受應變率影響之極軟弱砂岩力學行為及其模式

Abstract

台灣中北部之麓山帶地區之軟砂岩，在地質年代上屬於極年輕的地層，其特性為固結差、膠結不良及強度不高。因此會對工程之安全性產生影響。若於此等岩層中進行大地工程，應考慮因為強度與應力應變關係受到潛變等依時性力學行為之影響，工程上應採用的設計參數，亦應考慮時間效應之影響。因此於此種軟弱岩石的依時性行為，有必要加以掌握與了解。 本研究改良原有之傳統呆重式潛變試驗設備，使其適用於進行控制應變速率之載重試驗型態。研究中以台中縣大坑鄉附近之極軟弱砂岩為材料，針對軟砂岩的依時性行為加以探討，分析時間因素對其力學行為影響，並嘗試以連體損壞力學理論所建立的組合律模式為基礎，加入時間因素之考慮，而推導出依時性力學模式，來詮釋並預測軟岩材料受加載速率影響之強度及變形性。 由本研究之試驗結果發現，單壓試驗中尖峰強度與初始彈性模數會隨破壞時間增加或應變速率減少而下降，由配合單壓試驗所擷取之聲射累計數比較發現，應變速率愈低，則累計聲射事件數愈高，反映潛變效應造成材料累積損壞增加而影響力學性質，因此應該適合以連體損壞力學模式之概念來描述其依時性之力學行為。基於此觀點，本研究提出一受應變率影響之力學模式來描述軟砂岩之依時性行為。此模式為修改自原為非依時性連體損壞力學模式。由本文所提出之模式對試驗之模擬與比較，可看出此模式確實可合理描述極軟弱砂岩與應變率相關之應力應變行為。

The outcrops in the northern and central regions of Western Taiwan often contain very weak sandstone. These young formations are usually poorly cemented and their strengths lie within "very weak" and "extremely weak". Stability problem may be a concern for these materials. For very weak rock, the strength and stress-strain relations are likely affected by time. Engineering analysis and design, hence, should take into account the effect of timedependence. This project aims to study the time-dependent behavior of the very weak sandstone. Conventional dead-load type apparatus of for rock creep test was first modified to enhance rate-controlled tests on very weak rock material. Then, a series of strain-rate controlled uni-axial compression tests of were carried out to investigate the rate-dependent behavior of very weak sandstone. From the experimental results, it was observed that both the peak strength and the modulus of the very weak sandstone decrease as the failure-time increases or the strain-rate decreases. Also, a specimen loaded in a lower strain-rate tends to show more acoustic emission. On the other words, the creep effect seems to result in more micro-damage in the specimen. Consequently, a constitutive model on the basis of the continuum damage mechanics was developed to describe the rate-dependent mechanical behavior of the material. By comparing the simulated and experimental results of stress-strain relations, the proposed model seems to reasonably capture the rate-dependent behavior of very weak sandstone.