Research on muck conditioning for EPB shield tunnelling in composite formation

Yalong Jiang; Xin Tang; Bitang Zhu; Kai Zhou; Yucong Pan; Changjie Xu

doi:10.1038/s41598-024-81757-5

. 2024 Dec 28;14:30936. doi: 10.1038/s41598-024-81757-5

Research on muck conditioning for EPB shield tunnelling in composite formation

Yalong Jiang ^1,^2,^✉, Xin Tang ^1,², Bitang Zhu ^1,², Kai Zhou ⁴, Yucong Pan ³, Changjie Xu ^1,²

PMCID: PMC11680905 PMID: 39730664

Abstract

Compared with simple formations, EPB (earth pressure balance) shield tunnelling in composite formations encounters severe problems with muck conditioning and require improved muck conditioning technology to fulfil expectations for continuous and efficient excavation. In the Nanchang Metro Line 4 Project, a water-rich sand-argillaceous siltstone composite formation is encountered. With a high moisture content and complex composite formation ratio, it is quite difficult to determine the optimum muck conditioning scheme, and thus, muck spewing accidents frequently occur during the tunnelling process. In this study, laboratory muck conditioning tests are conducted for five ratios of stratum, from a full section water-rich sand formation to a water-rich sand-argillaceous siltstone composite formation. The muck conditioner ratio for EPB shield tunnelling in such composite formations is dynamically optimized based on an analysis of propulsion speed, penetration rate, thrust force and torque of the shield machine in the field. The following muck conditioning scheme is obtained: (1) For the full-section water-rich sand layer, foam and bentonite slurry should have similar proportions, namely, the foam injection ratio is 10–15% and the bentonite slurry injection ratio is 10%. (2) As the argillaceous siltstone content increases from 10 to 90%, the foam injection rate gradually increases from 20 to 50%, and the bentonite slurry injection rate gradually decreases from 20% to 0. (3) If the argillaceous siltstone content is 10%, then CMC polymer with an injection ratio of 1% is required. Once the argillaceous siltstone content exceeds 70%, it is necessary to add mudstone dispersant solution (1:5) with a volume ratio of 2 to 3%. According to the analysis of the field tunnelling parameters, including penetration rate, total shield thrust and cutterhead torque, the optimization of the muck conditioning scheme proposed based on laboratory tests is proved effective and deserves further application.

Keywords: Earth pressure balance shield, Muck conditioning, Composite stratum, Laboratory experiment, Field application

Subject terms: Environmental sciences, Solid Earth sciences

Introduction

With the rapid development of economy in China, there is great acceleration in the utilization of underground space in urban areas, such as urban metro networks, to solve the problems caused by heavy traffic congestion in densely populated areas and fulfil the ever-increasing demand for high traffic capacity^1,2. The shield tunnelling method, especially the earth pressure balance (EPB) shield method, has some incomparable advantages, such as high excavation efficiency, high construction safety, a friendly working environment, low project costs and low interference with surrounding ground and surface traffic activities^3,4. Therefore, EPB has become the main choice for urban metro tunnel excavation. However, due to geological and hydrogeological complexity, EPB shield tunnelling in different formations encounters different engineering problems directly related to muck conditioning effects, such as clogging of cutter, spewing from screw conveyor and wear of cutter tools^5–8 (as shown in Fig. 1). However, mucks produced in different formations usually require different muck conditioners and mixing schemes^4,9,10 (as shown in Table 1), which makes it more complicated in field applications.

Fig. 1 — Common engineering problems encountered during EPB shield tunnelling⁶. (a) Clogging of cutter, (b) Spewing from screw conveyor, (c) Wear of cutter tools.

Table 1.

Summary of soil conditioner adaptability for different formations.

Stratigraphic type	Stratigraphic characteristics	Prone to engineering problems	Modifier selection
Sand and pebble stratum	Grain size between 0.075–200 mm, lack of fine particles, permeability and very large internal friction angle of particles	Tool wear, shutdown to open the bin for tool change	Bentonite slurry, polymer
Coarse sand, gravel stratum	Grain size between 0.075–60 mm, insufficient fines in the formation, high permeability and high internal friction angle	Spiral conveyor spouting, large fluctuations in soil bin pressure, excessive ground deformation or even destabilization	Foam agent, bentonite slurry
Chalky sand, chalky ground	Particle size between 0.005–0.075 mm, with a certain viscosity	The soil bin is closed, the soil is not smooth, and the tool is “mud cake”	Foam agent, bentonite slurry, dispersant
Clay stratum	Particle size less than 0.005 mm, too viscous	The cutter opening is closed, the soil bin is not smooth, and the cutter tool is “mud cake”	Dispersant, foam agent

Open in a new tab

To ensure high excavation efficiency and safety for EPB shield tunnelling in different formations, many studies were conducted concerning muck conditioning schemes, muck conditioning effectiveness, and the physical properties of conditioned mucks. For example, Psomas¹¹ examined the improvement of sandy strata by foam through penetration, direct shear, and compression tests and found that the size of the particles was an important parameter in determining the soil properties. Peila et al.¹² evaluated the state of improved residual soils in sandy strata using slump tests and investigated the effect of water content and foam injection on the improvement of residual soils. Peila et al.¹³ also conducted tests on improved sandy soils with different gradations and found that the percentage of each particle size in the sandy soil had a great influence on the improvement effects of foam and water. They found that when the sandy soil had a high content of large particles, foam and water were unable to cause the residual soil to improve and reach the ideal state. Budach and Thewes¹⁴ proposed improvement schemes for sandy soils of different grain sizes by using foam mixed with additives and combined data from slump tests, infiltration tests, and direct shear tests. Borio and Peila¹⁵ tested the permeability properties of slurry soils modified with foam and found that foam was effective in regulating the permeability. They also found that the amount of percolation was related to the foam injection ratio. In addition to the properties of foam and muck, there are other factors in sandy soil that affects the muck conditioning effectiveness and the safety of shield tunnelling. For example, Bezuijen and Van Lottum¹⁶ studied the infiltration rate of foam in the face of the earth pressure balance shield machine, and found that in the sandy soil with low permeability, the infiltration rate of foam mainly depended on the groundwater flow rate, and had nothing to do with the properties of the foam. Bezuijen and Gerheim Souza Dias¹⁷ found that when the earth pressure balance shield machine was stationary, the water in the sand in the soil bin would flow to the face, causing the foam to dry and rupture, resulting in a decrease in the pressure in the soil chamber, which in turn affecting the stability of the face. In the study of cohesive strata, Vinai et al.¹⁸ applied water and foam to act on powdered clay and sandy soils, respectively; in simulations of a shield screw-out test, the amount of improver had an effect on the internal pressure of the screw as well as the screw torque. Ye et al.¹⁹ conducted slag improvement tests for a muddy siltstone formation with high clay mineral content, proposed an improvement scheme for this formation through slump tests, and derived an equation for the relationship between the slump of muddy siltstone and the foam ratio and water content. Zumsteg et al.²⁰ investigated the effect of different modifiers on the shear strength of clay soils and showed that the addition of dispersants or foams significantly reduced the shear strength of soil samples. Peila et al.²¹ studied clay adhesion and tested the effects of different modifiers on the clay. The results showed that the addition of foam and dispersant effectively reduced the adhesion of clay to metal interfaces. There were some general studies on the conditioning scheme, effectiveness and physical properties of mucks^22–26; however, most were based on tests and analyses for simple formations. Thus, there is still a lack of full studies on composite formations containing soft/hard interlayers with high water content, strong permeability and different physical properties and thus, a lack of deep studies on muck conditioning schemes and effectiveness analyses for dynamic composite formations.

In work on the tunnel section from Qili Station to Minyuanlu West Station in the Nanchang Metro Line 4 Project, complex formations are encountered. They include not only typical full section water-rich sand formations and argillaceous siltstone formations but also water-rich sand-argillaceous siltstone composite formations. For the latter formations, soft and hard interlayers with distinct area ratios and strength ratios occur simultaneously at the same tunnel face, and the chamber pressure fluctuates greatly with the changing tunnel face. In this case, it is quite difficult to determine the optimum muck conditioning scheme to solve muck flow problems and obtain stable shield operating parameters, which requires deep studies. This paper choses the aforementioned tunnel section as a case study. First, muck conditioner types are selected based on particle gradation analysis, and the commonly used bentonite and foam conditioners are subjected to performance testing to determine accurate mass ratios and concentrations. Then, muck conditioning tests are conducted on cases from full section water-rich sand formations to water-rich sand-argillaceous siltstone composite formations using these muck conditioners, and the optimum muck conditioner mixing scheme is determined by the muck properties obtained from slump tests, permeability tests and shear tests. Finally, the effectiveness of the muck conditioning scheme is verified by real muck conditioner usage and tunnelling performance recorded in the field. The conclusions of this study offer guidelines for EPB tunnelling in similar formation conditions.

Background

Engineering geological conditions

The tunnel section from Qili Station to Minyuanlu West Station in Nanchang Metro Line 4 Project is approximately 2007 m long, with a minimum curve radius of 350 m. EPB shield technique is adopted in tunnel digging with an outer diameter of 6.28 m. The burial depth of the tunnel section ranges from 10 to 29 m, and the underground water level ranges from 4 to 6 m from the surface. Starting from Qili Station, the EPB shield machine crosses complex and diverse formations: for ring no. 0 ~ 289 and 1124 ~ 1675, a full section of water-rich sand layer consisting of coarse sand and gravel sand, in total 1011 m; for ring no. 393 ~ 968, a full section of argillaceous siltstone, in total 692 m; for ring no. 289 ~ 393 and 968 ~ 1124, a changeable composite formation consisting of an overlying water-rich sand layer and an underlying argillaceous siltstone layer, in total 315 m. The longitudinal geological profile of the tunnel section is shown in (Fig. 2).

Fig. 2 — Tunnel section from Qili station to Minyuanlu West station in the Nanchang Metro line 4 project.

Muck particle gradation analysis

Since mucks are taken from different formations, the particle gradation characteristics have a great influence on the muck conditioner selection, conditioning scheme optimization and final conditioning effectiveness. Thus, muck particle gradation analysis should be undertaken first. The samples of sand and argillaceous siltstone are taken from the foundation pit of Minyuan Road West Station, the depth of which is within the same range as the tunnel face of the shield tunnel. The physical properties of sandy soil (mainly composed of coarse sand and gravelly sand) and argillaceous siltstone are shown in Table 2.

Table 2.

Physical and mechanical parameters of sandy soil and argillaceous siltstone.

		Density ρ(g/cm³	Specific weight of soil/G_s	Cohesive force c (kPa)	Internal friction angle φ(°)	Permeability coefficient k(m/s)
Sandy soil	Coarse sand	1.98	2.64	1	34	9.259 × 10⁻⁴
Sandy soil	Gravelly sand	2.00	2.63	1	36	1.157 × 10⁻³
Argillaceous siltstone		2.39		350	31	1.157 × 10⁻⁶

Open in a new tab

Figure 3 shows the particle gradation curves for the soils. For the full section water-rich sand layers consisting of coarse and gravel sands, the fine particle group (< 0.075 mm) is missing (i.e., the content is almost zero), and the average particle size d₅₀ of the sand layer is 0.5 mm. Tests show that the water content is between 21 and 35%, and the permeability coefficient is between 10⁻³ m/s and 10⁻⁴ m/s. According to the testing results, the sand sample shows excessive fluidity, insufficient cohesion and excessive permeability, which may easily cause soil spewing²⁷. Moreover, due to the high water content and permeability of this sandy muck, the use of foam as the only soil conditioner cannot ensure that the muck has good flow plasticity and anti-permeability. Therefore, the simultaneous use of bentonite to increase the content of the fine particle group and the anti-permeability of the muck is necessary²⁸.

As shown in Fig. 3, the content of the fine particle group in argillaceous soil samples is approximately 30%. It shows excessive cohesion and insufficient fluidity, which may easily cause mud cakes to form ahead of the cutterhead, further inducing cutter choke and flat wear, screw conveyor blockage, and then frequent chamber inspection and cutter replacement^19,21. Therefore, when the argillaceous muck is conditioned, more foam should be added to increase its fluidity, and less bentonite should be added to prevent cohesion that is too high. If the argillaceous muck still has excessive cohesion, then more dispersants should be added²⁹.

In particular, when tunnelling in a water-rich sand-argillaceous siltstone composite formation, it is extremely difficult to control the support pressure due to the large difference in the soil layer stiffnesses on the composite tunnel face. Moreover, it is quite likely to induce excessive ground settlement or uplifting. Meanwhile, the chamber pressure fluctuates greatly due to the dynamic change of the composite tunnel face and the large difference between the soil stiffnesses in the upper and lower tunnel sections. Therefore, it is quite difficult to adjust the shield attitude and maintain the tunnelling axis from offset³⁰.

Muck conditioner selection and conditioning effectiveness evaluation

Muck conditioner selection

From an economic viewpoint, foam and bentonite are the most widely used muck conditioners due to their low price and simple preparation^31,32. Therefore, in this study, performance tests are conducted on different foam and bentonite conditioners, and the optimum mass ratio and concentration of these muck conditioners are determined based on the test results.

Performance testing and selection of bentonite

When using bentonite slurry for soil conditioning, slurry concentration is an important consideration. When the slurry concentration is too high, its pumping performance is poor, and it is not easily transported to the soil chamber and the front of the cutterhead. When the slurry concentration is too low, it does not easily and effectively fill gaps within mucks³³. Therefore, before using bentonite mud, it is necessary to analyse the slurry proportions so as to obtain an appropriate slurry concentration.

According to the bentonite slurry proportions commonly used in the actual project construction, five groups of bentonite slurry proportions with bentonite to water mass ratios of 1:4, 1:6, 1:8, 1:10, and 1:12 are tested in this study with a type of sodium bentonite taken from the shield construction site. A rotary viscometer is used in the laboratory experiments to test the five groups of bentonite slurry, and the test results are shown in Fig. 4. It is indicated that the viscosity of the bentonite slurry decreases with the decrease of bentonite mass fraction. According to the slurry state and viscosity, when the bentonite slurry viscosity is kept between 20–30 mPa∙s, its fluidity and adhesion are better than at other viscosities, i.e., suitable for soil conditioning and pumping in the pipeline of the shield machine. When the bentonite to water mass ratio equals to 1:8, the slurry viscosity is 22 mPa∙s, which meets the viscosity requirement for shield construction³⁴. Therefore, a bentonite slurry ratio of 1:8 is selected for the following muck conditioning tests.

Fig. 4 — Performance testing of the bentonite slurry with different mass ratios.

Performance testing and selection of the foam

In addition to the foam generating system of the shield machine itself, laboratory studies on the performances of the shield foams and the conditioned mucks mainly employ foam generating equipment that is made in this laboratory. The foam generating equipment used in laboratory tests is made according to Thewes et al.³⁵, as shown in Fig. 5. The foaming agent (surface active agent) used in the tests is obtained from the construction site of Nanchang Metro Line 4.

Fig. 5 — Generation of foam in the laboratory. (a) Homemade foam preparation system, (b) Homemade foam.

The main technical indicators of the foam include volume flows of air and liquid of the foam, foam concentration, foam expansion ratio (FER), and half-life time (HLT). In the test, due to some technical problem, the volume flows of air and liquid of the foam generating system were not measure. For foam concentration, the influence on FER and HLT is not completely clear and understood. Some scholars pointed out that concentration may not have a significant effect on half-life time³⁶, some scholars^30,37 who believe that the foam concentration can significantly affect the expansion ratio and half-life time of the foam. In this paper, the appropriate foam concentration is identified using these two parameters. In the field, the appropriate foam concentration is usually determined according to manufacturer recommendations or field testing, which is generally around 0.5–5% by volume, while in the laboratory, it is usually determined according to these two parameters through performance testing.

Foam expansion ratio

The foam expansion ratio refers to the foam generated per unit volume of the foam solution, which is the most important parameter affecting the working performance of the foam. Foam stability is the ability of the foam to resist breakage under the influence of environmental factors, namely, the period of the foam survives after formation. Generally, the larger the foam expansion ratio is, the better the foam effect, but a foam expansion ratio that is too high can reduce the foam stability^38–40. To achieve a good foaming state and conditioning effect during shield tunnelling, a foam expansion ratio of 5–20 is usually required³⁷. The calculation formula is as follows:

where FER is the foam expansion ratio, Inline graphic is the foam volume after foaming, and is the volume of the foam solution.

The FER can be measured directly using a graduated cylinder (as shown in Fig. 6). In the present study, foam stability is measured using a 2152 type Ross-Miles foam metre (as shown in Fig. 7) according to standards ISO 696–1975.

Fig. 7 — Foam volume for foam solutions of different foam concentrations.

Figure 6 shows the FER for different foam concentrations (1–5%). While the foam concentration increases from 1 to 3%, the FER increases slowly, reaching a nearly stable value when the foam concentration is 3%. When the foam concentration is greater than 4%, the foam expansion ratio decreases slightly. Figure 7 shows the variation in the foam volume with time obtained using a Ross-Miles foam metre at five different foam concentrations. The foam volume produced at a foam concentration of 3–5% is significantly larger than that produced at a foam concentration of 1–2%, while the foam volume produced at a foam concentration of 3–5% is not notably different. It has been known from previous literatures that the foaming performance of the foaming agent solution is mainly determined by the foamability of the surfactant, and each surfactant has a critical micelle concentration^41,42. When the foaming agent solution is in a certain concentration range but lower than the critical micelle concentration, the surfactant molecules are adsorbed on the liquid film acting as single molecules. The adsorption amount on the surface increases with the increase of the concentration, and the hydrophobic groups attract and associate with each other to form micelles. The surface tension of the liquid as well as the surface energy of the system continue to decrease, and the foam continues to generate until the critical micelle concentration is reached. At this critical state, the surface active molecules cover the surface of the solution, meanwhile the surface tension and system energy no longer decrease, and the foaming performance no longer increases. At this time, foam formation and rupture are in equilibrium^43,44.

Half-life time

Half-life time (HLT) refers to the time required for the ratio of the dissipation to total foam masses of 50%, which is an important indicator of foam stability. Usually, the longer the half-life time is, the better the foam stability is, and the better the foam conditioning effect is^45,46. According to the actual shield tunnelling conditions, the general requirement of foam stability is that the half-life time of the foam should exceed 5 min³⁷.

In this study, a test apparatus mainly consisted of Ross-Miles foam metre containing a dropping pipette and a measuring cup is adopted to test the half-life time of the foam⁴⁷. The dropping pipette of the Ross-Miles foam metre is used as the decaying cylinder, and the liquid generated after foam rupture flows into a measuring cup. The foam mass (m₁) in the dropping pipette is measured at the start, and once the liquid mass of the ruptured foam in the measuring cup researches 1/2 m₁, the time t (unit: min) is recorded as the half-life time of the foam (as shown in Fig. 8).

Fig. 8 — The foam half-life time for different foam concentrations.

Figure 8 shows the half-life time of five different foam concentrations. As shown, the half-life time of the foam initially increases with increasing foam concentration. When the foam concentration reaches 3%, the half-life time of the foam decreases slightly with increasing foam concentration. From the test result it can be inferred that the solution concentration of 3% corresponds to an optimal half-life time of the foam. At this case, the viscosity of the solution is moderate, and the surface active molecules on the liquid film of the foam reach its peak amount, contributing to a relatively stable state^30,37,39,40.

With a combined consideration of the test results of FER and HLT, the optimal concentration of the foaming agent solution at a standard laboratory environment (temperature 20–25° C, humidity 40–60%) can be determined as 3%, which is also consistent with the previous research results of Galli et al.⁴⁵, Huang et al.²⁸, and Zhen et al.⁴⁸. Meanwhile, the foam FER with this concentration is 14, and the HLT is about 15 min. Foams with the aforementioned indicators will be used in subsequent indoor muck conditioning tests as well as field application.

Evaluation of muck conditioning effects

To determine whether the conditioned mucks meet the requirements of shield tunnelling, the muck conditioning effect is evaluated according to three aspects: flow plasticity, shear strength, and permeability of the conditioned muck. For flow plasticity, many scholars at home and abroad have introduced the slump test, which is originally designed to test the performance of fresh concrete, into the field of shield muck conditioning. In this case, the slump value is used to evaluate the flow plasticity of the conditioned muck, and the recommended slump value is 150–210 mm^5,18,42. It is worth noting that since grain size particles till 20–30 mm are used, millimeters are reported in the paper only for “completeness” of data (original data) and the rounded values are used for comparison. For shear strength, since the conditioned muck usually demonstrates fluidity, it cannot be measured through direct shear tests. Thus, the digital portable cross-plate shear metre is adopted to determine the undrained shear strength of the conditioned muck. According to the studies of Martinelli et al.⁴⁹; Messerklinger et al.⁵⁰; Milligan⁵¹, the undrained shear strength of the conditioned muck for shield tunnelling should be lower than 25 kPa. For permeability testing, an equipment is designed and used to measure muck permeability under pressed conditions, as shown in Fig. 9. This equipment can apply different water pressures to simulate groundwater pressure and apply different confining pressures to simulate the chamber pressure at different burial depths, which can accurately reproduce the mucking environment in a soil chamber. The specimen used for permeability tests has a height of 40 mm and a diameter of 70 mm. The specimen is placed into a pressure chamber, while permeable stones and filter papers are placed on the upper and lower surfaces of the specimen. The permeation pressure and the confining pressure are adjusted by a pressure regulator with a range of 0–0.8 MPa. According to the previous literatures, the general goal of muck conditioning is to control the permeability coefficient (k) of the conditioned muck to 10^–5 m/s or less^52–54.

Fig. 9 — Experimental equipment for pressed permeability tests.

Laboratory muck conditioning tests

Muck conditioning for the full-section water-rich sand layer

For the full-section sand layer, either foam or bentonite slurry can be used for muck conditioning during shield tunnelling, which can usually make the muck achieve good flow plasticity and permeability resistance. However, due to the high water content and permeability of the sand layer encountered in this shield tunnel case, in this study, single foam and bentonite slurry as well as foam-bentonite slurry composite conditioners are first tested for muck conditioning, based on which the test results are compared and analysed to determine the optimum muck conditioning scheme suitable for the full-section water-rich sand layer. Before the improvement, a special oven for geotechnical experiments is used to dry the sand sample, and then water is added to ensure that the water content is consistent with the in-situ formation water content (i.e., 20%). The average permeability of the unconditioned sandy soil is 1.14 × 10^–4 m/s according to the results of six permeability testing groups under a constant water head before muck conditioning. For the conditioned sandy soil, the permeability tests are conducted under a confining pressure of 200 kPa and permeation pressure of 150 kPa, which is consistent with the field condition.

Muck conditioning tests using single foam for the sandy soil layer

When using foam for muck conditioning during EPB shield tunnelling, the technical indicator is the usage amount of the foam, which is usually characterized by the foam injection ratio (FIR) calculated using the following formula:

where Inline graphic is the volume of injected foam and is the volume of the soil sample cut down by the cutterhead.

Water content of the test is always maintained at 20%, after muck conditioning with single foam, the permeability and slump of muck were measured by constant water head test and slump test, respectively. The variation in the permeability and slump value of the sandy soil with different foam injection ratios is shown in Fig. 10. The foam injection ratio ranges from 5 to 20% with an interval of 5%. It can be observed that the slump value increases as the foam injection ratio increases from 5 to 15%. When the foam injection ratio reaches 15%, the slump value changes slightly. When the foam injection ratio is in the 15–20% range, the slump value is generally 225 mm. At this case, the muck fluidity is relatively high and almost close to a flow state, which is harmful to the muck transportation of the screw conveyor. From the figure it is indicated that the foam plays a significant role in improving the fluidity of the sandy soil, and when the slump value is in a range of 150–210 mm, the muck reaches the optimum conditioning effect^18,24,42. Thus, it can be regarded that the slump value at FIR = 10% reaches the shield tunnelling requirements. Figure 10 also indicates that the sandy soil permeability first decreases sharply and then decreases slowly with increasing foam injection ratio. The permeability varies from 10⁻⁴ to 10⁻⁵ m/s but does not reach the permeability requirement (k < 10⁻⁵ m/s) to prevent muck spewing^53,55. Therefore, for the water-rich sandy soil layer, muck conditioning using single foam can achieve the muck fluidity requirement but cannot meet the muck permeability requirements.

Fig. 10 — Variation in the muck permeability and slump value with different foam injection ratios.

Muck conditioning tests using single bentonite slurry for the sandy soil layer

The usage amount of the bentonite slurry during EPB shield tunnelling is usually characterized by the bentonite slurry injection ratio (BIR), which is calculated as follows:

where Inline graphic is the injected mass of the bentonite slurry and is the mass of the soil cut down by the shield cutterhead.

The variation in the permeability and slump value of the sandy soil with different bentonite slurry (1:8) injection ratios after muck conditioning is captured and compared to that with different foam injection ratios in Fig. 11. The BIR ranges from 5 to 30% with an interval of 5%. It can be observed in Fig. 11 that the slump value shows an increase trend with increasing BIR. During this process, the slump value first increases slowly in the BIR range of 5–15%, and then increases sharply in the BIR range of 15–25%. After that, the slump value remains nearly unchanged in the BIR range of 25–30%. At this point, the slump value is approximately 225 mm, indicating that the muck fluidity is too large, and even close to a flow state¹². After that, the influence of the increase in BIR on the slump value is not significant. Meanwhile, From the analysis results it is clearly observed that the permeability of the sandy soil decreases sharply in the BIR range of 5–15% and decreases slowly in the BIR range of 15–30%. When the BIR reaches 10%, the permeability of the sandy soil is 2.9 × 10⁻⁶ m/s, and it is controlled below 10⁻⁵ m/s in the BIR range of 10%-30%, which meets the requirements for muck permeability (prevention of spewing) during EPB shield tunnelling¹⁴. For muck conditioning using single bentonite slurry, the slump value is between 150–210 mm when the BIR range is 20–25%. However, for muck conditioning using single foam, the muck fluidity meets the requirement when the FIR reaches 10%.

Fig. 11 — Variation of the muck permeability and slump value with bentonite slurry injection ratio.

In summary, the use of single bentonite slurry to condition sandy soils can meet the requirements of muck permeability and fluidity. However, it requires a high injection ratio, which would be an excessive and uneconomical workload for the formation of slurries in the field. Despite that bentonite slurry can obviously reduce the permeability of sandy soil to meet the requirement, its conditioning effect on the fluidity of sandy soil is not as significant as that of foam. That is, foam plays a dominant role in improving the fluidity of the sandy soil, while bentonite slurry plays a dominant role in improving the permeability of the sandy soil²⁸. Thus, the above two muck conditioners can be used simultaneously to improve the properties of sandy soil to achieve the ideal state. On this basis, the composite injection ratio of the foam to the bentonite slurry is determined in the following section.

Muck conditioning tests using foam and bentonite slurry composite for the sandy soil layer

From the analysis of the muck conditioning test results using single foam it is known that when the FIR is in the range of 10–15%, the corresponding slump value of the conditioned muck is 170–225 mm, which almost meets the requirement for muck transportation during shield tunnelling. From the analysis of the muck conditioning test results for single bentonite slurry it is shown that when the BIR reaches 10%, the sandy soil after conditioning meets the permeability requirement. Therefore, with a combined consideration of the above two muck conditioners’ characteristics as well as the muck conditioning test results, two sets of composited muck conditioning tests with FIR = 10%, BIR = 10%, and FIR = 15%, BIR = 10% are conducted for sandy soils. The slump value and permeability of the sandy soils under these two conditions are shown in Table 3 and Fig. 12.

Table 3.

Results of the muck conditioning test using foam and bentonite slurry composite for sandy soil.

Modifier proportioning parameters	Slump(mm)	Permeability coefficient (10⁻⁶ m/s)
FIR = 10%, BIR = 10%	185	2.40
FIR = 15%, BIR = 10%	210	2.08

Open in a new tab

Water content: 20–25%, Cf = 3%, FER = 14, bentonite-to-water mass ratio of 1:8

Fig. 12 — Slump tests of the muck conditioning test using foam and bentonite slurry composite for sandy soil. (a) FIR=10%, BIR=10%, (b) FIR=15%, BIR=10%.

From the analysis results given in Table 3, it is known that when the FIR is 10–15% and the BIR is 10%, the slump value is within the optimum muck conditioning range of 150–210 mm, and the permeability (k) is controlled below 10⁻⁵ m/s. As shown in Fig. 12, the sandy soil owns good fluidity, suitable plasticity, and low permeability after muck conditioning. Therefore, a foam and bentonite slurry composite with foam concentration of 3% and FIR of 10–15%, and bentonite to water mass ratio of 1:8 and BIR = 10% can meet the flow plasticity and permeability requirements both.

Muck conditioning tests using foam and bentonite slurry composite for the water-rich sand-argillaceous siltstone composite formation

To study the muck conditioning schemes and muck condition effectiveness for the dynamically changing tunnel face in the composite stratum, five different situations of the layers at the tunnel face are investigated. Then, suitable muck conditioners are selected according to their particle gradating characteristics in the muck. The optimum muck conditioning schemes for different tunnel faces are determined by analysing the changes in flow plasticity, permeability, and shear strength of the muck after conditioning.

Tunnel face selection and muck conditioning scheme determination

The soil composition of the tunnel face changes dynamically when the shield machine is digging in the composite stratum. Five tunnel faces with volume ratios of 10, 30, 50, 70, and 90% of the argillaceous siltstone section are selected, and the particle gradation analysis of these different tunnel faces is conducted, as shown in Fig. 13.

Fig. 13 — Particle gradation analysis of different tunnel faces.

The particle gradation curve analysis results show that the ratios of argillaceous siltstone to sandy soil ratios equals 9:1, 7:3, 1:1, 3:7, and 1:9 correspond to fine particle contents of approximately 3, 9, 15, 21, and 27%, respectively. The EPB shield is initially used for strata with a fine particle content (d < 0.075 mm) in the range of 20–30%⁵⁶. Thus, it is possible to calculate the usage amount of bentonite slurry based on this fine particle content. By comparing Fig. 13 with Thewes’ diagram, foam and polymers was selected as the muck conditioners.

Muck conditioning tests for the water-rich sand-argillaceous siltstone (volume ratio 9:1) composite formation

A composite formation with 10% argillaceous siltstone volume content is first prepared in the laboratory, and the water content is kept close to the in-situ formation water content (i.e., 20–25%). Because of the low content of argillaceous siltstone, the particle gradation curve shows that the content of fine particles of the muck is approximately 3%, similar to that of the full-section water-rich sand layer. Thus, the muck conditioning scheme should use bentonite slurry to add fine particles and foam to increase the fluidity of the muck. The properties of the conditioned muck are measured by slump tests, penetration tests and cross-plate shear tests.

During testing, when only water is added without any foam or bentonite slurry, the flow plasticity of the muck is not up to the expectation and the muck conditioning effect is poor, as shown in (Fig. 14a). When the foam (with FIR of 10%) is added, the muck after conditioning has a slump value of up to 246 mm and a permeability of 5.68 × 10⁻⁴ m/s. At this point, the conditioned muck shows high fluidity and is not good for transportation, and the permeability doesn’t meet the permeability requirement. Thus, another soil conditioner to increase the fine particle content is required in the conditioning scheme.

Fig. 14 — Slump value of the muck after conditioning for 10% argillaceous siltstone. (a) Unconditioning soil, (b) Ideal conditioning.

Different muck samples are made by adding carboxymethyl cellulose (CMC) polymer solution (the CMC polymer solution concentration is 1% by volume) with polymer solution injection ratio (PIR: ratio of polymer solution to muck sample in volume) of 0.5, 1, and 1.5% to soil and water. When PIR is 1%, the free water in the muck flocculates to form colloid, which inhabits the effect of high water content on the muck fluidity⁴⁸. With the increase in PIR, the free water in the muck flocculates excessively, and the muck fluidity becomes too low. Therefore, the polymer injection ratio is set as 1%, and foam is added with FIRs of 5, 10, 15, 20, 25, and 30%. It is worth noting that the order of adding the ingredients for conditioning is constant. Firstly, soil and water are mixed, and polymers are added to prepare a muck sample with a water content of 20–25%, and finally foam with different FIRs is added. When the FIR is 5–15%, the slump value of the muck after conditioning is less than 150 mm, and its flow plasticity is poor. When the FIR is 20%, the slump value of the muck after conditioning reaches 178 mm, showing a good flow plasticity. However, the permeability coefficient at this point is in the order of 10⁻⁴ m/s, which doesn’t meet the requirement. Thus, based on the above determined PIR and FIR values, bentonite slurry with BIRs of 5, 10, 15, 20, 25, and 30% is added. When the BIR is 5–15%, the permeability of the muck after conditioning is not reduced to the order of magnitude of 10⁻⁵ m/s. When the BIR is 20–25%, the slump value of the muck after conditioning reaches 172 mm, the details are shown in (Fig. 14b and Table 4). At this point, the permeability of the muck after conditioning is 8.72 × 10⁻⁶ m/s, and the cross-plate shear strength is 6.6 kPa. Thus, the muck after conditioning demonstrates good flow plasticity and suitable permeability, and meets the requirement for low shear strength. On this basis, for the water-rich sand-argillaceous siltstone (volume ratio 9:1) composite formation, the optimal muck conditioning scheme is 1% PIR (CMC solution), 20% FIR, and 20–25% BIR.

Table 4.

Slump and permeability coefficient of samples in different states.

Test procedure	Slump	Permeability coefficient	Parameter meet the requirement
Mix soil and water	246	5.68 × 10⁻⁴ m/s	Neither
Add CMC solution (C_CMC = 1%, PIR = 1%)	Too low	/	Neither
Add foam (FIR = 20%)	178	Approximate 10⁻⁴ m/s	Slump
Add bentonite slurry (BIR = 20–25%)	172	8.72 × 10⁻⁶ m/s	Both

Open in a new tab

Muck conditioning tests for the water-rich sand-argillaceous siltstone (volume ratios 7:3, 5:5, 3:7) composite formation

Composite formations with different argillaceous siltstone volume contents of 30, 50, and 70% are prepared in the laboratory, and the water content is close to the in-situ formation water content (i.e., 20–25%). As the content of the argillaceous siltstone gradually increases, the fine particle content of the muck increases. Hence, foam is mainly used in the muck conditioning scheme to increase the muck fluidity.

Through particle gradation analysis, the fine particle contents of the muck are 9, 15, and 21% when the argillaceous siltstone contents are 30, 50, and 70%, respectively. To ensure that the fine particle content of the muck after conditioning ranges between 20 and 30%⁵⁶, bentonite slurry of 11–21, 5–15, and 0–9% are added. At the same time, the foam is added with a gradient of 5%, and the FIR values are thus set as 5, 10, 15, 20, 25, and 30%. By analysing the slump value, permeability, and shear strength, the FIR and BIR values suitable for different argillaceous siltstone contents are determined.

When the argillaceous siltstone content is 30%, the BIR is 13–18%, and the FIR is approximately 25%. For the muck after conditioning, the slump value is 198 mm, the cross-plate shear strength is 5.5 kPa, and the permeability is 5.63 × 10⁻⁶ m/s.

When the argillaceous siltstone content is 50%, the BIR is 8–13%, and the FIR is approximately 30%. For the muck after conditioning, the slump value is 202 mm, the cross-plate shear strength is 5.1 kPa, and the permeability is 4.21 × 10^–6 m/s.

When the argillaceous siltstone content is 70%, the mudstone dispersant solution (1:5 sodium hexametaphosphate) injection ratio is 2%, and the FIR is approximately 40%. For the muck after conditioning, the slump value is 188 mm, the cross-plate shear strength is 3.1 kPa, and the permeability is 3.46 × 10⁻⁶ m/s. As shown in (Fig. 15a–c), all the mucks after conditioning shows good flow plasticity.

Fig. 15 — Slump values of the muck after conditioning for the 30, 50 and 70% argillaceous siltstones. (a) 30%, (b) 50%, (c) 70%.

Muck conditioning tests for the water-rich sand-argillaceous siltstone (volume ratio 1:9) composite formation

Due to the low content of the sand layer, the muck conditioning scheme under these conditions is similar to that for the full-section argillaceous siltstone¹⁹. That is, foam and water are mainly used to improve the muck fluidity, and a mudstone dispersant solution (sodium hexametaphosphate) is added if the muck conglomerates form a “mud cake”⁵⁷. With the FIR gradient set to 5%, the slump value of the muck after conditioning using different FIR values is measured.

When the FIR is between 5 and 40%, the slump value of the muck after conditioning is low, and it only reaches 113 mm when the FIR is 40%, as shown in (Fig. 16a). At this time, the flow plasticity of the muck after conditioning is poor, which can easily induce “mud cakes” on the cutterhead. When the FIR is between 50 and 60%, the fluid plasticity of the muck after conditioning changes insignificantly, and it is still bonded in a mass that adhered to the mixer blade, as shown in (Fig. 16b). Clearly, when the fine particle content in the muck is high and with strong viscidity, solely adding foam can not meet the fluid plasticity requirement. Therefore, for the muck which has been conditioned with the foam of 50–60% FIR, a 1:5 mudstone dispersant solution is chosen in this study. When its injection ratio is 3%, the muck after conditioning shows good flow plasticity, and its slump value reaches 198 mm. Additionally, the muck after conditioning demonstrates a cross-plate shear strength of 3.8 kPa and a permeability of 2.08 × 10⁻⁶ m/s, as shown in (Fig. 16c and Table 5). According to the above test results, the muck after conditioning has good flow plasticity, low shear strength, and low permeability.

Table 5.

Test result of the muck after conditioning for 90% argillaceous siltstone.

Slump value (mm)	permeability coefficient (m/s)	shear strength (kPa)
198 mm	2.08 × 10⁻⁶ m/s	3.8 kPa

Open in a new tab

Evaluation of the muck conditioning effectiveness

Flow plasticity

As shown in Fig. 17, to obtain good flow plasticity, the reasonable range of the slump value for the muck after conditioning is 150–210 mm. For water-rich sand-argillaceous siltstone composite formations with different section ratios, the optimal FIR is between 20 and 50%, and it increases gradually as the proportion of argillaceous siltstone increases. The reason is that the argillaceous siltstone itself owns high clay fraction, and it is easy to agglomerate the sand into agglomerates in the mixed muck, leading to poor muck fluidity²⁰. Therefore, when the proportion of argillaceous siltstone in the muck is large, the viscosity of the soil is relatively large. Although the bentonite slurry is no longer added and the amount of foam is increased, the fluidity of the muck still cannot meet the requirements. Therefore, a dispersant is usually required to weaken the connection between the clay particles and release the bound water, reducing the adhesion of muck and achieving good flow plasticity^57,58.

Shear strength

Before adding the modifier, the shear strength of the muck soil (after the sandy soil, argillaceous siltstone and water are fully mixed) is 10–15 kPa. After the injection of the modifier, it can be indicated from Table 6 that the shear strength of the mixed muck with different proportions is in the range of 3–7 kPa. Comparing the muck before and after improvement, it can be inferred that the addition of the modifiers significantly reduces the shear strength of the muck. This is mainly due to that after the foam is injected into the muck, certain tiny foams enter the pores of the muck, which change the contact state among the soil particles, weaken the bond strength and reduces the cohesion, thus reducing the shear strength of muck^20,32,59.

Table 6.

Evaluation of the muck conditioning effectiveness.

Argillaceous siltstone: Sandy soil	FIR (%)	BIR (%)	PIR (%)	1:5 mudstone dispersant solution injection ratio (%)	Slump value (mm)	Cross-plate shear strength (kPa)	Permeability coefficient (m/s)
1:9	20	20–25	1	0	172 (BIR = 20%)	6.6	8.72 10⁻⁶
3:7	25	13–18	0	0	198 (BIR = 15%)	5.5	5.63 10⁻⁶
1:1	30	8–13	0	0	202 (BIR = 10%)	5.1	4.21 10⁻⁶
7:3	40	0	0	2%	188	3.1	3.46 10⁻⁶
9:1	50	0	0	3%	198	3.8	2.08 10⁻⁶

Open in a new tab

Water content: 20–25%, Cf = 3%, FER = 14, bentonite-to-water mass ratio of 1:8, “1:5 mudstone dispersant solution injection ratio” is the ratio of 1:5 mudstone dispersant solution to muck sample in volume.

However, as the content of argillaceous siltstone increases, the adhesion in the mixed muck increases, and the use of foam alone cannot reduce the shear strength of the muck. Hence, it is necessary to add a dispersant that can reduce the adhesion between particles⁵⁹. The use of dispersant in clayey soil can not only effectively improve the flow plasticity of the muck, but also reduce the shear strength better than that of foam-bentonite slurry composite conditioner⁶⁰. For sandy soil layers, reducing the shear strength of the muck can significantly reduce the friction between the soil and the cutter head and cutter tools, hence to prolong their lifespans⁶¹. For clay soil layers, reducing adhesion and shear strength can prevent mud cake formation on the cutter disc and blockage in soil chamber as well as screw conveyors⁶².

Permeability

The permeabilities of the mucks after conditioning for the five composite formations are all below 10⁻⁵ m/s, meeting the requirements for no occurrence of spewing. By comparing different composite formations, with the increase of either the argillaceous siltstone composition ratio or the BIR, the permeability of the muck after conditioning all gradually decreases (as shown in Fig. 18). This is mainly due to that after bentonite slurry being mixed into the muck soil, viscous fine particles are added, increasing the viscous resistance of water during the seepage process in the muck. Besides, due to their water absorption and expansion characteristics, after the contact with water, the seepage channel is narrowed. At the same time, bentonite particles will flocculate and cement in contact with water, forming a low-permeability mud film^62,63. Therefore, as the injection ratio of bentonite slurry increases, the permeability coefficient of the muck decreases. The increase in the content of argillaceous siltstone can also reduce the permeability coefficient of the muck, mainly because that fine clay particles from argillaceous siltstone fragmentation improves the particle size distribution of the muck. The fine particles fill in the pores of the muck, blocking the passage of water^52,64, thus improving the permeability of the muck, which is similar to the improvement mechanism of bentonite slurry. However, it should be noted that limitations still exist in the permeability test since some factors, such as groundwater seepage, real pressure distribution in soil chamber, and the dynamic digging process of the shield machine, are not considered yet for their complexity^17,65,66.

Fig. 18 — Permeability of the muck after conditioning with different BIR values and fine particle contents.

Recommended muck conditioning schemes

Suggested muck conditioning parameters for the full-section water-rich sand layer

According to the above experimental study, for the full section water-rich sand layer, the coarse particle content in the muck is high, whereas the fine particle content is not sufficient. Thus, it is necessary to fill the particle voids using not only foam to improve the muck flow plasticity but also bentonite slurry to improve the muck permeability. The muck conditioning scheme is as follows: simultaneous use of 1:8 bentonite slurry and 3% foam solution with similar BIR and FIR values: FIR = 10–15%, BIR = 10%. After conditioning, the slump value of the muck is 150–210 mm, and the permeability is lower than the order of 10⁻⁵ m/s, meeting the requirements for good flow plasticity and low permeability.

Suggested muck conditioning parameters for the water-rich sand-argillaceous siltstone composite formation

According to the above experimental study, the muck conditioning parameters for the water-rich sand-argillaceous siltstone composite formation are shown in (Fig. 19). The optimum values of the muck conditioning parameters for water-rich sand-argillaceous siltstone composite formations with different section ratios are determined by linear interpolation.

Fig. 19 — Suggested muck conditioning parameters for the water-rich sand-argillaceous siltstone composite formation.

With an increasing proportion of argillaceous siltstone in the composite formation, the content of fine particles increases after cutter tool crushing, the FIR increases from 20 to 50%, and the BIR decreases from 20% to zero. The suggested muck conditioning parameters can be divided into three stages according to the proportion of argillaceous siltstone.

When the proportion of argillaceous siltstone is 10%, the content of the water-rich sand layer is relatively high. If there is too much water, polymers such as muck suspending agents must be added to accelerate flocculation. The optimal muck conditioning scheme is as follows: BIR of 20–25%, polymer (CMC) solution injection ratio of 1%, and FIR of approximately 20%.
When the proportion of argillaceous siltstone is between 10 and 70%, as the proportion of argillaceous siltstone increases, the content of fine particles in the muck increases, and thus, polymer is no longer needed for flocculation. However, the FIR should increase significantly to lubricate and disperse the muck. The optimal muck conditioning scheme is as follows: when the proportion of argillaceous siltstone is 30%, the BIR should be 13–18%, and the FIR should be approximately 25%; when the proportion of argillaceous siltstone is 50%, the BIR should be 8–13%, and the FIR should be approximately 30%.
When the proportion of argillaceous siltstone is higher than 70%, bentonite slurry is no longer needed as a plugging material to reduce the muck permeability. The muck itself has a high content of fine particles, and after cutterhead crushing and stirring, the voids between the particles are effectively filled to achieve a low permeability. At this time, the mudstone dispersant solution (sodium hexametaphosphate) should be added to disperse the muck to facilitate the mixing of the foam into the soil particles to improve the muck fluidity. The optimal muck conditioning scheme is as follows: when the argillaceous siltstone accounts for 70%, the injection ratio of 1:5 mudstone dispersant solution should be 2%, and the FIR should be approximately 40%; when the argillaceous siltstone accounts for 90%, the injection ratio of 1:5 mudstone dispersant solution should be 3%, and the FIR should be 40–50%.

Field application of optimized muck conditioning schemes

Analysis of field boring parameters of the full-section water-rich sand layers

During the shield tunnelling process in this case study, full-section water-rich sand layers composed of coarse sand and gravelly sand are encountered in rings 0 ~ 289 and 1125 ~ 1675. The FIR and BIR as well as machine parameters, including penetration rate, cutterhead torque, and total shield thrust, during the tunnelling process in rings 0–289 are analysed, as shown in Fig. 20, the values shown in Fig. 20 are average of per 5 rings.

Fig. 20 — Tunnelling parameters in the water-rich sand layer.

At the start-up phase during the tunnelling process in full-section water-rich sand layer, with the injection of foam and bentonite slurry, the penetration rate of the shield machine increases from 24 to 40 mm/min. The total thrust and cutterhead torque increase from 6100 Inline graphic and 1500 to 14000 and 2600 , respectively, at the normal operating level. Due to the lack of fine particles in the water-rich sand layer, the bentonite slurry amount is adjusted to identify a suitable injection ratio during the tunnelling process. At the commissioning phase, the penetration rate clearly fluctuates in the range of 30–40 mm/min when the BIR ranges in 7.4–16.7%, meanwhile the FIR is 44–60% (as shown in Fig. 20a). The total thrust force and cutter head torque also show high fluctuations.

During the muck conditioning phase, the muck conditioning scheme is adopted as follows: foam and bentonite slurry are used with similar injection ratios (i.e., FIR: 10–15%, BIR: 10%). However, the FIR value obviously exceeds the theoretical value, which is due to the complex field environment, some foam cannot play a role. More foam was used to ensure conditioning effect, resulting in a significant increase in FIR. In this process, as shown in Fig. 20b, the penetration rate increases from 34 mm/min to 45 mm/min, the total thrust force decreases from 15000–16000 Inline graphic to 14000–15000 , and the torque decreases from 2500–3000 to 2400–2700 , respectively. That is, comparing to the commissioning stage, the usages of the muck conditioners significantly decrease, corresponding to a decrease in total thrust and cutterhead torque, while an increase in penetration rate (as shown in Fig. 20c). According to the field test results and with a combined analysis of previous investigations in water-rich sand layers^28,48,67, the adopted muck conditioning scheme optimizes the shield tunnelling parameters and improves the tunnelling efficiency, which verifies that the laboratory muck conditioning scheme is reasonable and effective.

Analysis of field boring parameters of the water-rich sand layer-argillaceous siltstone composite formation

During the shield tunnelling process in rings 289–393, the shield machine crosses a composite formation with dynamically changing soft-hard interfaces, while the upper tunnel section is a water-rich sand layer and the lower tunnel section is argillaceous siltstone. The injection ratio of muck conditioners as well as machine parameters, including penetration rate, cutterhead torque, and total shield thrust, are analysed in this interval, the parameters of per ring are shown in (Fig. 21).

Muck conditioner usage: In the composite formation, when the shield machine advances, the argillaceous siltstone ratio of the tunnel face dynamically changes, and its content gradually increases. As shown in Fig. 21a,b, as the ring number increases, the FIR gradually increases, and the BIR gradually decreases to zero. This changing rule is consistent with the basic rule obtained from the laboratory test results: the FIR increases from 20 to 50%, and the BIR decreases from 20% to zero.
Penetration rate: As shown in Fig. 21c, with shield tunnelling, although the injection ratio of conditioner changes dynamically with the change of argillaceous siltstone content, the shield machine penetration rate fluctuates steadily in the range of 32–42 mm/min without drastic changes and finally maintains a normal working condition in this section of the composite formation. It is indicated that the muck conditioning scheme obtained from the laboratory tests can effectively solve the problem of penetration rate control when tunnelling in dynamically changing composite formations.
Total thrust force: as the shield machine advances, the total shield thrust decreases from 16000–20000 to 13,000–15,000 (as shown in Fig. 21d), indicating that with the muck conditioning scheme of increasing foam and decreasing bentonite slurry, the soil pressure in the soil chamber is moderate and not too large or too small. Thus, it can enable the shield machine to reduce the thrust while maintaining the normal penetration rate, reducing the energy consuming and improving the digging efficiency.
Cutterhead torque: The cutterhead torque first decreases from 2400–2600 to 1600–2000 and then increases to 2400–2800 when tunnelling from rings 350 to 393. This changing process shows that the mud cake ahead of the cutterhead is not obvious due to the muck conditioning effects in the early stage. When the argillaceous siltstone content in the excavation section increases, only using the mixture of foam and bentonite slurry as muck conditioning can no longer meet the requirements, which makes the cutterhead torque rise. Thus, at this stage, there is a need for muck conditioner to reduce the muck adhesion. In Sect. 5.2 it is proposed that when the argillaceous siltstone content is greater than 70%, it is necessary to add the mudstone dispersant solution to disperse the muck, and this muck conditioning scheme is a good match for the field conditions at this stage. Thus, the analysis of the field performance verifies that the laboratory muck conditioning scheme is suitable for the actual project.

Fig. 21 — Shield tunnelling parameters in the composite formation.

Conclusions

In this study, based on the real engineering EPB shield tunnelling case in Nanchang Metro Line 4, muck conditioning optimization for typical water-rich sand-argillaceous siltstone composite formation is investigated through laboratory tests and field applications. Some main conclusions are drawn as follows:

Test results demonstrate that bentonite slurry with viscosity of 22 mPa·s and a bentonite-to-water mass ratio of 1:8 gives the best performance in muck conditioning. Foam conditioner with FER of 14, HLT of 15 min, and concentration of 3% shows the most significant conditioning effect in laboratory experiments.
For a full-section water-rich sand layer, the muck conditioning effectivenesses of single foam, single bentonite slurry, and a composite of foam and bentonite slurry are compared, and their effects on the flow plasticity and permeability of the muck are evaluated. It is concluded that when using the foam and bentonite slurry composite as the muck conditioner, a reasonable scheme is obtained as: FIR = 10–15%, and BIR = 10% (bentonite-to-water mass ratio of 1:8). At this case, the optimum usage of foam and bentonite slurry can meet the requirements for muck flow plasticity and permeability.
For a water-rich sand layer-argillaceous siltstone composite formation, particle gradation analysis and laboratory muck conditioning tests show that as the argillaceous siltstone ratio increases from 10 to 90%, the FIR increases from 20 to 50%, and the BIR decreases gradually from 20% to zero. It is necessary to add mudstone dispersant solution to disperse the muck when the proportion of argillaceous siltstone is higher than 70%.
The muck conditioning scheme proposed based on laboratory muck conditioning tests are finally verified in the field application by analysing parameters such as the muck conditioner injection ratio and shield machine parameters. The research findings obtained in this study can provide guidance to other EPB shield tunnelling cases in similar full-section water-rich sand layers and water-rich sand layer-argillaceous siltstone composite formations.

Acknowledgements

The research work is supported by the National Natural Science Foundation of China (Grant No. 42377169, 42177140, 42267022), The Natural Science Foundation of Jiangxi Province (No.20232BCJ23004, 20232ACB214011), the State Key Laboratory of Performance Monitoring and Protecting of Rail Transit Infrastructure Foundation (No. HJGZ2021102).

Author contributions

Yalong Jiang: Conceptualization, Methodology, Investigation, Data Curation, Writing. Xin Tang: Methodology, Investigation, Data Curation, Writing. Bitang Zhu: Validation, Writing—Review& Editing. Kai Zhou: Investigation, Data Curation, Resources Yucong Pan: Data Curation, Resources, Funding acquisition. Changjie Xu: Supervision. ML Validation, Writing-Review& Editing, Methodology.

Data availability

Some or all data, models, or code that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1.Editorial Department of China Journal of Highway and Transport. Review on China’s tunnel engineering research: 2015. Ch. J. Highw. Transp.28(5), 1–65 (2015). [Google Scholar]
2.Editorial Department of China Journal of Highway and Transport. Review on China’s traffic tunnel engineering research: 2022. Ch. J. Highw. Transp.35(4), 1–40 (2022). [Google Scholar]
3.Edalat, K., Vahdatirad, M., Ghodrat, H., Firouzianbandpey, S. & Barari, A. Choosing TBM for tabriz subway using multi criteria method. J. Civ. Eng. Manag.16, 145–152 (2010). [Google Scholar]
4.Herrenknecht, M., Thewes, M. & Budach, C. The development of earth pressure shields: from the beginning to the present/Entwicklung der Erddruckschilde: Von den Anfängen bis zur Gegenwart. Geomechan. Tunn.4, 11–35 (2011). [Google Scholar]
5.Peila, D., Martinelli, D., Todaro, C. & Luciani, A. Soil conditioning in EPB shield tunnelling—An overview of laboratory tests. Geomechan. Tunn.12(5), 491–498 (2019). [Google Scholar]
6.Wang, S. et al. State-of-the-art on theories and technologies of soil conditioning for shield tunnelling. Ch. J. Highw. Transp.73, 244–251 (2020). [Google Scholar]
7.Todaro, C., Carigi, A., Peila, L., Martinelli, D. & Peila, D. Soil conditioning tests of clay for EPB tunnelling. Undergr. Sp.7(4), 483–497 (2022). [Google Scholar]
8.Onate, S. C. G., Martinelli, D., Todaro, C. & Pelia, D. Study of wear in conditioned granular soil by using a new test device ITA-AITES world tunnel congress 2016 WTC 2016. Soc. Min. Min. Metall. Explor.3, 2445–2454 (2016). [Google Scholar]
9.Peila, D., Oggeri, C. & Vinai, R. Screw conveyor device for laboratory tests on conditioned soil for EPB tunneling operations. J. Geotech. Geoenviron.133, 1622–1625 (2007). [Google Scholar]
10.Zumsteg, R., Plötze, M. & Puzrin, A. M. Effect of soil conditioners on the pressure and rate-dependent shear strength of different clays. J. Geotech. Geoenviron.138, 1138–1146 (2012). [Google Scholar]
11.Psomas, S. Properties of foam/sand mixtures for tunnelling applications. The University of Oxford (2001).
12.Peila, D., Oggeri, C. & Borio, L. Using the slump test to assess the behavior of conditioned soil for EPB tunneling. Environ. Eng. Geosci.15, 167–174 (2009). [Google Scholar]
13.Peila, D., Oggeri, C. & Borio, L. Influence of granulometry, time and temperature on soil conditioning for EPBS applications. Proceedings World Tunnel Congress 22–24. (2008).
14.Budach, C. & Thewes, M. Application ranges of EPB shields in coarse ground based on laboratory research. Tunn. Undergr. Sp. Tech.50, 296–304 (2015). [Google Scholar]
15.Borio, L. & Peila, D. Study of the permeability of foam conditioned soils with laboratory tests. Am. J. Environ. Sci.6(4), 365 (2010). [Google Scholar]
16.Bezuijen, A., Van Lottum, H. Referencing in Tunnelling. a decade of progress. geodelft 1995–2005. (2005).
17.Bezuijen, A. & Gerheim Souza Dias, T. EPB, chamber pressure dissipation during standstill. Proceeding IV International Conference on Computational methods in tunneling and subsurface engineering. Arbeitsbereich für Festigkeitslehre und Baustatik, 1, 225–231. (2017).
18.Vinai, R., Oggeri, C. & Peila, D. Soil conditioning of sand for EPB applications: A laboratory research. Tunn. Undergr. Sp. Tech.23, 308–317 (2008). [Google Scholar]
19.Ye, X. Y., Wang, S. Y., Yang, J. S., Sheng, D. C. & Xiao, C. Soil conditioning for EPB shield tunneling in argillaceous siltstone with high content of clay minerals: Case study. Int. J. Geomech.17(4), 05016002 (2017). [Google Scholar]
20.Zumsteg, R., Plötze, M. & Puzrin, A. M. Effects of dispersing foams and polymers on the mechanical behaviour of clay pastes. Géotechnique63(11), 920–933 (2013). [Google Scholar]
21.Peila, D., Picchio, A., Martinelli, D. & Negro, E. D. Laboratory tests on soil conditioning of clayey soil. Acta Geotech.11, 1061–1074 (2016). [Google Scholar]
22.Hollmann, F. S. & Thewes, M. Assessment method for clay clogging and disintegration of fines in mechanised tunnelling. Tunn. Undergr. Sp. Tech.37, 96–106 (2013). [Google Scholar]
23.Kim, T. H., Lee, I. M., Chung, H. Y., Park, J. J. & Ryu, Y. M. Application ranges of EPB shield TBM in weathered granite Soil: A laboratory scale study. Appl. Sci.11, 2995 (2021). [Google Scholar]
24.Martinelli, D., Peila, D. & Campa, E. Feasibility study of tar sands conditioning for earth pressure balance tunnelling. J. Rock Mech. Geotech.7, 684–690 (2015). [Google Scholar]
25.Zumsteg, R. & Langmaack, L. Mechanized tunneling in soft soils: Choice of excavation mode and application of soil-conditioning additives in glacial deposits. Engineering3, 863–870 (2017). [Google Scholar]
26.Thewes, M., Gramlich, N. & Langmaack, L. New Method for conditioning EPB shields in coarse-grained soils with foam containing solids. STUVA Conference, Munich 153–159. (2023).
27.Song, Y. R., He, J., Chen, G. F., Hong, M. & Li, P. B. Response of the spiral conveyor spewing in water-rich sand. Appl. Mechan. Mater.580, 1138–1142 (2014). [Google Scholar]
28.Huang, Z. et al. Conditioning experiment on sand and cobble soil for shield tunneling. Tunn. Undergr. Sp. Tech.87, 187–194 (2019). [Google Scholar]
29.Zhao, S., Li, S., Wan, Z. & Wang, M. Dispersant for reducing mud cakes of slurry shield tunnel boring machine in sticky ground. Adv. Mater. Sci. Eng.2021(1), 5524489 (2021). [Google Scholar]
30.Wang, H., Li, J., Wang, Z., Wang, D. & Zhan, H. Experimental investigation of the mechanism of foaming agent concentration affecting foam stability. J. Surfact. Deterg.20, 1443–1451 (2017). [Google Scholar]
31.Peña Duarte, M.A. Foam as a soil conditioner in tunnelling: physical and mechanical properties of conditioned sands. University of Oxford (2007).
32.Qiao, G. Development of new foam agent for EPB shield machine and foam-soil modification (Ph. D. Dissertation). China University of Mining & Technology, Beijing (2009).
33.Yang, Y., Wang, G., Li, H. & Huang, X. The new clay mud and its improvement effects of tunnels. Appl. Clay Sci.79, 49–56 (2013). [Google Scholar]
34.Lin, X., Zhou, X. & Yang, Y. A new soil conditioner for highly permeable sandy gravel stratum in EPBs. Appl. Sci.11, 2109 (2021). [Google Scholar]
35.Thewes, M., Budach, C., Bezuijen, A. Foam conditioning in EPB tunnelling. 7th International Symposium on Geotechnical Aspects of Underground Construction in Soft Ground. Crc Press-Taylor & Francis Group, Rome, ITALY, 127–135 (2011).
36.Carigi, A., Todaro, C., Martinelli, D. & Peila, D. A more comprehensive way to analyze foam stability for EPB tunnelling—Introduction of a mathematical characterization. Geosciences12(5), 191 (2022). [Google Scholar]
37.Sebastiani, D., Vilardi, G., Bavasso, I., Di Palma, L. & Miliziano, S. Classification of foam and foaming products for EPB mechanized tunnelling based on half-life time. Tunn. Undergr. Sp. Tech.92, 103044 (2019). [Google Scholar]
38.Monsalve, A. & Schechter, R. S. The stability of foams: Dependence of observation on the bubble size distribution. J. Coll. Interf. Sci.97, 327–335 (1984). [Google Scholar]
39.Pagureva, N., Tcholakova, S., Rusanova, K., Denkov, N. & Dimitrova, T. Factors affecting the coalescence stability of microbubbles. Coll. Surface A508, 21–29 (2016). [Google Scholar]
40.Samanta, S. & Ghosh, P. Coalescence of bubbles and stability of foams in aqueous solutions of Tween surfactants. Chem. Eng. Res. Des.89, 2344–2355 (2011). [Google Scholar]
41.Vandewalle, N. et al. Foam stability in microgravity. J. Phys. Conf. Ser.327(1), 012024 (2011). [Google Scholar]
42.Zhao, B., Liu, D. & Jiang, B. Soil conditioning of waterless sand–pebble stratum in EPB tunnel construction. Geotech. Geol. Eng.36(4), 2495–2504 (2018). [Google Scholar]
43.Carey, E. & Stubenrauch, C. Foaming properties of mixtures of a non-ionic (C12DMPO) and an ionic surfactant (C12TAB). J. Coll. Interf. Sci.346, 414–423 (2010). [DOI] [PubMed] [Google Scholar]
44.Exerowa, D. & Kruglyakov, P. M. Referencing in Foam and Foam Films: Theory, Experiment, Application (Elsevier, 1997). [Google Scholar]
45.Galli, M., Thewes, M. & Freimann, S. Soil Conditioning for EPB Tunnelling: Relevant Foam Properties and Characterisation 2139–2148 (CRC Press, 2019). [Google Scholar]
46.Wu, Y., Mooney, M. A. & Cha, M. An experimental examination of foam stability under pressure for EPB TBM tunneling. Tunn. Undergr. Sp. Tech.77, 80–93 (2018). [Google Scholar]
47.Wu, Y. L., Nazem, A., Meng, F. Y. & Mooney, M. A. Experimental study on the stability of foam-conditioned sand under pressure in the EPBM chamber. Tunn. Undergr. Sp. Tech.106, 103590 (2020). [Google Scholar]
48.Zhen, Z., Ge, X. & Zhang, J. Soil conditioning tests on sandy and cobbly soil for shield tunneling. KSCE J. Civ. Eng.25, 1229–1238 (2021). [Google Scholar]
49.Martinelli, D., Winderholler, R. & Peila, D. Undrained behaviour of granular soils conditioned for EPB tunnelling—A new experimental procedure. Geomechan. Tunn.10, 81–89 (2017). [Google Scholar]
50.Messerklinger, S., Zumsteg, R. & Puzrin, A. A new pressurized vane shear apparatus. Geotech. Test. J.34(2), 112–121 (2011). [Google Scholar]
51.Milligan, G. Lubrication and Soil Conditioning in Tunnelling, Pipe Jacking and Microtunnelling: A State-of-the-art Review (Geotechnical Consulting Group, 2000). [Google Scholar]
52.Huang, S., Wang, S., Xu, C., Shi, Y. & Ye, F. Effect of grain gradation on the permeability characteristics of coarse-grained soil conditioned with foam for EPB shield tunneling. KSCE J. Civ. Eng.23, 4662–4674 (2019). [Google Scholar]
53.Thewes, M. & Budach, C. Soil conditioning with foam during EPB tunnelling. Geomechan. Tunn.3, 256–267 (2010). [Google Scholar]
54.Wang, S. et al. Permeability stability calculation model of foam-conditioned soil based on the permeability constant. Int. J. Numer. Anal. Met.45, 540–559 (2021). [Google Scholar]
55.Wang, S. et al. Permeability characteristics of poorly graded sand conditioned with foam in different conditioning states. J. Test. Eval.49, 3620–3636 (2020). [Google Scholar]
56.Wei, Y. J., Wang, D. L., Li, J. G. & Jie, Y. X. Effects of soil conditioning on characteristics of a clay-sand-gravel mixed soil based on laboratory test. Appl. Sci.10(9), 3300 (2020). [Google Scholar]
57.Wang, S., Liu, P., Hu, Q. & Zhong, J. Effect of dispersant on the tangential adhesion strength between clay and metal for EPB shield tunnelling. Tunn. Undergr. Sp. Tech.95, 103144 (2020). [Google Scholar]
58.Liu, P. et al. Changes of Atterberg limits and electrochemical behaviors of clays with dispersants as conditioning agents for EPB shield tunnelling. Tunn. Undergr. Sp. Tech.73, 244–251 (2018). [Google Scholar]
59.Zumsteg, R., Plötze, M., Puzrin, A. Reduction of the clogging potential of clays: new chemical applications and novel quantification approaches. Bio-and Chemo-Mechanical Processes in Geotechnical Engineering: Géotechnique Symposium in Print 2013. pp. 44–54. ICE Publishing, (2014).
60.Zumsteg, R. & Puzrin, A. M. Stickiness and adhesion of conditioned clay pastes. Tunn. Undergr. Sp. Tech.31, 86–96 (2012). [Google Scholar]
61.Yang, Y., Li, X. & Li, X. Shear strength and compression coefficient for conditioned sand subjected to earth chamber stress levels. Adv. Mater. Sci. Eng.2018(1), 1759151 (2018). [Google Scholar]
62.Merritt, Stuart, A. Conditioning of clay soils for tunnelling machine screw conveyors. University of Cambridge (2005).
63.Wei, K. L. Microscopic mechanism analysis of foam and bentonite improved soil in earth pressure balance shield construction. Modern Tunn. Technol.44(1), 73–77 (2007). [Google Scholar]
64.Budach, C. Extended application ranges of EPB-Shields in coarse-grained soils. Doctoral thesis (2012).
65.Bezuijen, A., Pruiksma, J. P. & Van Meerten, H. H. Pore Pressures in Front of Tunnel Measurements Calculations and Consequences for Stability of Tunnel Face (CRC Press, 2020). [Google Scholar]
66.Bezuijen, A., Talmon, A., Joustra, J. & Grote, B. Pressure gradients and muck properties at the face of an EPB. Proceeding of Geotechnical aspects of underground construction in soft ground 195–201 (2006).
67.Li, S. et al. Soil conditioning tests on sandy soil for earth pressure balance shield tunneling and field applications. Tunn. Undergr. Sp. Tech.120, 104271 (2022). [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Some or all data, models, or code that support the findings of this study are available from the corresponding author upon reasonable request.

[CR1] 1.Editorial Department of China Journal of Highway and Transport. Review on China’s tunnel engineering research: 2015. Ch. J. Highw. Transp.28(5), 1–65 (2015). [Google Scholar]

[CR2] 2.Editorial Department of China Journal of Highway and Transport. Review on China’s traffic tunnel engineering research: 2022. Ch. J. Highw. Transp.35(4), 1–40 (2022). [Google Scholar]

[CR3] 3.Edalat, K., Vahdatirad, M., Ghodrat, H., Firouzianbandpey, S. & Barari, A. Choosing TBM for tabriz subway using multi criteria method. J. Civ. Eng. Manag.16, 145–152 (2010). [Google Scholar]

[CR4] 4.Herrenknecht, M., Thewes, M. & Budach, C. The development of earth pressure shields: from the beginning to the present/Entwicklung der Erddruckschilde: Von den Anfängen bis zur Gegenwart. Geomechan. Tunn.4, 11–35 (2011). [Google Scholar]

[CR5] 5.Peila, D., Martinelli, D., Todaro, C. & Luciani, A. Soil conditioning in EPB shield tunnelling—An overview of laboratory tests. Geomechan. Tunn.12(5), 491–498 (2019). [Google Scholar]

[CR6] 6.Wang, S. et al. State-of-the-art on theories and technologies of soil conditioning for shield tunnelling. Ch. J. Highw. Transp.73, 244–251 (2020). [Google Scholar]

[CR7] 7.Todaro, C., Carigi, A., Peila, L., Martinelli, D. & Peila, D. Soil conditioning tests of clay for EPB tunnelling. Undergr. Sp.7(4), 483–497 (2022). [Google Scholar]

[CR8] 8.Onate, S. C. G., Martinelli, D., Todaro, C. & Pelia, D. Study of wear in conditioned granular soil by using a new test device ITA-AITES world tunnel congress 2016 WTC 2016. Soc. Min. Min. Metall. Explor.3, 2445–2454 (2016). [Google Scholar]

[CR9] 9.Peila, D., Oggeri, C. & Vinai, R. Screw conveyor device for laboratory tests on conditioned soil for EPB tunneling operations. J. Geotech. Geoenviron.133, 1622–1625 (2007). [Google Scholar]

[CR10] 10.Zumsteg, R., Plötze, M. & Puzrin, A. M. Effect of soil conditioners on the pressure and rate-dependent shear strength of different clays. J. Geotech. Geoenviron.138, 1138–1146 (2012). [Google Scholar]

[CR11] 11.Psomas, S. Properties of foam/sand mixtures for tunnelling applications. The University of Oxford (2001).

[CR12] 12.Peila, D., Oggeri, C. & Borio, L. Using the slump test to assess the behavior of conditioned soil for EPB tunneling. Environ. Eng. Geosci.15, 167–174 (2009). [Google Scholar]

[CR13] 13.Peila, D., Oggeri, C. & Borio, L. Influence of granulometry, time and temperature on soil conditioning for EPBS applications. Proceedings World Tunnel Congress 22–24. (2008).

[CR14] 14.Budach, C. & Thewes, M. Application ranges of EPB shields in coarse ground based on laboratory research. Tunn. Undergr. Sp. Tech.50, 296–304 (2015). [Google Scholar]

[CR15] 15.Borio, L. & Peila, D. Study of the permeability of foam conditioned soils with laboratory tests. Am. J. Environ. Sci.6(4), 365 (2010). [Google Scholar]

[CR16] 16.Bezuijen, A., Van Lottum, H. Referencing in Tunnelling. a decade of progress. geodelft 1995–2005. (2005).

[CR17] 17.Bezuijen, A. & Gerheim Souza Dias, T. EPB, chamber pressure dissipation during standstill. Proceeding IV International Conference on Computational methods in tunneling and subsurface engineering. Arbeitsbereich für Festigkeitslehre und Baustatik, 1, 225–231. (2017).

[CR18] 18.Vinai, R., Oggeri, C. & Peila, D. Soil conditioning of sand for EPB applications: A laboratory research. Tunn. Undergr. Sp. Tech.23, 308–317 (2008). [Google Scholar]

[CR19] 19.Ye, X. Y., Wang, S. Y., Yang, J. S., Sheng, D. C. & Xiao, C. Soil conditioning for EPB shield tunneling in argillaceous siltstone with high content of clay minerals: Case study. Int. J. Geomech.17(4), 05016002 (2017). [Google Scholar]

[CR20] 20.Zumsteg, R., Plötze, M. & Puzrin, A. M. Effects of dispersing foams and polymers on the mechanical behaviour of clay pastes. Géotechnique63(11), 920–933 (2013). [Google Scholar]

[CR21] 21.Peila, D., Picchio, A., Martinelli, D. & Negro, E. D. Laboratory tests on soil conditioning of clayey soil. Acta Geotech.11, 1061–1074 (2016). [Google Scholar]

[CR22] 22.Hollmann, F. S. & Thewes, M. Assessment method for clay clogging and disintegration of fines in mechanised tunnelling. Tunn. Undergr. Sp. Tech.37, 96–106 (2013). [Google Scholar]

[CR23] 23.Kim, T. H., Lee, I. M., Chung, H. Y., Park, J. J. & Ryu, Y. M. Application ranges of EPB shield TBM in weathered granite Soil: A laboratory scale study. Appl. Sci.11, 2995 (2021). [Google Scholar]

[CR24] 24.Martinelli, D., Peila, D. & Campa, E. Feasibility study of tar sands conditioning for earth pressure balance tunnelling. J. Rock Mech. Geotech.7, 684–690 (2015). [Google Scholar]

[CR25] 25.Zumsteg, R. & Langmaack, L. Mechanized tunneling in soft soils: Choice of excavation mode and application of soil-conditioning additives in glacial deposits. Engineering3, 863–870 (2017). [Google Scholar]

[CR26] 26.Thewes, M., Gramlich, N. & Langmaack, L. New Method for conditioning EPB shields in coarse-grained soils with foam containing solids. STUVA Conference, Munich 153–159. (2023).

[CR27] 27.Song, Y. R., He, J., Chen, G. F., Hong, M. & Li, P. B. Response of the spiral conveyor spewing in water-rich sand. Appl. Mechan. Mater.580, 1138–1142 (2014). [Google Scholar]

[CR28] 28.Huang, Z. et al. Conditioning experiment on sand and cobble soil for shield tunneling. Tunn. Undergr. Sp. Tech.87, 187–194 (2019). [Google Scholar]

[CR29] 29.Zhao, S., Li, S., Wan, Z. & Wang, M. Dispersant for reducing mud cakes of slurry shield tunnel boring machine in sticky ground. Adv. Mater. Sci. Eng.2021(1), 5524489 (2021). [Google Scholar]

[CR30] 30.Wang, H., Li, J., Wang, Z., Wang, D. & Zhan, H. Experimental investigation of the mechanism of foaming agent concentration affecting foam stability. J. Surfact. Deterg.20, 1443–1451 (2017). [Google Scholar]

[CR31] 31.Peña Duarte, M.A. Foam as a soil conditioner in tunnelling: physical and mechanical properties of conditioned sands. University of Oxford (2007).

[CR32] 32.Qiao, G. Development of new foam agent for EPB shield machine and foam-soil modification (Ph. D. Dissertation). China University of Mining & Technology, Beijing (2009).

[CR33] 33.Yang, Y., Wang, G., Li, H. & Huang, X. The new clay mud and its improvement effects of tunnels. Appl. Clay Sci.79, 49–56 (2013). [Google Scholar]

[CR34] 34.Lin, X., Zhou, X. & Yang, Y. A new soil conditioner for highly permeable sandy gravel stratum in EPBs. Appl. Sci.11, 2109 (2021). [Google Scholar]

[CR35] 35.Thewes, M., Budach, C., Bezuijen, A. Foam conditioning in EPB tunnelling. 7th International Symposium on Geotechnical Aspects of Underground Construction in Soft Ground. Crc Press-Taylor & Francis Group, Rome, ITALY, 127–135 (2011).

[CR36] 36.Carigi, A., Todaro, C., Martinelli, D. & Peila, D. A more comprehensive way to analyze foam stability for EPB tunnelling—Introduction of a mathematical characterization. Geosciences12(5), 191 (2022). [Google Scholar]

[CR37] 37.Sebastiani, D., Vilardi, G., Bavasso, I., Di Palma, L. & Miliziano, S. Classification of foam and foaming products for EPB mechanized tunnelling based on half-life time. Tunn. Undergr. Sp. Tech.92, 103044 (2019). [Google Scholar]

[CR38] 38.Monsalve, A. & Schechter, R. S. The stability of foams: Dependence of observation on the bubble size distribution. J. Coll. Interf. Sci.97, 327–335 (1984). [Google Scholar]

[CR39] 39.Pagureva, N., Tcholakova, S., Rusanova, K., Denkov, N. & Dimitrova, T. Factors affecting the coalescence stability of microbubbles. Coll. Surface A508, 21–29 (2016). [Google Scholar]

[CR40] 40.Samanta, S. & Ghosh, P. Coalescence of bubbles and stability of foams in aqueous solutions of Tween surfactants. Chem. Eng. Res. Des.89, 2344–2355 (2011). [Google Scholar]

[CR41] 41.Vandewalle, N. et al. Foam stability in microgravity. J. Phys. Conf. Ser.327(1), 012024 (2011). [Google Scholar]

[CR42] 42.Zhao, B., Liu, D. & Jiang, B. Soil conditioning of waterless sand–pebble stratum in EPB tunnel construction. Geotech. Geol. Eng.36(4), 2495–2504 (2018). [Google Scholar]

[CR43] 43.Carey, E. & Stubenrauch, C. Foaming properties of mixtures of a non-ionic (C12DMPO) and an ionic surfactant (C12TAB). J. Coll. Interf. Sci.346, 414–423 (2010). [DOI] [PubMed] [Google Scholar]

[CR44] 44.Exerowa, D. & Kruglyakov, P. M. Referencing in Foam and Foam Films: Theory, Experiment, Application (Elsevier, 1997). [Google Scholar]

[CR45] 45.Galli, M., Thewes, M. & Freimann, S. Soil Conditioning for EPB Tunnelling: Relevant Foam Properties and Characterisation 2139–2148 (CRC Press, 2019). [Google Scholar]

[CR46] 46.Wu, Y., Mooney, M. A. & Cha, M. An experimental examination of foam stability under pressure for EPB TBM tunneling. Tunn. Undergr. Sp. Tech.77, 80–93 (2018). [Google Scholar]

[CR47] 47.Wu, Y. L., Nazem, A., Meng, F. Y. & Mooney, M. A. Experimental study on the stability of foam-conditioned sand under pressure in the EPBM chamber. Tunn. Undergr. Sp. Tech.106, 103590 (2020). [Google Scholar]

[CR48] 48.Zhen, Z., Ge, X. & Zhang, J. Soil conditioning tests on sandy and cobbly soil for shield tunneling. KSCE J. Civ. Eng.25, 1229–1238 (2021). [Google Scholar]

[CR49] 49.Martinelli, D., Winderholler, R. & Peila, D. Undrained behaviour of granular soils conditioned for EPB tunnelling—A new experimental procedure. Geomechan. Tunn.10, 81–89 (2017). [Google Scholar]

[CR50] 50.Messerklinger, S., Zumsteg, R. & Puzrin, A. A new pressurized vane shear apparatus. Geotech. Test. J.34(2), 112–121 (2011). [Google Scholar]

[CR51] 51.Milligan, G. Lubrication and Soil Conditioning in Tunnelling, Pipe Jacking and Microtunnelling: A State-of-the-art Review (Geotechnical Consulting Group, 2000). [Google Scholar]

[CR52] 52.Huang, S., Wang, S., Xu, C., Shi, Y. & Ye, F. Effect of grain gradation on the permeability characteristics of coarse-grained soil conditioned with foam for EPB shield tunneling. KSCE J. Civ. Eng.23, 4662–4674 (2019). [Google Scholar]

[CR53] 53.Thewes, M. & Budach, C. Soil conditioning with foam during EPB tunnelling. Geomechan. Tunn.3, 256–267 (2010). [Google Scholar]

[CR54] 54.Wang, S. et al. Permeability stability calculation model of foam-conditioned soil based on the permeability constant. Int. J. Numer. Anal. Met.45, 540–559 (2021). [Google Scholar]

[CR55] 55.Wang, S. et al. Permeability characteristics of poorly graded sand conditioned with foam in different conditioning states. J. Test. Eval.49, 3620–3636 (2020). [Google Scholar]

[CR56] 56.Wei, Y. J., Wang, D. L., Li, J. G. & Jie, Y. X. Effects of soil conditioning on characteristics of a clay-sand-gravel mixed soil based on laboratory test. Appl. Sci.10(9), 3300 (2020). [Google Scholar]

[CR57] 57.Wang, S., Liu, P., Hu, Q. & Zhong, J. Effect of dispersant on the tangential adhesion strength between clay and metal for EPB shield tunnelling. Tunn. Undergr. Sp. Tech.95, 103144 (2020). [Google Scholar]

[CR58] 58.Liu, P. et al. Changes of Atterberg limits and electrochemical behaviors of clays with dispersants as conditioning agents for EPB shield tunnelling. Tunn. Undergr. Sp. Tech.73, 244–251 (2018). [Google Scholar]

[CR59] 59.Zumsteg, R., Plötze, M., Puzrin, A. Reduction of the clogging potential of clays: new chemical applications and novel quantification approaches. Bio-and Chemo-Mechanical Processes in Geotechnical Engineering: Géotechnique Symposium in Print 2013. pp. 44–54. ICE Publishing, (2014).

[CR60] 60.Zumsteg, R. & Puzrin, A. M. Stickiness and adhesion of conditioned clay pastes. Tunn. Undergr. Sp. Tech.31, 86–96 (2012). [Google Scholar]

[CR61] 61.Yang, Y., Li, X. & Li, X. Shear strength and compression coefficient for conditioned sand subjected to earth chamber stress levels. Adv. Mater. Sci. Eng.2018(1), 1759151 (2018). [Google Scholar]

[CR62] 62.Merritt, Stuart, A. Conditioning of clay soils for tunnelling machine screw conveyors. University of Cambridge (2005).

[CR63] 63.Wei, K. L. Microscopic mechanism analysis of foam and bentonite improved soil in earth pressure balance shield construction. Modern Tunn. Technol.44(1), 73–77 (2007). [Google Scholar]

[CR64] 64.Budach, C. Extended application ranges of EPB-Shields in coarse-grained soils. Doctoral thesis (2012).

[CR65] 65.Bezuijen, A., Pruiksma, J. P. & Van Meerten, H. H. Pore Pressures in Front of Tunnel Measurements Calculations and Consequences for Stability of Tunnel Face (CRC Press, 2020). [Google Scholar]

[CR66] 66.Bezuijen, A., Talmon, A., Joustra, J. & Grote, B. Pressure gradients and muck properties at the face of an EPB. Proceeding of Geotechnical aspects of underground construction in soft ground 195–201 (2006).

[CR67] 67.Li, S. et al. Soil conditioning tests on sandy soil for earth pressure balance shield tunneling and field applications. Tunn. Undergr. Sp. Tech.120, 104271 (2022). [Google Scholar]

PERMALINK

Research on muck conditioning for EPB shield tunnelling in composite formation

Yalong Jiang

Xin Tang

Bitang Zhu

Kai Zhou

Yucong Pan

Changjie Xu

Abstract

Introduction

Fig. 1.

Table 1.

Background

Engineering geological conditions

Fig. 2.

Muck particle gradation analysis

Table 2.

Fig. 3.

Muck conditioner selection and conditioning effectiveness evaluation

Muck conditioner selection

Performance testing and selection of bentonite

Fig. 4.

Performance testing and selection of the foam

Fig. 5.

Foam expansion ratio

Fig. 6.

Fig. 7.

Half-life time

Fig. 8.

Evaluation of muck conditioning effects

Fig. 9.

Laboratory muck conditioning tests

Muck conditioning for the full-section water-rich sand layer

Muck conditioning tests using single foam for the sandy soil layer

Fig. 10.

Muck conditioning tests using single bentonite slurry for the sandy soil layer

Fig. 11.

Muck conditioning tests using foam and bentonite slurry composite for the sandy soil layer

Table 3.

Fig. 12.

Muck conditioning tests using foam and bentonite slurry composite for the water-rich sand-argillaceous siltstone composite formation

Tunnel face selection and muck conditioning scheme determination

Fig. 13.

Muck conditioning tests for the water-rich sand-argillaceous siltstone (volume ratio 9:1) composite formation

Fig. 14.

Table 4.

Muck conditioning tests for the water-rich sand-argillaceous siltstone (volume ratios 7:3, 5:5, 3:7) composite formation

Fig. 15.

Muck conditioning tests for the water-rich sand-argillaceous siltstone (volume ratio 1:9) composite formation

Fig. 16.

Table 5.

Evaluation of the muck conditioning effectiveness

Flow plasticity

Fig. 17.

Shear strength

Table 6.

Permeability

Fig. 18.

Recommended muck conditioning schemes

Suggested muck conditioning parameters for the full-section water-rich sand layer

Suggested muck conditioning parameters for the water-rich sand-argillaceous siltstone composite formation

Fig. 19.

Field application of optimized muck conditioning schemes

Analysis of field boring parameters of the full-section water-rich sand layers

Fig. 20.

Analysis of field boring parameters of the water-rich sand layer-argillaceous siltstone composite formation

Fig. 21.

Conclusions

Acknowledgements

Author contributions

Data availability

Declarations

Competing interests

Footnotes

References

Associated Data

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES