SEM, EDS and XPS Analysis of the Coatings Obtained on Titanium after Plasma Electrolytic Oxidation in Electrolytes Containing Copper Nitrate

In the paper, the Scanning Electron Microscopy (SEM) with Energy Dispersive X-ray Spectroscopy (EDS) and X-ray Photoelectron Spectroscopy (XPS) results of the surface layer formed on pure titanium after plasma electrolytic oxidation (micro arc oxidation) at the voltage of 450 V are shown. As an electrolyte, the mixture of copper nitrate Cu(NO3)2 (10–600 g/L) in concentrated phosphoric acid H3PO4 (98 g/mol) was used. The thickness of the obtained porous surface layer equals about 10 μm, and it consists mainly of titanium phosphates and oxygen with embedded copper ions as a bactericidal agent. The maximum percent of copper in the PEO surface layer was equal to 12.2 ± 0.7 wt % (7.6 ± 0.5 at %), which is the best result that the authors obtained. The top surface layer of all obtained plasma electrolytic oxidation (PEO) coatings consisted most likely mainly of Ti3(PO4)4∙nH3PO4 and Cu3(PO4)2∙nH3PO4 with a small addition of CuP2, CuO and Cu2O.


Material, Electrolyte and Setup
The plasma electrolytic oxidation of commercial purity titanium (CP Ti Grade 2) was performed at a voltage of 450 V for 3 min. The main elements of the setup were: a processing cell, a three-phase transformer with Graetz bridge, the electrodes and connecting wiring. The scheme of the set-up used was presented earlier in [38]. The titanium samples of dimensions of 30ˆ10ˆ1 mm were employed for the studies. Current density for constant voltage was not recorded, but it was observed to decrease over time. A cylinder made of AISI 316L stainless steel was used as a cathode.
For the studies, a mixture of 1000 mL of concentrated orthophosphoric H 3 PO 4 acid (85%) with 10 up to 600 g/L of copper nitrate Cu(NO 3 ) 2 electrolytes, were used. For each run, the electrolytic cell made of glass was used, containing up to 500 dm 3 of the electrolyte solution.

Set Ups for SEM and EDS
Scanning electron microscope (SEM), FEI Quanta 650 FEG (Field Electron and Iron Company (FEI), Hillsboro, OR, USA), equipped with energy-dispersive X-ray spectroscopy (EDS, FEI), for surface analysis was used. The microscope operated under the following conditions: voltage 15 kV, current 8-10 nA, beam diameter 6 µm, decreased vacuum in the chamber with the pressure of 50 Pa. The identification of spectral lines was performed by means of a spectral decomposition with holographic deconvolution using a targeted peak deconvolution function. All of the EDS data obtained by Statistica 10 [39] were processed.

XPS Studies
The XPS measurements on oxidized titanium samples were performed by means of the SCIENCE SES 2002 instrument (SCIENTA AB, ScientaOmicron, Uppsala, Sweden) using a monochromatic (Gammadata-Scienta) Al Kα (hν = 1486.6 eV) X-ray source (18.7 mA, 13.02 kV). Scan analyses were carried out with the measurement area of 3ˆ1 mm 2 and a pass energy of 500 eV with the energy step of 0.2 eV and the step time of 200 ms. The binding energy of the spectrometer has been calibrated by the position of the Fermi level on a clean metallic sample. The power supplies were stable and of high accuracy. The experiments were carried out in an ultra-high-vacuum system with a base pressure of about 6ˆ10´1 0 Pa. The XPS spectra were recorded in normal emission. For the XPS analyses, the Casa XPS 2.3.14 software (Shirley background type [11]) was used. All binding energy values presented in this paper were charge corrected to C 1s at 284.8 eV. All XPS spectra of titanium (Ti 2p), phosphorus (P 2p), copper (Cu 2p), as well as oxygen (O 1s) and carbon (C 1s) with 9 sweeps were performed. Additionally, for the copper Cu 2p spectra, the measurements were repeated by 36 sweeps to increase the signal-to-noise ratio to 6. For the interpretation of the deconvoluted spectra, the literature positions [40,41] were applied.

Results
In Figure 1, the SEM and EDS results of the surface layer formed on titanium after PEO at the voltage of 450 V in electrolyte consisting of 1 L H 3 PO 4 with the addition of (1) 10 g/L; (2) 300 g/L and (3) 600 g/L of Cu(NO 3 ) 2 are presented. All SEM photos show that the obtained surfaces are porous. Additionally, on the surface treated in the electrolyte consisting of 300 g/L of copper nitrate, the cracks in the surface layer are visible beyond the pores, which may be indicative of too rapid growth in volume of the layer. Interestingly, the increasing of the copper nitrate up to 600 g/L does not result in such cracks. Most likely, the copper ions inserted in the porous phosphorus-oxygen-titanium structure create the original modified surface layer. On the basis of EDS peaks, it can be concluded that, with the increase of copper nitrate in concentrated phosphoric acid, the amount of titanium in the surface layer decreases, whereas the increase of phosphorus and copper amounts is observed.
In Figure 2 and Table 1, the box and whisker plots with the descriptive statistics of the amount of copper in the surface layer formed on titanium after PEO are presented, respectively. One may notice that the copper amount in the surface layer increases with increasing the content of copper nitrate in the electrolyte. The minimum amount of copper in the surface layer was detected after the PEO performed in the electrolyte containing 10 g/L Cu(NO 3 ) 2 and was equal to 1.7˘0.4 wt % (1.2˘0.3 at %), and the maximum was found in the electrolyte containing 600 g/L Cu(NO 3 ) 2 in H 3 PO 4 , and equaling 12.2˘0.7 wt % (7.6˘0.5 at %). The significance tests in Statistica software (Version 10) [39] for all data were performed and show that the amounts of copper in the surface layer formed on titanium after PEO in the electrolytes containing various amounts of copper nitrate belong to different data populations. Therefore, it is possible to conclude that there are significant differences in the copper amount in the passive layer between the treated samples.
In Figure 3 and Table 2, the box and whisker plots with the descriptive statistics of the amount of phosphorus in the surface layer formed on titanium after PEO are shown, respectively. It is visible that the phosphorus amount (similar as copper) increases with the increase of the copper nitrate in the electrolyte. The minimum amount of phosphorus in the surface layer was detected after the PEO performed in the electrolyte containing 10 g/L Cu(NO 3 ) 2 and was equal to 8.9˘0.6 wt % (13.2˘0.9 at %), and the maximum was found for 600 g Cu(NO 3 ) 2 in 1000 mL H 3 PO 4 , equaling 42.9˘1.4 wt % (54.9˘1.1 at %). The significance tests presented in Table 2 show that the amounts of phosphorus in the surface layer formed on titanium after PEO in the electrolytes containing various amounts of copper nitrate belong to different data populations. Following that, it is possible to conclude that there are significant differences in the phosphorus amount in the passive layer between the treated samples. The study shows that the increasing concentration of copper nitrate in the phosphoric acid results in the increase of both copper, as well as phosphorus in the surface layer. These phenomena can be explained by the formation of chemical compounds between titanium, phosphorus, copper and oxygen, which are insoluble in the concentrated phosphoric acid.      In Figure 4 and Table 3, the box and whisker plots with the descriptive statistics of the amount of titanium in the surface layer formed on titanium after PEO are given, respectively. The maximum amount of titanium in the surface layer was detected after the PEO done in the electrolyte containing 10 g/L Cu(NO 3 ) 2 and was equal to 89.4˘0.9 wt % (85.6˘1.1 at %), whereas the minimum was found at 600 g/L Cu(NO 3 ) 2 , equaling to 42.2˘1.5 wt % (37.5˘1.4 at %). This may suggest that for 10 g/L Cu(NO 3 ) 2 in one liter of phosphoric acid, the coating is much thinner than the one formed in the electrolyte within 600 g/L of copper nitrate. The significance tests presented in Table 3 show that the amounts of oxygen in the surface layer formed on titanium after PEO in the electrolytes containing various amounts of copper nitrate mostly do not belong to different data populations. That finding allows for the conclusion that generally, there are no significant differences in the oxygen amounts in the passive layer between the treated samples.  For example, Figure 5 presents a cross-section of the surface layer obtained on pure titanium after PEO at the potential of 450 V in the electrolyte consisting of 300 g of copper nitrate dissolved in 1000 mL of concentrated phosphoric acid (98 g/mol). One may easily notice that the obtained surface layer is heterogeneous and porous, and its thickness is equal to about 10 μm. Additionally, one may presume that it is more biocompatible than a pure titanium surface and can be used as a transition layer between the titanium biomaterial and tissue.   For example, Figure 5 presents a cross-section of the surface layer obtained on pure titanium after PEO at the potential of 450 V in the electrolyte consisting of 300 g of copper nitrate dissolved in 1000 mL of concentrated phosphoric acid (98 g/mol). One may easily notice that the obtained surface layer is heterogeneous and porous, and its thickness is equal to about 10 µm. Additionally, one may presume that it is more biocompatible than a pure titanium surface and can be used as a transition layer between the titanium biomaterial and tissue. For example, Figure 5 presents a cross-section of the surface layer obtained on pure titanium after PEO at the potential of 450 V in the electrolyte consisting of 300 g of copper nitrate dissolved in 1000 mL of concentrated phosphoric acid (98 g/mol). One may easily notice that the obtained surface layer is heterogeneous and porous, and its thickness is equal to about 10 μm. Additionally, one may presume that it is more biocompatible than a pure titanium surface and can be used as a transition layer between the titanium biomaterial and tissue.  To find out the chemical composition of coatings formed during the PEO process, the XPS measurements were performed. With the help of that method, the oxidation stages of inter alia titanium, phosphorus and copper were possible to find.
In Figure 6, the fittings of C 1s spectra are shown. With this analysis, it is possible to determine how much oxygen is bound to the carbon contamination nano-layer and how much of this is bound with other chemical elements of the passive surface layer. All C 1s spectra were deconvoluted by four peaks: C-C/C-H (284. In Figure 6, the fittings of C 1s spectra are shown. With this analysis, it is possible to determine how much oxygen is bound to the carbon contamination nano-layer and how much of this is bound with other chemical elements of the passive surface layer. All C 1s spectra were deconvoluted by four peaks: C-C/C-H (284.8 eV), C-O-H/C-O-C (286.3 eV), C=O (287.7 eV), O-C=O (288.8 eV). On the basis of the obtained results, it was possible to separate oxygen bonded with carbon (contamination carbon layer) and with the chemical elements contained in the passive layer. In Table 4, there are enlisted results of the passive PEO layer obtained on pure CP titanium after the electrochemical plasma oxidation (PEO) in the electrolyte consisting of Cu(NO3)2 and H3PO4. It was found that in the PEO layer (of about 10-μm thick), mainly the titanium-copper phosphate compounds are visible. Assuming that all of the oxygen in the passive layer is bonded with phosphorus to form phosphates, then one may presume that not all of the phosphorus was used. In all of the PEO layers, a minimum of 7 at % of the phosphorus, which was not bonded with the oxygen, was found. Hence, the phosphorus must form anaerobic compounds with the titanium or copper. Based on Table 4, it can be noticed that the amount of copper in the PEO layer increases very slowly, but regarding the bactericidal properties, its amount is quite satisfactory. It should be noted that the XPS measurements allow one to study only the first 10 nm of the PEO layer; however, in accordance with the recent results revealed in [1], its thickness may be of a few micrometers.  Figure 7, the XPS high resolution spectra of titanium (Ti 2p), copper (Cu 2p), oxygen (O 1s) and phosphorus (P 2p) are presented. Concerning these results, it is possible to predict, with a high probability, what chemical compounds of titanium were formed after the PEO treatment. The P 2p spectra indicate that in the surface layer formed after the electrochemical PEO treatment with the addition of 300 g/L and 600 g/L Cu(NO3)2 to 1 L H3PO4, the highest/main peak has binding energy equal to about 133.9 eV, suggesting the presence of phosphates (PO4 3− ) in the studied surface layer. In the case of oxidation in the electrolyte containing 10 g/L Cu(NO3)2 in 500 mL H3PO4, apart from PO4 3− , an additional peak is visible, whose binding energy is about 135.2 eV. That can suggest the In Table 4, there are enlisted results of the passive PEO layer obtained on pure CP titanium after the electrochemical plasma oxidation (PEO) in the electrolyte consisting of Cu(NO 3 ) 2 and H 3 PO 4 . It was found that in the PEO layer (of about 10-µm thick), mainly the titanium-copper phosphate compounds are visible. Assuming that all of the oxygen in the passive layer is bonded with phosphorus to form phosphates, then one may presume that not all of the phosphorus was used. In all of the PEO layers, a minimum of 7 at % of the phosphorus, which was not bonded with the oxygen, was found. Hence, the phosphorus must form anaerobic compounds with the titanium or copper. Based on Table 4, it can be noticed that the amount of copper in the PEO layer increases very slowly, but regarding the bactericidal properties, its amount is quite satisfactory. It should be noted that the XPS measurements allow one to study only the first 10 nm of the PEO layer; however, in accordance with the recent results revealed in [1], its thickness may be of a few micrometers. In Figure 7, the XPS high resolution spectra of titanium (Ti 2p), copper (Cu 2p), oxygen (O 1s) and phosphorus (P 2p) are presented. Concerning these results, it is possible to predict, with a high probability, what chemical compounds of titanium were formed after the PEO treatment. The P 2p spectra indicate that in the surface layer formed after the electrochemical PEO treatment with the addition of 300 g/L and 600 g/L Cu(NO 3 ) 2 to 1 L H 3 PO 4 , the highest/main peak has binding energy equal to about 133.9 eV, suggesting the presence of phosphates (PO 4 3´) in the studied surface layer.
In the case of oxidation in the electrolyte containing 10 g/L Cu(NO 3 ) 2 in 500 mL H 3 PO 4 , apart from PO 4 3´, an additional peak is visible, whose binding energy is about 135.2 eV. That can suggest the presence of H 3 PO 4 acid molecules in the surface layer. In the case of Ti 2p 3/2 spectra, the maximum binding energies are noted in the points 460.4 eV, 459.9 eV, 459.6 eV referred to 10 g/L, 300 g/L, 600 g/L Cu(NO 3 ) 2 in 1000 mL H 3 PO 4 , respectively. The Ti 2p 3/2 spectrum combined with P 2p spectrum can suggest that the titanium detected in the surface layer is on the fourth oxidation stage and most likely appears as the Ti 3 (PO 4 ) 4¨n H 3 PO 4 compound. After the PEO oxidation in the electrolyte containing lower contents of the copper nitrate (10-300 g/L) in the phosphoric acid (1000 mL), the XPS signal of Cu 2p from the surface layer is on the noise level. Only by increasing the amount of the copper nitrate up to 600 g per 1000 mL of phosphoric acid, a distinct spectrum of Cu 2p was recorded.
spectrum can suggest that the titanium detected in the surface layer is on the fourth oxidation stage and most likely appears as the Ti3(PO4)4•nH3PO4 compound. After the PEO oxidation in the electrolyte containing lower contents of the copper nitrate (10-300 g/L) in the phosphoric acid (1000 mL), the XPS signal of Cu 2p from the surface layer is on the noise level. Only by increasing the amount of the copper nitrate up to 600 g per 1000 mL of phosphoric acid, a distinct spectrum of Cu 2p was recorded. In the case of a low signal-to-noise ratio, i.e., three for nine sweeps (Figure 8), the authors decided to repeat XPS high resolution scans for the Cu 2p3/2 region with 36 sweeps (signal-to-noise ratio equals six). Based on the obtained results, which are depicted in Figure 8, it should be noted that for the lowest amount of copper nitrate (10 g/L) in phosphoric acid, the signal is still very noisy; hence, it was not possible to perform the correct fitting. In the case of a low signal-to-noise ratio, i.e., three for nine sweeps (Figure 8), the authors decided to repeat XPS high resolution scans for the Cu 2p 3/2 region with 36 sweeps (signal-to-noise ratio equals six). Based on the obtained results, which are depicted in Figure 8, it should be noted that for the lowest amount of copper nitrate (10 g/L) in phosphoric acid, the signal is still very noisy; hence, it was not possible to perform the correct fitting. In Figure 8 and Table 5, the high resolution Cu 2p3/2 spectra for two high signals (300 g and 600 g of Cu(NO3)2 per 1000 mL H3PO4) are presented. The two first peaks at 932.3 eV and 932.9 eV (Figure 8a) and 932.4 eV (Figure 8b)   In Figure 8 and Table 5, the high resolution Cu 2p 3/2 spectra for two high signals (300 g and 600 g of Cu(NO 3 ) 2 per 1000 mL H 3 PO 4 ) are presented. The two first peaks at 932.3 eV and 932.9 eV (Figure 8a) and 932.4 eV (Figure 8b) and for both analyzed electrolytes can be interpreted most likely as CuP 2 and Cu 2 O, respectively. Other peaks of the ingrown energies are also responsible for the presence of Cu 2+ , i.e., 933.6, 934.1 eV are responsible mostly for CuO; 934.8, 935.1, 936.1, 937.2, 938.1 eV peaks may be responsible for the presence of Cu 2+ , most likely as the CuO¨Cu 3 (PO 4 ) 2¨n H 3 PO 4 compound in the PEO layer. Following that path, it should be noted that the more copper nitrate in electrolyte, the more anaerobic (CuP 2 ) and aerobic phosphorus-copper compounds (Cu 3 (PO 4 ) 2¨n H 3 PO 4 ) are in the PEO surface layer. The authors have also calculated the Cu/P, Cu/Ti and P/Ti ratios, and they are presented in Table 6. The PEO layer has been characterized by the dimensionless numbers. This is important because they can be used for the comparison of other surface layers containing phosphorus and copper incorporated. Table 6. Cu/P, Cu/Ti and P/Ti ratios gained on the basis of XPS results.

Conclusions
The plasma electrolytic oxidation of titanium as a biomaterial at a voltage of 450 V in electrolyte consisting of copper nitrate and concentrated phosphoric acid allows one to obtain the porous surface layer with the thickness of some micrometers. The most important fact is that inside of the surface layer, there are embedded copper ions, which are known to have a bactericidal effect. The biggest achievement of the work is gaining the percent of copper in the porous PEO surface layer reaching up to 12.2˘0.7 wt % (7.6˘0.5 at %), which should be considered as a good result. Increasing the copper nitrate amount in the solution resulted in the growth of the copper and phosphorus amounts in the produced surface layer. The study on the effect of copper ions on the composition and structure of the surface layer created on titanium during the PEO process will be further continued. The aim is to explain and understand the mechanisms of the surface layer formation. Most likely, the top surface layer of PEO coatings consists mainly of Ti 3 (PO 4 ) 4¨n H 3 PO 4 and Cu 3 (PO 4 ) 2¨n H 3 PO 4 with a small addition of CuP 2 , CuO and Cu 2 O. Physical and mechanical properties, as well as those related to the biocompatibility, of these porous coatings will be the subject of our future works.