Assembling Hybrid Energetic Materials with Controllable Interfacial Microstructures by Electrospray.

Constructing hybrid energetic materials (HEMs) consisting of nanothermites and organic high explosives is an efficient strategy to regulate the reactivity of energetic composites. To investigate the role of interfacial microstructures in determining the reactivity of HEMs, we employ electrospray, one ramification of electrohydrodynamic atomization, to assemble Al/CuO and hexanitrohexaazaisowurtzitane (CL-20) into composites with various morphologies from different solvent systems. The morphology and compositional information of the assembled clay-like or granular HEMs, which are obtained from ketone, ester, or mixtures of alcohol and ether, are confirmed by scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS). The phase transition of CL-20 due to the fast evaporation of charged droplets and insufficient time for recrystallization is studied by Fourier transform infrared spectroscopy (FTIR). Thermogravimetric-differential scanning calorimetry (TG-DSC) is applied to investigate the thermodynamic behaviors and synergistic effect of the nanothermite and high explosive. Enhancements in combustion performance and pressurization characteristics of the as-sprayed HEMs have been observed through open burn tests and pressure cell tests. Granular HEMs show high gas generation and high pressurization rate, while nitrocellulose (NC) fibers existing in the clay-like HEMs would weaken the reactivity to a certain extent. HEMs obtained from the mixture of n-propanol and diethyl ether, in which nano-CL-20 exists as independent particles rather than a matrix, exhibit high gas generation but low pressurization rate. The results indicate that the energy releasing performance of the prepared HEMs can be readily regulated by constructing various interfacial microstructures to satisfy the broad requirements of energy sources.

Introduction

Thermites, one class of pyrotechnics, which can release an enormous amount of heat rapidly from the redox reaction between an oxide and a fuel, are widely used in the fields of both the military and civilian, such as welding, antibiological, micropropulsion, and igniter applications. However, due to low gas generation and heterogeneous dispersion of reactants, the energy releasing performance of thermites is still feasible to be improved.

Increasing the contact area of oxide and fuel, scilicet shortening the characteristic lengths of heat and mass transport, is the most effective route to enhance reactivity and combustion efficiency.[1] Numerous strategies have been proposed to improve the reactivity of thermites. The simplest strategy is reducing the characteristic size of fuel and oxide to a nanoscale, namely, a nanothermite, resulting in the dramatic increase of the flame propagation rate up to several km/s. Because the sintering of nanoaluminum occurs much faster than reaction propagation,[2] energetic polymers, like nitrocellulose[3,4] and fluorine-containing polymer,[5,6] have been introduced to elevate combustion efficiency. Materials with high heat conductivities, i.e., carbon fibers, graphene, graphene oxide, and silicon,[7] are also used as positive additives to improve reactivity. Constructing hybrid energetic materials (HEMs) by combining organic high explosives, i.e., hexogen (RDX),[6,8−10] octogen (HMX),[11] and hexanitrohexaazaisowurtzitane (CL-20),[12−14] with nanothermites shows amazing enhancement of reactivity and even leads to the deflagration to detonation transition. Zhu[12] reported that integrating CL-20 into an Al/CuO array would lead to a more violent combustion with a much brighter and larger flame. Tests conducted by Yang[15] revealed that the addition of RDX would improve the max pressure of nano-Al/Fe2O3, ∼0.25 MPa vs 1.05 MPa. Meanwhile, hybrid energetic materials consisting of RDX and nano-Al/CuO could evenly generate detonation in a confined volume, reported by Thiruvengadathan[8] and Qiao,[10] respectively.

Compared to changing the formulation of thermites, it seems that assembling energetic materials with organized interfacial structures by new preparation methods is more efficient,[16] such as sol–gel,[17] incorporating with MOFs,[18] sputtering,[12,19] arrested reactive milling, and so on. Electrospray, one ramification of electrohydrodynamic atomization,[20] is a mature technology widely used in the realm of mass spectrometry, propulsion system of satellites, encapsulation, particle production, surface coating, and film deposition. In the most recent decade, electrospray has become a rapidly emerging route for the preparation of energetic materials with designed microstructures.[21,22] Micro- or nanoparticles, nanofibers,[23] and delaminate films[24] based on high explosives or nanothermites have been explored to increase energy releasing efficiency and reaction rate under control.[25,26]

According to the scale laws of electrohydrodynamic spraying,[27,28] operating parameters (such as applied voltage, flow rate, distance between the nozzle and substrate, and so on) and the properties of precursors (such as surface tension, viscosity, volatilization rate, electric conductivity, and so on) are key factors determining the microstructures of as-sprayed particles. In a typical electrospray process, as shown in Figure , a charged jet is emitted from the apex of a Taylor-cone, which is induced by the competition between surface tension and electrostatic force, which can be explained by the Rayleigh limit. The size of the droplet produced (Dd) in a cone-jet spraying can be predicted by eq where a is a constant, Q is the flow rate, ε0 is the dielectric constant in vacuum, ρs is the solution density, σ is the surface tension, and γ is the solution conductivity. After fast evaporation of solvents during flight to substrate, the diameter of deposited particles (Dp) can be predicted by eqs and 3where w is the mass fraction of the solute, ρp is the solid material density, and C is the precursor concentration. It should be noted that the solubility of CL-20 and nitrocellulose (NC) in different solvents would affect the process of recrystallization[29] or dissolution, leading to crystal changes or broadening of particle size distribution. So, changing the solvent system not only easily adjusts the diameter but also controls the microstructures of deposited HEMs.

Figure 1

Electrospray process of assembling HEMs consisting of Al/CuO and CL-20.

In this paper, we try to investigate the effect of interfacial microstructures on the reactivity of hybrid energetic materials (HEMs) consisting of Al/CuO and CL-20 from different solvents prepared by electrospray. To observe the synergistic effect between nanothermites and high explosives, the composition information and reactivity of as-sprayed samples with various morphologies are systematically characterized. The assembled HEMs show much more intense energy releasing performance determined by the interfacial microstructure than the physical mixtures and can meet the broad requirements of energy sources.

Results and Discussion

Morphologies and Compositional Information

The SEM images in Figure display the morphologies of the as-prepared HEMs obtained from different solvents. Compared to the disorder dispersion of Al/CuO, Al/CuO/CL-20, or Al/CuO/NC/CL-20(UM), shown in Figure S1, electrospray has efficiently assembled the energetic components into microparticles with narrow size distribution in the range of 1.5 to 10 μm. As summarized in Table , the microparticles of HEMs (ES, EAC) are the largest with an average diameter of 6.35 μm, while HEMs (ES, EAC/DMK-1:5) are the smallest with an average diameter of 2.46 μm. These particles are in various shapes, spherical (DMK), erythrocyte (EAC), elliptical spherical (EA/DEE-3:1), and irregular shapes (NPA/DEE-1:1 and mixtures of acetone and ethyl acetate). When ethyl acetate participates in the electrospray processes, the obtained materials are in the state of clay-like, attributed to the existence of NC fibers. As shown in Figure S1e,f, NC fibers in the clay-like HEMs bind microparticles as plasticine that can be pinched into any shapes, while the granular HEMs are loose powders that would easily scatter in collecting vials due to electrostatic force.

Figure 2

SEM images of the as-prepared HEMs obtained from various solvents, (a) acetone, (b) acetone/ethyl acetate (1:1), (c) ethyl acetate, (d) ethanol/diethyl ether (3:1), (e) n-propanol/diethyl ether (1:1). (f) Diagrams of different interfacial microstructures.

Table 1

Characteristic Sizes of the As-Sprayed HEMs without Consideration of NC Fibers

solvent	median diameter (μm)	diameter range (μm)	form of products
ethyl acetate	6.35	3.4–10.0	clay-like
acetone	3.22	1.4–5.3	granular
EA/DEE (3:1)	4.61	3.2–6.3	granular
NPA/DEE (1:1)	3.86	1.6–7.1	granular
EAC/DMK (3:1)	4.14	2.4–7.2	clay-like
EAC/DMK (1:1)	2.98	1.7–5.5	clay-like
EAC/DMK (1:2)	3.84	1.8–7.9	clay-like
EAC/DMK (1:3)	2.91	1.2–4.8	clay-like
EAC/DMK (1:5)	2.46	1.3–4.6	clay-like

The polymer fibers of NC bonded to the assembled microparticles are generated by accompanying electrospinning, which can be regarded as a variant of electrospray. Increasing the viscosity and viscoelasticity of precursors would generate sufficient chain entanglement, leading to a transition from electrospray to electrospinning. When operating temperature is higher than 25 °C, no fibers would appear, as shown in Figure S2. So, we can deduce that a low temperature would increase the viscosity of precursors containing ethyl acetate, further leading to the appearance of nanofibers. The rest of the assembled HEMs are in the state of loose powders without fibers (granular). We try to eliminate the fibers by introducing acetone into ethyl acetate. However, NC fibers still appear even when the volume ratio of acetone reached up to 83% (ES, EAC/DMK-1:5), as shown in Figure S3, leading to bad pressurization performances. So, it can be concluded that ethyl acetate is the key factor to producing NC fibers even though at relatively low concentrations.

Enlarged SEM images and EDS mapping images in Figure show that nanospheres of Al and CuO disperse evenly in the matrix of NC and CL-20. The surfaces of the assembled HEMs are really rough with the organized intimate contact of nAl and nCuO particles and have a number of voids due to the shrinking of droplets caused by the fast evaporation of solvents. Noticeably, no visible crystal of CL-20 appears on the surface of all samples due to the very short time and the presence of Al or CuO nanoparticles restraining the growth of the nucleus, as explained elsewhere.[13] Element maps can provide the detailed interfacial information of the assembled samples. Figure f illustrates the homogeneous dispersion of NC and CL-20 (distribution of element N), CuO (distribution of element Cu), and Al (distribution of element Al) on the surface and intimate contact of components. The bright spots in the Al map reveal the existence of excessively large nano-Al particles rather than the agglomeration of nano-Al, which is confirmed by SEM images.

Figure 3

Enlarged SEM images show detailed information of the as-prepared HEMs obtained from various solvents, (a) acetone, (b) acetone/ethyl acetate (1:1), (c) ethyl acetate, (d) ethanol/diethyl ether (3:1), (e) n-propanol/diethyl ether (1:1). (f) Elemental maps for Al, Cu, and N (N = green, Al = yellow, and Cu = blue) of HEMs (ES, DMK).

Compared to other polymorphs, ε-CL-20 is the most stable one with the highest energy density. However, the phase transition of CL-20 easily occurs during solvent treatment or a temperature rise, especially in the process of recrystallization.[26,29,30] XRD and FTIR are powerful tools to recognize the polymorph of CL-20. Nevertheless, the low content of CL-20 (5 wt %) results in no characteristic peak of CL-20, which can be recognized in the XRD pattern (Figure S4). As demonstrated in Figure , the raw CL-20 is in the ε-phase with a featured quartet in low frequency (738.2–758.2 cm–1). After recrystallization, the quartet of raw ε-CL-20 is broadened into a doublet implying a phase transition. Other researchers have confirmed that the raw ε-CL-20 would convert into a β-phase due to the ultrafast evaporation and insufficient recrystallization in similar electrospray processes.[26,29,30] However, in this study, the phase of CL-20 after electrospray cannot be concluded with no valid characteristic peaks under the interference of the same quantity of NC.

Figure 4

FTIR spectra of as-sprayed HEMs obtained from different solvents and raw CL-20.

Open Combustion Performance

Combustion behaviors in open air can evaluate the reactivity of energetic materials, as shown in Figure , and are recorded by a high-speed camera at a recording rate of 15000 fps. The leftmost images in each sequence of snapshots are labeled as 0 ms recording the first appearance of the visible flame. Meanwhile, the rightmost images record the most violent combustion flame of each sample. It seems that assembling CL-20 and NC into nano-Al/CuO by electrospray makes for a more efficient combustion. Compared to the raw CL-20 and the ultrasonic mixed samples, shown in Figure S5, including Al/CuO (hexane) and Al/CuO/CL-20 (hexane), the as-sprayed samples show more intense deflagration with shorter burn duration and faster reaction propagation. HEMs (ES, DMK) exhibit the most intense deflagration with a valid combustion duration of only 0.13 ms and complete burn without obvious ejected sparks. The clay-like HEMs (ES, EAC) possess similar combustion duration to HEMs (UM DMK), and a certain amount of unreacted materials is ejected out of the metal base by hot gaseous products, leading to a darker flame. However, some unreacted materials are also ejected and inflame outside in the case of as-sprayed loose powders (from the mixed solvent of NPA/DEE or EA/DEE).

Figure 5

Combustion processes of as-prepared HEMs are recorded by a high-speed camera under conditions of 15000 fps, min aperture, and 15 μs exposure. (a) Acetone, (b) ethyl acetate, (c) ethanol/diethyl ether (3:1), (d) n-propanol/diethyl ether (1:1), and (e) ultrasonically mixed products from acetone.

Pressure Cell Results

Figure shows the pressurization characteristics of the as-prepared materials, including the max pressure (Pmax) and pressurization rate (dP/dt), conducted using the pressure cell. The raw CL-20 combusts very gently when ignited by a heating wire and produces almost all gaseous products resulting in the highest Pmax = 1.17 MPa and the slowest dP/dt = 0.053 MPa/ms, entirely different from its detonation performance when initiated by a detonator. Incorporating CL-20 and NC into Al/CuO (0.50 MPa and 0.72 MPa/ms) by electrospray favors improved gas generation but slow reaction rate except HEMs (ES, DMK). The physical mixture of Al/CuO/NC/CL-20 shows the worst reactivity (0.53 MPa and 0.14 MPa/ms) due to its chaotic microstructure.

Figure 6

Pressurization characteristics of as-prepared Al/CuO/NC/CL-20 obtained from various solvents, (a) the max pressure and (b) the pressurization rate (dP/dt). The dashed lines exhibit the pressurization behaviors of raw CL-20, Al/CuO (UM), and HEMs (UM) as the contrast samples.

The pressurization performance of HEMs (ES, DMK) is superior to the other samples (ES or UM), with the highest Pmax of 0.71 MPa and the fastest dP/dt of 0.75 MPa/ms. The clay-like HEMs (ES, EAC, or a mixture of EAC/DMK) show lower Pmax, similar to the physical mixture (0.53 MPa), and moderate pressurization rate performance, listed in Figure S6, due to the existence of NC fibers among microparticles. It seems that the unsprayed CL-20 particles hinder the reactivity. HEMs (ES, NPA/DEE), in which the bulk of CL-20 exists as unchanged nanoparticles due to low CL-20 solubility in n-propanol, exhibit nearly the highest Pmax (0.71 MPa) but comparatively low pressurization rate (0.14 MPa/ms) to the physical mixture. These results indicate that the energy releasing performance of HEMs can be readily regulated by constructing various interfacial microstructures to satisfy the broad requirements of energy sources. For example, for a microthruster array based on a solid propellant, combustion of the filled propellant should produce abundant gaseous products at a mild rate. For the application of antibiological agents, high temperature products with rapid energy release rate to inactivate spores is a basic need. For the application of MEMS detonators, fast reactivity leading to a fast transition of deflagration to detonation is a requisite.

Thermal Analysis

Figure shows the typical DSC curves and TG curves of Al/CuO and HEMs conducted at a heating rate of 10 K/min prepared by ultrasonic mixing and electrospray, respectively. As shown in Figure a, only one doublet exothermic peak caused by aluminothermic reaction locates in the DSC curve of Al/CuO (UM). After introducing NC and CL-20 into Al/CuO by ultrasonic mixing or electrospray, four exothermic peaks appear, named PeakI, PeakII, PeakIII, and PeakIV. PeakI implies the merge of released heat caused by the decomposition of NC and CL-20.[6,15] The doublet exothermic peak of Al/CuO(UM) is divided into three steps, i.e., the peaks of II, III, and IV, similar to a study by Zhou et al.[31] However, the reaction with obvious weight loss occurring in PeakII is still unclear. The appearance of PeakII (from 350 to 450 °C) may be interpreted as the reaction of Al/CuO in low temperature, or further decomposition of residue decomposition products of NC and CL-20 for HEMs (UM), or the reaction between decomposition products of NC and CL-20 and nano-Al. PeakIII (from 530 to 650 °C) implies the solid–solid diffusion reaction of Al and CuO that occurs below 660 °C (the melting point of Al), while PeakIV (from 650 to 670 °C) can be attributed to the liquid–solid reaction of molten Al and solid CuO.[32]

Figure 7

(a) DSC curves and (b) TG curves of the as-prepared samples.

The total weight loss of Al/CuO(UM) is nearly negligible, only 1.8%. However, for HEMs (UM or ES), the process of weight loss can be divided into three stages, named LossI (from 150 to 230 °C), LossII (from 230 to 420 °C), and LossIII (from 450 to 700 °C), corresponding to the decomposition of NC and CL-20, the aluminothermic reaction at low temperature and at high temperature, respectively. When nanoparticles of Al and CuO are assembled into the matrix of NC and CL-20 by electrospray, a synergistic effect between nanothermites and high explosives or enhanced energy release performance appears with higher weight loss and higher weight loss rate than HEMs prepared by ultrasonic mixing, as shown in Figure b.

Figure and Figure S7 demonstrate the exothermic peak temperatures (T), heat release values (H), and weight losses of the as-prepared samples. It seems that the total released heat and weight loss of HEMs, as summarized in Table , have a positive correlation to their pressurization characteristics. Higher weight loss and larger heat release result in a higher pressurization rate and higher max pressure. The exothermic peak of (shown inFigure S8) raw CL-20 occurs at 238 °C, which is higher than that of HEMs. This is because both aluminum and coper oxide have an evident catalytic effect on the decomposition of CL-20,[12,33] leading to lower decomposition onset temperature and activation energy. Interestingly, compared to HEMs (UM), the exothermic peak of aluminothermic reaction for HEMs (ES) all shifts toward higher temperature, indicating more thermal stabilization. Intimate contact of components leads to the reaction of nanothermites that occurs in lower temperature as the PeakIV heat of HEMs (UM) is at least 2 times higher than (81.3 J/g vs 48.5 J/g) than that of HEMs (ES).

Figure 8

Thermal analysis results of prepared samples, (a) temperature of exothermic peaks and (b) weight loss in different stages.

Table 2

TG-DSC Results of Prepared Samples

sample	total heat (J/g)	weight loss (%)
Al/CuO	792.0	1.8
HEMs (UM)	721.7	12.2
HEMs (ES EAC)	650.1	21.2
HEMs (ES DMK)	762.6	17.4
HEMs (ES EAC/DMK)	778.7	20.2
HEMs (ES EA/DEE)	754.3	17.9
HEMs (ES NPA/DEE)	688.7	20.9

A weight loss of 12.2% is observed for HEMs (UM), which is about 10% higher than Al/CuO, whereas all of the HEMs prepared by electrospray have higher weight losses and higher weight loss rates during the overall heating process, implying more reaction completeness or synergistic effect between nanothermites and high explosives due to the controlled homogeneous distribution of components and larger contact area. It seems that the residue decomposition products of NC and CL-20 bring forward the reaction of Al and CuO at low temperature with more gaseous products. In the first stage, the weight loss of HEMs (UM) is only 5%, meaning the incomplete decomposition of NC and CL-20 (total mass ratio is 10%). LossII accounts for the majority of the total weight loss, which does not exist in Al/CuO(UM). In the range of 450 to700°C, for Al/CuO(UM), solid–solid reaction and liquid–solid reaction produced a tiny gaseous product with almost invisible mass loss. However, for HEMs (ES), 7 to 20% of the total weight loss occurs in LossIII. HEMs (ES) with arranged microstructures show enhanced reactivity with more gaseous products. The weight losses of HEMs (UM) in LossII and LossIII are much less than that of HEMs (ES), resulting in lower Pmax and dP/dt. For as-sprayed HEMs (ES), the max values of LossI, LossII, and LossIII reached up to 7.9, 9.5, and 4.2%, respectively. Moreover, the minimum total weight loss of HEMs (ES) reached up to 17.1%, which is 40% higher than HEMs (UM).

Thermal Decomposition Process of HEMs

Table S3 summarizes the temperatures of exothermic peaks (Tp), which are extracted from the nonisothermal DSC curves under heating rates of 5, 10, and 20 K/min. Then, the activation energies (Ea) of exothermic peaks calculated by the Kissinger method are listed in Table with a good fit coefficient (R2 > 0.97). The Ea of PeakIV cannot be efficiently calculated due to the corresponding tanglesome Tp values. The synergistic effect between nanothermites and high explosives can be proven by the decreasing Ea values. It is clear that the decomposition peaks of CL-20 in assembled HEMs all shift left, which could be attributed of the catalyst effect of CuO and Al nanoparticles. The activation energies of raw NC and CL-20 are 258.6 and 291.2 kJ/mol, respectively. The activation energy of the decomposition of CL-20 assembled in HEMs by electrospray is at least 29.7% lower than that of the raw CL-20.

Table 3

Activation Energies Ea (kJ/mol) of Prepared Samples Calculated by the Kissinger Method

sample	EaI	R2	EaII	R2	EaIII	R2
NC	258.6	0.9995
CL-20	291.2	0.9953
Al/CuO					332.6	0.9987
Al/CuO/CL-20	147.5	0.9985	177.7	0.9992	239.3	0.9992
HEMs (ES EAC)	168.1	0.9992	145.3	0.9986	165.5	0.9992
HEMs (ES DMK)	204.6	0.9931	200.2	0.9969	108.9	0.9743
HEMs (ES EA/DEE)	185.7	0.9999	176.3	0.9828	120.2	0.9999
HEMs (ES NPA/DEE)	150.7	0.9983	154.5	0.9790	175.2	0.9991

The significant decline in activation energy implies more powerful energy output. As mentioned above, the reaction of Al/CuO(UM) is divided into three regimes (PeakII, PeakIII, and PeakIV) in the case of HEMs (ES). It seems that PeakIII dominates the pressurization process, while PeakIII contributes the most to the released heat of Al/CuO solid–solid reaction. The EaIII of HEMs (ES DMK), which has the highest dP/dt (0.75 MPa/ms), is the lowest with only 108.9 kJ/mol, while the EaIII of HEMs (ES DPA/DEE), which has the slowest dP/dt (0.14 MPa/ms), is as high as 175.2 kJ/mol.

Effect of Interfacial Microstructures on the Reactivity of HEMs

The results of combustion and TG-DSC tests indicated that interfacial microstructures could affect the reactivity of HEMs with identical ingredients, obtained from different solvent systems. The properties of solvent, such as solubility, surface tension, electrical conductivity, viscosity, and vapor pressure, can affect the morphology of target materials. Acetone with the highest vapor pressure among the selected solvent systems would escape from charged droplets ultrafast, occurring at the solidification process in flight to substrate. Then the NC matrix coating would become fragmentary, and some Al or CuO nanoparticles are bare as shown in Figure a. The NC matrix can enhance the reactivity of nanothermites acting as gas agents that prevent nano-Al particles from quickly sintering. However, as reported in ref (22), exceeding mass of NC (>7.5 wt %) would decrease the combustion efficient, even worse than that of Al/CuO. At the initial combustion of assembled microparticles, the aggregate NC matrix needs more energy to be active accompanying a high Ea of 204.6 kJ/mol. Contributing to incomplete coating, unreacted bare nanoparticles could be excited directly by released heat, resulting in a more violent reaction than other granular HEMs (ES EA/DEE or DPA/DEE). This combustion process can offer an explanation about the highest dP/dt of HEMs (ES DMK). After initiation, the reductive decomposition products of NC and CL-20 would react complicatedly with Al/CuO[12] in the range of 350 to 450 °C. This reaction regime can be attributed to the accelerated reduction of CuO to Cu2O. The following steps are the solid–solid reaction (530 to 650 °C) and solid–liquid reaction (650 to 670 °C) of Al/CuO.

The high viscosity of the precursor containing ethyl acetate could lead to the presence of NC nanofibers at low operating temperature (<20 °C). NC is one powerful energetic material; however, the fiber–fiber interface would increase thermal resistance leading to a broader exothermic peak and higher onset temperature than the powder counterpart.[34] Combustion of the clay-like HEMs (ES, EAC) would weaken due to the existence of NC fibers, which act as barriers hindering the mass transport and heat exchange between oxide and fuel, leading to slower flame propagation and more ejected unreacted HEMs.

The surface morphology of HEMs (ES DPA/DEE) is similar to that of HEMs (ES EA/DEE), on which solid nanoparticles are coated by the NC matrix with a certain number of cavities. It is worth noting that only a small part of CL-20 in HEMs (ES DPA/DEE) undergo crystal transfer or recrystallization, leading to similar decomposition kinetics compared to Al/CuO/CL-20 (UM hexane), with roughly equivalent activate energy. The unchanged nano-CL-20 particles are excited slowly due to being sealed in the NC matrix coating of HEMs (ES DPA/DEE), leading to the lowest pressurization rate and longer combustion duration. However, the completeness of CL-20 decomposition can be promoted by the nanothermite reaction, which is the same as the case of Al/CuO/CL-20 (UM hexane), rather than in the case of HEMs (ES EA/DEE). This result indicates that the ε-CL-20 nanoparticles have superior combustion performance to the recrystallized ones.

Conclusions

We have assembled hybrid energetic materials consisting of nanothermites and high explosives with different morphologies by a one-way method of electrospray from various solvents, including acetone, ethyl acetate, and a mixture of the two (ethanol/diethyl ether and n-propanol/diethyl ether). Nanoparticles of Al and CuO with intimate contact dispersed uniformly in the matrix of CL-20 and NC and assembled into microparticles with narrow size distribution (1.5 to 10 μm). Introducing CL-20 and NC into nanothermites by electrospray led to superior reactivity with gas generation and pressurization rate. The granular HEMs (ES, DMK) exhibited violent combustion performance with ∼1.4 times higher Pmax (0.71 MPa) than that of Al/CuO and the highest dP/dt = 0.75 MPa/ms. Nanofibers of NC, appearing in the clay-like HEMs (ES, ethyl acetate or EAC/DMK), weakened the reactivity with low gas generation and insufficient combustion. Majority of CL-20 dispersed as suspended particles rather than a recrystallized matrix in HEMs (ES, NPA/DEE), leading to the high yield of gaseous products (0.71 MPa) but the lowest pressurization rate of 0.14 MPa/ms. The synergistic effect between the nanothermite and high explosive with arranged microstructures was also validated by TG-DSC results, which show higher weight loss, lower onset temperature, and lower activation energy of the decomposition of CL-20 and the reaction of nanothermites. This study provides a promising way to obtain HEMs with controllable combustion performance by constructing various interfacial microstructures.

Experimental Section

Preparation

Materials

Aluminum nanoparticles (Al, 100 nm) were purchased from Beijing Deke Daojin Science and Technology Co., Ltd. The active content of Al was about 74.2 wt %, determined by TG results conducted in an air purge. Copper oxide nanoparticles (CuO, 50 nm), colloidal solution (4–8 wt % nitrocellulose), and analytical reagents including ethyl acetate (EAC) and n-propanol(NPA) were purchased from Aladdin Reagent Industrial Corporation. Acetone (DMK), ethanol (EA), and diethyl ether (DEE) were supplied by Sinopharm Chemical Reagent Co., Ltd. ε-CL-20 with an average diameter of ∼260 nm was supplied by National Special Superfine Powder Engineering Research Center of China. All chemicals were used as received.

Precursor Preparation

The parameters of precursors, including mass loading, solvents, and content of the binder, have an obvious impact on keeping the stability of cone-jet mode. In this study, the total mass loadings of Al, CuO, NC, and CL-20 were kept at 125 mg/mL. 5 wt % NC was used as the energetic binder to increase the viscosity of the precursor preventing solid particles from quick sedimentation. 5 wt % CL-20 and 90 wt % Al/CuO were the other components. The equivalence ratio (φ) of Al/CuO was fixed at 1.4 considering the existence of the inherent Al2O3 shell. The range of CL-20 solubilities in different organic solvents was quite broad,[35,36] as summarized in Tables S1 and S2, and 6.3 mg/mL (5 wt %) CL-20 existed in different forms, suspension nanoparticles or solute.

Required amounts of NC and CL-20 were first dissolved in organic solvents to form transparent solutions, except the case of n-propanol/diethyl ether, in which the majority of CL-20 dispersed as nanoparticles. Nanoparticles of Al and CuO were added to form suspensions, which were then ultrasonically dispersed for 1 h followed by magnetic stirring for 24 h. The prepared precursors were then transferred into a syringe mounted on a syringe pump with a 0.43 mm metal flat nozzle.

Electrospray Assembly

The assembling process was conducted in a fume cupboard to prevent researchers from inhaling vapors of solvents and high voltage, in which temperatures of 18–21 °C and relative humidities of 10–30% were maintained. For electrospray, operation parameters to maintain a stable cone-jet mode were rather broad. Without regard to the effect of operation parameters on the reactivity of the HEMs, the feeding rate controlled by a syringe pump was set to 1.75 mL/h, 17 kV positive voltage was applied to the metal nozzle, and 3 kV negative voltage was applied to the foil collector. The distance between the nozzle and substrate was set as 15 cm to ensure the complete evaporation of organic solvents. As shown in Figure , during the process of electrohydrodynamic atomization, charged droplets underwent the processes of fission, evaporation, and recrystallization in flight to substrate, and then solid materials deposited on the foil receiver. Deposited solid materials were then scraped carefully from the foil by a plastic flake for further tests. All assembled samples are labeled as ES, i.e., Al/CuO/NC/CL-20 (ES, DMK).

Contrast samples of Al/CuO and Al/CuO/CL-20 were prepared by simple ultrasonic mixing for 60 min in hexane without recrystallization of CL-20. The contrast sample of Al/CuO/NC/CL-20 (acetone) was prepared by the same procedure as the precursors without further electrospray. All contrast samples are labeled as UM, i.e., HEMs (UM).

Material Characterization

Morphological Observation

The morphologies of the as-prepared HEMs were imaged by a scanning electron microscope (SEM, Sigma 500 ZEISS), and the distribution of components was evaluated by energy-dispersive X-ray spectroscopy (EDS, Xflash 6130 Bruker). The characteristic sizes of assembled materials were measured by a Nano Measure 1.2.

Compositional Information

Fourier transform infrared spectroscopy (FTIR, Bruker Vertex 80 V) was applied to verify the polymorph conversion of CL-20 on account of fast evaporation. FTIR spectra in the region of 1200 to 700 cm–1 were recorded, which were scanned at a resolution of 4 cm–1. The polymorph of the raw CL-20 was also confirmed by X-ray diffraction (XRD).

Measurement of Reactivity

Thermal Analysis

The thermal behaviors of samples under low heating rates were studied on a differential scanning calorimetry (DSC) instrument (TGA/DSC 3+, Mettler Toledo) under argon flow at heating rates of 5, 10, and 20 K/min in the range of room temperature to 800 °C. Samples are loaded into alumina crucibles. Activation energy (Ea) of exothermic peaks, which was used to evaluate the kinetics of as-prepared HEMs, can be calculated by the Kissinger method described by eq by varying the heating ratewhere β is the heating rate, Tp is the temperature of the exothermic peak, A is the pre-exponential factor, and R is the gas constant. The Ea of the exothermic peaks can be computed from the slope (−Ea/R) of the straight line, which is a linear fit of plotted ln(β/Tp2) against 1/Tp.

Combustion Cell Tests

The pressurization characteristics of the prepared HEMs were performed by a homemade combustion cell, which was equipped with a piezoelectric pressure sensor (CY-YD-205, Sinocera) coupled with a charge amplifier (YE5854A, Sinocera). In a typical test, 25 mg of samples was loaded in the confined cell with ∼15 mL volume and ignited by heating a nichrome (Ni-Cr) wire with a diameter of 0.2 mm, which was in direct contact with the surface of samples. After combustion, a pressure vs time curve was recorded by an oscilloscope. Maximum pressure and pressurization rate were extracted from the time-resolved pressure curves to evaluate the reactivity of samples. Data processing can be found in ref (22), and the schematic diagram of the pressure cell is shown in Figure S9.

Open Burning Tests

Combustion processes in an open environment were recorded by a high-speed camera at 15000 fps and 15 μs exposure time. For each test, about 3 mg of samples was loaded into a metal bowl-shaped container and ignited by heating a Ni-Cr wire similar to that of combustion cell tests.