Subject matter

Definition of the TC subject and of the studied material

The goal of this TC is to define dedicated testing procedure for stabilized and unstabilized earth as a building construction material (crude earth). A good understanding of the earthen constructions requires taking into account their large variability. A first reason of this variability is that the local soils are used as building materials. The local soils are variable depending on the pedology of the site, and the construction of earth buildings cannot be totally included in an industrial process. Then partly to adapt the building technique to the different soils, several constructions technics have been invented, which is the second reason of the variability. Among these technics the three most common ones are adobe blocks, cob and rammed earth (Hall, et al., 2012). Adobe blocks are produced or manufactured using wet earth based mixture molded which is removed before drying. Cob consists in stacking clods, made of a mix of plastic earth and fibres, in order to build a monolithic wall. Finally, rammed earth is formed by compacting raw earth (ideally with a water content which insures its highest density) within a framework and layered down in successive beds with a rammer. These definitions are however quite general and the exact protocols which are used strongly vary in function of the texture of the earth, the period and weather, the location and the mason. In addition, other construction technics like Compacted Earth Blocks, Extruded Earth Blocks, and Prefabricated Rammed Earth are gaining more and more interest.

On the other side, whatever the construction technics and the material used, some similarities exist. At first, earth walls are quite thick (30 to 50 cm) and are separated from the ground by a basement which is designed to prevent capillary rise. Earth materials are composed of clays, silts, sands and possibly gravels.

The continuous clayey matrix plays the role of the binder between all these constituents and thus provides cohesion and strength. But the connections between the constituents are not perfect and some small voids, called pores, are embedded within the solid material. The network formed by the connection of these voids, called the porous network, enables fluids (either liquid or gas) to flow through the material. Due to this property, earthen material is a porous medium with a quite low resistance to the vapor and liquid transfers. In addition, clays and fibers are by themselves porous media. Their high affinity with water molecules allows capillary condensations or adsorptions phenomena. Thanks to the combination of these two processes (vapor transport and in-pore water sorption/desorption), earthen materials are hygroscopic. In other words, raw earth has a great potential to regulate indoor relative humidity levels. The latent heat associate to the sorption/desorption phenomena can also strongly modify the thermal behavior of the material. Finally, this hygroscopic behavior impacts the mechanical characteristics of the material. For example the interactions of clay particles and/or fibers with the water molecules can be responsible for complexities in mechanical behavior, such as swelling and shrinkage (Van Damme, 2002; Lei, et al., 2014). In addition, the mechanical resistance of a soil decreases with its water content (i.e. mass of water on mass of dry solid) which is well known in soil mechanics but it also applies to earth buildings. It seems therefore that the water content in raw earth materials is crucial to predict its behavior.

In view of these complexities in behavior and these large variabilities, two main questions arise. The first one concerns the characteristics that must have an earth mixture to be used as a construction material. First answers have been provided by recommendation guides and standards. For example, BS1377 standards define a range of admissible values for the particle size distribution of earth. Along the same line, a proportion of 15-30% of clays, 10-30% of silts and 50-75% of sand is commonly recommended for the construction of adobe blocks (Garrison & Ruffner, 1983). In general, the clay content proportion should thus be sufficient to ensure a good material stiffness and strength, but the proportion of expansive clay must remain limited (lower than 50% of the total clay amount) in order to avoid cracking.

Therefore, the study of existing construction indicates that the problem is not so simple, and that the particle size distribution cannot be used as a discriminating parameter (Aubert, et al., 2014; Gomes et al., 2014; Champiré et al, 2016). In consequence, up to now, no clear objective rule exists to discriminate the ability of an earth to be used as a building material solely on the analysis of its granular and mineralogical properties.

The second main question concerns the evaluation of the performances (sustainability, strength, seismic resistance, thermal, moisture buffering, …) of the material. For the moment, the performance are estimated through the some common existing experimental methods: unconfined compressive strength to assess the mechanical, sorption/desorption curves and dry/wet cups for vapor adsorption and diffusion, hot wire or hot place for thermal conductivity, etc… These studies underline an important variability of the measured parameters. Numerous reasons of these disparities can be intuited like the lack of hygrometry and temperature control during the tests, the differences in samples manufacture, geometry, and in test protocol … As an example, in function of the experimental protocol used, an unconfined compressive test can lead to compressive strength up to 40MPa for earthen material (Aubert, et al. 2013), which is not realistic.

In addition, the knowledge of these common parameters is found to be insufficient to properly model the complex behavior of earthen walls (Soudani et al. 2016, Nowamooz and Chazallon, 2011, Bui et al., 2014).

In particular, the complexity in mechanical behavior of the material cannot be well described by the sole knowledge of a compressive strength and the apparent Young’s modulus of the first loading (Champiré et al, 2016), while its thermal insulation potential may be significantly underestimated if it is only based on its thermal conductivity and heat capacity (Soudani et al., 2016).

A first step in order to take into account material hygroscopicity in moisture buffering capabilities has been realized within the framework of the Nordtest project (Rode et al., 2005). It leads to the definition of test to determine the Moisture Buffering Value (based on the water mass uptake/release by the material when it is submitted to a daily cycle of relative humidity between 33%RH and 75%RH, which is representative of the average indoor RH daily variations in Northern European Countries. Although the principle of this test is interesting, it remains quite limited because of its lack of repeatability. In addition, its aim is to be applied to all hygroscopic materials, which is not necessarily a good option since the phenomena of sorption and desorption which occur within them can be strongly different. Finally, the applicability of this test in warmer European countries like Portugal or Italy, where the cycles of indoor relative humidity would rather be between 60%RH and 85%RH, is questionable.

In this context, the goal of this technical committee is to give an objective answer of these two questions by the definition of performance-oriented tests. These tests should depend on the use that will be made of the material, and their clear definition requires a good understanding of the specificities of earth, as well as the identification of the key parameters which drive its behavior.

Level of investigation

The objective is to define the minimal number of laboratory tests in order to provide an accurate assessment of the mechanical, thermal and hygroscopic performances of the material. To that aim, whenever possible, existing experienced tests will be used. However, due to the particularities of the material, development of new experimental protocols and/or the adaptation of existing ones may be necessary. These tests procedures and the way to analyse their results will be defined in the view of recent theoretical developments published in literature and/or developed elsewhere by the members of the TC. The behaviour of the material is strongly impacted by the external conditions (temperature and humidity variations, saline environment, etc…). In consequence, the definition of dedicated tests to assess the performance of the material in some specific environmental conditions may be necessary.

A main importance will be given to the repeatability and the robustness of the tests. Round robin tests will be performed, and, in function of the obtained results, the experimental protocols and even the nature of the test may be modified. The accuracy of the tests will be studied by comparing, at least qualitatively, their results to the on-site behaviour of earthen constructions. To allow this kind of confrontations, all the tested earths will come from existing constructions which are properly instrumented.

Limit of the scope

There is several ways to use earth in constructions (load-bearing walls, filling materials, plasters,…). This TC will be limited to load-bearing walls. However, several construction technics (rammed earth, CEB, adobe block, cob), made with unstabilized or lime/cement stabilized earths and with or without fibers will be studied. Whenever possible, the developed testing procedure would be such that they can be used for any of these constructions technics.

No standardisation activity, in its strict sense, will be made during the lifetime of this TC. However, all the experimental developments should be able to serve as a basis for the development of international/European standards on earth testing procedures

Finally, only experimental studies will be performed within the framework of this TC. All the theoretical developments required for this study have to be extracted from a bibliographic survey and/or realized elsewhere.

Terms of reference

This TC is mainly based on experimental researches, which are quite time consuming. Thus, an estimation of 5 years should be quite realistic.

Schedule

The TC should be mainly composed by academics (researchers, research engineers) and engineers. But strong interactions with architects and masons will be favoured by the realisation of dedicated workshops. The best number should be around 20 persons from at least 10 different laboratories.

Detailed working programme

The TC will be divided in two main working groups (WG). The first one will be focused on the thermal/hygrothermal characterisations and the second one the (hydro)-mechanical performances and seismic resistance. Each WG will follow the same general program as described above.

   1. Bibliographic study and initial set-up of key parameters

The first action of the TC will be the identification of the key parameters which must be measured in order to correctly assess the performance of the material.

To reach this goal, even before any additional development, an important action of this TC would be to bring together international recognized experts on earthen constructions. Then, the required theoretical knowledge on earthen materials will be provided by the TC’s members, coupled with an up-dated bibliographic survey.

This study will be formalised by the submission of a review paper in Materials and Structures and/or the redaction of a public report (D1).

In addition, an international conference on earth construction will be organised. The idea is to have a clear picture, at that stage, of all the newest development (both experimental and theoretical) on earthen constructions. In consequence, the proceedings of this conference (D2), which will be the second deliverable of the TC, will be completely in line with the objective of this first part.

   2. Initial set-up of laboratory tests

From the knowledge of the key parameters, a first set of experimental tests will be defined. Whenever possible, the use will be made of existing tests (like uniaxial compression test, hot plate test, etc…), but some new tests could be developed in order to take into account the specificity of the material (notably its strong interaction with water). Let us note that, the more rapid and cheaper tests will be preferred.

In addition, some dedicated tests for specifics environmental conditions will be developed. The definition of the specific conditions which are worth to be considered will be made on the basis of the knowledge of the TC’s members.

The results of this task will lead to the redaction of a confidential report (D3).

   3. Final set-up of laboratory tests

The accuracy of this first set of experimental tests will be tested in two ways.

The first one consists in a confrontation with on-site measurement. The idea is to verify that the performance of the material which is predicted from the laboratory tests is in concordance with the behaviour of constructions made with this material. To do so, all the tested material will come from existing constructions. Consequently, one of the first tasks of this TC will be to identify these constructions. This confrontation will be made on the bases of the temperature / hygrometry measurements inside/outside the construction, on the monitoring of energy consumption (when possible) and on the identification of possible pathologies.

The second one consists in assessing the robustness and repeatability of the tests. To do that, the idea is to realise round robin tests between all the involved laboratories. In function to the obtained results, some tests procedures will be modified.

In addition to scientific publications, the results of these tests will be formalised by a report defining the dedicated testing procedures for earth materials (D4).

A closure symposium will ultimately be organised in order to present the results of the TC to the scientific and the practitioner’s communities (D5).