Introduction

Energetic materials with large amounts of stored energy that can be released under specific conditions are of crucial importance and have many uses, including as explosives, propellants and pyrotechnics1,2,3,4,5,6,7. The most widely applied energetic materials are neutral compounds, such as 2,4,6-trinitrotoluene (TNT), 2,4,6-trinitrophenol (TNP), 1,3,5-trinitro-1,3,5-triazacyclohexane (RDX) and octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX)8, (Fig. 1), although these compounds suffer from their requirement for a multistep synthesis and their sensitivity9,10. In addition, the starting materials, e.g., methylbenzene, are non-renewable petrochemicals. Development and use of renewable starting materials has significant importance in sustainable development and recovery of waste resources11,12, and use of bio-renewable feed can avoid the complicated coal/petro-chemical process of precursors13,14,15. Biomolecules have been used as raw materials for the preparation of neutral bio-energetic materials (NBEMs), which are composed mainly of sugar alcohol-based energetic compounds (including nitroglycerin (NG), xylitol pentanitrate (XPN), mannitol hexanitrate (MHN), and nitrocellulose (NC))16 (Fig. 1). Bio-energetic materials have received renewed attention in recent years17,18,19,20,21. Certain NBEMs derived from sugar alcohols and glycine were reported by Klapötke et al.22,23. However, studies on this bio-based species focus primarily on neutral compounds.

Figure 1
figure 1

(a) Structures of NEMs from chemicals. I: TNT, II: PA, III RDX. (b) Structures of common IEMs from chemicals. IV: 1,5-Diamino-4-H−1,2,3,4-tetrazolium perchlorate, V: 1,2,3-triazolium nitrate, VI: 1-Allyl-3-methylimidazolium dicyanamide. (c) Structures of common NBEMs from biomolecules. VII: NG, VIII: XPN.

Ionic energetic materials (IEMs) including energetic salts and energetic ionic liquids have drawn extensive attention due to their designability and little/no vapor toxicity compared with the traditional NEMs24,25,26,27. IEMs, which are usually composed of heterocyclic cations (e.g., imidazolium, triazolium, tetrazolium cation, etc.) and energetic anions (e.g., nitrate (NO3 ), perchlorate (ClO4 ), dicyanamide (DCA) anion and bulky anions with one or more energetic groups, such as −NH2, −N3, and −CN, etc.)28,29,30,31,32, have demonstrated their excellent performance and great development potential in propellants and explosives (Fig. 1), but the formation materials are too sensitive and contain too many functional groups, which make their synthesis highly difficult and expensive33. Many IEM tasks remain to facilitate the development of promising energetic materials involving careful selection of precursors and synthetic protocols.

Amino acids, which are the building blocks of peptides and proteins, are common and low-cost bio-renewable molecules with amino (‒NH2) and carboxylic (‒COOH) functional groups. Production of amino acids can be accomplished by hydrolysis of proteins, of which approximately 108 tons are wasted each year34. Amino acids belong to the natural carbon pool, which is the intrinsic source of fuels and propellants. For example, glycine was previously incorporated as a fuel in the HAN-based propellant (HAN: hydroxylammonium nitrate) and also used as a raw material to construct neutral highly energetic oxidizers19,22,35,36,37. The energetic properties of amino acid salts/ionic liquids directly derived from amino acids have been rarely examined, although their characteristics as catalysts, absorbents and chiral reagents have been studied38,39,40,41,42,43. The possibility of developing ionic bio-energetic materials (IBEMs) from amino acid frameworks is systematically discussed in this work.

Results and Discussion

A series of IBEMs, including amino acid salts/ionic liquids [AA]ClO4 and [AA]NO3 (Fig. 2), were synthesized from natural amino acids and oxygen-rich perchloric acid (HClO4) or nitric acid (HNO3) in one step without any byproducts38. The synthesis protocol is a typical “atom economic reaction” in water at ambient temperature and pressure. This simple procedure and the use of biomaterials have the advantages of energy conservation, an abundant renewable source and low costs34. The resulting IBEMs [AA]ClO4 and [AA]NO3 could serve as a novel framework for safer bio-energetic materials. These materials were identified by nuclear magnetic resonance spectrometry (NMR) and Fourier transform infrared spectrometry (FTIR). Colorless crystals of [Gly]ClO4 and [Ala]NO3 suitable for single-crystal X-ray analysis were also obtained.

Figure 2
figure 2

Structures of energetic salts/ionic liquids [AA]ClO4 and [AA]NO3 derived from amino acids.

The combustibility test is an essential and simple test used to obtain a first impression of the energetic behavior of a compound44. The combustible characteristics of IBEMs, amino acids and several conventional NEMs were examined by heating 10 mg samples with a flame in air. The two IBEMs [AA]ClO4 and [AA]NO3 in this study can be ignited and display good combustibility (Fig. 3). Stable combustion was sustained until all samples were exhausted. It can be observed from the set of combustion images (Figures S6S10) that [AA]ClO4 exhibited quick and violent burning with notably intense fire. This fierce combustion might result from the high reactivity of ClO4 and the good oxygen balance of [AA]ClO4. Sustained and mild combustion was observed when [AA]NO3 was burned (Figures S11S13). In contrast, eight amino acids used as precursors in this work were difficult to ignite or produced faint combustion with tiny flames. The improved combustibility characteristic of [AA]ClO4 and [AA]NO3 should originate from the energetic anion species. Thus, the oxygen-rich energetic anions ClO4 and NO3 are favorable candidates for constructing IBEMs. Additionally, not all anions can improve the combustion of amino acid salts/ionic liquids, similar to the results from the contrast experiments. For example, prolinium hydrochloride ([Pro]Cl), prolinium hydrobromide ([Pro]Br), prolinium trifluoroacetate ([Pro]TFA), prolinium bisulfate.

Figure 3
figure 3

Combustion tests of 10 mg samples.

([Pro]HSO4) and prolinium trifluoromethanesulfonate ([Pro]OTf) were all assessed under the same condition, and all present results analogous to those of the amino acids, which indicates that the corresponding Cl, Br, TFA, HSO4 and OTf anions are not suitable for designing IBEMs. The combustibility performances of TNT, TNP, RDX and HMX were also tested and showed sustainable combustion with a mild flame, corresponding to their popular applications in propellants and explosives. As a comparison, the combustion of most IBEMs in this work is clearly more vigorous than that of TNT, TNP, RDX and HMX.

Thermal stability is vital to the performance of energetic materials. [AA]ClO4 and [AA]NO3 are thermally stable at room temperature and can be maintained without any decomposition for more than two years. [AA]ClO4 exhibit good thermal stability with decomposition temperatures (T d) greater than 210 °C, which exceed the criterion of 200 °C45,46 (Table 1). The highest thermal stability out of all samples is found for [Gly]ClO4 (263 °C). The T d values of [AA]NO3 are in the range of 125–183 °C, which are far beyond that of the typical NBEM and NG (50 °C). The melting points (T m) or glass transition temperatures (T g) of [AA]ClO4 and [AA]NO3 are between −46 °C and 159 °C. The initial melting point analysis reveals that [AA]ClO4 materials generally have significantly lower melting points than their nitrate analogues, among which many can be classified as ionic liquids. The thermal data also showed that certain IBEMs display a liquidus range at temperatures above 150 °C. The low T m and good thermal stability indicate that these materials are easily molded, similar to TNT. Differential thermal analysis (DTA) is a routine method used to characterize energetic materials17,18. Heat flow was measured under a nitrogen environment at a heating rate of 10 °C min−1. The DTA profile of [Gly]ClO4 and [Gly]NO3 is illustrated in Fig. 4. A notably large and sharp exothermic peak of [Gly]ClO4 is observed at a peak temperature of approximately 273 °C, which could be related to the oxidation-reduction reaction between glycine cation and ClO4 . The DTA curve of [Gly]NO3 shows a sharp exothermic peak near 182 °C and a sharp endothermic peak near 145 °C due to [Gly]NO3 melting, which corresponds to the melting point (149 °C) observed by DSC. [Gly]ClO4 displays a higher exothermic temperature than [Gly]NO3, probably because ClO4 is more stable than NO3 . The DTA curves of the remaining [AA]X are presented in the Supplementary Information.

Table 1 Physicochemical properties of IBEMs and reference compounds.
Figure 4
figure 4

DTA profile displaying the heat flow of [Gly]ClO4 and [Gly]NO3.

A high density value is a profitable reference for energetic materials. Interestingly, the obtained perchlorates or nitrates are denser than their precursors. The densities of [Ala]ClO4 and [Ala]NO3 are 1.69 and 1.47 g cm−3, respectively, which are higher than that of alanine (1.42 g cm−3) (Table 1). The densities of [Ala]ClO4 (1.69 g cm−3), [Ser]ClO4 (1.66 g cm−3) and [Asp]ClO4 (1.68 g cm−3) are superior to those of TNT (1.65 g cm−3) and NG (1.59 g cm−3). The density of [Gly]ClO4 can reach 1.89 g cm−3, which outperforms the currently used energetic materials, such as RDX (1.82 g cm−3) and TNP (1.76 g cm−3), and is similar to HMX (1.91 g cm−3).

The standard enthalpy of formation (Δf H°) and heats of combustion (Δc H°) of [AA]ClO4 and [AA]NO3 were calculated using the Gaussian 09 suite of programs47, as described in our previous studies48,49. The theoretical data −Δf H° range is 3.46 to 6.02 kJ g−1, whereas that of [Leu]ClO4 is the lowest (3.46 kJ g−1). The theoretical data −Δc H° range is 5.29 to 18.09 kJ g−1. The experimental heats of combustion of six amino acid salts/ionic liquids ([Gly]ClO4, [Ala]ClO4, [Val]ClO4, [Ile]ClO4, [Leu]ClO4, and [Pro]ClO4) were determined to verify the accuracy of the theoretical results, and the relative error is less than 5%. Therefore, the theoretical results are in good agreement with the experimental results. Most of the −Δc H° values of IBEMs exceed that of NG (6.73 kJ g−1). The −Δc H 0 values of TNP (12.47 kJ g−1) are lower than those of [Val]ClO4, [Pro]ClO4, [Val]NO3 and [Pro]NO3. [Ile]ClO4 (15.46 kJ g−1), [Leu]ClO4 (15.48 kJ g−1), [Ile]NO3 (18.07 kJ g−1), and [Leu]NO3 (18.09 kJ g−1) possesses high −Δc H° values above that of TNT (15.00 kJ g−1). The calculated detonation pressures (P) of [AA]ClO4 and [AA]NO3 lie between 14.04 GPa and 32.26 GPa. The calculated detonation velocities (D) lie in the range from 5844 to 8470 m s−1, certain of which are greater than those of TNT (6881 m s−1), TNP (7350 m s−1) and NG (7600 m s−1). [Gly]ClO4 has the highest detonation pressure of 32.26 GPa and the highest detonation velocity of 8470 m s−1, which are comparable to those of RDX (34.9 GPa, 8748 m s−1). Because of the different structures of the amino acid cations, diverse oxygen balances (OB) are based on CO in a range of −65.91 to 13.67. Perchlorate displays good oxygen balance, and the OBCO values of [Gly]ClO4 (13.67), [Ser]ClO4 (3.89) and [Asp]ClO4 (3.43) are positive. The sensitivities to impact of [AA]ClO4 and [AA]NO3 exceed 60 J. The two [AA]ClO4 and [AA]NO3 are stable at 360 N in the friction sensitivity test. The sensitivities of all IBEMs including [AA]ClO4 and [AA]NO3 are better than those of NBEM NG (0.2 J, 1 N). Based on the UN standard, all salts show both impact and friction insensitivity characteristics. The impact and friction sensitivities are remarkably good, especially for perchlorate salts because they are usually sensitive29,30.

The hydrogen bond plays an important role in increasing the density and decreasing the sensitivity of energetic compounds50,51. The structures of [Gly]ClO4 and [Ala]NO3 were characterized by single-crystal X-ray diffraction (Fig. 5). [Gly]ClO4 crystallizes in the monoclinic space group P21/c with four cation and four anion moieties in each unit cell. The structure is dominated by the interactions between cations and anions and hydrogen bonds (Fig. 5b). Each cation forms hydrogen bonds to three anions and another cation via the ammonium cation and carboxyl group. The donor-acceptor contact distances are in the range of 2.7422(18) to 2.9266(2) Å. The shortest H-acceptor distance starting from O(4) of the carboxyl group directly linked to O(8) of perchlorate is 2.7422(18) Å (O(4)−H(4)···O(8)). [Ala]NO3 crystallizes in the orthorhombic space group P212121 with four cation and four anion moieties in the unit cell. The packing structure of [Ala]NO3 is built up by hydrogen bonding and interactions between cations and anions along the a axis (Fig. 3d). Each NO3 anion is surrounded by three alaninium cations oriented toward the oxygen atoms (O3, O5). The donor-acceptor contact distance range is 2.618(6) to 2.893(6) Å.

Figure 5
figure 5

(a) Molecular structure of [Gly]ClO4. (b) Packing diagram of [Gly]ClO4 viewed down the a-axis. (c) Molecular structure of [Ala]NO3. (d) Packing diagram of [Ala]NO3 viewed down the a-axis. The unit cell is indicated, and the dashed lines represent hydrogen bonding. Thermal ellipsoids are drawn at the 50% probability level.

The localized orbital locator (LOL) that is dependent on the kinetic-energy density reveals the electronic shell structure of the compounds52. A large LOL value indicates that electrons are greatly localized, meaning that a covalent bond forms, with a lone pair or inner shells of an atom involved. The electron localization of [Gly]ClO4 and [Ala]NO3 cluster conformers where the NH3 + group connects directly to the ClO4 anion are shown in Figs 6 and 7, respectively53. Covalent regions (e.g., C–C, C–N, and N–H) have high LOL values. The regions between the NH3 + group and ClO4 anion have a low LOL value. indicating that [Gly]ClO4 is a classic ionic compound, and the low LOL value also implies that a hydrogen bond exists in the regions between the NH3 + group and ClO4 anion54,55. The electron localization of [Ala]NO3 cluster conformers is similar to that of [Gly]ClO4 cluster conformers.

Figure 6
figure 6

Color-filled map of localized orbital locator (LOL) of the cluster conformers in [Gly]ClO4.

Figure 7
figure 7

Color-filled map of localized orbital locator (LOL) of the cluster conformers in [Ala]NO3.

To explain the H-bonding, the quantum theory of atoms in molecules (QTAIM) analysis was used56,57,58. The topological criteria for the existence of hydrogen bonding is the bond path between the hydrogen atom and acceptor with the bond critical point (BCP) occurring at the minimum electron density ρ(r) (ρ BCP) and an appropriate gradient, the Laplacian of the electron density 2 ρ(r) (2 ρBCP) (Fig. 8). Table 2 lists the H-bonds as determined by the existence of a BCP and identifies these as weak and medium H-bonds in the [Gly]ClO4 crystal. The O4–H4–O8 bond with ρ BCP = 0.0247 e bohr−3 (2 ρBCP = 0.1074 e bohr−5) lies well within the medium H-bonding range. These hydrogen bonds in the crystal might contribute to closer packing to obtain the high density of [Gly]ClO4. Table 3 lists the H-bonds as determined by the existence of a BCP and identifies these as medium and strong H-bonds in the [Ala]NO3 crystal. The O6–H6–O3 bond with ρ BCP = 0.0424 e bohr−3 (2 ρBCP = 0.1323 e bohr−5) lies well within the strong H-bonding range. The natural bond orbital (NBO) analysis is also a key approach to the study of H-bonding57,58. The interactions between the empty acceptor-hydrogen (A–H) antibonding orbital (σ*) and the filled lone pair orbital (n) on the donor (D) are shown in Fig. 8. The orbital overlap represents the charge transfer from the occupied donor lone-pair orbital into the empty antibonding orbital. The stabilization energy E(2) (n→σ*) associated with the amount of electron density donated from the occupied donor lone-pair orbital to the empty antibonding orbital is obtained. The values for E(2) (n→σ*) in [Gly]ClO4 range from 10.71 to 30.19 kJ mol−1 (Table 3). The shortest hydrogen bond (O4–H4–O8) has the highest value of 30.19 kJ mol−1 in the [Gly]ClO4 crystal, indicating medium H- bonding, which is in agreement with the QTAIM analysis. The E(2) (n→σ*) value in [Ala]NO3 is 9.07 to 87.70 kJ mol−1, whereas that of O6–H6–O3 bond is the highest (87.80 kJ mol−1). In [Gly]ClO4, each anion forms hydrogen bonds to three glycinium cations with E(2) 30.19, 10.71 and 13.72 kJ mol−1, respectively, which might help to stabilize the irascible ClO4 and decrease the sensitivity of [Gly]ClO4. A similar situation is also present in [Ala]NO3.

Figure 8
figure 8

QTAIM and NBO analysis of the cluster conformers in [Gly]ClO4 (a) and [Ala]NO3 (b) crystal structure at the M062X/6–311 + + G(d,p) level. Larger spheres represent the atoms (C: grey, H: white, N: blue, O: red, Cl: light grey). Thick black lines indicate the bond paths of hydrogen bonds. The small red dots on the bond paths identify the BCPs. The natural bond orbitals isosurface is plotted at –0.05 a.u. and 0.05 a.u., respectively.

Table 2 Hydrogen bonds in [Gly]ClO4 crystal.
Table 3 Hydrogen bonds in [Ala]NO3 crystal.

The toxicity of energetic materials is an important concern for the public. To assess the acute lethal toxicity of IBEMs on aquatic animals, diluted aqueous solutions of selected high-quality energetic amino acid salts/ionic liquids were administered to the model species Macrobrachium nipponense. Macrobrachium nipponense is a dominant species in the freshwater and stream ecosystems and is a sensitive bio-indicator of pollutants59. The acute lethality curves of [Gly]ClO4, [Pro]ClO4, [Val]NO3 and [Ile]NO3 with respect to Macrobrachium nipponense are shown in Fig. 9. The mortality increases with the initial concentration of IBEMs. The median lethal concentrations (LC 50), the most important toxicological parameter, were calculated according to the lethal rates of Macrobrachium nipponense. The value of LC 50 for an examined compound is the dose required to kill half of the members of a tested population after a specified test period. It is obvious that the longer the test duration, the shorter are the median lethal concentrations (24 h, 48 h, 72 h and 96 h LC 50) of the four IBEMs to Macrobrachium nipponense, as presented in Table 4. The 96 h LC 50 values for [Gly]ClO4, [Pro]ClO4, [Val]NO3 and [Ile]NO3 to Macrobrachium nipponense are 367.65 (271.40–484.66), 648.80 (527.53–760.12), 471.86 (367.02–603.80) and 523.02 (412.76–647.49) mg L−1, which exceed the low toxicity critical value of 100 mgL−160. These IBEMs are apparently hypotoxic to Macrobrachium nipponense in the acute lethal toxicity assay, although in large amounts, perchlorate interferes with iodine uptake into the thyroid gland61,62.

Figure 9
figure 9

IBEM acute lethality curves of Macrobrachium nipponense determined at different IBEM concentrations in the exposed environment.

Table 4 Median lethal concentrations (LC 50) to Macrobrachium nipponense.

Conclusion

In conclusion, sixteen amino acid salts/ionic liquids, including [AA]ClO4 and [AA]NO3 materials, were systematically studied as ionic bio-energetic materials (IBEMs). We present a new approach for the development of bio-energetic materials. [AA]ClO4 and [AA]NO3 were easily prepared from bio-renewable amino acids and possessed high densities as well as low sensitivities. The combustion of these compounds is more intense than that of certain conventional NEMs. The energetic properties of several IBEMs are comparable to those of TNT, TNP, RDX and HMX. These IBEMs preserve the benign environment from the natural source and display excellent hypotoxicity. Our work suggests that IBEMs are promising candidates for bio-based energetic materials.

Experimental Section

General Methods

All chemicals were obtained commercially as analytical-grade materials and used as received. Solvents were dried using standard procedures. The ionic bio-energetic materials [AA]X were synthesized by direct acidification from amino acids and corresponding acids in water for 24 h and subsequently dried by evaporation of water in air followed by vacuum drying, as described in our previous studies38. Infrared spectra (IR) were recorded on a NEXUS 670 FT-IR spectrometer on KBr pellets. 1H and 13C NMR spectra were recorded on a Bruker 400 MHz nuclear magnetic resonance spectrometer operating at 400 and 100 MHz, respectively, with d 6-DMSO as the locking solvent. The 1H and 13C chemical shifts are reported in ppm relative to TMS. Coupling constants are given in Hertz. Elemental analyses (H, C, N) were performed on an Elementar Vario MICRO CUBE elemental analyzer. Decomposition temperatures were characterized using a thermogravimetric analyzer (TGA) on a NETZSCH TG 209F1 calorimeter. The heat flow of these materials was obtained by differential thermal analysis (DTA) using a NETZSCH TG 209F1 calorimeter. Measurements were accomplished by heating the samples at a heating rate of 10 °C min−1 from 25 to 600 °C. Melting points were determined by differential scanning calorimetry (DSC) on a TA Q20 calorimeter calibrated with standard pure indium. Measurements were performed at a heating rate of 10 °C min−1 with a nitrogen flow rate of 20 mL min−1. The reference sample was an Al container with nitrogen. The densities were measured at 25 °C using a pycnometer. The experimental enthalpy of combustion was measured using a Parr 6725 bomb calorimeter (static jacket) equipped with a Parr 207 A oxygen bomb. Initial safety testing of amino acids salts/ionic liquids with respect to impact and friction were performed using the BAM method. The sensitivity towards impact (IS) was tested by the action of a falling weight from different heights. The friction sensitivity (FS) was determined by rubbing a small amount of material between a porcelain plate and a pin with different contact pressures. Macrobrachium nipponense were obtained from local commercial suppliers and transported to a plastic aquarium equipped with an air pump to aerate the water. The temperature was maintained at 20.0 ± 0.5 °C. Macrobrachium nipponense were acclimated for one week and those with body length 5 ± 0.5 cm were used in acute toxicity tests in the initial experiments. Stock solutions were prepared in deionized water. Laboratory tests were conducted to determine the median lethal concentration (LC 50) for Macrobrachium nipponense. Ten animals were randomly sampled and placed in plastic beakers. After 24 h acclimatization, Macrobrachium nipponense were exposed to different concentrations of [Gly]ClO4, [Pro]ClO4, [Val]NO3 and [Ile]NO3 for 96 h. During the experiment, dead animals were removed over time. Mortality was recorded after 24, 48, 72 and 96 h. The LC 50 values of every test chemical with 95% confidence limits were calculated for Macrobrachium nipponense using the probit analysis63.

X-ray Crystallography

Single crystals of [Gly]ClO4 were removed from the flask, and a suitable crystal was selected and attached to a glass fiber. The data were collected by a New Gemini, Dual, EosS2 diffractometer with graphite-monochromated Mo-Kα radiation (λ = 0.71073 Å). Single crystals of [Ala]NO3 were removed from the flask, and a suitable crystal was selected and attached to a glass fiber. The data were collected by a New Gemini, Dual, EosS2 diffractometer with graphite-monochromated Cu-Kα radiation (λ = 1.54184 Å). The crystal was held at 293 K during data collection. Using Olex264, the structure was solved with the ShelXT65 structure solution program using Direct Methods and refined with the XL66 refinement package using least squares minimization. All non-hydrogen atoms were refined anisotropically, and hydrogen atoms were located and refined. No decomposition was observed during data collection. Crystal data and structure refinement for [Gly]ClO4 and [Ala]NO3 are given in Table 5. Details of the data are given in Tables S1S4.

Table 5 Crystal data and structure refinement for [Gly]ClO4 and [Ala]NO3.

Theoretical Study

Computations for the heat of combustion were performed using the Gaussian09 (Revision A.02) suite of programs47. The geometric optimization and frequency analyses were performed using Møller-Plesset second-order perturbation theory truncated at the second order (MP2) with the 6–311++G** basis set67,68. All of the optimized structures were characterized as true local energy minima on the potential energy surface without imaginary frequencies. The theoretical heats of combustion were simplified using the expression:

$${{\rm{\Delta }}}_{{\rm{c}}}H^\circ ([{\rm{AA}}]X,\,298\,{\rm{K}})={{\rm{\Sigma }}{\rm{\Delta }}}_{{\rm{f}}}H^\circ \,(\text{product},\,\text{298}.15\,{\rm{K}})-{{\rm{\Sigma }}{\rm{\Delta }}}_{{\rm{f}}}H^\circ (\text{reaction},\,\text{298}.15\,{\rm{K}})$$
(1)

The theoretical heats of formation of the cations and anions were computed using the method of protonation reactions (Figure S1). The sources of the energies of the parent ions in the isodesmic reactions were calculated from protonation reactions Δf H° (H+) = +1528.085 kJ mol−1 69. The enthalpies of reaction (Δr H°, 298.15) were obtained by combining the MP2/6–311++G** energy differences for the reaction, the scaled zero point energies, and other thermal factors. The theoretical heats of formation of the [AA]ClO4 and [AA]NO3 at T = 298.15 K were calculated based on a Born-Haber energy cycle (Figure S2). In a Born-Haber energy cycle, the heat of formation of an AAIL can be simplified by the following expression:

$${{\rm{\Delta }}}_{{\rm{f}}}H^\circ ([{\rm{AA}}]{\rm{X}},\,298\,{\rm{K}})={{\rm{\Sigma }}{\rm{\Delta }}}_{{\rm{f}}}H^\circ \,({\rm{cation}},\,298.15\,{\rm{K}})\,+\,{{\rm{\Sigma }}{\rm{\Delta }}}_{{\rm{f}}}H^\circ \,({\rm{anion}},\,298.15\,{\rm{K}})\,-\,{{\rm{\Delta }}}_{{\rm{f}}}{H}_{{\rm{L}}}$$
(2)

where Δf H L is the lattice energy of the ionic salt, and Δf H L (kJ mol−1) can be predicted by the formula suggested by Jenkins et al.70:

$${{\rm{\Delta }}}_{{\rm{f}}}{H}_{{\rm{L}}}={U}_{{\rm{POT}}}+[p({{\rm{n}}}_{{\rm{M}}}/2\,-\,2)+{\rm{q}}({{\rm{n}}}_{{\rm{X}}}/2\,-\,2)]RT$$
(3)

where nM and nX depend on the nature of the ions Mp+ and Xq−, respectively, and have a value of six for nonlinear polyatomic ions. The equation for lattice potential energy U POT has the form:

$${U}_{{\rm{POT}}}=\gamma {({\rho }_{{\rm{m}}}{/{\rm{M}}}_{{\rm{m}}})}^{1/3}+\delta $$
(4)

where ρ m is the density (g cm−3) and Mm is the chemical formula mass of the [AA]ClO4 and [AA]NO3. For MX (1:1) salts, γ is 1981.2 and δ is 103.8. Thus, U POT of [AA]ClO4 and [AA]NO3 is given by:

$${U}_{{\rm{POT}}}=1981.2{({\rho }_{{\rm{m}}}{/{\rm{M}}}_{{\rm{m}}})}^{1/3}+103.8$$
(5)

The LOL, QTAIM and NBO analysis of the cluster conformers in the [Gly]ClO4 crystal structure at the M062X/6–311++G (d, p) level was performed by the Gaussian09 (Revision A.02) suite of programs47. The Gaussian output wfn files were used as inputs for Multiwfn to perform the QTAIM analysis53. The LOL analysis map was drawn by Multiwfn. The AIM topological analysis diagram was drawn by Multiwfn and VMD71. The NBO was plotted by Multiwfn and VMD using the Gaussian output fch files.

Detonation property

The detonation parameters were calculated using the Kamlet-Jacobs equation72:

$$D=(1.01+1.312{\rho }_{{\rm{m}}}){{\rm{\varphi }}}^{1/2}$$
(6)
$$P=1.558{{\rho }_{{\rm{m}}}}^{2}{{\rm{\varphi }}}^{1/2}$$
(7)
$${\rm{\varphi }}={{\rm{NM}}}^{1/2}{{\rm{Q}}}^{1/2}$$
(8)

where P is the detonation pressure (GPa), D is the detonation velocity (km s−1), ρ m is the packed density (g cm−3), ϕ is the characteristic value of explosives, N is the moles of gas produced per gram of explosives (mol g−1), M is an average molar weight of detonation products (g mol−1), and Q is the maximum estimation heat of detonation (cal g−1).