Original Article
Microstructural characterization and hardness properties of electric resistance welding titanium joints for dental applications
Proc IMechE Part H: J Engineering in Medicine 2015, Vol. 229(6) 429–438 Ó IMechE 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0954411915585598 pih.sagepub.com
Lorella Ceschini1, Iuri Boromei1, Alessandro Morri1, Diego Nardi2, Gianluca Sighinolfi3 and Marco Degidi3
Abstract The electric resistance welding procedure is used to join a titanium bar with specific implant abutments in order to produce a framework directly in the oral cavity of the patient. This investigation studied the effects of the welding process on microstructure and hardness properties of commercially pure (CP2 and CP4) Ti components. Different welding powers and cooling procedures were applied to bars and abutments, normally used to produce the framework, in order to simulate the clinical intraoral welding procedure. The analyses highlighted that the joining process did not induce appreciable changes in the geometry of the abutments. However, because of unavoidable microstructural modifications in the welded zones, the hardness decreased to values lower than those of the unwelded CP2 and CP4 Ti grades, irrespective of the welding environments and parameters.
Keywords Titanium, electric resistance welding, scanning electron microscope, hardness
Date received: 30 December 2014; accepted: 13 April 2015
Introduction The use of electric resistance (or spot) welding to join two metallic parts without a filler material is well known and employed in dentistry.1,2 This process is used to join a titanium bar with specific implant abutments, so as to produce a framework directly in the oral cavity of the patient3 and is called intraoral welding. During the welding process, the bar and the abutment are overlapped and positioned between the copper electrodes. The heat necessary to obtain the weld is generated by passing low voltage/high amperage electrical current through the electrodes for a short period of time. This heat does not exceed the threshold limits above which tissue injury could occur.4 There are three stages in spot welding: (a) the electrodes are brought together against the metal and the pressure is applied; (b) the current is turned on; (c) the current is turned off, while the pressure is maintained (hold time). The welding should occur in solid-state condition, without extensive melting of the metal, which instead is forged and diffusion bonded. However, due to geometry and size of both abutments and electrodes, a high electrical
resistance is present at their counterfacing surfaces, thus causing heat generation and temperature increase, with consequent possible localized melting of the material. Electrodes having a wide contact area with the bars are required to obtain sound welds; however, no accurate joint surface preparation or filler is needed.1,2,5 Most current published data on welding of titanium and its alloys are focused on laser and electric arc welding techniques; moreover, the data often are obtained by analyzing and testing standard tensile or fatigue specimens, rather than joints between real components.5–10 Laser welding is considered the most efficient method 1
Department of Industrial Engineering (DIN), University of Bologna, Bologna, Italy 2 Department of Biomedical and Neuromotor Sciences (DIBINEM), University of Bologna, Bologna, Italy 3 Private Practice, Bologna, Italy Corresponding author: Diego Nardi, Department of Biomedical and Neuromotor Sciences (DIBINEM), University of Bologna, Viale Risorgimento 4, 40136 Bologna, Italy. Email: [email protected]
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
430
Proc IMechE Part H: J Engineering in Medicine 229(6)
Figure 1. Drafts of the (a) SynCone (Sy) Ti abutment and (b) Balance Base (BB) Ti abutment.
Figure 2. (a) Sy abutment, (b) BB abutment, (c) CP Ti bar–Sy joint and (d) CP Ti bar–BB joint.
for focusing thermal energy to small areas. The beam imparts energy to the metal, causing its evaporation and generating a cavity, called ‘‘keyhole.’’ When the heat source moves forward, the keyhole is rapidly filled with the molten metal produced around it and solidifies to form the joint.11,12 Any welding process affects mechanical properties of the material due to the microstructural modifications induced by the local temperatures and cooling rates. This investigation was focused on the effect of electric resistance welding on the microstructural features and, consequently, on the hardness properties of joints made with different commercially pure (CP) Ti grades. Different welding parameters and cooling procedures were applied to simulate the intraoral welding. The components and bars tested are used in everyday practice. Two different retentive systems, screwing
and conical couplings, were tested. The research was aimed at (a) evaluating the efficiency of this welding process in producing sound welds with limited microstructural defects that could negatively affect the performance of the final prosthesis, and (b) confirming that the welding process induces a safe localized temperature increase that could not lead to any detachment of molten metal.
Materials and methods The implants under investigation consist of CP titanium bars, welded to SynCone (Sy) and Balance Base (BB) Ti abutments (Figures 1 and 2). CP titanium, Grade 2 (CP2) and Grade 4 (CP4), was used for the bars (f = 2 mm), while CP4 was used for the abutments (Sy and BB). The nominal chemical
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
Ceschini et al.
431
Table 1. Nominal chemical compositions (wt%) of the CP Ti used for the bars. CP Ti
C
N
O
Fe
Others
Ti
Grade 2 Grade 4
Max 0.100 Max 0.080
Max 0.030 Max 0.050
Max 0.250 Max 0.400
Max 0.300 Max 0.500
Max 0.400 Max 0.400
Bal. Bal.
Table 2. Nominal mechanical properties of the supplied Ti bars. CP Ti
sp0.2% (MPa) min–max
UTS (MPa) min–max
A% min–max
HV min–max
Roughness Ra (mm) min–max
Grade 2 Grade 4
406–420 602–627
532–535 757–771
23.0–25.0 18.0–18.0
169.0–177.0 –
0.30–0.38 0.38–0.47
Proof stress (sp0.2%), ultimate tensile strength (UTS), elongation to failure (A%), Vickers hardness (HV) and arithmetic mean roughness (Ra).
Figure 3. Clinical use of the resistance welding unit.
compositions of the CP Ti are reported in Table 1. Table 2 shows their nominal mechanical properties, where sp0.2% is the proof stress, UTS is the ultimate tensile strength, A% is the elongation to failure, HV is the Vickers hardness and Ra is the arithmetic mean roughness. The superior strength of CP4 is due to its higher content of nitrogen, oxygen and iron. An electric resistance welding unit (WeldOneÒ; DENTSPLY Implants, Mannheim, Germany) (Figure 3) was used to produce different sets of welded specimens. The welding process was carried out both in air (climate-controlled room, temperature: 23 °C–24 °C, relative humidity: 50% 6 5%) and in water (controlled bath temperature: 30 °C), by changing the bars (different Ti grades), the abutments (Sy, BB) and the power settings. All the three pre-loaded power settings programs, 100% (Smart High), 75% (Smart Medium) and 65% (Smart Low), were used. Table 3 summarizes the analyzed welding settings. The analyses were carried out on metallographic specimens embedded in a cast acrylic resin (cold mounting) and polished with SiC emery papers and diamond paste (9, 3 mm and 1 mm), according to standard procedures.13 To reveal the microstructure, the samples were chemically etched using both modified Weck’s reagent (5 g NH4FHF, 0.5 mL HCl conc. in 100 mL
aqueous solution) and Kroll’s reagent (1–3 mL HF conc., 2–6 mL HNO3 conc. in 100 mL aqueous solution). Kroll is a standard, general purpose etching reagent for Ti and its alloys, while Weck’s tint etching produces colors in polarized light. The microstructural characterization was carried out by means of optical microscopy (OM), under bright field and polarized light. Image analysis on the optical micrographs was performed using the Image Pro-PlusÒ software (Media Cybernetics Inc., Rockville, MD, USA). Scanning Electron Microscope (SEM) equipped with an Energy Dispersive Spectrometer (EDS) was also used. Microhardness measurements were carried out according to ASTM E384-11e114 using an applied load of 0.98 N (HV0.1). The measurements were carried out on the CP titanium bars, on the Sy and BB abutments and on the welded implants (both in longitudinal and in transverse directions). Representative metallographic samples (as-polished and tint etched) of both Sy and BB abutments, welded to Ti bars (in longitudinal and transverse directions), are reported in Figure 4. The hardness tests were carried out along the red dotted line.
Results and discussion Microstructure The microstructure of metals strongly depends on the manufacturing processes, including production technologies of the base materials and welding, and affects their mechanical properties. The production cycle of CP Ti usually involves plastic deformation and recrystallization annealing to obtain components with homogeneous microstructure, good strength and toughness. The welding thermal cycle, heating up to very high localized temperatures followed by uncontrolled cooling, induced a modification of the initial alloy microstructure. The peculiar microstructural features observed in the joint center (JC) and heat-affected zone (HAZ) are
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
432
Proc IMechE Part H: J Engineering in Medicine 229(6)
Table 3. List of joints and corresponding welding settings. Ti bar grade
Ti abutment
Welding power, % full power
Quench media
CP2 CP2 CP2 CP2 CP4 CP4 CP2 CP2 CP4 CP4
BB BB BB BB BB BB Sy Sy Sy Sy
100 (Smart High) 100 (Smart High) 75 (Smart Medium) 65 (Smart Low) 100 (Smart High) 100 (Smart High) 100 (Smart High) 100 (Smart High) 100 (Smart High) 100 (Smart High)
Air Water Air Air Air Water Air Water Air Water
Figure 4. Representative color tint etched metallographic samples, for both (a, b) Sy and (c, d) BB welded abutments, in (a, c) longitudinal and (b, d) transverse directions. The red dotted lines represent the position of the indentation for microhardness measurements.
directly related to the temperatures and cooling conditions reached during welding. CP titanium undergoes an allotropic phase transformation when heated at temperatures higher than 882 °C, where the a-phase (hexagonal close-packed (HCP) crystal structure) transforms to the b-phase (body-centred cubic (BCC)). The cooling rate from the b-phase field has a controlling influence on the resulting microstructure, inducing an equilibrium
microstructure with equiaxed grains with slow cooling rates, or martensitic and metastable phases (Widmansta¨tten plates and acicular microstructure) with fast cooling.15 The observation of different microstructures in the welded zones is, therefore, indicative of the temperatures and cooling conditions underwent by the joints. Unwelded CP2 bars and CP4, bar and abutments, displayed a recrystallized microstructure typical of annealed CP Ti, with equiaxed a-phase grains and dispersed b-phase at the grain boundaries, due to the presence of elements, like Fe, which stabilize the b-phase.16,17 The transverse and longitudinal sections of the CP2 Ti showed larger grains on the surface of the bars (average value of 65 mm), with respect to the center (34 mm), as shown in Figure 5. The different grain size values in the outer and inner zones can be related to the production process of the bar. This induces larger plastic deformation on the surface, favoring and accelerating the primary and secondary stages of recrystallization during annealing, with a consequent grain growth.16 The transverse and longitudinal sections of the CP4 Ti showed instead a finer and more homogeneous microstructure with respect to CP2, with average grain size values of 18 and 9 mm, for the longitudinal and transverse directions, respectively, both on the surface and center of the bars (Figure 6). The different grain size values between CP2 and CP4 bars are probably due to both the different plastic deformation rates and the higher Fe content present in the CP4 Ti. Fe is rejected from the solid solution due to the low solubility in the a-phase to stabilize the b-phase (either during solidification or cooling). This process pins the a-grain boundaries hindering grain growth during recrystallization.16,17 The lower grain size in the CP4 Ti leads to a corresponding increase in hardness (HV), yield strength (sp0.2%) and UTS, according to the wellknown Hall–Petch relationship.16,18 However, the superior material strength and lower ductility make it more difficult to deform and shape during the joining process. Low magnification optical micrographs of resistance welded bar and abutment (Figure 4) highlighted that
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
Ceschini et al.
433
Figure 5. Low magnification optical micrographs of CP2 bar: (a) longitudinal view and (b) transverse view.
Figure 6. Low magnification optical micrographs of CP4 bar: (a) longitudinal view and (b) transverse view.
the procedure was effective in joining bars and abutments, independently of the size and shape of the latter. It can also be noted that the joining process did not induce appreciable modification in the geometry of the abutments. In particular, their different radii of curvature and thickness do not lead to appreciable changes in the size of the region that undergoes microstructural modification during welding. The very localized heating generated during welding, instead, led to a significant grain growth, both in the welded zone (Figure 7(a)) and at the electrodes/titanium interface (Figure 7(b)). Grain growth was observed from the base material to the interface with the electrodes (Figure 8(a)) and from the base material to the JC through the HAZ (Figure 8(b)). The high localized temperatures, above the b-transus, reached in the welded zones, mainly in the JC, led to the formation of b grains; inside these grains, during cooling, a substructure of a-plates colonies (typical Widmansta¨tten microstructure) developed, as shown in Figure 8(c) and (d). The presence of a-plates can be related to very high cooling rate (up to
100 °C/s)15 from the welding temperature. The similarity of microstructure in the HAZ and JC made it difficult to locate the welding line and exactly define the volume of material involved in welding.17,19–21 In the contact zones between electrode and titanium (Figures 7 and 8), the concurrent effects of pressure and temperature also induced local surface plastic deformation and oxygen enrichment, as shown by SEM-EDS analyses (Figure 9). It is worth noting that similar microstructures characterized these zones in the investigated samples, irrespective of the welding conditions, as well as bars and abutments. Moreover, the microstructural modifications induced by the welding heating were always observed in very limited zones of the joints, only few hundreds of microns far from the JC, confirming the very fast temperature drop, up to values that do not exceed the threshold limits above which tissue injury could occur.4 Most of the joints were defect free, as generally occurs in solid-state welding; however, in some samples, oxidized ‘‘drops’’, with a typical solidification microstructure and occluded gas pores, were observed (Figure
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
434
Proc IMechE Part H: J Engineering in Medicine 229(6)
Figure 7. Low magnification optical micrographs of a joint between CP2 bar and Sy abutment. Grain coarsening (a) in the welded zone and (b) in the interface between the electrode and the base material.
Figure 8. Interface between (a) the electrodes and the base material, (b) JC and HAZ in a joint between CP2 bar and Sy CP4. Widmansta¨tten microstructure in the (c) JC and (d) HAZ. JC: joint center; HAZ: heat-affected zone.
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
Ceschini et al.
435 and, consequently, high surface temperature inducing local melting. The presence of thick (about 3–5 mm) oxidized layers (Figure 10(b)) around the ‘‘drops’’ was confirmed by SEM-EDS analyses, as shown in Figure 11. Their presence hinders a continuous joint formation.
Hardness
Figure 9. SEM image of the contact zone between electrode and titanium, and EDS oxygen concentration profile.
10). No detachment of any of these drops was ever observed. Oxide layers were not observed in the microstructure of bars and abutments before welding; therefore, their presence in the joints suggests that localized partial melting of Ti occurred during welding. This is probably due to the low contact area during resistance welding, which can lead to high local electric resistance
In the investigated samples, hardness measurements showed a wide scatter, both in the base (unwelded) and welded materials, ranging between 160 and 190 HV0.1 for the CP2 bars, 261 and 300 HV0.1 for the CP4 bar and 222 and 270 HV0.1 for the CP4 abutments (Table 4). The hardness variations in the base materials are probably due to slightly different chemical compositions and also to the previously discussed different grain size values induced by the production process (it is worth highlighting that bars and abutments were supplied by different companies). The hardness profiles in the joints between CP4 bar and CP4 abutments (Figure 12(a) and (c)), and between CP2 bar and CP4 abutments (Figure 12(b) and (d)), showed a relevant hardness reduction in the welded zone with respect to the CP4 base metal, and a slight reduction with respect to the CP2 base metal. The lower
Figure 10. Local melting at the joint line. Drops with (a) internal gas pores and (b) highly oxidized surface in a joint between CP2 bar and BB CP4. (c, d) Solidification microstructure inside drop in a joint between CP4 bar and BB CP4.
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
436
Proc IMechE Part H: J Engineering in Medicine 229(6)
Figure 11. (a) SEM images of an oxidized drop and (b) EDS oxygen concentration profile on its surface.
Figure 12. Microhardness profiles in the welded samples. (a, c) CP4 bar–CP4 abutment; (b, d) CP2 bar–CP4 abutment. JC: joint center; HAZ: heat-affected zone; BM: base metal.
Table 4. Mean hardness of base materials and standard deviation. HV0.1
Sy CP Ti Grade 4 BB CP Ti Grade 4 Bar CP Ti Grade 4 Bar CP Ti Grade 2
Longitudinal
Transversal
222 6 18 270 6 23 300 6 21 190 6 12
– – 261 6 18 160 6 15
hardness values (between 140 and 180 HV0.1) measured in the joints were comparable with those of the CP2 base metal, while they were lower by more than 40% with respect to those of CP4 bar and abutments. Similar trends were observed in all the samples, irrespective of the welding parameters and environment, as well as of retentive systems. Hardness profiles along the joint cross-sections confirmed that the welded zone is the weakest part of the framework with local hardness
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
Ceschini et al.
437
generally lower than those of both CP2 and CP4 base materials. The differences in the microhardness values in the welded regions can be related to the different widths and crystallographic orientations of the a-lamellae and colonies of the Widmansta¨tten microstructure, rather than to the slight different chemical compositions between CP2 and CP4 grades. Several studies on HCP single crystals, in fact, have shown that indentation into the basal plane may be almost twice as hard as indentation perpendicular to it.16,21
Conclusion 1.
2.
3.
4.
5.
The joining process, independent of the CP titanium grades, as well as of different welding parameters or cooling procedures, led to similar microstructural modifications, consisting of (a) development of Widmansta¨tten microstructure in the JC and HAZ; (b) local melting in the JC of some joints; and (c) appreciable grain growth, in the contact zones between the copper electrode and titanium. The microhardness profiles along the cross-section joints highlighted a strength decrease in the JC and HAZ, with respect to the base metals, irrespective of the chemical composition (CP2/CP4 or CP4/ CP4), welding environments and parameters. This hardness decrease is commonly observed. The electric resistance welding did not induce appreciable modifications in the geometry of the abutments. The microstructural analyses highlighted that the temperature increase was only limited to few hundreds of microns from the JC, while it was negligible in the rest of the material. The heat generated during the welding process caused no detachment of any molten metal drop.
Acknowledgements The welding components were supplied DENTSPLY Implants, Mannheim, Germany.
by
Declaration of conflicting interests The authors declare that there is no conflict of interest. Funding This research received no other specific grants from any funding agency in the public, commercial or not-forprofit sectors. References 1. Iijima M, Brantley WA, Yuasa T, et al. Joining characteristics of b-titanium wires with electrical resistance welding. J Biomed Mater Res 2008; 85B: 378–384.
2. Nelson KR, Burstone CJ and Goldberg AJ. Optimal welding of beta titanium orthodontic wires. Am J Orthod Dentofacial Orthop 1987; 92: 213–219. 3. Degidi M, Nardi D and Piattelli A. A six-year follow-up of full-arch immediate restorations fabricated with an intraoral welding technique. Implant Dent 2013; 22: 224–231. 4. Degidi M, Nardi D, Sighinolfi G, et al. In vitro infrared thermography assessment of temperature peaks during the intra-oral welding of titanium abutments. Infrared Phys Tech 2012; 55: 279–283. 5. Sjo¨gren G, Andersson M and Bergman M. Laser welding of titanium in dentistry. Acta Odont Scand 1988; 46: 247–253. 6. James C, Taylor Steven O, et al. Effects of joint configuration for the arc welding of cast Ti-6Al-4V alloy rods in argon. J Prosthet Dent 1998; 79: 291–297. 7. Uysal H, Kurtoglu C, Gurbuz R, et al. Structure and mechanical properties of Cresco-Ti laser-welded joints and stress analyses using finite element models of fixed distal extension and fixed partial prosthetic designs. J Prosthet Dent 2005; 93: 235–244. 8. Srimaneepong V, Yoneyama T, Kobayashi E, et al. Comparative study on torsional strength, ductility and fracture characteristics of laser-welded a + b Ti–6Al– 7Nb alloy, CP Titanium and Co–Cr alloy dental castings. Dent Mater 2008; 24: 839–845. 9. Hart CN and Wilson PR. Evaluation of welded titanium joints used with cantilevered implant-supported prostheses. J Prosthet Dent 2006; 96: 25–32. 10. Fornaini C, Passaretti F, Villa E, et al. Intraoral laser welding: ultrastructural and mechanical analysis to compare laboratory laser and dental laser. Lasers Med Sci 2011; 26: 415–420. 11. Bertrand C, Laplanche O, Rocca JP, et al. Effect of the combination of different welding parameters on melting characteristics of grade 1 titanium with a pulsed Nd–Yag laser. Lasers Med Sci 2007; 22: 237–244. 12. Iwasaki K, Ohkawa S, Uo M, et al. Laser welding of titanium and dental precious alloys. Mater Trans 2004; 45: 1140–1146. 13. ASTM E3-11. Standard guide for preparation of metallographic specimens. West Conshohocken, PA: ASTM International, 2003, http://www.astm.org 14. ASTM E384-11e1. Standard test method for Knoop and Vickers hardness of materials. West Conshohocken, PA: ASTM International, 2003, http://www.astm.org 15. Lu¨tjering G and Williams JC. Titanium (Engineering Materials and Processes). Berlin: Springer, 2007, pp.177–201. 16. Lathabai S, Jarvis BL and Barton KJ. Comparison of keyhole and conventional gas tungsten arc welds in commercially pure titanium. Mat Sci Eng A 2001; 299: 81–93. 17. Greger M, Ern M, Kander L, et al. Structure and properties of titanium for dental implants. Metab 2009; 48: 249–252. 18. Danielson P, Wilson R and Alman D. Microstructure of titanium welds. Adv Mater Process 2003; 161: 39–42. 19. Orlowski de Garcia JA, de Lima GL, Bocalla˜o Pereira WD, et al. Characterization of titanium welded joints by the orbital gas tungsten arc welding process for aerospace application. J Aerosp Technol Manag 2010; 2: 211–218.
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
438
Proc IMechE Part H: J Engineering in Medicine 229(6)
20. Wiskott A, Doumas T, Scherrer SS, et al. Mechanical and structural characteristics of commercially pure grade 2 Ti welds and solder joints. J Mater Sci Mater Med 2001; 12: 719–725.
21. Learya RK, Mersonb E, Birminghamc K, et al. Microstructural and microtextural analysis of InterPulse GTCAW welds in Cp-Ti and Ti–6Al–4V. Mat Sci Eng A 2010; 527: 7694–7705.
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
Microstructural characterization and hardness properties of electric resistance welding titanium joints for dental applications
Proc IMechE Part H: J Engineering in Medicine 2015, Vol. 229(6) 429–438 Ó IMechE 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0954411915585598 pih.sagepub.com
Lorella Ceschini1, Iuri Boromei1, Alessandro Morri1, Diego Nardi2, Gianluca Sighinolfi3 and Marco Degidi3
Abstract The electric resistance welding procedure is used to join a titanium bar with specific implant abutments in order to produce a framework directly in the oral cavity of the patient. This investigation studied the effects of the welding process on microstructure and hardness properties of commercially pure (CP2 and CP4) Ti components. Different welding powers and cooling procedures were applied to bars and abutments, normally used to produce the framework, in order to simulate the clinical intraoral welding procedure. The analyses highlighted that the joining process did not induce appreciable changes in the geometry of the abutments. However, because of unavoidable microstructural modifications in the welded zones, the hardness decreased to values lower than those of the unwelded CP2 and CP4 Ti grades, irrespective of the welding environments and parameters.
Keywords Titanium, electric resistance welding, scanning electron microscope, hardness
Date received: 30 December 2014; accepted: 13 April 2015
Introduction The use of electric resistance (or spot) welding to join two metallic parts without a filler material is well known and employed in dentistry.1,2 This process is used to join a titanium bar with specific implant abutments, so as to produce a framework directly in the oral cavity of the patient3 and is called intraoral welding. During the welding process, the bar and the abutment are overlapped and positioned between the copper electrodes. The heat necessary to obtain the weld is generated by passing low voltage/high amperage electrical current through the electrodes for a short period of time. This heat does not exceed the threshold limits above which tissue injury could occur.4 There are three stages in spot welding: (a) the electrodes are brought together against the metal and the pressure is applied; (b) the current is turned on; (c) the current is turned off, while the pressure is maintained (hold time). The welding should occur in solid-state condition, without extensive melting of the metal, which instead is forged and diffusion bonded. However, due to geometry and size of both abutments and electrodes, a high electrical
resistance is present at their counterfacing surfaces, thus causing heat generation and temperature increase, with consequent possible localized melting of the material. Electrodes having a wide contact area with the bars are required to obtain sound welds; however, no accurate joint surface preparation or filler is needed.1,2,5 Most current published data on welding of titanium and its alloys are focused on laser and electric arc welding techniques; moreover, the data often are obtained by analyzing and testing standard tensile or fatigue specimens, rather than joints between real components.5–10 Laser welding is considered the most efficient method 1
Department of Industrial Engineering (DIN), University of Bologna, Bologna, Italy 2 Department of Biomedical and Neuromotor Sciences (DIBINEM), University of Bologna, Bologna, Italy 3 Private Practice, Bologna, Italy Corresponding author: Diego Nardi, Department of Biomedical and Neuromotor Sciences (DIBINEM), University of Bologna, Viale Risorgimento 4, 40136 Bologna, Italy. Email: [email protected]
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
430
Proc IMechE Part H: J Engineering in Medicine 229(6)
Figure 1. Drafts of the (a) SynCone (Sy) Ti abutment and (b) Balance Base (BB) Ti abutment.
Figure 2. (a) Sy abutment, (b) BB abutment, (c) CP Ti bar–Sy joint and (d) CP Ti bar–BB joint.
for focusing thermal energy to small areas. The beam imparts energy to the metal, causing its evaporation and generating a cavity, called ‘‘keyhole.’’ When the heat source moves forward, the keyhole is rapidly filled with the molten metal produced around it and solidifies to form the joint.11,12 Any welding process affects mechanical properties of the material due to the microstructural modifications induced by the local temperatures and cooling rates. This investigation was focused on the effect of electric resistance welding on the microstructural features and, consequently, on the hardness properties of joints made with different commercially pure (CP) Ti grades. Different welding parameters and cooling procedures were applied to simulate the intraoral welding. The components and bars tested are used in everyday practice. Two different retentive systems, screwing
and conical couplings, were tested. The research was aimed at (a) evaluating the efficiency of this welding process in producing sound welds with limited microstructural defects that could negatively affect the performance of the final prosthesis, and (b) confirming that the welding process induces a safe localized temperature increase that could not lead to any detachment of molten metal.
Materials and methods The implants under investigation consist of CP titanium bars, welded to SynCone (Sy) and Balance Base (BB) Ti abutments (Figures 1 and 2). CP titanium, Grade 2 (CP2) and Grade 4 (CP4), was used for the bars (f = 2 mm), while CP4 was used for the abutments (Sy and BB). The nominal chemical
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
Ceschini et al.
431
Table 1. Nominal chemical compositions (wt%) of the CP Ti used for the bars. CP Ti
C
N
O
Fe
Others
Ti
Grade 2 Grade 4
Max 0.100 Max 0.080
Max 0.030 Max 0.050
Max 0.250 Max 0.400
Max 0.300 Max 0.500
Max 0.400 Max 0.400
Bal. Bal.
Table 2. Nominal mechanical properties of the supplied Ti bars. CP Ti
sp0.2% (MPa) min–max
UTS (MPa) min–max
A% min–max
HV min–max
Roughness Ra (mm) min–max
Grade 2 Grade 4
406–420 602–627
532–535 757–771
23.0–25.0 18.0–18.0
169.0–177.0 –
0.30–0.38 0.38–0.47
Proof stress (sp0.2%), ultimate tensile strength (UTS), elongation to failure (A%), Vickers hardness (HV) and arithmetic mean roughness (Ra).
Figure 3. Clinical use of the resistance welding unit.
compositions of the CP Ti are reported in Table 1. Table 2 shows their nominal mechanical properties, where sp0.2% is the proof stress, UTS is the ultimate tensile strength, A% is the elongation to failure, HV is the Vickers hardness and Ra is the arithmetic mean roughness. The superior strength of CP4 is due to its higher content of nitrogen, oxygen and iron. An electric resistance welding unit (WeldOneÒ; DENTSPLY Implants, Mannheim, Germany) (Figure 3) was used to produce different sets of welded specimens. The welding process was carried out both in air (climate-controlled room, temperature: 23 °C–24 °C, relative humidity: 50% 6 5%) and in water (controlled bath temperature: 30 °C), by changing the bars (different Ti grades), the abutments (Sy, BB) and the power settings. All the three pre-loaded power settings programs, 100% (Smart High), 75% (Smart Medium) and 65% (Smart Low), were used. Table 3 summarizes the analyzed welding settings. The analyses were carried out on metallographic specimens embedded in a cast acrylic resin (cold mounting) and polished with SiC emery papers and diamond paste (9, 3 mm and 1 mm), according to standard procedures.13 To reveal the microstructure, the samples were chemically etched using both modified Weck’s reagent (5 g NH4FHF, 0.5 mL HCl conc. in 100 mL
aqueous solution) and Kroll’s reagent (1–3 mL HF conc., 2–6 mL HNO3 conc. in 100 mL aqueous solution). Kroll is a standard, general purpose etching reagent for Ti and its alloys, while Weck’s tint etching produces colors in polarized light. The microstructural characterization was carried out by means of optical microscopy (OM), under bright field and polarized light. Image analysis on the optical micrographs was performed using the Image Pro-PlusÒ software (Media Cybernetics Inc., Rockville, MD, USA). Scanning Electron Microscope (SEM) equipped with an Energy Dispersive Spectrometer (EDS) was also used. Microhardness measurements were carried out according to ASTM E384-11e114 using an applied load of 0.98 N (HV0.1). The measurements were carried out on the CP titanium bars, on the Sy and BB abutments and on the welded implants (both in longitudinal and in transverse directions). Representative metallographic samples (as-polished and tint etched) of both Sy and BB abutments, welded to Ti bars (in longitudinal and transverse directions), are reported in Figure 4. The hardness tests were carried out along the red dotted line.
Results and discussion Microstructure The microstructure of metals strongly depends on the manufacturing processes, including production technologies of the base materials and welding, and affects their mechanical properties. The production cycle of CP Ti usually involves plastic deformation and recrystallization annealing to obtain components with homogeneous microstructure, good strength and toughness. The welding thermal cycle, heating up to very high localized temperatures followed by uncontrolled cooling, induced a modification of the initial alloy microstructure. The peculiar microstructural features observed in the joint center (JC) and heat-affected zone (HAZ) are
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
432
Proc IMechE Part H: J Engineering in Medicine 229(6)
Table 3. List of joints and corresponding welding settings. Ti bar grade
Ti abutment
Welding power, % full power
Quench media
CP2 CP2 CP2 CP2 CP4 CP4 CP2 CP2 CP4 CP4
BB BB BB BB BB BB Sy Sy Sy Sy
100 (Smart High) 100 (Smart High) 75 (Smart Medium) 65 (Smart Low) 100 (Smart High) 100 (Smart High) 100 (Smart High) 100 (Smart High) 100 (Smart High) 100 (Smart High)
Air Water Air Air Air Water Air Water Air Water
Figure 4. Representative color tint etched metallographic samples, for both (a, b) Sy and (c, d) BB welded abutments, in (a, c) longitudinal and (b, d) transverse directions. The red dotted lines represent the position of the indentation for microhardness measurements.
directly related to the temperatures and cooling conditions reached during welding. CP titanium undergoes an allotropic phase transformation when heated at temperatures higher than 882 °C, where the a-phase (hexagonal close-packed (HCP) crystal structure) transforms to the b-phase (body-centred cubic (BCC)). The cooling rate from the b-phase field has a controlling influence on the resulting microstructure, inducing an equilibrium
microstructure with equiaxed grains with slow cooling rates, or martensitic and metastable phases (Widmansta¨tten plates and acicular microstructure) with fast cooling.15 The observation of different microstructures in the welded zones is, therefore, indicative of the temperatures and cooling conditions underwent by the joints. Unwelded CP2 bars and CP4, bar and abutments, displayed a recrystallized microstructure typical of annealed CP Ti, with equiaxed a-phase grains and dispersed b-phase at the grain boundaries, due to the presence of elements, like Fe, which stabilize the b-phase.16,17 The transverse and longitudinal sections of the CP2 Ti showed larger grains on the surface of the bars (average value of 65 mm), with respect to the center (34 mm), as shown in Figure 5. The different grain size values in the outer and inner zones can be related to the production process of the bar. This induces larger plastic deformation on the surface, favoring and accelerating the primary and secondary stages of recrystallization during annealing, with a consequent grain growth.16 The transverse and longitudinal sections of the CP4 Ti showed instead a finer and more homogeneous microstructure with respect to CP2, with average grain size values of 18 and 9 mm, for the longitudinal and transverse directions, respectively, both on the surface and center of the bars (Figure 6). The different grain size values between CP2 and CP4 bars are probably due to both the different plastic deformation rates and the higher Fe content present in the CP4 Ti. Fe is rejected from the solid solution due to the low solubility in the a-phase to stabilize the b-phase (either during solidification or cooling). This process pins the a-grain boundaries hindering grain growth during recrystallization.16,17 The lower grain size in the CP4 Ti leads to a corresponding increase in hardness (HV), yield strength (sp0.2%) and UTS, according to the wellknown Hall–Petch relationship.16,18 However, the superior material strength and lower ductility make it more difficult to deform and shape during the joining process. Low magnification optical micrographs of resistance welded bar and abutment (Figure 4) highlighted that
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
Ceschini et al.
433
Figure 5. Low magnification optical micrographs of CP2 bar: (a) longitudinal view and (b) transverse view.
Figure 6. Low magnification optical micrographs of CP4 bar: (a) longitudinal view and (b) transverse view.
the procedure was effective in joining bars and abutments, independently of the size and shape of the latter. It can also be noted that the joining process did not induce appreciable modification in the geometry of the abutments. In particular, their different radii of curvature and thickness do not lead to appreciable changes in the size of the region that undergoes microstructural modification during welding. The very localized heating generated during welding, instead, led to a significant grain growth, both in the welded zone (Figure 7(a)) and at the electrodes/titanium interface (Figure 7(b)). Grain growth was observed from the base material to the interface with the electrodes (Figure 8(a)) and from the base material to the JC through the HAZ (Figure 8(b)). The high localized temperatures, above the b-transus, reached in the welded zones, mainly in the JC, led to the formation of b grains; inside these grains, during cooling, a substructure of a-plates colonies (typical Widmansta¨tten microstructure) developed, as shown in Figure 8(c) and (d). The presence of a-plates can be related to very high cooling rate (up to
100 °C/s)15 from the welding temperature. The similarity of microstructure in the HAZ and JC made it difficult to locate the welding line and exactly define the volume of material involved in welding.17,19–21 In the contact zones between electrode and titanium (Figures 7 and 8), the concurrent effects of pressure and temperature also induced local surface plastic deformation and oxygen enrichment, as shown by SEM-EDS analyses (Figure 9). It is worth noting that similar microstructures characterized these zones in the investigated samples, irrespective of the welding conditions, as well as bars and abutments. Moreover, the microstructural modifications induced by the welding heating were always observed in very limited zones of the joints, only few hundreds of microns far from the JC, confirming the very fast temperature drop, up to values that do not exceed the threshold limits above which tissue injury could occur.4 Most of the joints were defect free, as generally occurs in solid-state welding; however, in some samples, oxidized ‘‘drops’’, with a typical solidification microstructure and occluded gas pores, were observed (Figure
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
434
Proc IMechE Part H: J Engineering in Medicine 229(6)
Figure 7. Low magnification optical micrographs of a joint between CP2 bar and Sy abutment. Grain coarsening (a) in the welded zone and (b) in the interface between the electrode and the base material.
Figure 8. Interface between (a) the electrodes and the base material, (b) JC and HAZ in a joint between CP2 bar and Sy CP4. Widmansta¨tten microstructure in the (c) JC and (d) HAZ. JC: joint center; HAZ: heat-affected zone.
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
Ceschini et al.
435 and, consequently, high surface temperature inducing local melting. The presence of thick (about 3–5 mm) oxidized layers (Figure 10(b)) around the ‘‘drops’’ was confirmed by SEM-EDS analyses, as shown in Figure 11. Their presence hinders a continuous joint formation.
Hardness
Figure 9. SEM image of the contact zone between electrode and titanium, and EDS oxygen concentration profile.
10). No detachment of any of these drops was ever observed. Oxide layers were not observed in the microstructure of bars and abutments before welding; therefore, their presence in the joints suggests that localized partial melting of Ti occurred during welding. This is probably due to the low contact area during resistance welding, which can lead to high local electric resistance
In the investigated samples, hardness measurements showed a wide scatter, both in the base (unwelded) and welded materials, ranging between 160 and 190 HV0.1 for the CP2 bars, 261 and 300 HV0.1 for the CP4 bar and 222 and 270 HV0.1 for the CP4 abutments (Table 4). The hardness variations in the base materials are probably due to slightly different chemical compositions and also to the previously discussed different grain size values induced by the production process (it is worth highlighting that bars and abutments were supplied by different companies). The hardness profiles in the joints between CP4 bar and CP4 abutments (Figure 12(a) and (c)), and between CP2 bar and CP4 abutments (Figure 12(b) and (d)), showed a relevant hardness reduction in the welded zone with respect to the CP4 base metal, and a slight reduction with respect to the CP2 base metal. The lower
Figure 10. Local melting at the joint line. Drops with (a) internal gas pores and (b) highly oxidized surface in a joint between CP2 bar and BB CP4. (c, d) Solidification microstructure inside drop in a joint between CP4 bar and BB CP4.
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
436
Proc IMechE Part H: J Engineering in Medicine 229(6)
Figure 11. (a) SEM images of an oxidized drop and (b) EDS oxygen concentration profile on its surface.
Figure 12. Microhardness profiles in the welded samples. (a, c) CP4 bar–CP4 abutment; (b, d) CP2 bar–CP4 abutment. JC: joint center; HAZ: heat-affected zone; BM: base metal.
Table 4. Mean hardness of base materials and standard deviation. HV0.1
Sy CP Ti Grade 4 BB CP Ti Grade 4 Bar CP Ti Grade 4 Bar CP Ti Grade 2
Longitudinal
Transversal
222 6 18 270 6 23 300 6 21 190 6 12
– – 261 6 18 160 6 15
hardness values (between 140 and 180 HV0.1) measured in the joints were comparable with those of the CP2 base metal, while they were lower by more than 40% with respect to those of CP4 bar and abutments. Similar trends were observed in all the samples, irrespective of the welding parameters and environment, as well as of retentive systems. Hardness profiles along the joint cross-sections confirmed that the welded zone is the weakest part of the framework with local hardness
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
Ceschini et al.
437
generally lower than those of both CP2 and CP4 base materials. The differences in the microhardness values in the welded regions can be related to the different widths and crystallographic orientations of the a-lamellae and colonies of the Widmansta¨tten microstructure, rather than to the slight different chemical compositions between CP2 and CP4 grades. Several studies on HCP single crystals, in fact, have shown that indentation into the basal plane may be almost twice as hard as indentation perpendicular to it.16,21
Conclusion 1.
2.
3.
4.
5.
The joining process, independent of the CP titanium grades, as well as of different welding parameters or cooling procedures, led to similar microstructural modifications, consisting of (a) development of Widmansta¨tten microstructure in the JC and HAZ; (b) local melting in the JC of some joints; and (c) appreciable grain growth, in the contact zones between the copper electrode and titanium. The microhardness profiles along the cross-section joints highlighted a strength decrease in the JC and HAZ, with respect to the base metals, irrespective of the chemical composition (CP2/CP4 or CP4/ CP4), welding environments and parameters. This hardness decrease is commonly observed. The electric resistance welding did not induce appreciable modifications in the geometry of the abutments. The microstructural analyses highlighted that the temperature increase was only limited to few hundreds of microns from the JC, while it was negligible in the rest of the material. The heat generated during the welding process caused no detachment of any molten metal drop.
Acknowledgements The welding components were supplied DENTSPLY Implants, Mannheim, Germany.
by
Declaration of conflicting interests The authors declare that there is no conflict of interest. Funding This research received no other specific grants from any funding agency in the public, commercial or not-forprofit sectors. References 1. Iijima M, Brantley WA, Yuasa T, et al. Joining characteristics of b-titanium wires with electrical resistance welding. J Biomed Mater Res 2008; 85B: 378–384.
2. Nelson KR, Burstone CJ and Goldberg AJ. Optimal welding of beta titanium orthodontic wires. Am J Orthod Dentofacial Orthop 1987; 92: 213–219. 3. Degidi M, Nardi D and Piattelli A. A six-year follow-up of full-arch immediate restorations fabricated with an intraoral welding technique. Implant Dent 2013; 22: 224–231. 4. Degidi M, Nardi D, Sighinolfi G, et al. In vitro infrared thermography assessment of temperature peaks during the intra-oral welding of titanium abutments. Infrared Phys Tech 2012; 55: 279–283. 5. Sjo¨gren G, Andersson M and Bergman M. Laser welding of titanium in dentistry. Acta Odont Scand 1988; 46: 247–253. 6. James C, Taylor Steven O, et al. Effects of joint configuration for the arc welding of cast Ti-6Al-4V alloy rods in argon. J Prosthet Dent 1998; 79: 291–297. 7. Uysal H, Kurtoglu C, Gurbuz R, et al. Structure and mechanical properties of Cresco-Ti laser-welded joints and stress analyses using finite element models of fixed distal extension and fixed partial prosthetic designs. J Prosthet Dent 2005; 93: 235–244. 8. Srimaneepong V, Yoneyama T, Kobayashi E, et al. Comparative study on torsional strength, ductility and fracture characteristics of laser-welded a + b Ti–6Al– 7Nb alloy, CP Titanium and Co–Cr alloy dental castings. Dent Mater 2008; 24: 839–845. 9. Hart CN and Wilson PR. Evaluation of welded titanium joints used with cantilevered implant-supported prostheses. J Prosthet Dent 2006; 96: 25–32. 10. Fornaini C, Passaretti F, Villa E, et al. Intraoral laser welding: ultrastructural and mechanical analysis to compare laboratory laser and dental laser. Lasers Med Sci 2011; 26: 415–420. 11. Bertrand C, Laplanche O, Rocca JP, et al. Effect of the combination of different welding parameters on melting characteristics of grade 1 titanium with a pulsed Nd–Yag laser. Lasers Med Sci 2007; 22: 237–244. 12. Iwasaki K, Ohkawa S, Uo M, et al. Laser welding of titanium and dental precious alloys. Mater Trans 2004; 45: 1140–1146. 13. ASTM E3-11. Standard guide for preparation of metallographic specimens. West Conshohocken, PA: ASTM International, 2003, http://www.astm.org 14. ASTM E384-11e1. Standard test method for Knoop and Vickers hardness of materials. West Conshohocken, PA: ASTM International, 2003, http://www.astm.org 15. Lu¨tjering G and Williams JC. Titanium (Engineering Materials and Processes). Berlin: Springer, 2007, pp.177–201. 16. Lathabai S, Jarvis BL and Barton KJ. Comparison of keyhole and conventional gas tungsten arc welds in commercially pure titanium. Mat Sci Eng A 2001; 299: 81–93. 17. Greger M, Ern M, Kander L, et al. Structure and properties of titanium for dental implants. Metab 2009; 48: 249–252. 18. Danielson P, Wilson R and Alman D. Microstructure of titanium welds. Adv Mater Process 2003; 161: 39–42. 19. Orlowski de Garcia JA, de Lima GL, Bocalla˜o Pereira WD, et al. Characterization of titanium welded joints by the orbital gas tungsten arc welding process for aerospace application. J Aerosp Technol Manag 2010; 2: 211–218.
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
438
Proc IMechE Part H: J Engineering in Medicine 229(6)
20. Wiskott A, Doumas T, Scherrer SS, et al. Mechanical and structural characteristics of commercially pure grade 2 Ti welds and solder joints. J Mater Sci Mater Med 2001; 12: 719–725.
21. Learya RK, Mersonb E, Birminghamc K, et al. Microstructural and microtextural analysis of InterPulse GTCAW welds in Cp-Ti and Ti–6Al–4V. Mat Sci Eng A 2010; 527: 7694–7705.
Downloaded from pih.sagepub.com at WEST VIRGINA UNIV on October 20, 2015
