Preliminary discussion on the ignition mechanism of exploding foil initiators igniting boron potassium nitrate

Haotian Jian , Guoqiang Zheng , Lejian Chen , Zheng Ning , Guofu Yin ,Peng Zhu ,*, Ruiqi Shen

a School of Chemistry and Chemical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China

b Micro-Nano Energetic Devices Key Laboratory of MIIT, Nanjing University of Science and Technology, Nanjing 210094, China

c Anhui Province Key Laboratory of Microsystem, the 43rd Research Institute of CETC, Hefei 230088, China

d Science and Technology on Applied Physical and Chemistry Laboratory, Shanxi Applied Physics and Chemistry Research Institute Xi'an 710061, China

Keywords:Exploding foil initiator PDV Plasma spectrum Ignition mechanism Boron potassium nitrate

ABSTRACT Exploding foil initiator (EFI) is a kind of advanced device for initiating explosives, but its function is unstable when it comes to directly igniting pyrotechnics.To solve the problem, this research aims to reveal the ignition mechanism of EFIs directly igniting pyrotechnics.An oscilloscope, a photon Doppler velocimetry, and a plasma spectrum measurement system were employed to obtain information of electric characteristics, impact pressure, and plasma temperature.The results of the electric characteristics and the impact pressure were inconsistent with ignition results.The only thing that the ignition success tests had in common was that their plasma all had a relatively long period of high-temperature duration (HTD).It eventually concludes that the ignition mechanism in this research is the microconvection heat transfer rather than the shock initiation, which differs from that of exploding foil initiators detonating explosives.Furthermore, the methods for evaluating the ignition success of semiconductor bridge initiators are not entirely applicable to the tests mentioned in this paper.The HTD is the critical parameter for judging the ignition success, and it is influenced by two factors: the late time discharge and the energy of the electric explosion.The longer time of the late time discharge and the more energy of the electric explosion,the easier it is to expand the HTD,which improves the probability of the ignition success.

1.Introduction

Electric energy is utilized to initiate energetic materials by electric heat or electric explosion whose representatives are exploding bridgewire (EBW) detonators [1,2], semiconductor bridge(SCB) igniters [3,4], and exploding foil initiators (EFIs) [5-7].The EBW and the SCB are widely used with their characteristics and mechanism have been deeply studied[8-12].The EFI is commonly adopted to detonate the explosive Hexanitrostilbene (HNS), and research on its mechanism is almost based on this function[13,14].With Boron Potassium Nitrate(BPN)as the output charge,explosive foil deflagration igniters (EFDIs) employ the EFI to ignite propellants in the field of aerospace [15].

According to differences in the structures and function mechanisms, the ignition modes of EFDIs are classified into two types[16]: Slapper-HNS-BPN and Slapper-BPN.The ignition mode of the Slapper-HNS-BPN type is that the explosive HNS detonated by the EFI ignites the pyrotechnic pellets BPN, which of the Slapper-BPN type is that the BPN is directly ignited by the EFI.It is obvious that the key to success to ignite the Slapper-HNS-BPN type is that the EFI detonates the HNS,and the explosive is initiated using a short-pulse shock wave produced by the high-velocity impact with the thin flyer plates launched by the EFI [13,14].

As for the Slapper-BPN type, its ignition mechanism is not as clear as that of the Slapper-HNS-BPN type.Few researchers[17,18]explained the ignition mechanism of the Slapper-BPN type through the detailed data obtained from the experiments.In practice, it is difficult for the EFI to stably ignite the BPN, and the standard deviation of its ignition energy is too large to be ignored.Due to the lack of sufficient research and detailed analysis of the ignition mechanism of the Slapper-BPN type, it is confusing how to adjust the ignition condition of the ignitor to ensure the stability of the ignition function.

In this paper, a series of experiments were designed to explore the ignition mechanism of the Slapper-BPN type.While the EFI ignites the BPN pellets, the plasma produced by the exploding foil which is excited by a large current pulse drives the flyer through the barrel to impact the pyrotechnic.The high-speed flyer and the high-temperature plasma both contribute to the interface reaction of the pyrotechnic ignition.Thus, obtaining the parameters of the flyer and plasma is beneficial to analyze the key factors of stable ignition.The characteristics of the current and voltage during the function time were respectively recorded by a Rogowski coil and a pair of high-voltage probes to display the process that the electric energy was deposited into the exploding foil.Furthermore, a photon Doppler velocimetry (PDV) as a novel measuring instrument was used to determine the velocity of the flyer to calculate the pressure that the flyer impacted the BPN.In order to evaluate the effects of thermal conduction on the ignition of the BPN, a plasma spectrum measurement system was utilized to observe the plasma emission characteristics generated by the electric explosion of the exploding foil, while the real-time temperature of the plasma was able to be calculated based on the Boltzmann multispectral slope method.The characteristics of the flyer and plasma during the process of the electric explosion were discussed to analyze the evolution law of the EFI function.Based on the analysis of the experiment data, several common standards and methods for the ignition judgment were considered not fully applicable, and the important factors of the success of the EFI to ignite the BPN pellets were pointed out.

2.Experimental devices and methods

2.1.Design of the exploding foil initiator

Fig.1 shows the essential structure of EFIs.The function process of the EFI is that the large current pulse passes through the metal layer and stimulates the narrow bridge area to generate the electric explosion which drives the flyer to a high speed through the barrel.In this research,the materials of the metal layer and the flyer were respectively copper (Cu) and polyimide (PI).It is obvious that the structure of the EFI has a significant influence on its ignition performance [19].The size of the Cu bridge is related to the energy consumed by the electric explosion.The height of the barrel is the distance of the flyer's acceleration, playing a decisive role in the muzzle velocity of the flyer.In this research,two types of EFIs with different structures were used in order to adjust the electric energy consumption and the flyer speed.The differences listed in Table 1 were the size of the Cu bridge, the internal diameter, and the height of the barrel.

Fig.1.Essential structure of the EFI.

Table 1 Structure sizes of the EFI.

2.2.Pyrotechnic composition

When it comes to pyrotechnic ignition,the composition cannot be ignored.For reducing the electrostatic sensitivity of the BPN pellets[20],the mixing ratio of boron/potassium nitrate was 50/50,and the phenolic resin with a 5%mass fraction was selected as the binder.The boron with a 0.9 μm average particle diameter was provided by Shanghai Xiaohuang Nano Technology Co.,Ltd,and the potassium nitrate with a 1.6 μm average particle diameter was homemade by the air-jet mill method.

2.3.Design of the experiment system

For measuring various parameters in the function process of the EFI,a corresponding experiment system was designed,as shown in Fig.2.This system consisted of a capacitor discharge unit(CDU),an oscilloscope with a Rogowski coil and a pair of voltage probes, a PDV,a plasma spectrum measurement system,and a synchronizer.The CDU comprised of a capacitor,a cold-cathode trigger tube,and an EFI, was the device to realize the electric explosion of the EFI.The capacitor reserved the electric energy provided by the highvoltage power supply (not shown in Fig.1) and provided the EFI with a large pulse current after the cold-cathode trigger tube was triggered.The resistance Rpand the inductance Lpwere respectively the sum of the resistance and inductance of the CDU.By changing the values of the capacitor,the charge voltage,the resistance Rpand the inductance Lp, the characteristics of the current and voltage were able to be adjusted to carry out the electric explosion of different degrees.For exploring the ignition mechanism of the Slapper-BPN type EFDI, six representative experiments were selected from dozens of them, and their conditions are listed in Table 2.

The oscilloscope, the PDV, and the plasma spectrum measurement system were utilized to collect data in this research.The oscilloscope (TELEDYNE LECROY wavesurfer 510) showed the current and voltage curves respectively recorded by the Rogowski coil(CWTUM, sensitivity: 1.0 mV/A) and a pair of the high-voltage probes (Tektronix P6015A, 1000X).The electric characteristics were of benefit to analyzing the information of the function process and energy consumption of the EFI.The PDV and the plasma spectrum measurement system will be introduced in detail later.The synchronizer set the corresponding trigger time for each measuring device to ensure that the data collected by different devices had the same time axis.Thus,the various parameters were able to be discussed comprehensively.

2.4.Photon Doppler velocimetry

Fig.2.Schematic representation of the design of the experiment system.

Table 2 Conditions and results of the tests.

Fig.3.Basic geometry of PDV based on self-mixing effect [22].

In this experiment, a homemade PDV based on the self-mixing effect was employed for measuring the flyer velocity [21].Fig.3[22] shows its schematic plot.A fixed frequency laser (λ0=1550 nm) was produced, which was divided into two lasers by a beam splitter.One laser was a reference laser and the other was a test laser, both at the same frequency as the original laser.After passing through an optical circulator and a laser probe, the test laser irradiated a moving object.The moving object reflected a signal laser which was received by the laser probe and entered the optical circulator.Because of the Doppler effect, the frequency of the signal laser had been changed.The optical circulator recognized the differences in the frequency between the signal laser and the test laser, and thus output the signal laser into a signal processing system.In the signal processing system,the reference laser and the signal laser generated a signal due to self-mixing interference.The signal after being amplified was recorded by a high-bandwidth oscilloscope.

2.5.Plasma spectrum measurement system

To achieve the information of the plasma generated by the electric explosion, a plasma spectrum measurement system(DH270-18F-03,Andor ICCD)was used to record the distribution of plasma spectrum intensity with the wavelength.As shown in Fig.2,an optical probe received the emitted light from the plasma which was then decomposed by the spectrometer into a series of discrete lights with different intensities.An intensified charge-coupled device(ICCD)that converted the optical signal to the electrical signal processed by computer software matched with the equipment.Finally,the image of the spectrum was shown on the computer.For observing the evolution of the plasma spectrum, it set a series of corresponding delay times for each test to obtain multiple spectrum images at different times during the electric explosion.With the spectrum images,the information of the plasma was able to be analyzed.

3.Results and discussion

3.1.Electric characteristics

The information of the current and voltage was respectively recorded by the Rogowski coil and the pair of high-voltage probes.The curves of the electric characteristics were able to explain the behavior of the electric energy during the electric explosion.The data of the current and voltage taken in Test 1 was selected as the representative to describe the electric explosion process and the data analysis method, while the data of the other tests would be listed in Table 3 and shown in subsection 3.4 with the flyer velocity and the plasma temperature.

Fig.4 shows the curves of the current and voltage of Test 1 which are typical electric explosion curves of EFIs.At the moment denoted as t0that the capacitor discharged the stored electric energy,the current grew up and heated the Cu bridge of the exploding foil to make it undergo phase transformation,such as melting and evaporation.The phase transformation reduced the conductivity of the Cu bridge and increased its resistance.During this process,the voltage on the exploding foil changed due to the current pulse and its dynamic resistance and finally rose to the peak, which was the moment marked as t1that the electric explosion occurred.The duration of the phase transformation was recorded as ΔtPT.After the electric explosion,the metallic vapor was heated and ionized by the increasing current.With the consumption of the electric energy,the current in the circuit and the voltage on the exploding foil gradually decreased,and the capacitor discharge was completed at t2.The process between the electric explosion and the end of the capacitor discharge was referred to as the late time discharge(LTD)phase,the duration of which was recorded as ΔtLTD.Overall,it was evident that the whole process of the electric explosion was able to be divided into two stages:the phase transformation stage and the LTD stage.The former stage mainly completed the phase transformation and the electric explosion of the Cu bridge, while the latter stage realized the formation and the heating of the plasma and the release of the remaining energy.

The concept of LTD was put forward in the research of SCB igniters[23].For the SCB igniter,it was considered that the ignition of energetic materials was related to the LTD stage.The SCB would ignite normally only when the current pulse was long enough to produce the LTD.When the current pulse was strong enough during the LTD stage, the palsma would be further heated to increase the plasma temperature and extend its duration,which was benefit of reliable SCB ignition.With the same plasma temperature, the longer the LTD time, the more conducive it is to the ignition of fireworks [24].Since the function process of the EFI was similar to that of the SCB, the information of the LTD was taken into consideration.Besides that, electric energy consumption was also regarded as a significant factor in pyrotechnic ignition.Thus,the energy of the whole process E0,the phase transformation stage E1and the LTD stage ELTDwere calculated by integrating the product of the current and voltage and listed in Table 3, with the durations.In addition,Δt0was the duration of the whole process of the electric explosion,ECwas the total energy deposited into the capacitor and ELwas the electric energy lost on the parasitic resistance of the circuit.

Fig.4.The current and voltage curves of Test 1.

Fig.5.The proportion of energy consumption in each stage to the total electric energy deposited in the capacitor.

Table 3 Energy consumption and durations of different stages of electric explosion.

Fig.5 is the proportion of energy consumption in each stage to the total electric energy deposited in the capacitor, based on the data in Table 3.The electric energy was mainly consumed on the parasitic resistance and the discharge in the LTD stage,only a little of which was used to realize the phase transformation of the Cu bridge.Compared with the test conditions of Tests 4,5,and 6,those of Tests 1, 2, and 3 were basically similar and only differed in the resistance and inductance of the circuit.It is worth paying attention to the energy distribution and the ignition results of Tests 1,2,and 3.With the LTD regarded as an important factor of the SCB initiation,it was obvious that,in these three tests,the LTD of Test 3 was the longest and the energy consumption of the LTD of Test 1 was the largest.However, neither Tests 1 nor 3 were able to ignite the pyrotechnic pellets while Test 2 succeeded.The only explanation for this confusing result was that even though the LTD stage was considered meaningful for the ignition, its duration or its energy consumption couldn't judge the ignition success independently.While the plasma produced by the electric explosion was heated during the LTD stage, it did not mean that the duration of the LTD was capable of replacing the heating time of the plasma to the pyrotechnic pellets.As for the LTD energy, it was also reasonable that it was unable to be directly used to evaluate the ignition success due to the differences between the EFI and the SCB in the function modes and the structures.While the SCB was closely connected with the pyrotechnic pellets,the Cu bridge of the EFI was separated from the BPN by the barrel.The BPN had no way to absorb all the LTD energy because of the heat conduction of the plasma to the substrate and the barrel.The phenomena and results of Tests 1,2,and 3 indicated that there were limitations when using the LTD to explain the ignition mechanism of EFIs.Although both the SCB and the EFI have the electric explosion, their ignition mechanisms are not completely the same.The ignition mechanism of EFIs has more complicated situations.

3.2.Flyer velocity and impact pressure

The flyer velocity was measured by the PDV and processed with a special data processing software program matched with the PDV[25,26].Fig.6 is the image of the flyer velocity of Test 2 which is selected as the representative to describe the data processing because of its clear image.There are two red lines that the horizontal one is the baseline produced by the reference laser and the other is the flyer real-time speed curve produced by the signal laser.The velocity of the flyer was obtained.The value of the baseline was subtracted from the flyer real-time speed curve to get the flyer velocity,as shown in Fig.7.At the moment of the electric explosion,the flyer got the maximum acceleration and its velocity continued to be improved in the barrel.In Test 2, the PDV measured the maximum velocity umaxwas 4322.8 m/s.However, while the EFI was used to ignite the pyrotechnic, the flyer generally could not achieve the maximum velocity since the limitation of the height of the barrel.The flight distance was able to be calculated by integrating the flyer velocity with time,and the time when the flyer left the barrel was determined based on the height of the barrel.Thus,the muzzle velocity of the flyer was able to be obtained with the interpolation method.As shown in Fig.7, the muzzle velocity uoutwas 3688.9 m/s in Test 2.

Fig.6.The image of the flyer velocity processed with the software.

Fig.7.The flyer velocity and the flight distance.

While the flyer velocity was commonly applied to evaluate the initiation capacity of EFIs, it was the pressure generated by the impact of the flyer that directly affected the energetic material.Therefore it was necessary to calculate the flyer impact pressure.Fig.8 shows the situations of the shock parameters at different times.At the moment before the flyer impact the BPN,as shown in Fig.8(a),the flyer was considered to impact the BPN at the muzzle velocity without the internal stress pf,0.The BPN not affected by the impact remained stationary (ue,0= 0) and had no internal stress(pe,0= 0).After the flyer impacted the BPN, their interaction was demonstrated in Fig.8(b).The subscripts f and e mean the parameters respectively belong to the flyer and the energetic material.It was assumed that the densities of the flyer(ρf,0)and the BPN(ρe,0)were not changed.Thus,for the flyer and the BPN,according to the momentum theorem, there were

where Afwas the impact area of the flyer and τ was the actuation duration of the shock wave.uswas the shock velocity and upwas the particle velocity.The relationship between usand upcould be described by the linear equation.

where c and λ both were the Hugoniot parameters of the material.At the interaction interface and the area of the shock fronts, there were:

Fig.8.The situations of the shock parameters at different times: (a) The moment before the flyer impacts the BPN; (b) The moment after the flyer impacts the BPN.

With the material parameters listed in Table 4,the impact pressure p was able to be calculated by using Eqs.(2)-(5) and listed in Table 5.

When the EFI was used to detonate the explosive, the impact pressure p and the actuation duration of the shock wave were commonly utilized together to judge the detonation success[13,14].However, generally speaking, it was more likely to detonate the explosive as long as the impact pressure was higher.In this research, when it came to employing the EFI to ignite the pyrotechnic, the situation seemed different from that when the EFI detonated the explosive.It was seen that Test 1 had the maximum impact pressure and Test 6 had the minimum impact pressure.But Test 1 failed to ignite the pyrotechnic while Test 6 succeeded.It was obviously different from the conventional situation.However, it was not immediately arbitrary to think that smaller impact pres-sure was more likely to ignite,which was able to be compared from the test results of Tests 2 and 3.It was worth noting that although the test conditions of Tests 4 and 5 had a lot of differences, their impact pressures were very approximate.With similar impact pressures,Test 5 succeeded to ignite the pyrotechnic and Test 4 did not.Based on the analysis of the data collected by the PDV and the ignition tests,it was apparent that the impact pressure was unable to be considered as a criterion for judging the ignition success.Furthermore,the theory of shock initiation is not fully applicable to the ignition mechanism of the Slapper-BPN type.

Table 4 The material parameters used for the impact pressure calculation.

Table 5 The muzzle velocities and the impact pressures of the tests.

3.3.Spectrum and temperature of plasma

Junying Wu et al.[29] studied the radiation properties of the plasma generated by the explosion of a composite Cu bridge foil and then proved that the plasma reached local thermodynamic equilibrium (LTE).In LTE, the electron and ion temperature are equal, and this is called the plasma thermodynamic equilibrium temperature.Thus,in this research,it was unnecessary to verify the assumption that the plasma generated by the electric energy reached the LTE, based on which the plasma temperature was concerned.The plasma temperatures at different times were able to be calculated using the Boltzmann multispectral slope method from the observed spectrum.

The plasma spectrum recorded by the plasma spectrum measurement system in Test 1 was shown in Fig.9.Combined with the electric characteristics of the electric explosion in Test 1 displayed in subsection 3.1, it could be seen the relationship between the plasma spectrum and the electric energy discharged.At 0.34 μs delay time, the measurement system observed the light of a little intensity without any clear characteristic peak.This was because,at that moment, the Cu bridge was heated during the phase transformation stage and the electric explosion hadn't occurred.So only the light was generated and the plasma was not.The emission spectrums at the delay times of 0.54 μs, 0.74 μs, and 0.94 μs were similar that they all had strong radiation intensity and some characteristic peaks.However, it was obvious that a lot of Cu atomic spectral lines were distorted and blurred due to the existence of strong radiation intensity in the observation wavelength range.According to the electric characteristics during the electric energy discharge, these three spectral curves happened at the LTD stage when the CDU provided a large current to heat and ionize the plasma.Therefore, the strong radiation intensity in the wide wavelength range was caused by the bright light generated by the large current.Toward the end of the LTD at the delay time of 1.14 μs,the characteristic spectral lines of Cu atoms were distinct and the emission spectrum of the Cu plasma was more similar to that of Ref.[29].The spectral intensity decreased gradually and almost disappeared at the delay time of 8.84 μs when the Cu plasma was cooled and became the Cu vapor.

In LTE condition, the electron temperature can be obtained by contrasting the relative intensities of two spectral lines [29].

Fig.10.The change of the plasma temperature in the tests.

where λmnis the transition wavelength of the spectral line, Imnis the integral intensity,Amnis the spontaneous emission probability of the transition, gmis the statistical weight of the upper energy level,Emis the value of the upper energy level,k is the Boltzmann's constant(0.6950417 cm/K),Teis the electron temperature,and C is the constant.Thus, the plasma temperature was able to be calculated from the slope of the straight line.Considering the balance between reducing the fitting standard deviation and the value credibility,three lines[Cu(I)]at 465.11,510.55,and 515.32 nm were adopted to calculate the plasma temperatures according to Eq.(6).The relevant parameters required for the calculation of these three lines are listed in Table 6.

The plasma temperatures at different delay times in the tests are shown in Fig.10.For the convenience of comparison,the time of theelectric explosion in each test is set at the origin of time.The blue curves belong to the tests which failed to ignite the pyrotechnics,while the red curves are those that succeeded.In contrast to the electric characteristics and the impact pressures, the plasma temperatures of the ignition failure group (Tests 1, 3, and 4) and the ignition success group(Tests 2,5,and 6)showed a stable difference.The time when the plasma temperature was higher than 90%of the maximum temperature was Dt90, and Dt80was also defined in this way.While the maximum temperatures in the tests were approximately similar, both < unknownfont fontname = "Symbol Tiger" fill = "FFCC99">Dt90and Dt80of the ignition success group were longer than these of the ignition failure group.For the convenience of discussion,the high-temperature durations including < unknownfont fontname = "Symbol Tiger"fill = "FFCC99">Dt90and Dt80are uniformly referred to as the hightemperature duration (HTD).The situation that the result of the HTD was consistent with that of the ignition had never happened on other parameters.Thus, the HTD deserved to be considered as one of the important parameters to judge the ignition success.The test phenomena of the spectrums and the plasma temperatures revealed that the heating time was a significant factor for the pyrotechnic ignition, and furthermore, indicated that the ignition mechanism of the Slapper-BPN type was more likely to be the micro-convection heat transfer rather than the shock initiation.

Table 6 Spectrum data of Cu atoms[29].

3.4.Behavior of EFI function

Since the behaviors of the electric energy discharge, the flyer acceleration and the plasma heating influenced each other in the process of the electric explosion, the evolution of various parameters was interrelated rather than independent.For analyzing the factors affecting the HTD to stably ignite the pyrotechnic, the data of the electric characteristics, the fly velocity, and the plasma temperature is shown together in Fig.11.

The appearance of the voltage peak was due to the electric explosion caused by the evaporation in the Cu bridge and driving the flyer to obtain a great acceleration.The flyer soon flew out of the barrel with a high velocity.The behavior of the fly was able to explain the phenomenon that the plasma temperature didn't continue to rise after the electric explosion, but remained unchanged or even decreased.This phenomenon had also been mentioned in Ref.[29]without a detailly explanation By analyzing the process of the EFI function,it was obvious that the motion of the flyer was because of the expansion of the plasma which hindered its temperature rise.There were two more reasons for this phenomenon.One is that the duration of the flyer acceleration was too short for the current to heat the plasma.And the other is that the electric power was not enough to compensate for the heat loss caused by the plasma expansion.This was able to be proved by the comparison of the two groups.One of the groups was Test 1 and Test 2,and the other was Test 4 and Test 5.In the former group,the expansion rate of Test 1 with the higher flyer velocity was higher than that of Test 2, and the plasma temperature of Test 1 should decrease.However, the current and voltage of Test 1 were higher,meaning it could release more electric energy to heat the plasma.In the latter group,the flyer velocities of Test 4 and Test 5 were similar,while the EFI used in Test 4 had a larger ratio of the barrel space to the mass of the Cu bridge.Thus, Test 4 had a higher plasma expansion rate.But the plasma temperature of Test 4 with the higher electric power didn't decrease as that in Test 5.

After being affected by the expansion of the plasma,the plasma temperature continued to be heated by the electric discharge in the LTD and rose up.Hence,there was an obvious lag between the peak of the plasma temperature and the current peak.For explaining how the factors affected the ignition,the electric energy consumed by the electric explosion was divided into the internal energy of the plasma EPIand the kinetic energy EKwhich included the kinetic energy of the flyer and the plasma as shown in Fig.12.Based on the assumption that the plasma was evenly distributed between the flyer and the substrate, the velocity of the plasma was half that of the flyer.

Fig.11.The behaviors of the electric characteristics, the flyer velocity, and the plasma temperature in each test.

Fig.12.The proportion of kinetic energy and internal energy to the total electric energy consumed by the electric explosion.

There were two factors influencing the HTD which was regarded as a critical parameter for the ignition success of the Slapper-BPN type.One was the LTD, and the other was the electric energy consumed by the electric explosion.The EPIs of Test 1 and Test 4 were higher than those of Test 2 and Test 6.However, Test 1 and Test 4 failed to ignite the pyrotechnic,while both Test 2 and Test 6 succeeded.The results indicated that the electric energy injected into the plasma too quickly was not of benefit to the ignition.After the Cu bridge became the copper vapor ionized into the plasma,the plasma had a very high temperature,so the electric energy released in the LTD stage was unable to make the plasma temperature significantly rise.Therefore, the main effect of the electric energy discharge was not to increase the plasma temperature, but to compensate for the heat loss caused by heat conduction, thermal radiation, and plasma expansion, so as to maintain the HTD for a longer time.For meeting this requirement, the electric energy discharge needed a relatively long LTD.However, to maintain the HTD,even with a long LTD,it still needed sufficient electric energy discharge.Test 3 had the longest LTD in the tests,but failed to ignite the pyrotechnic.Compared with the electric energy consumption in Test 1 and Test 2, Test 3 with the same capacitor and charged voltage had the least electric energy for the electric explosion,so it was difficult to make up for the subsequent heat loss.This was due to the inductors used in the tests whose resistances were too large with the result that they wasted too much electric energy.

4.Conclusions

In this research, the electric characteristics, the flyer velocity,and the plasma temperature in the EFI function process were recorded by various devices.The test data was detailly analyzed to reach several important conclusions about the ignition mechanism of the Slapper-BPN type.

(1) Although both the SCB and the EFI have the electric explosion,their ignition mechanisms are not completely the same.The result of the LTD which had been put forward when studying the SCB initiator was not consistent with that of the EFI ignition.The longer LTD was unable to make the probability of the ignition success of the Slapper-BPN type higher.Thus, the methods for evaluating the ignition success of the SCB initiator were not completely applicable to the Slapper-BPN type.

(2) The ignition mechanism of the Slapper-BPN type is not the shock initiation.It was found that the test with the higher impact pressure didn't have a higher probability of ignition success.The results of these tests were contrary to the characteristics of the shock initiation when it came to employing the EFI to detonate the HNS.

(3) The high-temperature duration is regarded as one of the important parameters to judge the ignition success, and the factors affecting the HTD are analyzed.It was discovered that the plasma temperatures of the tests which succeeded to ignite the pyrotechnics dropped more slowly than those of the ignition failure tests.Therefore,the HTD was regarded as one of the important parameters to judge the ignition success.The two factors influencing the HTD were analyzed:one was the LTD and the other was the electric energy consumed by the electric explosion.The longer LTD and the larger electric energy discharge were able to make up for the heat loss caused by heat conduction, thermal radiation, and plasma expansion to extend the HTD.

Negating several wrong guesses about the EFDI ignition, the ignition mechanism of the Slapper-BPN type is considered to be micro-convection heat transfer.The electric energy discharge and the HTD which are the critical factors of EFDI ignition are related to the electric characteristics.When it comes to the EFDI function,the pyrotechnics will limit the expansion of the plasma, and the other kinds of heat loss such as the heat conduction for the plasma to the barrel and the substrate can be calculated.Therefore, a terse method for evaluating the ignition success of the Slapper-BPN type based on the electric characteristics is expected to be developed in the following research.

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.