- Email: [email protected]
C) catalyst
Algal Research 24 (2017) 188–198
Contents lists available at ScienceDirect
Algal Research journal homepage: www.elsevier.com/locate/algal
Catalytic upgrading of fractionated microalgae bio-oil (Nannochloropsis oculata) using a noble metal (Pd/C) catalyst Hyungseok Nam a,b,⁎, Changkyu Kim c, Sergio C. Capareda a, Sushil Adhikari b a b c
Bio-Energy Test and Analysis Laboratory (BETA Lab), Department of Biological and Agricultural Engineering, Texas A&M University, College Station, TX 77843, USA Department of Biosystems Engineering, Auburn University, Auburn, AL 36849, USA Texas A&M Energy Institute, Texas A&M University, College Station, TX 77843, USA
a r t i c l e
i n f o
Article history: Received 24 October 2016 Received in revised form 19 February 2017 Accepted 24 March 2017 Available online xxxx Keywords: Distillation bio-oil Hydrotreatment Catalytic upgrading Microalgae Palladium
a b s t r a c t Pyrolytic bio-oil was chemically upgraded after physically distilled upgrades to meet the petroleum transportation fuel substitute. A Pd/C catalyst was used to upgrade the microalgae pyrolytic bio-oil to determine the effect of different distillation fractions and catalytic upgrading conditions on the yields and properties. The middle distillation fraction (F2) was upgraded under various temperature (130 to 250 °C) and pressure (4.1 to 8.3 MPa) conditions based on response surface methodology (RSM). The light distillation fraction (F1) and raw bio-oil were also catalytically upgraded for the comparison. The distillation step prior to catalytic upgrading led to a better quality of upgraded bio-oil compared to the direct bio-oil upgrades. Both the oxygen and hydrogen contents of light and middle fraction upgrades were improved, while the upgraded raw bio-oil showed limited improvement. The other properties of HHV and TAN with the middle fraction upgrades were improved to 42.9 MJ/kg and 1.09 mg KOH/g, respectively, at the severe condition as most of the ketones in upgrades were removed. Also, paraffin and aromatic chemical groups were significantly produced at the expense of the olefin groups through hydrogenation and hydrodeoxygenation. Thus, the catalytic upgrading after a distillation stage enhanced the quality of biofuel that can be a petroleum fuels substitute or additives. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Microalgae, as a source of third generation biofuels, have obtained much attention as an alternative energy source because they overcome the drawbacks of first and second generation biofuels. With simple conditions required for growth, light, sugar, CO2, N, P, and K, photosynthetic microalgae microorganisms grow rapidly even on non-arable land. The large amounts of lipids, proteins and carbohydrates in microalgae can be processed into biofuel, medicine, and food [1]. Many microalgae biofuels were studied in the form of biodiesel from lipid transesterification [2], bio-oil from the thermal conversion process of pyrolysis [3–5] and hydrothermal liquefaction (HTL) [6]. The transesterification process for biodiesel produced a large amount of residue, while the bio-oils showed high viscosity, high acidity, and high nitrogen and sulfur contents [7,8].
⁎ Corresponding author at: Bio-Energy Test and Analysis Laboratory (BETA Lab), Department of Biological and Agricultural Engineering, Texas A&M University, College Station, TX 77843, USA. E-mail addresses: [email protected], [email protected] (H. Nam), [email protected] (S.C. Capareda).
http://dx.doi.org/10.1016/j.algal.2017.03.021 2211-9264/© 2017 Elsevier B.V. All rights reserved.
An upgrading of thermally converted bio-oils is required for them to be used as potential fuel substitutes or additives. A physical method of distillation and a chemical method of catalytic upgrading were used to upgrade pyrolytic bio-oil by reducing the O/C ratio and increasing the H/C ratio, which also enhances the other fuel properties of viscosity, acid number, heating value, and water content. Distillation as a physical separation process upgrades bio-oils based on the relative volatility of chemicals at various temperatures. Several techniques were used to upgrade the microalgae bio-oil with direct, fractional, steam, vacuum, and molecular distillations. Bio-oil (Spirulina sp. and Tetraselmis sp.) from hydrothermal liquefaction (HTL) was fractionated using a vacuum distillation at temperature conditions of 300 °C and 350 °C [9]. The maximum yields of an HTL bio-oil and its distillates were 58% and 73%, respectively, whereas each HHV (higher heating value) were 32 MJ/kg and 40 MJ/kg with a 97% deoxygenation of bio-oil. In our previous study [8], the bio-oil of Nannochloropsis oculata microalgae was physically upgraded using fractional and vacuum distillation setups. Both the techniques showed similar distillate properties, while a higher product yield was obtained from fractional distillation when better separation was made with vacuum distillation. The lower operating temperature of vacuum distillation prevented polymerization in the heavy fraction. The HHV and the lowest total acid number (TAN) were
H. Nam et al. / Algal Research 24 (2017) 188–198
41.2 MJ/kg and 0.1 mg KOH/g obtained through deoxygenation and dehydration. Still, the fuel needed a further bio-oil upgrading step to be used as a petroleum fuel substitute due to the existence of oxygen and nitrogen contents, and high acidity in a certain bio-oil fraction. A chemical upgrading of catalytic hydrotreatment is performed by removing the oxygen, nitrogen, and sulfur contents from bio-oil, which is also used in the conventional petroleum refining process. Catalytic upgrading generally requires the reactions between hydrogen gas and bio-oil at high pressure and temperature with enough reaction time. Various alumina, silicate, and carbon based catalysts for bio-oil upgrading were evaluated from many studies. A catalyst screen study by Bai et al. [7] used 12 different catalysts for HTL bio-oil (Chlorella pyrenoidosa) upgrading. The highest product yield (77%) and a HHV of 45.3 MJ/kg for upgraded bio-oil were obtained with a mixed catalyst of Ru/C and Raney-Ni, and the lowest O/C (0.01) and N/C (0.017) were achieved with Ru/C and Raney-Ni, respectively. An HTL bio-oil (Nannochloropsis sp.) upgrading in supercritical water also conducted by Duan and Savage [10] to determine the effect of Pt/C, H2 pressure (3.5 MPa), and the pH of bio-oil on the bio-oil upgrades at 400 °C. The upgraded bio-oil showed a substantially decreased acid number, 25 mg KOH/g, with a decrease in O/C (0.041) and H/C (1.64) contents, which led to an increase the HHV (43 MJ/kg). They also conducted a similar study using Pd/C at the same temperature and pressure conditions to find the effect of reaction time (1–8 h) and catalyst loading (5–80%) on the hydrotreatment products [11]. The upgraded bio-oil properties with Pd/C upgrading showed 43.5 MJ/kg for HHV, 0.028 for O/C, and 1.79 H/C, which had a somewhat better result compared to the Pt/C upgrading. Barreir et al. [12] conducted HTL bio-oil upgrading (Scenedesmus Almeriensis and Nannochlopsis Gaditana) using Pt/Al2O3 and HZSM-5 at 400–450 °C with 4–8 MPa. The addition of water for the catalytic upgrading at a lower H2 pressure produced more oxygen chemicals compared to the upgrading under only H2 pressure. Also, Lin et al. [13,14] produced green diesel from hydrodeoxygenation of extracted microalgae oil using various catalysts (NiMo, Pt and Rh) in a continuous flow microreactor. The optimum yield of hydrocarbon ranging from C13 to C20 was 56% at operating conditions of 500 psi, 360 °C, and 1 s contact time. Maguyon [15] studied the pyrolysis bio-oil upgrading (Nannochloropsis oculata) with an HZSM-5 catalyst at a constant pressure (4.1 MPa) to determine the effect of temperature (200–300 °C) and reaction time (1–4 h). Mass balance was achieved based on the amount of upgraded bio-oil, coke, tar, and gas. The upgraded bio-oil showed a basic pH level (7.7–8.9) and the HHV ranged from 37.5–40 MJ/kg with a 1.46 H/C and 0.032 O/C. Even though many upgrading methods with various catalysts were studied based on the various operating conditions, few catalytic upgrading studies of distilled bio-oil have been reported. As a continuing distillation study of Nam et al. [8], the distilled bio-oil of light and middle fractions were used for chemical catalytic upgrading. The current study can expand the understanding of the effectiveness of the catalytically upgraded bio-oil from distilled bio-oil. The objectives of the study were to, 1) assess the effects of distilled bio-oils through Pd/C catalytic upgrading on the properties of bio-oil, 2) evaluate the performance of the catalyst at various conditions of temperature and pressure on the deoxygenation, hydrogen consumption, and turnover frequency, and. 3) to investigate the catalyst deactivation after catalytic upgrading. 2. Experimental 2.1. Sample preparation An algae species, Nannochloropsis oculata, from the Texas A&M AgriLife pond in Pecos, Texas, was obtained for pyrolysis and bio-oil distillation. The wet algae sample was first dried and processed through a batch type pyrolysis reactor, which was also used in other studies
189
[3,16]. The mass recovery and HHV (higher heating value) of bio-oil from pyrolysis was 20% and 38.6 MJ/kg. The bio-oil was then fractionated using a fractional distillation setup as shown in Fig. 1(b). The light (T b 120 °C) and middle (120 °C b T b 200 °C) fractions after fractional distillation were used in this study. The distillation mass recovery was 23% for the light fraction (fraction 1 or F1), 52% for the middle fraction (fraction 2 or F2) and 19% for the heavy fraction (fraction 3 or Fe), while the HHVs were 40.0 and 41.2 MJ/kg, respectively. A study by Nam et al. [8] showed a detailed fractional and vacuum distillation study for the fractionated microalgae bio-oils. 2.2. Experimental setup and procedure A batch type reactor (Micro Robinson-Mahoney, Autoclave Engineers) was used for the current catalytic upgrading study. The volume of the batch was 50 mL, which can handle a maximum pressure to 5000 psi at 343 °C or a maximum temperature to 538 °C as shown in Fig. 1(c). The temperature of the reactor was controlled by an electric heating band with a PID temperature controller. Two K type thermocouples were used; one for the temperature inside the reactor, and the other for the temperature of the electric heater. An internal stirrer was placed inside the batch to help in uniform reactions, and was set at around 300 rpm during the experiment. Each pressurized catalytic experiment was conducted with 7 g of bio-oil with 5% of 5%Pd/C catalyst (Sigma Aldrich, USA). Before the start of the experiment, hydrogen gas was used to flush as much air as possible from inside the reactor. Then the reactor was pressurized with 4.1, 6.2, or 8.3 MPa of hydrogen gas at room temperature. Once the experiment was ready to run, the heater jacket heated the inside reactor to 130, 190, or 250 °C with a stirring of 400 rpm. A constant reaction time was set at 4 h based on Duan and Savage [11], who state that the properties of upgraded bio-oil were not improved much after 4 hour reaction time. At the end of the 4 hour reaction time, the system was shut down and the temperature reduced to room temperature with the help of a water jacket. First, the pressure at room temperature after each experiment was noted to compute the amount of hydrogen gas used. The gas samples were then collected for analysis. Last, the upgraded bio-oil was collected after filtering out the catalyst using a pre-weighted Buchner glass filter. Then the remaining catalyst and bio-oil in the reactor were washed out thoroughly using acetone as solvent. Both the Buchner glass filter and washed bio-oil were then dried in a 60 °C oven to completely remove the acetone solvent. The weight of the dried catalyst in the glass filter after solvent wash was considered as coke. The solvent washed bio-oil was indicated as tar. The complete mass balance was made with upgraded bio-oil, coke, tar, and produced gas + loss. 2.3. Analytical methods The pyrolytic bio-oil and upgraded fractions were analyzed using the following analytical methods. A higher heating value (HHV) was obtained using a PARR bomb calorimeter 6200 in accordance with ASTM D 711. The ultimate analysis (for C, H, N and S) was obtained from a Vario MICRO Elemental analyzer based on the standards of ASTM D5373. The liquid products after upgrading were analyzed according to ASTM E203 for water content and ASTM D974 for total acid number (TAN). The pH of the bio-oil was measured using a digital pH meter. A Shimadzu IRAffinity-1 FTIR spectrophotometer was used to understand the changes of functional groups of the liquid products. The chemical compositions were analyzed using a GC-MS (Shimadzu QP2010Plus, ZB5MS 30 m × 0.25 mm diameter × 0.25 μm thick). The same temperature program of a previous distillation study was used [8]. The gas composition was analyzed using an SRI multiple gas chromatograph (GC) equipped with a TCD and a GC column containing a 6′ molecular sieve and a 6′ silica gel column. The size characterization of raw and used Pd/C catalyst was analyzed using a Quantachrome Instrument
190
H. Nam et al. / Algal Research 24 (2017) 188–198
Fig. 1. Schematic diagrams of (a) a pyrolysis batch reactor for bio-oil production [3], (b) a fractional distillation set-up [8], (c) and a pressure bio-oil upgrading catalytic reactor [15].
(Autosorb-iQ model). First, the samples were degassed for 3 h at 300 °C. The adsorption and desorption isotherms were then measured from the different relative pressures (0.0–1.0) of N2 adsorption under a −196 °C chamber. A surface area (m2/g) was also determined based on a BET model. The reducibility of catalysts was evaluated with temperature programmed reduction (TPR) under H2/Ar (10/90) gas. Initially the sample was flushed and dried at 120 °C for 30 min under nitrogen and cooled to 40 °C. Then the temperature was programmed to 800 °C at 10 °C/min. Thermogravimetric analysis (TGA) was performed using a PerkinElmer Pyris 1 TGA. About 5–6 mg of new, spent and CO2 regenerated catalysts were loaded with a 20 μL alumina crucible and heated from 25 °C to 900 °C at 10 °C/min under 20 mL/min of nitrogen. Then the averaged DTG curves from duplicates were obtained to understand the degree of volatiles over the temperature. 2.4. Experimental design and data analysis A CCD (central composite design) as a part of the RSM (response surface methodology) was used for the current study to understand the statistical correlations of temperature and pressure on the product yields and properties. The total number of runs in the CCD was 12 for catalytic upgrading of the middle fractions (denoted as F2) of bio-oil at conditions of temperature and pressure shown in Table 1. Three more runs were made for comparison between the direct bio-oil upgrading and the light fraction (denoted as F1) upgrading. F2 was collected from a fractional distillation ranging from 120 to 200 °C when F1 was also from distillation below 120 °C. The operating conditions in accordance with the CCD of the RSM were decided after preliminary
For comparison
Run #
Type of bio-oil
Temp. (°C)
Pressure (MPa)
1 2 3 4 5⁎ 6⁎ 7⁎ 8⁎
F2 F2 F2 F2 F2 F2 F2 F2 F2 F2 F2 F2 Bio-oil F1 F1
130 130 130 190 190 190 190 190 190 250 250 250 260 130 190
4.1 6.2 8.3 4.1 6.2 6.2 6.2 6.2 8.3 4.1 6.2 8.3 4.1 6.2 6.2
9 10 11 12 13 14 15
⁎ Center points of the CCD (central composite design).
Product yield ð%Þ ¼ Mass of product ðgÞ=ðInitial bio−oil ðgÞ þ Consumed H2 ðgÞÞ
ð1Þ
Mass of Product: weight of each bio-oil, coke and tar product (g) h i DOD ð%Þ ¼ oxygenfeed ð%Þ−oxygenproduct ð%Þ =oxygenfeed ð%Þ 100
ð2Þ
Hydrogen consumption ¼ ðPi =Ti Þ– X H2 ; f P f =T f Vgas =R
ð3Þ
Vgas: reactor batch volume occupying gas Pi or f: initial/final reactor pressure Ti or f: initial/final reactor temperature at 293 K XH2,f: hydrogen mole fraction in a produced gas at final stage R: gas constant (8.314 J/K·mol) TOF ðturn over frequencyÞ; moles=gcatalyst h ¼ ðPi =Ti Þ− X H2; f P f =T f ð4Þ Vgas =ðR wt of catalyst reaction timeÞ
Table 1 Experimental designs of algae bio-oil fuel catalytic upgrading.
RSM (CCD)
upgrading tests. The coded levels of 130, 190, and 250 °C for temperature and 4.1, 6.2, and 8.3 MPa for pressure were set as −1, 0, and + 1 respectively. Statistical analysis through ANOVA (analysis variance) was used to understand the fitness of the regression model at a 95% significance level. The experimental errors on the data at the center point of the CCD were obtained as a standard deviation. Each product yield of upgraded bio-oil, coke, tar, and loss + gas was obtained according to Eq. (1). The degree of deoxygenation (DOD) in the upgraded bio-oil was obtained by a different elemental composition according to Eq. (2). Hydrogen consumption during the catalytic upgrading was calculated based on Eq. (3), while turn over frequency (TOF) showing the amount of convertible initial bio-oil with a Pd/C catalyst before deactivation was obtained from Eq. (4).
3. Results and discussion 3.1. Product yields Microalgae bio-oil fractions were upgraded using a Pd/C catalyst to determine the effects of temperature and pressure on the catalyst activity in product yields and properties. Fig. 2 identifies the product yields of upgraded bio-oil, coke, tar, and loss/gas at each upgrading operating condition. The product yield of the upgraded middle fraction (F2 or fraction 2) microalgae bio-oil resulted in around 68% at 130 °C, 71% at 190 °C, and 50% at 250 °C. The statistical analysis indicated that the effect of temperature was more influential than pressure on the product yield. The best regression model for the product yield was the quadratic regression model with a p-value of 0.003. Two significant terms were
H. Nam et al. / Algal Research 24 (2017) 188–198
191
Fig. 2. Mass product yield after hydrotreatment with different fractions at various conditions.
the temperature terms of A (p-value = 0.0015) and A2 (p-value = 0.0012). Even though the pressure term on the yield is insignificant, the pressure (B) may affect the upgraded bio-oil yield when the pressure difference is large enough according to the coded equation (Eq. (5)). Similarly, the significant operating condition for the loss and gas yield was only the temperature terms of A (p-value = 0.0053) and A2 (p-value = 0.0033). The gas and loss yields at 250 °C were from 24 to 36% due to massive thermal and hydrocracking reactions. In contrast, the significant terms of the coke formation were mainly dependent on temperature (A, p-value = 0.013), pressure (B2, p-value = 0.037), and the interaction (AB, p-value = 0.008). This indicated that less coke was formed at a higher temperature and higher pressure condition, which can also be confirmed by the coded quadratic coefficients from Eq. (6). Li et al. [17,18] also observed that the total coke yield decreased to 2% at a higher hydrotreating temperature by converting the heavy chemicals to light one, which reduced the possibility of coke formation.
Upgraded liquid yield ¼ þ71−7:2A−2B−2AB−11:3A2 þ 1:3B2
ð5Þ
Coke formation ¼ þ6:04−0:5A þ 0:33B þ 0:75AB−0:12A2 –0:62B2 ð6Þ
Pd/C catalytic upgrading with raw bio-oil and a light fraction (F1 or fraction1) of bio-oil was carried out for comparison. The yield of the F1 upgrades (42–53%) was lower compared to the yield of F2 upgrades at the same operating conditions. As lighter hydrocarbon chemicals were included in the F1 bio-oil, many light chemicals were thermally cracked and converted into gaseous products through the catalytic reactions, resulting an increase in loss + gas yield (41–52%). Consequently, the tar content (1%) from the F1 upgrading was the lowest among the upgrades. On the other hand, the lowest product yield of raw bio-oil upgrading was obtained at 27%. The large production of coke and tar were mainly from a large amount of heavy chemicals in the bio-oil (mainly composed of paraffin waxes) [8]. 3.2. Physicochemical characteristics Microalgae pyrolytic bio-oil was distilled into three fractions (F1, F2, and F3) through a fractional distillation examined in a previous study [8]. The physical appearances of a fraction and its catalytic upgrades are illustrated in Fig. 3. The translucent dark-red color of fraction 2 was changed into translucent colors of dark-yellow and light-brown under the less severe conditions of 130 °C and 4.1 MPa. A cloudy darkgreen and dark-brown were obtained at more severe conditions of 190 °C and 250 °C. On the other hand, the physical appearance of light
Fig. 3. Pd/C catalytic upgraded microalgae bio-oil fractions.
192
H. Nam et al. / Algal Research 24 (2017) 188–198
fraction upgrades showed a translucent light-green color even after catalytic treatments. Physicochemical characteristics of produced bio-oil can be evaluated to understand the feasibility of bio-oils for fuel applications. Properties normally include TAN (total acid number), MC (moisture content), and HHV (higher heating value) as indicated in Table 2. Besides the properties in the current study, the standards of ASTM D6751 (biodiesel) and D7544 (pyrolysis biofuel) require a separate list of characteristics: kinematic viscosity, pH, density, ash content, cloud point, flash point, cetane number, oxidation stability, distillation temperature, etc. TAN of a bio-oil indicates the total acidity in bio-oil chemicals and is measured by the amount of basic chemical required to neutralize the given acidic bio-oil sample. The reduction in TAN is a very important factor if bio-oil is to be used in petroleum refineries. The biodiesel B100 standard of ASTM D6751 requires that the TAN be b 0.5 mg KOH/g. It is known that carboxylic acids, sugars, and extractives in a pyrolytic bio-oil increase the TAN level. From a previous study [8], the microalgae pyrolytic bio-oil showed 12.2 mg KOH/g, which increased to 17.2 mg KOH/g in a middle fraction (F2) and decreased to 0.12 mg KOH/g in a light fraction (F1). After Pd/C catalytic upgrades, the TAN of all samples were reduced. The light fraction (F1) upgrades were shown to be as low as b0.1 mg KOH/g from 0.12 mg KOH/g of the F1 distillate, while the middle fraction (F2) at 17.2 mg KOH/g was reduced to 1.09 mg KOH/g at 250 °C and 8.3 MPa. The raw bio-oil at 12.2 mg KOH/g was reduced to 2.44 mg KOH/g at 260 °C and 4.1 MPa. The improved TANs were lower than the TANs after catalytic upgrades in previous studies of the corn stover upgrades (5.3–20.8 mg KOH/g) and loblolly pine bio-oil upgrades (53–74 mg KOH/g) [19,20]. The high carboxylic acid composition of corn stover and loblolly pine biooil led to a relatively higher acid number. A quadratic model showed a significance on the TANs at various conditions (p-value b 0.000). Only the temperature terms of A (p-value b 0.000) and A2 (p-value = 0.001) were significant when the interaction term of temperature and pressure (AB) was close to significance (p-value = 0.053). The MC (moisture content) of pyrolysis bio-oil should be b30% in accordance with ASTM D7544, and cannot be directly used for a fuel substitute considering the moisture content from petro-fuel standards is b0.05% in B100 biodiesel, b 0.1% in gasoline, and b0.5% in crude oil [21, 22]. The microalgae bio-oil produced from slow pyrolysis contained a lower moisture content of 8.2% compared to other agricultural biomass bio-oils from jatropha, rice straw, and corn stover (10.2–20.3%) and the woody biomass of loblolly pine (21–32%) [3,4,20,23]. Two reasons for
Fraction 1a Upgrades B100 Gasoline Crude oil Fraction 3a a b
Conditions
TAN (mg KOH/g)
Water content (wt%)
HHV (MJ/kg)
Pyrolysis 260–4.1 Distillation 130–4.1 130–6.2 130–8.3 190–4.1 190–6.2b 190–8.3 250–4.1 250–6.2 250–8.3 Distillation 130–6.2 190–6.2 ASTM D6751 [22] [21] Distillation
12.2 2.44 17.2 9.35 9.64 12.6 1.86 3.18 ± 0.45 3.02 1.39 1.28 1.09 0.12 b0.1 b0.1 b0.5 – – –
8.21 2.16 1.25 0.65 0.92 1.14 1.09 0.82 ± 0.04 0.75 0.95 0.69 0.67 2.18 0.38 0.32 b0.05 b0.1 b0.5 –
38.6 40.6 41.2 42.5 41.5 40.0 41.9 42.0 ± 0.2 42.4 40.6 41.7 42.9 40.0 39.9 41.1 N/A 43–47 41–43 40.6
Imported from Nam et al. [8]. Average value of the center points.
3.3. Chemical compositions of upgraded fractions The chemical compositions of upgraded bio-oils were analyzed to better understand the mechanism of the catalytic upgrading process. Elemental compositions were initially obtained to observe the direct differences in elemental changes as shown in Table 3. The elemental changes of F2 upgrades at lower severity conditions of 130 and 190 °C were not as significant as were the changes in raw bio-oil and F1 upgrades. It can be inferred from this that the larger the difference in temperature conditions for upgrading and distillation, the more variation in elemental properties were obtained. Thus, only the F2 upgrades at 250 °C showed a relatively large change from F2 distillates. The 72% carbon and 9.7% hydrogen from raw bio-oil increased to a maximum of 78.7% carbon and 11.5% hydrogen for F2 upgrading at 250 °C and 8.3 MPa, Table 3 Elemental compositions of catalytic upgrades.
Table 2 Properties of catalytic upgrades of microalgae bio-oil.
Raw bio-oila Upgrades Fraction 2a Upgrades
the lower MC compared to lignocellulosic biomass bio-oil were a lower oxygen content and a higher amount of extractives in the dried microalgae [5,8]. After fractionation, the high MCs were reduced to 1.3–2.2% by forming an aqueous layer in a light fraction [8], which was comparable to the MCs of 1.8–3.9% of raw microalgae bio-oil upgrading by Monet [15]. In contrast, a catalytic upgrading with F1 and F2 bio-oils in the current study resulted in an additional reduction in the MC to as low as 0.32% through intensive hydrogenation and hydrodeoxygenation. An increase in pressure at 130 °C increased the MC of bio-oil, while the MC decreased at 250 °C. No significant statistical models for the moisture content were identified from ANOVA. HHV (higher heating value) is a representative fuel property defining the improvement of bio-oil. The degree of HHV enhancement from F2 distillates to the catalytic upgrading was lower than that from raw bio-oil (38.6 MJ/kg) to distillates (40–41 MJ/kg) as shown in Table 2. F2 distillate upgrades showed 40–43 MJ/kg when F1 upgrades were ranged from 40 to 41 MJ/kg. The ANOVA statistical analysis showed a significance with a quadratic model (p-value = 0.001) with the significant terms of temperature (A2, p-value = 0.01) and interaction term (AB, p-value = 0.000). The upgraded HHVs in the current study showed similar ranges compared to other upgrading studies. The distilled bio-oil showed 40.4 MJ/kg for Spirulina sp. and 41.2 MJ/kg for Nannochlropsis o. [8,9]. HHV of other microalgae upgrading studies indicated 37–40 MJ/kg for Nannochlropsis o. with zeolite catalytic upgrading [15], 42–43 MJ/kg for Nannochlropsis sp. with supercritical water upgrading [10], and 39.6–41.9 MJ/kg for Scenedesmus a. and 40.9–43.2 MJ/kg for Nannochloropsis g. with Pt/Al2O3 and HZSM-5 [12].
Raw bio-oila Upgrades Fraction 2a Upgrades
Conditions
C
H
N
S
O
H/C
Pyrolysis
72.2
9.7
6.0
0.220
11.9
1.61 0.123
260–4.1 78.1 Distillation 75.5
10.3 10.4
5.9 5.7
0.04 0.110
5.7 8.3
1.58 0.055 1.65 0.082
130–4.1 130–6.2 130–8.3 190–4.1 190–6.2b
10.9 10.7 10.4 11.2 11.4 ± 0.01 11.3 10.8 11.1 11.5 9.9
5.7 5.8 6.2 5.4 5.2 ± 0.2 5.3 5.8 5.4 4.6 4.6
0.135 0.215 0.146 0.154 0.12 ± 0.05 0.128 0.056 0.067 0.072 0.360
5.9 7.2 7.8 4.3 5.2 ± 0.6 5.7 5.4 4.6 5.1 16.1
1.69 1.69 1.66 1.71 1.76
0.057 0.071 0.078 0.041 0.050
1.75 1.67 1.68 1.76 1.73
0.055 0.052 0.044 0.049 0.175
11.4 11.7 8.4
4.9 4.6 5.6
0.12 0.14 0.05
6.4 5.2 6.3
1.77 0.062 1.75 0.050 1.27 0.06
77.4 76.1 75.4 78.9 78.1 ± 0.5 190–8.3 77.6 250–4.1 78.0 250–6.2 78.9 250–8.3 78.7 Distillation 69.0
Fraction 1a Upgrades 130–6.2 77.3 190–6.2 78.3 Distillation 79.7 Fraction a 3 a b
Imported from Nam et al. [8]. Average value of the center points.
O/C
H. Nam et al. / Algal Research 24 (2017) 188–198
and 78.3% C and 11.7% for F1 upgrading at 190 °C–6.2 MPa. When it comes to nitrogen content, there was no substantial reduction through the catalytic upgrading with two operating conditions (temperature from 130 °C to 250 °C and pressure from 4.1 to 8.3 MPa). The nitrogen and sulfur contents in bio-oil were also considered as important elements for the prevention of NOx and SOx contaminants during combustion reactions. The nitrogen content of F2 in the current study slightly decreased at increasing temperature even though it was not a substantial reduction. Also, F1 distillate showed a lower level of nitrogen content because most nitrogen and oxygen accumulated in the heavy asphaltene fractions determined by Biller et al. [24]. The best nitrogen regression model of 2FI (two factor interaction) was still not significant with a p-value of 0.1213. However, it indicated the effect of pressure and the interaction term showed significant with a 90% confidence level when temperature term was insignificant. Still, higher H2 pressures (13.8–15 MPa) for catalytic upgrading might not be a major condition for nitrogen removal according to previous studies [24]. The sulfur content in raw bio-oils was originally relatively low compared to the sulfur content (0.05–6%) in crude oil [21]. In addition, the sulfur content fell to below half or one-third of the initial sulfur content after catalytic upgrading. The reduction in sulfur content during F2 upgrading was only possible at the highest condition of severity, 250 °C. For the raw bio-oil catalytic upgrade, the elemental compositions are similar to those obtained in a heavy distillation fraction (C: 79.7%, N: 5.6%, and O: 6.3%) from a previous study [8] except for the hydrogen content (H: 8.4%), indicating active hydrogenation reactions occurred during the Pd/C upgrading. A Van Krevelen diagram was constructed with the H/C and O/C ratios for better determination of the elemental transitions through pyrolysis of algae, distillation and catalytic upgrading as shown in Fig. 4. Except for the direct raw bio-oil upgrading, the H/C range of F1 and F2 upgrades is similar to that of petroleum based fuels (1.65–1.80). The H/C ratio after F1 distillate upgrading showed the highest at around 1.77, while the O/C ranged between 0.04 and 0.08. The direction toward an increase in H/C and a decrease in O/C indicates an intensive decarboxylation or decarbonylation, while a decrease in H/C and a decrease in O/C points a dehydration reaction. A better upgrading performance was recognized from distilled bio-oil upgrading compared to raw biooil upgrading as a lower hydrogen content was obtained in the direct upgraded bio-oil. Similarly, the hydrogen contents decreased with raw bio-oil HZSM-5 upgrades from a previous study [15]. This can be inferred from the existence of heavy fraction. From the ANOVA analysis, both the H/C and O/C ratios showed significance on the quadratic model with a p-value = 0.013. The significance of terms on each ratio was different. The H/C ratio was significantly affected by temperature
193
(A2, p-value = 0.006). The interaction factors (AB, p-value = 0.089) and pressure term (B, p-value = 0.090) also showed significance with a 90% confidence level. In the case of the O/C ratio, only the temperature terms of A (p-value = 0.003) and A2 (p-value = 0.040) showed significant, while the interaction term (AB, p-value = 0.082) was significant with a 90% confidence level. From the statistical significance of the elemental transitions, both temperature and pressure need to be considered for a hydrogen increase, while the temperature term mainly influences oxygen removal. Fig. 5 shows the GC-MS chromatograms for the middle fraction and its Pd/C catalytic upgrades at different temperatures (130, 190, and 250 °C) and pressure (4.1 MPa). As the catalytic temperature increased, more and higher peaks in the upgrades were identified at a shorter retention time, indicating lower carbon number chemicals through hydrodeoxygenaton, hydrocracking and hydrogenantion. The upgrades at the shorter retention time were composed mainly of aromatic hydrocarbons and some new heteroatom containing chemicals (pyrrole and nitrile). Similar heights of relative peaks between olefin/alcohol and alkane groups in fraction 2 changed differently in upgrades. The saturated alkane group peaks were much higher at the highest temperature condition, for example, by converting1-undecene and 1-dodecanol into undecane and dodecane, through hydrogenation and hyrodeoxygenation. Also, active hydrocracking can be expected from the chemical transitions of nitrile groups as newly produced light nitrile peaks were found at the shorter retention time through a higher catalytic temperature, while some heavy nitriles substantially decreased. Categorized chemical groups from GC-MS were compared as shown in Fig. 6. The substantial hydrogenation conversion for olefin (alkene) and decarbonylation conversion for ketone led to an increase in saturated hydrocarbon paraffin (alkane) and naphthene (cycloalkane) groups. The higher temperature and pressure helped to increase the production of naphthene and aromatic and decrease the chemicals of unsaturated olefins and oxygenated ketone. There was a minimal effect of temperature at 130 °C on the conversions, which was a different trend compared to switchgrass bio-oil upgrading at 110–120 °C with a nickel catalyst [25]. The effect of pressure on the chemical compositions was substantial, especially at the temperature of 250 °C, considering an increase in the naphthene and aromatic hydrocarbons and a decrease in nitriles. Still, many nitrile compounds were present even after catalytic upgrading, which can be explained with a similar reaction between fatty acids and produced ammonia from other nitrile chemicals through denitrogenation [10]. Overall the relative chemical analysis of middle fraction upgrades showed over 70% paraffin, close to 10% naphthene and aromatics, and b 5% olefin, nitrile and ketone. Fig. S1 also shows the chemical group composition of the upgraded light fraction. A similar
Fig. 4. Van Krevelen diagram of upgraded distilled bio-oils.
194
H. Nam et al. / Algal Research 24 (2017) 188–198
Fig. 5. GC-MS chromatogram of fraction 2 and its upgrades at different temperatures.
trend of most chemical group transitions after catalytic upgrading were observed even though the aromatics in light fraction upgrades were not changed much compared to the original light fraction. The chemical transition through Pd/C catalytic upgrading was also identified using FTIR analysis as shown in Fig. 7. Only F2 samples with different temperatures were selected for the analysis. The adjacent aromatic hydrogen peak at 748 cm−1 showed an increased relative height in peaks as the catalytic temperature increased, which supports an increase of aromatics in upgrades. The peaks between 1350 and 1460 cm−1 for aliphatic hydrocarbons of CH3 and CH2 indicated an increase in a paraffin chemical group in upgrades. The double bonding oxygenate peak of C_O stretching ranging from 1650 to 1750 cm− 1 showed a decreased relative peak height due to the deoxygenated chemicals in upgrades. The alkene peaks of C_C stretching vibration
at 1630 to 1660 cm− 1 decreased at a higher temperature catalytic condition. 3.4. Gas analysis The gas products of produced hydrocarbon gases and unreacted hydrogen were collected after each catalytic upgrading as shown in Fig. 8. As the temperature increased at the same pressure, more CO, CH4, C2H6, and C3H8 gases were produced, indicating thermal and catalytic cracking. Similar results were reported by Li and Savage [26] that the light saturated hydrocarbons (C1–C4) increased at a higher temperature. Accordingly, the hydrogen consumption (Eq. (3)) increased as the temperature condition increased. On the other hand, the increased pressure at the constant temperature led to a reduction in gas yields
Fig. 6. Chemical composition functional groups of middle fraction (F2) catalytic upgrades.
H. Nam et al. / Algal Research 24 (2017) 188–198
195
Fig. 7. FTIR of upgraded bio-oils and Fraction 2 of pyrolytic algae bio-oil.
because the addition of a higher activation energy on the reactions helped converting the thermally cracked hydrocarbons into heavier saturated hydrocarbon chemicals. Consequently, hydrogen consumption increased during the cracking under higher pressure. The lower amount of gas product from direct bio-oil upgrading was due to a large amount of tar and coke formation in the reactor. Unlike other lignocellulosic based pyrolytic bio-oil upgrading, CO2 gas was rarely produced with the algae bio-oil. Similarly, Li and Savage [26] also experienced the absence of CO2 from the algae bio-oil upgrading with an HZSM-5 catalyst, while CO2 was obtained from their previous algae bio-oil upgrading in a supercritical water study [10]. Also, the absence of water in microalgae bio-oil prevented the possible reactions of a water-gas shift reaction (CO + H2O ⇌ CO2 + H2) and steam reforming (CH4 + H2O ⇌ CO + 3H2), which minimized CO2 gas production. At a high temperature of 250 °C, more hydrogen was produced in the gas compared to the consumed hydrogen gas, which can be from the in situ catalytic hydrogen production. Similarly, an increase in hydrogen content during upgrading was constructed with an increase in the amount of catalyst [11].
3.5. Hydrogen consumption and turnover frequency (TOF) The turnover frequency was obtained from Eq. (4), which represents the hydrogen consumption per amount of catalyst per reaction time. In other words, the degree of catalytic activity on bio-oil conversion is understandable during the upgrading. Table 4 clearly shows that the changes in TOF values are followed by hydrogen consumption. An increase in temperature and pressure conditions led to a tendency of increased TOF, which indicated higher catalytic activity. Both the temperature (p-value = 0.003) and pressure (p-value = 0.01) conditions significantly affected the TOF. A similar trend of activity on catalytic reactions at higher temperatures was also reported in a corn stover bio-oil with a Ru/C catalyst [19]. In contrast, the direct bio-oil upgrading showed a relatively low TOF (0.009 mol/g catalyst-h) even at a high temperature (260 °C) due to the presence of the heavy fraction, which, unlike F1 and F2 distillates, hinder the catalytic reactions. From the close TOF values of F1 and F2 at the same conditions, similar catalytic activity during the upgrading can be inferred.
Fig. 8. Produced gas composition during catalytic reactions of F2 and raw bio-oil upgrading.
196
H. Nam et al. / Algal Research 24 (2017) 188–198
Table 4 Hydrogen consumption and turnover frequency.
260–4.1 130–4.1 130–6.2 130–8.3 190–4.1 190–6.2* 190–8.3 250–4.1 250–6.2 250–8.3 130–6.2 190–8.3 *
Table 5 BET surface area and pore volume of Pd/C catalyst before and after use.
Initial fuel
H2 mmol/g bio-oil
TOF mol/g catalyst-h
Raw F2 F2 F2 F2 F2 F2 F2 F2 F2 F1 F1
2.8 0.6 0.8 2.8 3.4 4.3 9.7 4.6 6.1 7.9 1.3 8.2
0.009 0.004 0.005 0.019 0.023 0.029 0.065 0.032 0.041 0.052 0.008 0.055
Average value of the center points.
3.6. Catalyst characteristics The isotherm characteristics of new and used Pd/C were determined for understanding the degree of catalyst deactivation after the hydrotreatment of microalgae bio-oil as shown in Fig. 9(a). According to the IUPAC guidelines [27], the new Pd/C catalyst is close to Type I (a), shows a steep curve below 0.15 relative pressure (P/Po), and reaches a plateau, which indicate the presence of micropores. On the other hand, the used catalyst is similar to Type I (b) due to the round knee below, indicating the presence of wider micropores. Both the catalysts obtained Type H4 hysteresis (slit-shaped pores) even though the used catalyst shows a wider hysteresis loop. The appearance of hysteresis loops in the multilayer range of isotherms is associated with capillary condensation. Adsorption of chemicals and formation of cokes during the upgrading possibly helped the formation of capillary condensation on the catalyst pores. The ultimate analysis of new and used catalysts at 190 °C–6.2 MPa showed an increase in hydrogen (2.3%), nitrogen (1.52%), and sulfur (0.36%) contents of the used catalyst compared to the new catalyst (1.5% H, 0.19% N, and 0.07% S). Consequently, the surface area and the pore volume were reduced as indicated in Table 5. The least surface area of 447 m2/g was obtained at 250 °C upgrading from the new Pd/C catalyst of 1306 m2/g due to the deactivation causes including chemical/thermal degradation, fouling, coking and poisoning [28]. H2-TPR was performed to understand the reducibility of catalyst. New Pd/C initially showed a sharp negative peak at 62 °C, which indicated desorption of H2 on the surface of metallic Pd. Under
New Pd/C Used Pd/C at 130 °C Used Pd/C at 190 °C Used Pd/C at 250 °C
Surface area (m2/g)
Pore volume (cm3/g)
1306.3 700.4 539.6 447.8
0.638 0.383 0.330 0.268
hydrogen gas at below 100 °C, PdO and PdCl2 are easily reduced to metallic Pd and desorbed to PdHx as reported by others [29,30]. The second peak at 570 °C is corresponded to the reduction of the oxygen containing functional groups in carbon support. The used Pd/C catalyst showed no significant peaks as it reduced during the catalytic upgrading. Further regeneration stage would make the spent Pd/C catalyst available for more catalytic bio-oil upgrading. General regeneration methods for carbon based catalyst include thermal regeneration (700–1000 °C) under inert gas or oxidizing gas [31], solvent extractive regeneration, and reactive (acid-base or oxidative transformation) regeneration [32]. In the current study, a simple CO2 regeneration of spent catalyst was performed at 800 °C for an hour in a fixed bed reactor. The regenerated catalyst was evaluated by comparing with the used catalysts using a TGADTG as shown in Fig. 10. One peak from 70 °C to 320 °C from used catalyst was considered as tar or heavy fraction of bio-oil, which was not fully washed with solvent. The second peak rose from 360 °C due to the decomposition of carbon. Similar trend of catalyst weight loss was also reported with Ni, Pt, and Ru catalyst due to the waxy or intermediate compounds produced during hydrotreatment [33]. On the other hand, the CO2 regenerated catalyst does not show any peak under 400 °C, while the trend of weight reduction above 400 °C was similar. 3.7. Evaluation of mass, energy and carbon flows The Sankey diagram is used to evaluate the conversion flows from microalgae to bio-oil, to distilled bio-oil, and to catalytic upgrades as shown in Fig. 11. The pyrolytic bio-oil mass yield (21%) occupies half (50%) the energy of initial microalgae through pyrolysis due to enhance the HHV of bio-oil. The distillated fraction 1 and 2 (a potential fuel substitutes) showed 16% of mass yield and 40% of energy recovery over initial microalgae raw materials. After catalytic upgrading, the mass and energy yield of fraction 1 and 2 were 10.1% and 25.5%, respectively from 100% raw microalgae sample. Considering the mass and energy yields of direct bio-oil upgrades showed 5.7% and 14.2%, respectively,
Fig. 9. New and used Pd/C catalysts of (a) adsorption and desorption curves and (b) H2-TPR profiles.
H. Nam et al. / Algal Research 24 (2017) 188–198
Fig. 10. New and spent catalysts analysis for derivative thermogravimetric (DTG).
Fig. 11. Sankey diagram of mass, energy and carbon percent from dried algae to upgrades.
197
198
H. Nam et al. / Algal Research 24 (2017) 188–198
a higher mass and energy recovery can be achieved with an extra step of distillation as well as better physicochemical properties. The carbon distribution was also obtained, which shows a similar trend of energy flow. Similarly, the carbon recovery of direct bio-oil upgrades shows 15.2%, which is less than that of fraction 1 and 2 (27.3%). 4. Conclusion A catalytic upgrading of distilled microalgae fractions was performed using a noble metal catalyst Pd/C based on the response surface methodology (RSM). The upgrading conditions were temperatures from 130 to 250 °C and pressures from 4.1 to 8.3 MPa. The light fraction and raw pyrolysis bio-oil were also upgraded for comparison. Better upgraded bio-oil was obtained from distilled fraction catalytic upgrading compared to direct catalytic upgrading. Both the oxygen and hydrogen content of F1 and F2 upgrades were improved through deoxygenation, hydrogenation, and hydrocracking. HHV and TAN values were also improved accordingly. In contrast, the raw bio-oil upgrade showed a limited improvement, which showed a decrease in the H/C ratio. Temperature was a main influence on the improvement of bio-oil properties followed by pressure and the interaction of both. The ketones in upgrades were almost all removed at the catalytic upgrading condition above 190 °C, while nitrogen containing chemicals decreased slightly. Paraffin and aromatic chemical groups were significantly produced when the olefins decreased. The produced gas mainly included CO, CH4, C2H6, C3H8, and a trace of CO2, which showed a substantial catalytic conversion of bio-oil. The turnover frequency with hydrogen consumption indicated similar catalytic activity of F1 and F2 distillation upgrading at the same conditions, while low catalytic activity was determined in raw bio-oil upgrading. The spent catalyst showed a decrease in surface area and volume with an increase in nitrogen and sulfur contents. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.algal.2017.03.021. Acknowledgment The authors acknowledge the facilities, and the scientific and financial assistance of the Bio Energy Testing and Analysis Laboratory (BETA lab) and AgriLife Research at Texas A&M University. The work was also supported by the dissertation fellowship from OGAPS (Office of Graduate and Professional Studies) and undergraduate research scholarship from Honors & Undergraduate Research, Texas A&M University. References [1] L. Dufossé, P. Galaup, A. Yaron, S.M. Arad, P. Blanc, K.N.C. Murthy, G.A. Ravishankar, Microorganisms and microalgae as sources of pigments for food use: a scientific oddity or an industrial reality? Trends Food Sci. Technol. 16 (2005) 389–406. [2] Y. Chisti, Biodiesel from microalgae, Biotechnol. Adv. 25 (2007) 294–306. [3] H. Nam, S.C. Capareda, N. Ashwath, J. Kongkasawan, Experimental investigation of pyrolysis of rice straw using bench-scale auger, batch and fluidized bed reactors, Energy 93 (Part 2) (2015) 2384–2394. [4] J. Kongkasawan, H. Nam, S.C. Capareda, Jatropha waste meal as an alternative energy source via pressurized pyrolysis: a study on temperature effects, Energy 113 (2016) 631–642. [5] M.C.C. Maguyon, S.C. Capareda, Evaluating the effects of temperature on pressurized pyrolysis of Nannochloropsis oculata based on products yields and characteristics, Energy Convers. Manag. 76 (2013) 764–773.
[6] R. Shakya, J. Whelen, S. Adhikari, R. Mahadevan, S. Neupane, Effect of temperature and Na2CO3 catalyst on hydrothermal liquefaction of algae, Algal Res. 12 (2015) 80–90. [7] X. Bai, P. Duan, Y. Xu, A. Zhang, P.E. Savage, Hydrothermal catalytic processing of pretreated algal oil: a catalyst screening study, Fuel 120 (2014) 141–149. [8] H. Nam, J. Choi, S.C. Capareda, Comparative study of vacuum and fractional distillation using pyrolytic microalgae (Nannochloropsis oculata) bio-oil, Algal Res. 17 (2016) 87–96. [9] B.E. Eboibi, D.M. Lewis, P.J. Ashman, S. Chinnasamy, Hydrothermal liquefaction of microalgae for biocrude production: improving the biocrude properties with vacuum distillation, Bioresour. Technol. 174 (2014) 212–221. [10] P. Duan, P.E. Savage, Upgrading of crude algal bio-oil in supercritical water, Bioresour. Technol. 102 (2011) 1899–1906. [11] P. Duan, P.E. Savage, Catalytic hydrotreatment of crude algal bio-oil in supercritical water, Appl. Catal. B Environ. 104 (2011) 136–143. [12] D.L. Barreiro, B.R. Gómez, F. Ronsse, U. Hornung, A. Kruse, W. Prins, Heterogeneous catalytic upgrading of biocrude oil produced by hydrothermal liquefaction of microalgae: state of the art and own experiments, Fuel Process. Technol. 148 (2016) 117–127. [13] L. Zhou, A. Lawal, Hydrodeoxygenation of microalgae oil to green diesel over Pt, Rh and presulfided NiMo catalysts, Cat. Sci. Technol. 6 (2016) 1442–1454. [14] L. Zhou, A. Lawal, Evaluation of presulfided NiMo/γ-Al2O3 for hydrodeoxygenation of microalgae oil to produce green diesel, Energy Fuel 29 (2014) 262–272. [15] M. Maguyon, Technical Feasibility Study on Biofuels Production from Pyrolysis of Nannochloropsis oculata and Algal Bio-oil Upgrading, 2013. [16] H. Nam, S. Capareda, Experimental investigation of torrefaction of two agricultural wastes of different composition using RSM (response surface methodology), Energy 91 (2015) 507–516. [17] X. Li, R. Gunawan, Y. Wang, W. Chaiwat, X. Hu, M. Gholizadeh, D. Mourant, J. Bromly, C. Li, Upgrading of bio-oil into advanced biofuels and chemicals. Part III. Changes in aromatic structure and coke forming propensity during the catalytic hydrotreatment of a fast pyrolysis bio-oil with Pd/C catalyst, Fuel 116 (2014) 642–649. [18] W. Chaiwat, R. Gunawan, M. Gholizadeh, X. Li, C. Lievens, X. Hu, Y. Wang, D. Mourant, A. Rossiter, J. Bromly, C. Li, Upgrading of bio-oil into advanced biofuels and chemicals. Part II. Importance of holdup of heavy species during the hydrotreatment of bio-oil in a continuous packed-bed catalytic reactor, Fuel 112 (2013) 302–310. [19] J.A. Capunitan, S.C. Capareda, Hydrotreatment of corn stover bio-oil using noble metal catalysts, Fuel Process. Technol. 125 (2014) 190–199. [20] Y. Luo, V.K. Guda, E.B. Hassan, P.H. Steele, B. Mitchell, F. Yu, Hydrodeoxygenation of oxidized distilled bio-oil for the production of gasoline fuel type, Energy Convers. Manag. 112 (2016) 319–327. [21] J.G. Speight, The Chemistry and Technology of Petroleum, CRC Press, 2014. [22] Oak Ridge National Laboratory, Transportation Energy Data Book - Chapter 6 Alternative Fuel and Advanced Technology Vehicles and Characteristics, 2015. [23] J.A. Capunitan, S.C. Capareda, Assessing the potential for biofuel production of corn stover pyrolysis using a pressurized batch reactor, Fuel 95 (2012) 563–572. [24] P. Biller, B.K. Sharma, B. Kunwar, A.B. Ross, Hydroprocessing of bio-crude from continuous hydrothermal liquefaction of microalgae, Fuel 159 (2015) 197–205. [25] T. Imam, S. Capareda, Characterization of bio-oil, syn-gas and bio-char from switchgrass pyrolysis at various temperatures, J. Anal. Appl. Pyrolysis 93 (2012) 170–177. [26] Z. Li, P.E. Savage, Feedstocks for fuels and chemicals from algae: treatment of crude bio-oil over HZSM-5, Algal Res. 2 (2013) 154–163. [27] J. Rouquerol, F. Rouquerol, P. Llewellyn, G. Maurin, K.S. Sing, Adsorption by Powders and Porous Solids: Principles, Methodology and Applications, Academic Press, 2013. [28] M.D. Argyle, C.H. Bartholomew, Heterogeneous catalyst deactivation and regeneration: a review, Catalysts 5 (2015) 145–269. [29] P. Sangeetha, K. Shanthi, K.S.R. Rao, B. Viswanathan, P. Selvam, Hydrogenation of nitrobenzene over palladium-supported catalysts—effect of support, Appl. Catal. A Gen. 353 (2009) 160–165. [30] C. Tu, S. Cheng, Ceria-modified palladium/activated carbon as a high-performance catalyst for crude caprolactam hydrogenation purification, ACS Sustain. Chem. Eng. 2 (2014) 629–636. [31] H.F. Abbas, W.W. Daud, Thermocatalytic decomposition of methane for hydrogen production using activated carbon catalyst: regeneration and characterization studies, Int. J. Hydrog. Energy 34 (2009) 8034–8045. [32] M. Sheintuch, Y.I. Matatov-Meytal, Comparison of catalytic processes with other regeneration methods of activated carbon, Catal. Today 53 (1999) 73–80. [33] Z. Wang, S. Adhikari, P. Valdez, R. Shakya, C. Laird, Upgrading of hydrothermal liquefaction biocrude from algae grown in municipal wastewater, Fuel Process. Technol. 142 (2016) 147–156.
Contents lists available at ScienceDirect
Algal Research journal homepage: www.elsevier.com/locate/algal
Catalytic upgrading of fractionated microalgae bio-oil (Nannochloropsis oculata) using a noble metal (Pd/C) catalyst Hyungseok Nam a,b,⁎, Changkyu Kim c, Sergio C. Capareda a, Sushil Adhikari b a b c
Bio-Energy Test and Analysis Laboratory (BETA Lab), Department of Biological and Agricultural Engineering, Texas A&M University, College Station, TX 77843, USA Department of Biosystems Engineering, Auburn University, Auburn, AL 36849, USA Texas A&M Energy Institute, Texas A&M University, College Station, TX 77843, USA
a r t i c l e
i n f o
Article history: Received 24 October 2016 Received in revised form 19 February 2017 Accepted 24 March 2017 Available online xxxx Keywords: Distillation bio-oil Hydrotreatment Catalytic upgrading Microalgae Palladium
a b s t r a c t Pyrolytic bio-oil was chemically upgraded after physically distilled upgrades to meet the petroleum transportation fuel substitute. A Pd/C catalyst was used to upgrade the microalgae pyrolytic bio-oil to determine the effect of different distillation fractions and catalytic upgrading conditions on the yields and properties. The middle distillation fraction (F2) was upgraded under various temperature (130 to 250 °C) and pressure (4.1 to 8.3 MPa) conditions based on response surface methodology (RSM). The light distillation fraction (F1) and raw bio-oil were also catalytically upgraded for the comparison. The distillation step prior to catalytic upgrading led to a better quality of upgraded bio-oil compared to the direct bio-oil upgrades. Both the oxygen and hydrogen contents of light and middle fraction upgrades were improved, while the upgraded raw bio-oil showed limited improvement. The other properties of HHV and TAN with the middle fraction upgrades were improved to 42.9 MJ/kg and 1.09 mg KOH/g, respectively, at the severe condition as most of the ketones in upgrades were removed. Also, paraffin and aromatic chemical groups were significantly produced at the expense of the olefin groups through hydrogenation and hydrodeoxygenation. Thus, the catalytic upgrading after a distillation stage enhanced the quality of biofuel that can be a petroleum fuels substitute or additives. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Microalgae, as a source of third generation biofuels, have obtained much attention as an alternative energy source because they overcome the drawbacks of first and second generation biofuels. With simple conditions required for growth, light, sugar, CO2, N, P, and K, photosynthetic microalgae microorganisms grow rapidly even on non-arable land. The large amounts of lipids, proteins and carbohydrates in microalgae can be processed into biofuel, medicine, and food [1]. Many microalgae biofuels were studied in the form of biodiesel from lipid transesterification [2], bio-oil from the thermal conversion process of pyrolysis [3–5] and hydrothermal liquefaction (HTL) [6]. The transesterification process for biodiesel produced a large amount of residue, while the bio-oils showed high viscosity, high acidity, and high nitrogen and sulfur contents [7,8].
⁎ Corresponding author at: Bio-Energy Test and Analysis Laboratory (BETA Lab), Department of Biological and Agricultural Engineering, Texas A&M University, College Station, TX 77843, USA. E-mail addresses: [email protected], [email protected] (H. Nam), [email protected] (S.C. Capareda).
http://dx.doi.org/10.1016/j.algal.2017.03.021 2211-9264/© 2017 Elsevier B.V. All rights reserved.
An upgrading of thermally converted bio-oils is required for them to be used as potential fuel substitutes or additives. A physical method of distillation and a chemical method of catalytic upgrading were used to upgrade pyrolytic bio-oil by reducing the O/C ratio and increasing the H/C ratio, which also enhances the other fuel properties of viscosity, acid number, heating value, and water content. Distillation as a physical separation process upgrades bio-oils based on the relative volatility of chemicals at various temperatures. Several techniques were used to upgrade the microalgae bio-oil with direct, fractional, steam, vacuum, and molecular distillations. Bio-oil (Spirulina sp. and Tetraselmis sp.) from hydrothermal liquefaction (HTL) was fractionated using a vacuum distillation at temperature conditions of 300 °C and 350 °C [9]. The maximum yields of an HTL bio-oil and its distillates were 58% and 73%, respectively, whereas each HHV (higher heating value) were 32 MJ/kg and 40 MJ/kg with a 97% deoxygenation of bio-oil. In our previous study [8], the bio-oil of Nannochloropsis oculata microalgae was physically upgraded using fractional and vacuum distillation setups. Both the techniques showed similar distillate properties, while a higher product yield was obtained from fractional distillation when better separation was made with vacuum distillation. The lower operating temperature of vacuum distillation prevented polymerization in the heavy fraction. The HHV and the lowest total acid number (TAN) were
H. Nam et al. / Algal Research 24 (2017) 188–198
41.2 MJ/kg and 0.1 mg KOH/g obtained through deoxygenation and dehydration. Still, the fuel needed a further bio-oil upgrading step to be used as a petroleum fuel substitute due to the existence of oxygen and nitrogen contents, and high acidity in a certain bio-oil fraction. A chemical upgrading of catalytic hydrotreatment is performed by removing the oxygen, nitrogen, and sulfur contents from bio-oil, which is also used in the conventional petroleum refining process. Catalytic upgrading generally requires the reactions between hydrogen gas and bio-oil at high pressure and temperature with enough reaction time. Various alumina, silicate, and carbon based catalysts for bio-oil upgrading were evaluated from many studies. A catalyst screen study by Bai et al. [7] used 12 different catalysts for HTL bio-oil (Chlorella pyrenoidosa) upgrading. The highest product yield (77%) and a HHV of 45.3 MJ/kg for upgraded bio-oil were obtained with a mixed catalyst of Ru/C and Raney-Ni, and the lowest O/C (0.01) and N/C (0.017) were achieved with Ru/C and Raney-Ni, respectively. An HTL bio-oil (Nannochloropsis sp.) upgrading in supercritical water also conducted by Duan and Savage [10] to determine the effect of Pt/C, H2 pressure (3.5 MPa), and the pH of bio-oil on the bio-oil upgrades at 400 °C. The upgraded bio-oil showed a substantially decreased acid number, 25 mg KOH/g, with a decrease in O/C (0.041) and H/C (1.64) contents, which led to an increase the HHV (43 MJ/kg). They also conducted a similar study using Pd/C at the same temperature and pressure conditions to find the effect of reaction time (1–8 h) and catalyst loading (5–80%) on the hydrotreatment products [11]. The upgraded bio-oil properties with Pd/C upgrading showed 43.5 MJ/kg for HHV, 0.028 for O/C, and 1.79 H/C, which had a somewhat better result compared to the Pt/C upgrading. Barreir et al. [12] conducted HTL bio-oil upgrading (Scenedesmus Almeriensis and Nannochlopsis Gaditana) using Pt/Al2O3 and HZSM-5 at 400–450 °C with 4–8 MPa. The addition of water for the catalytic upgrading at a lower H2 pressure produced more oxygen chemicals compared to the upgrading under only H2 pressure. Also, Lin et al. [13,14] produced green diesel from hydrodeoxygenation of extracted microalgae oil using various catalysts (NiMo, Pt and Rh) in a continuous flow microreactor. The optimum yield of hydrocarbon ranging from C13 to C20 was 56% at operating conditions of 500 psi, 360 °C, and 1 s contact time. Maguyon [15] studied the pyrolysis bio-oil upgrading (Nannochloropsis oculata) with an HZSM-5 catalyst at a constant pressure (4.1 MPa) to determine the effect of temperature (200–300 °C) and reaction time (1–4 h). Mass balance was achieved based on the amount of upgraded bio-oil, coke, tar, and gas. The upgraded bio-oil showed a basic pH level (7.7–8.9) and the HHV ranged from 37.5–40 MJ/kg with a 1.46 H/C and 0.032 O/C. Even though many upgrading methods with various catalysts were studied based on the various operating conditions, few catalytic upgrading studies of distilled bio-oil have been reported. As a continuing distillation study of Nam et al. [8], the distilled bio-oil of light and middle fractions were used for chemical catalytic upgrading. The current study can expand the understanding of the effectiveness of the catalytically upgraded bio-oil from distilled bio-oil. The objectives of the study were to, 1) assess the effects of distilled bio-oils through Pd/C catalytic upgrading on the properties of bio-oil, 2) evaluate the performance of the catalyst at various conditions of temperature and pressure on the deoxygenation, hydrogen consumption, and turnover frequency, and. 3) to investigate the catalyst deactivation after catalytic upgrading. 2. Experimental 2.1. Sample preparation An algae species, Nannochloropsis oculata, from the Texas A&M AgriLife pond in Pecos, Texas, was obtained for pyrolysis and bio-oil distillation. The wet algae sample was first dried and processed through a batch type pyrolysis reactor, which was also used in other studies
189
[3,16]. The mass recovery and HHV (higher heating value) of bio-oil from pyrolysis was 20% and 38.6 MJ/kg. The bio-oil was then fractionated using a fractional distillation setup as shown in Fig. 1(b). The light (T b 120 °C) and middle (120 °C b T b 200 °C) fractions after fractional distillation were used in this study. The distillation mass recovery was 23% for the light fraction (fraction 1 or F1), 52% for the middle fraction (fraction 2 or F2) and 19% for the heavy fraction (fraction 3 or Fe), while the HHVs were 40.0 and 41.2 MJ/kg, respectively. A study by Nam et al. [8] showed a detailed fractional and vacuum distillation study for the fractionated microalgae bio-oils. 2.2. Experimental setup and procedure A batch type reactor (Micro Robinson-Mahoney, Autoclave Engineers) was used for the current catalytic upgrading study. The volume of the batch was 50 mL, which can handle a maximum pressure to 5000 psi at 343 °C or a maximum temperature to 538 °C as shown in Fig. 1(c). The temperature of the reactor was controlled by an electric heating band with a PID temperature controller. Two K type thermocouples were used; one for the temperature inside the reactor, and the other for the temperature of the electric heater. An internal stirrer was placed inside the batch to help in uniform reactions, and was set at around 300 rpm during the experiment. Each pressurized catalytic experiment was conducted with 7 g of bio-oil with 5% of 5%Pd/C catalyst (Sigma Aldrich, USA). Before the start of the experiment, hydrogen gas was used to flush as much air as possible from inside the reactor. Then the reactor was pressurized with 4.1, 6.2, or 8.3 MPa of hydrogen gas at room temperature. Once the experiment was ready to run, the heater jacket heated the inside reactor to 130, 190, or 250 °C with a stirring of 400 rpm. A constant reaction time was set at 4 h based on Duan and Savage [11], who state that the properties of upgraded bio-oil were not improved much after 4 hour reaction time. At the end of the 4 hour reaction time, the system was shut down and the temperature reduced to room temperature with the help of a water jacket. First, the pressure at room temperature after each experiment was noted to compute the amount of hydrogen gas used. The gas samples were then collected for analysis. Last, the upgraded bio-oil was collected after filtering out the catalyst using a pre-weighted Buchner glass filter. Then the remaining catalyst and bio-oil in the reactor were washed out thoroughly using acetone as solvent. Both the Buchner glass filter and washed bio-oil were then dried in a 60 °C oven to completely remove the acetone solvent. The weight of the dried catalyst in the glass filter after solvent wash was considered as coke. The solvent washed bio-oil was indicated as tar. The complete mass balance was made with upgraded bio-oil, coke, tar, and produced gas + loss. 2.3. Analytical methods The pyrolytic bio-oil and upgraded fractions were analyzed using the following analytical methods. A higher heating value (HHV) was obtained using a PARR bomb calorimeter 6200 in accordance with ASTM D 711. The ultimate analysis (for C, H, N and S) was obtained from a Vario MICRO Elemental analyzer based on the standards of ASTM D5373. The liquid products after upgrading were analyzed according to ASTM E203 for water content and ASTM D974 for total acid number (TAN). The pH of the bio-oil was measured using a digital pH meter. A Shimadzu IRAffinity-1 FTIR spectrophotometer was used to understand the changes of functional groups of the liquid products. The chemical compositions were analyzed using a GC-MS (Shimadzu QP2010Plus, ZB5MS 30 m × 0.25 mm diameter × 0.25 μm thick). The same temperature program of a previous distillation study was used [8]. The gas composition was analyzed using an SRI multiple gas chromatograph (GC) equipped with a TCD and a GC column containing a 6′ molecular sieve and a 6′ silica gel column. The size characterization of raw and used Pd/C catalyst was analyzed using a Quantachrome Instrument
190
H. Nam et al. / Algal Research 24 (2017) 188–198
Fig. 1. Schematic diagrams of (a) a pyrolysis batch reactor for bio-oil production [3], (b) a fractional distillation set-up [8], (c) and a pressure bio-oil upgrading catalytic reactor [15].
(Autosorb-iQ model). First, the samples were degassed for 3 h at 300 °C. The adsorption and desorption isotherms were then measured from the different relative pressures (0.0–1.0) of N2 adsorption under a −196 °C chamber. A surface area (m2/g) was also determined based on a BET model. The reducibility of catalysts was evaluated with temperature programmed reduction (TPR) under H2/Ar (10/90) gas. Initially the sample was flushed and dried at 120 °C for 30 min under nitrogen and cooled to 40 °C. Then the temperature was programmed to 800 °C at 10 °C/min. Thermogravimetric analysis (TGA) was performed using a PerkinElmer Pyris 1 TGA. About 5–6 mg of new, spent and CO2 regenerated catalysts were loaded with a 20 μL alumina crucible and heated from 25 °C to 900 °C at 10 °C/min under 20 mL/min of nitrogen. Then the averaged DTG curves from duplicates were obtained to understand the degree of volatiles over the temperature. 2.4. Experimental design and data analysis A CCD (central composite design) as a part of the RSM (response surface methodology) was used for the current study to understand the statistical correlations of temperature and pressure on the product yields and properties. The total number of runs in the CCD was 12 for catalytic upgrading of the middle fractions (denoted as F2) of bio-oil at conditions of temperature and pressure shown in Table 1. Three more runs were made for comparison between the direct bio-oil upgrading and the light fraction (denoted as F1) upgrading. F2 was collected from a fractional distillation ranging from 120 to 200 °C when F1 was also from distillation below 120 °C. The operating conditions in accordance with the CCD of the RSM were decided after preliminary
For comparison
Run #
Type of bio-oil
Temp. (°C)
Pressure (MPa)
1 2 3 4 5⁎ 6⁎ 7⁎ 8⁎
F2 F2 F2 F2 F2 F2 F2 F2 F2 F2 F2 F2 Bio-oil F1 F1
130 130 130 190 190 190 190 190 190 250 250 250 260 130 190
4.1 6.2 8.3 4.1 6.2 6.2 6.2 6.2 8.3 4.1 6.2 8.3 4.1 6.2 6.2
9 10 11 12 13 14 15
⁎ Center points of the CCD (central composite design).
Product yield ð%Þ ¼ Mass of product ðgÞ=ðInitial bio−oil ðgÞ þ Consumed H2 ðgÞÞ
ð1Þ
Mass of Product: weight of each bio-oil, coke and tar product (g) h i DOD ð%Þ ¼ oxygenfeed ð%Þ−oxygenproduct ð%Þ =oxygenfeed ð%Þ 100
ð2Þ
Hydrogen consumption ¼ ðPi =Ti Þ– X H2 ; f P f =T f Vgas =R
ð3Þ
Vgas: reactor batch volume occupying gas Pi or f: initial/final reactor pressure Ti or f: initial/final reactor temperature at 293 K XH2,f: hydrogen mole fraction in a produced gas at final stage R: gas constant (8.314 J/K·mol) TOF ðturn over frequencyÞ; moles=gcatalyst h ¼ ðPi =Ti Þ− X H2; f P f =T f ð4Þ Vgas =ðR wt of catalyst reaction timeÞ
Table 1 Experimental designs of algae bio-oil fuel catalytic upgrading.
RSM (CCD)
upgrading tests. The coded levels of 130, 190, and 250 °C for temperature and 4.1, 6.2, and 8.3 MPa for pressure were set as −1, 0, and + 1 respectively. Statistical analysis through ANOVA (analysis variance) was used to understand the fitness of the regression model at a 95% significance level. The experimental errors on the data at the center point of the CCD were obtained as a standard deviation. Each product yield of upgraded bio-oil, coke, tar, and loss + gas was obtained according to Eq. (1). The degree of deoxygenation (DOD) in the upgraded bio-oil was obtained by a different elemental composition according to Eq. (2). Hydrogen consumption during the catalytic upgrading was calculated based on Eq. (3), while turn over frequency (TOF) showing the amount of convertible initial bio-oil with a Pd/C catalyst before deactivation was obtained from Eq. (4).
3. Results and discussion 3.1. Product yields Microalgae bio-oil fractions were upgraded using a Pd/C catalyst to determine the effects of temperature and pressure on the catalyst activity in product yields and properties. Fig. 2 identifies the product yields of upgraded bio-oil, coke, tar, and loss/gas at each upgrading operating condition. The product yield of the upgraded middle fraction (F2 or fraction 2) microalgae bio-oil resulted in around 68% at 130 °C, 71% at 190 °C, and 50% at 250 °C. The statistical analysis indicated that the effect of temperature was more influential than pressure on the product yield. The best regression model for the product yield was the quadratic regression model with a p-value of 0.003. Two significant terms were
H. Nam et al. / Algal Research 24 (2017) 188–198
191
Fig. 2. Mass product yield after hydrotreatment with different fractions at various conditions.
the temperature terms of A (p-value = 0.0015) and A2 (p-value = 0.0012). Even though the pressure term on the yield is insignificant, the pressure (B) may affect the upgraded bio-oil yield when the pressure difference is large enough according to the coded equation (Eq. (5)). Similarly, the significant operating condition for the loss and gas yield was only the temperature terms of A (p-value = 0.0053) and A2 (p-value = 0.0033). The gas and loss yields at 250 °C were from 24 to 36% due to massive thermal and hydrocracking reactions. In contrast, the significant terms of the coke formation were mainly dependent on temperature (A, p-value = 0.013), pressure (B2, p-value = 0.037), and the interaction (AB, p-value = 0.008). This indicated that less coke was formed at a higher temperature and higher pressure condition, which can also be confirmed by the coded quadratic coefficients from Eq. (6). Li et al. [17,18] also observed that the total coke yield decreased to 2% at a higher hydrotreating temperature by converting the heavy chemicals to light one, which reduced the possibility of coke formation.
Upgraded liquid yield ¼ þ71−7:2A−2B−2AB−11:3A2 þ 1:3B2
ð5Þ
Coke formation ¼ þ6:04−0:5A þ 0:33B þ 0:75AB−0:12A2 –0:62B2 ð6Þ
Pd/C catalytic upgrading with raw bio-oil and a light fraction (F1 or fraction1) of bio-oil was carried out for comparison. The yield of the F1 upgrades (42–53%) was lower compared to the yield of F2 upgrades at the same operating conditions. As lighter hydrocarbon chemicals were included in the F1 bio-oil, many light chemicals were thermally cracked and converted into gaseous products through the catalytic reactions, resulting an increase in loss + gas yield (41–52%). Consequently, the tar content (1%) from the F1 upgrading was the lowest among the upgrades. On the other hand, the lowest product yield of raw bio-oil upgrading was obtained at 27%. The large production of coke and tar were mainly from a large amount of heavy chemicals in the bio-oil (mainly composed of paraffin waxes) [8]. 3.2. Physicochemical characteristics Microalgae pyrolytic bio-oil was distilled into three fractions (F1, F2, and F3) through a fractional distillation examined in a previous study [8]. The physical appearances of a fraction and its catalytic upgrades are illustrated in Fig. 3. The translucent dark-red color of fraction 2 was changed into translucent colors of dark-yellow and light-brown under the less severe conditions of 130 °C and 4.1 MPa. A cloudy darkgreen and dark-brown were obtained at more severe conditions of 190 °C and 250 °C. On the other hand, the physical appearance of light
Fig. 3. Pd/C catalytic upgraded microalgae bio-oil fractions.
192
H. Nam et al. / Algal Research 24 (2017) 188–198
fraction upgrades showed a translucent light-green color even after catalytic treatments. Physicochemical characteristics of produced bio-oil can be evaluated to understand the feasibility of bio-oils for fuel applications. Properties normally include TAN (total acid number), MC (moisture content), and HHV (higher heating value) as indicated in Table 2. Besides the properties in the current study, the standards of ASTM D6751 (biodiesel) and D7544 (pyrolysis biofuel) require a separate list of characteristics: kinematic viscosity, pH, density, ash content, cloud point, flash point, cetane number, oxidation stability, distillation temperature, etc. TAN of a bio-oil indicates the total acidity in bio-oil chemicals and is measured by the amount of basic chemical required to neutralize the given acidic bio-oil sample. The reduction in TAN is a very important factor if bio-oil is to be used in petroleum refineries. The biodiesel B100 standard of ASTM D6751 requires that the TAN be b 0.5 mg KOH/g. It is known that carboxylic acids, sugars, and extractives in a pyrolytic bio-oil increase the TAN level. From a previous study [8], the microalgae pyrolytic bio-oil showed 12.2 mg KOH/g, which increased to 17.2 mg KOH/g in a middle fraction (F2) and decreased to 0.12 mg KOH/g in a light fraction (F1). After Pd/C catalytic upgrades, the TAN of all samples were reduced. The light fraction (F1) upgrades were shown to be as low as b0.1 mg KOH/g from 0.12 mg KOH/g of the F1 distillate, while the middle fraction (F2) at 17.2 mg KOH/g was reduced to 1.09 mg KOH/g at 250 °C and 8.3 MPa. The raw bio-oil at 12.2 mg KOH/g was reduced to 2.44 mg KOH/g at 260 °C and 4.1 MPa. The improved TANs were lower than the TANs after catalytic upgrades in previous studies of the corn stover upgrades (5.3–20.8 mg KOH/g) and loblolly pine bio-oil upgrades (53–74 mg KOH/g) [19,20]. The high carboxylic acid composition of corn stover and loblolly pine biooil led to a relatively higher acid number. A quadratic model showed a significance on the TANs at various conditions (p-value b 0.000). Only the temperature terms of A (p-value b 0.000) and A2 (p-value = 0.001) were significant when the interaction term of temperature and pressure (AB) was close to significance (p-value = 0.053). The MC (moisture content) of pyrolysis bio-oil should be b30% in accordance with ASTM D7544, and cannot be directly used for a fuel substitute considering the moisture content from petro-fuel standards is b0.05% in B100 biodiesel, b 0.1% in gasoline, and b0.5% in crude oil [21, 22]. The microalgae bio-oil produced from slow pyrolysis contained a lower moisture content of 8.2% compared to other agricultural biomass bio-oils from jatropha, rice straw, and corn stover (10.2–20.3%) and the woody biomass of loblolly pine (21–32%) [3,4,20,23]. Two reasons for
Fraction 1a Upgrades B100 Gasoline Crude oil Fraction 3a a b
Conditions
TAN (mg KOH/g)
Water content (wt%)
HHV (MJ/kg)
Pyrolysis 260–4.1 Distillation 130–4.1 130–6.2 130–8.3 190–4.1 190–6.2b 190–8.3 250–4.1 250–6.2 250–8.3 Distillation 130–6.2 190–6.2 ASTM D6751 [22] [21] Distillation
12.2 2.44 17.2 9.35 9.64 12.6 1.86 3.18 ± 0.45 3.02 1.39 1.28 1.09 0.12 b0.1 b0.1 b0.5 – – –
8.21 2.16 1.25 0.65 0.92 1.14 1.09 0.82 ± 0.04 0.75 0.95 0.69 0.67 2.18 0.38 0.32 b0.05 b0.1 b0.5 –
38.6 40.6 41.2 42.5 41.5 40.0 41.9 42.0 ± 0.2 42.4 40.6 41.7 42.9 40.0 39.9 41.1 N/A 43–47 41–43 40.6
Imported from Nam et al. [8]. Average value of the center points.
3.3. Chemical compositions of upgraded fractions The chemical compositions of upgraded bio-oils were analyzed to better understand the mechanism of the catalytic upgrading process. Elemental compositions were initially obtained to observe the direct differences in elemental changes as shown in Table 3. The elemental changes of F2 upgrades at lower severity conditions of 130 and 190 °C were not as significant as were the changes in raw bio-oil and F1 upgrades. It can be inferred from this that the larger the difference in temperature conditions for upgrading and distillation, the more variation in elemental properties were obtained. Thus, only the F2 upgrades at 250 °C showed a relatively large change from F2 distillates. The 72% carbon and 9.7% hydrogen from raw bio-oil increased to a maximum of 78.7% carbon and 11.5% hydrogen for F2 upgrading at 250 °C and 8.3 MPa, Table 3 Elemental compositions of catalytic upgrades.
Table 2 Properties of catalytic upgrades of microalgae bio-oil.
Raw bio-oila Upgrades Fraction 2a Upgrades
the lower MC compared to lignocellulosic biomass bio-oil were a lower oxygen content and a higher amount of extractives in the dried microalgae [5,8]. After fractionation, the high MCs were reduced to 1.3–2.2% by forming an aqueous layer in a light fraction [8], which was comparable to the MCs of 1.8–3.9% of raw microalgae bio-oil upgrading by Monet [15]. In contrast, a catalytic upgrading with F1 and F2 bio-oils in the current study resulted in an additional reduction in the MC to as low as 0.32% through intensive hydrogenation and hydrodeoxygenation. An increase in pressure at 130 °C increased the MC of bio-oil, while the MC decreased at 250 °C. No significant statistical models for the moisture content were identified from ANOVA. HHV (higher heating value) is a representative fuel property defining the improvement of bio-oil. The degree of HHV enhancement from F2 distillates to the catalytic upgrading was lower than that from raw bio-oil (38.6 MJ/kg) to distillates (40–41 MJ/kg) as shown in Table 2. F2 distillate upgrades showed 40–43 MJ/kg when F1 upgrades were ranged from 40 to 41 MJ/kg. The ANOVA statistical analysis showed a significance with a quadratic model (p-value = 0.001) with the significant terms of temperature (A2, p-value = 0.01) and interaction term (AB, p-value = 0.000). The upgraded HHVs in the current study showed similar ranges compared to other upgrading studies. The distilled bio-oil showed 40.4 MJ/kg for Spirulina sp. and 41.2 MJ/kg for Nannochlropsis o. [8,9]. HHV of other microalgae upgrading studies indicated 37–40 MJ/kg for Nannochlropsis o. with zeolite catalytic upgrading [15], 42–43 MJ/kg for Nannochlropsis sp. with supercritical water upgrading [10], and 39.6–41.9 MJ/kg for Scenedesmus a. and 40.9–43.2 MJ/kg for Nannochloropsis g. with Pt/Al2O3 and HZSM-5 [12].
Raw bio-oila Upgrades Fraction 2a Upgrades
Conditions
C
H
N
S
O
H/C
Pyrolysis
72.2
9.7
6.0
0.220
11.9
1.61 0.123
260–4.1 78.1 Distillation 75.5
10.3 10.4
5.9 5.7
0.04 0.110
5.7 8.3
1.58 0.055 1.65 0.082
130–4.1 130–6.2 130–8.3 190–4.1 190–6.2b
10.9 10.7 10.4 11.2 11.4 ± 0.01 11.3 10.8 11.1 11.5 9.9
5.7 5.8 6.2 5.4 5.2 ± 0.2 5.3 5.8 5.4 4.6 4.6
0.135 0.215 0.146 0.154 0.12 ± 0.05 0.128 0.056 0.067 0.072 0.360
5.9 7.2 7.8 4.3 5.2 ± 0.6 5.7 5.4 4.6 5.1 16.1
1.69 1.69 1.66 1.71 1.76
0.057 0.071 0.078 0.041 0.050
1.75 1.67 1.68 1.76 1.73
0.055 0.052 0.044 0.049 0.175
11.4 11.7 8.4
4.9 4.6 5.6
0.12 0.14 0.05
6.4 5.2 6.3
1.77 0.062 1.75 0.050 1.27 0.06
77.4 76.1 75.4 78.9 78.1 ± 0.5 190–8.3 77.6 250–4.1 78.0 250–6.2 78.9 250–8.3 78.7 Distillation 69.0
Fraction 1a Upgrades 130–6.2 77.3 190–6.2 78.3 Distillation 79.7 Fraction a 3 a b
Imported from Nam et al. [8]. Average value of the center points.
O/C
H. Nam et al. / Algal Research 24 (2017) 188–198
and 78.3% C and 11.7% for F1 upgrading at 190 °C–6.2 MPa. When it comes to nitrogen content, there was no substantial reduction through the catalytic upgrading with two operating conditions (temperature from 130 °C to 250 °C and pressure from 4.1 to 8.3 MPa). The nitrogen and sulfur contents in bio-oil were also considered as important elements for the prevention of NOx and SOx contaminants during combustion reactions. The nitrogen content of F2 in the current study slightly decreased at increasing temperature even though it was not a substantial reduction. Also, F1 distillate showed a lower level of nitrogen content because most nitrogen and oxygen accumulated in the heavy asphaltene fractions determined by Biller et al. [24]. The best nitrogen regression model of 2FI (two factor interaction) was still not significant with a p-value of 0.1213. However, it indicated the effect of pressure and the interaction term showed significant with a 90% confidence level when temperature term was insignificant. Still, higher H2 pressures (13.8–15 MPa) for catalytic upgrading might not be a major condition for nitrogen removal according to previous studies [24]. The sulfur content in raw bio-oils was originally relatively low compared to the sulfur content (0.05–6%) in crude oil [21]. In addition, the sulfur content fell to below half or one-third of the initial sulfur content after catalytic upgrading. The reduction in sulfur content during F2 upgrading was only possible at the highest condition of severity, 250 °C. For the raw bio-oil catalytic upgrade, the elemental compositions are similar to those obtained in a heavy distillation fraction (C: 79.7%, N: 5.6%, and O: 6.3%) from a previous study [8] except for the hydrogen content (H: 8.4%), indicating active hydrogenation reactions occurred during the Pd/C upgrading. A Van Krevelen diagram was constructed with the H/C and O/C ratios for better determination of the elemental transitions through pyrolysis of algae, distillation and catalytic upgrading as shown in Fig. 4. Except for the direct raw bio-oil upgrading, the H/C range of F1 and F2 upgrades is similar to that of petroleum based fuels (1.65–1.80). The H/C ratio after F1 distillate upgrading showed the highest at around 1.77, while the O/C ranged between 0.04 and 0.08. The direction toward an increase in H/C and a decrease in O/C indicates an intensive decarboxylation or decarbonylation, while a decrease in H/C and a decrease in O/C points a dehydration reaction. A better upgrading performance was recognized from distilled bio-oil upgrading compared to raw biooil upgrading as a lower hydrogen content was obtained in the direct upgraded bio-oil. Similarly, the hydrogen contents decreased with raw bio-oil HZSM-5 upgrades from a previous study [15]. This can be inferred from the existence of heavy fraction. From the ANOVA analysis, both the H/C and O/C ratios showed significance on the quadratic model with a p-value = 0.013. The significance of terms on each ratio was different. The H/C ratio was significantly affected by temperature
193
(A2, p-value = 0.006). The interaction factors (AB, p-value = 0.089) and pressure term (B, p-value = 0.090) also showed significance with a 90% confidence level. In the case of the O/C ratio, only the temperature terms of A (p-value = 0.003) and A2 (p-value = 0.040) showed significant, while the interaction term (AB, p-value = 0.082) was significant with a 90% confidence level. From the statistical significance of the elemental transitions, both temperature and pressure need to be considered for a hydrogen increase, while the temperature term mainly influences oxygen removal. Fig. 5 shows the GC-MS chromatograms for the middle fraction and its Pd/C catalytic upgrades at different temperatures (130, 190, and 250 °C) and pressure (4.1 MPa). As the catalytic temperature increased, more and higher peaks in the upgrades were identified at a shorter retention time, indicating lower carbon number chemicals through hydrodeoxygenaton, hydrocracking and hydrogenantion. The upgrades at the shorter retention time were composed mainly of aromatic hydrocarbons and some new heteroatom containing chemicals (pyrrole and nitrile). Similar heights of relative peaks between olefin/alcohol and alkane groups in fraction 2 changed differently in upgrades. The saturated alkane group peaks were much higher at the highest temperature condition, for example, by converting1-undecene and 1-dodecanol into undecane and dodecane, through hydrogenation and hyrodeoxygenation. Also, active hydrocracking can be expected from the chemical transitions of nitrile groups as newly produced light nitrile peaks were found at the shorter retention time through a higher catalytic temperature, while some heavy nitriles substantially decreased. Categorized chemical groups from GC-MS were compared as shown in Fig. 6. The substantial hydrogenation conversion for olefin (alkene) and decarbonylation conversion for ketone led to an increase in saturated hydrocarbon paraffin (alkane) and naphthene (cycloalkane) groups. The higher temperature and pressure helped to increase the production of naphthene and aromatic and decrease the chemicals of unsaturated olefins and oxygenated ketone. There was a minimal effect of temperature at 130 °C on the conversions, which was a different trend compared to switchgrass bio-oil upgrading at 110–120 °C with a nickel catalyst [25]. The effect of pressure on the chemical compositions was substantial, especially at the temperature of 250 °C, considering an increase in the naphthene and aromatic hydrocarbons and a decrease in nitriles. Still, many nitrile compounds were present even after catalytic upgrading, which can be explained with a similar reaction between fatty acids and produced ammonia from other nitrile chemicals through denitrogenation [10]. Overall the relative chemical analysis of middle fraction upgrades showed over 70% paraffin, close to 10% naphthene and aromatics, and b 5% olefin, nitrile and ketone. Fig. S1 also shows the chemical group composition of the upgraded light fraction. A similar
Fig. 4. Van Krevelen diagram of upgraded distilled bio-oils.
194
H. Nam et al. / Algal Research 24 (2017) 188–198
Fig. 5. GC-MS chromatogram of fraction 2 and its upgrades at different temperatures.
trend of most chemical group transitions after catalytic upgrading were observed even though the aromatics in light fraction upgrades were not changed much compared to the original light fraction. The chemical transition through Pd/C catalytic upgrading was also identified using FTIR analysis as shown in Fig. 7. Only F2 samples with different temperatures were selected for the analysis. The adjacent aromatic hydrogen peak at 748 cm−1 showed an increased relative height in peaks as the catalytic temperature increased, which supports an increase of aromatics in upgrades. The peaks between 1350 and 1460 cm−1 for aliphatic hydrocarbons of CH3 and CH2 indicated an increase in a paraffin chemical group in upgrades. The double bonding oxygenate peak of C_O stretching ranging from 1650 to 1750 cm− 1 showed a decreased relative peak height due to the deoxygenated chemicals in upgrades. The alkene peaks of C_C stretching vibration
at 1630 to 1660 cm− 1 decreased at a higher temperature catalytic condition. 3.4. Gas analysis The gas products of produced hydrocarbon gases and unreacted hydrogen were collected after each catalytic upgrading as shown in Fig. 8. As the temperature increased at the same pressure, more CO, CH4, C2H6, and C3H8 gases were produced, indicating thermal and catalytic cracking. Similar results were reported by Li and Savage [26] that the light saturated hydrocarbons (C1–C4) increased at a higher temperature. Accordingly, the hydrogen consumption (Eq. (3)) increased as the temperature condition increased. On the other hand, the increased pressure at the constant temperature led to a reduction in gas yields
Fig. 6. Chemical composition functional groups of middle fraction (F2) catalytic upgrades.
H. Nam et al. / Algal Research 24 (2017) 188–198
195
Fig. 7. FTIR of upgraded bio-oils and Fraction 2 of pyrolytic algae bio-oil.
because the addition of a higher activation energy on the reactions helped converting the thermally cracked hydrocarbons into heavier saturated hydrocarbon chemicals. Consequently, hydrogen consumption increased during the cracking under higher pressure. The lower amount of gas product from direct bio-oil upgrading was due to a large amount of tar and coke formation in the reactor. Unlike other lignocellulosic based pyrolytic bio-oil upgrading, CO2 gas was rarely produced with the algae bio-oil. Similarly, Li and Savage [26] also experienced the absence of CO2 from the algae bio-oil upgrading with an HZSM-5 catalyst, while CO2 was obtained from their previous algae bio-oil upgrading in a supercritical water study [10]. Also, the absence of water in microalgae bio-oil prevented the possible reactions of a water-gas shift reaction (CO + H2O ⇌ CO2 + H2) and steam reforming (CH4 + H2O ⇌ CO + 3H2), which minimized CO2 gas production. At a high temperature of 250 °C, more hydrogen was produced in the gas compared to the consumed hydrogen gas, which can be from the in situ catalytic hydrogen production. Similarly, an increase in hydrogen content during upgrading was constructed with an increase in the amount of catalyst [11].
3.5. Hydrogen consumption and turnover frequency (TOF) The turnover frequency was obtained from Eq. (4), which represents the hydrogen consumption per amount of catalyst per reaction time. In other words, the degree of catalytic activity on bio-oil conversion is understandable during the upgrading. Table 4 clearly shows that the changes in TOF values are followed by hydrogen consumption. An increase in temperature and pressure conditions led to a tendency of increased TOF, which indicated higher catalytic activity. Both the temperature (p-value = 0.003) and pressure (p-value = 0.01) conditions significantly affected the TOF. A similar trend of activity on catalytic reactions at higher temperatures was also reported in a corn stover bio-oil with a Ru/C catalyst [19]. In contrast, the direct bio-oil upgrading showed a relatively low TOF (0.009 mol/g catalyst-h) even at a high temperature (260 °C) due to the presence of the heavy fraction, which, unlike F1 and F2 distillates, hinder the catalytic reactions. From the close TOF values of F1 and F2 at the same conditions, similar catalytic activity during the upgrading can be inferred.
Fig. 8. Produced gas composition during catalytic reactions of F2 and raw bio-oil upgrading.
196
H. Nam et al. / Algal Research 24 (2017) 188–198
Table 4 Hydrogen consumption and turnover frequency.
260–4.1 130–4.1 130–6.2 130–8.3 190–4.1 190–6.2* 190–8.3 250–4.1 250–6.2 250–8.3 130–6.2 190–8.3 *
Table 5 BET surface area and pore volume of Pd/C catalyst before and after use.
Initial fuel
H2 mmol/g bio-oil
TOF mol/g catalyst-h
Raw F2 F2 F2 F2 F2 F2 F2 F2 F2 F1 F1
2.8 0.6 0.8 2.8 3.4 4.3 9.7 4.6 6.1 7.9 1.3 8.2
0.009 0.004 0.005 0.019 0.023 0.029 0.065 0.032 0.041 0.052 0.008 0.055
Average value of the center points.
3.6. Catalyst characteristics The isotherm characteristics of new and used Pd/C were determined for understanding the degree of catalyst deactivation after the hydrotreatment of microalgae bio-oil as shown in Fig. 9(a). According to the IUPAC guidelines [27], the new Pd/C catalyst is close to Type I (a), shows a steep curve below 0.15 relative pressure (P/Po), and reaches a plateau, which indicate the presence of micropores. On the other hand, the used catalyst is similar to Type I (b) due to the round knee below, indicating the presence of wider micropores. Both the catalysts obtained Type H4 hysteresis (slit-shaped pores) even though the used catalyst shows a wider hysteresis loop. The appearance of hysteresis loops in the multilayer range of isotherms is associated with capillary condensation. Adsorption of chemicals and formation of cokes during the upgrading possibly helped the formation of capillary condensation on the catalyst pores. The ultimate analysis of new and used catalysts at 190 °C–6.2 MPa showed an increase in hydrogen (2.3%), nitrogen (1.52%), and sulfur (0.36%) contents of the used catalyst compared to the new catalyst (1.5% H, 0.19% N, and 0.07% S). Consequently, the surface area and the pore volume were reduced as indicated in Table 5. The least surface area of 447 m2/g was obtained at 250 °C upgrading from the new Pd/C catalyst of 1306 m2/g due to the deactivation causes including chemical/thermal degradation, fouling, coking and poisoning [28]. H2-TPR was performed to understand the reducibility of catalyst. New Pd/C initially showed a sharp negative peak at 62 °C, which indicated desorption of H2 on the surface of metallic Pd. Under
New Pd/C Used Pd/C at 130 °C Used Pd/C at 190 °C Used Pd/C at 250 °C
Surface area (m2/g)
Pore volume (cm3/g)
1306.3 700.4 539.6 447.8
0.638 0.383 0.330 0.268
hydrogen gas at below 100 °C, PdO and PdCl2 are easily reduced to metallic Pd and desorbed to PdHx as reported by others [29,30]. The second peak at 570 °C is corresponded to the reduction of the oxygen containing functional groups in carbon support. The used Pd/C catalyst showed no significant peaks as it reduced during the catalytic upgrading. Further regeneration stage would make the spent Pd/C catalyst available for more catalytic bio-oil upgrading. General regeneration methods for carbon based catalyst include thermal regeneration (700–1000 °C) under inert gas or oxidizing gas [31], solvent extractive regeneration, and reactive (acid-base or oxidative transformation) regeneration [32]. In the current study, a simple CO2 regeneration of spent catalyst was performed at 800 °C for an hour in a fixed bed reactor. The regenerated catalyst was evaluated by comparing with the used catalysts using a TGADTG as shown in Fig. 10. One peak from 70 °C to 320 °C from used catalyst was considered as tar or heavy fraction of bio-oil, which was not fully washed with solvent. The second peak rose from 360 °C due to the decomposition of carbon. Similar trend of catalyst weight loss was also reported with Ni, Pt, and Ru catalyst due to the waxy or intermediate compounds produced during hydrotreatment [33]. On the other hand, the CO2 regenerated catalyst does not show any peak under 400 °C, while the trend of weight reduction above 400 °C was similar. 3.7. Evaluation of mass, energy and carbon flows The Sankey diagram is used to evaluate the conversion flows from microalgae to bio-oil, to distilled bio-oil, and to catalytic upgrades as shown in Fig. 11. The pyrolytic bio-oil mass yield (21%) occupies half (50%) the energy of initial microalgae through pyrolysis due to enhance the HHV of bio-oil. The distillated fraction 1 and 2 (a potential fuel substitutes) showed 16% of mass yield and 40% of energy recovery over initial microalgae raw materials. After catalytic upgrading, the mass and energy yield of fraction 1 and 2 were 10.1% and 25.5%, respectively from 100% raw microalgae sample. Considering the mass and energy yields of direct bio-oil upgrades showed 5.7% and 14.2%, respectively,
Fig. 9. New and used Pd/C catalysts of (a) adsorption and desorption curves and (b) H2-TPR profiles.
H. Nam et al. / Algal Research 24 (2017) 188–198
Fig. 10. New and spent catalysts analysis for derivative thermogravimetric (DTG).
Fig. 11. Sankey diagram of mass, energy and carbon percent from dried algae to upgrades.
197
198
H. Nam et al. / Algal Research 24 (2017) 188–198
a higher mass and energy recovery can be achieved with an extra step of distillation as well as better physicochemical properties. The carbon distribution was also obtained, which shows a similar trend of energy flow. Similarly, the carbon recovery of direct bio-oil upgrades shows 15.2%, which is less than that of fraction 1 and 2 (27.3%). 4. Conclusion A catalytic upgrading of distilled microalgae fractions was performed using a noble metal catalyst Pd/C based on the response surface methodology (RSM). The upgrading conditions were temperatures from 130 to 250 °C and pressures from 4.1 to 8.3 MPa. The light fraction and raw pyrolysis bio-oil were also upgraded for comparison. Better upgraded bio-oil was obtained from distilled fraction catalytic upgrading compared to direct catalytic upgrading. Both the oxygen and hydrogen content of F1 and F2 upgrades were improved through deoxygenation, hydrogenation, and hydrocracking. HHV and TAN values were also improved accordingly. In contrast, the raw bio-oil upgrade showed a limited improvement, which showed a decrease in the H/C ratio. Temperature was a main influence on the improvement of bio-oil properties followed by pressure and the interaction of both. The ketones in upgrades were almost all removed at the catalytic upgrading condition above 190 °C, while nitrogen containing chemicals decreased slightly. Paraffin and aromatic chemical groups were significantly produced when the olefins decreased. The produced gas mainly included CO, CH4, C2H6, C3H8, and a trace of CO2, which showed a substantial catalytic conversion of bio-oil. The turnover frequency with hydrogen consumption indicated similar catalytic activity of F1 and F2 distillation upgrading at the same conditions, while low catalytic activity was determined in raw bio-oil upgrading. The spent catalyst showed a decrease in surface area and volume with an increase in nitrogen and sulfur contents. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.algal.2017.03.021. Acknowledgment The authors acknowledge the facilities, and the scientific and financial assistance of the Bio Energy Testing and Analysis Laboratory (BETA lab) and AgriLife Research at Texas A&M University. The work was also supported by the dissertation fellowship from OGAPS (Office of Graduate and Professional Studies) and undergraduate research scholarship from Honors & Undergraduate Research, Texas A&M University. References [1] L. Dufossé, P. Galaup, A. Yaron, S.M. Arad, P. Blanc, K.N.C. Murthy, G.A. Ravishankar, Microorganisms and microalgae as sources of pigments for food use: a scientific oddity or an industrial reality? Trends Food Sci. Technol. 16 (2005) 389–406. [2] Y. Chisti, Biodiesel from microalgae, Biotechnol. Adv. 25 (2007) 294–306. [3] H. Nam, S.C. Capareda, N. Ashwath, J. Kongkasawan, Experimental investigation of pyrolysis of rice straw using bench-scale auger, batch and fluidized bed reactors, Energy 93 (Part 2) (2015) 2384–2394. [4] J. Kongkasawan, H. Nam, S.C. Capareda, Jatropha waste meal as an alternative energy source via pressurized pyrolysis: a study on temperature effects, Energy 113 (2016) 631–642. [5] M.C.C. Maguyon, S.C. Capareda, Evaluating the effects of temperature on pressurized pyrolysis of Nannochloropsis oculata based on products yields and characteristics, Energy Convers. Manag. 76 (2013) 764–773.
[6] R. Shakya, J. Whelen, S. Adhikari, R. Mahadevan, S. Neupane, Effect of temperature and Na2CO3 catalyst on hydrothermal liquefaction of algae, Algal Res. 12 (2015) 80–90. [7] X. Bai, P. Duan, Y. Xu, A. Zhang, P.E. Savage, Hydrothermal catalytic processing of pretreated algal oil: a catalyst screening study, Fuel 120 (2014) 141–149. [8] H. Nam, J. Choi, S.C. Capareda, Comparative study of vacuum and fractional distillation using pyrolytic microalgae (Nannochloropsis oculata) bio-oil, Algal Res. 17 (2016) 87–96. [9] B.E. Eboibi, D.M. Lewis, P.J. Ashman, S. Chinnasamy, Hydrothermal liquefaction of microalgae for biocrude production: improving the biocrude properties with vacuum distillation, Bioresour. Technol. 174 (2014) 212–221. [10] P. Duan, P.E. Savage, Upgrading of crude algal bio-oil in supercritical water, Bioresour. Technol. 102 (2011) 1899–1906. [11] P. Duan, P.E. Savage, Catalytic hydrotreatment of crude algal bio-oil in supercritical water, Appl. Catal. B Environ. 104 (2011) 136–143. [12] D.L. Barreiro, B.R. Gómez, F. Ronsse, U. Hornung, A. Kruse, W. Prins, Heterogeneous catalytic upgrading of biocrude oil produced by hydrothermal liquefaction of microalgae: state of the art and own experiments, Fuel Process. Technol. 148 (2016) 117–127. [13] L. Zhou, A. Lawal, Hydrodeoxygenation of microalgae oil to green diesel over Pt, Rh and presulfided NiMo catalysts, Cat. Sci. Technol. 6 (2016) 1442–1454. [14] L. Zhou, A. Lawal, Evaluation of presulfided NiMo/γ-Al2O3 for hydrodeoxygenation of microalgae oil to produce green diesel, Energy Fuel 29 (2014) 262–272. [15] M. Maguyon, Technical Feasibility Study on Biofuels Production from Pyrolysis of Nannochloropsis oculata and Algal Bio-oil Upgrading, 2013. [16] H. Nam, S. Capareda, Experimental investigation of torrefaction of two agricultural wastes of different composition using RSM (response surface methodology), Energy 91 (2015) 507–516. [17] X. Li, R. Gunawan, Y. Wang, W. Chaiwat, X. Hu, M. Gholizadeh, D. Mourant, J. Bromly, C. Li, Upgrading of bio-oil into advanced biofuels and chemicals. Part III. Changes in aromatic structure and coke forming propensity during the catalytic hydrotreatment of a fast pyrolysis bio-oil with Pd/C catalyst, Fuel 116 (2014) 642–649. [18] W. Chaiwat, R. Gunawan, M. Gholizadeh, X. Li, C. Lievens, X. Hu, Y. Wang, D. Mourant, A. Rossiter, J. Bromly, C. Li, Upgrading of bio-oil into advanced biofuels and chemicals. Part II. Importance of holdup of heavy species during the hydrotreatment of bio-oil in a continuous packed-bed catalytic reactor, Fuel 112 (2013) 302–310. [19] J.A. Capunitan, S.C. Capareda, Hydrotreatment of corn stover bio-oil using noble metal catalysts, Fuel Process. Technol. 125 (2014) 190–199. [20] Y. Luo, V.K. Guda, E.B. Hassan, P.H. Steele, B. Mitchell, F. Yu, Hydrodeoxygenation of oxidized distilled bio-oil for the production of gasoline fuel type, Energy Convers. Manag. 112 (2016) 319–327. [21] J.G. Speight, The Chemistry and Technology of Petroleum, CRC Press, 2014. [22] Oak Ridge National Laboratory, Transportation Energy Data Book - Chapter 6 Alternative Fuel and Advanced Technology Vehicles and Characteristics, 2015. [23] J.A. Capunitan, S.C. Capareda, Assessing the potential for biofuel production of corn stover pyrolysis using a pressurized batch reactor, Fuel 95 (2012) 563–572. [24] P. Biller, B.K. Sharma, B. Kunwar, A.B. Ross, Hydroprocessing of bio-crude from continuous hydrothermal liquefaction of microalgae, Fuel 159 (2015) 197–205. [25] T. Imam, S. Capareda, Characterization of bio-oil, syn-gas and bio-char from switchgrass pyrolysis at various temperatures, J. Anal. Appl. Pyrolysis 93 (2012) 170–177. [26] Z. Li, P.E. Savage, Feedstocks for fuels and chemicals from algae: treatment of crude bio-oil over HZSM-5, Algal Res. 2 (2013) 154–163. [27] J. Rouquerol, F. Rouquerol, P. Llewellyn, G. Maurin, K.S. Sing, Adsorption by Powders and Porous Solids: Principles, Methodology and Applications, Academic Press, 2013. [28] M.D. Argyle, C.H. Bartholomew, Heterogeneous catalyst deactivation and regeneration: a review, Catalysts 5 (2015) 145–269. [29] P. Sangeetha, K. Shanthi, K.S.R. Rao, B. Viswanathan, P. Selvam, Hydrogenation of nitrobenzene over palladium-supported catalysts—effect of support, Appl. Catal. A Gen. 353 (2009) 160–165. [30] C. Tu, S. Cheng, Ceria-modified palladium/activated carbon as a high-performance catalyst for crude caprolactam hydrogenation purification, ACS Sustain. Chem. Eng. 2 (2014) 629–636. [31] H.F. Abbas, W.W. Daud, Thermocatalytic decomposition of methane for hydrogen production using activated carbon catalyst: regeneration and characterization studies, Int. J. Hydrog. Energy 34 (2009) 8034–8045. [32] M. Sheintuch, Y.I. Matatov-Meytal, Comparison of catalytic processes with other regeneration methods of activated carbon, Catal. Today 53 (1999) 73–80. [33] Z. Wang, S. Adhikari, P. Valdez, R. Shakya, C. Laird, Upgrading of hydrothermal liquefaction biocrude from algae grown in municipal wastewater, Fuel Process. Technol. 142 (2016) 147–156.