- Email: [email protected]
Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process
ARTICLE IN PRESS
JID: JTICE
[m5G;May 5, 2016;9:34]
Journal of the Taiwan Institute of Chemical Engineers 0 0 0 (2016) 1–11
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice
Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process Jyh-Shyong Chang a,∗, Jen-Chieh Cheng a, Tzong-Rong Ling b, Jia-Ming Chern a, Gow-Bin Wang c, Tse-Chuan Chou d, Chin-Tsou Kuo a a
Department of Chemical Engineering, Tatung University, 40 Chungshan North Road, 3rd Sec., Taipei, Taiwan Department of Chemical Engineering, I-Shou University, 1, Section 1, Hsueh-Cheng Road, Ta-Hsu Hsiang, Kaohsiung 84008, Taiwan Department of Chemical and Materials Engineering, Chang Gung University, 259 Wen-Hwa 1st Road, Kwei-Shan, Tao-Yuan 333, Taiwan d Department of Chemical Engineering, National Cheng Kung University, Tainan 701, Taiwan b c
a r t i c l e
i n f o
Article history: Received 9 March 2016 Revised 14 April 2016 Accepted 14 April 2016 Available online xxx Keywords: Low acid value Biodesel Waste cooking oil Fast pyrolysis
a b s t r a c t Continuous stable and fast pyrolysis of waste cooking oil (WCO) under isothermal conditions was achieved using catalyst pellets with a binder (bentonite) loaded with 25 wt% active materials including: Al2 (SO4 )3 , Na2 CO3 , NaOH, and CaO. The experiments were carried out in an instant evaporator pyrolytic system with pyrolysis reactor temperatures ranging from 450 o Cto 550 o C. The catalyst pellets were prepared by drying and calculating the extrudates made from the pastes of the active materials and the binder. Low acid value (AV < 0.5 mg KOH/g) pyrolytic oils (PO) were obtained using these base catalyst pellets. Long-term tests, performed with these base catalysts, showed that CaO pellets can produce a PO with a near zero AV in a cycle time of 12 h. The physical properties of the bio-gasoline oil (BGO) and the bio-diesel oil (BDO) when separated from the PO almost meets the specifications of commercial petroleum gasoline and diesel, except that the octane number of the BGO (76) was below the target (92). Soap-like matter, in addition to the coke, produced during the pyrolysis of WCO using the catalyst pellets (Na2 CO3 and NaOH) plugged the pyrolytic reactor, thereby preventing further operation. This soap-like matter was not found using the CaO catalyst pellet, suggesting that CaO was the most suitable base catalyst for WCO pyrolysis. The basic CaO pellets, able to reduce the fatty acids derived from thermal radical cracking reactions of WCO to aldehydes and ketones, produced a minimum aromatic hydrocarbon content in PO (from WCO) with a higher content of saturated fatty acids than from fresh edible soybean oil containing a higher content of unsaturated fatty acids. The resulting environmental friendly PO with a near zero AV, produced from WCO using the CaO pellets, proved the utility of the simple and one-step continuous pyrolytic process presented here. © 2016 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
1. Introduction Biomass, being the organic material produced by photosynthesis, is an abundant renewable energy source. In contrast to fossil fuels, biomass offers the benefit of being carbon dioxide neutral; however the use of land for biomass production, which could be used for growing food crops, is often challenged [1]. On the other hand, residues from agriculture and forestry, e.g. wastes from animals, wood, industry and the food industry possess a high content of organic compounds that are environmentally acceptable sources of energy [2]. These residues, or wastes, contain cellulose, hemicelluloses, lignin, triglycerides and fatty acids that can
∗
Corresponding author. Tel: +886 2 1822928-6266; fax: +886 2 5861939. E-mail address: [email protected] (J.-S. Chang).
be used, with appropriate thermo-chemical and biochemical processing, to produce renewable liquid fuels and valuable chemicals [3–6]. Waste cooking oil (WCO), from commercial catering establishments, which contain the triglyceride and fatty acid fractions that currently present a disposal problem are thus a promising feedstock alternative to expensive newly refined edible oils [7]. Without reclamation facilities, waste cooking oils and fats can give rise to significant disposal problems and in doing so create odor and pollution. Addressing this waste disposal problem, while creating a fuel substitute, potentially offers both economic and environmental benefits. Many developed countries have outlawed the disposal of WCO in domestic drainage systems [8]. The utilization of WCO, e.g. as a raw material or fuel, is important, as it not only mitigates risks to diners but also reduces raw material costs. The usual direct utilization of WCO is as a fuel [9–10] by esterification and/or transesterification [11–13] to produce bio-diesel. The major
http://dx.doi.org/10.1016/j.jtice.2016.04.014 1876-1070/© 2016 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE 2
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11 Table 1 Properties of feedstock and the catalysts prepared.
Oil
Acid value (mg of KOH/g)
Soybean oil WCO
0.08 9.50
Catalyst/ Physical properties Catalyst pellet∗ Catalyst composition Special surface area (m2 /g) Porous volume (mL/g) Average pore
Density (g/cm3 )
Gross heating value (kJ/g)
Kinematic viscosity at 40 o C (mm2 /s)
Water content (wt%)
0.92 0.92
31.92 39.34
32.61 37.22
0.04 0.32
C1 Al2 (SO)3 / Bentonite (75%) 10.0379
C2 Na2 CO3 / Bentonite (75%) 0.7336
C3 NaOH/ Bentonite (75%) 0.1368
C4 CaO/ Bentonite (75%) 10.9471
0.0506
0.0120
0.0072
0.0479
201.7702
653.2706
2111.8290
175.0352
0
diameter (A) ∗
Size of cylindrical pellet: O.D. = 0.5 cm; L/O.D = 1–1.6
factors inhibiting the utilization of this technology, for bio-diesel production, are the variation of the acid index, the solids content, the need for heavy equipment for large-scale production, and the alcohol required for the recovery and purification [5–7,9] Bio-diesel that leaves the production plant in good condition can become degraded during distribution without proper care and attention, due to e.g. oxidation, contact with water, and/or microbial activity [14]. The thermal or catalytic cracking of the triglycerides contained in WCO is a way of producing renewable bio-based products suitable for use in fuel and chemical applications. This approach has significant advantages over transesterification, including compatibility with engines and fuel standards; and feedstock flexibility: importantly, the final products are similar in composition to diesel fuel. In most triglyceride pyrolysis studies, molecular sieve catalysts, e.g. ZSM-5, MCM-41, and Y zeolite, were used [15–17]. There are also reports of triglyceride thermal cracking without a catalyst [18–19]. In general, partial oxidation or oxidation and cracking reactions of the bio-oil molecules occur with increasing temperatures during the pyrolysis process [20], leading to pyrolytic oils (PO) with high carboxylic acid contents and high acid values (AV) ranging from 17 to 142 [21–22]. These undesirable products have a large effect on the corrosion value, cold filter plugging point and freezing point of the biofuel. Acid values, which are a measure of the free fatty acid content, were determined using the method EN 14,104, GB-T 5530-2005. For the bio-diesel specification, the maximum allowable AV is 0.5 mgKOH/g in Europe and 0.8 mgKOH/g in American using ASTM D664 [23–25]. These undesirable acid products have a large effect on the corrosion value and cold flow properties. To decrease the AV of the pyrolytic oil (PO), by lowering its carboxylic acid content, esterification with methanol or deacidification with alkali solution can be used [26]. The results of Xu et al. [21] have shown that the catalytic cracking of woody oils generates fuels with a chemical composition similar to that of petroleum-based fuels. By using a basic catalyst (1.5–3% of the total amount of soybean oil), it is possible to obtain oils with good cold-flow properties and high heat values. Gas chromatography (GC) and Fourier transform infrared spectroscopic (FTIR) analyses of the products have shown that the distribution of the fractions can be modified by using a basic catalyst. The AV of the gasoline fraction of the pyrolytic oil was ∼ 30 while that of diesel fraction was ∼ 37. Esterification of either gasoline or diesel, derived from pyrolytic oils, using a solid acid catalyst can reduce the AV to ∼ 3. In this paper we chose to use WCO as a raw material because it is widely available from households and food companies. As WCO has potential in sustainable oil production, we re-
port a catalytic cracking study on WCO using base catalysts in a fixed bed reactor for producing low AV biofuel in a continuous process. It has been reported that low AV(< 5) PO can be produced with base catalysts using a process with a running time of ∼ 28 h. In this work, the catalytic cracking reaction was carried out under isothermal conditions using a tubular evaporator and a fixed bed reactor packed with basic pellet catalysts including: Na2 CO3 /bentonite, NaOH/bentonite and CaO/bentonite (all with bentonite in the sodium form). For comparison, Al2 (SO4 )/bentonite (ammonium form) was also used as an acidic catalyst. The resulting POs were characterized by elemental analysis and quantified by product yield. It is hoped to develop a simple, one-step continuous process for producing PO with low AVs to meet the requirements of commercial gasoline bio-diesel standards using basic catalyst pellets in this work. The basic catalysts can be regenerated by introducing air, or ozone [27], to burn off the coke after the AV of pyrolytic oil exceeds 5 mgKOH/g. 2. Experimental section 2.1. Materials and catalyst preparation WCO was provided from a local food processing company (Chant Oil Corporation, Taiwan). The important properties of WCO as well as the catalyst are listed in Table 1. Soybean oil was purchased locally (Taiwan Sugar Corporation). Typical fatty compositions of soybean oil and WCO are shown in Table 2. The main differences between these two feedstocks (soybean oil/WCO) are the ratios of the unsaturated fatty acids 85:69 and saturated fatty acids 15:31. Other chemicals such as Al2 (SO4 ), Na2 CO3 , NaOH, and CaO (all in fine powder form) were obtained from a local chemical
Table 2. Fatty acid composition of soybean oil and WCO. Fatty acid composition (wt%)
Soybean
WCO
Unsaturated fatty acids Oleic acid (C18:1) Linoleic acid (C18:2) α -linoleic acid (C18:2) Others Saturated fatty acids Palmitic acid (C16:0) Stearic acid (C18:0) Others
85 23 54 8 0 15 11 4 0
69 36 29 3 1 31 24 5 2
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
BG (8)
(7)
(4)
PO WCO
3
2.3. Product analysis The product compounds to be reported were identified using a GC–MS (Agilent 6890 GC/5973A MS) provided with a Agilent HP-5MS capillary column (l, 30 m; i.d., 0.25 mm; film thickness, 0.25 μm), with N2 as the carrier gas. A split ratio of 100:1 was used. The injector temperature was kept at 260 o C. The oven temperature increased from 50 o C (3 min) to 280 o C (10 min) at a heating rate of 5 o C/min. The mass spectra were obtained from m/z 50–800 at a scan rate of 1666 scans/s. The product mixture sample (1 μl) was injected automatically. Dynamic viscosity was measured with a CANNON miniAV, and the gross calorific value was measured using a Parr 1261 Calorimeter (ASTM D4809). The acid value of the PO was determined by auto-titration (Metrohm 877 Titrino Plus). The water content was measured using a Karl Fisher Titrator, KEM MKS-500. Density was measured with a densitometer (KEM DA-130 N).
(5) (6) (1)
(2)
(3)
Fig. 1. Schematic diagram of WCO pyrolysis system. (1): Feedstock vessel (2): Pump (3): Tubular evaporator (4): Pyrolysis reactor (5): Catalyst (6): Heater (7): Condenser (8): Insulator BG, bio-gas; PO, pyrolytic oil WCO, wasted cooking oil.
supplier (First Chemical in Taiwan), while the catalyst binder bentonite (sodium form) was obtained from the Whole Earth Trading Company (Taiwan). The bentonite was supplied in sodium form— this was mixed with a NH4 NO3 solution to transform it into the ammonium form to bind with Al2 (SO4 ). The catalyst was prepared by mixing bentonite (75 wt%) with one of Al2 (SO4 ), Na2 CO3 , NaOH, or CaO (each of these components was 25 wt%) in distilled water, to produce a paste. Once the mixture had been homogenized, the excess water was filtered and the paste extruded through a hole (0.5 cm diameter). The extrudates were dried for 24 h at ambient temperature, following which the cylindrical pellets (∼0.8 cm length and 0.4 cm diameter) were prepared. These pellets were dried at 110 °C for 24 h and then calcined at 600 °C for 4 h (ramp rate 5 °C/min) [28]. Note, CaO (25 wt%) was the maximum content that could be used successfully to make the catalyst pellet using the above procedure. 2.2. Pyrolysis reactor system In this work, WCO pyrolysis was carried out as shown in Fig. 1. The system comprised: a feed stock (1), metering pump (2), instant tubular evaporator (1.9 cm × 18 cm) (3), pyrolytic reactor (1.9 cm × 54 cm or 1.9 cm × 90 cm in length) (4) packed with catalyst bed (5) inside and an electrical resistances heater (4 × 1 kW) surrounding outside of the evaporator and the pyrolytic reactor (6). The pyrolytic vapor was passed through the gas–liquid separator (7) to obtain the PO and the bio-gas (BG) by maintaining the temperature of the separator vessel at 25 o C. The evaporator and the reactor were made of 3/4 inch OD stainless steel. The collected pyrolytic oil was separated into oil sludge (OS), bio-diesel oil (BDO) and bio-gasoline oil (BGO), respectively using a front-view distillation apparatus (Koehler K45200).
2.4. Experimental procedure Before the pyrolysis reaction, the packed catalysts were pretreated by purging with nitrogen gas at 500 o Cfor 2 h. Pyrolysis experiments with varied feed rates were carried out with fixed bed temperatures ranging from 450 to 550 o C. WCO was kept warm in feedstock at 60 o Cand pumped into the tubular evaporator that was then heated by an external electrical resistance (1 kW). The temperature of the tubular evaporator was controlled at temperatures ranging from 450 o Cto 550 o C, allowing the WCO to vaporize instantly prior to entering the catalyst bed. The evaporator is classified as an ‘instant evaporator’ as it is designed to ensure that no liquid can be accumulated during operation. After WCO pyrolysis, the pyrolytic vapor was condensed and separated to obtain the PO and the BG. The production rate of BG was measured by the water replacement method, prior to collection in a gas bag for composition analysis. The liquid products (PO, BGO, and BDO) were quantified by weighting, while the coke generated was estimated by mass balance calculations. All the samples were taken during 6–8 h after the startup of the system for each test to characterize the system. 3. Results and discussion Table 3 gives the short-term operating conditions for experiments using a constant feed rate and constant temperatures for the evaporator and pyrolysis reactor. Weight hourly space (WHSV) is calculated based on the liquid feed rate and the catalyst loading, excluding the bentonite binder. The WHSV of cases 1–6 (100 g catalyst pellets) was 1.56 h–1 while that of case 7 (150 g catalyst pellets) was 1.04 h–1 . By maintaining the tubular evaporator and the pyrolytic reactor at the desired temperature (500 o C), the gas residence time, shown in Table 3, was measured as the first bubble appearing in a glass container filled with water after an ‘feed Table 3 Operating conditions for the pyrolysis tests. Case no.
1
2
3
4
5
6
7
Feedstock Catalyst pellet Catalyst pellet loading (g) WHSV (h−1 )∗ Residence time of oil gas (s)
Soybean oil Blank
WCO Blank
WCO C1 100
WCO C2 100
WCO C3 100
WCO C4 100
WCO C4 150
18
18
1.56 7.5
1.56 7.5
1.56 7.5
1.56 7.5
1.04 10.2
Temperature of reactor, 500 °C; Temperature of vaporization, 500 °C; Temperature of condenser, 20 °C; Feed rate, 39 g/h. ∗ WHSV is based on catalyst loading not including the binder (bentonite).
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE 4
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11 Table 4 Properties of the pyrolytic oils obtained in cases 1–7. Case no. (Catalyst) 3
Density (g/cm ) Water content (wt %) Acid value (mg of KOH/g) Freezing point (◦ C) Elemental analysis (wt %) C H O
1 (none)
2 (none)
3 (C1)
4 (C2)
5 (C3)
6 (C4)
7 (C4)
0.81 0.41 107.87 4
0.81 0.41 115.73 5
0.82 0.38 32.43 –15
0.82 0.37 2.13 –16
0.82 0.33 0 –17
0.82 0.40 2.20 –17
0.82 0.39 0 –17
78.98 9.43 11.59
79.17 10.93 9.90
79.11 10.16 10.73
85.16 12.07 2.77
84.96 13.94 1.10
83.95 11.88 4.17
83.77 13.42 2.81
Table 5 Main fuel properties of WCO, PO, BGO, BDO, and commercial gasoline and diesel from the experiment case 7 (Table 3). Fuel properties
WCO
PO
BGO
BDO
92 Gasoline
Density (g/cm3 ) Kinematic viscosity at 40 ◦ C (mm2 /s) Gross heating value (kJ/g) Water content (wt %) Octane no. Cetane index Acid value (mg of KOH/g) Elemental analysis (wt %) C H O
0.91 37.22 39.34 0.32 – – 9.5
0.82 1.97 43.44 0.39 – – 0
0.77 0.75 42.83 0.42 76 – 1.2
0.84 2.11 43.99 0.37 – 51 0.5
0.78 1.0–1.9 44.00 – Min. 92 – 0
79.88 12.64 7.48
83.77 13.42 2.81
84.96 11.56 3.48
84.60 11.84 3.56
85.34 11.43 3.23
0.84 2.0–4.5 43.56 – – Min. 48 0 86.31 13.59 0.1
100
Case 90
1 2 3 4 5 6 7
80 70
Yield (%)
impulse’. It can be seen from Table 3, that with the same mass flow of the feedstock, the residence time of oil gas was different independently of the reactor being packed with a catalyst or not. The pyrolytic system achieved a steady operating state 6–8 h after startup, thus the experimental data in Tables 4 and 5 is the average of the performance during these 2 h. The factors addressed in Table 3 were used to evaluate the pyrolysis performance for the reactor bed: (a) with and without a catalyst, (b) packed with acidic and basic catalysts, and (c) packed with different quantities of base catalysts. The optimum catalyst, determined from the experiments based on Tables 3–5 was used to measure the effects of feed rates and evaporator and pyrolysis reactor temperatures on the properties of pyrolytic oils.
Diesel
60 50 40 30 20 10
3.1. Pyrolysis with and without catalyst
0
PO
Based on Table 3, the WCO pyrolysis product distribution (i.e. the yield of BGO, BDO, BG, OS, water, and coke) for the seven cases with catalyst (cases 3 to 7 in Table 3) and without (cases 1 and 2 in Table 3) at the same conditions of feed rate and operating temperatures of the evaporator and the pyrolitic reactor are shown in Fig. 2. Case 1 used fresh soybean oil (AV = 0.2) as the feed stock while cases 2–7 used WCO (AV = 9.5) as the feed stock. Table 4 gives the properties of the pyrolytic oils obtained from the experiments (cases 1–7 in Table 3). As shown in Table 4, high AV (around 110), high oxygen contents (about 10 wt%), and high freezing points (4–5 o C) were the main features of the PO obtained in the pyrolytic reactor operated without catalyst (cases 1 and 2) compared to those with catalysts (cases 3–7). Without a catalyst, the oil molecules might be partially oxidized into acid components. Thus, high acid values and high oxygen contents can be found in the experiments for cases 1 and 2. The reason for this result will be discussed in Section 3.7. The PO from the acid catalyst acid (case 3) gives a medium AV (32) but a high oxygen content (10 wt%). The cases using base catalyst pellets (Na2 CO3 , NaOH, and CaO), appear to promote the reaction of the fatty acids derived from the thermal decomposition of fresh edible oils or WCO to aldehydes or ketones [29–30]. This result is also supported by observing that the
BGO
BDO
BG
OS
Water
Coke
Product Fig. 2. Product distribution obtained from no-catalytic (cases 1–2) and catalytic pyrolysis (cases 3–7). The given conditions are shown in Tables 1 and 2.
oxygen content is reduced to below 5 wt% for cases 3–7 (Table 4). Fig. 2, shows that similar production rates for PO and its components BGO, BDO, BG, OS, water, and coke were obtained both with the fresh soybean oil and WCO by the thermal pyrolysis without catalyst (cases 1 and 2). However, the yields of PO without catalyst (cases 1 and 2) are relatively higher than those with catalyst (cases 3–7). As the editable soybean oil was only used for comparison, we shall only focus on the productivity of WCO. Because the total yield of BG, BGO, BDO, OS, water, and coke equals one, if less coke is formed, greater feedstock productivity can be obtained. The coke formation for non-catalytic pyrolysis (cases 1 and 2) was lower than that for catalytic pyrolysis (cases 3–7). In addition to thermal cracking, catalytic cracking also occurred in the catalytic pyrolysis. However, the high AV of the PO, produced by non-catalytic pyrolysis, counterbalances this benefit. Additionally, the coke formation of case 7 (18.85 wt%) is only slightly larger
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
than that of case 2 (15.51 wt%), suggesting that the base catalyst pellet (CaO/bentonite) is a potential choice to produce a near zero AV for PO. Based on the above discussion and the need to obtain pyrolytic oils with low acid values, pyrolysis without catalyst, as a process option, was rejected. 3.2. Comparison between acid and base catalysts Comparison of case 3 (acid catalyst) with cases 4–7 (base catalysts) from Table 4, shows that the AV and the oxygen content of the produced pyrolytic oils using base catalysts are much lower than those that result from the use of acid catalysts (Al2 (SO4 )). The deoxygenation capability using acid catalyst is much lower than that achieved using base catalysts, as highlighted by the fact that the AV of case 3 remains high (∼33) using the acid catalyst pellet (Al2 (SO4 )3 ). Further, coke formation using the acid catalyst is the highest among these seven cases as shown in Fig. 2; therefore, the use of base catalyst pellets (Na2 CO3 , NaOH, and CaO) is a preferential route to obtaining a PO with the sought after characteristics. 3.3. Comparison among the base catalysts for short-term operation Three base catalyst pellets (Na2 CO3 , NaOH, and CaO) were evaluated in this work. Observing the performance of cases 4– 6 as shown in Table 4 and Fig. 2, one can find PO with varying components (BGO, BDO, BG, OS, water, and coke). The coke yield ranged from 17 to 22 for cases 4–6. The BGO production rate using the catalyst pellets Na2 CO3 and NaOH (cases 4 and 5) was higher (∼38 wt%) than when using CaO catalyst pellet (case 6). The capability to transform free fatty acids to aldehydes or ketones using the Na2 CO3 and NaOH catalyst pellets seems to be similar to that when using CaO (AV of POs ranged from 0 to 2.2). Case 7 used a greater quantity of CaO catalyst pellets (decreasing WHSV) to achieve a near zero AV for the PO produced and a higher yield of BGO compared to that of case 6 using a lower quantity of CaO catalyst pellets. 3.4. Comparison of base catalysts for long-term operation In addition to the short-term performance of the catalyst pellets in the fixed-bed pyrolytic reactor, the long-term operation using the base catalyst pellets can provide more information about the capability of the base catalyst pellets. Fig. 3 depicts the development of the AV of pyrolytic oils over time for cases 4–7 (Table 3). The location of the last bar plotted shows that the fixed-bed reactor was blocked with deposited coke or soap-like matter along the bed and could not therefore be operated any longer. One can observe that the catalyst pellets (case 4 (Na2 CO3 ) and case 5 (NaOH)) cannot be operated after 16 h from start-up while the catalyst pellet (CaO) can maintain operation for 28 h. The soap-like matter produced was the special feature of the pyrolysis using the catalyst pellets (Na2 CO3 and NaOH) but was not found using the CaO catalyst pellet. Therefore, from the long-term operational viewpoint, the CaO catalyst pellet is a better choice than the catalyst pellets (Na2 CO3 and NaOH). Comparison of cases 6 and 7 shown in Fig. 3, reveals that a low WHSV (more CaO catalyst) was required to obtain a zero AV for the PO produced during a time interval of 12 h after startup. Therefore, setting the lifetime of CaO catalyst pellet to 12 h is appropriate when WHSV equals 1.04 h–1 and the catalyst needs to be regenerated. From previous comparisons (Sections 3.1–3.4), the CaO catalyst pellet was the best choice for achieving a low PO acid value when made from WCO using a fast pyrolysis process. Table 5 gives the main fuel properties of WCO, PO, BGO, BDO, and commercial gasoline and diesel from the experiment case 7 (Table 3). The properties of the obtained BDO were well matched with those of
5
commercial diesel. The kinematic viscosity of the produced BGO being lower than that of commercial gasoline means smaller components were obtained using the front-view distillation apparatus (Koehler K45200). The octane number 76 of the produced BGO is ‘off-specification’ and needs to be blended with an octane booster to meet the requirement of the specification. The properties shown in Table 5 are reflected in Fig. 4 which shows the GC-mass spectra of the pyrolytic products of case 7 for comparing the component distribution (C8-C24) of (a) diesel and PO, (b) gasoline and BGO, and (c) diesel and BDO. Light components exist in the GC-mass spectra of PO, BGO, and BDO when compared to those of commercial diesel and gasoline. 3.5. The effects of pyrolysis temperature and feed rates (WHSV) on the product distribution using the CaO catalyst pellet Using the operating condition (case 7 in Table 3), the effect of the pyrolysis temperature on the product distribution, using the CaO catalyst pellet at the same WHSV, is shown in Fig. 5. The coke formation yield is too high to be acceptable at 550 o C. Comparable liquid product (BGO and BDO) yields can be obtained for the pyrolytic temperatures 450 o C and 500 o C but with different yield distributions between BGO and BDO. A low coke yield but high OS yield was found at 450 o Ccompared to 500 o C; however, the AV of PO at 450 o C was 1.85 and zero at 500 o C, suggesting a pyrolytic temperature set to 500 o C seems to be most suitable. Using the operating conditions (case 7 in Table 3) with a varying the feed rate (WHSV), gives the yield distribution shown in Fig. 6. The AV of PO at a feed rate 60 g/h was 2.43 and zero at other feed rates. Because the smallest coke yield and highest yield of liquid products (BGO and BDO) was obtained at a feed rate of 39 g/h, operating the catalytic cracking system at WHSV(1.04 h−1 ) appears to be a good choice. 3.6. Catalyst regeneration In general, catalyst regeneration is needed to maintain PO product quality during long-term operation. Thus, reactivation of catalyst would be needed after running the process for a period of time. In this work, a catalyst regeneration process was carried out by burning out the coke deposits with air at 600 o C for 4 h. After regeneration of catalyst, the pyrolysis activity was increased to almost the initial state as shown in Fig. 7. 3.7. The roles of feedstocks, catalyst acidity and basicity on composition distribution of PO Several authors [7,31–32] agree that the initial triglyceride catalytic cracking step, using acidic catalysts, involves the initial thermal decomposition of the triglycerides to yield heavy oxygenated hydrocarbons by a free radical mechanism, which is independent of the catalyst’s characteristics. These high-molecular weight oxygenated compounds undergo secondary cracking reactions to give different products wherein the catalyst’s acidity plays a significant role [7,31–32]. Dupain et al. [30] carried out the catalytic cracking of rapeseed oil with commercial equilibrated FCC catalyst (Ecat) under realistic FCC conditions. They reported that triglyceride molecules are mainly converted within 50 ms between 485 and 585 °C into fatty acids through radical cracking reactions. The formation of radicals is enhanced by the catalyst’s external surface. These high-molecular weight oxygenated compounds then undergo secondary cracking reactions to give different products. Once the triglyceride has been thermally decomposed to highmolecular weight oxygenated hydrocarbons, their transformation into various products starts with deoxygenation, e.g. by decarboxylation or decarbonylation and then by secondary cracking reactions
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE 6
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
6
Case 4 (Table 2)
5 4 3 2 1 0 6
Case 5 (Table 2)
5 4 3 2
AV (mg KOH/g)
1 0 6
Case 6 (Table 2)
5 4 3 2 1 0 6
Case 7 (Table 2)
5 4 3 2 1 0 0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
Time (hr) Fig. 3. Long-term test of the catalyst pellets (Na2 CO3, NaOH, and CaO).
to produce short and long chain olefins and paraffins. The subsequent conversion of short-chain olefins (typically C2 –C4 olefins) into C2 –C10 olefins, aliphatic and aromatic hydrocarbons is promoted by catalytic mechanisms such as oligomerization, cyclization, and aromatization in the pores of the catalyst used. The formation of coke might result from the thermal polycondensation of triglycerides and/or the catalytic polymerization of aromatic hydrocarbons [7,31–32]. The results of this experimental work support the observations above, based on the GC-mass analysis, of the component distributions of the 7 cases and show that we can effectively improve the fast pyrolysis process for obtaining a good PO yield and quality from WCO using the information in this section. The composition distribution of gaseous products includes CO, CO2 , C4 –C6 , C3 H6 , C3 H8 , C2 H2 , C2 H4 , C2 H6 , CH4 , and H2 . The PO compositions are divided into 6 categories, including the com-
pounds that contain: (a) hydroxyl groups (alcohols), (b) carboxylic acid groups (carboxylic acids), (c) carbonyl groups (mainly aldehydes and ketones), (d) aromatic hydrocarbons (HCs), (e) alicyclic hy drocarbons, and (f) linear paraffins. The relative yield ζ (= i (yield (%) of gas product or PO × area (%) of GC-mass × quality (%)) of the identified component i) is adopted for the x-axis of Figs. 8–11. The y-axis of Figs. 8(a)–11(a) is the individual gas component, and that of Figs. 8(b)–11(b) is the individual category of PO product. The relative yield (ζ ) was expected to reflect the relative productivity of the individual gas components, or the individual category of the PO product, when the response factor for each GC-mass peak was not identified. Fig. 8 compares the composition distribution of fresh soybean oil (case 1) and WCO (case 2) using noncatalytic pyrolysis. The compositions of fresh soybean oil and WCO are shown in Table 2, which shows that the main difference lies
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
7
Diesel PO
a
C15 C14 C8
C16
C13
C19 C17 C18
C9 C11 C12
C20 C21 C22 C23C24
C10
Gasoline BGO
b
c Diesel BDO
5
10
15
20
25
30
35
40
45
50
55
60
Time (min) Fig. 4. GC-mass spectra of the pyrolytic products (case 7): (a) diesel and PO, (b) gasoline and BGO, and (c) diesel and BDO.
in the fractions of saturated and unsaturated fatty acids. The fresh soybean oil contained 85 wt% unsaturated fatty acids while WCO contained two-fold more saturated fatty acids than fresh soybean oil (31 wt% versus 15 wt%). Fig. 8(b) reveals comparatively high contents of carboxylic acids in PO in cases 3–7 (catalytic pyrolysis) obtained from both fresh soybean oil and WCO feedstock, which resulted in high AV as shown in Table 4. Aromatic HCs are much greater in fresh soybean oil than that in WCO. On the other hand,
alicyclic HCs and linear paraffins are relatively higher in WCO than in fresh soybean oil. Additionally, the composition distributions of gaseous products, obtained from fresh soybean oil, are relatively higher than those obtained from WCO (Fig. 8(a)). It has been reported that fatty acids with higher content of unsaturated alkyl chains exist in soybean oil and have a strong tendency to undergo a very fast aromatization process (via sequential cyclization, dehydrogenation and condensation reactions) to form (poly)aromatic
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE 8
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
Fig. 5. Pyrolysis product distribution of catalyst pellet CaO (case 7) at different temperatures (◦ C).
100
Feed rate (g/hr) 20 39 60 80
90 80
Yield (%)
70 60 50 40 30 20 10 0
PO
BGO
BDO
BG
OS
Water
Coke
Product Fig. 6. Pyrolysis product distribution of the catalyst pellet CaO (case 7) at different feed rates (g/h).
100
4. Conclusions
Cycle 90
1 2 3 4 5 6
80
Yield (%)
70 60 50 40 30 20 10 0
PO
BGO
BDO
BG
OS
Water
bodies with alkyl branches using acidic catalysts [30,33]. Therefore, higher contents of aromatic HCs can be obtained using fresh soybean oil. In the meantime, because fatty acids with higher saturated alkyl chain contents exist in WCO, higher contents of alicyclic HCs and linear paraffins can be produced using WCO in a thermal pyrolysis process. As it is friendlier to the environment to produce alicyclic HCs and linear paraffins than aromatic HCs, WCO with higher saturated fatty acid content is a better feedstock for thermal pyrolysis than valuable edible oils with higher unsaturated fatty acid contents. Fig. 9 compares the product distribution between non-catalytic pyrolysis (case 2) and pyrolysis using an acidic catalyst (case 3) with WCO as the feedstock. The results show that case 3 produced a much higher quantity of gaseous products using the acidic catalyst Al2 (SO)3 compared to non-catalytic pyrolysis (case 2), with coke formation (case 3) being much higher than for the other cases, resulting in the lowest production of PO among the seven cases (Fig. 9b). Fig. 10 compares the product distribution between a noncatalytic pyrolysis (case 2) and a catalytic pyrolysis using the basic CaO catalyst pellets (case 6) with WCO as the feedstock. A nearzero content of carboxylic acids and a high content of aldehydes and ketones was produced using the base catalyst (CaO) can be seen in Fig. 10 (b) this resulted in a very low AV of PO (AV = 2.2). Similar results can be found in the literature [29,34,35]. Catalytic decarboxylaion by removing naphthenic acid using the CaO catalyst was examined [34]. The bimolecular reaction mechanism converting aliphatic acid to ketones using CaO as catalyst was proposed in the work of Pestman et al. [29], while the oxygen content reduction in the bio-oil using catalytic CaO was evaluated in the work of Lin et al. [35]—which taken together suggest that the reduction of the fatty acids derived from radical cracking reactions of WCO to aldehydes and ketones can be assumed in experiments (case 6 and case 7) of this work. Furthermore, a very low aromatic HC content was obtained using the base catalyst. Using WCO as feedstock, Fig. 11 compares the product distribution between the pyrolysis using less CaO pellets (case 6), or more CaO catalyst pellets (case 7). With more CaO pellets, the aldehydes and ketones are converted to other PO product categories (alcohols, aromatic hydrocarbons (HCs), and alicyclic hydrocarbons) under the catalytic condition provided by the CaO pellets. Gaseous products increase with more CaO pellets (case 7) compared to that of case 6; however, the main result is that with enough CaO pellets the AV of the PO approaches zero, see Table 4 (case 7).
Coke
Product Fig. 7. Pyrolysis product distribution of the catalyst pellet CaO (case 7) at different catalyst regeneration cycles (recycle).
A pyrolytic system composed of an instant evaporator and a tubular pyrolytic reactor was successfully built. Fast pyrolysis of WCO was stably carried out in this pyrolytic system to obtain a near zero acid value oil using the optimized conditions WHSV = 1.04 h−1 and CaO pellet catalyst. The properties of the obtained BDO matched well with those of commercial diesel. The octane number 76 of the produced BGO is ‘off-specification’ and thus the product needs to be blended with an octane booster to meet the requirement of the specification of commercial gasoline. The degree coke formation (18.85 wt%) was only slightly larger than with non-catalytic thermal pyrolysis (15.51 wt%). The acidic catalyst pellets (Al2 (SO4 )3 ) and the related acidic zeolite based catalysts such as ZSM-5, MCM-41, and Y zeolite produced a PO with a high AV and a high content of aromatic HCs which are not friendly to the environment. Reduction of the fatty acids derived from thermal radical cracking reactions of WCO to aldehydes and ketones was the main role of the adopted CaO pellet catalyst in the catalytic pyrolysis of WCO. During the long running tests, the soap-like matter produced was the ‘special feature’ of the pyrolysis
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
a
9
b Case 1 Case 2
CO
Case 1 Case 2
Alcohols
CO2 C4-C6
Carboxylic acids
Propene Aldehydes, Ketones
Propane Ethyne
Aromatic HCs
Ethene Alicyclic HCs
Ethane Methane
Linear paraffins Hydrogen
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0
5
10
15
Relative yield
20
25
30
35
40
45
50
Relative yield
Fig. 8. Composition distribution of (a) gas product and (b) PO for comparison of case 1 and case 2.
a
b Case 2 Case 3
CO
Case 2 Case 3
Alcohols
CO2 C4-C6
Carboxylic acids
Propene
Aldehydes, Ketones
Propane Ethyne
Aromatic HCs
Ethene
Alicyclic HCs
Ethane Methane
Linear paraffins Hydrogen
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0
5
10
Relative yield
15
20
25
30
35
40
45
50
Relative yield
Fig. 9. Composition distribution of (a) gas product and (b) PO for comparison of case 2 and case 3.
a
b Case 2 Case 6
CO
Case 2 Case 6
Alcohols
CO2 C4-C6
Carboxylic acids
Propene Aldehydes, Ketones
Propane Ethyne
Aromatic HCs
Ethene Alicyclic HCs
Ethane Methane
Linear paraffins
Hydrogen 0.0
0.5
1.0
1.5
2.0
Relative yield
2.5
3.0
3.5
4.0
0
5
10
15
20
25
30
35
40
45
50
Relative yield
Fig. 10. Composition distribution of (a) gas product and (b) PO for comparison of case 2 and case 6.
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE 10
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
a
b Case 6 Case 7
CO
Case 6 Case 7
Alcohols
CO2 C4-C6
Carboxylic acids
Propene Aldehydes, Ketones
Propane Ethyne
Aromatic HCs
Ethene Alicyclic HCs
Ethane Methane
Linear paraffins Hydrogen
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Relative yield
0
5
10
15
20
25
30
35
40
45
50
Relative yield
Fig. 11. Composition distribution of (a) gas product and (b) PO for comparison of case 6 and case 7.
using the catalyst pellets (Na2 CO3 and NaOH) but was not found using the CaO catalyst pellet. Therefore, the CaO catalyst pellet is a better choice than the catalyst pellets (Na2 CO3 and NaOH). This point was revealed for the first time in this work. Regeneration of the used CaO pellets can return them to their initial activities. Although the price of pure CaO powder is very low, a way to increase the CaO content in the pellet still needs to be found in order to increase the process efficiency. In addition to the process advantages offered by the tubular pyrolysis system employing CaO catalyst pellets discussed above, the cost of building the system was much less than that of a comparable fluidized catalytic cracking reactor system. In addition to which the ability to operate the system without a carrier gas also results in a worthwhile cost saving. However, a significant disadvantage remains in that the maximum life-cycle of the catalyst pellets was only 28 h; therefore, methods need to be developed to extend the life-cycle of the catalyst pellets before scaling the process up in size to that of a commercial scale system. A possible way to address this problem may be to adopt a spouted bed with smaller sized catalyst pellets. Dedication This paper is dedicated to the memory of the late Prof. Y. C. Chao (ddd), who through his inspirational teaching, encouraged us all to continue in his path. Acknowledgments Funding for this work, provided for this research by the Ministry of Science and Technology of Republic of China (Grant No.: MOST 104-3113-E-036-001), is gratefully acknowledged. References [1] Canakci M, Van Gerpen J. A pilot plant to produce biodiesel from high free fatty acid feedstocks. J Am Soc Agri Biol Eng 2003;46(4):945–54. [2] Kulkarni MG, Dalai AK. Waste cooking oil-an economical source for biodiesel: a review. Ind Eng Chem Res 2006;45(9):2901–13. [3] Shi H, Hui Zhang. Waste oil and fat feedstocks for biodiesel production. Adv Petrol Explor Dev 2014;8(1):31–6. [4] Patil PD, Gude VG, Reddy HK, Mupaswy T, Deng S. Biodiesel production from waste cooking oil using sulfuric and microwave irradiation process. J Environ Protect 2012;3:107–13. [5] Wiggers VR, Meier HF, Wisniewski A Jr, Chivanga Barros AA, Wolf Maciel MR. Biofuel from continuous fast pyrolysis of soybean oil: a pilot plant study. Bioresour Technol 20 09;10 0:6570–7.
[6] Demirbas A. Biodiesel fuels from vegetable oils via catalytic and non-catalytic supercritical alcohol transesterificationa and other methods: a survey. Energy Convers Manage 2003;44:2093–109. [7] Maher KD, Bressler DC. Pyrolysis of triglyceride materials for the production of renewable fuels and chemicals. Bioresour Technol 2007;98:2351–68. [8] Xu J, Jiang J, Lu Y, Chen J. Liquid hydrocarbon fuels obtained by the pyrolysis of soybean oils. Bioresour Technol 20 09;10 0:4867–70. [9] Meier HF, Wiggers VR, Zonta GR, Scharf DR, Simionatto EL, Ender L. A kinetic model for thermal cracking of waste cooking oil based on chemical lumps. Fuel 2015;144:50–9. [10] Shadangi KP, Mohanty K. Thermal and catalytic pyrolysis of Karanja seed to produce liquid fuel. Fuel 2014;115:434–42. [11] Hoydonckx HE, Vos DEDe, Chavan SA, Jacobs PA. Esterification and transesterification of renewable chemicals. Topics Catal 2004;27(1):83–96. [12] Patel A, Narkhede N. Biodiesel synthesis via esterification and transesterification over a new heterogeneous catalyst comprising lacunary silicotungstate and MCM-41. Catal Sci Technol 2013;3:3317–25. [13] Lien YS, Hsieh LS, Wu JCS. Transesterification using oleophilic acid catalyst. Ind Eng Chem Res 2010;49(5):2118–21. [14] http://www.extension.org/pages/26636/transportation-and-storage-ofbiodiesel#.VdgYdbKqpHw. [15] Maher KD, Bressler DC. Pyrolysis of triglyceride materials for the production of renewable fuels and chemicals. Bioresour Technol 2007;98:2351–68. [16] Padmaja KV, Atheya N, Bhatnagar AK, Singh KK. Conversion of calotropis procera biocrude to liquid fuels using thermal and catalytic cracking. Fuel 2009;88:780–5. [17] Katikaneni SPR, Adjaye JD, Bakhshi NN. Performance of aluminophosphate molecular-sieve catalysts for the production of hydrocarbons from wood-derived and vegetable-oils. Energy Fuels 1995;9(6):1065–78. [18] Idem RO, Katikaneni SPR, Bakhshi NN. Thermal cracking of canola oil: reaction products in the presence and absence of steam. Energy Fuels 1996;10(6):1150–62. [19] Schwaba AW, Dykstrab GJ, Selkea E, Sorensonb SC, Prydea EH. Diesel fuel from thermal decomposition of soybean oil. JAOCS 1988;65(11):1781–6. [20] Hotová G, Slovák V. Effect of pyrolysis temperature and thermal oxidation on the adsorption properties of carbon cryogels. Thermochim Acta 2015;614:45–51. [21] Xu J, Jiang J, Sun Y, Chen J. Production of hydrocarbon fuels from pyrolysis of soybean oils using a basic catalyst. Bioresour Technol 2010;101:9803–6. [22] Xu J, Jiang J, Chen J, Sun Y. Biofuel production from catalytic cracking of woody oils. Bioresour Technol 2010;101:5586–91. [23] Oasmaa A, Elliott CE, Korhonen J. Acidity of biomass fast pyrolysis bio-oils. Energy Fuels 2010;24:6548–54. [24] Xu J, Xiao G, Zhou Y, Jiang J. Production of biofuels from high-acid-value waste oils. Energy Fuels 2011;25:4638–42. [25] ASTM D664-11a Standard test method for acid number of petroleum products by potentiometric titration. [26] Ding J, Xia Z, Lu J. Esterification and deacidification of a waste cooking oil (TAN 68.81 mg KOH/g) for biodiesel production. Energies 2012;5:2683–91. [27] Chen JC, Liu CY. Regeneration of spent catalysts with ozone. Int J Environ Chem Ecol Geol Geophys Eng 2013;7(7):414–17. [28] Arabiourrutia M, Olazar M, Aguado R, Lopez G, Barona A, Bilbao J. HZSM-5 and HY zeolite catalyst performance in the pyrolysis of tires in a conical spouted bed reactor. Ind Eng Chem Res 2008;47:7600–9. [29] Pestman R, Koster RM, Avan D, Pieterse JAZ, Ponec V. Reactions of carboxylic acids on oxides 2. bimolecular reaction of aliphatic acids to ketones. J Catal 1997;168:265–72.
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
JID: JTICE
ARTICLE IN PRESS J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
[30] Dupain X, Costa DJ, Schaverien CJ, Makkee M, Moulijn JA. Cracking of a rapeseed vegetable oil under realistic FCC conditions. Appl Catal B: Environ 2007;72:44–61. [31] Melero JA, Iglesias J, Garcia A. Biomass as renewable feedstock in standard refinery units. feasibility, opportunities and challenges. Energy Environ Sci 2012;5:7393–420. [32] Taufiqurrahmi N, Bhatia S. Catalytic cracking of edible and non-edible oils for the production of biofuels. Energy Environ Sci 2011;4:1087–112.
[m5G;May 5, 2016;9:34] 11
[33] Rao TV M, Dupain X, Makkee M. Catalytic cracking of edible and non-edible oils for the production of biofuels Fluid catalytic cracking: processing opportunities for fischer–tropsch waxes and vegetable oils to produce transportation fuels and light olefins. Micropor Mesopor Mater 2012;164:148–63. [34] Ding L, Rahimi P, Hawkins R, Bhatt S, Shi Y. Naphthenic acid removal from heavy oils on alkaline earth-metal oxides and ZnO catalysts. Appl Catal A: Gen 2009;371:121–30. [35] Lin Y, Zhang C, Zhang M, Zhang J. Deoxygenation of bio-oil during pyrolysis of biomass in the presence of CaO in a fluidized-bed reactor. Energy Fuels 2010;24:5686–95.
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
JID: JTICE
[m5G;May 5, 2016;9:34]
Journal of the Taiwan Institute of Chemical Engineers 0 0 0 (2016) 1–11
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice
Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process Jyh-Shyong Chang a,∗, Jen-Chieh Cheng a, Tzong-Rong Ling b, Jia-Ming Chern a, Gow-Bin Wang c, Tse-Chuan Chou d, Chin-Tsou Kuo a a
Department of Chemical Engineering, Tatung University, 40 Chungshan North Road, 3rd Sec., Taipei, Taiwan Department of Chemical Engineering, I-Shou University, 1, Section 1, Hsueh-Cheng Road, Ta-Hsu Hsiang, Kaohsiung 84008, Taiwan Department of Chemical and Materials Engineering, Chang Gung University, 259 Wen-Hwa 1st Road, Kwei-Shan, Tao-Yuan 333, Taiwan d Department of Chemical Engineering, National Cheng Kung University, Tainan 701, Taiwan b c
a r t i c l e
i n f o
Article history: Received 9 March 2016 Revised 14 April 2016 Accepted 14 April 2016 Available online xxx Keywords: Low acid value Biodesel Waste cooking oil Fast pyrolysis
a b s t r a c t Continuous stable and fast pyrolysis of waste cooking oil (WCO) under isothermal conditions was achieved using catalyst pellets with a binder (bentonite) loaded with 25 wt% active materials including: Al2 (SO4 )3 , Na2 CO3 , NaOH, and CaO. The experiments were carried out in an instant evaporator pyrolytic system with pyrolysis reactor temperatures ranging from 450 o Cto 550 o C. The catalyst pellets were prepared by drying and calculating the extrudates made from the pastes of the active materials and the binder. Low acid value (AV < 0.5 mg KOH/g) pyrolytic oils (PO) were obtained using these base catalyst pellets. Long-term tests, performed with these base catalysts, showed that CaO pellets can produce a PO with a near zero AV in a cycle time of 12 h. The physical properties of the bio-gasoline oil (BGO) and the bio-diesel oil (BDO) when separated from the PO almost meets the specifications of commercial petroleum gasoline and diesel, except that the octane number of the BGO (76) was below the target (92). Soap-like matter, in addition to the coke, produced during the pyrolysis of WCO using the catalyst pellets (Na2 CO3 and NaOH) plugged the pyrolytic reactor, thereby preventing further operation. This soap-like matter was not found using the CaO catalyst pellet, suggesting that CaO was the most suitable base catalyst for WCO pyrolysis. The basic CaO pellets, able to reduce the fatty acids derived from thermal radical cracking reactions of WCO to aldehydes and ketones, produced a minimum aromatic hydrocarbon content in PO (from WCO) with a higher content of saturated fatty acids than from fresh edible soybean oil containing a higher content of unsaturated fatty acids. The resulting environmental friendly PO with a near zero AV, produced from WCO using the CaO pellets, proved the utility of the simple and one-step continuous pyrolytic process presented here. © 2016 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
1. Introduction Biomass, being the organic material produced by photosynthesis, is an abundant renewable energy source. In contrast to fossil fuels, biomass offers the benefit of being carbon dioxide neutral; however the use of land for biomass production, which could be used for growing food crops, is often challenged [1]. On the other hand, residues from agriculture and forestry, e.g. wastes from animals, wood, industry and the food industry possess a high content of organic compounds that are environmentally acceptable sources of energy [2]. These residues, or wastes, contain cellulose, hemicelluloses, lignin, triglycerides and fatty acids that can
∗
Corresponding author. Tel: +886 2 1822928-6266; fax: +886 2 5861939. E-mail address: [email protected] (J.-S. Chang).
be used, with appropriate thermo-chemical and biochemical processing, to produce renewable liquid fuels and valuable chemicals [3–6]. Waste cooking oil (WCO), from commercial catering establishments, which contain the triglyceride and fatty acid fractions that currently present a disposal problem are thus a promising feedstock alternative to expensive newly refined edible oils [7]. Without reclamation facilities, waste cooking oils and fats can give rise to significant disposal problems and in doing so create odor and pollution. Addressing this waste disposal problem, while creating a fuel substitute, potentially offers both economic and environmental benefits. Many developed countries have outlawed the disposal of WCO in domestic drainage systems [8]. The utilization of WCO, e.g. as a raw material or fuel, is important, as it not only mitigates risks to diners but also reduces raw material costs. The usual direct utilization of WCO is as a fuel [9–10] by esterification and/or transesterification [11–13] to produce bio-diesel. The major
http://dx.doi.org/10.1016/j.jtice.2016.04.014 1876-1070/© 2016 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE 2
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11 Table 1 Properties of feedstock and the catalysts prepared.
Oil
Acid value (mg of KOH/g)
Soybean oil WCO
0.08 9.50
Catalyst/ Physical properties Catalyst pellet∗ Catalyst composition Special surface area (m2 /g) Porous volume (mL/g) Average pore
Density (g/cm3 )
Gross heating value (kJ/g)
Kinematic viscosity at 40 o C (mm2 /s)
Water content (wt%)
0.92 0.92
31.92 39.34
32.61 37.22
0.04 0.32
C1 Al2 (SO)3 / Bentonite (75%) 10.0379
C2 Na2 CO3 / Bentonite (75%) 0.7336
C3 NaOH/ Bentonite (75%) 0.1368
C4 CaO/ Bentonite (75%) 10.9471
0.0506
0.0120
0.0072
0.0479
201.7702
653.2706
2111.8290
175.0352
0
diameter (A) ∗
Size of cylindrical pellet: O.D. = 0.5 cm; L/O.D = 1–1.6
factors inhibiting the utilization of this technology, for bio-diesel production, are the variation of the acid index, the solids content, the need for heavy equipment for large-scale production, and the alcohol required for the recovery and purification [5–7,9] Bio-diesel that leaves the production plant in good condition can become degraded during distribution without proper care and attention, due to e.g. oxidation, contact with water, and/or microbial activity [14]. The thermal or catalytic cracking of the triglycerides contained in WCO is a way of producing renewable bio-based products suitable for use in fuel and chemical applications. This approach has significant advantages over transesterification, including compatibility with engines and fuel standards; and feedstock flexibility: importantly, the final products are similar in composition to diesel fuel. In most triglyceride pyrolysis studies, molecular sieve catalysts, e.g. ZSM-5, MCM-41, and Y zeolite, were used [15–17]. There are also reports of triglyceride thermal cracking without a catalyst [18–19]. In general, partial oxidation or oxidation and cracking reactions of the bio-oil molecules occur with increasing temperatures during the pyrolysis process [20], leading to pyrolytic oils (PO) with high carboxylic acid contents and high acid values (AV) ranging from 17 to 142 [21–22]. These undesirable products have a large effect on the corrosion value, cold filter plugging point and freezing point of the biofuel. Acid values, which are a measure of the free fatty acid content, were determined using the method EN 14,104, GB-T 5530-2005. For the bio-diesel specification, the maximum allowable AV is 0.5 mgKOH/g in Europe and 0.8 mgKOH/g in American using ASTM D664 [23–25]. These undesirable acid products have a large effect on the corrosion value and cold flow properties. To decrease the AV of the pyrolytic oil (PO), by lowering its carboxylic acid content, esterification with methanol or deacidification with alkali solution can be used [26]. The results of Xu et al. [21] have shown that the catalytic cracking of woody oils generates fuels with a chemical composition similar to that of petroleum-based fuels. By using a basic catalyst (1.5–3% of the total amount of soybean oil), it is possible to obtain oils with good cold-flow properties and high heat values. Gas chromatography (GC) and Fourier transform infrared spectroscopic (FTIR) analyses of the products have shown that the distribution of the fractions can be modified by using a basic catalyst. The AV of the gasoline fraction of the pyrolytic oil was ∼ 30 while that of diesel fraction was ∼ 37. Esterification of either gasoline or diesel, derived from pyrolytic oils, using a solid acid catalyst can reduce the AV to ∼ 3. In this paper we chose to use WCO as a raw material because it is widely available from households and food companies. As WCO has potential in sustainable oil production, we re-
port a catalytic cracking study on WCO using base catalysts in a fixed bed reactor for producing low AV biofuel in a continuous process. It has been reported that low AV(< 5) PO can be produced with base catalysts using a process with a running time of ∼ 28 h. In this work, the catalytic cracking reaction was carried out under isothermal conditions using a tubular evaporator and a fixed bed reactor packed with basic pellet catalysts including: Na2 CO3 /bentonite, NaOH/bentonite and CaO/bentonite (all with bentonite in the sodium form). For comparison, Al2 (SO4 )/bentonite (ammonium form) was also used as an acidic catalyst. The resulting POs were characterized by elemental analysis and quantified by product yield. It is hoped to develop a simple, one-step continuous process for producing PO with low AVs to meet the requirements of commercial gasoline bio-diesel standards using basic catalyst pellets in this work. The basic catalysts can be regenerated by introducing air, or ozone [27], to burn off the coke after the AV of pyrolytic oil exceeds 5 mgKOH/g. 2. Experimental section 2.1. Materials and catalyst preparation WCO was provided from a local food processing company (Chant Oil Corporation, Taiwan). The important properties of WCO as well as the catalyst are listed in Table 1. Soybean oil was purchased locally (Taiwan Sugar Corporation). Typical fatty compositions of soybean oil and WCO are shown in Table 2. The main differences between these two feedstocks (soybean oil/WCO) are the ratios of the unsaturated fatty acids 85:69 and saturated fatty acids 15:31. Other chemicals such as Al2 (SO4 ), Na2 CO3 , NaOH, and CaO (all in fine powder form) were obtained from a local chemical
Table 2. Fatty acid composition of soybean oil and WCO. Fatty acid composition (wt%)
Soybean
WCO
Unsaturated fatty acids Oleic acid (C18:1) Linoleic acid (C18:2) α -linoleic acid (C18:2) Others Saturated fatty acids Palmitic acid (C16:0) Stearic acid (C18:0) Others
85 23 54 8 0 15 11 4 0
69 36 29 3 1 31 24 5 2
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
BG (8)
(7)
(4)
PO WCO
3
2.3. Product analysis The product compounds to be reported were identified using a GC–MS (Agilent 6890 GC/5973A MS) provided with a Agilent HP-5MS capillary column (l, 30 m; i.d., 0.25 mm; film thickness, 0.25 μm), with N2 as the carrier gas. A split ratio of 100:1 was used. The injector temperature was kept at 260 o C. The oven temperature increased from 50 o C (3 min) to 280 o C (10 min) at a heating rate of 5 o C/min. The mass spectra were obtained from m/z 50–800 at a scan rate of 1666 scans/s. The product mixture sample (1 μl) was injected automatically. Dynamic viscosity was measured with a CANNON miniAV, and the gross calorific value was measured using a Parr 1261 Calorimeter (ASTM D4809). The acid value of the PO was determined by auto-titration (Metrohm 877 Titrino Plus). The water content was measured using a Karl Fisher Titrator, KEM MKS-500. Density was measured with a densitometer (KEM DA-130 N).
(5) (6) (1)
(2)
(3)
Fig. 1. Schematic diagram of WCO pyrolysis system. (1): Feedstock vessel (2): Pump (3): Tubular evaporator (4): Pyrolysis reactor (5): Catalyst (6): Heater (7): Condenser (8): Insulator BG, bio-gas; PO, pyrolytic oil WCO, wasted cooking oil.
supplier (First Chemical in Taiwan), while the catalyst binder bentonite (sodium form) was obtained from the Whole Earth Trading Company (Taiwan). The bentonite was supplied in sodium form— this was mixed with a NH4 NO3 solution to transform it into the ammonium form to bind with Al2 (SO4 ). The catalyst was prepared by mixing bentonite (75 wt%) with one of Al2 (SO4 ), Na2 CO3 , NaOH, or CaO (each of these components was 25 wt%) in distilled water, to produce a paste. Once the mixture had been homogenized, the excess water was filtered and the paste extruded through a hole (0.5 cm diameter). The extrudates were dried for 24 h at ambient temperature, following which the cylindrical pellets (∼0.8 cm length and 0.4 cm diameter) were prepared. These pellets were dried at 110 °C for 24 h and then calcined at 600 °C for 4 h (ramp rate 5 °C/min) [28]. Note, CaO (25 wt%) was the maximum content that could be used successfully to make the catalyst pellet using the above procedure. 2.2. Pyrolysis reactor system In this work, WCO pyrolysis was carried out as shown in Fig. 1. The system comprised: a feed stock (1), metering pump (2), instant tubular evaporator (1.9 cm × 18 cm) (3), pyrolytic reactor (1.9 cm × 54 cm or 1.9 cm × 90 cm in length) (4) packed with catalyst bed (5) inside and an electrical resistances heater (4 × 1 kW) surrounding outside of the evaporator and the pyrolytic reactor (6). The pyrolytic vapor was passed through the gas–liquid separator (7) to obtain the PO and the bio-gas (BG) by maintaining the temperature of the separator vessel at 25 o C. The evaporator and the reactor were made of 3/4 inch OD stainless steel. The collected pyrolytic oil was separated into oil sludge (OS), bio-diesel oil (BDO) and bio-gasoline oil (BGO), respectively using a front-view distillation apparatus (Koehler K45200).
2.4. Experimental procedure Before the pyrolysis reaction, the packed catalysts were pretreated by purging with nitrogen gas at 500 o Cfor 2 h. Pyrolysis experiments with varied feed rates were carried out with fixed bed temperatures ranging from 450 to 550 o C. WCO was kept warm in feedstock at 60 o Cand pumped into the tubular evaporator that was then heated by an external electrical resistance (1 kW). The temperature of the tubular evaporator was controlled at temperatures ranging from 450 o Cto 550 o C, allowing the WCO to vaporize instantly prior to entering the catalyst bed. The evaporator is classified as an ‘instant evaporator’ as it is designed to ensure that no liquid can be accumulated during operation. After WCO pyrolysis, the pyrolytic vapor was condensed and separated to obtain the PO and the BG. The production rate of BG was measured by the water replacement method, prior to collection in a gas bag for composition analysis. The liquid products (PO, BGO, and BDO) were quantified by weighting, while the coke generated was estimated by mass balance calculations. All the samples were taken during 6–8 h after the startup of the system for each test to characterize the system. 3. Results and discussion Table 3 gives the short-term operating conditions for experiments using a constant feed rate and constant temperatures for the evaporator and pyrolysis reactor. Weight hourly space (WHSV) is calculated based on the liquid feed rate and the catalyst loading, excluding the bentonite binder. The WHSV of cases 1–6 (100 g catalyst pellets) was 1.56 h–1 while that of case 7 (150 g catalyst pellets) was 1.04 h–1 . By maintaining the tubular evaporator and the pyrolytic reactor at the desired temperature (500 o C), the gas residence time, shown in Table 3, was measured as the first bubble appearing in a glass container filled with water after an ‘feed Table 3 Operating conditions for the pyrolysis tests. Case no.
1
2
3
4
5
6
7
Feedstock Catalyst pellet Catalyst pellet loading (g) WHSV (h−1 )∗ Residence time of oil gas (s)
Soybean oil Blank
WCO Blank
WCO C1 100
WCO C2 100
WCO C3 100
WCO C4 100
WCO C4 150
18
18
1.56 7.5
1.56 7.5
1.56 7.5
1.56 7.5
1.04 10.2
Temperature of reactor, 500 °C; Temperature of vaporization, 500 °C; Temperature of condenser, 20 °C; Feed rate, 39 g/h. ∗ WHSV is based on catalyst loading not including the binder (bentonite).
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE 4
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11 Table 4 Properties of the pyrolytic oils obtained in cases 1–7. Case no. (Catalyst) 3
Density (g/cm ) Water content (wt %) Acid value (mg of KOH/g) Freezing point (◦ C) Elemental analysis (wt %) C H O
1 (none)
2 (none)
3 (C1)
4 (C2)
5 (C3)
6 (C4)
7 (C4)
0.81 0.41 107.87 4
0.81 0.41 115.73 5
0.82 0.38 32.43 –15
0.82 0.37 2.13 –16
0.82 0.33 0 –17
0.82 0.40 2.20 –17
0.82 0.39 0 –17
78.98 9.43 11.59
79.17 10.93 9.90
79.11 10.16 10.73
85.16 12.07 2.77
84.96 13.94 1.10
83.95 11.88 4.17
83.77 13.42 2.81
Table 5 Main fuel properties of WCO, PO, BGO, BDO, and commercial gasoline and diesel from the experiment case 7 (Table 3). Fuel properties
WCO
PO
BGO
BDO
92 Gasoline
Density (g/cm3 ) Kinematic viscosity at 40 ◦ C (mm2 /s) Gross heating value (kJ/g) Water content (wt %) Octane no. Cetane index Acid value (mg of KOH/g) Elemental analysis (wt %) C H O
0.91 37.22 39.34 0.32 – – 9.5
0.82 1.97 43.44 0.39 – – 0
0.77 0.75 42.83 0.42 76 – 1.2
0.84 2.11 43.99 0.37 – 51 0.5
0.78 1.0–1.9 44.00 – Min. 92 – 0
79.88 12.64 7.48
83.77 13.42 2.81
84.96 11.56 3.48
84.60 11.84 3.56
85.34 11.43 3.23
0.84 2.0–4.5 43.56 – – Min. 48 0 86.31 13.59 0.1
100
Case 90
1 2 3 4 5 6 7
80 70
Yield (%)
impulse’. It can be seen from Table 3, that with the same mass flow of the feedstock, the residence time of oil gas was different independently of the reactor being packed with a catalyst or not. The pyrolytic system achieved a steady operating state 6–8 h after startup, thus the experimental data in Tables 4 and 5 is the average of the performance during these 2 h. The factors addressed in Table 3 were used to evaluate the pyrolysis performance for the reactor bed: (a) with and without a catalyst, (b) packed with acidic and basic catalysts, and (c) packed with different quantities of base catalysts. The optimum catalyst, determined from the experiments based on Tables 3–5 was used to measure the effects of feed rates and evaporator and pyrolysis reactor temperatures on the properties of pyrolytic oils.
Diesel
60 50 40 30 20 10
3.1. Pyrolysis with and without catalyst
0
PO
Based on Table 3, the WCO pyrolysis product distribution (i.e. the yield of BGO, BDO, BG, OS, water, and coke) for the seven cases with catalyst (cases 3 to 7 in Table 3) and without (cases 1 and 2 in Table 3) at the same conditions of feed rate and operating temperatures of the evaporator and the pyrolitic reactor are shown in Fig. 2. Case 1 used fresh soybean oil (AV = 0.2) as the feed stock while cases 2–7 used WCO (AV = 9.5) as the feed stock. Table 4 gives the properties of the pyrolytic oils obtained from the experiments (cases 1–7 in Table 3). As shown in Table 4, high AV (around 110), high oxygen contents (about 10 wt%), and high freezing points (4–5 o C) were the main features of the PO obtained in the pyrolytic reactor operated without catalyst (cases 1 and 2) compared to those with catalysts (cases 3–7). Without a catalyst, the oil molecules might be partially oxidized into acid components. Thus, high acid values and high oxygen contents can be found in the experiments for cases 1 and 2. The reason for this result will be discussed in Section 3.7. The PO from the acid catalyst acid (case 3) gives a medium AV (32) but a high oxygen content (10 wt%). The cases using base catalyst pellets (Na2 CO3 , NaOH, and CaO), appear to promote the reaction of the fatty acids derived from the thermal decomposition of fresh edible oils or WCO to aldehydes or ketones [29–30]. This result is also supported by observing that the
BGO
BDO
BG
OS
Water
Coke
Product Fig. 2. Product distribution obtained from no-catalytic (cases 1–2) and catalytic pyrolysis (cases 3–7). The given conditions are shown in Tables 1 and 2.
oxygen content is reduced to below 5 wt% for cases 3–7 (Table 4). Fig. 2, shows that similar production rates for PO and its components BGO, BDO, BG, OS, water, and coke were obtained both with the fresh soybean oil and WCO by the thermal pyrolysis without catalyst (cases 1 and 2). However, the yields of PO without catalyst (cases 1 and 2) are relatively higher than those with catalyst (cases 3–7). As the editable soybean oil was only used for comparison, we shall only focus on the productivity of WCO. Because the total yield of BG, BGO, BDO, OS, water, and coke equals one, if less coke is formed, greater feedstock productivity can be obtained. The coke formation for non-catalytic pyrolysis (cases 1 and 2) was lower than that for catalytic pyrolysis (cases 3–7). In addition to thermal cracking, catalytic cracking also occurred in the catalytic pyrolysis. However, the high AV of the PO, produced by non-catalytic pyrolysis, counterbalances this benefit. Additionally, the coke formation of case 7 (18.85 wt%) is only slightly larger
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
than that of case 2 (15.51 wt%), suggesting that the base catalyst pellet (CaO/bentonite) is a potential choice to produce a near zero AV for PO. Based on the above discussion and the need to obtain pyrolytic oils with low acid values, pyrolysis without catalyst, as a process option, was rejected. 3.2. Comparison between acid and base catalysts Comparison of case 3 (acid catalyst) with cases 4–7 (base catalysts) from Table 4, shows that the AV and the oxygen content of the produced pyrolytic oils using base catalysts are much lower than those that result from the use of acid catalysts (Al2 (SO4 )). The deoxygenation capability using acid catalyst is much lower than that achieved using base catalysts, as highlighted by the fact that the AV of case 3 remains high (∼33) using the acid catalyst pellet (Al2 (SO4 )3 ). Further, coke formation using the acid catalyst is the highest among these seven cases as shown in Fig. 2; therefore, the use of base catalyst pellets (Na2 CO3 , NaOH, and CaO) is a preferential route to obtaining a PO with the sought after characteristics. 3.3. Comparison among the base catalysts for short-term operation Three base catalyst pellets (Na2 CO3 , NaOH, and CaO) were evaluated in this work. Observing the performance of cases 4– 6 as shown in Table 4 and Fig. 2, one can find PO with varying components (BGO, BDO, BG, OS, water, and coke). The coke yield ranged from 17 to 22 for cases 4–6. The BGO production rate using the catalyst pellets Na2 CO3 and NaOH (cases 4 and 5) was higher (∼38 wt%) than when using CaO catalyst pellet (case 6). The capability to transform free fatty acids to aldehydes or ketones using the Na2 CO3 and NaOH catalyst pellets seems to be similar to that when using CaO (AV of POs ranged from 0 to 2.2). Case 7 used a greater quantity of CaO catalyst pellets (decreasing WHSV) to achieve a near zero AV for the PO produced and a higher yield of BGO compared to that of case 6 using a lower quantity of CaO catalyst pellets. 3.4. Comparison of base catalysts for long-term operation In addition to the short-term performance of the catalyst pellets in the fixed-bed pyrolytic reactor, the long-term operation using the base catalyst pellets can provide more information about the capability of the base catalyst pellets. Fig. 3 depicts the development of the AV of pyrolytic oils over time for cases 4–7 (Table 3). The location of the last bar plotted shows that the fixed-bed reactor was blocked with deposited coke or soap-like matter along the bed and could not therefore be operated any longer. One can observe that the catalyst pellets (case 4 (Na2 CO3 ) and case 5 (NaOH)) cannot be operated after 16 h from start-up while the catalyst pellet (CaO) can maintain operation for 28 h. The soap-like matter produced was the special feature of the pyrolysis using the catalyst pellets (Na2 CO3 and NaOH) but was not found using the CaO catalyst pellet. Therefore, from the long-term operational viewpoint, the CaO catalyst pellet is a better choice than the catalyst pellets (Na2 CO3 and NaOH). Comparison of cases 6 and 7 shown in Fig. 3, reveals that a low WHSV (more CaO catalyst) was required to obtain a zero AV for the PO produced during a time interval of 12 h after startup. Therefore, setting the lifetime of CaO catalyst pellet to 12 h is appropriate when WHSV equals 1.04 h–1 and the catalyst needs to be regenerated. From previous comparisons (Sections 3.1–3.4), the CaO catalyst pellet was the best choice for achieving a low PO acid value when made from WCO using a fast pyrolysis process. Table 5 gives the main fuel properties of WCO, PO, BGO, BDO, and commercial gasoline and diesel from the experiment case 7 (Table 3). The properties of the obtained BDO were well matched with those of
5
commercial diesel. The kinematic viscosity of the produced BGO being lower than that of commercial gasoline means smaller components were obtained using the front-view distillation apparatus (Koehler K45200). The octane number 76 of the produced BGO is ‘off-specification’ and needs to be blended with an octane booster to meet the requirement of the specification. The properties shown in Table 5 are reflected in Fig. 4 which shows the GC-mass spectra of the pyrolytic products of case 7 for comparing the component distribution (C8-C24) of (a) diesel and PO, (b) gasoline and BGO, and (c) diesel and BDO. Light components exist in the GC-mass spectra of PO, BGO, and BDO when compared to those of commercial diesel and gasoline. 3.5. The effects of pyrolysis temperature and feed rates (WHSV) on the product distribution using the CaO catalyst pellet Using the operating condition (case 7 in Table 3), the effect of the pyrolysis temperature on the product distribution, using the CaO catalyst pellet at the same WHSV, is shown in Fig. 5. The coke formation yield is too high to be acceptable at 550 o C. Comparable liquid product (BGO and BDO) yields can be obtained for the pyrolytic temperatures 450 o C and 500 o C but with different yield distributions between BGO and BDO. A low coke yield but high OS yield was found at 450 o Ccompared to 500 o C; however, the AV of PO at 450 o C was 1.85 and zero at 500 o C, suggesting a pyrolytic temperature set to 500 o C seems to be most suitable. Using the operating conditions (case 7 in Table 3) with a varying the feed rate (WHSV), gives the yield distribution shown in Fig. 6. The AV of PO at a feed rate 60 g/h was 2.43 and zero at other feed rates. Because the smallest coke yield and highest yield of liquid products (BGO and BDO) was obtained at a feed rate of 39 g/h, operating the catalytic cracking system at WHSV(1.04 h−1 ) appears to be a good choice. 3.6. Catalyst regeneration In general, catalyst regeneration is needed to maintain PO product quality during long-term operation. Thus, reactivation of catalyst would be needed after running the process for a period of time. In this work, a catalyst regeneration process was carried out by burning out the coke deposits with air at 600 o C for 4 h. After regeneration of catalyst, the pyrolysis activity was increased to almost the initial state as shown in Fig. 7. 3.7. The roles of feedstocks, catalyst acidity and basicity on composition distribution of PO Several authors [7,31–32] agree that the initial triglyceride catalytic cracking step, using acidic catalysts, involves the initial thermal decomposition of the triglycerides to yield heavy oxygenated hydrocarbons by a free radical mechanism, which is independent of the catalyst’s characteristics. These high-molecular weight oxygenated compounds undergo secondary cracking reactions to give different products wherein the catalyst’s acidity plays a significant role [7,31–32]. Dupain et al. [30] carried out the catalytic cracking of rapeseed oil with commercial equilibrated FCC catalyst (Ecat) under realistic FCC conditions. They reported that triglyceride molecules are mainly converted within 50 ms between 485 and 585 °C into fatty acids through radical cracking reactions. The formation of radicals is enhanced by the catalyst’s external surface. These high-molecular weight oxygenated compounds then undergo secondary cracking reactions to give different products. Once the triglyceride has been thermally decomposed to highmolecular weight oxygenated hydrocarbons, their transformation into various products starts with deoxygenation, e.g. by decarboxylation or decarbonylation and then by secondary cracking reactions
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE 6
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
6
Case 4 (Table 2)
5 4 3 2 1 0 6
Case 5 (Table 2)
5 4 3 2
AV (mg KOH/g)
1 0 6
Case 6 (Table 2)
5 4 3 2 1 0 6
Case 7 (Table 2)
5 4 3 2 1 0 0
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
Time (hr) Fig. 3. Long-term test of the catalyst pellets (Na2 CO3, NaOH, and CaO).
to produce short and long chain olefins and paraffins. The subsequent conversion of short-chain olefins (typically C2 –C4 olefins) into C2 –C10 olefins, aliphatic and aromatic hydrocarbons is promoted by catalytic mechanisms such as oligomerization, cyclization, and aromatization in the pores of the catalyst used. The formation of coke might result from the thermal polycondensation of triglycerides and/or the catalytic polymerization of aromatic hydrocarbons [7,31–32]. The results of this experimental work support the observations above, based on the GC-mass analysis, of the component distributions of the 7 cases and show that we can effectively improve the fast pyrolysis process for obtaining a good PO yield and quality from WCO using the information in this section. The composition distribution of gaseous products includes CO, CO2 , C4 –C6 , C3 H6 , C3 H8 , C2 H2 , C2 H4 , C2 H6 , CH4 , and H2 . The PO compositions are divided into 6 categories, including the com-
pounds that contain: (a) hydroxyl groups (alcohols), (b) carboxylic acid groups (carboxylic acids), (c) carbonyl groups (mainly aldehydes and ketones), (d) aromatic hydrocarbons (HCs), (e) alicyclic hy drocarbons, and (f) linear paraffins. The relative yield ζ (= i (yield (%) of gas product or PO × area (%) of GC-mass × quality (%)) of the identified component i) is adopted for the x-axis of Figs. 8–11. The y-axis of Figs. 8(a)–11(a) is the individual gas component, and that of Figs. 8(b)–11(b) is the individual category of PO product. The relative yield (ζ ) was expected to reflect the relative productivity of the individual gas components, or the individual category of the PO product, when the response factor for each GC-mass peak was not identified. Fig. 8 compares the composition distribution of fresh soybean oil (case 1) and WCO (case 2) using noncatalytic pyrolysis. The compositions of fresh soybean oil and WCO are shown in Table 2, which shows that the main difference lies
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
7
Diesel PO
a
C15 C14 C8
C16
C13
C19 C17 C18
C9 C11 C12
C20 C21 C22 C23C24
C10
Gasoline BGO
b
c Diesel BDO
5
10
15
20
25
30
35
40
45
50
55
60
Time (min) Fig. 4. GC-mass spectra of the pyrolytic products (case 7): (a) diesel and PO, (b) gasoline and BGO, and (c) diesel and BDO.
in the fractions of saturated and unsaturated fatty acids. The fresh soybean oil contained 85 wt% unsaturated fatty acids while WCO contained two-fold more saturated fatty acids than fresh soybean oil (31 wt% versus 15 wt%). Fig. 8(b) reveals comparatively high contents of carboxylic acids in PO in cases 3–7 (catalytic pyrolysis) obtained from both fresh soybean oil and WCO feedstock, which resulted in high AV as shown in Table 4. Aromatic HCs are much greater in fresh soybean oil than that in WCO. On the other hand,
alicyclic HCs and linear paraffins are relatively higher in WCO than in fresh soybean oil. Additionally, the composition distributions of gaseous products, obtained from fresh soybean oil, are relatively higher than those obtained from WCO (Fig. 8(a)). It has been reported that fatty acids with higher content of unsaturated alkyl chains exist in soybean oil and have a strong tendency to undergo a very fast aromatization process (via sequential cyclization, dehydrogenation and condensation reactions) to form (poly)aromatic
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE 8
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
Fig. 5. Pyrolysis product distribution of catalyst pellet CaO (case 7) at different temperatures (◦ C).
100
Feed rate (g/hr) 20 39 60 80
90 80
Yield (%)
70 60 50 40 30 20 10 0
PO
BGO
BDO
BG
OS
Water
Coke
Product Fig. 6. Pyrolysis product distribution of the catalyst pellet CaO (case 7) at different feed rates (g/h).
100
4. Conclusions
Cycle 90
1 2 3 4 5 6
80
Yield (%)
70 60 50 40 30 20 10 0
PO
BGO
BDO
BG
OS
Water
bodies with alkyl branches using acidic catalysts [30,33]. Therefore, higher contents of aromatic HCs can be obtained using fresh soybean oil. In the meantime, because fatty acids with higher saturated alkyl chain contents exist in WCO, higher contents of alicyclic HCs and linear paraffins can be produced using WCO in a thermal pyrolysis process. As it is friendlier to the environment to produce alicyclic HCs and linear paraffins than aromatic HCs, WCO with higher saturated fatty acid content is a better feedstock for thermal pyrolysis than valuable edible oils with higher unsaturated fatty acid contents. Fig. 9 compares the product distribution between non-catalytic pyrolysis (case 2) and pyrolysis using an acidic catalyst (case 3) with WCO as the feedstock. The results show that case 3 produced a much higher quantity of gaseous products using the acidic catalyst Al2 (SO)3 compared to non-catalytic pyrolysis (case 2), with coke formation (case 3) being much higher than for the other cases, resulting in the lowest production of PO among the seven cases (Fig. 9b). Fig. 10 compares the product distribution between a noncatalytic pyrolysis (case 2) and a catalytic pyrolysis using the basic CaO catalyst pellets (case 6) with WCO as the feedstock. A nearzero content of carboxylic acids and a high content of aldehydes and ketones was produced using the base catalyst (CaO) can be seen in Fig. 10 (b) this resulted in a very low AV of PO (AV = 2.2). Similar results can be found in the literature [29,34,35]. Catalytic decarboxylaion by removing naphthenic acid using the CaO catalyst was examined [34]. The bimolecular reaction mechanism converting aliphatic acid to ketones using CaO as catalyst was proposed in the work of Pestman et al. [29], while the oxygen content reduction in the bio-oil using catalytic CaO was evaluated in the work of Lin et al. [35]—which taken together suggest that the reduction of the fatty acids derived from radical cracking reactions of WCO to aldehydes and ketones can be assumed in experiments (case 6 and case 7) of this work. Furthermore, a very low aromatic HC content was obtained using the base catalyst. Using WCO as feedstock, Fig. 11 compares the product distribution between the pyrolysis using less CaO pellets (case 6), or more CaO catalyst pellets (case 7). With more CaO pellets, the aldehydes and ketones are converted to other PO product categories (alcohols, aromatic hydrocarbons (HCs), and alicyclic hydrocarbons) under the catalytic condition provided by the CaO pellets. Gaseous products increase with more CaO pellets (case 7) compared to that of case 6; however, the main result is that with enough CaO pellets the AV of the PO approaches zero, see Table 4 (case 7).
Coke
Product Fig. 7. Pyrolysis product distribution of the catalyst pellet CaO (case 7) at different catalyst regeneration cycles (recycle).
A pyrolytic system composed of an instant evaporator and a tubular pyrolytic reactor was successfully built. Fast pyrolysis of WCO was stably carried out in this pyrolytic system to obtain a near zero acid value oil using the optimized conditions WHSV = 1.04 h−1 and CaO pellet catalyst. The properties of the obtained BDO matched well with those of commercial diesel. The octane number 76 of the produced BGO is ‘off-specification’ and thus the product needs to be blended with an octane booster to meet the requirement of the specification of commercial gasoline. The degree coke formation (18.85 wt%) was only slightly larger than with non-catalytic thermal pyrolysis (15.51 wt%). The acidic catalyst pellets (Al2 (SO4 )3 ) and the related acidic zeolite based catalysts such as ZSM-5, MCM-41, and Y zeolite produced a PO with a high AV and a high content of aromatic HCs which are not friendly to the environment. Reduction of the fatty acids derived from thermal radical cracking reactions of WCO to aldehydes and ketones was the main role of the adopted CaO pellet catalyst in the catalytic pyrolysis of WCO. During the long running tests, the soap-like matter produced was the ‘special feature’ of the pyrolysis
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
a
9
b Case 1 Case 2
CO
Case 1 Case 2
Alcohols
CO2 C4-C6
Carboxylic acids
Propene Aldehydes, Ketones
Propane Ethyne
Aromatic HCs
Ethene Alicyclic HCs
Ethane Methane
Linear paraffins Hydrogen
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0
5
10
15
Relative yield
20
25
30
35
40
45
50
Relative yield
Fig. 8. Composition distribution of (a) gas product and (b) PO for comparison of case 1 and case 2.
a
b Case 2 Case 3
CO
Case 2 Case 3
Alcohols
CO2 C4-C6
Carboxylic acids
Propene
Aldehydes, Ketones
Propane Ethyne
Aromatic HCs
Ethene
Alicyclic HCs
Ethane Methane
Linear paraffins Hydrogen
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0
5
10
Relative yield
15
20
25
30
35
40
45
50
Relative yield
Fig. 9. Composition distribution of (a) gas product and (b) PO for comparison of case 2 and case 3.
a
b Case 2 Case 6
CO
Case 2 Case 6
Alcohols
CO2 C4-C6
Carboxylic acids
Propene Aldehydes, Ketones
Propane Ethyne
Aromatic HCs
Ethene Alicyclic HCs
Ethane Methane
Linear paraffins
Hydrogen 0.0
0.5
1.0
1.5
2.0
Relative yield
2.5
3.0
3.5
4.0
0
5
10
15
20
25
30
35
40
45
50
Relative yield
Fig. 10. Composition distribution of (a) gas product and (b) PO for comparison of case 2 and case 6.
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
ARTICLE IN PRESS
JID: JTICE 10
[m5G;May 5, 2016;9:34]
J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
a
b Case 6 Case 7
CO
Case 6 Case 7
Alcohols
CO2 C4-C6
Carboxylic acids
Propene Aldehydes, Ketones
Propane Ethyne
Aromatic HCs
Ethene Alicyclic HCs
Ethane Methane
Linear paraffins Hydrogen
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Relative yield
0
5
10
15
20
25
30
35
40
45
50
Relative yield
Fig. 11. Composition distribution of (a) gas product and (b) PO for comparison of case 6 and case 7.
using the catalyst pellets (Na2 CO3 and NaOH) but was not found using the CaO catalyst pellet. Therefore, the CaO catalyst pellet is a better choice than the catalyst pellets (Na2 CO3 and NaOH). This point was revealed for the first time in this work. Regeneration of the used CaO pellets can return them to their initial activities. Although the price of pure CaO powder is very low, a way to increase the CaO content in the pellet still needs to be found in order to increase the process efficiency. In addition to the process advantages offered by the tubular pyrolysis system employing CaO catalyst pellets discussed above, the cost of building the system was much less than that of a comparable fluidized catalytic cracking reactor system. In addition to which the ability to operate the system without a carrier gas also results in a worthwhile cost saving. However, a significant disadvantage remains in that the maximum life-cycle of the catalyst pellets was only 28 h; therefore, methods need to be developed to extend the life-cycle of the catalyst pellets before scaling the process up in size to that of a commercial scale system. A possible way to address this problem may be to adopt a spouted bed with smaller sized catalyst pellets. Dedication This paper is dedicated to the memory of the late Prof. Y. C. Chao (ddd), who through his inspirational teaching, encouraged us all to continue in his path. Acknowledgments Funding for this work, provided for this research by the Ministry of Science and Technology of Republic of China (Grant No.: MOST 104-3113-E-036-001), is gratefully acknowledged. References [1] Canakci M, Van Gerpen J. A pilot plant to produce biodiesel from high free fatty acid feedstocks. J Am Soc Agri Biol Eng 2003;46(4):945–54. [2] Kulkarni MG, Dalai AK. Waste cooking oil-an economical source for biodiesel: a review. Ind Eng Chem Res 2006;45(9):2901–13. [3] Shi H, Hui Zhang. Waste oil and fat feedstocks for biodiesel production. Adv Petrol Explor Dev 2014;8(1):31–6. [4] Patil PD, Gude VG, Reddy HK, Mupaswy T, Deng S. Biodiesel production from waste cooking oil using sulfuric and microwave irradiation process. J Environ Protect 2012;3:107–13. [5] Wiggers VR, Meier HF, Wisniewski A Jr, Chivanga Barros AA, Wolf Maciel MR. Biofuel from continuous fast pyrolysis of soybean oil: a pilot plant study. Bioresour Technol 20 09;10 0:6570–7.
[6] Demirbas A. Biodiesel fuels from vegetable oils via catalytic and non-catalytic supercritical alcohol transesterificationa and other methods: a survey. Energy Convers Manage 2003;44:2093–109. [7] Maher KD, Bressler DC. Pyrolysis of triglyceride materials for the production of renewable fuels and chemicals. Bioresour Technol 2007;98:2351–68. [8] Xu J, Jiang J, Lu Y, Chen J. Liquid hydrocarbon fuels obtained by the pyrolysis of soybean oils. Bioresour Technol 20 09;10 0:4867–70. [9] Meier HF, Wiggers VR, Zonta GR, Scharf DR, Simionatto EL, Ender L. A kinetic model for thermal cracking of waste cooking oil based on chemical lumps. Fuel 2015;144:50–9. [10] Shadangi KP, Mohanty K. Thermal and catalytic pyrolysis of Karanja seed to produce liquid fuel. Fuel 2014;115:434–42. [11] Hoydonckx HE, Vos DEDe, Chavan SA, Jacobs PA. Esterification and transesterification of renewable chemicals. Topics Catal 2004;27(1):83–96. [12] Patel A, Narkhede N. Biodiesel synthesis via esterification and transesterification over a new heterogeneous catalyst comprising lacunary silicotungstate and MCM-41. Catal Sci Technol 2013;3:3317–25. [13] Lien YS, Hsieh LS, Wu JCS. Transesterification using oleophilic acid catalyst. Ind Eng Chem Res 2010;49(5):2118–21. [14] http://www.extension.org/pages/26636/transportation-and-storage-ofbiodiesel#.VdgYdbKqpHw. [15] Maher KD, Bressler DC. Pyrolysis of triglyceride materials for the production of renewable fuels and chemicals. Bioresour Technol 2007;98:2351–68. [16] Padmaja KV, Atheya N, Bhatnagar AK, Singh KK. Conversion of calotropis procera biocrude to liquid fuels using thermal and catalytic cracking. Fuel 2009;88:780–5. [17] Katikaneni SPR, Adjaye JD, Bakhshi NN. Performance of aluminophosphate molecular-sieve catalysts for the production of hydrocarbons from wood-derived and vegetable-oils. Energy Fuels 1995;9(6):1065–78. [18] Idem RO, Katikaneni SPR, Bakhshi NN. Thermal cracking of canola oil: reaction products in the presence and absence of steam. Energy Fuels 1996;10(6):1150–62. [19] Schwaba AW, Dykstrab GJ, Selkea E, Sorensonb SC, Prydea EH. Diesel fuel from thermal decomposition of soybean oil. JAOCS 1988;65(11):1781–6. [20] Hotová G, Slovák V. Effect of pyrolysis temperature and thermal oxidation on the adsorption properties of carbon cryogels. Thermochim Acta 2015;614:45–51. [21] Xu J, Jiang J, Sun Y, Chen J. Production of hydrocarbon fuels from pyrolysis of soybean oils using a basic catalyst. Bioresour Technol 2010;101:9803–6. [22] Xu J, Jiang J, Chen J, Sun Y. Biofuel production from catalytic cracking of woody oils. Bioresour Technol 2010;101:5586–91. [23] Oasmaa A, Elliott CE, Korhonen J. Acidity of biomass fast pyrolysis bio-oils. Energy Fuels 2010;24:6548–54. [24] Xu J, Xiao G, Zhou Y, Jiang J. Production of biofuels from high-acid-value waste oils. Energy Fuels 2011;25:4638–42. [25] ASTM D664-11a Standard test method for acid number of petroleum products by potentiometric titration. [26] Ding J, Xia Z, Lu J. Esterification and deacidification of a waste cooking oil (TAN 68.81 mg KOH/g) for biodiesel production. Energies 2012;5:2683–91. [27] Chen JC, Liu CY. Regeneration of spent catalysts with ozone. Int J Environ Chem Ecol Geol Geophys Eng 2013;7(7):414–17. [28] Arabiourrutia M, Olazar M, Aguado R, Lopez G, Barona A, Bilbao J. HZSM-5 and HY zeolite catalyst performance in the pyrolysis of tires in a conical spouted bed reactor. Ind Eng Chem Res 2008;47:7600–9. [29] Pestman R, Koster RM, Avan D, Pieterse JAZ, Ponec V. Reactions of carboxylic acids on oxides 2. bimolecular reaction of aliphatic acids to ketones. J Catal 1997;168:265–72.
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014
JID: JTICE
ARTICLE IN PRESS J.-S. Chang et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2016) 1–11
[30] Dupain X, Costa DJ, Schaverien CJ, Makkee M, Moulijn JA. Cracking of a rapeseed vegetable oil under realistic FCC conditions. Appl Catal B: Environ 2007;72:44–61. [31] Melero JA, Iglesias J, Garcia A. Biomass as renewable feedstock in standard refinery units. feasibility, opportunities and challenges. Energy Environ Sci 2012;5:7393–420. [32] Taufiqurrahmi N, Bhatia S. Catalytic cracking of edible and non-edible oils for the production of biofuels. Energy Environ Sci 2011;4:1087–112.
[m5G;May 5, 2016;9:34] 11
[33] Rao TV M, Dupain X, Makkee M. Catalytic cracking of edible and non-edible oils for the production of biofuels Fluid catalytic cracking: processing opportunities for fischer–tropsch waxes and vegetable oils to produce transportation fuels and light olefins. Micropor Mesopor Mater 2012;164:148–63. [34] Ding L, Rahimi P, Hawkins R, Bhatt S, Shi Y. Naphthenic acid removal from heavy oils on alkaline earth-metal oxides and ZnO catalysts. Appl Catal A: Gen 2009;371:121–30. [35] Lin Y, Zhang C, Zhang M, Zhang J. Deoxygenation of bio-oil during pyrolysis of biomass in the presence of CaO in a fluidized-bed reactor. Energy Fuels 2010;24:5686–95.
Please cite this article as: J.-S. Chang et al., Low acid value bio-gasoline and bio-diesel made from waste cooking oils using a fast pyrolysis process, Journal of the Taiwan Institute of Chemical Engineers (2016), http://dx.doi.org/10.1016/j.jtice.2016.04.014