Fuel Processing Technology 91 (2010) 296305

Contents lists available at ScienceDirect

Fuel Processing Technology

Production and fuel properties of fast pyrolysis oil/bio-diesel blends

Department of Chemical Engineering, Monash University, Victoria 3800, Australia Department of Biological Systems Engineering, Washington State University, Pullman, WA, 99164, USA c Department of Chemical Engineering, Taiyuan University of Technology, Taiyuan, Shanxi, 030024, People's Republic of China d Curtin Center for Advanced Energy Science and Engineering, Curtin University of Technology, GPO Box U1987, Perth, WA 6845, Australia
b

a r t i c l e

i n f o

a b s t r a c t
This paper describes the production and fuel properties of fast pyrolysis oil/bio-diesel blends. The bio-oils used in this study were produced from the fast pyrolysis of woody biomasses, oil mallee and pine. The biodiesel employed was derived from canola vegetable oil. The conditions used to prepare the bio-oil/bio-diesel blends, as well as some of the fuel properties of the resulting bio-diesel rich phase, are reported. The experimental results show that the solubility of fast pyrolysis oils in bio-diesel is not as high as was previously reported for decanted oils obtained by Auger pyrolysis. The carboxylic acids, mono-phenols, furans and lignin derived oligomers were the compounds most soluble in bio-diesel, while the sugars, on the other hand, showed poor solubility. Although the presence of phenols enhances the oxidation stability of the bio-diesel rich phases, other fuel properties deteriorate. For example, the content of solid residues increased primarily because of the solubilisation of lignin derived oligomers, which were quantied by UVuorescence. Concentrations as high as 3.5 mass % of these compounds were observed in the bio-diesel rich phase. The solubility of bio-oil in bio-diesel was enhanced by using ethyl acetate/bio-diesel blends. Some fuel properties of the bio-diesel rich phase, after the removal of ethyl acetate, are reported. 2009 Elsevier B.V. All rights reserved.

Article history: Received 1 August 2009 Received in revised form 10 October 2009 Accepted 19 October 2009 Keywords: Bio-oil Bio-diesel Pyrolysis Fuel properties

1. Introduction High petroleum prices and the undesirable environmental impact of using fossil fuels, as well as the need to spur rural development, are the primary reasons for the increased interest in bio-fuels. While in the 1970s one half of the petroleum consumed was used to produce transportation fuels, by 2030 it is expected that more than two thirds of the petroleum produced will be employed for this purpose. Biomass has the potential to mitigate the use of petroleum since it is the only renewable alternative source for carbon based fuels and chemicals. Pyrolysis is an excellent technology, able to convert up to 70 mass % of the biomass into a crude bio-oil, which can be transported to centralized reneries for the production of transportation fuels and chemicals. The oil resulting from biomass pyrolysis is a dark liquid with an elemental composition very similar to that of the biomass. These oils are composed of 10 to 25 mass % of water and between 75 and 90 mass % organic compounds [1]. The small amounts of solid carbon and ash found in these oils are due to the poor separation efciencies of the cyclones commonly used to separate the charcoal from the pyrolytic vapors.

Corresponding author. Curtin Center for Advanced Energy Science and Engineering, Curtin University of Technology, 1 Turner Avenue, Technology Park, WA 6102 GPO Box U1987, Perth, WA 6845, Australia. Tel.: +61 8 9266 1131or1133; fax: +61 8 9266 1138. E-mail address: chun-zhu.li@curtin.edu.au (C.-Z. Li). 0378-3820/$ see front matter 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.fuproc.2009.10.012

Although, the production of crude bio-oils through pyrolysis has been the subject of intensive research for the past three decades [27], very little progress has been made to produce additives or transportation fuel extenders from these oils [8]. Bio-oil has been employed as a fuel to produce electricity in modied gas turbines and diesel engines [9]. These oils cannot, however, be used in their present form as transportation fuels. The high acidity, low thermal stability, low caloric value, high viscosity and poor lubrication are some of the properties limiting their direct use as transportation fuel. Thus there is an urgent need to develop new approaches to utilize these oils as sources of fuel additives or extenders [1,9]. Fully rening bio-oil to obtain transportation fuels and chemicals is a long term goal of many researchers. However, successful deployment of bio-oil reneries will depend on several new separation and reaction technologies which may take time to move from the laboratory to the industrial scale. In this context, creating blends of bio-oil with other transportation fuels could be a viable short-term alternative to utilize an important fraction of these oils [10]. Bio-diesels are methyl or ethyl esters of fatty acids with properties very comparable with those obtained for petroleum derived diesel. Bio-diesel is commonly derived from the trans-esterication of vegetable oils or animal fats with alcohols using acidic or basic catalysts [1113]. These fuels are completely miscible with petroleum derived fuels and have been proposed as an alternative to increase the

M. Garcia-Perez et al. / Fuel Processing Technology 91 (2010) 296305

297

lubricity of petroleum distilled fractions [1416]. Commercial biodiesels are commonly produced from soybean, canola, rapeseed, sunower and palm oil. Although these fuels are better solvents than petroleum derived diesel, they tend to have lower oxidation stability and comparatively poor cold ow properties [17,18]. The formation of crystals at relatively high temperatures is one of the major hurdles for the use of bio-diesel in cold weathers. The increase in the content of methyl and ethyl esters of saturated fatty acids in the range C16C18 deteriorates the cold ow properties of bio-diesel. Conversely, oil rich in unsaturated fatty acids has better cold ow properties, but exhibits lower oxidation stability. The oxidation stability of bio-diesels can be improved by the addition of phenolic compounds which are known to be excellent antioxidants. Reviews on the fuel properties of biodiesels can be found elsewhere [19,20]. The use of bio-diesel to extract some of the bio-oil fractions, particularly the high concentrations of phenolic compounds, could be a viable way to partially utilize bio-oils as additives for transportation fuels, while also improving the oxidation stability and cold ow properties of bio-diesels. Although it was previously reported that the decanted phase of the oil resulting from the Auger pyrolysis of pine is quite soluble in biodiesel [8], investigations of fast pyrolysis bio-oils as possible bio-diesel extenders have not been conducted. Thus, the main goal of this paper is to evaluate the solubility of fast pyrolysis bio-oils in bio-diesel and the fuel properties of the resulting bio-diesel rich phases. The solubility of lignin derived oligomers in the bio-diesel rich phase will be quantied. 2. Material and methods 2.1. Biomass samples Two different feedstocks (the woody fraction of mallee eucalyptus and pine) were employed to produce the bio-oils used in this study. The mallee was supplied by the Western Australian Department of Environment and Conservation. The pine pellets were obtained from the Southern Shaving Co, Cherryville, NC. Both materials were ground using a Fritsch Laboratory Cutting Mill Pulverizette 15. The resulting samples were sieved and particles with diameter between 90 and 600 m used in this study. The elemental compositions of each of the feedstocks are reported elsewhere [21]. 2.2. Fast pyrolysis A 1 kg/h fast pyrolysis unit was employed to obtain the bio-oils used in this study. A detailed description of the experimental set up and the conditions used to produce these oils has been previously reported [21]. Briey, the unit was composed of a system for biomass feeding, a uidized bed reactor, cyclones for charcoal collection, an electrical furnace to heat the carrier gas, bio-oil condensers and gas lters. The uidized bed reactor consisted of a cylindrical section 320 mm long with a 102 mm i.d. The inert bed material was silica sand with particle diameters between 351 and 401 m. Nitrogen, at 43 L/min (TPN), was used as uidization agent. This ow rate corresponds to two times the minimum uidization velocity (Umf = 0.1 m/s). The mass balances shown in Table 1 were carried out by quantifying the biomass pyrolysed as well as the bio-oils, charcoal
Table 1 Yield of products (mass %) at a pyrolysis temperature of 500 C. Products Char Bio-oil Gases Oil mallee (fast pyrolysis) 13.9 61.3 11.1 Pine pellets (fast pyrolysis) 13.8 64.3 13.8

and gases obtained. The yields of products were very similar for both feedstocks, with the fast pyrolysis of oil mallee woody biomass and pine yielding 61.3 and 64.3 mass % of oils, respectively. Around 14 mass % of charcoal was obtained for both feedstocks. 2.3. Analysis of bio-oils The bio-oils from mallee and pine were stored at 4 C before analysis. The caloric value was measured using a Leco AC350 calorimeter following Australian Standards (AS1038.5). Carbon, hydrogen and nitrogen were measured with the LECO CHN 100 analyser according to the Australian Standard AS2434.8. These analyses were carried out at HRL (Melbourne, Australia). The water content was determined by Karl Fischer titration (KF DL31 from Mettler-Toledo). A MooreVan Slyke-specic gravity bottle (Fischer Scientic) was used to determine the density of bio-oils. The measurement was carried out at 20 C (ASTM D-4809). The thermogravimetric analysis of the studied oils was performed to obtain an overall description of their boiling point distribution and cracking characteristic temperatures. The tests were carried out under nitrogen in a Perkin-Elmer Pyris 1 TGA. A 510 mg sample of bio-oil was placed in a platinum crucible in the TGA. The test was started as soon as the bio-oil fractions were added to the testing pan to avoid losses of very light fractions. The samples were heated from 30 to 500 C at a heating rate of 10 C/min. A total of 20 mL/min of nitrogen was used to obtain an inert atmosphere and to remove the evaporation and cracking products from the oven. The differential thermogravimetric (DTG) curves were resolved into six major families of compounds using a method described elsewhere [22,23]. The concentrations of targeted chemical species were determined by a Hewlett-Packard GC/MS (HP6890 series with an HP5973 MS detector) with a capillary column (HP-5MS, HP19091S-433). The content of lignin derived oligomers was determined by precipitation in cold water following the method described elsewhere [23]. Briey, 3 g of bio-oil were added dropwise to 300 mL of ice-cooled water under strong agitation. The water-insoluble fraction was removed by ltration. The solid was then washed with water for 5 min before being extracted with dichloromethane, the solid remaining in the lter was dried at 105 C for 1 h. The dichloromethane soluble fraction was rotary evaporated at 40 C and the remaining solid residue weighed and reported as the water-insoluble-CH2Cl2-soluble fraction. The content of lignin derived oligomers extracted by the bio-diesel was determined by UV-uorescence. The spectra were recorded using a Perkin-Elmer LS50B spectrometer and quartz cells with 10 mm light pass length. Solutions containing 4 ppm of each of the following were prepared: bio-oil, water-insoluble-CH2Cl2-soluble fractions, and the bio-diesel rich phases. These low concentrations are needed to avoid the effect of self-absorption. The synchronous uorescence spectrum was recorded with a constant energy difference of 1400 or 2800 cm 1 and a scan speed of 200 nm min 1. 2.4. Commercial bio-diesel The bio-diesel employed in this study was kindly supplied by Conservo (http://www.conservo.com.au/). This bio-diesel was produced by the trans-esterication of canola vegetable oil with methanol, and complies with all Australian Standards. (http://www. environment.gov.au/atmosphere/fuelquality/standards/biodiesel/ summary.html). 2.5. Preparation of and analysis of fast pyrolysis oil/bio-diesel blends Mallee and pine derived fast pyrolysis oils were blended separately with the commercial bio-diesel. A total of 30 g of blends containing 10, 20, 40 and 50 mass % of bio-oils were prepared. The blends were heated to 60 C for 30 min in 50 mL sealed vials. The vials were placed

298

M. Garcia-Perez et al. / Fuel Processing Technology 91 (2010) 296305 Table 3 Composition of bio-oil determined by GC/MS (mass % of bio-oil). No Compound 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Hydroxyacetaldehyde Acetic acid Propane, 2,2 dimethoxy 2-propanone, 1-hydroxy Propanoic acid Cyclopentanone 2-Furaldehyde Furfuryl alcohol Furan, tetrahydro-2,5-dimethoxy (cis) Furan, tetrahydro-2,5-dimethoxy (trans) 2 (5H)-Furanone 2 (3H)-Furanone, 5-methyl 2-Furanethanol, beta-methoxy Phenol 1,2-cyclopentanedione, 3-methyl Phenol 2-methyl Phenol 3-methyl (o-cresol) Phenol, 2-methoxy (guaiacol) 2(3H) Furanone, dihydro-3-hydroxy-4, 4-dimethyl Phenol, 2-methoxy-4-methyl Eugenol Vanillin 1,6 anhydro-beta-D-glucopyranose Hydroquinone Stilbene Phenol, 2,6 dimethoxy-4-(2-propenyl) (cis) (4-propenyl syringol) Phenol, 2,6 dimethoxy-4-(2-propenyl) (trans) (4-propenylsyringol) Syringaldehyde (benzaldehyde, 4-hydroxy-3,5 dimethoxy) Propanoic acid, 3-(4-hydroxy-3methoxyphenyl)2,5-dimethoxy-4-ethylbenzaldehyde Dibenzothiophene RT (min) 1.6 1.9 2.1 2.3 3.2 4.7 5.9 7.0 7.8 8.4 9.2 9.7 11.2 13.2 14.7 16.1 16.4 17.6 19.0 22.5 29.8 33.7 35.4 36.9 38.4 39.8 41.6 42.0 43.4 43.6 44.6 Oil mallee 3,71 5.73 0.59 2.16 1.82 0.12 0.33 0.02 0.34 0.29 0.03 0.07 0.70 0.57 0.04 0.05 0.02 0.12 0.07 0.05 0.51 1.46 6.49 0.09 0.12 0.56 0.41 1.16 0.99 0.31 0.17 Pine pellets 2.66 0.52 3.55 2.54 0.16 0.41 0.03 0.22 0.18 0.04 0.05 0.57 0.32 0.01 0.03 0.368 0.04 0.39 0.21 1.15 6.32 0.05

inside a water bath and left to cool down to room temperature for 24 h before decanting the partially soluble phases. The bio-diesel rich phase was separated using a 10 mL syringe and the weight of the biooil left in the vial determined. Blends of ethyl acetate/bio-diesel were also prepared to enhance the solubility of bio-oils in the bio-diesel rich phase. Briey, 500 mL of a solution containing 50 mass % of ethyl acetate and 50 mass % of biodiesel was prepared. This solution was used to prepare blends containing 10, 20, 40 and 50 mass % of bio-oils. The procedure followed was very similar to the one described in the previous paragraph to prepare bio-oil/bio-diesel blends. After the organic rich phase was separated from the aqueous phase, the ethyl acetate was removed using a rotary evaporator. Some of the fuel properties of the resulting bio-diesel rich phase have been determined. 3. Results 3.1. Bio-oil properties Table 2 compares some fuel properties of the bio-oils obtained from the fast pyrolysis of oil mallee woody biomass and pine pellets. The elemental compositions of these two oils are very similar. The values obtained are also comparable with those reported for other fast pyrolysis oils in the literature [24,25]. The density of bio-oil was approximately 1.2 g/cm3 and the elemental composition was similar to the elemental composition of both feedstocks. Although the gross caloric value of bio-oil per unit mass is only half that of gasoline (usually around 44 MJ/kg) the caloric value per unit volume, (approximately 26 MJ/L) is almost 81% that of gasoline (32 MJ/L). Thus, 1.23 L of bio-oil are needed to generate the same amount of energy as 1 L of petroleum derived fuels. This lower caloric value is mainly due to the high content of oxygen and water in these oils. The concentrations of some individual species found in these biooils, as quantied by GC/MS, are shown in Table 3. Although in general the composition of both oils was quite comparable, it was, as expected, not possible to identify syringols in the bio-oils derived from pine. The lack of syringols can be explained by the fact that pine is a softwood. Syringyl units are only found in hardwood lignin. In general the concentration of different compounds in the bio-oils was very similar to those reported for other fast pyrolysis oils reported in the literature [26,27]. It is noteworthy to mention that the relative high concentration of acetic acid (between 2 and 6 mass %) is responsible for the low pH reported for bio-oils (usually between 2 and 3). Fig. 1 shows the DTG curves for the bio-oils obtained from the fast pyrolysis of pine and oil mallee at 500 C. The DTG curves of bio-oils can be used as a good indication of the boiling point distribution of these oils and their overall composition. Six Gaussian curves representing six groups or families of compounds were used to t the DTG curves. The tting of DTG curves was carried out assuming that the families do not interact during evaporation and cracking. A more detailed explanation of the procedure used can be found elsewhere [22,28]. Whenever possible the values of activation energy (Ej) and reaction order (nj) used in the tting of bio-oil DTG curves were held constant. The values of zjo obtained are shown in Table 4. The values of zjo shown in Table 4 will be used in Section 3.4 to estimate the overall composition of each oil. The actual chemical

composition of each of the peaks tted in the DTG curve are not completely known, however certain compounds have been associated to the indicated peaks. For example, the rst group of compounds (group A) formed by organic compounds with boiling point below 100 C is known to contain important amounts of hydroxyacetaldehyde [23]. The second group (group B) is mainly formed by water, acetic acid and acetol. Mono-phenols and furans are the main species forming the third group (group C) (see Table 3). These compounds have boiling points similar to gasoline. Group D is mainly formed by levoglucosan and other sugars. Although families D and E are difcult to quantify by GC/MS, there is some experimental evidence suggesting that these families are composed of lignin derived oligomers and oligosugars. Peak E is the main peak associated with the lignin derived oligomers. The cracking of oligosugars is believed to be the main factor responsible for the formation of peak F. 3.2. Bio-diesel properties The GC/MS chromatogram for the commercial bio-diesel used in this study is shown in Fig. 2. It can be observed that the bio-diesel

Table 2 General properties of produced bio-oils at 500 C. Temperature (C) Oil mallee Pine pellet
a

Density (g/cm3) (20 C) 1.19 1.22

Elemental analysis (mass %, dry basis) C 52.4 52.0 H 6.8 6.8 N 0.07 0.01 O
a

Water content (mass %) 18.6 18.0

Gross caloric value (dry) (MJ/kg) 21.7 21.6

40.73 41.19

By difference.

M. Garcia-Perez et al. / Fuel Processing Technology 91 (2010) 296305

299

Fig. 1. DTG curves of fast pyrolysis bio-oils derived from pine and oil mallee.

derived from soy bean is mainly formed by ve methyl esters of fatty acids (hexadecanoic acid (palmitic acid), linoleic acid, oleoic acid, 7 Octadecedienoic acid and octadecanoic acid). The TG and DTG curves for the bio-diesel and bio-oils used in this study are plotted in Fig. 3. The thermal behavior of bio-oil and biodiesel differed substantially. While the peak of the DTG curve for the bio-oils is obtained at temperatures close to 100 C, the peak for biodiesel is located at a temperature close to 260 C. The boiling point of bio-diesel is very close to that of family D in the bio-oils (monosugars). Bio-oil is rich in compounds within the boiling range of gasoline. These differences in evaporation behavior are a good indication of major differences in the composition of both bio-fuels. 3.3. Solubility of bio-oil in bio-diesel Several blends with different bio-diesel/bio-oil ratios (1, 1.5, 4 and 9) were prepared. These ratios correspond to bio-oil concentrations of 50, 40, 20 and 10 mass % respectively. The bio-diesel/bio-oil ratio used to prepare the blends vs. the ratio of bio-diesel rich/bio-oil rich phases obtained after separating the resulting phases are shown in Fig. 4a. Fig. 4 also contains the results obtained when bio-dieselethyl acetate solutions were used to extract the bio-oil. The results obtained with bio-dieselethyl acetate solutions will be discussed in the Section 3.6. In Fig. 4a the slope of the straight line (K) resulting from a plot of the bio-diesel/bio-oil ratio against the ratio bio-diesel rich/bio-oil rich phases has been used as an indicator of the solubility of mallee and pine bio-oil in bio-diesel. As more bio-oil is extracted by the biodiesel, the mass of the bio-diesel rich phase collected will also increase and the amount of the bio-oil rich phase will decrease. Thus, as the fraction of bio-oil extracted by the bio-diesel increases, the values of K will also increase. For both fast pyrolysis oils the values of K obtained was around 1.2. This value is comparable to the one obtained for polar oils rich in sugars that were collected after the evaporation of the aqueous phase from the Auger pyrolysis of pine pellets (1.0 < K < 1.4), but was considerably lower than values of K obtained for decanted oils (2.1 < K < 3.1 ) produced in the same system [8]. Thus, it can be stated that the fast pyrolysis oils obtained from mallee and pine are much less soluble in bio-diesel than the decanted oils resulting from
Fig. 2. GC/MS of bio-diesel.

Table 4 Values of the mass fraction of volatile compounds and solid residue used to t the DTG curves. Temperature (C) Oil mallee Pine chip Mass fraction of volatiles zjo (mass fraction) A 0.11 0.14 B 0.35 0.29 C 0.20 0.20 D 0.12 0.12 E 0.13 0.19 F 0.09 0.07 Rs (mass %) 13.8 10.3

Fig. 3. TG and DTG curves of bio-oils and bio-diesel used.

300

M. Garcia-Perez et al. / Fuel Processing Technology 91 (2010) 296305

Auger pyrolysis [8]. These decanted oils are known to be very rich in lignin derived compounds (phenols and lignin derived oligomers) which show low polarity. The concentration of bio-oil derived compounds in the bio-diesel rich phase (cbio-oil) can be determined from the values of K and the mass of bio-oil and bio-diesel used to prepare the blend (mbio-diesel and mbio-oil). For this purpose the following equation was used: cbiooil = K 1 = K mbiodiesel = mbiooil + 1: 1

This equation was derived supposing that the solubility of biodiesel compounds in bio-oil is very limited. GC/MS analyses carried out in the bio-oil rich phase conrmed the validity of this assumption. Fig. 4(b) shows the concentration of bio-oil compounds in the biodiesel rich phase estimated using Eq. (1). Bio-oil concentrations in bio-diesel as high as 10 mass % were obtained when equal masses of bio-oil and bio-diesel were blended (biodiesel/bio-oil ratio = 1). The mass % of bio-oil extracted by the bio-diesel was higher than 15 mass % for a biodiesel/bio-oil ratio equal to 9 (see Fig. 4c). 3.4. Analysis of resulting bio-oil rich phases (extracted bio-oils) Fig. 5 shows the DTG curve of a bio-oil rich phase multiplied by the yield corresponding to this phase. Fig. 5 also presents the DTG curve corresponding to the fresh bio-oil. The deconvolution of this DTG curve was discussed in a previous section (see Fig. 1). Comparing the DTG curves of fresh oils with those of extracted oils suggest that the families B, C and E are those solubilising the most in the bio-diesel (phenols, furans and carboxylic acids). Family D and F (the sugars) are less soluble in the bio-diesel. The values of Ej, nj, Aj used to t the curves of bio-oil extracted were very similar to those used to t the curve of un-extracted oil. The values of zjo used to t each of these curves are shown in Table 5, and differ from those shown in Table 4. The values of zjo shown in Tables 4 and 5 can be used to estimate the concentration of each of the families or fractions in the oil. In order to quantify the content of each of these fractions it is necessary to suppose that the solid residues obtained were formed due to the contribution of families E and F. The contribution of each of these fractions was considered proportional to the value of zjo assigned to each group. The content of family A, B, C and D was calculated by multiplying the values of z jo corresponding to each of these fractions per the total content of volatiles measured by thermogravimetry (100 Rs). The contents of the families E and F were determined in the same way, but their contribution to the formation of solid residues was also taken into account. The content of each group in the fresh oils, and after extraction, are shown in Table 6. The values obtained for each of the fractions after extraction was

Fig. 4. Preparation of biomass bio-diesel blends.

Fig. 5. DTG curves of bio-oils after extraction.

M. Garcia-Perez et al. / Fuel Processing Technology 91 (2010) 296305 Table 5 Values of the mass fraction of volatile compounds and solid residue used to t the DTG curves (ratio bio-diesel/bio-oil = 9). Temperature (C) Mass fraction of volatiles zjo (mass fraction) Rs (mass %) B C D E F 0.2 0.2 0.15 0.11 0.13 0.11 0.12 0.11 0.08 (13.6) 0.09 (14.1)

301

Oil mallee after extraction 0.12 Pine chip after extraction 0.16

Table 6 Content of each of the groups before and after extraction with bio-diesel. Temperature (C) Composition (mass %) A Oil mallee (before extraction) Pine chip (before extraction) Oil mallee (after extraction) (ratio bio-oil/biodiesel 1/9) Pine chip (after extraction) (ratio bio-oil/biodiesel 1/9) 9.5 12.6 10.3 14.3 B 30.2 25.1 17.2 17.9 C 17.2 17.9 12.9 9.8 D 10.4 10.8 11.2 9.8 E 19.3 24.6 18.6 15.5 F 13.4 9.1 12.4 12.7

multiplied by the actual mass % of bio-oils extracted so that the results could be more easily compared. The results shown in Table 6 suggest that the compounds in family B, C and E are the ones more readily extracted by the bio-diesel. Between 15 and 20 mass % of the oil was extracted by the bio-diesel. Of this amount between 9 and 13 mass % corresponds to compounds from family B. These compounds are likely to be carboxylic acids (acetic and formic acids). Between 5 and 8 mass % correspond to

compounds associated to family C which is mainly formed by monolignols and furanic compounds. Between 1 and 9 mass % corresponds to lignin derived oligomers which are usually associated with family E. It must be point out that although the tting of DTG curves is an excellent way to get an overall understanding of the chemical composition of bio-oils, the precision associated with this method is limited (errors can be as high as 10%). There are some inconsistencies in Table 6 that need to be highlighted. For example, the content of family A and F for pine derived oils seems to be higher after extraction. This clear incongruence can be explained by the limitations of the experimental and analytical tting method used. Some of the fuel properties of the bio-diesel rich phase were also studied. The change in elemental composition and in gross caloric value for the bio-diesel rich phase is shown in Fig. 6. The content of carbon changes very little from almost 77 mass % for bio-diesel to 75 mass % for the bio-diesel rich phases prepared with bio-oil/biodiesel ratios equal 1. While the hydrogen content dropped from around 12.3 to approximately 11, the content of oxygen increased from 11 to 13 mass %. These changes in elemental composition have a limited impact on the caloric value of the resulting blend which decreased from 40 MJ/kg to around 39 MJ/kg. Some of the fuel properties of the bio-diesel rich phase resulting from a bio-diesel/bio-oil ratio equal to 4 (20 mass % of bio-oil) are presented in Table 7. The bio-diesel rich phase contains between 3 and 4 mass % of biooil. Although, in general most of the fuel properties tested remain within specications and the oxidation stability improves, other properties deteriorated. For example, the carbon residue increases from 0.01 to more than 0.10 mass %. The maximum value permissible by Australian Standards is 0.05 mass %. This increase is mainly due to

Fig. 6. Changes in the elemental composition and caloric value at different bio-oil concentrations.

302

M. Garcia-Perez et al. / Fuel Processing Technology 91 (2010) 296305

Table 7 Fuel properties of commercial biodiesel and the blends resulting their blending with pine and mallee bio-oils (20 mass %). Parameter Sulfated ash Carbon residue (100% distillation) Copper strip corrosion Oxidation stability Metals group 1 Sodium Potasium Metals group II Calcium Magnesium Acid value Water and sediments Cloud point Cold lter plugging (CFPP) Derived cetane number Ignition delay test Temperature Standard ASTM D874 ASTM D4530 ASTM D130 EN14112 Units mass % mass % hr Commercial bio-diesel 0.000 0.01 1a 10.2 Commercial bio-diesel/ pine bio-oil blends 0.002 0.13 1b 28.1 Commercial bio-diesel/ mallee bio-oil blends 0.002 0.10 1a 26.0 Australian standard 0.02 max 0.05 max 3 max 3 min

EN14108 EN14109

mg/kg mg/kg

1.0 1.0

2.2 1.4

1.0 1.4

5 max

EN14538 ASTM D664 ASTMD2709 IP 309 ASTMD890

mg/kg mg/kg mg KOH/g vol.% C C DNC ms C

1.0 1.0 0.43 0 3 14 57.17 3.540 559.3

1.0 1.0 0.97 0.08 4 14 55.51 3.656 559.3

1.0 1.0 1.25 0.01 3 13 56.07 3.616 559.9

5 max 0.5 max 0.05 max

47 min

the solubilisation of pyrolytic lignin in the bio-diesel. The acid number increased from 0.43 to more than 0.95 mg KOH/g. This increase is due to the solubilisation of carboxylic acids in the bio-diesel phase. The water and sediments increased to 0.08 vol.% exceeding the maximum permissible value which is 0.05 vol.%. This increase in the content of water and sediments is a good indication that a small amount of water solubilises in the bio-diesel matrix. It is noteworthy to point out that the oxidation stability of bio-diesel increased due to the extraction of phenolic compounds from the bio-oils. These compounds are known to be excellent antioxidants. Several of the properties exceeding specications are due to the solubilisation of lignin derived oligomers in the bio-diesel matrix. The solubilisation of these oligomers warrants further investigation. Thus, the next section is devoted to quantifying the content of lignin derived oligomers in bio-diesel.

3.5. Solubility of lignin derived oligomers from bio-oil in bio-diesel Although there are several methods to quantify the content of lignin derived oligomers in bio-oils, to our knowledge there is no method available to quantify the content of these compounds in bio-diesel. Figs. 7 and 8 show the synchronous spectra for the bio-diesel rich phase obtained at different bio-oil/bio-diesel ratios. The spectra were recorded

using energy differences of 1400 and 2800 cm 1. The spectra recorded correspond to the bio-oil fraction solubilised in the bio-diesel. In order to quantify the yield of lignin derived oligomers it is necessary to obtain the spectra corresponding to the water insolubleCH2Cl2 soluble fractions (lignin derived oligomers) of the bio-oils derived from mallee and pine. Fig. 9a) shows the spectrum corresponding to the lignin derived oligomers from pine derived oil and the one for the bio-diesel rich phase. In order to quantify the content of lignin derived oligomers in the bio-diesel rich phase it is necessary to nd a multiplication factor to t the tail of the UVuorescence spectra obtained for the lignin derived oligomers with the one obtained for the bio-diesel rich phase. Fig. 9b) shows how the tting was made. The factor used to achieve the best tting is then considered as a good estimation of the content of lignin derived oligomers in the bio-diesel rich phase. Fig. 10 shows the evolution of the content of lignin derived oligomers as a function of the ratio of bio-oil/bio-diesel used to prepare the blends. Fig. 10 also shows the total content of lignin derived oligomer compounds that solubilize in the bio-diesel. Concentrations as high as 3.5 mass % of oligomers in the bio-diesel can be obtained when the blends are prepared with bio-oil/bio-diesel ratios equal to 1. The lignin derived oligomers are known to be important precursors for the formation of carbonaceous residues. Thus, they must be carefully controlled, either during the production

Fig. 7. Constant energy ( 1400 cm 1 and 2800 cm 1) synchronous spectra of biodiesel rich phase obtained from blends with pine bio-oil. The uorescence intensity was recorded on the basis of the same bio-oil concentration 4 ppm.

M. Garcia-Perez et al. / Fuel Processing Technology 91 (2010) 296305

303

Fig. 8. Constant energy ( 1400 cm 1 and 2800 cm 1) synchronous spectra of bio-diesel rich phase obtained from blends with mallee derived bio-oil. The uorescence intensity recorded on the basis of the same bio-oil concentration 4 ppm.

Fig. 9. Constant energy ( 1400 cm 1) synchronous spectra of bio-diesel rich phase and lignin derived oligomers from pine bio-oils. The uorescence intensity is recorded on the basis of the same bio-oil concentration, 4 ppm.

of the bio-oil/bio-diesel blends, or by some post-treatment methods, for example distillation of the resulting blends. 3.6. Use of ethyl acetate to enhance the solubility of bio-oils in bio-diesel The solubility of fast pyrolysis oils in bio-diesel is relatively low. Thus, it was decided to enhance bio-oil solubility by adding a volatile

Fig. 10. Content of lignin derived oligomers in the Biodiesel rich phase.

solvent to the bio-diesel. The main hypothesis behind the proposed concept is that after removing the solvent, the resulting bio-diesel rich phase will be richer in bio-oil derived molecules. The solvent chosen was ethyl acetate because this solvent has been extensively used to extract the bio-oils [22,29]. Solutions containing 50 mass % of ethyl acetate and 50 mass % of bio-diesel were prepared. These solutions were used to create blends containing 10, 20, 40 and 50 mass % of biooils. The procedure employed to obtain the bio-diesel rich phase follows the one described in Section 3.3when bio-diesel alone was used as extracting media. Fig. 4 shows, the ratio of bio-diesel/bio-oil rich phase obtained, the concentration of bio-oil in the bio-diesel, and the mass % of bio-oil extracted for different bio-diesel/bio-oil ratios. As explained previously, the slope of the resulting straight line (K) can be used as an indicator of the solubility of bio-oil in bio-diesel. Using ethyl acetate improves the value of K from around 1.2 to between 1.4 and 2.2. This increase in the value of K is a good indication of the increase in the extraction capacity of the resulting bio-diesel ethyl acetate solution. Between 20 and 50 mass % of the bio-oil can be extracted if a biodiesel/ethyl acetate solution is used to extract the bio-oil compounds. Concentrations as high as 25 mass % of bio-oil in the bio-diesel rich phase can be obtained when using ethyl acetate. Fig. 11 shows the DTG curve of the bio-oil remaining after extraction with the biodiesel/ethyl acetate blend, multiplied by the yield of bio-oil remaining after extraction. In the case of the bio-oil from pine all bio-oil fractions

304

M. Garcia-Perez et al. / Fuel Processing Technology 91 (2010) 296305

Fig. 11. DTG curve of the bio-oils left after the extraction with bio-diesel/ethyl acetate blends.

were extracted by the bio-diesel rich phase. In the case of the bio-oil from mallee all the fractions were extracted except fraction D which is usually associated with the sugars. The ethyl acetate present in the bio-diesel rich phase was removed by rotary evaporation. It is noteworthy that after the removal of the ethyl acetate an important fraction of the bio-oil solubilised in biodiesel precipitated out. The precipitated tarry phase is likely to be formed by the most polar fractions solubilised in the bio-dieselethyl acetate phase. The elemental composition (content of carbon, hydrogen and oxygen) as well as the gross caloric value of the resulting bio-diesel rich phase are shown in Fig. 12. The more drastic change in the composition of the bio-diesel rich phase can be

considered as a good indicator that the concentration of bio-oil in the resulting bio-diesel phase is higher. The content of carbon dropped from 77 to approximately 72 mass % for both bio-oils (5 mass %). When the bio-diesel was used alone to extract the bio-oil the reduction in the content of carbon was of only 2 mass %). The content of hydrogen decreased by approximately 1 mass % compared with a reduction of only 0.4 mass % when the extraction was carried out in the absence of ethyl acetate. The oxygen content increased by more than 6 mass % with the addition of ethyl acetate to the bio-diesel, compared with just 2 mass % with the bio-diesel alone. The changes in elemental composition had a clear impact on the caloric value of these oils, with a reduction of almost 4 MJ/kg.

Fig. 12. Changes in the elemental composition and caloric values of blends resulting after the removal of ethyl acetate.

M. Garcia-Perez et al. / Fuel Processing Technology 91 (2010) 296305

305

The experimental results obtained in this paper conrm that the addition of volatile and polar compounds to the bio-diesel could be a viable approach to enhance the solubility of fast pyrolysis oils in bio-diesel. The study of the performance of other volatile organic solvents to enhance the solubility of bio-oil in bio-diesel warrants further investigation. 4. Conclusions In this study, blends of bio-oil/bio-diesel were prepared and the fuel properties of the resulting bio-diesel rich phase analysed. Extracting some of bio-oil compounds with Bio-diesel could be an alternative to utilize the phenolic fraction from bio-oils as fuel additives. This study proves that the oxidation stability of resulting bio-diesel rich phase improves while other fuel properties such as solid residue and the acid number deteriorate. The solubilisation of lignin derived oligomers in the bio-diesel is the main cause of the increase in solid residue observed. Concentrations of lignin derived oligomers as high as 3.5 mass % were quantied by a novel UV-uorescence technique proposed herein. The solubility of bio-oil in bio-diesel was enhanced by adding ethyl acetate to the bio-diesel. Our experimental results also show that the bio-diesel rich phase, containing bio-oil derived molecules, will require additional treatment such as distillation, to comply with the Australian bio-diesel Standards. Additional studies are needed to develop viable technologies to utilize the bio-oil phase insoluble in bio-diesel. This phase is known to be rich in oxygenated compounds derived from cellulose and hemicelluloses. Acknowledgements The authors are very thankful to the Australian Research Council (ARC Discovery Grant DP0556098) for the nancial support of this project. This project is also partially supported by the Commonwealth of Australia under the International Science Linkages program. The authors also thank Mr. Rick Giles of the Revegetation Unit, Department of Conservation and Lands, Western Australia, for kindly providing the biomass used. References
[1] C. Briens, J. Piskorz, F. Berruti, Biomass valorization for fuel and chemicals production a review, International Journal of Chemical Reactor Engineering 6 (2008) 1. [2] D.S. Scott, P. Majerski, J. Piskorz, D. Radlein, A second look at fast pyrolysis of biomassthe RTI process, Journal of Analytical and Applied Pyrolysis 51 (1999) 23. [3] J.P. Diebold, A.V. Bridgwater, Overview of fast pyrolysis of biomass for the production of liquid fuels, in: A. Bridgwater, et al., (Eds.), Fast Pyrolysis of Biomass: A Handbook, CPL Press, Berks, U.K, 1999, pp. 164188. [4] D. Meier, O. Faix, State of the art of applied fast pyrolysis of lignocellulosic materials review, Bioresource Technology 68 (1999) 71. [5] A.V. Bridgwater, D. Meier, D. Radlein, An overview of fast pyrolysis of biomass, Organic Geochemistry 30 (1999) 1479.

[6] A.V. Bridgwater, S. Czernik, J. Piskorz, An overview of fast pyrolysis, in: A.V. Bridgwater (Ed.), Progress in thermochemical biomass conversion. IEA Bioenergy, Black Sciences, 2001, pp. 977997. [7] S. Czernik, A.V. Bridgwater, Overview of applications of biomass fast pyrolysis oil, Energy & Fuels 18 (20) (2004) 590. [8] M. Garcia-Perez, T.T. Adams, J.W. Goodrum, D.P. Geller, K.C. Das, Production and fuel properties of pine chip bio-oil/biodiesel blends, Energy & Fuels 21 (4) (2007) 2363. [9] D. Chiaramonti, A. Oasmaa, Y. Solantausta, Power generation using fast pyrolysis liquid from biomass, Renewable & Sustainable Energy Reviews 11 (2007) 10561086. [10] M. Ikura, M. Stanciulescu, E. Hogan, Emulsication of pyrolysis derived bio-oil in diesel fuel, Biomass & Bioenergy 24 (2003) 221232. [11] C.-Y. Peng, C.-H. Lan, Y.-T. Dai, Specication and quantication of vapour phases in soy biodiesel and waste cooking oil bio-diesel, Chemosphere 65 (2006) 20542062. [12] M. Graboski, R.L. McCormick, Combusting of fast and vegetable oil derived fuels in diesel engines, Prog. Energy Combusti. Sci. 24 (1998) 125164. [13] T. Issariyakul, M.G. Kulkarno, L.C. Meher, A.K. Dalai, N.N. Bakhshi, Biodiesel production from mixtures of canola oil and used cooking oil, Chemical Engineering Journal 140 (2008) 7785. [14] M.G. Kulkarni, A.K. Dalai, N.N. Bakhshi, Transesterication of canola oil in mixed methanol/ethanol system and use of esters as lubricity additive, Bioresource Technology 98 (2007) 20272033. [15] J.H. Van Gerpen, S. Soylu, E.T. Mustafa, Evaluation of the lubricity of soybean oilbased additive in diesel fuel, ASAE, 1999, Paper No 996134. [16] D.P. Geller, J.W. Goodrum, Effects of specic fatty acid methyl esters on diesel fuel lubricity, Fuel 83 (2004) 23512356. [17] S.-K. Loh, S.-M. Chew, Y.-M. Choo, Oxidative stability and storage behavior of fatty acid methyl esters derived from used palm oil, JAOS, Journal of American Oil Chemists' Society 83 (11) (2006) 947. [18] R.O. Dunn, Effect of antioxidants on the oxidative stability of methyl soyate (biodiesel), Fuel Processing Technology 86 (2005) 10711085. [19] L.C. Meher, D.V. Sagar, S.N. Naik, Technical aspects of bio-diesel production by transesterication a review, Renewable & Sustainable Energy Reviews 10 (2006) 248268. [20] J.M. Marchetti, U.V. Miguel, A.F. Errazu, Possible methods for bio-diesel production, Renewable & Sustainable Energy Reviews 11 (2007) 13001311. [21] M. Garcia-Perez, X.-S. Wang, J. Shen, M.J. Rhodes, F. Tian, W.-J. Lee, H. Wu, C.-Z. Li, Fast pyrolysis of oil mallee woody biomass: effect of temperature on the yield and quality of pyrolysis products, Industrial & Engineering Chemistry Research 47 (2008) 18461854. [22] M. Garcia-Perez, A. Chaala, H. Pakdel, D. Kretschmer, C. Roy, Characterization of bio-oils in chemical families, Biomass & Bioenergy 31 (2007) 222242. [23] M. Garcia-Perez, S. Wang, J. Shen, M. Rhodes, W.-J. Lee, C.-Z. Li, Effect of temperature on the formation of lignin-derived oligomers during the fast pyrolysis of mallee woody biomass, Energy & Fuels 22 (2008) 20222032. [24] S.R.A. Kersten, X. Wang, W. Prins, W.P.M. van Swwaij, Biomass pyrolysis in a uidized bed reactor. Part 1. Literature review and model simulations, Industrial & Engineering Chemistry Research 44 (2005) 87738785. [25] X. Wang, S.R.A. Kersten, W. Prins, W.P.M. van Swaaij, Biomass pyrolysis in a uidized bed reactor. Part 2: Experimental validation of model results, Industrial & Engineering Chemistry Research 44 (2005) 8786. [26] C. Di Blasi, G. Signorelli, C. Di Russo, G. Rea, Product distribution from pyrolysis of wood and agricultural residues, Industrial & Engineering Chemistry Research 38 (1999) 2216. [27] C. Di Blasi, C. Branca, A. Galgano, Effects of diammonium phosphate on the yield and composition of products from wood pyrolysis, Industrial & Engineering Chemistry Research 46 (2007) 430. [28] C. Branca, C. Di Blasi, Multistep mechanism for the devolatilization of biomass fast pyrolysis oils, Industrial & Engineering Chemistry Research 45 (2006) 58915899. [29] K. Sipila, E. Kuoppale, L. Fagernas, A. Oasmaa, Characterization of biomass based ash pyrolysis oils, Biomass & Bioenergy 14 (2) (1998) 103113.