Taslema Sultana

Study of two different thin film coating methods in transmission laser micro-joining of thin Ti-film coated glass and polyimide for biomedical applications

Biomedical devices and implants require precision joining for hermetic sealing which can be achieved with low power lasers. The effect of two different thin metal film coating methods was studied in transmission laser micro-joints of titanium-coated glass and polyimide. The coating methods were cathodic arc physical vapor deposition (CA-PVD) and electron beam evaporation (EB-PVD). Titanium-coated glass joined to polyimide film can have neural electrode application. The improvement of the joint quality will be essential for robust performance of the device. Low power fiber laser (wave length = 1100 nm) was used for transmission laser micro-joining of thin titanium (Ti) film (approximately 200 nm) coated Pyrex borosilicate 7740 glass wafer (0.5 mm thick) and polyimide (Imidex) film (0.2 mm thick). Ti film acts as the coupling agent in the joining process. The Ti film deposition rate in the CA-PVD was 5-10 A/s and in the EB-PVD 1.5 A/s. The laser joint strength was measured by a lap shear test, the Ti film surfaces were analyzed by atomic force microscopy (AFM) and the lap shear tested joints were analyzed by optical microscopy and scanning electron microscopy (SEM). The film properties and the failure modes of the joints were correlated to joint strength. The CA-PVD produced around 4 times stronger laser joints than EB-PVD. The adhesion of the Ti film on glass by CA-PVD is better than that of the EB-PVD method. This is likely to be due to a higher film deposition rate and consequently higher adhesion or sticking coefficient for the CA-PVD particles arriving on the substrate compared to that of the EB-PVD film. EB-PVD shows poor laser bonding properties due to the development of thermal hotspots which occurs from film decohesion.

Performance of Laser Microjoints in Rat Brain

The stability of the laser bonded titanium coated glass/polyimide microjoints were studied in-vivo (by implanting on the rat brain surface for 10 days) and were compared with the earlier in-vitro (by soaking in artificial cerebrospinal fluid, CSF at 37°C for one week) data. In the current state, the strength of the joints were measured by a specially designed instrument called “pressure test” equipment where the samples were subjected to a variable pressure load (using high pressure nitrogen) controlled by a pressure regulator. The strength of the joints seems to degrade as a result of soaking in rat brain. The bond degradation in rat brain implants is similar compared to those soaked in CSF solution. Polyimide uptakes water through existing pores in it and also water gets in the joint region through the edges of the samples. Water might have caused oxidation of the chemical bonds which are thought to have formed by the laser fabrication process. Water availability at the joint region in implanted samples is less than that of the CSF soaked samples — which explains the better retention of the joint strength of the vivo samples. The average failure load from pressure test was found to be 1.4 N/mm, which is below the average tensile strength of the joint (7.3 N/mm) as published elsewhere by the authors. The difference between the joint strengths results from the difference in applied mechanical (joint peeling in pressure testing and joint shearing in tensile testing) loads and loading rates by the experiments.© 2008 ASME