| 摘要: | 本研究以峴港砂為水中回填之主要材料，經水泥改良後，探討水泥改良砂土經沉降堆積後之力學特性、抗液化機制及改良效果評估，並建立一較佳配比模式，可供填海造地工程實際應用與防制液化災害之參考。研究方法乃首先以初始乾單位重與初始相對密度為主要控制變因，再於不同水泥配比、養治時間、細料含量、水泥種類、人工海水鹽度與抗分散劑等條件下，分別採用濕搗法和水中沉降法製作改良土試體，進行一系列之無圍壓縮試驗與動力三軸試驗。 本研究之試驗結果顯示，當以濕搗法為準備試體之方式時，添加水泥於砂質土壤內，對於剪力強度之提升有顯著的效果，且改良土之無圍壓縮強度隨著初始乾單位重、水泥配比及養治時間之遞增而顯著地增強。另一方面，由於主要控制變因之差異與所含細料種類的不同，隨細料含量之增加，無圍壓縮強度則分別呈現持續上揚（控制初始乾單位重，添加峴港砂細料）或先升後降（控制初始相對密度，添加高嶺土或峴港砂細料）之不同趨勢。再者，當改良砂土所處環境之鹽度愈高，晚期無圍壓縮強度則有明顯降低之趨勢。 改良砂土於陸上回填時，具有很高之剪力強度，但就填海造地工程而言，改良砂土必須於水中進行回填，此時由於水泥與砂土顆粒產生分離現象，將導致改良土強度大幅損失約90 %以上，故本研究藉由添加抗分散劑之方式，利用其電荷吸附及架橋作用，可有效地將砂土顆粒與水泥包覆成團，進而同時沉降，減少水泥與砂土顆粒產生分離現象，促使水化作用得以有效發揮，可明確地有助於改良砂土剪力強度之增強。其中，抗分散劑種類之選擇，以聚丙烯醯胺類為最佳，可有效減少無圍壓縮強度之損失率，且損失率最低者可達41 %。 另外，以改良材料使用波特蘭水泥第I型，採用峴港砂細料為細料添加物之前提下，本研究提出一水泥改良土無圍壓縮強度預估模式，此模式藉由初始乾單位重、水泥配比、細料含量、鹽度與養治時間等五種參數，可準確預估水泥改良土之無圍壓縮強度。 在本研究試驗範圍內，當試體準備方式採用濕搗法者，於初始相對密度為40 %、高嶺土細料含量為10 %、鹽度為3.5 %，養治時間28天等條件，砂土添加水泥配比1.5 %者，或當採用水中沉降法進行試體準備時，以聚丙烯醯胺類為抗分散劑且添加率為344 mg/kg，水泥配比為6 %及養治時間為28天者，上述改良土之液化阻抗均可獲得明顯地改善。 有鑑於孔隙水壓比激發曲線預估模式大部分以純砂（少數以黏土）為對象，所建立相關預估模式未能充分呈現水泥改良土之孔隙水壓比特徵，故本研究提出一水泥改良土之修正超額孔隙水壓比激發曲線預估模式。此預估模式可有效兼具DeAlba等人（1975）和Sakai等人（1994）預估模式之優點，並且改善前述兩模式之缺陷。因此，若欲探討與模擬水泥改良土超額孔隙水壓比激發曲線可採本研究所提之模式較佳。 In this study, Danang Sand was used as the main material for backfilling in water. The purpose of this study is focused on the mechanical properties, liquefaction mechanisms and improved effects of cement treated sand after the procedure of water sedimentation. A optimum ratio model was established as a suggestion for reclamation in water or a prevention method of liquefaction hazards. The first step of research was to control initial dry weight or initial relative density, and then wet tamping method and water sedimentation method was used to prepare soil specimens under the condition of different cement content, curing time, fine content, cement type, salinity and coagulant, respectively. A series of unconfined compression tests and dynamic triaxial tests were carried out. The results of this study showed that the shear strength of cement treated sand prepared by wet tamping method increased significantly as the increase of initial dry unit weight, cement content and curing time, respectively. On the other hand, due to the difference between the control of main variable and types of fine materials, there exists different tendencies with the increase of fine content. Under the control of initial relative density and adding Danang fines in cement treated sand, the unconfined compression strength raised continuously with the increase of fine content. When the control of main variable was initial relative density, the unconfined compression strength increased first and then decreased with the increase of fine content, whether adding Danang fines or Kaolinite. Furthermore, the unconfined compression strength presented a clear trend to reduce with the increase of salinity of environment when curing time exceeded or was equal to 28 days. The cement treated sand can provide great shear strength for land reclamation. However, for the projects of reclamation in water, the cement treated sand will occur a separation between cement and sand particles, and that caused a significant loss of about 90 % of shear strength. Therefore, in this study, some coagulants were used and added into the cement treated sands. Because the coagulant can coat the sand particles and cement effectively, to prevent cement treated sand from separating by charge adsorption and bridging, and hydration of cement can act effectively and continuously. Then the shear strength of cement treated sand can be enhanced. Among them, the best type of coagulant is polyacrylamide which can effectively reduce the loss of unconfined compression strength, and the lowest loss rate can be reduced to 41 %. In addition, under the condition of using the Portland cement type I, Danang fine and Danang sand, this study recommends a model of unconfined compression strength for cement treated sand. This model includes five parameters, such as initial dry unit weight, cement content, fine content, salinity and curing time. The unconfined compression strength of cement treated sand can be assessed accurately. In the scope of this study, when the specimens were prepared by wet tamping method, under the conditions of initial relative density of 40 %, fine content of Kaolinite of 10 %, salinity of 3.5 %, curing time of 28 days and cement content of 1.5 % or when using the water sedimentation method, under the conditions of cement content of 6 %, adding ratio of polyacrylamide of 344 mg/kg and curing time of 28 days, the liquefaction resistance of the cement treated sand mentioned previously can be significantly improved. In view of most of the presented excess pore water pressure ratio curve models are established by pure sand, these models cannot present the characteristics of excess pore water pressure of cement treated sand properly. Therefore, this study recommends a modified model of excess pore pressure ratio curve which has the advantages of model of both DeAlba et al. (1975) and Sakai et al. (1994), and can improve the shortcomings of this aforementioned two models. So, the modified model mentioned in this study can offer a better exploration of the excess pore water pressure ratio of cement treated sand. |
