- Email: [email protected]
Production of bio-oil from fast pyrolysis of biomass using a pilot-scale circulating fluidized bed reactor and its characterization
Journal of Environmental Management 234 (2019) 138–144
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Research article
Production of bio-oil from fast pyrolysis of biomass using a pilot-scale circulating fluidized bed reactor and its characterization
T
Jo Yong Parka, Jae-Kon Kima,∗, Chang-Ho Ohb, Jong-Wook Parkb, Eilhann E. Kwonc,∗∗ a
Research Institute of Petroleum Technology, Korea Petroleum Quality & Distribution Authority, Cheongju, 28115, Republic of Korea Daekyung ESCO, Incheon, 21984, Republic of Korea c Department of Environment and Energy, Sejong University, 05006, Republic of Korea b
A R T I C LE I N FO
A B S T R A C T
Keywords: Fast pyrolysis Bio-oil Biomass Biofuel Circulating fluidized bed
To circumvent the adverse impacts arising from an excessive use of fossil fuels, bioenergy and chemical production from a carbon neutral resource (biomass) has drawn considerable attention over the last two decades. Among various technical candidates, fast pyrolysis of biomass has been considered as one of the viable technical routes for converting a carbonaceous material (biomass) into biocrude (bio-oil). In these respects, three biomass samples (i.e., sawdust, empty fruit bunch, and giant Miscanthus) were chosen as a carbon substrate for the pyrolysis process in this study. A pilot-scale circulating fluidized bed reactor was employed for the pyrolysis work, and biocrude from the fast pyrolysis process at 500 °C were characterized because the maximum yield of biocrude (60 wt% of the original sample mass) was achieved at 500 °C. The physico-chemical properties of biocrude were measured by the international standard/protocol (ASTM D7544 and/or EN 16900 test method) to harness biocrude as bioenergy and an initial feedstock for diverse chemicals. All measurements in this study demonstrated that the heating value, moisture content, and ash contents in biocrude were highly contingent on the type of biomass. Moreover, characterization of biocrude in this study significantly suggested that additional unit operations for char and metal removal must be conducted to meet the fuel standard in terms of biocrude as bioenergy.
1. Introduction Despite the diverse socio-economic benefits from harnessing fossil fuels as the primary energy resources, the excessive use of fossil fuels has brought forth global environmental problems (Naqvi et al., 2014). Note that carbon imbalance from combustion of fossil fuels has been regarded as one of the major driving forces for climate change and global warming in that total carbon inputs are far exceeding the full Earth's capacity to assimilate carbon through the natural carbon cycle (Tsai et al., 2006). To mitigate the global environmental burdens, obtaining bioenergy and chemicals from the carbon neutral resources (biomass) has gain considerable attention over the last two decades. Moreover, among various technical candidates, fast pyrolysis has been considered as one of the viable technical routes for converting biomass into the liquid-type fuels (Naqvi et al., 2014; Tsai et al., 2006; Qureshi et al., 2018). In detail, fast pyrolysis bio-oil (FPBO) converts biomass in biocrude under the moderate pyrolytic temperatures (450–650 °C) (Williams and Nugranad, 2000; Qureshi et al., 2018). In general, the
∗
yield of FPBO reaches up to 75 wt% of the original mass of biomass (dry mass basis), of which calorific value reaches up to 60–75% total calorific value of virgin biomass (dry basis) (Isahak et al., 2012). In addition to FPBO, the yield of char and non-condensable gas from fast pyrolysis of biomass reaches up to 10–20 and 10–25 wt% of the original sample mass, respectively. The compositional matrix of pyrogenic products from the fast pyrolysis is also highly contingent on the type of biomass. Thus, various biomasses (wood, corn stalks, agricultural waste, switch grass, forest waste, etc.) were tested for the fast pyrolysis process. The economic feasibility for the fast pyrolysis process of biomass done by other authors (Sembiring et al., 2015; Sulaiman and Abdullah, 2011) also suggested that fast pyrolysis of biomass using the small scale (50–100 tons per day) pyrolysis process shows the highest economic benefits in that a small-scale pyrolysis of biomass does not requires a high tipping fee for biomass and capital cost for a pyrolysis platform (Guedes et al., 2018; Qureshi et al., 2018). FPBO generally exhibits a complex compositional matrix, of which constituents include more than 300 chemical species that are derived
Corresponding author. Corresponding author. E-mail addresses: [email protected] (J.-K. Kim), [email protected] (E.E. Kwon).
∗∗
https://doi.org/10.1016/j.jenvman.2018.12.104 Received 31 August 2018; Received in revised form 22 December 2018; Accepted 26 December 2018 0301-4797/ © 2018 Elsevier Ltd. All rights reserved.
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
from depolymerization of the major component (i.e., cellulose, hemicellulose, and lignin) of biomass (Kanaujia et al., 2014). The typical oxygen content in biocrude reaches up to 40–50 wt% (Isahak et al., 2012). Therefore, biocrude has a low calorific value of 16–18 MJ/kg. Biocrude has the low pH value, of which pH is roughly equivalent to 2.5, and the viscosity of biocrude relative to fossil fuel is high (Sipilä et al., 1998). The moisture content in biocrude reaches up to 25–35 wt % (Isahak et al., 2012). Hence, the fuel properties of biocrude are inferior to that of diesel or gasoline. As a result, biocrude cannot be used directly in an internal combustion engine (Nogueira, 2012; Djokic et al., 2012; Maciel et al., 2016). Particularly, the high contents of water and oxygen, high viscosity, and the high content of ash lead to poor ignition and emission (Isahak et al., 2012). Such impurities also the main driving force for clogging (Buffi et al., 2018). Based on these rationales, biocrude is being used in an internal combustion engine or combustor after the proper upgrading steps (Chiaramonti et al., 2007; Buffi et al., 2018). FPBO was produced by employing a pilot-scale (1 barrel per day: bpd) circulating fluidized bed reactor with three different biomass samples (i.e., sawdust, empty fruit bunch (EFB), and giant Miscanthus) in this study. The produced FPBO was further characterized by the international stand (ASTM D 7544 and EN 16900 specifications). The further extraction of FPBO was carried out to figure out that the desired fuel properties for usefulness of biocrude was achieved in this study. 2. Materials and methods 2.1. Fast pyrolysis bio-oil production A circulating fluidized bed (CFB) reactor was used for fast pyrolysis of biomass, of which biomass feeding rate was adjusted as 42 kg h−1. The overall schematic layout for the pyrolysis system was presented in Fig. 1, and the major parts of the CFB reactor were labelled. In detail, as denoted in Fig. 1, the pyrolysis system is composed of a heating chamber surrounded by an electrical furnace, a biomass hopper, a screw feeder, a fluidized bed reactor, a cyclone, two condensers and an electrostatic precipitator. There are three processes in the CFB reactor: I) the fast pyrolysis reaction process, II) the oil recovery process, where the produced gas is rapidly condensed for oil recovery, and III) gas recirculation and excess gas incineration, where the non-condensed gas is recirculated, and excess gas is incinerated.
Fig. 1. (a) Schematic layout for a pilot-scale circulating fluidized bed (CFB) reactor, (b) a real pyrolytic system to produce FPBO.
Table 1 Operational parameters for the pilot-scale CFB reactor.
2.2. Biomass preparation Two biomass samples (sawdust and giant Miscanthus) were obtained from the Rural Development Administration (RDA) in Korea. EFB was imported from Indonesia. Prior to fast pyrolysis, three samples were mechanically milled, of which particle size was adjusted as 2–3 mm. 2.3. Fast pyrolysis of three different biomass samples The operational parameters for the pilot-scale CFB reactor was summarized in Table 1. Note that the production capacity is 1 barrel per day (1 bpd). To maintain the CFB conditions, 5 kg sand was added into the reactor. The gas flow rate was 3 m s−1, and biomass feeding rate was adjusted as 42 kg h−1.
Parameter
Average
Reaction temperature (˚C) Recycle ratio (%) Gas velocity (m s−1) Blower (Hz) Wind box press (mmAq) Top of reactor press (mmAq) Feed ratio (kg h−1) 1st cyclone temperature (˚C) 2nd cyclone temperature (˚C) Quencher temperature (˚C) EP temperature (˚C)
500 80 3 19 2100 50 42 450 350 33 23
2.5. FPBO extraction (purification) Diethyl ether was used to separate moisture from virgin FPBO from the primary and secondary oil recovery processes for extraction (Fig. 2). FPBO and the ether solution were mixed in a ratio of 1–3, and the diethyl ether and water layers were separated using a separatory funnel. The separated diethyl ether layer contains the hydrophilic components of the oil while the water layer contains moisture and the water-soluble components. The upper layer was carefully collected and dried in a rotary evaporator (30 °C and 30 min) to obtain the etherextracted bio-oil.
2.4. FPBO recovery FPBO was rapidly cooled to trap the condensable hydrocarbons. In this study, the indirect contact method was adopted. In detail, FPBO was recovered by cooling, and then capturing the remaining FPBO using an electric precipitator (EP). The temperatures in the primary and secondary cooling chambers were maintained at 20 °C. The gaseous products from the fast pyrolysis process are CO, CO₂, CH₄, and N2. While 80% of these non-condensed gases was recirculated, 20% of pyrolytic gases was incinerated to maintain the pyrolytic temperature.
2.6. FPBO characterization The properties of FPBO were characterized to measure the heating 139
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
Table 3 Elemental analyzer conditions. Item
C, H analysis
O analysis
Furnace temperature(°C) Oven temperature (°C) Carrier flow (mL min−1) Reference Flow (mL min−1.) Detector
L: 9000, R: 680 75 140 100 TCD
1060 65 130 100 TCD
was determined via titration with KOH and using phenolphthalein as the indicator. The samples were dissolved in neutralized methanol and titrated with a 0.1 N KOH solution. The flash point of the pyrolysis biooil was determined according to ASTM D 93 using a Pensky-Martens closed-cup tester (APM-7, TANAKA). The pour point of oil is an indication of the lowest temperature at which the oil can be pumped. The pour point of the pyrolysis bio-oil was determined according to ASTM D 97 suing a pour point tester (MPC 602, TANAKA). The elemental composition of the pyrolysis bio-oil was determined using a FLASH2000 Thermo-Elemental Analyzer. The operating conditions were summarized in Table 3. An FT-IR spectrometer (Nicolet 6700 manufactured by Thermo-Scientific) was used for the compositional analysis. The FT-IR spectra of the produced pyrolysis FPBO were recorded in the transmission mode between 4000 and 800 cm−1. Further, TGA was carried out with a Q500 TGA system (TA Instruments). Helium was used as the sweep gas at a flow rate of 100 mL min−1. Approximately 10 mg of the sample was loaded into an aluminum crucible, and the TGA system temperature was programmed to increase from 20 temperature to 800 °C at a rate of 10 °C min−1, followed by an isothermal run for 30 min at 800 °C.
Fig. 2. Liquid–liquid extraction using diethyl ether.
value, moisture content, and kinematic viscosity, of which measurements were in accordance with to the ASTM D 7544 or EN 16900 specification (Table 2). Metal contents in FPBO were quantified using an ICP unit. Also, TGA, IR, and elemental analyses of FPBO were carried out. The gross heating values of crude and dry FPBO were measured using a bomb calorimeter (Parr 6400, PARR Co. Ltd.), of which procedures are in accordance with ASTM D240. The water content of crude and dry FPBO was analyzed, of which protocols were in accordance with ASTM E203 using a Karl Fischer titrator (MKC-610, Kyoto Electronics Manufacturing Co., Ltd.). In detail, 100 g fraction of the sample was filtered at 40 °C through a pre-weighed membrane filter. The filter and residue were washed with methanol, dried, and weighed, and the solid content of the sample was calculated from the difference between their weights. The kinematic viscosity of the pyrolysis bio-oil was measured according to ASTM D445 using a viscometer (CAV2100, Cannon Instrument Company). The density of the pyrolysis bio-oil was determined using a density meter (DMA 5000m, Anton Parr) according to ASTM D4052. An X-ray sulfur analyzer (1800H, Horiba) was used to measure the sulfur content in the bio-oil according to ASTM D4294. To measure the ash content of the bio-oil, 10 g bio-oil was combusted at 700 °C for 5 h in a pre-weighed crucible. The crucible and residue were then cooled and weighed, and the ash content was calculated from difference in their masses. The total acid number of the pyrolysis bio-oil
3. Results and discussion 3.1. FPBO production Fast pyrolysis of biomass was carried out using the pilot-scale CFD reactor, and the compositional elemental matrix for each biomass sample were summarized in Table 4. As summarized in Table 4, the content of carbon in each biomass sample is not much different, but the carbon contents were in the following order: sawdust < EFB < giant Miscanthus. The moisture content was in the following order: sawdust < giant Miscanthus < EFB. Based on these characteristics (the contents of C and H), the heating values were in the following order: giant Miscanthus > sawdust > EFB. Fast pyrolysis of biomass was conducted from 400 to 600 °C, and the overall mass balance for three pyrogenic products (gas, liquid, and char) was illustrated in Fig. 3. The overall mass balance suggests that each mass portion of three pyrogenic products is highly affected by the
Table 2 Standard of FPBO in EU and US. Characteristic
ASTN D 7544 Grade G
Calorific value, MJ/kg Water content, wt% Solids content, wt% Kinematic viscosity, mm2/s Density at 20 °C, kg/dm3 Sulfur content, wt% Ash content, wt% pH Flash point, °C Pour Point, °C Nitrogen content, wt% Na, K, Ca, Mg, wt%
> 15 < 30 < 2.5 < 125 1.1–1.3 < 0.05 < 0.25 Report > 45 < -9 – –
prEN16900 Grade D
< 0.25
< 0.15
140
Method
Grade 1
D240 E203 D7579 D445 D4052 D4294 D482 E70 D93 D97 – –
> 14.0 < 30 < 2.5 < 125 < 1.3(at 15 °C) < 0.1 < 0.25 > 2.0 – < -9 report NA
Grade 2
< 0.5 < 50 < 0.05 < 0.05
< 0.02
Method DIN 51900-3 E203 D7579 EN ISO 3104 EN ISO 12185 EN ISO 20846 EN ISO 6245 E70 – ISO 3016 D5291 EN ISO 16476
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
Table 4 The proximate and ultimate analysis of different biomass. Characteristic(wt%)
EFB
Sawdust
Giant Miscanthus
Carbon Hydrogen Nitrogen Sulphur Chlorine Oxygen Volatile matter Ash Moisture Fixed carbon HHV(MJ/kg)
49.1 5.8 0.5 ND ND 41.4 65.0 1.5 15.2 18.3 16.4
43.2 7.9 1.0 0.6 ND 45.1 70.0 5.9 9.6 14.5 17.6
51.3 6.2 0.4 ND ND 42.2 67.5 1.5 11.2 19.8 19.2
operational temperatures. The decrease in the char yield with increasing temperature could be due to secondary decomposition of the char (Williams and Nugranad, 2000). Such observations are good agreement with the preliminary test and the general thermal degradation behaviors of biomass (Liu and Phillips, 1991; Adahchour et al., 2006; Sfetsas et al., 2011). The product yields from three biomass samples at 500 °C are almost identical to the case of sawdust, of which mass portion is equivalent to 60% liquid, 15% char and 25% noncondensable gases. Thus, all fast pyrolysis of three biomass samples was conducted at 500 °C to maximize the yield of FPBO. Fig. 4 depicts the compositional matrix for the non-condensed gas from 450 to 550 °C. Note that the optimal operating temperature for fast pyrolysis of biomass is 500 °C. Among pyrolytic gases, ≥ 80% of the total volume was composed of CO and CO2. Oxygen in FPBO was mostly converted to CO and CO2 when the reaction temperature was low (Williams and Nugranad, 2000). The contents of CO and CO2 were decreased with increasing temperature, while that of hydrogen and methane were increased. 3.2. Pyrolytic bio-oil properties The fuel properties of FPBO produced from sawdust, EFB, and giant Miscanthus at 500 °C were performed, of which measurements were followed by the ASTM D7544 or EN 16900 specification, and the results are shown in Table 5. In Europe and US, the quality standards for FPBO have been set with the goal of it being used for industrial boilers. The heating value of the FPBO produced is related to the water content, which is in the following order: giant Miscanthus > sawdust > EFB. The heating value of FPBO from EFB is lower than that required by the quality standards due to the high moisture content of the bio-oil. The solid content in the FPBO was measured without filtering the produced bio-oil. In the case of sawdust, it was found that very fine particles were not filtered by the cyclone and remained until the latter stage of the process. The shapes of these particles differ depending on the raw material. It has been reported that more small particles are produced when the raw material contains small-sized particles and when pyrolysis is performed with woody biomass rather than grassy biomass (Moraes et al., 2012). The kinematic viscosity varies greatly depending on the moisture content of the fuel, and the kinematic viscosity of EFB was the lowest, as it had the highest moisture content. The moisture content is the most important factor in pyrolysis. The moisture content of FPBO is related to that of the biomass. Regarding the total amount of acid, it has been reported that the carboxylic acid content increases for reactions under high moisture condition (Mantilla et al., 2014). EFB, which had the highest moisture content, also had the highest amount of acid, as shown in Table 5. The ash content of FPBO from sawdust is higher than that required by EN16900 standard due to the high solid content of the bio-oil. Functional group compositional analysis for the FPBO was conducted via FT-IR spectrometry and the results are shown in Fig. 5. The
Fig. 3. Yield of FPBO under various pyrolysis temperatures; (a) EFB, (b) sawdust, (c) Giant Miscanthus.
FPBO contains oxygenated groups. The peak at 3300–3600 cm−1 is due to the OH groups present in water, alcohols, and phenols. This is in agreement with the moisture contents reported in Table 5. Peaks attributable to C]O stretching vibrations between 1650 and 1850 cm−1 were observed in the FT-IR spectra of the FPBO. The CH, CH2, and CH3 functional groups present in alkanes or aromatic aliphatic oxides gave rise to absorption peaks at 2800–3000, 1470, and 1050–1100 cm−1, respectively. The peaks at 1650–1750 cm−1 indicate the presence of single-ring and polycyclic aromatic compounds (Lazzari et al., 2016). Therefore, these peaks are likely attributable to the alkyl groups 141
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
Fig. 6. TGA of FPBO; (a) EFB, (b) sawdust, (c) giant Miscanthus.
Fig. 4. Compositions of non-condensed gas from sawdust.
shown in Fig. 6. The FPBO prepared from sawdust and giant Miscanthus degraded up to 460 °C, and that produced from EFB degraded up to 560 °C. At 900 °C, 10% of the FPBO from giant Miscanthus and EFB, and 14% of the FPBO from sawdust remained as residue. Most of the volatile materials, primarily hemicellulose and cellulose, were degraded at approximately 500 °C. Loss in thermogravimetric mass below 300 °C was attributable to hemicellulose and while that above 300 °C was due to cellulose (Sulaiman and Abdullah, 2011). Lignin decomposed at temperatures above 400 °C (Sembiring et al., 2015).
Table 5 Properties of FPBO produced from different biomass. Characteristics
EFB
Sawdust
Giant Miscanthus
Calorific value, MJ/kg Water content, wt % Solids content, wt % Kinematic viscosity, mm2/s Density at 20 °C, kg/dm3 Sulfur content, wt % Ash content, wt % TAN, mg KOH/g Flash point, °C Pour point, °C
13.2 29.7 2.3 14.5 1.17 0.04 0.1 100.0 46 −22.5
15.9 22.4 4.3 21.7 1.15 0.05 1.3 75.6 49 −21.5
17.0 20.9 0.6 18.7 1.12 0.05 0.2 83.9 47 −17.5
3.3. Fuel properties based on extraction Organic and water-soluble layers can be distinguished when the FPBO is in liquid form. 60% of the FPBO was in the liquid phase. The organic layer must be separated from the water-soluble layer to upgrade the liquid FPBO. In this study, the liquid-liquid extraction method was used for separation (Sipilä et al., 1998). The organic layer of FPBO remaining after the water-soluble layer is removed is called dry FPBO. The quality of the dry FPBO was analyzed according to ASTM D 7544 or EN 16900 in Table 6. The heating value of the dry FPBO upgraded via the extraction was 21–25 MJ/kg. The moisture content of the dry FPBO can be reduced to below 16% through extraction, which is lower than that of FPBO. Further, the solid content of sawdust with high water content can be reduced via extraction, because the solid is separated along with the water-soluble layer. Dry FPBO has a higher kinematic viscosity than FPBO because the water-soluble layer, including polymers, is removed. However, the density of FPBO increases with the removal of moisture (Mantilla et al., 2014). EFB is known to have a high potassium content Table 6 The properties of dry FPBO according to extraction.
Fig. 5. FT-IR spectra of FPBO; (a) EFB, (b) sawdust, (c) giant Miscanthus.
attached to oxygenated compounds and complex oxygenated compounds with attached aromatic groups, which have been detected in all three biomass types in the following order: giant Miscanthus > sawdust > EFB. The TGA results of the FPBO produced from waste biomass are 142
Characteristics
EFB
Sawdust
Giant Miscanthus
Calorific value, MJ/kg Water content, wt % Solids content, wt % Kinematic viscosity, mm2/s Density at 20 °C, kg/dm3 Sulfur content, wt % Ash content, wt % TAN, mg KOH/g Flash point, °C Pour point, °C
21.9 15.5 0.1 16.8 1.20 0.04 0.02 132.9 65 −40.0
24.1 10.4 0.1 28.7 1.15 0.73 0.01 105.1 72 −42.0
22.3 14.2 0.1 19.3 1.19 1.65 0.01 133.5 68 −45.0
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
4. Conclusions
Table 7 Metal contents and microscopic images of FPBO from sawdust. Bio-oil
Images
Pyrolysis of biomass was performed domestically in a 1 bpd CFB reactor, and the properties of the FPBO and dry FPBO were analyzed according to the ASTM D7544 or EN 16900 standard. The optimal operating conditions were determined using sawdust at 500 °C and a quencher and EP recovery device. The non-condensable gas was composed of 80% of CO2 and CO and the FPBO properties were largely determined by the raw material used. In particular, the heating value, moisture content, and ash content of the FPBO were significantly affected by the biomass used. Among the three raw materials used, the heating value of giant Miscanthus was the highest, and the kinematic viscosity and total amount of acid were influenced by the moisture content. The moisture content of the raw materials was reduced to below 16% and the heating value was increased to above 20 MJ/kg via the extraction. Additionally, the solid content was significantly reduced. The suitability of the FPBO produced domestically according to ASTM D7544 or EN 16900 was investigated and it was found that the properties of the FPBO could be sufficiently improved via the extraction process. In order to use the domestically produced bio-oil, an additional upgrading process is necessary, but it was found that the FPBO satisfied the ASTM D7544 or EN 16900 for use in industrial boilers.
Metals Si
Cu
Ag
Fe
Ca
Ni
P
V
Al
FPBO
0.1
2.3
2.3
1.6
13.8
46.3
0.1↓
3.7
0.1↓
Dry FPBO
0.1↓
0.7
0.9
0.1↓
0.1↓
39.4
0.1↓
1.4
0.1↓
Table 8 The results of elemental analysis of dry FPBO. Biomass
Bio-oil
C
H
O
N
wt.% EFB Sawdust Giant Miscanthus
FPBO Dry FPBO FPBO Dry FPBO FPBO Dry FPBO
38.2 53.7 37.6 52.1 39.5 55.0
H/C
O/C
atom ratio 9.2 11.0 9.5 11.5 9.7 10.3
51.2 34.5 52.2 35.4 50.0 32.7
1.1 0.8 0.7 0.2 0.8 0.3
2.9 2.5 3.0 2.6 2.9 2.3
1.0 0.5 1.0 0.5 0.9 0.4
Acknowledgements This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20173010092430). References Adahchour, M., Beens, J., Vreuls, R.J.J., Brinkman, U.A.Th., 2006. Recent developments in comprehensive two-dimensional gas chromatography(GC × GC), I. Introduction and instrumental set-up. Trends Anal. Chem. 25, 438–454. Buffi, M., Cappelletti, A., Rizzo, A.M., Martelli, F., 2018. Combustion of fast pyrolysis biooil and blends in a micro gas turbine. Biomass Bioenergy 115, 174–185. Chiaramonti, D., Oasmaa, A., Solantausta, Y., 2007. Power generation using fast pyrolysis liquids from biomass. Renew. Sustain. Energy Rev. 11, 1056–1086. Djokic, M.R., Dijkmans, T., Yildiz, G., Prins, W., Geem, K.M.V., 2012. Quantitative analysis of crude and stabilized bio-oils by comprehensive two-dimensional gas-chromatography. J. Chromatogr. A 1257, 131–140. Dwiatmoko, A.A., Zhou, L., Choi, J.-W., Suh, D.J., Ha, J.-M., 2016. Hydrodeoxygenation of lignin-derived monomers and lignocellulose pyrolysis oil on the carbon-supported Ru catalysts. Catal. Today 265, 192–198. Guedes, R.E., Luna, A.S., Torres, A.R., 2018. Operating parameters for bio-oil production in biomass pyrolysis: a review. J. Anal. Appl. Pyrolysis 129, 134–149. Isahak, W.N.R.W., Hisham, M.W.M., Yarmo, M.A., Hin, T.Y., 2012. A review on bio-oil production from biomass by using pyrolysis method. Renew. Sustain. Energy Rev. 16, 5910–5923. Kanaujia, P.K., Sharma, Y.K., Garg, M.O., Tripathi, D., Singh, R., 2014. Review of analytical strategies in the production and upgrading of bio-oil derived from lignocellulosic biomass. J. Anal. Appl. Pyrolysis 105, 55–74. Kim, G., Seo, J., Choi, J.-W., Jae, J., Ha, J.-M., Suh, D.J., Lee, K.-Y., Jeon, J.-K., Kim, J.-K., 2018. Two-step continuous upgrading of sawdust pyrolysis oil to deoxygenated hydrocarbons using hydrotreating and hydrodeoxygenating catalysts. Catal. Today 303, 130–135. Lazzari, E., Schena, T., Primaz, C.T., Maciel, G.P.S., Machado, M.E., Cardoso, C.A.L., Jacques, R.A., Caramão, E.B., 2016. Production and chromatographic characterization of bio-oil from the pyrolysis of mango seed waste. Ind. Crop. Prod. 83, 529–536. Liu, Z.Y., Phillips, J., 1991. Comprehensive two-dimensional gas chromatography using an on-column thermal modulator interface. J. Chromatogr. Sci. 29, 227–231. Maciel, G.P.S., Machado, M.E., Barbar, J.A., Molin, D.D., Caramao, E.B., Jacques, R.A., 2016. GC - GC/TOFMS analysis concerning the identification of organic compounds extracted from the aqueous phase of sugarcane straw fast pyrolysis oil. Biomass Bioenergy 85, 198–206. Mantilla, S.V., Maradei, P.G., Gil, P.A., Cárdenas, S.T., 2014. Comparative study of bio-oil production from sugarcane bagasse and palm empty fruit bunch: yield optimization and bio-oil characterization. J. Anal. Appl. Pyrolysis 108, 284–294. Moraes, M.S.A., Georges, F., Almeida, S.R., Damasceno, F.C., Maciel, G.P.S., Zini, C.A., Jacques, R.A., Caramão, E.B., 2012. Analysis of products from pyrolysis of Brazilian sugar cane straw. Fuel Process. Technol. 101, 35–43. Naqvi, S.R., Uemura, Y., Yusup, S.B., 2014. Catalytic pyrolysis of paddy husk in a drop type pyrolyzer for bio-oil production: the role of temperature and catalyst. J. Anal. Appl. Pyrolysis 106, 57–62. Nogueira, J.M.F., 2012. Novel sorption-based methodologies for static micro extraction
Fig. 7. O/C ratio of crude and dry bio-oils.
and such alkali metals cause fouling on the internal walls of plant boilers. As shown in Table 7, the metal content in the dry FPBO decreases after extraction of FPBO (Chiaramonti et al., 2007; Mantilla et al., 2014), because the solid and heavy fraction is separated along with the water-soluble layer. The microscopic image of FPBO shows the presence of a large number of char particles, some larger than 400 μm (Table 7). These large particles appear to be agglomerates of smaller particles. Majority of the char particles were successfully removed from the FPBO by extraction process. These results are in agreement with the decrease in solid contents after extraction (Tables 5 and 6). Table 8 and Fig. 7 show the elemental and H/C ratio of FPBO after the moisture extraction process. The H/C ratio for petroleum is 2, with almost no oxygen. The H/C and O/C ratios for dry bio-oil decreased to 2.5 and 0.5 from 3 to 1, respectively. FPBO contains about 50% oxygen, while the extracted dry FPBO contains only about 35% oxygen. A catalytic process for upgrading bio-oil needs to be performed before it can be used as fuel in vehicles (Dwiatmoko et al., 2016; Kim et al., 2018).
143
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
spectrometry. J. Chromatogr. A 1228, 3317–3325. Sipilä, K., Kuoppala, E., Fagernäs, L., Oasmaa, A., 1998. Characterization of biomassbased flash pyrolysis oils. Biomass Bioenergy 14, 103–113. Sulaiman, F., Abdullah, N., 2011. Optimum conditions for maximising pyrolysis liquids of oil palm empty fruit bunches. Energy 36, 2352–2359. Tsai, W.T., Lee, M.K., Chang, Y.M., 2006. Fast pyrolysis of rice husk: product yields and compositions. Bioresour. Technol. 98, 22–28. Williams, P.T., Nugranad, N., 2000. Comparison of products from the pyrolysis and catalytic pyrolysis of rice husks. Energy 25, 493–513.
analysis: a review on SBSE and related techniques. Anal. Chim. Acta 757, 1–10. Qureshi, K.M., Lup, A.N.K., Khan, S., Abnisa, F., 2018. A technical review on semi-continuous and continuous pyrolysis process of biomass to bio-oil. J. Anal. Appl. Pyrolysis 131, 52–75. Sembiring, K.C., Rinaldi, N., Simanungkalit, S.P., 2015. Bio-oil from fast pyrolysis of empty fruit bunch at various temperature. Energy Proc. 65, 162–169. Sfetsas, T., Michailof, C., Lappas, A., Li, Q., Kneale, B., 2011. Qualitative and quantitative analysis of pyrolysis oil by gas chromatography with flame ionization detection and comprehensive two-dimensional gas chromatography with time-of-flight mass
144
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Research article
Production of bio-oil from fast pyrolysis of biomass using a pilot-scale circulating fluidized bed reactor and its characterization
T
Jo Yong Parka, Jae-Kon Kima,∗, Chang-Ho Ohb, Jong-Wook Parkb, Eilhann E. Kwonc,∗∗ a
Research Institute of Petroleum Technology, Korea Petroleum Quality & Distribution Authority, Cheongju, 28115, Republic of Korea Daekyung ESCO, Incheon, 21984, Republic of Korea c Department of Environment and Energy, Sejong University, 05006, Republic of Korea b
A R T I C LE I N FO
A B S T R A C T
Keywords: Fast pyrolysis Bio-oil Biomass Biofuel Circulating fluidized bed
To circumvent the adverse impacts arising from an excessive use of fossil fuels, bioenergy and chemical production from a carbon neutral resource (biomass) has drawn considerable attention over the last two decades. Among various technical candidates, fast pyrolysis of biomass has been considered as one of the viable technical routes for converting a carbonaceous material (biomass) into biocrude (bio-oil). In these respects, three biomass samples (i.e., sawdust, empty fruit bunch, and giant Miscanthus) were chosen as a carbon substrate for the pyrolysis process in this study. A pilot-scale circulating fluidized bed reactor was employed for the pyrolysis work, and biocrude from the fast pyrolysis process at 500 °C were characterized because the maximum yield of biocrude (60 wt% of the original sample mass) was achieved at 500 °C. The physico-chemical properties of biocrude were measured by the international standard/protocol (ASTM D7544 and/or EN 16900 test method) to harness biocrude as bioenergy and an initial feedstock for diverse chemicals. All measurements in this study demonstrated that the heating value, moisture content, and ash contents in biocrude were highly contingent on the type of biomass. Moreover, characterization of biocrude in this study significantly suggested that additional unit operations for char and metal removal must be conducted to meet the fuel standard in terms of biocrude as bioenergy.
1. Introduction Despite the diverse socio-economic benefits from harnessing fossil fuels as the primary energy resources, the excessive use of fossil fuels has brought forth global environmental problems (Naqvi et al., 2014). Note that carbon imbalance from combustion of fossil fuels has been regarded as one of the major driving forces for climate change and global warming in that total carbon inputs are far exceeding the full Earth's capacity to assimilate carbon through the natural carbon cycle (Tsai et al., 2006). To mitigate the global environmental burdens, obtaining bioenergy and chemicals from the carbon neutral resources (biomass) has gain considerable attention over the last two decades. Moreover, among various technical candidates, fast pyrolysis has been considered as one of the viable technical routes for converting biomass into the liquid-type fuels (Naqvi et al., 2014; Tsai et al., 2006; Qureshi et al., 2018). In detail, fast pyrolysis bio-oil (FPBO) converts biomass in biocrude under the moderate pyrolytic temperatures (450–650 °C) (Williams and Nugranad, 2000; Qureshi et al., 2018). In general, the
∗
yield of FPBO reaches up to 75 wt% of the original mass of biomass (dry mass basis), of which calorific value reaches up to 60–75% total calorific value of virgin biomass (dry basis) (Isahak et al., 2012). In addition to FPBO, the yield of char and non-condensable gas from fast pyrolysis of biomass reaches up to 10–20 and 10–25 wt% of the original sample mass, respectively. The compositional matrix of pyrogenic products from the fast pyrolysis is also highly contingent on the type of biomass. Thus, various biomasses (wood, corn stalks, agricultural waste, switch grass, forest waste, etc.) were tested for the fast pyrolysis process. The economic feasibility for the fast pyrolysis process of biomass done by other authors (Sembiring et al., 2015; Sulaiman and Abdullah, 2011) also suggested that fast pyrolysis of biomass using the small scale (50–100 tons per day) pyrolysis process shows the highest economic benefits in that a small-scale pyrolysis of biomass does not requires a high tipping fee for biomass and capital cost for a pyrolysis platform (Guedes et al., 2018; Qureshi et al., 2018). FPBO generally exhibits a complex compositional matrix, of which constituents include more than 300 chemical species that are derived
Corresponding author. Corresponding author. E-mail addresses: [email protected] (J.-K. Kim), [email protected] (E.E. Kwon).
∗∗
https://doi.org/10.1016/j.jenvman.2018.12.104 Received 31 August 2018; Received in revised form 22 December 2018; Accepted 26 December 2018 0301-4797/ © 2018 Elsevier Ltd. All rights reserved.
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
from depolymerization of the major component (i.e., cellulose, hemicellulose, and lignin) of biomass (Kanaujia et al., 2014). The typical oxygen content in biocrude reaches up to 40–50 wt% (Isahak et al., 2012). Therefore, biocrude has a low calorific value of 16–18 MJ/kg. Biocrude has the low pH value, of which pH is roughly equivalent to 2.5, and the viscosity of biocrude relative to fossil fuel is high (Sipilä et al., 1998). The moisture content in biocrude reaches up to 25–35 wt % (Isahak et al., 2012). Hence, the fuel properties of biocrude are inferior to that of diesel or gasoline. As a result, biocrude cannot be used directly in an internal combustion engine (Nogueira, 2012; Djokic et al., 2012; Maciel et al., 2016). Particularly, the high contents of water and oxygen, high viscosity, and the high content of ash lead to poor ignition and emission (Isahak et al., 2012). Such impurities also the main driving force for clogging (Buffi et al., 2018). Based on these rationales, biocrude is being used in an internal combustion engine or combustor after the proper upgrading steps (Chiaramonti et al., 2007; Buffi et al., 2018). FPBO was produced by employing a pilot-scale (1 barrel per day: bpd) circulating fluidized bed reactor with three different biomass samples (i.e., sawdust, empty fruit bunch (EFB), and giant Miscanthus) in this study. The produced FPBO was further characterized by the international stand (ASTM D 7544 and EN 16900 specifications). The further extraction of FPBO was carried out to figure out that the desired fuel properties for usefulness of biocrude was achieved in this study. 2. Materials and methods 2.1. Fast pyrolysis bio-oil production A circulating fluidized bed (CFB) reactor was used for fast pyrolysis of biomass, of which biomass feeding rate was adjusted as 42 kg h−1. The overall schematic layout for the pyrolysis system was presented in Fig. 1, and the major parts of the CFB reactor were labelled. In detail, as denoted in Fig. 1, the pyrolysis system is composed of a heating chamber surrounded by an electrical furnace, a biomass hopper, a screw feeder, a fluidized bed reactor, a cyclone, two condensers and an electrostatic precipitator. There are three processes in the CFB reactor: I) the fast pyrolysis reaction process, II) the oil recovery process, where the produced gas is rapidly condensed for oil recovery, and III) gas recirculation and excess gas incineration, where the non-condensed gas is recirculated, and excess gas is incinerated.
Fig. 1. (a) Schematic layout for a pilot-scale circulating fluidized bed (CFB) reactor, (b) a real pyrolytic system to produce FPBO.
Table 1 Operational parameters for the pilot-scale CFB reactor.
2.2. Biomass preparation Two biomass samples (sawdust and giant Miscanthus) were obtained from the Rural Development Administration (RDA) in Korea. EFB was imported from Indonesia. Prior to fast pyrolysis, three samples were mechanically milled, of which particle size was adjusted as 2–3 mm. 2.3. Fast pyrolysis of three different biomass samples The operational parameters for the pilot-scale CFB reactor was summarized in Table 1. Note that the production capacity is 1 barrel per day (1 bpd). To maintain the CFB conditions, 5 kg sand was added into the reactor. The gas flow rate was 3 m s−1, and biomass feeding rate was adjusted as 42 kg h−1.
Parameter
Average
Reaction temperature (˚C) Recycle ratio (%) Gas velocity (m s−1) Blower (Hz) Wind box press (mmAq) Top of reactor press (mmAq) Feed ratio (kg h−1) 1st cyclone temperature (˚C) 2nd cyclone temperature (˚C) Quencher temperature (˚C) EP temperature (˚C)
500 80 3 19 2100 50 42 450 350 33 23
2.5. FPBO extraction (purification) Diethyl ether was used to separate moisture from virgin FPBO from the primary and secondary oil recovery processes for extraction (Fig. 2). FPBO and the ether solution were mixed in a ratio of 1–3, and the diethyl ether and water layers were separated using a separatory funnel. The separated diethyl ether layer contains the hydrophilic components of the oil while the water layer contains moisture and the water-soluble components. The upper layer was carefully collected and dried in a rotary evaporator (30 °C and 30 min) to obtain the etherextracted bio-oil.
2.4. FPBO recovery FPBO was rapidly cooled to trap the condensable hydrocarbons. In this study, the indirect contact method was adopted. In detail, FPBO was recovered by cooling, and then capturing the remaining FPBO using an electric precipitator (EP). The temperatures in the primary and secondary cooling chambers were maintained at 20 °C. The gaseous products from the fast pyrolysis process are CO, CO₂, CH₄, and N2. While 80% of these non-condensed gases was recirculated, 20% of pyrolytic gases was incinerated to maintain the pyrolytic temperature.
2.6. FPBO characterization The properties of FPBO were characterized to measure the heating 139
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
Table 3 Elemental analyzer conditions. Item
C, H analysis
O analysis
Furnace temperature(°C) Oven temperature (°C) Carrier flow (mL min−1) Reference Flow (mL min−1.) Detector
L: 9000, R: 680 75 140 100 TCD
1060 65 130 100 TCD
was determined via titration with KOH and using phenolphthalein as the indicator. The samples were dissolved in neutralized methanol and titrated with a 0.1 N KOH solution. The flash point of the pyrolysis biooil was determined according to ASTM D 93 using a Pensky-Martens closed-cup tester (APM-7, TANAKA). The pour point of oil is an indication of the lowest temperature at which the oil can be pumped. The pour point of the pyrolysis bio-oil was determined according to ASTM D 97 suing a pour point tester (MPC 602, TANAKA). The elemental composition of the pyrolysis bio-oil was determined using a FLASH2000 Thermo-Elemental Analyzer. The operating conditions were summarized in Table 3. An FT-IR spectrometer (Nicolet 6700 manufactured by Thermo-Scientific) was used for the compositional analysis. The FT-IR spectra of the produced pyrolysis FPBO were recorded in the transmission mode between 4000 and 800 cm−1. Further, TGA was carried out with a Q500 TGA system (TA Instruments). Helium was used as the sweep gas at a flow rate of 100 mL min−1. Approximately 10 mg of the sample was loaded into an aluminum crucible, and the TGA system temperature was programmed to increase from 20 temperature to 800 °C at a rate of 10 °C min−1, followed by an isothermal run for 30 min at 800 °C.
Fig. 2. Liquid–liquid extraction using diethyl ether.
value, moisture content, and kinematic viscosity, of which measurements were in accordance with to the ASTM D 7544 or EN 16900 specification (Table 2). Metal contents in FPBO were quantified using an ICP unit. Also, TGA, IR, and elemental analyses of FPBO were carried out. The gross heating values of crude and dry FPBO were measured using a bomb calorimeter (Parr 6400, PARR Co. Ltd.), of which procedures are in accordance with ASTM D240. The water content of crude and dry FPBO was analyzed, of which protocols were in accordance with ASTM E203 using a Karl Fischer titrator (MKC-610, Kyoto Electronics Manufacturing Co., Ltd.). In detail, 100 g fraction of the sample was filtered at 40 °C through a pre-weighed membrane filter. The filter and residue were washed with methanol, dried, and weighed, and the solid content of the sample was calculated from the difference between their weights. The kinematic viscosity of the pyrolysis bio-oil was measured according to ASTM D445 using a viscometer (CAV2100, Cannon Instrument Company). The density of the pyrolysis bio-oil was determined using a density meter (DMA 5000m, Anton Parr) according to ASTM D4052. An X-ray sulfur analyzer (1800H, Horiba) was used to measure the sulfur content in the bio-oil according to ASTM D4294. To measure the ash content of the bio-oil, 10 g bio-oil was combusted at 700 °C for 5 h in a pre-weighed crucible. The crucible and residue were then cooled and weighed, and the ash content was calculated from difference in their masses. The total acid number of the pyrolysis bio-oil
3. Results and discussion 3.1. FPBO production Fast pyrolysis of biomass was carried out using the pilot-scale CFD reactor, and the compositional elemental matrix for each biomass sample were summarized in Table 4. As summarized in Table 4, the content of carbon in each biomass sample is not much different, but the carbon contents were in the following order: sawdust < EFB < giant Miscanthus. The moisture content was in the following order: sawdust < giant Miscanthus < EFB. Based on these characteristics (the contents of C and H), the heating values were in the following order: giant Miscanthus > sawdust > EFB. Fast pyrolysis of biomass was conducted from 400 to 600 °C, and the overall mass balance for three pyrogenic products (gas, liquid, and char) was illustrated in Fig. 3. The overall mass balance suggests that each mass portion of three pyrogenic products is highly affected by the
Table 2 Standard of FPBO in EU and US. Characteristic
ASTN D 7544 Grade G
Calorific value, MJ/kg Water content, wt% Solids content, wt% Kinematic viscosity, mm2/s Density at 20 °C, kg/dm3 Sulfur content, wt% Ash content, wt% pH Flash point, °C Pour Point, °C Nitrogen content, wt% Na, K, Ca, Mg, wt%
> 15 < 30 < 2.5 < 125 1.1–1.3 < 0.05 < 0.25 Report > 45 < -9 – –
prEN16900 Grade D
< 0.25
< 0.15
140
Method
Grade 1
D240 E203 D7579 D445 D4052 D4294 D482 E70 D93 D97 – –
> 14.0 < 30 < 2.5 < 125 < 1.3(at 15 °C) < 0.1 < 0.25 > 2.0 – < -9 report NA
Grade 2
< 0.5 < 50 < 0.05 < 0.05
< 0.02
Method DIN 51900-3 E203 D7579 EN ISO 3104 EN ISO 12185 EN ISO 20846 EN ISO 6245 E70 – ISO 3016 D5291 EN ISO 16476
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
Table 4 The proximate and ultimate analysis of different biomass. Characteristic(wt%)
EFB
Sawdust
Giant Miscanthus
Carbon Hydrogen Nitrogen Sulphur Chlorine Oxygen Volatile matter Ash Moisture Fixed carbon HHV(MJ/kg)
49.1 5.8 0.5 ND ND 41.4 65.0 1.5 15.2 18.3 16.4
43.2 7.9 1.0 0.6 ND 45.1 70.0 5.9 9.6 14.5 17.6
51.3 6.2 0.4 ND ND 42.2 67.5 1.5 11.2 19.8 19.2
operational temperatures. The decrease in the char yield with increasing temperature could be due to secondary decomposition of the char (Williams and Nugranad, 2000). Such observations are good agreement with the preliminary test and the general thermal degradation behaviors of biomass (Liu and Phillips, 1991; Adahchour et al., 2006; Sfetsas et al., 2011). The product yields from three biomass samples at 500 °C are almost identical to the case of sawdust, of which mass portion is equivalent to 60% liquid, 15% char and 25% noncondensable gases. Thus, all fast pyrolysis of three biomass samples was conducted at 500 °C to maximize the yield of FPBO. Fig. 4 depicts the compositional matrix for the non-condensed gas from 450 to 550 °C. Note that the optimal operating temperature for fast pyrolysis of biomass is 500 °C. Among pyrolytic gases, ≥ 80% of the total volume was composed of CO and CO2. Oxygen in FPBO was mostly converted to CO and CO2 when the reaction temperature was low (Williams and Nugranad, 2000). The contents of CO and CO2 were decreased with increasing temperature, while that of hydrogen and methane were increased. 3.2. Pyrolytic bio-oil properties The fuel properties of FPBO produced from sawdust, EFB, and giant Miscanthus at 500 °C were performed, of which measurements were followed by the ASTM D7544 or EN 16900 specification, and the results are shown in Table 5. In Europe and US, the quality standards for FPBO have been set with the goal of it being used for industrial boilers. The heating value of the FPBO produced is related to the water content, which is in the following order: giant Miscanthus > sawdust > EFB. The heating value of FPBO from EFB is lower than that required by the quality standards due to the high moisture content of the bio-oil. The solid content in the FPBO was measured without filtering the produced bio-oil. In the case of sawdust, it was found that very fine particles were not filtered by the cyclone and remained until the latter stage of the process. The shapes of these particles differ depending on the raw material. It has been reported that more small particles are produced when the raw material contains small-sized particles and when pyrolysis is performed with woody biomass rather than grassy biomass (Moraes et al., 2012). The kinematic viscosity varies greatly depending on the moisture content of the fuel, and the kinematic viscosity of EFB was the lowest, as it had the highest moisture content. The moisture content is the most important factor in pyrolysis. The moisture content of FPBO is related to that of the biomass. Regarding the total amount of acid, it has been reported that the carboxylic acid content increases for reactions under high moisture condition (Mantilla et al., 2014). EFB, which had the highest moisture content, also had the highest amount of acid, as shown in Table 5. The ash content of FPBO from sawdust is higher than that required by EN16900 standard due to the high solid content of the bio-oil. Functional group compositional analysis for the FPBO was conducted via FT-IR spectrometry and the results are shown in Fig. 5. The
Fig. 3. Yield of FPBO under various pyrolysis temperatures; (a) EFB, (b) sawdust, (c) Giant Miscanthus.
FPBO contains oxygenated groups. The peak at 3300–3600 cm−1 is due to the OH groups present in water, alcohols, and phenols. This is in agreement with the moisture contents reported in Table 5. Peaks attributable to C]O stretching vibrations between 1650 and 1850 cm−1 were observed in the FT-IR spectra of the FPBO. The CH, CH2, and CH3 functional groups present in alkanes or aromatic aliphatic oxides gave rise to absorption peaks at 2800–3000, 1470, and 1050–1100 cm−1, respectively. The peaks at 1650–1750 cm−1 indicate the presence of single-ring and polycyclic aromatic compounds (Lazzari et al., 2016). Therefore, these peaks are likely attributable to the alkyl groups 141
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
Fig. 6. TGA of FPBO; (a) EFB, (b) sawdust, (c) giant Miscanthus.
Fig. 4. Compositions of non-condensed gas from sawdust.
shown in Fig. 6. The FPBO prepared from sawdust and giant Miscanthus degraded up to 460 °C, and that produced from EFB degraded up to 560 °C. At 900 °C, 10% of the FPBO from giant Miscanthus and EFB, and 14% of the FPBO from sawdust remained as residue. Most of the volatile materials, primarily hemicellulose and cellulose, were degraded at approximately 500 °C. Loss in thermogravimetric mass below 300 °C was attributable to hemicellulose and while that above 300 °C was due to cellulose (Sulaiman and Abdullah, 2011). Lignin decomposed at temperatures above 400 °C (Sembiring et al., 2015).
Table 5 Properties of FPBO produced from different biomass. Characteristics
EFB
Sawdust
Giant Miscanthus
Calorific value, MJ/kg Water content, wt % Solids content, wt % Kinematic viscosity, mm2/s Density at 20 °C, kg/dm3 Sulfur content, wt % Ash content, wt % TAN, mg KOH/g Flash point, °C Pour point, °C
13.2 29.7 2.3 14.5 1.17 0.04 0.1 100.0 46 −22.5
15.9 22.4 4.3 21.7 1.15 0.05 1.3 75.6 49 −21.5
17.0 20.9 0.6 18.7 1.12 0.05 0.2 83.9 47 −17.5
3.3. Fuel properties based on extraction Organic and water-soluble layers can be distinguished when the FPBO is in liquid form. 60% of the FPBO was in the liquid phase. The organic layer must be separated from the water-soluble layer to upgrade the liquid FPBO. In this study, the liquid-liquid extraction method was used for separation (Sipilä et al., 1998). The organic layer of FPBO remaining after the water-soluble layer is removed is called dry FPBO. The quality of the dry FPBO was analyzed according to ASTM D 7544 or EN 16900 in Table 6. The heating value of the dry FPBO upgraded via the extraction was 21–25 MJ/kg. The moisture content of the dry FPBO can be reduced to below 16% through extraction, which is lower than that of FPBO. Further, the solid content of sawdust with high water content can be reduced via extraction, because the solid is separated along with the water-soluble layer. Dry FPBO has a higher kinematic viscosity than FPBO because the water-soluble layer, including polymers, is removed. However, the density of FPBO increases with the removal of moisture (Mantilla et al., 2014). EFB is known to have a high potassium content Table 6 The properties of dry FPBO according to extraction.
Fig. 5. FT-IR spectra of FPBO; (a) EFB, (b) sawdust, (c) giant Miscanthus.
attached to oxygenated compounds and complex oxygenated compounds with attached aromatic groups, which have been detected in all three biomass types in the following order: giant Miscanthus > sawdust > EFB. The TGA results of the FPBO produced from waste biomass are 142
Characteristics
EFB
Sawdust
Giant Miscanthus
Calorific value, MJ/kg Water content, wt % Solids content, wt % Kinematic viscosity, mm2/s Density at 20 °C, kg/dm3 Sulfur content, wt % Ash content, wt % TAN, mg KOH/g Flash point, °C Pour point, °C
21.9 15.5 0.1 16.8 1.20 0.04 0.02 132.9 65 −40.0
24.1 10.4 0.1 28.7 1.15 0.73 0.01 105.1 72 −42.0
22.3 14.2 0.1 19.3 1.19 1.65 0.01 133.5 68 −45.0
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
4. Conclusions
Table 7 Metal contents and microscopic images of FPBO from sawdust. Bio-oil
Images
Pyrolysis of biomass was performed domestically in a 1 bpd CFB reactor, and the properties of the FPBO and dry FPBO were analyzed according to the ASTM D7544 or EN 16900 standard. The optimal operating conditions were determined using sawdust at 500 °C and a quencher and EP recovery device. The non-condensable gas was composed of 80% of CO2 and CO and the FPBO properties were largely determined by the raw material used. In particular, the heating value, moisture content, and ash content of the FPBO were significantly affected by the biomass used. Among the three raw materials used, the heating value of giant Miscanthus was the highest, and the kinematic viscosity and total amount of acid were influenced by the moisture content. The moisture content of the raw materials was reduced to below 16% and the heating value was increased to above 20 MJ/kg via the extraction. Additionally, the solid content was significantly reduced. The suitability of the FPBO produced domestically according to ASTM D7544 or EN 16900 was investigated and it was found that the properties of the FPBO could be sufficiently improved via the extraction process. In order to use the domestically produced bio-oil, an additional upgrading process is necessary, but it was found that the FPBO satisfied the ASTM D7544 or EN 16900 for use in industrial boilers.
Metals Si
Cu
Ag
Fe
Ca
Ni
P
V
Al
FPBO
0.1
2.3
2.3
1.6
13.8
46.3
0.1↓
3.7
0.1↓
Dry FPBO
0.1↓
0.7
0.9
0.1↓
0.1↓
39.4
0.1↓
1.4
0.1↓
Table 8 The results of elemental analysis of dry FPBO. Biomass
Bio-oil
C
H
O
N
wt.% EFB Sawdust Giant Miscanthus
FPBO Dry FPBO FPBO Dry FPBO FPBO Dry FPBO
38.2 53.7 37.6 52.1 39.5 55.0
H/C
O/C
atom ratio 9.2 11.0 9.5 11.5 9.7 10.3
51.2 34.5 52.2 35.4 50.0 32.7
1.1 0.8 0.7 0.2 0.8 0.3
2.9 2.5 3.0 2.6 2.9 2.3
1.0 0.5 1.0 0.5 0.9 0.4
Acknowledgements This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (No. 20173010092430). References Adahchour, M., Beens, J., Vreuls, R.J.J., Brinkman, U.A.Th., 2006. Recent developments in comprehensive two-dimensional gas chromatography(GC × GC), I. Introduction and instrumental set-up. Trends Anal. Chem. 25, 438–454. Buffi, M., Cappelletti, A., Rizzo, A.M., Martelli, F., 2018. Combustion of fast pyrolysis biooil and blends in a micro gas turbine. Biomass Bioenergy 115, 174–185. Chiaramonti, D., Oasmaa, A., Solantausta, Y., 2007. Power generation using fast pyrolysis liquids from biomass. Renew. Sustain. Energy Rev. 11, 1056–1086. Djokic, M.R., Dijkmans, T., Yildiz, G., Prins, W., Geem, K.M.V., 2012. Quantitative analysis of crude and stabilized bio-oils by comprehensive two-dimensional gas-chromatography. J. Chromatogr. A 1257, 131–140. Dwiatmoko, A.A., Zhou, L., Choi, J.-W., Suh, D.J., Ha, J.-M., 2016. Hydrodeoxygenation of lignin-derived monomers and lignocellulose pyrolysis oil on the carbon-supported Ru catalysts. Catal. Today 265, 192–198. Guedes, R.E., Luna, A.S., Torres, A.R., 2018. Operating parameters for bio-oil production in biomass pyrolysis: a review. J. Anal. Appl. Pyrolysis 129, 134–149. Isahak, W.N.R.W., Hisham, M.W.M., Yarmo, M.A., Hin, T.Y., 2012. A review on bio-oil production from biomass by using pyrolysis method. Renew. Sustain. Energy Rev. 16, 5910–5923. Kanaujia, P.K., Sharma, Y.K., Garg, M.O., Tripathi, D., Singh, R., 2014. Review of analytical strategies in the production and upgrading of bio-oil derived from lignocellulosic biomass. J. Anal. Appl. Pyrolysis 105, 55–74. Kim, G., Seo, J., Choi, J.-W., Jae, J., Ha, J.-M., Suh, D.J., Lee, K.-Y., Jeon, J.-K., Kim, J.-K., 2018. Two-step continuous upgrading of sawdust pyrolysis oil to deoxygenated hydrocarbons using hydrotreating and hydrodeoxygenating catalysts. Catal. Today 303, 130–135. Lazzari, E., Schena, T., Primaz, C.T., Maciel, G.P.S., Machado, M.E., Cardoso, C.A.L., Jacques, R.A., Caramão, E.B., 2016. Production and chromatographic characterization of bio-oil from the pyrolysis of mango seed waste. Ind. Crop. Prod. 83, 529–536. Liu, Z.Y., Phillips, J., 1991. Comprehensive two-dimensional gas chromatography using an on-column thermal modulator interface. J. Chromatogr. Sci. 29, 227–231. Maciel, G.P.S., Machado, M.E., Barbar, J.A., Molin, D.D., Caramao, E.B., Jacques, R.A., 2016. GC - GC/TOFMS analysis concerning the identification of organic compounds extracted from the aqueous phase of sugarcane straw fast pyrolysis oil. Biomass Bioenergy 85, 198–206. Mantilla, S.V., Maradei, P.G., Gil, P.A., Cárdenas, S.T., 2014. Comparative study of bio-oil production from sugarcane bagasse and palm empty fruit bunch: yield optimization and bio-oil characterization. J. Anal. Appl. Pyrolysis 108, 284–294. Moraes, M.S.A., Georges, F., Almeida, S.R., Damasceno, F.C., Maciel, G.P.S., Zini, C.A., Jacques, R.A., Caramão, E.B., 2012. Analysis of products from pyrolysis of Brazilian sugar cane straw. Fuel Process. Technol. 101, 35–43. Naqvi, S.R., Uemura, Y., Yusup, S.B., 2014. Catalytic pyrolysis of paddy husk in a drop type pyrolyzer for bio-oil production: the role of temperature and catalyst. J. Anal. Appl. Pyrolysis 106, 57–62. Nogueira, J.M.F., 2012. Novel sorption-based methodologies for static micro extraction
Fig. 7. O/C ratio of crude and dry bio-oils.
and such alkali metals cause fouling on the internal walls of plant boilers. As shown in Table 7, the metal content in the dry FPBO decreases after extraction of FPBO (Chiaramonti et al., 2007; Mantilla et al., 2014), because the solid and heavy fraction is separated along with the water-soluble layer. The microscopic image of FPBO shows the presence of a large number of char particles, some larger than 400 μm (Table 7). These large particles appear to be agglomerates of smaller particles. Majority of the char particles were successfully removed from the FPBO by extraction process. These results are in agreement with the decrease in solid contents after extraction (Tables 5 and 6). Table 8 and Fig. 7 show the elemental and H/C ratio of FPBO after the moisture extraction process. The H/C ratio for petroleum is 2, with almost no oxygen. The H/C and O/C ratios for dry bio-oil decreased to 2.5 and 0.5 from 3 to 1, respectively. FPBO contains about 50% oxygen, while the extracted dry FPBO contains only about 35% oxygen. A catalytic process for upgrading bio-oil needs to be performed before it can be used as fuel in vehicles (Dwiatmoko et al., 2016; Kim et al., 2018).
143
Journal of Environmental Management 234 (2019) 138–144
J.Y. Park et al.
spectrometry. J. Chromatogr. A 1228, 3317–3325. Sipilä, K., Kuoppala, E., Fagernäs, L., Oasmaa, A., 1998. Characterization of biomassbased flash pyrolysis oils. Biomass Bioenergy 14, 103–113. Sulaiman, F., Abdullah, N., 2011. Optimum conditions for maximising pyrolysis liquids of oil palm empty fruit bunches. Energy 36, 2352–2359. Tsai, W.T., Lee, M.K., Chang, Y.M., 2006. Fast pyrolysis of rice husk: product yields and compositions. Bioresour. Technol. 98, 22–28. Williams, P.T., Nugranad, N., 2000. Comparison of products from the pyrolysis and catalytic pyrolysis of rice husks. Energy 25, 493–513.
analysis: a review on SBSE and related techniques. Anal. Chim. Acta 757, 1–10. Qureshi, K.M., Lup, A.N.K., Khan, S., Abnisa, F., 2018. A technical review on semi-continuous and continuous pyrolysis process of biomass to bio-oil. J. Anal. Appl. Pyrolysis 131, 52–75. Sembiring, K.C., Rinaldi, N., Simanungkalit, S.P., 2015. Bio-oil from fast pyrolysis of empty fruit bunch at various temperature. Energy Proc. 65, 162–169. Sfetsas, T., Michailof, C., Lappas, A., Li, Q., Kneale, B., 2011. Qualitative and quantitative analysis of pyrolysis oil by gas chromatography with flame ionization detection and comprehensive two-dimensional gas chromatography with time-of-flight mass
144