Waste mechanics might seem an odd choice of topic for a lecture organised by the International Geosynthetic Society (IGS). However, this notion was dispelled at the beginning of the lecture as Professor Jean-Pierre Gourc outlined the important role that the waste body plays in both the short- and long-term performance of landfill lining systems. Issues of differential settlement affecting cap integrity and slope instability were highlighted.
Gourc presented a summary of research work being carried out under the auspices of the French Environment Agency (ADEME). The project, funded by ADEME and sponsored by a number of French waste management companies and consultants, is based on field testing at landfills with laboratory work and numerical analyses.
The first part of the talk described research into the geomechanical behaviour of waste. In terms of landfill engineering, the physical characteristics of waste considered relevant are the density, moisture content, compressibility and shear strength.
Research includes development of a system of characterising waste material in French landfills in terms of 'soil like' and 'soil unlike'materials. This is needed to relate mechanical behaviour to waste constituents.
Typically, sorting indicated that 60% of the waste was 'soil unlike' with the rest 'soil like' The density of waste at the surface of several landfills was measured using a water replacement technique and the results showed a great deal of scatter, with values ranging from 5. 7kN/m 3to 16. 1kN/m 3for municipal solid waste materials. Gourc highlighted the problem of measuring water content of waste because of the absorption of water by some waste constituents (paper for example).
Compression characteristics of waste are important in terms of being able to predict the total landfill settlement and differential settlement above side slopes. Gourc said landfill settlement could be divided into two parts: primary settlement related to loading and secondary settlement dependent on time. Five settlement monitoring techniques were used in the research. Benchmarks or surface monitoring points are being used to measure total settlement. An example of the results from Lapouyade landfill is given in Figure 1.
It is interesting to note from this data that the settlements at the edge of the landfill, as a percentage of the waste thickness, are significantly higher than towards the centre. This is due to the horizontal component of settlement that occurs within the waste above the perimeter landfill slopes (Figure 2).
On two other landfills, total settlements were also measured using inclinometer tubing placed horizontally either just beneath the cap cover liner or buried in the waste body (with two exits at ground level).
This provided information along a cross-section of the landfill (Figure 3). Gourc said the system had proved to be robust and reliable.
'Floating balls' (a hydraulic system with head difference measured between the balls and ground surface) have been installed at a range of depths within the waste body. Plates with vertical rods have also been placed in the waste at different depths and extended with subsequent waste placement. Gourc said that at one landfill, only two of the six vertical rods placed to monitor settlements survived the waste compactor.
Gourc then described a novel system of burying steel plates horizontally within the waste, retro-drilling once waste had been placed and fitting the rods from the ground surface.
Results from the extensive monitoring programme are being used to develop tools for predicting future settlements. This is a notoriously difficult task and an extension of consolidation theory was presented using incremental calculation of total settlement. It was proposed that separate compression coefficients should be obtained for the primary and secondary components of settlement. A good fit had been obtained with observed settlements, although the ability to predict confidently the long-term settlements due to creep and degradation was as yet unproven.
Measurement of the shear strength of waste is important to assess the stability of landfill slopes and Gourc outlined two methods of measurement. First, shear strength can be back-analysed from the failure of a controlled excavated steep slope within the waste. Unfortunately, forcing a waste slope to fail is difficult due to high shear strengths that result in part from the reinforced nature of waste.
A second method using a large (1m by 1m) field shear box has been developed. The direct shear apparatus was taken to site and waste was placed in the box. Typical results are shown in Figure 4. The shear stress vs displacement curves do not show a peak because of the contractant behaviour of the waste during shearing (ie decreasing volume with associated increase in shear strength). These results show that the full shear strength of waste is mobilised at high shear strains.
Gourc presented the results of a full-scale test on the performance of two different landfill capping liner systems subjected to the formation of a void directly beneath the liner.
The capping liners comprised a 1. 2m thick clay liner and a 0. 7m thick clay liner reinforced with a geosynthetic layer. It was shown that the thinner reinforced clay capping liner performed better than the thicker clay liner. The importance of compaction practices on interlayer bonding and of layer rigidity were discussed.
Gourc then turned to the use of geomembrane capping systems and emphasised the importance of the shear strength developed at the various geosynthetic/geosynthetic and geosynthetic/soil interfaces.
Interface shear strength can be measured by both direct shear and tilting table methods; the use of the latter to investigate creep, hydraulic and dynamic interface strength as well as the conventional quasi-static strength was discussed. Tilting table tests (Figure 5) on polypropylene geomembranes suggested that the peak friction angle could vary from 16. 5º for a quasi-static test to 12º for the same interface if the test was left for two days. This suggests that some geosynthetic interfaces may lose their strength with time.
Results of field monitoring of the displacement performance of four capping systems were presented.
The capping systems were placed on a 50m long slope with a gradient of 1:3 (~18º) (Figure 6). Strain gauges and displacement markers were used to monitor the total and differential movement between the various components of the capping system. Initially, different interfaces suffered varying amounts of displacement depending on the interface shear strength between the various components. However, it was interesting to note that with time there was little differential movement between layers and the displacement of the complete capping system was due to the settlement of the waste.
After a lively question and answer session, the vote of thanks was given by BGA vice-chairman Tony Bracegirdle.