Waste Management xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Waste Management journal homepage: www.elsevier.com/locate/wasman

Characterisation of chemical composition and energy content of green waste and municipal solid waste from Greater Brisbane, Australia San Shwe Hla ⇑, Daniel Roberts CSIRO Energy, PO Box 883, Pullenvale, QLD 4069, Australia

a r t i c l e

i n f o

Article history: Received 19 December 2014 Accepted 30 March 2015 Available online xxxx Keywords: Australian urban wastes Chemical characterisation Moisture content Calorific value

a b s t r a c t The development and deployment of thermochemical waste-to-energy systems requires an understanding of the fundamental characteristics of waste streams. Despite Australia’s growing interest in gasification of waste streams, no data are available on their thermochemical properties. This work presents, for the first time, a characterisation of green waste and municipal solid waste in terms of chemistry and energy content. The study took place in Brisbane, the capital city of Queensland. The municipal solid waste was hand-sorted and classified into ten groups, including non-combustibles. The chemical properties of the combustible portion of municipal solid waste were measured directly and compared with calculations made based on their weight ratios in the overall municipal solid waste. The results obtained from both methods were in good agreement. The moisture content of green waste ranged from 29% to 46%. This variability – and the tendency for soil material to contaminate the samples – was the main contributor to the variation of samples’ energy content, which ranged between 7.8 and 10.7 MJ/kg. The total moisture content of food wastes and garden wastes was as high as 70% and 60%, respectively, while the total moisture content of non-packaging plastics was as low as 2.2%. The overall energy content (lower heating value on a wet basis, LHVwb) of the municipal solid waste was 7.9 MJ/kg, which is well above the World Bank-recommended value for utilisation in thermochemical conversion processes. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Australia’s urban waste streams are an untapped renewable energy resource. The space available for landfill is decreasing in our major cities, and the methane produced by landfilled municipal solid waste (MSW), green waste and biosolids is now recognised as a significant, long-term source of greenhouse gas emissions. Local authorities, state and federal governments, and the waste management industry now recognise opportunities in converting the energy in urban waste streams to renewable power or other energy products. There is a clear international precedent that modern waste-to-energy (WtE) plants are clean and efficient. With appropriate technology choice, it is technically feasible for a similar industry to be developed in Australia. This is evident by plans for two major projects in Western Australia to convert more than 200,000 tonnes of MSW into electricity annually (Pugh, 2014). Despite this encouraging activity, knowledge of the thermochemical properties of Australian waste streams is considerably lacking.

⇑ Corresponding author. Tel.: +61 7 3327 4125; fax: +61 7 3327 4455. E-mail address: [email protected] (S.S. Hla).

Such knowledge is critical for effective planning and development of WtE projects. Solid waste generated in Australia is usually classified into three main categories: municipal, commercial and industrial, and construction and demolition waste. An estimated 53 million tonnes of solid waste was generated from all sources in Australia during 2010–11 (Australian Bureau of Statistics, 2014). Of this, 27% was municipal (household) waste, equivalent to 14.3 million tonnes (with a per capita MSW generation rate of 660 kg/year). This is a significant increase compared with the per capita rate of 447 kg/year from just eight years earlier, in 2002–03 (Australian Bureau of Statistics, 2006). Globally, MSW is usually managed in four major ways: recycling, composting, landfilling, and WtE. Despite the significant amount of energy that could be recovered from urban waste streams as renewable energy, Australia uses only the first three methods to manage MSW. The country has no large-scale thermal treatment facilities for the disposal of non-hazardous MSW; the last MSW incineration plant shut down in 1997 (NSW Environment & Heritage, 2014), and an attempt to develop a solid waste energy recycling facility in Wollongong, New South Wales, failed, with the plant shut down in 2004 (URS Australia, 2010).

http://dx.doi.org/10.1016/j.wasman.2015.03.039 0956-053X/Ó 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Hla, S.S., Roberts, D. Characterisation of chemical composition and energy content of green waste and municipal solid waste from Greater Brisbane, Australia. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.03.039

2

S.S. Hla, D. Roberts / Waste Management xxx (2015) xxx–xxx

Nomenclature C Cl FW Gdn W GW H HHV LHV MC MSW N n

carbon content (wt.%) chlorine content (wt.%) food waste garden waste green waste hydrogen content (wt.%) higher heating value (MJ/kg) lower heating value (MJ/kg) moisture content (wt.%) municipal solid waste nitrogen content (wt.%) number of waste components used in Eq. (6) (9 for combustibles of MSW) NC non-combustibles O oxygen content (wt.%) O Plst other plastic OC other combustibles Pkg Plst packaging plastic Pkg Ppr packaging paper Pnt Ppr printing paper S sulphur content (wt.%)

Information about feedstock chemical properties and energy content is essential to the design and operation of any type of thermochemical conversion system, whether combustion or gasification-based. The energy content of MSW can be estimated based on average physical compositions using empirical models (Chang et al., 2007; Choi et al., 2008; Kathiravale et al., 2003; Lin et al., 2013; Liu et al., 1996). While this approach is quick and inexpensive, the downside is that the energy content of the type of organic waste in the country where the empirical model was developed is likely to differ significantly from that in the country where the model is applied. This variation is directly related to sociocultural properties; for example, differences in the amount and type of food wastes. Geographical and seasonal considerations also influence the quantity and type of waste generated in different countries. To avoid this uncertainty, waste samples should be systematically collected and prepared, and the energy content should be directly measured using standard laboratory apparatus such as a bomb calorimeter. Energy content can also be calculated from a sample’s ultimate analysis, which usually lists the carbon, hydrogen, oxygen, nitrogen, sulphur and ash content of the dry fuel on a weight percentage basis. Regular surveys have been conducted in Australia’s major cities to understand the physical composition of MSW, which is routinely managed by local councils. For example, Swales (2013) reported that an average MSW stream (samples collected from Brisbane City Council’s transfer stations in 2013) contained 53.3% of organic matter, 14.7% of plastic, 13% of paper, 4.2% of glass, 2.7% of metal, 11.6% of others and 0.5% of household hazardous. As MSW waste streams are landfilled according to the current waste management system, the surveys focus only on the quantity and distribution of wastes; determination of thermochemical characteristics is out of their scope. If WtE is to feature in strategic thinking and future planning, then the chemical characteristics of waste – in particular, the calorific value – become important. Researchers from developing and developed countries have reported their findings of chemical characteristics, including calorific values of MSW samples, via direct measurement; [e.g. Algeria (Guermoud et al., 2009), China (Zhou et al., 2014), Greece (Komilis et al., 2012), Greenland (Eisted and Christensen, 2011), India (Kumar and Goel, 2009), Jordan (Abu-Qudais and Abu-Qdais, 2000), Korea (Choi et al., 2008), Spain (Montejo et al., 2011),

Tex W WtE WW Xi

Yi Ymixture

Textiles weight of sample (g, kg) waste-to-energy wood waste weight fraction of respective component (dry basic fractions are used for dry-based values, wet basic fractions are used for wet-based values) property of each component of MSW property of mixed MSW

Subscript Af Ai c db f i Of Oi wb

air-dried sample (final) air-dried sample (initial) container dry basis final (including weight of container) initial (including weight of container) oven-dried sample (final) oven-dried sample (initial) wet basis

Taiwan (Lin et al., 2013), Turkey (Yildiz et al., 2013), UK (Parfitt and Bridgwater, 2008) and USA (Chin and Franconeri, 1980)]. However, Australian data for this research area are scarce. One of the challenges in analysing the chemical characteristics of MSW is the lack of a standard method for sample collection and preparation. While most researchers categorised and analysed the different physical components of MSW (e.g. food, paper, plastics, textiles, wood, glass, metals, etc.), Agrawal (1988) analysed only two fractions of MSW: combustible and non-combustible. Most researchers collected MSW samples directly from transfer stations, community bins and final disposal sites, sorting it into different categories later on (Brunner and Ernst, 1986; Chang et al., 2007; Gidarakos et al., 2006; Kumar and Goel, 2009; Yildiz et al., 2013), while some collected each category separately from different locations (Hanc et al., 2011; Katiyar et al., 2013; Komilis et al., 2012). The work presented here begins to address the lack of data for Australian waste streams by developing a method to characterise waste in terms of chemical composition and energy content, and applying the method to MSW and green waste from Brisbane, Australia.1 According to a Queensland Government’s report, in 2012 the city of Brisbane generated 780 thousands tons of MSW which was directed to landfill sites via transfer stations (Department of Environment and Heritage Protection, 2013). The green waste stream consists primarily of garden waste (e.g. prunings, grass clippings, trees, shrubs), and is particularly important given the large size of Brisbane’s catchment area and the region’s subtropical climate. Our work represents the first step towards understanding the relationship between the energy content and composition of MSW in Australia, as part of a wider characterisation of the WtE potential of priority urban waste streams. 2. Methods 2.1. Sampling of green waste Three samples of shredded green waste were collected from three Brisbane waste transfer stations over a 3-week period in February, which is towards the end of the warmer and wetter of 1 Brisbane is the capital city of Queensland, on Australia’s East Coast. Brisbane City Council is Australia’s largest local authority.

Please cite this article in press as: Hla, S.S., Roberts, D. Characterisation of chemical composition and energy content of green waste and municipal solid waste from Greater Brisbane, Australia. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.03.039

3

S.S. Hla, D. Roberts / Waste Management xxx (2015) xxx–xxx

Brisbane’s seasons. Sample collection followed the American Society for the Testing of Materials (ASTM) standard sampling procedure E 871-82. The gross sample was representative of the piles of shredded green waste, and samples were collected in nine increments from a foot or more below the surface at nine points covering the pile to eliminate effects of exposure to the environment. More than 30 kg of shredded green waste was collected for each sample. According to the standard, the quantity of sample should be large enough to be representative, but not less than 10 kg (ASTM, 2013). Samples were placed in airtight containers to prevent gains or losses in moisture from the atmosphere. 2.2. Sampling of MSW

2.3. Measurement of total moisture content of green waste The total moisture content of any solid wastes is one of the most significant variables affecting the actual heating value (energy content) of the material. The total moisture content of the green waste samples was determined using the two methods described below. In the first method, samples were immediately reduced to approximately 600–800 g using a ‘coning and quartering’ process. The operations of mixing, coning, and quartering are described in ASTM Practice D346 (ASTM, 2004). The moisture content of the reduced samples was determined according to ASTM standard method E 871-82 (ASTM, 2013), which involved placing the samples in open trays in an oven at 103 ± 1 °C until total weight changes were less than 0.2%. The standard states that samples should be placed in the oven for at least 16 h; in our study, the samples were left in the oven for at least 48 h to ensure they were completely dry. Following drying, the samples were removed from the oven and cooled in desiccators to room temperature. Samples were then weighed immediately to avoid moisture gain from the atmosphere. The process of heating, cooling and weighing the samples was repeated until total weight changes were less than 0.2%. For the green waste samples we tested, drying them in the oven for 48 h at 103 ± 1 °C was sufficient to remove all moisture. The moisture content (wet basis) of the green waste samples was calculated using Eq. (1):

ðW i  W f Þ  100 ðW i  W c Þ

Sample Material name type

Material details

Weight collected (kg)

01-FW Food waste 02-GW Garden waste 03-Pnt Printing Ppr paper 04-Pkg Packaging Ppr paper

Food scraps, kitchen waste Garden waste

8.79 1.59

Newspapers, magazines, books, printing & writing paper Package board, liquid paper containers, corrugated cardboard, disposable paper products, misc packaging, composite mostly paper High-density polyethylene, low-density polyethylene, polyethylene, polyvinyl chloride, polypropylene packaging and rigid polystyrene Expanded polystyrene, other plastic foam, other plastic film, composite mostly plastic, other plastic Clothing, textile, rags, leather Wood furniture, wood packaging, wood offcuts Nappies, rubber (footwear, tyres, tubes), cooking oil Glass, metals (ferrous and non-ferrous), other non-combustibles

3.43

05-Pkg Packaging Plst plastic

Because MSW is highly physically and chemically heterogeneous, sample collection is less straightforward than for green waste. As only a few grams of the sample is used for chemical analysis, it must be representative of the large waste pile in transfer stations. Samples are therefore collected as individual components, rather than entire MSW. In this study, MSW is classified into ten categories, which include nine combustible waste categories and a single group of non-combustibles, as shown in Table 1. This classification is based on analyses of MSW composition for the previous 13 years (Swales, 2013). Samples for each combustible category were randomly collected from five batches of MSW piles; the size of each batch was approximately 150 kg. Samples were stored in 25-L airtight plastic containers immediately after collection to prevent gains and losses in moisture from the atmosphere. Samples of inert materials (noncombustible) were not collected, but their quantity was recorded to adjust and calculate the chemical characteristics of the entire MSW stream. The total weight collected for all categories of MSW was approximately 33 kg.

MCð%wbÞ ¼

Table 1 Ten components of MSW categorised and collected in this study.

06-O Plst

Other plastic

07-Tex Textiles 08-WW Wood waste 09-OC 10-NC

Other combustibles Noncombustibles

Total (combustibles)

3.87

1.85

1.03

3.86 2.74 5.92 N/A 33.08

the amount of weight loss measured and recorded. The air-dried samples were then reduced to 600–800 g using the coning and quartering process described in ASTM Practice D346 (ASTM, 2004). The moisture content of the reduced shredded green waste sample was determined according to ASTM standard method E 871-82 (ASTM, 2013). The purpose of second method is to verify the reduction method (coning and quartering) applied in both methods. The moisture content (wet basis) of the green waste samples was calculated using Eq. (2):

2 4n MCð%wbÞ ¼

3 W Oi 1

ðW Ai W Af Þ ðW Ai Þ

n

o  W Of 5

W Oi 1

ðW Ai W Af Þ ðW Ai Þ

o

 100

ð2Þ

Step-by-step procedures for moisture content measurement for both methods are illustrated in Fig. 1. 2.4. Measurement of total moisture content for categorised MSW After the nine categories of MSW samples were collected, each was spread out and air-dried inside cages under cover to protect against wet weather and animals. After MSW samples were air-dried (from 3 to 28 days, depending on the components), the amount of weight loss was measured and recorded. The air-dried samples were cut into smaller pieces and dried further in an oven. The moisture content of oven-dried MSW samples was determined according to ASTM standard method E 871-82 (ASTM, 2013). The moisture content (wet basis) of each category of MSW sample was calculated using Eq. (2). 2.5. Sample preparation for laboratory analysis

ð1Þ

To measure moisture using the second method, the remainder of the sample was weighed and spread on a tarpaulin cloth inside a process bay. The samples were air-dried for at least two days and

Dried samples (500–800 g) of green waste and each category of MSW were milled using a cutting-type laboratory mill. Samples were reduced in size in four stages: dry samples were first milled into particles of 30 kg of total samples

Method 1 Method 2

Coning and Quartering

Remainder of the sample 600-800g of reduced sample Air-dry for at least 2 days

Coning and Quartering

Oven-dry for at least 2 days at 103±1°C

600-800g of reduced sample Calculate moisture content Oven-dry for at least 2 days at 103±1°C

Calculate moisture content

Fig. 1. Procedures of two different methods used to measure total moisture content of green waste samples.

Table 2 Chemical analyses and standard methods used in this study. Waste stream

Analysis

Parameters

Method of analysis

MSW

Proximate

Moisture Volatile matter Fixed carbon Ash C, H, N S O Cl

HRL Method 1.6, using a Leco MAC Analyser AS2434.2 Calculated [100(Moisture + Volatile Matter + Ash)] HRL Method 1.6, using a Leco MAC Analyser AS1038.6.4, using a Leco Truspec CHN Analyser CEN/TS 15289:2006(E) by ICP-OES Calculated [100(C + H + N + S + Cl + Ash)] CEN/TS 15289:2006(E) by titration

Moisture Volatile matter Fixed carbon Ash C, H, N S O Cl

SIS-CEN/TS 14774-3:2004 Standard I.S.CEN/TS 15148:2006 (modified) Standard Calculated I.S.CEN/TS 14775:2004 Standard ISO/TS 12902:2001 Standard CEN/TS 15289:2006(E) Standard by ICP-OES Calculated CEN/TS 15289:2006(E) Standard by ICP-OES

Ultimate

Green waste

Proximate

Ultimate

CEN = Committee for European Standardization; HRL = HRL Technology Pty Ltd; ICP–OES = inductively coupled plasma/optical emission spectrometry; I.S. = International Standard; ISO = International Organization for Standardization; SIS = the Swedish Standards Institute; TS = Technical specification.

milled into smaller sizes using 3, 2 and 1-mm sieves. Approximately 100 g of the sample passing the 1-mm sieve was then used for laboratory analysis.

2.6. Chemical analysis of waste samples The thermochemical characteristics of solid waste samples can be approximated by proximate analysis, ultimate analysis and determination of calorific value. Ultimate analysis is the measurement of carbon, hydrogen, nitrogen, oxygen and sulphur. As recommended by Liu and Liptak (1999), chlorine was also measured in this study, because it is the major contributor and the most important indicator for the formation of dioxins and furans. The analysis methods used are listed in Table 2. One of the main interests of this study is the heating (calorific) value of the waste samples. Any standard measurement of

heating value using laboratory apparatus such as a bomb calorimeter at constant volume returns the gross heating value (higher heating value on a dry basis, HHVdb). This is defined as the number of heat units evolved from complete combustion of a unit mass of sample, including the sensible heat and the latent heat of the water produced during combustion (CEN, 2010). In this study, the HHVdb of MSW samples was determined on a Leco AC350 calorimeter according to AS1038. The HHVdb of green waste samples was determined according to a method standardised by the Committee for European Standardization (CEN) (CEN, 2010). HHVdb data can be converted to data on a wet basis (HHVwb), or reported as a lower heating value (LHV; detailed below), using the moisture results from this work. To take moisture content into account, HHVwb is calculated using Eq. (3):

HHVwb ¼ HHVdb ð1  MC  0:01Þ

ð3Þ

Please cite this article in press as: Hla, S.S., Roberts, D. Characterisation of chemical composition and energy content of green waste and municipal solid waste from Greater Brisbane, Australia. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.03.039

5

S.S. Hla, D. Roberts / Waste Management xxx (2015) xxx–xxx

In most practical applications, the products of combustion leave the reactor at a temperature well above the condensation temperate of water vapour. As a result, the latent heat of condensation cannot be considered part of the energy content of the fuel. In addition, the gases evolving from combustion of solid fuel are usually able to expand, with the pressure of the system remaining constant. The heating value in this case, where the water remains as vapour under constant pressure, is designated as the LHV (CEN, 2010). LHV on a dry basis (LHVdb) is usually calculated from HHVdb using Eq. (4), according to the CEN standard method:

LHVdb ¼ HHVdb  0:2122  H  0:0008  ðO þ NÞ

ð4Þ

where H, O and N are mass percentages for hydrogen, oxygen and nitrogen from ultimate analysis, respectively, and units of heating values are in MJ/kg. When moisture content is taken into account, the LHV on a wet basis (LHVwb) can be calculated using Eq. (5):

LHVwb ¼ LHVdb ð1  MC  0:01Þ  2:443  MC  0:01

ð5Þ

Table 3 Moisture content (MC) of as-received green waste (GW) samples obtained from different transfer stations and measured using two methods. Sample

MC (%wb) by first method

MC (%wb) by second method

Closure (%)a

GW-01 GW-02 GW-03

28.6 34.9 45.7

28.7 35.5 46.0

99.6 98.3 99.4

a Percentage of closure; 100% closure would mean the two measurements matched exactly.

Table 4 Chemical analyses of as-received green waste samples. Analysis

Parameters

GW-01

GW-02

GW-03

Proximate

Total moisture (%wb) Volatile matter (%db) Fixed carbon (%db) Ash (%db)

28.7 65.5 20.3 14.2

35.5 71.1 21.0 7.9

46.0 68.1 20.4 11.5

Ultimate

C (%db) H (%db) N (%db) O (%db) S (%db) Cl (%db)

43.9 5.4 0.74 35.31 0.12 0.33

47.5 5.2 0.81 38.0 0.14 0.48

46 5.4 1.02 35.5 0.14 0.48

Energy content

HHVdb (MJ/kg) HHVwb (MJ/kg) LHVdb (MJ/kg) LHVwb (MJ/kg)

17.2 12.3 16.0 10.7

18.4 11.9 17.3 10.3

17.7 9.6 16.5 7.8

3. Results and discussion 3.1. Total moisture content of green waste As described in Section 2.3, the moisture content of green waste was measured using two different methods. The first is consistent with the standard ASTM method (ASTM E 871-82), while the second method was designed to check the accuracy of the representative sampling undertaken during the first method. Table 3 presents the moisture contents of three green waste samples measured by both methods. The agreement between the two methods is excellent, with a very low error (less than 2%). The total moisture content of green waste is generally controlled by two major factors: the original moisture content, and the variability in processing at the transfer station. The moisture content of three green waste samples collected during February, 2014 was found to be in the range of 29–46%. The total moisture content of GW-03 was higher than GW-01 and GW-02, because it was collected during a rainy week, highlighting the role and importance of external factors in these data.

Table 5 Chemical analyses of green waste samples on a dry, ash-free basis (daf). Analysis

Parameters

GW-01

GW-02

GW-03

Proximate

Volatile matter (%daf) Fixed carbon (%daf)

76.3 23.7

77.2 22.8

76.9 23.1

Ultimate

C (%daf) H (%daf) N (%daf) O (%daf) S (%daf) Cl (%daf)

51.17 6.29 0.86 41.15 0.14 0.38

51.57 5.65 0.88 41.23 0.15 0.52

51.98 6.10 1.15 40.09 0.16 0.52

Energy content

HHVdaf (MJ/kg) LHVdaf (MJ/kg)

20.05 18.68

19.98 18.75

20.00 18.67

3.2. Chemical analysis of green waste 3.3. Effect of moisture content on heating values of green waste Proximate analysis, ultimate analysis and energy content of green waste samples were determined using the methods described in Section 2. As expected, most chemical characteristics of green waste samples (presented in Table 4) are similar to those of other biomass fuels, such as wood chips (Reed and Das, 1988). However, the ash content of all green waste samples is significantly higher than most typical biomass fuels, which usually contain less than 2% ash. During sampling and sample preparation, we noticed that large amounts of soil were attached to the mulched green waste. The high ash (mineral matter) content in these green waste samples is therefore likely to be due to this soil material, rather than the samples’ inherent mineral matter. It is nonetheless important to recognise this material, because current green waste stream management systems do not remove soil material from green waste. To examine the chemical properties of green waste excluding the attached soil, the results are recalculated on a dry, ash-free basis (Table 5). Table 5 shows that chemical properties of the three different green wastes are very similar when the effect of ash content contributed by soil attachment is excluded, and are more consistent with other biomass fuels, such as wood chips or pellets.

Table 4 shows that the HHVdb of three different green wastes was in the range of 17.2–18.4 MJ/kg. When the energy content of three green wastes is presented in LHVwb, they are reduced accordingly to 7.8–10.7 MJ/kg. Although the energy content of GW-03 is slightly higher than that of GW-01 on a dry basis, it is lower than that of GW-01 on a wet basis. This reinforces the significant role of moisture content on the calorific value of solid wastes, and the importance of accurately accounting for variations arising from weather conditions, natural drying, or pre-treatment. This is consistent with the data presented by Komilis et al. (2014), who comprehensively investigated the effect of moisture content on the calorific value of several organic solid wastes. 3.4. Total moisture content and chemical analyse of MSW components As described in Section 2, the total moisture content of each component of MSW was measured using a two-step drying method. The time taken for air drying and the amount of sample used for oven drying vary according to the nature of the MSW components (e.g. moisture content, bulk density, odour) and the

Please cite this article in press as: Hla, S.S., Roberts, D. Characterisation of chemical composition and energy content of green waste and municipal solid waste from Greater Brisbane, Australia. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.03.039

6

S.S. Hla, D. Roberts / Waste Management xxx (2015) xxx–xxx

components, and as expected, the highest chlorine contents were measured in the plastics components. Food wastes were also found containing a significant amount of chlorine content. Heating values of each MSW component were measured and reported using the methods and equations described in Section 2.3. As expected, moisture content had a significant role to play: energy content of food waste was found to be reasonably high on a dry basis but was very low when moisture content was included in reporting of lower heating values. A similar relationship was observed for garden wastes due to their high moisture content. As expected, the energy content of the plastics categories are found highest because of their high carbon and hydrogen content, low ash content, and low moisture content especially for the ‘other plastics’ category. Higher heating value for paper categories were found lowest due to their low carbon content and high ash content.

Table 6 Raw data from measurement of total moisture content for components of MSW. Sample name

Material type

Time to air dry (days)

WAi (kg)

WAf (kg)

WOi (g)

01-FW 02-Gdn W 03-Pnt Ppr 04-Pkg Ppr 05-Pkg Plst 06-O Plst 07-Tex 08-WW 09-OC

Food waste Garden waste Printing paper

24 3 13

8.79 1.59 3.43

3.07 1.24 3.00

2272.00 1910.00 1144.01 590.48 708.79 649.47

Packaging paper Packaging plastic Other plastic Textiles Wood waste Other combustibles

10

3.87

3.13

1026.17

953.18

15

1.85

1.41

763.31

747.68

9 17 7 28

1.03 3.86 2.74 5.92

1.01 3.20 2.70 4.63

WOf (g)

906.52 903.33 778.66 750.64 1071.29 955.31 2886.00 2649.00

3.5. Chemical analyses and heating value of entire MSW amount of sample available (see Table 6). The total moisture content of the MSW samples as well as their chemical analyses are listed in Table 7. As expected, the highest moisture content was found in food waste (70%) followed by garden waste (60%). The moisture content of the other waste categories varied between 20% and 30%, except for the wood wastes (12%) and the other plastics waste, which had a particularly low moisture content of 2.2%. This is to be expected, because such plastics do not absorb any water or moisture. As presented in Table 7, volatile matter contents of most components were found between 70% and 80% except for plastic wastes which contain more than 90% volatile matter. The highest ash contents were found from the ‘other combustibles’ component, containing more than 20% ash mainly arising from rubber in footwear. High ash contents were also found in paper wastes and garden wastes. Proximate analyses give some insights into the major chemical components of waste streams and can give some indication of the likelihood of potentially troublesome behaviour in thermochemical systems. Carbon and hydrogen contents define the level of oxidation of a sample, and high amounts were found in both the plastic waste components (i.e. packaging plastics and other plastics). Oxygen contents of most components in MSW were found between 30% and 40% except for plastics wastes and the ‘other combustibles’ component. The nitrogen in MSW was mainly found in the organic based waste components such as food wastes and garden wastes, as well as in other plastics and textiles. Only small sulphur amounts were measured across all categories. Chlorine can be usually found in both or organic and inorganic MSW

To determine the overall characteristics of the entire MSW waste stream, all combustible components of the MSW sample were mixed according to their initial weight ratios using the data listed in Table 8. As this study only focuses on the combustible components of MSW, the relative component composition, excluding the noncombustibles, was also calculated (2nd rows of Table 8). Dried samples of MSW combustible components were mixed according to their weight ratios (last row of Table 8) and the chemical characteristics of the entire combustible mixture were then measured. In addition to this measurement, the chemical properties were calculated using their weight ratios and the individual properties of the components using Eq. (6):

Y mixture ¼

n X Y i  Xi

ð6Þ

i¼1

The measured and calculated results are compared in Table 9, along with the percentage of closure (100% closure means the calculation matched the measurement). The agreement between two methods was excellent, especially for the important heating values. Good agreements were also found for proximate and ultimate analyses, except for some minor components (e.g. N, S, Cl). This is due to their presence in very small amounts, and the resulting small denominators used to calculate the percentage of their closures. The HHVdb of the combustible portion of MSW was found to be as high as 22.5 MJ/kg, which is higher than many typical HHV values of wood and other organic fuels. This is mainly due to the

Table 7 Chemical analyses of each category of MSW samples. Analysis

Parameters

01-FW

02-GdnW

03-Pnt Ppr

04-Pkg Ppr

05-Pkg Plst

06-O Plst

07-Tex

08-WW

09-OC

Proximate

Total moisture (%wb) Volatile matter (%db) Fixed carbon (%db) Ash (%db)

70.6 77.1 16.9 6.1

59.9 69.9 21.8 8.3

19.8 74.8 15.5 9.7

25.0 75.4 12.4 12.2

25.6 91.5 4.6 3.9

2.2 97.8 0.9 1.3

20.1 81.7 16.9 1.4

11.9 80.8 18.2 1.0

28.2 72.3 7.3 20.4

Ultimate

C (%db) H (%db) N (%db) O (%db) S (%db) Cl (%db)

48.4 6.7 2.94 34.80 0.2 0.86

47.7 5.7 1.24 36.26 0.21 0.59

42.9 5.4 0.13 41.76 0.03 0.08

41.2 5.2 0.19 41.04 0.08 0.09

72.8 9.4 0.34 11.53 0.03 2.0

83.8 9.5 3.31 0.96 0.03 1.1

51.8 6.1 1.74 38.71 0.2 0.05

49.6 6.1 0.17 43.04 0.02 0.07

56.1 7.0 1.3 13.95 0.66 0.59

Energy content (MJ/kg)

HHVdb HHVwb LHVdb LHVwb

20.3 6.0 18.8 3.8

18.7 7.5 17.5 5.5

16.3 13.1 15.1 11.6

15.5 11.6 14.4 10.2

25.9 19.3 23.9 17.1

40.6 39.7 38.6 37.7

20.9 16.7 19.6 15.2

19.5 17.2 18.2 15.7

25.6 18.4 24.1 16.6

Please cite this article in press as: Hla, S.S., Roberts, D. Characterisation of chemical composition and energy content of green waste and municipal solid waste from Greater Brisbane, Australia. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.03.039

7

S.S. Hla, D. Roberts / Waste Management xxx (2015) xxx–xxx Table 8 Weight percentage makeup of reconstituted MSW sample. Parameters

01-FW

02-Gdn W

03-Pnt Ppr

04-Pkg Ppr

05-Pkg Plst

06-O Plst

07-Tex

08-WW

09-OC

10-NC

Composition (wt.%wb) Composition of combustibles (wt.%wb) Composition of combustibles (wt.%db)

29.44 34.80 20.26

26.92 31.85 25.27

7.31 8.53 13.54

4.46 5.27 7.83

1.79 2.12 3.11

6.20 7.33 14.20

2.41 2.85 4.51

2.63 3.12 5.43

3.47 4.11 5.84

15.47 N/A N/A

Table 9 Chemical characteristics of combustible mixtures of MSW sample (measured vs. calculated values). Analysis

Parameters

Measured from reconstituted sample

Calculated from individual sample

Closure (%)

Proximate

Total moisture (%wb) Volatile matter (%db) Fixed carbon (%db) Ash (%db)

N/A 77.4 15.1 7.6

49.5 78.4 14.5 7.2

N/A 98.7 104.1 105.6

Ultimate

C (%db) H (%db) N (%db) O (%db) S (%db) Cl (%db)

52.80 6.40 1.29 31.00 0.18 0.73

53.37 6.59 1.59 30.48 0.16 0.60

98.9 97.1 81.1 101.7 112.5 121.7

Energy content (MJ/kg)

HHVdb (MJ/kg) HHVwb (MJ/kg) LHVdb (MJ/kg) LHVwb (MJ/kg)

22.50 11.40 21.12 9.46

22.33 11.28 20.90 9.35

100.8 101.1 101.1 101.2

plastic wastes, which contain very high heating values, and their weight percentages in the mixture are moderate when it is calculated on a dry basis. As discussed, for practical applications, LHVwb should be used to calculate energy content of the waste. The LHVwb of the combustible portion of MSW was 9.46 MJ/kg (by direct measurement) and 9.35 MJ/kg (by calculation from individual properties). The energy content of the entire MSW (including non-combustible components) can therefore be calculated using Eq. (6) according to the components’ weight composition (first row of Table 8) and their LHVwb (last row of Table 7, assuming LHVwb of non-combustibles was zero). The energy content (LHVwb) of the entire MSW sample used in this study was found to be 7.9 MJ/kg. This is relatively high when compared with LHVwb of typical MSW from a range of different countries, e.g. 6 MJ/kg [Taiwan, (Chang et al., 2007)], 4.8 MJ/kg [India, (Kumar and Goel, 2009)], 2.85–6.71 MJ/kg [China, (Liu et al., 2006)] and [Algeria, 4.3 MJ/kg (Guermoud et al., 2009)]. It is also above the World Bank-recommended value (Rand et al., 2000), which suggests that the LHVwb of MSW should be on average of 7 MJ/kg, and must never fall below 6 MJ/kg for use in thermochemical conversion processes. However, it is found to be lower when it is compared with the average lower heating values of MSW reported from Japan (8.2–9.0 MJ/kg, (Tsukahara, 2012), Korea (8.16–11.92 MJ/kg, (Ryu and Shin, 2013), UK (9.22 MJ/kg, (Parfitt and Bridgwater, 2008) and USA (9.2 ± 0.96 MJ/kg, (Chin and Franconeri, 1980). It has to be noted that this study has been conducted in February which is wet season in Brisbane and high moisture content of MSW samples collected during this season affected significantly to LHVwb. 4. Conclusions To support the development and deployment of alternative, thermochemical strategies for urban waste management, green waste and MSW samples from a large local council were collected and analysed for their energy content and chemical composition. Apart from moisture and ash (mineral matter content), the chemical properties of the green waste samples collected from three different transfer stations were very similar. The energy content (LHVwb) of green waste ranged from 7.8–10.7 MJ/kg. The

heating values of green waste streams are mainly controlled by moisture content – which depends strongly on local weather conditions – and the amount of soil attached from the handling and sorting process, rather than by chemical composition. The chemical properties of the entire MSW sample and its nine combustible components were measured directly and compared with an overall MSW energy content determined using the weight proportion of different components and their chemical properties. The results obtained using both methods agreed very well. The major factors contributing to the energy content of MSW were moisture content, weight percentage of high-energy components (in this case, plastics) and weight percentage of non-combustibles. The energy content (LHVwb) of the MSW sample was found to be 7.9 MJ/kg, which is relatively high compared with average values of typical MSW from different countries, and above the World Bank-recommended energy minimum for waste-to-energy applications. This study only represents the sample of MSW collected in February. Waste compositions are known to be effected by weather and seasons. This type of survey should be done more frequently (preferable quarterly or at least twice a year) to obtain the year-round results. Acknowledgements The authors are grateful to Brisbane City Council for their financial support to conduct this study. The authors acknowledge that raw data used to calculate the initial weight ratios of MSW components was provided by EnvironCom (Brisbane). The authors are also thankful to EnvironCom (Brisbane) for their kind assistance and arrangement of MSW sample collection. The authors wish to thank Ms. Jing Xuan Yen for her assistance during sample collection and sample preparation of green wastes. References Abu-Qudais, M.d., Abu-Qdais, H.A., 2000. Energy content of municipal solid waste in Jordan and its potential utilization. Energy Convers. Manage. 41, 983–991. Agrawal, R.K., 1988. A rapid technique for characterization and proximate analysis of refuse-derived fuels and its implications for thermal conversion. Waste Manage. Res. 6, 271–280.

Please cite this article in press as: Hla, S.S., Roberts, D. Characterisation of chemical composition and energy content of green waste and municipal solid waste from Greater Brisbane, Australia. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.03.039

8

S.S. Hla, D. Roberts / Waste Management xxx (2015) xxx–xxx

ASTM, 2004. ASTM D346-04e1: Standard Practice for Collection and Preparation of Coke Samples for Laboratory Analysis. ASTM International, West Conshohocken, PA. ASTM, 2013. ASTM E871-82: Standard Test Method for Moisture Analysis of Particulate Wood Fuels. ASTM International, West Conshohocken, PA. Australian Bureau of Statistics, 2006. Australia’s Environment: Issues and Trends. (assessed on July, 2014). Australian Bureau of Statistics, 2014. Waste Account, Australia, 2010–11. (assessed on November, 2014). Brunner, P.H., Ernst, W.R., 1986. Alternative methods for the analysis of municipal solid waste. Waste Manage. Res. 4, 147–160. CEN, 2010. BS EN 14918:2009-Solid biofuels. Determination of calorific value. The Committee for European Standardization. Chang, Y.F., Lin, C.J., Chyan, J.M., Chen, I.M., Chang, J.E., 2007. Multiple regression models for the lower heating value of municipal solid waste in Taiwan. J. Environ. Manage. 85, 891–899. Chin, N., Franconeri, P., 1980. Composition and heating value of municipal solid waste in the spring creek area of New York City. In: 1980 National Waste Processing Conference. American Society of Mechanical Engineers, Washington D.C., USA, pp. 239–249. Choi, K.-I., Lee, S.-H., Lee, D.-H., Osako, M., 2008. Fundamental characteristics of input waste of small MSW incinerators in Korea. Waste Manage. 28, 2293– 2300. Department of Environment and Heritage Protection, 2013. State of waste and recycling in Queensland 2012. Queensland Government, pp. 36. Eisted, R., Christensen, T.H., 2011. Characterization of household waste in Greenland. Waste Manage. 31, 1461–1466. Gidarakos, E., Havas, G., Ntzamilis, P., 2006. Municipal solid waste composition determination supporting the integrated solid waste management system in the island of Crete. Waste Manage. 26, 668–679. Guermoud, N., Ouadjnia, F., Abdelmalek, F., Taleb, F., addou, A., 2009. Municipal solid waste in Mostaganem city (Western Algeria). Waste Manage. 29, 896–902. Hanc, A., Novak, P., Dvorak, M., Habart, J., Svehla, P., 2011. Composition and parameters of household bio-waste in four seasons. Waste Manage. 31, 1450– 1460. Kathiravale, S., Muhd Yunus, M.N., Sopian, K., Samsuddin, A.H., Rahman, R.A., 2003. Modeling the heating value of municipal solid waste. Fuel 82, 1119–1125. Katiyar, R.B., Suresh, S., Sharma, A.K., 2013. Characterisation of municipal solid waste generated by the City of Bhopal, India. Int. J. ChemTech Res. 5, 623–628. Komilis, D., Evangelou, A., Giannakis, G., Lymperis, C., 2012. Revisiting the elemental composition and the calorific value of the organic fraction of municipal solid wastes. Waste Manage. 32, 372–381.

Komilis, D., Kissas, K., Symeonidis, A., 2014. Effect of organic matter and moisture on the calorific value of solid wastes: an update of the Tanner diagram. Waste Manage. 34, 249–255. Kumar, K.N., Goel, S., 2009. Characterization of municipal solid waste (MSW) and a proposed management plan for Kharagpur, West Bengal, India. Resour. Conserv. Recycl. 53, 166–174. Lin, C.-J., Chyan, J.-M., Chen, I.M., Wang, Y.-T., 2013. Swift model for a lower heating value prediction based on wet-based physical components of municipal solid waste. Waste Manage. 33, 268–276. Liu, D.H.F., Liptak, B.G., 1999. Hazardous Waste and Solid. CRC Press, Taylor & Francis Group. Liu, J.I., Paode, R.D., Holsen, T.M., 1996. Modeling the energy content of multiple regression analysis. J. Air Waste Manag. Assoc. 46, 650–656. Liu, Z., Liu, Z., Li, X., 2006. Status and prospect of the application of municipal solid waste incineration in China. Appl. Therm. Eng. 26, 1193–1197. Montejo, C., Costa, C., Ramos, P., Márquez, M.d.C., 2011. Analysis and comparison of municipal solid waste and reject fraction as fuels for incineration plants. Appl. Therm. Eng. 31, 2135–2140. NSW Environment & Heritage, 2014. Environmental Issues: Management Strategies. (Assessed on November, 2014). Parfitt, J.P., Bridgwater, E., 2008. Municipal Waste Composition: A Review of Municipal Waste Component Analyses. Defra WR0119, London, UK. Pugh, J., 2014. Two large scale EfW projects in WA utilising MRF and low temperature gasification. In: National Energy from Waste Symposium, Lorne, Victoria, Australia. Rand, T., Haukohl, J., Marxen, U., 2000. Municipal Solid Waste Incineration – A Decision Makers’ Guide. The International Bank for Reconstruction and Development. The World Bank, Washington, D.C., USA. Reed, T.B., Das, A., 1988. Handbook of Biomass Downdraft Gasifier Engine System, second ed. The Biomass Energy Foundation Press, Golden, CO, USA. Ryu, C., Shin, D., 2013. Combined heat and power from municipal solid waste: current status and issues in South Korea. Energies 6, 45–57. Swales, L., 2013. 2013 Annual Waste Characterisation Survey. Waste Minimisation Team, Brisbane City Council. Tsukahara, M., 2012. Presentation of Japanese technology of waste to energy, JASEWorld Waste to Energy Sub WG. URS Australia, 2010. Final report: pre-feasibility assessment of a thermal conversion facility for the Australian Capital Territory. Department of the Environment, Climate Change, Energy and Water, ACT. Yildiz, S., Yaman, C., Demir, G., Ozcan, H.K., Coban, A., Okten, H.E., Sezer, K., Gorene, S., 2013. Characterization of municipal solid waste in Istanbul, Turkey. Environ. Prog. Sustain. Energy 32, 735–739. Zhou, H., Meng, A., Long, Y., Li, Q., Zhang, Y., 2014. Classification and comparison of municipal solid waste based on thermochemical characteristics. J. Air Waste Manag. Assoc. 64, 597–616.

Please cite this article in press as: Hla, S.S., Roberts, D. Characterisation of chemical composition and energy content of green waste and municipal solid waste from Greater Brisbane, Australia. Waste Management (2015), http://dx.doi.org/10.1016/j.wasman.2015.03.039

Characterisation of chemical composition and energy content of green waste and municipal solid waste from Greater Brisbane, Australia.

The development and deployment of thermochemical waste-to-energy systems requires an understanding of the fundamental characteristics of waste streams...
419KB Sizes 1 Downloads 14 Views