- Email: [email protected]
One-step microalgal biodiesel production from Chlorella pyrenoidosa using subcritical methanol extraction (SCM) technology
Biomass and Bioenergy 120 (2019) 265–272
Contents lists available at ScienceDirect
Biomass and Bioenergy journal homepage: www.elsevier.com/locate/biombioe
Research paper
One-step microalgal biodiesel production from Chlorella pyrenoidosa using subcritical methanol extraction (SCM) technology
T
Selvakumar Thiruvenkadam, Shamsul Izhar, Yoshida Hiroyuki, Razif Harun∗ Department of Chemical and Environmental Engineering, Universiti Putra Malaysia, Serdang, 43400, Malaysia
A R T I C LE I N FO
A B S T R A C T
Keywords: Low-lipid algae Reaction temperature Reaction time Methanol to algae ratio Central composite design Response surface methodology
In this work, we propose a one-step subcritical methanol extraction (SCM) process for biodiesel production from Chlorella pyrenoidosa. Therefore, the present study attempts to establish and determine the optimum operating conditions for maximum biodiesel yield from SCM of C. pyrenoidosa. A statistical approach, i.e. response surface methodology is employed in this study. The effects of three operational factors: reaction temperature (140–220 °C), reaction time (1–15 min) and methanol to algae ratio (1–9 wt.%) were investigated using a central composite design. A maximum yield of crude biodiesel of 7.1 wt.% was obtained at 160 °C, 3 min reaction time and 7 wt.% methanol to algae ratio. The analysis of variance revealed that methanol to algae ratio is the most significant factor for maximizing biodiesel yield. Regression analysis showed a good fit of the experimental data to the second-order polynomial model. With no requirement of catalyst nor any pretreatment step, SCM process is economically feasible to scale up the commercial biodiesel production from algae.
1. Introduction In recent years, there has been an increasing interest in algal biofuels. Algae are a unique biomass feedstock for sustainable production of biofuels. Algae, one of the fastest growing photosynthetic organisms on earth, have biomass productivity rates higher than terrestrial plants [1]. They have several advantages including tolerance to extreme environmental conditions, eco-friendly cultivation process, simple life cycle and resource availability for large-scale production [2]. Additional benefits of algae over food crops include fast growth rates, less water intake, adaptation to various water sources (fresh, seawater, saline or brackish, and wastewater), high photosynthetic efficiency, carbon dioxide (CO2) bio-sequestration, phytoremediation, inexpensive cultivation techniques using non-arable land and short harvesting periods. Notwithstanding these benefits, algal biofuel development faces a few drawbacks which include low biomass densities and high operating costs for biomass generation and conversion [3]. Numerous methods for the extraction of biochemical products from microalgae have been applied; but most common methods are expeller or oil press, solvent extraction, supercritical fluid extraction (SFE) and ultrasound techniques [4,5]. The oil from algae is usually extracted with organic solvent and then converted into biodiesel using a catalyst. The energy intake during drying and solvent extraction processes contributes to two-thirds of the total energy consumption of the entire
∗
process [6]. The cost of these extraction methods for commercial applications could be high. In view of the economic and environmental needs, it is desirable to explore the use of subcritical water as a fundamental research in recovering valuable materials from microalgal biomass. Many reports have documented the efficiency of subcritical water extraction (SCW) for the extraction of intercellular constituents of algae using water as a medium and also, a hydrolyzing agent [7–10]. The subcritical method is simpler, more environmentally friendly and can reach high extraction in a very short time. Depending on the operational conditions (temperature, residence time, particle size, moisture, and reactor configuration), SCW can cause diverse effects over the product yield and its quality. The bioactive products obtained are a valuable source of materials for the chemical, pharmaceutical, food and energy industries. Moreover, the use of hydrothermal processing in aquatic biomass (macro- and micro-algae) has been shown to be an interesting technology to produce bio-crude oil. The conventional method for biodiesel production from algae is a two-step process: (i) extraction of lipids from algae and (ii) conversion of lipids to biodiesel. Recent developments in subcritical studies utilize subcritical alcohol solvents as a single step process for extraction of lipids and then conversion of lipids to biodiesel. Thus, the production of in situ algal biodiesel using subcritical alcohol extraction, which circumvents expensive and complicated downstream processing steps, has gained
Corresponding author. E-mail address: [email protected] (R. Harun).
https://doi.org/10.1016/j.biombioe.2018.11.037 Received 30 June 2018; Received in revised form 19 November 2018; Accepted 22 November 2018 0961-9534/ © 2018 Published by Elsevier Ltd.
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
products and then the product mixture was transferred to a centrifuge tube. The reactor was rinsed with distilled water to recover products adhered to the reactor walls. The recovered products were added to the centrifuge tube. The supernatant and solids were separated by using centrifuge (KUBOTA 2420, Tokyo, Japan) operated at 2611 × g for 10 min. The supernatant carrying oil phase was thoroughly mixed with 1.5 ml of hexane and then left for 10 min for settling to separate hexane layer and water layer. The hexane layer containing oil extract was carefully pipetted into a pre-weighed glass bottle. The opened glass bottle was placed in the fume hood to evaporate the hexane from oil extract. The oil extract with the glass bottle was then weighed and the extracted oil amount is reported.
tremendous importance. Subcritical methanol extraction (SCM) has been used widely to produce biodiesel in situ from rice bran [11], rapeseed oil [12], castor oil [13] and soyabean oil [14]. However, few studies have investigated SCM of algal biomass, which includes Schizochytrium limacinum [15], Nannochloropsis gaditana [16], and Chlorella vulgaris [17]. In this paper, the extraction of fatty acid methyl esters (FAMEs) has been demonstrated through subcritical methanol extraction of Chlorella pyrenoidosa. This high-protein low-lipid microalga C. pyrenoidosa used in this study is known for its higher biomass productivity under various salinity conditions. This prospective study was designed to investigate the effect of reaction temperature, reaction time and methanol to algae ratio on the extraction of FAME. This study takes response surface methodology (RSM) approach to optimize the operating conditions for maximum extraction of FAME.
2.5. Characterization of oil The oil sample was transferred to a clean test tube, to which a solution (3 × 2 ml) of hexane:chloroform (4:1 v v−1) was added to extract fatty acid methyl esters (FAME) [20]. The sample was filtered using syringe filter (0.45 μm × 25 mm) prior to gas chromatography analysis. The FAME composition was analyzed by high-resolution Agilent 6890 Series GC system (Agilent Technologies, USA) coupled with a Zebron capillary column (ZB-WAX, 30 m length, 0.25 mm inner diameter, 0.25 μm film thickness). The initial oven temperature was held at 100 °C for 1 min. The temperature then ramped 5 °C min−1 to 230 °C and held for 20 min for a total run time of 47 min. Hydrogen gas flowing at 3 ml min−1 served as the carrier gas with a 2 μl injection volume in split-less mode. The injector temperature was 250 °C and the detector temperature was 260 °C.
2. Methods 2.1. Materials Dried and powdered biomass of the microalgae, Chlorella pyrenoidosa, was purchased from Sunrise Nutrachem Group Co, Ltd (Qingdao, China) and the biomass was stored inside a desiccator until further used. All solvents used in this study were of analytical grade quality and were obtained from Sigma Aldrich, Malaysia. 2.2. Characterization of biomass The samples were initially dried at 105 °C and then incinerated at 550 °C to measure the moisture and ash contents respectively [18]. Thermo-Gravimetric Analyzer (TGA) was used to obtain total volatile matter. Kjeldahl and Soxhlet extraction methods were carried out to determine total protein and total fat contents [18]. The carbohydrate content was calculated by subtracting from the total mass of moisture, ash, protein, and fat. The carbon, hydrogen, nitrogen, and sulfur contents of C. pyrenoidosa biomass were evaluated using a CHNS analyzer (model LECO True Spec CHNS628, USA). The oxygen content was calculated by difference from the total mass of carbon, hydrogen, and nitrogen. The higher heating value (HHV) of the alga was found by adapting Eq. (1) reported by Channiwala and Parikh [19];
2.6. Statistical analysis The purpose of response surface methodology (RSM) was to optimize the independent variables (factors) for an anticipated measured response. Oil yield is the response variable, also referred to as the dependent variable. RSM approach through central composite design (CCD) matrix was used for experimental design and optimization of independent factors. The independent factors studied in this work are the following: reaction temperature (°C), reaction time (min) and methanol to algae ratio (wt.%). The levels of independent factors used in this CCD matrix for optimization of crude biodiesel yield are presented in Table 1. A mathematical model on algal biodiesel production was derived with respect to transesterification conditions. Statistical analyses and model development were done using Design-Expert® 10.0.1 software (Stat-Ease Inc., Minneapolis, USA). The statistical data representing the correlation between independent factors and the response was used to fit the second-order polynomial equation Eq. (2).
HHV (MJ kg −1) = 0.3491C + 1.1783H + 0.1005S – 0.1034O – 0.0151N – 0.0211A
(1)
where C, H, N, S, O and A represent the mass of carbon, hydrogen, nitrogen, sulfur, oxygen, and ash, on a dry weight basis. 2.3. Subcritical methanol apparatus and procedure
Y = β0 + β1 A + β2 B + β3 C + β12 AB + β13 AC + β23 BC + β11 A2 + β22 B2
Custom-built stainless-steel reactors assembled from commercially available components (Swagelok Company, Japan) was used for SCM experiments. Methanol was added to the powdered C. pyrenoidosa cells to produce wet algae slurry of designated methanol to algae ratio according to the experimental runs. The total volume of the reactor was 35 ml while 70% of the reactor volume was filled with algae:solvent mixture. Argon gas was flushed into the headspace to remove residual air. The loaded reactor was sealed tightly and then placed into a preheated salt bath (180–350 °C; Thomas Kogaku Co Ltd.) or oil bath (100–180 °C; Thomas Kogaku Co Ltd.). After a desired period of time, the reactor was removed from the reaction bath and then submerged in cold water bath to stop the reaction. The experimental setup of the extraction process is illustrated in Fig. 1.
+ β33 C 2
(2)
where β0 is the model constant coefficient; β1, β2, β3 are linear coefficients; β12, β13, β23 are interaction coefficients; β11, β22, β33 are quadratic coefficients.
3. Result and discussion 3.1. C. pyrenoidosa characterization Table 2 presents the proximate, ultimate and biochemical analysis of C. pyrenoidosa. The microalgae were characterized with 22.8% carbohydrates, 62.7% proteins, 1.4% lipids and 5.6% moisture. In this work, C. pyrenoidosa cells were mixed with methanol according to the assigned quantities before being subjected to SCM extraction. The higher heating value of the algae was found to be 18.06 MJ kg−1.
2.4. Product separation and recovery The cooled reactor was opened inside a fume hood to release gas 266
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Fig. 1. Schematic diagram of the experimental apparatus, (a) salt bath, (b) batch reactor. Table 1 Levels of independent factors used for optimization. Independent factors
Levels
A: Reaction temperature (°C) B: Reaction time (min) C: Methanol to algae ratio (wt.%)
140 1 1
160 3 3
Table 3 Response values of the crude biodiesel yield for given levels of variables (reaction temperature, reaction time and methanol to algae ratio) in response surface methodology. 180 5 5
200 10 7
220 15 9
Run Number
Type
Table 2 Proximate, ultimate and biochemical analysis of C. pyrenoidosa. Characteristics
Value
Proximate (wt.%)
Moisture Ash
5.60 7.50
Biochemical (wt.%)
Organic content Carbohydrate Protein Lipid
86.90 22.80 62.70 1.40
Ultimate (wt.%)
Carbon Hydrogen Oxygen Nitrogen Sulfur
44.53 5.71 38.87a 9.80 1.09
HHV (MJ kg−1)
18.06
a
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
O (wt.%) = 100 – (C + H + N) (wt.%); HHV: higher heating value.
267
Factorial Axial Center Axial Factorial Factorial Factorial Center Axial Factorial Center Axial Center Center Center Factorial Factorial Axial Factorial Axial
Independent factors
Dependent variable
Reaction temperature, A (°C)
Reaction time, B (min)
Methanol to algae ratio, C (wt.%)
Crude biodiesel Yield, Y (kg kg−1 algae)
160 180 180 140 200 200 200 180 180 160 180 180 180 180 180 200 160 180 160 220
10 1 5 5 10 3 3 5 15 3 5 5 5 5 5 10 3 5 10 5
3 5 5 5 7 7 3 5 5 3 5 9 5 5 5 3 7 1 7 5
0.04177 0.04340 0.04529 0.06374 0.04595 0.04851 0.04590 0.04683 0.04264 0.04124 0.04684 0.06367 0.04725 0.04671 0.05116 0.05108 0.07094 0.05104 0.05430 0.05272
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Table 4 Selection of a suitable model for SCM system (Fit Summary). Source
Sum of squares
Sequential model sum of squares Mean 0.050 Linear 4.006E-004 2FI 3.499E-004 Quadratic 3.334E-004 Cubic 7.815E-005 Residual 1.971E-005 Total 0.051 Lack of fit tests Linear 7.615E-004 2FI 4.115E-004 Quadratic 7.815E-005 Cubic 0.000 Pure Error 1.971E-004 Model Summary statistics Source S.D. Linear 6.987E-003 2FI 5.760E-003 Quadratic 3.128E-003 Cubic 1.985E-003
d.f.
Mean square
1 3 3 3 5 5 20
0.050 1.335E-004 1.166E-004 1.111E-004 1.563E-005 3.942E-006 2.564E-003
11 8 5 0 5
6.922E-005 5.144E-005 1.563E-005
F
P>F
Remark
2.74 3.52 11.36 3.97
0.0779 0.0461 0.0015 0.0784
Suggested Aliased
17.56 13.05 3.97
0.0027 0.0058 0.0784
Pre. R2 −0.1410 0.2113 0.2646
PRESS 1.348E-003 9.321E-004 8.691E-004
Suggested Aliased
3.942E-006
2
R 0.3390 0.6351 0.9172 0.9833
Adj. R2 0.2151 0.4667 0.8427 0.9366
Remark
Suggested Aliased
Table 5 ANOVA of the regression model for the prediction of crude biodiesel yield. Source
Coefficient estimate
Model A B C A2 B2 C2 AB AC BC Residual Lack of fit Pure error Cor total Adeq. Prec.
−1.934E-003 −2.873E-003 4.910E-003 0.011 −2.513E-003 9.818E-003 9.091E-003 −0.022 −9.989E-003
d.f. 9 1 1 1 1 1 1 1 1 1 10 5 5 19
Standard error
2.073E-003 1.506E-003 2.073E-003 2.502E-003 2.699E-003 2.502E-003 4.234E-003 4.424E-003 4.234E-003
Sum of squares
Mean square
F-value
P-value
Remark
1.084E-003 8.518E-006 3.562E-005 5.489E-005 1.788E-004 8.487E-006 1.507E-004 4.511E-005 2.503E-004 5.448E-005 9.786E-006 7.815E-005 1.971E-006 1.182E-003
1.204E-004 8.518E-006 3.562E-005 5.489E-005 1.788E-004 8.487E-006 1.507E-004 4.511E-005 2.503E-004 5.448E-005 9.786E-006 1.563E-005 3.942E-006
12.31 0.87 3.64 5.61 18.27 0.87 15.40 4.61 25.58 5.57
0.0003 0.3728 0.0855 0.0394 0.0016 0.3737 0.0028 0.0573 0.0005 0.0400
Significant
3.97
0.0784
Not significant
12.258
weight ratio ranged between 1 and 9. The objective of response surface methodology is to quantify the interactions among the independent factors and the measured response. The central composite design matrix with 20 experimental runs of independent factors and response variable is presented in Table 3. The sequential model sum of squares, lack of fit tests and model summary statistics are examined to compare the sequential model fitting methods. The fit summary display (Table 4) compared four polynomial models, namely linear, interactive (2FI), quadratic and cubic models. From Table 4, it is evident that the quadratic model is the statistically suitable model for crude biodiesel yield by SCM and, thus, it has been utilized for further statistical studies. An ANOVA was performed to quantify the influence of independent factors and its interactions on the measured response. The ANOVA of the response surface quadratic model is given in Table 5 and it depicts the significance of the regression model test, individual model coefficient test and lack of fit test. The quadratic model F-value of 12.31 specifies the model is significant. “P > F” values lesser than 0.05 reflects the model terms are significant. Values greater than 0.1000 implies the model terms are not significant indicating these terms don't have a significant effect on the response. In this study, C, A2, C2 and interaction terms of AC and BC were significant model terms. The differences between the observed and predicted values can be credited to systematic error. Hence, the lack of fit test is used to differentiate the residual error and pure error from statistically design points. In this
Table 6 Major compounds in biocrude oil from C. pyrenoidosa at 160 °C, 3 min and 7 wt.% of methanol to algae ratio. Fatty acid methyl esters (FAMEs)
FAME content (%)
Methyl Methyl Methyl Methyl Methyl Methyl Methyl Methyl Methyl
1.03 2.19 31.24 5.14 1.85 0.94 13.08 7.30 2.08
myristate (C14:0) myristoleate (C14:1) palmitate (C16:0) palmitoleate (C16:1) stearate (C18:0) vaccenate (C18:1) linoleate (C18:2) linolenate (C18:3) docosahexaenoate (C22:6)
Saturated fatty acids (%) Monounsaturated fatty acids (%) Polyunsaturated fatty acids (%)
34.13 8.27 22.47
3.2. Response surface analysis based on central composite design The influence factors, which affect the crude biodiesel yield by SCM, are reaction temperature, reaction time and methanol to algae weight ratio. In order to examine the combined effects of these factors, experiments were carried out for different combinations of these independent factors, where reaction temperature varied from 140 to 220 °C, reaction time varied from 1 to 15 min and methanol to algae 268
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Here, Y is the response (crude biodiesel yield); A, B, and C denotes reaction temperature, reaction time and methanol to water weight ratio, respectively. The positive and negative values of the regression coefficients reflect the relation of the response to the independent variable. Positive values denote the synergistic effect, whereas negative values denote the antagonistic effect. The crude biodiesel yield was mainly influenced by methanol to algae ratio, both in linear and quadratic terms. The maximum biodiesel yield of 7.1 wt.% resulted from SCM operational conditions: 160 °C reaction temperature with 3 min reaction time and 7 wt.% of methanol to algae ratio. GC analysis of this biodiesel sample showed eight fatty acid methyl esters (FAMEs) as presented in Table 6. 3.3. Effect of process variables on biodiesel yield 3.3.1. Effect of temperature The biodiesel extraction of C. pyrenoidosa was carried out at different reaction temperatures ranging from 140 to 220 °C. The maximum biodiesel yield from this study was observed at 160 °C. The curvature in Fig. 3a shows a nonlinear relationship between reaction temperature and biodiesel yield. This nonlinear relationship is further confirmed from the three-dimensional contour plots shown in Fig. 2b and c. The coefficients of reaction temperature from the quadratic model indicated a negative linear and positive quadratic impact on the biodiesel yield. The interaction plots of reaction temperature with reaction time (Fig. 4) clearly shows that the biodiesel yield decreases with an increase in both factors. Reactions at low temperature are seen to have a steep decline in biodiesel yield with an increase in reaction time, compared to reactions at high temperature. It can also be seen that reaction temperature exhibits a positive slope with biodiesel yield at lower methanol to algae ratio. Contrary, a negative slope is seen for reactions at higher methanol to algae ratio. The increase in biodiesel yields at higher reaction temperatures for lower methanol to algae ratio was attributed to the increase in the diffusion coefficient between methanol and extracted biodiesel [11]. It is evident that, for reactions with higher methanol to algae ratio, a decrease in biodiesel yield is exhibited with an increase in temperature corresponding to lower quantity of algae loaded into SCM reactors. 3.3.2. Effect of reaction time The extraction experiments were performed for durations ranging from 1 to 15 min. Fig. 2a and b shows the effect of reaction time with methanol to algae ratio and reaction temperature, respectively. Biodiesel yield increased with an increase in reaction time, primarily for reactions with low methanol to algae ratio. The combined effect of reaction time with reaction temperature (Fig. 4) shows the decrease in biodiesel yield with an increase in both factors. The biodiesel yield increased with reaction time from 1 to 5 min, resulting in the maximum response for biodiesel yield occurred at 5 min (Fig. 3b). With further increase in reaction time, the biodiesel yields subsequently decreased. This outcome is similar to that of Zullaikah et al. [11] who found longer reaction time decreased oil yield due to degradation and polymerization of oil products.
Fig. 2. Response surface plots of biodiesel yield (kg kg−1 algae) at given (a) reaction temperature (°C), (b) reaction time (min) and (c) methanol to algae ratio (wt.%).
study, F-value of 3.97 representing lack of fit is not significant, compared to pure error, is desirable. “P > F” value of 0.0784 for lack of fit justifies the adequacy of the model for the experimental data at the 95% confidence level. Multiple regression analysis is used to correlate the response with independent factors. Based on the experimental data in Table 3, the correlation expressed in a non-linear regression method using a second order polynomial equation Eq. (3);
3.3.3. Effect of methanol to algae weight ratio Among three independent factors tested, methanol to algae ratio had a significant effect (P < 0.05) over biodiesel yield than reaction temperature and reaction time. It can be noted that methanol to algae ratio exhibits a positive linear and positive quadratic effect on biodiesel yield based on the results displayed in Table 5. Fig. 2a and b shows contour plots that explain the effects of methanol to algae ratio with reaction time and reaction temperature, respectively. One-factor plot (Fig. 3c) and interaction plots (Fig. 4) show that the biodiesel yield is directly proportional to low methanol to algae ratio; although biodiesel yield and high methanol to algae ratio are seen to be inversely
Y = 0.046 − 0.002 A − 0.003 B + 0.005 C + 0.009 AB − 0.022 AC − 0.010 BC + 0.011 A2 − 0.003 B2 + 0.010 C 2
(3) 269
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Fig. 3. The effect of (a) reaction temperature, (b) reaction time, and (c) methanol to algae ratio on the biodiesel yield.
By seeking from 87 starting points in response surface changes, the best optimum condition was to be at a reaction temperature of 140 °C, the reaction time of 3 min and methanol to algae loading of 1% corresponding to a maximum biodiesel yield of 7.94 wt.% was predicted (Fig. 5). Experiments were conducted using the aforementioned parameters. The result (7.57 wt.%) obtained was close with the predicted data of the software.
proportional. At low methanol to algae ratio, biodiesel yield increased with the increase in reaction temperature and reaction time. When the methanol to algae ratio is high, the biodiesel yield decreased with the increase in reaction temperature. It can be concluded that lower methanol to algae ratio contributes to higher biodiesel yield, at higher temperatures and longer reaction times, whereas higher methanol to algae ratio contributes to higher biodiesel yield at lower temperatures and shorter reaction times. These results are in accord with previous work of Phong et al. [21], who observed the decrease in FAME yield with an increase in methanol concentration during SCM of Chlorella sp. This is because higher methanol concentrations inhibit the interactions between nonpolar oil and methanol. Also, excess methanol concentration lowers the biodiesel yield due to the extraction of more polar compounds such as carbohydrates and proteins [11]. This differs from the findings observed during subcritical ethanol extraction of C. pyrenoidosa [22] and Nannochloropsis oceanica [23], whereby an increase in ethanol concentration increased oil yield.
4. Conclusion This study investigated the maximum biodiesel yield from SCM of C. pyrenoidosa. The oil is extracted and directly converted to FAME by transesterification with SCM without any catalyst. Optimization of operational factors was carried out using RSM to find the optimum conditions for maximum biodiesel recovery. The data obtained from the CCD matrix shows that the experimental data were fitted to a quadratic polynomial equation. Optimization of response showed that the reaction at 160 °C, 3 min reaction time and 7 wt.% methanol to algae loading provided the maximum biodiesel yield (7.1 wt.%). Among three input factors, methanol to algae ratio exhibited as the most significant factor on maximizing biodiesel yield. In general, SCM is the most promising technology that can be the first step to the fractionation and obtainment of products with high added-value according to the biorefinery concept.
3.4. Process optimization In order to find optimum conditions, we set the desired target for each factor and the response. A minimum reaction temperature, reaction time within range, methanol to algae ratio within range and maximum biodiesel yield, were set for maximum desirability. The importance of each target is set in improving fuel production economics. 270
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Fig. 4. The effects of process parameters interactions on biodiesel yield.
Fig. 5. Desirability ramp for numerical optimization of three selected goals.
271
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Acknowledgments
[11] S. Zullaikah, Y.T. Rahkadima, Y.-H. Ju, A non-catalytic in situ process to produce biodiesel from a rice milling by-product using a subcritical water-methanol mixture, Renew. Energy 111 (2017) 764–770. [12] J. Encinar, A. Pardal, G. Martínez, Transesterification of rapeseed oil in subcritical methanol conditions, Fuel Process. Technol. 94 (1) (2012) 40–46. [13] N. Sánchez, J.M. Encinar, G. Martínez, J.F. González, Biodiesel production from castor oil under subcritical methanol conditions, Int. J. Environ. Sustain Dev. 6 (1) (2015) 61. [14] J.-Z. Yin, Z. Ma, D.-P. Hu, Z.-L. Xiu, T.-H. Wang, Biodiesel production from subcritical methanol transesterification of soybean oil with sodium silicate, Energy Fuels 24 (5) (2010) 3179–3182. [15] Z. Bi, B.B. He, A.G. McDonald, Biodiesel production from green microalgae Schizochytrium limacinum via in situ transesterification, Energy Fuels 29 (8) (2015) 5018–5027. [16] W. Sitthithanaboon, H.K. Reddy, T. Muppaneni, S. Ponnusamy, V. Punsuvon, F. Holguim, B. Dungan, S. Deng, Single-step conversion of wet Nannochloropsis gaditana to biodiesel under subcritical methanol conditions, Fuel 147 (2015) 253–259. [17] Y.A. Tsigie, L.H. Huynh, S. Ismadji, A.M. Engida, Y.-H. Ju, In situ biodiesel production from wet Chlorella vulgaris under subcritical condition, Chem. Eng. J. 213 (2012) 104–108. [18] S. Jongrungruangchok, S. Bunrathep, T. Songsak, Nutrients and minerals content of eleven different samples of Moringa oleifera cultivated in Thailand, J. Health Res. 24 (3) (2010) 123–127. [19] S. Channiwala, P. Parikh, A unified correlation for estimating HHV of solid, liquid and gaseous fuels, Fuel 81 (8) (2002) 1051–1063. [20] T. Lewis, P.D. Nichols, T.A. McMeekin, Evaluation of extraction methods for recovery of fatty acids from lipid-producing microheterotrophs, J. Microbiol. Methods 43 (2) (2000) 107–116. [21] H.Q. Phong, T.D. Au, H.L. Huong, N. Van Dat, Biodiesel synthesis from algae (Chlorella sp.) in condition of subcritical methanol, Tạp chí Khoa học Trường Đại học Cần Thơ 4 (2016) 1–5. [22] X. Peng, X. Ma, Y. Lin, X. Wang, X. Zhang, C. Yang, Effect of process parameters on solvolysis liquefaction of Chlorella pyrenoidosa in ethanol–water system and energy evaluation, Energy Convers. Manag. 117 (2016) 43–53. [23] M.P. Caporgno, J. Pruvost, J. Legrand, O. Lepine, M. Tazerout, C. Bengoa, Hydrothermal liquefaction of Nannochloropsis oceanica in different solvents, Bioresour. Technol. 214 (2016) 404–410.
This work was supported by eScience Fund (5450818) under the Ministry of Science, Technology and Innovation, Malaysia. References [1] L.G. Alba, C. Torri, D. Fabbri, S.R. Kersten, D.W.W. Brilman, Microalgae growth on the aqueous phase from hydrothermal liquefaction of the same microalgae, Chem. Eng. J. 228 (2013) 214–223. [2] A. Parmar, N.K. Singh, A. Pandey, E. Gnansounou, D. Madamwar, Cyanobacteria and microalgae: a positive prospect for biofuels, Bioresour. Technol. 102 (22) (2011) 10163–10172. [3] R. Harun, J.W. Yip, S. Thiruvenkadam, W.A. Ghani, T. Cherrington, M.K. Danquah, Algal biomass conversion to bioethanol–a step‐by‐step assessment, Biotechnol. J. 9 (1) (2014) 73–86. [4] M. Mubarak, A. Shaija, T. Suchithra, A review on the extraction of lipid from microalgae for biodiesel production, Algal Res. 7 (2015) 117–123. [5] R. Halim, B. Gladman, M.K. Danquah, P.A. Webley, Oil extraction from microalgae for biodiesel production, Bioresour. Technol. 102 (1) (2011) 178–185. [6] L. Lardon, A. Helias, B. Sialve, J.-P. Steyer, O. Bernard, Life-cycle Assessment of Biodiesel Production from Microalgae, ACS Publications, 2009. [7] K. Anastasakis, A. Ross, Hydrothermal liquefaction of the brown macro-alga Laminaria saccharina: effect of reaction conditions on product distribution and composition, Bioresour. Technol. 102 (7) (2011) 4876–4883. [8] I. Rodríguez-Meizoso, L. Jaime, S. Santoyo, F. Señoráns, A. Cifuentes, E. Ibáñez, Subcritical water extraction and characterization of bioactive compounds from Haematococcus pluvialis microalga, J. Pharmaceut. Biomed. Anal. 51 (2) (2010) 456–463. [9] H.A. Ruiz, R.M. Rodriguez-Jasso, B.D. Fernandes, A.A. Vicente, J.A. Teixeira, Hydrothermal processing, as an alternative for upgrading agriculture residues and marine biomass according to the biorefinery concept: a review, Renew. Sustain. Energy Rev. 21 (2013) 35–51. [10] D.L. Barreiro, W. Prins, F. Ronsse, W. Brilman, Hydrothermal liquefaction (HTL) of microalgae for biofuel production: state of the art review and future prospects, Biomass Bioenergy 53 (2013) 113–127.
272
Contents lists available at ScienceDirect
Biomass and Bioenergy journal homepage: www.elsevier.com/locate/biombioe
Research paper
One-step microalgal biodiesel production from Chlorella pyrenoidosa using subcritical methanol extraction (SCM) technology
T
Selvakumar Thiruvenkadam, Shamsul Izhar, Yoshida Hiroyuki, Razif Harun∗ Department of Chemical and Environmental Engineering, Universiti Putra Malaysia, Serdang, 43400, Malaysia
A R T I C LE I N FO
A B S T R A C T
Keywords: Low-lipid algae Reaction temperature Reaction time Methanol to algae ratio Central composite design Response surface methodology
In this work, we propose a one-step subcritical methanol extraction (SCM) process for biodiesel production from Chlorella pyrenoidosa. Therefore, the present study attempts to establish and determine the optimum operating conditions for maximum biodiesel yield from SCM of C. pyrenoidosa. A statistical approach, i.e. response surface methodology is employed in this study. The effects of three operational factors: reaction temperature (140–220 °C), reaction time (1–15 min) and methanol to algae ratio (1–9 wt.%) were investigated using a central composite design. A maximum yield of crude biodiesel of 7.1 wt.% was obtained at 160 °C, 3 min reaction time and 7 wt.% methanol to algae ratio. The analysis of variance revealed that methanol to algae ratio is the most significant factor for maximizing biodiesel yield. Regression analysis showed a good fit of the experimental data to the second-order polynomial model. With no requirement of catalyst nor any pretreatment step, SCM process is economically feasible to scale up the commercial biodiesel production from algae.
1. Introduction In recent years, there has been an increasing interest in algal biofuels. Algae are a unique biomass feedstock for sustainable production of biofuels. Algae, one of the fastest growing photosynthetic organisms on earth, have biomass productivity rates higher than terrestrial plants [1]. They have several advantages including tolerance to extreme environmental conditions, eco-friendly cultivation process, simple life cycle and resource availability for large-scale production [2]. Additional benefits of algae over food crops include fast growth rates, less water intake, adaptation to various water sources (fresh, seawater, saline or brackish, and wastewater), high photosynthetic efficiency, carbon dioxide (CO2) bio-sequestration, phytoremediation, inexpensive cultivation techniques using non-arable land and short harvesting periods. Notwithstanding these benefits, algal biofuel development faces a few drawbacks which include low biomass densities and high operating costs for biomass generation and conversion [3]. Numerous methods for the extraction of biochemical products from microalgae have been applied; but most common methods are expeller or oil press, solvent extraction, supercritical fluid extraction (SFE) and ultrasound techniques [4,5]. The oil from algae is usually extracted with organic solvent and then converted into biodiesel using a catalyst. The energy intake during drying and solvent extraction processes contributes to two-thirds of the total energy consumption of the entire
∗
process [6]. The cost of these extraction methods for commercial applications could be high. In view of the economic and environmental needs, it is desirable to explore the use of subcritical water as a fundamental research in recovering valuable materials from microalgal biomass. Many reports have documented the efficiency of subcritical water extraction (SCW) for the extraction of intercellular constituents of algae using water as a medium and also, a hydrolyzing agent [7–10]. The subcritical method is simpler, more environmentally friendly and can reach high extraction in a very short time. Depending on the operational conditions (temperature, residence time, particle size, moisture, and reactor configuration), SCW can cause diverse effects over the product yield and its quality. The bioactive products obtained are a valuable source of materials for the chemical, pharmaceutical, food and energy industries. Moreover, the use of hydrothermal processing in aquatic biomass (macro- and micro-algae) has been shown to be an interesting technology to produce bio-crude oil. The conventional method for biodiesel production from algae is a two-step process: (i) extraction of lipids from algae and (ii) conversion of lipids to biodiesel. Recent developments in subcritical studies utilize subcritical alcohol solvents as a single step process for extraction of lipids and then conversion of lipids to biodiesel. Thus, the production of in situ algal biodiesel using subcritical alcohol extraction, which circumvents expensive and complicated downstream processing steps, has gained
Corresponding author. E-mail address: [email protected] (R. Harun).
https://doi.org/10.1016/j.biombioe.2018.11.037 Received 30 June 2018; Received in revised form 19 November 2018; Accepted 22 November 2018 0961-9534/ © 2018 Published by Elsevier Ltd.
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
products and then the product mixture was transferred to a centrifuge tube. The reactor was rinsed with distilled water to recover products adhered to the reactor walls. The recovered products were added to the centrifuge tube. The supernatant and solids were separated by using centrifuge (KUBOTA 2420, Tokyo, Japan) operated at 2611 × g for 10 min. The supernatant carrying oil phase was thoroughly mixed with 1.5 ml of hexane and then left for 10 min for settling to separate hexane layer and water layer. The hexane layer containing oil extract was carefully pipetted into a pre-weighed glass bottle. The opened glass bottle was placed in the fume hood to evaporate the hexane from oil extract. The oil extract with the glass bottle was then weighed and the extracted oil amount is reported.
tremendous importance. Subcritical methanol extraction (SCM) has been used widely to produce biodiesel in situ from rice bran [11], rapeseed oil [12], castor oil [13] and soyabean oil [14]. However, few studies have investigated SCM of algal biomass, which includes Schizochytrium limacinum [15], Nannochloropsis gaditana [16], and Chlorella vulgaris [17]. In this paper, the extraction of fatty acid methyl esters (FAMEs) has been demonstrated through subcritical methanol extraction of Chlorella pyrenoidosa. This high-protein low-lipid microalga C. pyrenoidosa used in this study is known for its higher biomass productivity under various salinity conditions. This prospective study was designed to investigate the effect of reaction temperature, reaction time and methanol to algae ratio on the extraction of FAME. This study takes response surface methodology (RSM) approach to optimize the operating conditions for maximum extraction of FAME.
2.5. Characterization of oil The oil sample was transferred to a clean test tube, to which a solution (3 × 2 ml) of hexane:chloroform (4:1 v v−1) was added to extract fatty acid methyl esters (FAME) [20]. The sample was filtered using syringe filter (0.45 μm × 25 mm) prior to gas chromatography analysis. The FAME composition was analyzed by high-resolution Agilent 6890 Series GC system (Agilent Technologies, USA) coupled with a Zebron capillary column (ZB-WAX, 30 m length, 0.25 mm inner diameter, 0.25 μm film thickness). The initial oven temperature was held at 100 °C for 1 min. The temperature then ramped 5 °C min−1 to 230 °C and held for 20 min for a total run time of 47 min. Hydrogen gas flowing at 3 ml min−1 served as the carrier gas with a 2 μl injection volume in split-less mode. The injector temperature was 250 °C and the detector temperature was 260 °C.
2. Methods 2.1. Materials Dried and powdered biomass of the microalgae, Chlorella pyrenoidosa, was purchased from Sunrise Nutrachem Group Co, Ltd (Qingdao, China) and the biomass was stored inside a desiccator until further used. All solvents used in this study were of analytical grade quality and were obtained from Sigma Aldrich, Malaysia. 2.2. Characterization of biomass The samples were initially dried at 105 °C and then incinerated at 550 °C to measure the moisture and ash contents respectively [18]. Thermo-Gravimetric Analyzer (TGA) was used to obtain total volatile matter. Kjeldahl and Soxhlet extraction methods were carried out to determine total protein and total fat contents [18]. The carbohydrate content was calculated by subtracting from the total mass of moisture, ash, protein, and fat. The carbon, hydrogen, nitrogen, and sulfur contents of C. pyrenoidosa biomass were evaluated using a CHNS analyzer (model LECO True Spec CHNS628, USA). The oxygen content was calculated by difference from the total mass of carbon, hydrogen, and nitrogen. The higher heating value (HHV) of the alga was found by adapting Eq. (1) reported by Channiwala and Parikh [19];
2.6. Statistical analysis The purpose of response surface methodology (RSM) was to optimize the independent variables (factors) for an anticipated measured response. Oil yield is the response variable, also referred to as the dependent variable. RSM approach through central composite design (CCD) matrix was used for experimental design and optimization of independent factors. The independent factors studied in this work are the following: reaction temperature (°C), reaction time (min) and methanol to algae ratio (wt.%). The levels of independent factors used in this CCD matrix for optimization of crude biodiesel yield are presented in Table 1. A mathematical model on algal biodiesel production was derived with respect to transesterification conditions. Statistical analyses and model development were done using Design-Expert® 10.0.1 software (Stat-Ease Inc., Minneapolis, USA). The statistical data representing the correlation between independent factors and the response was used to fit the second-order polynomial equation Eq. (2).
HHV (MJ kg −1) = 0.3491C + 1.1783H + 0.1005S – 0.1034O – 0.0151N – 0.0211A
(1)
where C, H, N, S, O and A represent the mass of carbon, hydrogen, nitrogen, sulfur, oxygen, and ash, on a dry weight basis. 2.3. Subcritical methanol apparatus and procedure
Y = β0 + β1 A + β2 B + β3 C + β12 AB + β13 AC + β23 BC + β11 A2 + β22 B2
Custom-built stainless-steel reactors assembled from commercially available components (Swagelok Company, Japan) was used for SCM experiments. Methanol was added to the powdered C. pyrenoidosa cells to produce wet algae slurry of designated methanol to algae ratio according to the experimental runs. The total volume of the reactor was 35 ml while 70% of the reactor volume was filled with algae:solvent mixture. Argon gas was flushed into the headspace to remove residual air. The loaded reactor was sealed tightly and then placed into a preheated salt bath (180–350 °C; Thomas Kogaku Co Ltd.) or oil bath (100–180 °C; Thomas Kogaku Co Ltd.). After a desired period of time, the reactor was removed from the reaction bath and then submerged in cold water bath to stop the reaction. The experimental setup of the extraction process is illustrated in Fig. 1.
+ β33 C 2
(2)
where β0 is the model constant coefficient; β1, β2, β3 are linear coefficients; β12, β13, β23 are interaction coefficients; β11, β22, β33 are quadratic coefficients.
3. Result and discussion 3.1. C. pyrenoidosa characterization Table 2 presents the proximate, ultimate and biochemical analysis of C. pyrenoidosa. The microalgae were characterized with 22.8% carbohydrates, 62.7% proteins, 1.4% lipids and 5.6% moisture. In this work, C. pyrenoidosa cells were mixed with methanol according to the assigned quantities before being subjected to SCM extraction. The higher heating value of the algae was found to be 18.06 MJ kg−1.
2.4. Product separation and recovery The cooled reactor was opened inside a fume hood to release gas 266
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Fig. 1. Schematic diagram of the experimental apparatus, (a) salt bath, (b) batch reactor. Table 1 Levels of independent factors used for optimization. Independent factors
Levels
A: Reaction temperature (°C) B: Reaction time (min) C: Methanol to algae ratio (wt.%)
140 1 1
160 3 3
Table 3 Response values of the crude biodiesel yield for given levels of variables (reaction temperature, reaction time and methanol to algae ratio) in response surface methodology. 180 5 5
200 10 7
220 15 9
Run Number
Type
Table 2 Proximate, ultimate and biochemical analysis of C. pyrenoidosa. Characteristics
Value
Proximate (wt.%)
Moisture Ash
5.60 7.50
Biochemical (wt.%)
Organic content Carbohydrate Protein Lipid
86.90 22.80 62.70 1.40
Ultimate (wt.%)
Carbon Hydrogen Oxygen Nitrogen Sulfur
44.53 5.71 38.87a 9.80 1.09
HHV (MJ kg−1)
18.06
a
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
O (wt.%) = 100 – (C + H + N) (wt.%); HHV: higher heating value.
267
Factorial Axial Center Axial Factorial Factorial Factorial Center Axial Factorial Center Axial Center Center Center Factorial Factorial Axial Factorial Axial
Independent factors
Dependent variable
Reaction temperature, A (°C)
Reaction time, B (min)
Methanol to algae ratio, C (wt.%)
Crude biodiesel Yield, Y (kg kg−1 algae)
160 180 180 140 200 200 200 180 180 160 180 180 180 180 180 200 160 180 160 220
10 1 5 5 10 3 3 5 15 3 5 5 5 5 5 10 3 5 10 5
3 5 5 5 7 7 3 5 5 3 5 9 5 5 5 3 7 1 7 5
0.04177 0.04340 0.04529 0.06374 0.04595 0.04851 0.04590 0.04683 0.04264 0.04124 0.04684 0.06367 0.04725 0.04671 0.05116 0.05108 0.07094 0.05104 0.05430 0.05272
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Table 4 Selection of a suitable model for SCM system (Fit Summary). Source
Sum of squares
Sequential model sum of squares Mean 0.050 Linear 4.006E-004 2FI 3.499E-004 Quadratic 3.334E-004 Cubic 7.815E-005 Residual 1.971E-005 Total 0.051 Lack of fit tests Linear 7.615E-004 2FI 4.115E-004 Quadratic 7.815E-005 Cubic 0.000 Pure Error 1.971E-004 Model Summary statistics Source S.D. Linear 6.987E-003 2FI 5.760E-003 Quadratic 3.128E-003 Cubic 1.985E-003
d.f.
Mean square
1 3 3 3 5 5 20
0.050 1.335E-004 1.166E-004 1.111E-004 1.563E-005 3.942E-006 2.564E-003
11 8 5 0 5
6.922E-005 5.144E-005 1.563E-005
F
P>F
Remark
2.74 3.52 11.36 3.97
0.0779 0.0461 0.0015 0.0784
Suggested Aliased
17.56 13.05 3.97
0.0027 0.0058 0.0784
Pre. R2 −0.1410 0.2113 0.2646
PRESS 1.348E-003 9.321E-004 8.691E-004
Suggested Aliased
3.942E-006
2
R 0.3390 0.6351 0.9172 0.9833
Adj. R2 0.2151 0.4667 0.8427 0.9366
Remark
Suggested Aliased
Table 5 ANOVA of the regression model for the prediction of crude biodiesel yield. Source
Coefficient estimate
Model A B C A2 B2 C2 AB AC BC Residual Lack of fit Pure error Cor total Adeq. Prec.
−1.934E-003 −2.873E-003 4.910E-003 0.011 −2.513E-003 9.818E-003 9.091E-003 −0.022 −9.989E-003
d.f. 9 1 1 1 1 1 1 1 1 1 10 5 5 19
Standard error
2.073E-003 1.506E-003 2.073E-003 2.502E-003 2.699E-003 2.502E-003 4.234E-003 4.424E-003 4.234E-003
Sum of squares
Mean square
F-value
P-value
Remark
1.084E-003 8.518E-006 3.562E-005 5.489E-005 1.788E-004 8.487E-006 1.507E-004 4.511E-005 2.503E-004 5.448E-005 9.786E-006 7.815E-005 1.971E-006 1.182E-003
1.204E-004 8.518E-006 3.562E-005 5.489E-005 1.788E-004 8.487E-006 1.507E-004 4.511E-005 2.503E-004 5.448E-005 9.786E-006 1.563E-005 3.942E-006
12.31 0.87 3.64 5.61 18.27 0.87 15.40 4.61 25.58 5.57
0.0003 0.3728 0.0855 0.0394 0.0016 0.3737 0.0028 0.0573 0.0005 0.0400
Significant
3.97
0.0784
Not significant
12.258
weight ratio ranged between 1 and 9. The objective of response surface methodology is to quantify the interactions among the independent factors and the measured response. The central composite design matrix with 20 experimental runs of independent factors and response variable is presented in Table 3. The sequential model sum of squares, lack of fit tests and model summary statistics are examined to compare the sequential model fitting methods. The fit summary display (Table 4) compared four polynomial models, namely linear, interactive (2FI), quadratic and cubic models. From Table 4, it is evident that the quadratic model is the statistically suitable model for crude biodiesel yield by SCM and, thus, it has been utilized for further statistical studies. An ANOVA was performed to quantify the influence of independent factors and its interactions on the measured response. The ANOVA of the response surface quadratic model is given in Table 5 and it depicts the significance of the regression model test, individual model coefficient test and lack of fit test. The quadratic model F-value of 12.31 specifies the model is significant. “P > F” values lesser than 0.05 reflects the model terms are significant. Values greater than 0.1000 implies the model terms are not significant indicating these terms don't have a significant effect on the response. In this study, C, A2, C2 and interaction terms of AC and BC were significant model terms. The differences between the observed and predicted values can be credited to systematic error. Hence, the lack of fit test is used to differentiate the residual error and pure error from statistically design points. In this
Table 6 Major compounds in biocrude oil from C. pyrenoidosa at 160 °C, 3 min and 7 wt.% of methanol to algae ratio. Fatty acid methyl esters (FAMEs)
FAME content (%)
Methyl Methyl Methyl Methyl Methyl Methyl Methyl Methyl Methyl
1.03 2.19 31.24 5.14 1.85 0.94 13.08 7.30 2.08
myristate (C14:0) myristoleate (C14:1) palmitate (C16:0) palmitoleate (C16:1) stearate (C18:0) vaccenate (C18:1) linoleate (C18:2) linolenate (C18:3) docosahexaenoate (C22:6)
Saturated fatty acids (%) Monounsaturated fatty acids (%) Polyunsaturated fatty acids (%)
34.13 8.27 22.47
3.2. Response surface analysis based on central composite design The influence factors, which affect the crude biodiesel yield by SCM, are reaction temperature, reaction time and methanol to algae weight ratio. In order to examine the combined effects of these factors, experiments were carried out for different combinations of these independent factors, where reaction temperature varied from 140 to 220 °C, reaction time varied from 1 to 15 min and methanol to algae 268
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Here, Y is the response (crude biodiesel yield); A, B, and C denotes reaction temperature, reaction time and methanol to water weight ratio, respectively. The positive and negative values of the regression coefficients reflect the relation of the response to the independent variable. Positive values denote the synergistic effect, whereas negative values denote the antagonistic effect. The crude biodiesel yield was mainly influenced by methanol to algae ratio, both in linear and quadratic terms. The maximum biodiesel yield of 7.1 wt.% resulted from SCM operational conditions: 160 °C reaction temperature with 3 min reaction time and 7 wt.% of methanol to algae ratio. GC analysis of this biodiesel sample showed eight fatty acid methyl esters (FAMEs) as presented in Table 6. 3.3. Effect of process variables on biodiesel yield 3.3.1. Effect of temperature The biodiesel extraction of C. pyrenoidosa was carried out at different reaction temperatures ranging from 140 to 220 °C. The maximum biodiesel yield from this study was observed at 160 °C. The curvature in Fig. 3a shows a nonlinear relationship between reaction temperature and biodiesel yield. This nonlinear relationship is further confirmed from the three-dimensional contour plots shown in Fig. 2b and c. The coefficients of reaction temperature from the quadratic model indicated a negative linear and positive quadratic impact on the biodiesel yield. The interaction plots of reaction temperature with reaction time (Fig. 4) clearly shows that the biodiesel yield decreases with an increase in both factors. Reactions at low temperature are seen to have a steep decline in biodiesel yield with an increase in reaction time, compared to reactions at high temperature. It can also be seen that reaction temperature exhibits a positive slope with biodiesel yield at lower methanol to algae ratio. Contrary, a negative slope is seen for reactions at higher methanol to algae ratio. The increase in biodiesel yields at higher reaction temperatures for lower methanol to algae ratio was attributed to the increase in the diffusion coefficient between methanol and extracted biodiesel [11]. It is evident that, for reactions with higher methanol to algae ratio, a decrease in biodiesel yield is exhibited with an increase in temperature corresponding to lower quantity of algae loaded into SCM reactors. 3.3.2. Effect of reaction time The extraction experiments were performed for durations ranging from 1 to 15 min. Fig. 2a and b shows the effect of reaction time with methanol to algae ratio and reaction temperature, respectively. Biodiesel yield increased with an increase in reaction time, primarily for reactions with low methanol to algae ratio. The combined effect of reaction time with reaction temperature (Fig. 4) shows the decrease in biodiesel yield with an increase in both factors. The biodiesel yield increased with reaction time from 1 to 5 min, resulting in the maximum response for biodiesel yield occurred at 5 min (Fig. 3b). With further increase in reaction time, the biodiesel yields subsequently decreased. This outcome is similar to that of Zullaikah et al. [11] who found longer reaction time decreased oil yield due to degradation and polymerization of oil products.
Fig. 2. Response surface plots of biodiesel yield (kg kg−1 algae) at given (a) reaction temperature (°C), (b) reaction time (min) and (c) methanol to algae ratio (wt.%).
study, F-value of 3.97 representing lack of fit is not significant, compared to pure error, is desirable. “P > F” value of 0.0784 for lack of fit justifies the adequacy of the model for the experimental data at the 95% confidence level. Multiple regression analysis is used to correlate the response with independent factors. Based on the experimental data in Table 3, the correlation expressed in a non-linear regression method using a second order polynomial equation Eq. (3);
3.3.3. Effect of methanol to algae weight ratio Among three independent factors tested, methanol to algae ratio had a significant effect (P < 0.05) over biodiesel yield than reaction temperature and reaction time. It can be noted that methanol to algae ratio exhibits a positive linear and positive quadratic effect on biodiesel yield based on the results displayed in Table 5. Fig. 2a and b shows contour plots that explain the effects of methanol to algae ratio with reaction time and reaction temperature, respectively. One-factor plot (Fig. 3c) and interaction plots (Fig. 4) show that the biodiesel yield is directly proportional to low methanol to algae ratio; although biodiesel yield and high methanol to algae ratio are seen to be inversely
Y = 0.046 − 0.002 A − 0.003 B + 0.005 C + 0.009 AB − 0.022 AC − 0.010 BC + 0.011 A2 − 0.003 B2 + 0.010 C 2
(3) 269
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Fig. 3. The effect of (a) reaction temperature, (b) reaction time, and (c) methanol to algae ratio on the biodiesel yield.
By seeking from 87 starting points in response surface changes, the best optimum condition was to be at a reaction temperature of 140 °C, the reaction time of 3 min and methanol to algae loading of 1% corresponding to a maximum biodiesel yield of 7.94 wt.% was predicted (Fig. 5). Experiments were conducted using the aforementioned parameters. The result (7.57 wt.%) obtained was close with the predicted data of the software.
proportional. At low methanol to algae ratio, biodiesel yield increased with the increase in reaction temperature and reaction time. When the methanol to algae ratio is high, the biodiesel yield decreased with the increase in reaction temperature. It can be concluded that lower methanol to algae ratio contributes to higher biodiesel yield, at higher temperatures and longer reaction times, whereas higher methanol to algae ratio contributes to higher biodiesel yield at lower temperatures and shorter reaction times. These results are in accord with previous work of Phong et al. [21], who observed the decrease in FAME yield with an increase in methanol concentration during SCM of Chlorella sp. This is because higher methanol concentrations inhibit the interactions between nonpolar oil and methanol. Also, excess methanol concentration lowers the biodiesel yield due to the extraction of more polar compounds such as carbohydrates and proteins [11]. This differs from the findings observed during subcritical ethanol extraction of C. pyrenoidosa [22] and Nannochloropsis oceanica [23], whereby an increase in ethanol concentration increased oil yield.
4. Conclusion This study investigated the maximum biodiesel yield from SCM of C. pyrenoidosa. The oil is extracted and directly converted to FAME by transesterification with SCM without any catalyst. Optimization of operational factors was carried out using RSM to find the optimum conditions for maximum biodiesel recovery. The data obtained from the CCD matrix shows that the experimental data were fitted to a quadratic polynomial equation. Optimization of response showed that the reaction at 160 °C, 3 min reaction time and 7 wt.% methanol to algae loading provided the maximum biodiesel yield (7.1 wt.%). Among three input factors, methanol to algae ratio exhibited as the most significant factor on maximizing biodiesel yield. In general, SCM is the most promising technology that can be the first step to the fractionation and obtainment of products with high added-value according to the biorefinery concept.
3.4. Process optimization In order to find optimum conditions, we set the desired target for each factor and the response. A minimum reaction temperature, reaction time within range, methanol to algae ratio within range and maximum biodiesel yield, were set for maximum desirability. The importance of each target is set in improving fuel production economics. 270
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Fig. 4. The effects of process parameters interactions on biodiesel yield.
Fig. 5. Desirability ramp for numerical optimization of three selected goals.
271
Biomass and Bioenergy 120 (2019) 265–272
S. Thiruvenkadam et al.
Acknowledgments
[11] S. Zullaikah, Y.T. Rahkadima, Y.-H. Ju, A non-catalytic in situ process to produce biodiesel from a rice milling by-product using a subcritical water-methanol mixture, Renew. Energy 111 (2017) 764–770. [12] J. Encinar, A. Pardal, G. Martínez, Transesterification of rapeseed oil in subcritical methanol conditions, Fuel Process. Technol. 94 (1) (2012) 40–46. [13] N. Sánchez, J.M. Encinar, G. Martínez, J.F. González, Biodiesel production from castor oil under subcritical methanol conditions, Int. J. Environ. Sustain Dev. 6 (1) (2015) 61. [14] J.-Z. Yin, Z. Ma, D.-P. Hu, Z.-L. Xiu, T.-H. Wang, Biodiesel production from subcritical methanol transesterification of soybean oil with sodium silicate, Energy Fuels 24 (5) (2010) 3179–3182. [15] Z. Bi, B.B. He, A.G. McDonald, Biodiesel production from green microalgae Schizochytrium limacinum via in situ transesterification, Energy Fuels 29 (8) (2015) 5018–5027. [16] W. Sitthithanaboon, H.K. Reddy, T. Muppaneni, S. Ponnusamy, V. Punsuvon, F. Holguim, B. Dungan, S. Deng, Single-step conversion of wet Nannochloropsis gaditana to biodiesel under subcritical methanol conditions, Fuel 147 (2015) 253–259. [17] Y.A. Tsigie, L.H. Huynh, S. Ismadji, A.M. Engida, Y.-H. Ju, In situ biodiesel production from wet Chlorella vulgaris under subcritical condition, Chem. Eng. J. 213 (2012) 104–108. [18] S. Jongrungruangchok, S. Bunrathep, T. Songsak, Nutrients and minerals content of eleven different samples of Moringa oleifera cultivated in Thailand, J. Health Res. 24 (3) (2010) 123–127. [19] S. Channiwala, P. Parikh, A unified correlation for estimating HHV of solid, liquid and gaseous fuels, Fuel 81 (8) (2002) 1051–1063. [20] T. Lewis, P.D. Nichols, T.A. McMeekin, Evaluation of extraction methods for recovery of fatty acids from lipid-producing microheterotrophs, J. Microbiol. Methods 43 (2) (2000) 107–116. [21] H.Q. Phong, T.D. Au, H.L. Huong, N. Van Dat, Biodiesel synthesis from algae (Chlorella sp.) in condition of subcritical methanol, Tạp chí Khoa học Trường Đại học Cần Thơ 4 (2016) 1–5. [22] X. Peng, X. Ma, Y. Lin, X. Wang, X. Zhang, C. Yang, Effect of process parameters on solvolysis liquefaction of Chlorella pyrenoidosa in ethanol–water system and energy evaluation, Energy Convers. Manag. 117 (2016) 43–53. [23] M.P. Caporgno, J. Pruvost, J. Legrand, O. Lepine, M. Tazerout, C. Bengoa, Hydrothermal liquefaction of Nannochloropsis oceanica in different solvents, Bioresour. Technol. 214 (2016) 404–410.
This work was supported by eScience Fund (5450818) under the Ministry of Science, Technology and Innovation, Malaysia. References [1] L.G. Alba, C. Torri, D. Fabbri, S.R. Kersten, D.W.W. Brilman, Microalgae growth on the aqueous phase from hydrothermal liquefaction of the same microalgae, Chem. Eng. J. 228 (2013) 214–223. [2] A. Parmar, N.K. Singh, A. Pandey, E. Gnansounou, D. Madamwar, Cyanobacteria and microalgae: a positive prospect for biofuels, Bioresour. Technol. 102 (22) (2011) 10163–10172. [3] R. Harun, J.W. Yip, S. Thiruvenkadam, W.A. Ghani, T. Cherrington, M.K. Danquah, Algal biomass conversion to bioethanol–a step‐by‐step assessment, Biotechnol. J. 9 (1) (2014) 73–86. [4] M. Mubarak, A. Shaija, T. Suchithra, A review on the extraction of lipid from microalgae for biodiesel production, Algal Res. 7 (2015) 117–123. [5] R. Halim, B. Gladman, M.K. Danquah, P.A. Webley, Oil extraction from microalgae for biodiesel production, Bioresour. Technol. 102 (1) (2011) 178–185. [6] L. Lardon, A. Helias, B. Sialve, J.-P. Steyer, O. Bernard, Life-cycle Assessment of Biodiesel Production from Microalgae, ACS Publications, 2009. [7] K. Anastasakis, A. Ross, Hydrothermal liquefaction of the brown macro-alga Laminaria saccharina: effect of reaction conditions on product distribution and composition, Bioresour. Technol. 102 (7) (2011) 4876–4883. [8] I. Rodríguez-Meizoso, L. Jaime, S. Santoyo, F. Señoráns, A. Cifuentes, E. Ibáñez, Subcritical water extraction and characterization of bioactive compounds from Haematococcus pluvialis microalga, J. Pharmaceut. Biomed. Anal. 51 (2) (2010) 456–463. [9] H.A. Ruiz, R.M. Rodriguez-Jasso, B.D. Fernandes, A.A. Vicente, J.A. Teixeira, Hydrothermal processing, as an alternative for upgrading agriculture residues and marine biomass according to the biorefinery concept: a review, Renew. Sustain. Energy Rev. 21 (2013) 35–51. [10] D.L. Barreiro, W. Prins, F. Ronsse, W. Brilman, Hydrothermal liquefaction (HTL) of microalgae for biofuel production: state of the art review and future prospects, Biomass Bioenergy 53 (2013) 113–127.
272