Abstract
The N5⁻ anion, known as pentazolate, represents a groundbreaking advancement in the field of energetic materials, offering promising applications in rocket propulsion, explosive devices, and pyrotechnics. Comprising five nitrogen atoms arranged in a cyclic structure with a negative charge, has captured significant interest due to its unique configuration and high energy potential. In this article, we provide a comprehensive overview of the N5⁻ anion’s potential as an energetic material, alongside the role of RMD simulations in elucidating its behavior. The ReaxFF forcefield was used to simulate the materials pyrolysis. The total energy behavior of different species containing pentazolate, across a range of temperatures (1500 K to 3000 K) revealed distinct trends and characteristics associated with the thermal dynamics and stability of the molecule under varying thermal conditions. Their mechanisms were elucidated, and the kinetic parameters were calculated, indicating that CNN5, with its low activation energy (39.14 kJ/mol), stands out as the most reactive, while PolyN5, with the highest activation energy (52.88 kJ/mol), is the most stable. Overall, the N5 - anion represents a promising avenue for the development of high-energy materials.
Key words
reactive molecular dynamics; pentazolate; decomposition mechanism; physical chemistry
INTRODUCTION
Energetic materials, which include explosives, propellants, and pyrotechnics, release substantial amounts of energy in the form of heat, light, and gas when triggered by stimuli such as heat, shock, or friction. This energy release is harnessed for a variety of applications, including rocket propulsion and military explosives (Mendonça &
Urgessa 2023). However, the inherent reactivity of these materials poses risks, making a deep understanding of their fundamental mechanisms crucial for safe and efficient use.
Stability is a concern in energetic materials area focusing on safe operations, reducing risk of handling. Nazin & Korsunskiy (2021) studied the stability of 11 standard explosives at temperatures below 100° C.
In recent years, computational methods like reactive molecular dynamics (RMD) simulations have emerged as powerful tools for exploring the behavior of energetic materials at the molecular level (Li et al. 2013, Gonçalves et al. 2022). These simulations facilitate the study of chemical reactions and energy transfer processes during detonation events, providing insights that can lead to the design of safer and more efficient materials (Hamilton et al. 2021, Mendonça et al. 2017). Wu et al. (2020) showed that femtosecond laser pulse can be used to prepare high purity energetic nanomaterials, working with a suitable selection of femtosecond laser pulse energy.
The N5⁻ anion, known as pentazolate, represents a groundbreaking advancement in the field of energetic materials (Hou et al. 2024), offering promising applications in rocket propulsion, explosive devices, and pyrotechnics (Lin et al. 2020, Steele & Oleynik 2017). This anion, comprising five nitrogen atoms arranged in a cyclic structure with a negative charge, has captured significant interest due to its unique configuration and high energy potential.
The reactivity of the N5⁻ anion stems from its cyclic structure, which contains significant strain energy (Lang et al. 2022). Under external stimuli such as heat or shock, it undergoes exothermic decomposition, releasing a large amount of energy in the form of gas and heat (Yang et al. 2021). The decomposition products, including molecular nitrogen, ammonia, and nitrogen oxide species, are less toxic and environmentally hazardous compared to the decomposition products of some conventional energetic materials.
The high energy content and distinctive structure of the N5⁻ anion make it an attractive material for various applications. In rocket propulsion, it could serve as a high energy density fuel, offering greater thrust and specific impulse than conventional fuels, and its high density could enhance payload capacity for space missions. In explosive devices, the N5⁻ anion could function as a primary explosive to initiate detonation or as a component of secondary explosives, providing high-energy gas to propel explosive products (Wu et al. 2020). Its application in pyrotechnics could lead to brighter and more visually appealing displays.
RMD simulations have significantly advanced the understanding of the N5⁻ anion’s reactivity and behavior. These simulations use interatomic potentials and reactive force fields derived from quantum mechanical calculations and experimental data to study the molecular interactions and chemical reactions within energetic materials. Recent advancements in RMD simulations have resulted in more accurate and efficient models, enabling the simulation of larger and more complex systems (Hamilton et al. 2021, Wu et al. 2020). Through RMD simulations, researchers have uncovered the complex decomposition process of the N5⁻ anion, involving multiple intermediates and products, and demonstrated that its reactivity is highly dependent on external stimuli such as temperature and pressure. These insights are crucial for developing safer and more efficient energetic materials (Li et al. 2013).
In this article, we provide a comprehensive overview of the N5⁻ anion’s potential as an energetic material, its synthesis, reactivity, and applications, alongside the role of RMD simulations in elucidating its behavior.
MATERIALS AND METHODS
Five different compounds were selected for studying the behavior of the N5 - anion. Figure 1 shows the molecular structure of CH3N5, CNN5, HN5, diN5 and polyN5, respectively.
Energetic compounds with pentazolate ion used in the study (CH3N5, CNN5, HN5, diN5 and polyN5, respectively).
The reactive molecular dynamics (RMD) simulations for the compounds were conducted using the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) software (Plimpton 1995), utilizing the ReaxFF force field specifically tailored for energetic materials (Senftle et al. 2016). ReaxFF, which can deal with the chemical reactions in large-scale systems in terms of the formation and dissociation of bonds, has been intensively developed for molecular dynamics (MD) simulation (Wu et al. 2020). ReaxFF is an empirical force field for reactive systems, and the total energy of the reaction system is described as a function of bond order, interatomic distance, valence, and torsion angles. ReaxFF also includes the energy of van der Waals and Coulomb interactions between atoms.
Particularly important applications of ReaxFF have been to energetic materials, which involve a complex sequence of reactions and multiple intermediates, many of which are difficult to detect, making it difficult to extract a mechanistic understanding. Thus, ReaxFF reactive dynamics (ReaxFF-RD) has been used successfully to study shock transformation for many energetic materials including hexahydro-1,3,5-trinitro-1,3,5-s-triazine (RDX), pentaerythritol tetranitrate (PETN), octahydro-1,3,5,7-tetranitro-1,3,5,7-
tetrazocine (HMX), 1,3,5-triamino-2,4,6-trinitrobenzene (TATB), triacetonetriperoxide (TATP), and nitromethane (NM) (Wu et al. 2020). The ReaxFF force field captures the intricate energy contributions and chemical reactions involved in the simulation, described by the following general energy function:
where Ebond represents the bond energy, Eover and Eunder denotes the over- and under-coordinated atom in the energy contribution, respectively, Eval, Epen, Etors are the valence angle term, penalty energy and torsion energy, respectively, Econj, EvdWaals and ECoulomb represent the conjugation effects to molecular energy, nonbonding van der Waals interactions and Coulomb interactions, respectively.
The ReaxFF force field assumes that bond orders and energies between atom pairs are derived from interatomic distances, incorporating contributions from sigma bonds, first and second pi bonds. The bond orders are corrected for overcoordination and residual bond orders in valence angles, which are then used to calculate bond energies.
The parameters for the ReaxFF force field were optimized using a successive one-parameter search technique based on quantum chemical calculations (DFT). This optimization ensured the accurate reproduction of heats of formation within 4.0 kcal/mol (Van Duin et al. 2001).
During each MD step, the force field dynamically updates bond orders, allowing the simulation to accurately model bond formation and breaking. ReaxFF’s accuracy extends to reproducing quantum mechanical data and providing detailed atomistic descriptions of complex chemical reactions.
Pyrolysis simulations were performed on the N5⁻ anion by placing molecules in a simulation cell of appropriate size to achieve realistic density values (Table I). The system was initially minimized at a low temperature (5 K) and then equilibrated at 298 K using an NVT ensemble (constant number of atoms, volume, and temperature) for 20-40 picoseconds (ps). For the production phase, the NVE ensemble (constant number of atoms, volume, and energy) was employed with a timestep of 0.1 femtoseconds (fs). The total simulation time for the production phase was 100 ps, with temperatures ranging from 1500 to 3000 K controlled by the Berendsen thermostat, featuring a temperature damping constant of 100 fs.
RMD simulations facilitated the calculation of kinetic parameters for the N5⁻ anion, demonstrating proven accuracy. 16 static temperatures between 1500 and 3000 K were applied to the system in an NVE ensemble, controlled by the Berendsen thermostat. This setup ensured smoother results with minimal temperature and energy fluctuations. First-order kinetics was assumed for the thermal decomposition of individual molecules.
The consumption rate of the molecules is dependent on the temperature; higher amounts of molecules decompose as the temperature increases. The rate of the decomposition indicates the activation energies. The rate constant at each temperature is determined with the integrated first order rate law equation:
where No is the initial number of molecules in the system and Ni is the number of molecules at the time t. This equation allows to calculate the reaction rate at every time-step. The rate constants calculated from different temperatures are then used to generate an Arrhenius plot for the fuel molecules. The activation energy (Ea) and the frequency factor (A) were calculated with the linear fit of the plot, according to the Arrhenius equation:
where R is the universal gas constant.
This methodology (Gonçalves et al. 2022) provided a comprehensive understanding of the kinetic behavior and reactivity of the N5⁻ anion under various conditions.
RESULTS AND DISCUSSION
Total energy
In RMD, the Hamiltonian (H) equations of motion return the total energy of the system, and thus fluctuation in the Hamiltonian can be related to fluctuations in the total energy of the system.
The simulations of the five species generated a large amount of data, therefore the results for the HN5 system will be presented in detail, and the other results may be seen in the supplementary material, as the behavior is similar.
The total energy behavior of HN5 across a range of temperatures (1500 K to 3000 K) over a simulation period of 100 picoseconds (Fig. 2) reveals distinct trends and characteristics associated with the thermal dynamics and stability of the molecule under varying thermal conditions. There are some distinct regions over which the results may be separated:
-
Initial Energy Increase (0-10 ps)
-
At the onset of the simulation, there is a rapid increase in total energy observed across all temperatures. This initial rise is due to the system reaching thermal equilibrium after the initial conditions and minimizations. The magnitude of the energy rise is more pronounced at higher temperatures, indicating a greater degree of molecular excitation and activation.
-
Energy Fluctuations and Peaks (10-30 ps)
-
Between 10 ps and 30 ps, the total energy exhibits noticeable fluctuations and reaches several peaks. These peaks correspond to transient states where the system is undergoing significant molecular rearrangements and bond breakage/forming events. Higher temperatures, such as 3000 K (light blue) and 2900 K (light green), show more substantial fluctuations, suggesting more dynamic and frequent reactions within the molecule.
-
Gradual Energy Decline (30-100 ps)
-
Post the peak period, the total energy gradually declines for all temperatures. This decline is associated with the system stabilizing and the exothermic decomposition reactions leading to the formation of stable products. The rate of energy decline is steeper at higher temperatures, implying more rapid reaction kinetics and faster decomposition rates.
Regarding the temperature-dependent behavior observed at the lower temperatures (1500 K to 2000 K), the total energy shows less pronounced fluctuations and a more gradual decline. This suggests slower reaction rates and fewer molecular collisions capable of inducing bond breakage. The energy curves at these temperatures remain relatively close to each other, indicating similar thermal behavior and energy dissipation rates.
As the temperature increases (2100 K to 2500 K), the energy fluctuations become more noticeable, and the peaks are higher. This is indicative of increased molecular activity and more frequent reaction events. The total energy starts to decline more sharply after reaching the peak, suggesting an increased rate of decomposition.
At the highest temperatures (2600 K to 3000 K), the energy fluctuations are most pronounced, and the peaks are the highest. The rapid and significant changes in energy reflect intense molecular activity, frequent bond breakages, and rapid decomposition processes. The decline in total energy is steep, indicating that the molecule quickly reaches a more stable state with the formation of reaction products.
The overall trend observed in the total energy behavior of HN5 highlights the strong dependence of molecular stability and reaction kinetics on temperature. Higher temperatures facilitate faster decomposition and more dynamic reactions, while lower temperatures result in more gradual and less pronounced energy changes. For the other studied systems (Supplementary Material - Figures S1 to S4), the observed behavior is similar to HN5, with only one exception; the total energy of the CNN5 system suffer gradual increases during the simulations, indicating lower velocities for the decomposition reactions. A larger number of simulation steps would be needed to observe the energy decrease of this system.
Thermal decomposition mechanism
The decomposition of HN5 at 3000K involves a variety of reactions that transform HN5 into different products. The reactions, based on their occurrences, provide insights into the primary and secondary pathways involved in the decomposition process.
Table II below presents the mechanism composed by reactions with at least 5 occurrences.
This mechanism presents some the key reactions, which are described below.
Primary Decomposition Reactions
1. HN5 → HN3 + N2 (96 occurrences)
- This reaction is the most frequent, suggesting that the primary decomposition pathway for HN5 at 3000K involves breaking down into HN3 and N2. This reaction likely serves as the initial step in the decomposition process, rapidly dissociating the HN5 molecule.
2. HN3 + N2 → HN5 (79 occurrences)
- This reversible reaction indicates that HN3 and N2 can recombine to form HN5. However, given the high temperature, the equilibrium likely favors the decomposition side, forming HN3 and N2 more frequently.
3. HN3 → N2 + HN (91 occurrences)
- The HN3 formed in the primary step further decomposes into N2 and HN. This step is critical as it leads to the generation of smaller, more stable molecules.
4. N2 + HN → HN3 (93 occurrences)
- Similar to the recombination of HN3 and N2 to form HN5, this reaction also shows reversibility. Yet, the high temperature would favor the decomposition of HN3 into N2 and HN.
5. N5 + H → HN5 (11 occurrences)
- Although less frequent, this reaction suggests a possible pathway for the formation of HN5 from N5 and H. At 3000K, the dissociation of larger clusters into smaller molecules is more favored.
Secondary Reactions and Further Decomposition
1. N4 → 2 N2 (520 occurrences)
- The formation and subsequent dissociation of N4 into two N2 molecules is highly frequent. This step is crucial in the breakdown of complex nitrogen species into simpler diatomic nitrogen.
2. N2 + HN → HN3 (112 occurrences)
- This reaction supports the intermediate formation of HN3 from HN and N2, indicating a cyclic interconversion among these species before complete decomposition.
3. HN5 → N2 + HN3 (104 occurrences)
- Another significant pathway where HN5 decomposes directly into N2 and HN3. This reaction further confirms the importance of HN3 and N2 as primary products in the decomposition mechanism.
4. HN2 → N2 + H (38 occurrences)
- The decomposition of HN2 into N2 and H suggests that nitrogen-hydrogen bonds are breaking, leading to the formation of diatomic nitrogen and atomic hydrogen.
Formation and Decomposition of Other Species
1. H2N3 → N2 + H2N (105 occurrences)
- The conversion between H2N3 and its decomposition products indicates the formation of complex intermediates that eventually break down into simpler molecules.
2. H2N + N2 → H2N3 (84 occurrences)
- This reversible reaction shows the potential for H2N and N2 to form H2N3, though decomposition into simpler components is generally favored at high temperatures.
3. HN4 → N2 + HN2 (71 occurrences)
- The breakdown of HN4 into N2 and HN2 is another pathway contributing to the formation of diatomic nitrogen and smaller nitrogen-hydrogen species.
At 3000K, HN5 primarily decomposes into HN3 and N2. The HN3 further breaks down into N2 and HN. These primary reactions are supported by secondary pathways involving the interconversion and decomposition of other nitrogen species such as N4, H2N3, and HN2. The high temperature favors the formation of stable diatomic nitrogen (N2) and smaller nitrogen-hydrogen fragments, driving the overall decomposition process. The frequent occurrence of N2 formation steps highlights the stability and prevalence of N2 as a final product in the thermal decomposition of HN5. Tables SI to SIV present the decomposition mechanism for the other species. Many of the elementary reactions can be seen in the other mechanisms, thus indicating an acceptable level of acuity for the simulations and results obtained.
Activation energy
The activation energy (Ea) is a critical parameter in understanding the stability and reactivity of chemical species. It represents the minimum energy required for a chemical reaction to occur. Along with the frequency factor (A), which relates to the number of successful collisions that lead to a reaction, these parameters help in predicting the rate at which a substance will decompose. The following discussion elaborates on the significance of the activation energies and frequency factors calculated for different N5 -containing structures, as presented in Table III. The Arrhenius plots for each system can be observed in Figure S5.
CH3N5 exhibits a relatively high activation energy of 49.61 kJ/mol. This high value suggests that CH3N5 is relatively stable under normal conditions, requiring substantial energy input to overcome the activation barrier and initiate decomposition. Once this energy threshold is surpassed, the high frequency factor of 2.74E+12 s⁻¹ indicates that the reaction proceeds at a rapid rate, implying a high probability of reaction once the molecules have acquired sufficient energy. The correlation coefficient (r² = 0.9273) further underscores the reliability of this kinetic data, indicating a strong conformity to the Arrhenius equation, which predicts the temperature dependence of reaction rates.
Among the structures studied, CNN5 has the lowest activation energy, 39.14 kJ/mol, implying that it is the least stable and most reactive of the group. This lower energy barrier means that CNN5 can decompose more readily, even under conditions where other structures might remain stable. The frequency factor, 1.82E+12 s⁻¹, is also relatively high, indicating that once the activation energy is overcome, the decomposition reaction occurs rapidly. The correlation coefficient (r² = 0.9017) is slightly lower than those of the other structures, but it still indicates a good fit to the Arrhenius model, suggesting reliable kinetic data with some variability in the reaction rate.
The activation energy for DiN5 is 51.43 kJ/mol, which is higher than that of CH3N5 and CNN5. This suggests that DiN5 is more stable and requires more energy to decompose, making it less reactive under similar conditions. The frequency factor, 4.47E+11 s⁻¹, is lower than those of CH3N5 and CNN5, indicating that the reaction rate is slower, even when the activation energy is overcome. This could mean that DiN5 molecules are less likely to react in a given period, potentially making them more suitable for applications where stability is desired. The correlation coefficient (r² = 0.9087) indicates a good fit to the kinetic model, further validating the reliability of these findings.
HN5 has a moderate activation energy of 45.91 kJ/mol, placing it between CNN5 and DiN5 in terms of stability and reactivity. This suggests that HN5 requires a reasonable amount of energy to decompose, making it moderately reactive. The frequency factor of 6.56E+11 s⁻¹ indicates a moderate reaction rate, suggesting that the decomposition is neither too slow nor too rapid once the activation energy is surpassed. The high correlation coefficient (r² = 0.9463) indicates very reliable kinetic data, supporting the consistency of the reaction’s behavior with the Arrhenius equation.
PolyN5 has the highest activation energy of 52.88 kJ/mol, indicating that it is the most stable of the N5 - containing structures studied. This high energy requirement for decomposition makes PolyN5 particularly resistant to reaction under normal conditions, which could make it valuable in applications where high stability is crucial. Despite its high activation energy, the frequency factor, 8.10E+11 s⁻¹, is moderate, suggesting that the reaction rate is reasonable once the activation barrier is overcome. The very high correlation coefficient (r² = 0.9584) indicates the most reliable kinetic data among the structures, reflecting the strong adherence of PolyN5’s reaction kinetics to the Arrhenius model.
When comparing these structures, it is evident that the activation energy and frequency factor play crucial roles in determining the reactivity and stability of the N5 -containing species. CNN5, with its low activation energy, stands out as the most reactive, while PolyN5, with the highest activation energy, is the most stable. CH3N5, DiN5, and HN5 fall between these extremes, with varying degrees of stability and reactivity. The correlation coefficients for all the structures suggest that the kinetic models used to calculate these parameters are highly reliable, providing confidence in the predictive power of these values for future studies and potential applications. The behavior of activation energy likely correlates with the electron-donating and withdrawing effects of CH₃ and CN groups. Electron-donating groups like CH₃ can increase the electron density on adjacent atoms, potentially lowering the stability and reducing the activation energy needed for reactions, which aligns with CH₃N₅ being less stable and more reactive than PolyN₅. Conversely, the CN group, as an electron-withdrawing group, tends to stabilize the structure by delocalizing electron density, which could reduce reactivity by increasing energy activation.
This reasoning supports the observed reactivity trends, with CNN₅ (having a CN group) exhibiting a lower activation energy and higher reactivity than structures stabilized by electron-donating groups. The varying activation energies reflect how the substituents impact stability and reactivity through their electronic effects, as seen in CH₃N₅, DiN₅, and HN₅.
CONCLUSIONS
The N5 - anion is a highly energetic material with the potential for use in various applications, such as rocket propulsion, explosive devices, and pyrotechnics. The unique structure and high energy content of the N5- anion make it an attractive material for these applications. However, the reactivity of the N5- anion also poses inherent risks, which must be carefully considered in the design and use of the material.
Reactive molecular dynamics simulations are a powerful tool for investigating the behavior of energetic materials at the molecular level. Recent advances in RMD simulations have led to a better understanding of the reactivity and behavior of the N5- anion. RMD simulations have shown that the decomposition of the N5- anion is a complex process that is highly dependent on external stimuli. The use of RMD simulations in the study of the N5 - anion can be used to predict the design of safer and more efficient energetic materials.
Overall, the N5 - anion represents a promising avenue for the development of high-energy materials, and further research into its properties and behavior will be important for its safe and effective use in various applications.
SUPPLEMENTARY MATERIAL
REFERENCES
-
GONÇALVES RFB, IHA BKV, ROCCO JAFF & KUZNETSOV AE. 2022. Reactive molecular dynamics of pyrolysis and combustion of alternative jet fuels: A ReaxFF study. Fuel 310B: 122157. https://doi.org/10.1016/j.fuel.2021.122157
» https://doi.org/10.1016/j.fuel.2021.122157 -
HAMILTON BW, STEELE BA, SAKANO, MN, KROONBLAWD MP, KUO IFW & STRACHAN A. 2021. Predicted Reaction Mechanisms, Product Speciation, Kinetics, and Detonation Properties of the Insensitive Explosive 2,6-Diamino-3,5-dinitropyrazine-1-oxide (LLM-105). J Phys Chem A 125(8): 1766-1777. https://doi.org/10.1021/acs.jpca.0c10946.
» https://doi.org/10.1021/acs.jpca.0c10946 -
HOU T, XU Y, XU Z, ZHANG X & LU M. 2024. Syntheses, crystal structures, and energetic properties of four cyclo-pentazolate salts [NHn(CH3)4-n]+N5− (n = 0~3). J Mol Struct 1303: 137500. https://doi.org/10.1016/j.molstruc.2024.137500.
» https://doi.org/10.1016/j.molstruc.2024.137500 -
LANG Q, JIANG S, XU Y & LU M. 2022. Theoretical study on N‐oxide pentazolate high‐energy‐density materials: Toward excellent energetic performance and good stability. J Phys Org Chem 35(7): 12-21. https://doi.org/10.1002/poc.4342.
» https://doi.org/10.1002/poc.4342 -
LI XH, ZHAO FQ, XU SY & JU XH. 2013. Molecular dynamics simulation on miscibility of trans-1,4,5,8-tetranitro-1, 4,5,8 -tetraazadecalin (TNAD) with some propellants. J Mol Model 19(6): 2391-2397. https://doi.org/10.1007/s00894-013-1786-z.
» https://doi.org/10.1007/s00894-013-1786-z -
LIN Q, WANG P, XU Y & LU M. 2020. Pentazolate Anion Cyclo-N5-: Development of a New Energetic Material. Engineering 6(9): 964-966. https://doi.org/10.1016/j.eng.2020.04.011.
» https://doi.org/10.1016/j.eng.2020.04.011 -
MENDONÇA FB, GONÇALVES RFB, URGESSA GS, IHA K, DOMINGUES M & ROCCO JAFF. 2017. Computational Chemistry Employment in Verification and Validation of the Detonation Pressure of Plastic Explosive-PBX. J Chem News 41(3): 310-314. https://doi.org/10.21577/0100-4042.20170169.
» https://doi.org/10.21577/0100-4042.20170169 - MENDONÇA FB & URGESSA GS. 2023. Deflagration to detonation: transitions during solid rocket motor burning. In: Rocco JAFF, Gonçalves RFB & Domingues MG (Eds), Engineering of Solid Rocket Motors - From Production to Firing, 1st ed., Cambridge Scholars Publishing, p. 224-246.
-
NAZIN GM & KORSUNSKIY BL. 2021. Determination of the Thermal Stability of Explosives. Russ J Phys Chem B 15(2): 271-277. https://doi.org/10.1134/S1990793121020093
» https://doi.org/10.1134/S1990793121020093 -
PLIMPTON S. 1995. Fast parallel algorithms for short-range molecular dynamics. J Comp Phys 117: 1-19. https://doi.org/10.1006/jcph.1995.1039.
» https://doi.org/10.1006/jcph.1995.1039 -
SENFTLE T ET AL. 2016. The ReaxFF reactive force-field: development, applications and future directions. npj Comput Mater 2: 15011. https://doi.org/10.1038/npjcompumats.2015.11.
» https://doi.org/10.1038/npjcompumats.2015.11 -
STEELE BA & OLEYNIK II. 2017. Pentazole and Ammonium Pentazolate: Crystalline Hydro-Nitrogens at High Pressure. J Phys Chem A 121(8): 1808-1813. https://doi.org/10.1021/acs.jpca.6b12900.
» https://doi.org/10.1021/acs.jpca.6b12900 -
VAN DUIN ACT, DASGUPTA S, LORANT F & GODDARD WA. 2001. ReaxFF: A Reactive Force Field for Hydrocarbons. J Phys Chem A 105(41): 9396-9409. DOI: https://doi.org/10.1021/jp004368u.
» https://doi.org/10.1021/jp004368u -
WU J, YANG L, LI Y, SULTAN M, GENG D & CHEN L. 2020. Microscopic Mechanisms of Femtosecond Laser Ablation of HMX from Reactive Molecular Dynamics Simulations. J Phys Chem C 124(21): 11681-11693. https://doi.org/10.1021/acs.jpcc.0c01567.
» https://doi.org/10.1021/acs.jpcc.0c01567 -
YANG C, CHEN L, WU W, ZHANG C, SUN C, DU Y & HU B. 2021. Investigating the Stabilizing Forces of Pentazolate Salts. ACS Appl Energy Mater 4(1): 146-153. https://doi.org/10.1021/acsaem.0c02052.
» https://doi.org/10.1021/acsaem.0c02052




