Introduction

Commercially pure titanium (CP Ti) has been recognized in dentistry for its biocompatibility, good mechanical properties and corrosion resistance (1-3). Due to this, the application of CP Ti in prosthodontics dates back to the 1980s and the works of Ida et al (4). Metal ceramic restorations have been a golden standard both for long term aesthetic and functional restorations (5). Using CP titanium as a base material for metal ceramic systems has been a challenge due to its high melting temperature (1668 ^oC) and a tendency to form a thick oxide surface layer. This is due to the phase transformation of titanium form α to β structure at a temperature of 880 ^oC (6-8). These properties cause high manufacturing costs and lower bond strength values in comparison to other base metals used in metal-ceramic systems (9-12).

Powder metallurgy (P/M) is a series of manufacturing processes which involve mixing, pressing and heating of various powdered metals and alloys into a desired shape. Powder metallurgy has been taken into account for lowering the cost of titanium parts (13). P/M titanium and its alloys can be classified, according to the adoption of raw powder, into three categories: pre-alloyed P/M Ti alloys, rapid solidified P/M Ti alloys, and blended elemental P/M Ti alloys. Use of blended elemental powder is much more cost-effective, due to cheap Ti and other elemental powders. Commonly used titanium powder includes sponge Ti fines and the hydrogenation and dehydrogenation (HDH) Ti powder (14). Among the powder metallurgy techniques, cold (CIP) and hot (HIP) isostatic pressure and hot deformation processes of metallic powder, which involve the simultaneous application of pressure and temperature, result in engineering components with higher relative density compared to the conventional sintering, or even fully dense materials (13, 14).

Electric discharge machining EDM utilizes a thermo electrical process in which material is removed from work piece by applying the heat energy of sparks. Electrical discharge is repeated between two electrodes (tool and work piece) in the presence of a dielectric fluid. The temperature of the area under spark increases. As a result, the materials melt and vaporize from localized area by using spark energy (15-17).

Machining CP Ti by conventional machining methods has some disadvantages such as high cutting temperature and high tool wear ratio (15). Thus, CP Ti is classified as “difficult-to-machine” material. Therefore, unconventional machining processes, including wire EDM (WEDM) are introduced for machining CP Ti (16-18).

Strong bonding to porcelain is essential for long term success of metal ceramic restorations (19). Bond strength between substructure metals and veneering ceramics can be tested using various methods. Several authors have proposed the use of the shear bond test for strength evaluation (20-23). However, since the International Organization for Standardization (ISO) standard for metal ceramics advocates a three-point bending test, almost all recent studies have used this method to evaluate bond strength (24-31). The three point bending test according to Schwickerath, described in the ISO 9693, requires metal samples with dimensions (25 ± 1) mm × (3 ± 0.1) mm × (0,5 ± 0.05) mm (24). So far, casting and milling of commercially available CP Ti has been used for manufacturing of such samples. Also, almost all of the studies found used grinding and polishing of the samples after manufacturing to obtain the required dimensions. The aim of the present study was to evaluate the effects of WEDM manufacturing process on the surface quality of the P/M CP Ti samples manufactured for ISO 9693 bond strength testing.

Materials and methods

CP Ti material was prepared using Ti 99.4% HDH powder with particle size <150 µm. In first step CIP was used at 200 MPa at room temperature, later hot vacuum pressed at 420 °C and 330 MPa and finally consolidated by direct extrusion at 500 °C to 13 x 13 mm bars using reduction area ratio 1:4.2. A total of 3 bars dimensions 150 x 13 x 13 mm were manufactured. From these bars samples were manufactured in the dimensions proposed by ISO 9693 using wire EDM. The basic parts of the WEDM machine consist of a wire electrode, a work table, and a servo control system, a power supply and dielectric supply system. In this study, copper wire of 0.25 mm diameter was used as a tool and distilled water was used as a dielectric fluid. The other input parameters important for surface quality after cutting were gap voltage (40 V), wire feed (5m/min), pulse on time (5s), and pulse off time (24s). These parameters remained constant for all samples during WEDM.

Eight samples were made and divided into 2 groups. The first group was untreated after WEDM, and the second was ground by use of silicon carbide papers P320-P4000 (Struers Inc., West Lake, US). Both groups were analysed using SEM, EDS and XDR.

Results

X-ray structural analysis

Two samples, a prepared sample (denoted S1) and a ground one, denoted S2, were analysed by X-ray powder diffraction (XRPD). Graphical result of Rietveld refinement for both samples is shown in Figure 1.

Figure 1 Graphical presentation of the final Rietveld refinement of room temperature data for samples S1 (up) and sample S2 (down). Experimental data are shown in red, the calculated pattern is blue while vertical marks, green, magenta and turquoise, represent the diffraction reflections for crystal phases TiO, α-Ti and β-Ti, respectively.

Rietveld refinements were carried out by using starting structural models from ICSD data base: 28955, 647556 and 151409 for TiO, α-Ti and β-Ti, respectively. Quantitative and qualitative analysis of sample S1 revealed that sample contained TiO as dominant phase (64.3 wt%), α-Ti was present in smaller amount (34.1 wt%) while β-Ti can was present only in traces (1.5 wt%). XRD pattern of sample S2 (obtained by mechanical treatment of S1) showed that its composition changed drastically upon mechanical treatment. Sample S2 was a completely single-phased sample containing only α-Ti (100 wt%) and no additional phases present. α-Ti crystallized in hexagonal crystal system, in space group P 63/mmc with unit-call parameters a= 2.9555(2) and c=4.6893 (3).

SEM and EDS analyses

Typical surface of P/M CP Ti samples machined by WEDM in cross section is shown in Figure 2. The base material of sample is covered by thin layer of material with different composition, arrow in Figure 2. In this layer a lot of fractures are visible, directed from the surface of material to the base (Figure 3). The results of EDS analysis of thin layer on cutting surface of Ti samples are presented in Figure 4 and show that the layer consists of Ti, Cu, C and O probably titanium and copper oxides and some carbide. The surface area of samples after grinding and removal of 20µm of surface material is shown in Figure 5. The Figure shows uniform composition of material and without visible fractures. The results of EDS analysis show a high level of pure Ti with no oxides (Figure 6). For better illustration purposes, the line of EDS analysis of Ti and Cu concentrations in cross section of untreated samples is shown in Figure 7.

Figure 2 SEM cross section surface of titanium samples machined by WEDM

Figure 3 SEM cross section and location of EDS analysis on surface layer

Figure 4 Results of EDS microanalyses of the layer’s chemical composition are presented in spectrum

Figure 5 SEM cross section and location of EDS analysis after grinding

Figure 6 Results of EDS microanalysis of surface layer’s chemical composition after grinding are presented in spectrum

Figure 7 Line analysis of surface layer of untreated samples with concentrations of Ti and Cu

Discussion

Bonding of CP Ti to dental ceramics is a difficult and sensitive process. Titanium surface structure and composition are crucial to establishing a good bond. Only a pure α-phase structure with a thin oxide layer is acceptable to obtain good bonding strength results (11). In the first part of this study, the WEDM manufactured samples of P/M CP Ti showed an inadequate surface layer. The layer had a large number of fractures and the composition showed traces of other elements (Cu) and oxides which could compromise the bond between titanium and dental ceramics. The cause of these fractures was most probably internal stress caused by very high temperature gradient in material during the WEDM. Also, the findings of Cu in the thin surface layer can be explained with the high temperature and the reaction between the cooper wire of the WEDM machine and the sample surface. High levels of oxides are also attributed to the high temperature which is higher (>880 ^oC) than the titanium phase transformation from α to β. The finding of C is attributed to sample preparation. XDR analysis of the untreated samples showed a high level of titanium oxides on the surface and a smaller amount of α and trace amounts of β-phase titanium. Because of its high reactivity with oxygen β-phase, titanium quickly forms a thick oxide layer on the surface. It can be assumed that all of the titanium oxide that was found on the surface of the samples came from the large amount of β-phase titanium. Several studies emphasised that β-phase and its reaction to form a thick oxide layer compromised the bond and caused adhesive and cohesive fractures between titanium and dental ceramics (7, 8, 11, 25-31).

After grinding the surface, quality significantly improved. Analysis showed no oxides or other impurities which presented a suitable surface for ceramic bonding. The effects of high temperature are limited to a thin layer on the surface (10 µm) (Figure 3). Such an improvement after removal of a 20 µm layer thickness of material (Figure 5) can be explained by the adequate cooling during the WEDM process and low heat conductivity of titanium (10). It also shows that WEDM is an excellent method of machining samples for ISO 9693 if accompanied by grinding and removing a thin surface layer.

Conclusion

Within the limitations of this study it can be concluded: