Comparison study of landfill gas emissions from subtropical landfill with various phases: A case study in Wuhan, China

Abstract

The compositions and annual variations of landfill gas (LFG) were studied at two large-scale sites of Chen-Jia-Chong Landfill. Seventy-six wells were built and used for the collection and measurement of LFG. The investigation revealed the similarities and differences of LFG components and variations at two sites with different phases. It was found that ambient temperature and rainfall exhibited strong correlations with LFG components at both sites. Methane (CH₄) contents showed excellent correlations with CO₂ at both sites. Notable correlations between hydrogen sulfide (H₂S) and major components (CH₄ and carbon dioxide [CO₂]) were only observed in unstable methane phase. Especially, the CH₄/CO₂ volumetric ratio could act as an excellent indicator for anaerobic reaction stage by judging its phasic variations. The study is beneficial for the efficient operation of LFG collection system and could shed light on gas purification and utilization.

Implications: The results in this paper could provide some beneficial information for landfill operators. Especially, the CH₄/CO₂ volumetric ratio could act as an excellent indicator for anaerobic reaction stage by judging its phasic variations. Moreover, the study could shed light on landfill gas purification and utilization.

Introduction

Landfill is the most predominant method for municipal solid waste (MSW) disposal in most countries. However, a number of previous studies in the last two decades have indicated that landfill may impose a threat to the environment and human health due to the generation of landfill gas and leachate. Municipal solid waste (MSW) in landfills decomposes and produces methane (CH₄) and carbon dioxide (CO₂) gases, trace amount of toxic substances, and bad odor, which are the by-products of the decomposition (Shin et al., 2005). Landfill gas is ranked as the third highest source of global anthropogenic methane emissions, responsible for approximately 9–12% of those emissions in 2005 (Intergovernmental Panel on Climate Change [IPCC], 2007). In particular, the potential contribution of methane to the global warming is 21 times higher than that of carbon dioxide (Gewald et al., 2012). Carbon emission reduction has become a global issue with the increase of methane and carbon dioxide emissions. In recent years, there are generally four approaches for dealing with methane. Firstly, modified-soil covers are applied to promote methane oxidation (Einola et al., 2007; Chiemchaisri et al., 2011; Sadasivam and Reddy, 2014) and aeration is used to reduce methane production (Ritzkowski and Stegmann, 2007; Raga and Cossu, 2013). Secondly, separate collection of organics before landfilling could efficiently reduce methane emission from landfills (Calabrò, 2009; Bernstad and Jansen, 2012). Thirdly, mechanical biological pretreatment was proven to reduce the gas generation potential and leachate strength of the residue (Bayard et al., 2008; Calabrò et al., 2011). In addition, landfill gas (LFG) is efficiently collected and used as bioenergy for electricity generation (Qin et al., 2001; Bove and Lunghi, 2006; Aguilar-Virgen et al., 2014). Overall, the application of LFG on power generation is more environmentally friendly and sustainable as a potential clean energy.

Although varieties of models were applied to estimate the potential energy utilization from landfills (Mackie and Cooper, 2009; Gregg, 2010; Amini and Reinhart, 2011; Penteado et al., 2012), the acknowledgment of the MSW field degradation is indispensable, especially for large-scale landfills. The LFG collection system efficiency varies by the configuration of the system installed (e.g., type, active or passive, number of wells) and by operation management (depression intensity) (Calabrò, 2009). The performance of MSW anaerobic degradation has been investigated using the simulation experiments (Mata-Alvarez et al., 2000; Agdag and Sponza, 2005; Valencia et al., 2009). There is still a great gap for compositions and collection efficiency between simulation experiments and field studies. This paper aims to investigate the characteristics and annual variations of LFG components (CH₄, CO₂, H₂, H₂S, and O₂) at two sites with various landfill ages. Meanwhile, climate parameters (ambient temperature and rainfall) are studied to figure out the effects on LFG generation. The emphasis is placed on the similarities and differences between two sites in different phases of anaerobic degradation.

Material and Methods

Site description

Chen-Jia-Chong Landfill (30°43′37.03″N, 114°32′35.09″E) is located in Wuhan City of central China. The landfill was opened in 2007 for municipal solid waste (MSW) dumping, acting as one of the biggest sanitary landfill with a high-density polyethylene (HDPE) cap in central China. During the landfilling process, HDPE membrane (1.5 mm) was used as the intermediate cover. Once each landfill cell was finished, the intermediate cover was laid to avoid odor and effects of flies. The average amount of waste disposed at this landfill reaches about 2400 t/day. The landfill is divided into six areas, two of which (A-1 and A-2) were chosen as the study area because of the effective landfill gas collection (Figure 1). The organic matter content of fresh waste ranged from 31% to 47%, and the average moisture content reached 53%. A landfill gas-fired power plant has been in operation based on the biogas from sites A-1 and A-2 since May 2009. The landfill age of site A-1 (from April 2007 to July 2008) was 1.5–3.5 yr and that of site A-2 (from August 2008 to September 2009) was 1–2 yr during the study period of 2010. The details of two sites are presented in Table 1.

Table 1. The details of the two study areas

Site	Period	Amount (million ton)	Area (ha)	Number of Wells
A-1	April 2007–July 2008	1.04	6.05	30
A-2	August 2008–September 2009	1.62	8.65	46

Figure 1. Study areas and LFG collection system in Chen-Jia-Chong Landfill.

Data collection and pattern

There were 30 and 46 wells built for LFG collection at sites A-1 and A-2, respectively. Sensors (biogas sensors; smartGAS, supplied by York Instument Corp., Beijing, China) were installed in each well for monitoring the gas composition (CH₄, CO₂, H₂, H₂S, and O₂). The data were transmitted to the control system and stored in the database. The depression of each well was constant. The biogas was collected and pumped into the extraction and purification system. Climate data (ambient temperature and rainfall) were also collected for analyzing their influences on LFG generation. Ambient temperature was measured every day. The rainfall data for each month were obtained from Wuhan Meteorological Bureau, Wuhan, China. Data of ambient temperature and rainfall in 2010 are shown in Figure 2a.

Figure 2. (a) Temperature data and rainfall data during the year 2010. (b) Gas quality during the “life time” of a landfill (Heyer et al., 2013). (c) Annual variations of CH₄, CO₂, and CH₄/CO₂ from both sites. (d) Annual variations of H₂S, H₂, and O₂ from both sites.

Statistical analysis

A number of gas components emitted as LFG were measured, and the emission characteristics of different sites were examined through statistics analysis. Correlation analysis using SPSS (Statistical Product and Service Solutions) 19.0 software (IBM, Armonk, NY, USA) was conducted to analyze the relationship between the gas components (CH₄, CO₂, H₂, H₂S, and O₂) and the climate parameters (ambient temperature and rainfall). Meanwhile, relationships among gas components were also analyzed. The differences presented were conformed by t test at 95% and 99% confidence levels.

Results and Discussion

Comparison of LFG component distribution

Shortly after MSW is landfilled, the organic components undergo biochemical reactions. The principal biochemical reaction in landfills is anaerobic degradation, which occurs in three main stages (hydrolysis, acetogenesis, and methanogenesis) that could be divided and extended further to several phases, as shown in Figure 2b (Themelis and Ulloa, 2007; Heyer et al., 2013). In general, LFG concentrations of the same landfill cell were highly variable as a result of the heterogeneity of solid waste in the investigated landfill sites (Chai et al., 2011). The LFG composition (CH₄, CO₂, H₂, and H₂S) of two sites with different ages were compared and analyzed in this section.

Variation of CH₄ and CO₂ contents

The variations of CH₄ and CO₂ at site A-1 and site A-2 are presented in Figure 2c. The CH₄ contents ranged from 45.48% to 62.91% and from 14.54% to 65.59% for site A-1 and site A-2, respectively. Both methane contents from A-1 and A-2 were in accordance with the data of Figure 2b and kept increasing. Moreover, it is observed that CH₄ content of A-2 showed a higher growing rate of 3.51 times than that of A-1 for 36.69% during 2010. As for carbon dioxide, the contents of CO₂ ranged from 21.74% to 31.60% and from 22.29% to 31.62% for site A-1 and site A-2, respectively. Both CO₂ contents of A-1 and A-2 showed slightly increasing trends. CO₂ content of A-1 displayed a more stable curve with smaller fluctuation values (26.03 ± 4.29%).

There may be several potential causes that may explain the differences between A-1 and A-2 for CH₄ and CO₂ variations. A lot of variations in the LFG emission are observed due to hydrogeology of the site and the methods of landfilling (i.e., open dumping or sanitary landfill) (Frid et al., 2010; Kumar et al., 2004). Chen et al. (2008) found that CH₄ and CO₂ emission rates depended on soil pH, moisture content, pressure, total organic carbon, type and age of the burial waste, type and depth of soil cover, and methane oxidation. In addition, it is reported that 2–3-yr-old landfill had the highest methane and carbon dioxide emission rates, whereas 5-yr-old landfill was the least (Hegde et al., 2003). As for Chen-Jia-Chong Landfill, hydrogeology, pressure, waste composition, and landfilling method were nearly the same for sites A-1 and A-2. HDPE cover was employed as the temporary and final caps, whereas methane oxidation was only effective in conventional soil covers (Einola et al., 2007; Park et al., 2010; Capaccioni et al., 2011; Chiemchaisri et al., 2011). Accordingly, soil characteristics were unlikely to be the key impact factors. Hence, landfill age could explain the phenomenon above and the differences may be attributed to the different stages of anaerobic reactions. Differences of oxygen concentrations may also explain the above phenomenon. CH₄/CO₂ of A-2 was found to be lower than that of A-1 from January to June (Figure 2c). MSW in A-2 with higher oxygen concentrations may be attributed to the shorter landfill age. Powell et al. (2006) found that air injection could cause a decrease of CH₄/CO₂. In other words, lower oxygen concentrations are favorable for higher CH₄/CO₂. In addition, no similarity of variations was observed between rainfall and CH₄/CO₂, which should be attributed to the isolation function of HDPE membrane covers (Zhang et al., 2013). In conclusion, site A-1 could be inferred in a stable methane phase and site A-2 was in an unstable methane phase (Figure 2b) approximately. In addition, the CH₄/CO₂ ratio showed different sensitivities at A-1 and A-2. As reported by Capaccioni et al. (2011), the total emission rate appeared well correlated with the CH₄/CO₂ volumetric ratio. Precisely, the CH₄/CO₂ volumetric ratio increases up to typical values for a mature LFG (Capaccioni et al., 2011). In this study, CH₄/CO₂ at A-1 kept a stable value of about 2.00, whereas that at A-2 increased up to nearly 2.00 and then got into a stable state, as shown in Figure 2c. The phenomenon observed in Figure 2c showed great accordance with related literatures. Accordingly, the CH₄/CO₂ volumetric ratio could act as an excellent indicator for the characteristics of LFG generation.

Variation of H₂S, H₂, and O₂ contents

The variations of H₂S, H₂, and O₂ at site A-1 and site A-2 are illustrated in Figure 2d. The concentrations of H₂S ranged from 47.08 to 120.01 ppm and from 81.22 to 215.27 ppm for sites A-1 and A-2, respectively. It was observed that H₂S of A-2 presented lower concentrations than that of A-1 from January to June, but higher concentrations from July to December. Overall, H₂S of A-2 presented an increasing trend, whereas that of A-1 showed a decreasing trend. In general, H₂S concentration is remarkably higher than all the other sulfur compounds and therefore chosen as one of the typical LFG components for investigation (Kim et al., 2005, 2006; Kim, 2006). H₂S is produced by sulfate-reducing microorganisms, with a sulfate ion acting as a terminal acceptor and the reaction proceeds at neutral or slightly alkaline pH (Farquhar and Rovers, 1973). As shown in Figure 2d, H₂S concentration at A-2 presented an increasing trend with low O₂ concentrations, on account of H₂S generation in the anaerobic stage reasonably. Moreover, landfill age of MSW could also affect H₂S generation. Kim (2006) compared emissions of reduced sulfur compounds from young and old landfill facilities. The results showed that H₂S generation had an evident correlation with the landfill age.

For H₂ concentration, H₂ concentrations from both of site A-1 and site A-2 showed decreasing trends. It is obvious that H₂ concentration of A-2 was significantly higher than that of A-1 during the whole year. H₂ is a typical product of the hydrolytic process in acid phase, as shown in Figure 2b, and forms CH₄ and H₂O with the addition of CO₂ (Farquhar and Rovers, 1973), that is, H₂ is produced during the nonmethanogenic stage but consumed during the methanogenic stage. This could explain the downtrend of H₂ in the study period for A-1 and A-2. The differences of H₂ concentration between A-1 and A-2 could be resulted from the landfill ages and the stages of anaerobic reaction.

In addition, O₂ of A-2 presented higher concentrations than that of A-1 from January to June, but lower concentrations from July to December. As shown in Figure 2b, O₂ would be gradually consumed over time; therefore, the O₂ concentrations presented a decreasing trend in both sites. Higher concentrations of O₂ at site A-2 from January to June could be attributed to shorter landfill age, whereas lower concentrations at site A-2 from July to December may be explained by fractional air leakage at site A-1, which was caused by the sealing property. Chen et al. (2011) found that the sealing property of the HDPE layer would affect the test results at a landfill of Ningbo, China. Usually, the HDPE leakage aggravated with time, and site A-1 had longer landfill age. Hence, It is necessary to keep the maintenance of HDPE to guarantee the sealing property in the long run.

Relationships among LFG components and the external factors

As a simple means to investigate the factors affecting LFG emissions, we conducted a correlation analysis of LFG components with the measurement of other parameters (including climate conditions) concurrently. The results of this analysis made for the A-1 and A-2 sites are shown in Table 2, respectively. Both similarities and differences were observed between the two sites. It indicated that the climate parameters (ambient temperature and rainfall) exhibited strong correlations (all of P values were over 0.05) with LFG components at both site A-1 and site A-2. There have been several literatures about climate effects on LFG emissions (Tecle et al., 2009; Uyanik et al., 2012). The influences of atmospheric pressure on landfill methane emissions at landfills with soil covers have been investigated, and notable effects were pointed out by some researchers (Czepiel et al., 2003; Nwachukwu and Anonye, 2013). For 36 hr following a rainfall event, both CH₄ and N₂O emissions were significantly correlated with moisture content of the cover soil at the two landfill sites without HDPE membrane cover (Zhang et al., 2013). However, the correlations between LFG components and climate parameters were poor in this study, which should be attributed to the isolation function of HDPE membrane covers. It is critical because HDPE can reduce heat loss and injection of water and oxygen into landfills. Chen et al. (2011) found that after welding the HDPE geomembranes together to form a whole airtight layer upon a larger area of landfill, the gas flow in the general pipe increased 25% comparing with the design where the HDPE geomembranes were not welded together, which means that the gas extraction ability improved. Moreover, CH₄ contents showed excellent correlations with CO₂ at both sites (r = 0.952 and 0.727). The similar phenomenon was also observed in other landfills (Uyanik et al., 2012). The reaction in the anaerobic decomposition releases a very small amount of heat and the product gas mainly contains about 54% methane and 46% carbon dioxide (Themelis and Ulloa, 2007), which could reasonably explain the strong correlations observed above. The significant correlations between oxygen and major components (CH₄ and CO₂) at both sites are obvious during anaerobic reaction. Landfilled organic waste generates landfill gas (LFG) when it degrades in anaerobic environment (Niskanen et al., 2013).

Table 2. Correlation matrix between different parameters at site A-1

Parameter		CH4	CO2	H2S	H2	O2
Site A-1
CH4	r	1
	P
	N	12
CO2	r	0.952**	1
	P	0.000
	N	12	12
H2S	r	0.125	−0.030	1
	P	0.699	0.925
	N	12	12	12
H2	r	−0.315	−0.459	0.798**	1
	P	0.319	0.134	0.002
	N	12	12	12	12
O2	r	−0.922**	−0.935**	0.133	0.512	1
	P	0.000	0.000	0.679	0.089
	N	12	12	12	12	12
CH4/CO2	r	−0.180	−0.472	0.570	0.444	0.359
	P	0.575	0.122	0.053	0.149	0.252
	N	12	12	12	12	12
AT	r	−0.297	−0.127	−0.327	−0.395	−0.084
	P	0.348	0.693	0.228	0.204	0.795
	N	12	12	12	12	12
Rainfall	r	−0.351	−0.307	−0.129	−0.003	0.087
	P	0.264	0.332	0.690	0.993	0.788
	N	12	12	12	12	12
Site A-2
CH4	r	1
	P
	N	12
CO2	r	0.727**	1
	P	0.007
	N	12	12
H2S	r	0.730**	0.628*	1
	P	0.007	0.029
	N	12	12	12
H2	r	−0.921**	−0.671*	−0.745**	1
	P	0.000	0.017	0.005
	N	12	12	12	12
O2	r	−0.976**	−0.692*	−0.590*	0.896**	1
	P	0.000	0.013	0.044	.000
	N	12	12	12	12	12
CH4/CO2	r	0.971**	0.547	0.643*	−0.892**	−0.963**
	P	0.000	0.066	0.024	0.000	0.000
	N	12	12	12	12	12
AT	r	0.365	0.195	−0.227	−0.190	−0.488
	P	0.244	0.543	0.479	0.553	0.108
	N	12	12	12	12	12
Rainfall	r	0.055	0.281	−0.186	0.075	−0.069
	P	0.864	0.376	0.563	0.817	0.832
	N	12	12	12	12	12

Notes: AT = ambient temperature.*Significant at 95% confidence level. **Significant at 99% confidence level.

Nevertheless, there were still several differences for composition variations between the two sites. Firstly, H₂S and H₂ variations showed positive correlations at site A-1, but negative correlations at site A-2. Hydrogen was consumed during the methanogenic stage (Themelis and Ulloa, 2007) and showed a downtrend at both sites (Figure 2d). H₂S was produced in the methanogenic stage and reached a peak at a time, then the concentration would get into a slow downward trend (Kim, 2006). Kim (2006) found that the mean value of H₂S from a young landfill was 139.07 ppm, whereas its counterparts in an old landfill being 3.86 ppb. It indicated that H₂S generation at site A-1 was decreasing and that at site A-2 was increasing during the study period. Secondly, the correlations between H₂S and the major components are significantly positive (r = 0.730 for CH₄ and r = 0.628 for CO₂) at site A-2, but they exhibited strong correlations (r = 0.125 for CH₄ and r = −0.030 for CO₂) at site A-1. As pointed out by Kim (2006), no vital correlations were found between H₂S and major components (CH₄ and CO₂) at 5–6-yr-old landfills (stable methane phase). It roughly indicated that H₂S acted as a sensitive indicator of LFG emission only in the unstable methane phase. Finally, the CH₄/CO₂ ratio showed different sensitivities at sites A-1 and A-2. No significant correlation was observed between CH₄/CO₂ and major components at site A-1, whereas significant correlation was found at site A-2. The correlation analysis confirmed that it was possible for the operators to judge the characteristics of LFG generation by CH₄/CO_2.

Conclusions

As methane gas can be trapped and used as a green energy source, as is practiced in most of the developed countries, it is important to understand the characteristics, regularities, and mechanisms of the large-scale landfills for power generation applications. In this study, characteristics and annual variations of two sites with different landfill ages were investigated. It was found that the ambient temperature (AT) and rainfall exhibited strong correlations with LFG components at both sites (Table 2). CH₄ contents showed excellent correlations with CO₂ at both sites. H₂ concentrations from both sites showed decreasing trends. In addition, H₂S could acted as a sensitive indicator of LFG emission only in unstable methane phase (A-2). Especially, the ratio of CH₄/CO₂ showed completely different sensitivities to different phases of anaerobic reaction and could act as an excellent indicator for the characteristics of LFG generation. The study of LFG characteristics from sanitary landfills provides information about the field degradation of MSW, which in turn will favor the design and operation of LFG collection and electricity generation facilities.

Funding

This work was supported by Program of Natural Science Foundation of China (No. 51278212) and the Fundamental Research Funds for the Central Universities (WUT: 2015IVA026).

Additional information

Funding

This work was supported by Program of Natural Science Foundation of China (No. 51278212) and the Fundamental Research Funds for the Central Universities (WUT: 2015IVA026).

Notes on contributors

Lie Yang

Lie Yang and Yanyan Liu are lecturers of School of Resources and Environmental Engineering, Wuhan University of Technology.

Zhulei Chen

Zhulei Chen works for the School of Environmental Science & Engineering in the Huazhong University of Science and Technology.

Xiong Zhang

Xiong Zhang and Ying Xie are affiliated with the Wuhan Environmental Investment and Development Co., Ltd.

Yanyan Liu

Lie Yang and Yanyan Liu are lecturers of School of Resources and Environmental Engineering, Wuhan University of Technology.

Ying Xie

Xiong Zhang and Ying Xie are affiliated with the Wuhan Environmental Investment and Development Co., Ltd.