0% found this document useful (0 votes)
14 views5 pages

Ferreira 2011

Uploaded by

MypatiVamsi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
14 views5 pages

Ferreira 2011

Uploaded by

MypatiVamsi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Technical Note

Impact of Bituminous Subballast on Railroad Track


Deformation Considering Atmospheric Actions
T. Moço Ferreira1; P. Fonseca Teixeira, Ph.D.2; and R. Cardoso, Ph.D.3

Abstract: High-speed ballasted track design standards usually require the use of sand and gravel layers as subballast to fulfill an accurate
Downloaded from ascelibrary.org by North Carolina State University on 05/14/15. Copyright ASCE. For personal use only; all rights reserved.

protection of the formation layers, not only against traffic loads, but also against the effects of weather. Seasonal changes in soil water content,
or suction changes, are responsible for cyclic volumetric strains on railroad trackbed layers, thus on the infrastructure. Being almost com-
pletely water-resistant when compared with granular-only layers, bituminous subballast offers a higher protection of the subgrade,
consequently improving its behavior along the infrastructure life cycle. This question is investigated through the comparison of the
performance of the track formation against atmospheric actions, taking into consideration the unsaturated state of the geomaterials. The
method adopted consists of modeling the vertical displacements of both bituminous and granular subballast designs through a finite-element
coupled thermo-hydro-mechanical (THM) analysis. The comparison of the two design solutions confirms that the adoption of a bituminous
subballast layer might allow important reductions in seasonal vertical displacements. DOI: 10.1061/(ASCE)GT.1943-5606.0000435.
© 2011 American Society of Civil Engineers.
CE Database subject headings: Railroad engineering; Railroad ballast; Railroad tracks; Deformation; Unsaturated soils.
Author keywords: Railroad engineering; Ballasted tracks; Bituminous subballast; Unsaturated soils.

Introduction Subballast layer functions are, among others, to protect


subgrade from atmospheric actions (Selig and Waters 1994).
Railroad track geometry deterioration attributed to both ballast and Usually unbound granular materials are adopted in most European
subgrade differential settlement is usually responsible for the most high-speed lines. However, some innovative solutions, such as the
relevant part of track maintenance costs. Recent studies have shown adoption of bituminous layers, may bring important benefits on
that the contribution of the subgrade settlement can play an impor- subgrade protection and railroad track performance (Teixeira et al.
tant part in this degradation phenomenon, even in the case of new 2006). Being almost completely impervious, the use of bituminous
high-speed lines (López Pita et al. 2007). subballast layers is expected to allow an optimal drainage of super-
In addition to the effect of cyclic loads attributed to passing ficial rainwater while contributing to minimize fluctuations of soil
trains, subgrade differential settlement might also occur because moisture content (Rose et al. 2000).
of atmospheric actions. Indeed, the existence of a great variety The motivation of this paper is to confirm the reduction of
of soils composing a railroad trackbed is responsible for important seasonal deformations of railroad trackbeds achieved by adopting
differential vertical displacements along the longitudinal develop- a bituminous subballast layer. The performance of a bituminous
ment. This evidence is related with the fact that different soils com- and a granular subballast solution is compared in terms of seasonal
posing a railroad trackbed shrink or swell in distinct ways, leading deformations. A thermo-hydro-mechanical (THM)–coupled analy-
to different vertical displacements between successive transversal sis is performed to model the interaction between unsaturated rail-
cross sections. These differential displacements contribute to a road soil layers and atmosphere. Previous integrated analyses of
gradual increase of the longitudinal defects associated with the road pavement performance with hydro-mechanical-coupled prob-
railroad track geometry. In the case of high-speed lines, this issue lems such as those presented in this study are scarce (Wallace 1977;
can be particularly critical since settlement tolerances are very Pufahl et al. 1990; McEnroe 1994; Alonso 1998). The adoption of
restrictive (on the order of millimeters). concepts from the mechanics of partially saturated soils in the im-
provement of high-speed railway track design has not been ex-
1
Ph.D. Student, CESUR, Dept. of Civil Engineering, Instituto Superior plored so far. This paper is expected to stimulate the discussion
Técnico (IST), Technical Univ. of Lisbon, Av. Rovisco Pais 1049-001 on the potential of such a type of analysis.
Lisboa, Portugal. E-mail: [email protected]
2
Assistant Professor, CESUR, Dept. of Civil Engineering,
Instituto Superior Técnico (IST), Technical Univ. of Lisbon, Av. Rovisco Numerical Model
Pais 1049-001 Lisboa, Portugal (corresponding author). E-mail: pft@ist
.utl.pt Atmospheric actions such as rainfall, changes in relative humidity,
3
Assistant Professor, ICIST, Dept. of Civil Engineering, Instituto and temperature are responsible for infiltration/evaporation
Superior Técnico (IST), Technical Univ. of Lisbon, Av. Rovisco Pais phenomena resulting in changes in the water content of the soil.
1049-001 Lisboa, Portugal. E-mail: [email protected]
For railroad trackbeds, soil layers are usually unsaturated and
Note. This manuscript was submitted on November 25, 2009; approved
on August 24, 2010; published online on September 2, 2010. Discussion variations in the saturation degree correspond to suction changes.
period open until August 1, 2011; separate discussions must be submitted Volume dependence on suction can be quantified only if an unsatu-
for individual papers. This paper is part of the Journal of Geotechnical and rated soil constitutive model is adopted. On wetting (suction
Geoenvironmental Engineering, Vol. 137, No. 3, March 1, 2011. ©ASCE, decrease), the associated deformations are swelling (volume incre-
ISSN 1090-0241/2011/3-288–292/$25.00. ment) under low confinement or collapse (volume decrease) under

288 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / MARCH 2011

J. Geotech. Geoenviron. Eng. 2011.137:288-292.


high confinement. Drying processes are related to shrinkage σij ¼ σij  Pg δ ij ð1Þ
deformations (volume decrease).
The deformation process of both granular and bituminous sub- ψij ¼ ψδ ij ¼ ðPg  Pl Þδ ij ð2Þ
ballast solutions is numerically analyzed using CODE_BRIGHT.
This finite-element (FE) code performs a THM analysis consider- The cross sections designed for this study simulate a 5 m high
ing soil as an unsaturated deformable porous media. CODE_- embankment and incorporate typical geometry and thicknesses of
BRIGHT solves the equations of conservation of solid mass, trackbed layers. To reach an equivalent structural behavior of a
balance of water and air masses, energy balance and overall granular subballast layer with 30 cm of thickness, the bituminous
mechanical equilibrium. Detailed considerations on the definition subballast layer is design with 12 cm (Teixeira et al. 2006).
of the balance equations are described in Olivella et al. (1996). Data for model calibration considering the thermal, hydraulic,
The code presents the advantage of working with “net” mean and mechanical problems to be solved simultaneously are difficult
stress σ [Eq. (1)] and suction ψ [Eq. (2)] as main state variables to find in the literature. For this reason, the present study considers
Downloaded from ascelibrary.org by North Carolina State University on 05/14/15. Copyright ASCE. For personal use only; all rights reserved.

instead of working with effective stresses. This approach simplifies synthetic materials in which properties are inspired in those
calculation as the definition of effective stresses for unsaturated described by Alonso (1998) in the analysis of roadway subgrades.
materials is an actual research topic (e.g. Gens et al. 2008). The main parameters for the constitutive laws necessary for the
“Net” mean stress is defined as the excess of total stress over THM analysis are in Table 1. For the particular case of the sub-
air pressure given by Eq. (1), where Pg = gas pressure (assumed grade, the properties adopted correspond to those of clayey soils
to be equal to the atmospheric pressure) and Pl = liquid moderately expansive for which suction changes may result in
pressure. Suction is defined as the difference between gas and volumetric deformations.
liquid pressure as shown by Eq. (2). Considering fully saturated The hydraulic constitutive equations controlling the balance of
conditions, ψ ¼ 0, Pl ¼ Pg , and net stress becomes Terzaghi’s water in the liquid and gas phases are described by Darcy’s law
effective stress. extended for unsaturated states in the calculation of conductive

Table 1. Material Models and Parameters (data from Alonso 1998)


Granular Bituminous Formation
Property Model Ballast subballast subballast and subgrade Legend
Water    1 λ λ ¼ 0:5 λ ¼ 0:55 λ ¼ 0:50 λ ¼ 0:55 Sls : maximum saturation
Pg Pl 1λ
retention Se ¼ SSlsl Srl
Srl ¼ 1 þ P0 P0 ¼ 0:01 MPa P0 ¼ 0:15 MPa P0 ¼ 0:35 MPa P0 ¼ 0:15 MPa Srl : residual saturation
curve Srl ¼ 0:0001 Srl ¼ 0:05 Srl ¼ 0:01 Srl ¼ 0:05 P0 : pressure for a
measured T
Sls ¼ 0:98 Sls ¼ 0:80 Sls ¼ 0:80 Sls ¼ 0:80 λ: shape function
Liquid flow ql ¼ ðkkrl =μl Þð∇Pl  ρl gÞ k: intrinsic
permeability (m2 )
λ ¼ 0:55 A ¼ 1:0; α ¼ 3:0 λ ¼ 0:55 μl : liquid viscosity
pffiffiffiffiffi 1
k rl ¼ Se ½1  ð1  Sλe Þλ 2 or, k ¼ 1:0 × 1012 k ¼ 0:5 × 1015 k ¼ 0:1 × 1019 k ¼ 0:5 × 1015 ρl : liquid density
Srl ¼ 0:15 Srl ¼ 0:01 Srl ¼ 0:15 ql : liquid advective flux
Sls ¼ 0:98 Sls ¼ 0:98 Sls ¼ 0:98 g: gravity
k rl ¼ AðSe Þα ; λ and α: power
A = constant
Air flow krg ¼ 1  krl krg and krl : coef.
for liquid
and gas relative
permeability
Conductive ic ¼ λ∇T λ ¼ 2 (W=mK) λ ¼ 2 (W=mK) λ ¼ 2 (W=mK) λ ¼ 2 (W=mK) ic : conductive heat flux
heat flow λ: thermal conductivity
Diffusion iwg ¼ ðnρg ð1  Sl ÞDwm IÞ∇ωwg iwg : nonadvective flux
of vapor
t ¼ 2:3; τ ¼ 1 t ¼ 2:3; τ ¼ 0:4 t ¼ 2:3; τ ¼ 1 t ¼ 2:3; τ ¼ 1 n: porosity
Dwm ¼ τ Dw0 ðT t =Pg Þ; I: identity matrix
Sg ¼ 1  Sl Dw0 ¼ 5:9 × 106 Dw0 ¼ 5:9 × 106 Dw0 ¼ 5:9 × 106 Dw0 ¼ 5:9 × 106 τ : tortuosity coefficient
Dw0 and t: parameters
Deformation Linear elastic strain E ¼ 150 MPa E ¼ 50 MPa E ¼ 3; 000 MPa E ¼ 50 MPa υ: Poisson coefficient
attributed to suction and υ ¼ 0:33 υ ¼ 0:33 υ ¼ 0:33 υ ¼ 0:33 E: Young modulus
temperature changes:
Δεv ¼ 3aψ ΔðPl  Pg Þ aψ ¼ 0:00 aψ ¼ 0:0025 aψ ¼ 0:00 aψ ¼ 0:0025 aψ : coef. suction changes
þ3bT ΔT bT ¼ 0:55 × 105 bT ¼ 0:55 × 105 bT ¼ 0:55 × 105 bT ¼ 0:55 × 105 bT : coef. temperature
changes

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / MARCH 2011 / 289

J. Geotech. Geoenviron. Eng. 2011.137:288-292.


flux, and Fick’s law in the calculation of advective flux. Darcy’s Climate data considered in the calculations are rain (water flow),
law is assumed to be valid for the materials under the hypothesis relative humidity (water in gas phase, corresponding to a given
of continuity of the liquid phase and because of the reduced veloc- gas density, ρvap ) and temperature. They are applied as boundary
ity (Olivella et al. 1996). Intrinsic permeability, k, matches the grain conditions in the soil surfaces in contact with the atmosphere. The
size distribution expected for the materials. Its dependence on boundary conditions, geometries of the different trackbed layers
porosity φ (independent from suction) is calculated by Kozeny’s and the considered FE mesh are in Fig. 1. Leakage is allowed, ac-
model. For the definition of the water retention curve (relationship tivated whenever liquid pressure generated from rain infiltration
between suction and water content/saturation degree) it is used reaches atmospheric pressure.
the expression proposed by Van Genuchten (1980). Water and Water vapor flow is controlled by a linear law [Eq. (3)], which
air permeability coefficients (k rl and krg , respectively) depend specifies that the evaporative flux jwg is proportional to the differ-
on saturation degree (Table 1). They are compatible with the veloc- ence between vapor density at the boundary of the embankment and
ity of pore pressure dissipation generated by seasonal changes in the water vapor density on the atmosphere.
Downloaded from ascelibrary.org by North Carolina State University on 05/14/15. Copyright ASCE. For personal use only; all rights reserved.

water content.
Table 1 also presents the equation used in the calculation of the jwg ¼ β g ðρ0vap  ρvap Þ ð3Þ
displacements. Elastic analysis is adopted and is justified because
small deformations are expected (order of millimeters) and yielding The initial water content considered, or initial suction, corre-
caused by the increment of vertical stress is not expected (Ferreira sponds to an equilibrium situation consistent with the water level
2007). Accordingly with the construction procedure of railway located at the lowest boundary of the domain. The value considered
tracks in which maximum vertical stress occurs at the construction (0.25 MPa) is similar to the values adopted in other works (Olivella
stage, further loading will not irreversibly change the soil state. It is et al. 1996; Alonso 1998; Mroueh and Shahrour 2002).
also assumed that maximum suction in the subgrade layers is
inferior to suction above which shrinkage is irreversible (Alonso
Analysis of the Results
et al. 1990).
Finally, the cross sections designed for this study simulate a In the field of railway infrastructures, the differential displacements
5 m high embankment and incorporate typical geometry and along a railway track are of great importance because of their con-
thicknesses of trackbed layers. To reach an equivalent structural sequences in the development of geometrical longitudinal defects.
behavior of a granular subballast layer with 30 cm of thickness, In the present study, the comparison between the bituminous and
the bituminous subballast layer is designed with 12 cm (Teixeira granular subballast solutions is made in terms of the evolution of
et al. 2006). vertical displacements calculated in several control points identified
in Fig. 2.
The atmospheric actions modeled in this analysis are connected
Consideration of Atmospheric Actions with infiltration/evaporation phenomena which are responsible for
water flow and variations in the water pressure. As shown by
The performance of granular and bituminous solutions was simu- Eq. (2), the unsaturated constitutive formulation used to model
lated considering a period of 5 years (1,800 days). This period the deformational behavior of both design solutions uses these
allows the comparison of both solutions in terms of their deforma- liquid pressure variations to compute suction (main state variable).
tion behavior during the early stages after construction. Wetting processes associated with suction decreases are responsible
A Mediterranean climate (Tarragona, Spain) was considered for swelling or collapse deformations; and adversely, shrinkage
in the analysis. It is characterized by warm and dry summers deformations occur when soils are drying. The amplitude of
and moderate winters (T av ¼ 18°C), the relative humidity has an deformations is proportional to the amplitude of suction changes
irregular distribution (RHav ¼ 71:3%) and high temperatures in experienced by the different soil layers.
summer enhance evaporation. For the purposes of a multiyear The impervious properties of the bituminous subballast have
simulation, the average year is repeated five times. a major impact in the zones immediately below this layer, which

Fig. 1. (a) Boundary flow rate conditions; (b) partial mesh for granular subballast design

290 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / MARCH 2011

J. Geotech. Geoenviron. Eng. 2011.137:288-292.


Downloaded from ascelibrary.org by North Carolina State University on 05/14/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 2. Evolution of vertical displacements—five years: (a) granular subballast design; (b) bituminous subballast design

maintains low saturation levels and small suction variations during (in contact with atmosphere). Furthermore, the inner points of the
the year. The analysis of the saturation degree evolution evidences embankment (Points C and F) have low amplitudes because they
that the bituminous subballast works as barrier against water infil- are located at depths not reached by infiltrated water. Adding to
tration (Ferreira et al. 2009). As a consequence of the protection this, the low moisture variations observed in the bituminous
performed by the bituminous material, the seasonal deformations subballast solution may have a positive impact in the deterioration
associated with this solution are significantly lower than the ones process of the subgrade.
from the granular design. This observation is shown in The amplitudes of the vertical displacement obtained for the two
Fig. 2(a) for the bituminous, and in Fig. 2(b) for the granular sol- solutions analyzed confirm the interest of using THM analysis to
utions, whereas the evolution of the vertical displacements along assess the consequences of different design solutions.
the five years in some control points is presented. Weather season-
ality is clearly reflected in the cyclic deformations obtained in all
Conclusions
points (A, B, C, D, E, and F). The cumulative values are explained
by water infiltration with consequent increment of the saturation The results of the performed thermo-hydro-mechanical analysis
degree (suction decrease). showed that bituminous subballast can allow a significant reduction
Point A can be taken as the most representative to study track in the seasonal amplitude of vertical displacements (almost 60% for
deformation. In this point, the amplitude of vertical displacements the case studied). In practice, this effect can have an impact in
(the difference between the maximum and the minimum for each reducing the longitudinal defects of a railroad high-speed track,
year) is practically constant for the 5 years, and as expected, is contributing to decrease the number of maintenance operations
lower for the bituminous solution (7 mm for the bituminous and and consequently, its costs.
12 mm for the granular). Presently, an ongoing research project aims to validate this par-
The analysis of the other points covering different depths and ticular advantage of bituminous subballast by monitoring moisture
distances from the slope confirms that the positive effect of the and deformations in an experimental test site with both granular and
bituminous layer increases with the distance from the slope surface bituminous subballast sections.

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / MARCH 2011 / 291

J. Geotech. Geoenviron. Eng. 2011.137:288-292.


Acknowledgments López, P., Teixeira, P. F., Casas, C., Ubalde, L., and Robusté, F. (2007).
“Evolution of track geometric quality in high-speed lines: Ten years
This research is being supported by the Portuguese National experience of the Madrid–Seville line.” Proc. Inst. Mech. Eng., Part
Science Foundation (FCT) since 2008 (Project Reference PTDC/ F, 221(F2), 147–155.
ECM/70571/2006). The writers would like to thank the support McEnroe, B. M. (1994). “Drainability of granular bases for highway
and collaboration of Dr. S. Olivella with CODE_BRIGHT. pavements.” Transp. Res. Rec., 1434, 23–28.
Mroueh, H., and Shahrour, I. (2002). “Three-dimensional finite element
analysis of the interaction between tunneling and pile foundations.”
References Int. J. Numer. Anal. Meth. Geomech., 26(3), 217–230.
Olivella, S., Gens, A., Carrera, J., and Alonso, E. E. (1996). “Numerical
Alonso, E. E. (1998).“Suction and moisture regimes in roadway bases formulation for simulator (CODE_BRIGHT) for coupled analysis of
and subgrades.” Simposio Internacional: Drenaje Interno de Firmes saline media.” Engng. Comput., 13(7), 87–112.
y Explanadas [Proc. Int. Symp. on Subdrainage in Roadway Pavements Pufahl, D. E., Lytton, R. L., and Liang, H. S. (1990). “Integrated computer
Downloaded from ascelibrary.org by North Carolina State University on 05/14/15. Copyright ASCE. For personal use only; all rights reserved.

and Subgrades], Granada, Spain, 57–104 model to estimate moisture and temperature effects beneath pave-
Alonso, E. E., Gens, A., and Josa, A. (1990). “A constitutive model for ments.” Transp. Res. Rec., 1286, 259–269.
partially saturated soils.” Géotechnique, 40(3), 405–430. Rose, J. G., Brown, E. R., and Osborne, M. L. (2000). “Asphalt
Ferreira, T. M. (2007) “Influence due to the incorporation of a bituminous trackbed technology development: The first 20 years.” Transp. Res.
subballast layer in terms of deformations on railway trackbeds.” Thesis Rec., 1713, 1–9.
for Integrated Master Degree in Civil Engineering, Instituto Superior Selig, E. T., and Waters, J. M. (1994). Track Geotechnology and
Técnico, Technical Univ. of Lisbon. Substructure Management, Thomas Telford Services, London.
Ferreira, T. M., Teixeira, P. F., and Cardoso, R. (2009). “Effects of Teixeira, P. F., López-Pita, A., Casas-Esplugas, C., Bachiller, A., and
incorporating a bituminous subballast layer on the deformation Robusté, F. (2006). “Improvements in high-speed ballasted track
of railway trackbeds.” Proc. 8th Int. Conf. on the Bearing design: Benefits of bituminous subballast layers.” Transp. Res. Rec.,
Capacity of Roads, Railways and Airfields, Vol. 2, Champaign, IL, 1943, 43–49.
1291–1301. Van Genuchten, M. T. (1980). “A closed-form equation for predicting
Gens, A., Guimarães, L., Sánchez, M., and Sheng, D. C. (2008). “Chapter the hydraulic conductivity of unsaturated soils.” Soil Sci. Soc.
3: Developments in modelling the generalised behaviour of unsaturated Am. J., 44(5), 892–898.
soils.” Unsaturated soils—Advances in geo-engineering, D. G. Toll, Wallace, K. B. (1977). “Moisture transients at the pavement edge:
C. E. Augarde, D. Gallipoli, and S. J. Wheeler, eds., Taylor & Francis, analytical studies of the influence of materials and shear design.”
Abingdon, UK. Géotechnique, 27(4), 497–516.

292 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / MARCH 2011

J. Geotech. Geoenviron. Eng. 2011.137:288-292.

You might also like