Rammed earth: experimental and numerical assessment as a construction material and its suitability for a seismic environment

Abstract

Rammed earth is one of the ancient construction techniques in which the moist soil is rammed in layers by manual, electrical or pneumatic rammers. The rammed earth has recently gained popularity because of its environment friendly behaviour and sustainability. Many historical rammed earth buildings around the world show its suitability but its behaviour in seismic regions is poorly investigated. The main aim of the thesis is to analyse the seismic performance of rammed earth walls in active seismic regions. Before investigating wall behaviour, it is important to acquire physical and mechanical properties of rammed earth material. As rammed earth is layered construction, the properties of both the material and the interlayers are required. In previous studies interlayer properties are usually assumed because of lack of experimental data. This study evaluates the physical and mechanical properties of rammed earth material and interlayer and uses these to analyse the seismic behaviour of rammed earth walls in three phases.

In the first phase, the physical properties were determined using geotechnical tests such as particle size distribution, consistency limits, optimum moisture content and maximum dry density. Unconfined compression test was undertaken under different environmental conditions to determine some of the mechanical properties such as unconfined compressive strength, Young's modulus and Poisson’s ratio along with the suction of rammed earth material through suction test. It has been found that the suction has major influence on the mechanical performance of rammed earth material. In addition, the unconfined compression test was simulated using concrete damaged plasticity for developing constitutive model for rammed earth material. The close approximation between experimental and numerical results showed that this model is reliable. The findings of this phase can be a useful aid to understand the effect of varying environment conditions on mechanical performance of rammed earth.

In the second phase, the shear parameters of rammed earth interlayer such as cohesion, frictional angle and shear stiffness were investigated using a recently developed test named shear wedge test. The results obtained were consistent with the previous results for unstabilised rammed earth. The simulation of shear wedge test was also carried out using surface based cohesive behaviour. This is the first time that the simulation of this test has ever been conducted. The results were in appropriate range of experimental results which showed the suitability of this model for interlayers. The use of the shear wedge test over existing shear testing approaches show that it is a robust way of finding shear parameters of interlayers as it is simpler and more cost effective.

In the third phase, an extensive numerical assessment was carried out using time history analysis with the properties obtained from the testing of specimens for the evaluation of the in-plane seismic behaviour of rammed earth walls for a design earthquake having peak ground acceleration of 0.32g for the Quetta region of Pakistan. The seismic behaviour of walls was evaluated with monolithic and layered approaches using solid wall and wall with openings. It was found that interlayers and openings influence the seismic response of the walls which highlights the importance of the inclusion of interlayers and openings in the simulation. In case of solid wall, damage was predominant in models with interlayers while it was prominent at the bottom and top of the wall when it was modelled as monolithic. In case of wall with openings, the damage was not only seen between the openings but also typical diagonal bands around the openings were seen with monolithic and layered approaches. The extended analysis with different opening sizes showed that increasing the opening sizes did not have pernicious influence on the seismic performance of the rammed earth wall.