Classification and comparison of municipal solid waste based on thermochemical characteristics

Abstract

Municipal solid waste (MSW) has been normally sorted into six categories, namely, food residue, wood waste, paper, textiles, plastics, and rubber. In each category, materials could be classified further into subgroups. Based on proximate and ultimate analysis and heating value, statistical methods such as analysis of variance (ANOVA) and cluster analysis were applied to analyze the characteristics of MSW in every subgroup and to try to distinguish their relative properties. The chemical characteristics analysis of MSW showed that polyethylene (PE), polypropylene (PP), and polystyrene (PS) had the highest volatile matter content, with almost no ash and fixed carbon, while polyethylene terephthalate (PET) had high carbon content but low hydrogen content. Bones and vegetables had the highest ash content, while nutshells and rubber had the highest fixed carbon content. Paper and starch food had the highest oxygen content, and wool and bones had the highest nitrogen and sulfur content. Polyvinyl chloride (PVC) had the highest chlorine content at about 55%. PE, PP, and PS had the highest heating value, followed by chemical products such as rubber and chemical fiber. Conversely, paper, vegetables and bones had the lowest heating value. The results of cluster analysis of MSW components showed that fruit peel, weeds, wood, bamboo, leaves and nutshells could be classified as the lignocellulose category; starch food, cotton, toilet paper, printing paper and cardboard could be classified as the glucose monomer category; wood and chemical fiber could be classified as the high nitrogen and sulfur category; and PE, PP, and PS could be cluster as the polyolefin category.

Implications: 

The yield of municipal solid waste (MSW) is constantly increasing and waste to energy (WTE) has been used extensively all over the world. During the processes of incineration, pyrolysis, or gasification, the impact of physical and chemical properties of MSW is of great significance. However, the traditional classification of MSW is too general to provide more detailed information in many investigations. It is necessary to perform the investigation of characteristics of combustible MSW to distinguish different categories of MSW and find out their subclassification.

Introduction

Municipal solid waste (MSW) generated in China has grown from 108.25 million tons in 1996 to 163.95 million tons in 2011 (National Bureau of Statistics of China [NBSC], 2012). Severe environmental problems will occur if the MSW cannot be disposed of properly (Tai et al., 2011). The traditional landfill method is facing a land shortage crisis (Dong et al., 2003), and waste to energy (WTE) methods such as incineration, pyrolysis, and gasification are drawing increasingly global concern (Liu and Liu, 2005).

Traditionally, MSW combustible fractions are divided to food residue, wood waste, paper, textiles, plastics, and rubber, six groups (MHUDC, 2009). Figure 1 shows the mean physical compositions of MSW in Chinese cities. The average physical combustible and noncombustible fractions of the MSW were 81.64% and 18.36%, respectively. In combustible MSW, the contents of food residue, plastics, paper, textiles, wood waste, and rubber, in decreasing order, were 55.86%, 11.15%, 8.52%, 3.16%, 2.94%, and 0.84%.

Figure 1. The mean physical compositions of MSW in China.

However, the classification of MSW is too general to provide more detailed information in many investigations. Many research studies were carried out using samples called food residue or plastics (Guo et al., 2001; Watanabe et al., 2004; Luo et al., 2010), while the groups themselves are very complex and may include subgroups of totally different properties. Plastics, for example, include polyethylene (PE), polystyrene (PS), and polyvinyl chloride (PVC), whose characteristics are completely different. Therefore, it is necessary to perform an investigation on characteristics of combustible MSW to distinguish different categories of MSW and find out their subclassification.

The proximate and ultimate analysis and heating value of MSW are fundamental parameters for incineration, pyrolysis, and gasification (Riber et al., 2009). This paper focuses on these thermochemical properties of the specific components of MSW. Analysis of variance (ANOVA) was adapted to analyze whether significant difference existed in MSW groups. Cluster analysis was carried out to classify MSW specific components. Therefore, some representatives can be selected during the experimental research of MSW.

Data and Discussion

Characterization and classification of specific components

Food residue

Food residue prevails in MSW categories, and is normally divided to five5 subgroups, namely, vegetables, fruit peel, bones, starch food, and nutshells. The proximate and ultimate analysis and heating value results of food residue are shown in Table 1. To eliminate the impact of moisture and ash, the ultimate analysis results were unified into dry ash free basis. Proximate analysis and higher heating value (HHV) were expressed on a dry weight basis.

Table 1. Chemical characteristics of food residue

		Proximate analysis (wt%)			Ultimate analysis (wt%)
Food residue subgroup and variety		A^d	V^d	FC^d	C^daf	H^daf	O^daf	N^daf	S^daf	Cl^daf	HHV^d (kJ/kg)	Reference used
Vegetable
1.	Chinese cabbage	20.01	71.76	8.23	49.68	5.08	40.51	4.07	0.66		15784	Zhang, 2005
2.	Potato				42.09	6.50	49.06	2.12	0.23		16912	Liu et al., 1999
3.	Pea				42.13	5.88	48.62	3.14	0.22		16533	Liu et al., 1999
4.	Scallion				48.12	6.27	41.74	3.09	0.78		18042	Liu et al., 1999
5.	Lettuce	17.19	80.45	2.36								Zhang et al., 2008a
6.	Vegetable	19.10	65.45	15.45	41.30	3.10	49.58	5.36	0.66			Li et al., 1999b
7.	Vegetable	19.96	65.06	14.99	41.14	4.85	48.46	4.39	1.16			Dai, 2007
8.	Vegetable				50.31	6.71	39.83	2.73	0.42			Zhao et al., 2011
Mean		19.07	70.68	10.25	44.97	5.49	45.40	3.56	0.59		16818
Minimum		17.19	65.06	2.36	41.14	3.10	39.83	2.12	0.22		15784
Maximum		20.01	80.45	15.45	50.31	6.71	49.58	5.36	1.16		18042
Fruit peel
9.	Orange peel	2.44	76.27	21.29	43.93	5.64	48.93	1.34	0.07	0.08	18097	Zhang et al., 2011b
10.	Orange peel	2.76	82.51	14.73	52.94	7.42	38.19	1.38	0.08		18231	Li et al., 1999b
11.	Orange peel	2.91	76.49	20.60	48.74	5.92	43.72	1.43	0.19		18470	Zhou et al., 2013
12.	Banana peel	8.81	75.37	15.82								Zhang et al., 2008a
13.	Fruit peel	3.13	77.68	19.20	50.15	5.82	42.69	1.14	0.20			Li et al., 2005
14.	Fruit peel	3.29	77.93	18.78	48.50	6.20	39.50	1.30	0.20		18638	Nie, 2000
15.	Fruit peel				47.58	6.17	45.47	0.78				Zhao et al., 2006
16.	Fruit peel										16887	Zou et al., 2006
Mean		3.89	77.71	18.40	48.64	6.19	43.08	1.23	0.15	0.08	18065
Minimum		2.44	75.37	14.73	43.93	5.64	38.19	0.78	0.07	0.08	16887
Maximum		8.81	82.51	21.29	52.94	7.42	48.93	1.43	0.20	0.08	18638
Bone
17.	Rib				55.92	8.83	25.63	9.29	0.32		17277	Liu et al., 1999
18.	Rib	38.22	61.56	0.23	51.61	6.38	31.91	9.48	0.62		14034	Li et al., 2001b
19.	Fish bone	39.82	56.25	3.93	63.87	8.01	19.08	8.39	0.64		15794	Shen et al., 2006
20.	Meat and bone	30.87	59.90	9.23	60.7	5.5	24.9	7.5	1.4			Ren et al., 2011
Mean		36.30	59.23	4.46	58.02	7.18	25.38	8.67	0.74		15701
Minimum		30.87	56.25	0.23	51.61	5.50	19.08	7.50	0.32		14034
Maximum		39.82	61.56	9.23	63.87	8.83	31.91	9.48	1.40		17277
Starch food
21.	Flour				42.78	6.19	48.39	2.48	0.15		18157	Liu et al., 1999
22.	Flour	0.62	90.59	8.79	43.47	6.74	46.76	2.93	0.1			Zhang et al., 2008b
23.	Rice	0.40	84.42	15.18	45.97	6.35	45.74	1.69	0.25		18140	*
24.	Rice	0.42	87.74	11.84	44.20	5.73	48.75	1.20	0.10	0.02	17972	Zhang et al., 2007a
25.	Rice				40.64	6.59	50.96	1.68	0.14		17802	Liu et al., 1999
26.	Rice	1.11	86.45	12.45	43.41	6.29	48.44	1.72	0.13		18095	Zhang, 2005
Mean		0.64	87.30	12.07	43.41	6.32	48.17	1.95	0.14	0.02	18033
Minimum		0.40	84.42	8.79	40.64	5.73	45.74	1.20	0.10	0.02	17802
Maximum		1.11	90.59	15.18	45.97	6.74	50.96	2.93	0.25	0.02	18157
Nutshell
27.	Chestnut shell	2.79	73.99	23.22							17785	Wang, 2007
28.	Peanut shell	1.82	73.88	24.31								Hu, 2010
29.	Peanut shell	2.10	74.34	23.57	51.98	6.88	40.10	0.83	0.21			Fang et al., 2010
30.	Peanut shell				53.60	6.70	38.30	1.20	0.20		20598	Zhang et al., 2011a
31.	Walnut shell	1.45	74.73	23.81	52.55	6.42	39.48	1.55				Wang and Li, 2008
Mean		2.04	74.23	23.73	52.71	6.67	39.29	1.19	0.20		19192
Minimum		1.45	73.88	23.22	51.98	6.42	38.30	0.83	0.20		17785
Maximum		2.79	74.73	24.31	53.60	6.88	40.10	1.55	0.21		20598

Notes: C: carbon content; H: hydrogen content; O: oxygen content; N: nitrogen content; S: sulfur content; Cl: chlorine content; A: ash content; V: volatile content; FC: fixed carbon content; daf: dry ash free basis; d: dry basis; *: the data without a reference were from our own research.

The proximate and ultimate analysis results of food residue varieties are plotted in Figure 2. Starch food, nutshells, and fruit peel had similar proximate analysis characteristics. The proximate analysis of vegetables and bones showed significant difference from other food residue subgroups. Bones had the highest ash content, followed by vegetables. Starch food had the highest volatile matter, followed by fruit peel. Based on elemental compositions, vegetables, fruit peel, starch food, and nutshells were close to each other, while bones had high C + H content, low O content, and high N + S + Cl content.

Figure 2. Chemical compositions of food residue specific components: (a) proximate analysis; (b) ultimate analysis.

The data in Table 1 are nearly normally distributed. Therefore, ANOVA was applied to investigate statistical significance of grouping the food residue, as shown in Table 2. The statistical significance level α was set as 0.05, which is commonly used in statistical analysis. Therefore, when the significance P was larger than α, there was no significant difference in food residue subgroups; when significance P was smaller than α, a significant difference existed among food residue subgroups (Agresti and Franklin, 2013). As shown in Table 2, there were significant differences in food residue subgroups for all variables.

Table 2. ANOVA of thermochemical properties in food residue subgroups

		Sum of squares	Degree of freedom	Mean square	F	Significance (P)
A	Between groups	2935.220	3	978.407	116.644	0.000
	Within groups	134.208	16	8.388
V	Between groups	1295.676	3	431.892	16.536	0.000
	Within groups	417.901	16	26.119
FC	Between groups	432.929	3	144.310	8.988	0.001
	Within groups	256.884	16	16.055
C + H	Between groups	597.594	3	199.198	8.813	0.000
	Within groups	542.456	24	22.602
O	Between groups	1324.556	3	441.519	22.209	0.000
	Within groups	477.131	24	19.880
N + S + Cl	Between groups	212.084	3	70.695	90.147	0.000
	Within groups	18.821	24	0.784
HHV	Between groups	2.147E7	3	7157786.070	4.409	0.021
	Within groups	2.435E7	15	1623398.498

Table 3. Chemical characteristics of wood waste subgroups and varieties.

Wood waste subgroup and variety		Proximate analysis (wt%)			Ultimate analysis (wt%)						HHV^d (kJ/kg)	Reference used
		A^d	V^d	FC^d	C^daf	H^daf	O^daf	N^daf	S^daf	Cl^daf
Wood
1.	China fir	3.53	82.04	14.44	51.47	6.89	41.08	0.40	0.16		20007	Song et al., 2005
2.	Dry wood	2.38	81.86	15.76	48.37	6.17	45.01	0.42	0.03			Li et al., 2002
3.	Pine	0.35	85.64	14.01	46.06	5.97	47.84	0.10	0.03			Wang et al., 2008
4.	Pine wood	0.95	83.50	15.54	50.51	5.95	43.39	0.11	0.03		19646	Tang et al., 2003b
5.	Poplar scraps	0.95	86.38	12.67	49.06	6.53	42.95	1.45	0.01			Wang and Li, 2008
6.	Sawdust	0.42	81.00	18.58	49.42	7.26	42.92	0.39	0.01		21178	Hu and Huang, 2009
7.	Sawdust	0.18	80.93	18.89	54.35	5.99	39.40	0.26	0.00			Li et al., 1996
8.	Sawdust	1.14	87.26	11.60	54.86	6.31	38.53	0.28	0.02			Zhou et al., 2001
9.	Sawdust	1.70	80.70	17.60	52.70	6.30	40.70	0.20	0.10			Zhou et al., 2007
10.	Sawdust	0.61	84.24	15.15	50.40	5.49	44.08	0.02	0.01			Jin et al., 2007
11.	Sawdust				51.63	4.62	42.66	0.83	0.17	0.09		Qing et al., 2005
12.	Spurce chips	0.36	85.00	14.64	47.40	6.10	46.40	0.10	0.10			Fang et al., 2010
13.	Tree branches	0.81	84.10	15.09	51.41	6.25	42.14	0.10	0.10			Li et al., 2005
14.	Wood	3.70	79.45	16.85	50.61	7.45	41.86	0.08	0.00		18126	Chen et al., 2012
15.	Wood	0.82	81.64	17.54	48.35	6.62	44.70	0.04	0.29		20697	Wu et al., 2009
16.	Wood				54.64	6.54	37.08	0.96	0.69	0.08		Jiang et al., 2004
17.	Wood				50.53	5.80	42.32	0.41	0.05	0.89		Liu, 2007
18.	Wood	4.20	80.50	15.30	46.49	6.56	45.67	1.00	0.29			Guo et al., 2001
19.	Wood chips	1.95	82.66	15.40	49.54	6.21	44.06	0.12	0.04	0.03	19163	Zhang et al., 2011b
20.	Wood chips	1.49	87.07	11.45	51.73	4.54	43.50	0.16	0.08		20186	Zheng et al., 2009
21.	Wood chips	3.45	81.50	15.05	49.03	5.69	44.98	0.22	0.07		18591	Li et al., 1999b
22.	Wooden chopsticks	2.18	83.45	14.37	48.79	5.16	45.70	0.30	0.04		18933	Xiao et al., 2009
Mean		1.64	83.10	15.26	50.33	6.11	43.04	0.36	0.11	0.27	19614
Minimum		0.18	79.45	11.45	46.06	4.54	37.08	0.02	0.00	0.03	18126
Maximum		4.20	87.26	18.89	54.86	7.45	47.84	1.45	0.69	0.89	21178
Bamboo
23.	Bamboo				47.81	5.55	45.89	0.70	0.06		19868	Liu et al., 1999
24.	Bamboo				46.88	6.11	46.73	0.28				Shao et al., 2006
25.	Bamboo	1.79	81.36	16.84	51.42	6.01	41.92	0.36	0.29		19616	Xiao et al., 2007
26.	Bamboo	0.69	81.03	18.27	50.46	6.32	42.73	0.22	0.10	0.16	19580	Liang et al., 2012
27.	Bamboo	1.96	80.56	17.48	50.76	5.91	42.98	0.28	0.07		19339	Yan et al., 2005
28.	Moso bamboo				48.89	5.98	44.86	0.27				Shao et al., 2006
Mean		1.48	80.99	17.53	49.37	5.98	44.18	0.35	0.13	0.16	19601
Minimum		0.69	80.56	16.84	46.88	5.55	41.92	0.22	0.06	0.16	19339
Maximum		1.96	81.36	18.27	51.42	6.32	46.73	0.70	0.29	0.16	19868
Leaves
29.	Chinar leaves				52.95	4.88	40.83	1.01	0.33		19120	Zhou et al., 2013
30.	Chinar leaves	7.78	75.35	16.87								Dong et al., 2002
31.	Fallen leaves	6.63	74.48	18.90								Li et al., 1998
32.	Leaves	9.43	74.32	16.25	47.18	5.61	46.35	0.18	0.68		18366	Luo, 2010
33.	Leaves	8.92	73.70	17.38	47.25	5.57	46.26	0.19	0.73		17198	Zhu, 2011
34.	Leaves	7.09	76.52	16.38							22599	Zhang, 2002
35.	Leaves	7.19	75.55	17.26								Zhang et al., 2008a
36.	Pine needles				52.57	6.30	40.44	0.54	0.16		20843	Xiao et al., 2005
37.	Poplar leaves				49.54	5.24	43.59	1.32	0.31		16850	Zhou et al., 2013
Mean		7.84	74.99	17.17	49.90	5.52	43.49	0.65	0.44		19163
Minimum		6.63	73.70	16.25	47.18	4.88	40.44	0.18	0.16		16850
Maximum		9.43	76.52	18.90	52.95	6.30	46.35	1.32	0.73		22599
Weeds
38.	King grass				48.37	6.30	44.58	0.49	0.25		22127	Pu et al., 2006
39.	King grass	7.44	74.12	18.43	46.91	5.89	46.30	0.70	0.21		17983	Teng et al., 2011
40.	Miscanthus	4.52	78.79	16.69	56.41	7.27	35.58	0.75				Fu et al., 2009
41.	Pennisetum	6.05	78.00	15.95	52.59	7.08	39.22	1.11				Fu et al., 2009
42.	Pennisetum				47.81	6.32	45.36	0.40	0.11			Liu et al., 2012
Mean		6.00	76.97	17.03	50.42	6.57	42.21	0.69	0.19		20055
Minimum		4.52	74.12	15.95	46.91	5.89	35.58	0.40	0.11		17983
Maximum		7.44	78.79	18.43	56.41	7.27	46.30	1.11	0.25		22127

Cluster analysis was applied to classify the food residue subgroups. The variables included ash content (A), volatile content (V), fixed carbon content (FC), C, H, O, N, S, and HHV. All the variables were standardized to a value between 0 and 1. The linkage between groups method was employed as the cluster method. Calculation of distance between any two objects was based on the Euclidean distance (Janowitz, 2010). The critical clustering coefficient was set as 1.5, which was suitable for MSW subgroups clustering after many tests. Therefore, the cluster happened with and only with a clustering coefficient less than 1.5. All statistical analyses were performed by SPSS software. The results showed that food residue could be classified into two characteristic groups: (i) vegetables, fruit peel, starch food, and nutshells, and (ii) bones.

Wood waste

Wood waste in MSW can be divided into four subgroups: wood, bamboo, leaves, and weeds. The proximate and ultimate analysis and heating value results of wood waste are shown in Table 3.

The proximate and ultimate analysis results of wood waste are plotted in Figure 3. Weeds and leaves had lower volatile matter and higher ash than wood and bamboo. The N + S + Cl content of wood waste varied from 0% to 5%. Significant difference in elemental composition was not observed for these subgroups. The HHV also showed little difference.

Figure 3. Chemical composition of wood waste specific components: (a) proximate analysis; (b) ultimate analysis.

The ANOVAs of proximate and ultimate analysis and HHV among wood waste subgroups are shown in Table 4. Under the 0.05 significance level, A, V, and N + S + Cl showed significant differences, and the most significant was for ash content. Cluster analysis indicated that each subgroup was a class and cannot cluster with others with a clustering coefficient less than 1.5.

Table 4. ANOVA of thermochemical properties in wood waste subgroups

		Sum of squares	Degree of freedom	Mean square	F	Significance (P)
A	Between groups	183.676	3	61.225	24.942	0.000
	Within groups	73.641	30	2.455
V	Between groups	303.329	3	101.110	15.635	0.000
	Within groups	194.001	30	6.467
FC	Between groups	21.227	3	7.076	1.286	0.297
	Within groups	165.027	30	5.501
C + H	Between groups	11.665	3	3.888	0.535	0.661
	Within groups	269.046	37	7.272
O	Between groups	11.457	3	3.819	0.478	0.699
	Within groups	295.492	37	7.986
N + S + Cl	Between groups	5.681	3	1.894	5.409	0.003
	Within groups	12.953	37	0.350
HHV	Between groups	1443928.171	3	481309.390	0.198	0.896
	Within groups	4.123E7	17	2425481.822

Paper

The paper in MSW can be divided into three subgroups, namely, printing paper (including newspapers, books, and magazines), cardboard, and toilet paper. The proximate and ultimate analysis and heating value results of paper are shown in Table 5.

Table 5. Chemical characteristics of paper subgroups and varieties

Paper subgroup and variety		Proximate analysis (wt%)			Ultimate analysis (wt%)						HHV^d (kJ/kg)	Reference used
		A^d	V^d	FC^d	C^daf	H^daf	O^daf	N^daf	S^daf	Cl^daf
Printing paper
1.	Blank A4 paper	10.69	79.33	9.98	45.12	5.31	48.91	0.38	0.28		13510
2.	Magazine	23.32	68.81	7.88							12742	Nie, 2000
3.	Magazine	29.49	62.44	8.07	41.04	8.99	49.15	0.42	0.40		11825	Zhang, 2002
4.	Magazine				46.44	6.54	46.33	0.16	0.30	0.23		Hu et al., 2011
5.	Newspaper	1.38	86.37	12.25	49.10	6.10	43.00	0.10	0.20		19734	Nie, 2000
6.	Newspaper	2.66	82.62	14.72	41.35	6.15	52.21	0.10	0.19			Li et al., 2002
7.	Newspaper	5.43	85.04	9.53	45.24	7.17	47.10	0.25	0.23		16270	Zhang, 2002
8.	Newspaper	8.07	79.54	12.39	48.01	5.71	45.82	0.33	0.13		17160
9.	Newspaper	9.47	78.19	12.33	46.66	6.25	46.86	0.13	0.09		15341	Shen, 2005
10.	Newspaper				49.86	6.34	43.24	0.15	0.26	0.15		Hu et al., 2011
11.	Printing paper	12.30	87.65	0.04	44.93	4.55	50.43	0.09	0.00		14236	Shen et al., 2006
12.	Printing paper				44.59	6.03	49.20	0.10	0.08		16146	Liu et al., 1999
13.	Writing paper				43.66	5.84	50.16	0.16	0.18		13690	Liu et al., 1999
Mean		11.42	78.89	9.69	45.50	6.25	47.70	0.20	0.20	0.19	15065
Minimum		29.49	87.65	14.72	49.86	8.99	52.21	0.42	0.40	0.23	19734
Maximum		1.38	62.44	0.04	41.04	4.55	43.00	0.09	0.00	0.15	11825
Cardboard
14.	Cardboard	10.77	78.70	10.53	47.09	6.00	46.63	0.18	0.11			Dai, 2007
15.	Cardboard	9.09	78.80	12.12	49.33	7.94	41.99	0.49	0.25		15992
16.	Cardboard	5.27	81.75	12.97	46.09	5.36	48.02	0.32	0.21		17278	Nie, 2000
17.	Cardboard				46.71	5.31	47.35	0.32	0.32		18367	Mi et al., 1993
18.	Cardboard	3.25	83.47	13.29							16002	Ke et al., 2006
19.	Carton	7.22	83.95	8.82	48.97	6.14	44.52	0.21	0.16		17099	Xiao et al., 2009
20.	Carton				47.32	6.53	45.55	0.14	0.30	0.15		Hu et al., 2011
21.	Paper board	14.07	72.27	13.67	49.51	6.29	43.71	0.27	0.22		15481	Li et al., 1999b
Mean		8.28	79.82	11.90	47.86	6.22	45.40	0.28	0.22	0.15	16703	Shen, 2005
Minimum		14.07	83.95	13.67	49.51	7.94	48.02	0.49	0.32	0.15	18367
Maximum		3.25	72.27	8.82	46.09	5.31	41.99	0.14	0.11	0.15	15481
Toilet paper
22.	Tissue paper	0.04	95.36	4.60	44.31	6.06	49.43	0.13	0.06		16855	Bai et al., 2007
23.	Toilet paper	0.52	90.47	9.01	45.18	6.13	48.32	0.25	0.11		17250
24.	Toilet paper				41.63	5.71	52.35	0.00	0.31			Na et al., 2008
Mean		0.28	92.91	6.81	43.71	5.97	50.03	0.13	0.16		17053
Minimum		0.04	90.47	4.60	41.63	5.71	48.32	0.00	0.06		16855
Maximum		0.52	95.36	9.01	45.18	6.13	52.35	0.25	0.31		17250

The proximate and ultimate analysis results of paper are plotted in Figure 4. Toilet paper had the highest V and the lowest A, while the proximate analysis of printing paper and cardboard showed no significant difference, as shown in Figure 4a. The values of N + S + Cl content of paper samples were all less than 0.25%. Compared with cardboard, toilet paper had higher O and lower C + H, and the elemental composition of printing paper was in between. As shown in Table 5, the HHV decreasing order was toilet paper > cardboard > printing paper.

Figure 4. Chemical composition of paper specific components: (a) proximate analysis; (b) ultimate analysis.

The ANOVAs of proximate and ultimate analysis and HHV among paper subgroups are shown in Table 6. Under the 0.05 significance level, C + H and O showed significant difference, and other variables showed no significance. Cluster analysis indicated that paper could be classified to two clusters: (i) printing paper and cardboard, and (ii) toilet paper.

Table 6. ANOVA of thermochemical properties in paper subgroups

		Sum of squares	Degree of freedom	Mean square	F	Significance (P)
A	Between groups	208.054	2	104.027	1.882	0.189
	Within groups	773.866	14	55.276
V	Between groups	332.100	2	166.050	3.506	0.058
	Within groups	663.096	14	47.364
FC	Between groups	42.669	2	21.334	1.731	0.213
	Within groups	172.592	14	12.328
C + H	Between groups	46.370	2	23.185	4.068	0.034
	Within groups	108.279	19	5.699
O	Between groups	49.721	2	24.860	3.802	0.041
	Within groups	124.247	19	6.539
N + S + Cl	Between groups	.116	2	0.058	1.554	0.237
	Within groups	0.710	19	0.037
HHV	Between groups	1.341E7	2	6704792.022	1.809	0.198
	Within groups	5.560E7	15	3706648.116

Textiles

The textiles in MSW can be divided into three subgroups, namely, cotton, chemical fibers, and wool. The proximate and ultimate analysis and heating value results of textiles are shown in Table 7.

Table 7. Chemical characteristics of textiles subgroups and varieties

Textiles subgroup and variety		Proximate analysis (wt%)			Ultimate analysis (wt%)
		A^d	V^d	FC^d	C^daf	H^daf	O^daf	N^daf	S^daf	Cl^daf	HHV^d (kJ/kg)	Reference used
Cotton
1.	Cotton	0.41	92.25	7.34	54.20	5.83	39.59	0.29	0.09		21374	Zhang, 2002
2.	Cotton	0.68	78.27	21.05	49.91	6.00	43.68	0.26	0.15		17532	Miao, 2006
3.	Cotton	0.20	96.20	3.59	44.92	9.00	45.86	0.19	0.03		15789	Yan et al., 2009
4.	Cotton	1.45	86.70	11.85	46.19	6.12	47.07	0.54	0.08		17246	Yan et al., 2005
5.	Cotton				43.18	6.10	50.04	0.52	0.16		16830	Liu et al., 1999
6.	Cotton	0.41	90.61	8.98								Li et al., 1998
7.	Cotton	1.06	81.42	17.51								Li and Zhang, 2004
8.	Cotton cloth	3.09	78.71	18.21	56.49	5.87	33.30	3.52	0.18	0.65	14211	Qing et al., 2005
9.	Cotton cloth	1.52	84.53	13.95	46.51	5.80	46.98	0.43	0.28		17430
10.	Cotton cloth	0.83	91.65	7.52	45.18	5.66	48.81	0.21	0.14		16251	Li et al., 1999b
11.	Cotton cloth	0.88	98.99	0.13	41.34	6.28	50.88	1.47	0.03			Zhang et al., 2008b
Mean		1.05	87.93	11.01	47.55	6.30	45.14	0.82	0.13	0.65	17083
Minimum		0.20	78.27	0.13	41.34	5.66	33.30	0.19	0.03	0.65	14211
Maximum		3.09	98.99	21.05	56.49	9.00	50.88	3.52	0.28	0.65	21374
Wool
12.	Wool				58.53	6.48	18.23	15.12	1.65		23632	Liu et al., 1999
13.	Wool	1.24	84.76	14.00	60.07	4.24	31.48	2.65	1.55		20920
Mean		1.24	84.76	14.00	59.30	5.36	24.85	8.89	1.60		22276
Minimum		1.24	84.76	14.00	58.53	4.24	18.23	2.65	1.55		20920
Maximum		1.24	84.76	14.00	60.07	6.48	31.48	15.12	1.65		23632
Chemical fibers
14.	Acrylic fiber	0.14	75.25	24.61	66.78	5.20	7.31	20.26	0.45		29770
15.	Chemical fiber				48.09	7.16	34.06	9.43	1.26		21959	Liu et al., 1999
16.	Chemical fiber	8.20	91.18	0.61								Zhang et al., 2008a
17.	Chemical fiber	1.48	89.78	8.75								Li et al., 1998
18.	Polyester taffeta	0.44	90.63	8.93	60.10	4.50	35.11	0.28	0.01		22080	Miao, 2006
19.	Terylene	0.49	88.60	10.91	62.16	4.14	33.12	0.29	0.28		20860
Mean		2.15	87.09	10.76	59.28	5.25	27.40	7.56	0.50		23667
Minimum		0.14	75.25	0.61	48.09	4.14	7.31	0.28	0.01		20860
Maximum		8.20	91.18	24.61	66.78	7.16	35.11	20.26	1.26		29770

The proximate analyses of different textiles samples were scattered, as shown in Figure 5a. Cotton had low A and various V and FC. The ash content of chemical fibers varied greatly. As shown in Figure 5b, the elemental composition of chemical fibers also showed great variance. The N content of acrylic fibers was as high as 20%, due to the monomer contained CN function group. The elemental composition of cotton was similar, and the N + S + Cl content was very low. The elemental composition of two wool samples varied greatly. As shown in Table 7, the HHV decreasing order was chemical fibers > wool > cotton.

Figure 5. Chemical composition of textiles specific components: (a) proximate analysis; (b) ultimate analysis.

The ANOVA of proximate and ultimate analysis and HHV among textiles subgroups are shown in Table 8. Under the 0.05 significance level, C + H, O and HHV showed significant difference, and other variables showed no significance. Cluster analysis indicated that each subgroup was a class and cannot cluster with others with a clustering coefficient less than 1.5.

Table 8. ANOVA of thermochemical properties in textiles subgroups

		Sum of squares	Degree of freedom	Mean square	F	Significance (P)
A	Between groups	4.041	2	2.021	0.495	0.621
	Within groups	53.070	13	4.082
V	Between groups	10.217	2	5.109	0.103	0.903
	Within groups	643.138	13	49.472
FC	Between groups	9.050	2	4.525	0.083	0.921
	Within groups	709.698	13	54.592
C + H	Between groups	414.587	2	207.293	7.460	0.008
	Within groups	333.452	12	27.788
O	Between groups	1251.787	2	625.893	8.544	0.005
	Within groups	879.082	12	73.257
N + S + Cl	Between groups	229.477	2	114.739	3.663	0.057
	Within groups	375.857	12	31.321
HHV	Between groups	1.310E8	2	6.551E7	8.612	0.006
	Within groups	8.368E7	11	7607225.057

Plastics

Unlike other MSW components, the plastic varieties tend to be pure. Five kinds of commonly used plastics include PE (including high-density polyethylene and low-density polyethylene), PP (polypropylene), PS, PVC, and PET (polyethylene terephthalate). The proximate and ultimate analysis and heating value results of plastics are shown in Table 9.

Table 9. Chemical characteristics of plastics subgroups and varieties

Plastics subgroup and variety		Proximate analysis (wt%)			Ultimate analysis (wt%)
		A^d	V^d	FC^d	C^daf	H^daf	O^daf	N^daf	S^daf	Cl^daf	HHV^d (kJ/kg)	Reference used
PE
1.	PE	0.06	99.94	0.00	86.66	13.26	0.00	0.06	0.02	0.00	37600	Li et al., 1999b
2.	PE	0.15	99.85	0.00	85.94	13.88	0.00	0.12	0.06		40983	He et al., 2009
3.	PE	0.15	99.85	0.00	85.45	14.32	0.00	0.16	0.07	0.00	46318	Zheng et al., 2009
4.	PE	0.30	99.70	0.00	84.97	14.30	0.00	0.70	0.02	0.00	46479	Chi et al., 2008
5.	PE	0.56	98.96	0.48	85.34	14.57	0.09	0.00	0.00			Zhang et al., 2008b
6.	HDPE	0.60	99.40	0.00	85.50	14.20	0.00	0.00	0.30	0.00		Zhou et al., 2009
7.	HDPE	0.58	99.42	0.00	85.44	14.24	0.32	0.00	0.00	0.00		Li et al., 2001a
8.	LDPE	0.30	99.70	0.00	85.50	14.30	0.00	0.00	0.20	0.00		Zhou et al., 2009
9.	LDPE	0.31	99.69	0.00	85.51	14.30	0.19	0.00	0.00	0.00		Li et al., 2001a
Mean		0.33	99.61	0.05	85.59	14.15	0.07	0.12	0.07	0.00	42845
Minimum		0.06	98.96	0.00	84.97	13.26	0.00	0.00	0.00	0.00	37600
Maximum		0.60	99.94	0.48	86.66	14.57	0.32	0.70	0.30	0.00	46479
PP
10.	PP	0.82	99.18	0.00	83.75	13.98	2.27	0.00	0.00	0.00		Li et al., 2001a
11.	PP	0.16	99.84	0.00	84.30	14.44	1.05	0.18	0.03		45769	Yan et al., 2005
12.	PP	0.08	99.79	0.13	84.62	15.23	0.00	0.14	0.01	0.00		Tang et al., 2003a
13.	PP				85.10	14.40	0.50	0.00	0.00			Zhao et al., 2012
14.	PP				85.22	13.70	1.08	0.00	0.00	0.00		Lan et al., 2012
15.	PP	0.02	99.97	0.01	85.41	12.51	1.85	0.23	0.00		46239	Bai et al., 2007
16.	PP				86.72	13.28	0.00	0.00	0.00			Feng et al., 2006
17.	PP	1.10	98.90	0.00								Zhou et al., 2009
Mean		0.44	99.54	0.03	85.02	13.93	0.96	0.08	0.01	0.00	46004
Minimum		0.02	98.90	0.00	83.75	12.51	0.00	0.00	0.00	0.00	45769
Maximum		1.10	99.97	0.13	86.72	15.23	2.27	0.23	0.03	0.00	46239
PS
18.	PS	0.04	99.57	0.39							38930
19.	PS	0.51	99.49	0.00	91.08	7.68	1.24	0.00	0.00	0.00		Li et al., 2001a
20.	PS	0.01	99.39	0.60							40116	Zhang et al., 2011b
21.	PS				89.20	9.00	1.80	0.00	0.00			Zhao et al., 2012
22.	PS				89.06	10.02	0.55	0.00	0.37			Feng et al., 2006
23.	PS				92.14	7.86	0.00	0.00	0.00	0.00		Miao, 2005
Mean		0.19	99.48	0.33	90.37	8.64	0.90	0.00	0.09	0.00	39523
Minimum		0.01	99.39	0.00	89.06	7.68	0.00	0.00	0.00	0.00	38930
Maximum		0.51	99.57	0.60	92.14	10.02	1.80	0.00	0.37	0.00	40116
PVC
24.	PVC	8.74	79.81	11.45	37.81	4.25	0.00	0.09	0.19	57.66		Jin et al., 2001
25.	PVC	14.36	65.17	20.47	38.82	5.10	0.00	0.06	0.30	55.72		Zhou et al., 2001
26.	PVC	14.99	65.06	19.95	40.39	4.54	0.21	0.20	0.09	54.56	15876	Li et al., 1999b
27.	PVC	0.32	91.25	8.43	41.79	4.84	0.00	0.09	0.02	53.26	21732	Zhu et al., 2008
28.	PVC	0.04	95.16	4.80	38.75	5.21	0.00	0.22	0.00	55.82	22566	Zhao et al., 2007
29.	PVC	0.00	94.93	5.07	38.34	4.47	0.00	0.23	0.61	56.35	20830
30.	PVC	0.08	93.96	5.96	40.38	5.83	0.00	0.07	1.25	52.46	22405	Jin, 2002
31.	PVC	0.10	94.50	5.40	38.83	4.57	0.00	0.01	0.00	56.60	21682	Zhang et al., 2011b
32.	PVC	14.12	71.39	14.49	40.91	4.80	0.00	0.06	0.01	54.23		Li et al., 2002
Mean		5.86	83.47	10.67	39.56	4.85	0.02	0.11	0.28	55.18	20848
Minimum		0.00	65.06	4.80	37.81	4.25	0.00	0.01	0.00	52.46	15876
Maximum		14.99	95.16	20.47	41.79	5.83	0.21	0.23	1.25	57.66	22566
PET
33.	PET				62.00	5.20	32.80	0.00	0.00			Lian et al., 2011
34.	PET				62.02	4.19	33.40	0.32	0.07			Feng et al., 2006
35.	PET	0.09	90.44	9.47	62.93	4.26	32.64	0.04	0.13		23090
36.	PET	0.31	94.09	5.60	62.26	4.06	33.69	0.00	0.00			Zhang et al., 2012
Mean		0.20	92.27	7.53	62.30	4.43	33.13	0.09	0.05		23090
Minimum		0.09	90.44	5.60	62.00	4.06	32.64	0.00	0.00		23090
Maximum		0.31	94.09	9.47	62.93	5.20	33.69	0.32	0.13		23090

The proximate and ultimate analysis results of plastics are plotted in Figure 6. The proximate analysis values of PE, PP, and PS were close to each other, with V nearly 100%. The proximate analysis of PVC varied greatly. Some PVC samples had little ash and some samples had an ash content as high as 15%. PET had more than 90% V and less than 10% FC. The C + H content of PE, PP, and PS was nearly 100%, and O, N, S, and Cl contents were almost zero, as shown in Figure 6b. The Cl content of PVC was between 50% and 60%, and the O content of PET was about 33%. The HHV in decreasing order was PP, PE, PS, PET, and PVC, and the HHV of PVC and PET was about half of that of PE, PP, and PS.

Figure 6. Chemical composition of plastics specific components: (a) proximate analysis; (b) ultimate analysis.

The ANOVA of proximate and ultimate analysis and HHV among plastics subgroups are shown in Table 10. Under the 0.05 significance level, all the variables except ash content showed significant difference. Cluster analysis indicated that plastics could be classified to three clusters: (i) PE, PP, PS, (ii) PVC, and (iii) PET.

Table 10. ANOVA of thermochemical properties in plastics subgroups

		Sum of squares	Degree of freedom	Mean square	F	Significance (P)
A	Between groups	187.415	4	46.854	2.700	0.056
	Within groups	399.092	23	17.352
V	Between groups	1530.396	4	382.599	6.278	0.001
	Within groups	1401.661	23	60.942
FC	Between groups	684.718	4	171.179	12.076	0.000
	Within groups	326.042	23	14.176
C + H	Between groups	19572.205	4	4893.051	5017.243	0.000
	Within groups	27.307	28	0.975
O	Between groups	3775.252	4	943.813	3607.887	0.000
	Within groups	7.325	28	0.262
N + S + Cl	Between groups	20117.213	4	5029.303	6840.447	0.000
	Within groups	20.586	28	0.735
HHV	Between groups	1.794E9	4	4.485E8	50.599	0.000
	Within groups	8.863E7	10	8863239.350

Rubber

Since rubber in MSW was mainly derived from waste tires, it was not subdivided for this paper. The proximate and ultimate analysis and heating value results of rubber are shown in Table 11.

Table 11. Chemical characteristics of rubber varieties

Rubber variety		Proximate analysis (wt%)			Ultimate analysis (wt%)							Reference used
		A^d	V^d	FC^d	C^daf	H^daf	O^daf	N^daf	S^daf	Cl^daf	HHV^d (kJ/kg)
1.	Rubber				80.05	7.99	6.00	1.79	4.17		21812	Chi et al., 2008
2.	Rubber	15.38	65.26	19.36	89.18	8.54	0.00	1.23	1.05		33402	Zhang, 2002
3.	Rubber	13.98	67.20	18.83	88.82	8.83	0.00	1.27	1.08			Zhang et al., 2003
4.	Rubber	13.50	58.80	27.70	85.75	11.49	0.55	0.55	1.66			Zhang et al., 2007b
5.	Rubber	8.36	84.77	6.86	77.72	10.12	7.42	0.00	2.66	2.08	25177	Qing et al., 2005
6.	Rubber	6.14	66.12	27.73	85.98	11.50	0.79	0.64	1.10			Liu, 2008
7.	Rubber	5.00	74.55	20.44	86.70	7.28	2.14	2.42	1.47		37299	Miao, 2006
8.	Tire	3.33	69.25	27.42	85.63	7.86	4.65	0.52	1.34		36757	Dai et al., 2001
9.	Tire	25.70	68.05	6.25	79.19	8.45	11.38	0.69	0.28		26491	Li et al., 2004
10.	Tire	19.27	63.11	17.61	88.56	8.52	0.88	0.75	1.29		30164	Li et al., 1999a
11.	Tire	16.02	59.92	24.06	84.21	7.73	5.29	0.61	2.16			Zhou et al., 2001
12.	Tire	14.56	80.70	4.74	86.13	6.94	3.47	0.58	1.73	1.16		Wang et al., 2003
13.	Tire	7.47	65.16	27.37	86.03	7.23	4.56	0.40	1.78			Su et al., 2008
14.	Tire	7.11	63.00	29.89	81.66	7.23	9.54	0.65	0.92			Deng et al., 2010
15.	Tire	4.89	68.13	26.99	82.65	7.55	7.45	1.67	0.67			Jin et al., 2007
16.	Tire	4.39	62.96	32.65	89.24	7.12	1.52	0.41	1.70			Ma et al., 2007
17.	Tire	4.19	65.44	30.37	83.92	6.83	7.55	0.78	0.92		38868	Zhang et al., 2005
18.	Tire	3.90	63.13	32.97	88.84	7.74	0.94	0.32	2.16		37929	Chen et al., 2012
Mean		10.19	67.39	22.43	85.01	8.27	4.12	0.85	1.56	1.62	31989
Minimum		3.33	58.80	4.74	77.72	6.83	0.00	0.00	0.28	1.16	21812
Maximum		25.70	84.77	32.97	89.24	11.50	11.38	2.42	4.17	2.08	38868

As shown in Table 11, the mean C content of rubber was 85.01%. The reason was that in addition to the high C content of the rubber polymer monomer, carbon black was usually added to tires to enhance wear resistance. Rubber had high H content, low O content, and high S and Cl content. Meanwhile, the HHV of rubber was as high as 31,989 kJ/kg. The proximate and ultimate analysis of different rubber samples varied significantly, as shown in Figure 7.

Figure 7. Chemical composition of rubber specific components (a) proximate analysis; (b) ultimate analysis.

Comparison and classification of MSW specific components

Comparison of MSW specific components

From 2.1.1 to 2.1.6, the proximate and ultimate analysis and HHV of MSW specific components were compared, as shown in Table 12. Bones and vegetables had the highest A, while PE, PP, PS, PET, and toilet paper had the highest V and the lowest A and FC. The FC of nutshells and rubber was the highest. PE, PP, PS, and rubber had the highest C and H, while the C and H of PVC were relatively low, due to high Cl. PET had high C but low H. Paper and starch food had the highest O, while PE, PP, and PS had no O and N. Wool was a keratin, a fibrous insoluble animal protein (Pine et al., 1981), and bone contained protein, so they had high N and S. Chemical fiber had the highest N, because of acrylic fiber. The HHVs of PE, PP, and PS were the highest and artificial polymers such as rubber and chemical fiber also had high HHVs, while HHV values of paper, vegetables, and bones were the lowest.

Table 12. The comparison of the characteristics of MSW specific components

Factors	Content listed in decreasing sequence
A	bones > vegetables > printing paper > rubber > cardboard > leaves > weeds > PVC > fruit peel > chemical fiber > nutshells > wood > bamboo > wool > cotton > starch food > PP > PE > toilet paper > PET > PS
V	PE > PP > PS > toilet paper > PET > cotton > starch food > chemical fiber > wool > PVC > wood > bamboo > cardboard > printing paper > fruit peel > weeds > leaves > nutshells > vegetables > rubber > bones
FC	nutshells > rubber > fruit peel > bamboo > leaves > weeds > wood > wool > starch food > cardboard > cotton > chemical fiber > PVC > vegetables > printing paper > PET > toilet paper > bones > PS > PE > PP
C	PS > PE > PP > rubber > PET > wool > chemical fiber > bones > nutshells > weeds > wood > leaves > bamboo > fruit peel > cardboard > cotton > printing paper > vegetables > toilet paper > starch food > PVC
H	PE > PP > PS > rubber > bones > nutshells > weeds > starch food > cotton > printing paper > cardboard > fruit peel > wood > bamboo > toilet paper > leaves > vegetables > wool > chemical fiber > PVC > PET
O	toilet paper > starch food > printing paper > vegetables > cardboard > cotton > bamboo > leaves > fruit peel > wood > weeds > nutshells > PET > chemical fiber > bones > wool > rubber > PP > PS > PE > PVC
N	wool > bones > chemical fiber > vegetables > starch food > fruit peel > nutshells > rubber > cotton > weeds > leaves > wood > bamboo > cardboard > printing paper > PET > toilet paper > PE > PVC > PP > PS
S	wool > rubber > bones > vegetables > chemical fiber > leaves > PVC > cardboard > nutshells > printing paper > weeds > toilet paper > fruit peel > starch food > bamboo > cotton > wood > PS > PE > PET > PP
Cl	PVC > rubber > cotton > wood > printing paper > bamboo > cardboard > fruit peel > starch food > PE = PP = PS
Q	PP > PE > PS > rubber > chemical fiber > PET > wool > PVC > weeds > wood > bamboo > nutshells > leaves > fruit peel > starch food > cotton > toilet paper > vegetables > cardboard > bones > printing paper

Classification of MSW specific components

Cluster analysis was applied to classify MSW specific components. The variables included A, V, FC, C, H, O, N, S, and HHV, as shown in Figure 8. Fruit peel, weeds, wood, bamboo, leaves, and nutshells could be clustered, as shown in Area A. The components of this cluster were composed of pectin, hemicellulose, cellulose, and lignin, with high FC and medium other indicators. Area B included starch food, cotton, toilet paper, printing paper, and cardboard. It seemed that their components were unrelated. However, the main ingredient of paper was cellulose (Wu et al., 2003), and cotton also contained 95% cellulose (Abidi et al., 2010), which had the same glucose monomer as starch food (Pine et al., 1981). Wool and chemical fiber were clustered as shown in Area C, which had low H and O, and high N, S, and HHV. PE, PP, and PS had the most similarity, as shown in Area D. The properties of this cluster were high V (nearly 100%), C, H, and HHV, and low O, N, A, and FC. Vegetables, PET, PVC, bones, and rubber had more complex composition, and thus they could not be classified into the four categories just described.

Figure 8. The cluster analysis of MSW specific components based on proximate and ultimate analysis and HHV.

Conclusion

The proximate and ultimate analysis and heating value of MSW specific components were analyzed and statistical methods such as ANOVA and cluster analysis were applied. Bones and vegetables had the highest ash content, and nutshells and rubber had the highest fixed carbon content. The carbon and hydrogen content of PVC was low and the chlorine content was very high, accordingly. PET had high carbon content and low hydrogen content. Paper and starch food had the highest oxygen content, and wool and bones had the highest nitrogen and sulfur content. The sulfur and chlorine content of rubber was relatively high. PE, PP, and PS had the highest heating value, and the heating values of chemical products such as rubber and chemical fiber were also very high. Conversely, paper, vegetables, and bones had the lowest heating value. The cluster analysis of MSW components was applied based on proximate and ultimate analysis and heating value results. The results showed that fruit peel, weeds, wood, bamboo, leaves, and nutshells could be classified as a category; starch food, cotton, toilet paper, printing paper, and cardboard could be classified as a category; wood and chemical fiber could be classified as a category; and PE, PP, and PS could be clustered as a category.

Funding

Financial support from the National Basic Research Program of China (973 Program, no. 2011CB201502) and National Natural Science Foundation of China (no. 21376134) is gratefully acknowledged.

Additional information

Notes on contributors

Hui Zhou

Yanguo Zhang is a professor, Qinghai Li is an associate professor, and Hui Zhou, Aihong Meng, and Yanqiu Long are Ph.D. candidates at the Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Thermal Engineering, Tsinghua University, Beijing, China.

Aihong Meng

Yanguo Zhang is a professor, Qinghai Li is an associate professor, and Hui Zhou, Aihong Meng, and Yanqiu Long are Ph.D. candidates at the Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Thermal Engineering, Tsinghua University, Beijing, China.

Yanqiu Long

Yanguo Zhang is a professor, Qinghai Li is an associate professor, and Hui Zhou, Aihong Meng, and Yanqiu Long are Ph.D. candidates at the Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Thermal Engineering, Tsinghua University, Beijing, China.

Qinghai Li

Yanguo Zhang is a professor, Qinghai Li is an associate professor, and Hui Zhou, Aihong Meng, and Yanqiu Long are Ph.D. candidates at the Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Thermal Engineering, Tsinghua University, Beijing, China.

Yanguo Zhang

Yanguo Zhang is a professor, Qinghai Li is an associate professor, and Hui Zhou, Aihong Meng, and Yanqiu Long are Ph.D. candidates at the Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Thermal Engineering, Tsinghua University, Beijing, China.