A Type 1 niobium (Nb) part and a grade 5 titanium (Ti6Al4V) part were joined by laser weld, to form a component to be used in implantable medical devices for electrical stimulation therapies. This component could be exposed to body fluid or particularly water which could penetrate through the medical adhesive seal over the time of implantation. To meet the device reliability, performance, and patient safety goals, electrochemical immersion testing was conducted to study corrosion of the base metals and their galvanic couples and to evaluate corrosion resistance of the weld joints. Samples of as-manufactured Ti6Al4V parts, Nb parts, and weld joints were prepared, cleaned and inspected for testing (Figure 1, Table 1). A wire made of the same material and mostly sealed by medical adhesive, was welded to each sample for electrical connection. The surface areas of samples were calculated. They were divided into groups and tested at 37°C in glass jars containing de-ionized (DI) water or phosphate buffered saline (PBS), respectively. Electrochemical tests including open circuit potential (OCP); cathodic polarization (CP) or anodic polarization (AP); galvanic corrosion or zero resistance ammeter (ZRA) testing; anodic cyclic polarization (ACP) and potentiostatic polarization (PP) were performed at different times during immersion of the samples. The potentiodynamic scan rate for CP, AP and ACP scans was 0.2 mV/s. The tests used three-electrode configuration consisting of a working electrode (WE) which was the testing sample; a reference electrode (Ref.) which was a Ag/AgCl electrode in 3 M NaCl solution; and a counter electrode (CE) which was a piece of platinum sheet, or for galvanic corrosion testing, which was the secondary working electrode of the Nb/Ti6Al4V galvanic couple. Visual and stereomicroscopic examinations were made during and after the testing with a total of immersion time of about 100-130 days. After 3-4 days of immersion, OCP was stabilizing with an average of about 0.08, 0.05 and 0.07 V vs. Ref. in DI water and 0.11, -0.02 and 0.06 V vs. Ref. in PBS, for the Ti6Al4V, Nb and joint samples, respectively. Though the OCP difference was small, it indicated Nb would act as anode when joined or coupled to Ti6Al4V. Galvanic corrosion could occur in Nb of the joint, especially considering the anode (Nb) to cathode (Ti6Al4V) area ratio was as small as 0.01. A galvanic current was estimated to be about 9-13 nA/cm2 in DI water and 24-42 nA/cm2 in PBS based on anodic polarization of Nb and cathodic polarization of Ti6Al4V in potentiodynamic scans (Figure 2a). Further, galvanic corrosion potential and current (Figures 2b and 2c) of the Nb/Ti6Al4V couples was measured by ZRA testing. After ~48 days of testing, the galvanic current was stabilized with an average of about 1 nA/cm2 in both DI water and PBS, indicating a stable passive film formed on Nb surface. The average galvanic corrosion rate of Nb of the couples was about 0.008 μm/year in both DI water and PBS, calculated from the measured galvanic current which was considered more accurate (as obtained in a more steady state) and much lower than the estimation from potentiodynamic scans. ACP curves of the joint and base metal samples showed similar polarization behavior in both DI water and PBS (Figure 2d). All samples displayed stable passivation with low current from OCP to a potential of Ep ≅0.8 V vs. OCP. Beyond Ep, current increased markedly, and small hysteresis formed in ACP scans for the joint and Nb samples. Additional potentiostatic polarization was thus performed on the joint and Nb samples in PBS for ~4 days at potentials of 1.2 V and 1.4 V vs. Ref. which were above Ep and in the hysteresis range. The current would stabilize at the increased levels for both potentials. Neither localized pitting or crevice corrosion attack nor active gas bubbling was observed on the joints and parts. Corrosion rate estimated from the increased current at 1.4 V vs. Ref. for Nb could be as high as 66 μm/year. Depending on the type of device design, an electrochemical potential possibly applied to the joint component was assessed under simulated electrical signals for stimulation, which was measured to be under 0.3 V vs. Ref. and smaller than Ep. Therefore, the joint materials of Nb and Ti6Al4V are stable and suitable for the device application with no concern of corrosion. Figure 1
Read full abstract