Abstract
Spray drying method was used to prepare cocrystals of hexanitrohexaazaisowurtzitane (CL-20) and cyclotetramethylene tetranitramine (HMX). Raw materials and cocrystals were characterized using scanning electron microscopy, X-ray diffraction, differential scanning calorimetry, Raman spectroscopy, and Fourier transform infrared spectroscopy. Impact and friction sensitivity of cocrystals were tested and analyzed. Results show that, after preparation by spray drying method, microparticles were spherical in shape and 0.5–5 µm in size. Particles formed aggregates of numerous tiny plate-like cocrystals, whereas CL-20/HMX cocrystals had thicknesses of below 100 nm. Cocrystals were formed by C–HO bonding between –NO2 (CL-20) and –CH2– (HMX). Nanococrystal explosives exhibited drop height of 47.3 cm, and friction demonstrated explosion probability of 64%. Compared with raw HMX, cocrystals displayed significantly reduced mechanical sensitivity.
1. Introduction
Hexanitrohexaazaisowurtzitane (CL-20) and cyclotetramethylene tetranitramine (HMX) are usually used in high explosives and some propellants [1–3]. However, these materials have high mechanical sensitivity that strongly influences safety of formulas. Nowadays, as effective method of reducing sensitivity, cocrystal explosive technology is extensively developed [4]. In 2012, Bolton et al. [5] successfully grew 2/1 cocrystal of CL-20/HMX by slow evaporation of saturated 2-propanol solution. Crystallographic density was 1.945 g/cm3 at room temperature, and predicted detonation velocity was 100 m/s higher than that of HMX. Due to intermolecular hydrogen bonding, impact sensitivity of cocrystal was not significantly different from that of HMX. Cocrystals have high oxygen balance of −13.65%, which is 8.03% higher than that of HMX and 2.70% lower than that of CL-20. Therefore, cocrystals were considered attractive ingredients in low-signature propellant formulations. Based on the study of Bolton et al., a series of researches were conducted on CL-20/HMX cocrystal explosive in recent years. Liu et al. [6] studied pressure-induced effects in CL-20/HMX cocrystal explosive by density functional theory with dispersion corrections. CL-20 plays significant role in electronic structure of cocrystals. Mechanical properties of cocrystal presented great anisotropy. Through molecular dynamics method and with CL-20/HMX cocrystal explosive as main ingredient, poly(ester urethane) (Estane 5703) as block copolymer, and hydroxyl-terminated polybutadiene (HTPB) as binder, two polymer-bonded explosives (PBXs) were explored by Sun et al. [7]. Stability and compatibility of PBX were good when containing small amount of Estane 5703. Zhang et al. [8] studied packing structure of CL-20/HMX cocrystal by using Hirshfeld surface and fingerprint plot from densities, packing coefficients, and molecular interaction contributions. OH played dominant role in CL-20/HMX cocrystal explosive. Anderson et al. [9] prepared scale-up CL-20/HMX cocrystals from acetonitrile/2-propanol solution using resonant acoustic mixing.
Mechanical sensitivity can be reduced by decreasing particle size of explosives [10, 11]. Ultrafine explosive particles have advantage over larger ones in perspective of improving burning rate of solid propellants [12–14]. Nowadays, ultrafine cocrystal explosives attract scholarly attention. Using spray flash evaporation process, Spitzer et al. [15] prepared nanosized and submicron-sized CL-20/HMX cocrystals. Gao et al. [16] prepared synthesized nano-CL-20/HMX cocrystals by ultrasonic spray-assisted electrostatic adsorption. Qiu et al. [17] prepared CL-20/HMX cocrystal particles with mean size below 200 nm by bead-milling aqueous suspension of CL-20 and HMX for an hour. Alhalaweh et al. [18, 19] used spray drying method to prepare pharmaceutical cocrystals. Meanwhile, in our previous studies [20, 21], spray drying method was also employed to prepare ultrafine CL-20/TNT and nano-HMX/TNT cocrystals with low sensitivity. To obtain CL-20/HMX cocrystal explosive with low mechanical sensitivity, nano-CL-20/HMX cocrystal explosive was prepared by spray drying method. Furthermore, the present study investigated in detail morphology, size, thermal decomposition properties, impact sensitivity, and friction sensitivity of nanococrystals.
2. Experimental Sections
2.1. Materials
Raw CL-20 was provided by Liaoning Qingyang Special Chemicals Co., Ltd., China. Raw HMX was provided by Gansu Yinguang Chemical Industry Co., Ltd., China. Acetone (analytical reagent) was purchased from Tianjin Tianda Chemicals Co., Ltd., China.
2.2. Preparation of Nano-CL-20/HMX Cocrystal Explosive
First, raw CL-20 (2.19 g) and raw HMX (0.74 g) (molar ratio of CL-20 to HMX 2/1) were dissolved in acetone (60 mL) to form uniform cosolution under appropriate temperature and stirring. Second, B-290 Mini Spray Dryer was used to produce cocrystal explosive microparticles. The equipment was made in Switzerland BUCHI Labortechnik AG. Temperatures of feedstock and discharge were 75 and 55°C, respectively. Pump rate, rate of feedstock, and flow of spray drying gas (N2) were set at 35 m3/h, 5 mL/min, and 300 L/h, respectively. Finally, cyclone was employed to separate spray drying gas (N2) and cocrystal explosive microparticles. CL-20/HMX cocrystal microparticles were collected in glass collector.
2.3. Characterization
S4800 field-emission scanning electron microscope was used to characterize particle size and morphology of raw CL-20, raw HMX, and prepared cocrystal sample. Equipment was manufactured by Hitachi Limited in Japan.
X-ray diffractometer was used to identify raw CL-20, raw HMX, and prepared cocrystal sample. It was manufactured by China Dandong Haoyuan Instrument Co., Ltd. Testing conditions included target material (Cu) with tube voltage of 40 kV, tube current of 30 mA, a 5° start angle, and 50° end angle.
Differential scanning calorimetry (DSC) was carried out with Setaram DSC 131 instrument, which was made by Setaram of France. In the test, each 0.7 mg sample was placed in closed aluminum crucible with 30 µL volume and hole in lid. Samples were measured with temperature profile of 30°C to 350°C with heating rate of 10 K/min in nitrogen atmosphere and flow of 30 mL/min. As reference, CL-20/HMX mixture with the same molar ratio was tested under the same conditions.
Raman spectroscopy (Raman) was performed to explore causes of formation of cocrystals using equipment manufactured by Bio-Rad Co., Ltd., USA. Raman excitation light was provided by 532 nm line of argon-ion laser, and 0.5 mW power was used for all measurements. For each sample, spectrum was obtained by averaging 20 scans with measurement time of 5 s per scan.
Perkin Elmer Spectrum 100 Fourier transform infrared spectroscopy (FT-IR) spectrometer was used to confirm mechanism of cocrystal formation. FT-IR equipment was made by Perkin Elmer in US. Samples were carried out in KBr pellet by mercury-cadmium-telluride detector at stand-off distance of 5 m using mid-IR supercontinuum light source. Samples were measured in 4000–400 cm−1 wave number range with 1 cm−1 resolution.
ERL type 12-drop hammer apparatus was used to conduct impact sensitivity test according to GJB-772A-97 standard method 601.3 [22]. Testing conditions consisted of drop weight of 2.500 ± 0.002 kg, sample mass of 35 ± 1 mg, and relative humidity of 50%. Test results were represented by critical drop height of 50% explosion probability (H50).
WM-1 pendulum friction apparatus was used to test the friction sensitivity based on the GJB-772A-97 standard method 602.1 [22]. Experimental conditions are shown as follows: pendulum weight, 1.5 kg; swaying angle, 90°; pressure, 3.92 Mpa; sample mass, 20 ± 1 mg; and test number, 25. Friction sensitivity is expressed as explosion probability ().
3. Results and Discussion
3.1. The Size and Morphology of Cocrystal Particles
Figure 1 shows scanning electron microscopy images of raw CL-20, raw HMX, and CL-20/HMX cocrystal explosive. As shown in Figure 1(a), raw CL-20 particles are spindle-shaped, and particle size is 30–300 µm with uneven distribution. As presented in Figure 1(b), raw HMX sample particles are polyhedron, and average particle size is 100 µm. Figure 1(c) displays spherical-shaped microparticles with size ranging from 0.5 µm to 5 µm. As indicated in Figure 1(d), CL-20/HMX microparticles are composed of numerous tiny cocrystals, which are plate-like in shape. Plate-like cocrystals have thicknesses below 100 nm.
(a)
(b)
(c)
(d)
3.2. XRD Analysis
XRD was used to evaluate raw CL-20, raw HMX, and CL-20/HMX cocrystal explosive. Figure 2 presents observed patterns. By using MDI Jade 9 and PDF 2009 software, crystal form of raw CL-20 was identified as ε-type (PDF#00-050-2045), and raw HMX exhibited β-type crystal form (PDF#00-045-1539). As shown in Figure 2, all diffraction peaks of raw HMX appeared in CL-20/HMX cocrystals pattern. However, diffraction peaks of raw CL-20 did not appear in pattern of CL-20/HMX cocrystal explosive at 12.5° and 13.8°. Compared with patterns of raw CL-20 and raw HMX, at 11.4° and 13.2°, respectively, new diffraction peaks appeared on CL-20/HMX cocrystals pattern. These XRD patterns are similar to research results of Qiu. Changes showed that prepared samples did not simply mix with CL-20 and HMX, but they interacted to form new cocrystals. In addition, peak of cocrystal explosive was much weaker and wider than those of raw samples, indicating nanoscale size of cocrystals from another angle.
(a)
(b)
(c)
3.3. Thermal Decomposition Characterization
Figure 3 shows DSC curves of CL-20/HMX mixture with molar ratio of 2/1 and nano-CL-20/HMX cocrystal explosive at heating rate of 10 K/min. Two curves were materially different. Two peak decomposition temperatures existing were observed in CL-20/HMX mixture: 250.84 (CL-20 [10]) and 284.46°C (HMX [11]). However, in cocrystal curve, one peak decomposition temperature was noted at 246.98°C. Formation of CL-20/HMX cocrystal was further confirmed by changing peak decomposition temperature.
(a)
(b)
3.4. Raman Analysis
Figure 4 presents Raman spectra of raw CL-20, raw HMX, and CL-20/HMX cocrystals. Table 1 lists assigned major bands of Raman spectra of samples. Figure 4 and Table 1 show that majority of peaks of CL-20 and HMX shifted markedly in cocrystal Raman spectrum. Interestingly, peaks disappeared; they represent symmetric –NO2 stretch vibration in CL-20 at 1595.1 cm−1 to 1577.0 cm−1. Similar phenomenon was observed with peaks at 3027.8 cm−1, representing asymmetric –CH2– stretching vibration in HMX. Symmetric –CH2– stretch vibration shifted from 2991.7 cm−1 to 3006.9 cm−1 in HMX. These changes can be attributed to formation of C–HO bonding between –NO2 (CL-20) and –CH2– (HMX).
3.5. FT-IR Analysis
To determine whether interactions existed between CL-20 and HMX molecules in cocrystals, FT-IR spectroscopy was conducted on raw CL-20, raw HMX, and CL-20/HMX cocrystals. Figure 5 presents FT-IR spectra of raw CL-20, raw HMX, and CL-20/HMX cocrystals. Table 2 lists assigned major bands of FT-IR spectra of samples. Absorptions of cocrystals are similar to those of raw materials, but some peaks of CL-20 and HMX shifted visibly, as shown in cocrystals FT-IR spectrum. For CL-20/HMX cocrystals, absorptions at 1607, 1590, and 1568 cm−1, which represent asymmetric stretching of –NO2 of CL-20, shifted to 1604, 1576, and 1527 cm−1, respectively, and absorptions at 1383, 1329, and 1285 cm−1, which represent symmetric stretching of –NO2 of CL-20, shifted to 1395, 1331, and 1275 cm−1, respectively. Meanwhile, peaks at 3036 and 3027 cm−1, which represent –CH2– stretching of HMX, weakened and shifted to 3033 and 2856 cm−1, respectively. Per above analysis, we inferred that hydrogen bond interactions possibly exist between CL-20 (–NO2) and HMX (–CH2–) molecules. Conclusion of FT-IR spectra agrees with results of Raman spectra.
(a)
(b)
(c)
3.6. Mechanical Sensitivity Characterization
Mechanical sensitivity is challenge in explosive detonation under mechanical actions, such as impact and friction. Impact and friction sensitivity tests were performed on raw CL-20, raw HMX, CL-20/HMX mixture with molar ratio of 2/1, and nano-CL-20/HMX cocrystal explosives, and Table 3 provides test results. Nano-CL-20/HMX cocrystal explosive demonstrated remarkable decrease in impact and friction sensitivities in comparison with that of mixture. Drop height of nanococrystal explosive increased by 31.9 cm compared with CL-20/HMX mixture, and explosion probability of friction was 20% lower than that of raw HMX. Result suggests that nano-CL-20/HMX cocrystal explosive is difficult to detonate under impact and friction conditions.
4. Conclusions
Spray drying method was used to prepare nano-CL-20/HMX cocrystal explosive in 2/1 molar ratio. XRD and DSC patterns revealed that cocrystal explosive was not the intimate mixture of CL-20/HMX. Raman and FT-IR spectra imply that formation of CL-20/HMX cocrystals originates from C–HO hydrogen bonding between –NO2 (CL-20) and –CH2– (HMX). CL-20/HMX cocrystals are plate-like in shape with thicknesses below 100 nm. These tiny cocrystals do not disperse individually but rather agglomerate into microparticles ranging from 0.5 µm to 5 µm in size. Peak decomposition temperature of cocrystal explosives is 246.98°C, which is 3.86°C lower than that of CL-20. Compared with that of HMX, nanococrystals manifested significant decrease in impact and friction sensitivities, drop height increased by 141%, and friction probability reduced by 20%. Nano-CL-20/HMX cocrystal explosive exhibits insensitivity to mechanical action.
Competing Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
Acknowledgments
This work was funded by the Graduate Education Innovation Project of Shanxi Province (no. 2016BY119) and the 13th Graduate Student of Science and Technology in North University of China (no. 20161325).