Experimental investigation on the expansive characteristics of Gansu clay-sulfate rocks in China subjected to confined compression and cyclic drying-wetting

doi:10.21203/rs.3.rs-2063123/v1

Download PDF

Short Report

Experimental investigation on the expansive characteristics of Gansu clay-sulfate rocks in China subjected to confined compression and cyclic drying-wetting

https://doi.org/10.21203/rs.3.rs-2063123/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Gansu clay-sulfate rock is a special red-bed rock that formed in an evaporation environment with strong oxidation. Due to the high content of hydrophilic minerals (including clay mineral, sulfate, and carbonate), tunnel in clay-sulfate rock stratum has a potential risk of large expansive deformation after the erosion of groundwater. In this study, a loadable cyclic drying-wetting instrument was designed to simulate the occurrence of the surrounding rock mass of a tunnel experiencing a periodic change in groundwater level. Firstly, a series of drying-wetting cycle tests with different normal pressure were carried on to obtain the expansion regulation of clay-sulfate rock. Then, computed tomography (CT) and scanning electron microscope (SEM) tests were carried on to explain the corresponding mechanism in mesoscopic. The test results show that large expansive deformation was generated in the first water immersion and only part of the deformation can be recovered during the later drying process. During the 2–5 cycles, expansion-shrink deformation is much less than in the first cycle, and the volume of the sample tends to be constant. Meanwhile, the degeneration of the clay-sulfate rock was observed in the first cycle by the CT and SEM test. Then the mesostructure of the sample is stable during the 2–5 cycle. It indicates the clay rock, in large part, degenerated to clay soil during the first drying-wetting cycle. Structural degradation and hydration of anhydrite is the main source of irreversible expansive deformation. Moreover, normal stress and confined restraint can reduce or even eliminate the expansive deformation, however, cannot avoid the degeneration of mesostructure. The shear strength of Gansu clay-sulfate rock also decreased after the first drying-wetting cycle, which manifested as the attenuation in both cohesion and internal friction angle.

drying-wetting cycle

clay-sulfate rock

coupling effect

mesostructure

expansive deformation

In tunnel construction or mineral excavation, the cyclic changes in hydrogeological conditions may cause engineering hazards, especially in the regions of sulfate-bearing rock (Kwon et al. 2009; Butscher et al. 2011). However, the mechanism of the water-rock effect remains poorly understood because the expansion of clay-sulfate rocks is controlled by coupled hydraulic, chemical, and mechanical processes.

Over the past decades, researchers conducted numerous laboratory tests to investigate water-rock interaction, including water-softening, expansive (or swelling), slaking (or disintegration) and etc. In most of the tests, clay-bearing rocks were regarded as water-sensitive (Erguler et al. 2009; Liu et al. 2018; Sadisun et al. 2005; Gautam et al. 2013; 2016). Some clay-bearing rock even exhibits behavior intermediate between sedimentary rock and clay. This kind of semi-rock tends to fully disintegrate when exposed to water directly (Cripps et al. 1981; Alonso et al. 2010; Gasc-Barbier et al. 2007). The clay-sulfate rock collected in Gansu is a typical red-bed semi-rocks that belongs to both clay-bearing rock and sulfate-bearing rock. The sulfate in the rocks is partly hydrated and the cementation of the mineral is weak, which is different from the clay-sulfate rocks with high content of anhydrite in Germany and Switzerland (Steiner, 1993; Pimentel, et al. 2013; Butscher, et al. 2018). Meanwhile, its hydraulic properties are more complex, because the content of carbonate minerals is high. In sum, Gansu clay-sulfate rock is an expansive semi-rocks with a special composition.

In underground engineering, confined expansive deformation is more common than slaking of unconstrained rock due to the restraint of supporting structures or surrounding rocks (Li et al. 2016), and drying-wetting cycles caused by the change in water level are more common than long-term soaking. A better understanding of the physical and mechanical properties of constrained rock subjected to periodic water erosion remains an ongoing research issue. Some novel instruments and test methods were proposed (Hu et al. 2019; Zeng et al. 2021) in recent years. The mineral arrangement and mesostructure are the internal factors that lead to the expansion of rock samples and CT and SEM tests can identify mesoscopic and internal damage in rock materials (Jia et al. 2013; Kuang et al. 2022). Considering the sample is special and the lake of systematical investigation, the meso-mechanism of the confined expansive phenomenon and its main impact factor (maybe expansive mineral content, cementation structure, or others) is not understood clearly. The experimental and theoretical investigation of expansive mechanisms needs to be supplemented.

The main objective of this research is to understand the water-induced degeneration of clay-sulfate rock after cyclic drying-wetting treatments. In the section “Materials and test methodology”, the physical and hydraulic properties of the rock sample and the theory of the loadable cyclic drying-wetting tests were introduced. Then, the drying-wetting cycle tests were carried on and expansive (swelling) characteristics of clay-sulfate rocks were analyzed in the section “Test results”. In the section “Mesoscopic expansive mechanism”, the influence of structural degeneration on the expansion regulation of the clay rock was analyzed based on CT and SEM tests. The last section presents the discussion about the influence of drying-wetting cycles on shear strength and conclusions were presented.

Material Properties

The rock sample, which is a Neogene red-bed argillaceous rock, was collected from a tunnel under construction of the G8513 expressway, Gansu, China (Fig. 1). The basic physical properties test shows that the rock sample has low moisture content and high expansive pressure, as indicated in Table 1.

Table 1

Physical properties of rock samples
Porosity (%)	Dry density (g·cm^− 3)	Natural density (g·cm^− 3)	Moisture content (%)	Free expansion ratio	Expansive Pressure kPa
11.2 ~ 14.6	2.07 ~ 2.24	2.05 ~ 2.46	5.80 ~ 7.66	2.28*	525

To understand the source of expansion pressure and the mineral composition of the rock, the result of X-ray diffraction (XRD) tests were performed (Table 2).

Table 2

Mineral composition of rock samples (%)
sulfate		clay mineral		carbonate		silicate (non-clay)			others
gypsum	anhydrite	illite	chlorite	dolomite	calcite	quartz	muscovite	feldspar	others
17	6	19	7	4	9	18	3	15	2
*Other minerals include hematite, mirabilite, biotite, et al. whose content is less than 1%

As shown in Table 2, The sample consists of 36% non-clay silicate, 26% clay mineral, 23%sulfate and 13% carbonate. It is notable that 18% of the content is gypsum. It means 30% of sulfate is still not hydrated and the rock belongs to the clay-sulfate rock (Butscher et al. 2016; 2018). The expansive potential of anhydrite (about 61%) and clay mineral is the main cause of the disintegrating (slaking) behavior of clay-sulfate rocks. The slake durability test (ISTM 1981), was performed to assess the water resistance of undisturbed clay-sulfate rock. Five drying-wetting cycles were completed instead of the usual two cycles to obtain long-term stability in drying-wetting cycles. The slaking process and history curve of residual mass is shown in Fig. 2.

Based on the 2nd cycle slake durability index (I_d2), the n^th cycle slake durability index can be obtained from Eq. (4).

$${I_{{\text{dn}}}}=\frac{{{M_{\text{n}}}}}{{{M_0}}} \times 100\%$$

(4)

where I_dn is the n^th cycle slake durability index, M_n is the residual mass (whose grain size is larger than 2mm) of the sample after n cycles, M₀ is the initial mass of the sample, m₀ = 497.75g in this test.

As shown in Fig. 2-(b), the residual mass is a convex function about cycles, which reflects an accelerated slake tendency. The 2nd cycle slakes durability index is 0.798, and the 5th cycle slakes durability index is only 0.206.

Instrument and method of the drying-wetting cycle test

Stress state and drying-wetting transmission are the two key variables studied in this paper. A cyclic drying-wetting instrument was designed and created to simulate the occurrence of in-situ rocks, which are subject to a confined compression state. The sketch and photo of the instrument are shown in Fig. 3.

As shown in Fig. 3, the instrument consists of three main parts: the loading system (part 1), the drying-wetting chamber (part 2), and the confined restraint device (part 3). The pressure system is modified from a multi-function compression test machine with a stress sensor. The test chamber is made of high-strength glass with valves on top and bottom. The cyclic drying-wetting is controlled by a bottom valve (2–5 in Fig. 3) which is connected to a pressure water pump. Meanwhile, to promote the absorption of rock, vacuum pressure is necessary. The pressure in the chamber is controlled by a top valve (2–2 in Fig. 3) connected to a vacuum pump by a rubber tube. The function of the confined restraint device is to restrict the lateral deformation of the test sample. The circumferential expansion was completely constrained and axial expansion is adjustable.

The function of the instrument in Fig. 3 extended beyond the cyclic drying-wetting cycle coupled with normal compression and confined restrict. The expansive potential and expansive pressure could be measured without removing the sample from the instrument. In terms of the expansive stress test, by limiting the normal displacement, the history of expansive force would be recorded by the stress sensor. The expansive pressure can be calculated by Eq. (6)

$${p_n}=\frac{{{F_n} - {G_0}}}{{{A_0}}}$$

(6)

where, p_n is the expansive pressure of the sample after n times drying-wetting cycle, and G₀ is the gravity of the indenter and permeable stone above the sample. A₀ is the cross-section area of the sample, and F₀ is the normal force recorded by the stress sensor of the instrument.

In terms of the confined expansive ratio test, the history of expansive strain would be calculated by the data of normal displacement of the indenter. Due to the confined function of the instrument, the sectional area of the specimen does not change, the expansion strain can be calculated from Eq. (5) by the height of specimens:

$${\varepsilon _{\text{n}}}=\frac{{\Delta h}}{{{h_0}}}$$

(5)

where Δh is the increment of specimen height; h₀ is the initial height of the specimen.

To avoid the impact of mineral hydration and weather, dry-coring and wrapping measures (plastic film) were used during the core collection. Moreover. removing the weathering layer was also necessary before the test. The processing of the undisturbed sample is shown in Fig. 4. A Φ100mm rock core was cut into 30 ± 5mm pieces. Then, the pieces were re-cored horizontally by a Φ64.8mm×40mm ring cutter. Finally, the specimens were polished into Φ61.8mm×20mm standard by the grinder and installed into the confined restraint device.

To study the coupling effect of the confined compression and cyclic drying-wetting transmission on the clay-sulfate rock, five groups of tests were conducted under the normal pressure of 0, 100, 200, 400, and 600kPa, respectively. The maximum normal stress 600kPa was greater than expansive stress.

Considering the environment of underground rock, the drainage consolidation method was used instead of the oven evaporation method in drying-wetting cycles. In the tests, the preset normal pressure was applied and maintained until the test over. When the normal stress had been stable, the samples were soaked for 12h and then consolidated for another 36 h as a drying-wetting cycle. In total, 5 cycles were repeated.

Expansive characteristics in the drying-wetting cycle

The history curve of expansive volume strain during the first cycle is presented in Fig. 5.

When the normal stress is below the expansive pressure, the undisturbed samples experienced a short decrease in volume and then started to expand during the immersion process. The compression deformation indicates that the soft phenomena were represented early than expansion. In the consolidation process, some of the expansive deformations recovered, however, a large amount of deformation was irrecoverable. This phenomenon is due to the irreversible hydration of anhydrite and the consolidated pressure is limited. The maximum strain and consolidated strain (minimum strain in the consolidation process) during the first drying-wetting cycle move down when the normal stress increases. When the normal stress reached 600kPa, which is larger than expansive pressure, the expansive behavior disappeared in the immersion process. This observation shows that the normal stress could suppress or even eliminate the expansion of clay-sulfate rock in macroscopic.

The maximum and consolidated volume strain of the clay-sulfate rock during the five drying-wetting cycles under different normal pressure is present in Fig. 6.

It is observed that variation of both maximum and consolidated volume is less than 5% during the second to the fifth cycle. When the cycles are the same, the bigger the normal stress is, the smaller the consolidated volume strain is. The recoverable expansion is positively related to normal stress.

Mesoscopic Expansive Mechanism

The expansion of clay-sulfate rocks involves both “clay expansion” and “sulfate expansion”. In terms of clay expansion, the osmotic water inflow between the surfaces of clay may be the major resource. In addition, sulfate expansion may belong to inner crystalline expansion, which is caused by sulfate hydration (from anhydrite to gypsum). The osmotic water can be dried out by pressure and clay expansion may be recoverable. However, the combined water cannot be dehydrated spontaneously by pressure. To accurately analyze the reason for irreversible expansive strain, the undisturbed sample, and sample after one drying-wetting cycle were scanned by scanning electron microscope (SEM) and computer tomography (CT) machine. The typical mesoscopic cementation modes are shown in Fig. 7.

According to Fig. 7(a), based on the right CT picture, some high-density anhydrite minerals particles with low gray levels distributed in clay minerals can be observed. Meanwhile, micro-crack and micro-pore can be observed in the scanning face of the left SEM picture. The bonded structure of the mineral skeleton and its boundary (contact face) can be identified clearly. Therefore, undisturbed clay-sulfate rock may belong to the weak cementation rock, namely semi-rock. The CT and SEM picture of the 600kPa normal stress group is shown in Fig. 7(b). In the right CT picture, the non-uniformity of gray distribution is strong and some macroscopic fractures appear on the outside boundary of the sample. The mesostructure in the left SEM picture has also changed significantly, and scale-like clay minerals are flocculent accumulated. We also observed the skeleton of the larger mineral particles is not clear. It was attributed that they may be covered by clay minerals. The micro-crack in the sample after a drying-wetting cycle is smaller and denser than the undisturbed sample. Finally, the CT and SEM picture of clay rock that experienced one and five drying-wetting cycles have the same mesostructure. The micro-crack in the SEM picture and macro-crack in the CT picture seems generated and propagated in this process.

Compared with the initial cement structure, the flocculent structure is more common in compacted clay. Therefore, the intrinsic of the sample after the first drying-wetting cycle may be clay (soil) rather than clay rock. Once, the normal stress is up to the sample’s expansive pressure, the expansive deformation was eliminated. However, the degeneration process of rock still occurred.

The influence of the drying-wetting cycle on the expansive properties of samples can be divided into two stages. The first stage only corresponds to the first drying-wetting cycle, which is mainly manifested in the structural change in mesoscopic. In this process, the clay-sulfate rock degenerated from semi-rock to clay with the increase in volume. Due to the changes in mesostructure and unrecoverable expansive deformation, the expansive potential and expansive pressure are decreased significantly. The next stage corresponds to the second to fifth drying-wetting cycles, which are characterized by a gradual decrease in both volume expansive potential and expansive pressure. In this process, the volume of samples is reduced slowly and tends to be constant.

Discussion

In addition to expansibility, shear strength is also a crucial index to evaluate the engineering property of swelling rock or expansive clay (Zeng et al. 2021; Pineda et al. 2014; Xu et al. 2020). It determines the bearing capacity of the surrounding rock directly. To study the effect of the drying-wetting cycle on shear strength, shear tests were carried out by direct shear apparatus (strain type), in which the normal load keeps constant during the former drying-wetting cycle and the later shear test. The change regulation of shear strength, cohesion, and internal friction angle with the cycles are shown in Fig. 8.

As shown in Fig. 8-(a), the shear strength decreased with the increase in the number of drying-wetting cycles, especially in the first cycle. After the first cycle, the shear strength decreased by 76.9%, 73.7%, 75.6%, and 68.6% in the groups with 100kPa, 200kPa, 300kPa, and 400kPa normal pressure respectively. The sharp decreases may also be caused by the degeneration of the specimen (from rock to clay), which could be proved by the SEM photo in Fig. 6. Then the shear strength decreases slowly during the second to fifth cycle. In this stage, the influences of the changes in mesostructure are not the governing factor. According to the shear test results, the evolution of cohesion and internal friction angle during drying-wetting cycles can be calculated and the results are shown in Fig. 8-(b). The cohesion and internal friction angle of the sample decreased by 79.8% and 55.1%after the first drying-wetting cycle. We also observe that in the latter drying-wetting cycles, the cohesion continued to decrease very slowly, while the internal friction angle remained stable at 15°.

Conclusion

A self-designed instrument was created and used to conduct the drying-wetting cycle experiments on confined compression rock. The coupled effect of stress state and cyclic drying-wetting transmission on the expansive properties of soft rock was studied. The following conclusions can be drawn from the experimental results.

(1) Gansu clay-sulfate rock is a special red-bed argillaceous rock that is sedimented in a drying and strong oxidizing environment. Its sulfate content is 23%, and the hydration ratio of sulfate is 70%. Due to the high content of clay minerals and sulfate, the rock samples are water-sensitive. The samples show accelerated slake tendency in freely drying-wetting cycles, and the 2^nd and 5^th cycle slakes durability index is 0.798 and 0.206.

(2) With a confined restrain device and controllable normal force, the self-designed instrument achieves the coupling between the cyclic drying-wetting transmission and a constant stress state in the experiment. When the normal stress is below the expansive stress, the volume increased significantly in the first wetting process, and only a part of the deformation can be recovered in the first drying process. Later, in the 2^nd to 5^th drying-wetting cycles, the expansive-shrinking deformation of rock is little and the volume of specimen tends to be constant.

(3) In the first drying-wetting cycle, the mesostructure of the test sample changes from particle cementation to flocculent accumulation, which indicated the clay-sulfate rock, in a large part, degenerated from semi-rock to clay. During the 2^nd to 5^th drying-wetting cycles, there is no essential structural change. Moreover, the normal pressure may decrease and even eliminate the expansive deformation of clay-sulfate rock, however, which cannot avoid the degeneration process.

(4) Same as the expansion pressure, the shear strength, as well as cohesion, and the internal friction angle of clay-sulfate rock decreased after the drying-wetting cycle. We note a decrease of 79.8% and 55.1% in the cohesion and internal friction angle at the end of the first cycle. During the second to fifth cycles, there was a slight slow decrease in internal friction angle and cohesion.

Redit authorship contribution statement

Shufan Cheng: Conceptualization, Methodology, Investigation, Data curation; Writing- original draft.

Yang Ye: Test, Data processing, Funding acquisition.

Yawu Zeng: Supervision, Funding acquisition, Writing - review & editing.

Rui Gao: Funding acquisition, Formal analysis, Writing - review & editing.

Data Availability Statement

Some or all data, models, or codes generated or used during the study are available from the corresponding author at reasonable request.

Acknowledgments

This work was supported by the National Natural Science Foundation of China, (Grant Nos. 42107175, held by Prof. Yang Ye), (Grant Nos. 51178358, held by Prof. Gui Gao), and Fundamental Research Funds for the Central Universities (Grant Nos. 2042020kf1035). We would like to express our sincere appreciation to the anonymous reviewers for their valuable suggestions.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests that could influence the work reported in this paper.

Alonso E, Pineda J, Cardoso R (2010). Degradation of marls; two case studies from the Iberian Peninsula Geological Society, Engineering Geology Special Publications, London, U. K. http://dx.doi.org/10.1144/EGSP23.5
Butscher C, Huggenberger P, Zechner E, Einstein HH (2011). Relation between hydrogeological setting and swelling potential of clay-sulfate rocks in tunneling. Eng. Geol. 122(3):204-214. https://doi.org/10.1016/j.enggeo.2011.05.009.
Butscher C, Breuer S, and Blum P (2018). Swelling Laws for Clay-Sulfate Rocks Revisited. B. Eng. Geol. Environ. 77(1):399-408. https://doi.org/10.1007/s10064-016-0986-z.
Butscher C, Mutschler T, and Blum P (2016). Swelling of Clay-Sulfate Rocks: A Review of Processes and Controls. Rock Mech. Rock Eng. 49(4), 1533-1549. https://doi.org/10.1007/s00603-015-0827-6.
Cripps JR, Taylor RK (1981). The Engineering properties of mudrocks. Q. J. Eng Geol. 14(4):325-346. https://doi.org/10.1144/GSL.QJEG.1981.014.04.10.
Erguler ZA, Ulusay R (2009). Assessment of physical disintegration characteristics of clay-bearing rocks: disintegration index test and a new durability classification chart. Eng. Geol. 105(1):11-19. https://doi.org/10.1016/j.enggeo.2008.12.013.
Gasc-Barbier M, Tessier D (2007) Structural modifications of a hard deep clayey rock due to hygro-mechanical solicitations. Int. J. Geomech. 7(3), 227-235. https://doi.org/10.1061/(ASCE)1532-3641(2007)7:3(227).
Gautam TP, Shakoor A (2013). Slaking behavior of clay-bearing rocks during a one-year exposure to natural climatic conditions. Eng. Geol. 166(Nov):17-25. https://doi.org/10.1016/j.enggeo.2013.08.003.
Gautam TP, Shakoor A (2016). Comparing the slaking of clay-bearing rocks under laboratory conditions to slaking under natural climatic conditions. Rock Mech. Rock Eng. 49(1):19-31. https://doi.org/10.1007/s00603-015-0729-7.
Hu M, Liu Y, Ren J, Wu R, Zhang Y (2019). laboratory test on crack development in mudstone undaer the action of dry-wet cycles. B. Eng. Geol. Environ, 78(1):543-556. https://doi.org/10.1007/s10064-017-1080-x.
ISRM (1981). ISRM Suggested methods: rock characterization, testing and monitoring Brown, E.T. (Ed.). Pergamon Press, London.
Jia L, Chen M, Jin Y (2014). 3D imaging of fractures in carbonate rocks using X-ray computed tomography technology. Carbonate Evaporites 29, 147–153. https://doi.org/10.1007/s13146-013-0179-9
Kuang H, Wang P, Ai X, et al. (2022). Study of the karst development rate of the Jinping porous carbonate rock formation based on porosity analysis of rock images. Carbonates Evaporites 37: 55. https://doi.org/10.1007/s13146-022-00801-5
Kwon S, Lee C S, Cho S J, Jeon S W, Cho W J (2009). An investigation of the excavation damaged zone at the KAERI underground research tunnel. Tunn. Undergr. Sp. Tech. 24(1):1-13. https://doi.org/10.1016/j.tust.2008.01.004.
Li Q, Shi W, Yang R (2016). Deformation mechanisms in a coal mine roadway in extremely swelling soft rock. SpringerPlus. 5(1):1310. https://doi.org/10.1186/s40064-016-2942-6.
Liu D, Wang Z, Zhang X, Wamg Y, Zhang X, Li D (2018). “Experimental investigation on the mechanical and acoustic emission characteristics of shale softened by water absorption.” J. Nat. Gas. Sci. Eng. 50(Feb):301-308. https://doi.org/10.1016/j.jngse.2017.11.020.
Pimentel E, Anagnostou G (2013). New apparatus and experimental setup for long-term swelling tests on sulphatic claystones. Rock Mech Rock Eng.46, 1271–1285. https://doi.org/10.1007/s00603-013-0396-5
Sadisun IA, Shimada H, Ichinose M, Matsui K (2005). Study on the physical disintegration characteristics of Subang claystone subjected to a modified slaking index test. Geotech. Geol. Eng. 23(3):199-218. https://doi.org/10.1007/s10706-003-6112-6.
Steiner W (1993). Swelling rock in tunnels-rock characterization, effect of horizontal stresses and construction procedures. Int J Rock Mech Min. 30, 361–380. https://doi.org/10.1016/0148-9062(93)91720-4
Zeng L, Luo J, Liu J, Gao Q (2021). Disintegration characteristics and mechanisms of carbonaceous mudstone subjected to load and cyclic drying-wetting. J. Mater. Civil Eng. 33(8):04021195. https://doi.org/10.1061/(ASCE)MT.1943-5533.0003817.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this older preprint version

Read the latest preprint version →

Experimental investigation on the expansive characteristics of Gansu clay-sulfate rocks in China subjected to confined compression and cyclic drying-wetting

Status:

Version 1

Abstract

Figures

Introduction

Materials And Test Methodology

Material Properties

Instrument and method of the drying-wetting cycle test

Test Results

Expansive characteristics in the drying-wetting cycle

Mesoscopic Expansive Mechanism

Discussion And Conclusion

Discussion

Conclusion

Declarations

Redit authorship contribution statement

Data Availability Statement

Acknowledgments

Declaration of Competing Interest

References

Additional Declarations

Status:

Version 1