Improving the damage potential of W-Zr reactive structure material under extreme loading condition

Lu-yao Wang,Jian-wei Jiang,Mei Li,Jian-bing Men,Shu-you Wang

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

Keywords:Reactive structure material Extreme loading Structural integrity Damage potential

ABSTRACT Projectiles made of reactive structure materials(RSM)can damage the target with not only kinetic but also chemical energy,but the enhanced damage potential of RSM may become compromised if extreme loading condition disintegrates the projectile before the target is reached.In this work,a ductile coating of Ni was introduced to a tungsten-zirconium(W-Zr)alloy,a typical brittle RSM,to preserve the damage potential of the projectile.Detonation driving tests were carried out with X-ray photography and gunpowder deflagration driving tests were carried out with high-speed photography for the coated and uncoated RSM samples,respectively.The craters on the witness target were analyzed by scanning electron microscopy and X-ray diffraction.The Ni coating was found to effectively preserve the damage potential of the W-Zr alloy under extreme loading conditions,whereas the uncoated sample fractured and ignited before impacting the target in both detonation and deflagration driving.The crack propagation between the reactively brittle core and the ductile coating was analyzed based on the crack arrest theory to mechanistically demonstrate how the coating improves the structural integrity and preserves the damage potential of the projectile.Specifically,the Ni coating envelops the W-Zr core until the coated sphere penetrates the target,and the coating is then eroded and worn to release the reactive core for the projectile to damage the target more intensively.

1.Introduction

Reactive structure material(RSM)refers to material whose structural strength approaches or exceeds that of steel and undergoes severe chemical exothermic reactions under extreme conditions(e.g.,high temperature,high pressure,laser loading,etc.)[1].The fragments of RSM projectiles possess both kinetic and chemical damage capacity and can thus generate greater damage than steel fragments with same mass[2].For example,Xiao tested RSM and steel fragments under identical impacting conditions.After the steel witness plate is penetrated,compared with the steel fragments,RSM fragments achieve more than five times higher shock-wave peak overpressure behind the target[2].Besides,when a 3 mm thick aluminum is placed 200 mm behind the steel witness plate to evaluate the damage capacity of the fragments,the area of perforation on the aluminum plate is 8-37 times larger when the RSM fragments are used[3].Prefabricated warhead with RSM fragments should effectively damage incoming missiles,aircraft,etc.,as the fragments not only penetrate the targets with their kinetic energy but also accelerate the initiation of munition or trigger oil tank fire with their chemical energy.Nevertheless,RSM fragments may be broken and induced to react under the detonation pressure of warheads,and as of today only some proof-of-principle trials have been reported on warhead with RSM fragments[4-6].If RSM fragments are initiated upon the detonation of warheads,the chemical reaction of RSM fragment will take place over the course of the flight of the projectile,and much less,if any,chemical energy will remain available for incurring damage to the target upon the eventual impact.

Some researchers have attempted to install a thick buffer layer between the RSM and the charge[7]or fill the RSM in a thick inert shell[8]to prevent the loading condition from sabotaging the damage potential of the RSM.A buffer layer of such kind may help protect the RSM fragments but it significantly decreases the charge mass,thus reducing the velocity of fragments and kinetic energy output.Alternatively,some other researchers try to increase the yield strength of the RSM.Existing RSM can be mainly listed in three categories with rising yield strength:(1)polymer-active metal composite[1,9,10](e.g.,PTFE-Al-W,16.46 MPa yield strength[11]);(2)reactive alloy[12-14](e.g.,W-Zr,1060 MPa yield strength[13]);(3)high-entropy alloy(HAE)[15,16](e.g.,Zr41.25Ti13.75Cu12.5Ni10Be22.5,1.8 GPa yield strength[16]),yet each has its own caveat.Studies have shown that RSM fragments of the first two cannot maintain full structural integrity under detonation loading[13-16].Ironically,RSM of the third kind(i.e.,HAE)has high enough yield strength to maintain full structural integrity under detonation loading but find difficulty in achieving thorough activation upon impacting the target[15],and as a result the damage potential from the chemical energy may not be fully realized.

According to Split-Hopkinson Pressure Bar tests[12],W-Zr alloy is a brittle material with high strength.Cleavage fractures develop along the grain boundaries of W-Zr after external pressure is applied,and zirconia is produced at the fracture surface from the exothermic chemical reactions.The thermal stress then accelerates the generation and propagation of new cracks.Therefore,preventing the W-Zr from generating a fractured surface under extreme loading may retard the contact of the reactive material with atmospheric oxygen and thus preserve the damage potential.Koçak[18]proposed that crack propagation in a brittle material may be arrested by applying a material with higher fracture toughness on the surface of the brittle material.Hence,in this work,we applied Ni coating to the surface of W-Zr spheres by electroplating,and then compared the structural integrity and damage potential of the coated and uncoated spheres under detonation and gunpowder deflagration driving.Results and mechanisms were examined and analyzed by X-ray photography,high-speed photography,and scanning electron microscopy.

2.Materials and methods

2.1.Equipment

The crystalline phase components were examined on a D8 Advance(Bruker,Germany)X-ray diffraction(XRD)instrument.Surface morphology was determined with an S-4800(Hitachi,Japan)scanning electron microscope(SEM)apparatus.

Samples driven by deflagration loading were observed by a FASTCAM SA-Z high-speed photography system(PHOTRON,Japan)at a frame rate of 10000 fps.Samples driven by detonation loading were observed by an X-ray photography system(1200 kV pulse,Sweden)to record their states at two moments approximately 20 μs apart.

2.2.Preparation of materials

The W-Zr reactive alloy comprises 52 wt% tungsten,43 wt%zirconium,and 5 wt%nickel.The alloy was prepared from tungsten powder(CAS 7440-33-7,99.98% purity,d50=1-5μm),zirconium powder(CAS 7440-67-7,99.5%purity,stored in water,～200 mesh),and nickel powder(FNiTZ-101,99.5% purity,d50=3-5μm),all of which purchased from Aladdin.Fig.1 illustrates the morphology of three powders observed by SEM analysis.The dendritic feature of the nickel powder helps ensure uniform and tight integration with tungsten and zirconium during sintering.

Fig.1.SEM images of(left)tungsten powder(scale size 2μm),(middle)zirconium powder(scale size 20μm),and(right)nickel powder(scale size 5μm).

2.3.Samples and microstructural characterization

The uncoated spheres were prepared by powder mixing,ball milling,vacuum sintering,and mechanical reshaping.Specifically,the W,Zr,and Ni powders were submerged in double distilled water(used to prevent Zr powder from oxidation)inside a wet ball mixer machine along with carboxymethyl cellulose as the binder.After thorough mixing,the combined powders were sintered at 1533°C and 8×10-3Pa in a high temperature furnace.The sintered body was mechanically reshaped to give crude spheres of W52Zr43Ni5with a diameter of 9.2 mm and a density of 10.23 g/cm3.

Crude spheres were further mechanically ground to a diameter of 8.8 mm before the introduction of the Ni coating.The polished spheres were electroplated in a solution containing nickel sulfate(260 g/L),nickel chloride(40 g/L)and boric acid(45 g/L)at pH=4.0-4.5 to give the coated spheres of 9.2 mm diameter and 10.06 g/cm3.

Fig.2 shows the photos of the crude and coated spheres,along with a diagram of the coated sphere.The uncoated spheres appear black because zirconium was oxidized from exposure to atmosphere during storage.In contrast,the coated spheres have a bright metallic luster.

Fig.2.Photos of the crude uncoated(left)and coated(middle)spheres,and a diagram of the coated sphere(right).

For the coated sphere,Fig.3a shows a clear boundary between the core and the coating(marked with white dotted line)on the SEM image of the cross-section.The area above the boundary is dominated by nickel,as is evident from the line scanning and map scanning results in XRD analysis(Fig.3b).Elements are evenly distributed in the core(Fig.3c),with the content of W,Zr,Ni,C,and O being 49.23 wt%,42.81 wt%,4.32 wt%,1.91 wt%,and 1.73 wt%,respectively.The residual carbon and oxygen resulted from the incomplete removal of the initial atmosphere during the sintering process.

Fig.3.Microstructure and element distribution of the coated sphere.(a)SEM image of the cross-section;(b)line scanning of XRD;(c)map scanning of XRD.

2.4.Detonation driving setup

Detonation driving was carried out for the uncoated and coated spheres to determine how well the Ni coating can preserve the damage potential of the W-Zr alloy.The detonation firing unit(Fig.4)includes the booster pellet,the detonation holder,the hollow cylinder,the positioning device,and the main charge.These components are assembled by applying AB glue.The positioning device(Fig.5)is made of ABS plastic by 3D printing and helps ensure that the fired sphere and the charge are coaxial.Table 1 lists the parameters of the key components.

Fig.4.Schematic(left)and picture(right)of components in the detonation firing unit.

Fig.6 shows the detonation driving setup in full.The firing unit is suspended from the cantilever 4 m above the witness target by a cotton rope to ensure vertical firing of the sphere.The witness target consists of a Q235 steel plate(6 mm thick)that covers a steel box filled with cotton yarn.A pulsed photography system,consisting of two parallel X-ray tubes and a film,is adopted to capture the state of the fired sphere at two moments after detonation.The detonation initiates a trigger to synchronize the X-ray signals so that the first tube emits a pulsed signal at time T1after detonation and the second tube emits at time T2.Two reference steel cubes 50 mm apart are adhered to the film to calibrate the size of the sphere captured on the film.To protect the system from the disturbance of shock wave and fragments,a 10 mm thick aluminum shield is placed before the film,and a protective box made of 10 mm thick Q235 steel surrounds the detonation firing unit.Fig.7 shows the photography assembly,the detonation firing unit,and the witness target.

2.5.Gunpowder deflagration driving setup

The spheres are fired off from a smooth-bore gun(12.7 mm inner diameter)to the witness target,a 6 mm thick Q235 steel plate.The firing velocity of the sphere is manipulated by adjusting the mass of the propellant in the barrel.An on-off velocity measuring system,which consists of a timer along with on-off signal sensors,records the interval for the projectile to pass different sensors.A high-speed camera that points to the witness target from the side records the trajectory and the energy-releasing behavior of the projectile before and after the target is penetrated(Fig.8).

Fig.5.Schematic(left)and photo(right)of the positioning device.

Table 1Components in the detonation firing unit.

Fig.6.Schematic diagram of detonation driving setup.

3.Results

3.1.Detonation driving

Fig.9 shows the X-ray photos captured at two typical moments T1and T2after detonation.The velocity was 745 m/s and 845 m/s for the uncoated and coated sphere,respectively.Whereas the uncoated sphere disintegrated during its movement,the structure of the coated sphere remained intact under identical detonation loading conditions(Fig.9).

Fig.10 shows the damage to the witness target by the uncoated and coated spheres.The uncoated sphere,which disintegrated before hitting the witness plate,produced on the surface of the steel witness plate a moon loamy uneven blackened pit(Fig.10a).On the contrary,the coated sphere penetrated the cover plate to give a crater with smooth walls(Fig.10b),and the diameter of the through-hole on the front and back was 12.11 mm and 12.09 mm,respectively.The cotton yarn in the steel box did not burn when the uncoated sphere was fired(Fig.10c)but burned for more than 2 min when the coated sphere was tested(Fig.10d).

Fig.7.Experimental setup of the detonation driving test.

Fig.8.Deflagration driving experimental setup(left:schematic diagram;right:photo).

Fig.9.X-ray photos of the coated and uncoated spheres at two typical moments after detonation.

3.2.Gunpowder deflagration driving

Fig.11 shows the video frames from high-speed photography.The firing velocity was 1323 m/s for the uncoated sphere and 1267 m/s for the coated sphere.Before hitting the witness target,the uncoated sphere exhibited a trajectory of a bright line(Fig.11U1,2),whereas the coated sphere did not have a clearly visualized trajectory(Fig.11C1,2).After penetrating the target plate(Fig.11U3-8 and Fig.11C3-8),both spheres were broken into small debris and reacted rapidly with air to release substantial heat[17]to give a gradually expanding fireball.The fireball of the uncoated sphere was always smaller than that of the coated sphere at the same instant.Fig.12 shows the craters on the witness target.The uncoated sphere generated two holes with different size on the plate,whereas the coated sphere cleanly penetrated the target and created only a smooth hole.

Fig.10.Damages to the witness target by the spheres after detonation driving.

4.Discussion

4.1.Damage pattern

Deflagration driving gives kinetic energy to the projectile with the pressure inside the barrel from the combustion of gunpowder that takes place over a longer time scale than denotation,whereas detonation driving gives the sphere a higher transient pressure but a lower initial velocity.With detonation driving,the uncoated sphere got fractured during the firing process because of the ultrahigh detonation pressure and failed to eventually penetrate the covering steel plate.In the case of gunpowder deflagration driving,although the uncoated sphere also disintegrated upon firing,it did penetrate the witness target and left on it two irregular holes.The coated sphere remained intact after firing by either detonation or deflagration driving and very well penetrated the witness target in both cases.

The smooth crater left by the coated sphere on the witness plate was incised for microscopic analysis(Fig.13).The 1/4 crater after inlaying was spray-coated with gold powder to assist the analysis of the SEM image and XRD line scanning(Fig.14).The areas A,B,and C in Fig.14 represent the inner wall of the crater,the crater-target interface,and the target,respectively.Since the combustion of zirconium particles accompanies sustainable heat release,the reaction products of the W-Zr alloy give spherical granular structures in area A(Fig.14)under thermal stress and surface tension[16].Area B appears pale grey and its main components are Ni,Zr and W according to EDS results.Moreover,the strength of the diffraction signal reaches 40,25,and 50 CPS for Ni,W and Zr,respectively.It is noteworthy that the content of Ni detected was much higher than that proportion of Ni in the initial RSM core.It is further confirmed that the Ni coating remained adhered to the W-Zr core until the sphere penetrated the target,and the coating was eroded by the deformation resistance of the target.This conclusion is consistent with the results of X-ray and high-speed photography.Under extreme loading,the coated sphere maintained structural integrity and released chemical energy to enhance damage.In other words,because of the protection from the Ni coating,the W-Zr core debris is released and reacts only behind the witness target.

4.2.Crack arrest analysis

The cracking of a brittle material can be arrested to some extent,when it is wrapped by a new material with higher fracture toughness[18].Fig.15a shows a simplified diagram of crack propagation at the interface from the brittle core to the protective coating,in which the crack is simplified as an ellipsoid moving and expanding in a single direction with the long axis of ellipsoid defined as the crack length.

Fig.15b plots the relationship between the driving force of crack(G)and the crack length(L)according to the theory proposed by Koçak[18,19].Specifically,crack propagating only in the core is represented by the black dotted curve ABC1D.The crack stops propagating when the area of ABC1is equal to that of C1DE,and the corresponding final crack length extends to point E.In the presence of coating,crack propagating from the core with a lower fracture toughness of KIC1to the coating with a higher fracture toughness of KIC2is represented by the red curve,and the crack stops propagating when the area of ABF1C2is equal to the area of IHGF2,thus restricting the crack only to point J.

The driving force for the crack propagation can be expressed as G=kL/E,where k is a function ofσandτ;L is the crack length,and E is the Young’s modulus of the material[19].For the studied spheres,the protective Ni coating has E=68 GPa and KIC=80-100 MPa/m0.5[20].The brittle W-Zr core has E=11 GPa[21]and its KICvalue can be estimated from Eq.(1)based on the Vickers hardness test[22]:

where,HVis the Vickers hardness,E is the Young’s modulus,a is half the length of the indentation diagonal,and l is half the length of the Palmqvist crack.

Fig.11.Typical high-speed photographs of(U)uncoated sphere and(C)coated sphere penetrating the witness target(6 mm thickness Q235 steel).Firing velocity is 1323 m/s for(U)and 1267 m/s for(C).

Fig.12.Craters on the witness target(6 mm thick Q235 steel)generated by the uncoated and coated spheres.

Fig.13.Specimens prepared for microscopic analysis.

Fig.14.SEM and EDS results of section of 1/4 crater.

Fig.16 shows the SEM image of the W-Zr core in the Vickers hardness test.The W-Zr core was found to have KIC=4.35 MPa/m0.5from the results of the Vickers hardness test(Table 2).

That is,the Ni coating has about 18-22 times higher KICthan the W-Zr core,as well as a lower crack driving force than in core due to the difference of Young’s modulus.Therefore,when the crack on the core reaches the interface at point F1,the crack propagation is slowed and stops when the area of IHGF2equals that of ABF1C2.Hence,in the presence of the nickel coating,the crack becomes far smaller and only extends to point J,instead of going to point E when the W-Zr sphere does not have any protective coating.

4.3.Protection of damage potential by the Ni layer:a mechanistic analysis

Both X-ray and high-speed photography show that fragmentation and combustion occur and reinforce each other as soon as the uncoated sphere ignites,even if the projectile has flown away from the detonation products and has not reached the target.In contrast,under identical loading conditions,the coated sphere experiences fragmentation and combustion only after the penetration of the target thanks to the protective effect of the Ni coating,which prevents formation of initial W-Zr fragments and contact with oxygen.

The flight and penetration can be further compared between the uncoated and coated spheres to reveal how crack arrest in the initial stage helps preserve the damage potential of the W-Zr core.Upon initial impact,the uncoated sphere generates fragments,and temperature increases in the atmosphere around the fragments because of both shock and combustion.The zirconium in the reactive W-Zr reacts with oxygen in air to release extensive heat upon combustion(Equation(2)),and this reaction quickly generates a zirconia film that has a high melting point of 2670°C on the surface of the fragments.The combustion of metal should require a temperature higher than the melting point of the zirconia film for the unoxidized metal to sustain the combustion[23].However,Liu XJ found that the combustion temperature of W-Zr fragments in air is only 1348°C[21],far lower than the melting point of the zirconia film.

Fig.15.(a)Schematic diagram of crack propagation at the interface between different medium.(b)Relationship between the crack driving force G and the crack length L.Black dotted line is the crack propagation in the brittle core in the absence of coating,and red solid line is the progression of the crack from the core to the coating.Reproduced from Refs.[18,19].

Fig.16.The W-Zr core in the Vickers hardness test.The shade area in the left image is Palmqvist crack zone in the test.The SEM observation of the indentation is given in the image on the right.

Table 2Vickers hardness test results of the W-Zr core.

We speculate this is possibly because the zirconia film and the enclosed W-Zr alloy appear to undergo continuous cracking and burning when the mismatch of thermal stress in the temperature field breaks apart the zirconia film.A simplified model analysis is given as follows,in which W-Zr fragments are assumed spherical,and zirconia films are assumed to have uniform thickness.Besides,the difference of particle size in the temperature field is neglected,the stress of particles in the initial state is considered zero,and the temperature becomes uniform instantaneously.In this way,all stress distributions of particles during temperature rise have one dimensional spherical symmetry.

Fig.17 shows a simplified model of the W-Zr fragments,where black dotted circles represent unreacted W-Zr particles and green shades represent zirconia.In this model,rcis the radius of unreacted W-Zr particles,which are covered by zirconia to form an overall spherical fragment particle with a radius of rp.For the fanshaped microelement of the particle(red region in Fig.17),the radial stress isσrand the tangential stress isσθ.Other parameters referring to the W-Zr core and the zirconia film adopt a subscript of 1 and 2,respectively.

Fig.17.Simplified stress distribution model of W-Zr fragments in the temperature field.

Analysis of the fan-shaped microelements gives[24]:

The radial strain and the tangential strain need to satisfy Eqs.(4)and(5):

whereεris the radial strain,εθis the tangential strain,E is the elastic modulus,νis the Poisson’s ratio,a is the coefficient of thermal expansion,T is the temperature rise of the particle,and u represents the radial displacement components.The radial displacement can be evaluated by combining(4)and(5)to give(6):

of which the general solution is

It can be seen from(7)that the radial and tangential stresses must satisfy the following:

where k1and k2are constants and x is the lower limit of integration.The solution of the fragment particle consists of two parts.The part for unreacted W-Zr is obtained by integrating from r=0 to r=rc:

The part for zirconia is obtained by integrating from r=rcto r=rp:

Therefore,k11,k21,k22must satisfy(11)because of the boundary conditions of the solutions(i.e.,u1|r=rc=u2|r=rc，σ1r|r=rc=σ2r|r=rc，σ2r|r=rp=0):

At the outer edge of the film(i.e.,r=rp),the film break apart under thermal stress when the yield limit|σ2r-σ2θ|=σ2yis reached.

Table 3 lists the physical properties of both zirconia and the unreacted W-Zr particles.It can thus be derived by combining(11)and(12)that the temperature for the zirconia film to attain its yield limit is 670°C.

Table 3Physical properties of the W-Zr alloy and zirconia[20].

The preceding analysis shows that the zirconia film is broken at 670°C,at which point the enclosed W and Zr are exposed to oxygen and start exothermic combustion reactions.The combustion in turn further increases the temperature.In this way,as soon as the uncoated sphere is activated,fragmentation and combustion will mutually reinforce in a positive feedback.

The coated spheres under extreme loading conditions also experience temperature rise and internal cracking,but the results in Section 3 have shown that the coating effectively stops the cracks from propagating.In other words,even if some cracks are formed in the core,subsequent oxidative combustion does not ensue because the coating effectively isolates the W-Zr fragments from the oxygen in the atmosphere.Because the sphere is not subject to external loading during its flight,the partially cracked core is enclosed in the coating and approaches the target until the eventual impact.When the sphere penetrates the target plate,the coating is eroded and the enclosed W-Zr then gets to contact oxygen,and continuous fragmentation and combustion reactions then follow.

Fig.18 shows the behavior of uncoated and coated W-Zr spheres at different stages(e.g.,initial loading,flight,penetration).The coating effectively prevents premature reaction between the core and oxygen under extreme driving,and thus ensures that the sphere during its flight is free of sustained fragmentation and combustion.In this way,the damage potential of the RSM is well preserved and only release after the target is impacted.

Fig.18.Schematic illustration of the behavior of coated and uncoated spheres at individual stages.

5.Conclusion

We applied ductile Ni coating to the surface of W-Zr reactive material by electroplating,and carried out detonation driving and gunpowder deflagration driving experiments for both coated and uncoated W-Zr spheres to study their motion state,damage pattern,and energy-releasing behavior by high-speed photography,X-ray photography,and microscopic analysis.The coating was found to effectively preserve the damage potential of the W-Zr reactive material.

The uncoated sphere already generates fragments upon initial loading,and the fragmentation are in contact with oxygen since initiation.Since temperature readily exceeds 670°C along the path of the sphere’s flight,the zirconia film on the surface of the fragments breaks apart due to the mismatch of thermal stress and the unreacted alloy in the fragments continue to react,hence consuming the precious chemical energy of the uncoated sphere before it hit the target.In contrast,the Ni coating successfully arrests the propagation of cracks that originate from the W-Zr core,and the alloy is protected from contact oxygen while the sphere is still in flight.In this way,premature consumption of chemical energy from the fragments doesn’t occur for the coated spheres.

Declaration of competing interest

We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work,there is no professional or other personal interest of any nature or kind in any product,service and/or company that could be construed as influencing the position presented in,or the review of,the manuscript entitled.

Acknowledgments

National Natural Science Foundation of China.Grant ID:11872123.