Some unsaturated soil-related aspects in railway substructures Yu-Jun CUI Laboratoire Navier - Ecole des Ponts ParisTech 1
Acknowledgements Ecole des Ponts ParisTech (ENPC): - Pr Pierre Delage - Dr Anh Minh Tang - Dr Trong Vinh Duong - Dr Jean-Paul Karam - Dr Viet Nam Trinh - Dr Yang Chao - Dr Jean-Claude Dupla - Dr Jean Canou The French Railway Company (SNCF): - Mr Jean-Marc Terpereau - Mr Gilles Marchadier - Ms Yahel Szerman - Mr Nicolas Calon - Mr Alain Robinet 2/36
The French Railway network Most of railway structures were constructed in late XIX century First LGV was constructed in 1970 The network has not changed much since then 94% conventional lines over 32 000 km Railway network in late XIX century Current railway network 3/36
New lines vs conventional lines New lines: 6% Conventional lines: 94% 4/36
Problems related to non-saturation New lines: Special sub-grade soils such as Collapsible soils Swelling soils Conventional lines: Sub-grade soils Hydro-mechanical behaviour of interlayer soils 5/36
Outline Introduction Case of loess along northern TGV line Case of marl along Mediterranean TGV line Interlayer soil in the sub-structure of conventional lines Concluding remarks 6/36
Locations of the case studies Case loess Case Interlayer soils Case Marl 7/36
Case of loess along northern TGV line 8/36
9/36
10/36
Sampling 11/36
Geotechnical properties Soil S1 (1.20m) S2 (2.20m) S3 (3.50m) S4 (4.90m) 100 d (Mg/m 3 ) 1.52 1.39 1.54 1.55 C ca (%) 5 6 15 9 % < 2 (µm) 20 16 16 18 w p 21 22 20 21 Percentage finer (%) Tamisat (%) 90 80 70 60 50 40 30 20 10 h = 1.2 m h = 2.2 m h = 3.5 m h = 4.9 m Ip 9 6 6 9 Suction s 0 (kpa) 20 34 28 14 0 1 10 100 Grain Diamètre size équivalent (µm)(mm) 12/36
Observation at SEM Soil 2.20 m 13/36
Collapse behaviour of soil-2.2 m Déformation verticale (%) Vertical strain (%) -10-5 0 5 10 15 w o = 18 % ; s o ~30 kpa Vertical stress (kpa) Contrainte verticale (kpa) 1 10 100 1000 Saturé Wetting sous under 3 kpa Saturé Wetting à under 200 200 kpa kpa Teneur Constant en water eau content constante 20 14/36
Liquefaction resistance Load by train 15/36
Liquefaction resistance - (soil 2.20 m) ' c 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0 w=34% w=33% w=32,2% w=30% 0 500 1000 1500 2000 N 16/36
Case of marl along Mediterranean TGV line 17/36
Damage to the drainage system Chabrillan site; line constructed in 1998; damage observed in 2001 18/36
Profiles of heave of Line 1 (to Paris) Heave (mm) 100 80 60 40 20 June 2002 July 2003 March 2004 March 2005 April 2006 March 2007 0-20 0 200 400 600 Distance from Km 530-000 (m) 19/36
Plan of investigation boreholes and the excavation in 2003 Cross-section A-A Natural land (before 1998) Excavation 2003 Axe of the line Excavation 1998 A 0 50 m Boreholes Line 1 (to Paris) Line 2 (to Marseilles) Excavation 2003 B1-1 B1-2 B1-3 B1-4 B1-5 B1-6 Km 530+000 To Paris Km 530+100 B2-1 B2-2 B2-3 Km 530+200 Km 530+300 A Km 530+400 Km 530+500 To Marseilles 20/36
Profiles of boreholes 21/36
Evolution of swell potential 1.4 1.2 Under the reproduced afterexcavation in-situ stress Experimental results Swell (%) 1.0 0.8 0.6 0.4 0.2 Measurements in 2002 Measurements in 2007 0.0 1995 2000 2005 2010 2015 2020 22/36
Interlayer soils 23/36
Conventional railway substructures Were initially built with ballast and sub-grade without separation layer Complex substructure : The interlayer was created mainly by naturally mixing ballast and sub grade Variability of interlayer soils depending on the nature of sub grade Three functions expected for the interlayer when innovating the lines : Separation Support Drainage Sub grade 24/36
Function of interlayer Separation Support Track-bed drainage Rail Sleeper Ballast Interlayer Water + Fine particles Sub-grade Mud pumping zone 25/36
Function of interlayer Separation Support Track-bed drainage Rail Sleeper Ballast Interlayer Water + Fine particles Sub-grade 26/36
Function of interlayer Separation Support Track-bed drainage Rail Sleeper Ballast Interlayer Sub-grade 27/36
Experimental devices Large-scale triaxial cell Soil specimen: D = 300 mm H = 600 mm Loading frequency: < 20 Hz 28/36
Experimental devices Large-scale infiltration column Soil specimen: D = 300 mm H = 600 mm Instrumentation: - 5 TDR for volumetric water content - 5 tensiometers for soil suction 29/36
Soils sampling Interlayer 30/36
Soils studied Percentage finer (%) 100 80 60 40 20 ITL 0 ITL 5 ITL 10 Sub-soil ITL -10 0 10-3 10-2 10-1 1 10 10 2 Grain size (mm) ITL 0 : Natural interlayer soil +5 % sub soil ITL 5 +10 % sub soil ITL 10-10 % sub soil ITL -10 31/36
Sample preparation in laboratory Soil dry density: - In situ: 2.40 Mg/m 3 - Laboratory: 2. 00 Mg/m 3 (compaction in layers) 32/36
Water retention property 100 Degree of saturation (%) 90 80 70 60 50 40 ITL 10 Fines 30 0.1 1 10 100 1000 Matric suction (kpa) 33/36
Hydraulic conductivity vs suction, fines, macropores Hydaulic conductivity (m/s) 10-4 ITL 10-5 10 - Saturated state Fines Fines - Saturated state ITL 10 10-6 10-7 10-8 10-9 10-10 10-11 van Genuchten's model 10-12 0.1 1 10 100 1000 Matric suction (kpa) 34/36
Permanant strain vs water content and fine fraction Permanent axial strain (%) 5 4 3 2 ITL -10 w4c ITL -10 w6c ITL -10 w12c Water content increase (a) 1 0 0 30 60 90 120 150 180 210 Permanent axial strain (%) 5 ITL -10 w4c 4 ITL 0 w4c 3 ITL 10 w4c 2 1 (a) Fine content increase 0 0 30 60 90 120 150 180 210 Permanent axial strain (%) Permanent axial strain (%) 5 ITL 4 0 w4c ITL 0 w6c 3 ITL 0 w12c (b) Water content increase 2 1 0 0 30 60 90 120 150 180 210 5 ITL 10 w4c 4 ITL 10 w6c 3 ITL 10 w12c 2 (c) Water content increase 1 0 0 30 60 90 120 150 180 210 q max (kpa) Permanent axial strain (%) Permanent axial strain (%) 5 ITL 4-10 w6c ITL 0 w6c 3 ITL 5 w6c ITL 2 10 w6c Fine content increase (b) 1 0 0 30 60 90 120 150 180 210 5 ITL -10 w12c 4 ITL 0 w12c 3 ITL 10 w12c 2 1 Fine content increase (c) 0 0 30 60 90 120 150 180 210 q max (kpa) 35/36
Concluding remarks Unsaturated soil mechanics is involved in both new and conventional lines New lines - Special sub-soils o o Loess: collapse upon wetting; liquefaction in saturated state, and the liquefaction resistance can be increased significantly with a slight decrease of water content Marl: swelling upon water infiltration; excavation or unloading is the origin of swelling; possibility to estimate the time needed to reach a new equilibrium Conventional lines Interlayer soils o o Governing role of fine fraction for the hydraulic conductivity of interlayer soils Effects of water content and fine content to be accounted for together when analyzing the mechanical behavior 36/36