Bioresource Technology 181 (2015) 3239Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior techBiodiesel synthesis by direct transesterification of microalgaBotryococcus braunii with continuous methanol refluxhttp://dx.doi.org/10.1016/j.biortech.2015.01.0470960-8524/ 2015 Elsevier Ltd. All rights reserved.

Corresponding author at: Scientific and Technological Bioresources Nucleus,Universidad de La Frontera, Casilla 54-D, Temuco, Chile. Tel.: +56 45 2325477; fax:+56 45 2732402.

E-mail addresses: p.hidalgo02@ufromail.cl (P. Hidalgo), gustavo.ciudad@ufrontera.cl (G. Ciudad), si.schober@uni-graz.at (S. Schober), martin.mittelbach@uni-graz.at (M. Mittelbach), rodrigo.navia@ufrontera.cl (R. Navia).Pamela Hidalgo a, Gustavo Ciudad a,b, Sigurd Schober c, Martin Mittelbach c, Rodrigo Navia a,b,d,a Scientific and Technological Bioresources Nucleus, Universidad de La Frontera, Casilla 54-D, Temuco, ChilebDepartament of Chemical Engineering, Universidad de La Frontera, Casilla 54-D, Temuco, Chilec Institute of Chemistry, Working Group Chemistry and Technology of Renewable Resources, NAWI Graz, University of Graz, Heinrichstrae 28, A-8010 Graz, AustriadCentre for Biotechnology and Bioengineering (CeBiB), Universidad de La Frontera, Casilla 54-D, Temuco, Chile

h i g h l i g h t s

Reflux extraction reactor promotes high fatty acid methyl esters yield (FAME). One step process combining lipid extraction and lipid conversion into FAME. Biomass is repeatedly contacted with pure solvent enhancing FAME yield. Hexane use as co-solvent enhances FAME yield.a r t i c l e i n f o

Article history:Received 20 November 2014Received in revised form 8 January 2015Accepted 9 January 2015Available online 17 January 2015

Keywords:Botryococcus brauniiDirect transesterificationAcyl acceptorCo-solventHexanea b s t r a c t

Direct transesterification of Botryococcus braunii with continuous acyl acceptor reflux was evaluated. Thismethod combines in one step lipid extraction and esterification/transesterification. Fatty acid methylesters (FAME) synthesis by direct conversion of microalgal biomass was carried out using sulfuric acidas catalyst and methanol as acyl acceptor. In this system, once lipids are extracted, they are contactedwith the catalyst and methanol reaching 82% wt of FAME yield. To optimize the reaction conditions, a fac-torial design using surface response methodology was applied. The effects of catalyst concentration andco-solvent concentration were studied. Hexane was used as co-solvent for increasing lipid extraction per-formance. The incorporation of hexane in the reaction provoked an increase in FAME yield from 82% (puremethanol) to 95% when a 47% v/v of hexane was incorporated in the reaction. However, the selectivitytowards non-saponifiable lipids such as sterols was increased, negatively affecting biodiesel quality.

2015 Elsevier Ltd. All rights reserved.1. Introduction

Recent interest in the production of microalgae is mainlyrelated to the use of microalgal biomass as renewable source forbiofuels and bioproducts development. Microalgae have high lipidcontent which can be used for aquaculture, human nutrition orbiofuel production. A lipid yield between 58,000 L/ha and136,000 L/ha has been estimated. On the other side, oil from oil-seeds such as rapeseed or soybean present oil yields of 1190 L/haand 446 L/ha, respectively (Chen et al., 2011; Halim et al., 2010).The lipid extraction process is one of the most limiting steps forthe development of biodiesel production industry based in micro-algae. In fact, lipid extraction from microalgae is mainly performedby organic solvents and not by conventional physical methods, dueto difficulties in breaking the cell wall, making microalgae-basedbiodiesel production unfeasible at industrial scale (Ehimen et al.,2010; Hidalgo et al., 2013).

Simultaneous lipid extraction and esterification/transesterifica-tion is a technique of great value for biodiesel production frommicroalgae, as it allows extracting and converting fatty acids intofatty acid methyl esters (FAME) in a single step bypassing theuse of large quantities of organic solvents used in lipid extraction(Jin et al., 2014; Wahlen et al., 2011).

This process has been traditionally performed by direct contactbetween the biomass, the catalyst and the acyl-acceptor, where ahigh quantity of acyl acceptor is necessary to promote lipidsdiffusion from inside the cell (Griffiths et al., 2010). In this process,

http://crossmark.crossref.org/dialog/?doi=10.1016/j.biortech.2015.01.047&domain=pdfhttp://dx.doi.org/10.1016/j.biortech.2015.01.047mailto:p.hidalgo02@ufromail.clmailto:gustavo.ciudad@ufrontera.clmailto:gustavo.ciudad@ufrontera.clmailto:si.schober@uni-graz.atmailto:martin.mittelbach@uni-graz.atmailto:martin.mittelbach@uni-graz.atmailto:rodrigo.navia@ufrontera.clhttp://dx.doi.org/10.1016/j.biortech.2015.01.047http://www.sciencedirect.com/science/journal/09608524http://www.elsevier.com/locate/biortech

Table 2Lipids composition of B. braunii microalgae.

Lipid composition Content (%) Content (%)(on the basis of total lipids)

Saponifiable lipidsTriglycerides 1.7 0.3 1Diglycerides 2.5 0.5 1.5

P. Hidalgo et al. / Bioresource Technology 181 (2015) 3239 33lipids pass through a polar lipid bilayer by simple diffusion follow-ing the concentration gradient.

If the biomass is in direct contact with the acyl acceptor, lipidsdiffusion from cells to the acyl acceptor could be limited due to adecreasing concentration gradient in time. Thereby, high acylacceptor volume is necessary to increase lipids extraction. How-ever, a high acyl acceptor volume will decrease the acid strengthof the catalyst and thus decrease the reaction yield (Hidalgoet al., 2014). Additionally, the acid catalyst performs a dual role,as catalyst of the reaction and as a cell wall disruptor agent(Ozgul-Yucel and Turkay, 2002). Therefore, a reduction of acidstrength of the catalyst due to the increase of acyl acceptor volumecould have a negative impact on the total reaction yield.

The traditional configuration used in biomass direct transesteri-fication corresponds to a closed reaction vessel containing thereaction mixture where the acyl acceptor is in direct contact withthe biomass (Haas and Wagner, 2011). In this configuration bothtemperature and agitation are maintained during a specified per-iod of time (Ehimen et al., 2010; El-Shimi et al., 2013; Hidalgoet al., 2013). This system is of easy implementation, but lipidsextraction will be limited due to a decreasing concentration gradi-ent of lipids (outside and inside the cell), requiring a subsequentstep for separating the biomass from the reaction product(Hidalgo et al., 2014). Normally, high acyl acceptor volume forincreasing lipids diffusion and FAME yield has been used in bio-mass transesterification (Ehimen et al., 2010). Acyl acceptor excessplays also a role as extraction solvent, improving the contactbetween catalyst and biomass, altering the permeability of thesolid substrate (Haas et al., 2007). Besides, acyl acceptor excess isresponsible for breaking linkages between glycerin and fatty acidsduring the reaction (Hidalgo et al., 2013). Siler-Marinkovic andTomasevic (1998) used a wide range of methanol:lipid molarratios, ranging between 100:1 and 300:1, for the transesterificationof macerated sunflower seeds (Siler-Marinkovic and Tomasevic,1998). Moreover, Ehimen et al. (2010) used a methanol:lipid molarratio between 105:1 and 524:1 for the direct transesterification ofChlorella microalgae (Ehimen et al., 2010).

Solvent reflux processes have been traditionally carried out forlipid extraction using a continuous or discontinuous (soxhlet)extractor. In this sense, the continuous process could have applica-tion in the production of biodiesel frommicroalgae and until today,only few studies tackle this issue in the literature. In the imple-mentation of this system, microalgal biomass is contacted repeat-edly with fresh portions of the acyl acceptor and in so, lipids can beextracted and esterified in the presence of a high acyl acceptor vol-ume, hence favoring product formation. Although in this system noapplication of shear stress is used to produce microalgae cell walldisruption, highest FAME yields were obtained using this configu-ration, probably enhanced by high lipids extraction yields due todiffusion (Hidalgo et al., 2014). According to this previous study,the use of solvent reflux increased FAME yield in direct transeste-rification of microalgal biomass. A FAME yield close to 80% wasachieved using this configuration in contrast to the traditionalconfiguration in batch reactor (FAME yield of only 60%) wherethe biomass is in direct contact with solvent and catalyst. Althoughbetter results were obtained in this system, only few studiesTable 1Independent variables and levels of the experimental design.

Independent variable Levels

1.47 1 0 1 1.47Co-solvent (% v/v) 34 40 55 70 76

(3.4) (3.1) (2.3) (1.5) (1.2)Catalyst (% wt)* 59.6 75 112.5 150 165.4

* On the basis of total fatty acids. Values in parentheses correspond to thepolarity index (PI) of the mixture evaluated according to (Snyder, 1978).related to the use of solvent reflux for the direct transesterificationof microalgal biomass have been published yet. Therefore itappears interesting to test this technique and its operationalconditions when using Botryococcus braunii as a lipid source formicroalgae-based biodiesel production.

Hence, the aim of this work was to carry out direct transesteri-fication of microalgal biomass in a reflux extraction reactor (RER), adifferent configuration compared to the used techniques wherelipid diffusion is limited. In RER, the microalgae sample is repeat-edly contacted with pure solvent, thus increasing the lipid extrac-tion and conversion yield to FAME. To optimize the operationalconditions in this system, an experimental design using surfaceresponse methodology was developed to find out the influence ofthe operational variables and the interaction among them on FAMEyield. The variables studied were co-solvent and catalyst concen-tration. Hexane was used as co-solvent in the reaction.2. Methods

2.1. Lipid quantification in B. braunii

The microalga used in this study was supplied by DesertBioenergy S.A., Chile. The moisture content was 7.8% wt and sizedistribution was

34 P. Hidalgo et al. / Bioresource Technology 181 (2015) 3239biomass is repeatedly contacted with fresh portions of solvent. Thesystem operates at the solvents boiling temperature for maintain-ing a constant reflux (Luque de Castro and Priego-Capote, 2010). Inthe experiments, the system was coupled to a continuous flowextractor, where the biomass was located. In RER, the extractedlipids were contacted with the catalyst and acyl-acceptor for itsconversion to FAME during 5 h. Methanol was used as acyl-acceptor and sulfuric acid (H2SO4) was used as acid catalyst.

After direct transesterification, methanol was removed by dis-tillation in a rotary evaporator and after that hexane (5 mL) wasadded for FAME extraction while the supernatant was removed.The supernatant was washed three times with distilled water(5 mL) to remove traces of catalyst, methanol and hydrophiliccomponents. Afterwards, the sample was centrifuged and thesupernatant (or non-polar phase) was separated and distilled forits gravimetric quantification.

The solvent free sample was stored at 4 C for FAME determina-tion by GCFID. Heptadecanoic acid methyl ester (C17:0) ofchromatographic purity was used as internal standard for thequantification.

FAME yield was calculated using Eq. (1):

FAMEyield %wt :FAME % wtTFA % wt 100 1

where FAME (% wt) represents the quantification by GCFID of fattyacid methyl esters and TFA corresponds to the gravimetricquantification of transesterifiable lipids.2.2.1. Evaluation of the effect methanol:total fatty acids molar ratio onFAME yield

The evaluation was performed at five levels of methanol:totalfatty acids molar ratios (36:1, 76:1, 151:1, 227:1 and 303:1). Allthe experiments were performed using 75% acid catalyst (on thebasis of total fatty acids). Additionally, the RER system wascompared with a control transesterification system (CTS) whichconsisted of a flask with a condenser and a heating plate withmagnetic stirrer to maintain a homogeneous mixture during the0

20

40

60

80

100FAME Yield in RERFAME yield in CTS

FAM

E Yi

eld

(% w

t)

Methanol:fatty acid molar ratio

A A AB

C

D D

EF

G

38 76 151 227 303:1 :1 :1 :1 :1

Fig. 1. Effect of methanol:total fatty acids molar ratio on FAME yield in the directtransesterification. Reaction conditions: 75% acid catalyst (on the basis of total fattyacids) and reaction time of 5 h. Data represents the mean values of duplicates anderror bars show the standard deviations. The different letters indicate a significantdifference at p < 0.05.reaction. Thereby in CTS, the biomass is directly contacted withthe acyl-acceptor and catalyst.

2.2.2. Spent microalgal biomass characteristics in RERThe removal of total organic carbon (TOC), protein, lipids and

pigments of the biomass was evaluated before and after the reac-tion. The results of RER were compared with the CTS. The experi-ments were carried out using 75% acid catalyst (on basis of totalfatty acids) and the methanol:total fatty acids molar ratio definedin Section 2.2.1.

TOC was evaluated in an organic carbon analyzer (TOC-VCPH,SHIMADZU). Protein content was determined using Kjeldahlmethod (Dupont et al., 2011). Lipid content of biomass after reac-tion was quantified by lipid extraction using methanol:chloroformaccording to the already described methodology. The estimation ofcarotenoids and chlorophylls were evaluated according to the pro-cedure of Lichtenthaler (Lichtenthaler, 1987).

2.2.3. Optimization of direct transesterification in a RERAn experimental design using surface response methodology

was applied to find out the influence of the operational variableson FAME yield during direct transesterification of B. brauniibiomass. The independent variables used in this study were co-solvent and catalyst concentration. Heaxane was selected as co-solvent. Although hexane is less efficient compared to otherorganic solvents in lipid extraction from microalgae (Nagle andLemke, 1990), it has minimal affinity towards polar lipids andhydrocarbons (Grima et al., 1994), and its use in mixture withmethanol may extract a FAME fraction with higher purity, thusdiminishing further purification steps.

The independent variables and levels of the experimentaldesign are shown in Table 1. FAME yield was used as response var-iable. In these experiments, hexane was used as co-solvent in thereaction.

2.3. Chromatographic methods

2.3.1. FAME quantification and distributionAn Agilent Technologies 7890A GC system with FID and a polar

capillary column (J&W 122-7031, 30 m 250 lm 0.15 lm) wereused for FAME identification and quantification. Helium was usedas carrier gas (0.7 mL min1) and the sample was injected (1 lL)with a split ratio (ratio 100:1). The following temperature profilewas used: 60 C for 2 min, then an increase of temperature up to200 C at a rate of 10 C min1, finally a rise until 240 C at a rateof 5 C min1.

2.3.2. Acyl-glycerol quantificationA Hewlett Packard 6890 series GC system with FID and a polar

capillary column (J&W 123-5711, 15 m 320 lm 0.10 lm) wereused for acyl-glycerol analysis. Helium was used as carrier gas(1 mL min1). A temperature gradient of 15 C min1 from 50 to180 C, then 7 C min1 from 180 to 230 C and finally an increaseat a rate of 10 C min1 to 370 C were applied in thisdetermination.

2.3.3. Lipid profileA Hewlett Packard HP 6890 series GC system coupled to a Hew-

lett Packard 5973 mass spectrometry detector was used in theidentification and quantification of lipid profile. A capillary column(DB-5MS UI, 30 m 250 lm 0.25 lm) and helium (1 mL min1)were used. The sample was injected (1 lL) with split injection(ratio 50:1:1). The following temperature program was used:50 C for 3 min and then increasing temperature at a rate of10 C min1 up to 300 C.

0

10

20

30

40

Microalgae Treatment in RER Treatment in CTS

TOC

(%

wt o

n ba

sis

of d

ry b

iom

ass)

(b)A

B

C

0

2

4

6

8

10

12

14

Microalgae Treatment in RER Treatment in CTS

Tota

l fat

ty a

cids

(%

wt o

n ba

sis

of d

ry b

iom

ass)

A

B

C

(a)

0

5

10

15

20

25

30

35

Microalgae Treatment in RER Treatment in CTS

Prot

ein

cont

ent

(% w

t on

basi

s of

dry

bio

mas

s)

A

B

C

(c)

0

2

4

6

8

10

Microalgae Treatment in RER Treatment in CTSPi

gmen

ts

(mg/

g on

bas

is o

f dry

bio

mas

s)

A

B

B

(d)

Fig. 2. Microalgal biomass characterization before and after direct transesterification in RER and CTS. (a) Lipids content (as total fatty acids) (b) total organic carbon (TOC)content (c) protein content and (d) pigments (as carotenoids and chlorophylls). Reaction conditions: 75% acid catalyst (on the basis of total fatty acids), methanol:total fattyacids molar ratio of 151:1, reaction time of 5 h. Data represents the mean values of two samples and the error bars show the standard deviations. The different letters indicatea significant difference at p < 0.05.

P. Hidalgo et al. / Bioresource Technology 181 (2015) 3239 352.4. Statistical analysis

Variance analysis and Tukeys test were performed using JMP-9software (SAS Institute Inc., Cary, NC). A p-value below 0.05 wasconsidered significant.3. Results and discussion

A 26.3% wt of lipid content (of dry microalgae) extracted bymethanol/chloroform mixture using a soxhlet extractor was foundfor B. braunii. From the extracted lipids, a 58.8% corresponded totransesterifiable lipids defined as total fatty acids (TFA). TFA iscomposed by acyl-glycerols, free fatty acid, esters, phospholipidsand glycolipids (Halim et al., 2012). A high content of free fatty acid(46.7% wt on the basis of saponifiable lipids), a low content of acyl-glycerols (1% triglyceride, 1.5% diglyceride, 0.5% monoglyceride onbasis to total lipids) were observed (as shown in Table 2).

The fatty acids profile of B. braunii lipids is mainly composed byoleic acid (C18:1, 54.9%), followed by palmitic acid (C16:0, 12.2%),linolenic acid (C18:3, 5.5%), stearic acid (C18:0, 3.9%) and linoleicacid (C18:2, 3.6%). A high saturated and monounsaturated fattyacids content is suitable to enhance oxidative stability of the sam-ple (Demirbas, 2009). The high C18:1 content in this microalgaehas been already reported, and is related to the biosynthesis of bio-polymers derived from the polymerization of long-chain fatty acidsknown as algaenans (Laureillard et al., 1988).

A high USLs content was also found (41.2% of total extracted lip-ids). In general, high USLs content has been reported in lipids frommicroalgae (Wang and Wang, 2012). USLs are mainly formed by

0

10

20

30

40

50

60

70

0

10

20

30

40

50

0 1 2 3 4 5

FAME yield (% wt)FFA (% wt)Acylglycerides(% wt)

FAM

E yi

eld

(% w

t)

FFA or A

cylglycerides content (%

wt)

Time (h)

(a)

0

20

40

60

80

100

0

10

20

30

40

50

0 1 2 3 4 5

FAME yield (% wt)FFA (% wt)Acylglycerides(% wt)

FAM

E yi

eld

(% w

t)

FFA or A

cylglycerides content (%

wt)

Time (h)

(b)

Fig. 3. FAME yield and reduction of FFA and acyl-glycerides during directtransesterification of microalgal biomass (a) CTS (b) RER. Reaction conditions:75% acid catalyst (on the basis of total fatty acids), methanol:total fatty acids molarratio of 151:1. Data represents the mean values of two samples and the error barsshow the standard deviations.

Table 3Experiment matrix for the factorial design and conversion values (experimental andpredicted).

Run Co-solvent(% v/v) A

Catalyst(% wt) B

FAME yield (% wt)experimental

FAME yield(% wt) predicted

1 55(0) 165.4(1.41) 92.76 90.382 55(0) 112.5(0) 90.42 90.823 55(0) 112.5(0) 93.52 90.824 40(1) 150(1) 92.29 91.905 70(1) 75(1) 61.03 61.376 40(1) 75(1) 93.02 89.227 55(0) 112.5(0) 92.12 90.828 55(0) 112.5(0) 88.52 90.829 70(1) 150(1) 67.70 71.4410 34(1.41) 112.5(0) 85.30 88.2611 55(0) 59.6(1.41) 78.95 81.3912 55(0) 112.5(0) 89.52 90.8213 76(1.41) 112.5(0) 57.10 54.20

Values in parentheses correspond to codified terms.

36 P. Hidalgo et al. / Bioresource Technology 181 (2015) 3239sterols, phytols, fatty alcohols and hydrocarbons (Allard andTemplier, 2000). The major components of USLs found in B. brauniiwere hydrocarbons (13.4% on the basis of total lipids), alcohols(phytols: 10.4% on the basis of total lipids) and sterols (campester-ol and b-sitosterol: 8.1% on the basis of total lipids). B. braunii isknown to producer high amounts of hydrocarbons, triterpenes,sterols among other lipids which cannot be transesterified to FAMEduring the reaction (Hidalgo et al., 2014; Metzger and Largeau,2005).

3.1. Effect of methanol:total fatty acids molar ratio on FAME yield

Different experiments were performed using methanol:totalfatty acids molar ratios ranging between 38 and 303. As shownin Fig. 1, a significant increase of FAME yield up to 82% (48.2% onthe basis of total lipids) was observed for RER compared to CTSand was directly proportional to the increase in the methanol:totalfatty acids molar ratio until 151:1. At higher molar ratios, FAMEyield remained constant. In this configuration, the main lipidextraction mechanism is diffusion, as it involves percolation ofthe solvent through the biomass, thereby allowing the diffusionof internal lipids out of the cell (Hidalgo et al., 2014).

On the opposite, for CTS an increase of the methanol:total fattyacids molar ratio from 38:1 to 227:1 provoked an increment inFAME yield from 17% (10% on the basis of total lipids) to 53% wt(31% on the basis of total lipids). However, an increase in the meth-anol:total fatty acids molar ratio above 227:1 produced a decreasein FAME yield. This reduction could be related to a dilution effect ofthe acid catalyst at high methanol:total fatty acids molar ratios(Velasquez-Orta et al., 2012).

3.2. Spent microalgal biomass characteristics in RER

Higher FAME yields were observed in RER compared to CTS,suggesting a higher lipid extraction yield from biomass. After reac-tion, microalgal biomass diminished its total fatty acids, organiccarbon, proteins and pigments content (as shown in Fig. 2). A hightotal fatty acids extraction yield was observed in RER, as shown inFig. 2a. This result is coincident with the highest FAME yieldobtained in RER. The lesser removal of total organic carbon inRER (Fig. 2b) after direct transesterification can be related to alower extraction yield of microalgae constituents such as protein,pigments and carbohydrates. In fact, methanol and sulfuric acidhave a dual role during direct transesterification. Methanol actsas acyl acceptor as well as a polar nature cell constituents solvent,while sulfuric acid acts as catalyst as well as hydrolyzing agent ofcell constituents, process that is more relevant in CTS configurationas in RER the catalyst is not in contact with the biomass.Hydrophilic components such as pigments and proteins may formhydrogen bonds with methanol used as acyl-acceptor in the reac-tion (Halim et al., 2012; Henriques et al., 2007). Thereby, methanolplays a key role in the extraction of pigments, such as chlorophyllsand carotenoids, as well as in the removal of proteins and polar lip-ids of cell wall. Methanol has been used in protein extractionreplacing conventional membrane proteins extraction techniques(which include the use of detergents, chaotropic agents andorganic acids) that require subsequent sample post-treatment,such as clean-up or pH adjustment (Zhang et al., 2007). From theobtained results, a high decrease of both pigments and proteinswas observed in microalgal biomass after reaction in CTS (Fig. 2cand d). The higher protein removal may be related to the affinityof the acyl acceptor to peptides present in the cell membrane.Besides, the catalyst can also promote side reactions, such as pro-tein hydrolysis (producing peptides), as well as carbohydrates andother cell constituents hydrolysis, producing a higher proteinextraction yield in CTS compared to RER. Acid catalyst is efficient

Table 4Model ANOVA.

Source Coefficient estimatea Sum of squares df Mean square F-value p-Value

Model 1938.70 5 387.74 36.88

40 48

55 63

70

7594

113 131

150

61

70

78

87

95

FAM

E yie

ld (%

wt)

Co-solvent (% v/v) Catalyst (% w t) 40 48 55 63 70

75

94

113

131

150FAME yield (%wt)

Co-solv ent (% v /v )

Cat

alys

t (%

wt)

67.0

72.678.283.8

89.491.8

(a) (b)

Fig. 4. Response surface of FAME yield with respect to catalyst and co-solvent dosage. (a) Two-dimensional contour plot. (b) Three-dimensional surface plot.

38 P. Hidalgo et al. / Bioresource Technology 181 (2015) 3239co-solvent) and 3.7% of the total variations are not explained by themodel. Besides, lack of fit is not significant, thereby this is a suit-able model to describe the effect of the independent variables onFAME yield.

The regression coefficients of the co-solvent (A) and catalyst (B)linear terms and the quadratic term A2 have a significant effect onFAME yield (p-value

P. Hidalgo et al. / Bioresource Technology 181 (2015) 3239 39Halim, R., Danquah, M., Webley, P.A., 2012. Extraction of oil from microalgae forbiodiesel production: a review. Biotechnol. Adv. 30 (3), 709732.

Halim, R., Gladman, B., Danquah, M.K., Webley, P.A., 2010. Oil extraction frommicroalgae for biodiesel production. Bioresour. Technol. 102 (1), 178185.

Henriques, M., Silva, A., Rocha, J., 2007. Extraction and quantification of pigmentsfrom a marine microalga: a simple and reproducible method. In:Communicating Current Research and Educational Topics and Trends inApplied Microbiology. F. Publishers, Spain, pp. 586593.

Hidalgo, P., Toro, C., Ciudad, G., Navia, R., 2013. Advances in directtransesterification of microalgal biomass for biodiesel production. Rev.Environ. Sci. Biotechnol. 12 (2), 179199.

Hidalgo, P., Toro, C., Ciudad, G., Schober, S., Mittelbach, M., Navia, R., 2014.Evaluation of different operational strategies for biodiesel production by directtransesterification of microalgal biomass. Energy Fuels 28, 38143820.

Jin, B., Duan, P., Xu, Y., Wang, B., Wang, F., Zhang, L., 2014. Lewis acid-catalyzed insitu transesterification/esterification of microalgae in supercritical ethanol.Bioresour. Technol. 162, 341349.

Kates, M., Baxter, R., 1962. Lipid composition of mesophilic and psychrophilic yeasts(Candida species) as influenced by environmental temperature. Can. J. Biochem.Physiol. 40 (9), 12131227.

Laureillard, J., Largeau, C., Casadevall, E., 1988. Oleic acid in the biosynthesis of theresistant biopolymers of Botryococcus braunii. Phytochemistry 27 (7), 20952098.

Lichtenthaler, H.K., 1987. Chlorophylls and carotenoids: pigments of photosyntheticbiomembranes. Methods Enzymol. 148, 350382.

Luque de Castro, M.D., Priego-Capote, F., 2010. Soxhlet extraction: past and presentpanacea. J. Chromatogr. A 1217 (16), 23832389.Metzger, P., Largeau, C., 2005. Botryococcus braunii: a rich source for hydrocarbonsand related ether lipids. Appl. Microbiol. Biotechnol. 66 (5), 486496.

Nagle, N., Lemke, P., 1990. Production of methyl ester fuel from microalgae. Appl.Biochem. Biotechnol. 24, 25 (1), 355361.

Ozgul-Yucel, S., Turkay, S., 2002. Variables affecting the yields of methyl estersderived from in situ esterification of rice bran oil. J. Am. Oil Chem. Soc. 79 (6),611614.

Siler-Marinkovic, S., Tomasevic, A., 1998. Transesterification of sunflower oil in situ.Fuel 77 (12), 13891391.

Snyder, L.R., 1978. Classification off the solvent properties of common liquids. J.Chromatogr. Sci. 16 (6), 223234.

Tsugita, A., Scheffler, J.-J., 1982. A rapid method for acid hydrolysis of protein with amixture of trifluoroacetic acid and hydrochloric acid. Eur. J. Biochem. 124 (3),585588.

Velasquez-Orta, S.B., Lee, J.G.M., Harvey, A., 2012. Alkaline in situ transesterificationof Chlorella vulgaris. Fuel 94, 544550.

Wahlen, B.D., Willis, R.M., Seefeldt, L.C., 2011. Biodiesel production by simultaneousextraction and conversion of total lipids from microalgae, cyanobacteria, andwild mixed-cultures. Bioresour. Technol. 102 (3), 27242730.

Wang, G., Wang, T., 2012. Characterization of lipid components in two microalgaefor biofuel application. J. Am. Oil Chem. Soc. 89 (1), 135143.

Zhang, H., Lin, Q., Ponnusamy, S., Kothandaraman, N., Lim, T.K., Zhao, C., Kit, H.S.,Arijit, B., Rauff, M., Hew, C.-L., Chung, M.C.M., Joshi, S.B., Choolani, M., 2007.Differential recovery of membrane proteins after extraction by aqueousmethanol and trifluoroethanol. Proteomics 7 (10), 16541663.

http://refhub.elsevier.com/S0960-8524(15)00067-X/h0075http://refhub.elsevier.com/S0960-8524(15)00067-X/h0075http://refhub.elsevier.com/S0960-8524(15)00067-X/h0080http://refhub.elsevier.com/S0960-8524(15)00067-X/h0080http://refhub.elsevier.com/S0960-8524(15)00067-X/h0085http://refhub.elsevier.com/S0960-8524(15)00067-X/h0085http://refhub.elsevier.com/S0960-8524(15)00067-X/h0085http://refhub.elsevier.com/S0960-8524(15)00067-X/h0085http://refhub.elsevier.com/S0960-8524(15)00067-X/h0090http://refhub.elsevier.com/S0960-8524(15)00067-X/h0090http://refhub.elsevier.com/S0960-8524(15)00067-X/h0090http://refhub.elsevier.com/S0960-8524(15)00067-X/h0095http://refhub.elsevier.com/S0960-8524(15)00067-X/h0095http://refhub.elsevier.com/S0960-8524(15)00067-X/h0095http://refhub.elsevier.com/S0960-8524(15)00067-X/h0100http://refhub.elsevier.com/S0960-8524(15)00067-X/h0100http://refhub.elsevier.com/S0960-8524(15)00067-X/h0100http://refhub.elsevier.com/S0960-8524(15)00067-X/h0105http://refhub.elsevier.com/S0960-8524(15)00067-X/h0105http://refhub.elsevier.com/S0960-8524(15)00067-X/h0105http://refhub.elsevier.com/S0960-8524(15)00067-X/h0110http://refhub.elsevier.com/S0960-8524(15)00067-X/h0110http://refhub.elsevier.com/S0960-8524(15)00067-X/h0110http://refhub.elsevier.com/S0960-8524(15)00067-X/h0115http://refhub.elsevier.com/S0960-8524(15)00067-X/h0115http://refhub.elsevier.com/S0960-8524(15)00067-X/h0120http://refhub.elsevier.com/S0960-8524(15)00067-X/h0120http://refhub.elsevier.com/S0960-8524(15)00067-X/h0125http://refhub.elsevier.com/S0960-8524(15)00067-X/h0125http://refhub.elsevier.com/S0960-8524(15)00067-X/h0130http://refhub.elsevier.com/S0960-8524(15)00067-X/h0130http://refhub.elsevier.com/S0960-8524(15)00067-X/h0135http://refhub.elsevier.com/S0960-8524(15)00067-X/h0135http://refhub.elsevier.com/S0960-8524(15)00067-X/h0135http://refhub.elsevier.com/S0960-8524(15)00067-X/h0140http://refhub.elsevier.com/S0960-8524(15)00067-X/h0140http://refhub.elsevier.com/S0960-8524(15)00067-X/h0145http://refhub.elsevier.com/S0960-8524(15)00067-X/h0145http://refhub.elsevier.com/S0960-8524(15)00067-X/h0150http://refhub.elsevier.com/S0960-8524(15)00067-X/h0150http://refhub.elsevier.com/S0960-8524(15)00067-X/h0150http://refhub.elsevier.com/S0960-8524(15)00067-X/h0155http://refhub.elsevier.com/S0960-8524(15)00067-X/h0155http://refhub.elsevier.com/S0960-8524(15)00067-X/h0160http://refhub.elsevier.com/S0960-8524(15)00067-X/h0160http://refhub.elsevier.com/S0960-8524(15)00067-X/h0160http://refhub.elsevier.com/S0960-8524(15)00067-X/h0165http://refhub.elsevier.com/S0960-8524(15)00067-X/h0165http://refhub.elsevier.com/S0960-8524(15)00067-X/h0170http://refhub.elsevier.com/S0960-8524(15)00067-X/h0170http://refhub.elsevier.com/S0960-8524(15)00067-X/h0170http://refhub.elsevier.com/S0960-8524(15)00067-X/h0170

Biodiesel synthesis by direct transesterification of microalga Botryococcus braunii with continuous methanol reflux1 Introduction2 Methods2.1 Lipid quantification in B. braunii2.2 Experimental set up2.2.1 Evaluation of the effect methanol:total fatty acids molar ratio on FAME yield2.2.2 Spent microalgal biomass characteristics in RER2.2.3 Optimization of direct transesterification in a RER

2.3 Chromatographic methods2.3.1 FAME quantification and distribution2.3.2 Acyl-glycerol quantification2.3.3 Lipid profile

2.4 Statistical analysis

3 Results and discussion3.1 Effect of methanol:total fatty acids molar ratio on FAME yield3.2 Spent microalgal biomass characteristics in RER3.3 Optimization of FAME yield in RER

4 ConclusionAcknowledgementsReferences