Loess, a grayish yellow or brownish yellow, clastic, highly porous, brittle Quaternary aeolian sediment, covers approximately 10% of the Earth’s land surface 1,2. The Loess Plateau in mid-north China is the greatest bulk accumulation of loess on the earth 3. It covers an area of 624,641 km2 and spans seven provinces in China, making up 20% of arable land and supporting 17% of its population 4,5. As loess is a highly erosion-prone soil, the unique hilly-gully landscape was formed under the influence of human and natural activities for thousands of years 6. Due to inappropriate soil conditions and multi-hazard scenarios, local farmers were in poverty as their efforts can hardly be paid off. The tense relationship between nature and human beings developed gradually.
With the development of economic construction in recent years, urban areas on the Loess Plateau have expanded rapidly, and the restriction of mountainous terrain on available construction land is becoming increasingly prominent 7. To alleviate the nature-human tension and solve the shortage of construction land resources, local governments proposed the Mountain Excavation and City Construction (MECC) project 8. In cities such as Chongqing, Yichang, Lanzhou and Yan'an, tens of square kilometers of new lands have been created. As one of the largest MECC projects, the amount of earthwork in Yan’an new district exceeded 600 million m3 to create 78.5 km2 of flat land 9. Dozens of hills with 100–150 m in height were removed to fill in valleys. Compaction, the most cost-effective way 10, was used to improve the manual filling loess body. Its basic principle is to increase the density of loess by rolling or heavy tamping. The height of compacted loess filled in the valley could finally exceed 80 m. Besides the unprecedented scale, the coexistence of intact and compacted loess is another feature that differs from previous projects. Recent work by engineers has established that the two kinds of foundations have obvious differences 7,11−13. The intact loess is uniform in nature and has good engineering geological conditions. By contrast, the compacted loess has a different degree of compressibility and collapsibility. These differences make the two kinds of foundations prone to uneven settlements, and ground fissure often occurs at the contact place. Hence, the comparative study of these two kinds of loess is of great value to the stability analysis of the site. This will be directly related to the long-term stability of this kind of large artificial loess project and whether it can meet the demand of urban construction.
Recently, a large number of scholars have carried out research on the properties of intact and compacted loess, and have found that compacted loess is quite different from intact loess 11,13−16. Many laboratory experiments proved have shown that despite the compaction measures, remolded loess still has collapsibility which can be even more serious than original loess 17–20. Besides, water retentions of the compacted loess and the intact loess are also different 12,21. Compared to re-compacted loess, intact loess exhibits a more pronounced hysteresis in the soil-water retention curve (SWRC) for low suctions. However, re-compacted loess exhibits larger hysteresis than intact loess for high suctions. It is worth mentioning that, the macro-level hydrological and mechanical properties of soils either in their compacted state or intact state is profoundly affected by the microstructure, especially the pore structure 12,14,21−25. As the compacted loess has a distinct pore-structure in comparison to the intact loess, it behaves differently from the intact loess in response to hydrological and mechanical change 21,26,27. Thus, qualitative and quantitative research on the soil pore structure of compacted loess and intact loess is of great significance for agricultural cultivation, water and soil conservation and engineering construction.
Mercury intrusion porosimetry (MIP) and scanning electron microscopy (SEM) are the most widely adopted techniques in such pore studies. The investigations have mainly concentrated on the determination of the plane morphology, two-dimensional characteristics of the loess pores, pore size distribution (PSD) and corresponding microstructural parameters 28–32. However, SEM can only obtain 2D images and the results are highly dependent on the choice of observing directions, which will result in non-unique shape descriptors for the pores 33. Because of the principle that forcing a non-wetting liquid (often mercury) to intrude into a material at high pressure to get the PSD, MIP is destructive so that the results may not be true 34,35, and it cannot directly visualize the pore structure 11. Fortunately, with the development of new observation apparatuses and processing technologies, X-ray computed tomography (CT) has been used to investigate the pore structure of loess in recent years 1,26,36−40. It can provide the images of each layer inside the sample without interfering with the test process and obtain 3D results 26. Nonetheless, due to the exorbitant price and the limited scanning pixels, researchers can only get 3D reconstructed pore structures in size of hundreds of microns, when the CT scan resolution is up to 1µm 39,40. And such high-resolution CT studies are extremely limited. Considering the heterogeneity of loess, especially compacted loess, the accuracy of such studies needs to confirm whether the tiny 3D pore structure can represent the properties of conventional size samples.
Representative volume element (RVE) is an important concept in the mechanics and physics of random heterogeneous materials 41,42. The RVE is usually regarded as a volume of heterogeneous material that is sufficiently large to be statistically representative of the microstructural heterogeneities 43. If the volume is relatively small, a slight increase in size may result in a notable change in characteristics. When the volume is larger than a critical value, the characteristics would no longer change with an increase in size. This critical value is called the representative volume element (RVE) size 44. Knowledge of the RVE size has been widely applied to accurate analyses of three-dimensional fractures network of jointed rock mass 45–48. Similarly, it also contributes to the correct and accurate analysis of soil microstructure. This means that studies based on CT scans are representative if the tiny 3D volume corresponds to the RVE size. However, little research in this direction has been carried out.
Overall, the comparative studies on the 3D pore microstructure between intact loess and compacted loess based on the non-destructive CT scanning technology, especially at high resolution (1µm), is very limited. Moreover, whether such tiny three-dimensional pore structures are representative needs to be verified urgently.
In this paper, 3D pore microstructures of intact and compacted loess from the Loess Plateau were established based on serial CT images with a voxel size of 1 µm3. Quantitative analysis and comparison were carried out to characterize the parameters of 3D structure volumes with different sizes. And geometrical RVE sizes of the microstructure of intact and compacted loess were determined. This research is expected to provide an in-depth and comprehensive understanding of the 3D pore geometrical characteristics of intact and compacted loess and insights into the study of the mechanical mechanism of loess behavior.