classification of municipal solid waste components for thermal conversion in waste-to-energy research

1. Classification of municipal solid waste components for thermal
conversion in waste-to-energy research
Hui Zhou, YanQiu Long, AiHong Meng, QingHai Li, YanGuo Zhang ⇑
Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Thermal Engineering, Tsinghua University, Beijing 100084, PR China
h i g h l i g h t s
MSW components are classified by
proximate and ultimate analyses.
MSW components are classified by TG
characteristics using cluster analysis.
TG characteristics of 26 kinds of MSW
components are discussed.
Typical components can be selected
for thermal conversion research.
g r a p h i c a l a b s t r a c t
Chinar
leaf
Poplar
leaf
Poplar
wood
Ginkgo
leaf
Celery
Pakchoi
Chinese
cabbage
Banana
peel
Spinach
Potato
Rice
Tangerine
peel
Orange
peel
Blank
printing
paper
Newspaper
Rubber
Tissue
paper
Cotton
cloth
Absorbent
cotton
gauze
PVC
Terylene
PET
HDPE
LDPE
PP
PS
a r t i c l e i n f o
Article history:
Received 6 August 2014
Received in revised form 30 November 2014
Accepted 4 December 2014
Available online 15 December 2014
Keywords:
Municipal solid waste
Proximate analysis
Ultimate analysis
Heating value
TGA
Classification
a b s t r a c t
Different researches selected different municipal solid waste (MSW) components to study their thermal
chemical characteristics in waste-to-energy research. Therefore, a specific classification is needed for the
research of thermal conversion of MSW components. In this paper, based on 26 kinds of MSW component
samples, MSW components were classified using cluster analysis method according to the proximate and
ultimate analyses and heating value results, as well as thermogravimetric (TG) characteristics. The clas-
sification groups include vegetables (including banana peel), starch food, orange peel, wood waste, print-
ing paper, cellulose, PVC, PET (including terylene), PE/PP, PS, and rubber. According to the above
classification, typical components can be selected for thermal conversion in waste-to-energy research.
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
Increasing amounts of municipal solid waste (MSW) are gener-
ated in recent years [1]. In China, the amount of municipal solid
waste has grown to 170.81 million tons in 2012 [2]. Traditional
landfill method is facing many problems such as land shortage
and underground water pollution [3]. Therefore, waste to energy
(WTE) methods, including incineration, pyrolysis and gasification
are drawing global attentions [4].
Municipal solid waste is a complicated mixture of food residue,
paper, plastics and some other components [5]. The research about
real MSW mixture is difficult to be repeated, because of the varia-
tion of MSW from region to region and time to time [6]. Besides,
lab-scale research usually takes gram-scale amount of MSW
[7,8], which means there may be some errors about the sampling
http://dx.doi.org/10.1016/j.fuel.2014.12.015
0016-2361/Ó 2014 Elsevier Ltd. All rights reserved.
⇑ Corresponding author. Tel.: +86 10 62783373; fax: +86 10 62798047 801.
E-mail address: zhangyg@tsinghua.edu.cn (Y. Zhang).
Fuel 145 (2015) 151–157
Contents lists available at ScienceDirect
Fuel
journal homepage: www.elsevier.com/locate/fuel

2. process. Therefore, many researches try to study the characteristics
of MSW single components [9–12].
Even the components such as food residue or plastics are to too
general for lab-scale research, the groups themselves are very com-
plex and may include subgroups with totally different properties
and need to be investigated [13]. For example, plastics may include
PE, PP, PS, PVC, and PET, whose thermochemical characteristics are
completely different [14]. Therefore, a more specific classification
is needed for the research of thermal conversion of MSW
components.
Different researchers choose different kinds of components for
thermal conversion research [9,11,15]. Li AM et al. selected paper,
paperboard, wood-chip, cotton cloth, vegetal, orange husk, PE, PVC,
and rubber to study the effects of heating methods, moisture con-
tents and size of waste on pyrolysis gas yields and compositions, as
well as heating values [15]. Sorum et al. chose 11 MSW compo-
nents (newspaper, cardboard, recycled paper, glossy paper, spruce,
plastics, HDPE, LDPE, PP, PS, PVC, multi-material, and juice carton)
to study the pyrolysis characteristics [16]. Jiang et al. chose plas-
tics, paper, cloth, wood, rubber, and leaves of vegetables to study
the combustion characteristics of MSW [17]. However, some kinds
of important components may be missed, and some components
may have similar characteristics that extra tests could be saved.
Therefore, typical MSW components should be selected for the
studies.
Proximate and ultimate analyses as well as heating value are
fundamental parameters for incineration, pyrolysis, and gasifica-
tion [18]. Proximate analysis provides the moisture, ash, volatile
and fixed carbon content, which impacts the drying, ignition and
ash disposal process of MSW. Ultimate analysis provides the ele-
mental compositions of fuel, which determines the gas products
of thermal conversion. Heating value is also very important for
the design of incinerators. Thermogravimetric analyzer (TGA) is
one of the most common techniques used to investigate the
thermal behavior of small fuel samples, with no limitations in heat
and mass transfer at low heating rates [19–22]. Obtained results
can be used to determine reactivity, which includes pyrolysis rate
(mass loss per time unit) and mass loss kinetics of the fuels. Mean-
while, the results of TGA can be easily obtained and usually have
very good repeatability.
Traditionally, MSW combustible fractions are divided into six
groups, i.e. food residue, wood waste, paper, textiles, plastics, and
rubber based on physical sources [23]. However, there is no classi-
fication of MSW based on thermochemical characteristics. In this
paper, 26 kinds of municipal solid waste components are selected.
According to the proximate and ultimate analysis and heating
value results, as well as TG characteristics, MSW components were
classified into several groups by cluster analysis method. There-
fore, typical MSW components can be selected for the research of
thermal conversion process.
2. Materials and methods
2.1. Materials
26 kinds of MSW components from six groups (food residue,
wood waste, paper, textiles, plastics, and rubber) are selected for
this study. The proximate and ultimate analyses as well as heating
value were carried out by China Coal Research Institute (CCRI) and
The Lab of Thermal Engineering, Tsinghua University. The proxi-
mate analysis is performed referring to GB/T 212. Carbon and
hydrogen content is measured according to GB/T 476; nitrogen
content is measured according to GB/T 19227; and total sulfur con-
tent is measured according to GB/T 214. The oxygen content is cal-
culated as the difference between 100% and the sum of other
elements. The heating value is determined according to GB/T
213. The proximate and ultimate analyses as well as heating value
Table 1
Proximate and ultimate analyses of MSW components.
Groups Samples Proximate analysis (wt%) Ultimate analysis (wt%) HHV (MJ/kg)
Ad
Vd
FCd
Cdaf
Hdaf
Odaf
Ndaf
Sdaf
Food residue Chinese cabbage (CHC) 9.91 67.60 22.49 47.49 5.88 41.79 4.11 0.73 16.99
Rice (RI) 0.40 84.42 15.18 45.97 6.35 45.74 1.69 0.25 18.14
Potato (PO) 3.15 79.52 17.33 44.41 5.33 47.82 1.81 0.64 17.10
Tangerine peel (TP) 2.91 76.49 20.60 48.74 5.92 43.83 1.43 0.08 18.47
Banana peel (BP) 10.85 64.38 24.77 35.80 4.79 54.93 4.37 0.10 16.39
Pakchoi (PA) 18.44 63.97 17.59 43.37 5.93 48.64 1.25 0.81 18.90
Celery (CE) 14.58 65.36 20.06 38.46 6.16 54.52 0.21 0.65 13.57
Orange peel (OP) 2.15 77.93 19.92 40.28 6.12 52.46 1.08 0.06 17.10
Spinach (SP) 15.97 65.26 18.77 47.58 6.48 43.93 1.57 0.43 17.08
Wood waste Poplar wood (PW) 7.54 73.85 18.61 51.36 5.89 41.00 1.52 0.22 18.50
Poplar leaf (PL) 15.69 68.74 15.57 49.54 5.24 43.30 1.32 0.59 16.85
Chinar leaf (CL) 9.23 69.74 21.03 52.95 4.88 40.51 1.01 0.65 19.12
Gingko leaf (GL) 11.62 73.19 15.19 41.35 5.54 50.88 1.36 0.87 15.28
Paper Blank printing paper (BPP) 10.69 79.33 9.98 45.12 5.31 48.91 0.38 0.28 13.51
Tissue paper (TIP) 0.52 90.47 9.01 45.18 6.13 48.32 0.25 0.11 17.25
Newspaper (NE) 8.07 79.54 12.39 48.01 5.71 45.86 0.33 0.09 17.16
Textiles Cotton cloth (CC) 1.52 84.53 13.95 46.51 5.80 46.98 0.43 0.28 17.43
Absorbent cotton gauze (ACG) 0.14 94.85 5.01 46.74 5.69 47.23 0.27 0.08 16.82
Terylene (TE) 0.49 88.60 10.91 62.16 4.14 33.12 0.29 0.28 20.86
Plastics PS 0.04 99.57 0.39 86.06 6.27 1.93 5.73 0.00 38.93
LDPE 0.00 99.98 0.02 85.98 11.20 2.61 0.21 0.00 46.48
HDPE 0.18 99.57 0.25 85.35 12.70 1.90 0.05 0.14 46.36
PVC 0.00 94.93 5.07 38.34 4.47 56.96a
0.23 0.00 20.83
PP 0.02 99.98 0.00 83.51 10.64 5.63 0.22 0.00 45.20
PET 0.09 90.44 9.47 63.01 4.27 32.69 0.04 0.00 23.09
Rubber Rubber (RU) 10.24 62.83 26.93 89.53 6.70 1.07 0.69 2.02 35.74
A: ash; V: volatile; FC: fixed carbon; HHV: high heating value; d: dry basis; daf: dry ash free basis.
a
It is Cl for PVC.
152 H. Zhou et al. / Fuel 145 (2015) 151–157

3. are shown in Table 1. To eliminate the impact of moisture, the
proximate analysis and high heating value (HHV) were expressed
as dry basis (dry at 105 °C). The ultimate analysis results were uni-
fied as dry ash free basis.
2.2. Experimental apparatus
The TGA experiments were performed by a NETZSCH STA 409C/
3/F with a flow rate of 100 ml min1
of N2. Temperature rose from
room temperature to 1000 °C at a heating rate of 10 °C min1
.
Repeated experiments showed that TG curves had good
reproducibility.
Before pyrolysis, the samples were dried at 105 °C to eliminate
the moisture. Powder samples were ground and sieved into parti-
cles with the diameter less than 250 lm; fiber samples were cut to
less than 5 mm, which are small enough to prevent heat transfer
effect in TG experiments.
3. Results and discussion
3.1. Proximate and ultimate analyses and heating value
The proximate and ultimate analyses of MSW components are
shown in Fig. 1. According to Fig. 1(a), these samples can be classi-
fied into five groups. The first group includes PS, LDPE, HDPE, and
PP, with almost 100% volatile and no ash or fixed carbon. The same
proximate results of PE, PP, PS, and PVC were also reported by oth-
ers [11,15,16]. The second group includes PVC, absorbent cotton
gauze, PET, tissue paper, terylene, cotton cloth, rice, and potato,
with low ash content and high volatile. Newspaper and blank
printing paper can be classified as a group, with approximate
80% volatile and 10% ash content. Orange peel and tangerine peel
can be classified as a group, with less than 3% ash content and
approximate 20% fixed carbon content. The others, including some
food residue samples, wood waste samples and rubber can be clas-
sified as the last group. The ash content varies from 7% to 19%; the
volatile varies from 63% to 74%; and the fixed carbon content varies
from 15% to 26%. The vegetal sample reported by Li AM et al. has
similar proximate composition [15].
The ternary chart of ultimate analysis results is shown in
Fig. 1(b). The carbon and hydrogen content is combined together;
and the nitrogen, sulfur and chlorine content are combined
together, as reported by Vassilev et al. [24]. The components can
be classified into 4 groups. PVC itself is one group, with 56.96%
chlorine content. LDPE, HDPE, PP, PS, and rubber are a group, with
very high carbon (more than 83%) and hydrogen (more than 6%)
content. Accordingly, the percentages of other elements are quite
low. The same elemental composition of PE and PP was also
reported by others [11,15,16]. Terylene and PET could be classified
as a group, with approximate 62% carbon, 4% hydrogen, 33% oxy-
gen, and a small amount of nitrogen and sulfur (less than 0.3%).
In fact, terylene and PET are different forms of the same chemical
structure. The other components, including food residue, wood
waste, paper, and textiles compose the last group, with 40–60%
C + H, 40–60% O, and less than 5% N + S + Cl. It is a group of biomass
or natural polymer. The newspaper, paperboard, wood-chip, and
cotton cloth reported in other research had similar characteristics
[15,16].
The classification of HHV of MSW components is shown in
Table 2. The HHV of all the samples varies from 10 to 50 MJ kg1
.
PP, HDPE, and LDPE have the highest HHV (more than 40 MJ kg1
).
The HHVs of rubber and PS are also very high (from 30 to
40 MJ kg1
). The HHV of PVC, terylene, and PET locate between
20 and 25 MJ kg1
. For other components, including food residue,
Chinese cabbage
Rice
Potato
Banana peel
Pakchoi
Celery
Orange peel
Poplar wood
Chinar leaf
Tissue paper
Terylene
PS
LDPE
HDPE
PVC
PP
PET
Rubber
Gingko leaf
0.4 0.5 0.6 0.7 0.8 0.9 1.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.0
0.2
0.4
0.6
Blank printing paper
Newspaper
Absorbent cotton gauze
Cotton cloth
Newspaper
Poplar leaf
Tanerine peel
Spinach
Banana peel
Orange peel
O
N
+
S
+
C
l
C+H
Celery
Chinese cabbage
Rice
Potato
Tangerine peel
Banana peel Pakchoi
Celery
Orange peel
Spinach
Poplar wood
Poplar leaf
Chinar leaf
Blank printing paper
Tissue paper
Newspaper
Cotton cloth
Terylene
PS
LDPE
HDPE
PVC
PP
PET
Rubber
Gingko leaf
0.0 0.1 0.2 0.3 0.4
0.0
0.1
0.2
0.3
0.4
0.6
0.7
0.8
0.9
1.0
V
F
C
A
PS, LDPE, HDPE, PP
Absorbent cotton gauze
(b) Ultimate analysis
(a) Proximate analysis
Fig. 1. Chemical compositions of MSW components.
Table 2
Classification of HHV of MSW components.
HHV 40 MJ kg1
HHV
30–40 MJ kg1
HHV 20–30 MJ kg1
HHV
10–20 MJ kg1
PP, HDPE, LDPE Rubber, PS PVC, terylene, PET Blank printing paper, celery, banana peel, absorbent cotton gauze, poplar leaf,
Chinese cabbage, spinach, potato, orange peel, newspaper, tissue paper,
cotton cloth, rice, tangerine peel, poplar wood, pakchoi, chinar leaf
H. Zhou et al. / Fuel 145 (2015) 151–157 153

4. wood waste, paper, and textiles, the HHV is in the range of 10–
20 MJ kg1
. The classification of HHV is similar to the classification
of ultimate analysis, as shown in Fig. 1(b).
3.2. Cluster analysis based on thermogravimetric characteristics
MSW components samples were tested using TGA to obtain the
thermogravimetric characteristics. Based on TG characteristics (TG
curves), the MSW components were classified using cluster analy-
sis method. N sets of data were exported from the TG curve, which
was regarded as a vector X or Y. N denotes the number of mass data
from a TG curve; X and Y denote the data group of mass between
100% and 0%, i.e.
X ¼ ðx1; x2; x3; . . . XNÞ; Y ¼ ðy1; y2; y3; . . . yNÞ ð1Þ
Euclidean distance was introduced to measure the difference
between TG curves quantitatively [25]. Euclidean distance is the
most common distance measure, which means the absolute dis-
tance of spots in multi-dimensional space.
distðX; YÞ ¼
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
X
N
i¼1
ðxi yiÞ2
v
u
u
t ð2Þ
During the cluster analysis process, data from 100 to 1000 °C
were exported from TG curves every 10 °C, which was good
enough to reflect the characteristics of TG characteristics. The
detailed Euclidean distance values between TG curves of different
components are shown in Table 3.
Between-groups linkage was adopted to perform the cluster
analysis. The cluster analysis was performed step by step: similar
samples will be clustered at first, and samples with largest differ-
ence will be clustered at last. All the analyses were performed by
SPSS software. The classification is shown in Fig. 2. The compo-
nents can be classified into 11 groups.
As shown in Fig. 2, four kinds of wood waste (chinar leaf, poplar
leaf, poplar wood, and ginkgo leaf) can be classified into a group.
The derivative thermogravimetric (DTG) curves of these compo-
nents are shown in Fig. 3(a). The main peak of these four compo-
nents is presented at 320–350 °C, and there is a shoulder peak at
Table 3
Detailed Euclidean distance values between TG curves of different components.
Euclidean distance (%) CHC RI PO TP BP PA CE OP SP PW PL CL GL BPP TIP NE CC ACG TE PS LDPE HDPE PVC PP PET RU
CHC
RI 79
PO 97 47
TP 49 61 63
BP 16 71 90 43
PA 46 103 131 92 50
CE 38 84 113 76 37 25
OP 100 81 52 58 98 143 128
SP 22 81 104 65 28 31 27 114
PW 53 52 77 55 49 71 60 92 50
PL 37 69 96 63 35 43 36 110 28 30
CL 44 76 104 71 43 44 42 117 33 33 11
GL 58 62 76 59 58 79 72 85 55 27 41 43
BPP 109 82 94 104 110 123 115 107 104 69 90 90 62
TIP 225 171 156 195 223 251 238 157 227 184 212 214 178 136
NE 137 95 96 120 135 156 146 106 134 93 119 119 86 41 99
CC 191 137 117 160 189 221 206 119 194 157 184 187 149 115 54 84
ACG 285 233 216 253 282 312 300 212 288 244 273 274 238 199 67 160 114
TE 189 181 190 190 190 194 196 194 181 156 166 159 145 149 200 145 199 240
PS 326 291 280 305 325 346 341 276 325 287 309 306 275 255 190 225 217 177 188
LDPE 326 312 308 319 329 336 339 303 321 296 308 303 280 272 264 257 274 278 180 203
HDPE 319 309 309 316 323 326 331 305 314 291 301 296 275 269 274 257 282 293 179 225 35
PVC 232 179 146 197 228 265 251 150 237 202 226 230 190 174 101 143 86 132 216 205 260 272
PP 322 302 297 311 324 335 335 293 318 289 303 299 273 263 241 243 256 249 158 153 73 101 240
PET 226 218 224 225 228 232 235 224 219 195 205 198 182 183 217 175 220 249 42 175 148 151 229 122
RU 112 135 165 136 110 92 100 178 99 94 85 76 104 132 246 155 231 298 143 303 302 293 274 294 182
Chinar leaf
Poplar leaf
Poplar wood
Ginkgo leaf
Celery
Pakchoi
Chinese cabbage
Banana peel
Spinach
Potato
Rice
Tangerine peel
Orange peel
Blank printing paper
Newspaper
Rubber
Tissue paper
Cotton cloth
Absorbent cotton gauze
PVC
Terylene
PET
HDPE
LDPE
PP
PS
a
b
c
d
e
f
g
h
i
j
k
Fig. 2. Dendrogram of the cluster analysis of MSW components based on TG
characteristics.
154 H. Zhou et al. / Fuel 145 (2015) 151–157

5. approximate 280 °C. Another peak at 700 °C is also identified.
Similar TG characteristics of spruce were also reported [16]. Wood
is composed of hemi-cellulose, cellulose, and lignin. The first peak
(shoulder peak) is derived from the decomposition of hemi-cellu-
lose [26], and the main peak is derived from the decomposition
of cellulose, which is also the main ingredient of wood [27]. The
peak at 700 °C is derived from the decomposition of lignin [19].
Five kinds of food residue components can be classified into a
group, as shown in Fig. 2. Celery, pakchoi, Chinese cabbage, and
spinach are vegetables, and banana peel is fruit peel. This group
has a peak at 295–320 °C, due to the decomposition of hemi-cellu-
lose and cellulose, and one or two shoulder peaks at 200–260 °C,
because the pyrolysis of hemi-cellulose [19].
Potato and rice can be classified as one group. They have only one
peak at approximate 300 °C. In fact, the main composition of potato
and rice is starch [28]. Two fruit peel, tangerine peel and orange peel
can be classified into one group. They have two peaks, as shown in
Fig. 3(d). The first peak is at 210–230 °C, because of the decomposi-
tion of pectin and hemi-cellulose [29]. The second peak is at 331–
333 °C, because of the decomposition of cellulose [30,31]. Similar
results of orange waste pyrolysis were also reported by Lopez-
Velazquez et al. [32]. Blank printing paper and newspaper can be
classified as one group, with the main peak at approximate 350 °C,
because of the decomposition of cellulose, the main composition
of paper [33–35]. Similar characteristics of uncoated printing and
writing paper were also reported by Chang et al. [36]. However,
blank printing paper has a peak at 721.7 °C, because of the decom-
position of CaCO3, which is added to smooth the paper [37,38].
The decomposition of rubber is different from that of other compo-
nents, with the main peak at 378.4 °C. Tissue paper, cotton cloth,
and absorbent cotton gauze can be classified as one group, with a
single peak at 333–353 °C, due to the decomposition of cellulose
[39,40].
PVC is special because of chlorine, as shown in Fig. 1(b). The
pyrolysis of PVC can be divided into two main stages. The first
stage is from 250 to 375 °C with a peak at 286.3 °C, because of
the dehydrochlorination process; the second stage is from 375 to
500 °C, with a peak at 469.7 °C (Fig. 3(h)), because of the decompo-
sition of hydrocarbon residue [12]. The similar result of PVC pyro-
lysis was also reported by Sorum et al. [16]. The chemical structure
of terylene and PET is the same, therefore their TG characteristics
are similar. They have a main peak at approximate 440 °C, with
very high intensity (20% min1
). Similar characteristics of PET
pyrolysis were also reported by other researches [41,42]. It should
be noted that terylene also has a tiny peak at 304.9 °C, maybe
because of the decomposition of impurity. HDPE, LDPE, and PP
have similar TG characteristics. Their pyrolysis process is simple,
with only one peak at 455–485 °C. Particularly, the characteristics
Fig. 3. DTG curves of different categories of MSW components based on thermogravimetric characteristics.
H. Zhou et al. / Fuel 145 (2015) 151–157 155

6. of HDPE and LDPE have little difference, which was also reported
by Sorum et al. [16]. The pyrolysis of PS is different with other plas-
tics, with the single peak at 413.9 °C (Fig. 3(k)), lower temperature
than that of HDPE, LDPE, and PP.
According to the results in Fig. 1, Table 2, and Fig. 2, the classi-
fication of MSW components can be obtained based on proximate
and ultimate composition, heating value, and thermogravimetric
characteristics. It should be noted that Fig. 1 and Table 2 provide
a more general classification, while TG characteristics provide a
more specific classification, which means TG characteristics give
more information about the thermochemical properties of the
components. TG characteristics may act as ‘‘fingerprint’’ to distin-
guish the components.
Since pyrolysis is a more fundamental process of thermal con-
version, the importance of the pyrolytic characteristics of a given
fuel relies not only on the pyrolytic products but also on the fact
that pyrolysis is the first chemical step in gasification or combus-
tion [27,43]. Therefore, Fig. 2 provides a classification for the
lab-scale research of not only pyrolysis, but also incineration and
gasification. Typical samples from the 11 groups in Fig. 2 can be
selected for the research of the thermal conversion of MSW, which
almost covers all the components in real MSW. However, gas prod-
ucts were not considered in this process, therefore more work
should be carried out in the future.
4. Conclusions
Generally, combustible MSW are classified into food residue,
wood waste, paper, textiles, plastics, and rubber six physical groups.
According to the proximate and ultimate analyses and heating value
results, as well as TG characteristics, MSW components are classi-
fied into 11 groups using cluster analysis method. For food residue,
vegetables and banana peel are a group, with multiple peaks below
320 °C; starch food is a group, with single peak at 300 °C; and tan-
gerine peel and orange peel are a group, with less than 3% ash con-
tent and approximate 20% fixed carbon content. Wood waste is a
group, with weak peak of hemi-cellulose, strong peak of cellulose
and another peak of lignin. Printing paper (including blank printing
paper and newspaper) is a group, with approximate 80% volatile and
10% ash content and the main DTG peak at 350 °C. Rubber itself is a
group. Tissue paper, cotton cloth, and absorbent cotton gauze are a
cellulose group. PVC is a group, because of the high chlorine content
and two stages pyrolysis. Terylene and PET are a group, with the
main peak at 440 °C. PE and PP are a group, with a very strong peak
at 455–485 °C, and PS is a group. According to the above classifica-
tion, typical components can be selected for thermal conversion in
waste-to-energy research.
Acknowledgements
The financial support from National Basic Research Program of
China (973 Program, No. 2011CB201502) and National Natural Sci-
ence Foundation of China (No. 21376134) are gratefully
acknowledged.
References
[1] He MY, Xiao B, Liu SM, Hu ZQ, Guo XJ, Luo SY, et al. Syngas production from
pyrolysis of municipal solid waste (MSW) with dolomite as downstream
catalysts. J Anal Appl Pyrol 2010;87:181–7.
[2] National Bureau of Statistics of China. China statistical yearbook,
2013. Beijing: China Statistics Press; 2013.
[3] Dong CQ, Jin BS, Li DJ. Predicting the heating value of MSW with a feed forward
neural network. Waste Manage 2003;23:103–6.
[4] Liu YS, Liu YS. Novel incineration technology integrated with drying, pyrolysis,
gasification, and combustion of MSW and ashes vitrification. Environ Sci
Technol 2005;39:3855–63.
[5] Zhou H, Meng A, Long Y, Li Q, Zhang Y. An overview of characteristics of
municipal solid waste fuel in China: physical, chemical composition and
heating value. Renew Sust Energ Rev 2014;36:107–22.
[6] Boer DE, Jedrczak A, Kowalski Z, Kulczycka J, Szpadt R. A review of municipal
solid waste composition and quantities in Poland. Waste Manage
2010;30:369–77.
[7] Zhang Y, Chen Y, Meng A, Li Q, Cheng H. Experimental and thermodynamic
investigation on transfer of cadmium influenced by sulfur and chlorine during
municipal solid waste (MSW) incineration. J Hazard Mater 2008;153:309–19.
[8] Velghe I, Carleer R, Yperman J, Schreurs S. Study of the pyrolysis of municipal
solid waste for the production of valuable products. J Anal Appl Pyrol
2011;92:366–75.
[9] Li AM, Li XD, Li SQ, Ren Y, Chi Y, Yan JH, et al. Pyrolysis of solid waste in a rotary
kiln: influence of final pyrolysis temperature on the pyrolysis products. J Anal
Appl Pyrol 1999;50:149–62.
[10] Luo SY, Xiao B, Hu ZQ, Liu SM. Effect of particle size on pyrolysis of single-
component municipal solid waste in fixed bed reactor. Int J Hydrogen Energ
2010;35:93–7.
[11] Zheng J, Jin YQ, Chi Y, Wen JM, Jiang XG, Ni MJ. Pyrolysis characteristics of
organic components of municipal solid waste at high heating rates. Waste
Manage 2009;29:1089–94.
[12] Zhou H, Meng A, Long Y, Li Q, Zhang Y. Interactions of municipal solid waste
components during pyrolysis: a TG-FTIR study. J Anal Appl Pyrol
2014;108:19–25.
[13] Zhou H, Meng A, Long Y, Li Q, Zhang Y. Classification and comparison of
municipal solid waste based on thermochemical characteristics. J Air Waste
Manage 2014;64:597–616.
[14] Williams PT, Williams EA. Interaction of plastics in mixed-plastics pyrolysis.
Energ Fuel 1999;13:188–96.
[15] Li AM, Li XD, Li SQ, Ren Y, Shang N, Chi Y, et al. Experimental studies on
municipal solid waste pyrolysis in a laboratory-scale rotary kiln. Energy
1999;24:209–18.
[16] Sorum L, Gronli MG, Hustad JE. Pyrolysis characteristics and kinetics of
municipal solid wastes. Fuel 2001;80:1217–27.
[17] Jiang F, Liu S, Wang HG, Pan ZG. Experimental study on combustion
characteristics of municipal solid waste. J Environ Sci-China 2003;15:482–6.
[18] Riber C, Petersen C, Christensen TH. Chemical composition of material
fractions in Danish household waste. Waste Manage 2009;29:1251–7.
[19] Zhou H, Long Y, Meng A, Li Q, Zhang Y. The pyrolysis simulation of five biomass
species by hemi-cellulose, cellulose and lignin based on thermogravimetric
curves. Thermochim Acta 2013;566:36–43.
[20] Chen T, Wu J, Zhang J, Wu J, Sun L. Gasification kinetic analysis of the three
pseudocomponents of biomass-cellulose, semicellulose and lignin.
Bioresource Technol 2014;153:223–9.
[21] Zhang J, Chen T, Wu J, Wu J. A novel Gaussian-DAEM-reaction model for the
pyrolysis of cellulose, hemicellulose and lignin. RSC Adv 2014;4:17513.
[22] Wu J, Chen T, Luo X, Han D, Wang Z, Wu J. TG/FTIR analysis on co-pyrolysis
behavior of PE, PVC and PS. Waste Manage 2014;34:676–82.
[23] Ministry of Housing and Urban-Rural Development of China. CJ/T 313-2009
sampling and analysis methods for domestic waste. 2009.
[24] Vassilev SV, Baxter D, Andersen LK, Vassileva CG. An overview of the chemical
composition of biomass. Fuel 2010;89:913–33.
[25] Janowitz MF. Ordinal and relational clustering. Singapore: World Scientific
Publishing Co., Pte. Ltd.; 2010.
[26] Yang H, Yan R, Chen H, Lee DH, Zheng C. Characteristics of hemicellulose,
cellulose and lignin pyrolysis. Fuel 2007;86:1781–8.
[27] Skreiberg A, Skreiberg Ø, Sandquist J, Sørum L. TGA and macro-TGA
characterisation of biomass fuels and fuel mixtures. Fuel 2011;90:2182–97.
[28] Dumitriu S. Polysaccharides: structural diversity and functional versatility.
2nd ed. Boca Raton, Florida: CRC Press; 2004.
[29] Fisher T, Hajaligol M, Waymack B, Kellogg D. Pyrolysis behavior and kinetics of
biomass derived materials. J Anal Appl Pyrol 2002;62:331–49.
[30] Zapata B, Balmaseda J, Fregoso-Israel E, Torres-García E. Thermo-kinetics study
of orange peel in air. J Therm Anal Calorim 2009;98:309–15.
[31] Miranda R, Bustos-Martinez D, Blanco CS, Villarreal MHG, Cantú MER.
Pyrolysis of sweet orange (Citrus sinensis) dry peel. J Anal Appl Pyrol
2009;86:245–51.
[32] Lopez-Velazquez MA, Santes V, Balmaseda J, Torres-Garcia E. Pyrolysis of
orange waste: a thermo-kinetic study. J Anal Appl Pyrol 2013;99:170–7.
[33] Soares S, Camino G, Levchik S. Comparative study of the thermal
decomposition of pure cellulose and pulp paper. Polym Degrad Stabil
1995;49:275–83.
[34] Wu CH, Chang CY, Tseng CH. Pyrolysis products of uncoated printing and
writing paper of MSW. Fuel 2002;81:719–25.
[35] Wu CH, Chang CY, Tseng CH, Lin JP. Pyrolysis product distribution of waste
newspaper in MSW. J Anal Appl Pyrol 2003;67:41–53.
[36] Chang CY, Wu CH, Hwang JY, Lin JP, Yang WF, Shih SM, et al. Pyrolysis kinetics
of uncoated printing and writing paper of MSW. J Environ Eng-Asce
1996;122:299–305.
[37] Wu CH, Chang CY, Lin JP. Pyrolysis kinetics of paper mixtures in municipal
solid waste. J Chem Technol Biot 1997;68:65–74.
156 H. Zhou et al. / Fuel 145 (2015) 151–157

7. [38] Zhao D, Dai Y, Chen K, Sun Y, Yang F, Chen K. Effect of potassium inorganic and
organic salts on the pyrolysis kinetics of cigarette paper. J Anal Appl Pyrol
2013;102:114–23.
[39] Abidi N, Cabrales L, Haigler CH. Changes in the cell wall and cellulose content
of developing cotton fibers investigated by FTIR spectroscopy. Carbohyd Polym
2013.
[40] Yin C, Li J, Xu Q, Peng Q, Liu Y, Shen X. Chemical modification of cotton
cellulose in supercritical carbon dioxide: synthesis and characterization of
cellulose carbamate. Carbohyd Polym 2007;67:147–54.
[41] Çepelioğullar Ö, Pütün AE. Thermal and kinetic behaviors of biomass and
plastic wastes in co-pyrolysis. Energ Convers Manage 2013;75:263–70.
[42] Girija BG, Sailaja RRN, Madras G. Thermal degradation and mechanical
properties of PET blends. Polym Degrad Stabil 2005;90:147–53.
[43] Di Blasi C. Combustion and gasification rates of lignocellulosic chars. Prog
Energ Combust 2009;35:121–40.
H. Zhou et al. / Fuel 145 (2015) 151–157 157