Tunability of mechanical and biodegradation properties of zinc-based biomaterial with calcium Micronutrient alloying

Biodegradable metals are being investigated as temporary implants that dissolve safely in the body after bone regeneration. Zinc (Zn) has an intermediate biodegradation rate between magnesium and stainless steels, yet its degradation rate is too slow to function as a temporary orthopedic implant. Alloying with nutrient elements is considered a strategy to tune its mechanical properties and in vivo biodegradability. Zn/calcium (Zn/Ca) alloys (with 0.5, 1, and 2 wt% Ca) were processed by spark plasma sintering and their microstructure, mechanical, and biodegradation properties were investigated. Ca was distributed in the grain boundary regions of Zn due to its low miscibility in Zn. Furthermore, the corrosion rates of Zn/Ca alloys determined from linear polarization measurements (0.164 – 0.325 mm/yr) accelerated by at least 10% compared with pure sintered Zn (0.149 mm/yr) with simultaneous dissolution of Zn and Ca, as verified from X-ray diffraction analysis of the corrosion products. The alloy specimens exhibited hardness (52 – 58 HV) and compressive strength (93 – 119 MPa) comparable with those of human cortical and cancellous bones (49 HV; 90 – 209 MPa). This study demonstrated the tunability of the mechanical and biodegradation properties of Zn-based materials by alloying them with a nutrient element for potential application as temporary orthopedic implants.


Introduction
Biodegradable metallic implants show a promising potential to emerge as temporary medical devices that dissolve safely in the body after bone regeneration.Therefore, their use will minimize the need to carry out costly and sometimes life-threatening additional surgeries to replace or remove current permanent implants (Heiden et al., 2015).In clinical practice, permanent metallic implants made from alloys of titanium (Shamsolhodaei et al., 2021;Zuo et al., 2022), stainless steels (Fu et al., 2020;Lodhi et al., 2019;Zeng et al., 2018), and cobalt-based biomaterials (Chen and Thouas, 2015;Hedberg et al., 2014;Petit et al., 2016) have been used for orthopedic applications.Although these metal alloys exhibit good mechanical properties and durability, their non-degradability, higher moduli of elasticity compared with natural bone, and long-term clinical complications that may require another removal surgery have spurred researchers to look for viable alternatives (Chen and Thouas, 2015).
Several alloys of iron (Fe)-based, and magnesium (Mg)-based materials have been extensively studied as absorbable metallic implants.Mgbased implant materials exhibit high biocompatibility and low thrombogenicity (Hernández-Escobar et al., 2019).However, some technical challenges militating against the application of Mg-based material include its rapid degradation rate in the physiological environment (Heiden et al., 2015;Mostaed et al., 2018;Zheng et al., 2014) although this can be partially mitigated.This can result in premature deterioration of mechanical integrity, which does not allow entire tissue remodeling.Furthermore, rapid degradation results in the accumulation of large amounts of hydrogen gas, which can lead to wound interface cavitation and tissue necrosis (Heiden et al., 2015).In addition, excess concentrations of degradation products above the normal daily concentrations have been a cause for concern (Heiden et al., 2015).
Compared to Mg, Fe has a slow biodegradation rate that is not suitable for practical resorption applications (Heiden et al., 2015).Some studies have shown that Mg-based stent implant is completely degraded after 6 months in vivo, whereas a stainless steel stent in a similar condition retains more than 90% structural integrity (Fu et al., 2020).Another study reports the complete degradation of a Mg-based stent within 3 months, whereas an Fe-based stent shows minimal degradation after 12 months (Peuster et al., 2006)).In clinical practice, the structural integrity of a cardiovascular implant material is expected to have degraded by about half after 6 months of evaluation to allow for a gradual remodeling of the surrounding aortic tissue.Therefore, Fe-based implants, without optimized structural modification, may not be suitable for bio-resorbable applications (Fu et al., 2020;Peuster et al., 2006).Moreover, the degradation of Fe produces iron oxide in large amounts, which may not be adequately metabolized in the body (Bowen et al., 2013;Mostaed et al., 2018).Due to their magnetic nature, Fe-based implants can interfere with some medical equipment (e.g., MRI) (Heiden et al., 2015).Consequently, some recent research efforts have considered zinc (Zn) as a potentially viable alternative to both Mg and Fe, primarily associated with its moderate corrosion rate in simulated body fluids (Fu et al., 2020;Hernández-Escobar et al., 2019).
Zn is needed for normal growth and wound healing possesses (Mostaed et al., 2018).It also possesses anti-inflammatory, antibacterial, anti-fungal, and anticancer properties (Heiden et al., 2015;Su et al., 2019).Furthermore, research has shown that Zn plays a significant role in the growth, formation, and mineralization of bone (Heiden et al., 2015;Su et al., 2019).In bone fracture healing, the adequate mechanical strength of the temporary metallic implant should be sustained between 12 and 24 weeks (Zheng et al., 2014).Despite the intermediate corrosion rate of Zn, its degradation rate is slow for a potential temporary bone replacement application.For instance, corrosion rates ranging from 0.014 to 0.073 mm/yr (Lin et al., 2019;Yang et al., 2020Yang et al., , 2018) ) have been reported in the literature.At these rates, a hypothetical 2 mm Zn implant will be completely degraded in about 27 years.Such prolonged degradation may elicit inappropriate host response due to a mismatch between implant degradation and the bone remodeling process.Therefore, it is imperative to design Zn-based alloys with tunable degradation rates.One strategy is to design alloys with nutrient elements, such as magnesium (Mg), calcium (Ca), iron (Fe), manganese (Mn), and molybdenum (Mb) (Heiden et al., 2015).As the human body contains some of these trace metallic elements, they are usually selected for alloying with Zn (Li et al., 2015;Yang et al., 2020;Zou et al., 2018).
Some studies have shown that bioceramic phases, such as hydroxyapatite (Yang et al., 2018), and nutrient elements including Cu (Tang et al., 2017), Mg (Li et al., 2015), Mg/Cu (Lin et al., 2019), Mn/Fe/Ag (Yang et al., 2020), Ca (Li et al., 2015;Zou et al., 2018) and Sr (Li et al., 2015;Yang et al., 2020) could enhance the degradation of Zn-based materials.Only limited studies exist on the Zn-Ca system.Li et al. examined the effect of mechanical working on cast Zn-1Ca alloy (Li et al., 2015).They reported that rolling and extrusion enhanced microhardness, tensile and compression properties.Ca accelerated the degradation of the alloy by about 19 and 12% during electrochemical and immersion corrosion, respectively (Li et al., 2015).The authors, however, did not examine the effect of Ca on the microstructure of the alloy.Similar observations were documented in (Zou et al., 2018) for as-cast Zn/Ca alloys.Contrary to the report in (Li et al., 2015;Zou et al., 2018), a recent study showed that Ca had limited influence on the tensile strength, compressive strength, and microhardness of extruded Zn-0.8Ca although the corrosion rate increased after immersion in simulated body fluid (Yang et al., 2020).Considering the vital role of Ca in ensuring bone health and preventing osteoporotic fractures (Zhu and Prince, 2012), more research works on the Zn/Ca alloy system are required to further understand their biocorrosion and mechanical behaviors, especially in the as-prepared state to isolate the effect of mechanical working.
Powder metallurgy processing is ideal for processing a wide array of alloys and near-net forming of parts (Akinwekomi et al., 2016).When combined with the spark plasma sintering (SPS) technique, rapid sintering can be achieved while minimizing porosity and defects (Alvi and Akhtar, 2019).In this work, we investigated the effect of Ca as an alloying element to modify the mechanical and biodegradation properties of Zn alloys as potential temporary implant materials.The alloys were processed via powder metallurgy by elemental mixing of Zn and Ca powder and subsequent sintering of the powder mixture using the SPS technique.Experimental results showed that Ca could be used to modify the mechanical properties of the alloys to match those of human cortical and cancellous bones while accelerating the biodegradation process for bone remodeling.The results showed that the alloys are potentially applicable as temporary biodegradable implants for cancellous and cortical bone healing.

Powder preparation and spark plasma sintering
Zinc powder (Zn, 99.9% purity − 140 + 325 mesh; Alfa Aesar) and calcium granules (Ca, 99.9% purity 2-3 mm; Merck, Germany) were used as the starting materials.At first the Ca granules were crushed in an agate mortar and passed through a sieve with an aperture of 150 μm.Zn and Ca powder were mixed using a roller mill for 1 h at a ball-to-powder ratio of 10:1.Ca weight fractions (wt.%) were 0.5, 1, and 2 wt% and samples designated as Zn, Zn-0.5Ca,Zn-1Ca, and Zn-2Ca.All powders were handled in a glove box under a high-purity argon atmosphere.Pure Zn (control sample) and mixed powders were loaded into a graphite die of diameter 12 mm and sintered in a spark plasma sintering (SPS) equipment (Dr.Sinter 2050; Sumitomo Coal Mining Co., Ltd., Japan).Several combinations of sintering parameters (pressure and temperature) were experimented with before selecting the optimal combination of 300 • C under a uniaxial pressure of 40 MPa.The heating rate was 20 • C/min with a soaking time of 10 min at the sintering temperature followed by furnace cooling to the ambient temperature.

Density, microstructural, and X-ray diffraction characterizations
General sample preparation involved successive grinding and polishing on silicon carbide papers of increasing grit sizes.Cylindrical samples of dimensions Ø 12 mm × 7 mm were used for density measurements by Archimedes principle.The microstructural examination was conducted on a scanning electron microscopy (SEM, JSM-IT300LV, JEOL GmbH, Germany) equipped with an energy-dispersive X-ray spectroscopy (EDS).X-ray diffraction analyses were undertaken on a PANalytical Empyrean machine operating at 40 kV and 40 mA using Cu-Kα radiation from 2θ = 20 • -80 • with a scan step size of 0.026.

Microhardness and compression tests
Vickers microhardness (Duramin-40 AC3, Struers, Germany) of the samples (Ø 12 mm × 3 mm) were obtained at room temperature under an applied load of 5 gf and a loading duration of 10 s.The average and standard deviations of at least ten readings were reported.Room temperature compression tests were done on a Gleeble 3800 (Dynamic Systems Inc, USA) at a strain rate of 0.001 s − 1 .Samples were wire-cut into Ø 5 mm × 10 mm and a minimum of three replicates from each composition were tested.Nickel pastes and graphite sheets were placed at the end of each sample to minimize friction and barreling effects.Fractured surfaces were examined under the SEM.

Electrochemical and immersion tests
An electrochemical test at ambient temperature was done using a standard three-electrode setup in phosphate-buffered saline (PBS, pH 7.4, Sigma-Aldrich, Germany) on a Gamry Instruments Reference 3000 (Gamry Instruments, Warminster, PA, USA) instrument.Zn samples were used as the working electrode, while a graphite rod and Ag/AgCl saturated with KCl were used as the counter and reference electrodes, respectively.Open circuit potential (OCP) was established for 3600 s followed by linear polarization measurement at a perturbation amplitude of ±10 mV versus OCP at 0.2 mV/s.Thereafter, the potential dynamic polarization test was carried out from − 0.3 to +0.3 V vs OCP at a scan rate of 0.2 mV/s.For both measurements, at least three samples were tested.Corrosion rate, CR LPR (mm/y) from the linear polarization measurement was determined according to ASTM Standard G59-97 (ASTM Standard G59-97 (2020), 2020), while the potentiodynamic corrosion rate, CR pd (mm/y) was evaluated from the relation (ASTM Standard G102-89, 2010): where K 1 , i cor , ρ, and EW represent the corrosion constant (3.27 × 10 − 3 mm g/μA cm y), corrosion current density (μA/cm 2 ), density (g/cm 3 ), and equivalent alloy weight, respectively.Long-term corrosion characteristics of the samples (Ø 12 mm × 3 mm) were assessed in PBS for 4 weeks.After the test, the samples were rinsed with distilled water and dried in an oven at 50 • C. Changes in surface morphology and composition after corrosion were analyzed on JSM-IT300LV SEM.Furthermore, corrosion products were analyzed using X-ray diffraction technique on the corroded test coupons.Thereafter, corrosion products were removed using 100 g ammonium persulfate ((NH 4 ) 2 S 2 O 8 ) dissolved in 1000 mL of distilled water.Corroded samples were washed for 5 min at 20-25 • C (ASTM Standard G1-03, 2011).The mass loss of samples was measured on an electronic balance.An average of three measurements were taken for each batch of samples.Mass loss corrosion rate (CR m , mm/yr) was calculated as (ASTM-NACE TM0169/G31 − 12a, 2012): where K, W l , ρ, A, and t represent corrosion constant (8.76 × 10 4 ), mass loss (g), density (g/cm 3 ), surface area (cm 2 ) of a sample, and immersion time (h), respectively.

Density, microstructural, and X-ray diffraction characterizations
The sintered densities of Zn-Ca alloys are shown in Table 1.All the alloys sintered at 300 • C exhibit densification of more than 93% of the theoretical density of Zn (7.13 g/cm 3 (Rumble, 2004)), an indication of the suitability of the SPS technique in processing Zn-based materials.Fig. 1 shows the microstructure of the sintered Zn and Zn/Ca samples.For the pure Zn, initial particle morphology is indistinguishable in the microstructure with no discernible pores.This shows that the sintered Zn metal is dense and conforms with the value of density (7.07 g/cm 3 , relative density 99.99% of the theoretical) shown in Table 1.
The densities of the Zn/Ca alloys decrease with increasing wt.% of Ca addition.In Fig. 1(b)-(d), individual Zn particles are distinguishable and interparticle pores become comparatively larger with increasing Ca content, which is reflected in the decrease in the sintered density reported in Table 1.Ca and Zn exhibit almost no solubility in each other (Brubaker and Liu, 2001;Wasiur-Rahman and Medraj, 2009), hence Ca is mainly found in the interparticle regions of Zn, as shown in Fig. 1.EDX mapping (Fig. 1(e)) of Zn-2Ca, as a representative Zn/Ca sintered material, further confirms the separation of Ca into the interparticle regions of Zn.A similar observation is reported for Zn-hydroxyapatite (HA) composites, where HA particles are segregated into the interparticle regions of Zn (Yang et al., 2018).
The X-ray diffractograms of the sintered specimens are shown in Fig. 2.Only the diffraction peaks belonging to Zn are observed, while other likely stoichiometric compound phases, such as Ca 3 Zn, Ca 5 Zn 3 and CaZn 13 (Wasiur-Rahman and Medraj, 2009), and elemental Ca could not be detected.Although Li et al. (2015) have reported the presence of CaZn 13 in a melt-processed Zn-1Ca alloy at about 2θ = 33 • , no such peak is detected in this study.Such a compound may be present in the authors' work considering that the processing temperature is 657 • C for 30 min.This temperature is favorable for the formation of Zn/Ca compounds (Wasiur-Rahman and Medraj, 2009).However, in our study, two probable reasons may be responsible for the absence of such compounds: (i) the lowest temperatures of formation of some stoichiometric Zn/Ca compounds, for instance, Ca 3 Zn and Ca 5 Zn 3 , are 393 • C and 687 • C, respectively (Brubaker and Liu, 2001;Wasiur-Rahman and Medraj, 2009), while the eutectic decomposition of liquid Zn/Ca into Zn + CaZn 13 occurs at 419 • C (Brubaker and Liu, 2001).These temperatures are, at least, 93 • C higher than the processing temperature (300 • C) in this study.Therefore, it is highly plausible that these stoichiometric compounds are not formed during SPS processing.(ii) X-ray diffraction technique has a limitation in detecting elements or compounds with low concentrations.For instance, Khan et al. ( 2020) mention a limit of detection of about 2% per phase.Due to the low weight fraction of Ca used in alloying in this study (max. 2 wt%), reflections from any Zn/Ca compounds or elemental Ca may be too low to be statistically distinguishable from the background counts.XRD limit of detection is given as N reflection > N background + 3σ background , where N is measured number of counts and σ background (= ) is the absolute standard deviation of the background count (Ermrich and Opper, 2013).

Microhardness and compressive characteristics
Microhardness values are presented in Table 1.The highest value of 58.20 HV is exhibited by pure Zn, which is approximately 5% higher than that of Zn-0.5Ca.Generally, the microhardness of the samples decreases as Ca wt.% increases.This observation is related to the higher porosity structure found in the microstructure of the Zn-Ca samples (Fig. 1).These volume defects will decrease resistance to localized deformation of the matrix during indentation (Akinwekomi et al., 2016), thus resulting in lowering the values of microhardness.Nevertheless, the microhardness values reported in this study supersede those of as-cast Zn (Zou et al., 2018) and compare favorably with those of as-cast and as-rolled Zn-1Ca (Li et al., 2015).Further, comparison with similar SPS-processed and extruded Zn ( Čapek et al., 2018) and Zn/hydroxyapatite composites (Yang et al., 2018), shows that Zn/Ca alloy in this work exhibits improved microhardness values.This may be attributed to the optimized processing parameters (pressure, temperature, and sintering time) used in this work.For instance, a longer sintering time activates sintering and facilitates solid-state interparticle diffusion (Akinwekomi et al., 2020), which is evident from the microstructural images presented in Fig. 1.Literature data show that the microhardness values that have been reported for both human cancellous (48.92 HV) and cortical bones (49.18 HV) (Boivin et al., 2008) are comparable to the values measured in this work.This suggests that the Zn/Ca alloys are potentially applicable as temporary implant materials for remodeling cancellous and cortical bones.
Representative compression stress-strain curves are presented in Fig. 3.During the compression test, pure Zn specimens do not fracture but continue to deform up to 50% strain.On the other hand, Zn/Ca samples fracture at less than 10% strain.Therefore, the ultimate compressive strengths (UCS) of all samples are reported at 6% total strain in compliance with the recommendation of ASTM E9-09 (ASTM Standard E9-09, 2009) for samples without a sharp-kneed stress-strain characteristic curve.Pure Zn exhibits the highest ultimate compressive strength (UCS) and yield strength (CYS).Alloying with Ca, however, causes a decline in these properties.For instance, there is a decrease of 3%, 13%, and 22% in UCS as Ca wt.% increases from 0.5 to 2 wt% when compared with pure Zn.Microstructural characterization earlier presented in Fig. 1 offers some insight into this observation.Due to the limited solubility of Ca in Zn, Ca particles are segregated along the interparticle regions of Zn with the formation of numerous pores.These pores act as stress concentrators, which reduces both the UCS and CYS of the alloys ( Čapek et al., 2018).The values of UCS of these samples fall in the range of the reference UCS for human cortical bone (Ginebra, 2009).
Therefore, they have the potential as temporary bone substitute materials while eliminating the challenge of stress shielding or loosening of implants associated with high-strength materials, such as stainless steel and titanium-based alloys.

Electrochemical and immersion behaviors
Transient electrochemical behavior evaluated as the potentiodynamic electrochemical test in PBS is shown in Fig. 4. As corrosion potential increases, the anodic part of the polarization curve of pure Zn shows three distinct regions: anodic dissolution, passivation region, and trans-passivation.The anodic regions of Zn/Ca samples, however, indicate only dissolution.All the samples have similar cathodic behaviors.Tafel extrapolation is used to estimate the corrosion potential, E corr , corrosion current densities, i corr , and the corresponding potential dynamic corrosion rate, CR pd are shown in Table 2. Except for Zn-0.5Ca,i corr generally increases with increasing Ca content, an indication of more susceptibility to corrosion degradation.
For instance, the corrosion current densities of Zn-1Ca and Zn-2Ca  exceed that of pure Zn by 16 and 29 times, respectively.Zn-1Ca and Zn-2Ca have higher CR pd of 0.158 and 0.357 mm/yr than pure Zn (0.201 mm/yr).Since Ca has no miscibility with Zn (Brubaker and Liu, 2001), it is segregated in the Zn interparticle regions that form pores with adjacent Zn particles.These voids, therefore, increase the surface area exposed to the electrolyte and facilitate its percolation deep into the samples to enhance degradation.As seen in the XRD analyses of the corrosion products, both Ca and Zn are dissolved in the PBS solution.
Similarly, corrosion rate (CR LPR ) determined from linear polarization experiments are shown in Table 2.The results follow a similar trend as that of CR pd , where an increase in Ca content accelerates corrosion.These results agree with an earlier study that shows that hydroxy apatite (Yang et al., 2018) and alloying elements, such as Mg and Cu (Lin et al., 2019;Tang et al., 2017), enhance the corrosion of Zn.Further, as Zn is nobler than Ca, it is possible that galvanic couples may be established at different sites on the alloys, which may contribute to accelerating their corrosion (Yang et al., 2020;Zou et al., 2018).
Similarly, long-term immersion test results suggest an increase in corrosion rate with increasing Ca content.After 4 weeks of immersion, fine corrosion precipitates are seen in the microstructural images shown in Fig. 5.For the pure Zn specimen, a homogeneous distribution of filiform or fern-like corrosion precipitates could be observed.However, these corrosion products increase in size and morphology as the Ca content increases.XRD analysis (Fig. 5 (Fu et al., 2020;Yang et al., 2018) and simulated body fluid ( Čapek et al., 2018), which possess similar ions as PBS (Chen et al., 2016).Representative microstructural images of the corroded surfaces after the removal of the corrosion products are shown as insets in Fig. 5. Pure Zn presents uniform surface degradation with some superficial cavities, whereas Zn-Ca samples exhibit extensive cavities/grooves indicating severe corrosion.The corrosion of Zn in PBS can be summarized in the following steps (Chen et al., 2016).
Step This corrosion product (Zn(OH) 2 ) is thought to provide some form of protection that is seen as the temporary passivation region in the anodic portion of the Tafel plot of Zn (Fig. 4).However, with a prolonged immersion term, Cl-in PBS attacks Zn(OH) 2 to release further zinc ions that combine with the phosphate ions in PBS to yield a more thermodynamically stable zinc phosphate (Chen et al., 2016;Yang et al., 2018).Zn (OH) 2 may also decompose into zinc oxide and water (Yang et al., 2018) Finally, excess Zn 2+ may also react with Ca 2+ from PBS to yield hydrated calcium zincate:

Conclusions
We investigated the effect of the nutrient element Ca as an alloying element to modify the mechanical and biodegradation properties of Zn as a potential temporary implant material.Samples were synthesized via powder metallurgy and consolidated using the spark plasma sintering technique and their microstructure, mechanical and bio-corrosion properties were investigated.The density of samples was greater than 93% highlighting the suitability of the spark plasma sintering technique as a viable approach for the fabrication of temporary orthopedic Znbased implants.Ca was distributed in the interparticle regions of Zn due to its limited solubility in Zn.The compressive strength (93-119 MPa) and hardness (52-58 HV) properties of the samples were within the range required for human cortical and cancellous (90-209 MPa; 49 HV) bones.In addition, the corrosion rates of Zn/Ca alloys determined from linear polarization scans in PBS accelerated by at least 10% compared with pure Zn.A similar trend was also observed from the immersion test.This study demonstrated the tunability of the mechanical and biodegradation properties of Zn-based materials by alloying them with a nutrient element for potential application as temporary orthopedic devices.Further studies are, however, required to explore the in vivo and tribo-corrosion characteristics of these alloys.Resources, Project administration, Funding acquisition.

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.
and ZnO are all detected as corrosion products shown in the XRD presented in Fig.5(e).

Fig. 4 .
Fig. 4. Representative potentiodynamic polarization curves of Zn and Zn/Ca alloys in PBS.Inset is the OCP for each sample.

Table 1
Density and mechanical properties of pure Zn, Zn/Ca alloys sintered at 300 • C, and human bones.
(e)) of the degradation products shows the presence of hydrated zinc phosphates i.e., Zn(HPO 4 ) 2 ⋅3H 2 O ⋅2H 2 O), and zinc oxide (ZnO).Similar corrosion products have been identified in the immersion tests of Zn in Hank's 1: Oxidation of elemental Zn to Zn ions at the anode: Zn → Zn 2+

Table 2
Tafel electrochemical parameters of Zn and Z/Ca alloys in PBS solution.