Landfill 2000 - A Field Trial of Accelerated Waste StabilisationProject Objectives
This study investigated alternative landfill management procedures designed to accelerate the stabilisation processes occurring in municipal solid wastes (MSW). The Landfill 2000 test cells were located at Dewsbury Landfill, UK. The work was carried in two purpose built prototype waste processing cells each containing 1000 tonnes of MSW. The research objectives were to:
Technical Background to the Project
- Establish optimum cell conditions that would give a three year waste degradation period
- Quantify the methane generating potential of the waste/sewage sludge mix
- Determine the biological condition of the waste digestate after a three year period and assess its suitability as an environmentally friendly soil conditioning medium
The project investigated a concept proposed by West Yorkshire Waste Management termed ‘Landfill 2000’. The concept was based upon anaerobic digestion of MSW in purpose built waste processing cells. For such a concept, the prime objective was to recover the usable gas potential of the waste under controlled but accelerated anaerobic conditions. The resulting digestate was either removed to landfill or further processed to produce a soil conditioning medium. In any subsequent landfill operation, inorganic contaminants are flushed from the material using a lower level of environmental protection than is commonly expected for un-processed municipal solid waste. In order to achieve more rapid waste stabilisation, effort was required in two areas: the first involving the development of control procedures over the stabilisation processes in a landfill environment, and the second being more focused attention on the quality and characteristics of waste going to landfill. Overall, an alternative operational strategy for landfill was sought to treat the landfill as a process and not simply a ‘black box’ repository for waste. The work was undertaken over the period April 1991 to March 1995.Overview of the Work Programme
The study examined the operational features of the Landfill 2000 concept, which made the ambitious assumption that the usable gas extraction period can be contracted into a three year waste incubation period. Two key design features were adopted in the design of the trials to reflect this objective. These included the use of:
- Digested sewage sludge in a mixture with municipal solid waste (to provide a source of essential moisture and microbial innocula)
- Additional liquid to enable leachate recirculation (to promote microbial movement in the wastes including the mobilisation of degradable substrates and essential nutrients)
Plans for constructing two experimental reactors at the Lower Spen Valley Landfill at Ravensthorpe, West Yorkshire were agreed in 1990 and a high-density polyethylene (HDPE) flexible membrane liner was chosen as the waste containment medium. Two cells, each 36-metre-long by 23-metre-wide with a maximum depth of 5 metres were excavated, and the sides and base lined with a 1.5 millimetre sheet. A 1 millimetre sheet was used as the closure cover. The liner material was manufactured and installed by Agru of Austria, with quality assurance and quality control aspects of the installation overseen by Golder Associates (UK) Ltd.
Approximately 1000 tonnes of MSW and Sewage sludge mix (10:1) overlaid leachate drainage media and a collection pipework system in the base of each reactor. Additional pipework was laid in a protective cover layer over the waste, allowing gas collection and leachate recirculation. One reactor (Cell 2) was set up as a control to compare standard landfill practice and operational procedures with a more manipulative approach in a second reactor (Cell 1), based on the ability to recirculate leachate. Imported liquor was introduced and recirculated in Cell 1 to provide the required moisture content and nutrient mobility for the process to thrive. Gas quality and temperature readings within each reactor were monitored from a number of discrete probes along with the ability to measure gas pressure and gas flow rate (device patented, Bartington and Reynolds
). Atmospheric pressure and ambient temperature measurements were also included. Data was recorded from all installed sensors using data-logging equipment housed in a site cabin positioned between the two reactor cells. The physical chemical and biological condition of the waste was determined at the end of a four year incubation period, by carrying out a sample recovery exercise using a percussion drilling technique. Waste characterisation work included category and distribution assays, interstitial liquid analyses, other physical and chemical analyses, including determination of biochemical methane potential (BMP).Key Issues Relevant to the Development of Sustainable Landfill Practices
- Examination of the static water balance for the recirculation cell showed that substantial movement of moisture took place through the body of the waste, rather than being confined to specific ‘by-pass’ routes. Lithium tracer tests indicated a mean hydraulic retention time of about six months. The water balance data suggest a total bed volume of about 640m3 , of which nearly 380m3 was assumed to be mobile water within the wastes. At an effective recirculation rate of close to 3 m-3d-1, we estimated that the system was operating at a mean hydraulic retention time of between 130 and 210 days.
- Biological stabilisation processes within the experimental cells generally progressed in a bi-modal fashion. Examination of leachate composition within the drainage blanket and interstitial liquid within the wastes provided the evidence to support this conclusion. Volatile acid rich leachates, typical of acidogenic conditions within landfilled MSW, predominated within the wastes of both cells, although leachate recirculation led to a more even distribution of these dissolved organics throughout the body of the waste. By comparison, the sump leachates were typically methanogenic.
- Further evidence of bi-modal behaviour was obtained through the analysis of low level hydrogen in the gas phase immediately above the saturated zone (2.5-3.0 metres depth) and at 1.5 metres in the unsaturated waste. In the recycle cell, hydrogen levels remained unperturbed by induced organic flow between the waste and the saturated zone, indicating that the methanogenic potential of the drainage media was well developed by a steady flow of substrate, induced by leachate recirculation. On the other hand, hydrogen levels in the non-recycle cell remained high, following the removal of standing leachate in the cell. We concluded that the induced flow of organic rich liquid into the saturated zone of this cell provided a shock loading of substrate far in excess of the capability of methanogens to remove it efficiently. Consequently, the methanogenic potential in the base of the non-recycle cell was not as well developed as in the recycle cell.
- The quality of landfill gas, in terms of the methane content, was higher in the leachate recycle cell (50-60% CH4 v/v compared with 45-50% CH4 v/v in the non-recycle cell). Specific landfill gas (LFG) yields from the two cells were estimated from measured flow data and range from around 20 to 9 m3 LFG per tonne ‘as received’ MSW (wet weight) per year, for the leachate recycle and non-recycle cells, respectively.
- Assuming a maximum LFG yield of around 200 m3. dry tonne-1 MSW, calculations show that around 60% of the readily available gas had been generated in the leachate recycle cell in comparison with 23% in the non-recycle cell. Data in our report showed that the readily degradable fraction of MSW accounted for around 18% w/w of the total mass, with 12% classified as ‘moderately’ degradable, 31% classified as ‘slowly’ degradable and 40% biologically inert. Category and distribution assays of recovered samples from both cells supported these claims and showed the wastes to be largely unaltered by the three year incubation in the experimental cells.
- Laboratory tests carried out to determine the residual biochemical methane potential (BMP) of the recovered waste samples from the cells yielded results which, when equated with the measured gas production figures, compared favourably with more recent estimates of MSW methane yield cited in the literature.
In view of the comparatively high gas yields obtained in the trials, we concluded that the stabilisation processes within the waste of both experimental cells had been advanced significantly by the addition of water (in the form of digested sewage sludge) at the time of waste emplacement. Further improvements had been achieved by leachate recycling activity.
The gas yields recorded were particularly remarkable in view of the low temperature conditions maintained in the shallow waste cells (6-17o
C) when compared with temperatures normally achieved in deeper more insulated operational landfills (20-40o
C). By assessing the Biochemical Methane Potential (BMP) of recovered waste samples and extrapolating data for volumetric gas flow under passive venting from the cells, we were able to make broad predictions about MSW stabilisation periods. For the leachate recycle cell, a predicted stabilisation period of around 7 years compared with approximately 17 years in the non-recycle cell. These stabilisation periods were estimated on the basis of removal of readily degradable organics and did not take account of the removal of more recalcitrant carbon in the waste.
Hydraulic behaviour within the recirculation cell was difficult to interpret. However, retrospective examination of water balance data indicated that, at an effective recirculation rate of close to 3m3
, a mean hydraulic retention time (HRT) of between 130 and 210 days has been successfully achieved.
We concluded that leachate recirculation techniques could be developed at full scale operational landfills where mean HRTs of between 1 and 5 years are required to meet current guidance for flushing bioreactor conditions. However, changes in effective porosity of saturated basal waste layers should not be discounted. Our calculations suggested that following an initial waste porosity of around 25% at the start of recirculation, this rapidly diminished in the saturated layers of waste to around 3%, within 30 months. Our assumption had been that stabilisation activity within this saturated layer led to accelerated settlement and compaction in this region. The implication we derive is that the saturated basal layer of a flushing bioreactor landfill should be constructed to minimise this effect in the short to medium term. We recommended that investigations be carried out on potential methods for retaining waste porosity characteristics in saturated waste layers at the base of landfills.
Although leachate recirculation had been responsible for a more uniform distribution of soluble organics and the development of more neutral pH conditions within the wastes, bimodal activity predominated in both cells, with the saturated zone presenting significant opportunities for optimising methanogenic removal of dissolved organic carbon. We recommended that further work be undertaken to investigate the use of low level hydrogen analysis as a means of monitoring, and hence managing, the performance of this zone in bioreactor landfills.
We also concluded that there was little scope for recovering the usable gas potential of landfilled wastes within a three year period, as proposed in the original ‘Landfill 2000’ concept. Although stabilisation of the wastes had been accelerated through leachate recirculation activity, other waste characteristics changed little over the four year period of the trial and therefore the waste material would be unsuitable as a soil conditioning medium, without further pre-treatment.
The trials demonstrated successfully that there is substantial scope for accelerating the removal of the readily degradable fraction of MSW in the landfill environment. The trials also provided some insight into the development of control procedures and drew attention to the changes in structural characteristics of waste under conditions of significant leachate recirculation.