Anisotropic Impact Sensitivity of Metal-Free Molecular Perovskite High-Energetic Material (C6H14N2)(NH2NH3)(ClO4)3 by First-Principles Study.

Density functional theory simulations were carried out to investigate energetic molecular perovskite (C6H14N2)(NH2NH3)(ClO4)3 which was a new type energetic material promising for future application. The electronic properties, surface energy, and hydrogen bonding of (100), (010), (011), (101), (111) surfaces were studied, and the anisotropic impact sensitivity of these surfaces were reported. By comparing the values of the band gaps for different surface structures, we found that the (100) surface has the lowest sensitivity, while the (101) surface was considered to be much more sensitive than the others. The results for the total density of states further validated the previous conclusion obtained from the band gap. Additionally, the calculated surface energy indicated that surface energy was positively correlated with impact sensitivity. Hydrogen bond content of the surface structures showed distinct variability according to the two-dimensional fingerprint plots. In particular, the hydrogen bond content of (100) surface was higher than that of other surfaces, indicating that the impact sensitivity of (100) surface is the lowest.

Introduction

Energetic materials are a class of special reactive substances storing large amounts of chemical energy, including explosives, propellants, and pyrotechnics.[1,2] With the rapid development of modern military science and technology and aerospace industry, the requirements for energetic materials are also increasing. In order to adapt to a harsher environment and cope with various extreme conditions in the universe, the new generation of energetic materials must have good comprehensive performance including high energy, low sensitivity, low cost, and functional diversity.[3] Driven by both synthetic and theoretical research, new structures of energetic materials are emerging.[4,5] However, the high detonation performance and low sensitivity are mutually exclusive for most of explosives. Few explosives could meet the practical applications of weapons and equipment effectively. Therefore, designing new structures and exploring new concepts of energetic compounds are still the main research hotspots in the field of energetic materials.[6,7]

In recent years, molecular perovskites have played an important role in many areas, including solar cells,[8−10] thermoelectricity,[11] ferroelectrics,[12] piezoelectricity,[13] lasers,[14] and so on due to their unique structures and excellent properties.[15] Lately, Chen’s team[16−18] has designed and synthesized a series of high-energy-density compounds with molecular perovskite structure characteristics, which created a precedent for molecular perovskite in the field of energetic materials. Different from traditional energetic compounds, these energetic molecular perovskites are composed of special oxidative ions (like ClO4–) and organic cations (as fuel part). Most importantly, these energetic molecular perovskites have the advantages of excellent detonation performances, good stability, and low cost. The emergence of molecular perovskites high-energetic materials is undoubtedly of great significance to the development of energetic materials. Among them, the metal-free molecular perovskite (H2dabco)(NH4)(ClO4)3] (DAP-4 for short) has attracted great attention to researchers because of the high exothermic effect and high amount of gas.[19−24] Compared to DAP-4, the metal-free molecular perovskite (H2dabco)(NH2NH3)(ClO4)3 (DAP-7 for short)[18] not only possesses more excellent detonation performances and better thermal stability but has great potential to be used as a heat-resistant explosive.

Sensitivity is an important index to measure the stability of explosives, which is usually defined as the degree of difficulty of decomposition or explosion for energetic materials under external stimulation (impact, friction, etc.).[25] According to external stimulus factors, the sensitivity of explosives could be sorted into impact sensitivity, friction sensitivity, and electrostatic spark sensitivity. Among them, impact sensitivity is studied frequently in experiment and theory. Since it was discovered that different crystal orientations of pentaerythritol tetranitrate (PETN) correspond to different impact sensitivity,[26] people began to focus on investigating anisotropic impact sensitivity using theoretical simulations.[25,27−29] Theoretical studies have shown that starting from the surface structure was a very effective method to study the anisotropy of impact sensitivity.[30,31]

The anisotropy of impact sensitivity is relevant to the thermodynamics and chemical reactions of explosives, which is essential to understand the reaction/explosion mechanism.[26,32,33] However, there has not been any report on anisotropic impact sensitivity of DAP-7 since it was synthesized. The goal of this study was to explore anisotropic impact sensitivity of DAP-7 based on density functional theory. The (100), (010), (011), (101), and (111) surfaces were studied from three aspects of electronic properties, surface energy, and hydrogen bonding in this work. Moreover, the corresponding properties of bulk crystal for DAP-7 were also investigated to form a comparison with the surface structures. This research could provide important theoretical supports for designing and developing newly energetic materials.

Computational Methods

All calculations in this paper were performed by using the Vienna Ab initio Simulation Package (VASP) code[34,35] in the framework of density functional theory (DFT). The projector augmented wave (PAW) method was used to describe interaction between the core and valence electrons.[36] Here, the model starting structure adopted the experimental structure of DAP-7 crystallized in a monoclinic lattice with P21/m space group.[18] The unit cell of DAP-7 featured a hexagonal perovskite-type structure with the formula ABX3, where A-site cation was H2dabco2+, B-site cation was NH2NH3+, and X-site anion was ClO4–. The unit cell and ions structure of DAP-7 were displayed in Figure . The H(1s1), C(2s22p2), N(2s22p3), O(2s22p6), and Cl(3s23p5) were considered as the valence electrons.

Figure 1

(a) Crystal structure of DAP-7. (b) Ion composition. H atom in light gray, C atom in dark gray, N atom in blue, O atom in red, and Cl atom in green.

First of all, the experimental crystal structure of DAP-7 performed full geometry optimizations which allow lattice parameters and atomic positions to be fully relaxed. The cutoff energy of plane waves was set to 450 eV. Gamma-centered Monkhorst–Pack k-point mesh of 3 × 4 × 3 was used for sampling the Brillouin zone in geometry optimization. The convergence criteria of energy and force was less than 10–5 eV and 0.03 eV/Å, respectively. In the case of hybrid perovskite materials, more accurate results could be obtained by considering the van der Waals (vdW) correction.[37] In terms of exchange correlation function, there were four functions tested to choose a more appropriate method in geometry optimization, including the Perdew, Burke, and Ernzerh of (PBE) functional,[38] the PBE functional revised for solids (PBEsol),[39] PBEsol plus the addition of Grimme’s D3 dispersion correction (PBEsol+D3),[40] and the PBE functional plus the addition of D3 (PBE+D3).[41] The results are shown in Table . After that, they were cut along the (100), (010), (011), (101), and (111) crystal plane of the geometry-optimized crystal structure of DAP-7 and periodic conditions were applied to these surface models. The vacuum layer was set to 15 Å along the Z-direction, which was thick enough to avoid any interaction between the surfaces. To that end, the periodic surface models containing the (100), (010), (011), (101), and (111) surfaces could be obtained.

Table 1

Experimental and Calculated Lattice Parameters of DAP-7

methods	a (Å)	b (Å)	c (Å)	B (deg)	volume (Å3)
experiment[18]	10.378	8.050	10.587	117.99	781.034
PBE	10.783	8.042	11.0766	119.16	838.794
	(+3.90%)	(−0.10%)	(+4.63%)	(+0.98%)	(+7.39%)
PBE+D3	10.353	7.945	10.608	118.66	765.622
	(−0.24%)	(−1.31%)	(+0.20%)	(+0.56%)	(−1.97%)
PBEsol	10.486	7.839	10.753	119.00	773.057
	(+1.04%)	(−2.62%)	(+1.57%)	(+0.85%)	(−1.02%)
PBEsol+D3	10.191	7.799	10.432	118.62	727.850
	(−1.80%)	(−3.12%)	(−1.46%)	(+0.53%)	(−6.81%)

During the geometry optimizations of surface models, the cell parameters were fixed and only the atomic coordinates were optimized. Except for the K-point settings, the other parameters for geometry optimization, electronic properties, and surface energy calculations in the surface models were the same as those for the bulk crystal. The K points were set as Gamma-centered Monkhorst–Pack k-point mesh of 4 × 3 × 1, 3 × 3 × 1, 3 × 3 × 1, 3 × 4 × 1, and 3 × 3 × 1 during the geometry optimization of the surface models of (100), (010), (011), (101), and (111), respectively.

Results and Discussion

Crystal Structures and Surface Structures

The unit cell parameters of DAP-7 by experiment and DFT relaxed are tabulated in Table . It was found that the optimized structure by PBE+D3 was in good accordance with the experimental structure, whereas the PBE seemed to overestimate the lattice constants more drastically due to ignoring the van der Waals (vdW) correction. The errors of the lattice constants obtained by the methods of PBEsol and PBEsol+D3 were generally smaller than that of PBE but still larger than that of PBE+D3. Overall, the PBE+D3 functional was chosen as the optimal exchange correlation function in this paper. The unit cell structure optimized by PBE+D3 was used as the optimal structure. The optimized surface structures by PBE+D3 including (100), (010), (011), (101), and (111) were shown in Figure .

Figure 2

Different surface structures of DAP-7.

Electronic Structure

Band gap, the energy difference between the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO), is an important parameter to characterize the electronic structure of crystals.[42] In the field of energetic materials, the band gap was closely related to the impact sensitivity, which was reflected in the fact that the initiation of energetic materials was accompanied by rupture of chemical bonds requiring population of antibonding state. Therefore, the electronic excitation generally resulted in a weakening of chemical bonds. Hence, the smaller the band gap was, the easier the electronic was to reach antibonding orbital, and the easier the chemical bond was to break.[43,44] According to the first-principles band gap criterion, a smaller band gap corresponds to a higher impact sensitivity for the energetic crystal with similar structure or thermal decomposition mechanism.[45−47] The first-principles band gap criterion has been widely used to predict impact sensitivity of various energetic compounds.[46−49]

To investigate the influence of surfaces on impact sensitivity of DAP-7, the band gap of bulk crystal and (100), (010), (011), (101), and (111) surfaces were calculated, and the results are given in Table . The sequence of band gap in surface structures were (101) < (010) < (011) < (111) < (100), where the band gap value of the (100) surface was the largest and the band gap of the (101) surface was the smallest among these surfaces. The difference of band gap in surface structures of DAP-7 suggested that the impact sensitivities exhibit distinct anisotropic characteristics. According to the first-principles band gap criterion, it can be predicted that the impact sensitivities of surfaces in descending order were (101) > (010) > (011) > (111) > (100). The results showed that the (100) surface has the lowest impact sensitivity, whereas the (101) surface has the highest impact sensitivity. In addition, we found that the band gap values of these surfaces were smaller than that of the bulk crystal, suggesting that these surface structures show higher impact sensitivity than bulk crystal. This may result from the fact that the energetic molecules on the exposed surface show higher activities due to the reduced activation barriers.[50]

Table 2

Calculated Band Gap of Different Surfaces for DAP-7

contents	bulk	(100)	(010)	(011)	(101)	(111)
band gap (eV)	5.069	4.976	4.225	4.560	3.878	4.934

Electronic structure was an intrinsic factor that was critical in determining the impact sensitivities and chemical reactions of energetic materials. For the purpose of seeking more information about the impact sensitivity anisotropy from the electronic structure, the total density of states (TDOS) of bulk crystal and each surface of DAP-7 were investigated in this work, as depicted in Figure . Compared to the bulk crystal, the TDOS of these surfaces migrated toward the lower energy region. The conduction bands of different surface structures were arranged from the low-energy region to the high-energy region as (101), (010), (011), (111), and (100) respectively. Observing the band gaps of the five surface structures mentioned above, it could be demonstrated that the more the conduction bands of the TDOS migrate to the lower energy region, the smaller the band gap is. By taking into account the relationship between the impact sensitivity and band gap, it was presumed that the more the conduction bands of the TDOS move toward lower energy, the more sensitive the surface is.

Figure 3

DOS for bulk crystal and different surfaces of DAP-7.

Surface Energy

The formation of surfaces could change the periodic potential field of the material, which may affect the anisotropic impact sensitivity.[30] Here, the surface energy of different surfaces was calculated and then compared with the impact sensitivity quantified by the band gap to study the relationship between the surface energy and impact sensitivity.

The surface energy can be applied to evaluate the stability of each kind of low index surfaces,[51] and it can be calculated by the following equation[52−55]where Esurface indicates the surface energy, Etotal is the total energy of the surface model, Ebulk refers to the total energy of unit cell of DAP-7, A is the surface area of the surface, and N is the number of surface layers. The calculation results of surface energies in (100), (010), (011), (101), and (111) surfaces were 0.0482, 0.0808, 0.0678, 0.1437, and 0.0716 J/m2, respectively. The sequence of surface energy of five surface structures in descending order was as follows: (101) > (010) > (111) > (011) > (100). Additionally, it could be found that the sequence of the surface energy was broadly consistent with the sequence of the impact sensitivity discussed in the previous section. Therefore, we presumed that the greater surface energy is, the more sensitive the surface would be, because increasing surface energy could result in the more active molecules favorable to the decomposition and explosion of explosives.

Hydrogen Bond

There exists a large number of H···O (or O···H) hydrogen bonding interactions in the crystal of DAP-7, and these hydrogen bonding interactions facilitated the face-sharing B(ClO4)6 octahedrons to be connected thus forming a hexagonal packing structure.[18] Hence, hydrogen bonding played a predominant role in the formation and stabilities of DAP-7.

It is well-known that hydrogen bonding was an important factor affecting the structural stability, thermodynamic properties, and molecular packing of energetic compounds.[56,57] The content of hydrogen bond could affect the binding energy between molecules. The more hydrogen bond content there is, the more impact energy is required by explosives to break the hydrogen bond net.[58] Therefore, the higher the hydrogen bond content is, the lower the impact sensitivity is.[57−59] Two-dimensional fingerprint plots could display the type and proportion of intermolecular interactions, providing important information for predicting the impact sensitivity of energetic materials.[48,60−62]

Figure displays the fingerprint plots of bulk crystal and five surface structures. The percentage of O···H (H···O) interactions in the bulk crystal was higher than that of all surface structures, and the reason was that the existence of a vacuum layer in surfaces could weaken the intermolecular interactions. The percentages of O···H (H···O) in (100), (010), (011), (101), and (111) surfaces were 72.8%, 70.2%, 72.3%, 68.3%, and 63.0%, respectively. According to the relationship between hydrogen bonding and structural stability, it could be inferred that the sequence of impact sensitivity corresponding to the five surface structures is (111) > (101) > (010) > (011) > (100). Additionally, we could find that the result in this part was roughly consistent with that in the preceding part.

Figure 4

Fingerprint plots for bulk crystal and different surface structures of DAP-7.

Conclusions

Density functional theory has been performed to investigate the electronic properties, surface energy, and hydrogen bonding in different surface structures of DAP-7. The lattice constants of the optimized bulk crystal showed that the simulation results obtained by the PBE+D3 method were in good agreement with the experimental results. The calculated band gap values of the surface structures demonstrated that the impact sensitivity of different surfaces showed significant anisotropy. Especially, the impact sensitivity of (100) surface was considered to be the lowest among all surfaces. An analysis of the total density of states further validated the anisotropic properties obtained from the band gap, which was reflected by the varying magnitudes of migration of the total density of states on different surfaces relative to that of the bulk crystal. Since the order of surface energy was roughly consistent with the sequence of impact sensitivity quantified by band gap, we speculated that a larger surface energy corresponded to a higher impact sensitivity. The two-dimensional fingerprint plots indicated that the contents of hydrogen bonds in different surface structures showed significant variability. The (100) surface contained the most hydrogen bonds so that its impact sensitivity was the lowest. To summarize, the research on electronic properties, surface energy, and hydrogen bonding suggested that the impact sensitivity of different surface structures presents obvious anisotropy and the impact sensitivity of (100) surface is the lowest among all surface structures.