Australasian Science Magazine

Example of a modern rammed earth house in Western Australia. Photo: Stephen Dobson

By Daniela Ciancio

Building and maintaining houses in remote Aboriginal communities is difficult and expensive, but engineering improvements to rammed earth constructions offer a viable alternative.

Building a house in any of Australia’s remote communities can be a challenging project due to the isolation of the site and hence the difficulties and costs involved in the transportation of the construction materials and labour force. Sadly, the majority of Australia’s remote communities are Aboriginal, and are less prepared to sustain the expensive costs of a dwelling.

A University of Western Australia (UWA) research team of mostly engineers has recently discovered that the ancient building technique of rammed earth might be the solution for an affordable and sustainable housing program in remote communities.

The UWA team is currently working on a project that aims to further improve this building technique and promote its use. The driving force of the project is the discovery that a rammed earth house in a remote area built with earth collected from the same site of the house has the potentialities to be far cheaper than other construction techniques. But this is not everything! The same study discovered other environmental and social benefits in the use of rammed earth.

Housing in Remote Areas
In general, the cost of a house is mainly determined by the cost of the construction materials and the cost of the labour force working on the construction site. In a remote community, it is often the case that neither construction materials nor skilled labour are readily available on-site. For this reason, two more expenses must be added to the list: the transportation of the materials from the closest supply centres to the remote community, and the accommodation of the skilled labour force brought on-site.

Hence the overall construction cost of a house in a remote area is always higher than the cost of the same house built in a metropolitan zone. For this and other reasons, in some remote communities Aboriginal families that cannot afford the expenses of building their own house find themselves in an endless waiting list to obtain a house from the government.

But the construction cost is not the only issue with housing in remote zones. The most preferred construction technique in remote areas is the steel-framed house. Steel panels and frames are much lighter to transport than bricks or timber boards. As a result, a truck can carry the materials to build a small steel-framed house in a single journey, whereas it would need more trips to carry the material if the same house is made from bricks or timber.

Unfortunately, in tropical climates a steel framed house is not able to create a comfortable indoor environment without the installation of an air conditioning system. If this breaks down the occupants need to find an air-conditioning technician who is willing to make the long journey to reach the house and fix the problem. While this is not impossible, it is unlikely to happen and very expensive.

In 2006, 30% of the total permanent dwellings (mainly steel-framed houses) in indigenous communities required major repairs or replacement. According to the Australian Bureau of Statistics, in that year $37.4 million was spent on the repair and maintenance of indigenous housing. Likewise the Sydney Morning Herald reported in 2009 that on another occasion the Australian government spent $80 million to inspect or fix 2900 houses in the Northern Territory alone.

It is clear that not only construction but also maintenance costs must be considered when assessing the cost-effectiveness of any housing scheme.

An Affordable and Sustainable Alternative Material
Rammed earth is an historic form of building where moist soil is compacted in layers inside temporary formwork. Once the soil has been adequately compacted, the formwork is removed, leaving the finished wall to dry out.

Traditional rammed earth mixes contain a small amount of clay (up to 30%) and lime or bitumen as bonding agents. Modern “stabilised” rammed earth uses cement to increase its strength and durability.

Sustainable, cost-effective and durable houses made of rammed earth can overcome the expense of housing in remote areas. Soil can be sourced on-site at zero or almost zero cost, so the transportation cost of the construction materials is significantly reduced. Furthermore, rammed earth walls do not require painting or other wall treatments, so there is little ongoing maintenance costs.

In 1933, as part of the National Industrial Recovery Act in the USA, a total of seven rammed earth houses were built in Gardendale, Alabama. Architect and engineer Thomas Hibben successfully taught unskilled labourers to build a rammed earth house. Fourteen men needed 5 weeks to build the walls for the first house but only 5 days to build the last of the seven. The original houses are still occupied today.

Similarly, in June 1997 the Arrillhjere house project in the west of Alice Springs was successfully completed. Unskilled Aboriginal labourers were employed to build the rammed earth foundation and the mud brick walls. They gained knowledge of the appropriate technologies and building and then took those experiences back to their own communities.

The bulk of rammed earth construction is very straightforward. It is necessary to have only one experienced rammed earth contractor on-site during construction when sufficient (even semi-skilled) labour is available. As a result, not only can local jobs be created, but the overall cost of the construction may be reduced by eliminating the need for expensive accommodation for labour brought from outside the community.

The economic and social benefits of rammed earth are not the only attractive features of this material. Earthen techniques in general and rammed earth in particular have environmental and sustainable benefits. One study conducted in France found that the use of locally sourced materials for rammed earth construction had a lower environmental impact compared with the use of construction materials sourced far away and transported to the building site. The energy consumed in transportation can be reduced by 85% when comparing a rammed earth to a standard concrete house.

In another project in India, the use of soil and cement to create unfired masonry blocks resulted in a 62% reduction in energy compared with a reinforced concrete framed structure and a 45% reduction when compared with burnt clay brick masonry and reinforced concrete solid slab construction.

But if rammed earth is an affordable construction material, why has it not been extensively used so far?

Obstacles to Overcome
Although increasing environmental awareness is motivating interest in rammed earth structures, rammed earth currently accounts for only a small proportion of the new building market in Australia and the rest of the world. Engineers, architects and builders are reluctant to use this construction technique because the engineering knowledge of the structural and material properties of rammed earth lags significantly behind that of more common building materials, such as concrete, masonry and steel. Although scientific methods have recently been applied to the study of rammed earth, the knowledge base of this material remains founded on rules of thumb and empirical assumptions.

Soil is a mixture of particles of various sizes and natures, ranging from tiny clay particles to silt, sand and finally gravel. The percentage breakdown of each component in the mix is called the particle size distribution. For instance, a soil can be composed by 20% of clay, 0% silt, 30% sand and 50% of gravel.

The particle size distribution indicates whether a certain soil can be used as rammed earth. In general, a suitable soil should have all sizes of material present, from the tiniest clay particle to the biggest gravel particle. This characteristic will facilitate the compaction of the soil when it is compressed by a pneumatic hammer.

High percentages of clay in the soil mix are unfavourable. The higher the clay content, the greater the amount of water trapped between clay particles. As the rammed earth wall dries out, the soil matrix will contract and cracks will develop. On the other hand, a minimum amount of clay is needed to act as a bonding agent.

It is easy to understand that soils can have infinite combinations of particle size distributions, and that the above-mentioned recommendations provide only qualitative descriptions of the rammed earth structure that can be obtained from a certain soil. For this reason, one of the still-unanswered questions is: what soil is appropriate to be used as rammed earth and what strength and durability will characterise that final product?

The assessment of factors affecting the strength of rammed earth still requires further investigation. What is clear is that several factors affect the strength of rammed earth, the first being the type of soil, with each particle size distribution corresponding to different mechanical characteristics and hence different strength.

The second factor is the amount of water added to the mix. Sufficient water must be added to lubricate the particles and to help the compaction process. The optimal water content will result in the maximum density of the mix during compaction. A mixture that is too dry or too wet will not compact properly as it will not be dense enough, and hence not strong enough.

Another factor is the energy used in the compaction process. Too little energy will not achieve the maximum density, but too much energy will crush the gravel particles, changing the particle size distribution and hence the properties of the soil.

Sometimes, cement is used in small amounts to improve the performance of rammed earth. It is added dry to the soil mix before compaction. Historical rammed earth structures containing no cement in the soil mix are still standing around the world, including some parts of the Great Wall of China – an example that good soil and an accurate ramming method can result in longevity to the structure.

While soil alone can be strong and durable, cement is used these days as an additive because building codes and construction rules on-site must adhere to quality control procedures. In short, all of the materials used to build a house must be characterised by accurate indices of strength and durability. In most cases, not only does the addition of cement improve the strength and durability of rammed earth, but it also makes it easier to standardise these qualities.

One of the most common myths about rammed earth is that a wall might be washed away by the rain. For rammed earth walls with no cement added to the mix, a large eave is usually provided to protect the surface of the wall from the rain and to avoid the formation of shrinkage cracks. While this protects the wall against the rain, it does not protect it from the erosion of the wind, so some maintenance is required to extend the life of the wall. When cement is added, the durability indices (capacity to be impermeable to water, to be resistant to erosion, to not shrink) significantly improve.

Working Towards a Solution
In 2008 I founded a rammed earth research group at UWA. More recently the group has collaborated with the Western Australian Department of Housing, through which it has gained new members with expertise in housing thermal assessment, Aboriginal training, project management and finance.

At the moment we are working on different research fronts. Some, but not all, of them are as follows:

• When adding cement to the rammed earth mix we increase its strength and durability properties, but we also increase the environmental impact of the final product. We are trying to find out the right amount of cement so that the gain in terms of longevity of the structure is higher than the loss in terms of environmental sustainability.

• Currently rammed earth is mainly used to build walls. We are investigating the possibility to use this material in other structural components like lintels, foundations and ground floor slabs.

• The thermal performance of a house made of rammed earth has not been exhaustively investigated. The perception of the dwellers is of comfort, but there are not enough studies to prove that a rammed earth dwelling can be comfortable environment with little use of cooling or heating systems. We are working to scientifically prove that while its insulating properties are not high, rammed earth walls make a house comfortable.

• Determining which soil is appropriate to be used as rammed earth and what strength and durability will characterise that final product remains an unanswered question. We are working to understand the source of strength of rammed earth by analysing samples under the microscope.

This research project is motivated by the recognition of the potential benefits of using rammed earth as a construction material in remote areas and has the goal of improving the housing program in remote Aboriginal communities in Western Australia. Our group is developing a comprehensive program of laboratory and on-site experimental tests with the aim of establishing a theory for the structural use of rammed earth as an engineering construction material. Our work will lead to the development of a draft design code for rammed earth that will be submitted in 3 years time to Standards Australia.

The success of the project also depends on the level of engagement of the indigenous communities participating in the project and their willingness to be actively involved in the house construction process.