Dissociative Chemisorption ofO2Inducing Stress Corrosion Cracking in Silicon Crystals

Abstract

Fracture experiments to evaluate the cleavage energy of the ð110Þ½11 0 and ð111Þ½1 12 cleavage systems in silicon at room temperature and humidity give 2:7 AE 0.3 and 2.2 AE 0.2 J=m 2 , respectively, lower than any previous measurement and inconsistent with density functional theory (DFT) surface energy calculations of 3.46 and 2.88 J=m 2 . However, in an inert gas environment, we measure values of 3.5 AE 0.2 and 2.9 AE 0.2 J=m 2 , consistent with DFT, that suggest a previously undetected stress corrosion cracking scenario for Si crack initiation in room conditions. This is fully confirmed by hybrid quantum-mechanicsmolecular-mechanics calculations.

Low Speed Crack Propagation via Kink Formation and Advance on the Silicon (110) Cleavage Plane

“…These two conjectures have not been universally confirmed by experiments, some of which have detected cracks moving at less than 1% of the Rayleigh speed at both room temperature and 77 K [16]. We note, however, that these experiments were carried out in air, and the measured speed is low enough to enable propagation by stress corrosion [17].…”

“…Namely, the energy-speed relationship was examined for the low energy regime on the ð110Þ½110 low energy cleavage system of silicon. The experiments were carried out in an inert argon environment to prevent chemically activated ("stress corrosion") cracking [17]. We evaluated the crack speed from the fracture surface features using the Wallner lines method, calculating the energy release rates G at each point of the specimen by finite element analyses [39].…”

In situ stable crack growth at the micron scale

“…As the experiment is performed in displacement control, the driving force decreases with crack length and therefore cracking is inherently stable 17–19 . Furthermore, in situ testing allows the atmosphere to be controlled, reducing concerns of environmentally induced effects such as stress corrosion cracking 20 . In our test, stable crack growth is demonstrated through a prolonged displacement hold (between 150 and 300 s) with no observable increase in crack length.…”

Fatigue in Nanometric Single-Crystal Silicon Layers and Beams

“…Finally, the observed increase of the SiO 2 thickness after fatigue cycles is another validating point [17], [21]. Poor initial evidences of static fatigue were in early paper taken as a proof against this model: however recent works markedly showed the existence of static fatigue [25]- [28] with stress According to the second model, fatigue is instead a result of crack propagation in the Silicon itself [7], [8], [19]. Roughness, wear, cusps and debris, resulting from the manufacturing of the device or from the cyclic stress, can produce a lever effect inside native surface cracks in Silicon: during the compressive phase of the load, these imperfections can generate a stress amplification at the crack tip, inducing the crack propagation.…”