(Co-)combustion of additives, water hyacinth and sewage sludge:
Thermogravimetric, kinetic, gas and thermodynamic modeling analyses
Jingyong Liu
a,⇑, Limao Huang
a, Guang Sun
a, Jiacong Chen
a, Shengwei Zhuang
a, Kenlin Chang
a,
Wuming Xie
a, Jiahong Kuo
a, Yao He
a, Shuiyu Sun
a, Musa Buyukada
b, Fatih Evrendilek
b,ca
Guangzhou Key Laboratory of Environmental Catalysis and Pollution Control, School of Environmental Science and Engineering, Institute of Environmental Health and Pollution Control, Guangdong University of Technology, Guangzhou 510006, China
b
Department of Environmental Engineering, Abant Izzet Baysal University, 14052 Bolu, Turkey c
Department of Environmental Engineering, Ardahan University, 75002 Ardahan, Turkey
a r t i c l e i n f o
Article history: Received 26 May 2018 Revised 16 September 2018 Accepted 17 September 2018 Available online 10 October 2018 Keywords: Sewage sludge Water hyacinth Co-combustion Additives TG-MS
a b s t r a c t
Additives and biomass were co-combusted with sewage sludge (SS) to promote SS incineration treatment and energy generation. (Co-)combustion characteristics of sewage sludge (SS), water hyacinth (WH), and 5% five additives (K2CO3, Na2CO3, Mg2CO3, MgO and Al2O3) were quantified and compared using thermogravimetric-mass spectrometric (TG-MS) and numerical analyses. The combustion performance of SS declined slightly with the additives which was demonstrated by the 0.03-to-0.25-fold decreases in comprehensive combustibility index (CCI). The co-combustion performed well given the 0.31-fold increase in CCI. Kinetic parameters were estimated using the Ozawa-Flynn-Wall (OFW) and Kissinger-Akahira-Sunose (KAS) methods. Apparent activation energy estimates by OFW and KAS were consistent. The addition of K2CO3and MgCO3decreased the weighted average activation energy of SS. Adding K2CO3 to the blend reduced CO2, NO2, SO2, HCN and NH3emissions. CO2, NO2and SO2emissions were higher from WH than SS. Adding WH or K2CO3to SS increased CO2, NO2and SO2but HCN and NH3emissions. Based on both catalytic effects and evolved gases, K2CO3was potentially an optimal option for the cat-alytic combustion among the tested additives.
Ó 2018 Elsevier Ltd. All rights reserved.
1. Introduction
The increasing waste stream of sewage sludge (SS) stems from the rapid growth rates of global urbanization and human popula-tion (Sebestyén et al., 2017; Liu et al., 2017a,b) and amounted to about 30 mega tons (Mt) (with 80% moisture content) in China in 2015 (Huang et al., 2016). Contaminated with (in)organic pollu-tants, SS poses harmful effects on the public and environment when inappropriately disposed. The conventional methods of SS disposal include landfills, agricultural applications, and thermal processing (Cieslik et al., 2015;Xie et al., 2018a; Liu et al., 2015). Today’s perception is changing from seeing SS as an unwanted waste to seeing it as a beneficial biofuel, thus helping to reduce environmental pollution and costs (Liu et al., 2016b; Sebestyén et al., 2017; Xie et al., 2018b). Therefore, (co-)combustion of SS has been applied as one of the most feasible treatments (Otero et al., 2002; Jiang et al., 2016; Zhuo et al., 2017). Since the recent operative costs of industrial (co-)combustion are relatively high,
there still exist rooms for significant improvements to enhance co-combustion characteristics of SS and to reduce associated emis-sions (Cieslik et al., 2015). In so doing, several methods have been adopted such as the optimization of SS co-combustion (Han and Bollas, 2016), the use of catalysts or additives (e.g., ZSM-5, HZSM-5, Y-zeolite, FCC, and metal oxides) (Utton et al., 2001; Syed-Hassana et al., 2017), and the addition of biofuels.
There is an increasing amount and variety of biofuel feedstocks from which to obtain energy such as biocrops, silvicultural and agricultural wastes, and treatments of by-products (Chen et al., 2017; Huang et al., 2016). Water hyacinth (WH) (Eichhornia crassipes) was reported as one of the world’s top ten ‘‘invasive grasses” due to its rapid proliferation (Villamagna and Murphy, 2010). However, its rapid and excessive spread paves the way for its use in biofuel production (Zimmels et al., 2009), bioremediation (Gangulya et al., 2012), bioethanol and gas production (Aswathy et al., 2010; Mishima et al., 2008), feed production and adsorbent preparation (Guerrero-Coronilla et al., 2015), and pyrolysis (Luo et al., 2011). Only a few studies have explored the co-combustion process of WH.
https://doi.org/10.1016/j.wasman.2018.09.030
0956-053X/Ó 2018 Elsevier Ltd. All rights reserved.
⇑ Corresponding author.
E-mail address:[email protected](J. Liu).
Contents lists available atScienceDirect
Waste Management
Sewage sludge is known as a high ash material on a dry basis. The main inherent ash forming elements are Al, Ca, Fe, K, Mg, Na, P, S, and Si, together with trace amounts of Cl, Cu, Ti, and Zn. They could exist as oxides, silicates, carbonates, sulfates, chlorides, and phosphates (Shao et al., 2010). Previous studies (Yang et al., 2006; Yu et al., 2009; Shao et al., 2010; Abbasi-Atibeh and Yozgatligil, 2014; Fang et al., 2016) showed that mineral matters may have positive or negative effects on the thermal degradation process of solid fuels. Yang et al. (2006)found that most of the mineral additives (KCl, Na2CO3, CaMg(CO3)2, Fe2O3, and Al2O3)
demonstrated negligible effects on the pyrolysis of palm oil wastes, while K2CO3was found to inhibit the pyrolysis of hemicellulose but
promote the degradation of cellulose.Yu et al. (2009)stated that the four catalysts of MgO, CuO, NiO, and CaO enhanced the ignition and combustion of rice straw.Shao et al. (2010)reported that the metal oxides (Al2O3, CaO, Fe2O3, TiO2, and ZnO) accelerated its
ini-tial pyrolysis decomposition of SS, with pyrolysis time decreased by Al2O3 and TiO2 and prolonged by CaO, Fe2O3, and ZnO.
Abbasi-Atibeh and Yozgatligil (2014)showed that K-based catalyst had high char reactivity and decreased the burnout temperature of lignite.Fang et al. (2016)reported that MgO was the best one out of the MgO, Al2O3 and ZnO additives, given the contrast of the
pyrolysis characteristics. However, there existed only a few studies about the co-combustion behaviors of SS and WH combined with additives by using thermogravimetry and mass spectrometry (TG-MS) analyses.
Thermogravimetry and mass spectrometry can provide the real-time monitoring of flue gas changes with temperature and the qualitative comparison of relevant combustion parameters and gas emission patterns for practical applications. Therefore, this study aimed at thermogravimetric, kinetic, gas and thermody-namic analyses of SS, WH and SW (SS + WH = 80% + 20% blend) co-combustion with the five additives (K2CO3, Na2CO3, MgCO3,
MgO, and Al2O3) based on the TG-MS, iso-conversional and
numer-ical analyses.
2. Materials and methods
2.1. Materials
SS samples were continuously collected at intervals of 0.5 and 8 h from a terminal conveyor belt of a wastewater treatment plant in Guangzhou. WH samples were gathered from the canals of Guangz-hou University Mega Center in the Guangdong Province, China. They were allowed to dry naturally in one week at room temperature in laboratory, crushed and sieved into the desired particle size of less than 74
l
m. They were then dried at 105°C until reaching a constant weight and kept in a desiccator for subsequent analyses. Their ultimate, proximate, calorific and ash analyses are presented in our previous study ofHuang et al. (2018)(SeeTable 1).The following five inexpensive and colorless additives were directly acquired: K2CO3(purity 99.0%), Na2CO3(purity 99.8%),
MgCO3(purity between 83.68 and 98.32%), MgO (purity 98.0%)
and Al2O3 (purity 98.0%). The SW ratio of 80% SS to 20% WH
was adopted for comparisons with or without the use of the addi-tives according to findings about optimal SW blends byHuang et al. (2016 and 2018).
2.2. TG-MS analyses
TG-MS analyses were carried out using the three heating rates of 10, 20 and 40°C min1, the flow rate of 50 ml min1, and a
DSC–TGA (NETZSCH STA 409 PC) in the range of 30–1000°C. Each analysis involved a sample of about 8 ± 0.5 mg in alumina crucible. Initially, experiments without samples were designed to measure the baseline to eliminate instrumental systematic errors. Replications in triplicates were used to ensure that random errors fell within ±2%. TG curve and its derivative TG (DTG) curve were obtained using the kinetic software of NETZSCH-T4.
Emissions evolved under 20°C min1 at 300 ml min1 were
monitored using a TG-MS (Rigaku Thermo Mass Photo, Japan)
Table 1
Ultimate, proximate, calorific value and ash composition analyses of SS and WH on an air-dry basis (Huang et al., 2018).
Sample Ultimate analyses (%) Proximate analyses (%) Qnet,da(MJ kg1)
C H Ob N S Mc Vd Ae FCf SS 24.13 3.94 12.49 4.50 0.74 7.57 40.22 46.63 5.58 10.79 WH 36.62 5.28 27.49 3.01 0.25 9.95 56.30 17.40 16.35 14.77
Ash composition analyses (%)
Na2O MgO P2O5 Al2O3 SiO2 K2O CaO TiO2 MnO Fe2O3
SS 0.14 0.25 0.23 1.59 2.81 0.04 11.20 0.20 0.05 14.28
WH 0.18 / 0.35 0.18 0.41 14.11 0.23 0.01 0.05 0.10
aQ
net, d, higher heating value on an air-dried basis. b O, calculated by O = 100%–C–H–N–S–MA. c M, moisture. d V, volatile matters. e A, ash. f FC, fixed carbon. Nomenclature Symbols/abbreviations Ti ignition temperature (°C) Tp peak temperature (°C) Mr residual mass at 1000°C (%)
a
mass conversion degree (%) Ea activation energy (kJ mol1)Em weighted average activation energy (kJ mol1)
Rp reaction rate at the peaks (% min1)
Tb peak temperature of 98% conversion (°C)
Tm DTGmaxpeak temperature (K)
Rv average reaction rate ranging from ignition temperature
to final temperature (% min1)
under the 20%O2/80%He atmosphere. The data were normalized for
each m/z to have a response factor (Otero et al., 2011; Huang et al., 2018) and for the samples to be comparable to one another in terms of molecule intensity even if they had the same parents such as CO2, NO2, NO, SO2, HCN (hydrogen cyanide), and NH3.
2.3. Kinetic analyses
Heterogeneous solid-state reaction rates are generally stated using Eq.(1):
d
a
dt¼ k Tð Þf ð
a
Þ ð1ÞGiven the below Arrhenius equation:
k Tð Þ ¼ AeðEa=RTÞ ð2Þ
where
a
= conversion degree; t = time; T = reaction temperature; A = pre-exponential factor; Ea = apparent activation energy; and R = universal gas constant (8.314 J/K mol1).a
was described thus (Chen et al., 2013):a
¼ m0mtm0 m1 ð3Þ
wherem0, m1 and mt= sample masses that were initial, final and
at time t, respectively.
Eqs.(1) and (2)were combined thus:
d
a
dt¼ AeðEa=RTÞfð
a
Þ ð4ÞWhen the following heating rateðbÞ (°C s1) was introduced: b ¼dT
dt ð5Þ
Eq.(4)became thus:
bddT
a
¼ AeðEa=RTÞfða
Þ ð6ÞKinetic analyses of reaction behaviors are pivotal to the estab-lishment of robust processes as biomass (co-)combustion behavior is closely coupled to its thermal decomposition mechanism (Amanda and Leandro, 2016). The integral iso-conversional model of Ozawa-Flynn-Wall (OFW) was adopted to predict average acti-vation energy (Ea) as was expressed in Eq.(7)(Kim et al., 2010):
lnð Þ ¼ Cb aRE Ta ð7Þ
where Ca= function of
a
; and T = absolute temperature (K). The integral iso-conversional linear model of Kissinger-Akahira-Sunoseis (KAS) was expressed as follows (Mishra and Bhaskar, 2014):ln b T2
¼ CaRE Ta ð8Þ
For a constant
a
, Eawas estimated using a linear fit to the plot of ln (b/T2) versus 1/T. Influence of the additives on the kinetic parame-ters was determined estimating the weighted average activation energy (Em) thus (Shao et al., 2010):
Em¼ E1F1þ E2F2þ . . . þ EnFn ð9Þ
where E1to En= Eavalues at each stage; and F1to Fn= weight losses.
Co-combustion parameters quantified from the TG-DTG curves with/without the additives were ignition temperature (Ti), peak
temperature (Tp), burnout temperature (Tb), maximum weight loss
rate (Rp), and average weight loss rate (Rv). Also, comprehensive
combustibility index (CCI) was used to assess co-combustion
per-formance. CCI was based on characteristic temperatures and reac-tion rates thus (Huang et al., 2016):
CCI¼ Rp
Rð vÞ
T2iTb
ð10Þ
2.4. Equilibrium predictive modeling
Reducing the formation of gases containing Cl, K and Na can effectively abate deposit formation, slagging, corrosion, and emis-sions during the thermal conversion of biofuels. To better under-stand the transformations of the additives between 400 and 1800 K, the thermodynamic model of FactSage 7.1 was used. Based on the minimization of the Gibbs energy of a given system, Fact-Sage uses thermodynamic equilibrium calculations for the thermal behaviors of the metal compounds (Liao et al., 2015). The transfor-mations of the alkali metals occur at high temperatures where kinetic limitations, mass transports, and chemical potentials all influence the calculated phases. The proximate and ultimate anal-yses of the fuels and the five additives (5%) (Table 1) were used as inputs in the calculations. In so doing, 14 chemical elements were identified for the co-combustion. The four main elements (C, H, N, and O), the eight minerals (Na2O, MgO, P2O5, Al2O3, SiO2, K2O, CaO,
and Fe2O3) (Table 1), and the five additives were considered. The
gas and condensed phases were assumed to be ideal and pure, respectively. The co-combustion temperature varied between 400 and 1800 K, and the pressure was set to 1 atm (1.013 105Pa).
Fig. 1. TG-DTG curves of SS, SS + K2CO3, SS + Na2CO3, SS + MgCO3, SS + MgO, SS + Al2O3, and SS + WH (Huang et al., 2018) under air atmosphere at the heating rate of 20°C min1.
The temperature step was chosen as 100 K, while the excess air ratio (k) was set to 1.5 for all the calculations, a typically set value for SS combustion. The air was assumed to consist (mole fractions) of 79% N2 and 21% O2. The final equilibrium state was obtained
determining all the possible species derived from K, Na, Mg, and Al. The thermodynamic analyses served to determine the target species among the matrix components under specific co-combustion conditions and to predict their possible interactions (seeTable 6).
3. Result and discussion
3.1. Thermogravimetric analyses of SS with or without additives
Fig. 1a and b present the TG and DTG curves with or without the five additives at 20°C min1. In order to carry on the deep compar-ison, the TG and DTG curves of SW (SS + WH) in our previous stud-ies (Huang et al., 2018) were also added inFig. 1. The TG–DTG curves of SS with and without the additives had similar trends and showed four distinct stages of mass loss. Regardless of the samples, the initial stage of mass loss occurred approximately between 25 and 175°C due to evaporation. The second stage (175–390°C) involved the emissions of organic volatiles, whereas the third stage (390–700°C) was related to the combustion of more complex and thermally stable structures, and char oxidation (Liu et al., 2016a). During these stages, the SS mass decreased rapidly by about 92.78%. The last step (> 700°C) was the slow decomposition rate of carbonaceous residues with a slight mass loss. The second peak of the DTG curves of SS + MgCO3pointed
pri-marily to the breakdown of (MgCO3)4Mg(OH)2at 350°C into MgO.
Similar results were reported for the SW co-combustion (Huang et al., 2018).
The additives had different effects on the (co-)combustion parameters of SS (Table 2). The additives lowered Tiand Tpto the
ranges of 1–6°C and 2–8 °C, respectively, which indicated better catalytic activity (Shen and Qin, 2006). The TG curve of SW dipped into a lower temperature range than did that of SS+additives. The volatile combustion stage coincided with the maximum weight loss rate for both SS and SW. However, SW had a greater maximum weight loss rate (5.39% min1) than did SS (3.96% min1). This case indicated that WH boosted the SS combustion and evaporation vig-orously (Huang et al., 2016). The residual weights of SS, SS + K2CO3,
SS + Na2CO3, SS + MgCO3, SS + MgO, SS + Al2O3and SW were
esti-mated at 48.63, 49.65, 50.12, 48.46, 51.47, 52.76 and 42.30%, respectively. The reason for this may be attributed to the difference in weight loss rate between WH and SS which in turn positively
impacted the SS co-combustion (Huang et al., 2016). The residual mass (Mr) increased with the additives (Muhammad et al., 2012).
The CCI estimates were used to quantify the (co-)combustion behaviors of SS, SS+additives and SW (Table 2). The larger CCI val-ues pointed to a more vigorous burning and faster char burnout. The combustion performance of SS declined slightly with the addi-tives which was demonstrated by the 0.03-to-0.25-fold increase in CCI. However, the co-combustion performed well considering the 0.31-fold increase in CCI.
Fig. 2. TG–DTG curves of SS with K2CO3at three heating rates. Table 2
Characteristic parameters based on TG–DTG curves at the heating rate of 20°C min1.
Sample Tia(°C) Tpb(°C) Rpc(% min1) Rvd(% min1) Mre(%) Tbf(°C) CCIg(108)
Tp1 Tp2 Tp3 Rp1 Rp2 Rp3 SS 237 288 528 / 3.96 3.29 / 1.092 48.63 647.78 2.565 SS + K2CO3 231 283 526 / 3.89 4.01 / 1.071 49.65 860.17 1.932 SS + Na2CO3 236 285 518 / 3.70 3.32 / 1.052 50.12 644.52 2.329 SS + MgCO3 236 285 413 526 3.87 2.62 3.11 1.086 48.86 653.18 2.482 SS + MgO 234 285 527 / 3.77 2.98 / 1.028 51.47 659.31 2.285 SS + Al2O3 235 286 527 / 3.66 3.09 / 1.002 52.76 608.93 2.332 SW (Huang et al., 2018) 242 296 524 / 5.39 3.07 / 1.229 42.30 645.59 3.360 a
Tiis ignition temperature for volatile release. b
Tpis temperature associated with Rp. c R
pis maximum weight loss rate. d R
vis average mass loss rate. e
Mris residual mass at 1000°C. f
Tbis temperature of 98% conversion. g
CCI is comprehensive combustion index, unit is %2
3.2. Effects of heating rate on SS combustion with K2CO3
The TG and DTG curves of SS + K2CO3at 10, 20 and 40°C min1
are shown inFig. 2.Table 3shows the effect of heating rate on the combustion characteristic parameters clearly. The heating rate affected the combustion behavior of SS+K2CO3significantly in that
the peaks shifted to the right with the increased heating rate. The peak temperatures of 273, 285 and 293°C belonged to 10, 20 and 40°C min1, respectively. The reason for these findings may be related to the difference between external and internal tures of the sample in response to the rapidly increased tempera-ture of the reaction system with the increased heating rate. The non-isothermal heating due to the heat diffusion (Chen et al., 2017; Kan et al., 2014a; Kan et al., 2014b). The low heating rate caused a gradual heating, and thus, a better heat transfer, whereas the high heating rate generated a steep temperature gradient, and thus, a reaction lag (Chen et al., 2017).
As shown inTable 3, with the increased heating rate, Ti, Tp1, Tp2
and Tb, rose from 220 to 235°C, 273 to 293 °C, 499 to 550 °C, and
862.83 to 889.46°C, respectively. The Rp of all of the steps (Rp1
and Rp2) and Rv markedly increased asb increased from 10 to
40°C min1. Asb increased, the TG curves shifted to a higher tem-perature gradually, with the final masses of 50.10, 49.65 and 49.73% at 10, 20 and 40°C min1, respectively. Thus, the heating
rate did not affect Mrsignificantly. In addition, the combustion
per-formance of SW was evaluated by the combustion index of CCI. When b increased from 10 to 40 °C min1, the CCI grew from
0.511 to 7.664 108 %2
K3min2. This relationship illustrates that increasing the heating rate was beneficial to the separation and combustion of the volatiles. Similar findings were also
reported about the combustion, kinetic and thermodynamic analy-ses of spent mushroom substrate (Huang et al., 2018).
3.3. Kinetic analyses
The coefficients of determination (r2) values for the OFW- and
KAS-based Eaestimates ranged from 95.23 to 99.99%, indicating
that mass conversion accounted for most of the variability in Ea.
In order to obtain the results of comparative analysis, the Eaof
SW (Huang et al.,2018) was added to theFig. 3also. As shown in
Fig. 3, the Eaestimates and trends by KAS and OFW were close to
one another. Eawas highly variable during the(co-)combustion as
the SS reaction involved a non-linear multiple-step (instead of a simple one-step) mechanism (Xu and Chen, 2013). The Eavalues
varied with the increased mass conversion rate, possessing almost the same tendency of variation with the additives (Fig. 3). For example, Ea for SS + K2CO3 decreased steeply from 191.3 to
120.6 kJ mol1at 0.2 <
a
< 0.6 and slowly increased ata
> 0.6 (with thermal decomposition and volatile degradation ata
= 0.1–0.6, and fixed carbon combustion ata
> 0.65).The SS(co-)combustion with and without the additives had the two major stages with a relative mass loss (Fig. 3). As shown in
Table 4, the Eavalue of the first stage was higher than that of the
second stage. In the low temperature range, the SS mass loss varied between 45.45 and 51.76% with the additives. In the high temper-ature region, the SS mass loss ranged from 42.34 to 47.00%. The OFW-based Emestimates for SS, SS + 5%K2CO3, SS + 5%Na2CO3, SS
+ 5%MgCO3, SS + 5%MgO, and SS + 5%Al2O3 were estimated at
154.7, 149.1, 218.3, 136.4, 245.0 and 181.2 kJ mol1, respectively. This indicated that K2CO3and MgCO3decreased Em.
Fig. 3. Changes in Eaas a function of mass conversion rate according to the OFW and KAS methods. Table 3
Combustion characteristic parameters of SS + K2CO3at three heating rates (b in °C min1).
b Ti(°C) Tp(°C) Rp(% min1) Rv(% min1) Mr(%) Tb(°C) CCI (108)
Tp1 Tp2 Rp1 Rp2
10 220 273 499 1.91 1.75 0.516 50.10 862.83 0.511
20 227 285 526 3.88 4.01 1.071 49.65 860.17 1.932
Table 4
Kinetic parameters at the heating rate of 20°C min1.
Samples Temperature range (°C) Ea(kJ/mol) F(%) Em(kJ/mol)
OFW KAS OFW KAS
SS 175–398 202.2 192.2 47.13 154.7 144.3 398–674 130.8 118.2 45.42 SS + 5%K2CO3 152–393 184.0 174.2 47.19 149.1 139.1 393–662 147.1 134.5 42.34 SS + 5%Na2CO3 188–393 312.2 312.0 45.45 218.3 217.9 393–667 165.6 165.1 46.13 SS + 5%MgCO3 168–386 154.6 154.2 46.14 136.4 135.8 386–690 138.4 137.6 47.00 SS + 5%MgO 180–387 211.0 210.0 46.18 161.3 160.5 387–682 139.1 138.7 45.79 SS + 5%Al2O3 180–412 354.5 354.0 51.04 245.0 244.7 412–692 150.5 150.2 42.60 SW (Huang et al., 2018) 175–403 216.04 216.0 51.76 181.2 180.8 403–683 153.96 153.6 45.13
Fig.4. Emission curves of SS, WH, SS + K2CO3and two samples (SW and SW + K2CO3) (Huang et al., 2018) for: (a) CO2, (b) NO2, (c) NO, (d) SO2, (e) NH3, and (f) HCN at the heating rate of 20°C min1.
3.4. Evolved gas analyses
In the (co-)combustion of SS, WH and SS + K2CO3, the main
ion-ized fragments with m/z = 17, 18, 28, 30, 44, 46, 60 and 64 showed NH3, H2O, CO, NO, CO2, NO2, COS and SO2as the most probable
par-ent molecules. In order to carry on the deep comparison, our pre-vious studies including the curves of two samples (SW, SW + K2CO3) and their gas analysis results (Huang et al., 2018) were
also added in this paper. Particular care must be taken when reporting some ions as they could belong to various compounds. For example, ions with m/z = 30 were related to the evolution of such compounds as NO, primary amines (CH4N) and C2H6. Ions
with m/z = 27 could be assigned to HCN or C2H3 ions, Ions with
m/z = 28 could be assigned to CO or C2H4ions, and m/z = 17
repre-sented OH fragment of H2O in addition to NH3 (Huang et al.,
2018).
In this section, the main gas pollutants SO2, NO2, CO2, NO, HCN
and NH3are mainly discussed (Otero et al., 2002; Miranda et al.,
2012; Huang et al., 2018). The (co-)combustion occurred continu-ously during which CO2, NO, NH3, and HCN were majorly released
(Fig. 4). The trends of the DTG curves (Fig. 1) were consistent with those of the ion intensities of the emissions evolved from the two stages (Fig. 4). Similar results were given for biomass pyrolysis and (co-)combustion (Huang et al., 2011; Miranda et al., 2012; Huang et al., 2018). There were continuous CO2and NOxemissions during
the whole process but primarily in the (co-)combustion stage with a higher temperature range (450–700°C) (Fig. 4a–c). The addition of WH to SS increased NOxemissions which peaked towards a high
temperature range due to the high nitrogen content of WH, and higher temperatures needed for the SS (co-)combustion. SO2and
NH3were released primarily in the (co-)combustion stage with a
lower temperature range (200–400°C) (Fig. 4d–e). SO2emissions
were higher from WH than SS and SW. SO2emissions from SW
began at 200°C and peaked at 320–390 °C. The release of HCN (Fig. 4f) from SS occurred during the whole combustion process. HCN may adversely affect human health due to its toxicity (Jonathan et al., 2008).
Table 5indicates the normalized intensities of CO2, NO2, SO2,
HCN and NH3 emissions. CO2 emissions were higher than SO2
and NO2emissions as supported by the proximate and ultimate
analyses (Table 1). The addition of K2CO3 to SW decreased the
emissions which peaked in the low temperature region (Fig. 4). This suggests that the addition of K2CO3to SW was very effective
in the pollutant removal from the entire co-combustion process (Huang et al., 2018). CO2, NO2and SO2emissions were higher from
WH than SS. The addition of WH or K2CO3to SS increased CO2, NO2
and SO2emissions but reduced HCN and NH3emissions.
3.5. FactSage numerical analyses
The main K compounds formed once K2CO3was added (Fig. 5a)
were K2SO4, KAlSiO4, and K2Ca2(CO3)3between 400 and 800 K and
KAlO2, K2SO4, KOH(g), and KAlSiO4 between 800 and 1600 K. At
above 1100 K, KOH(g) was generated. At above 1550 K, KOH(g) accounted for 50% of K. At above 1300 K, a small amount of
Table 5
Normalized intensities at the heating rate of 20°C min1. Samples Normalized intensities (106A/g)
SO2 NO2 CO2 NO HCN NH3 SS 1.5118 1.915 471.61 31.84 5310 33,890 WH 172.3 1120 267,200 11,500 2220 43,040 SW (Huang et al., 2018) 140 159 36,850 1560 1407 6874 SS + K2CO3 127.67 418 105,487 4691 2170 24,035 SW + K2CO3 (Huang et al., 2018) 0.96 1.17 283.13 14.09 4.14 66.34
Fig.5. Thermodynamic equilibrium distributions of K, Na, Mg and Al with the addition of (a) K2CO3, (b) Na2CO3, (c) MgCO3, (d) Al2O3and (e) MgO.
K2SO4(g) was generated. At above 1600 K, K2SO4(g) peaked
accounting for 25% of K.
Having added Na2CO3(Fig. 5b), the main Na compounds formed
were Na2SO4, NaAlSiO4, and Na2Ca2(CO3)3between 400 and 800 K
and NaNO3, Na2CO3, Na2SO4, and Na2CO3between 800 and 1600 K.
At above 1200 K, NaOH(g) was generated. NaOH(g) accounted for 15% at 1500 K and 32% at 1600 K of Na. At above 1400 K, a small amount (less than 5%) of Na2SO4(g) was generated.
Likewise, Mg compounds were primarily Mg2SiO4, MgAl2O4,
(MgO)(Fe2O3), and MgOCa3O3Si2O4 between 400 and 800 K. The
major Al compounds were Al2O3, MgAl2O4, and Mg5Al2Si3O10(OH)8
between 400 and 800 K. The main Mg compounds appeared as (MgO)(Fe2O3), Mg2SiO4, MgAl2O4, and MgOCa3O3Si2O4 between
400 and 800 K and as (MgO)(Fe2O3), MgO, MgOCa3O3Si2O4, and
MgAl2O4between 800 and 1800 K.
Overall, with the five additives, only K and Na had the gaseous form of KOH(g) at the high temperature, with all the other prod-ucts being in the solid state. The alkali metals deposited on the superheater were reported as the cause of the corrosion (Back, et al., 2017; Li et al., 2015). Alkali chloride was generated easily at the high temperature, and subsequently, condensed on the heat-ing surface (Blaesing and Mueller, 2013). Alkali metal might be formed in the molten state at a lower temperature which made it easier to adhere to the heating surface causing corrosion (Qi et al., 2017; Xiao, 2006; Li et al., 2012).
The forms of K-, Na-, Mg- and Al-containing compounds corre-sponding to Ti, Tp1, Tp2, and Tbwere also observed (Table 6). The
compounds corresponding to Tp1and Tp2were different most likely
due to the intermediate products of the additives (Back et al., 2017; Li et al., 2015). With the alkali metal additives, the additives were uniformly attached to carbon of SS so as to increase its active sur-face. The additives promoted the oxidation reaction to produce an unstable intermediate metal oxide which was used as a carrier to transfer oxygen to the carbon atoms of SS (Putun et al., 2008; Yan et al., 2018). During the entire catalytic combustion process, the reduced metal element reacted with oxygen again, and the indefinite, unstable intermediate oxide played a significant role in the oxygen transfer (Miró et al., 1999; Querini et al., 1999). In the late period of burning SS, some unstable intermediate metal oxides were converted into stable oxides, and the catalytic activity
decreased (Serra et al., 1997; Jiménez et al., 2006a; Jiménez et al., 2006b). The catalytic effect of the alkali metal was not obvious after the late stage of the SS combustion (after the second peak of weight loss) which was consistent with the TG-DTG curve (Fig. 1) (Table 6).
4. Conclusion
The additives and water hyacinth were co-combusted with sewage sludge using the thermogravimetric-mass spectrometric and numerical analyses. The combustion performance of SS declined slightly with the additives according to the 0.03-to-0.25-fold decrease in CCI, while the co-combustion performed well considering the 0.31-fold increase in CCI. The Eaestimates by the
OFW and KAS methods were consistent with one another. K2CO3
and MgCO3decreased the Emestimate of SS by OFW. The addition
of K2CO3to SW reduced CO2, NO2, SO2, HCN and NH3emissions
from the entire co-combustion process. CO2, NO2and SO2
emis-sions were higher from WH than SS. Adding WH or K2CO3to SS
increased CO2, NO2 and SO2 emissions but decreased HCN and
NH3emissions. Based on both catalytic effects and evolved gases,
K2CO3was potentially an optimal option for the catalytic
combus-tion among the tested additives.
Acknowledgments
This work was financially supported by the Scientific and Technological Planning Project of Guangzhou, China (No. 201704030109; 2016201604030058), and Guangdong Special Support Program for Training High Level Talents (No. 2014TQ01Z248), and the Science and Technology Planning Project of Guangdong Province, China (No. 2017A050501036; 2017A040403059).
References
Abbasi-Atibeh, E., Yozgatligil, A., 2014. A study on the effects of catalysts on pyrolysis and combustion characteristics of Turkish lignite in oxy-fuel conditions. Fuel 115, 841–849.
Amanda, A.D.M., Leandro, C.M., 2016. Kinetic parameters of red pepper waste as biomass to solid biofuel. Bioresour. Technol. 204, 157–163.
Table 6
The forms of K, Na, Mg and Al at Ti, Tp1, Tp2and Tbcorresponding to TDG curves at the heating rate of 20°C min1.
Ti Tp1 Tp2 Tb
K2CO3 K2SO4 K2SO4 K2SO4 KAlO2
KAlSiO4 KAlSiO4 KAlSiO4 K2SO4
KAlSi2O6 KAlSi2O6 K2Ca2(CO3)3
KMg3AlSi3O10(OH)2 KMg3AlSi3O10(OH)2 KMg3AlSi3O10(OH)2
Na2CO3 NaNO3 NaNO3 Na2SO4 Na2SO4,
Na2CO3 Na2CO3 Na2CO3 Na4CaSi3O9
Na2Ca3Si6O16 Na2Ca3Si6O16 Na2Ca2Si3O9 Na3PO4,
NaAl9O14 NaAl9O14 NaAl9O14 NaOH(g)
Na2SO4 Na2SO4
Na2Ca2Si3O9 Na2Ca2Si3O9
MgCO3 Mg5Al2Si3O10(OH)2 Mg5Al2Si3O10(OH)2 (MgO)(Fe2O3) MgOCa3O3Si2O4
CaMg(CO3)2 CaMg(CO3)2 Mg2SiO4 MgO
KMg3AlSi3O10(OH)2 KMg3AlSi3O10(OH)2 MgAl2O4 MgAl2O4
Mg2SiO4 Mg2SiO4
MgAl2O4 MgAl2O4
Al2O3 Al2O3 Al2O3 MgAl2O4 MgAl2O4
MgAl2O4 MgAl2O4
Mg5Al2Si3O10(OH)8 Mg5Al2Si3O10(OH)8
MgO Mg5Al2Si3O10(OH)2 Mg5Al2Si3O10(OH)2 MgOCa3O3Si2O4 MgOCa3O3Si2O4
CaMg(CO3)2 CaMg(CO3)2 (MgO)(Fe2O3) MgO
Mg2SiO4 Mg2SiO4 MgAl2O4 MgAl2O4
(MgO)(Fe2O3) (MgO)(Fe2O3)
Aswathy, U.S., Sukumaran, R.K., Devi, G.L., Rajasree, K.P., Singhania, R.R., Pandey, A., 2010. Bio-ethanol from water hyacinth biomass: an evaluation of enzymatic saccharification strategy. Bioresour. Technol. 101, 925–930.
Back, S.K., Yoo, H.M., Jang, H.N., 2017. Effects of alkali metals and chlorine on corrosion of super heater tube in biomass circulating fluidized bed boiler. Appl. Chem. Eng. 28, 29–34.
Blaesing, M., Mueller, M., 2013. Release of alkali metal, sulphur, and chlorine species from high temperature gasification of high- and low-rank coals. Fuel Process. Technol. 106, 289–294.
Chen, J.C., Liu, J.Y., He, Y., Huang, L.M., Sun, S.Y., Sun, J., Chang, K.L., Kuo, J.H., Huang, S.S., Ning, X.A., 2017. Investigation of co-combustion characteristics of sewage sludge and coffee grounds mixtures using thermogravimetric analysis coupled to artificial neural networks modeling. Bioresour. Technol. 225, 234–245.
Chen, C., Lu, Z., Ma, X., Long, J., Peng, Y., Hu, L., Lu, Q., 2013. Oxy-fuel combustion characteristics and kinetics of microalgae Chlorella vulgaris by thermogravimetric analysis. Bioresour. Technol. 144, 563–571.
Cieslik, B.M., Snik, J.N., Konieczka, P., 2015. Review of sewage sludge management: standards, regulations and analytical methods. J. Clean. Prod. 90, 1–15.
Fang, S., Yu, Z., Lin, Y., Lin, Y., Fan, Y., Liao, Y., Ma, X., 2016. Effects of additives on the co-pyrolysis of municipal solid waste and paper sludge by using thermogravimetric analysis. Bioresour. Technol. 209, 265–272.
Gangulya, A., Chatterjee, P.K., Dey, A., 2012. Studies on ethanol production from water hyacinth – a review. Renew. Sustain. Energy Rev. 16, 966–972.
Guerrero-Coronilla, I., Morales-Barrera, L., Cristiani-Urbina, E., 2015. Kinetic, isotherm and thermodynamic studies of amaranth dye biosorption from aqueous solution onto water hyacinth leaves. J. Environ. Manage. 152, 99–108.
Han, L., Bollas, G.M., 2016. Dynamic optimization of fixed bed chemical-looping combustion processes. Energy 2, 1107–1119.
Huang, Y.F., Kuan, W.H., Chiueh, P.T., Lo, S.L., 2011. Pyrolysis of biomass by thermal analysis-mass spectrometry (TA-MS). Bioresour. Technol. 102, 3527–3534.
Huang, L.M., Liu, J.Y., He, Y., Sun, S.Y., Chen, J.C., Sun, J., Chang, K.L., Kuo, J.H., Ning, X. A., 2016. Thermodynamics and kinetics parameters of co-combustion between sewage sludge and water hyacinth in CO2/O2atmosphere as biomass to solid
biofuel. Bioresour. Technol. 218, 631–642.
Huang, L.M., Xie, C.D., Liu, J.Y., Zhang, X.C., Chang, K.L., Kuo, J.H., Sun, J., Xie, W.M., Zheng, L., Sun, S.Y., Buyukada, M., Evrendilek, F., 2018. Influence of catalysts on co-combustion of sewage sludge and water hyacinth blends as determined by TG-MS analysis. Bioresour. Technol. 247, 217–225.
Jiang, L.B., Yuan, X.Z., Li, H., Chen, X.H., Xiao, Z.H., Liang, J., Leng, L.J., Guo, Z., Zeng, G. M., 2016. Co-pelletization of sewage sludge and biomass: thermogravimetric analysis and ash deposits. Fuel Process. Technol. 145, 109–115.
Jiménez, R., García, X., Cellier, C., Ruiz, P., Gordon, A.L., 2006a. Soot combustion with K/MgO as catalyst. Appl. Catal. A Gen. 297, 125–134.
Jiménez, R., García, X., Cellier, C., Ruiz, P., Gordon, A.L., 2006b. Soot combustion with K/MgO as catalyst: II. Effect of K-precursor. Appl. Catal. A Gen. 314, 81–88.
Jonathan, J.D., Pendlebury, I., Richard, J.A., Shehr, B., Kathleen, J., Ennifer, J.M., Schneider, S.U., 2008. Respiratory control in neonatal rats exposed to prenatal cigarette smoke. Am. J. Respir. Crit. Care Med. 177, 1255–1261.
Kan, T., Strezov, V., Evans, T., Kan, T., Grierson, S., Nys, R.D., Strezov, V., 2014b. Comparative assessment of the thermochemical conversion of freshwater and marine micro- and macroalgae. Energy Fuels 28, 104–114.
Kan, T., Strezov, V., Evans, T., 2014a. Catalytic pyrolysis of coffee grounds using NiCu-impregnated catalysts. Energy Fuels 28, 228–235.
Kim, Y.S., Kim, Y.S., Kim, S.H., 2010. Investigation of thermodynamic parameters in the thermal decomposition of plastic waste-waste lube oil compounds. Environ. Sci. Technol. 44, 5313–5317.
Li, X.J., Wang, Y.Z., Yue, M.Z., 2012. Experimental study on the corrosion characteristics during co-firing of cornstalk and lean coal. Renew. Sustain. Energy 347–353, 2622–2625.
Li, W.Q., Wang, L.Y., Qiao, Y., 2015. Effect of atmosphere on the release behavior of alkali and alkaline earth metals during coal oxy-fuel combustion. Fuel 139, 164–170.
Liao, Y.F., Wu, S.M., Chen, T., Cao, Y.W., Ma, X.Q., 2015. The alkali metal characteristic during biomass combustion with additives. Energy Procedia 75, 124–129.
Liu, J.Y., Huang, S.J., Sun, S.Y., Ning, X.A., Li, X.M., Chen, T., Luo, G.Q., Xie, W.M., Wang, Y.J., Yang, Z.Y., 2015. Effects of sulfur on lead partitioning during sludge combustion based on experiments and thermodynamic calculations. Waste Manage. 38, 336–348.
Liu, J.Y., Huang, L.M., Ning, X.A., Kuo, J., Sun, J., Chang, K.L., Wang, Y.J., Chen, J.C., Chen, J.N., 2016a. Analysis of co-combustion characteristics of sewage sludge and water hyacinth. Huanjing Kexue Xuebao. 36, 2955–2967 (In Chinese).
Liu, J.Y., Zeng, J.J., Sun, S.Y., Huang, S.S., Kuo, J.H., Chen, N.W., 2016b. Combined effects of FeCl3and CaO conditioning on SO2, HCl and heavy metals emissions
during the DDSS incineration. Chem. Eng. J. 299, 449–458.
Liu, J.Y., Xie, W.H., Zhuo, Z.X., Musa, B., Fatih, E., 2017b. Thermochemical behaviors of textile dying sludge, paper mill sludge, and their blends during (co-)combustion. Thermochim Acta 655, 101–105.
Liu, J.Y., Huang, L.M., Musa, B., Fatih, E., 2017a. Response surface optimization, modeling and uncertainty analysis of mass loss response of co-combustion of sewage sludge and water hyacinth. Appl. Therm. Eng. 125, 328–335.
Luo, G.E., Strong, P.J., Wang, H.L., Ni, W.Z., Shi, W.Y., 2011. Kinetics of the pyrolytic and hydrothermal decomposition of water hyacinth. Bioresour. Technol. 102, 6990–6994.
Miranda, T., Román, S., Montero, I., Nogales-Delgado, S., Arranz, J.I., Rojas, C.V., González, J.F., 2012. Study of the emissions and kinetic parameters during combustion of grape pomace: dilution as an effective way to reduce pollution. Fuel Process. Technol. 103, 160–165.
Miró, E., Ravelli, F., Ulla, M., Cornaglia, L., Querini, C.A., 1999. Catalytic combustion of diesel soot on Co, K supported catalysts. Catal. Today 53, 631–638.
Mishima, D., Kuniki, M., Sei, K., Soda, S., Ike, M., Fujita, M., 2008. Ethanol production from candidate energy crops: water hyacinth (Eichhornia crassipes) and water lettuce (Pistia stratiotes L.). Bioresour. Technol. 99, 2590–2600.
Mishra, G., Bhaskar, T., 2014. Non isothermal model free kinetics for pyrolysis of rice straw. Bioresour. Technol. 169, 614–621.
Muhammad, S.B., Peter, A.J., Flemming, F., Stig, W., Kim, D.J., Johan, W., Søren, T.P., 2012. Ash transformation and deposit build-up during biomass suspension and grate firing: full-scale experimental studies. Fuel Process. Technol. 97, 93– 106.
Otero, M., Díez, C., Calvo, L.F., García, A.I., Morán, A., 2002. Analysis of the co-combustion of sewage sludge and coal by TG-MS. Biomass Bioenerg. 22, 319–329.
Otero, M., Sánchez, M.E., Gómez, X., 2011. Co-firing of coal and manure biomass: a TG-MS approach. Bioresour. Technol. 102, 8304–8309.
Putun, E., Ates, F., Putun, A.E., 2008. Catalytic pyrolysis of biomass in inert and steam atmospheres. Fuel 87, 815–824.
Qi, J.H., Han, K.H., Wang, Q., 2017. Carbonization of biomass: effect of additives on alkali metals residue, SO2and NO emission of chars during combustion. Energy
130, 560–569.
Querini, C.A., Cornaglia, L.M., Ulla, M.A., Miró, E.E., 1999. Catalytic combustion of diesel soot on Co, K/MgO catalysts. Effect of the potassium loading on activity and stability. Appl. Catal. B Environ. 20, 165–177.
Sebestyén, Z., Barta-Rajnai, E., Bozi, J., Blazsó, M., Jakab, E., Miskolczi, N., Sója, J., Czégény, Z.S., 2017. Thermo-catalytic pyrolysis of biomass and plastic mixtures using HZSM-5. Appl. Energy 207, 114–122.
Serra, V., Saracco, G., Badini, C., Specchia, V., 1997. Combustion of carbonaceous materials by Cu–K–V based catalysts. 2. Reaction mechanism. Appl. Catal. B Environ. 11, 329–346.
Shao, J.G., Yan, R., Chen, H.P., Yang, H.P., Lee, D.H., 2010. Catalytic effect of metal oxides on pyrolysis of sewage sludge. Fuel Process. Technol. 91, 1113–1118.
Shen, B.X., Qin, L., 2006. Study on MSW catalytic combustion by TGA. Energ. Conver. Manage. 47, 1429–1437.
Syed-Hassana, S.S.A., Wang, Y., Hu, S., Su, S., Xiang, J., 2017. Thermochemical processing of sewage sludge to energy and fuel: fundamentals, challenges and considerations. Renew. Sustain. Energy Rev. 80, 888–913.
Utton, D., Elleher, B., Ross, J.R.H., 2001. Review of literature on catalysts for biomass gasification. Fuel Process. Technol. 73, 155–173.
Villamagna, A.M., Murphy, B.R., 2010. Ecological and socio-economic impacts of invasive water hyacinth (Eichhornia crassipes): a review. Freshwater Biol. 55, 282–298.
Xiao, Q.M., 2006. Effect of additives on behavior of alkali metals during straw combustion. J. Zhejiang Univ. (Eng. Sci.) 40 (4), 599–604.
Xie, W.H., Huang, J.L., Liu, J.Y., Zhao, Y.J., Xie, W.M., Kuo, J.H., He, Y., Sun, J., Heng, L., Chang, K.L., Sun, S.Y., Musa, B., Fatih, E., 2018a. Assessing thermal behaviors and kinetics of (co-)combustion of textile dyeing sludge and sugarcane bagasse. Appl. Therm. Eng. 131, 874–883.
Xie, C.D., Liu, J.Y., Zhang, X.C., Xie, W.M., Sun, J., Chang, K.L., Kuo, J.H., Xie, W.H., Liu, C., Sun, S.Y., Musa, B., Fatih, E., 2018b. Co-combustion thermal conversion characteristics of textile dyeing sludge and pomelo peel using TGA and artificial neural networks. Appl. Energ. 212, 786–795.
Xu, Y., Chen, B., 2013. Investigation of thermodynamic parameters in the pyrolysis conversion of biomass and manure to biochars using thermogravimetric analysis. Bioresour. Technol. 146, 485–493.
Yan, Y.F., Feng, S., Zhang, L., 2018. Experimental research on catalytic combustion characteristics of inferior coal and sludge mixture. J. Energy Resour. Technol. Trans. ASME 140, 3.
Yang, H., Yan, R., Chen, H., Zheng, C., Lee, D.H., Liang, D.T., 2006. Influence of mineral matter on pyrolysis of palm oil wastes. Combust. Flame 146, 605–611.
Yu, Z.S., Ma, X.Q., Liu, A., 2009. Thermogravimetric analysis of rice and wheat straw catalytic combustion in air-and oxygen-enriched atmospheres[J]. Energy Convers. Manage. 50, 561–566.
Zhuo, Z., Liu, J., Sun, S., Sun, J., Kuo, J., Chang, K., Fu, J., Wang, Y., 2017. Thermogravimetric characteristics of textile dyeing sludge, coal and their blend in N2/O2and CO2/O2atmospheres. Appl. Therm. Eng. 111, 87–94.
Zimmels, Y., Kirzhner, F., Kadmon, A., 2009. Effect of circulation and aeration on wastewater treatment by floating aquatic plants. Separat. Purif. Technol. 66, 570–577.
