Solid–solid phase transition study of ε-CL-20/binder composites

Abstract

The solid–solid phase transition of 2,4,6,8,10,12-hexanitro-2,4,6,8,10,12-hexaazaisowurtzitane (CL-20)-based composites is important for understanding the effect of additives in the application of explosives. In comparison to the phase transition in the solvent, the solid–solid phase transition process is more complex, and rarely studied. To reveal the effect of binders such as glycidyl azide polymer (GAP), isocyanate desmodur (N100), polyethylene (PE), polyethylenimine (PEI), and polyvinylpyrrolidone (PVP), which were added to the CL-20 explosive, we employed in situ X-ray diffraction with temperatures from 30 to 185 °C, X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FT-IR), differential scanning calorimetry and thermogravimetry (DSC/TG), together with field emission scanning electron microscopy (FE-SEM) and particle size distribution. The in situ XRD shows that all the five CL-20-based explosive composites had regular particle sizes similar to raw CL-20. The PE/CL-20 composite has a similar phase transition temperature (T_PT) to raw CL-20 (T_PT = 160 °C), while, for GAP/CL-20 and N100/CL-20, the T_PT appeared beyond 15 and 20 °C, respectively, compared with raw CL-20. For PVP/CL-20 and PEI/CL-20, the T_PT are delayed to 185 and 180 °C, respectively. The results are in agreement with the DSC and TG curves. In addition, we also investigated the molecular interactions between CL-20 and binders by FT-IR. It shows that no, or very weak, chemical bonds were formed between Cl-20 and PE; nevertheless, some chemical interaction was found between CL-20 and other binders (GAP, N100, PVP and PEI).

---

1. Introduction

Hexanitro-hexa-azaisowurtzitane (HNIW or CL-20) has been characterized as a high energy density compound in a number of high-performance explosive, propellant, and propelling agent compositions, and it can be regarded as a high-energy material for the next generation.^1,2 It has six N–NO₂ groups in its polycyclic structure. Both spatial orientation of these nitro groups, with respect to the five-member and six-member rings in the cage, and the differences in crystal lattice packing (as well as the number of molecules per unit cell) define four experimentally isolated polymorphs: α, β, γ and ε, as shown in Fig. 1b.^3,4 Different polymorphs lead to different physical and chemical properties such as thermal stability, sensitivity, density, and detonation velocity, which determine its application. Among these, the ε-phase is the most thermodynamically stable and also the one with the highest density.^5–7 The high density makes the ε-phase the most interesting as an energetic material. When heated to high temperatures (T = 160–170 °C), a solid–solid phase transition of CL-20 occurs from the ε-phase to the γ-phase. The abovementioned fast phase conversion results in volume expansion (ε → γ by 6%) and stress-cracking.^8–11 The resulting defect could be a source of hot spots and lead to super-rate burning that sensitizes the explosive. The effect of internal defect structure of the explosive crystal on the mechanical non-shock stimulus and shock sensitivity of a polymer bonded explosive is detrimental.

Fig. 1 (a) Ball-and-stick model of ε-CL-20; (b) the four polymorphs of CL-20; (c)–(e) preparation of CL-20-based composites.

As an energetic material, its sensitivity is one of the most important properties and determines its application. Therefore, it is necessary to study the phase transitions of CL-20. To date, related studies on phase transitions of CL-20 are concentrated on its recrystallization in solution or on the basis of theoretical calculations. The thermal property of CL-20 is an important parameter related to safety performance and applications. However, the study of the solid–solid phase transition of CL-20 has rarely been reported, this could be due to the difficult mechanism of this transition. Energetic materials and binders are the main ingredients in PBXs and propellant, and heating is usually employed during the course of manufacturing.^12–17 Cl-20-based composite explosives (including propellant and PBXs) have potential applications in many munitions.⁴ The solid–solid phase transitions of CL-20 may occur during the process.^18–20 Therefore, it is necessary to study the influence between binders and the phase transition of CL-20 in the heating process.

The objective of this study is to investigate the effects of five different binders (GAP, N100, PE, PVP, and PEI in Fig. 2) on the phase transition of CL-20. The weight ratio of binders/explosive in most ingredients of PBXs and propellant is about 10 : 90; thus, this weight ratio was selected in this study.^14,21,22 The binders were added into the CL-20 explosives by a simple mechanical mixed treatment. Furthermore, we tried to find the possible reasons of the abovementioned effects by investigating the molecular interactions between CL-20 and binders using Fourier transform infrared spectroscopy. It could provide effective theoretical basis for controlling the phase of CL-20 and reduce the risk of unintended phase transformation of CL-20. It might also be helpful for the real application of the CL-20-based explosive composites in military and civilian areas.

Fig. 2 Molecular structures of the binders.

2. Experimental section

2.1. Materials and chemicals

Explosive: 2,4,6,8,10,12-hexanitro-2,4,6,8,10,12-hexaazaiso-wurtzitane (C₆H₆N₁₂O₁₂, ε-CL-20, mean particle size 1.3 μm) was provided by Qingyang Chemical Industry Corporation. Binders: PE, GAP, N100, PVP, and PEI were all provided by Shenyang Chemical Works. Dimethylbenzene (AR grade) was provided by China National Pharmaceutical Group. Deionized water was produced by Nanjing Yipuyida Technology Corporation.

2.2. Preparation of CL-20-based explosive composites

The CL-20-based explosive composites with five binders (GAP, N100, PE, PVP, and PEI) were prepared (Fig. 1c) by adding CL-20 (binder : CL-20 = 10 : 90, wt%) into the suitable solvent (GAP, N100, and PE for dimethylbenzene and PVP and PEI for deionized water). After the binders were fully dissolved at appropriate respectively (Fig. 1d), CL-20-based explosive composites were obtained after stirring and air-drying at 25 °C (Fig. 1e).

2.3. Characterizations

2.3.1 X-ray diffraction. The phase content of CL-20 was determined by X-ray diffraction (XRD, Bruker D8 Advanced diffractometer, Germany) analysis using Cu-Kα (λ = 1.540598 Å) radiation at 40 kV and 40 mA and a graphite diffracted-beam monochromatic. The samples were packed into an amorphous silicon holder and the diffraction angle (2θ) was scanned from 5° to 80°, the scanning rate was 12° min⁻¹.

2.3.2 Fourier transform infrared spectroscopy (FT-IR). FT-IR spectrum was recorded on a Bruker-Tensor 27 spectrometer (FT-IR, Bruker, Germany). About 150 mg of KBr was ground in a mortar and pestle and 1 wt% of the solid sample was ground with KBr and the spectra of the mixture was recorded in the 4000–400 cm⁻¹ region.

2.3.3 Field emission scanning electron microscopy (FE-SEM). The morphology and surface appearance of the CL-20 particles were characterized by field emission scanning electron microscopy (FE-SEM, Ultra-55, Carl Zeiss, Germany) at an acceleration voltage of 10 kV after gold sputtering coating under the vacuum degree of 10⁻⁶ Pa for 50 s.

2.3.4 Particle size and size distribution. The average size and particle size distribution of CL-20 were calculated by a laser particle size analyzer.

2.3.5 In situ heat synchrotron X-ray diffraction. The measurements were carried out from 30 to 185 °C, at a heating rate of 0.1 °C s⁻¹. The temperature in the cell was monitored over time using a temperature transducer (Newport Electronics GmbH, Germany) before maintaining a constant temperature for 10 min. The diffraction angle (2θ) was scanned from 5° to 80° with steps of 12° min⁻¹.

The scanning data were collected at intervals of 10 °C from 30 to 140 °C and at intervals of 5 °C from 140 to 185 °C. After the scanning data was collected at 185 °C, the temperature was drawn back to 30 °C at a cooling rate of 0.5 °C s⁻¹. Before collecting the data, the temperature was maintained for at least 2 min.

2.3.6 Differential scanning calorimetry and thermogravimetry. Samples of raw CL-20 and prepared CL-20-based explosive composites were analyzed with differential scanning calorimetry and thermogravimetry (TG) (DSC/TG, PerkinElmer Diamond, USA). The conditions were as follows: sample mass: 3.00 mg, heating rate: 10 K min⁻¹, and nitrogen atmosphere (flow rate: 30 mL min⁻¹).

3. Results and discussion

3.1 Morphology and particle size characterization

The morphology and particle size of raw CL-20 (Fig. 3a) and the prepared CL-20-based explosive composites (GAP/CL-20, N100/CL-20, PE/CL-20, PVP/CL-20, and PEI/CL-20 shown in Fig. 3b–f) were revealed by FE-SEM. The morphologies of the CL-20-based explosive composites with five binders were spheroidic, which was similar to raw CL-20. It was evident from Fig. 3 that the particles with sizes of 1–2 μm for CL-20-based explosive composites were formed, and Cl-20 was well-distributed in the five binders. Fig. 4f shows the size distribution of raw CL-20 by a laser particle size analyzer, and it was in good agreement with the FE-SEM image shown in Fig. 3a.

Fig. 3 FE-SEM images of CL-20 and Cl-20-based explosive composites: (a) CL-20, (b) PE/CL-20, (c) GAP/CL-20, (d) N100/CL-20, (e) PVP/CL-20, and (f) PEI/CL-20.

Fig. 4 FT-IR spectra of Cl-20-based explosive composites: (a) PE/CL-20, (b) GAP/CL-20, (c) N100/CL-20, (d) PVP/CL-20, and (e) PEI/CL-20. (f) Size distribution of raw CL-20.

3.2 Molecular interactions

The mechanism of phase transition in solid state is very complex and difficult. Unlike material recrystallization in solution, there is no solvent in the solid-state polymorphic transition. New crystal seed must overcome a much higher energy barrier to nucleate and grow in the solid state; thus, phase transition usually occurs at high temperature or under pressure conditions.²³ Molecular interactions between CL-20 and the binders were likely to occur because of the large specific surface and close contact. For the foregoing reasons, the phase transition temperature of Cl-20 may change.

The interactions between CL-20 and the binders were studied by means of FT-IR spectroscopy. Fig. 4 shows the infrared spectra of pure CL-20 and the five CL-20-based composites at room temperature. In Fig. 4a, we observed that the PE/CL-20 composite did not show any new peaks, indicating that no or very weak chemical bonds were formed between Cl-20 and PE, because of the stable chemical properties of CH₂ functional groups of PE.

As we can see in Fig. 4b, the peak at 3043 cm⁻¹ is attributed to the stretching vibration of the C–H bond of CL-20. When mixed with GAP, it exhibits a jump toward a higher wavenumber (2 cm⁻¹) compared with that of the pure CL-20.^24,25 The wavenumber jump coincides with a possible interaction between the CH group of CL-20 and GAP, mainly because of the high reactivity of the CH group caused by the two adjacent ammonium nitrate molecules. Fig. 4c shows the spectral region between 2540 and 1950 cm⁻¹, corresponding to the region of the stretching vibrations of the N=C=O bonds of the N100. The resonant regions for the mixture shrink with respect to those for the pure N100 in the FT-IR, indicating possible chemical bonds between the abovementioned group and CL-20. The shift in PVP/CL-20 composite in Fig. 4d is higher (18 cm⁻¹) than that in the purity PVP, indicating possible chemical bonds with CH groups of CL-20 and C=O groups of PVP. The peak at 3043 cm⁻¹ is assigned to CL-20, while the broad peak at about 3000 to 3750 cm⁻¹ is assigned to the vibration of the NH₂ and NH, as shown in Fig. 4e.²⁶ While mixed, the CH bond is displaced to a higher wavenumber (2 cm⁻¹) and the broad peak becomes narrower, because some of the NH₂ or NH groups probably had chemical forces with the NO₂ group of CL-20.²⁷

Fig. 5 XRD patterns of the conformers from four polymorphs of CL-20.

Fig. 4 shows that the molecular interactions between CL-20 and binders depend on the nature of the binders. When heated, the abovementioned interactions may affect spatial orientation of the nitro groups and cause differences in crystal lattice packing of CL-20; thus, it is likely to lead to a different phase transition temperature for CL-20 (¹H LNMR and ¹H SNMR were employed to study the chemical interaction of CL-20 and binders, but the results are not shown here).

3.3 X-ray diffraction characterization

Fig. 5 shows the four typical XRD patterns of CL-20 crystal forms that could be divided into α, β, γ and ε forms (Fig. 5a–d). For the ε and γ forms of CL-20, there was a region of diffraction angle (2θ) at 12.0–14.5°, as shown in the inner magnified image of Fig. 5. The small 2θ range between 12° and 14.5° is chosen because the material showed the strongest reflection in this range. Clear differences could also be found: the peaks at 12.8° and 13.8° of ε-CL-20 were replaced by new peaks at 12.9°, 13.3° and 14.2° of γ-CL-20. The disappearance and new formation of reflections indicated a structural change. Thus, changes in the diffraction pattern due to structural changes could be pointed out most sensitively in this region. Therefore, we can study the difference in the five explosive composites by in situ XRD to investigate the crystal forms of Cl-20.

3.4 In situ X-ray diffraction

The crystal forms of CL-20 in Cl-20-based explosive composites are greatly affected by temperature. Thus, it was possible to monitor changes at different temperatures in the diffraction pattern. Fig. 6a shows the in situ XRD patterns of pure CL-20 powder. It can be concluded that at a low temperature, the structure is the same as that of ε-CL-20 shown in Fig. 5a. There are two strong reflections (2θ = 12.6° and 13.8°) and a weaker one (12.8°). The whole curve shifts left with increasing temperature, and is due to the expansion of the sample, when heated. The crystal form did not change during this period. Beginning at 160 °C, a new reflection appears at 12.8–12.9°, it is the characteristic diffraction peak of γ-CL-20 (Fig. 5d). The formation of new reflection indicates a structural change (ε → γ).²¹ The new peak becomes more apparent with continuous increase in temperature, it is concluded that more and more ε-CL-20 changed to γ-CL-20. The curve has no noticeable change when the temperature decreases from 185 to 30 °C; this shows that the change in crystal form is an irreversible process under this condition. Similarly, as shown in Fig. 6c and d, the crystal transition temperatures of GAP/CL-20 and N100/CL-20 composites are 145 and 140 °C, respectively. The transition temperatures decrease by 15 and 20 °C, respectively, compared to that of pure CL-20. In addition, we can see that the transition temperature of the PE/CL-20 composite (Fig. 6b) is 160 °C, the same as for pure CL-20. No new peaks appear before the highest temperature, 185 °C, in the PVP/CL-20 composite in Fig. 6e; this shows that the ε-CL-20 does not change during the whole heating process. Thus, the phase transition temperature increased by about 25 °C in the PVP/CL-20 composite. Fig. 6f shows that the transition temperature of PEI/CL-20 composite increased by about 20 °C compared to that for pure CL-20. Just like pure Cl-20, the transitions for ε → γ crystal forms of all the abovementioned composites are irreversible process under the heating process. All the in situ XRD patterns of CL-20 and its composites are consistent with the FT-IR curves.

Fig. 6 In situ XRD of CL-20 and Cl-20-based explosive composites: (a) CL-20, (b) PE/CL-20, (c) GAP/CL-20, (d) N100/CL-20, (e) PVP/CL-20, and (f) PEI/CL-20.

3.5 Thermal stability of the materials

Fig. 7 shows the DSC and TG results of pure CL-20 and Cl-20-based composites. The DSC curve of pure CL-20 in Fig. 7a shows an endothermic peak at 166.9 °C and an exothermic peak at 231.3 °C (curve 1). The TG profile of pure CL-20 in Fig. 7b did not show any weight loss at 166.9 °C, ruling out any possible loss of its fragments (curve 1). There is a sharp decline in peak on TG curves at about 231.3 °C due to the decomposition of CL-20, which is consistent with the DSC curve. Based on a previous study,⁹ the endothermic peak at 166.9 °C is the phase transform temperature (ε → γ) of CL-20.

Fig. 7 DSC (a) and TG (b) curves of CL-20 and Cl-20-based explosive composites.

Compared to the pure CL-20, the endothermic and exothermic peaks of the abovementioned composites have been affected by different binders to different degrees. The endothermic peak of PE/CL-20 remains consistent with pure CL-20, and does not show any apparent shifts (Fig. 7a (curve 2)). However, other endothermic peaks of different composites shift to lower temperatures or disappear. As we can see in Fig. 7a (curves 3 and 4), the phase transition temperatures of N100/CL-20 and GAP/CL-20 decreased by 25 and 4.5 °C, respectively, compared with that of the pure CL-20. Fig. 7a (curves 5 and 6) shows that the endothermic peaks of both PVP/CL-20 and PEI/CL-20 disappeared. It shows that ε-CL-20 does not change before its respective decomposition temperatures in the abovementioned two composites. The phase-transform temperature of all the abovementioned composites in DSC and TG was consistent with the result from in situ XRD and FT-IR.

Exothermal reactions have been one of the most important explosive properties. The exothermal peak can be elucidated by the emergence of nitro-to-nitrite rearrangement in the molecule, which leads to the destruction of the conjugated system and thus the fracture of nitro group, resulting in the formation of nitrogen monoxide (NO₂) in the CL-20 molecule. The exothermal peaks in DSC show that the five CL-20-based composites efficiently retain their energetic properties, even when 10 wt% binders were added into explosives. However, the phase-transition temperatures were changed in varying degrees by different binders, causing different exothermal temperatures. The results indicate that there are some chemical or other effects between CL-20 and binders molecules in the solid state. The details of the related temperatures of CL-20 and CL-20-based explosive composites with five binders are shown in Table 1.

Table 1 Related temperatures of CL-20 and Cl-20-based explosive composites^a

	CL-20	PE/CL-20	GAP/CL-20	N100/CL-20	PVP/CL-20	PEI/CL-20
a T1: initial phase transition temperature, T2: phase transition temperature, and T3: exothermal temperature.
T1/°C	162.6	158.9	154.4	135.4	—	—
T2/°C	166.9	165.4	162.4	141.9	—	—
T3/°C	231.3	235.2	222.9	199.5	230.7	234.2

---

4. Conclusions

In summary, five CL-20-based explosive composites were successfully prepared with a weight ratio of 10 : 90. The CL-20 particles of GAP, N100, PE, PVP, and PEI were well-distributed with binders. The results showed that the PE/CL-20 composite has the similar phase-transition temperature as that of the raw CL-20, while, for GAP/CL-20 and N100/CL-20, the T_PT was observed beyond 15–20 °C. For PVP/CL-20 and PEI/CL-20, the T_PT are delayed to relatively high temperatures by 20–25 °C. FT-IR shows that no, or very weak, chemical bonds were formed between Cl-20 and PE; nevertheless, some chemical interaction were found in the GAP/CL-20, N100/CL-20, PVP/CL-20 and PEI/CL-20 composites, which may lead to different phase transition temperature of CL-20. The study may provide an effective theoretical basis for controlling the phase of CL-20 and reducing the risk of unintended phase transformation of CL-20.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (No. 11172276 and 11502242), Development Foundation of CAEP (No. 2014B0302041), and Young Talent Foundation of Institute of Chemical Materials (KJCX-201405, KJXZ-201403), and Doctoral Program of Southwest University of Science and Technology (Grant No. 14zx7105), and CAEP (No. 15zh0058).

Notes and references

1. R. Simpson, P. Urtiew, D. Ornellas, G. Moody, K. Scribner and D. Hoffman, Propellants, Explos., Pyrotech., 1997, 22, 249–255.
2. N. B. Bolotina, M. J. Hardie, R. L. Speer and A. A. Pinkerton, J. Appl. Crystallogr., 2004, 37, 808–814.
3. P. Goede, N. V. Latypov and H. Östmark, Propellants, Explos., Pyrotech., 2004, 29, 205–208.
4. R. Sivabalan, G. Gore, U. Nair, A. Saikia, S. Venugopalan and B. Gandhe, J. Hazard. Mater., 2007, 139, 199–203.
5. M. Xiang, Q. Jiao, Y. Zhu, J. Yu and L. Chen, J. Therm. Anal. Calorim., 2014, 116, 1159–1163.
6. O. Ordzhonikidze, A. Pivkina, Y. Frolov, N. Muravyev and K. Monogarov, J. Therm. Anal. Calorim., 2011, 105, 529–534.
7. S. Jin, Q. Shu, S. Chen and Y. Shi, Propellants, Explos., Pyrotech., 2007, 32, 468–471.
8. J. Li and T. B. Brill, Propellants, Explos., Pyrotech., 2007, 32, 326–330.
9. R. Turcotte, M. Vachon, Q. S. Kwok, R. Wang and D. E. Jones, Thermochim. Acta, 2005, 433, 105–115.
10. J. C. Gump and S. M. Peiris, J. Appl. Phys., 2008, 104, 083509.
11. Q. Tian, G. Yan, G. Sun, C. Huang, L. Xie, B. Chen, M. Huang, H. Li, X. Liu and J. Wang, Cent. Eur. J. Energ. Mater., 2013, 10, 359–369.
12. R. M. Vrcelj, H. G. Gallagher and J. N. Sherwood, J. Am. Chem. Soc., 2001, 123, 2291–2295.
13. A. Sikder and N. Sikder, J. Hazard. Mater., 2004, 112, 1–15.
14. D. Drodge, D. M. Williamson, S. Palmer and W. Proud, Shock Compression Of Condensed Matter-2011: Proceedings of the Conference of the American Physical Society Topical Group on Shock Compression of Condensed Matter, 2012.
15. C. Zhang, J. Li, Y.-J. Luo and B. Zhai, J. Energ. Mater., 2015, 33, 305–314.
16. D. Bhowmik, V. S. Sadavarte, S. M. Pande and B. S. Saraswat, Cent. Eur. J. Energ. Mater., 2015, 12, 145–158.
17. A. Azoug, A. Constantinescu, R. Nevière and G. Jacob, Fuel, 2015, 14, 39–47.
18. U. Nair, R. Sivabalan, G. Gore, M. Geetha, S. Asthana and H. Singh, Combust., Explos. Shock Waves, 2005, 41, 121–132.
19. R. M. Vrcelj, J. N. Sherwood, A. R. Kennedy, H. G. Gallagher and T. Gelbrich, Cryst. Growth Des., 2003, 3, 1027–1032.
20. H. G. Gallagher, R. M. Vrcelj and J. N. Sherwood, J. Cryst. Growth, 2003, 250, 486–498.
21. R. B. Patel, V. Stepanov, S. Swaszek, A. Surapaneni and H. Qiu, Propellants, Explos., Pyrotech., 2015, 40, 210–214.
22. D. Drodge, D. Williamson, S. Palmer, W. Proud and R. Govier, J. Phys. D: Appl. Phys., 2010, 43, 355–403.
23. P. Zhang, J.-j. Xu, X.-y. Guo, Q.-j. Jiao and J.-y. Zhang, J. Therm. Anal. Calorim., 2014, 117, 1001–1008.
24. A. Abdelghany, M. S. Mekhail, E. Abdelrazek and M. Aboud, J. Alloys Compd., 2015, 646, 326–332.
25. H. Zhong, Z. Huang, G. Zhao, S. Wang, G. Liu and Z. Cao, J. Mater. Res. Technol., 2015, 4, 151–161.
26. S. M. Husssein, O. H. Shihab, S. S. Ibrahim and N. M. Ahmed, World J. Nano Sci. Eng., 2015, 5, 1–9.
27. Z. Liu, E. Abedin, S. Zein and F. Endres, ChemPhysChem, 2015, 16, 970–977.

---