CO2 atmosphere

Applied Thermal Engineering 128 (2018) 662–671

Contents lists available at ScienceDirect

Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng

Research Paper

Study of the pyrolysis of municipal sludge in N2/CO2 atmosphere Zihuan Wang, Xiaoqian Ma ⇑, Zhongliang Yao, Quanheng Yu, Zhao Wang, Yousheng Lin Guangdong Key Laboratory of Efficient and Clean Energy Utilization Institutes, School of Electric Power, South China University of Technology, Guangzhou 510640, China School of Electric Power, South China University of Technology, No. 381, Wushan Road, Tianhe District, Guangzhou 510640, China

h i g h l i g h t s
The pyrolysis properties of wet and dried municipal sludge are studied.
With the existence of moisture, the releasing of volatiles is enhanced and a large amount of CO2 and NH3 generates earlier.
CO2 makes a difference to the pyrolysis behavior in high temperature.
Kinetics parameters are calculated by using Coast-Redfern method.

a r t i c l e

i n f o

Article history: Received 5 December 2016 Revised 6 September 2017 Accepted 10 September 2017 Available online 12 September 2017 Keywords: Pyrolysis Wet sewage sludge CO2 TG-FTIR Gaseous product

a b s t r a c t The present study aims to explore the pyrolysis characteristics and gaseous product of municipal sludge including dry sample (DS) and wet sample (WS). The experiments were performed under atmospheres of different N2/CO2 ratios using TG-FTIR. All of the thermal process could be divided into two stages. With the existence of moisture, the releasing of volatiles was enhanced and a large amount of CO2 and NH3 generated earlier. CO2 was inert in first stage and became reactive in second stage. The char gasification by CO2 resulted in more mass loss and macropores generation. As the CO2 concentration increased, the corresponding peak shifted to lower temperature. The kinetic parameters were calculated by CoastRedfern method and agreed with the pyrolysis properties observed. The activation energy and preexponential factor increased (or decreased) simultaneously, illustrating kinetic compensation effect. Ó 2017 Published by Elsevier Ltd.

1. Introduction Municipal sludge is the inevitable by-product of municipal wastewater treatment process. Along with the urbanization process, the yield of municipal sludge is increasing rapidly. It is estimated that the total production of municipal sludge in China will have reached 60 million–90 million tons by 2020. Municipal sludge is a heterogeneous mixture, rich in organic matter and N, P, K nutrient elements, also containing heavy metals and pathogenic organisms. Due to its content of hazardous substances, it would result in environmental problem and human disease without proper disposal. The disposal of sludge is a great challenge. The common used methods such as landfill and dumping possibly caused secondary pollution to groundwater and soil. As the laws and regulations of sludge management is getting stricter, their application would be limited [1]. Pyrolysis is regarded as one of the most promising ⇑ Corresponding author at: School of Electric Power, South China University of Technology, No. 381, Wushan Road, Tianhe District, Guangzhou 510640, China. E-mail address: [email protected] (X. Ma). http://dx.doi.org/10.1016/j.applthermaleng.2017.09.044 1359-4311/Ó 2017 Published by Elsevier Ltd.

treatment of sludge. Its distinct advantages reflect in generating valuable product including syngas, bio-oil and biochar as well as reducing volume, eliminating pathogens. The syngas and bio-oil of high caloric value with further processing can be used as fuel [2,3]. Bio-char also has many application such as soil conditioner, adsorbent and construction materials [4,5]. In addition, the majority of heavy metals are concentrated in pyrolysis residues except volatile elements Hg and Cd [6,7]. The pyrolysis of sludge have already been implemented. The Sewage sludge utilization plant in Balingen, German has been in continuous operation since 2002 [8]. The plant converts 1250 tons sludge per year into syngas that is used as fuel for a thermal power plant and thus recovers energy as heat and electricity. The relevant research about pyrolysis of sludge has intensified in recent years. In most of these experiments, dried sludge was usually used as the object of study [9,10]. However, initial sludge contains a large amount of moisture. And it consumes a lot of energy to remove the moisture content [11]. In the process of pyrolysis, the existence of moisture in wet sludge would make a difference to pyrolysis characteristics. As pointed out in several papers [12,13], the presence of water in the sludge increases the

663

Z. Wang et al. / Applied Thermal Engineering 128 (2018) 662–671

ated to DS and WS respectively. The ultimate and proximate analyses of the sample on dry basis were shown in Table 1.

production of gases and also favors the generation of hydrogen. So far little information is available about the thermal decomposition characteristics of wet sludge. Therefore, both wet and dry municipal sludge are taken as experiment samples in our work to explore the possibility of combining cost reduction and pyrolysis characteristics improvement together. Many research have reported that the pyrolysis process and product were affected by many factors such as particle size, temperature, heating rate and the composition of sludge [14–17]. However, few studies focused on the effect of CO2 atmosphere on sludge pyrolysis. CO2 accounts for a large fraction of the pyrolytic gas. Recycling gas product stream and then providing CO2 as carrier gas instead of N2 would result in the enrichment of CO2 in the flue gas, which helps carbon capture and storage (CCS) and reducing the greenhouse gas emission [18–20]. Such study also could assess a better understanding to oxy-fuel combustion, which is one of the most primary technologies for CO2 capture. Some research about coal, biomass have shown that the participation of CO2 made a great difference to the pyrolysis behaviors. TGFTIR analysis of coal pyrolysis showed that replacing N2 with CO2 enhanced the volatile releasing rate and prevented the calcite from decomposing [21]. The study of lump coal pyrolysis indicated that CO2 atmosphere was favorable to the release of H2O and absorbing gases. The fast pyrolysis process of biomass showed that CO2 impacted the final gas yield and composition, as well as the char yield and properties [22]. The reaction between CO2 and the nascent char above 600 °C occurred at a considerable high rate correlated with thermal cracking and resulted in an extra mass loss [23]. In other words, CO2 may play the role as reactant or product and thus changes the pyrolysis behavior. It is worthy to study the impact of CO2 concentration on the sludge pyrolysis further. The objective of the present work is to explore the pyrolysis characteristics and gaseous product of municipal sludge in different N2/CO2 atmosphere. The experiments of the wet/dry sludge in different atmospheres were carried out using Thermogravimetric Analyzer and Fourier Transform Infrared Spectrometry (TGFTIR). TG is widely applied to monitor mass change of a solid sample as function of temperature at given heating rates, reflecting the reaction process. The recorded TG data was used to study pyrolysis characteristics and evaluate the kinetic parameters, which is useful for optimization of actual operation condition. FTIR could identify the evolved gas with time and deepen the understanding of the pyrolysis process. Furthermore, Scanning Electron Microscopy and Energy Dispersive Spectrometer (SEM-EDS) were used as auxiliary means to get the surface morphology and element distribution of the solid residue.

2.2. Experiment The pyrolysis experiments were carried out in a Thermogravimetric Analyzer (METTLER TOLEDO TGA/DSC 1/1600). According to the moisture content, the initial weight of sample was maintained at 7 ± 1 mg for DS and 10 ± 1 mg for WS respectively to ensure the same quality of dry basis. To investigate the effect of the atmosphere, the carrier gases were N2, CO2, 75%CO225%N2, 50%CO250%N2, 25%CO275%N2 respectively, with a total flow rate of 80 mL/min1. The samples were heated from ambient temperature to 1000 °C at the rate of 10/30/50 °C/min, and then remained at 1000 °C for 5 min to complete reaction. Blank experiments were carried out using empty crucible under different atmosphere at different heating rate to obtain the baselines, which would be deducted in the experiments with sample. Gaseous products from TGA passed through a transfer line into Fourier Transform Infrared Spectrometry (NicoletTM iSTM 10 FT-IR spectrometer). The transfer line should be preheated to 225 C and kept at the temperature during the experiment, in order to prevent gas condensation. FTIR scan wave number range is from 4000 to 500 cm1, with resolution better than 0.4 cm1. FTIR spectra was used to analyze the evolved gas under N2 atmosphere in real time. Before FTIR scanning, pass N2 for 5 min and correct the background baseline. In CO2 atmosphere, however, the high absorbance of CO2 would cause interference due to the limitation of FTIR technique and thus the untrusted result should be discarded. 2.3. Kinetic analysis The basic dynamic equation for heterogeneous solid-state reactions is generally given by:

da 1 ¼ kðTÞf ðaÞ dT b

ð1Þ

where a is the conversion rate of the sample, t (min) is time, b is the heating rate, T (K) is the absolute temperature, k(T) is the temperature-dependent constant, f (a) is the reaction mechanism function. Expression of conversion rate is as follows:

a ¼ mo  mt =mo  mf

ð2Þ

where mo, mt, mf are initial mass, mass at time t and final mass of the sample respectively. k(T) is usually described by the famous Arrhenius equation:

k ¼ A expðE=RTÞ

ð3Þ

where A is the pre-exponential factor, E is the apparent activation energy, R is the universal gas constant. For nth-order reaction, f (a) is expressed asf ðaÞ ¼ ð1  aÞn : Integration of the function f (a) is defined as:

2. Methods 2.1. Material

GðaÞ ¼

The municipal sludge from Liede Wastewater Treatment Plant in Guangzhou city, Guangdong province of China was used as the raw material. The dry sludge sample was prepared at 105 °C in an air dry oven for 24 h until the mass no longer changed. The wet sludge sample after drying in the same oven for 4 h was obtained with moisture content of 30%. The samples were abbrevi-

Z

0

a

da A ¼ ð1  aÞn b

Z

T

T0

  E exp  dT RT

ð4Þ

G (a) has no analytical solution and only get numerical and approximate solutions. An approximate equation can be obtained by Coats-Redfern method and rearranged into the following form [24,25]:

Table 1 Ultimate and proximate analyses of sample on dry basis. Sample

Municipal sludge

Ultimate analyses (wt%, d)

Proximate analyses (wt%, d)

C

H

O

N

S

Volatile

Fixed carbon

Ash

26.52

4.556

19.434

4.32

0.6

48.82

6.61

44.57

664

Z. Wang et al. / Applied Thermal Engineering 128 (2018) 662–671

"Z ln 0

a

#

da

ð1  aÞ T

n 2

¼ ln

   AR 2RT E 1  bE E RT

ð5Þ

3.1. Thermogravimetric analysis

Generally for temperatures and activation energies, 2RT/E < < 1, which can be ignored. So Eq. (5) can be converted and written as:

ln

" # 1  ð1  aÞ1n

ln

   lnð1  aÞ

2

T ð1  nÞ

T

2

¼ ln

 ¼ ln

  AR E  bE RT

 AR E  bE RT

ðfor n – 1Þ

ðfor n ¼ 1Þ

3. Results and discussion

ð6Þ

ð7Þ

Plotting the left side of Eqs. (6) and (7) versus 1/T, the activation energy E and the pre-exponential factor A can be calculated from the slope and intercept of the straight lines respectively. The reaction order n can be evaluated through linear correlation coefficient R2.

The thermogravimetric (TG) curves and the derivative thermogravimetric (DTG) curves of DS and WS in N2 atmosphere at the heating rate of 10 °C/min are presented in Fig. 1(a). The mass loss before 150 °C was attributed to the vaporization of moisture. The DTG curve of DS presented a humble peak at around 100 °C due to chemical bonded water and equilibrium water while the peak of WS was distinct because of its great moisture content. However, it was inconvenient to compare the decomposition rate between DS and WS from the initial curves because of the different proportion of dry basis in the samples. To analyze the decomposition process, 150 °C was taken as the onset temperature of decomposition and the TG data was normalize to sample size. The curves obtained are illustrated in Fig. 1(b). Two stages could be distinguished during the pyrolysis process of DS. The first stage (SI) was in range of 150–600 °C, which was

Fig. 1. The TG and DTG curves (a) origin, (b) normalized of DS and WS at 10 °C/min in N2 atmosphere.

Z. Wang et al. / Applied Thermal Engineering 128 (2018) 662–671

regarded as the de-volatilization stage [26]. SI could be further divided into two parts by the boundary temperature 400 °C and a shoulder peak was observed on the right side. The overall mass change of this stage was 43.87 wt%, consistent with the volatile matter content of 48.82 wt% (shown in Table 1). The maximum decomposition rate appeared at 286 °C. In the second stage (SII), the mass rate was much smaller. It reached only 8.8 wt% mass loss of this interval. A closer inspection revealed there were two small parallel peaks. It was attributed to the degradation of minerals like calcium carbonate and the gasification of partial char respectively [18,27]. The TG and DTG curves of municipal sludge in relevant literatures have something different in the shape due to its complex composition [10,28]. In case of WS, two decomposition stages were also observed. The difference between WS and DS lay in the early period of the de-volatilization stage. A shoulder peak in range of 160–260 °C was observed on the left side, which was inexistent for DS. The maximum peak also centered at 291 °C. And no remarkable difference was found in the next section between the curves referring to

665

position, shape and height. In summary, the moisture in the sample enhanced the decomposition of organic matter below 260 °C but didn’t affect the reaction above 260 °C. According to relevant literatures, the typical degradation temperature range of hemicelluloses was 160–360 °C and that of cellulose was 240–390 °C while the lignocellulosic-based materials presented a wide degradation range (180–900 °C) [27]. Low stability organic components, whose major constituent is often starch, reached the highest decomposition rate at around 300 °C. All plastics except PVC reached the maximum decomposition rate between 410 and 515 °C [29]. The decomposition of the hydrocarbon chains of the major lipid constituents was ranging from 250 to 500 °C. All proteins samples went through weight loss within 200– 500 °C [30]. Municipal sludge is composed of various components and the mass loss peaks of them might overlap with each other during the decomposition process. Furthermore, there is a great possibility that catalytic effects and synergistic effects take place. Therefore, it is hard to identify the degradation interval of each single

Fig. 2. Evolution characteristics of different gas at 30 °C/min heating rate from (a) DS, (b) WS.

666

Z. Wang et al. / Applied Thermal Engineering 128 (2018) 662–671

compound through the TGA data alone. However, it is reasonable to assume that the interval located at 150–400 °C involved the decomposition of more reactive compound such as starch, hemicelluloses and cellulose, together with the preliminary degradation of other less reactive matters such as proteins and lipids. And the right shoulder peak in the range of 400–600 °C was associated with the further degradation of intermediates produced from lipids, proteins and so on.

3.2. FTIR analysis The evolution of gaseous products in N2 atmosphere was distinguished by the characteristic wave numbers. In SI, the main gaseous products were detected comprising CO2 (2358 cm1), CH4 (3011 cm1), C@O (1600–1850 cm1), H2O (1509 cm1), NH3 (964 cm1) [31,32]. And in SII, the main released gas were CO2 and CO (2180 cm1). H2 can’t be distinguished due to the limitation of FTIR technique. The absorbance of different gas with temperature for DS and WS are shown in Fig. 2. The emission of CO2 from pyrolysis of DS was of largest quantity due to the high C and O content of sample. The CO2 emission profile showed two peaks. The first one took the prominent role, much stronger than the second one. The production of CO2 in range of 150–600 °C was attributed to the break and recombining of thermolabile oxygen-containing functional groups such as carboxyl (ACOOH), carbonyl (C@O), ether groups(R-O-R) and hydroxyl(AOH) [33]. And a small amount of CO2 released in SII was from decomposing of carbonate NH3 took a considerable fraction of the evolved gases, which was the source of unpleasant smell. The NH3 emission profile showed one peaks in the range of 150–600 °C. It was possibly from the decomposing of protein or amino acid. The C@O emission profiles showed the similar trend as NH3. C@O group present in ketones, aldehydes, carboxylic acids, amides, esters could be split from amino acid, fatty acid and et al. The absorption of CO2, NH3, C@O all reached the maximum at around 300 °C, corresponding to the maximum peak on the DTG curve. The absorbance of CH4 reached the maximum value at 465 °C, corresponding to the right shoulder peak on the DTG curve. CH4 required higher temperature to generate and evolved later than CO2, C@O and NH3. It was inferred that CH4 came from the cracking of hydrocarbon

chains and it had to overcome higher energy obstacle before the reaction took place. The CO emission profile displayed two peak. There was a minor hump from 200 °C to 500 °C, indicating only a small quantity of CO released in the de-volatilization stage. The producing rate of CO speeded up at 750 °C and reach maximum at 960 °C, a little later than the second absorption peak of CO2. That is to say, most fraction of CO was contributed to the reaction between carbonaceous residue and CO2. From Fig. 2(b), there were differences between the gas production properties of WS and DS. More gas released at 225 °C, in accordance with the left shoulder peak at on the DTG curve. The emission profiles of C@O, CH4, CO from WS were similar to those from DS. However, the emission profiles of CO2 and NH3 for WS showed double peaks at 225 °C and 302 °C respectively. From the area formed by the profiles and the X axis, it can be inferred that a larger amount of CO2 and NH3 generated from WS qualitatively. High moisture content in the WS produces a steam-rich atmosphere in the reactor which reacted with both the organic vapors (steam reforming) and the solid residue (steam gasification) [12]. And it also pointed out in relevant research that long residence times at high temperature combined with high heating rate and low flow of carrier gas favored the reaction [13]. In the condition of our experiment, the moisture converted into steam and moved away with the flow gas. The contact and reaction time between situ steam and the volatile compounds was relatively short. The evolution characteristics of H2O with time for DS and WS are shown in Fig. S1, illustrating H2O evolved from WS in the first 4 min. Therefore, the pyrolysis process was influenced by the moisture content only at an early stage. Also, the hydrolysis mechanism might promotes the pyrolysis of WS. Organic components including carbohydrates, proteins and lipids occurred hydrolysis reaction in the water existence and heating condition and resulted in some functional groups of low bond energy [34]. For example, the depolymerization reaction of protein formed amino acid, which further occurred deamination and decarboxylation reaction, and thus released CO2 and NH3. The typical hydrolysis reaction formulas were as follows:

H  ½NH2 CH2 COn  OH þ nH2 O ! nNH2 CH2 COOH

ð8Þ

ðC6 H10 O5 Þn þ nH2 O ! nC6 H12 O6

ð9Þ

Fig. 3. The TG and DTG curves of DS in N2 atmosphere at different heating rates.

Z. Wang et al. / Applied Thermal Engineering 128 (2018) 662–671

3.3. The influence of heating rate TG and DTG curves at different heating rates (10/30/50 °C/ min) for DS in N2 atmosphere are shown in Fig. 3. Along with the increase of heating rate, the reaction during the pyrolysis process was more violent. The maximum mass loss rate increases from 2.04%/min to 10.56%/min as b varies from 10 to 50 °C/min. Meanwhile, the postponement of thermal progress occurred and the peaks deviated to high temperature. There were only one peak in SII at 30/50 °C/min heating rate while two peaks were distinguished at 10 °C/min heating rate. At higher heating rate there wasn’t enough time for minerals decomposition before char gasification occurred, therefore two interval overlapped with each other. At the same temperature, the conversion rate was lower under the high heating rate. The thermal hysteresis phenomenon is due to the presence of heat resistance [35]. Heat couldn’t transfer into the bulk sample inside instantly so the temperature in the core of a particle is lower than that on its sur-

667

face. High heating rate enlarged the temperature difference and the inside of sample couldn’t get enough energy to decomposition in time. 3.4. The influence of CO2 concentration The experimental results of DS at 10 °C/min heating rate was take as example to study the effects of replacing N2 with CO2 on pyrolysis behavior. The TG and DTG curves of DS in different atmospheres are presented in Fig. 4. When comparing the curves, no significant differences were revealed below 600 °C. Just mass loss rate and mass change after SI increased slightly with CO2 participation because of the different thermophysical properties between CO2 and N2. Namely, CO2 had limited effect on the decomposition of volatiles. However, when temperature rose to above 650 °C (SII), the influence of CO2 was obvious. The peak of this stage in CO2 atmosphere became distinct. The maximum mass loss rate increased

Fig. 4. Pyrolysis of DS in different N2/CO2 atmospheres at 10 °C/min: (a) TG curves, (b) DTG curves.

668

Z. Wang et al. / Applied Thermal Engineering 128 (2018) 662–671

Fig. 5. SEM imagines of solid products from (a) N2 atmosphere of low magnification, (b) CO2 atmosphere of low magnification, (c) N2 atmosphere of high magnification, (d) CO2 atmosphere of high magnification.

and the peak appeared earlier in CO2 environment. The residual mass loss ratio was 44.15% in CO2 atmosphere compared to 47.33% in N2 atmosphere due to the Boudouard reaction (C þ CO2 ! 2CO) [36]. The reaction was reported to be thermodynamically favored in the temperature regime above 720 °C in the study of macroalgae [37]. The SEM images of the solid products obtained in N2 and CO2 atmosphere are shown in Fig. 5. The surface of the residue obtained in N2 was rough and had granular deposits on the edge. The pores were of irregular shapes and varying sizes. The morphology revealed the sign of softening, melting and fusing. The residue obtained in CO2 showed smoother surface and higher porosity. The structure was more rounded cavity. More holes and macropores were visible while less mesopores and micropores were found from the image of higher magnification. The char gasification was responsible for this phenomenon. It facilitated the developing and widening of pores, and further, micropores coalesced into pores of larger size. It was in good agreement with relevant literatures. It was reported that the char gasification by CO2 removes carbon atoms from the interior of the particle, enlarging the opened micropores [38]. The Brunauer–Emmett–T eller (BET) surface area and total pore volume of char under CO2 atmosphere were much larger than those of char under the N2 atmosphere [38,39]. Wang et al. also found that more disor-

dered char was formed with the pyrolysis in CO2 than that in N2 [40]. In N2 atmosphere, mass loss of this interval was mainly attributed to the thermal decomposition of minerals so the residue after pyrolysis was char and ash, while in CO2, the char gasification by CO2 occupied a more important role and therefore only ash remained. The semi-quantitative EDS results (shown in Table S1 in Supplement Materials) confirmed that the carbon content in the residue of N2 atmosphere was 10.88 wt% while that of CO2 atmosphere was 3.3 wt%. The Boudouard reaction is heterogeneous between char (solid phase) and CO2 (gas phase). The increasing of CO2 concentration facilitates gas diffusion into char pores and expedites char gasification. As the CO2 concentration increased, the peak of SII shifted to lower temperature without significant change in its shape. The peak value hardly varied with the CO2 concentration, indicating when CO2 concentration is 25%, it already had high reactivity. 3.5. Kinetics parameters From the foregoing, the pyrolysis process can be divided into two stages. The described method and regression analysis are applied for the TG data to calculate kinetic parameters of each stage.

Table 2 The kinetic parameters of SI. Atmosphere

N2 CO2

Sample

DS WS DS WS

10 °C/min

30 °C/min

50 °C/min

E (kJ/mol)

ln A (s1)

R2

E (kJ/mol)

ln A (s1)

R2

E (kJ/mol)

ln A (s1)

R2

43.82 41.18 43.93 40.99

2.70 2.31 2.70 2.29

0.985 0.985 0.988 0.984

45.20 41.51 45.11 41.74

3.90 3.32 3.85 3.38

0.990 0.987 0.990 0.987

46.08 41.88 45.95 41.89

4.50 3.84 4.47 3.83

0.990 0.987 0.990 0.987

Kinetic model

Emean (kJ/mol)

(1x)2 (1x)2 (1x)2 (1x)2

45.04 41.53 45.00 41.54

669

Z. Wang et al. / Applied Thermal Engineering 128 (2018) 662–671

3.5.1. The first stage The kinetic parameters and the corresponding correlation coefficient R2 of SI in N2 and CO2 atmosphere are shown in Table 2. The activation energy E and the pre-exponential factor A for different atmospheres are nearly the same, indicating that the CO2 concentration had little effect on the first stage. Therefore, the results of other atmospheres are not listed out for reducing the excessive tables. It was observed that the most appropriate kinetic model was always (1x)2. The mean values of activation energy (Emean) of DS was 45.0 kJ/mol. The Emean of WS was 41.5 kJ/mol, lower than DS, which hinted the facilitation owing to moisture presence. That is to say, remaining certain amount of moisture in the sludge could promote the pyrolysis process as well as save drying cost. It can found from Fig. 6 that activation energy E and preexponential factor A at different heating rate increased (or decreased) simultaneously, exhibiting kinetic compensation effect (KCE). Many examples of such behavior have been reported for diverse reaction systems, illustrating the widespread KCE [41– 43]. The relationship between E and ln A corresponding to the

two samples are shown in Fig. 6(a). Two linear regression expressions are respectively:

DS

ln A ¼ 0:8368E  33:981 R2 ¼ 0:9929

ð10Þ

WS

ln A ¼ 1:8058E  71:841 R2 ¼ 0:9506

ð11Þ

1

where E is expressed in kJ/mol and A in s . The high correlation coefficients also validated the correctness of the kinetic parameters indirectly [35,44]. 3.5.2. The second stage The difference between DS and WS in second stage is not substantial and this section would focus on the pyrolysis characteristics in different atmosphere. Only the results of kinetics analysis of DS in SII are listed in Table 3 as an illustration. As shown in Table 3, the kinetic models were (1x)3/2 in N2 atmosphere and (1x) with CO2 participation, illustrating different reaction mechanism [45]. And the activation energy in CO2 atmosphere was smaller than that in pure N2 atmosphere. Compared with the first stage, the

Fig. 6. Relationship between ln A and E: (a) SI, (b) SII.

670

Z. Wang et al. / Applied Thermal Engineering 128 (2018) 662–671

Table 3 The kinetic parameters of SII. Sample

DS

Atmosphere

N2 25%CO275%N2 50%CO250%N2 75%CO225%N2 CO2

10 °C/min

30 °C/min

E (kJ/mol)

ln A (s

242.35 240.11 235.99 228.50 217.38

18.17 18.50 18.25 17.62 16.52

1

)

R

2

0.973 0.963 0.964 0.963 0.967

50 °C/min

E (kJ/mol)

ln A (s

263.47 204.60 202.24 200.07 193.40

21.78 15.11 15.03 14.92 14.36

activation energy of this stage was much higher, reflecting the activated molecule level. Meanwhile, as the CO2 concentration increased, the activation energy at the same heating rate lowered down with small extent. These results were supported by the observation of the DTG curves shown in Fig. 4. Similar findings were also pointed out in the research of macroalgae [37] and lignocellulosic substrate [18]. There was kinetic compensation effect not only for different heating rate, but also for different CO2 concentration. The linear relationship between activation energy and pre-exponential factor was represented in Fig. 6(b) and expressed by Eq. (10):

ln A ¼ 0:092E  3:5242 R2 ¼ 0:997

ð12Þ

The KCE provides a possible means to extrapolate the kinetics parameters to other operation conditions for different CO2 concentration. However, the prediction still need further and cautious verification. 3.6. Practical implications This study discussed the effect of moisture content and CO2 atmosphere on sludge pyrolysis. There are some differences on the evolution of gaseous products from DS and WS, illustrating the possibility of drying cost reduction as well as pyrolysis characteristics improvement. It also reveals that providing CO2 as carrier gas instead of N2 is a feasible method for carbon capture and storage (CCS) and carbon emission reduction. Also, the results of the kinetics parameters can provided useful information and specific guidance for the practical operation.

4. Conclusion Two stages were distinguished during the pyrolysis process of municipal sludge. CO2, CO, NH3, CH4, C@O were detected as the main gaseous products. With the existence of moisture, the releasing of volatiles was enhanced and more gas generated in the early stage. The introduction of CO2 caused more mass loss and different characteristic of solid residue by Boudouard reaction. Along with increasing of CO2 concentration, the activation energy reduced. KCE was observed in the cases of different heating rate as well as different CO2 concentration. Acknowledgements This work was supported by the National Natural Science Foundation of China (51406058, 51476060); Guangdong Key Laboratory of Efficient and Clean Energy Utilization (2013A061401005); Guangdong Natural Science Foundation (2015A030311037); the Fundamental Research Funds for the Central Universities (2015ZZ015); the Key Laboratory of Efficient and Clean Energy Utilization of Guangdong Higher Education Institutes, South China University of Technology (KLB10004); China Postdoctoral Science Foundation (2015M582382).

1

)

R

2

0.969 0.977 0.979 0.981 0.982

E (kJ/mol)

ln A (s

247.92 189.35 186.24 184.13 178.72

20.46 13.75 13.63 13.45 13.03

1

)

R

Kinetic model

Emean (kJ/mol)

(1x)3/2 (1x) (1x) (1x) (1x)

251.25 211.36 208.16 204.23 196.50

2

0.984 0.975 0.977 0.985 0.983

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.applthermaleng. 2017.09.044.

References [1] G. Yang, G. Zhang, H. Wang, Current state of sludge production, management, treatment and disposal in China, Water Res. 78 (2015) 60–73. [2] R. Han, C. Zhao, J. Liu, A. Chen, H. Wang, Thermal characterization and syngas production from the pyrolysis of biophysical dried and traditional thermal dried sewage sludge, Bioresour. Technol. 198 (2015) 276–282. [3] A. Dominguez, Y. Fernandez, B. Fidalgo, J.J. Pis, J.A. Menendez, Bio-syngas production with low concentrations of CO2 and CH4 from microwaveinduced pyrolysis of wet and dried sewage sludge, Chemosphere 70 (2008) 397–403. [4] J. Alvarez, G. Lopez, M. Amutio, J. Bilbao, M. Olazar, Preparation of adsorbents from sewage sludge pyrolytic char by carbon dioxide activation, Process Saf. Environ. 103 Part A (2016) 76-86. [5] S. Thangalazhy-Gopakumar, W.M. Al-Nadheri, D. Jegarajan, J.N. Sahu, N.M. Mubarak, S. Nizamuddin, Utilization of palm oil sludge through pyrolysis for bio-oil and bio-char production, Bioresour. Technol. 178 (2015) 65–69. [6] F. Chen, Y. Hu, X. Dou, D. Chen, X. Dai, Chemical forms of heavy metals in pyrolytic char of heavy metal-implanted sewage sludge and their impacts on leaching behaviors, J. Anal. Appl. Pyrol. 116 (2015) 152–160. [7] H. Jin, R.O. Arazo, J. Gao, S. Capareda, Z. Chang, Leaching of heavy metals from fast pyrolysis residues produced from different particle sizes of sewage sludge, J. Anal. Appl. Pyrol. 109 (2014) 168–175. [8] C.D. Benjamin Wiechmann, Dr. Christian Kabbe, Simone Brandt, Dr. Ines Vogel, Dr. Andrea Roskosch, Sewage sludge management in germany, Umweltbundesamt, Germany, 2015, pp. 25. [9] N. Gao, J. Li, B. Qi, A. Li, Y. Duan, Z. Wang, Thermal analysis and products distribution of dried sewage sludge pyrolysis, J. Anal. Appl. Pyrol. 105 (2014) 43–48. [10] W. Xiaohua, J. Jiancheng, Effect of heating rate on the municipal sewage sludge pyrolysis character, Energy Procedia. 14 (2012) 1648–1652. [11] M. Kurt, A. Aksoy, F.D. Sanin, Evaluation of solar sludge drying alternatives by costs and area requirements, Water Res. 82 (2015) 47–57. [12] B. Zhang, S. Xiong, B. Xiao, D. Yu, X. Jia, Mechanism of wet sewage sludge pyrolysis in a tubular furnace, Int. J. Hydrogen Energy 36 (2011) 355–363. [13] A. Domínguez, J.A. Menéndez, J.J. Pis, Hydrogen rich fuel gas production from the pyrolysis of wet sewage sludge at high temperature, J. Anal. Appl. Pyrol. 77 (2006) 127–132. [14] Q. Dai, X. Jiang, G. Lv, X. Ma, Y. Jin, F. Wang, Y. Chi, J. Yan, Investigation into particle size influence on PAH formation during dry sewage sludge pyrolysis: TG-FTIR analysis and batch scale research, J. Anal. Appl. Pyrol. 112 (2015) 388– 393. [15] A.G. Barneto, J.A. Carmona, J.E.M. Alfonso, J.D. Blanco, Kinetic models based in biomass components for the combustion and pyrolysis of sewage sludge and its compost, J. Anal. Appl. Pyrol. 86 (2009) 108–114. [16] X. Huang, J.-P. Cao, P. Shi, X.-Y. Zhao, X.-B. Feng, Y.-P. Zhao, X. Fan, X.-Y. Wei, T. Takarada, Influences of pyrolysis conditions in the production and chemical composition of the bio-oils from fast pyrolysis of sewage sludge, J. Anal. Appl. Pyrol. 110 (2014) 353–362. [17] J.M. Reckamp, R.A. Garrido, J.A. Satrio, Selective pyrolysis of paper mill sludge by using pretreatment processes to enhance the quality of bio-oil and biochar products, Biomass Bioenergy 71 (2014) 235–244. [18] Y. He, X. Ma, Comparative investigation on non-isothermal kinetics for thermo-degradation of lignocellulosic substrate and its chlorinated derivative in atmospheres with CO2 participation, Bioresour. Technol. 189 (2015) 71–80. [19] J. Davison, Performance and costs of power plants with capture and storage of CO2, Energy 32 (2007) 1163–1176. [20] M.B. Toftegaard, J. Brix, P.A. Jensen, P. Glarborg, A.D. Jensen, Oxy-fuel combustion of solid fuels, Prog. Energ. Combust. 36 (2010) 581–625. [21] C.Z. Lunbo Duan, Wu Zhou, Chengrui Qu, Xiaoping Chen, Investigation on coal pyrolysis in CO2 atmosphere, Energy Fuel 23 (2009) 3826–3830.

Z. Wang et al. / Applied Thermal Engineering 128 (2018) 662–671 [22] C. Guizani, F.J. Escudero Sanz, S. Salvador, Effects of CO2 on biomass fast pyrolysis: reaction rate, gas yields and char reactive properties, Fuel 116 (2014) 310–320. [23] Y. Bai, P. Wang, L. Yan, C. Liu, F. Li, K. Xie, Effects of CO2 on gas evolution and char structure formation during lump coal pyrolysis at elevated pressures, J. Anal. Appl. Pyrol. 104 (2013) 202–209. [24] J.P.R.A.W. Coats, Kinetic parameters from thermogravimetric data, Nature 201 (1964) 68–69. [25] R.S. Chutia, R. Kataki, T. Bhaskar, Thermogravimetric and decomposition kinetic studies of mesua ferrea L. deoiled cake, Bioresour. Technol. 139 (2013) 66–72. [26] Y. Liao, X. Ma, Thermogravimetric analysis of the co-combustion of coal and paper mill sludge, Appl. Energy 87 (2010) 3526–3532. [27] P. Thipkhunthod, V. Meeyoo, P. Rangsunvigit, T. Rirksomboon, Describing sewage sludge pyrolysis kinetics by a combination of biomass fractions decomposition, J. Anal. Appl. Pyrol. 79 (2007) 78–85. [28] J. Chen, L. Mu, H. Yin, X. Song, A. Li, Thermogravimetric analysis and kinetics of the combustion of refining and chemicals wastewater in different oxygen concentrations, J. Therm. Anal. Calorim. 119 (2015) 2205–2219. [29] J.M. Heikkinen, J.C. Hordijk, W. de Jong, H. Spliethoff, Thermogravimetry as a tool to classify waste components to be used for energy generation, J. Anal. Appl. Pyrol. 71 (2004) 883–900. [30] K. Kebelmann, A. Hornung, U. Karsten, G. Griffiths, Intermediate pyrolysis and product identification by TGA and Py-GC/MS of green microalgae and their extracted protein and lipid components, Biomass Bioenergy 49 (2013) 38–48. [31] Y. Lin, Y. Liao, Z. Yu, S. Fang, Y. Lin, Y. Fan, X. Peng, X. Ma, Co-pyrolysis kinetics of sewage sludge and oil shale thermal decomposition using TGA–FTIR analysis, Energy Convers. Manage. 118 (2016) 345–352. [32] X. Peng, X. Ma, Y. Lin, Z. Guo, S. Hu, X. Ning, Y. Cao, Y. Zhang, Co-pyrolysis between microalgae and textile dyeing sludge by TG–FTIR: Kinetics and products, Energy Convers. Manage. 100 (2015) 391–402. [33] L. Tian, B. Shen, H. Xu, F. Li, Y. Wang, S. Singh, Thermal behavior of waste tea pyrolysis by TG-FTIR analysis, Energy 103 (2016) 533–542.

671

[34] W.P.F. Barber, Thermal hydrolysis for sewage treatment: a critical review, Water Res. 104 (2016) 53–71. [35] G. Mishra, J. Kumar, T. Bhaskar, Kinetic studies on the pyrolysis of pinewood, Bioresour. Technol. 182 (2015) 282–288. [36] M.F. Irfan, M.R. Usman, K. Kusakabe, Coal gasification in CO2 atmosphere and its kinetics since a brief review, Energy 36 (2011) (1948) 12–40. [37] E.E. Kwon, Y.J. Jeon, H. Yi, New candidate for biofuel feedstock beyond terrestrial biomass for thermo-chemical process (pyrolysis/gasification) enhanced by carbon dioxide (CO2), Bioresour. Technol. 123 (2012) 673–677. [38] C. Jindarom, V. Meeyoo, B. Kitiyanan, T. Rirksomboon, P. Rangsunvigit, Surface characterization and dye adsorptive capacities of char obtained from pyrolysis/gasification of sewage sludge, Chem. Eng. J. 133 (2007) 239–246. [39] X. Liu, M. Xu, J. Si, Y. Gu, C. Xiong, H. Yao, Effect of sodium on the structure and reactivity of the chars formed under N2 and CO2 atmospheres, Energy Fuels 26 (2012) 185–192. [40] B. Wang, L. Sun, S. Su, J. Xiang, S. Hu, H. Fei, Char structural evolution during pyrolysis and its influence on combustion reactivity in air and oxy-fuel conditions, Energy Fuels 26 (2012) 1565–1574. [41] G. Skodras, G. Nenes, N. Zafeiriou, Low rank coal – CO2 gasification: experimental study, analysis of the kinetic parameters by Weibull distribution and compensation effect, Appl. Therm. Eng. 74 (2015) 111–118. [42] M.C. Wilson, A.K. Galwey, Compensation effect in heterogeneous catalytic reactions including hydrocarbon formation on clays, Nature 243 (1973) 402– 404. [43] G. Skodras, Catalysis and compensation effect of K2CO3 in low-rank coal - CO2 gasification, Cent. Eur. J. Chem. 11 (2013) 1187–1200. [44] Z. Yao, X. Ma, Y. Lin, Effects of hydrothermal treatment temperature and residence time on characteristics and combustion behaviors of green waste, Appl. Therm. Eng. 104 (2016) 678–686. [45] C. Jindarom, V. Meeyoo, T. Rirksomboon, P. Rangsunvigit, Thermochemical decomposition of sewage sludge in CO2 and N2 atmosphere, Chemosphere 67 (2007) 1477–1484.