The mixture of sewage sludge and biomass waste as solid biofuels:
Process characteristic and environmental implication
Jiacong Chen
a,1, Yao He
a,1, Jingyong Liu
a,*, Chao Liu
a, Wuming Xie
a, Jiahong Kuo
a,
Xiaochun Zhang
a, Shoupeng Li
a, Jialin Liang
a, Shuiyu Sun
a, Musa Buyukada
b,
Fatih Evrendilek
c,daGuangzhou Key Laboratory of Environmental Catalysis and Pollution Control, Guangdong Key Laboratory of Environmental Catalysis and Health Risk Control, School of Environmental Science and Engineering, Institute of Environmental Health and Pollution Control, Guangdong University of Technology, Guangzhou, 510006, China
bDepartment of Chemical Engineering, Bolu Abant Izzet Baysal University, Bolu, 14052, Turkey cDepartment of Environmental Engineering, Bolu Abant Izzet Baysal University, Bolu, 14052, Turkey dDepartment of Environmental Engineering, Ardahan University, Ardahan, 75002, Turkey
a r t i c l e i n f o
Article history: Received 31 May 2018 Received in revised form 4 December 2018 Accepted 31 January 2019 Available online 4 February 2019
Keywords: Oxy-fuel (O2/CO2) Solid biofuels Synergistic effect TGA-MS XRF
a b s t r a c t
Oxy-fuel and air atmospheres were used to (co-)combust sewage sludge (SS) and biomass waste, coffee ground residues (CG) via thermogravimetric analysis (TGA). The combustion behavior of SS did not differ significantly in both atmospheres. The replacement of 79%N2by 79%CO2significantly influenced the char
combustion of CG. Synergistic effect of the blends in the oxy-fuel was weaker than air condition. Acti-vation energy of the co-combustion in the second stage was lower in the oxy-fuel than air atmosphere. The gaseous emissions during blend fuel combustion were investigated by online mass spectrometry (MS). Results show that the gas emissions of different fuels show different stage characteristics. CH3, H2O,
C2H2and NO emissions peaked from the volatiles combustion of CG, while the co-combustion led to SO2
increment. Besides, the composition of the solid residues was examined by X-rayfluorescence spec-trometer (XRF), and their impact on environment was evaluated. The compositions of the solid residues pointed to the ability of SS to lower the fouling and slagging risks of CG. This investigation aimed to afford a fully understanding for the co-combustion progress of SS and CG under air and oxy-fuel envi-ronments and its implication for environment.
© 2019 Elsevier Ltd. All rights reserved.
1. Introduction
Globally, the rapid growth rates of population and consumption have generated such large quantities of sewage sludge (SS) that its treatment has become an urgent matter to be dealt with [1]. Its high salt, nutrient, heavy metal, pathogen and organic pollutant contents pose a significant threat to the well-being and health of
humans and ecosystems unless disposed properly [2,3].
Mono-combustion has been used as one of the economically efficient
and environmentally friendly ways to reduce the SS waste stream, and its hazards [4,5]. However, its high ash content fails to lead to a complete combustion, thus forming a variety of secondary organic
pollutants such as polycyclic aromatic hydrocarbons, and dioxins. The common industrial practice to avoid this issue has been to co-combust SS with a feedstock with higher energy density such as coal or biomass waste [6e8].
One such feedstock of biomass waste is coffee ground residues (CG) generated from the treatment of raw coffee powder at the ratio of CG to produced instant coffee of 0.91 kg [9,10]. Its major constituents are sugars polymerized into (hemi)cellulose structures (45.3% w/w), protein (13.6% w/w), ashes, and minerals such as K, P, Mg, and Ca [11]. Chen et al. [12] found that thefixed carbon com-bustion of CG was easier than that of pine or anthracite. The low ash content, high volatile matters, and high heating value of CG make it as a promising biofuel to avoid the thermal cycle problem [13]. The main challenges that the co-combustion technology faces closely relate to ignition and burnout temperatures,flue gas emissions, and ash-related issues (e.g., slagging and fouling). These issues can be
* Corresponding author.
E-mail address:Liujy@gdut.edu.cn(J. Liu). 1 These authors contributed to the work equally.
Contents lists available atScienceDirect
Renewable Energy
j o u r n a l h o m e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / r e n e n e
https://doi.org/10.1016/j.renene.2019.01.119
solved through the design and optimization of the co-combustion systems in terms of better performance and lower environmental impacts [14,15]. Gong et al. [16] reported that the co-utilization of microalgae residue and oil sludge reduced NOxand SO2emissions.
Sung et al. [17] demonstrated that there existed a synergistic effect
on the NOxreduction during co-combustion of woody biomass and
coal. Many researchers have studied the ash composition and the tendency of slagging and fouling [18e20]. However, there exists no study about analyzing main gaseous emissions and ash deposits during combustion of SS and CG.
The oxy-fuel co-combustion technology in the N-free atmo-sphere has emerged as a main carbon capture and storage strategy to reduce greenhouse gas emissions and to produce highly concentrated CO2steam in theflue gas [21e23]. The application of
this technology for waste thermal disposal gains an advantage over that of the traditional one in terms of carbon emission. Scientists have found that the substitution of CO2for N2has exerted a
vari-ation in characteristic of the combustion behavior and composition of the solid residue. For example, Buhre et al. [24] reported that the flame ignition of pulverized coal was delayed in the O2/CO2
at-mosphere relative to the O2/N2atmosphere. Kastanaki and
Vam-vuka [25] reported a greater synergistic effect of biomass on the co-combustion of lignite than hard coal-derived chars. Shaddix and Molina [26] stated a significant increase in the char particle tem-peratures of (sub-)bituminous coals in the O2/N2atmosphere upon
the replacement of N2with CO2. However, sparse reports can be
found with respect to the topic of interactions between co-processing SS and CG by oxy-fuel combustion technology.
The present study aimed at assessing the co-combustion po-tential of CG with SS for heat or power production in terms of its combustion performances and environmental effects. The objec-tives of this experimental study were to (1) quantify the (co-) combustion behaviors of SS, CG, and their four blends in the oxy-fuel versus air atmosphere; (2) determine their performances based on combustion index, interaction deviation, and apparent activation energy; (3) identify gas evolutions and their environ-mental implications; and (4) evaluate the potential risk of ash slagging based on the empirical indices. These conclusions will provide a theoretical reference for the co-firing SS with CG in en-gineering application.
2. Materials and methods 2.1. Sample preparation
SS samples were collected from an urban wastewater treatment plant located in Guangzhou, China. CG samples were obtained from an instant coffee processing factory in China. Both sample types were air-dried, ground in a grinder and then sieved to obtain par-ticles less than 74
m
m in diameter. All the samples were dried in an oven at 105C for 24 h and stored in a desiccator. The blends of SS and CG were based on the following four ratios of 9:1, 8:2, 7:3 and 6:4 on a dry weight basis and coded as SC91, SC82, SC73, and SC64, respectively. The proximate and ultimate analyses of SS and CG were obtained from Chen et al. [27] (SeeTable 1).2.2. Thermogravimetric analysis
The thermal decompositions of SS, CG, and their four blends were carried out using a thermogravimetric (TG) analyzer (Mettler Toledo TGA/DSC 1). 10 mg samples were placed in an alumina crucible, placed into a furnace and heated from a room temperature to 1000C at the three heating rates of 10, 20 and 40C/min. The two (co-)combustion atmospheres considered in the present study were the air (O2/N2¼ 21/79) and oxy-fuel (O2/CO2¼ 21/79) ones. A
flow rate of 100 mL/min and a holding time of 5 min were adopted to ensure an identical temperature distribution and thermal equi-librium during the experiments. Each experiment was replicated three times in order to reduce random experimental errors and to compute standard deviations.
2.3. Kinetic analysis
Kinetic behavior of the thermal decomposition of the different constituents was expressed using the following rate of conversion:
d
a
dt¼ kðTÞf ð
a
Þ (1)where t (min) was time; T was the absolute temperature; f(
a
)represented the function of reaction mechanism; and
a
was thedegree of conversion that can be calculated thus:
a
¼W0 WiW0 Wf
(2)
where W0, Wiand Wfreferred to initial, instantaneous andfinal
masses, respectively. k(T) was a temperature-dependent rate con-stant expressed by the Arrhenius law thus:
kðTÞ ¼ A exp Ea RT (3)
where A was the pre-exponential factor; Ea(J/mol) was activation
energy of the reaction; and R was the universal gas constant, 8.314 J/mol/K. Substituting Eq.(3)in Eq.(1)leads to the following:
d
a
dt¼ AeðEa
RTÞf ð
a
Þ (4)Taking into account that the temperature was a function of time and increases with the constant heating rate
b
(K/s),b
can be re-written thus:b
¼dT dt¼ dT da
da
dt (5)Eqs.(4) and (5)can be combined and rearranged as follows:
gð
a
Þ ¼ ða 0 da
fða
Þ¼ Ab
ðT 0 exp Ea RT dT (6)where g(
a
) was the integrated form of conversion-dependentfunction f(
a
), andR0TexpEa RT
dT on the right-hand side of Eq.
(6) was called the temperature integral, which did not have an
analytical solution.
If it is assumed that x¼ Ea/RT, thenR0Texp
Ea RT dT is reduced toEa R R∞ x e x x2dx ¼ Ea
RPðxÞ, where P(x) was an infinite function of x.
Therefore, substituting the values of temperature integral (P(x)) in Eq.(6)yields: gð
a
Þ ¼ ða 0 da
fða
Þ¼ AEab
RPðxÞ (7)As the value of Ea/R was constant, the evaluation ofR0afdðaaÞwas
dependent on the function P(x). In this study, the iso-conversional
FlynneWalleOzawa (FWO) were used to approximate apparent activation energy (Ea). The KAS method [28,29] based on the
Coats-Redfern [30] approximation was expressed thus:
ln
b
T2 ¼ ln AR Eagða
Þ Ea RT (8)where the plot of ln(
b
/T2) against 1/T resulted in a line whose slope was used to determine Ea.Doyle [31] evaluated P(x) and suggested its value to be
logðPðxÞÞ ¼ 2:315 0:4567x over the range of 20 x 60. Using the Doyle (1962) approximation and the logarithmic form of Eq.(6),
the following linear equation of the FWO method [32,33] was
obtained: logð
b
Þ ¼ log AEa gða
ÞR 2:315 0:4567 Ea RT (9)Upon changing the base 10 (log) to natural logarithm (ln), Eq.(9)
can be rewritten thus:
lnð
b
Þ ¼ ln AEa gða
ÞR 5:331 1:052 Ea RT (10)where Eafor different conversion values was calculated from the
slope of ln
b
versus 1/T.2.4. Thermogravimetric-mass spectrometric analysis
Thermogravimetric-mass spectrometric (TG-MS) analysis was performed using a Thermo Mass Photo (Rigaky-Tokyo, Japan). Three-mg samples were used, while the temperature was raised from a room temperature to 1000C at a heating rate of 20C/min in the O2/He (21/79) atmosphere. The primary focus of this study
was on such small molecules as CH3, H2O, C2H2, NO, CO2, and SO2,
with the mass/charge (m/z) values of 15, 18, 26, 30, 44, and 64, respectively.
2.5. Ash composition analysis
The major elements of the SS and CG ashes were analyzed using
a ZSX Primus II sequential X-rayfluorescence (XRF) spectrometer
(Rigaku, Japan). To obtain ash samples, fuels were combusted at 575± 25C in an electric furnace according to ASTM E1755. A low
ash temperature instead of its typical temperature of 900C was used to ensure that no transformation of alkali metals, in particular, K occurred [34].
3. Results and discussion 3.1. (Co)-combustion behaviors
The (D)TG curves for the (co-)combustions of SS, CG, and their four blends at a heating rate of 20C/min in the air versus oxy-fuel atmosphere are presented inFig. 1. For an in-depth analysis, the (D) TG curves of pure SS , CG and SC64 in the O2/CO2(21/79)
atmo-sphere were also added toFig. 1according to our previous study [27]. The higher volatiles-to-fixed carbon (V/FC) ratio of SS (21.0) than CG (3.4) pointed to the dominance of the volatiles combustion of SS. The two main stages of the SS combustion were determined as follows: (1) the devolatilization and combustion of organic compounds, and thus, the char formation in the range of 185e410C; and (2) the combustion of char from 410 to 635C [35].
No significant change in the combustion parameters of SS was
detected when N2was replaced by CO2(Table 2). In other words,
CO2exerted a minor effect on SS reaction during the entire
com-bustion process. The identical TG curves until 290C of the CG combustion in both atmospheres suggested that CO2in the oxy-fuel
atmosphere behaved as an inert atmosphere in this region. The two main peaks of the DTG curve in the O2/N2atmosphere at 300 and
430C for CG corresponded to the devolatilization and
char-combustion stages, respectively, the latter of which was distin-guished by a shoulder on the right of the second peak.
Unlike in the air atmosphere, the char combustion stage was continuous with a wide peak in the oxy-fuel atmosphere most
probably due to the commencement of the char-CO2reaction. The
char-combustion behaviors of SS in the second stage were similar in both atmospheres. However, the peak values for CG were lower in the oxy-fuel than air atmosphere due to the poorer properties of CO2than N2[36]. This case revealed that the replacement of N2by
CO2changed the weight loss mechanisms, but the effect of the fuel
type played a greater role than did the atmosphere type [37]. The higher maximum rate of weight loss for CG (37.6%/min) than SS (3.8%/min) (Fig. 1,Table 2) pointed to the higher CG than SS reactivity in the air atmosphere. Thefinal residual masses of SS and CG were estimated at 0.8% and 49.6%, respectively, in the air at-mosphere. Therefore, CG appeared to be a promising additive that enhanced the SS combustion performance due to their different compositions and structures. The switch from the air to the oxy-fuel atmosphere for CG decreased the ignition temperature (Ti) by
5.0C but increased the burnout temperature (Tb) by 27.5C since Ti
was controlled by the fuel ejection of volatiles (Table 2). However, Tbshowed a strong relationship with the oxidation ingredients. The
maximum weight loss rate of CG was 22.2%/min in the oxy-fuel atmosphere, a 15.4%/min decrease relative to the DTGmaxvalue in
the air atmosphere. Thus, replacing N2with CO2at the same oxygen
levels adversely influenced the CG combustion.
All the co-combustion profiles at 20C/min varied in between
the mono-combustion ones in both atmospheres (Fig. 1). The
Table 1
Proximate and ultimate analyses and heating values of SS and CG on an air-dried basis [27].
Sample Ultimate analyses (wt %) Proximate analyses (wt%) HHVa
C H Of N S Mb Vc Ad FCe (MJ/kg)
SS 24.13 3.94 23.31 4.50 0.74 5.50 48.80 43.38 2.32 10.67
CG 57.17 7.10 32.80 2.31 0.06 2.69 74.82 0.56 21.93 24.81
aHHV, higher heating value on an air-dried basis. b M, moisture.
c V, volatile matters. d A, ash.
e FC,fixed carbon, calculated by FC ¼ 100-M-V-A. f O, calculated by O¼ 100-A-C-H-N-S.
increased CG fraction grew the weight loss rate of the main peak, and the occurrence of the total burnout at lower temperatures
pointed to the enhanced co-combustion efficiency. A less weight
loss rate, and a slightly higher burnout temperature in the oxy-fuel atmosphere indicated a slightly delayed co-combustion. The higher
Tmof the blends than the individual fuels in both atmospheres
(Table 2) suggested that their decomposition and emission were not linearly related to the CG ratio, regardless of the atmosphere type. Tmof SS and CG increased by 6.7 and 4.7C in the oxy-fuel
atmosphere relative to the air atmosphere, respectively. The less reduction in Tmobserved for each blend in the O2/CO2than O2/N2
atmosphere appeared to stem from the different heating values of
Fig. 1. TG (a) and DTG (b) curves for combustion of SS, CG and their mixtures in O2/N2and O2/CO2atmospheres (TG and DTG curves of SS (O2/CO2¼ 21/79) , CG (O2/CO2¼ 21/79) and SC64 (O2/CO2¼ 21/79) cited in Ref. [27]).
the fuels, and the different specific heats of CO2and N2.
Comprehensive combustion characteristic index (S) was used to evaluate the (co-)combustion performances of the fuels [38]:
S¼ðdW=dtÞmaxðdW=dtÞmean
T2iTf (11)
where (dW/dt)maxrefers to the maximum mass loss rate; and (dW/
dt)meanis the average mass loss rate. The higher the S value is, the
more vigorously the samples are burned, and the more rapidly the char is burned out. The CG combustion performed best in both atmospheres (Table 2). The S values of the individual fuels were lower in the O2/CO2than O2/N2atmosphere. Especially, the change
from O2/N2to O2/CO2decreased the S value (from 19.64 107to
11.33 107%2/min2/C3), thus indicating the worse combustion
performances of the individual CG fuels in the O2/CO2atmosphere.
The elevated CG ratio of the blends resulted in a nonlinear growth of the S value due to a synergistic effect. The S values in both at-mospheres declined with the blends less than the individual fuels. The S value of SC64 remained the same in both atmospheres owing to the synergistic interaction between SS and CG.
3.2. Synergistic effects of the blends
In order to explore the synergistic behaviors of the blends, their theoretical DTG curves were plotted using the sum of the weight loss rates of SS and CG fractions under the same conditions (Fig. 2). Under the assumption of no interaction between the blends and the other materials, the predicted DTG curve of the blends was ob-tained using the following equation [39]:
DTGmixture¼ xSSDTGSSþ xCGDTGCG (12)
where DTGSSand DTGCGare the observed weight loss rates of SS
and CG, respectively. xSSand xCGare the blend fractions of SS and
CG, respectively.
From thefirst peak at 308C in the air atmosphere, the slower
experimental reaction rate, and the smaller experimental weight loss rate than the calculated ones were indicative of the interaction between SS and CG. The synergistic effects of the blends were weaker in the oxy-fuel atmosphere. A pronounced shift of the second peak to higher temperatures was observed. The two peaks
expected at 430C and 465C merged to produce one single peak at 490C, thus revealing the synergistic behaviors of the blends.
To further explore the synergistic effect during the
co-combustion,△W(△W ¼ TGexp-TGcal) was introduced as the
dif-ference in the weight losses between the experimental and calcu-lated data whose results are depicted inFig. 2c and d. The negative values showed a synergistic effect towards the char formation, while the positive values implied a synergistic effect towards the volatiles formation. For the air atmosphere in the range of
35e250C, the observed deviation of 0.15% showed a minimum
interaction between SS and CG with a slightly higher release of volatiles than expected. Between 250 and 585C, the larger posi-tive deviations of the weight loss rates were observed with the two peaks at 308 and 490C. Thefirst synergistic peak with a maximum deviation of 4.2% coincided with the peak DTG of the co-combustion (Table 2). This pointed to a significant synergistic mechanism that further promoted the cracking of char in SS and CG, thus forming more volatiles. The second peak with a maximum deviation of 6.2% occurred at the critical temperature where CG
degraded together with the second degradation stage of SS.△W
peaks rose with the increased CG fraction (Fig. 2b). CG had more volatiles than did SS and decomposed mainly in the range of 180e550C, as discussed in Section 3.1.1. The increased CG ratio
may have trapped a large quantity of volatiles in the voids of the particles or on the particle surfaces of SS and CG, thus preventing the release of volatiles. Therefore, TGexp was higher than TGcal.
△Wmaxpeaked (6.2%) for SC64 where the two materials interacted
most. The decompositions of SS, CG and their blends speeded up by the increased temperature were completed by 610C. The consis-tently observed deviation of 0.2% during the entire conversion process pointed to the synergistic effects between the reactive volatiles and/or between the solids that promoted the reactions of degradation or suppressed the reactions of re-condensation and char formation, as was also reported by Gunasee et al. [40]. To capture the synergistic effect, the root mean square (RMS) values of
the deviation between the experimental and calculated (△W)
values were also used by Wu et al. [41]. The RMS values of△W
were estimated at 0.65, 1.00, 1.37 and 1.70 for SC91, SC82, SC73, and SC64, respectively, and were ten times those reported in related literature [42,43]. This case suggested a remarkable synergy during the co-combustion which in turn yielded smaller solid residues.
In the oxy-fuel atmosphere, no difference in the weight loss was found (△W < 0.01%) at below 250C. In the range of 280e585C,
the two peaks observed in Fig. 2d were related to the two
consecutive positive synergistic effects. The first peak at 334C
corresponded to the main overlapping decomposition zone of SS and CG. The reactions between the co-combustion intermediates due to the devolatilization formed more thermally stable com-pounds than did the individual fuels. The second peak with a
maximum deviation of 4.1% at 510C coincided with the second
degradation stage of SS and CG (Fig. 1). At above 610C, negligible deviations (△W < 0.06%) were depicted (Fig. 2d). The RMS values of the△W were estimated at 0.34, 0.68, 0.94, and 1.17 for SC91, SC82, SC73, and SC64 in the O2/CO2atmosphere, respectively. The
lower RMS values of the△W in the O2/CO2than O2/N2atmosphere
indicated weaker synergistic effects in the O2/CO2atmosphere. The
reason for this may be the fact that CO2diminished the
concen-trations of the reactive radicals due to its tendency to inhibit the radical formation, and the enhancement of the recombination re-actions such as HþO2þM ¼ HO2þM [44].
3.3. Activation energy
In order to gain insights into the synergistic effects, the Eavalues
estimated with the different
a
values in both atmospheres areTable 2
(Co-)combustion characteristic parameters of SS and CG in O2/N2 and O2/CO2 atmospheres. Experiment Tia Tbb Tmc DTGmaxd Mfe Sf(107) O2/N2 SS 237.3 645.2 290.8 3.85 49.67 1.10 SC91 248.6 625.4 309.2 5.08 44.97 1.50 SC82 258.4 609.2 312.0 6.76 40.00 2.07 SC73 268.7 593.6 311.2 8.95 35.11 2.81 SC64 273.9 580.0 308.0 11.61 30.33 3.85 CG 285.9 484.4 302.2 37.64 0.82 19.64 O2/CO2 SS [27] 240.0 646.8 297.5 3.84 49.65 1.08 SC91 244.6 630.0 307.8 4.93 44.70 1.50 SC82 258.7 610.4 309.4 6.77 39.95 2.06 SC73 265.2 596.2 309.0 8.59 35.00 2.76 SC64 [27] 272.7 584.7 308.5 10.71 30.14 3.58 CG [27] 280.9 511.9 306.9 22.20 0.95 11.33 aT i, ignition temperature,C. b T b, burnout temperature,C. c T m, peak temperature,C. d DTG
max, maximum mass loss rate, %/min. e M
f, residual mass, %.
presented inTables 3 and 4. The three heating rates of 10, 20 and 40C/min were used to evaluate the relationships between Eaand
a
using the KAS and FWO methods. The Eaestimates had relatively
high coefficients of determination (R2) between 0.9314 and 1
(Tables 3 and 4) and did not significantly differ between the KAS and FWO methods.
In both atmospheres, the increased blend ratio did not mono-tonically increase Ea. SC82 led to the lowest Eavalue of 250 kJ/mol
(KAS) and 247.99 kJ/mol (FWO) in the air atmosphere and 249.49 kJ/mol (KAS) and 247.53 kJ/mol (FWO) in the oxy-fuel at-mosphere followed by SC91, SC73, and SC64, respectively. This result suggested an accelerative impact of 20% CG on the SS devo-latilization. The lower Eavalue observed in the oxy-fuel than air
atmosphere showed that replacing N2by CO2at the same oxygen
level changed the degradation pathway to the reaction of less en-ergy requirement [37].
A noticeable increasing trend in Eafor thefirst degradation stage
occurred with the increased
a
. The Eavalue of the blends peaked ata
¼ 0.3 in both atmospheres at 310C. This accounted for thechange in the DTG curve of the blend at the same temperature from the energy perspective. The Eavalues in turn decreased persistently.
The co-combustion reactions needed less energy with the decreased Ea. The lower Eavalues in the air than oxy-fuel
atmo-sphere with 40% CG when
a
< 0.4 indicated that the decomposition of the blend volatiles in the air atmosphere may be enhanced by the significant radical interactions. Considering the respective Eavaluesdetermined at each
a
, the second combustion stage required less energy than did thefirst one. Similar results were reported for theco-combustion of bamboo [45]. Some contradictory results were
also reported about Eavalues of the blends of distillation residues
and lignite in related literature [46,47]. The differences between
our results and thosefindings may be caused by the different types of the kinetic models and biomass materials used. The Eavalues of
the blends for the second stage were lower in the O2/CO2than O2/
N2atmosphere. These lower values were confirmed by the positive
synergy effects described earlier in Section3.3. Although the reason for this cannot be inferred from the current experiments and ana-lyses, one possible explanation can be that the additional char-CO2
reaction enhanced the fuel reactivity in the O2/CO2 atmosphere.
However, the competing CO2and O2chemisorptions lowered the
combustion reactivity of the O2/CO2 atmosphere. These two
competitive effects on the combustion reactivity were also reported for the oxy-fuel atmosphere in related literature [48]. Further studies still remain to be conducted to clarify under what condi-tions one prevails another over time [37,49e51].
3.4. Gas products
Fig. 3shows the evolutions of the six common fragments from the co-combustion in the O2/He atmosphere. The organic matter of
SS was mainly composed of fat, protein, carbohydrate, and cellulose [52]. Although SS and CG were heterogeneous, their gas evolutions
can confirm their thermal degradation stages [53]. When the
relative abundances of some selected molecular ions (i.e. CH3, H2O,
C2H2, NO, CO2, and SO2) common to SS, CG, and SC64 were
compared, H2O and CO2were found to be released more during the
co-combustion. 3.4.1. H2O
H2O releases were characterized by the molecular ion fragment
in 18 amu and occurred at several stages. The three MS peaks of
H2O observed during the SS combustion were in close agreement
Fig. 2. Comparison of calculated versus experimental DTG curves for blends in O2/N2(a) and O2/CO2(c) atmospheres, and variations of△W in O2/N2(b) and O2/CO2(d) atmospheres.
with its DTG profile (Fig. 1b). Thefirst peak at below 200C cor-responded to the dehydration. In the following temperature range,
H2O release was related to the degradation of various
oxygen-containing groups such as hydroxyl groups, and glycosylic units [35,40,54]. The highest intensity of the second peak of H2O
dis-played by CG was consistent with its high hydrogen content (Table 1). From the blends, H2O was evolved over the entire
tem-perature range by 700C (Fig. 3b). The evolution of H2O at a low
temperature from the co-combustion was attributed to the release of free and physically bounded water. The highest peak at 320C was related to the degradations of SS and CG. At above 350C, the H2O formation curve of SC64 displayed a larger area (the enhanced
H2O release) than that of the individual biofuels. This result was
indicative of the extent of△W where high positive deviations were obtained in this temperature range.
3.4.2. CO2
The elevated temperature began to break the unstable C]O of
the macromolecular organic compounds and generated CO2. The
magnitude of the CO2evolution was generally 40 to 70 times that of
CH3, C2H2, and NO. CO2(m/z¼ 44) were released during the two
stages as a result of the decarboxylation and decarbonylation re-actions. For SS and CG, the second stage with higher CO2emissions
was more prominent than thefirst stage which suggested that CO2
was released more from the combustion of chars than volatiles. The
recurrent records of the higher intensity of CO2þ ion for CG
confirmed its rich carbon content (Table 1). At above 500C, low
CO2emissions were observed for SS probably due to the
polymer-ization reaction of coking [55]. As can be seen inFig. 3e, the lower peak intensity of SC64 than CG indicated that the addition of SS reduced the CO2emissions.
3.4.3. CxHy
The fragment ions of CH3þ(m/z¼ 15) and C2H2þ(m/z¼ 26) are
mainly representative of the evolution of hydrocarbons such as biodegradable materials, and organic species in cells and sewage
polymers [56]. At high temperatures, CeC and CeH bonds were
ruptured to form free radicals which were further recombined into
small molecular compounds. Since the bond energy of CeC
(346.9 kJ/mol) is lower than that of CeH (413.8 kJ/mol) [57], CeC bond is easily broken theoretically. Hence, the cleavage of alkanes with more carbon atoms forms new kinds of alkanes. Furthermore, the formed free radicals can be mutually combined freely with hydrogen atoms [58]. Both fragments for SS were detected mainly
from 220 to 650C and produced continuously over a wide
tem-perature range. High ion concentrations were observed in thefirst stage during the combustion of CG during which a single narrow peak for CH3þ was centered at 330C most probably due to the
cracking of methyl and methylene groups in (hemi)celluloses. A
minor hump in the range of 410e505C stemmed most probably
from the degradation of lignin. With SC64, the two peaks were observed for CH3þand C2H2þat 330 and 500C, respectively. Thefirst
peak was due to the volatile hydrocarbons from both SS and CG. However, the higher intensity of the second peak than the
Table 3
Activation energies for co-combustion of SS and CG in O2/N2atmosphere according to FWO and KAS methods.
Samples a KAS method FWO method E (kJ/mol) R2 E (kJ/mol) R2 SC64 0.2 334.3 0.9985 326.8 0.9986 0.3 420.7 0.9988 409.2 0.9988 0.4 420.2 0.9931 409.0 0.9934 0.5 322.1 0.9896 316.1 0.9903 0.6 177.2 0.9977 179.1 0.9979 0.7 174.6 0.9955 177.5 0.9960 0.8 175.1 1.0000 178.5 1.0000 Average 289.2 285.2 SC73 0.2 345.4 0.9997 337.3 0.9997 0.3 411.5 0.9969 400.5 0.9970 0.4 408.6 0.9983 398.0 0.9984 0.5 329.9 1.0000 323.6 1.0000 0.6 185.7 0.9984 187.2 0.9986 0.7 178.4 0.9977 181.1 0.9980 0.8 176.7 0.9992 180.2 0.9993 Average 290.9 286.8 SC82 0.2 321.0 1.0000 314.0 1.0000 0.3 354.3 0.9963 346.1 0.9964 0.4 332.2 0.9928 325.4 0.9932 0.5 262.8 0.9966 259.9 0.9968 0.6 159.4 0.9989 162.3 0.9990 0.7 159.5 0.9985 163.3 0.9987 0.8 160.5 0.9981 164.8 0.9983 Average 250.0 247.9 SC91 0.2 327.1 0.9974 319.7 0.9975 0.3 403.5 0.9703 392.8 0.9716 0.4 395.4 0.9579 385.4 0.9599 0.5 266.8 0.9881 263.7 0.9889 0.6 165.8 0.9994 168.6 0.9994 0.7 179.2 0.9951 182.0 0.9957 0.8 175.0 0.9954 178.7 0.9960 Average 273.3 270.1 Table 4
Activation energies for co-combustion of SS and CG in O2/CO2atmosphere according to FWO and KAS methods.
Samples a KAS method FWO method E (kJ/mol) R2 E (kJ/mol) R2 SC64 0.2 346.6 0.9407 338.4 0.9435 0.3 484.4 0.9948 469.8 0.9468 0.4 448.4 0.9679 435.8 0.9692 0.5 286.4 0.9692 282.2 0.9712 0.6 149.5 0.9834 152.8 0.9855 0.7 154.4 0.9931 158.3 0.9940 0.8 144.6 0.9997 149.7 0.9997 Average 287.8 283.8 SC73 0.2 321.3 0.9576 314.4 0.9598 0.3 427.1 0.9504 415.3 0.9524 0.4 390.3 0.9679 380.6 0.9599 0.5 252.1 0.9872 249.6 0.9882 0.6 141.7 0.9970 145.5 0.9974 0.7 150.1 0.9977 154.3 0.9980 0.8 142.3 0.9999 147.5 0.9999 Average 260.7 258.2 SC82 0.2 307.5 0.9454 301.2 0.9483 0.3 385.2 0.9839 375.4 0.9846 0.4 355.4 0.9919 347.4 0.9924 0.5 244.7 0.9932 242.6 0.9937 0.6 149.1 0.9961 152.6 0.9965 0.7 154.2 0.9966 158.2 0.9970 0.8 150.1 0.9996 155.0 0.9997 Average 249.4 247.5 SC91 0.2 319.4 0.9605 312.4 0.9626 0.3 388.6 0.9578 378.7 0.9597 0.4 369.3 0.9314 360.6 0.9347 0.5 240.2 0.9765 238.4 0.9783 0.6 148.1 0.9936 151.7 0.9944 0.7 159.6 0.9957 163.4 0.9962 0.8 158.0 0.9994 162.6 0.9994 Average 254.7 252.5
individual fuels confirmed the high reactivity of their intermediates at higher temperatures promoting the decomposition of heavy non-volatiles into CH3þ and C2H2þ. At this temperature, positive
deviations were observed (Fig. 3a and c) for the decomposition of SC64. This showed synergistic effects on the secondary reactions such as the second cracking of C-R groups, and the combination of C and H.
3.4.4. NO
NOx emissions occurred at below 1000C [59]. Since NOx
emissions are mainly composed of NO [59], this study focused on the NO emission patterns. The NO emission profile of SS showed the first peak in the range of 200e415C and the prominent peak
be-tween 415 and 715C. The second peak with a strong emission was
twice thefirst one which indicated that the char N-combustion
required a relatively higher ratio than did the volatile N-combus-tion. Unlike SS, CG showed two continuous peaks in the second stage (415e515C) whose intensities were lower than the second
peak of SS. Biomass char was reported to be more reactive in the reduction of NO than coal char. This result can be related to NO-char
reactions as described below [60]:
2NOþ C/N2þ CO2 (13)
NOþ C/1
2N2þ CO (14)
As shown in Fig. 3d, the addition of CG to SS lowered NO
emissions. There appear to be the following three reasons for this. Firstly, the reaction of the NO precursors of the co-combustion with
O2 was suppressed due to O2 deficiency, thus reducing NO, in
particular, produced by the volatile N-combustion [61]. Secondly, CG produced more CO2than did SS, while the gasification reaction
between CO2and C produced a certain amount of CO. Therefore,
homogeneous reactions between CO and NO reduced the NO emissions [59]. Finally, NO can be reduced on the surface or in the pores of the char to form N2. As can be seen inTable 1, the higher
fixed carbon of CG than SS suggested that more char formed during the co-combustion than the SS combustion. Song et al. [62] pointed out that tar compounds from biomass including acetic acids, toluene, phenol, naphthalene, and 1-hydroxy-naphthalene had a considerable ability to reduce NO emissions. Overall, the cumula-tive and interaction effects of the above factors may make the co-combustion process produce less NO emissions.
3.4.5. SO2
SO2emissions present two peaks for all the fuels (Fig. 3f). The
main reason for the higherfirst than second peak intensity was the volatiles combustion of organic S. SO2evolution was earlier from SS
than CG. As was mentioned in Section3.1, the lower Tiof SS than CG
caused SS volatiles to be rapidly burned to produce SO2. The
detection range of SO2 peaks of CG was only in the range of
200e500C, much narrower than that of SS and SC64. The peak
intensity of SO2among the samples was as follows: SC64> SS > CG.
Not only was the S content of SS higher than that of CG (Table 1), but also theirflammable S origins were different in that SS had mainly organic S and pyrite S, while CG had mainly organic S and sulfate in the main structure. As shown in 3.1, in the main stage of the CG
combustion, a large amount of volatiles was released and burnt rapidly, thus forming a partly oxygen-poor environment. In this case, incompletely combusted S led to some pyrolysis products such as H2S and COS, thus reducing the conversion rate of S to SO2
[63]. Also, the higher contents of alkaline metal (K2O, and Na2O)
and earth metal (CaO, and MgO) of CG than SS promoted sulfate formation when reacted with SO2and reduced SO2emission. SC64
led to the higher intensity of SO2emission than did the individual
fuels which suggested that synergistic effect of the blend. 3.5. Chemical compositions of SS and CG ashes
The ash compositions of SS and CG determined by XRF are presented inFig. 4. SS was mainly composed of clay minerals. SiO2
was the most abundant element of the SS ash which accounted for 50.5% by weight (Table S1). The ash fusion temperature (AFT) may be elevated with the high SiO2content that forms stable SiO2-like
networks. The higher quantity of Al2O3of the SS than CG ash played
a supporting role in the skeleton structure due to its high melting point and raised the AFT [64,65]. Thefluxing effect of CaO on the ash fusion behavior was also observed. The higher CaO content of the CG ash decreased the AFT (Fig. 4). The volatilizing and melting temperatures of SiO2were lowered with the presence of carbonates
[66]. The high level of CaO formed the silicate eutectic with a low
fusion temperature [67]. SiO2 may react with CaO to form Ca
compounds, thus lowering free CaO [68]. To some extent, the low CaO content is advantageous for alleviating the issues of fouling and slagging.
The AFT, and the numerous empirical indices derived from chemical properties have been well accepted as the common in-dustrial guidelines in the practical applications. These include based/acid ratio (Rb/a, Rb/(aþp)), silicon/alumina (SiO2/Al2O3), iron/
calcium ratio (Fe2O3/CaO), slagging index (RS), viscosity index (SR),
fouling index (Fu), and fusion temperature index (F) [64,69e72].
Some of the above indices are shown inTable 5. The oxides of the SS ash such as SiO2, Al2O3, and TiO2with high ionic potential acted as
the network formers and had higher concentration than did those of the CG ash. The CG ash had the higher contents of K2O, Na2O,
CaO, MgO, and Fe2O3 with a low ionic potential weakening the
network than did the SS ash. Thus, the more the SS acidic oxides exist, the smaller the SS Rb/avalue is, as can be seen from Eq.(11).
The calculated Rb/aindicated that SS had a low slagging propensity.
Slagging is generally divided into the low (SR> 78), medium
(66.1< SR< 78) and high (SR< 66.1) levels as defined by Eq.(12).
The higher SRvalue of the SS than CG ash indicated a lower risk of
slagging. The F value expressed in Eq.(13)showed a low slagging potential with SS. The data suggested that blending of SS with CG can overcome the slagging risk.
Rb=a¼Fe2O3þCaO þ MgO þ Na2Oþ K2O SiO2þAl2O3þTiO2
(15)
SR¼ SiO2
SiO2þFe2O3þ CaO þ MgO,100
(16)
F¼SiO2þK2Oþ P2O5
CaOþ MgO (17)
4. Conclusions
Replacement of 79% N2by 79% CO2had a pronounced effect on
the CG combustion such as the smaller maximum weight loss rate, the delay of the peak temperature, the worse burnout performance, and the decreased comprehensive combustion index. The differ-ences between the experimental and calculated TG values pointed to the positive synergistic effects in the oxy-fuel and air atmo-spheres. The positive synergy effects of the blends led to the higher Eavalues for thefirst stage and the lower Eavalues for the second
stage in both atmospheres. The Ea values of the blends for the
second stage were lower in the oxy-fuel than air atmosphere. The CH3, H2O, C2H2and SO2evolutions (except for NO and CO2) peaked
at the range of 250e400C during the volatiles combustion stage.
The addition of CG to SS lowered the NO emission in the char combustion stage. The SO2evolution of SC64 was much higher than
the individual fuels due to the interaction effect. The empirical slagging indices indicated that co-combustion reduced the slagging potential of the CG ash.
Acknowledgements
This study was financially supported by the Scientific and
Technological Planning Project of Guangzhou, China
(No.201704030109; 2016201604030058), the National Natural Science Foundation of China (No.51806040; 51608129), and the Science and Technology Planning Project of Guangdong Province, China (No. 2019B020208017; 2018A050506046; 2017A040403044; 2017A050501036).
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.renene.2019.01.119. References
[1] B.M. Cieslik, J. Namiesnik, P. Konieczka, Review of sewage sludge
manage-ment: standards, regulations and analytical methods, J. Clean. Prod. 90 (2015) 1e15.
[2] J. Nadziakiewicz, M. Koziol, Co-combustion of sludge with coal, Appl. Energy 75 (2003) 239e248.
[3] A. Kijo-Kleczkowska, K. Sroda, M. Kosowska-Golachowska, T. Musial, K. Wolski, Combustion of pelleted sewage sludge with reference to coal and biomass, Fuel 170 (2016) 141e160.
[4] A. Magdziarz, M. Wilk, Thermogravimetric study of biomass, sewage sludge and coal combustion, Energy Convers. Manag. 75 (2013) 425e430. [5] J.Y. Liu, J.J. Zeng, S.Y. Sun, S.S. Huang, J.H. Kuo, N.W. Chen, Combined effects of
FeCl3and CaO conditioning on SO2, HCl and heavy metals emissions during
the DDSS incineration, Chem. Eng. J. 299 (2016) 449e458.
[6] S.S.A. Syed-Hassan, Y. Wang, S. Hu, S. Su, J. Xiang, Thermochemical processing of sewage sludge to energy and fuel: fundamentals, challenges and consid-erations, Renew. Sustain. Energy Rev. 80 (2017) 888e913.
[7] L.M. Huang, J.Y. Liu, Y. He, S.Y. Sun, J.C. Chen, J. Sun, K.L. Chang, J.H. Kuo, X.A. Ning, Thermodynamics and kinetics parameters of co-combustion be-tween sewage sludge and water hyacinth in CO2/O2atmosphere as biomass to
solid biofuel, Bioresour. Technol. 218 (2016) 631e642.
[8] K.K. Agniezka, S. Katarzyna, K.G. Monika, M. Tomasz, W. Krzysztof, Experi-mental research of sewage sludge with coal and biomass co-combustion, in pellet form, Waste Manag. 53 (2016) 165e181.
[9] M.A. Silva, S.A. Nebra, M. Silva, C.G. Sanchez, The use of biomass residues in the Brazilian soluble coffee industry, Biomass Bioenergy 14 (1998) 457e467. [10] S.J. Mussatto, E.M.S. Machado, S. Martins, A. Teixeira, Production, composition, and application of coffee and its industrial residues, Food Bioprocess Technol. 4 (2011) 661e672.
[11] S.J. Mussatto, L.M. Carneiro, J.P.A. Sliva, I.C. Roberto, J.A. Teixeira, A study on chemical constituents and sugars extraction from spent coffee grounds, Car-nojyd. Polym. 83 (2011) 368e374.
[12] N.W. Chen, J. Ren, Z.W. Ye, Q.Z. Xu, J.Y. Liu, S.Y. Sun, Kinetics of coffee in-dustrial residue pyrolysis using distributed activation energy model and components separation of bio-oil by sequencing temperature-raising pyrol-ysis, Bioresour. Technol. 221 (2016) 534e540.
[13] J.C. Chen, J.Y. Liu, Y. He, L.M. Huang, S.Y. Sun, J. Sun, K.L. Chang, J.H. Kuo, S.S. Huang, X.A. Ning, Investigation of co-combustion characteristics of sewage sludge and coffee grounds mixtures using thermogravimetric analysis coupled to artificial neural networks modeling, Bioresour. Technol. 225 (2017) 234e245.
[14] T. Kupka, M. Mancini, M. Irmer, R. Weber, Investigation of ash deposit for-mation during co-firing of coal with sewage sludge, saw-dust and refuse derived fuel, Fuel 87 (2008) 2824e2837.
[15] D. Vamvuka, S. Sfakiotakis, S. Saxioni, Evaluation of urban wastes as promising co-fuels for energy productione a TG/MS study, Fuel 147 (2015) 170e183. [16] Z.Q. Gong, L. Wang, Z.B. Wang, T. Wang, Y. Xu, F.Q. Sun, Z.Q. Sun, Z.Z. Liu,
L.Y. Zhu, Experimental study on combustion and pollutants emissions of oil sludge blended with microalgae residue, J. Energy Inst. 91 (2018) 877e886. [17] Y. Sung, S. Lee, C. Kim, D. Jun, C. Moon, C. Choi, D. Kim, Synergistic effect of
co-firing woody biomass with coal on NOxreduction and burnout during
air-staged combustion, Exp. Therm. Fluid Sci. 71 (2016) 114e125.
[18] P. Teixeira, H. Lopes, I. Gulyurtlu, N. Lapa, P. Abelha, Evaluation of slagging and fouling tendency during biomass co-firing with coal in a fluidized bed, Biomass Bioenergy 39 (2012) 192e203.
[19] F.H. Guo, Z.P. Zhong, Co-combustion of anthracite coal and wood pellets: thermodynamic analysis, combustion efficiency, pollutant emissions and ash slagging, Environ. Pollut. 239 (2018) 21e29.
[20] L.B. Jiang, X.Z. Yuan, H. Li, X.H. Chen, Z.H. Xiao, J. Liang, L.J. Leng, Z. Guo, G.M. Zeng, Co-pelletization of sewage sludge and biomass: thermogravimetric analysis and ash deposits, Fuel Process. Technol. 145 (2016) 109e115. [21] P.A. Bejarano, Y.A. Levendis, Single-coal-particle combustion in O2/N2and O2/
CO2environments, Combust. Flame 153 (2008) 270e287.
[22] T. Wall, Y.H. Liu, C. Spero, L. Elliott, S. Khare, R. Rathnam, F. Zeenathal, B. Moghtaderi, B. Buhre, C.D. Sheng, R. Gupta, T. Yamada, K. Makino, J.L. Yu, An overview on oxyfuel coal combustion-State of the art research and technology development, Chem. Eng. Res. Des. 87 (2009) 1003e1016.
[23] M.B. Toftegaard, J. Brix, P.A. Jensen, P. Glarborg, A.D. Jensen, Oxy-fuel com-bustion of solid fuels, Prog. Energy Combust. Sci. 36 (2010) 581e625. [24] B.J.P. Buhre, L.K. Elliott, C.D. Sheng, R.P. Gupta, T.F. Wall, Oxy-fuel combustion
technology for coalefired power generation, Prog. Energy Combust. Sci. 31 (2005) 283e307.
[25] E. Kastanaki, D. Vamvuka, A comparative reactivity and kinetic study on the combustion of coalebiomass char blends, Fuel 85 (2006) 1186e1193. [26] C.R. Shaddix, A. Molina, Particle imaging of ignition and devolatilization of
pulverized coal during oxy-fuel combustion, Proc. Combust. Inst. 32 (2009) 2091e2098.
Table 5
Indices of slagging and fouling for SS and CG ashes.
Ash sample Rb/aa Rb/(aþp)b S/Ac I/Cd SRe Ff SS 0.258 0.394 2.776 2.393 78.629 10.838 CG 2.497 2.662 3.229 0.481 31.728 1.297 aR
b/a¼ (Fe2O3þCaOþMgOþNa2OþK2O)/(SiO2þAl2O3þTiO2). bR
b/(aþp)¼ (Fe2O3þCaOþMgOþNa2OþK2OþP2O5)/(SiO2þAl2O3þTiO2). c S/A¼ SiO
2/Al2O3. d I/C¼ Fe
2O3/CaO. eS
R¼ (SiO2/(SiO2þFe2O3þCaOþMgO))$100. f F¼ (SiO
[27] J.C. Chen, C.D. Xie, J.Y. Liu, Y. He, W.M. Xie, X.C. Zhang, K.L. Chang, J.H. Kuo, J. Sun, L. Zheng, S.Y. Sun, M. Buyukada, F. Evrendilek, Co-combustion of sewage sludge and coffee grounds under increased O2/CO2 atmospheres:
thermodynamic characteristics, kinetics and artificial neural network modeling, Bioresour. Technol. 247 (2018) 230e238.
[28] H.E. Kissinger, Reaction kinetics in differential thermal analysis, Anal. Chem. 29 (1957) 1702e1706.
[29] T. Akahira, T. Sunose, Joint convention of four electrical institutes, Sci. Technol. 16 (1971) 22e31.
[30] A.W. Coats, J.P. Redfern, Kinetic parameters from thermogravimetric data, Nature 201 (1964) 68e69.
[31] C.D. Doyle, Estimating isothermal life from thermogravimetric data, J. Appl. Polym. Sci. 6 (1962) 639e642.
[32] T. Ozawa, A new method of analyzing thermogravimetric data, Bull. Chem. Soc. Jpn. 38 (1965) 1881e1886.
[33] J.H. Flynn, L.A. Wall, A. quick, Direct method for the determination of acti-vation energy from thermogravimetric data, Polym. Lett. 4 (1966) 323e328. [34] S. Konsomboon, S. Pipatmanomai, T. Madhiyanon, S. Tia, Effect of kaolin
addition on ash characteristics of palm empty fruit bunch (EFB) upon com-bustion, Appl. Energy 88 (2011) 298e305.
[35] C. Peng, Y.B. Zhai, Y. Zhu, B.B. Xu, T.F. Wang, C.T. Li, G.M. Zeng, Production of char from sewage sludge employing hydrothermal carbonization: char properties, combustion behavior and thermal characteristics, Fuel 176 (2016) 110e118.
[36] M.F. Irfan, A. Arami-Niya, M.H. Chakrabarti, W.M.A.W. Daud, M.R. Usman, Kinetics of gasification of coal, biomass and their blends in air (N2/O2) and
different oxy-fuel (O2/CO2) atmospheres, Energy 37 (2012) 665e672.
[37] Y.T. Tang, X.Q. Ma, Z.Y. Lai, Y.X. Fan, Thermogravimetric analyses of co-combustion of plastic, rubber, leather in N2/O2 and CO2/O2 atmospheres,
Energy 90 (2015) 1066e1074.
[38] L.J. Yu, S. Wang, X.M. Jiang, N. Wang, C.Q. Zhang, Thermal analysis studies on combustion characteristics of seaweed, J. Therm. Anal. Calorim. 93 (2008) 611e617.
[39] K.M. Lu, W.J. Lee, W.H. Chen, T.C. Lin, Thermogravimetric analysis and kinetics of co-pyrolysis of raw/torrefied wood and coal blends, Appl. Energy 105 (2013) 57e65.
[40] S.D. Gunasee, M. Carrier, J.F. Gorgens, R. Mohee, Pyrolysis and combustion of municipal solid wastes: evaluation of synergistic effects using TGA-MS, J. Anal. Appl. Pyrol. 121 (2016) 50e61.
[41] Z.Q. Wu, S.Z. Wang, J. Zhao, L. Chen, H.Y. Meng, Synergistic effect on thermal behavior during co-pyrolysis of lignocellulosic biomass model components blend with bituminous coal, Bioresour. Technol. 169 (2014) 220e228. [42] A.K. Sadhukhan, P. Gupta, T. Goyal, R.K. Saha, Modelling of pyrolysis of
coal-biomass blends using thermogravimetric analysis, Bioresour. Technol. 99 (2008) 8022e8026.
[43] P. Yangali, A.M. Celaya, J.L. Goldfarb, Co-pyrolysis reaction rates and activation energies of West Virginia coal and cherry pit blends, J. Anal. Appl. Pyrol. 108 (2014) 203e211.
[44] C.X. Chen, Z.G. Lu, X.Q. Ma, J. Long, Y.N. Peng, L.K. Hu, Q. Lu, Oxy-fuel com-bustion characteristic and kinetics of microalgae Chlorella vulgaris by ther-mogravimetric analysis, Bioresour. Technol. 144 (2013) 563e571.
[45] W. Yang, H. Wang, M. Zhang, J.Y. Zhu, J. Zhou, S.J. Wu, Fuel properties and combustion kinetics of hydrochar prepared by hydrothermal carbonization of bamboo, Bioresour. Technol. 205 (2016) 199e204.
[46] W.X. Wu, Y.F. Mei, L. Zhang, R.H. Liu, J.M. Cai, Kinetics and reaction chemistry of pyrolysis and combustion of tobacco waste, Fuel 156 (2015) 71e80. [47] H. Li, S.Q. Xia, P.S. Ma, Thermogravimetric investigation of the co-combustion
between the pyrolysis oil distillation residue and lignite, Bioresour. Technol. 218 (2016) 615e622.
[48] Y.G. Kin, J.D. Kim, B.H. Lee, J.H. Song, Y.J. Chang, C.H. Jeon, Experimental investigation into combustion characteristics of two sub-bituminous coals in O2/N2and O2/CO2environments, Energy Fuels 24 (2010) 6034e6040.
[49] H. Liu, Combustion of coal chars in O2/CO2and O2/N2mixtures: a comparative
study with non-isothermal thermogravimetric analyzer (TGA) tests, Energy Fuels 23 (2009) 4278e4285.
[50] S. Vyazovkin, Kinetic concepts of thermally stimulated reactions in solids: a view from a historical perspective, Int. Rev. Phys. Chem. 19 (2000) 45e60. [51] Y. Liu, P.F. Fu, B. Zhang, F. Yue, H.C. Zhou, C.G. Zeng, Study on the surface active
reactivity of coal char conversion in O2/N2and O2/CO2atmospheres, Fuel 181
(2016) 1244e1256.
[52] A. Kapanen, M. Vikman, J. Rajas€arkk€a, M. Virta, M. It€avaara, Biotests for environmental quality assessment of composted sewage sludge, Waste Manag. 33 (2013) 1451e1460.
[53] D. Pujol, C. Liu, J. Gominho, M.A. Olivella, N. Fiol, I. Villaescusa, H. Pereira, The chemical composition of exhausted coffee waste, Ind. Crop. Prod. 50 (2013) 423e429.
[54] D. Lopez-Gonzalez, M. Fernandez-Lopez, J.L. Valverde, L. Sanchez-Silva,
Thermogravimetric-mass spectrometric analysis on combustion of lignocel-lulosic biomass, Bioresour. Technol. 143 (2013) 562e574.
[55] A. Magdziarz, S. Werle, Analysis of the combustion and pyrolysis of dried sewage sludge by TGA and MS, Waste Manag. 34 (2014) 174e179. [56] X.Y. Xu, B. Zhao, M.L. Sun, X. Chen, M.C. Zhang, H.B. Li, S.C. Xu, Co-pyrolysis
characteristics of municipal sewage sludge and hazelnut shell by TG-DTG-MS and residue analysis, Waste Manag. 62 (2017) 91e100.
[57] L. Sørum, M.G. Grønli, J.E. Hustad, Pyrolysis characteristics and kinetics of municipal solid wastes, Fuel 80 (2001) 1217e1227.
[58] Z.H. Chen, M. Hu, B.H. Cui, S.M. Liu, D.B. Guo, B. Xiao, The effect of bioleaching on sewage sludge pyrolysis, Waste Manag. 48 (2016) 383e388.
[59] L.M. Shao, S.S. Fan, H. Zhang, Q.S. Yao, P.J. He, SO2and NOxemission from
sludge combustion in a CO2/O2atmosphere, Fuel 109 (2013) 178e183.
[60] L. Dong, S.Q. Gao, W.L. Song, G.W. Xu, Experimental study of NO reduction over biomass char, Fuel Process. Technol. 88 (2007) 707e715.
[61] Y.S. Lin, X.Q. Ma, X.X. Ning, Z.S. Yu, TGA-FTIR analysis of co-combustion characteristics of paper sludge and oil-palm solid wastes, Energy Convers. Manag. 89 (2015) 727e734.
[62] Y. Song, Y. Wang, W. Yang, C.B. Yao, G.W. Xu, Reduction of NO over biomass tar in micro-fluidized bed, Fuel Process. Technol. 118 (2014) 270e277. [63] Q. Zhang, H. Liu, X.J. Zhang, H.X. Xing, H.Y. Hu, H. Yao, Novel utilization of
conditioner CaO for gas pollutants control during co-combustion of sludge and coal, Fuel 206 (2017) 541e545.
[64] J.X. Liu, X.M. Jiang, Y.C. Zhang, J. Shen, H. Zhang, Size effects on the thermal behavior of superfine pulverized coal ash, Energy Convers. Manag. 123 (2016) 280e289.
[65] B.T. Zhao, Z.X. Zhang, X.J. Wu, Prediction of coal ash fusion temperature by least-squares support vector machine model, Energy Fuels 24 (2010) 3066e3071.
[66] R.W. Bryers, Fireside slagging, fouling, and high-temperature corrosion of heat-transfer surface due to impurities in steam-raising fuels, Prog. Energy Combust. Sci. 22 (1996) 29e120.
[67] P. Xing, L.I. Darvell, J.M. Jones, L. Ma, M. Pourkashanian, A. Williams, Experi-mental and theoretical methods for evaluating ash properties of pine and El Cerrejon coal used in co-firing, Fuel 183 (2016) 39e54.
[68] M.B. Folgueras, R.M. Diaz, J. Xiberta, Sulphur retention during co-combustion of coal and sewage sludge, Fuel 83 (2004) 1315e1322.
[69] T.G. Bridgeman, L.I. Darvell, J.M. Jones, P.T. Williams, R. Fahmi, A.V. Bridgewater, T. Barraclough, I. Shield, N. Yates, S.C. Thain, I.S. Donnison, Influence of particle size on the analytical and chemical properties of two energy crops, Fuel 86 (2007) 60e72.
[70] A. Kazagic, I. Smajevic, Synergy effects of co-firing wooden biomass with Bosnian coal, Energy 34 (2009) 699e707.
[71] G. Wang, R.B. Silva, J.L.T. Azevedo, S. Martins-Dias, M. Costa, Evaluation of the combustion behaviour and ash characteristics of biomass waste derived fuels, pine and coal in a drop tube furnace, Fuel 117 (2014) 809e824.
[72] R. García, C. Pizarro, A. Alvarez, A.G. Lavín, J.L. Bueno, Study of biomass combustion wastes, Fuel 148 (2015) 152e159.
