- Email: [email protected]
A comparative study of nitrogen conversion during pyrolysis of coconut fiber, its corresponding biochar and their blends with lignite
Bioresource Technology 151 (2014) 85–90
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
A comparative study of nitrogen conversion during pyrolysis of coconut fiber, its corresponding biochar and their blends with lignite Zhengang Liu, Rajasekhar Balasubramanian ⇑ Department of Civil and Environmental Engineering, National University of Singapore, Singapore 117576, Singapore
h i g h l i g h t s Coconut fiber (CF), its biochar and their blends with lignite were pyrolyzed. CF and its biochar showed different nitrogen partitioning between HCN and NH3. Interactions between CF/biochar and lignite decreased HCN and NH3 yields. Less HCN+NH3 and more N2 were generated from biochar/lignite pyrolysis. Enhanced benefits can be achieved from biochar/lignite co-processing.
a r t i c l e
i n f o
Article history: Received 19 August 2013 Received in revised form 12 October 2013 Accepted 14 October 2013 Available online 22 October 2013 Keywords: Nitrogen distribution Raw biomass Biochar Synergistic interactions
a b s t r a c t In the present study, the conversion of fuel-N to HCN and NH3 was investigated during rapid pyrolysis of raw biomass (coconut fiber), its corresponding biochar and their blends with lignite within a temperature range of 600–900 °C. The results showed that the raw biomass and the biochar showed totally different nitrogen partitioning between NH3 and HCN. HCN was the dominant nitrogen pollutant from pyrolysis of raw biomass, while for the biochar pyrolysis the yield of NH3 was slightly higher than that of HCN. Synergistic interactions occurred within both raw biomass/lignite and biochar/lignite blends, especially for the biochar/lignite blend, and resulted in reduced yields of HCN and NH3, decreased the total nitrogen percentage retained in the char and promoted harmless N2 formation. These findings suggest that biochar/lignite co-firing for energy production may have the enhanced benefit of reduced emissions of nitrogen pollutants than raw biomass/lignite. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Biomass and coal co-firing for heat and power production in existing coal-fired power plants is viewed as a key strategy of combining the utilization of biomass energy with the substantial reduction of CO2 discharged into the atmosphere. In recent years, extensive studies associated with biomass and coal co-processing have been carried out. These studies have been mostly focused on thermal conversion characteristics of biomass/coal co-processing systems (Haykiri-Acma and Yaman, 2007; Jones et al., 2005; Kwong et al., 2007; Lu et al., 2013; Park et al., 2010; Vuthaluru, 2004). For example, the addition of sawdust to coal changed the pyrolysis product streams significantly, resulting in decreased char and tar yields, and increased gas product yield (Park et al., 2010). The pyrolysis of three different types of coal in the presence of pinewood decreased aromatics and increased phenols in the resultant pyrolysis oil (Jones et al., 2005). Compared with coal combus⇑ Corresponding author. Tel.: +65 65165135; fax: +65 67744202. E-mail address: [email protected] (R. Balasubramanian). 0960-8524/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2013.10.043
tion alone, the biomass/coal co-firing process had lowered combustion temperature and reduced energy output (Kwong et al., 2007). However, raw biomass is not an ideal solid fuel, and large-scale applications of biomass/coal co-firing are restricted by inherent properties of biomass materials such as poor grindability and low energy density (Ibrahim et al., 2013; Khan et al., 2009). The poor grindability, caused by the fibrous structure of biomass feedstock, creates a challenge for fuel preparation and feeding to the process unit. The low energy density resulting from high oxygen content and high moisture content leads to significant costs for the transportation and storage. It is, therefore, necessary to enhance the fuel quality of raw biomass prior to co-processing with coal. Low temperature pyrolysis is one of attractive technologies to upgrade biomass feedstocks. The resultant biochar exhibits improved fuel quality compared to raw biomass including elevated energy density and increased grindability, addressing the key problems that hinder the use of raw biomass as a direct fuel (Fisher et al., 2012; Ibrahim et al., 2013; Park et al., 2012). As an example, energy consumption for the grinding of torrefied biomass was
86
Z. Liu, R. Balasubramanian / Bioresource Technology 151 (2014) 85–90
significantly decreased compared to that of untreated biomass (Phanphanich and Mani, 2011). In addition, with the improved physicochemical properties, favorable thermal conversion characteristics have been observed by the use of biochar. For example, the gasification of pyrolytic biochar produced more H2 and CO than that of raw biomass (Chen et al., 2013). Compared to raw biomass pyrolysis, H2 and CH4 were enriched in the syngas, and the increased concentrations of phenols and sugars and reduced concentrations of guaiacols and furans were reported from the pyrolysis of torrefied biomass (Ren et al., 2013). Apart from thermal conversion efficiency, as a direct fuel, emission of gaseous pollutants from the combustion of raw biomass materials is also one of the key issues in the context of developing advanced clean and renewable energy technology (Khan et al., 2009). Pyrolysis is the initial step for the combustion; therefore, understanding nitrogen conversion during pyrolysis is essential to efficiently control the formation of nitrogen pollutants, NOx and N2O, from subsequent combustion. Some studies have reported total nitrogen distribution in product streams during individual biomass and coal pyrolysis. It appears that nitrogen conversion depends on many factors including pyrolysis conditions and starting biomass feedstocks (Glarborg et al., 2003). For example, due to high contents of lignin in lignocellulosic biomass, much more HCN than NH3 was formed from the pyrolysis of lignocellulosic biomass than aquatic biomass (Yuan et al., 2010, 2011). Coal pyrolysis indicated that the coal type played an important role in the distribution of nitrogen among HCN, NH3 and N2, and Fe-containing materials served as a catalyst to promote N2 formation (Wu and Ohtsuka, 1997a). As for biomass/coal co-pyrolysis, synergistic interactions between biomass and coal were reported, leading to decreased char-N yield and increased volatile-N yield. However, the total amount of NH3 and HCN, in volatile decreased from biomass/coal co-pyrolysis (Yuan et al., 2010, 2011). It was also reported that the addition of hydrothermally pre-treated biomass in lignite suppressed the formation of HCN and NH3, and promoted the nitrogen conversion into harmless N2 at temperatures P800 °C (Liu et al., 2013a). A number of studies have been reported on nitrogen conversion during coal pyrolysis, while studies dealing with nitrogen conversion during raw biomass pyrolysis alone and raw biomass/coal copyrolysis are relatively sparse (Hansson et al., 2004; Glarborg et al., 2003; Tian et al., 2007; Yuan et al., 2010, 2011). It was reported that the interactions occurring between biomass feedstocks and coal are associated with the similarity of their fuel ranks during raw biomass and coal co-pyrolysis (Moon et al., 2013). With the enhanced fuel quality, the interactions occurring between biochar and coal are expected to differ from those between raw biomass and coal, leading to different nitrogen containing emissions from raw biomass pyrolysis (Moon et al., 2013). With the increasing importance of biomass energy, large amounts of biochar will be produced in the near future. To the best of our knowledge, no information is currently available for the formation and relative distributions of HCN and NH3 during pyrolysis of biochar and its blends with coal. In the present study, the conversion of fuel-N was investigated during pyrolysis of raw biomass, its corresponding biochar and their blends with lignite to assess the environmental benefits of biochar/coal co-processing.
2. Methods 2.1. Materials Coconut fiber (CF) was selected as a representative biomass due to its abundance in tropical areas in the present study. The biochar was produced from typical pyrolysis upgrading of raw biomass and
detailed procedures were reported elsewhere (Liu and Balasubramanian, 2013a). Briefly, CF was heated from room temperature to 350 °C under an inert atmosphere. A temperature of 350 °C was chosen because previous experiments have shown it to be suitable for biochar production as solid fuel (Liu and Balasubramanian, 2013a; Park et al., 2012). After maintaining at 350 °C for 20 min, the reactor was cooled down to room temperature. The biochar was recovered as a solid residue and labeled as PY-CF. The lignite sample was obtained from Pasar, Indonesia. All the fuel samples were dried at 105 °C for 24 h and crushed to less than 125 lm for the subsequent pyrolysis. The fuel blends were prepared using lignite weight fraction of 75%. Table 1 presents ultimate and proximate analysis results of each fuel used in the present study. 2.2. Pyrolysis A horizontal fixed-bed tubular quartz reactor (60 mm i.d., 1600 mm long) connected to a gas trap was used for the pyrolysis experiments. In each run, about 0.6 g of the fuel sample was loaded into an alumina boat that was then placed inside the quartz tube. Argon gas (flow rate 100 ml/min) was introduced into the pyrolysis system to maintain an inert atmosphere around the fuel sample. After the furnace was heated to the pre-determined temperature, the fuel sample was positioned within the heated area by pushing the quartz tube further into the furnace. The fuel sample was pyrolyzed at the preset temperature (600–900 °C) for 20 min, followed by keeping it out of the heating area and then cooling it down to the room temperature. Gaseous HCN and NH3 produced during pyrolysis were collected by absorption in 20 mmol/L NaOH and 2 mmol/L HNO3 solutions in gas traps, respectively. Separate runs were carried out for collecting NH3 and HCN produced because HCN and NH3 have non-negligible solubilities in HNO3 and NaOH solutions, respectively. Three replicates of each pyrolysis experiment were performed to ensure reproducibility of the emissions. 2.3. Quantification The carbon, hydrogen, sulfur and nitrogen contents were determined using a Vario MACRO cube Elemental Analyzer (USA). As for the metal content analysis, the fuel sample was digested in mixed acid solutions and the metal concentration (K, Na, Ca and Fe) in the solution was quantified by ICP-OES Perkin Elmer 3000DV (USA). The char fraction that remained in the alumina boat was weighted and its nitrogen content (char-N) was determined. The quantity of NH3 was measured using a Flow Injection Analyzer (FIA)
Table 1 Ultimate analysis, proximate analysis and major metal content analysis of CF, PY-CF and lignite. Fuel
CF
PY-CF
Lignite
Ultimate analysis (dry ash free, wt%) C 47.75 H 5.61 N 0.90 S 0.23 O (By difference) 45.51
67.51 3.95 1.01 0.37 27.16
61.64 5.72 1.74 0.77 30.13
Proximate analysis (dry base, wt%) Volatile matter 80.69 Fixed carbon 10.26 Ash 9.05
30.02 60.81 9.18
48.06 40.68 9.26
3.13 0.69 0.24 0.04
0.04 2.86 0.10 0.59
Major metal content analysis (dry base, wt%) K 1.61 Ca 0.34 Na 0.12 Fe 0.02
Z. Liu, R. Balasubramanian / Bioresource Technology 151 (2014) 85–90
(Quickchem 8500, Lachat Instrument). For HCN, its concentration was determined using an ion chromatograph (ICS-2000, Dionex) equipped with an amperometer detector (An IonPac AS7 Column was used and the eluent was CH3COONa (100 mmol/l) with the flow rate of 10 ll/min). The yields of char-N, HCN and NH3 were calculated as a percentage of total nitrogen in the fuel sample. 3. Results and discussion We have recently reported the nitrogen conversion during the pyrolysis of lignite within a temperature range of 600–900 °C (Liu et al., 2013a). This work focused on the comparative evaluation of CF and PY-CF, and the influences of their additions to lignite on the total nitrogen distribution during co-pyrolysis. 3.1. Pyrolysis of individual CF and PY-CF HCN and NH3 are commonly believed to be two important precursors of NOx and N2O resulting from solid fuel combustion; therefore, it is very important to determine the yields of HCN and NH3 from CF and PY-CF pyrolysis to achieve a cleaner combustion. As can be seen from Fig. 1, CF and PY-CF showed totally different patterns of total nitrogen partitioning between HCN and NH3. HCN was the dominant nitrogen pollutant from the pyrolysis of CF, and the yield of HCN was much higher than that of NH3, especially at low temperatures. For example, HCN and NH3 accounted for 8.5 and 2.8% of total nitrogen at 600 oC pyrolysis, respectively. In contrast, NH3 was the main pollutant from PY-CF pyrolysis and the difference of the yields between HCN and NH3 was only marginal at entire temperature range, similar to that of lignite pyrolysis (Liu et al., 2013a). The yields of HCN and NH3 were 2.4 and 2.5% at 600 °C, and 1.8 and 1.9% at 900 °C from PY-CF pyrolysis. In addition, the temperature played an important role in the nitrogen partitioning between HCN and NH3 for both CF and PY-CF. The yields of HCN and NH3 decreased with the increasing pyrolysis temperatures for both CF and PY-CF with the exception of NH3 yield from CF pyrolysis. In the case of the NH3 yield from CF pyrolysis, it increased with the increasing temperatures, reaching a maximum of 3.2% at 800 °C and then its yield decreased with the further increasing temperatures up to 900 °C. In comparison to PY-CF, more HCN and NH3 were generated from CF pyrolysis at all tested temperatures. The most important source of N2O is believed to originate from oxidation of the gaseous HCN (Kilpinen and Hupa, 1991). Therefore, more NOx and N2O are expected from CF than those of PY-CF during subsequent combustion. 9.0 HCN HCN HCN NH3 NH3 NH3
Nitrogen distribution (%)
7.5
6.0
CF PY-CF Lignite CF PY-CF Lignite
4.5
3.0
1.5
0.0 600
700 800 o Temperature ( C)
900
Fig. 1. Nitrogen partitioning of HCN and NH3 from CF and PY-CF at different pyrolysis temperatures (lignite for comparison).
87
Nitrogen conversion during biomass pyrolysis is known to be closely related to the chemical composition of biomass materials (Yuan et al., 2010, 2011). From the structural analysis, cellulose and hemicellulose originally contained in CF were totally decomposed /carbonized and a part of lignin was also decomposed during pyrolysis upgrading (Liu et al., 2013b). Therefore, the different nitrogen conversions for CF and PY-CF can mainly be ascribed to different chemical compositions of CF and PY-CF. In addition, it was reported that alkali and alkaline earth metal (AAEM) can suppress the formation of HCN, and catalyze the conversion from fuelN to NH3 during coal pyrolysis (Ohtsuka et al., 1997). Our previous study revealed that all the metals originally contained in raw biomass remain in the biochar during biomass pyrolysis upgrading (Liu and Balasubramanian, 2013b). As a consequence, the AAEM contents in PY-CF are higher than those of CF as confirmed in Table 1 (generally the contents of Ca, K and Na in PY-CF are two times higher than those in CF). Therefore enriched AAEM in PY-CF also partly contributes to different fuel-N partitioning of HCN and NH3 during pyrolysis compared to CF. Lastly, HCN is believed to be mainly generated from the thermal cracking of unstable N-containing compounds in the volatiles while NH3 is formed from more stable N-containing compounds in the nascent char during pyrolysis (Hansson et al., 2004; Yuan et al., 2010). Due to sharply lower volatile content of PY-CF (80.69% and 30.02% for CF and PY-CF, respectively), less polymerization reactions among volatiles presumably occurred during the pyrolysis of PY-CF than that of CF, resulting in less HCN formed during PY-CF pyrolysis. Increased cavity in PY-CF also could lead to less polymerization, contributing to the low HCN yield from PY-CF pyrolysis. It is worth noting that the results obtained in this study are distinctly different from those from previous reports in which increased lignin contents in biomass resulted in the increased HCN yield from pyrolysis (Yuan et al., 2010, 2011). From the above analysis, it is known that PY-CF has higher lignin contents than CF. However, NH3 was the dominant nitrogen pollutant from pyrolysis of PY-CF, not HCN. These results support the conclusion made in our previous study: besides the total lignin content, the state of lignin in biomass materials is also important in governing the nitrogen partitioning of NH3 and HCN during pyrolysis (Liu et al., 2013a). In the present study, the total nitrogen in pyrolysis products was divided into three groups: char-N, (HCN + NH3)-N, (tar + N2)-N. The amount of (tar + N2)-N from each experiment was calculated by difference between the total fuel-N and the sum of char-N and (HCN + NH3)-N due to the difficulty of the collection and quantification of tar condensed inner side of tube. Fig. 2 shows the total fuel-N partitioning of each product group from CF and PY-CF pyrolysis. It was clear that more nitrogen retained in the char and less (tar + N2)-N was formed from PY-CF pyrolysis than those of CF pyrolysis. For both CF and PY-CF pyrolysis, the yields of char-N and (HCN + NH3)-N decreased with the increasing temperature and the total nitrogen was mainly in the form of tar + N2 at temperatures higher than 700 °C for CF and 900 °C for PY-CF, respectively. For PY-CF pyrolysis, the char-N yield decreased with the increasing temperature and dominated at the temperature range of 600–800 °C. The yields of (HCN + NH3)-N and (tar + N2)-N were minimal at temperatures 6700 °C. The combined amounts of these two groups only accounted for 11% and 18% of total nitrogen from pyrolysis at 600 and 700 °C, respectively. However, a sharp increase of (tar + N2)-N yield was observed when the pyrolysis temperature was higher than 800 °C. The yield of (tar + N2)-N increased from 5.9% to 26.8% with the temperature increase from 600 to 800 °C, and with a further increase to 48.3% at 900 °C. In the case of CF pyrolysis, similar effect of temperature on the nitrogen distribution was observed. When compared to PY-CF pyrolysis,
Z. Liu, R. Balasubramanian / Bioresource Technology 151 (2014) 85–90
Char-N Char-N (HCN+NH3)-N (HCN+NH3)-N (Tar+N2)-N (Tar+N2)-N
100
Nitrogen distribution (%)
80
5.0
CF PY-CF CF PY-CF CF PY-CF
4.0 Nitrogen distribution (%)
88
60
40
20
HCN HCN NH3 NH3
Theoretical Experimental Theoretical Experimental
A
CF/Lignite blend
3.0
2.0
1.0
0 600
700
800
0.0
900
600
o
Temperature ( C)
700 800 o Temperature ( C)
900
Fig. 2. Nitrogen distributions of CF and PY-CF at different pyrolysis temperatures.
3.2. Pyrolysis of CF/lignite and PY-CF/lignite blends
4.0
Nitrogen distribution (%)
the char-N yield was significantly lower while the yield of (HCN + NH3)-N was higher for CF pyrolysis at the entire temperature range. The less pollutant emissions together with improved combustion behaviors of PY-CF indicate that as a solid fuel, except higher thermal efficiency, the additional enhanced environmental benefits can also be achieved by using pyrolytic biochars compared to raw biomass.
HCN HCN NH3 NH3
B
Theoretical PY-CF/Lignite blend Experimental Theoretical Experimental
3.0
2.0
1.0
Fig. 3 presents the total nitrogen partitioning of HCN and NH3 from pyrolysis of CF/lignite and PY-CF/lignite blends. The yields of NH3 and HCN varied with the temperature for both blends. The addition of CF and PY-CF to lignite showed similar effects on HCN and NH3 yields from co-pyrolysis: the yields of HCN and NH3 increased initially with the increasing temperature reaching a maximum and then decreased with a further increase of pyrolysis temperature. For CF/lignite blend, the yields of HCN and NH3 reached their maximum at 800 °C, with 2.8 and 2.9% for HCN and NH3, respectively. In the case of PY-CF/lignite pyrolysis, HCN and NH3 reached their maximum yield at 700 °C (1.4%) and 800 °C (2.0%), respectively. Based on individual proportions of each fuel in the blends, with the assumption that there is no interaction between each component during co-pyrolysis, theoretical yields of HCN and NH3 from CF/lignite and PY-CF/lignite were calculated by the following equation (Eq. (1)):
YieldBlend ¼ 25% YieldCF=PY-CF þ 75% YieldLignite
ð1Þ
The experimental and theoretical yields of HCN and NH3 from the blends pyrolysis are shown in Fig. 3. PY-CF addition notably suppressed the formation of both HCN and NH3. Though less pronounced, similar trends were observed for the addition of CF to lignite. For PY-CF and lignite co-pyrolysis, the experimental yield of NH3 decreased by 29.6% and 67.8% at 600 and 900 °C, respectively (relative to the theoretical yields), and the yield of HCN decreased by 12.6% and 48.4% at 600 and 900 °C, respectively. In the case of CF and lignite co-pyrolysis, the experimental yield of NH3 decreased by 11.3% and 31.8% at 600 and 900 °C, respectively, and that of HCN decreased by 2.3% and 11.0% at 600 and 900 °C, respectively. The experimental yields of both HCN and NH3 were lower than those of theoretical values from CF/lignite and PY-CF/lignite pyrolysis, implying the environmental benefit of biomass/coal and biochar/coal co-pyrolysis than raw biomass, biochar and coal pyrolysis alone.
0.0 600
700 800 o Temperature ( C)
900
Fig. 3. Theoretical and experimental nitrogen partitioning of HCN and NH3 from (A) CF/lignite and (B) PY-CF/lignite blends at different pyrolysis temperatures.
The experimental and theoretical nitrogen distributions from CF/lignite and PY-CF/lignite co-pyrolysis at different temperatures are shown in Fig. 4. For both blends, a monotonic trend was observed for the yields of char-N and (tar + N2)-N: char-N yield decreased while the (tar + N2)-N yield increased with the increasing temperature. Minimum and maximum yields of char-N and (tar + N2)-N were obtained at 900 °C pyrolysis: 37.8 and 56.9% for CF/lignite and 34.5 and 63.4% for PY-CF/lignite, respectively. While for (HCN + NH3)-N, its yield increased with the increasing pyrolysis temperature, reaching a maximum at a temperature of 800 °C for both CF/lignite and PY-CF/lignite blends (5.7 and 3.3% for CF/lignite and PY-CF/lignite, respectively) and with further increased temperature, the yields decreased to 5.3 and 2.1% for CF/lignite and PY-CF/ lignite, respectively. In general, the effects of the addition of CF and PY-CF on (HCN + NH3)-N yields were minimal compared to the yields of char-N and (tar + N2)-N. The obvious deviation between theoretical and experimental nitrogen distributions confirms that some interactions between each component of the blends do occur during co-pyrolysis of CF and PY-CF blended with lignite. For both CF/lignite and PY-CF/lignite blends, the experimental char-N yields were lower than the theoretical values over the entire experimental temperature range, indicating that the interactions decreased the nitrogen partitioning in char-N. In addition, the experimental (HCN + NH3)-N yields are lower than the theoretical values for both blends at all tested temperatures, implying that the interactions suppressed the formation of HCN + NH3 during co-pyrolysis. This observation is in agreement with previous report in that the addition of blue-green algae and water hyacinth to coal significantly
89
Z. Liu, R. Balasubramanian / Bioresource Technology 151 (2014) 85–90
75
A
CF/Lignite blend
Nitrogen distribution (%)
60
Char-N Char-N (HCN+NH3)-N (HCN+NH3)-N (Tar+N2)-N (Tar+N2)-N
Theoretical Experimental Theoretical Experimental Theoretical Experimental
45
30
15
et al., 2001; Wu and Ohtsuka, 1997b). As for the higher N2 yield from PY-CF/lignite pyrolysis, it most likely originated from the increased catalysis due to the higher content of these metals in PY-CF compared to those in CF. Given the low nitrogen content of biochar, reduced yields of HCN and NH3 and increased yield of N2 during pyrolysis, the blending of biochar with lignite may be an effective way to minimize emissions of nitrogen-containing pollutants. To further compare the effects of the addition of CF and PY-CF on nitrogen distributions, the percent change for the difference (PCD) between experimental and theoretical distributions was calculated according to the following equation (Eq. (2)):
PCD ð%Þ ¼
ðExperimental yield Theoretical yieldÞ 100 Theoretical yield
ð2Þ
0 600
B
700 800 o Temperature ( C)
PY-CF/Lignite blend
Nitrogen distribution (%)
80
900
Char-N Char-N (HCN+NH3)-N (HCN+NH3)-N (Tar+N2)-N (Tar+N2)-N
Theoretical Experimental Theoretical Experimental Theoretical Experimental
60
40
20
0 600
700 800 o Temperature ( C)
900
Fig. 4. Theoretical and experimental nitrogen distributions of (A) CF/lignite and (B) PY-CF/lignite blends at different pyrolysis temperatures.
decreased the yields of HCN and NH3 during co-pyrolysis (Yuan et al., 2011). As for the (tar + N2)-N, its experimental yields were higher than theoretical values for both blends pyrolysis at all tested temperature, especially for PY-CF/lignite. Although the tar was difficult to quantify, visual observation of the condensates inside the quartz tube, however, suggested that less tar was formed with the increasing temperature for both blends and the PY-CF/lignite produced less tar than the CF/lignite under identical pyrolysis conditions. Therefore, one can conclude that more nitrogen in the PYCF/lignite blend was converted into harmless N2 at high temperatures. It has been reported that Fe, Ca and K can promote the conversion from fuel-N to N2. Therefore, the increased N2 at elevated temperatures could be ascribed to the catalytic effect of these metals in the blends in the present study (as shown in Table 1, the lignite was rich in Ca and Fe, and the PY-CH contained considerable amount of K) (Mori et al., 1996; Ohshima et al., 2012; Tsubouchi
As shown in Table 2, as a direct comparison, the deviations were less for CF/lignite compared to PY-CF/lignite at all pyrolysis temperatures, indicating increased interactions occurred between PYCF and lignite than those between CF and lignite during co-pyrolysis. The negative PCD values of char-N and (HCN + NH3)-N for CF/ lignite and PY-CF/lignite reveal that the addition of both raw biomass and the biochar decreased the nitrogen partitioning of char-N and (HCN + NH3)-N. The deviation increased with increasing pyrolysis temperature for char-N and (HCN + NH3)-N from PY-CF/lignite pyrolysis while for CF/lignite pyrolysis, only the deviation of char-N increased with the increasing temperature. Compared to theoretical char-N yields, the PCD values increased from 12.2 to 30.0% for PY-CF/lignite and 9.7 to 21.3% for CF/lignite, respectively. The interactions within CF/lignite and PY-CF/lignite during combustion processes were also studied and similar interactions were observed between CF and lignite, and PY-CF and lignite (Liu and Balasubramanian, 2013a). However, it is clear that more significant interactions occurred between PY-CF and lignite than that between CF and lignite during co-pyrolysis in this study. Therefore, different interactions are present between biochar and lignite concerning their thermal conversion and the nitrogen partitioning of the blend. Moreover, the difference also implies the complex nature of interactions and the overall interactions require further investigations. The above analysis confirmed that the significant interactions resulting from the addition of PY-CF to lignite suppressed the conversion from fuel-N to HCN and NH3 (the precursors of NOx and N2O), promoted the fuel-N partitioning of harmless N2, indicating the additional environmental benefits from pyrolysis of biochar/ lignite than that of raw biomass/lignite. 4. Conclusion HCN was the dominant nitrogen containing pollutant from raw biomass pyrolysis while the yield of NH3 was slightly higher than that of HCN from biochar pyrolysis. Compared to raw biomass, the biochar produced less HCN and NH3 and a higher percent of nitrogen retained in the char. Synergistic interactions were observed within both raw biomass/lignite and biochar/lignite blends, resulting in decreased char-N and (HCN + NH3)-N yields, and
Table 2 The deviations between theoretical and experimental values of nitrogen distributions for CF/lignite and PY-CF/lignite pyrolysis. CF/lignite
PY-CF/lignite
Temperature (°C)
600
700
800
900
600
700
800
900
Char-N (HCN + NH3)-N (Tar + N2)-N
9.7 6.1 37.7
16.1 8.8 39.5
19.1 20.7 37.7
21.3 8.3 23.2
12.2 23.5 81.8
16.6 30.9 62.5
20.1 46.3 53.2
30.0 59.5 39.1
90
Z. Liu, R. Balasubramanian / Bioresource Technology 151 (2014) 85–90
promoted N2 formation. More significant interactions occurred between biochar and lignite, implying that enhanced environmental benefits can be achieved by using biochar to co-process with lignite than raw biomass. Acknowledgement This work was financially supported by M3TC (Minerals, Materials and Metals Technology Centre), NUS. References Chen, W.H., Chen, C.J., Huang, C., Shen, C.H., Hsu, H.W., 2013. A comparison of gasification phenomena among raw biomass, torrefied biomass and coal in an entrained-flow reactor. Appl. Energy 112, 421–430. Fisher, E.M., Dupont, C., Darvell, L.I., Commandre, J.M., Saddawi, A., Jones, J.M., Grateau, M., Nocquet, T., Salvador, S., 2012. Combustion and gasification characteristics of chars from raw and torrefied biomass. Bioresour. Technol. 119, 157–165. Glarborg, P., Jensen, A.D., Johnsson, J.E., 2003. Fuel nitrogen conversion in solid fuel fired systems. Prog. Energy Combust. Sci. 29, 89–113. Hansson, K.M., Samuelsson, J., Tullin, C., Amand, L.E., 2004. Formation of HNCO, HCN, and NH3 from the pyrolysis of bark and nitrogen-containing model compounds. Combust. Flame 137, 265–277. Haykiri-Acma, H., Yaman, S., 2007. Synergy in devolatilization characteristics of lignite and hazelnut shell during co-pyrolysis. Fuel 86, 373–380. Ibrahim, R.H.H., Darvell, L.I., Jones, I.M., Williams, A., 2013. Physicochemical characterisation of torrefied biomass. J. Anal. Appl. Pyrolysis 103, 21–30. Jones, J.M., Kubacki, M., Ross, A.B., Williams, A., 2005. Devolatilisation characteristics of coal and biomass blends. J. Anal. Appl. Pyrolysis 74, 502–511. Khan, A.A., Jong, W., Jansens, P.J., Spliethoff, H., 2009. Biomass combustion in fluidized bed boilers: potential problems and remedies. Fuel Process. Technol. 90, 21–50. Kilpinen, P., Hupa, M., 1991. Homogeneous N2O chemistry at fluidized bed combustion conditions: a kinetic modeling study. Combust. Flame 85, 94–104. Kwong, P.C.W., Chao, C.Y.H., Wang, J.H., Cheung, C.W., Kendall, G., 2007. Cocombustion performance of coal with rice husks and bamboo. Atmos. Environ. 41, 7462–7472. Liu, Z., Balasubramanian, R., 2013a. A comparison of thermal behaviors of raw biomass, corresponding biochar and their blends with lignite. Bioresour. Technol. 111, 201–210. Liu, Z., Balasubramanian, R., 2013b. Upgrading of waste biomass by hydrothermal carbonization (HTC) and low temperature pyrolysis (LTP): a comparative evaluation. Appl. Energy. http://dx.doi.org/10.1016/j.apenergy.2013.06.027.
Liu, Z., Quek, A., Parshetti, G., Jain, A., Srinivasan, M.P., Hoekman, S.K., Balasubramanian, R., 2013a. A study of nitrogen conversion and polycyclic aromatic hydrocarbon (PAH) emissions during hydrochar-lignite co-pyrolysis. Appl. Energy 108, 74–81. Liu, Z., Quek, A., Hoekman, S.K., Balasubramanian, R., 2013b. Production of solid biochar fuel from waste biomass by hydrothermal carbonization. Fuel 103, 943–949. Lu, K.M., Lee, W.J., Chen, W.H., Lin, T.C., 2013. Thermogravimetric analysis and kinetics of co-pyrolysis of raw/torrefied wood and coal blends. Appl. Energy 105, 57–65. Moon, C., Sung, Y., Ahn, S., Kim, T., Choi, G., Kim, D., 2013. Effect of blending ratio on combustion performance in blends of biomass and coals of different ranks. Exp. Thermal Fluid Sci. 47, 232–240. Mori, H., Asami, K., Ohtsuka, Y., 1996. Role of iron catalyst in fate of fuel nitrogen during coal pyrolysis. Energy Fuels 10, 1022–1027. Ohshima, Y., Tsubouchi, N., Ohtsuka, Y., 2012. Iron-catalyzed nitrogen removal as N2 from PAN-derived activated carbon. Appl. Catal. B: Environ. 111–112, 614–620. Ohtsuka, Y., Wu, Z., Furimsky, E., 1997. Effect of alkali and alkaline earth metals on nitrogen release during temperature programmed pyrolysis of coal. Fuel 76, 1361–1367. Park, D.K., Kim, S.D., Lee, S.H., Lee, J.G., 2010. Co-pyrolysis characteristics of sawdust and coal blend in TGA and a fixed bed reactor. Bioresour. Technol. 101, 6151– 6156. Park, S.W., Jang, C.H., Baek, K.R., Yang, J.K., 2012. Torrefaction and low-temperature carbonization of woody biomass: evaluation of fuel characteristics of the products. Energy 45, 676–685. Phanphanich, M., Mani, S., 2011. Impact of torrefaction on the grindability and fuel characteristics of forest biomass. Bioresour. Technol. 102, 1246–1253. Ren, S., Lei, H., Wang, L., Bu, Q., Chen, S., Wu, J., Julson, J., Ruan, R., 2013. The effect of torrefaction on compositions of bio-oil and syngas from biomass pyrolysis by microwave heating. Bioresour. Technol. 135, 659–664. Tian, F.J., Yu, J.L., Mckenzie, L.J., Hayashi, J.I., Li, C.Z., 2007. Conversion of fuel-N into HCN and NH3 during pyrolysis and gasification in steam: a comparative study of coal and biomass. Energy Fuels 21, 517–521. Tsubouchi, N., Ohshima, Y., Xu, C., Ohtsuka, Y., 2001. Enhancement of N2 formation from the nitrogen in carbon and coal by calcium. Energy Fuels 15, 158–162. Vuthaluru, H.B., 2004. Investigations into the pyrolytic behaviour of coal/biomass blends using thermogravimetric analysis. Bioresour. Technol. 92, 187–195. Wu, Z., Ohtsuka, Y., 1997a. Nitrogen distribution in a fixed bed pyrolysis of coals with different ranks: formation and source of N2. Energy Fuels 11, 477–482. Wu, Z., Ohtsuka, Y., 1997b. Key factors for formation of N2 from low-rank coals during fixed bed pyrolysis: pyrolysis conditions and inherent minerals. Energy Fuels 11, 902–908. Yuan, S., Zhou, Z., Li, J., Chen, X., Wang, F., 2010. HCN and NH3 released from biomass and soybean cake under rapid pyrolysis. Energy Fuels 24, 6166–6171. Yuan, S., Chen, X.L., Li, W., Liu, H., Wang, F., 2011. Nitrogen conversion under rapid pyrolysis of two types of aquatic biomass and corresponding blends with coal. Bioresour. Technol. 102, 10124–10130.
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
A comparative study of nitrogen conversion during pyrolysis of coconut fiber, its corresponding biochar and their blends with lignite Zhengang Liu, Rajasekhar Balasubramanian ⇑ Department of Civil and Environmental Engineering, National University of Singapore, Singapore 117576, Singapore
h i g h l i g h t s Coconut fiber (CF), its biochar and their blends with lignite were pyrolyzed. CF and its biochar showed different nitrogen partitioning between HCN and NH3. Interactions between CF/biochar and lignite decreased HCN and NH3 yields. Less HCN+NH3 and more N2 were generated from biochar/lignite pyrolysis. Enhanced benefits can be achieved from biochar/lignite co-processing.
a r t i c l e
i n f o
Article history: Received 19 August 2013 Received in revised form 12 October 2013 Accepted 14 October 2013 Available online 22 October 2013 Keywords: Nitrogen distribution Raw biomass Biochar Synergistic interactions
a b s t r a c t In the present study, the conversion of fuel-N to HCN and NH3 was investigated during rapid pyrolysis of raw biomass (coconut fiber), its corresponding biochar and their blends with lignite within a temperature range of 600–900 °C. The results showed that the raw biomass and the biochar showed totally different nitrogen partitioning between NH3 and HCN. HCN was the dominant nitrogen pollutant from pyrolysis of raw biomass, while for the biochar pyrolysis the yield of NH3 was slightly higher than that of HCN. Synergistic interactions occurred within both raw biomass/lignite and biochar/lignite blends, especially for the biochar/lignite blend, and resulted in reduced yields of HCN and NH3, decreased the total nitrogen percentage retained in the char and promoted harmless N2 formation. These findings suggest that biochar/lignite co-firing for energy production may have the enhanced benefit of reduced emissions of nitrogen pollutants than raw biomass/lignite. Ó 2013 Elsevier Ltd. All rights reserved.
1. Introduction Biomass and coal co-firing for heat and power production in existing coal-fired power plants is viewed as a key strategy of combining the utilization of biomass energy with the substantial reduction of CO2 discharged into the atmosphere. In recent years, extensive studies associated with biomass and coal co-processing have been carried out. These studies have been mostly focused on thermal conversion characteristics of biomass/coal co-processing systems (Haykiri-Acma and Yaman, 2007; Jones et al., 2005; Kwong et al., 2007; Lu et al., 2013; Park et al., 2010; Vuthaluru, 2004). For example, the addition of sawdust to coal changed the pyrolysis product streams significantly, resulting in decreased char and tar yields, and increased gas product yield (Park et al., 2010). The pyrolysis of three different types of coal in the presence of pinewood decreased aromatics and increased phenols in the resultant pyrolysis oil (Jones et al., 2005). Compared with coal combus⇑ Corresponding author. Tel.: +65 65165135; fax: +65 67744202. E-mail address: [email protected] (R. Balasubramanian). 0960-8524/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biortech.2013.10.043
tion alone, the biomass/coal co-firing process had lowered combustion temperature and reduced energy output (Kwong et al., 2007). However, raw biomass is not an ideal solid fuel, and large-scale applications of biomass/coal co-firing are restricted by inherent properties of biomass materials such as poor grindability and low energy density (Ibrahim et al., 2013; Khan et al., 2009). The poor grindability, caused by the fibrous structure of biomass feedstock, creates a challenge for fuel preparation and feeding to the process unit. The low energy density resulting from high oxygen content and high moisture content leads to significant costs for the transportation and storage. It is, therefore, necessary to enhance the fuel quality of raw biomass prior to co-processing with coal. Low temperature pyrolysis is one of attractive technologies to upgrade biomass feedstocks. The resultant biochar exhibits improved fuel quality compared to raw biomass including elevated energy density and increased grindability, addressing the key problems that hinder the use of raw biomass as a direct fuel (Fisher et al., 2012; Ibrahim et al., 2013; Park et al., 2012). As an example, energy consumption for the grinding of torrefied biomass was
86
Z. Liu, R. Balasubramanian / Bioresource Technology 151 (2014) 85–90
significantly decreased compared to that of untreated biomass (Phanphanich and Mani, 2011). In addition, with the improved physicochemical properties, favorable thermal conversion characteristics have been observed by the use of biochar. For example, the gasification of pyrolytic biochar produced more H2 and CO than that of raw biomass (Chen et al., 2013). Compared to raw biomass pyrolysis, H2 and CH4 were enriched in the syngas, and the increased concentrations of phenols and sugars and reduced concentrations of guaiacols and furans were reported from the pyrolysis of torrefied biomass (Ren et al., 2013). Apart from thermal conversion efficiency, as a direct fuel, emission of gaseous pollutants from the combustion of raw biomass materials is also one of the key issues in the context of developing advanced clean and renewable energy technology (Khan et al., 2009). Pyrolysis is the initial step for the combustion; therefore, understanding nitrogen conversion during pyrolysis is essential to efficiently control the formation of nitrogen pollutants, NOx and N2O, from subsequent combustion. Some studies have reported total nitrogen distribution in product streams during individual biomass and coal pyrolysis. It appears that nitrogen conversion depends on many factors including pyrolysis conditions and starting biomass feedstocks (Glarborg et al., 2003). For example, due to high contents of lignin in lignocellulosic biomass, much more HCN than NH3 was formed from the pyrolysis of lignocellulosic biomass than aquatic biomass (Yuan et al., 2010, 2011). Coal pyrolysis indicated that the coal type played an important role in the distribution of nitrogen among HCN, NH3 and N2, and Fe-containing materials served as a catalyst to promote N2 formation (Wu and Ohtsuka, 1997a). As for biomass/coal co-pyrolysis, synergistic interactions between biomass and coal were reported, leading to decreased char-N yield and increased volatile-N yield. However, the total amount of NH3 and HCN, in volatile decreased from biomass/coal co-pyrolysis (Yuan et al., 2010, 2011). It was also reported that the addition of hydrothermally pre-treated biomass in lignite suppressed the formation of HCN and NH3, and promoted the nitrogen conversion into harmless N2 at temperatures P800 °C (Liu et al., 2013a). A number of studies have been reported on nitrogen conversion during coal pyrolysis, while studies dealing with nitrogen conversion during raw biomass pyrolysis alone and raw biomass/coal copyrolysis are relatively sparse (Hansson et al., 2004; Glarborg et al., 2003; Tian et al., 2007; Yuan et al., 2010, 2011). It was reported that the interactions occurring between biomass feedstocks and coal are associated with the similarity of their fuel ranks during raw biomass and coal co-pyrolysis (Moon et al., 2013). With the enhanced fuel quality, the interactions occurring between biochar and coal are expected to differ from those between raw biomass and coal, leading to different nitrogen containing emissions from raw biomass pyrolysis (Moon et al., 2013). With the increasing importance of biomass energy, large amounts of biochar will be produced in the near future. To the best of our knowledge, no information is currently available for the formation and relative distributions of HCN and NH3 during pyrolysis of biochar and its blends with coal. In the present study, the conversion of fuel-N was investigated during pyrolysis of raw biomass, its corresponding biochar and their blends with lignite to assess the environmental benefits of biochar/coal co-processing.
2. Methods 2.1. Materials Coconut fiber (CF) was selected as a representative biomass due to its abundance in tropical areas in the present study. The biochar was produced from typical pyrolysis upgrading of raw biomass and
detailed procedures were reported elsewhere (Liu and Balasubramanian, 2013a). Briefly, CF was heated from room temperature to 350 °C under an inert atmosphere. A temperature of 350 °C was chosen because previous experiments have shown it to be suitable for biochar production as solid fuel (Liu and Balasubramanian, 2013a; Park et al., 2012). After maintaining at 350 °C for 20 min, the reactor was cooled down to room temperature. The biochar was recovered as a solid residue and labeled as PY-CF. The lignite sample was obtained from Pasar, Indonesia. All the fuel samples were dried at 105 °C for 24 h and crushed to less than 125 lm for the subsequent pyrolysis. The fuel blends were prepared using lignite weight fraction of 75%. Table 1 presents ultimate and proximate analysis results of each fuel used in the present study. 2.2. Pyrolysis A horizontal fixed-bed tubular quartz reactor (60 mm i.d., 1600 mm long) connected to a gas trap was used for the pyrolysis experiments. In each run, about 0.6 g of the fuel sample was loaded into an alumina boat that was then placed inside the quartz tube. Argon gas (flow rate 100 ml/min) was introduced into the pyrolysis system to maintain an inert atmosphere around the fuel sample. After the furnace was heated to the pre-determined temperature, the fuel sample was positioned within the heated area by pushing the quartz tube further into the furnace. The fuel sample was pyrolyzed at the preset temperature (600–900 °C) for 20 min, followed by keeping it out of the heating area and then cooling it down to the room temperature. Gaseous HCN and NH3 produced during pyrolysis were collected by absorption in 20 mmol/L NaOH and 2 mmol/L HNO3 solutions in gas traps, respectively. Separate runs were carried out for collecting NH3 and HCN produced because HCN and NH3 have non-negligible solubilities in HNO3 and NaOH solutions, respectively. Three replicates of each pyrolysis experiment were performed to ensure reproducibility of the emissions. 2.3. Quantification The carbon, hydrogen, sulfur and nitrogen contents were determined using a Vario MACRO cube Elemental Analyzer (USA). As for the metal content analysis, the fuel sample was digested in mixed acid solutions and the metal concentration (K, Na, Ca and Fe) in the solution was quantified by ICP-OES Perkin Elmer 3000DV (USA). The char fraction that remained in the alumina boat was weighted and its nitrogen content (char-N) was determined. The quantity of NH3 was measured using a Flow Injection Analyzer (FIA)
Table 1 Ultimate analysis, proximate analysis and major metal content analysis of CF, PY-CF and lignite. Fuel
CF
PY-CF
Lignite
Ultimate analysis (dry ash free, wt%) C 47.75 H 5.61 N 0.90 S 0.23 O (By difference) 45.51
67.51 3.95 1.01 0.37 27.16
61.64 5.72 1.74 0.77 30.13
Proximate analysis (dry base, wt%) Volatile matter 80.69 Fixed carbon 10.26 Ash 9.05
30.02 60.81 9.18
48.06 40.68 9.26
3.13 0.69 0.24 0.04
0.04 2.86 0.10 0.59
Major metal content analysis (dry base, wt%) K 1.61 Ca 0.34 Na 0.12 Fe 0.02
Z. Liu, R. Balasubramanian / Bioresource Technology 151 (2014) 85–90
(Quickchem 8500, Lachat Instrument). For HCN, its concentration was determined using an ion chromatograph (ICS-2000, Dionex) equipped with an amperometer detector (An IonPac AS7 Column was used and the eluent was CH3COONa (100 mmol/l) with the flow rate of 10 ll/min). The yields of char-N, HCN and NH3 were calculated as a percentage of total nitrogen in the fuel sample. 3. Results and discussion We have recently reported the nitrogen conversion during the pyrolysis of lignite within a temperature range of 600–900 °C (Liu et al., 2013a). This work focused on the comparative evaluation of CF and PY-CF, and the influences of their additions to lignite on the total nitrogen distribution during co-pyrolysis. 3.1. Pyrolysis of individual CF and PY-CF HCN and NH3 are commonly believed to be two important precursors of NOx and N2O resulting from solid fuel combustion; therefore, it is very important to determine the yields of HCN and NH3 from CF and PY-CF pyrolysis to achieve a cleaner combustion. As can be seen from Fig. 1, CF and PY-CF showed totally different patterns of total nitrogen partitioning between HCN and NH3. HCN was the dominant nitrogen pollutant from the pyrolysis of CF, and the yield of HCN was much higher than that of NH3, especially at low temperatures. For example, HCN and NH3 accounted for 8.5 and 2.8% of total nitrogen at 600 oC pyrolysis, respectively. In contrast, NH3 was the main pollutant from PY-CF pyrolysis and the difference of the yields between HCN and NH3 was only marginal at entire temperature range, similar to that of lignite pyrolysis (Liu et al., 2013a). The yields of HCN and NH3 were 2.4 and 2.5% at 600 °C, and 1.8 and 1.9% at 900 °C from PY-CF pyrolysis. In addition, the temperature played an important role in the nitrogen partitioning between HCN and NH3 for both CF and PY-CF. The yields of HCN and NH3 decreased with the increasing pyrolysis temperatures for both CF and PY-CF with the exception of NH3 yield from CF pyrolysis. In the case of the NH3 yield from CF pyrolysis, it increased with the increasing temperatures, reaching a maximum of 3.2% at 800 °C and then its yield decreased with the further increasing temperatures up to 900 °C. In comparison to PY-CF, more HCN and NH3 were generated from CF pyrolysis at all tested temperatures. The most important source of N2O is believed to originate from oxidation of the gaseous HCN (Kilpinen and Hupa, 1991). Therefore, more NOx and N2O are expected from CF than those of PY-CF during subsequent combustion. 9.0 HCN HCN HCN NH3 NH3 NH3
Nitrogen distribution (%)
7.5
6.0
CF PY-CF Lignite CF PY-CF Lignite
4.5
3.0
1.5
0.0 600
700 800 o Temperature ( C)
900
Fig. 1. Nitrogen partitioning of HCN and NH3 from CF and PY-CF at different pyrolysis temperatures (lignite for comparison).
87
Nitrogen conversion during biomass pyrolysis is known to be closely related to the chemical composition of biomass materials (Yuan et al., 2010, 2011). From the structural analysis, cellulose and hemicellulose originally contained in CF were totally decomposed /carbonized and a part of lignin was also decomposed during pyrolysis upgrading (Liu et al., 2013b). Therefore, the different nitrogen conversions for CF and PY-CF can mainly be ascribed to different chemical compositions of CF and PY-CF. In addition, it was reported that alkali and alkaline earth metal (AAEM) can suppress the formation of HCN, and catalyze the conversion from fuelN to NH3 during coal pyrolysis (Ohtsuka et al., 1997). Our previous study revealed that all the metals originally contained in raw biomass remain in the biochar during biomass pyrolysis upgrading (Liu and Balasubramanian, 2013b). As a consequence, the AAEM contents in PY-CF are higher than those of CF as confirmed in Table 1 (generally the contents of Ca, K and Na in PY-CF are two times higher than those in CF). Therefore enriched AAEM in PY-CF also partly contributes to different fuel-N partitioning of HCN and NH3 during pyrolysis compared to CF. Lastly, HCN is believed to be mainly generated from the thermal cracking of unstable N-containing compounds in the volatiles while NH3 is formed from more stable N-containing compounds in the nascent char during pyrolysis (Hansson et al., 2004; Yuan et al., 2010). Due to sharply lower volatile content of PY-CF (80.69% and 30.02% for CF and PY-CF, respectively), less polymerization reactions among volatiles presumably occurred during the pyrolysis of PY-CF than that of CF, resulting in less HCN formed during PY-CF pyrolysis. Increased cavity in PY-CF also could lead to less polymerization, contributing to the low HCN yield from PY-CF pyrolysis. It is worth noting that the results obtained in this study are distinctly different from those from previous reports in which increased lignin contents in biomass resulted in the increased HCN yield from pyrolysis (Yuan et al., 2010, 2011). From the above analysis, it is known that PY-CF has higher lignin contents than CF. However, NH3 was the dominant nitrogen pollutant from pyrolysis of PY-CF, not HCN. These results support the conclusion made in our previous study: besides the total lignin content, the state of lignin in biomass materials is also important in governing the nitrogen partitioning of NH3 and HCN during pyrolysis (Liu et al., 2013a). In the present study, the total nitrogen in pyrolysis products was divided into three groups: char-N, (HCN + NH3)-N, (tar + N2)-N. The amount of (tar + N2)-N from each experiment was calculated by difference between the total fuel-N and the sum of char-N and (HCN + NH3)-N due to the difficulty of the collection and quantification of tar condensed inner side of tube. Fig. 2 shows the total fuel-N partitioning of each product group from CF and PY-CF pyrolysis. It was clear that more nitrogen retained in the char and less (tar + N2)-N was formed from PY-CF pyrolysis than those of CF pyrolysis. For both CF and PY-CF pyrolysis, the yields of char-N and (HCN + NH3)-N decreased with the increasing temperature and the total nitrogen was mainly in the form of tar + N2 at temperatures higher than 700 °C for CF and 900 °C for PY-CF, respectively. For PY-CF pyrolysis, the char-N yield decreased with the increasing temperature and dominated at the temperature range of 600–800 °C. The yields of (HCN + NH3)-N and (tar + N2)-N were minimal at temperatures 6700 °C. The combined amounts of these two groups only accounted for 11% and 18% of total nitrogen from pyrolysis at 600 and 700 °C, respectively. However, a sharp increase of (tar + N2)-N yield was observed when the pyrolysis temperature was higher than 800 °C. The yield of (tar + N2)-N increased from 5.9% to 26.8% with the temperature increase from 600 to 800 °C, and with a further increase to 48.3% at 900 °C. In the case of CF pyrolysis, similar effect of temperature on the nitrogen distribution was observed. When compared to PY-CF pyrolysis,
Z. Liu, R. Balasubramanian / Bioresource Technology 151 (2014) 85–90
Char-N Char-N (HCN+NH3)-N (HCN+NH3)-N (Tar+N2)-N (Tar+N2)-N
100
Nitrogen distribution (%)
80
5.0
CF PY-CF CF PY-CF CF PY-CF
4.0 Nitrogen distribution (%)
88
60
40
20
HCN HCN NH3 NH3
Theoretical Experimental Theoretical Experimental
A
CF/Lignite blend
3.0
2.0
1.0
0 600
700
800
0.0
900
600
o
Temperature ( C)
700 800 o Temperature ( C)
900
Fig. 2. Nitrogen distributions of CF and PY-CF at different pyrolysis temperatures.
3.2. Pyrolysis of CF/lignite and PY-CF/lignite blends
4.0
Nitrogen distribution (%)
the char-N yield was significantly lower while the yield of (HCN + NH3)-N was higher for CF pyrolysis at the entire temperature range. The less pollutant emissions together with improved combustion behaviors of PY-CF indicate that as a solid fuel, except higher thermal efficiency, the additional enhanced environmental benefits can also be achieved by using pyrolytic biochars compared to raw biomass.
HCN HCN NH3 NH3
B
Theoretical PY-CF/Lignite blend Experimental Theoretical Experimental
3.0
2.0
1.0
Fig. 3 presents the total nitrogen partitioning of HCN and NH3 from pyrolysis of CF/lignite and PY-CF/lignite blends. The yields of NH3 and HCN varied with the temperature for both blends. The addition of CF and PY-CF to lignite showed similar effects on HCN and NH3 yields from co-pyrolysis: the yields of HCN and NH3 increased initially with the increasing temperature reaching a maximum and then decreased with a further increase of pyrolysis temperature. For CF/lignite blend, the yields of HCN and NH3 reached their maximum at 800 °C, with 2.8 and 2.9% for HCN and NH3, respectively. In the case of PY-CF/lignite pyrolysis, HCN and NH3 reached their maximum yield at 700 °C (1.4%) and 800 °C (2.0%), respectively. Based on individual proportions of each fuel in the blends, with the assumption that there is no interaction between each component during co-pyrolysis, theoretical yields of HCN and NH3 from CF/lignite and PY-CF/lignite were calculated by the following equation (Eq. (1)):
YieldBlend ¼ 25% YieldCF=PY-CF þ 75% YieldLignite
ð1Þ
The experimental and theoretical yields of HCN and NH3 from the blends pyrolysis are shown in Fig. 3. PY-CF addition notably suppressed the formation of both HCN and NH3. Though less pronounced, similar trends were observed for the addition of CF to lignite. For PY-CF and lignite co-pyrolysis, the experimental yield of NH3 decreased by 29.6% and 67.8% at 600 and 900 °C, respectively (relative to the theoretical yields), and the yield of HCN decreased by 12.6% and 48.4% at 600 and 900 °C, respectively. In the case of CF and lignite co-pyrolysis, the experimental yield of NH3 decreased by 11.3% and 31.8% at 600 and 900 °C, respectively, and that of HCN decreased by 2.3% and 11.0% at 600 and 900 °C, respectively. The experimental yields of both HCN and NH3 were lower than those of theoretical values from CF/lignite and PY-CF/lignite pyrolysis, implying the environmental benefit of biomass/coal and biochar/coal co-pyrolysis than raw biomass, biochar and coal pyrolysis alone.
0.0 600
700 800 o Temperature ( C)
900
Fig. 3. Theoretical and experimental nitrogen partitioning of HCN and NH3 from (A) CF/lignite and (B) PY-CF/lignite blends at different pyrolysis temperatures.
The experimental and theoretical nitrogen distributions from CF/lignite and PY-CF/lignite co-pyrolysis at different temperatures are shown in Fig. 4. For both blends, a monotonic trend was observed for the yields of char-N and (tar + N2)-N: char-N yield decreased while the (tar + N2)-N yield increased with the increasing temperature. Minimum and maximum yields of char-N and (tar + N2)-N were obtained at 900 °C pyrolysis: 37.8 and 56.9% for CF/lignite and 34.5 and 63.4% for PY-CF/lignite, respectively. While for (HCN + NH3)-N, its yield increased with the increasing pyrolysis temperature, reaching a maximum at a temperature of 800 °C for both CF/lignite and PY-CF/lignite blends (5.7 and 3.3% for CF/lignite and PY-CF/lignite, respectively) and with further increased temperature, the yields decreased to 5.3 and 2.1% for CF/lignite and PY-CF/ lignite, respectively. In general, the effects of the addition of CF and PY-CF on (HCN + NH3)-N yields were minimal compared to the yields of char-N and (tar + N2)-N. The obvious deviation between theoretical and experimental nitrogen distributions confirms that some interactions between each component of the blends do occur during co-pyrolysis of CF and PY-CF blended with lignite. For both CF/lignite and PY-CF/lignite blends, the experimental char-N yields were lower than the theoretical values over the entire experimental temperature range, indicating that the interactions decreased the nitrogen partitioning in char-N. In addition, the experimental (HCN + NH3)-N yields are lower than the theoretical values for both blends at all tested temperatures, implying that the interactions suppressed the formation of HCN + NH3 during co-pyrolysis. This observation is in agreement with previous report in that the addition of blue-green algae and water hyacinth to coal significantly
89
Z. Liu, R. Balasubramanian / Bioresource Technology 151 (2014) 85–90
75
A
CF/Lignite blend
Nitrogen distribution (%)
60
Char-N Char-N (HCN+NH3)-N (HCN+NH3)-N (Tar+N2)-N (Tar+N2)-N
Theoretical Experimental Theoretical Experimental Theoretical Experimental
45
30
15
et al., 2001; Wu and Ohtsuka, 1997b). As for the higher N2 yield from PY-CF/lignite pyrolysis, it most likely originated from the increased catalysis due to the higher content of these metals in PY-CF compared to those in CF. Given the low nitrogen content of biochar, reduced yields of HCN and NH3 and increased yield of N2 during pyrolysis, the blending of biochar with lignite may be an effective way to minimize emissions of nitrogen-containing pollutants. To further compare the effects of the addition of CF and PY-CF on nitrogen distributions, the percent change for the difference (PCD) between experimental and theoretical distributions was calculated according to the following equation (Eq. (2)):
PCD ð%Þ ¼
ðExperimental yield Theoretical yieldÞ 100 Theoretical yield
ð2Þ
0 600
B
700 800 o Temperature ( C)
PY-CF/Lignite blend
Nitrogen distribution (%)
80
900
Char-N Char-N (HCN+NH3)-N (HCN+NH3)-N (Tar+N2)-N (Tar+N2)-N
Theoretical Experimental Theoretical Experimental Theoretical Experimental
60
40
20
0 600
700 800 o Temperature ( C)
900
Fig. 4. Theoretical and experimental nitrogen distributions of (A) CF/lignite and (B) PY-CF/lignite blends at different pyrolysis temperatures.
decreased the yields of HCN and NH3 during co-pyrolysis (Yuan et al., 2011). As for the (tar + N2)-N, its experimental yields were higher than theoretical values for both blends pyrolysis at all tested temperature, especially for PY-CF/lignite. Although the tar was difficult to quantify, visual observation of the condensates inside the quartz tube, however, suggested that less tar was formed with the increasing temperature for both blends and the PY-CF/lignite produced less tar than the CF/lignite under identical pyrolysis conditions. Therefore, one can conclude that more nitrogen in the PYCF/lignite blend was converted into harmless N2 at high temperatures. It has been reported that Fe, Ca and K can promote the conversion from fuel-N to N2. Therefore, the increased N2 at elevated temperatures could be ascribed to the catalytic effect of these metals in the blends in the present study (as shown in Table 1, the lignite was rich in Ca and Fe, and the PY-CH contained considerable amount of K) (Mori et al., 1996; Ohshima et al., 2012; Tsubouchi
As shown in Table 2, as a direct comparison, the deviations were less for CF/lignite compared to PY-CF/lignite at all pyrolysis temperatures, indicating increased interactions occurred between PYCF and lignite than those between CF and lignite during co-pyrolysis. The negative PCD values of char-N and (HCN + NH3)-N for CF/ lignite and PY-CF/lignite reveal that the addition of both raw biomass and the biochar decreased the nitrogen partitioning of char-N and (HCN + NH3)-N. The deviation increased with increasing pyrolysis temperature for char-N and (HCN + NH3)-N from PY-CF/lignite pyrolysis while for CF/lignite pyrolysis, only the deviation of char-N increased with the increasing temperature. Compared to theoretical char-N yields, the PCD values increased from 12.2 to 30.0% for PY-CF/lignite and 9.7 to 21.3% for CF/lignite, respectively. The interactions within CF/lignite and PY-CF/lignite during combustion processes were also studied and similar interactions were observed between CF and lignite, and PY-CF and lignite (Liu and Balasubramanian, 2013a). However, it is clear that more significant interactions occurred between PY-CF and lignite than that between CF and lignite during co-pyrolysis in this study. Therefore, different interactions are present between biochar and lignite concerning their thermal conversion and the nitrogen partitioning of the blend. Moreover, the difference also implies the complex nature of interactions and the overall interactions require further investigations. The above analysis confirmed that the significant interactions resulting from the addition of PY-CF to lignite suppressed the conversion from fuel-N to HCN and NH3 (the precursors of NOx and N2O), promoted the fuel-N partitioning of harmless N2, indicating the additional environmental benefits from pyrolysis of biochar/ lignite than that of raw biomass/lignite. 4. Conclusion HCN was the dominant nitrogen containing pollutant from raw biomass pyrolysis while the yield of NH3 was slightly higher than that of HCN from biochar pyrolysis. Compared to raw biomass, the biochar produced less HCN and NH3 and a higher percent of nitrogen retained in the char. Synergistic interactions were observed within both raw biomass/lignite and biochar/lignite blends, resulting in decreased char-N and (HCN + NH3)-N yields, and
Table 2 The deviations between theoretical and experimental values of nitrogen distributions for CF/lignite and PY-CF/lignite pyrolysis. CF/lignite
PY-CF/lignite
Temperature (°C)
600
700
800
900
600
700
800
900
Char-N (HCN + NH3)-N (Tar + N2)-N
9.7 6.1 37.7
16.1 8.8 39.5
19.1 20.7 37.7
21.3 8.3 23.2
12.2 23.5 81.8
16.6 30.9 62.5
20.1 46.3 53.2
30.0 59.5 39.1
90
Z. Liu, R. Balasubramanian / Bioresource Technology 151 (2014) 85–90
promoted N2 formation. More significant interactions occurred between biochar and lignite, implying that enhanced environmental benefits can be achieved by using biochar to co-process with lignite than raw biomass. Acknowledgement This work was financially supported by M3TC (Minerals, Materials and Metals Technology Centre), NUS. References Chen, W.H., Chen, C.J., Huang, C., Shen, C.H., Hsu, H.W., 2013. A comparison of gasification phenomena among raw biomass, torrefied biomass and coal in an entrained-flow reactor. Appl. Energy 112, 421–430. Fisher, E.M., Dupont, C., Darvell, L.I., Commandre, J.M., Saddawi, A., Jones, J.M., Grateau, M., Nocquet, T., Salvador, S., 2012. Combustion and gasification characteristics of chars from raw and torrefied biomass. Bioresour. Technol. 119, 157–165. Glarborg, P., Jensen, A.D., Johnsson, J.E., 2003. Fuel nitrogen conversion in solid fuel fired systems. Prog. Energy Combust. Sci. 29, 89–113. Hansson, K.M., Samuelsson, J., Tullin, C., Amand, L.E., 2004. Formation of HNCO, HCN, and NH3 from the pyrolysis of bark and nitrogen-containing model compounds. Combust. Flame 137, 265–277. Haykiri-Acma, H., Yaman, S., 2007. Synergy in devolatilization characteristics of lignite and hazelnut shell during co-pyrolysis. Fuel 86, 373–380. Ibrahim, R.H.H., Darvell, L.I., Jones, I.M., Williams, A., 2013. Physicochemical characterisation of torrefied biomass. J. Anal. Appl. Pyrolysis 103, 21–30. Jones, J.M., Kubacki, M., Ross, A.B., Williams, A., 2005. Devolatilisation characteristics of coal and biomass blends. J. Anal. Appl. Pyrolysis 74, 502–511. Khan, A.A., Jong, W., Jansens, P.J., Spliethoff, H., 2009. Biomass combustion in fluidized bed boilers: potential problems and remedies. Fuel Process. Technol. 90, 21–50. Kilpinen, P., Hupa, M., 1991. Homogeneous N2O chemistry at fluidized bed combustion conditions: a kinetic modeling study. Combust. Flame 85, 94–104. Kwong, P.C.W., Chao, C.Y.H., Wang, J.H., Cheung, C.W., Kendall, G., 2007. Cocombustion performance of coal with rice husks and bamboo. Atmos. Environ. 41, 7462–7472. Liu, Z., Balasubramanian, R., 2013a. A comparison of thermal behaviors of raw biomass, corresponding biochar and their blends with lignite. Bioresour. Technol. 111, 201–210. Liu, Z., Balasubramanian, R., 2013b. Upgrading of waste biomass by hydrothermal carbonization (HTC) and low temperature pyrolysis (LTP): a comparative evaluation. Appl. Energy. http://dx.doi.org/10.1016/j.apenergy.2013.06.027.
Liu, Z., Quek, A., Parshetti, G., Jain, A., Srinivasan, M.P., Hoekman, S.K., Balasubramanian, R., 2013a. A study of nitrogen conversion and polycyclic aromatic hydrocarbon (PAH) emissions during hydrochar-lignite co-pyrolysis. Appl. Energy 108, 74–81. Liu, Z., Quek, A., Hoekman, S.K., Balasubramanian, R., 2013b. Production of solid biochar fuel from waste biomass by hydrothermal carbonization. Fuel 103, 943–949. Lu, K.M., Lee, W.J., Chen, W.H., Lin, T.C., 2013. Thermogravimetric analysis and kinetics of co-pyrolysis of raw/torrefied wood and coal blends. Appl. Energy 105, 57–65. Moon, C., Sung, Y., Ahn, S., Kim, T., Choi, G., Kim, D., 2013. Effect of blending ratio on combustion performance in blends of biomass and coals of different ranks. Exp. Thermal Fluid Sci. 47, 232–240. Mori, H., Asami, K., Ohtsuka, Y., 1996. Role of iron catalyst in fate of fuel nitrogen during coal pyrolysis. Energy Fuels 10, 1022–1027. Ohshima, Y., Tsubouchi, N., Ohtsuka, Y., 2012. Iron-catalyzed nitrogen removal as N2 from PAN-derived activated carbon. Appl. Catal. B: Environ. 111–112, 614–620. Ohtsuka, Y., Wu, Z., Furimsky, E., 1997. Effect of alkali and alkaline earth metals on nitrogen release during temperature programmed pyrolysis of coal. Fuel 76, 1361–1367. Park, D.K., Kim, S.D., Lee, S.H., Lee, J.G., 2010. Co-pyrolysis characteristics of sawdust and coal blend in TGA and a fixed bed reactor. Bioresour. Technol. 101, 6151– 6156. Park, S.W., Jang, C.H., Baek, K.R., Yang, J.K., 2012. Torrefaction and low-temperature carbonization of woody biomass: evaluation of fuel characteristics of the products. Energy 45, 676–685. Phanphanich, M., Mani, S., 2011. Impact of torrefaction on the grindability and fuel characteristics of forest biomass. Bioresour. Technol. 102, 1246–1253. Ren, S., Lei, H., Wang, L., Bu, Q., Chen, S., Wu, J., Julson, J., Ruan, R., 2013. The effect of torrefaction on compositions of bio-oil and syngas from biomass pyrolysis by microwave heating. Bioresour. Technol. 135, 659–664. Tian, F.J., Yu, J.L., Mckenzie, L.J., Hayashi, J.I., Li, C.Z., 2007. Conversion of fuel-N into HCN and NH3 during pyrolysis and gasification in steam: a comparative study of coal and biomass. Energy Fuels 21, 517–521. Tsubouchi, N., Ohshima, Y., Xu, C., Ohtsuka, Y., 2001. Enhancement of N2 formation from the nitrogen in carbon and coal by calcium. Energy Fuels 15, 158–162. Vuthaluru, H.B., 2004. Investigations into the pyrolytic behaviour of coal/biomass blends using thermogravimetric analysis. Bioresour. Technol. 92, 187–195. Wu, Z., Ohtsuka, Y., 1997a. Nitrogen distribution in a fixed bed pyrolysis of coals with different ranks: formation and source of N2. Energy Fuels 11, 477–482. Wu, Z., Ohtsuka, Y., 1997b. Key factors for formation of N2 from low-rank coals during fixed bed pyrolysis: pyrolysis conditions and inherent minerals. Energy Fuels 11, 902–908. Yuan, S., Zhou, Z., Li, J., Chen, X., Wang, F., 2010. HCN and NH3 released from biomass and soybean cake under rapid pyrolysis. Energy Fuels 24, 6166–6171. Yuan, S., Chen, X.L., Li, W., Liu, H., Wang, F., 2011. Nitrogen conversion under rapid pyrolysis of two types of aquatic biomass and corresponding blends with coal. Bioresour. Technol. 102, 10124–10130.