Hydrothermal synthesis and characterization of calcium phosphate-based coatings on AZ31 magnesium alloy

ORIGINAL ARTICLE KOM – Corrosion and Material Protection Journal 67 8-13 (2023) DOI: 10.2478/kom-2023-0002 8 INTRODUCTION Magnesium as a light metal is used in industry since the 1930s when it was first industrially produced. Together with its alloys, magnesium is one of the lightest metallic construction materials. At the same time, its ability to degrade into non-toxic products in biological systems allows Mg materials to be used in medicine, particularly in orthopedics for implants. One of the major limitations of Mg use in this field is its excessive chemical reactivity, which can cause the development of an organism’s inflammatory and release of large amounts of hydrogen gas into the immediate environment of the implant. To avoid these disadvantages and to enable the use of Mg materials in medicine, they tend to be various biocompatible coatings or protective layers are applied to their surface. For this purpose, in recent years, phosphatebased coatings (CaP) have been applied, most often in the form of hydroxyapatite (HAp). These coatings effectively increase the corrosion resistance of Mg alloys and at the same time provide better biocompatibility through binding with bone tissue [1,2]. The preparation of CaP coatings can be accomplished by a number of methods, including chemical vapor deposition, solution precipitation, sol-gel method, etc. Nowadays, in particular, hydrothermal synthesis seems to be very attractive. This method is technologically and economically inexpensive, but at the same time allows the formation of high-quality layers. However, prior to the actual deposition process, the substrate must be suitably pretreated so that the coating on its surface achieves the best properties (chemical, physical and mechanical) [1]. The critical factors of hydrothermal synthesis affecting the structure and properties of CaP coatings are mainly temperature, pH, deposition time, etc [1-6]. As demonstrated by studies, the correct selection of precursors, and their concentration or pretreatment of the substrate surface are also among the key parameters (deposition time, pH of the reaction mixture, and concentration of precursors in the reaction mixture) for the resulting coating properties. This study investigates the influence of hydrothermal synthesis parameters on the morphology, phase composition and corrosion behavior of CaP coatings. EXPERIMENTAL The samples of AZ31 magnesium alloy were cut to dimensions of 20 × 20 mm. The elemental composition of AZ31 magnesium alloy was the same as in our previous studies [7,8]. Magnesium samples were ground using #1200 SiC paper. Then, the samples were rinsed in distilled water and ethanol and air-dried. Hydrothermal synthesis and characterization of calcium phosphate-based coatings on AZ31 magnesium alloy Horáková L., Doskočil L., Wasserbauer J., Buchtík M. Materials Research Centre, Faculty of Chemistry, Brno University of Technology, Purkyňova 464/118, 612 00 Brno, Czech Republic E-mail: [email protected] This study aims to analyze the influence of process parameters used for hydrothermal synthesis of CaP coatings on their properties and to improve their corrosion resistance and biocompatibility compared to the substrat AZ31. The parameters monitored were deposition time, pH of the reaction mixture, and concentration of precursors in the reaction mixture. For the deposited CaP coatings on AZ31 magnesium alloy, the surface morphology and the number of structural defects were evaluated using scanning electron microscopy. Electrochemical corrosion properties were evaluated using polarization techniques in Hank’s solution. The results showed that the best properties were obtained for the sample prepared in a reaction mixture at 120 °C, pH 5 for a deposition time of 120 min, when the concentration of precursors in the reaction mixture was 0.30 mol/l Ca(NO3)2·4H2O and 0.28 mol/l NH4H2PO4. Under these conditions, the best electrochemical corrosion properties were achieved. Hydrothermal synthesis and characterization of calcium phosphate-based coatings... Horáková L., et. al KOM – Corrosion and Material Protection Journal 67 8-13 (2023) DOI: 10.2478/kom-2023-0002 9 To unify the surface, and to prevent the excessive dissolution of the Mg substrate during the hydrothermal synthesis, a stable and compatible Mg(OH)2 layer was deposited on the surface of ground samples. Mg(OH)2 layer was prepared under hydrothermal conditions in 2 mol/l NaOH solution at 120 °C for 12 hours. After this, the samples were rinsed in distilled water and ethanol and air-dried. The preparation of the reaction mixture was carried out at laboratory temperature. The reaction mixture was prepared by mixing 175 ml of solution marked as A in this study, containing 0.07, 0.15 or 0.30 mol/l Ca(NO3)2·4H2O. Then, 125 ml of solution marked as B in this study,containing 0.07, 0.14 or 0.28 mol/l of NH4H2PO4 was dropwise added into intensively stirred solution A. The pH value was adjusted in the range 2-7 using 1 mol/l NaOH solution. Magnesium samples were hung on a special holder coated with Teflon tape and were transferred into 250 ml Teflon-lined stainless steel pressure vessel containing 150 ml of the prepared reaction mixture. The pressure vessel was placed into the Memmert UF55 (Memmert GmbH + Co.KG, Büchenbach, Germany) electric oven heated up to 120 °C for 15 300 minutes. Coated magnesium samples were rinsed with distilled water, and alcohol, and air-dried. The surface morphology and chemical composition of the prepared coatings were analyzed using Zeiss EVO LS-10 scanning electron microscope (SEM) (Carl Zeiss Ltd., Cambridge, UK) with energy dispersive spectrometer (EDS) Oxford Instruments Xmax 80 mm2 (Oxford Instruments plc, Abingdon, UK) and AZtec software (version 2.4, Oxford Instruments, High Wycombe, UK) was used to analyze the obtained data. Potentiodynamic measurements were performed using the Bio-Logic VSP-300 potentiostat/galvanostat (BioLogic, Seyssinet-Pariset, France) in Hank’s Balanced Salt Solution (HBSS) without Ca2+, Mg2+ ions at laboratory temperature. The measured samples were monitored in open circuit potential (OCP) for 60 min (when it was established) before beginning the electrochemical measurements. Electrochemical potentiodynamic measurements were performed in a potential range from −200 mV to +500 mV vs. open circuit potential (OCP) at a scan rate of 1 mV/s.. The range was chosen with regard to previous study [8]. The measurements took place in a standard three-electrode cell connection with a saturated calomel electrode (reference electrode), a platinum mesh (counter electrode) and a working electrode (measured sample with an exposed area of 1 cm2). Results were evaluated using Biologic EC-Lab® software. RESULTS AND DISCUSSION Mg(OH)2 interlayer Fig. 1a shows the surface of the AZ31 magnesium alloy after grinding (without the coating). One-directional grooves resulting from the grinding process are visible. From a macroscopic point of view, the surface is homogeneous, without visible oxidized areas and defects. Based on the previous studies [9-11], it was necessary to treat the Mg alloy surface with a protective layer of Mg(OH)2 before the deposition of CaP coatings to prevent its degradation during the hydrothermal process. Fig. 1b shows the pre-treated surface of AZ31 magnesium alloy. In the SEM image (Fig. 1b), no difference in the surface morphology was evident compared to the ground surface of the AZ31 alloy sample. The presence of the Mg(OH)2 layer could be demonstrated by EDS detecting a higher oxygen content on the surface compared to the ground one (Fig. 1). The thickness of the Mg(OH)2 layer was determined from a cross-cut of the sample, and it was approximately 3 µm. The formation of a compact Mg(OH)2 layer over the entire substrate surface was also indicated by the color change of the samples surface (Fig. 1c,d). a) b) c) d) Fig. 1. Analysis of the Mg substrate, a) ground AZ31 alloy, SEM, b) ground AZ31 alloy with hydrothermally prepared Mg(OH)2 layer, SEM, c) sample of ground AZ31 alloy, d) sample of ground AZ31 alloy with hydrothermally prepared Mg(OH)2 layer Hydrothermal synthesis and characterization of calcium phosphate-based coatings... Horáková L., et. al KOM – Corrosion and Material Protection Journal 67 8-13 (2023) DOI: 10.2478/kom-2023-0002 10 Effect of pH of the reaction mixture and deposition time on the characteristics of CaP coatings The pH of the CaP reaction mixture was set to 2, 3 5 and 7. The deposition time of the CaP coatings was set to 15, 30, 60, 120, 180 and 300 minutes. The concentration of precursors was 0.30 mol/l Ca(NO3)2·4H2O and 0.28 mol/l NH4H2PO4 and the preparation time of interlayer Mg(OH)2 was 24 h. The criterion for assessing the quality of the coating was the evaluation of the surface morphology of the coating and the results of potentiodynamic tests in the Hank’s solution. The surface morphology of the prepared CaP coatings was analyzed using SEM (Fig. 2). At pH 2 and 3, the substrate surface was non-uniformly covered with CaP crystals after 30 minutes of deposition. Due to the local formation of coarse crystals, the substrate was significantly corroded between the CaP crystals [12]. In the case of the sample deposited at pH 2, a more uniform layer covering the surface did not form even after a longer deposition time. Corrosion of the Mg substrate j) pH 7, 30 min d) pH 3, 30 min g) pH 5, 30 min a) pH 2, 30 min l) pH 7, 300 min f) pH 3, 300 min i) pH 5, 300 min c) pH 2, 300 min k) pH 7, 120 min e) pH 3, 120 min h) pH 5, 120 min b) pH 2, 120 min Fig. 2. Surface morphology of CaP coatings Hydrothermal synthesis and characterization of calcium phosphate-based coatings... Horáková L., et. al KOM – Corrosion and Material Protection Journal 67 8-13 (2023) DOI: 10.2478/kom-2023-0002 11 occurred mainly due to the acidic pH and the presence of aggressive ions of (NO3 −, PO43−) in the reaction mixture. Locally, coarse CaP crystals were present on the surface. In the case of the sample deposited at pH 3, there was partial coverage of the Mg substrate with increasing deposition time. Consequently, the surface was still significantly corroded due to the low pH and aggressive ions of the reaction mixture. At deposition times longer than 120 min, significant coarsening of the CaP crystals occurred, however, the corrosion of the substrate between the CaP crystals remained. In the case of samples deposited at pH 5, a uniform layer on the surface was not achieved as well. After 120 minutes of deposition, the surface was covered with fine CaP crystals (flake-like crystals with different lengths). As the reaction time increased, the coating crystals became coarser and the Mg substrate corroded even easier. The formation of CaP coating with a different morphology was observed at pH 7 when compared to the coatings deposited at lower pH values. The coating was formed by very fine crystals. After 30 minutes of deposition, local peeling of the coating was observed. Local separation of the coating from the substrate also contributed to the structural heterogeneity and thus reduced protection of the coating against corrosion attack. This was probably due to the poor adhesion of the CaP coating to the substrate material. As the deposition time increased, the CaP crystals coarsened. Above mentioned phenomena can be explained as follows. During the exposure of Mg alloy with Mg(OH)2 layer in the CaP reaction mixture, two opposing reactions occur. The first (undesired) reaction in this case is the dissolution of the Mg(OH)2 layer due to the acidic environment and the presence of aggressive ions. The second competitive process is the formation of the CaP coating. Thus, different reactions can occur depending on the reaction conditions of the present ions. First, the dissociation of Ca(NO3)2 takes place. Ca2+ ions react in an aqueous solution with H2PO4 − ions. The resulting monohydrate of calcium dihydrogen phosphate subsequently reacts with other calcium ions and water to form CaHPO4·2H2O (dicalcium phosphate dihydrate), (Eq. 1) [13]. Ca2+ + H2PO4 − +2 H2O → CaHPO4 ∙ 2 H2O + H+ (1) Dicalcium phosphate dihydrate can react with calcium ions to form HAp (Eq. 2-3) or lower calcium phosphates – octacalcium phosphate. It depends on the condition of the reaction mixture. 6 CaHPO4 ∙ 2 H2O + 2 Ca2+ + 4 OH− → → Ca8(HPO4)2(PO4)4 ∙ 5 H2O + 11 H2O (2) 6 CaHPO4 ∙ 2 H2O + 4 Ca2+ → → Ca10(PO4)6(OH)2 + 8 H+ (3) Potentiodynamic measurements From the dependence of the corrosion current density of CaP coatings on the preparation time log icorr = f (t) (Fig. 3a), it can be seen that during the low deposition time the corrosion current density (icorr) was also lower. At higher deposition times, it was observed that there was an increase in icorr probably due to the breaking of the Mg(OH)2 interlayer. It was also observed that with the increasing pH value of the reaction mixture, a decrease in icorr occurs at the same deposition times. In the case of the samples prepared at pH 2 and 3, the dissolution of the protective Mg(OH)2 interlayer and corrosion of the Mg substrate occurred predominantly due to the acidic environment and the instability of Mg(OH)2 in the acidic environment. As the pH increased to 5, there was a more pronounced decrease in icorr. On the other hand, the value of icorr with Mg(OH)2 and CaP coating prepared at pH 5 reached approx. an order of magnitude higher compared to the Mg alloy coated with Mg(OH)2 layer only. a) b) Fig. 3. The dependence of potentiodynamic parameters on deposition time and pH value of the deposition mixture: a) log icorr = f(t), b) Ecorr = f(t) Hydrothermal synthesis and characterization of calcium phosphate-based coatings... Horáková L., et. al KOM – Corrosion and Material Protection Journal 67 8-13 (2023) DOI: 10.2478/kom-2023-0002 12 In the Fig. 3b, it can be seen that the most negative value of corrosion potential Ecorr (–1633 mV) has AZ31 substrate without Mg(OH)2 interlayer. After Mg(OH)2 deposition, the Ecorr value shifted to 1359 mV. Deposition of CaP coating at pH of the reaction mixture 3, 5 and 7 resulted in a furthershift of the Ecorr to more positive values. However, with increasing exposure time, there was a gradual decrease in the Ecorr value. The most significant decrease in corrosion potential was observed between deposition time 120 and 180 minutes, which was probably due to partial disruption of the Mg(OH)2 interlayer and corrosion of the Mg substrate (AZ31). Effect of concentration of precursors in the reaction mixture The preparation was performed at pH 5, for 120 minutes at a temperature of 120 °C with different concentrations of precursors. The SEM analysis of the coatings (Fig. 4) showed a coarsening of the CaP crystals with a decreasing concentration of precursors in the reaction mixture. Also, the distance between CaP crystals increased with decreasing concentration of precursors in the reaction mixture. This increased space between the CaP crystals led to the phenomenon where the corrosive medium easily passes to the surface of the substrate. This is reflected in a decreased corrosion resistance (Tab. 1). Fig. 5. Elemental analysis of optimized CaP coating on AZ31 magnesium alloy a) c)b) Fig. 4. The morphology of CaP coatings prepared at different concentrations of precursors, a) 0.30 mol/l Ca(NO3)2·4H2O; 0.28 mol/l NH4H2PO4, b) 0.15 mol/l Ca(NO3)2·4H2O; 0.14 mol/l NH4H2PO4, c) 0.07 mol/l Ca(NO3)2·4H2O; 0.07 mol/l NH4H2PO4 Tab. 1. Results of potentiodynamic measurements, effect of precursor concentration on Ecorr and icorr of CaP-coated Mg substrates Concentration of precursors (mol/l) 0.30 mol/l Ca 0.28 mol/l H2PO4 0.15 mol/l Ca 0.14 mol/l H2PO4 0.07 mol/lCa 0.07 mol/l H2PO4 Ecorr (mV) –1470 –1504 –1558 icorr (µA·cm-2) 0.20 ± 0.05 0.20 ± 0.03 0.30 ± 0.05 Hydrothermal synthesis and characterization of calcium phosphate-based coatings... Horáková L., et. al KOM – Corrosion and Material Protection Journal 67 8-13 (2023) DOI: 10.2478/kom-2023-0002 13 Considering the surface morphology, the coverage of the Mg substrate and the number of defects, pH 5 and a deposition time of 120 min were chosen as optimal conditions. Fig. 5 shows the optimized CaP coating. Based on EDS analysis, the optimized CaP coating has a uniform distribution of Ca, P and O elements (Fig. 5). The average thickness of optimized CaP coating ranged between 15 and 20 µm. The EDS analysis also revealed that the coating contains 70.7 at.% of oxygen, 15.4 at.% of calcium, 13.7 at.% of phosphorus and 0.2 at.% of magnesium. It can be assumed that it was probably DCPD (dibasic calcium phosphate dihydrate) with a small amount of other phosphates, e.g. octacalcium phosphate or HAp [14-15]. The possible presence of Mg may be due to the partial substitution of Mg2+ ions for Ca2+ ions in CaP, or its presence may indicate a lower coating thickness when the X-ray beam interacted with the Mg substrate [14-15]. The DCPD is a compound similar to HAp in terms of acceptability with bone tissue. The DCPD phase has an ideal atomic representation of 12.5 at.% Ca, 12.5 at.% P and 75.0 at.% O. CONCLUSIONS Based on the above presented findings, the following conclusions can be drawn: ● There is no uniform formation of CaP coating at all in an acidic environment at pH 2 and 3. At pH 2 and 3, there is significant corrosion of the Mg substrate. ● At pH 7, fine CaP crystals are formed on the surface of the Mg substrate. The CaP coating deposited under these conditions has poor adhesion to the substrate and peeling occurs even after a short deposition time. With the increasing time of preparation, the CaP crystals become coarser as in the case of pH 2 and 3. ● The most suitable conditions for deposition appear to be as follows: temperature 120 °C, pH 5, deposition time 120 minutes, precursor concentration in reaction mixture 0.30 mol/l Ca(NO3)2·4H2O and 0.28 mol/l NH4H2PO4. ● The optimized deposited CaP coating reached a thickness of 15-20 µm and it improves corrosion resistance and biokompatibility of AZ31. ● The best electrochemical corrosion properties, namely corrosion potential Ecorr was determined to −1470 mV and the corrosion current density icorr was 0.20 ± ± 0.05 µA·cm-2. Acknowledgement This work was supported by Specific University Research at FCH BUT, Project Nr. FCH-S-22-8012, Ministry of Education, Youth and Sports of the Czech Republic. REFERENCES 1. Tan L. Preparation and characterization of Ca-P coating on AZ31 magnesium alloy. Transactions of Nonferrous Metals Society of China. 2010, 20, s 648-654. 2. Parida P. et al. Classification of Biomaterials used in Medicine. International Journal of Advances in Applied Sciences. 2012, 1(3), 125-129. 3. Esmaily M. et al. Fundamentals and advances in magnesium alloy corrosion. Progress in Materials Science. 2017, 89, 92-193. 4. León B., JANSEN J. Thin Calcium Phosphate Coatings for Medical Implants. New York, NY: Springer New York, 2009. 5. Chuprunov K. et al. The pH Level Influence on Hydroxyapatite Phase Composition Synthesized with Hydrothermal Method. IOP Conference Series: Materials Science and Engineering. 2020, 731(1). 6. Zhu Y. et al. Growth and characterization of Mg(OH)2 film on magnesium alloy AZ31. Applied Surface Science. 2011, 257(14), 6129-6137. 7. Doskočil L. et al. Characterization of Prepared Superhydrophobic Surfaces on AZ31 and AZ91 Alloys Etched with ZnCl2 and SnCl2. Coatings. 2022, 12(10). 8. Dziková J. et al. Characterization and Corrosion Properties of Fluoride Conversion Coating Prepared on AZ31 Magnesium Alloy. Coatings. 2021, 11(6). 9. Tomozawa M., Hiromoto S. Growth mechanism of hydroxyapatite-coatings formed on pure magnesium and corrosion behavior of the coated magnesium. Applied Surface Science. 2011, 257(19), 8253-8257. 10. Ali A. et al. Hydrothermal deposition of high strength calcium phosphate coatings on magnesium alloy for biomedical applications. Surface and Coatings Technology. 2019, 357, 716-727. 11. Jeong H., Yoo Y. Synthesis and characterization of thin films on magnesium alloy using a hydrothermal method. Surface and Coatings Technology. 2015, 284, 26-30. 12. Earl J. S. et al. Hydrothermal synthesis of hydroxyapatite. Journal of Physics: Conference Series. 2006, 26, 268-271. 13. Su Y. et al. Enhancing the corrosion resistance and surface bioactivity of a calcium-phosphate coating on a biodegradable AZ60 magnesium alloy via a simple fluorine posttreatment method. RSC Advances. 2015, 5(69), 5600156010. 14. Sasikumar Y. et al. Fabrication of Brushite Coating on AZ91D and AZ31 Alloys by Two-Step Chemical Treatment and Its Surface Protection in Simulated Body Fluid. Journal of Materials Engineering and Performance. 2019, 28(6), 3803-3815. 15. Tang H. et al. Fabrication and Characterization of Mg(OH)2 Films on AZ31 Magnesium Alloy by Alkali Treatment. International Journal of Electrochemical Science. 2017, 1377-1388.