Fracture Strain of Lpcvd Polysilicon

A new polysilicon bridge-slider structure (Fig. 1), in which one end of the bridge is fixed and the other is connected to a plate sliding in two flanged guideways, is designed and fabricated to study the strain at fracture of LPCVD polysilicon. In the experiments, a mechanical probe is used to push against the plate end, compressing and forcing the bridge to buckle until it breaks. The distance that the plate needs to be pushed to break the bridge is recorded. Nonlinear beam theory is then used to interpret the results of these axially-loaded-bridge experiments. The measured average fracture strain of as-deposited LPCVD polysilicon is 1.72%. High-temperature annealing of the bridgesliders at 1000 °C for one hour decreases the average fracture strain to 0.93%. INTRODUCTION Polysilicon has been demonstrated to have useful applications for sensors and actuators. Moreover, complex polysilicon micromechanisms have been shown to be feasible [1,2] using pin joints, gears, springs, and cranks. As a result, new devices like micro-motors and micro optical shutters are becoming possible. Proper design of these structures is hindered, however, by the lack of precise knowledge of the mechanical properties of polysilicon. Of these parameters, strain at fracture is particularly important. For single-crystal silicon, Eisner [3] reported a maximum fracture strain of 2.03% measured on whiskers about one micrometer in diameter under tension. Pearson, Read, and Feldmann [4] reported a maximum fracture strain of 2.6% for silicon whiskers both grown from vapor and cut from bulk silicon. Moreover, it is found that below 600 °C there is little or no plastic flow in silicon whiskers about 20 Jlm in diameter before fracture. It is expected that polysilicon will also behave linearelastically before fracture at room temperature because grain boundaries in polysilicon can greatly block dislocation motion [6] and make the polysilicon more like an ideal brittle material. Based on this assumption, Fan, Tai, and Muller [5] reported preliminary experiments which determined an experimental fracture strain of polysilicon to be 1.7% using a spiral-spring-restrained pin-joint structure. However, such a spring-restrained pin-joint structure is not optimal for the fracture experiment. Moreover, it is generally accepted that a statistical method should be used to study the fracture strength in brittle materials [6]. We report here a systematic method to study the fracture strain of polysilicon. We introduce a new, easily implemented method using a bridge-slider structure to avoid the difficulties of handling small samples as reported in the silicon whisker experiments [3,4]. The bridge-slider structure is specially designed and processed to improve experimental accuracy. BRIDGE-SLIDER STRUCTURE Figure 1 shows an SEM photograph of the newly designed bridge-slider structure. The right end of the free-standing bridge, shown in Fig. 1, is anchored to the silicon substrate, while the left end is connected to a sliding plate guided by two flanges. The outer edges of the slider flanges are sawtooth shaped with a tooth pitch of 4 Jlm to provide scales for locating the end of the slider. This sawtooth feature greatly simplifies our experiments. Figure 2 shows a cross section of the slider to demonstrate its translational freedom of motion. Clearly, the slider, made of second-layer polysilicon, is fully separated from the restraining elements that are made of first-layer polysilicon. The flanges one of which is circled in Fig. 2 allow the slider only to slide in and out of the plane of Fig. 2. Figure 1 A bridge-slider structure. The bridge-slider is made of second-layer polysilicon. Integrated sawtooth scales are defined at the outer side of flanges. The marker is 30 Jlm in length. TH0215-4/88/0000-0088 is) 1988 IEEE