1 December 2019 in Géotechnique
Experiments on bentonite slurry infiltration into saturated sand have been carried out in a modified laboratory set-up that provides a hydraulic gradient comparable to real tunnels. Test series 1 investigated the characteristics of mud spurt and filter cake formation during water–bentonite slurry infiltration, before and after removing the external filter cake formed on the original boundary between the slurry and the sand. Test series 2 examined the conditions of filter cake formation during water–bentonite–sand slurry infiltration. The experimental results of series 1 indicate that, in the second infiltration, a new external filter cake will be formed for low concentrations of slurries (40 and 50 g/l), whereas hardly any new external filter cake formed for high concentration of slurry (60 g/l) owing to the formation of an internal filter cake in the sand during the first infiltration. The formation of this internal filter cake meant that the infiltration distance was much smaller for the second infiltration than that in the first for 60 g/l slurry. The experimental results of series 2 with a water–bentonite–sand slurry showed no external filter cake formation, but depth filtration was observed for high-density slurry (1300 and 1500 kg/m3). For low-density slurries (1050 and 1100 kg/m3), a thicker external filter cake, but with a higher permeability, than that in series 1 was observed on the sand surface, because the cake consisted to a large extent of sand particles. Mechanisms of mud spurt and filter cake formation are discussed, with a new solution for mud spurt and the theory of depth filtration.
1 April 2019 in Géotechnique
Infiltration of pressurised bentonite slurry into saturated sand has been investigated in two laboratory set-ups. The first set-up is comparable to set-ups described in the literature for slurry infiltration tests. The second one results in a hydraulic gradient in the soil comparable to what can be expected in front of a tunnel-boring machine. The different characteristics of mud spurt and filter cake formation were clearly identified in the experimental results. The transition from mud spurt to filter cake formation depends on the infiltration velocity according to the literature. However, this was not found in these tests.
26 September 2018 in Environmental Earth Sciences
1 April 2022 in Géotechnique
Infiltration of pressurised foam into saturated sand was investigated in a laboratory set-up that provided an hydraulic gradient comparable to real tunnels. Four different stages were identified in the infiltration: ‘foam spurt’ (the foam infiltration into the pores of the sand); water flow with reduced permeability; water flow with increased permeability; and a residual stage. It appears that the infiltration depth of foam into sand decreases with increasing foam expansion ratio (FER) of the foam until a limit FER (approximately 15 for the foaming agent and sand used in this experiment). For higher FERs the permeability remains more or less constant. It was also found that a ‘dry’ foam (FER = 20) is not essential for formation of a low-permeability layer at the sand surface; even a ‘wet’ foam (FER = 5) can form a low-permeability layer. Furthermore, a foam-infiltrated zone is essential to achieve a low-permeability layer. In these tests, the flow velocities were lower than the normal excavation velocity of a tunnel-boring machine. Thus there was no foam-infiltrated zone and consequently no zone with low permeability during drilling. By comparing the pore pressures measured using pore pressure transducers with and without hydrophilic filter papers, the capillary pressure in the foam-infiltrated sand was found.
23 July 2022 in KSCE Journal of Civil Engineering
For the safety and efficiency of earth pressure balance (EPB) shield tunnelling, the dynamic characteristics of the muck in the excavation chamber and the screw conveyor should be considered. However, the velocity and pressure distribution of the muck in the EPB shield, reflecting muck flowability and face stability, are rarely observed at the jobsite. Therefore, a computational fluid dynamics (CFD) model with a modified inlet boundary was established to investigate the muck characteristics in the chamber and the screw conveyor during tunnelling in a full chamber mode. A compressible Bingham model was adapted to describe the muck compressibility and rheology. The time-dependent tunnelling parameters obtained from the simulation excellently matched the field-measured data. The simulation results show the nonuniform velocity distribution of the muck at the excavation face and the low-velocity zones where clogging potentially occurs. The muck pressure reaches a peak at the excavation face and then decreases along the cutterhead openings, chamber and screw conveyor for inflow. The simulation also captures the pressure imbalance at the cutterhead and bulkhead, which may destabilize the excavation face. The proposed discharge efficiency is 56.7% in the simulated condition, indicating that tunnelling with these muck properties has no risk of muck spewing.
1 April 2022 in Géotechnique
Experiments on infiltration of pressurised soil–foam mixture into saturated sand have been carried out in a laboratory set-up that provides a hydraulic gradient comparable to real tunnels. The soil–foam mixture used was comparable to the excavated soil (soil–foam mixture) that can be expected in the excavation chamber of an earth pressure balance (EPB) shield. It appears that a higher effective foam expansion ratio (FERm) is more effective to form a low-permeability layer in the sand. Since there will be an interaction between air, water and sand in the soil–foam mixture, the infiltration characteristics of the soil–foam mixture are a little different from that of the foam. It was shown that a larger air fraction at a given solid fraction reduces the foam flow in the sand. The water permeability of foam-infiltrated sand decreases with an increasing air fraction and a decreasing liquid fraction for a given solid fraction, and also decreases with an increasing air fraction and a decreasing solid fraction for a given liquid fraction. The situation with high sand fraction should be avoided in the field because there will be no impermeable or low-permeability layer formed at the tunnel face. Furthermore, the water permeability of foam-infiltrated sand decreases with the foam injection ratio (FIR) and an appropriate FIR is important for EPB shield tunnelling in saturated sand.