A novel bis-(triazolyl)-furoxan energetic compound—graphene-like crystal structure for balancing energy and sensitivity

Abstract

A novel bis-(1,2,4-triazolyl)-furoxan structure for balancing energy and sensitivity has been successfully achieved. Theoretical calculations and impact sensitivity tests show that bis-(2-methy-3-nitroamine-1,2,4-triazolyl)-furoxan and bis-(2-methy-3-amine-1,2,4-triazolyl)-furoxan of two energetic compounds exhibit outstanding denotation performance (8.1 km s⁻¹) and excellent impact sensitivity properties (>25 J). It is demonstrated that the graphene-like layered stacking structure of novel bis-(1,2,4-triazole)-furoxan compounds is due to intramolecular interactions and intermolecular interactions by combining structure description with wavefunction analysis (Hirshfeld surface analysis, IRI analysis, IGMH analysis, AIM analysis, etc.). Numerous experiments have shown that 3-nitro-5-chloroxime-1,2,4-triazole or 3-nitroamino-5-chloroxime-1,2,4-triazole of two intermediate compounds is unable to obtain bis-triazole-furoxan under alkaline conditions. N-methylation of the triazole was accomplished by protecting the amino group with dimethylformamide dimethylacetal, facilitating furoxan synthesis. This study first introduces the electrostatic potential isosurface distribution of clusters within crystals in analysing energetic compounds. The results show that the electrostatic potential distribution of clusters is more regular than individual molecules, which could be significant for predicting and assessing the sensitivity of energetic compounds.

---

Introduction

The balance between sensitivity and detonation performance has always been a popular area of research for energetic compounds. These compounds tend to have opposite trends of sensitivity and detonation performance. However, there is no conclusive evidence that the low sensitivity and their high detonation performance cannot coexist in a paradoxical relationship. In recent years, furoxan¹ has had a structural building block by linking aromatic compounds such as tetrazoles,^2,3 oxadiazoles,^4,5 oxatriazoles,⁶ and benzenes^7,8 to energetic sites or by self-forming dimers⁹ and trimers^10,11 through coupling, resulting in a wide variety of energetic compounds, which has made significant progress in the development and research of energetic compounds.^12,13 Studies have shown that furoxan has a distinct advantage due to its unique structure concerning other 5-membered heterocycles.¹⁴ The furoxan ring can form a latent nitroxide inner ring structure with two reactive oxygen atoms, increasing the target compound's oxygen content, enthalpy, and crystal density, which results in a higher detonation performance. 1,2,4-Triazole energetic compounds are more advantageous in balancing the sensitivity and detonation performance due to their excellent backbone structure (high enthalpy of formation, aromaticity, uniform carbon and nitrogen distribution, structural symmetry, etc.) compared with the 5-membered rings of oxadiazoles (furazan, furoxan) and unstable nitrogen-linked compounds (1,2,3-triazoles, tetrazoles). Some energetic compounds based on 1,2,4-triazole are well-known for their superb energy and usefulness. Examples include 3-nitro-5-keto-(1,2,4)-triazole (NTO), 3,5-dinitro-1-methyl-1,2,4-triazole (MDNT), and 3-amino-5-nitro-1,2,4-triazole (ANT).

In recent research, the combination of 1,2,4-triazole with furoxan as a bicyclic energetic compound has gained the attention of researchers.^15–17 However, previous studies have only focused on binding a single triazole to the furoxan molecule.¹⁸ Furoxan has two carbon modification sites, which connect the triazole molecules on both sides and can lead to the development of a novel tricyclic energetic compound.^19,20 The structure of bis-(1,2,4-triazolyl)-furoxan is identical to that of bis-(tetrazole)-furoxan previously published.²¹ However, the tetrazole's N–N-linked structure is unstable, much like azide compounds. Replacing bis(tetrazole) with bis(1,2,4-triazole) can enhance the safe properties of the furoxan compound. This enhancement increases its energy and decreases sensitivity by modifying the C-site and N-site groups of the triazole. Only one report has been published on the similar structure of bis-(1,2,4-triazolyl)-furoxan from 2012 until now.²² However, this structure could be enhanced at a molecular level. Introducing a nitro group into triazole at the N position can increase energy, but it also makes it impossible to increase energy further, as shown in Fig. 1. Therefore, it is necessary to continue exploring synthetic methods, detonation, and crystal properties of bis-(1,2,4-triazolyl)-furoxan-like energetic compounds. This work delves into the synthesis methods of three novel tricyclic structures of bis-(1,2,4-triazolyl)-furoxan energetic compounds 13, 14, and 15. A combination of crystal analysis and wavefunction analysis investigates the advantages of the bis-(triazole)-furoxan properties for analysing their molecule structural features and crystal stacking.

Fig. 1 The structure of bis-(1,2,4-triazolyl)-furoxan of energetic compounds.

Experimental section

Materials and measurements

All the chemicals and solvents were purchased from Sigma-Aldrich and used without further purification. Compound 1 can be synthesised through three steps, as reported.²³ Compounds 3 and 8 can be synthesised from compound 2 and compound 7, as reported.²⁴ Compounds 4, 6 and 9 cannot be obtained through numerous experiments, as shown in Fig. 2. For spectroscopic characterisation, a small quantity of the solid compound was dissolved in DMSO-d₆, and NMR data were acquired on a Bruker AVANCE (400 MHz for ¹H and 600 MHz for ¹³C) spectrometer. FTIR spectra were measured using an FTIR PerkinElmer Frontier spectrometer. The thermal decomposition data were obtained from a PerkinElmer DSC 4000. Compounds 12, 13, and 15 were structurally characterised by single-crystal X-ray diffraction at an ambient temperature of 298 K using a Bruker D8 VENTURE single-crystal diffractometer with a graphite monochromator controlled using APEX2. The structures were solved using direct methods using the Shelxtl software. An Elementar Vario is used for elemental analysis (EA).

Fig. 2 Failed synthetic routes of compounds 4, 6 and 9.

Calculation methods. The prediction for the density of the compounds (13–15) was carried out using Peter Politzer's method.²⁵ The enthalpy of the compounds (13–15) was calculated using the PBE0-D3(BJ)/6-311G(**)//PBE0-D3(BJ)/6-31G(**) level via Gaussian16; the detonation parameters were predicted via the K–J equation;²⁶ the Hirshfeld surface analysis and the energy framework analysis were carried out via the CrystalExplorer software;²⁷ AIM (Bader's theory of ‘atoms in molecules) analysis,²⁸ electrostatic analysis,^29–31 IRI analysis³² and IGMH analysis³³ were performed by combining the Multiwfn³⁴ and the VMD software; the wavefunction files about the electrostatic potential isosurface, IRI and IGMH were obtained through the B3LYP-D3(BJ)/6-311G(**) level calculation. The wavefunction files from the AIM analysis were obtained through the B3LYP-D3(BJ)/ma-TZVPP//B3LYP-D3(BJ)/TZVP level calculation.

Synthesis of compound 1. a) Take 3.2 g (22.8 mmol) of 3-formyl-5-amine-1,2,4-triazole and dissolve it in 50 ml of ethanol. Keep the temperature at 0 °C and slowly add 5 ml of thionyl chloride (SOCl₂). After heating the reaction mixture under reflux to 90 °C for 2 hours, cool the reaction down to room temperature. Neutralise to neutral with sodium carbonate (Na₂CO₃). Filter and collect the white products (3.2 g, 20.5 mmol). Yield 90%. Elemental anal. calcd. (%) for C 38.46, H 5.12, N 35.89, O 20.51; found: C 38.86, H 5.25, N 34.80, O 20.57. IR (KBr, cm⁻¹): 3449, 3139, 1723, 1635, 1456, 1240, 1118, 1018, 712.

b) Dissolve 1 g (6.4 mmol) of the product from step a) in 10 ml of 25% ammonia and heat the mixture to 70 °C under closed conditions for 2 hours. Then, cool the reaction down to room temperature. Add acetic acid to the mixture to acidify it until the pH reaches 4. Collect the resulting precipitate by filtration to obtain a white product known as 5-amino-1H-1,2,4-triazole-3-carboxamide (0.6 g 4.7 mmol). Yield 72%. Elemental anal. calcd. (%) for C 27.90, H 3.87, N 54.26, O 13.95; found: C 28.34, H 3.94, N 54.90, O 11.70. IR (KBr, cm⁻¹): 3340, 3142, 1691, 1646, 1580, 1483, 1415, 1297, 1121, 1069, 763, 687.

c) Add 3 g (23.2 mmol) of product from step b) and 20 g of P₂O₅ into 300 ml of anhydrous acetonitrile simultaneously. Stir the mixture for 72 hours at room temperature. Then, 100 ml of distilled water was added to the reaction vessel. Evaporate the acetonitrile in a vacuum and extract the resulting clear solution with ethyl acetate (3 × 100 mL). Dry the combined organic layers over magnesium sulfate. Finally, remove the solvent in a vacuum to obtain a colourless powder (2 g, 18.0 mmol). Yield 77%. Elemental anal. calcd. (%) for C 32.43, H 4.50, N 63.06, O 0; found: C 31.76, H 2.87, N 59.76, O 5.74. IR (KBr, cm⁻¹): 3431, 3289, 3147, 2259, 1635, 1557, 1490, 1356, 1120, 1010, 728, 675.

Synthesis of compound 11. 0.6 g (5.4 mmol) of compound 1 and 1.2 g of dimethylformamide dimethyl acetal were added simultaneously to 18 ml of acetonitrile. The temperature was gradually raised to 40 °C for 1 hour. The resulting solution was bright orange. The solution was then poured into a tray, and the solvent was left to air-dry naturally to obtain a sample of needle-shaped crystals. Then, 1 g of needle-shaped crystals was added to 2 ml of distilled water. Then, 5 ml of 50% hydroxylamine solution was slowly poured into the mixture, and the reaction was kept for 30 minutes. After adding 15 ml of hydrochloric acid and 3 g of sodium nitrite to the mix, the temperature was maintained at 0 °C. Finally, the yellow powder was filtered (0.6 g, 3.0 mmol). Yield 56%. IR (KBr, cm⁻¹): 1597, 1537, 1466, 1394, 1116, 1075, 1013, 942, 887, 825, 727. Elemental anal. calcd. (%) for C 42.63, H 5.58, N 42.63, O 9.13; found: C 42.51, H 5.57, N 42.52, O 9.51. ¹H NMR (400 MHz, DMSO) δ 12.97 (s, 1H), 8.62 (s, 1H), 3.70 (s, 3H), 3.19 (s, 3H), 3.09 (s, 3H). ¹³C NMR (101 MHz, DMSO) δ 159.13, 156.56, 148.32, 127.15, 35.16, 34.45, 34.37.

Synthesis of compound 12. 1 g (5 mmol) of compound 11 was dissolved in 100 ml tetrahydrofuran. Then, 0.3 grams of sodium carbonate (Na₂CO₃) was added directly to the reaction system, and the reaction was carried out at 25 °C for 24 hours. The solution changed from cloudy to clear. Afterwards, the impurities were filtered out, leaving the solution with the dissolved products behind. The colourless product slowly precipitated and was collected (1 g 2.6 mmol). Yield 53%. IR (KBr, cm⁻¹): 1620, 1527, 1468, 1392, 1337, 1294, 1109, 1063, 1029, 959, 798. Elemental anal. calcd. (%) for C 44.90, H 5.34, N 44.91, O 4.81; found: C44.54, H 5.69, N 43.37, O 4.02. ¹H NMR (400 MHz, DMSO) δ 8.39 (s, 1H), 8.35 (s, 1H), 3.66 (d, J = 16.7 Hz, 6H), 3.11 (d, J = 5.5 Hz, 6H), 3.02 (s, 6H). ¹³C NMR (101 MHz, DMSO) δ 159.32, 157.57, 150.25, 146.83, 143.64, 34.21, 24.98.

Synthesis of compound 13. 0.7 g (1.8 mmol) of compound 12 was dissolved in 50 ml of 11% hydrochloric acid solution and reacted 48 hours at 25 °C. The white product slowly precipitated and was collected (0.4 g 1.5 mmol).

Yield 84%. IR (KBr, cm⁻¹): 3413, 3322, 3187, 1650, 1574, 1499, 1432, 1293, 1260, 1221, 1075, 964, 837, 767, 667. Elemental anal. calcd. (%) for C 36.36, H 3.70 N 53.03, O 6.81; found: C 37.29, H 3.54, N 54.36, O 6.16. ¹H NMR (400 MHz, DMSO) δ 6.53 (d, J = 5.7 Hz, 4H), 3.57 (d, J = 3.8 Hz, 6H). ¹³C NMR (101 MHz, DMSO) δ 156.21, 150.16, 146.47, 143.24, 108.82, 33.58.

Synthesis of compound 14. At a temperature of −5 °C, 0.2 g (0.7 mmol) of compound 13 was added to 3 ml of 30% hydrogen peroxide, and then 1.5 ml of concentrated sulfuric acid was slowly added. The temperature was kept below 0 °C. After the addition, the temperature was raised to room temperature. Then 0.5 g of sodium tungstate was added in batches, and the feeding temperature was maintained at room temperature for 4 hours after each batch addition. Finally, the mixture was extracted with ethyl acetate (0.1 g 0.3 mmol). Yield 44%. IR (KBr, cm⁻¹): 3087, 2926, 1610, 1509, 1398, 1360, 1282, 1247, 1072, 1043, 964, 896, 822, 747, 709. Elemental anal. calcd. (%) for C 36.36, H 1.85, N 43.20, O 24.69; found: C 36.25, H 2.13, N 43.11, O 24.22. ¹H NMR (400 MHz, DMSO) δ 3.58 (d, J = 3.3 Hz, 6H). ¹³C NMR (101 MHz, DMSO) δ 154.85, 148.32, 144.49, 141.34, 107.91, 34.23.

Synthesis of compound 15. 0.2 g (0.7 mmol) of compound 13 was added gradually to 10 ml of 98% concentrated sulfuric acid. Subsequently, 1 ml of fuming nitric acid was slowly added to the reaction mixture, which was carried out at 0 °C for one hour. The yellow-coloured reaction solution was poured into ice water, separating a white product. The product was collected through filtration, and a yield of 0.24 grams was obtained. (0.24 g 6.8 mmol). Yield 97%. IR (KBr, cm⁻¹): 3446, 3075, 1805, 1617, 1551, 1500, 1458, 1413, 1379, 1309, 1237, 1144, 1043, 972, 935, 801, 719. Elemental anal. calcd. (%) for C 27.27, H 1.85, N 43.20, O 24.69; found: C 27.00, H 1.70, N 42.12, O 25.51. ¹H NMR (400 MHz, DMSO) δ 7.54 (d, 2H), 3.85 (s, 3H), 3.81 (s, 3H). ¹³C NMR (101 MHz, DMSO) δ 148.93, 145.98, 143.64, 138.74, 106.05, 35.35, 35.29.

Results and discussion

Initially, a synthetic route resulted in the formation of two furoxan energetic compounds 9 and 10via a conventional 1,3-dipolar cycloaddition reaction³⁵ between nitrile oxides of compound 3 (3-nitroamine-5-chloroxime-(1,2,4-triazole)) and compound 8 (3-nitro-5-chloroxime-(1,2,4-triazole)) under weak-base conditions. Unfortunately, numerous experiments have shown that this synthesis route of compounds 9 and 10 is impossible, as shown in Fig. 2. Different weak bases, such as K₂CO₃, Na₂CO₃, Ag₂CO₃, MnCO₃, CuCO₃, ZnCO₃, (NH₄)₂CO₃, pyridine, etc., were tested with various solvents, which include tetrahydrofuran, ethyl acetate, butyl acetate, dioxane, toluene, ether and many ratios of the mixture. The failure can be attributed to three main aspects. In the first aspect, weak bases such as ZnCO₃, (NH₄)₂CO₃, MnCO₃ and pyridine have difficulty reacting with compounds 3 and 8. In the second aspect, too many impurities are difficult to separate and purify. Our tests have demonstrated that one equivalent of Na₂CO₃, K₂CO₃, and Ag₂CO₃ can help two types of chloroxime (compounds 3 and 8) react using tetrahydrofuran or ethyl acetate as solvents. The products still contain excessive impurities that are difficult to separate. The third aspect is that compounds 3 and 8 form a residue that is difficult to dissolve when excessive weak bases are used. Only methanol and ethanol may separate organic compounds. However, the furoxan of products is not detected upon solvent evaporation. When the products generated by using Ag₂CO₃ conditions are added to excess NH₃OH–HCl, NH₄Cl, NaCl, KCl, and BaCl₂ through substitution reaction, it is not possible to obtain more stable furoxan products in solvents such as methanol, ethanol, acetone, and tetrahydrofuran. Additionally, forming furoxan was abandoned through compound 6 (3-amino-5-chloroxime-(1,2,4-triazole)) because the direct conversion of compound 5 to compound 6 was impossible, as shown in Fig. 2.

It has been concluded that the synthetic route mentioned cannot be achieved due to the nitramine and nitro groups on the 1,2,4-triazole. These groups may cause the entire chloroxime triazole to become highly acidic, making it impossible for the intermediate nitrile oxides to undergo a single 1.3-dipole cycloaddition reaction under weak alkaline conditions. This leads to the production of unwanted by-products. Alternatively, the route can be modified to synthesise the furoxan first and then convert the amine to nitramine and nitro groups. Therefore, protecting the amino group on compound 1 is essential before synthesising the bis-(1,2,4-triazolyl)-furoxan.

As depicted in Fig. 3, the amino group of compound 3 was protected according to the method reported in the literature.³⁶ However, it was discovered that the nitrogen atom of the triazole containing active hydrogens can act as a nucleophile and attack the carbon atoms of dimethylformamide dimethylacetal. It is demonstrated that the product was not unique when analysed using TLC chromatography. The amino group of the triazole can form a Schiff base, while the imine nitrogen atom on the triazole is protected. It was previously believed that compound 1 could be converted to compound 10 by converting the cyane to chloroxime in a two-step process. It was discovered that methylation of the N atom of the triazole led to the conversion of compound 1 to compound 11, which was surprising. Unfortunately, compound 10 cannot be obtained. The mechanical reason for the methylation of nitrogen atoms on triazoles remains unclear. The result of N-methylation removes the active hydrogen and reduces the acidity of compound 11. This is likely the primary reason that compound 11 precipitates in acidic solutions without extraction in contrast to compounds 3 and 8.

Fig. 3 The synthetic route of compound 13.

Compound 11 was successfully converted to compound 12 (furoxan) with Na₂CO₃ in tetrahydrofuran. Under 11% dilute hydrochloric acid conditions, the Schiff base of compound 12 was hydrolysed to produce compound 13, as shown in Fig. 3. The synthesis of the structural framework of the bis-(1,2,4-triazole)-furoxan energetic compound was completed. The nitro group was introduced via nitration and oxidation, forming two higher energetic compounds: compounds 14 and 15, as depicted in Fig. 4. The single-crystal structures of compounds 12, 13 and 15 are shown in Fig. 5, and crystal parameters and CCDC numbers are provided in Table S1.† The H NMR spectra, C NMR spectra, IR spectra, and DSC thermal analysis diagram for compounds 13–15 are available in Fig. S1–S21 in the ESI.†

Fig. 4 The synthetic route of compounds 14 and 15.

Fig. 5 The single-crystal structure view of compounds 12, 13, and 15 corresponds to a), b), and c). Thermal ellipsoids are drawn at 50% probability.

The physical and chemical properties of the compounds (13–15)

The table indicates that the density of compound 13 is 1.68 g cm⁻³, slightly higher than that of H₂BTF, which has a density of 1.62 g cm⁻³. However, the explosive performance of compound 13 is significantly lower than that of H₂BTF. This discrepancy is mainly attributed to the higher nitrogen content in H₂BTF. On the other hand, compound 13 demonstrates a sensitivity exceeding 30 J, indicating a much greater impact energy than H₂BTF, which has an impact energy of less than 2 J in terms of sensitivity. The presence of tetrazolyl in H₂BTF enhances its explosive properties due to the N–N linked group. Replacing tetrazolyl with triazolyl in furoxan could significantly improve safety and detonation performance.

The detonation performance of compounds 14 and 15 is slightly lower than that of BNTAF. This difference is attributed to the lower density and enthalpy of formation in compounds 14 and 15 compared to BNTAF. However, their impact sensitivity is much higher than that of BNTAF. Although the impact energy for BNTAF has not been reported, it is noted that the impact sensitivity for BNTAF has an H₅₀ of 7.2 cm according to the report in the GJB772A-1997-601-2 method. In contrast, the H₅₀ for LLM-105 under 2 kg drop hammer conditions exceeds 100 cm. The impact sensitivity of compound 15 is comparable to that of LLM-105, which exceeds 25 J, indicating that compound 15 has significantly better impact sensitivity than BNTAF. The trade-off of sacrificing some detonation energy is justified to significantly reduce sensitivity in compound 15. Unfortunately, compound 15 exhibits weak thermal stability, with decomposition occurring at around 144 °C as opposed to the decomposition temperature of compound 13 at 278 °C, which indicates a significant reduction in heat resistance. This may be due to the instability of the furoxan ring, which could be causing the higher inductive effect of the nitro groups on the triazole ring. Compounds 13 and 15 have better impact sensitivities compared to compound 14, both exceeding 25 J. This is due to the lower donation heat and the graphene structure of crystal stacking, which can be demonstrated in the next part of the discussion. The crystalline structure of graphene enhances safety by reducing hot spot generation in external impact stimulation through sliding molecular layers (Table 1).

Table 1 The physical and chemical properties of the compounds (13–15) compared to H₂BTF (ref. 21) and BNTNF (ref. 22)

a Decomposition temperature (DSC, 5 °C min^−1). b Predicted by ref. 25. c Both nitrogen and oxygen. d Heat of formation calculated with Gaussian16. e Calculated detonation velocities via K–J equation. f Calculated detonation pressure. g Detonation heat. h Impact sensitivity. i Oxygen balance (OB, %) for CaHbNcOd OB = 1600 × (d − a − b/2)/Mw.
Comp.	13	14	15	BNTAF	H2BTF	LLM-105
T d [°C]	278	138	144	139	229	350
ρ [g cm^−3]	1.677	1.77	1.77	1.84	1.62	1.85
N + O^c [%]	40	56	53	67	67	57
ΔHf^d [kJ mol^−1]	114.83	185.45	210.95	841.5	−133.76	−12.95
D [km s^−1]	6.00	8.10	8.10	8.49	7.80	8.28
P [GPa]	15.02	26.77	26.92	32.40	23.80	26.77
Q max [kcal g^−1]	0.55	1.05	1.09	—	—	1.17
IS^h [J]	>30	15	>25	—	<2	28.7
OB^i [%]	−0.63	−0.26	−0.23	−0.05	−0.216	−0.216

---

Structural description

(1) Structure description of molecules. The dihedral angle of the compounds was measured using CrystalExplorer software, as shown in Fig. 6. The (b) diagram was created by combining the VMD and the Multiwfn. This was achieved by analysing the interaction region indicator (IRI = 1.1). In the (b) diagram, the covalent bonds between atoms appear in dark blue, steric effects in red, and weak interactions such as hydrogen bonding in light blue and van der Waals interactions in green. It has been observed that compounds 13 and 15 are almost in the same plane, as evidenced by three dihedral angles (−3.95°, 1.7°, −1.13°) and (7.89°, 5.89°, 0.53°, −0.77°, −3.70°) at five dihedral angles respectively in Fig. 6(a). The nitrosamine groups are in one plane with the triazole ring, as evidenced by two dihedral angles (7.89°, −3.70°) in compound 15. Although compounds 13 and 15 are not strictly fused rings, they are almost in the same plane because of the often neglected weak interactions between intramolecular groups. The triazoles on both sides of compounds 13 and 15 are not symmetrical. The triazole on the furoxan side can be obtained by rotating the other by 180°. This lack of symmetry may be due to the intramolecular interactions between the nitrogen atoms and the methyl and amino groups, constraining all atoms in the same plane in compounds 13 and 15. This is supported by the IRI analyses in Fig. 6(b).

Fig. 6 Molecular structure in crystals (a); IRI analysis (b) of compounds 13 and 15.

(2) Structural description for the cluster of compound 13. Fig. 7 indicates that several water molecules surrounding the central sulfate molecule spatially combine with the NH₂ and CH₃ of compound 13 through H-bonding interactions. To describe the stacking of compound 13, the four oxygen atoms connected to the sulfate molecules of the tetrahedral structure could be grouped into two pairs: (O5^a, O6) and (O5, O6^a). Additionally, it can be observed that the cluster on either side of the sulfate molecule forms a V-shape, as shown in Fig. 7(b). Fig. 7 shows the H-bonds with cyan dashed lines to understand better how hydrogen bonding interactions result in stacking. Table S2† also provides the hydrogen bonds' group classes and bond lengths. In Fig. 7(a), the oxygen atoms of sulfate can form two groups of hydrogen bonds with the N1 atom of compound 13: O5⋯H–N1^b (1.9605 Å) and O6⋯H–N1 (1.9435 Å). As a result, the four oxygen atoms of the sulfate molecule can form a cluster conformation with five compound 13 molecules through hydrogen bonding interactions, resulting in a six-molecule cluster. The three water molecules around the sulfate stabilise the crystal stacking. O7 and (O3, O4) are located on both sides of the sulfate molecule in Fig. 7(a). Moreover, O7 can form a hydrogen bond with the H atom of methyl (C1^c) in (O7⋯H–C1^c), and O3 can form a hydrogen bond with the NH₂ in (O3⋯H–N7^d). In Fig. 7(b), the stacking of compound 13 in layers may be attributed to the hydrogen bonding interactions of the three water molecules within the crystals. Compound 13 in 2D is arranged linearly in a combination of M in the horizontal axis through π⋯π interactions between molecules, and compound 13 molecules stacked in layers on both sides are connected by sulfate and water molecules, as shown in Fig. S30.† This results in a crystal stacking from 2-dimensional to 3-dimensional. To better observe this stacking structure, the structural diagram of the supramolecule from a, b, and c three-axis directions is given in Fig. S33.†

Fig. 7 The cluster structure of compound 13 observed from (a) and (b) two angles.

(3) Structural description for the cluster of compound 15. The cluster arrangement of compound 15 can be seen in Fig. 8. Black transparent double arrows are included to more conveniently describe the stacking. Fig. 8(b) shows that the arrows point to the angular alignment axes within the crystal lattice. Fig. 8(a) demonstrates that four compound 15 molecules exhibit centre symmetry on both sides of the double arrows. Compound 15 stacks in the following manner: it rotates around the axis of the black double arrows and spirals in a counterclockwise or clockwise direction, either moving up or down about the double arrows. This rotation and spiral movement completes one entire rotational cycle involving four molecules. Fig. 8(a) clearly illustrates the hydrogen bonding of compound 15 with water and acetonitrile molecules. Specific details about the groups and lengths of the H-bonds depicted are provided in Table S3.† Two hydrogen atoms on the O7 atom form two sets of hydrogen bonds with the O2 atom of compound 15 and the N13 atom of acetonitrile (O2⋯H–O7, N13⋯H–O7). The hydrogen on O7^b forms a hydrogen bond (O5^a⋯H–O7^b) with O5^a located on the other side of compound 15. The oxygen atom O4^a of the furoxan is bonded to the H atom on the methyl group C8 of another compound 15 (O4^a⋯H–C8). The H atom on C6^a is bonded to the O1 atom on the nitro group (O1⋯H–C6^a). Fig. 8(b) illustrates how the molecules of compound 15 can form a stacked layer in the crystal space. Compound 15 is arranged horizontally, crossing from left to right. Compound 15 molecules are planar molecules in their spatial structure, unfolding a planar state similar to graphene in Fig. S31.† To better observe this stacking structure, the structural diagram of the supramolecule from a, b, and c three axial directions is given in Fig. S34.†

Fig. 8 (a and b) Molecular cluster diagrams of compound 15 observed from two different angles.

Wavefunction analysis

(1) Electrostatic potential distribution analysis. The research about the electrostatic potential distribution of clusters of energetic compounds usually has more physical significance than that of a monomer. However, the electrostatic potential distribution of clusters has been ignored in previous studies. Compound 13 of the monomer does not show a pattern of positive and negative charges in the overall distribution. Fig. 9(b) shows two humps in the positive potential area (20 kcal mol⁻¹) and the negative potential area (−20 kcal mol⁻¹) of cluster 13 of electrostatic potential. This indicates that the cluster's electrostatic potential is polarised, which may affect the sensitivity of energetic compounds. When comparing the electrostatic potential of compound 13 in Fig. 9(c and d), it is observed that the positive potential (20 kcal mol⁻¹) increases dramatically, which may be caused by the intense negativity of the nitramine group. The polarisation of the cluster is more severe than that of the monomer on compound 15. The diagrams of the electrostatic potential isosurface of the monomer and clusters are provided in Fig. S28 and S29.† Fig. S29† shows the electrostatic potential isosurface of the penetration effect between individual molecules within each cluster of compounds 13 and 15, also representing the weak interactions in the cluster. The electrostatic potential of clusters can provide physically meaningful guidance for understanding the effect of sensitivity.

Fig. 9 The electrostatic potential distribution of the monomer (a and c) and clusters (b and d) of compound 13 and compound 15.

(2) Hirshfeld surface analysis. The Hirshfeld surfaces mapped on d_norm of compounds 13 and 15 were analysed for colour variations by using CrystalExplorer (the blue colour represents weaker interactions on the surface, while the red colour represents more vital interaction and the white colour represents weaker interactions in general, such as pi–pi stacking, weaker hydrogen bonds, etc.) in Fig. 10. The Hirshfeld surfaces are mapped using the shape index and the curvedness provided in Fig. S24.† Decomposed fingerprint plots of compounds 13 and 15 are provided in Fig. S22 and S23.† The area (d_e > d_i) represents a hydrogen bond donor on the surface of the molecule on the right-hand side of the fingerprint, and the area with (d_e < d_i) represents a hydrogen bond acceptor on the surface of the molecule. The multiple hydrogen atoms of the amino and methyl groups can provide more hydrogen bonding due to compound 13 having two amino groups of methyl on both sides. Additionally, the O atoms of the furoxan and the N atom of the triazole can provide more acceptor sites for H-bonding.

Fig. 10 The fingerprint of the Hirshfeld surface mapped on d_norm of compounds 13 and 15.

Compound 13 exhibits more red markings than compound 15, which suggests a higher presence of strong interactions. The O atoms of the nitro group in compound 13 could provide many strong hydrogen bond acceptors. The area of compound 13's fingerprints is significantly larger than that of compound 15 in Fig. 10. This confirms that compound 15 has fewer strong interactions. It also can be speculated that compound 13 has more surface-strong interactions, which could result in a tighter crystal arrangement. In addition, the Hirshfeld surfaces of compounds 13 and 15 exhibit more white and less blue surfaces. The white surfaces suggest that the molecule has relatively weak interactions, such as weak hydrogen bonding and van der Waals interactions. The blue surfaces indicate that the electron density tends to be zero. The less blue surfaces of compounds 13 and 15 suggest that the crystal's pores may be relatively small. This property is significant as it may decrease the impact sensitivity of energetic compounds.

The inside O atoms of compound 13 contribute up to 13.1% on the Hirshfeld surface. The interaction of the inside O atoms with the outside O atoms (O–O) is only 0.4%, making up less than 3% of the total contribution of oxygen atoms. Meanwhile, the inside oxygen of compound 15 can contribute up to 34% due to the oxygen atom of the NO₂ group. The interaction of the inside oxygen atom with the outside oxygen atom is only 4.5%, which is less than 15% of the overall contribution of oxygen atoms. In some cases, having fewer interactions between oxygen atoms can increase the safety of energetic compounds. The negative charge of the internal and external oxygen atoms can cause electrostatic repulsion and the formation of hot spots when the molecule in the crystal twists or slips during external simulations—compounds 13 and 15 exhibit lower N–N, C–C, and N–O interaction percentages. The percentages are 7.2%, 2.6%, and 1.2% for N–N, C–C, and N–O interactions in compound 13 and 6.6%, 0.4%, and 4.7% in compound 15, respectively in Fig. 11. Compound 13 contains more hydrogen atoms due to amino and methyl groups, resulting in increased H–H interactions of 23.2%, compared to compound 15, which only reaches 13.4%. Compound 13 not only has less O–O, N–N, and other strong mutual solid repulsion but can also form a more robust H-bonding network, which reduces hot spots in the crystal. It can be inferred that compound 13 is a relatively safer energetic compound.

Fig. 11 The interaction statistics between inside and outside atoms of compounds 13 and 15.

(3) The enrichment ratio analysis. ³⁷ The enrichment ratio is a significant calculation derived from analysing interactions between two elements. It indicates the tendency of interaction between two elements in crystal packing. The enrichment ratio can be studied as a function of the surface (%) value. A higher surface (%) for an element suggests increased interaction opportunities on the Hirshfeld surface. Given a specific surface (%), if the enrichment ratio for two elements exceeds 1, it provides strong evidence of heightened interaction. Conversely, a lower value indicates a low tendency of disfavored contacts. The enrichment ratios, surface, and random contacts of the main intermolecular interactions for compounds 13 and 15 are summarised in Table S4.† The crystal structures of both compounds mainly feature H⋯O type contacts. Compound 13 has well-enriched contacts with E_HO = 1.67, while compound 15 has E_HO = 1.35. This enrichment is due to the high presence of S_H (49.5%) and S_O (17.1%) in compound 13 and S_H (41.85%) and S_O (29.10%) in compound 15 about the total Hirshfeld surface area. Both compounds also have relatively high E_NN = 1.42 and E_NN = 1.46 due to the high S_N with an abundance of S_N (21.20%) in compound 13 and S_N (22.50%) in compound 15, respectively. Additionally, heterocycles of compounds 13 and 15 containing N atoms have a higher propensity to form π⋯π stacking due to δ+ and δ− polarized atoms, which can form interacting partners contributing to the high E_NN.

The fundamental structure distinction between compound 13 and compound 15 is the replacement of the hydrogen atoms of the amino groups on both sides with nitro groups. As a result, the S_O (17.1%) of compound 13 is lower than the So (29.1%) of compound 15. Compound 13 also has a lower E_CO = 0.82 (<1), while compound 15 has an E_CO = 1.64 (>1). This indicates that the C⋯O interaction plays a more significant role in the crystal stacking of compound 15. In π⋯π stacking, the oxygen atoms interact with positively charged carbon atoms bonded to heteroatom atoms. Compound 15 has the highest E_CC = 2.82, indicating the presence of C⋯C interactions in the π⋯π stacking within the heteroaromatic ring across the crystal stacking. This suggests a solid aromatic fused ring in compound 15. In contrast, compound 13 has an E_CC = 0.66 (<1), indicating relatively fewer C–C interactions in the π packing species. The enrichment ratios for H⋯H, C⋯H, O⋯O, and N⋯O in the two compounds are below 1 (see Table S4†), indicating a lower probability of forming crystal packing. Moreover, the contribution of N⋯H hydrogen bonding in compound 15 (E_NH = 1.03) is higher than that in compound 13 (E_NH = 0.99), which suggests that N⋯H-bond contacts play a more significant role in compound 15.

(4) AIM (atoms in molecule) analysis. Fig. 12 illustrates ten yellow orbs (hydrogen bonding critical points) highlighted in green for compounds 13 and 15 using AIM calculations. By comparing the values of the Laplacian of electron density, the eta index, and |V_(BCP)|/G_(BCP) of these critical points could examine the nature and strength of hydrogen bonding. Additionally, the yellow path through these vital points represents weak interactions. The bond energies of the hydrogen bonds in the essential points were also calculated using the equation E_int = −223.08 × ρ(rBCP) + 0.7423.³⁸

Fig. 12 Bond critical points in clusters of compound 13 (a and b) and compound 15 (c and d).

A sulfate molecule and three water molecules connect compound 15 through hydrogen bonding, as shown in Fig. 12(a and b). When the Laplacian of electron density (∇²ρ) is >0, it can be generally assumed that non-covalent interactions are formed. The ∇²ρ values of A–J are all greater than 0 in Table S5.† This indicates that these critical point regions could form hydrogen-bonding interactions. This judgment is based on the |V_(BCP)|/G_(BCP) criterion for determining the type of bonding. When |V_(BCP)|/G_(BCP) is <1, interactions in a chemical system are characteristic of closed-shell interactions. When 1 < |V_(BCP)|/G_(BCP) < 2, they are intermediate. Three |V_(BCP)|/G_(BCP) of critical points B (0.8211), I (0.7083), and J (0.6363) are less than 1, while the other seven critical points are >1 and <2. This indicates that the weak interactions of the three critical points (B, I, J) differ from the other seven critical points. When η(r) is <1, it indicates non-covalent interactions, with a larger value indicating stronger strength. Comparing the values of η(r) and |V_(BCP)|/G_(BCP) can indicate that the weak interactions of B, I, and J are weaker hydrogen bonds. Judging by the molecular distances between the layers in Fig. 12(a and b), it can be inferred that the weak interactions of the critical points I and J represent weak hydrogen bonding between the molecules of layers. The relatively weak hydrogen bonding at point B is due to the stronger N–H⋯O interaction of the amino group. The other seven weak interactions are relatively strong. It can be demonstrated that B (2.7170 kcal mol⁻¹), I (0.1772 kcal mol⁻¹), and J (0.3094 kcal mol⁻¹) exhibit lower energies compared to the other critical points, which indicates that these are weak hydrogen bonding points by analysing the strength of hydrogen bonding in terms of electron density and calculating the bond energies (|E_int|) in Table S5.†

Compound 15 has fewer hydrogen bonds than compound 13, with only five critical points (A–E) in Fig. 12(c and d). This suggests that the interactions within the cluster molecules of compound 15 are relatively weak by the Laplacian of electron density, the eta index, and |V_(BCP)|/G_(BCP) of these critical points. The |E_int| value indicates that critical points A (1.6479 kcal mol⁻¹) and D (1.4043 kcal mol⁻¹) represent the hydrogen bonding interactions in compound 15 in Table S5.† In contrast, critical points B and C represent interactions involving a single acetonitrile molecule, a water molecule, and interactions in compound 15. These interactions are also part of weak hydrogen bonding.

(5) IGMH and energy framework analysis. The thin green layer is a π–π stacking layer formed by van der Waals interaction between the layers in Fig. 13(a and b). This demonstrated the presence of hydrogen bonding between layers and van der Waals interactions in the crystal stacking of compounds 13 and 15. Compound 15 is stabilised by a water molecule that acts as a hydrogen bonding donor and acceptor (within the purple dashed circle) in Fig. 13(b). A hydrogen bond between the water molecule and the N atom on the triazole in the in-plane further enhances the interactions. Hydrogen bonding can effectively tighten the individual molecules per unit volume, increasing the crystal density and reducing the voids in the lattice. The scattered graphs of IGMH are provided in Fig. S27 and S28.† The distribution of dots in the upper left of the scatter graphs indicates that van der Waals interactions and weak hydrogen bonding interactions are spread around compounds 13 and 15.

Fig. 13 IGMH analysis of clusters of compound 13 (a) and compound 15 (b).

Conclusion

Three novel bis-(1,2,4-triazolyl)-furoxan structures of energetic compounds (13–15) were synthesised in this work. However, the experiment demonstrated that compounds 9 and 10 failed to form. The successful synthesis of furoxan may be attributed to the N-methylation of active hydrogen on triazole. Although N-methylation reduces the energy of bis-(triazole)-furoxan, its significance can be highlighted in three key ways. Firstly, methylation facilitates the 1,3-dipole reaction, allowing it to proceed smoothly. Secondly, the methyl group acts as an electron-donating group, enabling the amino group to participate effectively in subsequent nitration and oxidation reactions. Finally, methylation introduces a spatial hydrogen-bonding effect in the crystal structure, contributing to the crystals' stacking. The relatively low decomposition temperature of compound 15 (144 °C) may be attributed to the introduction of the nitramine group.

The impact energy of compound 13 (>30 J) is significantly higher than that of the previously reported energetic compound H₂BTF (<2 J). Compound 15 has a detonation velocity of 8.10 km s⁻¹, lower than that of BNTAF, measured at 8.49 km s⁻¹. However, the sensitivity of compound 15 is similar to that of LLM-105, with an estimated sensitivity more significant than 25 J. This indicates that replacing the tetrazolyl group with a 1,2,4-triazolyl group on the furoxan structure maintains the energy level while significantly reducing sensitivity. This also shows that the structure of the bis-triazole-furoxan is crucial for balancing energy output and sensitivity. The excellent sensitivity of compound 15 is partly due to the molecular structure and primarily due to the graphene-like crystal stacking. The structure description of compound 13 and 15 clusters with IRI, IGMH and AMI noncovalent interaction analysis revealed that compound 13 and 15 are planar molecules due to intramolecular interactions. The planar stacking of the molecules is caused by weak hydrogen bonding and intermolecular interactions between the cluster molecules. Analysis of the electrostatic potential distribution shows that by replacing single molecules with clustered molecules to examine the electrostatic potential of molecular surfaces, clustered molecules exhibit a more regular electrostatic potential, which is significant for subsequent studies assessing the sensitivity of energetic compounds.

Data availability

The data supporting this study's findings are available from the corresponding author upon reasonable request.

Author contributions

Conceptualisation: Yucun Liu; Jiapeng Wang. Software: Jianhua Wang; Jiapeng Wang. Writing – original draft: Jiapeng Wang; Jianhua Wang. Investigation: Junming Yuan; Yanwu Yu. Supervision: Yankang Zhang. Resources: Xuejian Yan.

Conflicts of interest

There are no conflicts of interest to declare.

Acknowledgements

All of the co-authors want to express their gratitude to Professor Yucun Liu for his guidance, academic encouragement, and friendly critique. Jiapeng Wang would also like to thank his family and friends for their support and care throughout his research.

References

1. G. Romeo, U. Chiacchio, M. H. Keshavarz, H. Motamedoshariati, R. Moghayadnia, M. Ghanbarzadeh and J. Azarniamehraban, Modern Heterocyclic Chemistry, 2018, vol. 38, 1047-1252-1102.
2. L. L. Fershtat, M. I. Struchkova, A. S. Goloveshkin, I. S. Bushmarinov and N. N. Makhova, Heteroat. Chem., 2014, 25, 226–237.
3. L. X. Liang, K. Wang, C. M. Bian, L. M. Ling and Z. M. Zhou, Chem. – Eur. J., 2013, 19, 14911–14919.
4. E. C. Johnson, E. J. Bukowski, J. J. Sabatini, R. C. Sausa, E. F. C. Byrd, M. A. Garner and D. E. Chavez, ChemPlusChem, 2019, 84, 319–322.
5. E. C. Johnson, J. J. Sabatini, D. E. Chavez, L. A. Wells, J. E. Banning, R. C. Sausa, E. F. C. Byrd and J. A. Orlicki, ChemPlusChem, 2020, 85, 237–239.
6. E. S. Zhilin, D. M. Bystrov, I. V. Ananyev, L. L. Fershtat and N. N. Makhova, Chem. – Eur. J., 2019, 25, 14284–14289.
7. J. W. Fronabarger, M. D. Williams, W. B. Sanborn, D. A. Parrish and M. Bichay, Propellants, Explos., Pyrotech., 2011, 36, 459–470.
8. C. Jovené, M. Jacquet, J. Marrot, F. Bourdreux, M. E. Kletsky, O. N. Burov, A.-M. Gonçalves and R. Goumont, Eur. J. Org. Chem., 2014, 2014, 6451–6466.
9. I. Gospodinov, T. Hermann, T. M. Klapotke and J. Stierstorfer, Propellants, Explos., Pyrotech., 2018, 43, 355–363.
10. D. Fischer, T. M. Klapötke and J. Stierstorfer, Eur. J. Inorg. Chem., 2014, 2014, 5808–5811.
11. C. L. He, H. X. Gao, G. H. Imler, D. A. Parrish and J. M. Shreeve, J. Mater. Chem. A, 2018, 6, 9391–9396.
12. L. L. Fershtat and N. N. Makhova, ChemPlusChem, 2020, 85, 13–42.
13. L. L. Fershtat, I. V. Ovchinnikov, M. A. Epishina, A. A. Romanova, D. B. Lempert, N. V. Muravyev and N. N. Makhova, ChemPlusChem, 2017, 82, 1315–1319.
14. E.-C. Koch, Propellants, Explos., Pyrotech., 2016, 41, 526–538.
15. N. S. Aleksandrova, K. A. Monogarov and A. B. Sheremetev, Chem. Heterocycl. Compd., 2020, 56, 619–622.
16. J. Ma, H. Yang, J. Tang, G. Zhang, Z. Yi, S. Zhu and G. Cheng, Dalton Trans., 2020, 49, 4675–4679.
17. A. K. Chinnam, R. J. Staples and J. N. M. Shreeve, J. Org. Chem., 2021, 86, 7781–7786.
18. L. L. Fershtat, M. A. Epishina, I. V. Ovchinnikov, V. V. Kachala and N. N. Makhova, Chem. Heterocycl. Compd., 2015, 51, 754–759.
19. A. B. Sheremetev, N. S. Aleksandrova, S. S. Semyakin, K. Y. Suponitsky and D. B. Lempert, Chem. – Asian J., 2019, 14, 4255–4261.
20. A. A. Larin, A. N. Pivkina, I. V. Ananyev, D. V. Khakimov and L. L. Fershtat, Front. Chem., 2022, 10 DOI:10.3389/fchem.2022.1012605.
21. H. Huang, Z. Zhou, L. Liang, J. Song, K. Wang, D. Cao, C. Bian, W. Sun and M. Xue, Z. Anorg. Allg. Chem., 2012, 638, 392–400.
22. Y. A. Li, Z. Z. Zhang, Z. X. Ge, B. Z. Wang, W. P. Lai and Y. F. Luo, Chin. J. Chem., 2013, 31, 520–524.
23. A. A. Dippold, D. Izsák and T. M. Klapötke, Chem. – Eur. J., 2013, 19, 12042–12051.
24. A. A. Dippold and T. M. Klapötke, Chem. – Asian J., 2013, 8, 1463–1471.
25. P. Politzer, J. Martinez, J. S. Murray and M. C. Concha, Mol. Phys., 2010, 108, 1391–1396.
26. L. Kazandjian and J. F. Danel, Propellants, Explos., Pyrotech., 2006, 31, 20–24.
27. P. R. Spackman, M. J. Turner, J. J. McKinnon, S. K. Wolff, D. J. Grimwood, D. Jayatilaka and M. A. Spackman, J. Appl. Crystallogr., 2021, 54, 1006–1011.
28. R. F. W. Bader, The Quantum Mechanical Basis of Conceptual Chemistry, Monatsh. Chem., 2005, 136, 819–854.
29. J. Zhang and T. Lu, Phys. Chem. Chem. Phys., 2021, 23, 20323–20328.
30. S. Manzetti and T. Lu, J. Phys. Org. Chem., 2013, 26, 473–483.
31. T. Lu and S. Manzetti, Struct. Chem., 2014, 25, 1521–1533.
32. T. Lu and Q. Chen, Chem.:Methods, 2021, 1, 231–239.
33. T. Lu and Q. Chen, in Comprehensive Computational Chemistry, 2024, pp. 240–264,  DOI:10.1016/b978-0-12-821978-2.00076-3.
34. T. Lu and F. Chen, J. Comput. Chem., 2011, 33, 580–592.
35. J. Plumet, ChemPlusChem, 2020, 85, 2252–2271.
36. V. G. Andrianov, Chem. Heterocycl. Compd., 1997, 33, 1115–1119.
37. C. Jelsch, K. Ejsmont and L. Huder, IUCrJ, 2014, 1, 119–128.
38. S. Emamian, T. Lu, H. Kruse and H. Emamian, J. Comput. Chem., 2019, 40, 2868–2881.

---

Footnote

† Electronic supplementary information (ESI) available: CCDC 2358185–2358187. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4ce01116c

---