Thermal and ignition properties of hexanitrostilbene (HNS)microspheres prepared by droplet microfluidics

Rui-shan Han ,Fei-peng Lu ,Fang Zhang ,Yan-lan Wang ,Mi Zhou ,Guo-sheng Qin ,Jian-hua Chen ,Hai-fu Wang ,En-yi Chu ,**

a State Key Laboratory of Explosion and Technology, Beijing Institute of Technology, Beijing,100081, China

b Science and Technology on Applied Physical Chemistry Laboratory, Shaanxi Applied Physics-Chemistry Research Institute, Xi'an, 710061, Shaanxi, China

Keywords:Microfluidics HNS microspheres Thermal stability Ignition threshold

ABSTRACT HNS-IV (Hexanitrostilbene-IV) is the main charge of the exploding foil initiators (EFI),and the microstructure of the HNS will directly affect its density,flowability,sensitivity,and stability.HNS microspheres were prepared using droplet microfluidics,and the particle size,morphology,specific surface area,thermal performance,and ignition threshold of the HNS microspheres were characterized and tested.The results shown that the prepared HNS microspheres have high sphericity,with an average particle size of 20.52 μm (coefficient of variation less than 0.2),and a specific surface area of 21.62 m2/g(6.87 m2/g higher than the raw material).Without changing the crystal structure and thermal stability of HNS-IV,this method significantly enhances the sensitivity of HNS-IV to short pulses and reduces the ignition threshold of the slapper detonator to below 1000 V.This will contribute to the miniaturization and low cost of EFI.

1.Introduction

Exploding foil initiators(EFI),also called slapper detonators,are explosive devices that initiate when a high velocity polymeric flyer impacts the explosive material[1].Since the slapper detonator has no sensitive primary explosive,it can work safely in harsh environments such as strong radiation and stray current.However,the insensitive design also makes it require a very high voltage for initiation,which greatly limits its development towards miniaturization and low cost [2].

Therefore,reducing the ignition threshold of the EFI has always been the direction pursued in pyrotechnic research.The current methods include: 1.Exploring new methods for integrated manufacturing of EFI [3-8],2.Researching and developing more efficient explosive foil chips[9-20],3.Optimizing the material and structure design of the flyer and accelerating chamber [21-23],4.Improving the refined preparation process of hexanitrostilbene(HNS) [24-27],5.Compound with other explosives [28-30],etc.However,the structural design and optimization of the EFI have approached the maximum.Without using other methods,it is difficult to reduce the ignition threshold of the EFI to below 1000 V simply by adjusting the structural design.

In terms of explosives research,HNS is an insensitive highexplosive with good heat resistance.Especially HNS-IV,it has the characteristics of high purity,sensitivity to short pulses,low mechanical and electrostatic sensitivity,and good vacuum stability,which makes it the main charge for EFI [31-33].How to further improve the short pulse duration shock waves sensitivity of HNS-IV and reduce the ignition threshold without reducing the thermal safety of the main charge is a problem that needs to be solved urgently.

Different microstructures in the same material usually lead to different macroscopic properties.For explosives,the microstructure directly affects the density,flowability,sensitivity,and stability[34-39].Common methods for adjusting the microstructure of HNS include recrystallization [40-43],spray drying [25,44],desolvation [45,46],mechanical grinding [26,47],etc.The essence of these methods was to improve the sensitivity by changing the microstructure,increasing the specific surface area and porosity of HNS.Therefore,in this paper,the droplet microfluidic technology was used to aggregate the raw HNS-IV to prepare highly spherical HNS microspheres to obtain high specific surface area and porosity.The influence of structure on the performance of HNS was studied.It was found that the HNS microspheres not only conducive to improving the flowability and bulk density of HNS-IV,thereby improving the charging and pressing properties of HNS-IV,but also enhancing HNS-IV's short pulse duration shock waves sensitivity and reducing the ignition threshold of the EFI without changing the thermal stability of HNS-IV.

2.Experiment

2.1.Materials

Ethyl acetate (EC) and sodium dodecyl sulfate (SDS) were commercially purchased (A.R.) and used without further purification.Nitrocellulose (NC) (viscosity: 1/4 s) was purchased from Sichuan Nitrocell Co.Ltd.HNS(average particle diameter:200 nm)was produced by North University of China.The flow-focusing microfluidic chips and the droplet collection device were produced by Suzhou Wenhao Microfluidic Technology Co.Ltd.

2.2.Preparation

The first step was to prepare the dispersed phase and continuous phase solutions.The dispersed phase was the NC/EA solution in which 20 wt% HNS-IV particles were suspended,and the concentration of the NC solution was 1%.The continuous phase was an aqueous solution with 2 wt% SDS dissolved.The flow-focusing microfluidic chip channel had a depth of 100 μm,the width at the intersection point was 250 μm,the width at the exit point was 400 μm.The syringe pumps were used to inject the dispersed phase and the continuous phase solutions into the channels.The dispersed phase flow rate(Qd)and continuous phase flow rate(Qc)were set as 50 and 2000 μL/min,respectively.When the two-phase solutions were subjected to the action of a flow-focusing microfluidic chip,the oil-in-water(O/W)microdroplets with suspended HNS particles were generated.These droplets undergo continuous phase extraction and became semi-solidified HNS microspheres in the microchannels.After entering the droplet collection device,it was further solidified to form HNS microspheres.The entire HNS microsphere preparation system was shown in Fig.1.

Fig.1.The HNS microsphere preparation device.

2.3.Characterization

The morphology of the powder was investigated using a scanning electron microscope (SEM,mod.Vega TS5136XM,Tescan,Czech).And the size and distribution of HNS microspheres were measured and analyzed using the particle size analysis software Nano Measurer.

The X-ray diffraction (XRD) characterization was performed using D8 advance X-ray Diffraction (Bruker,Germany) measurement,at a resolution of 0.02°.The scanning range was 5°-70°,the scanning rate was 10°/min.

The HNS microspheres and raw HNS were dispersed in pure water and ethanol respectively by ultrasonic treatment,and the particle size distribution of HNS microspheres and raw HNS was performed using a Malvern MasterSize 5004 laser particle size analyzer.

Thermogravimetric (TG) and differential thermogravimetric(DTG) tests were carried out by a Thermal Gravimetric Analyzer TG209F1 from Netzsch (Germany).Powder samples with a weight of about 0.2 mg were crimped into an aluminum pan.The analyses were conducted in nitrogen(flow rate 50 mL/min),by heating the sample from 25 to 550°C with heating rate 5°C/min,10°C/min,15°C/min and 20°C/min,respectively.

The specific surface area (SSA) of HNS microspheres was measured using Quanta NOVA 2200e gas adsorption analyzer(USA).

The cook-off test was carried out using the cook-off test system developed by our laboratory,which was composed of a temperature control module,a heating module,and a data acquisition module,as shown in Fig.2(a).The test conditions refer to the U.S.military standard MIL-DTL-23659F Appendix A.In this study,the HNS microspheres pellets were first heated to 250°C at a heating rate of 1°C/min and kept for 10 min.Subsequently,the temperature was increased to 350°C at a heating rate of 3.3°C/h and kept for 30 min.

The initiation threshold of HNS microspheres was determined by the device of electrically exploding foil-driven flyer plate developed in our laboratory,and the schematic of the device was shown in Fig.2(b).The initiation threshold,expressed as initiation voltage,was determined by a sensitivity test using the Langlie method.The testing conditions were as follows: capacitance,0.22 μF;polyimide flyer plate,25 μm;copper foil,0.3 mm（L）×0.3 mm（W）×4.6 μm（H）;HNS charge,50±1 mg,Φ 3.4 mm×3.5 mm,and density 1.57 g/cm3.

3.Results and discussion

3.1.Morphology, particle size, crystal and specific surface area characterization

The SEM images of the raw HNS and its particle size distribution were shown in Fig.3(a)and Fig.3(d),respectively.The particle size was about 200 nm.The morphology of the HNS microspheres prepared by droplet microfluidic was shown in Fig.3(b).The spherical degree of HNS microspheres was very high.In order to quantitatively characterize the sphericity of HNS microspheres,the relative standard deviation of the particle diameters in different directions was calculated[48,49].Twenty HNS microspheres were randomly selected on the scanning electron microscope photograph,and their particle sizes in 4 different directions were measured using Nano Measurer software,as shown in Fig.3(c).After calculation,the average relative standard deviation of those HNS microspheres was 2.02%,which indicated that the HNS microspheres have a high degree of sphericity.The particle size distribution of the HNS microspheres was measured using a laser particle size analyzer,as shown in Fig.3(d).Furthermore,the particle size of all 163 HNS microspheres on the scanning electron microscope photograph was measured,and the statistical results are shown in Table 1.The average particle size of the HNS microspheres was 20.48 μm,and the coefficient of variation (CV) was 19.33%.

Table 1 The particle size statistics of HNS microspheres.

Fig.3.(a)Scanning electron micrograph of raw HNS;(b)Scanning electron micrograph of HNS microspheres;(c)Schematic diagram of measuring sphericity of HNS microspheres;(d) The particle size distribution of HNS microspheres and raw HNS;(e) XRD spectrum of HNS microspheres and raw HNS;(f) The adsorption and desorption curves of HNS microspheres and raw HNS.

XRD was used to characterize the crystal form of HNS microspheres and raw HNS.It can be seen from Fig.3(e) that the peak positions and shapes of HNS microspheres and raw HNS were basically the same,which were the same as the standard XRD pattern of HNS.This indicated that the spheroidization process did not change the crystal structure of HNS.

The adsorption and desorption curves of HNS microspheres and raw HNS were shown in Fig.3(f).It can be seen from the adsorption and desorption curves that in the low pressure section(0-0.2)and the medium pressure section (0.2-0.8),the adsorption and desorption curves are slowly rising and falling.This indicated that the HNS has less intermolecular force with N2and adsorbed slowly.In the high pressure section (0.8-1),the adsorption capacity increased rapidly,which was caused by the slit pores formed by particle aggregation,as shown in Fig.3(b).The adsorption capacity of HNS microspheres was higher than that of raw HNS in the whole pressure range,indicating that HNS microspheres contain more pores in both the nanopores and the slit pores.The specific surface areas of HNS microspheres and raw HNS were measured by the multipoint BET method in the pressure range of 0.05-0.3,they were 26.44 m2/g and 13.09 m2/g,respectively.It also meant that the HNS microspheres have more micropores after spheroidization.

3.2.Thermal performance and activation energy analysis

Probing the thermodynamic and kinetic parameters of explosives is an important way to master its thermal decomposition properties.The TG and DTG curves of the raw HNS and HNS microspheres at a heating rates of 5°C/min,10°C/min,15°C/min,and 20°C/min were obtained by thermal gravimetric analyzer,as shown in Fig.4(a)and Fig.4(b).It can be seen from the TG and DTG curves that both the raw HNS and HNS microspheres start to decompose at about 273°C.With the increase of heating rate,the thermal decomposition temperature of raw HNS and HNS microspheres increased gradually.Moreover,the TG curves were all smooth inverted S-shaped,and the DTG curves were all unimodal.This indicated that the thermal decomposition of raw HNS and HNS microspheres in nitrogen atmosphere was a single process andfollowed a single decomposition mechanism.

Fig.4.(a)The TG and DTG curves of raw HNS;(b)The TG and DTG curves of HNS microspheres;(c)The fitting curves of raw HNS at different decomposition ratios;(d)The fitting curves of HNS microspheres at different decomposition ratios.

The Friedman method was used to calculate the apparent activation energy(Ea)of HNS microspheres and raw HNS at 0.1,0.2,0.3,and 0.4 decomposition ratios,and the Friedman formula was shown in Eq.(1).

where β is the heating rate;α is the decomposition ratio;Tis the temperature at different decomposition ratios;Ais the preexponential factor;nis the reaction order;Ris the gas constant 8.314 J·mol-1·K-1.

According to Eq.(1),took 1/Tas the abscissa and ln(β·dα/dT)as the ordinate to make a scatter plot.The fitting curves of raw HNS and HNS microspheres at different decomposition ratios were obtained by linear fitting,as shown in Fig.4(c) and Fig.4(d),and the slope of the curve was-E/R.TheEadata were listed in Table 2.It can be seen that theEagradually decreases with the proceeds of the reaction.And theEaof HNS microspheres was lower than that of raw HNS,with an average reduction of about 20.87%.It indicated that HNS microspheres have higher reactivity compared to raw HNS.This is because nitrocellulose is also an energetic material with an exothermic peak temperature of about 210°C.A small amount of addition will increase the reactivity of HNS.

Table 2 The Ea results of raw HNS and HNS microspheres under different decomposition ratios.

3.3.Cook-off test

With reference to the charge structure of the slapper detonator,a cook-off test was carried out on the HNS microsphere pellets.The pellets before and after the test were shown in Fig.5(a) and Fig.5(b).The HNS charge shell was intact as before after the cookoff test.The HNS pellets turned into a black residue,but they still maintain a regular cylindrical shape.Their volume was reduced from the original Φ 3.4 mm×3.5 mm to Φ 2.35 mm×2.25 mm,and the mass was reduced to 6.4 mg.The black residue should be the carbon compound remaining after the slow thermal decomposition of HNS.

Fig.5.(a) Photograph of HNS microsphere pellets before cook-off test;(b) Photograph of HNS microsphere pellets after cook-off test;(c) Temperature-time curve of the cook-off test;(d) X-ray energy spectra of HNS microsphere pellets before and after the cook-off test.

Fig.5(c)shown the temperature-time curve of the cook-off test when the heating rate was 3.3°C/h.It can be seen from the figure that the measured temperature can better match the preset temperature.This was because the lower heating rate made the temperature gradient between the furnace and the sample smaller,and the test system was close to the thermal equilibrium state.During the whole test,no strong exothermic peak was detected,whichindicated that HNS only had a slow thermal decomposition,and no detonation or deflagration occurred.During the thermal decomposition of HNS,the released energy could be slowly diffused into the surrounding environment,and there was not enough heat accumulation inside the pellet.

The samples before and after the cook-off test were analyzed by energy dispersive X-ray spectroscopy,and the spectrum was shown in Fig.5(d).The sample before the test had a higher content of N and O elements.However,after the cook-off test,the content of C element in the sample increased significantly,and the content of N and O elements decreased.This also shown that HNS undergone a thermal decomposition reaction under the cook-off test.

3.4.Initiation threshold test

To study the initiation threshold of the HNS microspheres,we used a self-made electrically exploding foil-driven flyer plate device to conduct the ignition test of the HNS pellets with a size of Φ3.4 mm × 3.5 mm.The Langlie method was used to predict the test step length and process the test data.The ignition results of the pellets were shown in Fig.6(a) and Fig.6(b),and the ignition voltage and threshold were shown in Table 3.Under the same test conditions,the 50% initiation voltages of HNS microspheres and raw HNS were 913.7 V and 1003.6 V,respectively,which was about 90 V lower than raw HNS.This indicated that after raw HNS were prepared into microspheres by using flow-focusing microfluidic method,the sensitivity of short pulse duration shock waves was significantly improved.This was beneficial to reduce the initiation voltage of the EFI.

Table 3 Initiation threshold test results of HNS microspheres and Raw HNS by Langlie method.

Fig.6.(a) The pellets that were successfully ignition;(b) The pellets that were not successfully ignition;(c) The surface of the raw HNS pellets;(d) The surface of the HNS microsphere pellets.

In order to explore the regulation mechanism of the HNS microspheres that improve the sensitivity of short pulse duration shock waves,the surface of the pellets was characterized by SEM.The microscopic images were shown in Fig.6(c)and Fig.6(d).When the raw HNS or HNS microspheres were pressed into pellets,their surfaces were relatively flat.However,the pellets pressed by HNS microspheres could be clearly seen the squeezed boundary between the microsphere particles,and still maintain the morphology of HNS microspheres.Under the impact of high-speed slappers,HNS microspheres were subjected to short pulse duration shock waves.The air in the voids of the HNS microspheres were adiabatically compressed,so that their temperature rose rapidly,forming “hot spots”.When the temperature of the “hot spot”exceeded the critical of initiation,an explosion was triggered.It was because of these increased nanopores within HNS microspheres that enhanced the slapper sensitivity.In addition,the addition of a small amount of NC could also increase the reactivity of HNS,reduced the temperature of the critical point of initiation,and made the pellets of HNS microspheres easier to detonate.

4.Conclusions

(1) Highly spherical HNS microspheres can be prepared quickly and efficiently with the droplet microfluidic method.

(2) Without changing the crystal structure and thermal stability of HNS,the short pulse duration shock wave initiation threshold of the HNS was reduced to below 1000 V,and the regulation mechanism was explored.This is a great significance for expanding the application range and reducing the volume and price of the EFI.

(3) Moreover,this method also broadens people's research ideas on the HNS short pulse duration shock waves sensitivity control mechanism,and provides a new direction for the design of EFI.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The project was financially supported by a foundation item from the China People's Liberation Army General Armaments Department.