非擾動粉土細砂試體隻Ko壓密三軸試驗

Abstract

於台灣西部平原在台中以南、台北盆地以及東部的宜蘭與花蓮，皆有厚層的粉土質砂土，在此等土壤上進行的土木工程不計其數。但過去國內外有關的研究中，很少有使用台灣粉土質砂土作為試驗對象，因此對於其工程上的行為認知十分缺乏，有些甚至是錯誤的。雖然數量有限，但是所有研究結果都顯示重模試體並不能足以能夠代表不擾動粉土質砂的力學行為。因此若要取得台灣西部粉土質砂的力學行為，必須使用非擾動試體才能有代表性。在以往研究中三軸試驗的壓密方式，大都是進行等向壓密，鮮少進行Ko壓密。但是沉積土壤在大自然中壓密的狀態，是最接近沒有側向應變，也就是在Ko狀態下壓密。所以為了了解土壤真正在自然界中壓密過後的各種力學行為，在實驗室中的三軸試驗，以Ko狀態進行壓密也是有其必要性的。為了有效的解釋台灣粉土質砂土之現地或是室內試驗結果，必須首先能夠掌握這些地質背景以及顆粒特性對於工程性質的影響，並且必須以不擾動試體作試驗其結果才有代表性。因此本論文之目的為以嘉義太保所取冰凍不擾動試體，使用交通大學大地工程試驗室自行研發的側向應變量側設備，進行三軸Ko壓密試驗，且於壓密後進行靜態三軸試驗，期盼能夠明瞭具有細粒料砂土之工程特性，以提供台灣西部地區非塑性粉土細大地工程分析之用。試驗結果顯示Ko壓密重複性試驗有良好之重複性，代表研發之側向量測系統能準確量測側向應變。但以不擾動冰凍試體施作試驗之結果因無明顯規律性，無法作出合理解釋。原因可能為試體於採樣、冰凍或解凍時，已產生擾動；或者此取樣地層土壤並不均勻，有許多互層，造成試驗結果差異甚大。比較不擾動與重模試體之試驗結果，可明顯發現其工程特性的確相差許多。當試體受到剪動時，不擾動試體會呈現超額孔隙水壓，而重模試體為正的；強度部分，不擾動試體所能承受之極限軸差應力比重模試體高出許多。

Numoerous civil engineering projects have been built on the silty sand layer on the western plain in Taiwan. Due to the lack of research on silty sand, our knowledge on the engineering behavior of this type of material is limited and sometimes misleading. Though the numbers are limited, the available results have indicated that tests on reconstituted samples could not fully reflect the behavior of undisturbed sand. Also, to better simulate the field state of stress, it is necessary to consolidate the specimen under Ko conditions in the triaxial tests. The objective of this research is to further our understading of silty sand by performing Ko consolidated triaxial compression tests on undisturbed specimens. Soil samples were taken from Chia-yi cownty using a Densison Sampler. A triaxial testing device developed at the National Chiao Tung University Geotechnical Laboratory was used for the experiments. Upon triaxial tests, the undisturbed specimens were reconstituted to similar densities and tested again for comparison purposes. The results showed reasonable repeatability in Ko values and stress-strain relationships from tests on reconstituted specimens. Significant scattering was noticed form tests on undisturbed specimens. The scatering may be a result of variability or layering of natural sands. The undisturbed specimens developed negative excess pore pressure while the reconstituted specimens tested under the same conditions had positive excess pore pressure. The peak stress differences from reconstituted specimens were consistantly lower than those of undisturbed specimens.