Electrical technique for monitoring crack growth in thin-film fracture mechanics specimens
- PDF / 1,458,938 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 45 Downloads / 224 Views
An accurate and reliable electrical technique for continuous monitoring of crack growth in fracture specimens containing technologically relevant thin-film device structures has been developed. Both adhesive and cohesive crack growth measurements are reported using a SiO2 passivation layer and a conducting titanium film deposited on the side face of fracture specimens. Crack velocity measurements approaching 10−12 m/s were achieved, representing nearly an order of magnitude improvement over commonly used compliance-based techniques. The technique may be particularly useful for elucidating near threshold crack velocity behavior, which is important for thin-film reliability.
Fracture and crack growth in thin-film structures is of significant fundamental and technological interest for current and emerging device technologies. In particular, the phenomenon of subcritical crack growth associated with processes of stress-corrosion cracking or cyclic fatigue crack growth in thin-film structures has recently received considerable attention.1–11 This time or loadingcycle dependent mode of failure typically occurs at loads well below those required to cause critical fracture. Crack-growth rates, da/dt, are generally characterized as a function of the applied strain energy release rate, G (J/m2).1,3,11 Growth rates may span many orders of magnitude, from approximately 10−3 m/s to threshold growth rates less than 10−10 m/s. The threshold represents the value of G below which crack growth is essentially dormant and is particularly important because the majority of the lifetime of defects in device structures is spent at or near threshold. Accurate measurement of the crack growth rate in thin-film structures presents significant experimental challenges, particularly at very low growth rates where limited crack extension occurs. Crack length is generally obtained from elastic compliance relations for the test specimen, which increases with crack extension or by monitoring the crack length optically.12,13 Optical observation of the crack length is experimentally arduous and complicated in thin-films structures containing interfaces and layers of different optical contrast. Furthermore, this method becomes increasingly inaccurate at both high and
a)
Address all correspondence to this author. e-mail: [email protected] DOI: 10.1557/JMR.2004.0421 J. Mater. Res., Vol. 19, No. 11, Nov 2004
http://journals.cambridge.org
Downloaded: 06 Apr 2015
low growth rates. At high growth rates it is difficult to manually track the crack tip, and automated imaging techniques are often unreliable and analysis times are slow. Low growth rates are generally even more difficult to determine due to extremely limited crack extension that is often below the optical resolution of the microscope, even after extensive testing times. Post test fracture surface analysis to determine crack length precludes in situ modification of the testing parameters. Compliance-based techniques provide a far more convenient and accurate measure of crack extension an
Data Loading...
