Accepted Manuscript Combustion characteristics and retention-emission of Selenium during co-firing of torrefied biomass and its blends with high ash coal Habib Ullah, Guijian Liu, Balal Yousaf, Muhammad Ubaid Ali, Qumber Abbas, Chuncai Zhou PII: DOI: Reference:

S0960-8524(17)31458-X http://dx.doi.org/10.1016/j.biortech.2017.08.144 BITE 18753

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

13 July 2017 19 August 2017 22 August 2017

Please cite this article as: Ullah, H., Liu, G., Yousaf, B., Ali, M.U., Abbas, Q., Zhou, C., Combustion characteristics and retention-emission of Selenium during co-firing of torrefied biomass and its blends with high ash coal, Bioresource Technology (2017), doi: http://dx.doi.org/10.1016/j.biortech.2017.08.144

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Combustion characteristics and retention-emission of Selenium during co-firing of torrefied biomass and its blends with high ash coal Habib Ullah1,2, Guijian Liu1,2*, Balal Yousaf1, Muhammad Ubaid Ali1, Qumber Abbas1, Chuncai Zhou1 1

CAS-Key Laboratory of Crust-Mantle Materials and the Environments, School of Earth and Space Sciences,

University of Science and Technology of China, Hefei 230026, PR China 2

State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, The Chinese

Academy of Sciences, Xi’an, Shaanxi 710075, China

---------------------------------Email IDs Dr. Habib Ullah: [email protected] Prof. Dr. Guijian Liu: [email protected] Dr. BalalYousaf: [email protected] Dr. Muhammad Ubaid Ali: [email protected] Dr. Qumber Abbas: [email protected] Dr. Chuncai Zhou: [email protected] * Corresponding author: Professor Guijian Liu [email protected] Tel: +86-551-63603714 fax: +86-551-63621485

1

Abstract The combustion characteristics, kinetic analysis and selenium retention-emission behavior during co-combustion of high ash coal (HAC) with pine wood (PW) biomass and torrefied pine wood (TPW) were investigated through a combination of thermogravimetric analysis (TGA) and laboratory-based circulating fluidized bed combustion experiment. Improved ignition behavior and thermal reactivity of HAC were observed through the addition of a suitable proportion of biomass and torrefied. During combustion of blends, higher values of relative enrichment factors in fly ash revealed the maximum content of condensing volatile selenium on fly ash particles, and depleted level in bottom ash. Selenium emission in blends decreased by the increasing ratio of both PW and TPW. Higher reductions (48%) in the total Se volatilization were found for HAC/TPW than individual HAC sample, recommending that TPW have the best potential of selenium retention. The interaction among selenium and fly ash particles may cause the retention of selenium. Key words: High ash coal, pine wood, torrefied pine wood, co-combustion, selenium retentionemission. 1. Introduction Selenium [Se] is one of the environmentally-sensitive and most toxic element, demonstrated in CAAAs (Clean Air Act Amendments) of the USA, 1990 (Tang et al., 2012). Increased emissions of Se and other trace elements has produced a thorough counterpunch for both the environment as well as human health (Liu et al., 2007; Tang et al., 2012; Yousaf et al., 2016a). In feed coal, besides selenium other toxic elements like As, Co, Ni, V, Mn, Zn, Cd, Cr, Pb etc. are also present in enormous amount. Selenium could release and enter the environment from feed coal, as a result of sufficient heat, and complicated physical and chemical reaction during combustion

2

process (Yu et al., 2007). In several studies, positive correlation of toxic elements with ash have been reported, which shows an inorganic affinity (Liu et al., 2004). Recently, the most important anthropogenic cause for Se and other trace elements emission accepted is coal combustion for the production of electricity generation. Selenium and other trace elements present in coal are emitted and reallocated during combustion processes into fly ash, bottom ash, and gaseous state (Tang et al., 2012; Itskos et al., 2010). Volatilization of Se occurs during combustion of coal, in the form of SeO2 and elemental Se0. Gaseous SeO2 in the boiler, can be adsorbed chemically on the surface of fly ash or persist in the gaseous state (Diaz-Somoano et al., 2004). According to Tian et al. (2011), the emissions of Se in central and eastern provinces of China, are usually much larger as compared to those present in the west. In Anhui province, the entire emissions of Se from coal-fired power plants in 2007, were determined at 50.96 t. In general, the determination of total Se and other toxic element levels in both feed coal and burning residues may act as a baseline data on toxic elements volatilization (Matjie et al., 2011). In combustion processes of coal, reasonably comprehensive work had been performed on the conversion behaviors of toxic elements (Furimsky, 2000). Some results demonstrated that the partitioning behaviors regarding toxic elements were assigned to a great level, via transformation of solid/gas stage during combustion process, which in turn determined through fuel composition, toxic element volatilization and their way of existence in coal, conditions of combustion and the interaction with different components of ash (Wang et al., 2014; Miller et al., 2003). The major inorganic components like Ca, Fe and Al present in fly ash can also play a vital role due to their chemical reaction, for determining the partitioning of toxic elements among the particulate phases and gas phase (Lopez-Anton et al., 2006). Considering this mechanism, many solid sorbents such as silica, limestone, sand, bauxite, kaolinite, aluminum oxide, Si-Al-based sorbent,

3

CaO, apatite and Ca-based sorbents were appeared and combine with coal in drop tube furnaces, fixed bed and fluidized bed incinerators as an additive (Yao et al., 2004). The most effective sorbent (Ca based), reduce toxic element after its addition either to combustion boiler or through injection of sorbent into gases during the processes of combustion (Mahuli et al., 1997). The process of adsorption for Ca-based sorbents was considered to be a mixture of physical adsorption and chemical reaction (Mahuli et al., 1997). Biomass is regarded as high alkali earth and alkali contents (Zhou et al., 2014a, 2016). Therefore, co-combustion of high ash content with biomass and torrefied biomass may offer a substitute control process for the retention of selenium. The design of recently used combustion systems are in agreement with the feedstock combustion characteristic and ash yield. The behavior of combustion is totally changed from coal because of the huge differences in the fixed carbon, volatile matter, ash component and mineral structure (Zhou et al., 2014a). Several studies reported different combustion behaviors for various biomass samples (Idris et al., 2010). Thus, it is necessary to study the thermochemical behaviors of coal/biomass blends in order to investigate the combustion characteristics in recently actual combustion schemes. Torrefaction is a type of slow pyrolysis carried out under low-temperature range (200300 °C) with a residence period of 15-30 minutes (Guizani et al., 2016; Kopczynski et al., 2017). A number of advantages such as the rise of heating rate, biological balance, easier milling, lower expenditure of storage and transportation associated with the process of torrefaction, were emphasized (Tchapda and Pisupati, 2014). The process of torrefaction provides a charry substance that contains a low content of moisture with greater calorific value in contrast to the raw biomass. The lower content of moisture improves the deposition time of the torrefied

4

material. Moreover, the grindability of processed biomass is significantly increased by torrefaction (Toptas et al., 2015). As a pretreatment option (torrefaction), has a rewarding response for the reduction of tar pioneers which is extremely useful to get constant gasification (Chew and Doshi, 2011). In recent past, oxygenated torrefaction (under oxidizing envelope) has been competently explored by virtue of propitious utilization of combustion gas just like a torrefaction activator (Lasek et al., 2017). In literature, during the process of co-combustion, torrefied fuels application either as a primary fuel or secondary fuel, is recommended firmly by various authors (Tchapda and Pisupati, 2014; Mun et al., 2016). In the co-combustion process, it is quite obvious that the usage of torrefied material can be a desired preference with coal, rather than raw biomass. In order to know the combustion nature of blends of coal with the torrefied material, still more research works are desired (Goldfarb and Liu, 2013). In this study, the torrefied material of pine wood and raw pine wood were chosen as the biomass fuels. Due to its cost-effective and massive availability, the torrefied and raw pine biomass fuels are considered to have an outstanding prospect for the usage in co-combustion schemes for the production of energy. Thus, the core aims of this study were to investigate (1) the combustion characteristics of high ash coal, pine wood, torrefied pine wood, and their blends through thermogravimetric analysis, (2) selenium partitioning behavior during co-combustion of high ash coal with pine wood and torrefied pine wood. (3) The retention-emission mechanism of selenium during high ash coal co-combusted with biomass and torrefied. It is expected that the results of present work could provide useful information for the capture of selenium and the rational application of both torrefied and biomass. 2. Experimental

5

2.1. Sample selection and preparation In this work, the constituents chosen for sample preparation, were high ash coal (HAC) from Huainan Coal Mine, the biomass of pine wood (PW) and torrefied pine wood (TPW). The samples were firstly air-dried for 24 h, in an oven at 110 °C, crushed and passed through an 80mesh sieve to homogenize for the subsequent analysis. The ultimate samples were kept in the desiccators. Raw biomass and Torrefied were blended with high ash coal to make different binary blends of different proportion of HAC: PW and HAC: TPW. High ash coal: biomass ratios in varying blend compositions were chosen as 90:10, 80:20, 70:30, 60:40 and 50:50, which has been designated as HAC90PW10, HAC80PW20, HAC70PW30, HAC60PW40, and HAC50PW50, respectively. For high ash coal–torrefied biomass combination, composition ratios of HAC: torrefied biomass taken were, 90:10, 80:20, 70:30, 60:40, and 50:50 which have been selected as, HAC90TPW10, HAC80TPW20, HAC70TPW30, HAC60TPW40, and HAC50TPW50, respectively. High ash coal, raw biomass, torrefied and all the blend samples were further crushed to a size of 75 µm, for further study. Proximate and ultimate analysis of the selected samples were measured according to the standard methods of ASTM, and the results were outlined in Table 1. 2.2. Torrefaction Torrefaction experiment of raw PW was conducted within the range of 200-300 °C with the desire to yield a carbon rich and densified solid, under an inert environment. The thermal procedure was conducted in a gas-tight heated electric muffle furnace fixed with a glass tube reactor at a continuous heating rate of 10 °C /min under argon gas at 300 ml/min. About 100 g of the raw PW sample was placed in the glass tube reactor and heated to the fixed temperature (300 °C), and held for 30 minutes at this temperature. The sample was removed after the

6

treatment process, cooled and enclosed with a container to avoid oxidation, and afterward milled to 75 µm size fraction. Blends of HAC and torrefied biomass were then prepared at different ratios based on weight for further investigation. 2.3. Thermo-gravimetric analysis The combustion behavior of HAC, PW, TPW, as well as their blended samples in different proportions were studied by means of thermogravimetric analysis. In order to avoid the interference of mass and heat transfer, about 10 mg sample was applied and carried out under a 75 ml/min air flow from room temperature to 800 °C. The DTG (differential thermogravimetric) curves, were attained at a continuous heating rate of 20 0C/min, from room temperature to 800 °C and retained till there is steadiness in weight loss. The combustion characteristics, such as IT (ignition temperature), combustion reactivity and BT (burnout temperature) were determined according to literature from TG/DTG curve (Park and Jang, 2012; Toptas et al., 2015). 2.4. Trace elements analysis The laboratory-scale CFB (circulating fluidized bed) furnace was used to study the retention features of selenium during the co-combustion of HAC with PW and TPW. The CFB furnace is composed of a stainless steel and covered with a material which is electrically resistant to thermal insulation. The temperature for the reaction was set at 800 °C. The desired samples of bottom ash, fly ash and feedstock were collected at each site. For the preparation of each sample, 0.2 g of the desired sample was first weighed and then added to a digestion flask which was precleaned. The samples were pre-digested using a solution of HNO3, H2SO4, and HClO4 with a ratio of 5:3:2. The samples were left in a fume hood for 24 hours, then placed on a hot plate and were heated from 200-250 °C as far as digested near to dry condition. Then the samples were left to cool down after removing from the hot plate to room temperature. After cooling, 5 ml of pure

7

deionized water were added to the samples and kept at a temperature of 100 °C on the hot plate before evaporated close to a dry condition (Yousaf et al., 2016b; Ali et al., 2017). Samples were again placed for cooling, then diluted by adding 20 ml of clean deionized water. Finally, the digested solutions of samples were filtered via paper membrane for ultimate analysis. The concentration of some trace elements (Se, As, Cd, Cr, Co, Cu, Ni, Mn, Zn, Pb, Fe, Sr, and V) in HAC, PW, TPW and their blends, also in their ash samples, were measured by ICP-MS. 3. Results and discussion 3.1. The physicochemical property of high ash coal, biomass and torrefied The properties of HAC, PW, TPW and their respective blends regarding proximate and ultimate analyses were outlined in Table 1. Higher contents of fixed carbon were found in TPW than HAC and PW. However, the highest volatile matter, hydrogen oxygen in the selected PW and TPW suggesting an improved thermal reactivity and easier ignition (Demirbas, 2004). Therefore, combustion process and ignition property of HAC were expected to promote with torrefied and biomass addition. The variation of N and S mass portion present in the fuel show a direct connection with NOx SO2 emission, respectively (Sommersacher et al., 2011; Kopczynski et al., 2017). Moreover, nitrogen and sulfur contents in the selected PW and TPW samples were less than the HAC, demonstrating that during the co-combustion process of blends, the NOx and SO2 (atmospheric pollutants) emission, could be reduced (Sami et al., 2001; Zhou et al., 2016). 3.2. The thermochemical results of individual fuel The combustion characteristics depicted through thermogravimetric analyzer may provide important information regarding the reactivity and combustibility of the solid fuels in recent combustion amenities (Varol et al., 2010). The results of TGA were outlined in Fig. 1, which indicated the TG and DTG curves, depicted through the temperature programmed burning

8

of HAC, PW, TPW and their blends at 20 °C /min heating rate. Moreover, the thermal parameters including temperature interval, highest temperature, and rate of mass loss (DTGmax) obtained during co-combustion of high ash coal with desired biomass and torrefied biomass were shown in Table 2. As presented in Fig. 1, increased weight loss was found due to the thermal disintegration of organic materials with the increasing combustion temperature. The combustion parameters amongst the selected materials are numerous, however combustion process could be categorized into three stages, that is (Stage 1) moisture dehydration, (Stage 2) discharges of volatiles and (Stage 3) combustions of char (Limayem and Ricke, 2012). Generally, the evolution of stage I ranges from room temperature to 150°C, is attributed to the moisture evaporation in fuels, which will not be expressed in this study. Different thermal behaviors were found between HAC, PW and TPW samples. In Stage 1 for both biomass and torrefied, the weight loss and rate of maximum weight loss were greater as compared to HAC, which was persistent along the moisture content of the designated samples. An important combustion stage for HAC was presented between 329 and 666°C with a maximum weight loss at 462°C which was assigned to the immediate burning of volatile substances and char. The oxidizing combustion associated to release of volatile substances is unremarkable, which was coinciding with the outcomes depicted by earlier investigations (Li et al., 2011). The DTG curves of PW and TPW was fairly changed as compared to HAC, as shown in (Fig. 1). As a lignocellulosic substance, biomass is primarily comprises of three components named hemicellulose, cellulose and lignin. The thermal behavior of biomass addressed in many studies indicated that the emission of volatile substances stage in air current is assigned mainly to the hemicellulose decomposition at temperature range of 220–315°C, cellulose 315 to 400°C and portion of lignin from 137 to 667°C (Limayem and Ricke, 2012).

9

It was perceived that PW and TPW display separate particular combustion behaviors during release and burning of volatile substances (Stage 2). The peak temperature (Tmax) for PW of highest weight loss rate and DTG(max) (377 °C and 0.902%/ °C) were higher than TPW (361°C and 0.605%/ °C). Further, there were two noticeable peaks (225-393°C and 394-671°C) in TPW during volatile matters emission regimes. The biomass and torrefied had a lower ignition temperature as compared to HAC during the process of combustion, which suggested that cocombustion of HAC with biomass could improve the ignition performance (Zhou et al., 2016). For TPW, the ignition temperature increased than PW, which was attributed mainly to the thermal degradation of chemical structure. The maximum temperature of oxidative pyrolysis and char burning of TPW moved to lower temperature demonstrating that it had improved combustibility. The T(max) of TPW (412°C) for Stage 3, was lower than HAC (462 °C), whereas the DTG(max) and weight loss (WL) of TPW (0.312 %/ °C and 26.74 %, respectively) was higher than that of HAC(0.180%/ °C and 23.39%, respectively). 3.3. Combustion characteristics of fuel blends Co-combustion of PW and TPW under certain circumstances may provide a costeffective use of these materials in current coal-fired power plants. Perhaps the higher volatile content of both PW and TPW and the higher carbon content of HAC may counterbalance one another during co-combustion and can deliver an improved burning process as compared to the single fuel combustion. So, the co-combustion study of HAC with PW and TPW were carried out in order to verify the connections amongst them. Figure 1 presented the TG curves of the blends of HAC with different proportions of PW and TPW. With the increase of combustion temperature, samples degradation could be observed along with weight losses after the process of moisture evaporation. Numerous weight losses were found regarding the blend samples amongst

10

the changed adding amounts of biomass and torrefied biomass. TG curves of coal and biomass blends had been focused in several studies demonstrated that the curves were positioned successively among the separate materials (Zhou et al., 2015), whereas different phenomenon was observed in this work. An earlier weight loss occurred in the combustion of blends as compared to coal sample. This stage was attributed to the earlier emission and burning of the volatile substance in both biomass and torrefied material. Mainly two peaks were found from the DTG curves of the blends consistent with the emission of volatile matters in biomass and torrefied, and burning of coal, torrefied and biomass char. The Tmax of PW and TPW decreased from 381°C to 376°C, and from 371°C to 362°C respectively, with the addition of up to 50 wt. % of PW and TPW content in the blend. Concurrently, the intenseness of the maximum DTG for PW and TPW increased from 0.165 to 0.209%/ °C, and from 0.064 to 0.248%/ °C, respectively, and is attributed to the burning of increased volatiles contents during Stage 2. For blends of CG with PW and TPW, the initial temperature decreased from 210 to 185°C and 250 to 228°C (Table 2), with the addition of PW and TPW respectively, in the samples, demonstrating that much better ignition and combustion is promoted. Another peak found due to combustion behavior in the temperature range of 400–670°C, for both the blends of HAC with PW and TPW, which could be attributed to instantaneous char combustion in coal, PW and TPW, and volatile matters present in biomass with high-molecular-mass (Zhang et al., 2013). The intensity of the DTG peaks of Stage 3 decreased with the increase of PW and TPW proportions in the blends, which were assigned to coal char combustion. However, at Stage 3 the T(max) was moved to lower temperature with PW and TPW addition. For HAC and PW blends, the Tmax decreased from 465 to 459°C with the DTG(max) declined from 0.150 to 0.118%/ °C, whereas, for HAC and TPW, the T(max) decreased from 462 to 451°C. In various studies, the

11

reactivity found in solid fuels was positively associated with DTG(max) while inversely associated to Tmax (Vamvuka et al., 2003). Thus, the addition of an appropriate amount of biomass and torrefied could improve the thermic reactivity of HAC. 3.4. The kinetic study of coal, biomass, torrefied and their blends The kinetic parameters like pre-exponential factor and activation energy on solid reaction can be calculated by means of integral and differential techniques (Bamford and Tipper, 1980; Zhou et al., 2014a). The integral technique was manipulated in this work, in order to assess the kinetic characteristics of high ash coal, biomass and torrefied, during the process of cocombustion. Generally, for all kinetics studies, the most accepted major rate equation can be defined as Eq. (1): (1)

dx/dt = K(T). ƒ(x) wherex, represents the ratio of mass conversion during burning, f(x) denote the

suppositious model, determined through the reaction process, while t, is the combustion period (min) and T absolute temperature (K). K(T) indicate reaction rate which can be calculated via Arrhenius calculation as Eq. (2): K(T) =A . exp(−E/RT)

(2)

whereA, denote pre-exponential factor (min-1 ), E is activation energy (kJ/mol), and R is the universal gas constant (8.314 J/(K mol)). The heating rate (H, k/min) is constant when combustion is employed using an anisothermal thermogravimetric study and can be determined by the relationship given below as Eq. (3): H= dT/dt = (dx/dt). (dT/dx)

(3)

Replacement of Eq. (3) in Eq. (1) and adjustments toward Eq. (4): dx/ƒ(x) = (A/H) . exp(−E/RT) . dT

(4)

12

Combining two sides of Eq. (4) results in Eq. (5), as under; 



g(x) = /ƒ(x) = (A/H) ₒ exp −  

(5)

whereg(x), represents the fundamental conversion function. Meanwhile, E/RT≫ 1, substituting the right end of Eq. (5) gives Eq. (6):





(A/H) ₒ exp −   ≅ (ART2/HE) . (1−2RT/E) × exp − 

(6)

By putting Eq. (6) in Eq. (5) and taking logarithms on each side of Eq. (5) following equation as Eq. (7), is obtained: ln(g(x)/T2 = ln (AR (1− 2RT/E)/HE) – E/RT

(7)

As the right side of Eq. (7) is determined as basically constant for maximum values of E and range of temperature for burning. So, a straight line must be attained with a slope of –E/R via plotting the values of ln(g(x)/T2) against 1/T, while E(activation energy), could be determined using the slope of that line. In Table 3, the results regarding kinetic parameters assessed through the first-order chemical reaction for the desired samples were presented, in which R2 described the correlation coefficient. It was quite obvious from the maximum R2 values that the model of the first-order chemical reaction showed agreeable fitting. The activation energy results depicted for the individual samples were consistent with the results attained from preceding works (Idris et al., 2010; Sami et al., 2001). The activation energy for both PW and TPW at volatiles burning profile as well as char burning profile were lower than that of HAC, proposing that both these materials were more susceptible to ignition and burning. The outcomes were agreeable to the earlier stated proximate and ultimate analysis of this study. Mostly, a gradual increase took place in activation 13

energy by adding increased proportion of biomass with high ash coal at Stage 2 (volatiles burning profile), whereas in case of char combustion profile (Stage 3) this phenomenon is reversed, and results were consistent with the previous study (Zhou et al., 2015). The current tendency found in activation energies for the blends of high ash coal/biomass seemed to be in decent agreement to those depicted in other studies (Vuthaluru, 2004; Ismail et al., 2005a). The activation energy as energy barrier provides the knowledge of dire energy required for the initiation of a reaction. Therefore, to confirm lower activation energy, lower temperatures needed for the advancement of accelerative combustion response, the blends having lower activation energy has been suggested. Nevertheless, only those fuel mixture should be taken into consideration, the heating value of which is higher than 20 MJ/kg, in order to make sure flame stability (Biagini et al., 2002; Teixeira et al., 2012). On the basis of this restriction, it was concluded that, based on the lowermost activation energy on coal thermal progress regime and the tolerable heating value limit, the kinetic behaviors of HAC could be improved by adding both PW and TPW in suitable proportions. 3.5. The retention-emission behavior of toxic elements during co-combustion coal with biomass and torrefied As mentioned earlier, biomass is considered as lower heating values with a maximum alkali and alkali-earth elements, which can result in unstable heating and fouling when used in the recent boiler. To make sure the auto-thermal combustion and evade fouling process, more than 20 MJ/kg of the heating values, with < 0.5 base/acid ratios of blends should be mandatory (Teixeira et al., 2012). Keeping these concerns in view, different proportion of additive PW and TPW was supposed to be feasible. So, the HAC/PW and HAC/TPW blends of 90:10, 80:20, 70:30, 60:40 and 50:50 were conducted to study the selenium retention-emission behavior.

14

A number of physical and chemical reactions such as volatilization, decomposition, carbonization, dehydration, dihydroxylation, crystallization, oxidation, reduction, takes place in feedstock during the combustion process, allowing the toxic elements to be transformed. In feed stock, Se was released and redistributed during combustion, amongst the fly ash, bottom ash, and flue gas (Zhou et al., 2014b). The total selenium concentrations in the feedstock, bottom ash and fly ash were presented in Table 4. Highest selenium concentration was found in HAC (5.02 mg/kg) than both the TPW and PW (1.12, 1.22 mg/kg respectively), which was consistent with previous studies (Zhou et al., 2015). Lower selenium concentration of 1.84 mg/kg was found in bottom ash as compared to the corresponding HAC, while its concentration was highly enhanced in fly ash (7.04 mg/kg), which revealed that during combustion process, the volatilization and condensation of selenium on surfaces of fly ash took place. According to the relative enrichment factor (EF) values, selenium was found to be volatile and was enriched in fly ash while depleted in the bottom ash. Higher EF values in fly ash indicated the presence of higher concentration of condensing selenium on fly ash components (Tang et al., 2012; Meij, 1994). Higher selenium concentration was found in both bottom and fly ash for PW as compared to the feedstock, which could be assigned to the volatile matter emission and oxidation of char in PW. In the case of blend samples for both PW and TPW, selenium concentrations in fly ash were higher in comparison with the feedstock. In order to know the behavior of selenium during combustion, the EF may grant some valuable details. Selenium can be located on the fly ash through absorption, chemical reaction and condensation mechanisms (Meij, 1994). Selenium recovery in fly and bottom ash largely relies on the bottom slag alkalinity and process of combustion. During combustion process, trace elements which are related to organic matter and inorganic sulfides e.g. selenium, could volatilize more freely than other elements associated with aluminosilicate

15

minerals and transfer more preferably to fly ash (Sager, 2006). The relative enrichment factor could be calculated as: EF = Cs/Cf × A

(8)

where Cs, is the selenium concentration in the bottom or fly ash, Cf represents selenium concentration of feedstock and A, is the ash yield. The EF of selenium in both bottom and fly ash were determined using Eq. (8), and presented in (Table 4). From the table, it could be noticed that selenium was enriched in fly ash while enervated in the bottom ash. Higher values for EFs of selenium were observed in fly ash, which recommends that selenium was volatile and estimated to be associated with fine particles. In the selected torrefied pine wood sample, the values of EFs for both bottom ash and fly ash (0.33, 1.71 respectively) were greater than HAC (0.21), suggesting the capturing of selenium by torrefied pine wood during combustion. In the case of blend samples of TPW, it was clearly observed that with the addition of torrefied material, the EFs of fly ash increased, indicating that the addition of torrefied biomass could reduce the concentration of volatile selenium. For the evaluation of selenium volatilization level during the process of co-combustion, Vr (volatilization ratio) was carried out according to the equation calculated by (Zhou et al., 2014b) as: Vr = 100 − .  + . /

(9)

In the above equation Cb, Ca and Cf, denotes selenium concentration in bottom ash, fly ash and feedstock, respectively. Both Fa and Fb parameters are the output rates of ash yield (%) in fly ash and bottom ash, respectively. For blend samples, the calculated selenium volatile ratios were determined by its volatilization ratio in separate material and is described as follow, VCal =Rhac × Vhac + Rpw × Vpw

(10) 16

whereRhac and Rpw represents the proportion of high ash coal and pine wood/torrefied pine wood in the blends, respectively. While, Vhac and Vpw are the selenium volatilization ratio during combustion of HAC and pine wood/torrefied pine wood, respectively. In Fig. 2, both calculated and experimental volatile ratios of selenium were plotted. From the figure, it was observed that selenium volatilize ration for both calculated and experimental data decreased gradually with TPW and PW addition. The results recommended that reduction in selenium volatilization ratio can be obtained through co-combustion of HAC with TPW and PW. Higher reductions in the total volatilization were found for HAC/TPW than the HAC/PW blends, which demonstrated different retention-emission behavior of selenium among TPW and PW samples. The maximum reductions for blends of HAC/TPW (almost 48%) recommending that TPW have the best potential of selenium capturing. The comparatively high redistributed behavior of typically high volatile selenium element in fly ash particles, may be illustrated via its affinity towards calcium oxide, which behaves like a scavenging scheme for it in the fly ash (Querol et al., 1995; Tang et al., 2012), and through CaSeO4 formation (Clemens et al., 1999). This relation has been suggested to be responsible for selenium retention in fly ash particles during coal combustion (Yudovich and Ketris, 2005). The EDX spectrum of the raw samples showed a higher content of Ca, Mg, Al, K and Na with lower content of S and Si, suggesting that these elements could play a vital role in the retention of selenium. In HAC, selenium could be linked with both organic and inorganic substances, sometimes it may exist in both forms. In previous studies, it has been described that trace elements linked with sulfides and organic substance are highly volatilized (Querol et al., 1995). Selenium element is primarily associated with sulfide minerals in coal (Zhou et al., 2014b). In general, the dominant forms of Se are organic and sulfidic in coal. Irrespective of selenium types

17

in coal, it is vaporized preferentially in the form of elemental or oxide during combustion of coal, which ultimately condenses on the surfaces of prevailing particles through different physiochemical reactions in post-burning areas, or directly leak out into the air. This partitioning and redistribution behavior of selenium may be influenced by several factors like the initial content of selenium in coal, conditions of combustion and properties of ash. (Xu et al., 2003). Furthermore, chemical composition of ash also affects selenium behavior during process of combustion, particularly for calcium in fly ash. At moderate temperature (about 600 °C), calcium hydroxide (Ca(OH)2) show maximum proficiency for selenium capturing than other minerals like silica, kaolinite and alumina (Ghosh-Dastidar et al., 1996). On the contrary, minerals having calcium such as mono-calcium silicate, di-calcium silicate and calcium oxide, also indicate a different capturing capability for selenium at temperature 600-1000 °C in presence of both air and nitrogen atmosphere (Sterling and Helble, 2003). As mentioned earlier, a feedstock enriched with calcium (Ca), magnesium (Mg), sodium (Na), potassium (K), and aluminum (Al) contents with lower content of Sulphur (S) and silica (Si), would have better ability in the retention of selenium. Therefore, the retention of selenium as well as the thermal reactivity could be improved through the addition of torrefied and biomass in appropriate proportions. 4. Conclusions The combustion characteristics such as ignition efficiency and thermal reactivity during co-combustion could be improved by adding suitable proportions of PW and TPW with HAC. The kinetic features could be upgraded through reasonable PW and TPW addition. During combustion, Se in blends was released and redistributed amongst the bottom ash, fly ash and flue gas. Se volatilization ration reduced gradually with PW and TPW addition. The total reductions

18

in volatilization of HAC/TPW blends were higher than HAC/PW, which revealed different retention-emission behavior of Se. High reductions for HAC/TPW suggesting that TPW have the best potential of selenium capturing. Acknowledgments The authors acknowledge the support from the National Basic Research Program of China (973 Program, 2014CB238903), the National Natural Science Foundation of China (NO. 41672144, 41173032). Chinese scholarship council greatly acknowledged for the provision of PhD fellowship. Authors are also acknowledged editors and reviewers for polishing the language of the paper and for in-depth discussion.

References 1. Ali, M.U., Liu, G., Yousaf, B., Abbas, Q., Ullah, H., Munir, M.A.M., Fu, B., 2017. Pollution characteristics and human health risks of potentially (eco) toxic elements (PTEs) in road dust from metropolitan area of Hefei, China. Chemosphere 181, 111–121. 2. Bamford, C.H., Tipper, C.F.H. (Eds.), 1980. Decomposition reactions of solids. In: Green, N. (Ed.), Comprehensive Chemical Kinetics, vol. 22. Elsevier, New York, pp. 115–246. 3. Biagini, E., Lippi, F., Petarca, L., Tognotti, L., 2002. Devolatilization rate of biomasses and coal-biomass blends: an experimental investigation. Fuel 81, 1041–1050. 4. Chew, J.J., Doshi, V., 2011. Recent advances in biomass pretreatmenttorrefaction fundamentals and technology. Renew. Sustain. Energy Rev. 15(8), 4212-22. 5. Clemens, A.H., Damiano L.F, Gong, D, Matheson, T.W., 1999. Partitioning behavior of some toxic volatile elements during stoker and fluidized bed combustion of alkaline subbituminous coal. Fuel 78: 1379–85. 6. Demirbas, A., 2004. Combustion characteristics of different biomass fuels. Prog. Energy Combust. Sci. 30 (2), 219–230. 7. Diaz-Somoano, M., López-Antón, M.A., Martínez-Tarazona, M.R., 2004. Retention of arsenic and selenium during hot gas desulfurization using metal oxide sorbents. Energy Fuels 18(5):1238–42. 8. Furimsky, E., 2000. Characterization of trace element emissions from coal combustion by equilibrium calculations. Fuel Process. Technol. 63 (1), 29–44. 9. Ghosh-Dastidar, A., Mahuli, S., Agnihotri, R., Fan, L.-S., 1996. “Selenium capture using sorbent powers: Mechanism of sorption by hydrated lime”, Environ. Sci. Technol. 30, 447-452. 10. Goldfarb, J.L., Liu, C., 2013. Impact of blend ratio on the co-firing of a commercial torrefied biomass and coal via analysis of oxidation kinetics. Bioresour. Technol. 149, 208–215. 11. Guizani, C., Haddad, K., Jeguirim, M., Colin, B., Limousy, L., 2016. Combustion characteristics and kinetics of torrefied olive pomace. Energy 107, 453-63. 12. Idris, S.S., Rahman, N.A., Ismail, K., Alias, A.B., Rashid, Z.A., Aris, M.J., 2010. Investigation on thermochemical behaviour of low rank Malaysian coal, oil palm biomass

19

and their blends during pyrolysis via thermogravimetric analysis (TGA). Bioresour. Technol. 101 (12), 4584–4592. 13. Itskos, G., Koukouzasa, N., Vasilatos, C., Megremib, I., Moutsatsouc, A., 2010. Comparative uptake study of toxic elements from aquatic media by the different particle size fractions of fly ash. J. Hazard. Mater. 183, 787–92. 14. Kopczynski, M., Lasek, J.A., Iluk, A., Zuwała, J., 2017. The co-combustion of hard coal with raw and torrefied biomasses (willow (Salix viminalis), olive oil residue and waste wood from furniture manufacturing). Energy 1–10. doi:10.1016/j.energy.2017.04.036. 15. Lasek, J.A., Kopczy, nski M., Janusz, M., Iluk, A., Zuwała, J., 2017. Combustion properties of torrefied biomass obtained from flue gas-enhanced reactor. Energy 119, 362-8. 16. Limayem, A., Ricke, S.C., 2012. Lignocellulosic biomass for bioethanol production: current perspectives, potential issues and future prospects. Prog. Energy Combust. Sci. 38 (4), 449–467. 17. Liu, G.J., Zheng, L.G., Duzgoren-Aydin, N.S., Gao, L.F., Liu, J.H., Peng, Z.C., 2007. Health effects of arsenic, fluorine, and selenium from indoor burning of Chinese coal. Rev. Environ. Contam. Toxicol. 189, 89–106. 18. Liu G.J., Zhang, H.Y., Gao, L.F., Zheng, L.G., Peng, Z.C., 2004. Petrological and mineralogical characterizations and chemical composition of coal ashes from power plants in Yanzhou mining district, China. Fuel Process Technol. 85, 1635–46. 19. Lopez-Anton, M.A., Diaz-Somoano, M., Spears, D.A., Martinez-Tarazona, M.R., 2006. Arsenic and selenium capture by fly ashes at low temperature. Environ. Sci. Technol. 40 (12), 3947–3951. 20. Mahuli, S., Agnihotri, R., Chauk, S., GhoshDastidar, A., Fan, L.S., 1997. Mechanism of arsenic sorption by hydrated lime. Environ. Sci. Technol. 31 (11), 3226–3231. 21. Matjie, R.H., French, D., Ward, C.R., Pistorius, P.C., Li, Z., 2011. Behaviour of coal mineral matter in sintering and slagging of ash during the gasification process. Fuel Process. Technol. 92, 1426–33. 22. Meij, R., 1994. Trace-element behavior in coal-fired power-plants. Fuel Process. Technol. 39 (1–3), 199–217. 23. Miller, B., Dugwell, D.R., Kandiyoti, R., 2003. The influence of injected HCl and SO2 on the behavior of trace elements during wood-bark combustion. Energy Fuels 17 (5), 1382–1391. 24. Mun, T-Y., Tumsa, T.Z., Lee, U., Yang, W., 2016. Performance evaluation of co-firing various kinds of biomass with low rank coals in a 500 MWe coal-fired power plant. Energy 115(Part 1), 954-62. 25. Park, S.W., Jang, C.H., 2012. Effects of pyrolysis temperature on changes in fuel characteristics of biomass char. Energy 39, 187–195. 26. Querol, X., Fernández-Turiel, J.L., López-Soler, A., 1995. Trace elements in coal and their behaviour during combustion in a large power station. Fuel 74(3), 331–43. 27. Sager, M., 2006. Selenium in agriculture, food, and nutrition. Pure Appl. Chem. 78, 111– 133. doi:10.1351/pac200678010111. 28. Sami, M., Annamalai, K., Wooldridge, M., 2001. Co-firing of coal and biomass fuel blends. Prog. Energy Combust. Sci. 27 (2), 171–214.

20

29. Sommersacher, P., Brunner, T., Obernberger, I., 2011. Fuel indexes: a novel method for the evaluation of relevant combustion properties of new biomass fuels. Energy Fuels 26(1), 380-90. 30. Sterling, R. O., Helble, J. J., 2003. “Reaction of arsenic vapor species with fly ash compounds: kinetics and speciation of the reaction with calcium silicates” Chemosphere 51, 1111- 1119. 31. Tang, Q., Liu, G., Yan, Z., Sun, R., 2012. Distribution and fate of environmentally sensitive elements (arsenic, mercury, stibium and selenium) in coal-fired power plants at Huainan, Anhui, China. Fuel 95, 334–339. 32. Tchapda, A., Pisupati, S., 2014. A review of thermal Co-Conversion of coal and biomass/ waste. Energies 7(3), 1098. 33. Teixeira, P., Lopes, H., Gulyurtlu, I., Lapa, N., Abelha, P., 2012. Evaluation of slagging and fouling tendency during biomass co-firing with coal in a fluidized bed. Biomass Bioenergy 39, 192–203. 34. Tian, H.Z., Wang, Y., Xue, Z.G., Qu, Y.P., Chai, F.H., Hao, J.M., 2011. Atmospheric emissions estimation of Hg, As, and Se from coal-fired power plants in China, 2007. Sci Total Environ 409(16), 3078–81. 35. Toptas, A., Yildirim, Y., Duman, G., Yanik, J., 2015. Combustion behavior of different kinds of torrefied biomass and their blends with lignite. Bioresour. Technol. 177, 328– 336. 36. Vamvuka, D., Kakaras, E., Kastanaki, E., Grammelis, P., 2003. Pyrolysis characteristics and kinetics of biomass residuals mixtures with lignite. Fuel 82 (15–17), 1949–1960. 37. Varol, M., Atimtay, A.T., Bay, B., Olgun, H., 2010. Investigation of co-combustion characteristics of low quality lignite coals and biomass with thermogravimetric analysis. Thermochim. Acta 510, 195–201. doi:10.1016/j.tca.2010.07.014 38. Wang, X.Y., Huang, Y.J., Zhong, Z.P., Yan, Y.P., Niu, M.M., Wang, Y.X., 2014. Control of inhalable particulate lead emission from incinerator using kaolin in two addition modes. Fuel Process. Technol. 119, 228–235. 39. Xu, M., Yan, R., Zheng, C., Qiao, Y., Han, J., Sheng, C., 2003. “Status of trace element emission in a coal combustion process: a review”, Fuel Proc. Technol. 85, 215-237. 40. Yao, H., Mkilaha, I.S.N., Naruse, I., 2004. Screening of sorbents and capture of lead and cadmium compounds during sewage sludge combustion. Fuel 83 (7–8), 1001–1007. 41. Yousaf, B., Amina., Liu, G., Wang, R., Imtiaz, M., Rizwan, M.S., Zia-ur-Rehman, M., Qadir, A., Si, Y. 2016a. The importance of evaluating metal exposure and predicting human health risks in urban-periurban environments influenced by emerging industry. Chemosphere. 150, 79-89. 42. Yousaf, B., Liu, G., Wang, R., Imtiaz, M., Zia-ur-Rehman, M., Munir, M.A.M., Niu, 2016b. Bioavailability evaluation, uptake of heavy metals and potential health risks via dietary exposure in urban-industrial areas. Environ. Sci. Pollut. Res. 23, 22443-22453. 43. Yu, J.L., Lucas, J.A., Wall, T.F., 2007. Formation of the structure of chars during devolatilization of pulverized coal and its thermoproperties: a review. Prog. Energy Combust. Sci. 33, 135–70 44. Yudovich, Y.E., Ketris, M.P., 2005. Arsenic in coal a review. Int. J. Coal Geol. 61, 141– 96.

21

45. Zhang, K.H., Zhang, K., Cao, Y., Pan, W.P., 2013. Co-combustion characteristics and blending optimization of tobacco stem and high-sulfur bituminous coal based on thermogravimetric and mass spectrometry analyses. Bioresour. Technol. 131, 325–332. 46. Zhou, C., Liu, G., Fang, T., Lam, P.K.S., 2015. Investigation on thermal and trace element characteristics during co-combustion biomass with coal gangue. Bioresour. Technol. 175, 454–462. 47. Zhou, C., Liu, G., Wang, X., Qi, C., 2016. Co-combustion of bituminous coal and biomass fuel blends: Thermochemical characterization, potential utilization and environmental advantage. Bioresour. Technol. 218, 418–427. 48. Zhou, C.C., Liu, G.J., Cheng, S., Fang, T., Lam, P.K.S., 2014a. Thermochemical and trace element behavior of coal gangue, agricultural biomass and their blends during cocombustion. Bioresour. Technol. 166, 243-251. 49. Zhou, C.C., Liu, G.J., Fang, T., Wu, D., Lam, P.K.S., 2014b. Partitioning and transformation behavior of toxic elements during circulated fluidized bed combustion of coal gangue. Fuel 135, 1–8.

22

23

Fig. 1. The TG and DTG curves of the selected high ash coal (HAC), pine wood (PW), torrefied pine wood (TPW) and their blends at heating rate of 20 °C /min.

24

Fig. 2. A comparison between the calculated and experimental volatilize ratios of selenium.

25

Table 1. Proximate analysis on air dried basis (wt. %), Ultimate analysis (wt. %) of high ash coal, biomass, torrefied biomass and their blends.

Sample HAC

Proximate analysis, dry basis (wt. %) MC A VM FC 1.21 58.2 16.5 24.09

Ultimate analysis, daf (wt. %)

PW

4.01

5.23

76.4

14.36

49.2

6.2

43.86

0.14 0.6

TPW

3.01

22.5

49.4

27.02

66.01 5.7

27.84

0.85 0.6

HAC90PW10

1.62

56.4

20.22 21.76

--

--

--

--

--

HAC80PW20

1.83

47.4

31.5

19.27

--

--

--

--

--

HAC70PW30

1.83

39

45.5

15.67

--

--

--

--

--

HAC60PW40

1.91

35.54

53.5

15.05

--

--

--

--

--

HAC50PW50

2.04

30.4

66.5

13.06

--

--

--

--

--

HAC90TPW10

1.41

57.1

35

11.49

--

--

--

--

--

HAC80TPW20

1.41

51.4

38

9.19

--

--

--

--

--

HAC70TPW30

1.51

46.4

40.5

12.09

--

--

--

--

--

HAC60TPW40

1.62

44

45

9.79

--

--

--

--

--

HAC50TPW50

1.80

43.4

47

5.59

--

--

--

--

--

Trace elements (mg/kg) HAC

Se

As

Cr

Ni

V

Co

Zn

Pb

C 64

H 4.37

Oa N S 24.33 1.13 1.15

5.02

4.56 215.5

44.57 127.3

28.23 148.2

36.47

PW

1.22

3.43 374.3

25.34

14.08

20.82

TPW

1.12

1.13 337.4

8.27

Note: a calculated by difference

26

58.93

2.59 1.34

65

58.29 1.83

Table 2. The temperature range, highest temperature, maximum degree of weight loss and rate of weight loss for the selected samples at the heating rate of 20 °C /min. Sample

HAC

Stage 2 Range(°C) --

T(max) --

Stage 3 Range(°C)

T(max)

--

329-666

462

0.180

23.39

DTG(max) % /°C --

WL %

DTG(max)

WL %

HAC90PW10

210-400

381

0.165

13.80

401-663

465

0.150

19.37

HAC80PW20

206-405

378

0.181

16.02

406-661

461

0.147

19.03

HAC70PW30

201-408

377

0.194

19.73

409-659

459

0.133

18.86

HAC60PW40

195-412

376

0.203

23.42

413-655

460

0.124

17.68

HAC50PW50

185-413

376

0.209

26.39

414-649

459

0.118

17.28

PW

150-417

377

0.902

72.24

HAC90TPW10

250-393

371

0.064

6.76

398-694

462

0.162

21.89

HAC80TPW20

239-396

363

0.112

9.62

400-671

460

0.158

21.48

HAC70TPW30

235-400

361

0.172

14.91

401-664

456

0.147

20.63

HAC60TPW40

230-401

361

0.256

17.87

392-662

451

0.174

22.64

HAC50TPW50

228-402

362

0.248

20.86

394-668

451

0.170

22.47

TPW

225-393

361

0.605

28.67

394-671

412

0.312

26.74

27

--

--

--

--

Table 3. Kinetic characteristics of high ash coal (HAC), pine wood (PW), torrefied pine wood (TPW) and their blends at heating rate of 20 °C /min. Sample HAC

Stage 2 E (KJ mol-1) --

r --

Stage 3 E (KJ mol-1) 139.67

r 0.9946

HAC90PW10

65.98

0.9962

110.76

0.9991

HAC80PW20

70.23

0.9971

105.56

0.9934

HAC70PW30

74.54

0.9951

99.43

0.9994

HAC60PW40

78.09

0.9983

95.74

0.9996

HAC50PW50 PW

81.74 85.89

0.995 0.9968

93.44 --

0.9981 --

HAC90TPW10

80.95

0.9988

126.50

0.9962

HAC80TPW20

83.23

0.9878

117.47

0.9943

HAC70TPW30

88.45

0.9994

113.35

0.9968

HAC60TPW40

92.12

0.9986

111.34

0.9928

HAC50TPW50

95.64

0.9998

103.45

0.9932

TPW

98.82

0.9964

70.20

0.995

Notes: E = activation energy; r = the correlation coefficient.

28

Table 4. Selenium concentration in Feedstock, bottom ash, fly ash (mg/kg) and relative enrichment factors (EF) for both bottom and fly ash. Sample

Feed stock

HAC

5.02

Bottom Ash 1.84

HAC90PW10

4.17

HAC80PW20

Fly ash

EF (bottom ash)

EF (fly ash)

7.04

0.21

0.81

1.28

6.02

0.17

0.81

3.39

1.22

6.82

0.17

0.95

HAC70PW30

4.99

0.48

5.98

0.03

0.44

HAC60PW40

5.22

1.15

5.98

0.06

0.33

HAC50PW50

5.88

1.65

5.13

0.05

0.16

PW

1.22

1.71

3.87

0.07

0.16

HAC90TPW10

6.35

1.36

5.34

0.12

0.44

HAC80TPW20

4.22

1.58

5.44

0.19

0.66

HAC70TPW30

3.69

0.47

5.82

0.05

0.73

HAC60TPW40

3.21

0.91

5.25

0.12

0.75

HAC50TPW50

3.51

1.26

6.28

0.15

0.78

TPW

1.12

1.68

0.98

0.33

1.71

29

•

The combustion characteristics of HAC, PW and TPW were assessed.

•

The kinetic study of high ash coal co-combusted with PW and TPW, were depicted.

•

The retention-emission behavior of selenium during combustion process was studied.

•

Se emission was decreased during co-combustion of TPW with HAC.

•

The high affinity of selenium with fly ash particles may cause Se-retention.

30