criterion

Fig. 1. MMH/NTO/He (with 53 mol% of He) hypergolicity diagram as a function of total mixture pressure and equivalence ratio obtained by assuming rapid mixing. : no ignition; ●: ignition.

2.1. Proof-of-concept procedure Traditionally, flammability limits for a given fuel/oxidizer system are determined by pass-or-fail flame-test results for known compositions of experimentally prepared pre-mixtures of the nonreactive gaseous components that are subjected to a standardized external ignition source [7]. For a given initial condition of temperature and pressure, the region between the lower and upper flammability limit defines the envelope of chemical compositions, which when ignited will support the propagation of a flame. Chemical equilibrium calculations can be used to estimate flammability envelopes where the use of a theoretical flame temperature threshold is applied as a criterion to estimate lower and upper flammability limits. By following the works of Egerton [8], White [9] and both Zabetakis et al. [10] and Burgess and Wheeler [11] have shown that the adiabatic flame temperature, at the lower and upper flammability limit, remains approximately constant in hydrocarbon/air mixtures, and that the reported experimental flammability limits correspond to fuel/oxidizer ratios for which in general, the adiabatic flame temperature is in the 10 0 0– 1500 K range. Melhem and Shanley [12] have found a flame temperature of 1400 K in the flammability limits determination for hydrocarbons (e.g., ethane/air mixture). However, a flame temperature equal to 1200 K was recommended by the authors for safety issues. In 1999 and 20 0 0, Mashuga and Crowl [13] also proposed a flame temperature criterion of 1200 K in order to determine flammability limits of hydrocarbon–air mixtures. In 2003, Hshieh et al. [14] proposed a different set of correlations for flammability limits via the measurement of gross (or net) heat of combustion of the compound. Heats of combustion and flame temperatures are related properties and therefore this latter idea is not fundamentally different from that of previous studies. Further works are available in this field, which are all based on the same premise [15–21]. Occasionally, a higher critical flame temperature has been proposed such as 1600 K. Correlation methods are also available and reviewed in the literature to predict flammability limits [22– 26]. The assumption that the limits of hypergolic mixtures can be similarly rationalized with a simple thermochemical criterion is examined and validated in Section 2.2 and extended to other combinations in Section 2.3.

Fig. 2. MMH/NTO/He (with 60 mol% of He) hypergolicity diagram as a function of total mixture pressure and equivalence ratio obtained by assuming rapid mixing. : no ignition; ●: ignition.

2.2. Approach One difficulty in determining a threshold flame temperature for hypergolic systems is due to the sparsity of experimental data. Previously, we have reported experimental hypergolicity diagrams at room temperature in the gas phase combination of NTO and MMH in helium. Because premixed mixtures were not feasible, the assumption of rapid adiabatic mixing was applied to obtain hypergolicity envelopes from the experimental data reported by Catoire et al. [6]. Figures 1–5 summarize these results for hypergolicignition and non-ignition as a function of total pressure and equivalence ratio at different mol% of the helium diluent. Here we define equivalence ratio to be (fuel/oxidant)/stoichiometry ratio. By

J. Sabard et al. / Combustion and Flame 176 (2017) 547–553

Fig. 3. MMH/NTO/He (with 70 mol% of He) hypergolicity diagram as a function of total mixture pressure and equivalence ratio obtained by assuming rapid mixing. : no ignition; ●: ignition.

considering the stoichiometric equation,

CH6 N2 + α ( β NO2 + γ N2 O4 ) → CO2 + 3H2 O + with

α=

5 2



1

β + 2γ



9 N2 4

and γ + β = 1

it is possible to define the equivalence ratio  from the composition according to:

=

PMMH / PNTO 1/α

where PMMH and PNTO are, respectively, the partial pressure of MMH and NTO in the mixture MMH/NTO/He. The composition at time zero is such that the proportions of NO2 and N2 O4 in NTO are determined from equilibrium thermodynamics. NO2 –N2 O4 equilibrium is reached in less than a millisecond, a negligible time compared to the reaction time. For mixtures with 53 mol% of helium, a minimum in hypergolicity domain is observed at an equivalence ratio equal to 2.5 and a total pressure equal to 3 kPa. When the amount of helium increases to 60 mol%, a minimum in hypergolicity domain is observed at a total pressure of 6 kPa and an equivalence ratio of 1.5. When the amount of helium was increased to 70 and 80 mol%, the minimum in hypergolicity domain was observed, respectively, at equivalence ratio of 1.07 and 0.95, and a total pressure of 10 and 23 kPa. For mixtures with 85 mol% of helium, no ignition was found for the entire domain tested. For the MMH/NTO/He mixtures studied, an increase in the mol% of helium leads to both a decrease of the minimum in the hypergolicity curve with respect to equivalence ratio and an increase in the minimum total pressure for ignition to occur. In order to establish a consistent kinetics criterion, for example, with the flammability diagram reported in Fig. 1, we have used CHEMKIN-II [27] and numerically considered several mixtures. The procedure involves the consideration of mixtures some of which, according to Figs. 1–4, are experimentally flammable and others which are experimentally non-flammable to compute the corresponding ignition delays, and from the observed ignition delay trend, the kinetics criterion is determined. For these calculations, a detailed MMH/NTO kinetics model described by Catoire

549

et al. [28] and updated since its publication was used. It consists of 403 equilibrated reactions among 82 species. This mechanism has been benchmarked with theoretical data available in the literature and the agreement between theory and predictions was found to be good. The homogeneous mixture at a given composition is assumed to remain at constant volume in an adiabatic setup. As previously explained, the quick establishment of the NO2 /N2 O4 equilibrium acts as an unwanted heat sink. This is shown in Fig. 6 where a very weak temperature drop (here 2 degrees) at around time zero is computed. A summary of the numerical procedure is presented as Supplementary material. For the case of no ignition, three computed temperature profiles are observed depending on the equivalence ratio. For low total pressures and adiabatic conditions, temperature profiles were computed for a very lean (Fig. 6), lean (Fig. 7) and rich (Fig. 8) MMH/NTO/He mixtures. The temperature profiles computed either exhibit maximum temperatures that are not representative of a mixture able to support a flame (Figs. 6 and 7) or the computed ignition delay is too long (Fig. 8). In contrast, for an igniting stoichiometric mixture at low total pressure, the temperature profile reported in Fig. 9 exhibits a temperature of about 30 0 0 K, which is typical for the flame temperature in MMH/NTO. Moreover, this profile is for a point located right on the ignition limit curve (pressure of 6 kPa and equivalence ratio of 1.0). This and other calculations at various pressure and equivalence ratios allow for the establishment of a simple numerical criterion to predict ignition events for non-adiabatic conditions, which are more representative of the actual experimental conditions: a rapid one-step or two-step temperature profile leading to a flame temperature of at least equal to 2900 K. Near the limit, in the non-ignition zone just below the lines in Figs. 1–4, flame temperatures of about 270 0–280 0 K are computed, and under adiabatic experimental conditions, transient flames are also likely to be observed. The temperature criterion determined here will be dependent on the experimental setup used (i.e., rapid adiabatic mixing). The heat capacity, surface-to-volume ratio, etc. of the apparatus could also be factors in the experimental determination of igniting versus non-igniting hypergolic mixtures. For that reason, as it is usual in the safety field, a precautionary margin is hereafter considered and a criterion of 2700 K selected. This numerical criterion was then validated by determining the theoretical hypergolicity envelopes and by comparing them to the experimental data, as seen in Figs. 10 and 11, where the lines in the figures are the experimental limits from Figs. 1 and 3. Since a NTO/MMH kinetics model exists, the determination of a kinetics criterion in terms of both ignition delay and temperature is possible. However, for the other systems under consideration here, only sparse detailed chemical kinetics models exist and the kinetics approach of ignition delay is therefore not feasible, instead only the temperature criterion is used as discussed below. Figures 10 and 11 show that the use of this numerical criterion is quite consistent with the experimental results, which was expected for Fig. 10 because the criterion was established according to experimental results of Fig. 1. The good agreement observed for Fig. 11, and for other data not reported here, shows that the criterion can be extrapolated to other mixtures with a high degree of confidence. 2.3. Other hypergolic mixtures Other fuel/oxidizer combinations are known to be hypergolic. However, for these combinations, limited data is available on hypergolicity limits, see for instance, Gray and Spencer [29], Perlee et al. [30], Zabetakis [31], Furno et al. [32] and Seamans et al. [33]. The critical flame temperature considered here is 2700 K. For all of the systems, various fuel/oxidizer combina-

550

J. Sabard et al. / Combustion and Flame 176 (2017) 547–553

330

Temperature (K)

320

310

300

290 0

Fig. 4. MMH/NTO/He (with 80 mol% of He) hypergolicity diagram as a function of total mixture pressure and equivalence ratio obtained by assuming rapid mixing. : no ignition; ●: ignition.

0.2

0.4

0.6

Time (s)

0.8

1

Fig. 6. Typical temperature profile computed under adiabatic conditions for a nonigniting, very lean MMH/NTO/He (with 53 mol% of He) mixture (equivalence ratio < 0.1 and low total pressure < 8 kPa).

1400

Temperature (K)

1200 1000 800 600 400 200 0

0.02

0.04

0.06

Time (s)

0.08

0.1

Fig. 7. Typical temperature profile computed under adiabatic conditions for a nonigniting, lean MMH/NTO/He (with 53 mol% of He) mixture (equivalence ratio between 0.1 and 0.5 and low total pressure < 8 kPa).

Fig. 5. MMH/NTO/He (with 85 mol% of He) hypergolicity diagram as a function of total mixture pressure and equivalence ratio obtained by assuming rapid mixing. : no ignition.

tions are considered from 0 to 1 in terms of mole fractions, with X(fuel) + X(oxidizer) = 1, where X is the mole fraction. In the present calculations, the reactants are always considered to form gaseous homogeneous mixtures at room temperature and atmospheric pressure conditions. Calculations with liquid reactants, forming homogeneous liquid mixtures, would have led to similar flame temperature values. No diluent was considered here. Because our criterion was shown to be valid over a wide dilution range (from 53 mol% up to 85 mol%), it is assumed here that the critical flame temperature is the same regardless of the dilution considered. The calculated lower and upper limits for these hypergolic mixtures are given in Table 1, which were calculated in a similar manner to the limits determined in Figs. 10 and 11. Thermodynamic data for reactants and products at equilibrium are from various thermodynamic tables. For completeness and because high

flame temperatures are expected for space fuels, ions have been considered at equilibrium. However, even at the highest flame temperature computed, these ions have been found to be present at low levels and they do not impact the value of the computed flame temperature. In the calculations reported in Table 1, Aerozine 50 is a 50/50 mix by weight of hydrazine and unsymmetrical dimethylhydrazine (UDMH). Therefore, Aerozine 50 is composed of 65.2 mol% of hydrazine and 34.8 mol% of UDMH. Red fuming nitric acid (RFNA) is composed of 80.5 mol% of HNO3 and 19.5 mol% of NO2 and white fuming nitric acid (WFNA) is composed of 100 mol% of HNO3 . H2 O2 is considered pure although it is generally available as a H2 O2 /H2 O mixture. This can change the limits but the effect is expected to be negligible as it is the case here for limits reported in Table 1 for WFNA and RFNA. For a given fuel, UHL is largest with HNO3 (or NO2 ) and smallest with H2 O2 . Similarly, LHL is smallest with H2 O2 and largest with HNO3 (or NO2 ). Limits in WFNA and RFNA are about the same if one considers the results for UDMH and for 2-furanmethanol re-

J. Sabard et al. / Combustion and Flame 176 (2017) 547–553

551

Table 1 Calculated lower and upper hypergolicity limits for hypergolic fuel/oxidizer mixtures expressed in mole fraction X of fuel in fuel/oxidant mixtures. Mole percentages are obtained by multiplying mole fraction by 100. Fuel

Oxidizer

Lower hypergolicity limit (LHL)

Upper hypergolicity limit (UHL)

Hydrazine Hydrazine Hydrazine MMH MMH MMH UDMH UDMH UDMH UDMH Aerozine 50 Aniline Aniline 2-Furanmethanol 2-Furanmethanol

H 2 O2 NO2 HNO3 H2 O2 NO2 HNO3 H2 O2 NO2 HNO3 RFNA NO2 H2 O2 HNO3 HNO3 RFNA

0.07 < X(Hydrazine) < 0.08 0.16 < X(Hydrazine) < 0.17 0.31 < X(Hydrazine) < 0.32 0.03 < X(MMH) < 0.04 0.07 < X(MMH) < 0.08 0.16 < X(MMH) < 0.17 0.02 < X(UDMH) < 0.03 0.04 < X(UDMH) < 0.05 0.11 < X(UDMH) < 0.12 0.09 < X(UDMH) < 0.1 X(Aerozine 50) ≈ 0.09 0.01 < X(Aniline) < 0.02 0.06 < X(Aniline) < 0.07 0.08 < X(C5 H6 O2 ) < 0.09 0.07 < X(C5 H6 O2 ) < 0.08

0.72 < X(Hydrazine) < 0.73 0.83 < X(Hydrazine) < 0.84 0.84 < X(Hydrazine) < 0.85 0.42 < X(MMH) < 0.43 0.58 < X(MMH) < 0.59 0.59 < X(MMH) < 0.60 0.30 < X(UDMH) < 0.31 0.44 < X(UDMH) < 0.45 0.46 < X(UDMH) < 0.47 0.456 < X(UDMH) < 0.459 X(Aerozine 50) ≈ 0.64 0.16 < X(Aniline) < 0.17 0.27 < X(Aniline) < 0.28 0.36 < X(C5 H6 O2 ) < 0.37 0.36 < X(C5 H6 O2 ) < 0.37

40

2000

Total pressure (kPa)

Temperature (K)

1600

1200

800

30

20

10

400

0 0

0 0

1

2

3

Time (s)

4

5

Fig. 8. Typical temperature profile computed under adiabatic conditions for a nonigniting, rich MMH/NTO/He (with 53 mol% of He) mixture (equivalence ratio > 2.0 and low total pressure < 2.5 kPa).

4000

Temperature (K)

3000

2000

1

2 3 4 Equivalence ratio

5

6

Fig. 10. Computed hypergolicity diagram (mol% He = 53). The line is the experimental limit between ignition and non-ignition regions given in Fig. 1. : no ignition; : ignition.

Table 2 Mixture ratios reported in the literature for selected hypergolic combinations [36–38]. Bipropellant combination

Mixture ratio (MR)

NTO/hydrazine NTO/MMH NTO/UDMH NTO/Aerozine 50 RFNAa /hydrazine RFNAa /MMH RFNAa /UDMH H2 O2 b /hydrazine

1; 1.08; 1.20; 1.34 1.5; 1.6; 1.73; 1.75; 2.19 2.10; 2.60 1.56; 1.59; 1.62; 2.00 1.28 2.13 2.60 2.15

a b

86% HNO3 + 14% NTO (wt%). 85% H2 O2 + 15% water (wt%).

1000

0 0

0.02

0.04

0.06

Time (s)

0.08

0.1

Fig. 9. Typical temperature profile computed under adiabatic conditions for an igniting, stoichiometric MMH/NTO/He (with 53 mol% of He) mixture (equivalence ratio 1.0 and low total pressure of 6 kPa).

ported in Table 1. Mixtures such as stoichiometric CH4 /O2 or stoichiometric H2 /O2 exhibit adiabatic flame temperatures at constant volume higher than the criterion considered here but they are not hypergolic, since these mixtures at room temperature would have exhibited extremely long calculated ignition delays. This is consistent with the statement that both kinetics and thermodynamic criteria are needed to ensure that a combination is hypergolic. The criterion proposed must only be used for systems either known to be hypergolic or suspected to be hypergolic.

552

J. Sabard et al. / Combustion and Flame 176 (2017) 547–553 Table 3 Mole percentage of fuel in selected hypergolic combinations at selected mixture ratio, MR values. Bipropellant combination

mol% of fuel

NTO/hydrazine NO2 /hydrazine NTO/MMH NO2 /MMH NTO/UDMH NO2 /UDMH NTO/Aerozine 50 NO2 /Aerozine 50 RFNAb /hydrazine RFNAb /MMH RFNAb /UDMH

74 at MR = 1 68 at MR = 1.34 59a at MR = 1 52a at MR = 1.34 57 at MR= 1.5 48 at MR = 2.19 40a at MR = 1.5 31a at MR = 2.19 37 at MR = 2.6 42 at MR = 2.1 22.7a at MR = 2.6 26.7a at MR = 2.1 58.5 at MR= 1.56 52.3 at MR = 2 41.4a at MR = 1.56 35.5a at MR = 2 61.6 at MR = 1.28 40.2 at MR = 2.13 29.7 at MR = 2.6

LHL–UHL (mol% of fuel) according to Table 1.

Average of LHL and UHL (mol% of fuel)

16–84

50

7–59

33

4–45

24.5

9–64 31–85c 16–60c 9–46

36.5 58 38 27.5

a Calculations performed from literature MR assuming that 1 mol of liquid NTO gives 2 mol of gaseous NO2 and that 1 mol of liquid fuel gives 1 mol of gaseous fuel. b RFNA is 86% HNO3 + 14% NTO (wt%). c Data reported are the ones calculated for WFNA instead of RFNA. See text for comments.

Total pressure (kPa)

40

30

20

10

0 0.0

0.5

1.0 1.5 Equivalence ratio

2.0

2.5

Fig. 11. Computed hypergolicity diagram (mol% He = 70). The line is the experimental limit between ignition and non-ignition regions given in Fig. 3. : no ignition; : ignition.

The only LFL and UFL data available in the literature for many of the fuels listed above is for fuel/air mixtures which are not hypergolic, see for example: IPCS International Programme on Chemical Safety, the International Chemical Safety Cards [34] (explosive limits vol% in air: hydrazine; 4.7–100, MMH; 2.5–97, UDMH; 2.4–20, aniline; 1.2–11.0, and 2-furanmethanol; 1.8–16.3), or CDC NIOSH Pocket Guide to Chemical Hazards [35] (explosive limits vol% in air:hydrazine; 2.9–98, MMH; 2.5–92, UDMH; 2–95, aniline; 1.3–11.0, and 2-furanmethanol; 1.8–16.3). The LFL and UFL envelopes for fuel mixtures with the oxidants shown above are qualitatively in agreement with the available air safety/hazard literature. Although it is difficult to quantitatively assess the calculated limits, nonetheless, our criterion applied here allows for the prediction of potentially new hypergolic combinations, such as ionic liquids/oxidizer systems if the thermodynamic approach described here is validated. For this task it is possible to consider the mixture ratio (MR) values reported in spacecraft propulsion applications, where MR is the weight ratio of oxidizer to fuel consumed and is given by [36]:

MR = wo /wf where wo is the oxidizer weight flow rate and wf the fuel weight flow rate. MR reported in the literature are for liquid systems (liquid NTO and liquid fuel). In the combustion chamber, both liquids are vaporized and therefore the corresponding composition of the gas phase for applications is derived for liquid/liquid MR as-

suming that 1 liquid mole of NTO gives 2 moles of gaseous NO2 . MR values for NTO/MMH combinations and for other bipropellant hypergolic combinations of interest are shown in Table 2. Lots of MR values are possible for real applications, in fact, all of the ones comprised between LHL and UHL. For instance, for the MMH/NTO combination the mixture ratio for maximum performance is about 1.6 [36], although a mixture ratio of 1.5 is frequently used. Although the lowest MR available is probably not representative of the LHL and the highest MR ratio available is probably not representative of the UHL, all of the MR values reported are in between LHL and UHL, and it is more than probable that the MR values used in practice are far away from both LHL and UHL to avoid instability problems and quenching, and they lie essentially in the middle of the hypergolicity range. The mixture ratio values given in Table 2, which are obviously in between LHL and UHL, are converted to mole percentage of fuel to allow for direct comparison with the limits reported in Table 1. Results are presented in Table 3. Calculations reported in Table 3 show that the average of the LHL and UHL matches quite well with the mol% of fuel typically used. This result cannot be obtained fortuitously for all of the systems examined here, and one has to conclude that the pass-or-fail tests have enabled engineers and scientists (together with theoretical and thermochemical considerations) to recommend mixtures for applications which are approximately right in the middle of the flammability domain determined here using the criterion described in this paper. This result is potentially of importance for new bipropellant combinations such as in ionic liquid/oxidant systems. 3. Conclusion In this proof-of-concept demonstration, the study of MMH/NTO/He gas phase numerical chemical kinetics has been shown to be consistent with the observed experimental reactivity of such mixtures at room temperature. The method provides reliable MMH/NTO/He hypergolicity diagrams showing the conditions for which auto-ignition or hypergolic-ignition is possible, which is based on a simple thermochemical criterion of achieving a minimum temperature threshold condition of 2700 K. The same criterion was applied to other combinations known to be hypergolic, and their lower and upper limits were predicted. In a companion paper, we extend this method to mixtures of ionic liquid fuels with common oxidizers such as HNO3 , NTO, and H2 O2 . This method can be utilized to systematically screen for suitable ionic liquid/oxidizer candidates for potential hypergolic propellant applications provided that their chemical kinetics model for ignition [39] is sufficiently developed to determine

J. Sabard et al. / Combustion and Flame 176 (2017) 547–553

temperature profiles to assess that “reasonable” i.e., not too long, ignition delays are expectable. Acknowledgments This research was supported by the European Office of Aerospace Research & Development (EOARD) under contract FA8655-13-1-3035. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.combustflame.2016.11. 020. References [1] G.P. Sutton, Rocket propulsion elements, seventh ed., Wiley, New York, 2001. [2] S. Schneider, T. Hawkins, M. Rosander, G. Vaghjiani, S. Chambreau, G. Drake, Ionic liquids as hypergolic fuels, Energy Fuels 22 (2008) 2871–2872. [3] Q. Zhang, J.M. Shreeve, Energetic ionic liquids as explosives and propellant fuels: a new journey of ionic liquid chemistry, Chem. Rev. 114 (2014) 10527–10574. [4] N.V. Plechkova, K.R. Seddon, Applications of ionic liquids in the chemical industry, Chem. Soc. Rev. 37 (2008) 123–150. [5] J.K. Niemeier, D.P. Kjell, Hydrazine and aqueous hydrazine solutions: evaluating safety in chemical processes, Org. Process Res. Dev. 17 (2013) 1580–1590. [6] L. Catoire, N. Chaumeix, S. Pichon, C. Paillard, Visualizations of gas-phase NTO/MMH reactivity, J. Propuls. Power 22 (2006) 120–126. [7] (a) G Zabetakis, A.L. Furno, G.W. Jones, Minimum spontaneous ignition temperatures of combustibles in air, Ind. Eng. Chem. 46 (1954) 2173–2178; (b) M. Molnarne, V. Schröder, Comparison of calculated data for the flammability and the oxidation potential according to ISO 10156 with experimentally determined values, J. Loss Prev. Process Ind. 24 (2011) 900–907; (c) C. Ding, Y. He, J. Yin, W. Yao, D. Zhou, J. Wang, Study on the pressure dependence of boiling point, flashpoint, and lower flammability limit at low ambient pressure, Ind. Eng. Chem. Res. 54 (2015) 1899–1907; (d) S. Li, Y. Zhang, X. Qiu, B. Li, H. Zhang, Effects of inert dilution and preheating temperature on lean flammability limit of syngas, Energy Fuels 28 (2014) 3442–3452. [8] A.C. Egerton, Limits of inflammability, Sym. (Int.) Combust. 4 (1953) 4–13. [9] (a) A.G. White, Limits for the propagation of flame at various temperatures in mixtures of ammonia with air and oxygen, J. Chem. Soc. Trans. 121 (1922) 1688–1695; (b) A.G. White, Limits for the propagation of flame in inflammable gas–air mixtures. Part III. The effects of temperature on the limits, J. Chem. Soc. Trans. 127 (1925) 672–684. [10] M.G. Zabetakis, S. Lambiris, G.S. Scott, Flame temperatures of limit mixtures, Sym. (Int.) Combust. 7 (1958) 484–487. [11] M.J. Burgess, R.V. Wheeler, The lower limit of inflammation of mixtures of the paraffin hydrocarbons with air, J. Chem. Soc. Trans. 99 (1991) 2013–2030. [12] (a) G.A. Melhem, E.S. Shanley, On the estimation of hazard potential for chemical substances, Process Saf. Prog. 15 (1996) 168–172; (b) G.A. Melhem, A detailed method for estimating mixture flammability limits using chemical equilibrium, Process Saf. Prog. 16 (1997) 203–218. [13] (a) C.V. Mashuga, D.A. Crowl, Flammability zone prediction using calculated adiabatic flame temperatures, Process Saf. Prog. 18 (1999) 127–134; (b) C.V. Mashuga, D.A. Crowl, Derivation of Le Chatelier’s mixing rule for flammable limits, Process Saf. Prog. 19 (20 0 0) 112–117.

553

[14] F.Y. Hshieh, D.B. Hirsch, H.D. Beeson, Using heat of combustion data to estimate or assess some threshold flammability properties of materials, Fire Mater. 27 (2003) 267–273. [15] M. Vidal, W.J. Rogers, J.C. Holste, M.S. Mannan, A review of estimation methods for flash points and flammability limits, Process Saf. Prog. 23 (2004) 47–55. [16] M Vidal, W Wong, W.J. Rogers, M.S. Mannan, Evaluation of lower flammability limits of fuel-air-diluent mixtures using calculated adiabatic flame temperatures, J. Hazard. Mater. 130 (2006) 21–27. [17] J.G. Quintiere, A theoretical basis for flammability properties, Fire Mater. 30 (2006) 175–214. [18] T. Ma, A thermal theory for estimating the flammability of a mixture, Fire Saf. J. 46 (2011) 558–567. [19] I.A. Zlochower, Experimental flammability limits and associated theoretical flame temperatures as a tool for predicting the temperature dependence of these limits, J. Loss Prev. Process Ind. 25 (2012) 555–560. [20] T. Ma, Q. Wang, M.D. Larranaga, Correlations for estimating flammability limits of pure fuels and fuel-inert mixtures, Fire Saf. J. 56 (2013) 9–19. [21] T. Ma, Using critical flame temperature for estimating lower flammability limits of a mixture, Process Saf. Prog. 32 (2013) 387–392. [22] Y. Pan, J. Jiang, R. Wang, H. Cao, Y. Cui, Prediction of the upper flammability limits of organic compounds from molecular structures, Ind. Eng. Chem. Res. 48 (2009) 5064–5069. [23] L.G. Britton, Using heats of oxidation to evaluate flammability hazards, Process Saf. Prog. 21 (2002) 31–54. [24] T.A. Albahri, Flammability characteristics of pure hydrocarbons, Chem. Eng. Sci. 58 (2003) 3629–3641. [25] F. Gharagheizi, A new group contribution-based model for estimation of lower flammability limit of pure compounds, J. Hazard. Mater. 170 (2009) 595–604. [26] M. Vidal, W.J. Rogers, J.C. Holste, M.S. Mannan, A review of estimation methods for flash points and flammability limits, Process Saf. Prog. 23 (2004) 47–55. [27] Chemkin II, Reaction Design, San Diego, 2013. [28] L. Catoire, N. Chaumeix, C. Paillard, Chemical kinetic model for monomethylhydrazine/nitrogen tetroxide gas-phase combustion and hypergolic ignition, J. Propuls. Power 20 (2004) 87–92. [29] P. Gray, M. Spencer, Combustion of unsymmetrical dimethyl hydrazine supported by oxygen or nitrous oxide: laminar flame propagation, Combust. Flame 8 (1964) 29–36. [30] H.E. Perlee, A.G. Imhof, M.G. Zabetakis, Flammability characteristics of hydrazine fuels in nitrogen tetroxide atmospheres, J. Chem. Eng. Data 7 (1962) 377–379. [31] M.G. Zabetakis, Flammability characteristics of combustible gases and vapors, Bureau of Mines, U.S. Dept. of the Interior, Washington, 1965 Bulletin # 627. [32] A.L. Furno, G.H. Martindill, M.G. Zabetakis, Limits of flammability of hydrazine-hydrocarbon vapor mixtures, J. Chem. Eng. Data 3 (1962) 375–376. [33] T.F. Seamans, M. Vanpee, V.D. Agosta, Development of a fundamental model of hypergolic ignition in space-ambient engines, AIAA J. 5 (1967) 1616–1624. [34] IPCS International Programme on Chemical Safety, the International Chemical Safety Cards. http://www.inchem.org/pages/icsc.html (accessed October 2015). [35] CDC NIOSH Pocket Guide to Chemical Hazards. http://www.cdc.gov/niosh/npg/ (accessed October 2015). [36] Brown, C.D.; Spacecraft propulsion, second printing, American Institute of Aeronautics and Astronautics, Reston, V.A. (1996) [37] G.P. Sutton, Rocket propulsion elements, sixth ed., John Wiley & Sons, Inc., New York, 1992. [38] Rocket and Space Technology: www.braeunig.us/space (accessed January 21, 2016). [39] L. Catoire, S.D. Chambreau, G.L. Vaghjiani, Chemical kinetics interpretation of hypergolicity of some ionic liquid-based systems, Combust. Flame 159 (2012) 1759–1768.