Sie sind auf Seite 1von 7

Available online at www.sciencedirect.


Waste Management 28 (2008) 15281534

Landll gas upgrading with countercurrent water wash

S. Rasi *, J. La ntela , A. Veijanen, J. Rintala
University of Jyva skyla , Department of Biological and Environmental Science, P.O. Box 35, FI-40014 Jyva skyla , Finland Accepted 31 March 2007 Available online 11 September 2007

Abstract A pilot-scale countercurrent absorption process for upgrading landll gas to produce vehicle fuel was studied using absorption and desorption units and water as absorbent. The height-to-diameter ratio of the absorption column used was 3:1 instead of the more conventionally used 20:1 ratio, and a higher pressure was used along with a lower water ow rate. The eects of pressure (1030 bar) and water and gas ow-ratios on the upgrading process were studied. Methane content in the product gas increased to near or above 90% with both gas ows (50 and 100 l/min) used at over 20 bar pressure with 10 l/min water ow and at 30 bar pressure with 5 l/min water ow. Carbon dioxide content with these upgrading parameters ranged from 3.2% to 4.8%. The remaining fraction of the product gas was nitrogen (from 6% to 7%), while hydrogen sulphide was removed to below the detection limit with all of the upgrading parameters used. The methane content of exhaust gas increased with increasing pressure. In conclusion, the pilot-scale gas upgrading process studied here appears to be able to produce gas with high energy content (>90% methane), apparently suitable as vehicle fuel, from landll gas. 2007 Elsevier Ltd. All rights reserved.

1. Introduction Landll gas is a signicant greenhouse gas (GHG) which, due to its high energy potential, also has potential for energy production. The collection and subsequent use of landll gas for heat and electricity production is on the increase in many countries (Bove and Lunghi, 2006; Henigin and Eymann, 1996; Reinicke, 1996), both to reduce GHG emissions from landlls and to replace fossil fuels in energy production. Upgrading landll gas for biofuel enhances the energy density of the fuel and minimizes the transportation cost of the landll gas, thereby enabling better utilization of the recovered gas in cases where landlls are located at a long distance from the heat-consuming infrastructure. The main components of landll gas are methane (from 40% to 60%), carbon dioxide (from 35% to 50%), nitrogen (from 0% to 20%), oxygen (from 0% to 1%) and hydrogen sulphide (from 50 to 200 ppm) (Bove and Lunghi, 2006). Landll gas can also contain trace compounds such as ali*

Corresponding author. Tel.: +358 14 2602340; fax: +358 14 2602321. E-mail address: saija.rasi@bytl.jyu. (S. Rasi).

phatic and aromatic hydrocarbons, halogenated compounds and silicon-containing compounds up to a total concentration of 2000 mg/m3 (Schweigkoer and Niessner, 2001). Some of these compounds may cause corrosion or other damage to engines if they are not removed from the gas before utilization. So far, studies and applications on upgrading landll gas are relatively few, while the upgrading of biogases from sewage and biowaste digesters is rapidly increasing in some countries (Tynell, 2005). Upgrading landll gas to replace fossil vehicle fuels has been practised in some countries (Environmental Agency, 2004). In Reykjavik (Iceland), landll gas upgrading was started in 2000 with water scrubber technology (Halldorsson, 2006). The methane content in the product gas has been around 95%, and a landll can provide fuel for 30004000 vehicles. The main purpose of the project has been to sell energy from landll gas, and vehicle fuel has been seen as a reasonable option in a country where all of the energy used for heating and electricity is already produced from renewable sources. Countercurrent water wash, which is often used in upgrading biogas to vehicle fuel, is considered to be a simple, ecient and versatile method capable of treating

0956-053X/$ - see front matter 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.wasman.2007.03.032

S. Rasi et al. / Waste Management 28 (2008) 15281534


corrosive compounds at relatively low cost (Hunter and Oyama, 2000), especially in cases when wastewater can be used as an absorbent in the process. The countercurrent water wash process is commonly used to produce biogas in sewage plants or biogas plants treating manure and other organic waste. Biogas from these sources usually contains methane (from 55% to 70%), carbon dioxide (from 30% to 40%) and hydrogen sulphide (from 10 to 2000 ppm), while oxygen or nitrogen contents are typically below 1% (Jo nsson et al., 2003). Water absorption is an eective upgrading process for these biogases, as carbon dioxide (3.37 g/l, 1 atm, 0 C) and hydrogen sulphide (7.1 g/l, 1 atm, 0 C) are more soluble in water than methane (0.039 g/l, 1 atm, 2 C) (Air Liquide, 2006) and thus it is possible to produce high quality methane gas. Landll gas contains some compounds that are more complex to remove, such as nitrogen, halogenated compounds and organic silicon compounds, because of their low solubility in water (Air Liquide, 2006; Schweigkoer and Niessner, 2001). Even though biogas in increasingly upgraded for vehicle use, no literature is available on factors aecting product gas content in the absorption process and neither is there much literature available on upgrading of landll gas. Biogas upgrading processes need to be designed to avoid or minimize emissions. Some of the methane can be absorbed to an absorption liquid and be released into the air with exhaust gas. Hydrogen sulphide, other reduced sulphur compounds and halogenated compounds in exhaust gas can have environmental and health risks if released into the air untreated. Water used in the process should also be treated, for example, with other wastewaters (Environmental Agency, 2004). The objective of this study was to determine the feasibility of a countercurrent absorption process with a new type of design (with a small height-to-diameter ratio) to upgrade landll gas to vehicle fuel. For that purpose, the eects of pressure and of water and gas ows on product gas composition were studied. 2. Materials and methods 2.1. Upgrading equipment The pilot-scale upgrading facility contained an absorption column (500 l, height 185 cm, diameter 60 cm), desorption column (500 l, height 155 cm, diameter 65 cm) and water storage tank (500 l). Both the absorption column and desorption column were (randomly) lled with pallrings (4 4 cm) to create a large contact surface between the gas and water. The pilot-scale upgrading plant can treat 6 m3 of raw gas per hour and is built inside a movable container. The pilot plant was located at the Mustankorkea landll in central Finland, where landll gas is collected from a 12extraction well system over an area of 16 ha and used for district heating (Mustankorkea Ltd., 2005). Raw gas was

taken for the upgrading process from a collector pipe in the landll gas collection system. The raw gas was dried and then channelled into the rst gas storage tank and then compressed to the upgrading pressure. Compressed raw gas and absorbent were then channelled into the absorption column in which the pressure was maintained. The absorbent used in the process was tap water (measured pH 7). Water was sprayed from the top of the column (at a temperature of ca. 10 C), and raw gas was channelled from the bottom of the column (Fig. 1). After the absorption column, water and the compounds absorbed into it were channelled to the desorption column. In this column, water was regenerated at vacuum pressure (0 and 0.3 bar), exhaust gas was released from the column and the product gas (upgraded landll gas) was piped to a landll gas collection pipe. 2.2. Upgrading parameters Upgrading eciency was studied using dierent pressures (10, 15, 20, 25 and 30 bar) and water (5 and 10 l/ min) and gas (50 and 100 l/min) ows in trials which typically lasted 30 min each. The conditions were initially selected on the basis of preliminary studies which showed, e.g., that at pressure below 10 bar the methane content of the product gas was less than 80%. The water ow of 10 l/min and the gas ow of 100 l/min were the maximum operating ows with this equipment. Methane enrichment was calculated from the results, EnrichmentCH4 100 Cp Cr ; Cr

where Cp, methane content in product gas and Cr, methane content in raw gas. 2.3. Sampling and analyses Gas samples were taken from the raw gas, product gas and exhaust gas. The raw gas was obtained from the landll gas pumping station after condensation (considered dried gas), while the product and exhaust gas were from the pilot system without drying. Gas samples for the methane, carbon dioxide, oxygen, nitrogen and hydrogen sulphide measurements were collected in TECOBAG (PETP/AL/PE 12/12/75) aluminium gas bags (volume 10 l). Samples for the measurement of volatile organic compounds (VOCs) were collected in Nalophan NA bags. The samples were drawn from the bags into sampling tubes at a rate of 90 ml/min using a Gillian Personal Air Sampler (LFS-113DC). The adsorption material in the sampling tubes was Tenax GR. The tubes were capped and stored at 18 C until analysed. The adsorbed VOC samples were analysed using thermal desorption gas chromatograph mass spectrometry (Tekmar Purge & Trap Concentrator 3000/Agilent 6890+/5973 N MSD). Methane, carbon dioxide, oxygen and nitrogen were measured with PerkinElmer Autosystem XL gas chro-


S. Rasi et al. / Waste Management 28 (2008) 15281534

Fig. 1. Upgrading process used in this study.

matograph by using a thermal conductivity detector (TCD). Injector, oven and detector temperatures were 50, 45 and 160 C, respectively. Hydrogen sulphide was measured with an infra-red gas analyser (GA 94) equipped with an electro-chemical cell, Rae systems gas detection tubes (range 2.560 ppm and 25250 ppm) and a portable gas chromatograph (Photovac GC/PID). Water samples were collected before and after the desorption column; and pH, dissolved sulphide and VOCs were determined. The VOC samples were analysed using thermal desorption gas chromatographmass spectrometry (Tekmar Purge & Trap Concentrator 3000/Agilent 6890+/ 5973 N MSD). Dissolved sulphide was determined by the methylene blue method (Tru per and Schlegel, 1961) using an UV/VIS spectrophotometer (Hitachi U-1500). 3. Results 3.1. Eects of pressure and water and gas ow on product gas During the trials, the methane, carbon dioxide, oxygen and nitrogen contents in the raw landll gas were 53.2 1.4%, 40.8 1.0%, 0.4 0.3% and 4.9 1.2%, respectively. Methane content in the product gas was close to or above 90% at over 20 bar pressure with 10 l/min water ow and at 30 bar pressure with 5 l/min water ow

(Fig. 2). Over 68% methane enrichment and over 90% carbon dioxide removal eciencies were obtained with 10 l/ min water ow at over 20 bar pressure. Similar upgrading eciencies were obtained with lower water ow (5 l/min) at higher pressures (Fig. 2). Oxygen and nitrogen were not removed under any of the conditions studied, and nitrogen concentration in the product gas increased by about 2% over that in the raw gas (data not shown). During the trials, hydrogen sulphide content in raw landll gas ranged from 30 to 120 ppm, and it was removed from the product gas to below the detection limit (0.1 ppm) in all conditions studied (data not shown). The eects of the gas/water ow-ratio (ratios 520, gas ows 50 and 100 l/min, water ows 5 and 10 l/min) on methane enrichment and carbon dioxide removal eciencies were studied at 30 bar. The highest pressure was selected for these trials due to its high upgrading eciency in the present studies. In all of the gas/water ow-ratios studied, the product gas contained over 90% methane and less than 4.5% carbon dioxide, indicating that the gas and water ows applied had no eect on upgrading eciency (Table 1). Removal of VOCs in the raw gas was studied at 30 bar pressure, water ow 10 l/min and gas ow 100 l/min. The raw and product gases contained mainly aliphatic and aromatic hydrocarbons, but reduced sulphur compounds, hal-

S. Rasi et al. / Waste Management 28 (2008) 15281534

100 90 93 90 87


Efficiency %

80 70 60 50 40 30 5 10 15 20 25 30 35

Methane %

84 81 78 75 72 5 10 15 20 25 30 35

Pressure bar
CH4, water 10 l/min CO2, water 10 l/min CH4, water 5 l/min CO2, water 5 l/min

Pressure bar
Water 10 l/min Water 5 l/min

Fig. 2. The eects of pressure and water ow rates of 5 and 10 l/min on carbon dioxide removal and methane enrichment eciencies (left) and methane content of the product gas (right).

Table 1 The eect of gas and water ow-ratio on upgrading eciency and contents of main components of product gas at 30 bar pressure Water ow rate (l/min) 10 10 5 5 Gas ow rate (l/min) 100 50 100 50 Gas/water ow-ratio 10 5 20 10 Product gas CH4 (%) 90.5 1.2 91.5 2.5 90.5 1.2 90.1 1.4 CO2 (%) 3.7 0.9 4.5 1.3 4.5 0.8 4.2 1.7 O2 (%) 0.5 0.2 0.5 0.2 0.5 0.3 0.5 0.1 N2 (%) 7.0 0.5 6.8 0.7 7.1 1.0 7.4 1.0 CH4 enrichment eciency (%) 68.0 4.7 68.4 5.6 70.4 5.5 68.4 5.2 CO2 removal eciency (%) 91.1 2.3 88.2 3.3 89.0 2.0 89.7 4.1

ogenated compounds and organic silicon compounds were also detected in both gases. Organic silicon compounds or aliphatic and aromatic hydrocarbons other than benzene were not removed. Reduced sulphur compounds and halogenated compounds were removed from the product gas (Table 2). 3.2. Eects of pressure, water and gas ow on exhaust gas The eects of pressure (10, 15, 20, 25 and 30 bar) and water ow (5 and 10 l/min) on methane content in exhaust gas was studied using pressures of 0 and 0.3 bar (vacuum) in the desorption column. Methane content in exhaust gas increased with increasing pressure up to 15% at 30 bar pressure (Fig. 3). Carbon dioxide content in exhaust gas ranged from 73% to 82%, and hydrogen sulphide content was over 100 ppm in all of the conditions studied.
Table 2 The total amounts of VOCs (TVOCs), organic silicon compounds, benzene, reduced sulphur and halogenated compounds in raw gas and product gas (30 bar pressure, 10 l/min water ow, 100 l/min gas ow) Raw gas (mg/m3) TVOC Organic silicon compounds 24.2149.4 0.32.8 Raw gas (mg/m3) Benzene Reduced sulphur compounds Halogenated compounds 1.0 0.5 0.9 0.4 3.0 1.4 Product gas (mg/m3) 33.8142.8 0.53.1 Removal (%) 45 9 90 10 96 4

17 15

Methane %

13 11 9 7 5 5 10 15 20 25 30 35

Pressure bar
Water 10 l/min, 0 bar Water 10 l/min, -0.3 bar Water 5 l/min, 0 bar Water 5 l/min, -0.3 bar

Fig. 3. The eect of pressure and water ow on methane content in exhaust gas.

VOCs in exhaust gas were analysed at pressure of 30 bar, water ow of 10 l/min, gas ow of 100 l/min and pressure in the desorption column of 0 bar. Concentrations of TVOCs, organic silicon compounds, benzene, reduced sulphur and halogenated compounds in the exhaust gas were 46.2 17.1 mg/m3, 1.0 0.3 mg/m3, 1.1 0.5 mg/ m3, 1.0 0.1 mg/m3 and 3.1 1.4 mg/m3, respectively. 3.3. Eect of desorption on absorbent The eect of desorption on absorbent water quality was studied. It was not possible to separate the eect of the absorption or desorption unit conditions on water quality because of the slow water circulation in the process. TVOC


S. Rasi et al. / Waste Management 28 (2008) 15281534

Table 3 TVOC, benzene, organic silicon, reduced sulphur and halogenated compounds in process water before and after desorption (0 and 0.3 bar) unit Before (lg/l) TVOC Benzene Organic silicon compounds Reduced sulphur compounds Halogenated compounds 7301380 12.025.3 9.213.2 2.622.2 18.095.8 After (lg/l) 5151250 10.314.1 2.619.0 4.08.0 11.375.2

5.2 5.1 5.0 4.9

4.8 4.7 4.6 4.5 4.4 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 After


Fig. 4. pH of absorbent water before and after desorption.

concentration increased in the water samples during the studied period, but it could be seen from the dierent water samples that the VOC concentrations were lower after than before desorption (Table 3). The pH of the water increased after desorption (Fig. 4), while the dissolved sulphide concentration after desorption was below the detection limit of 0.01 mg/l in all of the water samples. 4. Discussion 4.1. Eects of pressure, water and gas ow on product gas The present results suggest that at over 20 bar pressure, the studied countercurrent process is a feasible method for the production of gas with high methane content (>90%) from landll gas. With 10 l/min water ow, the optimal operating pressure was 20 bar as higher pressures did not have any signicant eect on upgrading eciencies (Fig. 2). The lower methane content in product gas at 25 bar pressure than at 20 bar pressure was probably due to variation in methane content in raw gas. At the highest upgrading eciency the product gas contained, besides methane, about 4% carbon dioxide and about 7% nitrogen, which means that if the raw gas did not contain nitrogen and with the same carbon dioxide removal eciency, the methane content of the product gas could reach >95%. It also was shown that besides the pressure applied (10 30 bar), the absorbent ow rate aected the upgrading eciency (as carbon dioxide removal). Carbon dioxide removal eciency increased with increasing pressure at a constant absorbent ow rate, as expected according to

Henrys law (Hunter and Oyama, 2000), until 20 bar pressure was attained at a water ow of 10 l/min after which maximum removal eciency was achieved. On the other hand, a higher absorbent ow rate could compensate for lower pressure when the gas and water phase contact surface is higher; for example, at 20 bar pressure at an absorbent ow of 10 l/min a similar upgrading eciency was obtained as at 30 bar with ow rate of 5 l/min. The eects of pressure and absorbent ow on methane enrichment were not as straightforward because the methane content of the raw landll gas varied. The variation in carbon dioxide removal eciency was smaller, as was the variation in carbon dioxide content in the raw gas. In this study variation in raw gas content was not signicant and had a minor eect on the upgrading process. A higher variation, for example in methane content, which has been reported in other landll gas studies (Bove and Lunghi, 2006; Shin et al., 2002), can have a greater eect on the upgrading process. The quality of the product gas in this case study is suitable for vehicle fuels as the methane content was >86%, which is the minimum natural gas quality (G25 reference test fuel) for natural gas vehicles (European Commission, 2001), and as compounds harmful to engines, as classied by the Environmental Agency (2004), were not found in high concentrations. On the other hand, the content of methane, nitrogen and carbon dioxide in the product gas did not fulll the Swedish quality demands for vehicle fuel, namely methane >95% and carbon dioxide, oxygen and nitrogen combined <3% (Wellinger and Lindberg, 2000). Although carbon dioxide and nitrogen in the product gas do not harm car engines, the energy content of the gas is lower (Jo nsson et al., 2003). The higher relative nitrogen content in product gas than raw gas indicates nitrogen enrichment in the upgrading process, which can be prevented by adding a separate upgrading unit, such as an activated carbon adsorption process (Yang, 1987) for nitrogen or by lowering the nitrogen content in the raw gas, e.g., through changes in the landll gas collection system. On the other hand, an additional adsorption unit would complicate the process. Another potential concern is that organic silicon compounds found in the raw gas could not be removed from the product gas under the conditions tested. 4.2. Implications for eld applications The results indicate that with absorbent columns with a height-to-diameter ratio of 3:1 and liquid ow rates of 0.03 and 0.017 m3/min per m2 of the column area can be used to produce methane-rich gas alongside the more conventionally used 20:1 ratio and a water ow rate from 0.04 to 2 m3/min per m2 of the column area (Hunter and Oyama, 2000; Tynell, 2005), as shown by the fact that in the present upgrading process over 90% methane content gas was achieved. Absorption columns used in water absorption processes are typically 10 m in height to achieve maximum


S. Rasi et al. / Waste Management 28 (2008) 15281534


contact surface between the gas and water phase, and upgrading is done at 912 bar pressure (Tynell, 2005). In this study higher pressure compensated for the lack of column height. The dependence on the upgrading eciency of water and gas ow-ratios in this study was negligible. In previous reports, where column designs are dierent, water and gas ow-ratios are more important to upgrading eciency than the dimensions of the absorption column (Kapdi et al., 2005). The high gas ow rate can cause ooding in the column, and with a high liquid ow rate ooding can occur even at lower gas ows (Schnelle and Brown, 2002). In this study ooding was not experienced at any of the water or gas ow rates. Carbon dioxide removal eciency was highest at the highest water and gas ows (10 l/min and 100 l/min, respectively), but the dierences between the dierent water/gas ratios were small. The high pressure used in the process may be the main reason for the lack of an eect of high the gas/water ratios. At 20 bar and lower pressures, methane enrichment and carbon dioxide removal eciencies were more dependent on the water ow rate than at 30 bar pressure. With the higher water ow, the highest carbon dioxide removal was achieved at lower pressures and with nitrogen in the raw gas higher methane content could not be achieved. This means that at 30 bar pressure, less water is needed for gas upgrading; and because water is one of the major costs in this process, if pure water is used without water circulation, decreasing the amount of water used should be considered. In cases where landll leachate or other wastewater can be used in the process, upgrading at a lower pressure and at a higher water ow rate is probably more economical. It is also possible that the life of the equipment decreases and apparent that pumping and compression energy cost increases when higher pressure is used. Both water circulation and use of wastewater increases the possibility of biological growth on the packing material, and can lower the upgrading eciency if packing material is not regularly cleaned (Tynell, 2005). Because of high solubility of hydrogen sulphide in water over a wide pressure range, pressure or water ow rates had no eect on hydrogen sulphide removal as its concentration in the product gas was, in all conditions studied, below the detection limit of 0.1 ppm. The present upgrading process removed halogenated and sulphur compounds from the landll gas even though these concentrations in the raw gas were low compared to those found in other studies (Shin et al., 2002; Stoddart et al., 1999) and lower than the concentrations of over 250 mg organic Cl/N m3 and over 1500 ppm of hydrogen sulphide considered damaging (corrosive) to gas engines (Environmental Agency, 2004; Stoddart et al., 1999). For vehicle use, the limits are stricter, <3 ppm for hydrogen sulphide and <1 mg/m3 for halogenated hydrocarbons (Jo nsson et al., 2003; Wellinger and Lindberg, 2000), which were achieved with the present upgrading process. Most of the VOCs present in the raw gas, mostly aliphatic and aromatic hydrocarbons, were not removed in the upgrading process and neither were organic silicon

compounds. In this study the amounts of organic silicon compounds in the raw gas were less than 2.8 mg/m3, which was lower than reported in some previous studies (22 mg/ m3 and up to 50 mg/m3) (Schweigkoer and Niessner, 1999; Stoddart et al., 1999). The limits for the concentration of silicon compounds in landll gas given by dierent gas engine manufacturers range from 0.03 to 28 mg/m3. Silicon compounds can cause scung to engines but are also reported to deactivate catalysts (Wheless and Pierce, 2006). Because amounts of organic silicon compounds can vary in dierent locations, upgrading needs have to be estimated separately for every landll. Because of the low solubility of silicon compounds (except trimethylsilanol), absorbents other than water or another upgrading method could be considered, e.g., an eective method for the removal of silicon compounds is silica gel adsorption (Schweigkoer and Niessner, 2001). Other techniques for removing silicon compounds are absorption to concentrated sulphuric acid or nitric acid, but these are not practical processes because of the potential environmental and health risks of acids to humans (Schweigkoer and Niessner, 2001). The results of this study show that, for example, methane, hydrogen sulphide and VOCs can be emitted to air from exhaust gas and absorbent if there is no exhaust gas or wastewater treatment. Especially when higher pressures are used in absorption columns, exhaust gas should be channeled back to the upgrading process as more methane is lost in higher pressures (Fig. 3). During the upgrading process, carbon dioxide and hydrogen sulphide are absorbed to water and some of the carbon dioxide molecules form carbonic acid (H2CO3), which lowers the pH of the water, making it slightly acidic (Tynell, 2005). After desorption, the pH of the water was higher than before desorption, but still acid, hence some of the dissolved compounds leave the water during regeneration and become part of the exhaust gas. The pressure in the desorption column also aected the methane content of the exhaust gas at higher absorption pressures, being lower when 0.3 bar pressure was used in desorption. The dierences were negligible, and a longer measurement time in each trial would have been needed to evaluate the total eect of desorption pressure. With the water absorption process used in Jo nko ping, Sweden, methane losses are calculated to be about 3% (Persson, 2003). In the present study, product or exhaust gas ows could not be measured and thus total methane loss was not calculated. Dierent upgrading parameters or different desorption pressures had no eect on the hydrogen sulphide content of the exhaust gas. Samples were taken over a short period, hence variation in exhaust gas composition may not have had time to occur. To minimize emissions into the air, the compounds removed from the raw gas during the landll gas upgrading process can be burnt or collected and channelled back to the upgrading process. If the exhaust gas is not treated, some of the released compounds, such as hydrogen sulphide and halogenated compounds, may have environmental or health eects on humans. Emissions from landll gas aring can contain


S. Rasi et al. / Waste Management 28 (2008) 15281534 Halldorsson, B., 2006. Experience of landll gas upgrading to vehicle fuel. In: Proceedings, Nordic Biogas Conference 89.2.2006, Helsinki, Finland. Henigin, P.L.A., Eymann, U., 1996. Landll gas use in reciprocating engines and turbines in Germany. In: Christensen, T.H., Cossu, R., Stegmann, R. (Eds.), Landlling of Waste: Biogas. E&FN Spon, London, pp. 621632. Hunter, P., Oyama, S.T., 2000. Control of volatile organic compound emissions. Conventional and Emerging Technologies. John Wiley & Sons Inc., New York, USA. Jo nsson, O., Polman, E., Jensen, J.K., Eklund, R., Schyl, H., Ivarsson, S., 2003. Sustainable gas enters the European gas distribution system. Danish Gas Technology Center. Kapdi, S.S., Vijay, V.K., Rajesh, S.K., Prasad, R., 2005. Biogas scrubbing, compression and storage: perspective and prospectus in Indian context. Renewable Energy 3, 11951202. Mustankorkea Ltd., 2005. Biokaasun energiaka yto n kokeiluvaihe ylitti odotukset: Kaatopaikan biokaasun riitta a 500 omakotitalon la mmitykseen Jyva skyla ssa (Energy use of biogas exceed expectations: landll gas covers heating for 500 houses in Jyva skyla , original in Finnish). www.mustankorkea.. Persson, M., 2003. Utva rdering av uppgraderingstekniker fo r biogas. Report SGC 142, Swedish Gas Centre (Evaluation of upgrading techniques for biogas, original in Swedish). Reinicke, B., 1996. Landll gas in reciprocating engines: operational experiences. In: Christensen, T.H., Cossu, R., Stegmann, R. (Eds.), Landlling of Waste: Biogas. E&FN Spon, London, pp. 633643. Schnelle, K.B., Brown, C.A., 2002. Air Pollution Control Technology Handbook. CRC Press, Washington, USA. Schweigkoer, M., Niessner, R., 1999. Determination of siloxanes and VOC in landll gas and sewage gas by Canister sampling and GCMS/ AES analysis. Environmental Science & Technology 33, 36803685. Schweigkoer, M., Niessner, R., 2001. Removal of siloxanes in biogases. Journal of Hazardous Materials B83, 183196. Shin, H.-C., Park, J.-W., Park, K., Song, H.-C., 2002. Removal characteristics of trace compounds of landll gas by activated carbon adsorption. Environmental Pollution 119, 227236. Stoddart, J., Zhu, M., Staines, J., Rothery, E., Lewicki, R., 1999. Experience with halogenated hydrocarbons removal from landll gas. In: Proceedings Sardinia 1999, Seventh International Waste Management and Landll Symposium, vol. II. CISA, Cagliari, Sardinia, pp. 489498. Tru per, H.G., Schlegel, H.G., 1961. Sulphur metabolism in thlorhodaceae I. Quantitative measurements on growing cells of Chromatium Okenll. Antonie van Leeuwenhoek 30, 225238. ., 2005. Microbial growth on pall-rings a problem when Tynell, A upgrading biogas with the technique absorption with water wash. Svenska biogasfo reningen. Wellinger, A., Lindberg, A., 2000. Biogas Upgrading and Utilization IEA Bioenergy Task 24. International Energy Association, Paris, France. Wheless, E., Pierce, J., 2006. Siloxanes in landll and digester gas update. SCS Engineers. Wiemer, K., Widder, G., 1996. Emissions from landll gas thermal treatment plants. In: Christensen, T.H., Cossu, R., Stegmann, R. (Eds.), Landlling of Waste: Biogas. E&FN Spon, London, pp. 727 749. Yang, R.T., 1987. In: Gas Separation by Adsorption Processes, vol. 1. Butterworth Publishers, London, Copyright 1997 by Imperial College Press, Michigan.

high levels of NOx, CO and PAHs and, in some cases, vinyl chloride and dioxin (Wiemer and Widder, 1996). On the other hand the health risk from a gas upgrading plant is not dierent from other landll gas utilization plants. In general, the amount of TVOCs in the water increased during the measurements, which indicates some enrichment of VOCs during the upgrading process. The amounts of VOCs were lower after than before desorption, except for organic silicon compounds. The amounts of organic silicon compounds varied during the measurements, and there was no dierence between samples taken after and before. The eect on the environment of the water used can be decreased if the water is regenerated and used again in the process. Gas upgrading plants with water absorption and no water circulation are mainly used in the context of wastewater treatment plants where treated wastewater can be used as an absorption liquid and channelled back to the wastewater treatment process after use (Tynell, 2005). In landlls the water used can be channelled to the water treatment unit and treated with leachate or it can be used for irrigation of the landll. This research demonstrates the potential for high pressure water absorption systems to enrich landll gases. Flow rate data were not collected for the product and exhaust gases, and more research is needed to conduct a conclusive mass balance analysis of the system. Further work is also needed into the tradeos related to energy requirements and wastewater treatments needed for low and high pressure absorption systems. Acknowledgements This work was funded by the Maj and Tor Nessling Foundation, National Technology Agency of Finland (Tekes, Grant no. 70045/03), Grenovia Ltd. and Tampere Regional Solid Waste Management Ltd. Thanks go to Mustankorkea Waste Management Ltd. for providing the site for this study. References
Air Liquide, 2006. Gas encyclopedia. Encyclopedia.asp?languageid=11. Bove, R., Lunghi, P., 2006. Electric power generation from landll gas using traditional and innovative technologies. Energy Conversion and Management 47, 13911401. Environment Agency and Scottish Environment Protection Agency, 2004. Guidance on gas treatment technologies for landll gas engines, Bristol. gas_treatment_tech.pdf. European Commission, 2001. Commission directive 2001/27/EC. Ocial Journal of the European Communities L107/10. eur-lex/pri/en/oj/dat/2001/l_107/l_10720010418en00100023.pdf.