- Email: [email protected]
Complementary effects of torrefaction and co-pelletization: Energy consumption and characteristics of pellets
Accepted Manuscript Complementary effects of torrefaction and co-pelletization: Energy consumption and characteristics of pellets Liang Cao, Xingzhong Yuan, Hui Li, Changzhu Li, Zhihua Xiao, Longbo Jiang, Binbin Huang, Zhihong Xiao, Xiaohong Chen, Hou Wang, Guangming Zeng PII: DOI: Reference:
S0960-8524(15)00224-2 http://dx.doi.org/10.1016/j.biortech.2015.02.045 BITE 14618
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
21 November 2014 9 February 2015 11 February 2015
Please cite this article as: Cao, L., Yuan, X., Li, H., Li, C., Xiao, Z., Jiang, L., Huang, B., Xiao, Z., Chen, X., Wang, H., Zeng, G., Complementary effects of torrefaction and co-pelletization: Energy consumption and characteristics of pellets, Bioresource Technology (2015), doi: http://dx.doi.org/10.1016/j.biortech.2015.02.045
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.
Complementary effects of torrefaction and co-pelletization : Energy consumption and characteristics of pellets Liang Cao a, b, c, Xingzhong Yuan a, b,*, Hui Li c,d,*, Changzhu Li c,d, Zhihua Xiao a, b, Longbo Jiang a, b
, Binbin Huang a, b, Zhihong Xiao d, Xiaohong Chen e, Hou Wang a, b, Guangming Zeng a, b
a College of Environmental Science and Engineering, Hunan University, Changsha 410082, PR China b Key Laboratory of Environmental Biology and Pollution Control (Hunan University), Ministry of Education, Changsha 410082, PR China c Institute of Biological and Environmental Engineering, Hunan Academy of Forestry, Changsha 410004, PR China d Institute of Bio-energy, Hunan Academy of Forestry, Changsha 410004, PR China e School of Business, Central South University, Changsha 410083, P.R. China
* Corresponding author. Tel.: + 86-731-88821413；Fax：+ 86-731-88821413.E-mail address: [email protected] (X.Z. Yuan). Tel.: +86-731-85578765；Fax： +86-731-85578765.E-mail address: [email protected] (H. Li). 1
Abstract In this study, compelmentary of torrefaction and co-pelletization for biomass pellets production was investigated. Two kinds of biomass materials were torrefied and mixed with oil cake for co-pelletization. The energy consumption during pelletization and pellet characteristics including moisture absorption, pellet density, pellet strength and combustion characteristic, were evaluated. It was shown that torrefaction improved the characteristics of pellets with high heating values, low moisture absorption and well combustion characteristic. Furthermore, co-pelletization between torrefied biomass and cater bean cake can reduce several negative effects of torrefaction such as high energy consumption, low pellet density and strength. The optimal conditions for energy consumption and pellet strength were torrefied at 270 o
C and a blending with 15% castor bean cake for both biomass materials. The present
study indicated that compelmentary performances of the torrefaction and co-pelletization with castor bean cake provide a promising alternative for fuel production from biomass and oil cake. Keywords: Oil cake, Torrefaction, Co-pelletization, Energy consumption, Combustion
2
1. Introduction Biomass pellets have become a sustainable and environmental energy resource as its attractive characteristics, such as renewable, none net CO2 emission and low production cost (Du et al., 2014; Jiang et al., 2014; Nunes et al., 2014; Xu et al., 2014)). In general, pelletization of biomass makes its density 4 to10 times higher than those without pelletization which can significantly reduce the transportation/storage costs (Li et al., 2012; Liu et al., 2014). However, due to the high moisture absorption, low energy density and rapid combustion reactivity of biomass, all of those significantly hinder the large-scale application of biomass pellets (Chin et al., 2013; Zhou, C., 2014). Therefore, the advancements of the properties of biomass pellets have gained widespread attention. Among the technologies for improving the properties of biomass, torrefaction was regarded as one of the efficient and feasible method (Chen et al., 2014; Du et al., 2014; Li et al.,). Wen et al.(2014) reported that pretreatment of torrefaction is a pyrolytic process at 200-300 oC in an inert atmosphere. It is well known that torrefied biomass is a material characterized by hydrophobicity, grindability and high heat value because of the removal of hemicellulose, lignin, protein, starch and fatty compounds in biomass (Chen et al., 2014; Li et al., 2012; Mišljenovic et al., 2014). Moreover, the combustion performance of pellets produced from torrefied biomass approaches to that of coal, which can be due to the removal of free water and volatile components during torrefaction (Du et al., 2014). However, compaction and extrusion energy consumptions during pelletization were remarkably augmented for the 3
torrefied biomass materials, while the density and strength of the pellets made from torrefied biomass were decreased compared with those of untreated biomass (Li et al., 2012). Different methods and other various torrefaction and pelletization conditions were adopted in order to resolve the disadvantages mentioned above (Chen et al., 2014; Du et al., 2014; Li et al., 2012). Among the existing technologies, co-pelletization process has become a widespread used technology to improve the characteristics of pellet (Stahl et al., 2011; Shang et al., 2013;Jiang et al., 2014; Nunes et al., 2014). The major advantage of co-pelletization is to eliminate the negative properties of pellets with the aid of additives. For example, Jiang et al. (2014) investigated that biomass mixed with sludge can enhance the pellet density and strength as the protein in sludge can act as a binder between biomass particles. Stahl et al. (2011) studied that energy consumption decreased with an increasing content of rapeseed cake, which was due to the reduction of friction during pelletization caused by the presence of rape oil in rapeseed cake. Zhou et al. (2014) managed to improve the combustion progress by blending biomass with coal to increase the content of fixed carbon. Therefore, several studies have investigated the effects of torrefaction and co-pelletizing on the pellet properties. It was carried out by Reza et al. (2014) that pellets produced from dry torrefied and hydrothermal carbonization (HTC) biomass blends possessed higher pellets density and better durability than those from torrefied biomass. Shang et al., (2013) investigated the blending of rapeseed oil with torrefied biomass can reduce the energy consumption during pelletization (Shang et al., 2013). However, no further study was 4
carried out to reveal the potential effect of rapeseed oil on pellets’ strength (Shang et al., 2013). In addition, the higher price of additives compared with conventional biomass resulting in the economic infeasibility of biomass were seldom considered (Ahn et al., 2014b; Chau et al., 2009; Reza et al., 2014). In this respect, more attention should be paid on the low-cost additives during co-pelletization. In China, oil cake is a main byproduct in the oil expression process with a total annual amount of 16-20 Mt. Nevertheless, large amount of oil cake carries hazardous substances such as sinapic acid, saponin and tannins; and poor palatability and nutritional imbalance make it difficult to be utilized as animals feed. Meanwhile, several specific features of oil cake, such as the higher energy content and functional components (starch, protein and bio-oil, etc.) in oil cake reveal that oil cake could be utilized as fuel and additives in the co-pelletization with torrefied biomass. To our best knowledge, few studies are available on the complementary effects of torrefaction and co-pelletization of oil cake on energy consumption and pellets characteristics. In this study, two kinds of biomass (cedarwood and camphorwood) were pretreated at different torrefaction temperature (240 oC, 270 oC, and 300 oC), respectively. Caster bean cake, one kind of high-production oil cakes in East Asia, was selected as the additive. The torrefied biomass materials were blended with caster bean cake for co-pelletization. Subsequently, the effects of torrefaction and caster bean cake on the energy consumption and pellet properties were investigated. The primary objective of this study was to reveal the compelmentary mechanisms between torrefaction and co-pelletization for biomass pellet preparation. 5
2. Materials and methods 2.1Materials Two types of wood sawdust were selected in the present study: cedarwood (CED) and camphorwood (CAM), which were collected from local forest in Changsha and further grounded into fractions with particle size below 1 mm (Table 1). The selected oil cake was castor bean cake (CAS) obtained from the expression process of castor bean in a pilot of Hunan Academy of Forestry. The selected materials were dried (40 o
C, 48 hours) and stored in plastic containers at 4 oC for further study.
2.2 Torrefaction The torrefaction reactor is a tubular furnace which belongs to a bench-scale fixed bed reactor (Fig.1). The reactor contains a tubular unit, an electrical heater and a gas supply. The biomass was heated up to the desired temperature with a rate of 10 o
C/min-1 and a nitrogen flow of 100 mL/min-1 as a carrier gas was used during
torrefaction. The set torrefaction conditions were 240 oC (light torrefaction), 270 oC (medium torrefaction) and 300 oC (heavy torrefaction) with a residence time of 30 min, respectively. Table 1 records the weight loss and particle size of the torrefied samples at different temperature with respect to raw samples. 2.3 Co-pelletization The pellets were prepared using a DWD-10 pressure unit installed with a cylinder die and a piston (Fig.1). The cylinder die encircled with a heating tap had an opening of 7 mm in diameter and 120 mm in length. The end of the die was a removable plug. The piston with sizes of 6.8 mm in diameter and 80 mm in length 6
was installed to supply the pressure for the sample. In this work, the additive (CAS) was added into raw and torrefied samples with the proportion of 0, 15 and 30%, respectively. The loaded sample (around 0.7 g) was compressed with a rate of 2 mm/min until the desired pressure was achieved. The maximum pressure was 4000 N with a residence time of 30 s. Prior to pelletization, the cylinder die was preheated up to 150
o
C. The transducer was used to measure the compression force and
displacement and the relevant data were recorded by a computer. Compression and extrusion energy consumption were calculated by the formulation as following. (1)
= ∑ ∙
Where EC is the energy consumption (J), F is the compression force (N), S is the displacement (mm). 2.4 Moisture absorption Prior to the moisture absorption, the pellets were dried at 105 oC for 48 hours. The tests were carried out in a humidity chamber (30 oC and 90% relative humidity). The corresponding weights of the pellets were determined every 15 min for the first five hours. The equilibrium moisture content was measured after 2 days. Two pellets were carried out in this analysis. The kinetics of moisture absorption is presented by the ASABE formulation as following (Lam et al., 2012): (M-Me)/(Mi-Me)=e-k t
(2)
Where M is the instantaneous moisture content (decimal, dry basis), Me is the equilibrium moisture content (decimal, dry basis), and Mi is the initial moisture content (decimal, dry basis). The coefficient k is an absorption constant (min-1), and t 7
is the exposure time (min). 2.5 Pellets strength The pellets strength was also measured by the DWD-10 pressure unit. The strength was determined by the force crushing the pellet to the length of pellet and the depth of indentation as described in an earlier work (Li et al., 2012). Each pellet was horizontally laid down on a flat surface. An increasing force was applied at the center of the pellet until crushing. Three pellets were applied to observe the representative data. The pellet strength was determined using the pattern of Meyer hardness tester (L.G. Tabil et al., 2002). The Meyer hardness (MS), defined as the static indentation strength, was calculated using the following equation (L.G. Tabil et al., 2002): MS=F/[π(Dh-h2)]
(3)
Where MS is Meyer hardness (N/mm2); F is maximum pressure (N); D is pellet diameter (mm) and h is the depth of indentation (mm). 2.6 Thermogravimetric analysis (TG) Thermogravimetric analysis was performed using the TG-60 analyzer (JAP) in order to evaluate the pellets’ combustion characteristic. The analyzer can provide continuous recording of the weight loss (TG) and the rate of weight loss (DTG) curves. In this study, around 5mg of sample (shown in Table 1) loaded into the platinum crucible was placed inside the thermal analyzer. The experiment run was carried out at heating rates of 10 oC/min with temperature ranging from 30oC to 800oC. Once the heating temperature reaching 105oC, it was hold for 5 min to remove the free water in sample.Nitrogen and air were used as purging gas with a flow rate of 8
50 mL/min and 100 mL/min, respectively. The experiment under given condition was carried out more than twice. 3. Results and discussion 3.1 Compositional analysis In Table 2, the contents of volatile matters were reduced with the increasing of torrefaction temperature for CED and CAM, whereas higher fixed carbon contents in biomass were observed. It was claimed by previous study that fixed carbon release higher energy compared to volatile matters (Du et al., 2014). Moreover, Poddar et al. (2014) reported that fixed carbon can burn steadily, whereas volatile matters have higher thermal reactivity. Torrefied biomass had higher proportion of fixed carbon in comparison with untreated biomass, indicating that torrefaction had a positive influence for biomass towards higher calorific value and sustainability of combustion. However, an increasing trend of ash content was observed, which can be explained by the mass loss during torrefaction. This mass loss can be attributed to the drying of bound water and the removal of volatile matters (Chen et al. 2014). In ultimate analysis, the oxygen content was decreased from 44.49% to 33.49% for CED and from 45.03% to 32.81% for CAM with the increased torrefaction temperature, respectively. The decrease of oxygen contents for the torrefied biomass may be due to the decomposition of carbohydrate, carbonyl and hydroxyl groups in samples (Chen et al., 2014). In addition, the increased of carbon fraction can be ascribed to the dehydration and decarbonylation during torrefaction (Chen et al., 2014). As shown in Table 2, the reduction in O/C ratio was in accordance with the increment of HHV, 9
suggesting that the dehydration and decarbonylation enhanced the heating value of biomass. It was also reported by Poddar et al. (2014) that the contents of C-C and C-H bonds having higher energy were accumulated after the removal of C-O bond with the increase of torrefaction temperatures. Furthermore, the reduction in nitrogen content was observed after torrefaction, which may be contributed to the slight formation of NOx emissions. (Li et al., 2012) In addition, around 67% and 41% of extractives were decomposed for CED and CAM, respectively (Table 3). The extractives contents of both biomass materials decreased from 24.12% to 7.96% and from 15.91% to 9.37% for CED and CAM which were torefied at 240 oC, whereas only slight reduction was observed at temperature of 240-300 oC. Kaliyan et al. (2010) and Peng et al. (2012) reported that fatty compounds, protein, starch, and water soluble carbohydrates in biomass would decompose easily in light temperature. Thus, those extractives of CED and CAM decomposed in the light torrefaction might be fatty compounds, protein, starch, and water soluble carbohydrates in this study. The hemicellulose decomposition of CAM and CED occurred at temperature of 240-300 oC. The main reduction of hemicellulose in CAM was observed at the temperature range of 240-270 oC. The contents of cellulose and lignin were consequently accumulated at the temperature lower than 270 o
C. This is due to the higher maximum decomposition temperature of cellulose and
lignin than that of hemicellulose. However, significant accumulation of lignin occurred at torrefaction temperature of 270-300 oC. This may be due to the large reduction of cellulose contents at torrefaction temperature of 300 oC as shown in 10
Table 3. This can also be confirmed by the thermal analysis under nitrogen combustion condition shown in Fig. 2a and b, which indicated that the samples torrefied at 300 oC had a less contents of cellulose than other samples of both materials. These are consistent with the observations were reported in previous studies (Pelaez-Samaniego et al., 2014; Oyedun et al., 2014). 3.2 Energy consumption In Figure 3, an increase in compaction energy consumption was obtained for pellets without CAS with the increasing severity of torrefacion. The compaction energy consumption increased from 31.95 J/g to 35.62 J/g for CED and from 26.72 J/g to 40.32 J/g for CAM with the torrefaction temperature increasing to 300 oC, respectively. The compaction energy consumption was considerably influenced by the plasticity and elasticity characteristic of material, the sliding friction and the pressure input during pelletization (Shang et al., 2013). The high ash contents of torrefied biomass increased static friction among particles, resulting in higher compaction energy consumption (Shang et al., 2013). In addition, the removal of hydrogen bonding from the decomposition of extractives, hemicellulose and lignin could reduce the plasticity of samples and consequently increased the energy consumption during compaction (Li et al., 2012). As mentioned in Section 3.1, the protein, starch, fatty compounds and water soluble carbohydrates were removed primarily below the torrefation temperature of 240 oC, which resulted in the removal of available hydrogen bonding. Meanwhile, the decomposition of hemicellulose at torrefation temperature of 240 oC also resulted in the loss of hydrogen bonding. As shown in Fig. 11
3a and b, the CED torrefied at temperature of 240 oC had similar compaction energy consumption with CED torrefied at 270 oC, whereas a notable increase was obtained for the CAM torrefied at temperature between 240 oC and 270 oC. In Table 3, the extractives, cellulose and lignin content changed only slightly at temperature from 240 oC to 270 oC except hemicellulose indicating that the noticeable increment of compaction energy consumption for CAM was due to the considerable decomposition of hemicellulose at that temperature (Table 3). Biomass torrefied at temperature of 300 oC had the highest energy consumption which may be attributed to the highest ash content and the decomposition of cellulose (Fig. 3 and Table 2). Andreola et al. (2013) reported that the major component of ash is SiO2 (around 90%) which contribute to the high friction and deep abrasion. Moreover, Peng et al. (2012) indicated cellulose provides biomass particles’ strength with its fibrous structure and strong hydrogen bonding. In this work, the cellulose decomposed mainly at temperature of 270 oC to 300 oC, indicating that the grindability of samples torrefied at 300 oC was enhanced. The samples torrefied at 300 oC were easily to break into smaller particles than raw biomass during pelletization, thus resulting in larger contact area and higher frictional energy consumption than other samples. In addition, the residue of cellulose degradation may coat the surface of hemicellulose and lignin, which reduced the plasticity of samples (Li et al., 2012). It was widely accepted that oil can be added into biomass as a lubricant in order to reduce the energy consumption during compaction (Ahn et al., 2014b; Shang et al., 2013). In this study, the caster bean cake (a kind of oil-compressed byproduct) was 12
added into the untreated and torrefied samples as a potential additive. An obvious reduction of energy consumption was obtained in Fig. 3a and b. The residual oil in CAS could act as a lubricant, and thus reduced the frictions among sample particles during pelletization. In addition, starch and protein in CAS can provide hydrogen bonding enhancing the plasticity of sawdust (Li et al., 2012). Those are responsible for the decrement of energy consumption during pelletization. Similar observations were found in the previous studies, where the additives of rapeseed oil and rapeseed cake were added as lubricant (Stahl et al., 2011; Shang et al. 2014). Moreover, Fig. 3a shows that the largest reduction of energy consumption was observed for the CED torrefied at 300 oC by adding CAS. Energy consumption is ascribed to the barrier that overcomes friction and crushes particles. The torrefaction can lower particles strength resulting in the reduction of the crush particles energy input (Peng et al., 2012). Furthermore, bio-oil, starch and protein in CAS can decrease the energy consumption by reducing the friction among particles and enhancing the plasticity of sawdust. The large reduction of energy consumption indicated that torrefaction and co-pelletization with CAS had complementary effect on reducing the energy consumption markedly. As shown in Fig. 3c and d, energy consumption of extrusion increased from 0.07 J/g to 0.38 J/g for CED and 0.01 J/g to from 0.27 J/g for CAM with the increasing degree of torrefaction, respectively. As a result, torrefied biomass had higher energy consumption of extrusion than those of untreated biomass under the same condition. The increase of extrusion energy consumption can be explained by the higher friction mentioned in compaction energy consumption between particles and die. In our study, 13
the results show that adding CAS can reduce energy consumption for extrusion (Fig. 3c and d). This indicates that adding CAS can reduce the friction during pelletization by squeezing the oil into gaps between pellet and die as a lubricant. 3.3 Moisture absorption In the pellets from untreated materials, the equilibrium moisture absorption (EMA, shown in Tbale 4) was 16.47% for CED and 13.81% for CAM, respectively. Decreasing trends of the EMA were observed for CED and CAM through torrefaction, indicating that the torrefied biomass was characterized by hydrophobicity (Kaliyan & Morey, 2010). The EMA was reduced with increasing of the torrefaction temperature in the range of 240-300 oC. This may be due to the removal of extractives (protein, starch and carbonhydrates) and the decomposition of hemicellulose and lignin, thus leading to the loss of hydrotropic substances during torrefaction. In addition, the decomposed intermediate compounds coating on the hydrotropic substances were also regarded as another reason for the decrease of EMA (Kaliyan & Morey, 2010). As shown in Table 4, the EMA was increased slightly with the increasing content of CAS, while the moisture absorption rates decreased with the increasing CAS content. The increase of EMA can be explained by the presence of protein and starch in CAS, which belong to hydrotropic substances and can absorb water with its hydrophilic groups (Kaliyan & Morey, 2010). Meanwhile, the protein and starch in CAS were squeezed into the gaps and voids between particles to form the close-knit structure, which prevented water
14
vapour from entering into the pellets in the first five hours. The close-knit structure of pellets was further comfirmed by the SEM (Fig. S1 in Supplementary data). 3.4 Pellets density Pellets density is a critical factor affecting the energy density and the storage/transportation cost of biomass pellets. As shown in Fig. 4, the pellets density significantly decreased from 1g/cm3 to 0.86 g/cm3 for CED and from 1.05 to 0.84 g/cm3 for CAM with the increase of torrefaction temperature, respectively. In addition, CED torrefied at 300 oC was hard to be compacted into pellet under set condition. The loss of chemically bonded water and low-melting point compounds which acted as binding agents were considered as the reasons for lower densities of pellets from torrefied biomass. This can also be confirmed by the SEM of torrefied pellets shown in Fig. S1c and e (in Supplementary data), in which clear gaps were identified. As shown in Fig. 4, the pellets density increased with the increasing content of CAS. Furthermore, larger increment of pellets density for torrefied samples was awarded in comparison to the untreated samples. As discussed in section 3.3, the close-knit structure reinforced by CAS led to the elimination of gaps and voids between particles (Fig. S1 in Supplementary data). Therefore, compelmentary of torrefaction and co-pelletization can resist the negative effect of torrfaction on pellets density. 3.5 Pellets strength In order to evaluate the effect of torrefaction and co-pelletization on pellets strength, collapsing force and Meyer hardness for selected pellets were investigated (Fig. 5 and Table 4). Meyer hardness of the raw CED and CAM pellets were 3.57 15
N/mm2 and 4.65 N/mm2, respectively. The clear decreasing trends of Meyer hardness were observed for both biomass materials through torrefaction, which were consistent with the previous work (Li et al., 2012). The pellets made of CED torrefied at 300 oC could not be achieved under the previous condition, while the Meyer hardness of CAM pellets was 2.9 N/mm2. Furthermore, collapsing forces of the pellets without CAS were significantly decreased with the increasing torrefaction temperature. In this study, the lower Meyer hardness of torrefied pellets indicated the lack of binders and solid bridges between torrefied particles, which could be well explained by the following three aspects. Firstly, after biomass torrefaction, the decomposition of hemicellulose/lignin and the removal of extractives (proteins, starch and carbonhydrates) reduced the hydrogen bonding and solid bridges among particles resulting in the internal voids and spaces in the pellets. During strength test, those voids and spaces promoted the relative movement of particles in the matrix of pellets, leading to the lower strength of pellet. In addition, the residual ash and the intermediate products formed from decomposition of hemicellulose/lignin and extractives deposited on the surface of biomass particles. The deposition of those ash and intermediate products could prevent the formation of solid bridge and interlock, which can be confirmed by the gaps and voids in Fig. S1c and e (in Supplementary data). Secondly, the moisture in biomass played an important role in the attractive forces of adjacent particles (Ahn et al., 2014a). The moisture adsorption capacity of biomass decreased considerably for biomass through torrefaction (Table 4), leading to the quantitative decrement of weak hydrogen bonding sites during pelletization. 16
Lastly, the grindability of biomass was enhanced through torrefaction as the decomposition of hemicellulose, cellulose and lignin took place (Arias et al., 2008). It was thus more easily for the torrefied biomass to be crushed into fines during pelletization (Arias et al., 2008). As a result, the joints between adjacent torrefied particles were enlarged in comparison to that of raw samples. The enlarged joints leaded to the increase of gaps and voids because of the absence of binder and the weak of attractive force in the joints. However, the collapsing force for the CAM pellets produced from sawdust torrfied at 240 oC was increased to 48.23 N, compared to 40.46 N for the raw CAM pellets. The removal of fatty compounds compounds during torrefaction at 240 oC may reduce the relative movement of particles within the pellets. In a previous report, the strength of pellets produced from mixture of rapeseed cake and biomass decreased with an increased amount of rapeseed cake, which was ascribed to the lubrication capacity of extractive matters in rapeseed cake (Stahl et al., 2011). Nevertheless, a different result was achieved in the present work as the addition of CAS into biomass (in Table 4 and Fig. 5). Compared with the liquid fraction of rapeseed oil, the soft-solid protein and starch compounds in CAS can be compressed into the gaps and voids of particles during pelletization. Those compressed protein and starch compounds can act as the binders and form solid bridge and hydrogen bonding, leading to an improved close-knit structure of matrix in pellets. This can be confirmed by the SEM (Fig. S1 in Supplementary data) and the higher density of pellets (Fig. 4). Furthermore, the enlarged joints between adjacent 17
particles were filled and connected by the proteins and starch in CAS, contributing to a further enhancement of solid bridge and attractive forces among particles (Jiang et al., 2014; Kaliyan & Morey, 2009). Therefore, the compelmentary of torrefaction and co-pelletization can improve the pellets strength. 3.6 Combustion characteristics The combustion characteristic is subjected to thermal analysis under air combustion conditions, which is necessary in evaluating the application properties of fuels for future applications. Higher weight loss was observed for raw samples than torrefied samples during the initial stage of combustion in the temperatures range of 30 to 105 oC (Table 5). Generally, the initial mass loss is attributed to the dehydration in samples. The torrefied samples consequently contain less moisture in comparison to raw samples due to the initial dehydration of biomass during torrefaction (Yi et al., 2012). Meanwhile, as depicted in Fig. 2c and d, different curves were shown between raw samples and torrefied samples, especially for the heavy torrefied samples, which were consistent with the previous study (Chen & Kuo, 2011). It can be due to the compositions variations of hemicellulose, celluloses and lignin after torrefaction (Table 3). It was suggested by Peng et al. (2012) that the main loss of hemicellulose occurred at 220-315 oC compared with the loss of cellulose primarily at 315-400 oC. The lignin decomposed slowly over a wide temperature range with the maximum decomposition rate obtained at around 460 oC (Zhou et al., 2014). As shown in Table 5, the ignition temperature and final combustion temperature increased with the increasing degree of torrefaction. The maximum DTG decreased at stage 1 and then 18
increased at stage 2 with the increasing torrefaction temperature (Biswas et al., 2014). As shown in Table 3, the change of maximum DTG at stage 2 is in consistent with the variation of carbon fraction with the increase of torrefaction temperature. In Table 5, the ignition temperature and maximum DTG of pellets decreased with the increasing CAS ratio. High maximum DTG had a harmful effect on combustibility (Yi et al., 2012). As a result, a more stable combustion was achieved by adding CAS. However, in stage 1, the maximum DTG was increased with the increasing content of CAS for the untreated and lightly torefied samples at 240 oC. It was indicated that the residual oil in CAS was able to accelerate the burning at stage 1, in which hemicellulose and cellulose were the primary combustible components (Zhou et al., 2014). It was found that the combustion characteristics were highly affected by the degree of torrefaction and the ratio of CAS. Table 5 shows that higher residue was left with the increasing content of CAS, which is due to the high ash contents in CAS (8.79%). 4. Conclusions The process of torrefaction can improve the hydrophobicity of biomass pellets as the decomposition of protein and hemicellulose occured. Meanwhile, co-pelletization with caster bean cake can reduce the high energy consumption and weak pellets strength caused by torrefaction. The reduction of energy consumption is attributed to the well grindability of torrefied biomass and the lubricant of caster bean cake. Moreover, the high density and Meyer hardness of pellets by co-pelletization of torrefied biomass were caused by the enlarged contact area and thus enhanced attractive forces between particles. The economic cost budget should be investigated 19
in further large-scale production. Acknowledgements The authors gratefully acknowledge the financial support provided by the National Natural Science Foundation of China (No. 21407046, 31470594), the Natural Science Foundation of Hunan Province, China (No. 13JJ4118), Key Laboratory of Renewable Energy Chinese Academy of Sciences (No.y407k91001), and
Collaborative
Innovation
Center
of
Resource-Conserving
Environment-Friendly Society and Ecological Civilization.
20
&
References Ahn, B.J., Chang, H.S., Lee, S.M., Choi, D.H., Cho, S.T., Han, G.S., Yang, I., 2014a. Effect of binders on the durability of wood pellets fabricated from Larix kaemferi C. and Liriodendron tulipifera L. sawdust. Renewable Energy, 62, 18-23. Andreola, F., Martı´n, M.I., Ferrari, A.M., Lancellotti, I., Bondioli, F., Rinco´ n, J.M., Romero, M., Barbieri, L., 2013. Technological properties of glass-ceramic tiles obtained using rice husk ash as silica precursor. Ceramics International, 39, 5427-5435. Arias, B., Pevida, C., Fermoso, J., Plaza, M.G., Rubiera, F., Pis, J.J., 2008. Influence of torrefaction on the grindability and reactivity of woody biomass materials. Fuel Processing Technology, 89(2), 169-175. Bach, Q. V., Tran, K. Q., Skreiberg, Ø., Khalil, R.A., Phan, A.N., 2014. Effects of wet torrefaction on reactivity and kinetics of wood under air combustion conditions. Fuel, 137, 375-383. Basu, P., Sadhukhan, A.K., Gupta, P., Rao, S., Dhungana, A., Acharya, B., 2014. An experimental and theoretical investigation on torrefaction of a large wet wood particle. Bioresour Technol, 159, 215-22. Biswas, A.K., Rudolfsson, M., Broström, M., Umeki, K. 2014., Effect of pelletizing conditions on combustion behaviour of single wood pellet. Applied Energy, 119, 79-84. Chau, J., Sowlati, T., Sokhansanj, S., Preto, F., Melin, S., Bi, X., 2009. Economic sensitivity of wood biomass materials utilization for greenhouse heating application. Applied Energy, 86(5), 616-621. Chen, W. H., Kuo, P. C., 2011. Torrefaction and co-torrefaction characterization of hemicellulose, cellulose and lignin as well as torrefaction of some basic constituents in biomass 21
materials. Energy, 36(2), 803-811. Chen, W.H., Cheng, W.Y., Lu, K.M., Huang, Y.P., 2011. An evaluation on improvement of pulverized biomass materials property for solid fuel through torrefaction. Applied Energy, 88(11), 3636-3644. Chen, Y., Yang, H., Yang, Q., Hao, H., Zhu, B., Chen, H., 2014. Torrefaction of agriculture straws and its application on biomass materials pyrolysis poly-generation. Bioresour Technol, 156, 70-7. Chin, K.L., H’ng, P.S., Go, W.Z., Wong, W.Z., Lim, T.W., Maminski, M., Paridah, M.T., Luqman, A.C., 2013. Optimization of torrefaction conditions for high energy density solid biofuel from oil palm biomass materials and fast growing species available in Malaysia. Industrial Crops and Products, 49, 768-774. Du, S.W., Chen, W.H., Lucas, J.A., 2014. Pretreatment of biomass materials by torrefaction and carbonization for coal blend used in pulverized coal injection. Bioresour Technol, 161, 333-9. Iroba, K.L., Tabil, L.G., Sokhansanj, S., Dumonceaux, T., 2014. Pretreatment and fractionation of barley straw using steam explosion at low severity factor. Biomass materials and Bioenergy, 66, 286-300. Jiang, L., Liang, J., Yuan, X., Li, H., Li, C., Xiao, Z., Huang, H., Wang, H., Zeng, G., 2014. Co-pelletization of sewage sludge and biomass materials: The density and hardness of pellet. Bioresour Technol, 166, 435-43. Kaliyan, N., Morey, R. 2009. Factors affecting strength and durability of densified biomass materials products. Biomass materials and Bioenergy, 33(3), 337-359. 22
Kaliyan, N., Morey, R.V., 2010. Natural binders and solid bridge type binding mechanisms in briquettes and pellets made from corn stover and switchgrass. Bioresour Technol, 101(3), 1082-90.Lam, P.S., Sokhansanj, S., Bi, X.T., Lim, C.J., Larsson, S.H., 2012. Drying characteristics and equilibrium moisture content of steam-treated Douglas fir (Pseudotsuga menziesii L.). Bioresour Technol, 116, 396-402. Li, H., Liu, X.H., Legros, R., Bi, X.T., Lim, C.J., Sokhansanj, S., 2012. Pelletization of torrefied sawdust and properties of torrefied pellets. Applied Energy, 93, 680-685. Li, J., Brzdekiewicz, A., Yang, W., Blasiak, W., 2012. Co-firing based on biomass torrefaction in a pulverized coal boiler with aim of 100% fuel switching. Applied Energy, 99, 344-354. Liu, Z., Quek, A., Balasubramanian, R., 2014. Preparation and characterization of fuel pellets from woody biomass materials, agro-residues and their corresponding hydrochars. Applied Energy, 113, 1315-1322. Medic, D., Darr, M., Shah, A., Rahn, S., 2012. Effect of Torrefaction on Water Vapor Adsorption Properties and Resistance to Microbial Degradation of Corn Stover. Energy & Fuels, 26(4), 2386-2393. Mišljenović, N., Bach, Q. V., Tran, K. Q., Salas-Bringas, C., Skreiberg, Ø., 2014. Torrefaction Influence on Pelletability and Pellet Quality of Norwegian Forest Residues. Energy Fuels, 28, 2554-2561. Nunes, L.J., Matias, J.C., Catalão, J.P., 2014. Mixed biomass materials pellets for thermal energy production: A review of combustion models. Applied Energy, 127, 135-140. Oyedun, A.O., Tee, C.Z., Hanson, S., Hui, C.W., 2014. Thermogravimetric analysis of the pyrolysis characteristics and kinetics of plastics and biomass blends. Fuel Processing 23
Technology, 128, 471-481. Pelaez-Samaniego, M.R., Yadama, V., Garcia-Perez, M., Lowell, E., McDonald, A.G., 2014. Effect of temperature during wood torrefaction on the formation oflignin liquid intermediates. Journal of Analytical and Applied Pyrolysis, 109, 222-233. Peng, J.H., Bi, X.T., Lim, J., Sokhansanj, S., 2012. Development of Torrefaction Kinetics for British Columbia Softwoods. International Journal of Chemical Reactor Engineering, 10. Poddar, S., Kamruzzamana, M, Sujan, S.M., Hossain, M., Jamal, M.S., Gafur, M.A. Khanam, M., 2014. Effect of compression pressure on lignocellulosic biomass materials pellet to improve fuel properties: Higher heating value. Fuel, 131, 43-48. Reza, M. T., Uddin, M. H., Lynam, J. G., Coronella, C. J., 2014. Engineered pellets from dry torrefied and HTC biochar blends. Biomass and Bioenergy, 63, 229-238. Saleh, S.B., Hansen, B.B., Jensen, P.A., Johansen, K.D., 2013. Efficient Fuel Pretreatment: Simultaneous Torrefaction and Grinding of Biomass. Enegy Fules, 27, 7531-7540. Shang, L., Nielsen, N.P.K., Stelte, W., Dahl, J., Ahrenfeldt, J., Holm, J.K., Arnavat, M.P., Bach, L.S., Henriksen, U.B., 2013. Lab and Bench-Scale Pelletization of Torrefied Wood Chips—Process Optimization and Pellet Quality. BioEnergy Research, 7(1), 87-94. Tabil, L.G., Sokhansanj, S., Crerar1, W.J., Patil, R.T., Khoshtaghaza, M.H., Opoku1, A., 2002. Physical characterization of alfalfa cubes: I. Hardness. Canadian biosystem engineering, 44,55-63. Wen, J.L., Sun, S.L., Yuan, T.Q., Xu, F., Sun, R.C., 2014. Understanding the chemical and structural transformations of lignin macromolecule during torrefaction. Applied Energy, 121, 1-9. 24
Xu, S.C., He, Z.X., Long, R.Y., 2014. Factors that influence carbon emissions due to energy consumption in China: Decomposition analysis using LMDI. Applied Energy, 127, 182-193. Yi, Q., Qi, F., Cheng, G., Zhang, Y., Xiao, B., Hu, Z., Liu, S., Cai, H., Xu, S., 2012. Thermogravimetric analysis of co-combustion of biomass materials and biochar. Journal of Thermal Analysis and Calorimetry, 112(3), 1475-1479. Zhou, C., Liu, G., Cheng, S., Fang, T., Lam, P.K., 2014. Thermochemical and trace element behavior of coal gangue, agricultural biomass materials and their blends during co-combustion. Bioresour Technol, 166, 243-51.
25
Figure captions Fig. 1 The schematic and flow diagram of torrefaction and co-pelletization. Fig. 2 TG and DTG curves of raw and torrefied pellets under a nitrogen flow rate of 50 mL/min and a air flow rate of 100 mL/min at a heating rate of 15 oC/min. Fig. 3 The energy consumption of compaction and extrusion associated with the degree of torrefaction and the additive ratios CAS. Fig. 4 Effects of the CAS ratio and the degree of torrefaction on pellets density. Fig. 5 Meyer hardness of pellets as a function of the mixing ratio of CAS and the torrefaction temperature.
Fig. 1 The schematic and flow diagram of torrefaction and co-pelletization.
26
Fig. 2 TG and DTG curves of raw and torrefied pellets under a nitrogen flow rate of 50 mL/min and a air flow rate of 100 mL/min at a heating rate of 15 oC/min. (a) CED pellets in nitrogen ; (b) CAM pellets in nitrogen (c) CED pellets in air ; (d) CAM pellets in air.
27
Fig. 3 The energy consumption of compaction and extrusion associated with the degree of torrefaction and the additive ratios CAS. (a) compaction for CED pellets; (b) compaction for CAM pellets (c) extrusion for CED pellets; (d) extrusion for CAM pellets.
28
Fig. 4 Effects of the CAS ratio and the degree of torrefaction on pellets density. (a) CED pellets; (b) CAM pellets.
29
Fig. 5 Meyer hardness of pellets as a function of the mixing ratio of CAS and the torrefaction temperature. (a) CED pellets; (b) CAM pellets.
30
Table 1 Weight loss and particle size distribution of raw and torrefied samples. Sample
Weight loss
Size distribution 1000-840um
840-420um
420-180um
<180um
CED Raw
0
5.73
42.32
38.49
13.46
240oC
7.69
4.28
37.89
42.33
15.51
270oC
19.92
3.78
35.18
42.56
18.49
300oC
33.54
2.91
38.05
42.82
16.22
Raw
0
8.09
59.68
27.48
4.76
240oC
13.72
5.93
57.14
31.34
5.58
270oC
16.81
7.16
56.30
30.90
5.63
300oC
34.00
2.91
38.05
42.82
16.22
CAM
31
Table 2 The proximate and ultimate analysis of raw and torrefied samples expressed on a dry basis. CED
Raw
o
CAM o
o
o
240 C
270 C
300 C
Raw
240 C
270oC
300oC
CAS
Proximate analysis (%)
VM
81.86
81.96
79.25
65.34
85.51
82.62
79.20
70.91
69.8
FC
17.25
17.12
18.76
32.64
13.42
16.21
19.36
26.55
21.41
A
0.89
0.92
1.99
2.02
1.07
1.17
1.44
2.54
8.79
Elemental analysis (%) C
48.95
54.74
53.41
61.02
48.18
51.02
53.05
61.94
44.21
H
5.91
5.76
6.02
5.28
6.09
6.19
6.06
5.02
5.93
O
44.49
39.22
40.33
33.49
45.03
42.37
40.60
32.81
41.94
N
0.65
0.28
0.25
0.22
0.70
0.41
0.30
0.23
7.92
O/C ratio
0.91
0.72
0.76
0.54
0.93
0.83
0.77
0.53
0.95
HHV (MJ/kg)
19.57
19.87
20.61
23.17
18.86
18.90
20.45
22.35
19.12
.
32
Table 3 Fiber analysis of the raw and torried biomass using cellulose tester. Sample
Composition H
a
b
c
Composition change d
C
L
E
H
a
Cb
Lc
Ed
CED
Raw
12.05
36.22
27.61
24.12
0
0
0
0
240 oC
10.51
45.84
35.69
7.96
-12.78
26.56
29.26
-67.00
270 oC
7.37
46.59
39.59
6.45
-38.84
28.63
43.39
-73.26
300 oC
4.04
24.52
66.38
5.06
-66.47
-32.30
140.42
-79.02
Raw
20.82
38.87
24.4
15.91
0
0
0
0
240 oC
18.39
48.47
23.77
9.37
-11.67
24.70
-2.58
-41.11
270 oC
9.7
52.69
27.77
9.84
-53.41
35.55
13.81
-38.15
300 oC
4.63
40.22
49.58
5.57
-77.76
3.47
103.20
-64.99
CAS
28.98
20.73
23.75
44.54
CAM
a
Hemicelluloses Cellulose c Lignin d Extractives b
33
Table 4 Data obtained in connection with the equilibrium moisture absorption, moisture absorption rates (k) and collapsing force of raw and torrefied pellets. CED Samples
Raw
CAM
EMA (%)c
k (min-1)
Collapsing force (N)
EMA (%)c
k (min-1)
Collapsing force (N)
16.47
0.00534
56.36
13.81
0.01028
40.46
a
Raw
17.34
0.00535
63.46
13.91
0.0095
55.33
Rawb
17.09
0.00479
73.63
13.33
0.00851
66.36
o
12.86
0.00554
35.53
11.86
0.00788
48.23
13.31
0.00566
43.3
11.26
0.00768
52.2
14.49
0.00485
48.21
11.94
0.00682
72.06
240 C 240 oC a o
240 C
b
o
270 C
12.00
0.00504
19.23
9.52
0.00773
47.9
o
a
13.86
0.00459
44.83
10.37
0.00777
64.6
o
b
14.44
0.00425
58.26
10.49
0.0069
79.13
10.46
0.00414
0d
9.21
0.00755
31.73
a
11.49
0.00422
19.46
10.21
0.0073
67.1
300 oC b
13.01
0.00445
35.96
10.25
0.00708
128.61
270 C 270 C 300 oC o
300 C
a
Mixed with 15% CAS (dry basis)
b
Mixed with 30% CAS (dry basis)
c
Equilibrium moisture absorption d Pellet cannot be compact under150 oC, 4000N
34
Table 5 Combustion characteristics of raw biomass, torrefied biomass and their blends with CAS. Stage 1 Samples
Stage 2
Residue (%)
Range (oC)
Tmax (oC)
DTGmax (% s-1)
Range (oC)
Tmax (oC)
DTGmax (% s-1)
Raw
202-402
340
0.54
402-487
467
0.53
0.94
Rawa
200-378
332
0.33
378-516
429
0.46
3.09
Rawb
215-374
346
0.98
374-506
452
0.04
1.67
240 C
238-405
343
0.47
405-498
478
0.61
1.46
240 oC a
226-368
357
0.97
368-507
476
0.15
1.87
240 oC b
229-367
347
1.06
367-505
465
0.09
3.35
270 C
247-402
340
0.37
402-506
471
0.65
2.33
270 oC a
216-388
338
0.47
388-511
444
0.55
1.92
270 oC b
224-368
355
0.94
368-507
463
0.09
2.56
300 C
253-388
329
0.30
388-508
458
0.79
1.35
300 oC a
227-379
326
0.21
379-525
445
0.36
3.46
300 oC b
222-383
326
0.25
383-523
441
0.38
2.32
Raw
222-391
334
0.37
391-500
451
0.34
0.37
Rawa
206-363
346
0.88
363-482
454
0.08
0.7
b
Raw
216-392
335
0.29
392-506
457
0.33
1.46
240 oC
223-396
331
0.36
396-510
450
0.20
1.97
240 oC a
220-357
334
0.74
357-491
442
0.08
1.66
240 oC b
223-383
333
0.37
383-494
445
0.39
3.62
270 C
236-384
332
0.56
384-513
440
0.18
1.14
270 oC a
233-357
349
0.91
357-508
453
0.11
1.57
270 oC b
228-350
330
0.58
350-523
438
0.36
3.60
300 C
233-375
326
0.33
375-508
450
0.68
1.87
300 oC a
233-360
331
0.45
360-493
445
0.49
4.38
300 oC b
224-376
328
0.22
376-520
446
0.18
3.48
CED
o
o
o
CAM
o
o
a
Mixed with 15% CAS (dry basis)
b
Mixed with 30% CAS (dry basis)
35
Highlights
Caster bean cake improves the features of pellets produced from torrefied biomass.
Grindability and lubricant are available to decrease the energy consumption.
Complementary effect of protein and enlarged contact area improve pellet strength.
Co-pelletizing biomass torrefied at 270 oC with 15% CAS are the optimal conditions.
36
S0960-8524(15)00224-2 http://dx.doi.org/10.1016/j.biortech.2015.02.045 BITE 14618
To appear in:
Bioresource Technology
Received Date: Revised Date: Accepted Date:
21 November 2014 9 February 2015 11 February 2015
Please cite this article as: Cao, L., Yuan, X., Li, H., Li, C., Xiao, Z., Jiang, L., Huang, B., Xiao, Z., Chen, X., Wang, H., Zeng, G., Complementary effects of torrefaction and co-pelletization: Energy consumption and characteristics of pellets, Bioresource Technology (2015), doi: http://dx.doi.org/10.1016/j.biortech.2015.02.045
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.
Complementary effects of torrefaction and co-pelletization : Energy consumption and characteristics of pellets Liang Cao a, b, c, Xingzhong Yuan a, b,*, Hui Li c,d,*, Changzhu Li c,d, Zhihua Xiao a, b, Longbo Jiang a, b
, Binbin Huang a, b, Zhihong Xiao d, Xiaohong Chen e, Hou Wang a, b, Guangming Zeng a, b
a College of Environmental Science and Engineering, Hunan University, Changsha 410082, PR China b Key Laboratory of Environmental Biology and Pollution Control (Hunan University), Ministry of Education, Changsha 410082, PR China c Institute of Biological and Environmental Engineering, Hunan Academy of Forestry, Changsha 410004, PR China d Institute of Bio-energy, Hunan Academy of Forestry, Changsha 410004, PR China e School of Business, Central South University, Changsha 410083, P.R. China
* Corresponding author. Tel.: + 86-731-88821413；Fax：+ 86-731-88821413.E-mail address: [email protected] (X.Z. Yuan). Tel.: +86-731-85578765；Fax： +86-731-85578765.E-mail address: [email protected] (H. Li). 1
Abstract In this study, compelmentary of torrefaction and co-pelletization for biomass pellets production was investigated. Two kinds of biomass materials were torrefied and mixed with oil cake for co-pelletization. The energy consumption during pelletization and pellet characteristics including moisture absorption, pellet density, pellet strength and combustion characteristic, were evaluated. It was shown that torrefaction improved the characteristics of pellets with high heating values, low moisture absorption and well combustion characteristic. Furthermore, co-pelletization between torrefied biomass and cater bean cake can reduce several negative effects of torrefaction such as high energy consumption, low pellet density and strength. The optimal conditions for energy consumption and pellet strength were torrefied at 270 o
C and a blending with 15% castor bean cake for both biomass materials. The present
study indicated that compelmentary performances of the torrefaction and co-pelletization with castor bean cake provide a promising alternative for fuel production from biomass and oil cake. Keywords: Oil cake, Torrefaction, Co-pelletization, Energy consumption, Combustion
2
1. Introduction Biomass pellets have become a sustainable and environmental energy resource as its attractive characteristics, such as renewable, none net CO2 emission and low production cost (Du et al., 2014; Jiang et al., 2014; Nunes et al., 2014; Xu et al., 2014)). In general, pelletization of biomass makes its density 4 to10 times higher than those without pelletization which can significantly reduce the transportation/storage costs (Li et al., 2012; Liu et al., 2014). However, due to the high moisture absorption, low energy density and rapid combustion reactivity of biomass, all of those significantly hinder the large-scale application of biomass pellets (Chin et al., 2013; Zhou, C., 2014). Therefore, the advancements of the properties of biomass pellets have gained widespread attention. Among the technologies for improving the properties of biomass, torrefaction was regarded as one of the efficient and feasible method (Chen et al., 2014; Du et al., 2014; Li et al.,). Wen et al.(2014) reported that pretreatment of torrefaction is a pyrolytic process at 200-300 oC in an inert atmosphere. It is well known that torrefied biomass is a material characterized by hydrophobicity, grindability and high heat value because of the removal of hemicellulose, lignin, protein, starch and fatty compounds in biomass (Chen et al., 2014; Li et al., 2012; Mišljenovic et al., 2014). Moreover, the combustion performance of pellets produced from torrefied biomass approaches to that of coal, which can be due to the removal of free water and volatile components during torrefaction (Du et al., 2014). However, compaction and extrusion energy consumptions during pelletization were remarkably augmented for the 3
torrefied biomass materials, while the density and strength of the pellets made from torrefied biomass were decreased compared with those of untreated biomass (Li et al., 2012). Different methods and other various torrefaction and pelletization conditions were adopted in order to resolve the disadvantages mentioned above (Chen et al., 2014; Du et al., 2014; Li et al., 2012). Among the existing technologies, co-pelletization process has become a widespread used technology to improve the characteristics of pellet (Stahl et al., 2011; Shang et al., 2013;Jiang et al., 2014; Nunes et al., 2014). The major advantage of co-pelletization is to eliminate the negative properties of pellets with the aid of additives. For example, Jiang et al. (2014) investigated that biomass mixed with sludge can enhance the pellet density and strength as the protein in sludge can act as a binder between biomass particles. Stahl et al. (2011) studied that energy consumption decreased with an increasing content of rapeseed cake, which was due to the reduction of friction during pelletization caused by the presence of rape oil in rapeseed cake. Zhou et al. (2014) managed to improve the combustion progress by blending biomass with coal to increase the content of fixed carbon. Therefore, several studies have investigated the effects of torrefaction and co-pelletizing on the pellet properties. It was carried out by Reza et al. (2014) that pellets produced from dry torrefied and hydrothermal carbonization (HTC) biomass blends possessed higher pellets density and better durability than those from torrefied biomass. Shang et al., (2013) investigated the blending of rapeseed oil with torrefied biomass can reduce the energy consumption during pelletization (Shang et al., 2013). However, no further study was 4
carried out to reveal the potential effect of rapeseed oil on pellets’ strength (Shang et al., 2013). In addition, the higher price of additives compared with conventional biomass resulting in the economic infeasibility of biomass were seldom considered (Ahn et al., 2014b; Chau et al., 2009; Reza et al., 2014). In this respect, more attention should be paid on the low-cost additives during co-pelletization. In China, oil cake is a main byproduct in the oil expression process with a total annual amount of 16-20 Mt. Nevertheless, large amount of oil cake carries hazardous substances such as sinapic acid, saponin and tannins; and poor palatability and nutritional imbalance make it difficult to be utilized as animals feed. Meanwhile, several specific features of oil cake, such as the higher energy content and functional components (starch, protein and bio-oil, etc.) in oil cake reveal that oil cake could be utilized as fuel and additives in the co-pelletization with torrefied biomass. To our best knowledge, few studies are available on the complementary effects of torrefaction and co-pelletization of oil cake on energy consumption and pellets characteristics. In this study, two kinds of biomass (cedarwood and camphorwood) were pretreated at different torrefaction temperature (240 oC, 270 oC, and 300 oC), respectively. Caster bean cake, one kind of high-production oil cakes in East Asia, was selected as the additive. The torrefied biomass materials were blended with caster bean cake for co-pelletization. Subsequently, the effects of torrefaction and caster bean cake on the energy consumption and pellet properties were investigated. The primary objective of this study was to reveal the compelmentary mechanisms between torrefaction and co-pelletization for biomass pellet preparation. 5
2. Materials and methods 2.1Materials Two types of wood sawdust were selected in the present study: cedarwood (CED) and camphorwood (CAM), which were collected from local forest in Changsha and further grounded into fractions with particle size below 1 mm (Table 1). The selected oil cake was castor bean cake (CAS) obtained from the expression process of castor bean in a pilot of Hunan Academy of Forestry. The selected materials were dried (40 o
C, 48 hours) and stored in plastic containers at 4 oC for further study.
2.2 Torrefaction The torrefaction reactor is a tubular furnace which belongs to a bench-scale fixed bed reactor (Fig.1). The reactor contains a tubular unit, an electrical heater and a gas supply. The biomass was heated up to the desired temperature with a rate of 10 o
C/min-1 and a nitrogen flow of 100 mL/min-1 as a carrier gas was used during
torrefaction. The set torrefaction conditions were 240 oC (light torrefaction), 270 oC (medium torrefaction) and 300 oC (heavy torrefaction) with a residence time of 30 min, respectively. Table 1 records the weight loss and particle size of the torrefied samples at different temperature with respect to raw samples. 2.3 Co-pelletization The pellets were prepared using a DWD-10 pressure unit installed with a cylinder die and a piston (Fig.1). The cylinder die encircled with a heating tap had an opening of 7 mm in diameter and 120 mm in length. The end of the die was a removable plug. The piston with sizes of 6.8 mm in diameter and 80 mm in length 6
was installed to supply the pressure for the sample. In this work, the additive (CAS) was added into raw and torrefied samples with the proportion of 0, 15 and 30%, respectively. The loaded sample (around 0.7 g) was compressed with a rate of 2 mm/min until the desired pressure was achieved. The maximum pressure was 4000 N with a residence time of 30 s. Prior to pelletization, the cylinder die was preheated up to 150
o
C. The transducer was used to measure the compression force and
displacement and the relevant data were recorded by a computer. Compression and extrusion energy consumption were calculated by the formulation as following. (1)
= ∑ ∙
Where EC is the energy consumption (J), F is the compression force (N), S is the displacement (mm). 2.4 Moisture absorption Prior to the moisture absorption, the pellets were dried at 105 oC for 48 hours. The tests were carried out in a humidity chamber (30 oC and 90% relative humidity). The corresponding weights of the pellets were determined every 15 min for the first five hours. The equilibrium moisture content was measured after 2 days. Two pellets were carried out in this analysis. The kinetics of moisture absorption is presented by the ASABE formulation as following (Lam et al., 2012): (M-Me)/(Mi-Me)=e-k t
(2)
Where M is the instantaneous moisture content (decimal, dry basis), Me is the equilibrium moisture content (decimal, dry basis), and Mi is the initial moisture content (decimal, dry basis). The coefficient k is an absorption constant (min-1), and t 7
is the exposure time (min). 2.5 Pellets strength The pellets strength was also measured by the DWD-10 pressure unit. The strength was determined by the force crushing the pellet to the length of pellet and the depth of indentation as described in an earlier work (Li et al., 2012). Each pellet was horizontally laid down on a flat surface. An increasing force was applied at the center of the pellet until crushing. Three pellets were applied to observe the representative data. The pellet strength was determined using the pattern of Meyer hardness tester (L.G. Tabil et al., 2002). The Meyer hardness (MS), defined as the static indentation strength, was calculated using the following equation (L.G. Tabil et al., 2002): MS=F/[π(Dh-h2)]
(3)
Where MS is Meyer hardness (N/mm2); F is maximum pressure (N); D is pellet diameter (mm) and h is the depth of indentation (mm). 2.6 Thermogravimetric analysis (TG) Thermogravimetric analysis was performed using the TG-60 analyzer (JAP) in order to evaluate the pellets’ combustion characteristic. The analyzer can provide continuous recording of the weight loss (TG) and the rate of weight loss (DTG) curves. In this study, around 5mg of sample (shown in Table 1) loaded into the platinum crucible was placed inside the thermal analyzer. The experiment run was carried out at heating rates of 10 oC/min with temperature ranging from 30oC to 800oC. Once the heating temperature reaching 105oC, it was hold for 5 min to remove the free water in sample.Nitrogen and air were used as purging gas with a flow rate of 8
50 mL/min and 100 mL/min, respectively. The experiment under given condition was carried out more than twice. 3. Results and discussion 3.1 Compositional analysis In Table 2, the contents of volatile matters were reduced with the increasing of torrefaction temperature for CED and CAM, whereas higher fixed carbon contents in biomass were observed. It was claimed by previous study that fixed carbon release higher energy compared to volatile matters (Du et al., 2014). Moreover, Poddar et al. (2014) reported that fixed carbon can burn steadily, whereas volatile matters have higher thermal reactivity. Torrefied biomass had higher proportion of fixed carbon in comparison with untreated biomass, indicating that torrefaction had a positive influence for biomass towards higher calorific value and sustainability of combustion. However, an increasing trend of ash content was observed, which can be explained by the mass loss during torrefaction. This mass loss can be attributed to the drying of bound water and the removal of volatile matters (Chen et al. 2014). In ultimate analysis, the oxygen content was decreased from 44.49% to 33.49% for CED and from 45.03% to 32.81% for CAM with the increased torrefaction temperature, respectively. The decrease of oxygen contents for the torrefied biomass may be due to the decomposition of carbohydrate, carbonyl and hydroxyl groups in samples (Chen et al., 2014). In addition, the increased of carbon fraction can be ascribed to the dehydration and decarbonylation during torrefaction (Chen et al., 2014). As shown in Table 2, the reduction in O/C ratio was in accordance with the increment of HHV, 9
suggesting that the dehydration and decarbonylation enhanced the heating value of biomass. It was also reported by Poddar et al. (2014) that the contents of C-C and C-H bonds having higher energy were accumulated after the removal of C-O bond with the increase of torrefaction temperatures. Furthermore, the reduction in nitrogen content was observed after torrefaction, which may be contributed to the slight formation of NOx emissions. (Li et al., 2012) In addition, around 67% and 41% of extractives were decomposed for CED and CAM, respectively (Table 3). The extractives contents of both biomass materials decreased from 24.12% to 7.96% and from 15.91% to 9.37% for CED and CAM which were torefied at 240 oC, whereas only slight reduction was observed at temperature of 240-300 oC. Kaliyan et al. (2010) and Peng et al. (2012) reported that fatty compounds, protein, starch, and water soluble carbohydrates in biomass would decompose easily in light temperature. Thus, those extractives of CED and CAM decomposed in the light torrefaction might be fatty compounds, protein, starch, and water soluble carbohydrates in this study. The hemicellulose decomposition of CAM and CED occurred at temperature of 240-300 oC. The main reduction of hemicellulose in CAM was observed at the temperature range of 240-270 oC. The contents of cellulose and lignin were consequently accumulated at the temperature lower than 270 o
C. This is due to the higher maximum decomposition temperature of cellulose and
lignin than that of hemicellulose. However, significant accumulation of lignin occurred at torrefaction temperature of 270-300 oC. This may be due to the large reduction of cellulose contents at torrefaction temperature of 300 oC as shown in 10
Table 3. This can also be confirmed by the thermal analysis under nitrogen combustion condition shown in Fig. 2a and b, which indicated that the samples torrefied at 300 oC had a less contents of cellulose than other samples of both materials. These are consistent with the observations were reported in previous studies (Pelaez-Samaniego et al., 2014; Oyedun et al., 2014). 3.2 Energy consumption In Figure 3, an increase in compaction energy consumption was obtained for pellets without CAS with the increasing severity of torrefacion. The compaction energy consumption increased from 31.95 J/g to 35.62 J/g for CED and from 26.72 J/g to 40.32 J/g for CAM with the torrefaction temperature increasing to 300 oC, respectively. The compaction energy consumption was considerably influenced by the plasticity and elasticity characteristic of material, the sliding friction and the pressure input during pelletization (Shang et al., 2013). The high ash contents of torrefied biomass increased static friction among particles, resulting in higher compaction energy consumption (Shang et al., 2013). In addition, the removal of hydrogen bonding from the decomposition of extractives, hemicellulose and lignin could reduce the plasticity of samples and consequently increased the energy consumption during compaction (Li et al., 2012). As mentioned in Section 3.1, the protein, starch, fatty compounds and water soluble carbohydrates were removed primarily below the torrefation temperature of 240 oC, which resulted in the removal of available hydrogen bonding. Meanwhile, the decomposition of hemicellulose at torrefation temperature of 240 oC also resulted in the loss of hydrogen bonding. As shown in Fig. 11
3a and b, the CED torrefied at temperature of 240 oC had similar compaction energy consumption with CED torrefied at 270 oC, whereas a notable increase was obtained for the CAM torrefied at temperature between 240 oC and 270 oC. In Table 3, the extractives, cellulose and lignin content changed only slightly at temperature from 240 oC to 270 oC except hemicellulose indicating that the noticeable increment of compaction energy consumption for CAM was due to the considerable decomposition of hemicellulose at that temperature (Table 3). Biomass torrefied at temperature of 300 oC had the highest energy consumption which may be attributed to the highest ash content and the decomposition of cellulose (Fig. 3 and Table 2). Andreola et al. (2013) reported that the major component of ash is SiO2 (around 90%) which contribute to the high friction and deep abrasion. Moreover, Peng et al. (2012) indicated cellulose provides biomass particles’ strength with its fibrous structure and strong hydrogen bonding. In this work, the cellulose decomposed mainly at temperature of 270 oC to 300 oC, indicating that the grindability of samples torrefied at 300 oC was enhanced. The samples torrefied at 300 oC were easily to break into smaller particles than raw biomass during pelletization, thus resulting in larger contact area and higher frictional energy consumption than other samples. In addition, the residue of cellulose degradation may coat the surface of hemicellulose and lignin, which reduced the plasticity of samples (Li et al., 2012). It was widely accepted that oil can be added into biomass as a lubricant in order to reduce the energy consumption during compaction (Ahn et al., 2014b; Shang et al., 2013). In this study, the caster bean cake (a kind of oil-compressed byproduct) was 12
added into the untreated and torrefied samples as a potential additive. An obvious reduction of energy consumption was obtained in Fig. 3a and b. The residual oil in CAS could act as a lubricant, and thus reduced the frictions among sample particles during pelletization. In addition, starch and protein in CAS can provide hydrogen bonding enhancing the plasticity of sawdust (Li et al., 2012). Those are responsible for the decrement of energy consumption during pelletization. Similar observations were found in the previous studies, where the additives of rapeseed oil and rapeseed cake were added as lubricant (Stahl et al., 2011; Shang et al. 2014). Moreover, Fig. 3a shows that the largest reduction of energy consumption was observed for the CED torrefied at 300 oC by adding CAS. Energy consumption is ascribed to the barrier that overcomes friction and crushes particles. The torrefaction can lower particles strength resulting in the reduction of the crush particles energy input (Peng et al., 2012). Furthermore, bio-oil, starch and protein in CAS can decrease the energy consumption by reducing the friction among particles and enhancing the plasticity of sawdust. The large reduction of energy consumption indicated that torrefaction and co-pelletization with CAS had complementary effect on reducing the energy consumption markedly. As shown in Fig. 3c and d, energy consumption of extrusion increased from 0.07 J/g to 0.38 J/g for CED and 0.01 J/g to from 0.27 J/g for CAM with the increasing degree of torrefaction, respectively. As a result, torrefied biomass had higher energy consumption of extrusion than those of untreated biomass under the same condition. The increase of extrusion energy consumption can be explained by the higher friction mentioned in compaction energy consumption between particles and die. In our study, 13
the results show that adding CAS can reduce energy consumption for extrusion (Fig. 3c and d). This indicates that adding CAS can reduce the friction during pelletization by squeezing the oil into gaps between pellet and die as a lubricant. 3.3 Moisture absorption In the pellets from untreated materials, the equilibrium moisture absorption (EMA, shown in Tbale 4) was 16.47% for CED and 13.81% for CAM, respectively. Decreasing trends of the EMA were observed for CED and CAM through torrefaction, indicating that the torrefied biomass was characterized by hydrophobicity (Kaliyan & Morey, 2010). The EMA was reduced with increasing of the torrefaction temperature in the range of 240-300 oC. This may be due to the removal of extractives (protein, starch and carbonhydrates) and the decomposition of hemicellulose and lignin, thus leading to the loss of hydrotropic substances during torrefaction. In addition, the decomposed intermediate compounds coating on the hydrotropic substances were also regarded as another reason for the decrease of EMA (Kaliyan & Morey, 2010). As shown in Table 4, the EMA was increased slightly with the increasing content of CAS, while the moisture absorption rates decreased with the increasing CAS content. The increase of EMA can be explained by the presence of protein and starch in CAS, which belong to hydrotropic substances and can absorb water with its hydrophilic groups (Kaliyan & Morey, 2010). Meanwhile, the protein and starch in CAS were squeezed into the gaps and voids between particles to form the close-knit structure, which prevented water
14
vapour from entering into the pellets in the first five hours. The close-knit structure of pellets was further comfirmed by the SEM (Fig. S1 in Supplementary data). 3.4 Pellets density Pellets density is a critical factor affecting the energy density and the storage/transportation cost of biomass pellets. As shown in Fig. 4, the pellets density significantly decreased from 1g/cm3 to 0.86 g/cm3 for CED and from 1.05 to 0.84 g/cm3 for CAM with the increase of torrefaction temperature, respectively. In addition, CED torrefied at 300 oC was hard to be compacted into pellet under set condition. The loss of chemically bonded water and low-melting point compounds which acted as binding agents were considered as the reasons for lower densities of pellets from torrefied biomass. This can also be confirmed by the SEM of torrefied pellets shown in Fig. S1c and e (in Supplementary data), in which clear gaps were identified. As shown in Fig. 4, the pellets density increased with the increasing content of CAS. Furthermore, larger increment of pellets density for torrefied samples was awarded in comparison to the untreated samples. As discussed in section 3.3, the close-knit structure reinforced by CAS led to the elimination of gaps and voids between particles (Fig. S1 in Supplementary data). Therefore, compelmentary of torrefaction and co-pelletization can resist the negative effect of torrfaction on pellets density. 3.5 Pellets strength In order to evaluate the effect of torrefaction and co-pelletization on pellets strength, collapsing force and Meyer hardness for selected pellets were investigated (Fig. 5 and Table 4). Meyer hardness of the raw CED and CAM pellets were 3.57 15
N/mm2 and 4.65 N/mm2, respectively. The clear decreasing trends of Meyer hardness were observed for both biomass materials through torrefaction, which were consistent with the previous work (Li et al., 2012). The pellets made of CED torrefied at 300 oC could not be achieved under the previous condition, while the Meyer hardness of CAM pellets was 2.9 N/mm2. Furthermore, collapsing forces of the pellets without CAS were significantly decreased with the increasing torrefaction temperature. In this study, the lower Meyer hardness of torrefied pellets indicated the lack of binders and solid bridges between torrefied particles, which could be well explained by the following three aspects. Firstly, after biomass torrefaction, the decomposition of hemicellulose/lignin and the removal of extractives (proteins, starch and carbonhydrates) reduced the hydrogen bonding and solid bridges among particles resulting in the internal voids and spaces in the pellets. During strength test, those voids and spaces promoted the relative movement of particles in the matrix of pellets, leading to the lower strength of pellet. In addition, the residual ash and the intermediate products formed from decomposition of hemicellulose/lignin and extractives deposited on the surface of biomass particles. The deposition of those ash and intermediate products could prevent the formation of solid bridge and interlock, which can be confirmed by the gaps and voids in Fig. S1c and e (in Supplementary data). Secondly, the moisture in biomass played an important role in the attractive forces of adjacent particles (Ahn et al., 2014a). The moisture adsorption capacity of biomass decreased considerably for biomass through torrefaction (Table 4), leading to the quantitative decrement of weak hydrogen bonding sites during pelletization. 16
Lastly, the grindability of biomass was enhanced through torrefaction as the decomposition of hemicellulose, cellulose and lignin took place (Arias et al., 2008). It was thus more easily for the torrefied biomass to be crushed into fines during pelletization (Arias et al., 2008). As a result, the joints between adjacent torrefied particles were enlarged in comparison to that of raw samples. The enlarged joints leaded to the increase of gaps and voids because of the absence of binder and the weak of attractive force in the joints. However, the collapsing force for the CAM pellets produced from sawdust torrfied at 240 oC was increased to 48.23 N, compared to 40.46 N for the raw CAM pellets. The removal of fatty compounds compounds during torrefaction at 240 oC may reduce the relative movement of particles within the pellets. In a previous report, the strength of pellets produced from mixture of rapeseed cake and biomass decreased with an increased amount of rapeseed cake, which was ascribed to the lubrication capacity of extractive matters in rapeseed cake (Stahl et al., 2011). Nevertheless, a different result was achieved in the present work as the addition of CAS into biomass (in Table 4 and Fig. 5). Compared with the liquid fraction of rapeseed oil, the soft-solid protein and starch compounds in CAS can be compressed into the gaps and voids of particles during pelletization. Those compressed protein and starch compounds can act as the binders and form solid bridge and hydrogen bonding, leading to an improved close-knit structure of matrix in pellets. This can be confirmed by the SEM (Fig. S1 in Supplementary data) and the higher density of pellets (Fig. 4). Furthermore, the enlarged joints between adjacent 17
particles were filled and connected by the proteins and starch in CAS, contributing to a further enhancement of solid bridge and attractive forces among particles (Jiang et al., 2014; Kaliyan & Morey, 2009). Therefore, the compelmentary of torrefaction and co-pelletization can improve the pellets strength. 3.6 Combustion characteristics The combustion characteristic is subjected to thermal analysis under air combustion conditions, which is necessary in evaluating the application properties of fuels for future applications. Higher weight loss was observed for raw samples than torrefied samples during the initial stage of combustion in the temperatures range of 30 to 105 oC (Table 5). Generally, the initial mass loss is attributed to the dehydration in samples. The torrefied samples consequently contain less moisture in comparison to raw samples due to the initial dehydration of biomass during torrefaction (Yi et al., 2012). Meanwhile, as depicted in Fig. 2c and d, different curves were shown between raw samples and torrefied samples, especially for the heavy torrefied samples, which were consistent with the previous study (Chen & Kuo, 2011). It can be due to the compositions variations of hemicellulose, celluloses and lignin after torrefaction (Table 3). It was suggested by Peng et al. (2012) that the main loss of hemicellulose occurred at 220-315 oC compared with the loss of cellulose primarily at 315-400 oC. The lignin decomposed slowly over a wide temperature range with the maximum decomposition rate obtained at around 460 oC (Zhou et al., 2014). As shown in Table 5, the ignition temperature and final combustion temperature increased with the increasing degree of torrefaction. The maximum DTG decreased at stage 1 and then 18
increased at stage 2 with the increasing torrefaction temperature (Biswas et al., 2014). As shown in Table 3, the change of maximum DTG at stage 2 is in consistent with the variation of carbon fraction with the increase of torrefaction temperature. In Table 5, the ignition temperature and maximum DTG of pellets decreased with the increasing CAS ratio. High maximum DTG had a harmful effect on combustibility (Yi et al., 2012). As a result, a more stable combustion was achieved by adding CAS. However, in stage 1, the maximum DTG was increased with the increasing content of CAS for the untreated and lightly torefied samples at 240 oC. It was indicated that the residual oil in CAS was able to accelerate the burning at stage 1, in which hemicellulose and cellulose were the primary combustible components (Zhou et al., 2014). It was found that the combustion characteristics were highly affected by the degree of torrefaction and the ratio of CAS. Table 5 shows that higher residue was left with the increasing content of CAS, which is due to the high ash contents in CAS (8.79%). 4. Conclusions The process of torrefaction can improve the hydrophobicity of biomass pellets as the decomposition of protein and hemicellulose occured. Meanwhile, co-pelletization with caster bean cake can reduce the high energy consumption and weak pellets strength caused by torrefaction. The reduction of energy consumption is attributed to the well grindability of torrefied biomass and the lubricant of caster bean cake. Moreover, the high density and Meyer hardness of pellets by co-pelletization of torrefied biomass were caused by the enlarged contact area and thus enhanced attractive forces between particles. The economic cost budget should be investigated 19
in further large-scale production. Acknowledgements The authors gratefully acknowledge the financial support provided by the National Natural Science Foundation of China (No. 21407046, 31470594), the Natural Science Foundation of Hunan Province, China (No. 13JJ4118), Key Laboratory of Renewable Energy Chinese Academy of Sciences (No.y407k91001), and
Collaborative
Innovation
Center
of
Resource-Conserving
Environment-Friendly Society and Ecological Civilization.
20
&
References Ahn, B.J., Chang, H.S., Lee, S.M., Choi, D.H., Cho, S.T., Han, G.S., Yang, I., 2014a. Effect of binders on the durability of wood pellets fabricated from Larix kaemferi C. and Liriodendron tulipifera L. sawdust. Renewable Energy, 62, 18-23. Andreola, F., Martı´n, M.I., Ferrari, A.M., Lancellotti, I., Bondioli, F., Rinco´ n, J.M., Romero, M., Barbieri, L., 2013. Technological properties of glass-ceramic tiles obtained using rice husk ash as silica precursor. Ceramics International, 39, 5427-5435. Arias, B., Pevida, C., Fermoso, J., Plaza, M.G., Rubiera, F., Pis, J.J., 2008. Influence of torrefaction on the grindability and reactivity of woody biomass materials. Fuel Processing Technology, 89(2), 169-175. Bach, Q. V., Tran, K. Q., Skreiberg, Ø., Khalil, R.A., Phan, A.N., 2014. Effects of wet torrefaction on reactivity and kinetics of wood under air combustion conditions. Fuel, 137, 375-383. Basu, P., Sadhukhan, A.K., Gupta, P., Rao, S., Dhungana, A., Acharya, B., 2014. An experimental and theoretical investigation on torrefaction of a large wet wood particle. Bioresour Technol, 159, 215-22. Biswas, A.K., Rudolfsson, M., Broström, M., Umeki, K. 2014., Effect of pelletizing conditions on combustion behaviour of single wood pellet. Applied Energy, 119, 79-84. Chau, J., Sowlati, T., Sokhansanj, S., Preto, F., Melin, S., Bi, X., 2009. Economic sensitivity of wood biomass materials utilization for greenhouse heating application. Applied Energy, 86(5), 616-621. Chen, W. H., Kuo, P. C., 2011. Torrefaction and co-torrefaction characterization of hemicellulose, cellulose and lignin as well as torrefaction of some basic constituents in biomass 21
materials. Energy, 36(2), 803-811. Chen, W.H., Cheng, W.Y., Lu, K.M., Huang, Y.P., 2011. An evaluation on improvement of pulverized biomass materials property for solid fuel through torrefaction. Applied Energy, 88(11), 3636-3644. Chen, Y., Yang, H., Yang, Q., Hao, H., Zhu, B., Chen, H., 2014. Torrefaction of agriculture straws and its application on biomass materials pyrolysis poly-generation. Bioresour Technol, 156, 70-7. Chin, K.L., H’ng, P.S., Go, W.Z., Wong, W.Z., Lim, T.W., Maminski, M., Paridah, M.T., Luqman, A.C., 2013. Optimization of torrefaction conditions for high energy density solid biofuel from oil palm biomass materials and fast growing species available in Malaysia. Industrial Crops and Products, 49, 768-774. Du, S.W., Chen, W.H., Lucas, J.A., 2014. Pretreatment of biomass materials by torrefaction and carbonization for coal blend used in pulverized coal injection. Bioresour Technol, 161, 333-9. Iroba, K.L., Tabil, L.G., Sokhansanj, S., Dumonceaux, T., 2014. Pretreatment and fractionation of barley straw using steam explosion at low severity factor. Biomass materials and Bioenergy, 66, 286-300. Jiang, L., Liang, J., Yuan, X., Li, H., Li, C., Xiao, Z., Huang, H., Wang, H., Zeng, G., 2014. Co-pelletization of sewage sludge and biomass materials: The density and hardness of pellet. Bioresour Technol, 166, 435-43. Kaliyan, N., Morey, R. 2009. Factors affecting strength and durability of densified biomass materials products. Biomass materials and Bioenergy, 33(3), 337-359. 22
Kaliyan, N., Morey, R.V., 2010. Natural binders and solid bridge type binding mechanisms in briquettes and pellets made from corn stover and switchgrass. Bioresour Technol, 101(3), 1082-90.Lam, P.S., Sokhansanj, S., Bi, X.T., Lim, C.J., Larsson, S.H., 2012. Drying characteristics and equilibrium moisture content of steam-treated Douglas fir (Pseudotsuga menziesii L.). Bioresour Technol, 116, 396-402. Li, H., Liu, X.H., Legros, R., Bi, X.T., Lim, C.J., Sokhansanj, S., 2012. Pelletization of torrefied sawdust and properties of torrefied pellets. Applied Energy, 93, 680-685. Li, J., Brzdekiewicz, A., Yang, W., Blasiak, W., 2012. Co-firing based on biomass torrefaction in a pulverized coal boiler with aim of 100% fuel switching. Applied Energy, 99, 344-354. Liu, Z., Quek, A., Balasubramanian, R., 2014. Preparation and characterization of fuel pellets from woody biomass materials, agro-residues and their corresponding hydrochars. Applied Energy, 113, 1315-1322. Medic, D., Darr, M., Shah, A., Rahn, S., 2012. Effect of Torrefaction on Water Vapor Adsorption Properties and Resistance to Microbial Degradation of Corn Stover. Energy & Fuels, 26(4), 2386-2393. Mišljenović, N., Bach, Q. V., Tran, K. Q., Salas-Bringas, C., Skreiberg, Ø., 2014. Torrefaction Influence on Pelletability and Pellet Quality of Norwegian Forest Residues. Energy Fuels, 28, 2554-2561. Nunes, L.J., Matias, J.C., Catalão, J.P., 2014. Mixed biomass materials pellets for thermal energy production: A review of combustion models. Applied Energy, 127, 135-140. Oyedun, A.O., Tee, C.Z., Hanson, S., Hui, C.W., 2014. Thermogravimetric analysis of the pyrolysis characteristics and kinetics of plastics and biomass blends. Fuel Processing 23
Technology, 128, 471-481. Pelaez-Samaniego, M.R., Yadama, V., Garcia-Perez, M., Lowell, E., McDonald, A.G., 2014. Effect of temperature during wood torrefaction on the formation oflignin liquid intermediates. Journal of Analytical and Applied Pyrolysis, 109, 222-233. Peng, J.H., Bi, X.T., Lim, J., Sokhansanj, S., 2012. Development of Torrefaction Kinetics for British Columbia Softwoods. International Journal of Chemical Reactor Engineering, 10. Poddar, S., Kamruzzamana, M, Sujan, S.M., Hossain, M., Jamal, M.S., Gafur, M.A. Khanam, M., 2014. Effect of compression pressure on lignocellulosic biomass materials pellet to improve fuel properties: Higher heating value. Fuel, 131, 43-48. Reza, M. T., Uddin, M. H., Lynam, J. G., Coronella, C. J., 2014. Engineered pellets from dry torrefied and HTC biochar blends. Biomass and Bioenergy, 63, 229-238. Saleh, S.B., Hansen, B.B., Jensen, P.A., Johansen, K.D., 2013. Efficient Fuel Pretreatment: Simultaneous Torrefaction and Grinding of Biomass. Enegy Fules, 27, 7531-7540. Shang, L., Nielsen, N.P.K., Stelte, W., Dahl, J., Ahrenfeldt, J., Holm, J.K., Arnavat, M.P., Bach, L.S., Henriksen, U.B., 2013. Lab and Bench-Scale Pelletization of Torrefied Wood Chips—Process Optimization and Pellet Quality. BioEnergy Research, 7(1), 87-94. Tabil, L.G., Sokhansanj, S., Crerar1, W.J., Patil, R.T., Khoshtaghaza, M.H., Opoku1, A., 2002. Physical characterization of alfalfa cubes: I. Hardness. Canadian biosystem engineering, 44,55-63. Wen, J.L., Sun, S.L., Yuan, T.Q., Xu, F., Sun, R.C., 2014. Understanding the chemical and structural transformations of lignin macromolecule during torrefaction. Applied Energy, 121, 1-9. 24
Xu, S.C., He, Z.X., Long, R.Y., 2014. Factors that influence carbon emissions due to energy consumption in China: Decomposition analysis using LMDI. Applied Energy, 127, 182-193. Yi, Q., Qi, F., Cheng, G., Zhang, Y., Xiao, B., Hu, Z., Liu, S., Cai, H., Xu, S., 2012. Thermogravimetric analysis of co-combustion of biomass materials and biochar. Journal of Thermal Analysis and Calorimetry, 112(3), 1475-1479. Zhou, C., Liu, G., Cheng, S., Fang, T., Lam, P.K., 2014. Thermochemical and trace element behavior of coal gangue, agricultural biomass materials and their blends during co-combustion. Bioresour Technol, 166, 243-51.
25
Figure captions Fig. 1 The schematic and flow diagram of torrefaction and co-pelletization. Fig. 2 TG and DTG curves of raw and torrefied pellets under a nitrogen flow rate of 50 mL/min and a air flow rate of 100 mL/min at a heating rate of 15 oC/min. Fig. 3 The energy consumption of compaction and extrusion associated with the degree of torrefaction and the additive ratios CAS. Fig. 4 Effects of the CAS ratio and the degree of torrefaction on pellets density. Fig. 5 Meyer hardness of pellets as a function of the mixing ratio of CAS and the torrefaction temperature.
Fig. 1 The schematic and flow diagram of torrefaction and co-pelletization.
26
Fig. 2 TG and DTG curves of raw and torrefied pellets under a nitrogen flow rate of 50 mL/min and a air flow rate of 100 mL/min at a heating rate of 15 oC/min. (a) CED pellets in nitrogen ; (b) CAM pellets in nitrogen (c) CED pellets in air ; (d) CAM pellets in air.
27
Fig. 3 The energy consumption of compaction and extrusion associated with the degree of torrefaction and the additive ratios CAS. (a) compaction for CED pellets; (b) compaction for CAM pellets (c) extrusion for CED pellets; (d) extrusion for CAM pellets.
28
Fig. 4 Effects of the CAS ratio and the degree of torrefaction on pellets density. (a) CED pellets; (b) CAM pellets.
29
Fig. 5 Meyer hardness of pellets as a function of the mixing ratio of CAS and the torrefaction temperature. (a) CED pellets; (b) CAM pellets.
30
Table 1 Weight loss and particle size distribution of raw and torrefied samples. Sample
Weight loss
Size distribution 1000-840um
840-420um
420-180um
<180um
CED Raw
0
5.73
42.32
38.49
13.46
240oC
7.69
4.28
37.89
42.33
15.51
270oC
19.92
3.78
35.18
42.56
18.49
300oC
33.54
2.91
38.05
42.82
16.22
Raw
0
8.09
59.68
27.48
4.76
240oC
13.72
5.93
57.14
31.34
5.58
270oC
16.81
7.16
56.30
30.90
5.63
300oC
34.00
2.91
38.05
42.82
16.22
CAM
31
Table 2 The proximate and ultimate analysis of raw and torrefied samples expressed on a dry basis. CED
Raw
o
CAM o
o
o
240 C
270 C
300 C
Raw
240 C
270oC
300oC
CAS
Proximate analysis (%)
VM
81.86
81.96
79.25
65.34
85.51
82.62
79.20
70.91
69.8
FC
17.25
17.12
18.76
32.64
13.42
16.21
19.36
26.55
21.41
A
0.89
0.92
1.99
2.02
1.07
1.17
1.44
2.54
8.79
Elemental analysis (%) C
48.95
54.74
53.41
61.02
48.18
51.02
53.05
61.94
44.21
H
5.91
5.76
6.02
5.28
6.09
6.19
6.06
5.02
5.93
O
44.49
39.22
40.33
33.49
45.03
42.37
40.60
32.81
41.94
N
0.65
0.28
0.25
0.22
0.70
0.41
0.30
0.23
7.92
O/C ratio
0.91
0.72
0.76
0.54
0.93
0.83
0.77
0.53
0.95
HHV (MJ/kg)
19.57
19.87
20.61
23.17
18.86
18.90
20.45
22.35
19.12
.
32
Table 3 Fiber analysis of the raw and torried biomass using cellulose tester. Sample
Composition H
a
b
c
Composition change d
C
L
E
H
a
Cb
Lc
Ed
CED
Raw
12.05
36.22
27.61
24.12
0
0
0
0
240 oC
10.51
45.84
35.69
7.96
-12.78
26.56
29.26
-67.00
270 oC
7.37
46.59
39.59
6.45
-38.84
28.63
43.39
-73.26
300 oC
4.04
24.52
66.38
5.06
-66.47
-32.30
140.42
-79.02
Raw
20.82
38.87
24.4
15.91
0
0
0
0
240 oC
18.39
48.47
23.77
9.37
-11.67
24.70
-2.58
-41.11
270 oC
9.7
52.69
27.77
9.84
-53.41
35.55
13.81
-38.15
300 oC
4.63
40.22
49.58
5.57
-77.76
3.47
103.20
-64.99
CAS
28.98
20.73
23.75
44.54
CAM
a
Hemicelluloses Cellulose c Lignin d Extractives b
33
Table 4 Data obtained in connection with the equilibrium moisture absorption, moisture absorption rates (k) and collapsing force of raw and torrefied pellets. CED Samples
Raw
CAM
EMA (%)c
k (min-1)
Collapsing force (N)
EMA (%)c
k (min-1)
Collapsing force (N)
16.47
0.00534
56.36
13.81
0.01028
40.46
a
Raw
17.34
0.00535
63.46
13.91
0.0095
55.33
Rawb
17.09
0.00479
73.63
13.33
0.00851
66.36
o
12.86
0.00554
35.53
11.86
0.00788
48.23
13.31
0.00566
43.3
11.26
0.00768
52.2
14.49
0.00485
48.21
11.94
0.00682
72.06
240 C 240 oC a o
240 C
b
o
270 C
12.00
0.00504
19.23
9.52
0.00773
47.9
o
a
13.86
0.00459
44.83
10.37
0.00777
64.6
o
b
14.44
0.00425
58.26
10.49
0.0069
79.13
10.46
0.00414
0d
9.21
0.00755
31.73
a
11.49
0.00422
19.46
10.21
0.0073
67.1
300 oC b
13.01
0.00445
35.96
10.25
0.00708
128.61
270 C 270 C 300 oC o
300 C
a
Mixed with 15% CAS (dry basis)
b
Mixed with 30% CAS (dry basis)
c
Equilibrium moisture absorption d Pellet cannot be compact under150 oC, 4000N
34
Table 5 Combustion characteristics of raw biomass, torrefied biomass and their blends with CAS. Stage 1 Samples
Stage 2
Residue (%)
Range (oC)
Tmax (oC)
DTGmax (% s-1)
Range (oC)
Tmax (oC)
DTGmax (% s-1)
Raw
202-402
340
0.54
402-487
467
0.53
0.94
Rawa
200-378
332
0.33
378-516
429
0.46
3.09
Rawb
215-374
346
0.98
374-506
452
0.04
1.67
240 C
238-405
343
0.47
405-498
478
0.61
1.46
240 oC a
226-368
357
0.97
368-507
476
0.15
1.87
240 oC b
229-367
347
1.06
367-505
465
0.09
3.35
270 C
247-402
340
0.37
402-506
471
0.65
2.33
270 oC a
216-388
338
0.47
388-511
444
0.55
1.92
270 oC b
224-368
355
0.94
368-507
463
0.09
2.56
300 C
253-388
329
0.30
388-508
458
0.79
1.35
300 oC a
227-379
326
0.21
379-525
445
0.36
3.46
300 oC b
222-383
326
0.25
383-523
441
0.38
2.32
Raw
222-391
334
0.37
391-500
451
0.34
0.37
Rawa
206-363
346
0.88
363-482
454
0.08
0.7
b
Raw
216-392
335
0.29
392-506
457
0.33
1.46
240 oC
223-396
331
0.36
396-510
450
0.20
1.97
240 oC a
220-357
334
0.74
357-491
442
0.08
1.66
240 oC b
223-383
333
0.37
383-494
445
0.39
3.62
270 C
236-384
332
0.56
384-513
440
0.18
1.14
270 oC a
233-357
349
0.91
357-508
453
0.11
1.57
270 oC b
228-350
330
0.58
350-523
438
0.36
3.60
300 C
233-375
326
0.33
375-508
450
0.68
1.87
300 oC a
233-360
331
0.45
360-493
445
0.49
4.38
300 oC b
224-376
328
0.22
376-520
446
0.18
3.48
CED
o
o
o
CAM
o
o
a
Mixed with 15% CAS (dry basis)
b
Mixed with 30% CAS (dry basis)
35
Highlights
Caster bean cake improves the features of pellets produced from torrefied biomass.
Grindability and lubricant are available to decrease the energy consumption.
Complementary effect of protein and enlarged contact area improve pellet strength.
Co-pelletizing biomass torrefied at 270 oC with 15% CAS are the optimal conditions.
36