Ajit Kumar,Pulak M Pandey

Department of Mechanical Engineering,Indian Institute of Technology,Delhi,New Delhi 110016,India

Abstract The present work investigates the effect of ultrasonic power(%)and the time of ultrasonic vibration on the sintered density and ultimate compressive strength(UCS)of Mg15Nb3Zn1Ca fabricated using ultrasonic assisted conventional sintering(UACS).The customized UACS setup was designed and manufactured to conduct the experimentations.A customized ultrasonic stepped horn assembly was used for providing vibrations to the sample during sintering.Further,to evaluate the efficacy of ultrasonic vibration parameters,the developed setup was used to sinter Mg15Nb3Zn1Ca composite material.The study unveiled an increased sintered density and UCS of the fabricated sample by the increase in ultrasonic power(%).Moreover,a decrease in sintered density and UCS was observed with an increase in the time of ultrasonic vibration beyond a certain limit.Samples sintered with the assistance of ultrasonic vibration at 100% ultrasonic power,and 20 min of ultrasonic vibration resulted in a sintered density of 1.928±0.062 g·cm?3 and UCS of 234.9±12.3MPa.The obtained mechanical properties of the fabricated sample were comparable to the properties of cortical bone.The surface morphology and elemental compositions of samples fabricated using UACS declared a fair dispersion of reinforcement in the matrix containing merely the source elements.The results of corrosion test have showed that the assistance of ultrasonic vibration suppressed the degradation behaviour of the sintered sample after performing electrochemical study of samples using 3-electorde cell voltammetry.Mg15Nb3Zn1Ca fabricated using UACS showed a 50.18%and 9.08% of reduction in corrosion rate over conventionally sintered pure Mg and Mg15Nb3Zn1Ca respectively.In addition,electrochemical impedance spectroscopy(EIS)results indicated an enhanced corrosion resistance of the Mg15Nb3Zn1Ca composite material when fabricated at 100% of ultrasonic power with 20 min of vibration time.Apart from that,electrochemical equivalent circuits also resulted in good fitting of the experimental data obtained from EIS.

Keywords: Ultrasonic assisted sintering;Magnesium;Corrosion;EIS;Mechanical properties.

1.Introduction

In recent decades,biodegradable materials,appearing as“smart”implantable material,got significant attention from researchers in the field of clinical applications such as trauma and orthopaedic surgery.The major impelling force which initiates to develop the biodegradable implant is their degradation behaviour in biological fluids.Biodegradable material is a class of materials that possesses a superior characteristic to act as a substitute for long-term implantable material.It may degrade in the physiological environment after serving the successful tissue healing process.The biodegradable property of implants will minimize the burden of additional expenses,treatment duration,and pain from the patients by eliminating post-surgery requirements to remove the implant next to the adequate healing of tissue [1,2].

Amongst highly studied biodegradable metallic material,namely Magnesium(Mg),Zinc(Zn),and Iron(Fe),Mg and its alloys have notable perks over the other two materials.The study of Zn and Fe based alloys as biodegradable implant material is limited to a few groups worldwide [3–7].The significant mismatching of mechanical properties and young’s modulus of Fe and Zn compared to the bone causes stress shielding effects and also results in a reduction of density and thinning of bone [8].Moreover,in contrast to renowned metallic bio-inert materials(Titanium(Ti),stainless steel(SS),cobalt(Co)and chromium(Cr))and their alloys,the corrosion product of Mg and its alloys exhibit the unique features of being nontoxic and biologically absorbable [9,10].In addition to these features,the purpose of accelerating the progress of fresh bone tissue [11–13]also boosts their candidacy for implantable materials [14].However,the excessively corrosive nature of Mg in the biological fluids results in their rapid degradation rate and deteriorates the mechanical integrity,which hinders its clinical applications [13,15].So,the necessity arises to develop a novel Mg alloys/composite or to make existing Mg alloys/composite suitable for the clinical use by diminishing their corrosive behaviour in the biological fluids.

Till date,several efforts have been given to improve the mechanical and anti-corrosion properties of Mg and its alloys/ composite,and it continues still either by addressing novel alloys/composite [16–21]or by implementing a new fabrication strategy on the existing/ novel Mg and its alloys/composite.Apart from the research in elemental alloying and fabrication strategy,surface modification is another approach to increase the anti-corrosion properties of magnesium alloys.Significant studies [22–24]have reported the advancement in surface modification on biodegradable Mg alloys/composites which demonstrated improved anti-corrosive properties of the selected Mg without scarifying its bioactivity and biocompatibility features.

Aydin et al.[25]used the powder metallurgy technique to fabricate B4C reinforced Mg matrix composite.Subsequently,the microstructure,wear,and mechanical properties of the fabricated sample were investigated.They have found a uniform distribution of B4C in the magnesium matrix.Moreover,increasing of B4C content to matrix led to significant improvement in the hardness and compressive yield strength of the fabricated sample.Turan et al.[26]fabricated carbonaceous reinforced AZ91 matrix composites using powder metallurgy and found that the incorporation of carbonaceous reinforcement significantly improved the hardness of the fabricated composite.In another work,Aydin et al.[27]used hot-press sintering to synthesized AZ91/TiC composite.They have reported that the addition of TiC to AZ91 matrix considerably yielded in the improvement of hardness as well as wear resistance of the composite material.

In our prior work [28],a novel Mg-based biomaterial,namely Mg15Nb3Zn1Ca,has been developed by conventional sintering route.A significant improvement in the anti-corrosive behaviour and mechanical properties of Mg15Nb3Zn1Ca was obtained over other existing Mg and its alloys.In the past few years,ultrasonic applications have been extensively employed in most of the manufacturing areas,such as ultrasonic-assisted drilling [29],ultrasonic-assisted EDM process[30],ultrasonic welding[31],ultrasonic-assisted turning [32],ultrasonic-assisted milling [33],ultrasonicassisted finishing [34],ultrasonic sintering [35],etc.However,the major challenge that researchers generally face is the design of a horn that could effectively transmit the vibration energy at the end of the horn.Transmission of vibration energy from one end of the horn to the other end is only possible when the horn is in resonance with the transducer[36].This needs a proper attention to design and manufacturing an effective horn.The material property plays a vital role in horn design,and it is preferred to use a material that possesses low acoustic losses and high fatigue strength.Different metallic materials have been used to manufacture the ultrasonic horn.The most common materials for horn fabrication are Ti6Al4V,SS304/304L,Al2004/7075,Monel alloys,etc.as these materials provide suitable acoustic property as well as fatigue strength [36].In the current work,as per cost concern and temperature sustainability,SS304 was taken as horn material as it is cheaper than both Ti6Al4V and Monel alloys material and can sustain a maximum temperature of sintering,i.e.,640 °C.Other reports [37,38]have also suggested to select stainless steel as the horn material for ultrasonic sintering.Additionally,the shape of the horn also plays an utmost important role in amplifying the input amplification.The crosssection of sonotrodes for ultrasonic machining or ultrasonic assisted machining has mostly circular shape.Different horn shapes like straight,exponential,catenoidal and stepped are used for transmitting ultrasonic vibration energy.Whereas to provide the maximum gain over other shaped horn,selection of stepped shaped horn is recommended in the prior report[35].In this research wok,the shape and the material chosen for the horn are cylindrical shaped stepped and SS304,respectively.

Only limited work has been carried to evaluate the effect of ultrasonic vibration assistance on the sintering process.In a study reported by Abedini et al.[39]the effect of ultrasonic vibration on hot powder pressing of Ti6Al4V alloy has been investigated.Their experimental results have showed that the assistance of ultrasonic-assisted in hot powder pressing imparts better mechanical properties as compared to conventional sintering.Similarly,the ultrasonic-assisted sintering of Cu-CNT by hot isostatic pressing demonstrated the improved relative density and mechanical properties over conventional sintering [40].

Wei et al.[41]applied the ultrasonic vibrations using stainless steel horn to execute the solid-state reaction between Fe2O3and CaO to form CaFe2O4(CF).The incorporation of ultrasonic vibration showed the formation of CaFe2O4(CF)even at a lower temperature than conventional sintering.Abedini et al.[42]incorporated the transmission of ultrasonic vibration to the AA1100 aluminium powder,which was already under the hot environment along with the additional pressure of 20 MPa.It was reported that ultrasonic vibration during hot pressing led to an improvement in the relative density of AA1100 aluminium over hot pressing.Other research has been conducted in the same direction by Feng et al.[43],where AlN ceramic powders were synthesized using a hot pressing sintering system assisted by vibration.The provision of vibration during hot processing of AlN powders showed better properties than traditional hot pressing techniques.Recently,Singh et al.[35]have consolidated the loose copper powders with the help of ultrasonic assisted pressure less sintering.Authors have reported that the incorporation of vibration to the powders during sintering improved the distribution of the particles and led to a reduced porosity.Thereby producing a significant improvement in the relative density of the sintered part.

It is evident from the available reports that as of now,no one has implemented the ultrasonic technique to analyse the effect of ultrasonic power and time of vibration on the mechanical and corrosive properties of Mg-based composites.Moreover,no work is reported on the sintering of the Mgbased green compact assisted by ultrasonic vibration.Therefore,the current work presents an attempt to further improve the properties of developed Mg-based biomaterial by transmitting the vibrational energy to the sample during sintering.A transducer connected to the ultrasonic horn was used to transmit the vibrational energy to the sample.The prior work used conventional sintering(CS)to fabricate the compacted sample of Mg15Nb3Zn1Ca.In contrast,the current work shows a first effort by any researcher to sinter a compact sample using ultrasonic assisted conventional sintering(UACS).The present work also emphasizes on sintering of Mg15Nb3Zn1Ca compact using UACS to get better mechanical and corrosive properties over prior work.

2.Materials and method

2.1.Materials

Stainless steel(SS)304 was chosen as horn material for propagating the ultrasonic vibration.The specimen holder used to hold the sample was made of the same material as that of the horn,whereas a green compacts of Mg15Nb3Zn1Ca was used as sample material.The coolant jacket,to circulate the water over the surface of the horn was made of aluminium.Stud,which was serve to connect mould to the horn was made of the same material as that of the horn.The average size of powder particles(such as Mg,Ca,Zn,and Nb)used for blending was in the range of 45 μm with the following composition(in weight fraction):81%Mg,3%Zn,1%Ca,and 15% Nb.

2.2.Development of experimental setup

Prior to performing the experimentations,the design and validation of horn-holder assembly were needed to successfully conduct the ultrasonic assisted conventional sintering experiments.This section provides detailed information behind selecting a specific stepped horn and specimen holder and the finite element(FE)analysis of the horn-holder assembly.The FE analysis of designed stepped horn-holder assembly was carried out using ANSYS(v2015)environment to calculate its natural frequency and working amplitude of vibration.

2.2.1.Dimensions of stepped horn and specimen holder

The stepped circular horn was designed by selecting the inner diameter(D1)as 32mm,which was reduced to 16 mm at the outer diameter(D2)from the middle of the horn.The inner diameter was opted based on the diameter of the transducer to cover the horn from one end for minimizing the loss of energy.These diameters were selected to amplify the vibration at the end of outer diameter by four times as per stepped horn amplification ratio/gain,which is[44].The stepped horn attached with a customized specimen holder was designed and fabricated to amplify the vibrational energy received from the transducer end(input)(～5 μm)to the sample end(output)of the horn.The step by step drafting of stepped horn-specimen holder assembly is presented in Fig.1.Initially,the length of the horn was calculated by the analytical method.Subsequently,a FEM simulation was executed to make the necessary changes in the horn design for achieving the required performance of the stepped horn.Subsequently,a specimen holder with suitable dimensions was tuned to get the required functioning of horn-specimen holder assembly.Next to simulation,the stepped horn,specimen holder,and other essential fittings of UACS set up were manufactured according to design requirements.Usually,half wavelength(λ/2)horn consisted of a single nodal point that was designed,which transforms its cross-section at the length ofλ/4 to amplify the input amplitude of vibration.As per the specific requirement,a horn with multi-nodes(n=number of nodes)point could be designed by increasing the overall horn length[45,46].A specimen holder of rectangular shape havingλ/2 length was designed and developed to mount on the output end of the stepped horn.The overall length of the stepped horn was calculated by selecting five nodal points as per the analytical equation mentioned below.

Where,Lis the overall length of horn;nis the number of nodal points=5;crefers to sound velocity in the selected material(SS304)=4912 m/s;fis the natural frequency=20 kHz.

So,

A modal analysis was further performed to fine-tune the dimensions of the horn and specimen holder so that it can have a longitudinal mode shape at the natural frequency of 20±0.5 kHz.The fine-tuning will allow the designed horn to have maximum amplitude gain at 20±0.5 kHz.The constraint for opting the 20±0.5 kHz frequency is due to an electric generator(which is used as the power supply for the ultrasonic system),which can only produce the mentioned frequency range.The final dimension of the designed horn with specimen holder and their modal analysis results have been shown in Fig.1(a)(CAD model)and Fig.3,respectively.The specimen holder’s pictorial presentation with their dimensions has been depicted in Fig.1(b).Based on the analysis outcome,the total length of the designed set up(stepped horn+specimen holder)can be written as

Fig.1.(a)Design of rescaled horn along with customized specimen holder and(b)pictorial presentation of the specimen holder.

Where,L′is the overall length of the horn and specimen holder assembly.

2.2.2.Modal analysis of the horn-holder assembly

The modal analysis was carried out at a specific ultrasonic frequency range to observe the response of horn,i.e.,mode shapes and natural frequency at each step.The mesh independence analysis was carried out to determine the least simulation time in respect of the element size.Tetragonal element shape was selected for the analysis as it is fit for this structural analysis to grid the complex 3D geometry without difficulty [47].It is clear from Fig.2(a)that solution converges at element size 4 and 5.Beyond this element size,the solution gives erroneous results as the increase of element size decrease the approximation values.The element size of 4 mm showed the optimum value owing to the least simulation time at the required natural frequency(refer to Fig.2a).To run the simulation,the specimen holder was treated as a single solid body,thereby showed a longitudinal propagation of the ultrasonic wave,as shown in Fig.2(b).The assembled horn-specimen holder configuration confirmed a longitudinal mode of vibration at the natural frequency of 20,831Hz,as shown in Fig.3.Different frequencies were generated at each mode.The desired mode,at which horn assembly generated a linear mode shape vibration,was the 47th mode.The different frequencies and their corresponding modes have been shown in Fig.4(a-e).

2.2.3.Harmonic analysis of the horn-holder assembly

Harmonic analysis of horn-specimen holder was performed based on the obtained natural frequency from the modal analysis.In Harmonic analysis,～5 μm was given(as the amplitude of vibration)at the input end of the horn in order to approximately evaluate the expected amplitude at the output end(at 20±0.5 kHz).Based on the harmonic analysis of the whole structure(horn and specimen holder assembly;Fig.5),it is clear that a maximum amplitude of 15.358 μm is obtained near the end of a smaller cross-section of the horn,which is further transmitted to the specimen.The obtained amplitude showed a substantial effect on the improvement of density and UCS of sintered samples after experimentation.The loss in amplitude could be due to the presence of friction/damping between the horn and specimen holder,as both of them were attached through the screw and thread mechanism.From Fig.6,it could be declared that the equivalent stress(100.24 MPa)generated in the horn-holder assembly is lower than the yield stress of SS304(205 MPa),which indicated that the designed horn-holder assembly is safe and suitable to conduct the experiments.

Fig.2.Illustration of(a)mesh independency analysis and(b)longitudinal propagation of the ultrasonic wave in simulated horn.

Fig.3.Modal analysis result of stepped horn assembled with specimen holder.

3.Fabrication of experimental setup for UACS

A schematic diagram and pictorial presentation of the customized UACS setup are depicted in Fig.7(a)and(b),respectively.It consisted of the ceramic tube furnace,stepped horn-specimen holder assembly,an ultrasonic generator connected with piezoelectric crystal transducer,inert gas inletoutlet connections,and a cooling system.The cooling system comprised of water chiller along with aluminium jacket which surrounded the horn and allowed circulation of chilled water over the horn surface to dissipate the heat generation.The customized specimen holder was having a cylindrical cavity of dimension ～?10 mm×15 mm and a small hole at the bottom(refer to Fig.7).The cavity was used to hold the sample firmly where hole at the bottom was provided to extract the sample from cavity after the completion of the sintering process.The specimen holder was assembled rigidly with the output end of the stepped horn.The compact cylindrical sample of Mg15Nb3Zn1Ca was forcefully inserted in the holder’s cavity,and the vibrational amplitude of the horn was examined using a laser vibrometer(model:Polytech OFV 303).A bulky metallic stand(S.No.9 of Fig.7a)was used to clamp the stepped horn and to provide similar experimental settings such that no motion in lateral as well rotary direction could occur.To evaluate the natural frequency of the horn-holder assembly,the impact hammer test was executed.The vibration at the end of the specimen holder was recorded with the help of a Polytec OFV 303 laser vibrometer.The measured resonance frequency was 20.45 kHz and confirmed that the horn-holder assembly oscillated near the simulated longitudinal frequency value.

Fig.4.Illustration of different mode shape for the horn assembly using modal analysis(??required frequency).

Fig.5.Harmonic analysis results of stepped horn assembled with specimen holder.

Fig.6.Illustration of stress generated in the stepped horn-specimen holder assembly.

4.Experimentation procedure

As the name suggests,ultrasonic assisted conventional sintering,therefore,the implementation of ultrasonic vibration at the sample during conventional sintering was attempted.Firstly,in the author’s previous work [28],the conventional sintering parameters were optimized,and the sample sintered at the optimized parameters showed significantly improved results over pure Mg.Now,this work was supposed to implement the ultrasonic vibration on the sample by selecting all optimized sintering parameters from the prior work.Authors obtained the green compact cylindrical samples(?10 × 15 mm2)by following the similar procedures which were used in their other work [28,48].In the present work,the experiments were started by inserting the sample in the cavity of specimen holder then placing the horn-holder assembly inside the hollow ceramic tube of the furnace.Subsequently,both ends of the tube were sealed with the caps.A couple of holes were provided on one of the cap to insert both the gas inlet lid,and the output end of the horn,whereas another cap contains a single hole for the gas outlet(refer to Fig.7a).Prior to sintering,purging was done to remove the available gases from the tube.The optimized heating rate=25 °C/min,and sintering temperature=640 °C were programmed in the high temperature furnace.Different time of vibration were used for the samples at constant ultrasonic power(%)with the sintering temperature of 640 °C.Similarly,at the same sintering temperature,i.e.,640 °C,individual ultrasonic power was supplied for a fixed time of vibration.After accomplishing the required sintering operation,the sample was immediately pulled out from the hot environment similar to the prior work to get increased cooling rate.The effect of different ultrasonic conditions such as ultrasonic power(%)and time of vibration on the physical,mechanical and degradation properties of the sintered sample have been studied.The optimum conditions of both ultrasonic variables,i.e.,power(%)and time of vibration,were simultaneously applied on the Mg15Nb3Zn1Ca compact to achieve the optimized result.In order to avoid any damage in ultrasonic assisted sintering setup due to the rise in temperature,a provision of continuous circulation of chilled water(～9 °C)over the horn’s surface was given to dissipate the heat generated in the ultrasonic horn.

5.Characterizations and testing procedure

5.1.Material characterization

The surface morphology and elemental configuration of fabricated samples were observed by scanning electron microscopy(SEM)and energy dispersive X-ray spectrometer(EDX)analysis.The instruments,namely TM3000 table top scanning electron microscope(Hitachi;USA)equipped with swifted 3000EDX software(Oxford Instruments;France)was used to carry out the SEM and EDX analysis.An optical microscope(Leica ICC50 HD;Germany)was used to capture the microstructure images of Mg15Nb3Zn1Ca.The samples were gradually polished up to 4000 grit using SiC abrasive paper,and ultimately the final touch was given by polishing the samples using colloidal silica(MasterPolish suspension,Buehler,USA).Subsequently,to unveil the microstructure,the polished surface was immediately immersed in the 10% Nital solution for 5–7 s [49,50].Furthermore,the XRD analysis was done on the corrosion product of the sample obtained after accomplishing its corrosion study.The XRD spectra were obtained by Rigaku Ultima IV having CuKα(λ=1.5406 ?A)monochromatic radiation source operated at 15 mA and 40 kV.The analysis was done with a scanning speed of 2° per minute over a range of 2θfrom 30 to 80°The step size used throughout the scan was 0.1 μm.

5.2.Determination of ultimate compressive strength and density of Mg15Nb3Zn1Ca

In order to obtain the ultimate compressive strength(UCS)of sintered samples,the compression test was executed using a 100 kN universal testing machine(Instron 5582;USA).A set of three samples were tested for repeatability of the results.Each sample with the dimension of ?10 mm × 15 mm was prepared as per ISO 13,314 [51].The uniaxial compression test was conducted in quasi-static mode with a crosshead speed of 1 mm per minute.

The density of samples was obtained in distilled water at the ambient condition as per Archimede’s principle based on ASTM B962–17 [52].The sintered density of Mg15Nb3Zn1Ca was recorded by weighing the sample in air and then in water with the help of electronic balance having accuracy of±0.1 mg.The apparatus(GR-202,make-A&D,United Kingdom)was used to record the experimental density of sintered samples.

Fig.7.(a)Schematic diagram of ultrasonic assisted conventional sintering experimental setup(b)Pictorial presentation of ultrasonic assisted conventional sintering experimental setup.

5.3.Electrochemical measurements

A potentiodynamic polarization test was conducted to obtain the corrosion rate of the samples in the simulated body fluid(SBF)solution.The preparation of the SBF solution was done as per the description of Kokubo [53].Table 1 depicts the reagents with their corresponding composition used to prepare 1L SBF solution.A 3 electrode cell configuration was chosen to perform the electrochemical test.The electrodes namely reference electrode,a counter electrode,and working electrode were respectively assigned to saturated calomel electrode(SCE),graphite rod,and test sample.The one end of Mg15Nb3Zn1Ca sample,which was supposed to expose to SBF was kept open and moulded in the epoxy resin.The sample was made as a working electrode by connecting copper(Cu)wire to the opposite end of the sample.The opened surface of the sample was polished until the grit size of 3000 using SiC abrasive paper.The potentiodynamic polarization and electrochemical impedance spectroscopy(EIS)measurements were performed after stabilization of open circuit potential(OCP).A 20 min of a time span was observed to stabilize the OCP for the test sample.The potential range of±200 mV with reference to OCP was applied while performing the polarization measurements.The EIS measurements were done at 10 mV perturbation with a sweeping rate of 0.5 mV/s in the frequency range of 100,000 to 0.01 Hz.

Table 1 Reagents used to prepare simulated body fluid(1L,pH=7.4,37±0.5 °C).

5.4.Immersion corrosion tests

As per ASTM G31–72,[54]the static immersion test of the prepared samples was performed in the SBF solution.The immersion test configuration(SBF containing dipped sample)was maintained at 37°C temperature during the process of the immersion test.The testing configuration was kept inside the incubator to maintain its required temperature.The samples with 5 mm×? 10 mm as shown in Fig.1a was taken for immersion test.Prior to the test,weight and surface area of all the samples were measured.Three set of all samples was taken for immersion measurement.The samples were then immersed in SBF solution(150 ml)for 2,4,8,and 12 h,respectively,to estimate the pH variation and weight loss.The weight loss of the samples were obtained after cleaning the samples in an ultrasonic bath with acetone followed by the distilled water.Then the samples were allowed to get dried at the ambient temperature.Subsequently,the samples were rinsed with chromic acid remove the corrosion products from the samples.Lastly,the sample was kept in the vacuum oven for 1 day to acquire complete drying of the sample.Finally,the weight loss of the sample was estimated with the help of weighing scale.The pH value of the sample was regularly monitored using pH metre(Model:pH tester 30,accuracy 0.01).

6.Results and discussion

This section addresses the physical,mechanical,and corrosion properties of ultrasonically assisted sintered samples prepared under the influence of different ultrasonic power(%)and time of vibration.Also,the present part discusses the fractured behaviour of ultrasonic assisted conventionally sintered samples.

6.1.Effect of time of vibration on sintered density and UCS

Fig.8(a)illustrates the compressive engineering stressstrain behaviour of Mg15Nb3Zn1Ca fabricated at different time of vibration.Whereas the effects of the time of vibration on UCS and sintered density of Mg15Nb3Zn1Ca have been shown in Fig.8(b).The sintered density and UCS of samples found to be increased by increasing the time of vibration till 20 min.The initial improvement in both sintered density and UCS could be attributed to particles neck growth as well as the reorientation and movement of particles within the hot matrix.A combination of increased holding time and time of vibration for the samples may inferred to the formation of large neck growth as well as reorientation and movement of particles with the matrix.Neck growth of particles increased as the holding time increased owing to the absorption of heat for an extended time [55].Moreover,the ultrasonic vibration provided to the sample driven the reorientation and movement of particles by minimizing the friction between inter-particles[56].As the inter-particles friction minimized,the significant movement of particles may have occurred to fill the available pores within the matrix,thereby led to the formation of strong inter-particles bonds.Therefore,the improvement in the density of Mg15Nb3Zn1Ca compact was expected due to the initiation of larger neck growth as well as reduction in the friction between the particles.The reduction of porosity level and the formation of a strong bond in the sintered sample correspondingly resulted in improvement in sintered density and UCS of the samples.

Fig.8.(a)Engineering stress-strain diagram of Mg15Nb3Zn1Ca fabricated at different time of vibration and(b)Effect of time of vibration on the UCS and sintered density of the Mg15Nb3Zn1Ca sintered at 640 °C with 25 °C/min at the ultrasonic power of 100%.

Fig.9.SEM images(2000X)of sample sintered with 100% ultrasonic power at different time of vibration(a)0 min(b)20 min(c)30 min and(b)60 min.

Furthermore,a reduction in the results of both sintered density and UCS were found by increasing the vibration time beyond 20 min,as can be seen in Fig.8(b).This could probably be due to the absorption of heat energy by the sample for a relatively prolonged duration,which led to the softening of the matrix.In addition to this,ultrasonic vibration further aid the softening of the matrix by increasing the temperature of the sample.Moreover,the inter-particles friction may be excessively reduced by applying the ultrasonic vibration for a longer time,which turned in the settling of heavier particles to the bottom.The settling down of particles resulted in an agglomeration of particles by leaving some random pores inside the matrix,which yielded in the reduction of both sintered density and the UCS.Fig.9 showed the corresponding SEM images of sample fabricated using UACS by applying 0 min,20 min,30 min and 60 min of vibration time.It is evident from Fig.9 that sample fabricated at 0 min of time of vibration possessed only limited diffusion of particles which led to increased porosity.The diffusion particles between particles increased by allowing the sample to fabricate at relatively increased time of vibration.That is why sample fabricated at 20 min of time of vibration exhibited lesser porosity level.However,as discussed earlier increasing time of vibration beyond a certain limit allowed the sample to receive prolonged heat as well as vibration.The prolonged heat and vibration for sample correspondingly led to the softening to matrix and agglomeration of particles.The agglomeration of particles within matrix produced porosity inside the matrix by leaving some random pores inside it which can be seen in Fig.9(c-d).

6.2.Effect of the ultrasonic power(%)on sintered density and UCS

Fig.10(a)shows the compressive stress-strain curve of Mg15Nb3Zn1Ca fabricated under the influence of different ultrasonic power(%).Whereas the effect of ultrasonic power(%)on sintered density and UCS has been represented as Fig.10(b).As reported earlier [56],vibration plays a substantial role in minimizing the friction between the inter-particles and permits the reorientation and movement of particles.The increase in sintered density and UCS was obtained with the increase in ultrasonic powder(%)(refer to Fig.10(b)).The morphology of the samples sintered at 20%,60%,and 100%of ultrasonic power have been shown in Fig.11(a-c),respectively.The ultrasonic displacement amplitude increased proportionally with the increase in power(%)that resulted in higher neck growth of particles.This led to the formation of strong bonds between the particles to improve the strength of the sintered sample [55,57].The increase in ultrasonic power improved the displacement amplitude of the sample and accelerated the material flow rate during the sintering stage [55].From Fig.11,it is evident that the increase in ultrasonic power(%)led to reducing the porosity which led to the smoothing of morphology.This effect in particular,may be attributed to improvement in particles diffusion as well as a reduction in inter-particles friction,which accelerated the material flow rate during the sintering stage.Consequently,the vacant spaces within the sample were found to be nearly occupied by the expansion of the particles to result in a more compacted sample.Fig.12 shows another SEM images of sample sintered at different power(%).In this case,the green compact was containing a high porosity level as the sample was compacted at very little pressure(～20 MPa).After performing the sintering of green compact with 20% and 100% of ultrasonic power,a substantial particles diffusion with reduced porosity was obtained in the sample sintered with higher ultrasonic power(%)(Refer to Fig.12b).

Fig.10.(a)Engineering stress-strain curve of Mg15Nb3Zn1Ca fabricated at different ultrasonic power and(b)Effect of ultrasonic power on UCS and sintered density of the Mg15Nb3Zn1Ca sintered at 640 °C with 25 °C/min for 20 min time of vibration.

Fig.11.SEM images(1000X)of Mg15Nb3Zn1Ca sintered for 20 min time of vibration at ultrasonic power of(a)20%,(b)60% and(c)100%.

6.3.Fracture analysis of Mg15Nb3Zn1Ca sample prepared using UACS

Next to performing the compression test,the mode of fracture for the Mg and Mg15Nb3Zn1Ca fabricated using ultrasonic assisted conventional sintering is depicted in Fig.13(b)and(c),respectively.Splitting of Both Mg and Mg15Nb3Zn1Ca into two parts can be observed in Fig.13.The sample being uniaxilly compressed is shown in Fig.13(a).A brittle mode of fracture could be noticed in the Mg sample,whereas Mg with reinforced particulates indicated a combined brittle and shear mode of fracture.It was found that Mg15Nb3Zn1Ca sample is separated into two parts at an angle of approximately 45° to the direction of compressive loading.The particular type of fracture is owing to the presence of the simultaneous effect of brittle and shear mode.Though as per the available literature[48,58],the fracture of pure magnesium under uniaxial compression has mainly occurred due to the brittle mode.However,the failure of Mg15Nb3Zn1Ca,due to the combined brittle and shear mode could be attributed to the incorporation of blenders to the Mg matrix.The SEM images of the fractured surface for both Mg and Mg15Nb3Zn1Ca have been respectively shown in Fig.14(a)and(b).It could be elucidated from Fig.14(a)that a minor or no shear regions are visible for Mg,which refers to dominant brittle mode fracture.In contrast,significant shear regions can be seen in Fig.14(b),which indicates a combined brittle and a shear mode of fractures for Mg15Nb3Zn1Ca.The failure of samples owing to the presence of the shear effect performed a foremost character to boost of strength of such materials [59].This showed noticeable progress towards the improvement of compressive strength.

Fig.12.SEM images(5000X)of Mg15Nb3Zn1Ca(compaction pressure～20MPa)sintered at ultrasonic power(a)20% and(b)100% for 20 min time of vibration.

Fig.13.Uniaxial compression of the samples(a)before the test;and after the test(b)Mg;(c)Mg15Nb3Zn1Ca.

Fig.14.SEM images(2000X)of fractured surfaces(a)Mg and(b)Mg15Nb3Zn1Ca.

6.4.Comparison of UCS and sintered density for UACS with CS

The sample fabricated using UACS at optimum ultrasonic parameters,i.e.,20 min time of vibration and 100% of ultrasonic power,revealed a significant improvement in UCS and sintered density of Mg15Nb3Zn1Ca over conventionally sintered(optimized)sample [28].Also,the aforesaid biomaterial developed by UACS showed a comparable density and higher strength than cortical bone [60,61].Fig.15 illustrates a comparison of UCS and sintered density by bar graph for ultrasonic assisted conventionally sintered Mg15Nb3Zn1Ca with cortical bone and CS(optimized).

6.5.Characterization of Mg15Nb3Zn1Ca fabricated using UACS

Similar to the preceding work [28],no contamination residues were obtained in the XRD plot for Mg15Nb3Zn1Ca fabricated using UACS(refer to Fig.16).Besides,the XRD analysis of the corrosion sample after electrochemical testing was performed to evaluate the phase constituents of corrosion products on the surface of the sample.The detailed discussion of the XRD result for corrosion products has been presented in the forthcoming section.The EDS equipped SEM analysis of the fabricated Mg15Nb3Zn1Ca declared that the matrix composite owns mainly the source elements with minor carbon and oxygen content(refer to Fig.17).Moreover,a substantial dispersion of reinforcements within the matrix was also found and has been presented in the rightmost part of Fig.17(a).

Fig.15.Comparison of(a)UCS and(b)sintered density for Mg15Nb3Zn1Ca fabricated by UACS with cortical bone CS(optimized).

Fig.16.XRD peaks of Mg15Nb3Zn1Ca fabricated using UACS.

7.In-vitro degradation analysis of the fabricated sample

This section presents the in-vitro degradation behaviour of Mg15Nb3Zn1Ca fabricated using ultrasonic assited conventional sintering at different processing parameters.For carrying out the in-vitro degradation study,the electrochemical and immersion testing have been performed.

7.1.Potentiodynamic polarization test

The polarization curves of samples sintered at different ultrasonic power(0%,20%,40%,60%,80% and 100%)are shown in Fig.18.The curve for Mg15Nb3Zn1Ca(blue line),sintered at optimized conventional sintering parameters is also included in Fig.18 for the comparison purpose.The corresponding electrochemical data are tabulated in Table 2.It was found that the sample fabricated with 100% ultrasonic power possessed the highest corrosion potential(Ecorr)and lowest corrosion current density(icorr)amongst the group(Table 2).The lower value of icorris the indication of improvement in anti-corrosive properties of samples which drives by the generation of passivation region at the surface of the sample.Higher the passivation region refers to the movement of corrosion potential towards the positive direction,thereby indicating higher corrosion potential to result in lesser corrosion susceptibility.Pores morphology of the sintered sample predominantly plays a vital role in initiating the corrosion mechanism.As discussed earlier that the sample sintered under the influence of ultrasonic vibration showed a higher particles diffusion.It further resulted in a reduction in porosity and formation of stronger grips on the reinforced Nb particulates.In order to support the statement,microstructural images of Mg15Nb3Zn1Ca fabricated using UACS with different power(%)have been shown in Fig.19.The sample sintered with 100% ultrasonic power exhibited a finer microstructure with a minimum number of microvoids as compared to 0% ultrasonic power(refer to Fig.19).

Table 2 Corrosion parameters for ultrasonic assisted conventionally sintered Mg15Nb3Zn1Ca estimated through Tafel analysis.

Fig.17.Illustration of EDX spectra of(a)Mg15Nb3Zn1Ca sample and(b)Nb particulates distributed within Mg alloys matrix.

Fig.18.Tafel plot of Mg15Nb3Zn1Ca fabricated at different ultrasonic power using UACS and optimised conventional sintered sintering process.

According to an available report [62],the microstructure of material plays an essential role in corrosion behaviour.It has been reported that the oxide film retains increased compactness with finer microstructure and demonstrates improved anti-corrosion.According to Fig.19(b),it could be inferred that the sample sintered without the influence of ultrasonic vibration(i.e.,at 0% power intensity)possessed a large number of microvoids,thus lacking in producing a finer microstructure.The compacts having low porosity promotes pore closure during sintering leads to reduction in pores size,thereby results in more resistant to crevice corrosion attack [63].On the contrary,the sample sintered without the influence of ultrasonic vibration possessed the higher porosity and was more to prone pores/crevice corrosion.It is referred to the narrow and isolated wide-open channels in highly porous samplesto provide a large number of sites capable of trapping liquid that eventually promoted crevice corrosion.Therefore,it is established that the sample fabricated with 100% intensity of ultrasonic power exhibited higher corrosion resistance as compared to 0%.

Fig.19.Optical micrograph of Mg15Nb3Zn1Ca sintered using UACS with(a)100% ultrasonic power(less porous)and(b)0% ultrasonic power(highly porous).

Previous researchers [28,63,64]have shown that a porous specimen is more sensitive to local corrosion when compared with the sample without pores owing to the larger true surface area than the nominal surface area.Hence,a porous specimen would suffer more corrosion than a less porous one.

7.2.Immersion test

A significant variations in the corrosion behaviour of the samples fabricated at 0%,60% and 100% ultrasonic power were obtained during in vitro static immersion test(refer to Fig.20).The foremost factor of these variations was implementation of increased ultrasonic power during UACS of Mg15Nb3Zn1Ca.The pH deviations of the samples relative to immersion time is depicted in Fig.20b.At the 12 h of immersion time,the maximum pH value for all the samples was recorded in the range of 8.32–8.62.After removal of corrosion product from the surface of samples,which were fabricated at 0%,60% and 100% ultrasonic power,the corresponding weight loss of samples are shown in Fig.20a.It is clear from the data,sample fabricated at increased power(%)led to reduced weight loss whereas in contrast,the increased weight loss was observed for the sample fabricated at lower ultrasonic power.This could be owing to the presence of relatively more microvoids on the sample fabricated at lower ultrasonic power(refer to Figs.11 and 19).These microvoids significantly activated the penetration of the corrosive fluid in the composite samples and allowed to initiate crevice corrosion attack,thereby increasing the corrosion rate of the composite samples fabricated at 0% ultrasonic power.Aydin et al.[27]reported that the presence of reinforcement particulates within the matrix accelerates the corrosion rate as existed particulates act as cathode and matrix acts as anode.As both dissimilar metals(Nb and Mg)possesses different electrode potential,the existence of potential difference between reinforcements i.e.Nb and Mg matrix is the driving force for the destructive attack on the anode(Mg matrix).Generally,the matrix which acts as anode corrodes more as compared to cathode when current flows through the conductive medium.

Fig.20.Static immersion test results of Mg15Nb3Zn1Ca fabricated at 0%,60% and 100% ultrasonic power for(a)weight loss and(b)pH variation.

Fig.21.SEM images(1000X)of the corroded samples after 4 h and 12 h of immersion time for(a),(b)0%,(b),(c)60% and(e),(f)100% ultrasonic power.

Apart from crevice corrosion,the galvanic corrosion significantly accelerated the corrosion rate of the sample fabricated at 0% ultrasonic power.This is owing to the intrusion of SBF into the matrix,which allowed the SBF to interact with more reinforced particulates(Nb)and produce more galvanic corrosion that further resulted in increased corrosion rate.The simultaneous effect of crevice corrosion and galvanic corrosion led to the higher corrosion rate for the sample fabricated with 0% ultrasonic power.The morphology of the corroded surfaces of Mg15Nb3Zn1Ca obtained after immersion time of 4 h and 12 h is shown in Fig.20.It is evident that microvoids of sample fabricated at 0% ultrasonic power allowed the SBF to enter inside the matrix and generated adequate number of pits(refer Fig.21a)which later expedited the degradation of matrix(refer Fig.21b)and led to increased corrosion rate by combined influence of crevice and galvanic corrosion.The reduction of microvoids was observed by increasing the ultrasonic power(refer to Fig.11).The reduced microvoids tends to improved compactness of the sample,which exhibit closure of pores.These types of pores do not allow the fluids to enter within the matrix and thereby results in more resistant to crevice corrosion attack.No pits were found on the sample fabricated at 100% of ultrasonic power even after 4 h of immersion time(refer Fig.21e)and owing to this no severe degradation in the sample was observed even after 12 h of immersion time as can be seen in Fig.21(f).The degradation in the Mg15Nb3Zn1Ca fabricated at 100% ultrasonic power was merely susceptible due to galvanic corrosion.Also,the formation of highly stable oxide layers(Nb2O5,Ca10(PO4)6(OH)2)on the surface of sample(refer Fig.20)fabricated at 100%ultrasonic power may be the reason for reduced degradation of the sample.

Fig.22.XRD spectra of corrosion products of corroded Mg15Nb3Zn1Ca.

7.3.Characterization of corrosion products for Mg15Nb3Zn1Ca fabricated using UACS

The XRD analysis of the corroded sample after electrochemical testing was performed to analyse the formation of corrosion products on the sample’s surface.Fig.22 depicts the XRD pattern of corrosion products for Mg15Nb3Zn1Ca after accomplishing electrochemical measurements in the SBF solution.The red and black colours of XRD patterns in Fig.22 correspond to the sample fabricated at 100% and 0% ultrasonic power(%)during UACS.It is evident that the surface of the corroded sample was mainly consisted of Mg(OH)2,Mg,Ca10(PO4)6(OH)2,CaCO3,and Nb2O5.It has been reported in a prior study [65]that the oxide film of magnesium presents loose and porous features.In contrast,an oxide film of rare-earth elements demonstrates relatively dense and chemically stable characteristics.Sample fabricated at 100% ultrasonic power showed a noticeable formation of Nb2O5on its surface by immersing it into the SBF solution and depicted a new protective layer of corrosion products on the surfaces of Mg matrix composite which imparted high corrosion resistance to the composite.According to the XRD pattern,it could be inferred that Mg15Nb3Zn1Ca fabricated at 100% ultrasonic power possessed a highly protective layer owing to the significant formation of niobium oxide on the surface.This is attributed to the highly chemically stable nature of Nb2O5,as well as its outstanding anti-corrosion properties over Mg oxide/ hydroxide.Moreover,as the fluid came in contact with the highly dense surface,the dense structure of the sample surface did not allow the SBF to enter inside the matrix due to the formation of a cumulative oxide layer on the sample surface.However,a reverse corrosion behaviour was noticed for the sample fabricated at 0% ultrasonic power,which exhibited a lower corrosion resistance in comparison to the sample fabricated at 100% ultrasonic power.The main reason,in particular,for this behaviour was the initiation of crevice corrosion due to the entry of fluids inside the matrix.A marginal decrease in corrosion resistance of conventionally sintered Mg15Nb3Zn1Ca was attributed to the presence of relatively higher porosity than the sample fabricated using UACS with 100% ultrasonic power.

XRD analysis of the corrosion products did not reveal any phase that could create complications to the host tissues.Moreover,in the corrosion product,a bone-like apatite Ca10(PO4)6(OH)2was detected on the sample surface,which supports the proliferation of osteoblasts [66].The formation of Ca10(PO4)6(OH)2on the surface of sample during immersion supports the restoration of the bone tissues.Moreover,the surface of Mg15Nb3Zn1Ca gets a temporary protection by the deposition of corrosion products.

7.4.Electrochemical impedance spectroscopy and equivalent electrical circuit fitting

Fig.23 depicts the fitting results of the electrochemical impedance spectroscopy(EIS)data.Fig.23(a)shows the Nyquist diagram,whereas Fig.23(b-d)show the combined bode diagrams for the magnitude of impedance and phase angle.All the samples represented a similar trend with highfrequency capacitive loop,medium frequency capacitive loop,and low-frequency inductive loop.Nevertheless,the distinct difference still existed amongst all the samples.It could be observed from the Nyquist diagram that the sample fabricated under the influence of 100% ultrasonic power presented a larger diameter of semicircle suggesting a higher corrosion resistance as compared to the samples fabricated under the influence of lower ultrasonic power.The cause may be attributed to the formation of a relatively strong protective film on the surface of the sample fabricated at higher ultrasonic power(refer to red colour XRD pattern of Fig.22).As discussed earlier that UACS assists in forming highly dense product,which hinders the fluid to enter into the matrix and thus reduced the corrosion activity.Moreover,a highly dense structure could form a strong oxide layer on the surface as the SBF used to come in contact with the exposed surface.The oxide layers which probably contain the niobium oxide(Nb2O5)and magnesium hydroxide(Mg(OH)2)was strong enough to protect the surface from the corrosion.Amongst the protective layers,Nb2O5played an important role as it is almost five times stronger than Mg(OH)2in the aqueous solutions.A marginal decrease in corrosion resistance of optimized conventionally sintered Mg15Nb3Zn1Ca was attributed to the presence of relatively higher porosity than the sample fabricated using UACS(100% ultrasonic power).However,the same material demonstrated a relatively poor corrosion resistance when fabricated at 0% ultrasonic power.This is owing to the presence of highly porous structure on the surface of sample which allowed the SBF to enter inside the matrix and prompted both galvanic and crevice corrosion attack.

Fig.23.EIS spectra in the form of(a)Nyquist and(b),(c),and(d)Bode diagrams for magnitude of impedance and phase angle.

Fig.23.Continued

The results obtained from EIS are represented in the form of Nyquist and Bode plots.To support the results of Nyquist diagram,bode plots are shown in Fig.23(b-d).The inference of the Bode plots is similar to that of the Nyquist plots.Therefore,the Bode plots for only 0% and 100% ultrasonic power have been shown.Additionally,for comparison purpose Bode plot for the sample fabricated at optimised CS parameters has also been shown in Fig.23(b-d).As per bode plot presented in Fig.23(b-d),it could also be observed that Mg15Nb3Zn1Ca fabricated using UACS with 100% ultrasonic power possessed increased impedance modulus(|Z|)in the low-frequency region.This trend of bode plot indicated that Mg15Nb3Zn1Ca fabricated using UACS(100% ultrasonic power)has improved anti-corrosive feature than other two [67].To elucidate the corrosion mechanisms of selected composite material,the plots of phase angle are also presented in Fig.23(b-d).The surface condition as well as structural transformation can be seen by the plots of phase angle which indicate a capacitive nature over the frequency range [68].The shifting of phase angle towards more negative directions at low frequency range(～10?2–1Hz),signifies the formation of a passive films on the surface exposed to SBF which effectively inhibit the corrosion by adopting UACS(100% ultrasonic power).The equivalent circuit models for Mg15Nb3Zn1Ca sintered at 0% and 100% ultrasonic power is depicted in Fig.24(a)and(b)respectively.

The values of the elements in the equivalent circuits obtained from fitting are given in Table 3.The fitting of experimental data was executed with the help of Z view software.In the equivalent circuit diagram,Rsand Rfrefer to the solution and film resistance respectively,CPE corresponds to the constant phase element comprised of a capacitance(assigned as CT)and associated phase angle(assigned as CP).R0indicates the solution resistance in the corrosion pits,Rtcorresponds to the charge transfer resistance and L is the associated inductive element.The chi-squared value indicated the square of the standard deviation between the original data and the fitting results.It is clear that Mg15Nb3Zn1Ca(100% ultrasonic power),with a large arc diameter,has a higher value of Rtand Rf,implying the composites show the best corrosion resistance.The results of the EIS fitting showed an excellent agreement with potentiodynamic polarization results.

Fig.24.Equivalent circuits used to fit EIS experimental data for Mg15Nb3Zn1Ca sintered using(a)UACS(0% ultrasonic power)and(b)for both CS(optimized)and UACS(100% ultrasonic power).

Table 3 Parameters of the circuit model fitted from the EIS experimental results.

8.Conclusions

i.For the very first time,ultrasonic assisted conventional sintering(UACS)was adopted to sinter green compact samples without applying external pressure during sintering.The incorporation of ultrasonic vibration during conventional sintering inferred an improved sintered density,UCS,and anti-corrosive characteristics of sintered Mg15Nb3Zn1Ca over conventional sintering(CS),which was recently used to develop the same material.

ii.A reduction in the porosity of the sintered sample was observed by increasing ultrasonic power(%).The sintered sample comprised of reduced porosity resulted in improved sintered density and UCS.Also,increasing the time of vibration during sintering up to a certain extent(i.e.,20 min)resulted in minimization of porosity,while beyond the aforesaid limit,the porosity was obtained to be increased.

iii.At an optimum set of ultrasonic condition(100% ultrasonic power;20 min time of vibration),UACS of Mg15Nb3Zn1Ca demonstrated improved properties(Sintered density=1.928 g·cm?3,UCS=234.9 MPa,Corrosion rate=1.093 mmpy)over the properties Mg15Nb3Zn1Ca sintered using optimized CS.

iv.EIS studies revealed that samples with reduced porosity demonstrated a highly stable oxide film(as per XRD analysis),thereby resulted in improved anticorrosion properties.Moreover,the chemical composition of corrosion products exhibited a comparable composition to that of human bones and was favourable for the recovery of bone tissues.

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 authors are grateful for the financial supports provided by the Department of Science and Technology-Science and Engineering Research Board(DST-SERB),New Delhi,India(Grant reference no.EMR/2017/001550)to carry out this work.