1635
Effectiveness of In-soil Seismic Isolation taking into account of Soil-Structure
Interaction
Efficacité d’ Isolement sismique dans le Sol tenant compte de l’interaction du Sol avec la Structure
Tsatsis A.K., Anastasopoulos I.C., Gelagoti F.L., Kourkoulis R.S.
Laboratory of Soil Mechanics, National Technical University of Athens
ABSTRACT: In the present study an innovative seismic isolation method is proposed that introduces a sliding surface within the
foundation soil.The sliding surface comprises of two synthetic liner layers at contact with each other creating an interface of small
friction that enfolds the foundation soil. The effectiveness of the isolation system is explored as a function of the earthquake intensity
accounting for soil-structure-interaction phenomena. It is shown that the proposed system serves as a fuse mechanism within the soil
and substantially reduces the acceleration transmitted onto the structure. The isolated structure may be subjected to increased
differential lateral displacement, due to sliding at the isolation interface – something that has to be considered in the design.
RÉSUMÉ : Dans cette étude une méthode innovante d'isolement sismique est proposé, composé d’une surface de glissement dans le
sol. La surface de glissement se compose de deux couches de revêtement synthétique, caractérisé d’une résistance réduite. L'efficacité
du système proposé est explorée en fonction de l'intensité sismique. Il est démontré que le système proposé acte comme un
mécanisme fusible dans le sol, réduisant considérablement l'accélération transmise sur la structure. La structure isolée peut être
soumis à un déplacement différentiel latérale, en raison de glissement à l'interface d'isolation – quelque chose que doit être pris en
compte dans le désign.
KEYWORDS: soil-structure interaction ; in-soil isolation ; synthetic liner
1 INTRODUCTION
In the present study an isolation method is proposed that
introduces a sliding surface within the foundation soil. The
sliding surface comprises two layers of a smooth synthetic liner
in contact with each other. Yegian et al. (2004a) were the first to
propose the application of synthetic liners just below the
foundation with the intention to introduce, an interface of small
friction coefficient upon witch the structure would slide as a
rigid block. To determine the properties of the interface,
shaking table tests were conducted, concluding that the static
and dynamic friction coefficient is of the order of 0.10 and 0.07,
respectively. The same researchers (Yegian et al., 2004b)
investigated the idea of such a sliding surface within the
foundation soil, forming an isolated soil prism of ellipsoidal
shape.
Based on this idea, Georgarakos & Gazetas (2006)
parametrically investigated the effect of the sliding surface
geometry on the seismic response. Several geometries were
investigated, ranging from cyndrical, to basin-shaped,
trapezoidal, and trapezoidal with wedges. The latter was found
to be the optimum solution, providing the restoring force of the
cylindrical surface, while being significantly easier to construct.
The systems functionality is based on the ability of the isolated
soil to slide on the synthetic liner, while the two wedges offer
the necessary restoring force through their weight. The response
of this system can be seen as mechanically analogous to a mass
sliding on a horizontal surface, being restrained by two springs
that work only when compressed.
The investigated system is schematically illustrated in
Figure 1. The geometry of the isolation system is trapezoidal,
with isolated wedges on the two sides. The synthetic liners are
placed at a depth H = 2 m under the surface. The slope of the
excavation trench is assumed equal to 1:1 – a realistic
assumption for relatively competent soi. The isolated
embankment comprises a dense gravel layer. The latter is
modeled with a nonlinear constitutive model, with a Mohr-
Coulomb failure criterion and non-associative flow rule. A
rather large Young’s modulus E = 500 MPa is assumed, while
the friction and dilation angles are equal to
φ
= 48
o
and ψ = 15
o
,
respectively. The two wedges are filled with pumice, a
lightweight material of density
ρ
= 1 Mg/m
3
and relatively small
stiffness
E
= 10 MPa, in order to impose the minimum possible
resistance to the sliding motion of the embankment.
The superstructure, an idealized bridge pier (for simplicity),
is placed on top of the isolated embankment. The bridge pier is
designed according to EC8, assuming a design acceleration
a
gr
= 0.24g and behavior factor
q
= 2. Having an elastic natural
period
T
= 0.48 sec, the design spectral acceleration is equal to
SA = 0.3g. In order to undertake the resulting design bending
moment
M
D
= 43 MNm, a longitudinal reinforcement of
100
Φ
32 is required, combined with transversal reinforcement of
Φ
32/8cm.
1
1
1
1
h
= 12 m
d
= 3 m
B
= 11 m
στρώσεις γεωμεμβράνης
κλίση
πρανών
2 m
E, ρ
E, ρ
19 m
Figure 1. Schematic illustration of the in-soil isolation system under
consideration.
