|Abstract: ||利用核能發電已經為世界各國主要電力來源之一，但是核子燃料具有高放射性，且半衰期長達數千年甚至數十萬年，而在衰變過程中產生之衰變熱，會對周圍產生溫度場效應，如何將用過核子燃料與生物圈完全隔絕，是各核能先進國家之共同目標。目前對於用過核子燃料處置方式一致較推崇「深地層處置」(Deep Geologic Disposal)概念，用以隔絕與減緩放射性核種釋放遷移。而在深地層處置場主要受到四大因素影響，包含T熱學(Thermal)、H水力(Hydraulic)、M力學(Mechanical)、C化學(Chemical)因素，稱為T-H-M-C耦合效應，通常為兩項或兩項以上交互作用，進而影響最終處置場之預期功能。|
本研究針對熱-水-力耦合試驗分為小型試驗及數值模擬比對，首先小型試驗延續林柏吾(2017)小型熱-水耦合試驗並改變方式進行試驗，以先加熱後進水方式，此方式較符合於實際情況，當緩衝材料受到衰變熱影響之溫度變化進而再受到地下水入侵之模式，以時域反射法(time domain reflectometry, TDR)，同時試驗不同初始乾密度(1.4 g/cm3、1.5 g/cm3、1.6 g/cm3)小型試驗，並建立視介電常數-溫度-體積含水量三相圖。但由於試驗之膨潤土具有較高的回脹壓力及回脹潛能，吸水後將產生回脹，對試體內部產生擠壓，造成孔隙改變，即乾密度的變化，可以觀察到試體底部因回脹向上擠壓，同時又將TDR感測器往上推。由於所建立之視介電常數-溫度-含水量三相圖在換算時需對應其乾密度，在乾密度改變的情況下，將會造成計算上誤差，無法反應出確實的體積含水量，因此在分層體積含水量計算上出現不符合實際情況之歷時含水量變化曲線。同時為驗證數值模擬之可信度，本研究除考慮回脹應變設定外，包含因不同溫度對對內部試體進水程度的差異，透過實際乾密度量測檢定，數值模擬之結果與量測值很相近，顯示本研究提出之數值模擬有一定的可信度，未來可針對深地層處置場情況需求進行模擬，並建立大型之數值模擬，以符合處置場的實際狀況。
;The use of nuclear power generation has become one of the major sources of electricity in the world, but the nuclear fuel waste is tricky to treat due to its radioactive characteristic that the half-life of decay is as long as hundreds of thousands of years. Additionally, the decay heat generated during the decay process will have a temperature field effect on surroundings. How to completely isolate the used nuclear fuel from the biosphere is the common goal of all advanced nuclear energy countries. At present, the treatment for used nuclear fuel is based on the concept of “Deep Geologic Disposal”, which intends to isolate migration of radioactive species in deep formations. However, it is mainly affected by Thermal, Hydro, Mechanical, and Chemical factors, called T-H-M-C coupling effect, and usually two or more factors interact, leading the unexpected function of the final disposal site.
To access the THM coupling effects, this study proposed a small-scale physical modeling and the related numerical simulation. Firstly, the small-scale test continues Lin (2017) T-H coupling test, and the test scenario was applied in terms of heating specimen first and water intrusion subsequently. This is much in accordance with the actual situation when the buffer material is subjected to the temperature change caused by the decay heat initially and then subjected to the groundwater intrusion in field. Time domain reflectometry (TDR) was used as the core method to observe the water intrusion, and different initial dry density sets (1.4 g/cm3, 1.5 g/cm3, 1.6 g/cm3) were applied in the small-scale test. Three-dimensional diagrams of apparent dielectric constant, temperature, and volumetric water content were established accordingly. Due to the high swelling pressure of the tested bentonite, it produced re-expansion after water absorption. The inside of the specimen was squeezed, causing the changes of pores and making the dry density be changed consequently. Furthermore, it was observed that the bottom of the specimen was squeezed upward because the fixed boundary, while the TDR sensor was pushed up. Since the volumetric water content of the three-dimensional diagrams needs to correspond to dry density, calculation error produced in the case of a change in dry density. Therefore, the current observations of water content variation did not meet the actual situation.
To verify the feasibility of the numerical simulation, this study adopt influence factors of dry density changes, swelling due to saturation, and the infiltration rate as function of temperature of the internal sample. The calculations of each dry density sample were performed and similar to those of the manual sampling, indicating a certain degree of confidence in the numerical simulation. Possible future demand for the case of deep geologic disposal field simulation, and the establishment of large-scale numerical simulation to be satisfied the actual situation of the disposal field