Degree Name

Doctor of Philosophy


Department of Civil and Mining Engineering


This thesis is concerned with the behaviour of reinforced earth model walls under different conditions. The following investigations have been carried out as part of this thesis:

(a) Experimental investigation of the effect of backfill thrust on the vertical normal stress within the reinforced zone. The reinforced zone was 300mm high (H = 300); the length determined by the length of the reinforcement strips was 370mm (L = 370); and the width was 800mm (B = 800).

(b) Experimental investigation of the effect of the width of reinforcement strips on the vertical normal stress within the reinforced zone and the backfill.

(c) Experimental investigation of the influence of reinforcement on the shear strength of sand in direct shear using a shear box specially built for the purpose. The effect of submergence on the shear strength of reinforced sand was included as part of this investigation.

(d) Numerical analyses of reinforced earth model walls, using a finite element method and considering the reinforced zone as a composite material. The influence of the stiffness ratio between the reinforced zone and backfill was included as a part of the investigation.

In the experimental investigations the overall effect of backfill thrust was an increase in the vertical normal stress along a horizontal plane close to the base of the reinforced zone. This behaviour is a reflection of the contribution of two components of lateral pressure, namely, (a) The horizontal lateral stress component and, (b) The tangential or frictional component acting vertically downwards along the vertical interface between the reinforced zone and the backfill.

If the effect of the frictional force is separated from the overall effect, the influence of the horizontal component of lateral stress can be identified. This is found to be a decrease in vertical normal stress at the back of the wall (near the interface) and an increase in pressure in the middle region . This behaviour is partly but only qualitatively similar to that predicted by the trapezoidal theory since the observed increase at the front of the wall (near the facing) is not maximum but smaller than that in the middle region. The wall obviously behaves as a flexible structure and there is clear evidence of an arching phenomenon as well (arching in the length direction - L direction)

A series of model reinforced earth walls were built where the width of the reinforcement strips was varied. The length and number of strips were kept constant. Measurement of the vertical pressure close to the base, showed that the width of reinforcement affects the shape of the vertical pressure distribution along the reinforced zone. It also affects the vertical pressure across the wall and arching in the width direction (B direction) appears to influence the normal stress distribution.

Direct shear tests were performed using a purpose-built, 300mm square box. Several parameters were tested to investigate their effect on the shear strength of reinforced sand. The results show that the relationship between the strip width and the strength of the soil is non-linear. Moreover, there is an optimum width which results in a composite material (reinforced soil) with the highest strength. The influence of the type of reinforcement was tested and the efficiency of ribbed strip over smooth strip was quantified. The behaviour of reinforced sand under submerged condition was also studied. Under submergence, the efficiency of reinforcement is the same as in the dry condition.

Numerical analyses, using the finite element method, were performed to simulate the reinforced earth model wall. The parameters were related considering the reinforced zone to be a composite material. The effect of the 'stiffness' of the reinforced zone on the vertical normal stress, horizontal normal stress, shear stress and deformations were studied. Comparisons were made between the experimental results and the results from finite element analyses. In some respects the results were similar but in other respects, the results differed significantly. The importance of boundary conditions and of the relative stiffness between reinforced zone and backfill was found to be significant.

Considering reinforced earth as an anisotropic material, the 'elastic' (or stiffness) parameters were evaluated on the basis of an available theory. These parameters were used in additional finite element analyses. Some of the analyses were performed considering full-scale model walls.

In the final chapter of this thesis the important results of all the investigations are summarised and the possible implications of these findings for design are discussed.