The Effects of Acid Etching on the Morphology and Properties of Medical Grade Alumina and Zirconia Surfaces for Prosthetics

In this study we modify and functionalize the surface of alumina and zirconia ceramics for medical applications using chemical etching with mixtures of sulfuric, nitric, hydrouoric acids and peroxide. After etching, the impact of processes on surface development, chemical composition, and topography is studied to select the most effective process of surface development. Medical grade alumina and zirconia ceramic powders have been chemically etched with selected three kinds of acidic solutions : 1.sulfuric and nitric acid, 2.sulfuric acid and peroxide, 3.uoric acid various diluted aqueous solution during the selected time periods. Following heat treatment was performed and the samples characterization were undertaken: morphology and chemical composition , phase composition, functional group determination, and the specic surface area and porosity evaluation.. Comparing the results raised from acidic etching, it was noticed that the use of H 2 SO 4 :HNO 3 solutions causes sulphur residues in ceramic in the form of sulphates. The application of HF negatively affects the structure of the material and cause agglomeration. The most advantageous modication of ceramic powders was application of piranha solution, the obtaining surface development was achieved, satisfactory degree of agglomeration and post-process pollution.

SS and Ti6Al4V alloy [14][15][16]. In ammations, bone lose, are problems can be avoided with the use of ceramic biomaterials.
Despite the numerous advantages, the application of ceramics brings some issues that arise from the physicochemical nature of this group of materials. This is analogous to the zirconium drawbacks connected with metastable nature of this materials, which a signi cant amount of work has been dedicated to, such as the work of Guo, Gremillard and Chevalier [17,18]. In the work of Chevalier and Gremillard, the problem of the in uence of the body environment on the structure is addressed, i.e. the structural changes caused by the interaction between the zirconium material with saliva and moisture [19]. These structural changes are described with the term low thermal degradation (LTD) and are a wellknown mechanism that destructively affects dentures and prostheses made from ZrO 2 [19,20]. The main problem that results from LTD is swelling of zirconia grains, falling out of the structure and, as a result, the formation of critical fractures of the dentures and prostheses [17]. The mechanisms associated with the formation of these adverse phenomena have been described by Guo, who presents kinetic models which are con rmed experimentally by other authors [21,22]. Other disadvantages of ceramics are their high fragility and a relatively time-consuming and expensive production process [23,24]. The proper selection of process parameters when milling dental crowns and bridges creates many problems, for the case of zirconia, special attention needs to be paid to the cooling of the material during milling to prevent overheating of the structure, which negatively affects the durability of the dentures [23,25]. Besides, to the best of our knowledge, there is no clear position as to whether water or oils used during cooling initiate the uncontrolled phase transformation and in consequence LTD which cause premature ageing and destruction of dentures [17,18]. These and other problems i.e. metastable phases stabilisation, dopant selection have been previously addressed in the literature [26][27][28]. In summary, biomedical ceramics are an interesting class of materials but require a lot of attention and further development to aforementioned de ciencies.
Polymer-ceramic composites i.e. ZrO 2 -PEEK (polyether ether ketone), SiO 2 -PMMA (poly(methyl0 methacrylate) as biomaterials are an interesting alternative to ceramics and metals. These materials aim to combine the required properties of the ceramic and polymer groups and minimize occurrence of structural defects, as much as possible. Particular interesting to this type of composite is connected with durability and tribological resistance of ceramics, and low costs and the resistance to brittle fracture of polymers. Previous literature has focused on oxide ceramics combined with PMMA and PEEK [29,30]. The main problem associated with obtaining valuable polymer-ceramic composites is achieving a secure connection between the ceramic and the polymer [31,32]. Depending on the surface treatment of the ceramic grains, a chemical or mechanical bond can be formed which is the adhesive between the ceramic grains and the polymer resin. Therefore, surface modi cation of ceramics is one of the most important issues in these composites. An important way to improve the bond between the ceramic and the polymer is to obtain a chemical bond between the chemically inert ceramic and the polymer. It is possible through the functionalization of the ceramic surface with functional groups and coupling agents i.e. -OH, -CH 2 HN 2 , -NH 2 , -CF 3 , -COOH, H 3 PO 4 , aminosilanes and uorosilanes [33,34]. The main reason of application of coupling agents is to prevent to the high differences between the surface energy of the hydrophilic ceramic and that of the hydrophobic polymer matrix, which leads to agglomeration and poor dispersion of the ceramic ller particles and in the results preventing in the formation of voids and interfacial defects [34]. There are many ways to create a su ciently large surface area development on the ceramic surface: oxidation, application of characteristic functional groups, silanization, thermal treatment in a gas atmosphere, melt in ltration, ionic liquid etching, sol-gel process, and co-precipitation [35][36][37][38][39]. For example, the chemical etching processes have a large amount of literature focused on chemical etching concerns the preparation of the ZrO 2 surface in a way that enables permanent bonding to hard dental tissues or composite materials [39]. One of the important aspects required for the success of ZrO 2 ceramics is the establishment of proper adhesion between substrate and adherent [39,40]. The gold-standard protocol for resin bonding to glass-ceramics is etching with hydro uoric acid followed by the application of a silane coupling agent (chemical and mechanical adhesion) [39,40]. Acid etching (various concentration and times) has been shown to change the surface micro-morphology of glass and oxide ceramics (many surface defects) [39], the resin adhesion, the increase in HF acid concentration, and the etching time associated with an increase of the surface area available to adhesion with resin [39].
Here, we report a simple and low-cost surface modi cation process for medical-grade ZrO 2 and Al 2 O 3 ceramics using chemical acid in the form of an acid-perchlorate bath. This treatment was carried out in order to prepare ceramic llers for mixing with selected polymers in future, and in consequence preparation of polymer-ceramic composites for biomedical applications. For etching process we use sulphuric acid, nitric acid, perchlorate, and hydro uoric acid, for comparison, baths. This performed research concerns the assessment of the impact of the prepared baths on surface development, chemical and phase composition, as well as the occurrence of characteristic functional groups, i.e. hydroxyl, which positively affect the chemical connection with the polymer species in composite.

Samples Preparation
The test samples were prepared from ceramic powders of alumina (Sigma Aldrich) and zirconia (Acros Organics) with 99.9% submicron powder. The samples were prepared in 3-gram batches. Powder samples without any additional treatment were chemically etched in the following solutions: (I) hot, fresh Piranha solution, (II) sulfuric and nitrogen acid mixture, and (III) hydro uoric acid. The etching solution concentrations and ratios are shown in Table 1. All reagents were purchased from Avantor Chemicals. The following were selected as process variables: concentration and volume ratios of the individual reagents and the etching times. The times were set to 30, 60 and 120 seconds. The ceramic powder samples were placed in a beaker with an etching solution and mixed with a magnetic stirrer. The sample names, process times, and etching baths are given in Table 1. After etching, the samples were quantitatively transferred to a paper lter and washed with deionized water and ltered under pressure. After etching, the samples were quantitatively transferred to a paper lter and washed with deionized water (3 × 200 cm 3 ) and ltered under pressure through lter with a ceramic membrane to remove the remaining etching solution. The samples were then transferred to a dryer and dried without any forced air ow at 80 °C for 24 hrs. After cooling the sample to disintegrate the agglomerates, they were added to a beaker with 2-propanol (Acros Orgnics) and placed in an ultrasonic bath to homogenize them for 15 minutes. The samples prepared in this way were then used for materials characterization. The samples were then again transferred to a dryer and dried without any forced air ow at 80 °C for 24 hrs The gure Fig. 1. shows the entire sample preparation process.

Characterisation
The prepared samples were tested in terms of their morphology, characteristic functional group content, phase composition, and speci c surface development. The morphology and initial elemental composition analysis were performed using scanning electron microscopy equipped with energy-dispersive X-ray detector (SEM/EDX). A high resolution scanning electron microscopes FEI INSPECT f50 (ThermoFisher Scienti c, USA) and JEOL JSM-7610F+ (JEOL, Japan) with Schottky cathode and microanalyser Aztec Ultima Max 65 (Oxford Instruments, UK) were used for these measurements. Both secondary and back scattered electrons were detected for imaging, mapping, and analysing the selected areas. Samples were prepared on stubs covered with carbon tape and coated with gold to optimise resolution and avoid possible charging.
The analysis of characteristic functional groups was performed using Fourier transform infrared (FTIR) spectroscopy. The FTIR spectra of all samples were measured using the attenuated total re ectance (ATR) technique. The samples were pressed by a pressure device on the single-re ection diamond ATR crystal, and spectra were collected using an FT-IR Nicolet iS50 (Thermo Scienti c, USA) spectrometer with a Smart Orbit ATR accessory and a deuterated triglycine sulphate (DTGS) detector. The measurement parameters were as follows: spectral region = 4000 − 400 cm − 1 ; spectral resolution = 4 cm − 1 ; 64 scans; Happ-Genzel apodisation.
The X-ray powder diffraction (XRD) analysis was performed using a RIGAKU Ultima IV diffractometer, with a scintillation detector, CuKα radiation source, NiKβ lter, and Bragg-Brentano arrangement. Samples were measured in ambient atmosphere using re exion mode (conditions: 40 kV, 40 mA, 2°/min, 0.05 step). The database used for qualitative phase analysis was ICDD PDF-2/Release 2011 RDB.
Speci c surface development was determined by constructing nitrogen adsorption/desorption isotherms, which were measured using Autosorb (Quantachrome Instruments iQ2). Prior to the analysis, all samples were degassed for 3.6 h at 300 °C. The speci c surface area was calculated using the Brunauer-Emmett-Teller (BET) method. Porosity and pore size distribution was determined using the Barrett-Joyner-Halenda (BJH) method, applying the software from the producer of the apparatus. Approximate pore shapes were speci ed based on de Boer adsorption hysteresis classi cation. Samples with relatively high degrees of agglomeration or aggregation, and those containing residues from etching agents were eliminated from further study.
The zirconia samples displayed a general tendency to agglomerate independently of the etching agent. Among all etching agents, the largest degrees of agglomeration and aggregation were observed when using a mixture of sulphuric and nitric acids; less agglomeration was noted following the use of hydro uoric acid. The smallest degree of agglomeration was observed in powders that were etched with Piranha solution. In general, less agglomeration was observed in alumina samples, because these powders were more dispersed and therefore less susceptible to agglomeration. However, the dependence of alumina agglomeration on the etching agent was similar to that observed for the zirconia samples. Speci cally, most alumina agglomerates and aggregates were observed following treatments with a mixture of sulphuric acid and nitric acid, less were observed for hydro uoric acid, and the least for Piranha solution. Figure 3 presents a comparison of the degree of agglomeration in zirconia and alumina samples for all etching agents.
The second important factor that was assessed using microscopic observations was the presence of impurities/residues from the etching process. After analysing all sample variants, it was discovered that only samples that had been etched with a sulphuric and nitric acid mixture contained some residue, and these samples contained only sulphur residue after etching. Depending on the type of ceramic, sulphur was found in various forms. In the case of zirconia, sulphur-containing impurities took the form of layerlike precipitates on grains, whereas, in the case of alumina, sulphur impurities exhibited more diverse geometries. Figures 4 display examples showing sulphur contamination for zirconia and alumina, and the weight percentage of sulphur detected using EDS semi-quantitative analysis in samples ZrO2_SN1_60 and Al2O3_SN1_60 were 0.5 and 2.3, respectively.
The effect of etching agent on the development of zirconia and alumina surfaces was examined under electrons. For zirconia, the most visible impact of etching on the material's morphology occurred using Piranha solution, which also caused the least agglomeration. Additionally, there was a signi cant degree of surface structural changes following etching using a mixture of sulphuric acid and nitric acid, while surprisingly, the least-altered zirconia structure was observed upon hydro uoric acid etching. The situation was different in the alumina samples, which showed the least structural change after etching with Piranha solution, partial etching of the structure caused by a mixture of sulphuric and nitric acid, and the largest changes in the structure were observed after etching with hydro uoric acid.
Based on the observations for zirconia powders, it was determined that the Piranha solution at a concentration of Cp = 100% and an etching time of 120 s etched the surface the most, causing the largest surface structural changes. As the concentration of the etching agent decreased or the time was reduced, the zirconia was less etched. In the case of a mixture of nitric and sulphuric acid, the zirconia surface was generally disrupted poorly, regardless of time and concentration. After etching with hydro uoric acid, the samples exhibited greater surface changes with increasing concentration and time. As mentioned above more developed surface was obtained for Piranha solution. For alumina, the Piranha solution caused a similar degree of change to the surface of the samples, regardless of time and concentration; the morphology of alumina samples was changed, but no trend is clear.
The situation was similar after etching alumina samples with a mixture of sulphuric and nitric acid, although these surfaces were etched to a lesser degree than those subjected to Piranha solution. The best results were obtained applying treatment with hydro uoric acid at a concentration of Cp = 15% and an exposure time of 120 s. In general, longer times and higher HF concentrations caused the grain surface to change more, and the original spatial structure became distorted. Figure 6 shows the most etched structures that were obtained in these experiments. In these images, the degree of structural change after etching is clear in both the zirconia and alumina samples. Both images display etched and non-etched structures, the latter of which is especially visible in the alumina photo, where at hexagonal crystals can be observed (Fig. 5b). In contrast, the ball-like grains in the zirconia sample have changed to a sponge-like form after etching (Fig. 5a).

Structural characterization
The FTIR spectra for all ten samples are shown in Fig. 6 [41][42][43]. In addition, it was determined that the bands at 450-550 cm − 1 and 750-950 cm − 1 correspond to tetragonal zirconia. The monoclinic phase cannot be determined decisively from the FTIR spectrum [40]. 3.3. Surface size and porosity properties Figure 9 shows the adsorption and desorption isotherms, as well as the multi-point BET for chosen samples. In addition, Table 2 presents the data from measuring speci c surface area, porosity, and pore shapes of selected samples. Samples etched in mixture of sulfuric acid and nitric acid were excluded because they contained impurities in the form of residual sulphur from sulphuric acid, which could not be removed from the ceramic grains. In general, the zirconia samples underwent greater surface development than alumina after etching with any of the respective acids tested herein, although the results show a large range of values. As for zirconia, the samples' pore volumes tended to be greater than 0.4 cm 3 g − 1 . Overall, the observations regarding surface developments indicated that the etching agents have ful lled their task. However, it is di cult to compare the obtained results with published data from other researchers, because limited information could be found related to the effects of any pickling baths on the development of the speci c surfaces and/or the formation of functional groups. Only one report from Aida et al. refers to surface treatment of zirconia in a similar way, but the results are only relevant for ceramic blocks [47]. The alumina samples in this study exhibited a lesser degree of surface development in each case. This can be explained by the fact that the initial zirconia has a lower susceptibility to etching and greater neness than the original alumina. In the case of sample ZrO2_HF0.15_120 (Table 2), no reliable pore data was obtained to allow comparison to other samples.

Conclusions
The main objective of this study was to use etching processes to achieve the surface development of medical grade zirconia and alumina for use as llers in polymer-ceramic composites. Such development of the ceramic surfaces is intended to allow for a permanent mechanical connection with polymer resins, which is a current challenge in the eld. For this purpose, three etching agents were used; (I) fresh and hot Piranha solution, (II) a mixture of 98% H 2 SO 4 and 65 Cp% HNO 3 acids, and (III) 45 Cp% hydro uoric acid.
The rst two acidic etching solutions were diluted in water to ratios of 100, 50, and 25, and HF was diluted in water to ratios of 15, 10, and 5 to provide various experimental conditions. For each acid solution, etching times of 30, 60, and 120 seconds were selected. These experimental durations were chosen because, based on available literature data, it was concluded that they should be su cient to modify the surface of the ceramic grains to an extent that they can make connections with polymers [49]. Based on the observed morphologies, and the chemical and phase composition studies, the following relationships were elucidated: Impact on morphology and surface development. SEM images of zirconia and alumina revealed that the most bene cial surface structural effects are obtained after using Piranha solution, whereas the H 2 SO 4 /HNO 3 solutions had the least in uence on surface development. In the case of HF, the effect is indirect. However, the acid type clearly has a strong impact on ceramic grain agglomeration.
Depending on the acid's oxidation power, different degrees of agglomeration are introduced, and as the oxidation power decreases, the least agglomeration was observed (using HF in both zirconia and alumina samples). Regardless of the acid used, a larger extent of agglomeration was always observed in zirconia, most likely due to a higher degree of short-range weak van der Waals-type interactions, relative to alumina. Based on BET speci c surface development measurements and their correlation with SEM results, the highest speci c surface development was observed for zirconia etched in HF. Slightly lower values were obtained after using Piranha solution, while samples etched in H 2 SO 4 and HNO 3 were not considered at all due to sulphur impurities that could not be removed with the available methods. For alumina, the BET results revealed a more complicated relationship. Since these ceramic powders were already larger than zirconia, the alumina tiles were broken down into agglomerates, and it was di cult to disintegrate them. Therefore, even after grinding the samples with an agate mortar and pestle, and conducting the BET analysis, it was di cult to obtain a concrete conclusion.
Effects on chemical composition and content of characteristic functional groups. After etching and EDX testing, sulphur impurities were detected in samples etched with a mixture of H 2 SO 4 and HNO 3 .
Etched ceramics took different forms (Fig. 4) depending on the type and conditions. The compounds formed on the surface of zirconia and alumina (i.e., inorganic salts sulphate), may have been formed by reactions with dissociated nitric acid. FTIR analysis was carried out in order to verify the presence of characteristic functional groups on the ceramic surface, which can positively in uence the chemical interaction between ceramics and polymers in a future composite. It was possible to identify hydroxyl groups in some of the samples, especially those that were etched in Piranha solution. The occurrence of hydroxyl groups permanently bound to the surface of ceramics positively in uences their abilities to chemically interact with phenolic and polyphenolic formaldehyde resins, and after dehydration (i.e., losing water), they favour formation of carbonyl bonds.
Impact on phase composition. After acid etching, no major changes to the samples' phase compositions were observed that would prevent the use of modi ed alumina and zirconia in medical applications. The phase composition was only slightly altered after etching samples with Piranha solution for 120 seconds. Content of new sulphur-containing crystalline phases on these samples was small but it was detected by SEM/EDX.