Hong-qi Nie,Hay Yee Chan,Sreekumar Pisharath,Huey Hoon Hng
Energetics Research Institute,Nanyang Technological University,Singapore
Keywords:Nanothermite Bimetal Combustion efficiency Aging Thermochemical calculations
ABSTRACT Nanothermites have been employed as fuel additives in energetic formulations due to their higher energy density over CHNO energetics.Nevertheless,sintering and degradation of nanoparticles significantly limit the practical use of nanothermites.In this work,combustion characteristic and aging behavior of aluminum/iron oxide(Al/Fe2O3)nanothermite mixtures were investigated in the presence of micron-scale nickel aimed to produce bimetal thermite powders.The results showed that the alumina content in the combustion residue increased from 88.3% for Al/Fe2O3 nanothermite to 96.5% for the nanothermite mixture containing 20 wt%nickel.Finer particle sizes of combustion residue were obtained for the nanothermite mixtures containing nickel,indicative of the reduced agglomeration.Both results suggested a more complete combustion in the bimetal thermite powders.Aging behavior of the nanothermite mixture was also assessed by measuring the heat of combustion of the mixture before and after aging process.The reduction in heat of combustion of nanothermite mixtures containing nickel was less severe as compared to a significant decrease for the nanothermite mixture without nickel,indicating better aging resistance of the bimetal thermite powders.
Nanothermites belong to a class of energetic materials comprising of a metal fuel and a metal oxidizer in nano-scale that are well-known to undergo a rapid and exothermic redox reaction[1,2].Compared with traditional energetic materials at micron-size scale,nanothermites are featured with higher energy density and significantly enhanced reactivity.It can be attributed to the dominance of kinetic control mechanism over diffusion control in determining the reaction rate between the two constituents[3-6].A variety of applications have been realized for nanothermites primarily as energetic additives in explosives and propellants[7-10],as well as other applications such as gas generators[11],nano-scale welding[12,13],micropropulsion[14]and electric igniters[15].
To take full advantage of nanoparticles for the aforementioned applications,existing problems such as particle sintering[16]resulting in dramatically reduced materials’effectiveness,as well as a propensity to degradation by aging[17,18]causing deterioration of reactive properties,remain unsolved.For instance,high heating rate dynamic transmission electron microscopy(DTEM)investigation of aluminum/copper(Al/CuO)nanocomposites has indicated rapid loss of the nanostructure[19].It reveals that particles in nano-scale become susceptible to morphological changes and would be thermodynamically driven into nanoparticle aggregates[20-22]when they are exposed to high temperatures and rapid heating scenarios.Incomplete reactions[23]have also been found to occur in nanothermite composites,especially for Al/CuO,owing to the severe sintering of the reactants before the reaction proceeds to completion.
Experimental studies on the burning of aluminized propellants have indicated that coalescence process of aluminum particles starts at~780 K,which is well below the aluminum melting point(933 K).The coalescence process is enhanced by the formation of“bridges”across the particles.At this temperature,the alumina shell remains solid and retards fast oxidation and self-sustained combustion of the particles[24].Ignition of the particles takes place when the particles achieve a temperature of about 1300-1500 K.When the temperature of alumina melting point(2300 K)is reached,coalescence of adjacent particles is accomplished resulting in the formation of larger agglomerates[25].
Sintering of the aluminum particles could be reduced by minimizing the residence time of the molten aluminum particles at the burning surface[26].This is achieved by adding another metal that reacts with the aluminum core well below the aluminum melting temperature.Nickel-aluminum(Ni-Al)mixtures have received significant attention in this regard because of the possible occurrence of Ni-Al reactions.Several research works[27-29]on Al-Ni reactions had experimentally indicated that,intermediate reaction products in metastable phase are formed at~480 K[29]well below the temperature[19]at which the coalescence of n-Al particles is most likely to occur.It has been observed that nickel coated aluminum can reduce the agglomerate diameter by 50% in comparison to comparable sized aluminum[30].Furthermore,it was reported[31]that bimetal Ni-Al powders undergo oxidation at a slower rate as compared to n-Al,indicative of a lower tendency towards aging for bimetal thermite powders.These aspects have not been investigated thoroughly in the context of nanothermites.
The objective of this work is to investigate how the addition of Ni powders to Al/Fe2O3nanothermites is capable of:a)mitigating the sintering issues of aluminum particles and hence promote reaction completeness;and b)alleviating the effect of aging on the performance of nanothermite system.In this work,the mass weightage of Ni is varied from 5 to 20 wt%with respect to the mass of Al to prepare the bimetal thermites with Fe2O3as oxidizer.The prepared bimetal thermite powders were characterized by electron microscopy,X-ray diffraction and thermal analysis techniques.Combustion efficiency was evaluated by quantitatively analyzing the combustion products.Powder morphology and particle size distribution of combustion residues were analyzed by scanning electron microscopy.Aging study was carried out based on isothermal micro-calorimetry measurements using a thermal activity monitor(TAM)under controlled humidity for both the bimetal thermite powders and a standard Al/Fe2O3nanothermite as reference.The effect of aging on heat of combustion was also investigated.
Nano-aluminum(n-Al)powder supplied by Armament Research,Development and Engineering Centre(ARDEC)in U.S.and iron oxide(n-Fe2O3)nanopowder procured commercially from Alfa Aesar were the major constituents of the nanothermite.The nickel(Ni)powder purchased from Strem Chemicals was selected as the secondary metal component to yield the bimetal thermite powders.In order to exclude the possibility of particle sintering in nano-scale Ni powder due to its high surface free energy[32],the Ni powder in micron-scale was considered.All starting materials were used as received without any modifications.
Field Emission Scanning Microscope(JEOL JSM-7600F-FESEM)operated at accelerating voltage of 2 kV was employed to acquire the SEM images for the starting materials.n-Al and n-Fe2O3powders were well dispersed in isopropanol through ultrasonication for 10 min before cast drying onto a silicon wafer in order to reduce agglomeration.Ni powder was directly cast onto a silicon wafer as the dry powder.All samples were coated with platinum at 20 mA for 30 s prior to SEM observation.Fig.1 shows that n-Al appears with spherical morphology while n-Fe2O3and Ni appear as irregularly shaped particles.In addition,small particles residing on the surface of larger ones are observed,particularly in the Ni powder.The composing content in small particles is believed to be as same as the larger ones.The small particles are likely to provide more surfaces available for the oxidation reaction in comparison with the larger counterparts.
The particle size distribution of the starting materials was measured using theParticaLA-950 laser scattering particle size analyzer by HORIBA.The samples were dispersed in isopropanol with the help of ultrasonication to minimize particle agglomeration prior to the measurement.The measured particle size distribution in Fig.2 shows that the average particle size of n-Fe2O3powder is smaller(~180 nm)with a narrower variance as compared to that of n-Al(~600 nm).The average particle size of Ni used to prepare the bimetal thermite powder is(~18μm).
The bimetal thermite powders(Al/Fe2O3/Ni)were prepared through physically mixing using ultrasonication.The ultrasonication is a simple method that enables one to obtain the well mixed powder system by dispersion of the reactants followed by evaporation of solvent.All the ultrasonication processes were carried out in Elma S60H Elmasonic with ultrasonic frequency and power of 37 kHz and 150 W,respectively.Al and Fe2O3nanopowders were weighed in the respective amounts to prepare the nanothermite mixture at equivalence ratio of 1.2.Optimum combustion performance was observed at this equivalence ratio in our previous work[33].The active content of Al is around 70% as determined by thermogravimetric analysis(TGA)which was considered in the calculation of equivalence ratio.The mass gain(Δm)in TGA is attributed to oxidation of aluminum,and thus the aluminum content(c)can be estimated using the following Eq.(1)based on a mass balance and the mass ratio of aluminum to oxygen in Al2O3[34]:
The two metallic powders were first ultra-sonicated in 20 mL of isopropanol for 15 min independently,resulting in the breaking up of initial agglomerates in each powder.Further ultrasonication was applied after mixing the two components together for an additional 60 min in isopropanol.The Al/Fe2O3nanothermite powders were used as a benchmark to be compared with the bimetal thermite powders.To produce the bimetal thermite powders,Ni was introduced to the mixture of Al/Fe2O3nanothermites in the range of 5-20 wt% with respect to the mass of Al,by performing an additional 30 min of ultrasonication.The bimetal thermite powders were recovered by dry casting on a flat steel pan and gently collected using an anti-static brush.
2.3.1.Bomb calorimetry
The bimetal thermite powders with varying amounts of Ni in the range of 5-20 wt%with respect to the mass of Al were prepared and experimentally characterized for their total energy output using Parr 6200 bomb calorimeter.The atmosphere in the bomb vessel was argon keeping the pressure at 250 psi.
2.3.2.Thermo-analytical measurement
The measurements for kinetic analysis were focused on Al and bimetal(Al/Ni)powders since aging and reactivity of the thermite mixture is governed by Al fuel[35,36].The weight gain during the oxidation of n-Al mixed with 10 and 20 wt%of Ni was measured at multiple ramp rates ranging from 5 to 20 K/min by Shimadzu DTG-60H equipment.The powders were heated to 700°C in purified air atmosphere at 50 mL/min flow rate.Activation energy was derived from the differential thermal analysis(DTA)curves by using Kissinger analysis.The oxidation of n-Al was treated as a control in this measurement.
Fig.1.FESEM micrographs of the starting materials of(a)n-Al,(b)n-Fe2O3 and(c)Ni.
2.3.3.Isothermal measurement
Heat flow calorimetry measurements were conducted using TA Instruments TAMIII microcalorimeter at a constant temperature of 60°C for the purpose of studying the effect of aging on the reactivity of bimetal thermite powders.The Al/Fe2O3nanothermite powders served as a benchmark.In aging experiment,six independent calorimeters were used simultaneously.Approximately 10 mg of bimetal thermite powders with 10 and 20 wt%of Ni with respect to the weight of n-Al were placed inside separate glass vials.The relative humidity(RH)of the air in the glass vials was controlled by placing ampoules containing saturated solutions of potassium fluoride(for 20%RH)and sodium chloride(for 75%RH).The aging process was monitored isothermally for 20 days.The aged samples were recovered and characterized by XRD,SEM and bomb calorimetry.
2.3.4.X-ray diffraction analysis
XRD analysis of the combustion residues were recorded using a Bruker D8 Advance Powder X-ray Diffractometer featuring a filtered Cu-Kαradiation(λ=1.5418?).Phase identification was done using the Match!software which is integrated with the latest International Centre for Diffraction Data(ICDD)database.Quantitative analyses of the sample XRD pattern were also performed using Bruker TOPAS Version 3.0.
Thermochemical calculations were performed to obtain the heat of combustion of the bimetal thermite powders with increasing Ni content.The thermochemical equilibrium code,EXPLO5[37]was used in this work to calculate the equilibrium compositions and combustion characteristics of the bimetal thermite,under isochoric conditions.
The combustion calculations are based on the free energy minimization technique[38].It assumes that when the chemical potential of reaction products is equal to that of initial reactants in the equilibrium state,the free energy of product approaches its minimum value.The code uses either the virial or ideal gas equation of state for gaseous combustion products,while condensed products(solid and liquids)are considered incompressible.
The combustion of a thermite mixture is an irreversible process in which it transforms into products which may be in condensed liquid or gaseous phase.Naturally,the reaction is accompanied by heat liberation.The calculation of thermodynamic properties of combustion products under isochoric conditions is based on the assumptions:a)the combustion proceeds adiabatically and b)the state of equilibrium establishes in the products.Hence,only the initial participating reactants and the final combustion products are important,as the state of chemical equilibrium rapidly establishes in combustion products.The calculation goes as follows:
(i)The system of equations,mathematically describing the state of equilibrium in the combustion products,is formed on the basis of the free energy minimization method.These equations are solved applying modified Newton’s steepest descent method.
(ii)The state of gaseous combustion products is described by the virial or ideal gas equation of state.
(iii)Condensed phase products are considered incompressible.
(iv)Thermodynamic functions of combustion gases,as real gases,are derived applying thermodynamic laws and the virial equation of state.
(v)Thermodynamic functions of combustion products in the standard state are derived from the enthalpy.
The heat of combustion(Qv)or reaction internal energy(ΔE0rxn)is calculated applying Hess’s law as the difference between sum of the energies of formation of combustion products and energy of formation of reactant(energetic material),shown in Eq.(2):
Fig.2.Particle size distribution of the starting materials of n-Al,n-Fe2O3 and Ni powders.
Fig.3.Experimentally measured and computed energy outputs of bimetal thermite powders with varying Ni content of 0-35 wt%(with respect to the mass of Al).
Fig.4.The evolution of weight changes(a)and corresponding DTA traces(b)for Al and bimetal powders with respect to temperature.Samples were heated at a heating rate of 5 K/min.
whereniis the mole number of theith combustion product and(ΔfE0298)iis the energy of formation ofith combustion product.
The heats of combustion obtained from the bomb calorimeter were compared to that calculated using EXPLO5.The experimental of 3949 J/g is closer to the theoretical value of 3954 J/g reported by Fischer and his co-worker[1].The addition of 20 wt%Ni decreases the heat of combustion of Al/Fe2O3nanothermite by 5.7%.The dilution in energy is due to the replacement of the more energy releasing Al with micron scale Ni powder in the mixture.
The calculated heats of combustions are shown in Fig.3 for comparison.The measured and computed energy output display a similar trend when the Ni content is increased from 0 to 20 wt%.Thermochemical calculations show the heats of combustion lower than experimental output for all Ni content.A larger drop in the heat of combustion with increasing Ni content(up to 35 wt%)is also observed as compared to that recorded in the experimental studies.
The lowered heats of combustion calculated from EXPLO5 are likely attributed to that the calculations were performed assuming conditions of the combustion front,where temperature ranges from 2000 to 4000 K.Under such conditions,Ni would be in liquid or gaseous(l/g)form since Ni melts at~1700 K and boils at about 3100 K.In the bomb calorimeter experiments,products are cooled to a temperature below which no change in temperature is observed.It suggests that products such as Ni and Al2O3are likely to be in the solid phase.Comparing experimentally measured and calculated heat of combustion shows that the calculated heat was much lower than measured value.It could be possibly attributed to a certain amount of energy(specifically enthalpy of fusion and vaporization)has been“consumed”during the simulation to convert the solid products to l/g.Therefore,the calculated heats will be much lower as long as there are l/g products.The observed gap can be narrowed if the exact products at the combustion front can be known,but in-situ analysis remains an experimental challenge for now.Nonetheless,both calculated and experimentally measured heats of combustion show that the addition of Ni,at least up to 20 wt %,does not have a detrimental effect on the energy output.
The thermogravimetric analysis(TGA)traces representing the weight changes of the studied samples as a function of temperature are shown in Fig.4(a).A slight weight loss at low temperatures is observed for all samples,particularly in bimetal powders,which can be ascribed to the decomposition of solvent residues adsorbed on the metal surfaces.The noticeable mass increases due to oxidation begin above 500°C for all samples.In addition,the corresponding DTA curves shown in Fig.4(b)indicated that the onset temperatures shift to higher temperatures as the Ni content is increased from 0 to 20 wt%,indicating a greater thermal stability possessed by the bimetal powders.
DTA plots shown in Fig.4(b)were used to derive activation energies according to the kinetic analysis method[39]using the DTA peak maximaTm,according to Eq.(3):
whereβis the heating rate,Tmis the peak maximum temperature,Ais the pre-exponential factor,Ris the gas constant,nis the empirical order of reaction,mdenotes the maximum,αis the reaction progress andEais the activation energy.
Fig.5.Kissinger plot comparing the activation energy obtained for n-Al powder and bimetal powders in air.
Fig.6.Comparison of total energy outputs obtained for the fresh and aged bimetal powders with 0-20 wt.% of Ni additions under 20% and 75% humidity levels.
The activation energyEacan be obtained by plotting lnversus 1/Tmfor a series of runs at different heating ratesβ.Fig.5 shows the plot of lnversus 1/Tmfor n-Al powder and the bimetal powder.The slope of the two lines,which were fitted using linear regression and minimization of least squares,givesEa/R,from which the activation energyEacan be derived.It can be found that the activation energies obtained for the bimetal powders are reduced,indicative of a higher reactivity in comparison with n-Al powder without Ni additions once the reaction is initiated.The reduced activation energy could be attributed to the interaction between Ni and Al that is a highly exothermic reaction.The generated heat increases the reaction temperature possibly resulting in creation of cracks on the oxide shell of Al as a result of its polymorphic phase change[40].The exposure of Al surface to the oxygen is facilitated due to the formation of cracks,thereby enhancing the reaction kinetics.
The aged samples subjected to two RH levels in air under an isothermal environment(~60°C)were recovered and their respective total energy outputs were re-evaluated with bomb calorimetric measurements.The measured total energy outputs for aged and fresh samples with varying Ni content were plotted and compared in Fig.6.In general,reduced energy outputs are observed for all aged samples as compared to the fresh powders.The degree of reduction in energy outputs are found to be comparable across the Ni range studied for the powders subjected to a moderate humidity level at 20%.A substantial reduction in energy output of 33%is observed for the Al/Fe2O3nanothermite powder without Ni at the extreme humidity level of 75%.
In contrast,the reduction in energy outputs is less severe,at 13%and 12%for the bimetal thermite powders with 10 and 20 wt.%Ni,respectively.This indicates that the energy output is mostly preserved as Ni is increasingly introduced to form the bimetal thermite powders.The susceptibility of n-Al powder to degrade under humid conditions at relative high temperatures is greatly alleviated.
It was reported[41]that,under a hot and humid condition,the Al particles grow bayerite layer because of the hydration reaction which later sinter together forming continuous bayerite-aluminum composites.The sintered Al composites have exhibited a decreased energy output that could be ascribed to the significant reduction in the reacting interface of Al.On the other hand,with addition of Ni,the surface of Ni particles provides the site allowing nano Al particles to reside on.The hydration process can be taken place on the interface between Ni and Al.It has been observed[42]that the hydration layer of Ni can readily undergoes the decomposition from the Ni surface at temperatures far below that of Al combustion.This decomposition process allows to free the nano Al particles from the Ni surface,ensuring the Al surfaces are mostly available for the subsequent oxidation,thereby mitigating the aging effect on the combustion.
Fig.7.XRD patterns of combustion products recovered for the nanothermite(n-Al/Fe2O3)and the bimetal thermite powders.
Table 1Quantitative analysis of combustion products of nanothermite and bimetal thermite powder.
Fig.8.SEM micrographs of combustion products analyzed for(a)n-Al/Fe2O3,(b)n-Al/Fe2O3/10Ni and(c)n-Al/Fe2O3/20Ni.
The combustion products of nanothermite and bimetal thermite powders were collected after the bomb calorimetric measurements conducted in argon.The recovered combustion products were analyzed using XRD to quantitatively determine the residual compositions.Fig.7 shows the XRD patterns of the products recovered after combustion.
Fig.9.Particle size distribution obtained from processing of SEM images taken for the combustion residues collected from(a)n-Al/Fe2O3,(b)n-Al/Fe2O3/10Ni and(c)n-Al/Fe2O3/20Ni.
The main products were identified for the n-Al/Fe2O3and the bimetal thermite powders as alumina(Al2O3and Al2.66O4)and iron(Fe)from the resulting XRD patterns,which is consistent with that reported in literature[43].It could be clearly seen that an increase in the intensity of Al2O3peaks(labeled by the circles in Fig.7)when Ni is added to the nanothermite powder.Since the composition of the nanothermite is only slightly fuel-rich(equivalence ratio of 1.2),the intermetallic phase of Fe3Al which is supposed to be found in the over-aluminized mixture is not present in the combustion products for both powders.Instead,Fe3O4is found in the combustion products,which likely resulted from the reduction of Fe2O3to Fe3O4and FeO.The reaction proceeds further between the remaining Al and FeO with the formation of alumina and elemental Fe.Additionally,the presence of elemental Ni in the combustion products of bimetal thermite powders suggests that the intermetallic of Ni/Al oxidizes further with oxygen to form Al2O3and elemental Ni.Results of the quantitative analysis of the combustion products are tabulated in Table 1.
Quantitative analyses of the sample XRD pattern were also performed using Bruker TOPAS Version 3.0.The results are representative of the weight percentage of the combustion products which are detectable by XRD.The amount of iron(Fe)is relatively low that could be due to the product of Fe is mostly in amorphous state.Moreover,an improvement in combustion efficiency for bimetal thermite powders is indicated by the higher amount of alumina(94.7 and 96.5 wt.%)present in the combustion products of bimetal thermite powders with 10 and 20% Ni addition as compared to that of nanothermite powders(88.3 wt.%)as shown in Table 1.This finding is further strengthened by the lower activation energy found for the oxidation of bimetal powders,versus that obtained for Al powder alone,indicating a higher reactivity of nanothermite powders when Ni is added.
SEM micrographs in Fig.8 show less agglomeration of the combustion products when Ni is present(Fig.8(b)and(c))as compared to nanothermite powder alone(Fig.8(a)),suggesting that less sintering has taken place for the bimetal thermite powders during combustion.It is plausible that the heat release from the intermetallic reaction between Ni and Al enhances the reaction temperature that leads to the increase in the formation of gaseous products,e.g.,Fe.The gaseous products are likely to aid in breakup of the Al agglomerates in the combustion process[23].
Furthermore,particle size distributions of collected combustion residues were determined by processing multiple SEM images at different magnifications ensuring the entire range of particle sizes is included using ImageJ(a freely available software package).This approach has been detailed in recent published work[44,45].The obtained particle size distributions are sorted into logarithmic bins and presented in Fig.9.The particle size distribution appears to be relatively broad for Al/Fe2O3nanothermite powder spreading over 10-100 nm.In contrast,the particle size distributions of bimetal thermite powders are skewed towards the finer particle sizes as compared to that of Al/Fe2O3nanothermite powder.The results confirm that a less agglomeration in the combustion of Al/Fe2O3nanothermite in the presence of Ni,leading to the higher combustion efficiency.
In this study,the bimetal thermite powders were prepared by physically mixing Al/Fe2O3nanothermite powders with nickel powder at micron-scale ranging from 0 to 20 wt% with respect to the mass of Al.Based on the results of characterization,the following conclusions have been drawn:
1.Both experimental and calculated heats of combustion indicated a slight reduction in the total energy output with increasing Ni content.
2.The vulnerability of nanothermite powders to humidity at relative high temperatures was substantially reduced when Ni is added,and an extended shelf life of bimetal thermite powders can be expected.
3.An enhancement in combustion efficiency in bimetal thermite powders was attained and supported by observations of large amounts of alumina as well as finer particle sizes in the combustion products,hence indicative of higher reaction completeness.
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.
Acknowledgement
The authors would like to thank Ms Tan Meng Lu and Dr Chong Che Chang for the assistance on sample characterization.This research did not receive any specific grant from funding agencies in the public,commercial,or not-for-profit sectors.