Quantum Chemical Insights into the Initial Reaction Pathways Governing the Self-Ignition of Hypergolic Ionic Liquids with H2O2 Authors: Miori Nogamida, Hirotoshi Mori Published: 2026 ABSTRACT Hypergolic ionic liquids (HILs), characterized by low volatility and improved safety, have attracted increasing attention as next-generation green propellants. However, the ignition mechanism of HILs with hydrogen peroxide (H2O2) remains poorly understood at the molecular level. In this study, we investigate the initial ignition processes of representative HIL anions reacting with H2O2 by combining automated reaction path exploration, kinetic analysis, energy density analysis (EDA), and natural bond orbital (NBO) analysis. The reactivity of each anion is systematically evaluated based on activation free energies. The results show that [BH4]− and [BH3CN]− exhibit hypergolic behavior, whereas [BH2(CN)2]− and [N­(CN)2]− do not, in qualitative agreement with experimental trends. Among the systems studied, [BH4]− exhibits the lowest activation barrier (20.6 kcal/mol) and the largest heat release (−105.4 kcal/mol), while cyano substitution progressively reduces reactivity. Further analysis reveals that ignition behavior is primarily governed by the activation barrier of the initial reaction, which is controlled by the efficiency of electron donation from the anion bonding orbitals to the antibonding orbital of H2O2. These findings establish an activation-controlled framework for understanding self-ignition in HIL/H2O2 systems and provide molecular-level design principles for next-generation green hypergolic propellants. Introduction Hypergolic propellants spontaneously ignite upon contact with an oxidizer and have been widely used in aerospace applications because of their rapid and reliable ignition characteristics. Conventional systems, such as hydrazine-based fuels combined with oxidizers like HNO3 or N2O4, exhibit short ignition delay times and stable performance, but suffer from severe drawbacks including high volatility, acute toxicity, and carcinogenicity. , To address these issues, hypergolic ionic liquids (HILs) were introduced as safer alternatives by Schneider and co-workers. These salts, composed of tunable combinations of cations and anions, offer low volatility and high energy density, making them promising candidates for environmentally benign propellants. , While fuming nitric acid was initially employed as the oxidizer, ,− concentrated H2O2 has recently emerged as a safer and more sustainable alternative, particularly because of its hypergolic reactivity with boron-containing anions such as [Al­(BH4)4]−, [BH4]−, and [BH3CN]−. − Despite these advances, the ignition mechanism of HIL/H2O2 systems remains poorly understood, especially regarding the origin of differences in hypergolic behavior among various anions. Conventional transition-state-based approaches often rely on predefined reaction pathways and may fail to identify the primary initial reactions governing ignition without prior mechanistic assumptions. , Self-ignition is macroscopically observed as a rapid temperature rise, yet its origin lies in elementary reactions occurring immediately after contact between fuel and oxidizer. − Therefore, understanding the relative ignition behavior of different anions requires a quantitative comparison of the activation barriers associated with the primary initial reactions. In this study, we employ the artificial force induced reaction (AFIR) method to systematically explore the initial reaction pathways between HIL anions and H2O2, enabling unbiased comparative identification of the relevant transition states without predefined mechanistic assumptions. Based on these pathways, activation barriers and reaction energetics are evaluated, and kinetic simulations are performed to achieve semiquantitative kinetic discrimination between hypergolic and nonhypergolic systems. Furthermore, EDA and NBO analyses are conducted to elucidate the thermochemical and orbital-level factors governing reactivity, thereby linking donor orbital character and donor–acceptor interactions to activation-barrier control. The central objective of this work is to establish an activation-controlled framework for self-ignition in HIL/H2O2 systems through comparative pathway analysis, semiquantitative kinetic discrimination, and electronic-structure interpretation, and thereby to provide molecular design principles that balance low activation barriers with sufficient heat release. Computational Details Reaction pathways between HIL anions and hydrogen peroxide (H2O2) were explored using the AFIR method implemented in GRRM17, , with Gaussian16 used as the electronic structure calculation engine. Based on previous studies and our preliminary calculations (Figure S1), cations exert only a minor influence on the initial reaction pathways. ,, Therefore, the present study focuses on anion–H2O2 interactions. The representative anions considered in this study are shown in Figure . These four anions were selected as representative and widely studied HIL anions. Experimental ignition data using H2O2 are available for these systems, thereby enabling a comparative analysis of the molecular factors governing the initial ignition step. ,,, Initial pathway searches were performed at the UωB97XD/6–31G­(d) level of theory. Pathways with collision energy parameters below 50 kcal/mol were further refined at the UωB97XD/aug-cc-pVDZ level through geometry optimization and frequency analysis. Transition states were characterized by vibrational frequency analysis, confirming the presence of exactly one imaginary frequency. The connectivity of each transition state was verified by intrinsic reaction coordinate (IRC) calculations, and final energies were obtained via single-point calculations at the UCCSD­(T)/aug-cc-pVDZ level. Reported ΔG and ΔH values at 298 K include zero-point vibrational and thermal corrections derived from the frequency calculations at the Uω B97XD/aug-cc-pVDZ level. Representative hypergolic anions investigated in this study for reactions with H2O2. Rate constants were computed using transition state theory (TST) and expressed in modified Arrhenius form. Calculated Arrhenius parameters for the elementary reactions used in the reduced kinetic model are summarized in Tables S16–S19. Thermodynamic properties were represented using NASA7 polynomials, and kinetic simulations were conducted using Cantera. Simulations were performed at 300 K and 1 atm with an anion/H2O2 ratio of 1:4, and temporal profiles of temperature and species concentrations were analyzed. It should be noted that the present kinetic model is a reduced model focusing on initial reactions; thus, it is intended to capture relative trends in ignition behavior rather than absolute ignition delay times. Energy density analysis (EDA) , was used to investigate the origin of heat release, and natural bond orbital (NBO) analysis was employed to characterize the orbital interactions governing initial reactivity. Results and Discussion Semiquantitative Kinetic Discrimination of Relative Ignition Propensity To examine whether the quantum-chemically identified initial reactions can account for the experimentally observed ignition behavior, kinetic simulations were conducted under identical conditions. As shown in Figure , temperature evolution was tracked for each system at 300 K and 1 atm with an anion/H2O2 ratio of 1:4. Time evolution of temperature obtained from kinetic simulations at 300 K and 1 atm with an anion/H2O2 ratio of 1:4: (a) [BH4]− + H2O2, (b) [BH3CN]− + H2O2, (c) [BH2(CN)2]− + H2O2, and (d) [N­(CN)2]− + H2O2. Nonhypergolic systems ([BH2(CN)2]−, [N­(CN)2]−) exhibited only modest initial temperature increases to 379 and 447 K, respectively, followed by thermal decay. This indicates that although the initial reactions are exothermic, the associated reaction rates are too slow to sustain thermal runaway under ambient conditions. The absence of a sustained temperature rise is consistent with experimentally observed nonhypergolic behavior. , In contrast, hypergolic systems ([BH4]−, [BH3CN]−) showed rapid temperature escalation after the initial stage, indicating the onset of sustained exothermic reactions. The ignition delay times, defined here as the time required to reach 1000 K, were estimated to be 0.30 s for [BH4]− and 4.68 s for [BH3CN]−. Although these values are longer than the corresponding experimental measurements ([BH4]−: 18.5 ms; [BH3CN]−: 390 ms), the relative trend is correctly reproduced. , This discrepancy is most likely attributable to the reduced kinetic model employed in the present study, which considers only the initial reactions and does not explicitly include subsequent chain-propagating combustion processes. These kinetic results indicate that only the [BH4]− and [BH3CN]− systems possess sufficiently accessible product-forming pathways to trigger sustained temperature rise, whereas the corresponding pathways in the nonhypergolic systems remain too slow under ambient conditions. This trend motivates a closer examination of the corresponding reaction energy profiles. For clarity, the net stoichiometries corresponding to the dominant initial processes captured in the reduced kinetic model are summarized in Table S15. For the nonhypergolic systems, these dominant processes correspond to prereactive-complex formation rather than subsequent product-forming steps. Reaction Energy Profiles Governing Ignition The corresponding reaction energy profiles are summarized in Figure . The product-forming pathways identified by AFIR commonly involve hydrogen-bonded prereactive complex formation, followed by O–O bond cleavage and oxygen transfer. Thus, the initial reaction is not a free hydride transfer or outer-sphere electron transfer, but a prereactive-complex-mediated oxygen-transfer process driven by donor-to-σ *­(O–O) charge transfer. Reaction energy profiles for the initial reactions of each anion with H2O2. Free energies at 298 K are shown, with enthalpies in parentheses; all values include zero-point vibrational and thermal corrections. Despite this common mechanistic framework, the activation barriers increase systematically with cyano substitution [BH4]−<[BH3CN]−<[BH2(CN)2]−<[N(CN)2]− For the [N­(CN)2]− system, the product-forming pathway identified by AFIR begins with formation of a hydrogen-bonded complex accompanied by a heat release of ΔH r = −14.8 kcal/mol, followed by O–O bond cleavage and oxygen transfer to yield H2O. However, the calculated activation Gibbs energy is as high as 38.8 kcal/mol, which is consistent with its nonhypergolic behavior under ambient conditions. In contrast, for the borohydride-type anions [BH4]−, [BH3CN]−, and [BH2(CN)2]−, the product-forming pathways identified by AFIR involve analogous reaction sequences consisting of hydrogen-bonded complex formation, O–O bond cleavage, and oxygen transfer to boron. Their activation Gibbs energies are 20.6, 22.6, and 24.5 kcal/mol, respectively, showing a progressive increase with increasing cyano substitution. This trend indicates that cyano substitution suppresses the initial reactivity of borohydride-based anions. A more subtle difference emerges between [BH3CN]− and [BH2(CN)2]−. Although their activation Gibbs energies are similar, their ignition behaviors diverge. This difference is not captured by the intrinsic rates at 298 K alone, but by the extent of the initial heat release. The [BH3CN]− system is more exothermic both upon prereactive-complex formation (ΔH = −15.6 vs −12.5 kcal mol–1) and over the overall initial product-forming process (ΔH = −103.4 vs −100.8 kcal mol–1). Consequently, in the kinetic simulation, the temperature rises to 465 K for [BH3CN]− but only to 379 K for [BH2(CN)2]−, and the rate constants for the corresponding first product-forming step reach approximately 5.40 s–1 and 2.50 × 10–4 s–1, respectively. Thus, when the activation barriers are similar, the decisive factor is whether the initial heat release is sufficient to shift the system into a temperature regime in which the corresponding product-forming step can proceed at an appreciable rate. These observations support an activation-controlled interpretation of ignition, in which the initial barrier determines whether the reaction can enter a thermally accelerating regime, while heat release acts as a secondary amplification factor once the reaction is initiated. Thus, the energetic trends presented here provide the molecular basis for the kinetic behaviors described above. Electronic and Thermochemical Origins of Reactivity To understand the origin of the differences in heat release in greater detail, EDA was carried out for the oxidized products. As representative examples, we focus here on the [BH4]− system, which exhibits the highest exothermicity, and the [N­(CN)2]− system, which shows the lowest reactivity. When each oxidized product is partitioned into the oxidized anion fragment and H2O, the stabilization contributed by H2O is found to be nearly identical in both systems, at approximately −74 kcal/mol. Therefore, the difference in overall exothermicity mainly originates from the extent of stabilization of the oxidized anion fragment. In the oxidized [N­(CN)2]− system, oxidation reduces the extent to which the original resonance stabilization can be retained, whereas the oxidized [BH4]− system can gain additional stabilization upon oxidation. Thus, the thermochemical differences are governed by the intrinsic electronic structure of the anion and its ability to stabilize oxidized configurations. The spatial distributions of the heat-release contributions are shown in Figure . Spatial distributions of heat-release contributions upon oxidation for the representative anion systems. NBO analysis further reveals that reactivity is governed by both orbital energy alignment and spatial overlap. This reactivity is driven by donation from the anion donor orbital into the σ *­(O–O) orbital of H2O2 within the prereactive complex, indicating an intracomplex charge-transfer-assisted oxygen-transfer mechanism. The orbital energy of the acceptor NBO of H2O2 is 0.171 au, whereas the donor orbital energies for [BH4]−, [BH3CN]−, [BH2(CN)2]−, and [N­(CN)2]− are −0.217, −0.256, −0.295, and −0.262 au, respectively. The corresponding donor orbital energies and occupancies are summarized in Table . 1Orbital Energies (ε) and Occupancies (n) for the Electron-Donating NBOs of [BH4]−, [BH3CN]−, [BH2(CN)2]−, and [N­(CN)2]−, Together with the Electron-Accepting NBO of H2O2 Orbital Energies (ε) and Occupancies (n) for the Electron-Donating NBOs of [BH4]−, [BH3CN]−, [BH2(CN)2]−, and [N­(CN)2]−, Together with the Electron-Accepting NBO of H2O2 The high reactivity of [BH4]− can be attributed to the B–H bonding orbital, which contains a nearly spherical H 1s component and can overlap effectively with the O–O antibonding orbital of H2O2. In other words, [BH4]− benefits from both a relatively small donor–acceptor energy gap and favorable spatial overlap, facilitating efficient electron transfer in the initial step. In [BH3CN]− and [BH2(CN)2]−, increasing cyano substitution stabilizes the donor orbitals, thereby increasing the energy gap with the acceptor orbital of H2O2 and suppressing electron transfer. This trend is consistent with the increase in activation barrier. In [N­(CN)2]−, the relevant donor orbital also has a smaller occupancy than those of the borohydride-type anions, which further diminishes its ability to transfer electron density into the antibonding orbital of H2O2. Taken together, these results show that thermochemical stabilization and electronic structure cooperatively govern ignition behavior, with the activation barrier emerging as the primary descriptor. These results provide a direct electronic-structure basis for predicting hypergolic reactivity. Conclusion This study demonstrates that the self-ignition behavior of HIL/H2O2 systems is primarily governed by the activation barrier of the initial reaction. While heat release and orbital interactions contribute, the activation barrier emerges as the primary factor. These findings provide a clear molecular design guideline: minimize the initial activation barrier while maintaining sufficient exothermic stabilization. Future studies incorporating cation effects, condensed-phase or solvent environments, and reaction pathways beyond the primary initial processes identified here may help bridge the gap between the present model focused on the primary initial reactions and a more quantitative description of experimentally observed self-ignition behavior. These insights are expected to guide the rational design of safer and more efficient HIL-based hypergolic propellant systems. Supplementary Material