The purpose of this study was to measure the porosity in different laser welded cast alloys non-destructively using X-ray micro-focus computerized tomography (micro-CT) and to evaluate the effect of porosity on the tensile strength of the welded joints.
Materials and methods
The welding procedure was conducted in rectangular cast metals, CoCr, Ti and platinum added gold alloy (AuPt). The metal plates were butted CoCr to CoCr (CoCr/CoCr) or Ti to Ti (Ti/Ti) for welding of similar metals and Ti to AuPt (Ti/AuPt) for welding of dissimilar metals. Specimens were welded under several laser-welding conditions; with groove (normal), without groove (no groove), spatter, crack, or no overlapped welding (no overlap) ( n = 5). Porosity in the welded area was evaluated using a micro-CT. Tensile strength of the welded specimens was measured at a crosshead speed of 1 mm/min. Multiple comparisons of the group means were performed using ANOVA and Fisher’s multiple comparisons test ( α = .05). The relationship between the porosity and the tensile strength was investigated with a regression analysis.
Three-dimensional images of Ti/AuPt could not be obtained due to metal artifacts and the tensile specimens of Ti/AuPt were debonded prior to the tensile test. All other welded specimens had porosity in the welded area and the porosities ranged from 0.01% to 0.17%. The fractures of most of the CoCr/CoCr and Ti/Ti specimens occurred in the parent metals. Joint strength had no relationship with the porosity in the welded area ( R 2 = 0.148 for CoCr/CoCr, R 2 = 0.088 for Ti/Ti, respectively).
The small amount of porosity caused by the laser-welding procedures did not affect the joint strength. The joint strength of Ti/AuPt was too weak to be used clinically.
Welding is an important process for fabricating dental prostheses. In the early 1990s, laser-welding was introduced in Germany . In recent years, laser-welding has become an essential technique to repair and add separate metal segments of the frame work for metal prostheses .
The advantages of laser-welding are that the parent metals can be welded without solder and that the parent metal can be used as a solder, if necessary. Corrosion resistance is not reduced by using the same metal for the solder as for the parent metal, and the mechanical strength of the weld joint will not suffer . Zupancic et al. reported that the corrosion resistance of laser-welded joints was better than of brazed ones, primarily due to differences in passivation ability. According to the potentiodynamic curves obtained, the curves of intact CoCr alloy and laser-welded joint CoCr were similar and indicated a nearly passive state, while the potentiodynamic curves corresponding to the brazed joint CoCr showed no distinct passive region. Since laser energy can be concentrated in a small area, there are fewer effects of heating and oxidation on the area surrounding the spot to be welded . It is easy to weld dental prostheses using this method since no additional materials, such as investment material or gas torches that are used for conventional dental soldering, are needed . Laser welding has accelerated the use of Ti and its alloys for prostheses since conventional soldering techniques cannot be used for joining Ti materials because of oxygen contamination during the soldering procedure .
The wave length, peak pulse power, pulse energy, output energy, pulse duration, pulse frequency and spot diameter of the laser affect the mechanical strength of joints . In most of the laser-welding machines used for dentistry, the output energy (current or voltage), pulse duration and spot diameter of the laser can be adjusted. The success or failure of the laser-welding procedure is affected not only by these physical parameters but also by the operator’s experience, dexterity, eyesight and knowledge of the machine . Eyesight influences the adjustment of the laser beam focusing. Currently operators have to make fine adjustments during laser welding, although manufacturers claim the laser-welding technique to be a new, rapid, economic and accurate method for joining metal.
Edge preparation, penetration welding, two-side welding and overlapped welding are common techniques used in welding dental prostheses. Iwasaki et al. reported that one-sided welding caused distortion of specimens and two-side welding decreased this. Watanabe et al. reported that double-welded specimens showed significantly greater joint strength compared to single-welded specimens. Two-side welding is a common technique that does not cause distortion of specimen . The CoCr/CoCr (normal) and Ti/Ti (normal) are commonly used by dental laboratory technicians. Unorthodox techniques, such as no edge preparation and no overlapped welding were added as welding conditions evaluated in the current study. There are some situations when a metal framework is too thin to prepare an edge. Specimens with intentional welding defects, “crack” and “spatter”, were also fabricated. These welding defects can be reproduced by controlling parameters such as current, pulse duration and spot diameter. Cracks appear in a welded area due to thermal residual stress during the welding procedure and/or changes in microstructure that affect the quality of the welded parts .
The presence of inclusions and voids in the welded area is a critical issue during the welding process . In SEM micrographs of fracture surfaces after tensile testing, many small pores were observed in welded regions . Internal defects and porosity in cast structures have been assessed by radiographic and metallographic analyses . Radiographic analysis evaluates only the presence of porosity, but the quantity of porosity cannot be determined. Metallographic analysis of the specimens cannot be performed non-destructively. These analyses are two-dimensional evaluation methods. In contrast, an X-ray micro-focus computerized tomography (micro-CT) can evaluate the porosity three-dimensionally (3D) . Micro-CT, a high-resolution variant of medical CT, allows non-invasive mapping of the microstructure in 3D with spatial resolution approaching that of optical microscopy.
The purpose of this study was to evaluate the porosity of specimens welded under various welding conditions non-destructively using micro-CT. This study was based on the following two research hypotheses. The first was that many air bubbles would not combine in the laser-welded area. The second was that porosity in the laser-welded area would not affect the joint strength of laser-welded alloys.
Materials and methods
Preparation of cast plate for laser-welding
Three alloys, cobalt–chrominium (CoCr), titanium (Ti), and platinum added gold alloy (AuPt), and three joining metals for welding were evaluated ( Table 1 ). Rectangular metal plates were prepared by the lost-wax casting method. The acrylic resin (ShinkoLite; Mitsubishi Rayon Co., Ltd., Tokyo, Japan) patterns (1 × 10 × 20 mm) were invested in casting rings with an investment material for the subsequent bond strength test. For Ti and CoCr plates, an investment material (Multi-Vest; Dentsply Intl. Inc., York, PA) was used. Ti and CoCr were cast using a centrifugal casting machine (Ticast Super R; Selec Co., Ltd., Osaka, Japan). For AuPt, an investment material (Uni-Vest-Silky; Shofu Inc., Kyoto, Japan) was used. AuPt was cast using a casting machine with a gas pressure attached vacuum system (Patcaster; Reburn Ohara, Osaka, Japan). All castings were divested from the molds and the sprues of cast plates were removed. Ti cast plates were pickled with acid solution (Reburn Ohara, Osaka, Japan) for 2 min and CoCr cast plates were electro-polished using an electro-polisher (Elepika; Shofu Inc., Kyoto, Japan) for 2 min. After pickling or electro-polishing, each cast specimen was refined using carborundum points (Shofu Inc., Kyoto, Japan) and airborne-particle-abraded with 50 μm alumina particles (Fujilundum A80; Fuji Manufacturing Co., Ltd., Tokyo, Japan).
|Alloy||Ti ingot JS2||Ti||Selec Co., Ltd., Osaka, Japan||Ti(99.485), H(0.015), O(0.20), N(0.05), Fe(0.25)|
|Filler metal||Laser wire Ti||Ti||Selec Co., Ltd., Osaka, Japan||Ti(99.485), H(0.015), O(0.20), N(0.05), Fe(0.25)|
|Alloy||Lasernium||CoCr||Selec Co., Ltd., Osaka, Japan||Co(58.0), Cr(30.0), Mo(6.0), Fe, Mn, Si, W|
|Filler metal||Lasernium||CoCr||Selec Co., Ltd., Osaka, Japan||Co(68.0), Cr(25.0), Mo(5.0), Fe, Si, Al|
|Alloy||Aurofluid 3||AuPt||Metalor-Technologies.S.A., Neuchatel, Switzerland||Au(71.0), Pt(2.0), Pd(2.0), Ag(9.0), Cu(14.5), Zn(1.5)|
|Filler metal||PGA-12||AuPt||Ishifuku Metal Industry Co., Ltd., Tokyo, Japan||Au(70.0), Pt(3.0), Pd(2.0), Ag(4.7), Cu(20.0)|
Laser parameters and welding conditions
Two experienced dental technicians in laser-welding welded the cast metal plates. The plates were matched and approximated Ti to Ti (Ti/Ti) or CoCr to CoCr (CoCr/CoCr) for the welding of similar metals and Ti to AuPt (Ti/AuPt) for dissimilar welding. The two cast plates were butted against one another longitudinally using a jig and weld-bonded using a Nd:YAG laser-welding machine (ALP-50; Alpha-Laser GmbH, Puchheim, Germany). CoCr and Ti cast plates were divided into 4 and 2 groups, respectively ( n = 5). The welding conditions followed are shown in Table 2 . Welding conditions from CoCr/CoCr (normal) to CoCr/CoCr (crack) and Ti/Ti (normal) and Ti/Ti (no overlap) were for CoCr/CoCr and Ti/Ti, respectively. Welding condition Ti/AuPt (normal) was for Ti/AuPt specimens. In welding conditions, CoCr/CoCr (normal), Ti/Ti (normal) and Ti/AuPt (normal) were the normal condition, and provided edge preparations as shown in Fig. 1 a . Initially the spot welding was conducted at three points; at the center, and one-fourth of the distance bilaterally as a pre-welding procedure. These spot welds were penetration welds, that is, the weld penetration depth was extended over the entire thickness of the specimens. After the pre-welding procedure, the same number of spots was filled up with filler metal in turn from both sides. Welding condition CoCr/CoCr (no groove) was only a butt joint without edge preparations as shown in Fig. 1 b. The laser-welds were overlapped except for welding condition Ti/Ti (no overlap), and the rate of overlap was approximately 90% as shown in Fig. 1 c. Welding conditions CoCr/CoCr (spatter) and CoCr/CoCr (crack) included intentional defects in each specimen. In welding condition CoCr/CoCr (spatter) the current of the spot welding was increased to increase the welding depth and generate the spatters easily. In CoCr/CoCr (crack) the power density for laser-welding was decreased in order to generate cracks. Welding condition Ti/Ti (no overlap) was not overlapped as shown in Fig. 1 d.
|No.||Condition||Current (A)||Pulse duration (ms)||Spot diameter (mm)|
|Filled with filler metal wire||210||5.4||0.7|
|2||CoCr/CoCr||No groove||Spot welding||210||5.4||0.5|
|Filled with filler metal wire||210||5.4||0.7|
|Filled with filler metal wire||210||5.4||0.7|
|Filled with filler metal wire||210||8.0||0.8|
The welding area (1.96 × 10 mm) of all the specimens was observed and analyzed using a micro-CT device (inspeXio SMX-90CT; Shimadzu Co., Kyoto, Japan) before the tensile test. The tube voltage and current of the X-ray generator were 65 kV and 100 μA respectively. X-ray image acquisition was performed with the CCD camera equipped in the micro-CT. One hundred and one sliced images of the specimens were obtained horizontally along the length of the specimens. The thickness of a slice was 0.0196 mm. These images were reconstructed in 3D using image analysis software (TRI/3D-Bon; Ratoc Engineering, Tokyo, Japan). The sizes, numbers and total volume ratios of the bubbles which existed in the specimens are also determined using this software. The diameters of the bubbles are calculated by assuming the pore shape to be spherical. The porosity was calculated from the total volume ratios of the bubbles ( n = 5).
The efficiency of the welding process was measured with a tensile test. Tensile testing was conducted at a crosshead speed of 1 mm/min and gauge length of 10 mm with a universal testing machine (AG-20kN; Shimadzu Co., Kyoto, Japan). The maximum load (N) was recorded when the specimen fractured and the tensile strength (MPa) and their averages and standard deviations were calculated ( n = 5). The modes of failure were visually evaluated and categorized as cohesive failure, interfacial failure, or mixed.
Multiple comparisons of the group means were made using one-way analysis of variance and Fisher’s multiple comparisons test and the Student’s t -test ( α = .05). A regression analysis was performed to evaluate the relationship between the porosity and the joint strength.