A study for the prediction of precursory slips based on the contact mechanics of asperities

本研究包括以下三个项目。压痕实验。我们主要对铝、铜、黄铜等金属样品进行了一系列的压痕划痕测试。我们发现，随着压头水平运动的开始，压头深入到样品中更深，随着水平运动的进行，压头向上浮动。用模型进行了模拟，结果与观测结果吻合较好。我们关注的是初始摩擦力总是略高于稳态摩擦力的事实。这可能与静摩擦大于动摩擦的事实有关。通过断层上的透射波来探测前兆滑动的试验。使用东京大学ERI的双轴夹层型仪器，我们进行了一系列的粘滑实验，其中P波和s波通过断层传播。我们发现透射波对断层的接触状态很敏感。随着剪切应力的增大，透射波的振幅显著增大。然而，随着前兆滑移的发生，增加速率略有下降。这被解释为更换粗糙触点的影响。实验研究了断层几何形状（粗糙体高度、粗糙体大小、间距等）对透射波的影响。弹性波通过人工断层传播。断层表面形成了不同宽度和深度的平行沟槽。我们发现，波的透射率完全由沟槽深度/宽度与波长的比值决定。我们还建立了一个基于断层刚度的数学模型。该模型很好地解释了当波长为接触颗粒尺寸的4倍时的观测结果。

This study consists of the following three projects.1. Indentation experiment. We performed a series of indentation scratch tests mainly on metal samples, such as aluminum, copper, and brass. We found that the indenter penetrated deeper into the sample with the onset of horizontal movement of the indenter, and that it floated up with the progress of the horizontal movement. A simulation was performed with a model, which well agreed with the observation. We focused on the fact that the initial friction force was always slightly higher than the friction at steady state. This may correlate the fact that static friction is larger than the dynamic friction.2. An experimental trial to detect precursory slips by transmission waves across a fault. Using a biaxial sandwich type apparatus at ERI, the University of Tokyo, we performed a series of stick-slip experiments in which P- and S-waves were transmitted across the fault. We found that the transmitted waves are sensitive to the contact states of the fault. With the increase in shear stress, the amplitude of the transmitted waves significantly increased. However, the increasing rate was slightly decreased with the onset of the precursory slips. This is interpreted as the effects of the replacement of asperity contacts.3. An experiment to investigate the effects of fault geometry (asperity height, asperity size, spacing etc.) on the transmission waves. Elastic waves were transmitted across an artificial fault. Parallel grooves with various widths and depths were cut on the fault surfaces. We found that the transmissivity of the waves is solely determined by the ratio of the groove depth/width to wavelength. We have also made a mathematical model based on the stiffness of the fault. The model well explains the observation when the wavelength is 4 times longer than the size of contacting asperities.