Fate of toxic phorbol esters in Jatropha curcas oil by a biodiesel fuel production process

O R I G I N A L P A P E RFate of toxic phorbol esters in Jatropha curcas oil by a biodiesel fuel production process Duong Huu Huy1,2• Kiyoshi Imamura3•Le Tu Thanh2•Phuong Duc Luu4• Hoa T

O R I G I N A L P A P E R

Fate of toxic phorbol esters in Jatropha curcas oil by a biodiesel

fuel production process

Duong Huu Huy1,2• Kiyoshi Imamura3•Le Tu Thanh2•Phuong Duc Luu4•

Hoa Thi Truong5• Hanh Thi Ngoc Le1• Boi Van Luu4•Norimichi Takenaka1•

Yasuaki Maeda3

Received: 24 September 2015 / Accepted: 6 March 2016

Ó Springer-Verlag Berlin Heidelberg 2016

Abstract Biodiesel fuel (BDF) is an important alternative

fuel because of the carbon neutral nature of biomass and

the exhaustion of fossil fuel resources Jatropha curcas oil

(JCO) produced from J curcas seeds contains toxic

phor-bol esters that can cause cancer The behaviors of toxic

phorbol esters were investigated during BDF production

Liquid chromatography–tandem mass spectrometry and

photodiode array analyses revealed that the phorbol esters

contained in JCO had a tigliane skeleton The partition

coefficients of phorbol esters between methanol (MeOH)

and the oil (KMeOH/oil) ranged from 2.4 to 20 As a result,

the phorbol esters in the JCO were largely partitioned into

the MeOH phase The phorbol esters in the oil were

con-verted stoichiometrically into phorbol and the

corre-sponding fatty acid methyl esters via a transesterification

reaction in a potassium hydroxide (KOH)/methanol

(MeOH) solution The phorbol produced predominantly

partitioned into the glycerin phase A small amount of phorbol residue contained in the BDF could be removed by washing with water These results suggest that it is safe to use BDF produced by the aforementioned transesterifica-tion reactransesterifica-tion and purificatransesterifica-tion process However, phorbol contamination of glycerin and wastewater from the pro-duction process should not be ignored

Keywords Phorbol esters Phorbol  Jatropha curcas oil (JCO)  Transesterification  Participation

Introduction Biodiesel fuel (BDF) is an alternative fuel produced from renewable vegetable oils (Thanh et al 2010b, 2013; Chakraborty et al 2015), animal fats (Halek et al 2013; Thanh et al 2013; Gurusala and Selvan 2015), recycled cooking oil (Thanh et al 2010a; Chuah et al 2015; Dela-vari et al.2015), and biomass waste (Caetano et al.2014), and it has drawn significant attention because diminishing petroleum reserves and increasing environmental concerns that favor the use of carbon neutral fuels (Glaser 2009) Presently, more than 10 million tonnes of BDF have been produced commercially from vegetable oil, and about three million tonnes have been produced from waste cooking oils

in the European Union (EU), which have reduced air pol-lution and the net emission of greenhouse gases (Freedman

et al 1984; Shay 1993; Ma and Hanna1999; Yuen-May and Ah-Ngan 2000; Parawira2010)

A variety of edible oils, such as rapeseed, soybean, palm, coconut, sunflower, and peanut oils, can be used as raw materials for BDF; however, the use of edible oils for BDF production competes with that for food production in the marketplace This increases the costs of BDF products and

& Kiyoshi Imamura

k_imamura@riast.osakafu-u.ac.jp

1 Graduate School of Engineering, Osaka Prefecture

University, 1-1 Gakuen-cho, Naka-ku, Sakai-shi,

Osaka 599-8531, Japan

2 Faculty of Environmental Science, University of Science,

Vietnam National University - Ho Chi Minh City, 227

Nguyen Van Cu St., Dist 5, Ho Chi Minh City, Vietnam

3 Research Organization for University-Community

Collaborations, Osaka Prefecture University, 1-2

Gakuen-cho, Naka-ku, Sakai-shi, Osaka 599-8531, Japan

4 Faculty of Chemistry, Vietnam National University, Hanoi,

19 Le Thanh Tong St., Hanoi, Vietnam

5 Danang Environmental Technology Center, Institute of

Environmental Technology, Vietnam Academy of Science

and Technology, Tran Dai Nghia Road, Ngu Hanh Son

District, Da Nang, Vietnam

DOI 10.1007/s10098-016-1149-4

disturbs the stable supply of food products Therefore, it is

necessary to identify other raw materials that have high

yields and lower prices than edible oils In this context,

non-edible oils, such as jatropha, neem, karanja, rubber, and

tobacco oils are prominent candidates for BDF production

Jatropha curcas, an oil-bearing shrub, can grow at high

elevations in dry regions, as well as on wastelands, and is

widely distributed in Asian, American and African

coun-tries The seed kernels contain up to 60 % oil that is

composed of triglycerides, but the seeds and seed oil (JCO)

cannot be used as nutrients because they are toxic and

co-carcinogenic to humans and animals (Makkar et al.1998;

Ahmed and Salimon2009; Li et al.2010) As a result of the

sudden increase in the price of crude and edible oils in

2008, the plantation area of J curcas expanded to a few

tens of thousands of hectares (ha) in developing countries,

including those in West Africa and India, to increase BDF

production (Iiyama2012) Siang (2009) reported that the

expected worldwide land area for J curcas cultivation will

be 33 million ha in 2017 according to an estimate by the

International Jatropha Organization, which will result in

the production of 160 million tonnes of seeds

Phorbol esters have been identified as the major toxic

compounds in JCO, and their contents are less than a few

percent in the seed kernels (Makkar et al.1997) Phorbol is

a naturally occurring tigliane diterpene, and it contains four

rings (A, B, C, and D) that are substituted with five

hydroxyl (OH) functional groups The epimeric isomer of

the beta OH group at the C4 position is biologically active,

while that of the alpha OH group is inactive (Silinsky and

Searl 2003) The esterification of phorbol at different

positions with various kinds of carboxylic acids leads to the

formation of a large variety of phorbol ester compounds

Many kinds of phorbol esters and deoxyphorbol esters have

been identified using liquid chromatography–tandem mass

spectrometry (LC/MS/MS) measurements (Vogg et al

1999) Six 12-deoxy-16-hydroxy phorbol esters have been

isolated from JCO, and their structures and toxicities have

been characterized (Haas et al 2002; Goel et al 2007)

(Fig.1)

Phorbol esters are well known as cancer-promoting

materials that exert a plethora of biological effects,

including inflammation, tumor promotion, cell

prolifera-tion, and differentiation (Mentlein1986; Goel et al.2007;

Li et al 2010) Devappa et al (2010) reported that the

toxicity (EC50, half maximal effective concentration) of

phorbol esters extracted from JCO using methanol (MeOH)

was 330 lg/L (phorbol 12-myristate-13-acetate (TPA)

equivalent) by the snail bioassay and 26.5 mg/L by the

Artemia assay Roach et al (2012) reported the EC50sof

six compounds, named Jatropha factors C1 to C6, which

were isolated and purified from J curcas seeds The EC50s

of factor C3 were 280 lg/L by the snail test and 44.6 mg/L

by the Artemia test; the EC50sof factor C2 were 270 lg/L and 487 mg/L; the EC50sof factor C1 were 170 lg/L and 17.6 mg/L; and the EC50sof a C4&C5 mixture were 90 lg/

L and 1.8 mg/L, respectively

During BDF production, the dry seeds of J curcas are chopped into pieces and pressed and heated to make the oil Crude BDF is produced from the oil by a transesterification reaction in a MeOH solution in the presence of alkaline (KOH and/or NaOH) (Berchmans and Hirata2008; Thanh

et al 2010a,b) The final BDF product is obtained after purification with water washes, followed by distillation under reduced pressure to remove the water Homogeneous transesterification process that is catalyzed by KOH using acetone as a co-solvent had been developed, and the reaction using the co-solvent method terminates within a few minutes to produce crude BDF and glycerin (Maeda

et al 2011; Thanh et al.2013; Luu et al.2014) Recently, the heterogeneous catalyst reaction has been developed for both esterification of FFAs and transesterification of triglyceride in a single step (Singh et al.2015) and the use

of a helicoidal reactor with ultrasound-assisted for contin-uous biodiesel production (Delavari et al.2015)

As a huge increase in BDF production from JCO is expected in the latter half of this decade, the behaviors of the toxic compounds of phorbol esters in JCO should be examined during BDF production to prevent harmful effects to humans, to minimize the contamination of the environment via the emission of waste materials, and to ensure the safety of BDF as a commercial product The objectives of this study are to investigate the behaviors of toxic phorbol esters during the production of BDF, and to remove phorbol ester contaminants from the BDF products

In addition, the distributions of phorbol and phorbol esters into the glycerin and FAME phases by the transesterifica-tion of JCO and into the FAME and wastewater phases by the clean-up process of crude BDF are investigated in order

to prospect the fate of toxic phorbol and phorbol esters under the process of BDF production from JCO

Materials and methods Jatropha curcas oil

In this study, JCO produced from J curcas seeds harvested

in Son La, Vietnam was used JCO was produced by compressing dry seeds containing 38 wt% of oil The physical and chemical properties were as follows: density, 0.913 g/cm3; acid value, 9.67 mg KOH/g oil; water con-tent, 0.1 wt%; and the components of fatty acid methyl esters (FAME) after transesterification were methyl palmitate (16.2 wt%), methyl stearate (7.4 wt%), methyl oleate (35.5 wt%), and methyl linoleate (37.1 wt%) The

estimated average molecular weight of the JCO was

840 g/mol

Reagents and standards

Standards of phorbol and five kinds of phorbol esters

(PDA, phorbol 12-, 13-diacetate; PDBu, phorbol 12-,

13-dibutyrate; PDB, phorbol 12-, 13-dibenzoate; TPA, and

PDD, phorbol 12-, 13-didecanoate) were purchased from

Wako Pure Chemicals (Osaka, Japan) MeOH, ethanol,

acetonitrile, and tetrahydrofuran (THF) were a

high-per-formance liquid chromatography (HPLC) analytical

gra-de, and isopropanol, acetone, KOH and phosphoric acid

were analytical grade They were purchased from Wako

Pure Chemicals (Osaka, Japan)

The alkaline solution (KOH/MeOH) for the

transesteri-fication reaction was prepared by dissolving 3.6 g of KOH

in 100 mL of MeOH

Preparation of standard and stock solutions Individual stock standard solutions of phorbol, PDA, PDBu, PDB, TPA, and PDD were prepared at a concen-tration of 1000 lg/mL Oxygen in the atmosphere of the MeOH solutions was purged for 10 min with nitrogen gas Standard concentrations were estimated by measuring the difference in the container weight before and after disso-lution of the standards A mixture of six compounds was prepared by mixing the individual standard solutions and diluting them at concentrations ranging from 0.1 to 100 lg/

mL All standard solutions were stored at 4°C

Measurement of phorbol, phorbol esters, and fatty acid methyl esters

The HPLC system for quantitative analysis of phorbol and its esters consisted of a series GL 7400 (GL Sciences Inc.,

4 13

O C

H

H OH

OH C

H3

C

H3 16

O O

CH3

CH2

4 13

O C

H

H OH

OH C

H3

C

H3 16

CH2

O O O

O

CH3

4 13

O

C

H

H OH

OH C

H3

C

H3 16

O O

C

H2

CH3

4 13

O C

H

H OH

OH C

H3

C

H3 16

O O O

O

C

H2

CH3

4 13

O C

H

H OH

OH C

H3

C

H3 16

O O O

O

CH3

10 9

4 8

5 6 7 1

2 3

14 13 12

11 15

O C

H3 19

20 OH

H

H OH

OH C

H3 18

C

H3 17 16 O

OH

A B C D

Jatropha factor C2

Fig 1 Structures of 12-deoxy-16-hydroxy phorbol and six phorbol esters named Jatropha factor C1 to C6 in Jatropha curcas oil (Haas et al.

2002 )

Saitama, Japan) equipped with a UV–Vis detector

(GL-7450, GL Sciences Inc.) and a photodiode array (PDA)

detector (GL-7452A, GL Sciences Inc.) For the analysis

using the UV–Vis detector, an Inertsil ODS-4 analytical

column (particle size 3 lm, 250 mm 9 3 mm i.d.) was

used Analytical conditions were as follows: the mobile

phase was water and acetonitrile, operated in a gradient

mode, with an initial water to acetonitrile volume ratio of

60:40, followed by a 50:50 ratio for 10 min, a 25:75 ratio

for 30 min, a 0:100 ratio for 15 min, and a 60:40 ratio for

10 min Finally, the column was washed with solvent

containing 75 % THF and 25 % acetonitrile The

separa-tion process was conducted at a column temperature of

30°C, and the flow rate was 0.4 mL/min The UV-VS

detector was operated at wavelength of 280 nm The

injection volume was 20 lL

For the analysis using the PDA detector, a cartridge

guard column E was mounted on an Inertsil ODS-4 column

(particle size 3 lm, 100 mm 9 3 mm i.d.) Analytical

conditions were as follows: the initial mobile phase was a

mixture of water and acetonitrile (95:5 ratio), followed by a

50:50 ratio for 10 min, a 25:75 ratio for 15 min, and a

0:100 ratio for 15 min at a column temperature of 30°C

The injection volume was 50 lL

A LC/MS/MS system for qualitative analysis of phorbol

esters consisting of a GC 7400 HPLC (GL Sciences Inc.,

Saitama, Japan) and an Applied Biosystems API 4000

QTrapÒ LC/MS/MS system (Thermo Fisher Scientific,

Waltham, MA, USA) equipped with an electron spray was

used The analytical conditions of the HPLC system were

the same as those of the PDA analysis described

previ-ously The LC/MS/MS system was operated in multiple

reactions monitoring (MRM)-positive mode with

collision-induced dissociation The characteristic precursor ion was

monitored simultaneously with one of its fragment

prod-ucts, such as m/z 313 to m/z 295 (313/295) and m/z 295 to

m/z 267 (295/267) for monitoring the ingenane type of

phorbol, while m/z 311 to m/z 293 (311/293) and m/z 293 to

m/z 265 (293/265) were used for monitoring the tigliane

type (Vogg et al.1999)

A group of peaks eluted from 35 to 40 min according to

the PDA analysis of the HPLC data, and the peaks that

eluted from 40 to 45 min according to a precursor scan

analysis of the LC/MS/MS with m/z 311 to m/z 293 (311/

293) were assigned as components of the tigliane-type

phorbol esters

The gas chromatograph (GC) system for FAME analysis

was a Hewlett Packard HP 6890 (Agilent Technologies,

Santa Clara, CA, USA) equipped with a flame ionization

detector (FID) The analytical column was a SPTM-2380

(30 m 9 0.25 mm i.d., 0.2-lm film thickness) (Supelco,

Bellefonte, PA, USA) The column temperature was held at

50°C for 1 min, and it was programmed to increase to

250 °C at a rate of 10 °C/min and held for 5 min The injection temperature was 250°C, and the helium gas flow rate was 1.0 mL/min The gas flow rates for the FID detector were as follows: hydrogen, 40 mL/min; air,

450 mL/min; and the carrier gas supply (helium), 45 mL/ min A 1–2 lL sample was injected by split mode with a split ratio of 1:50

Transesterifications Transesterification of JCO was performed as follows: 2.8 mL of MeOH containing 0.1 g KOH was added to 10 g

of JCO (molar ratio of MeOH to oil, 6:1; KOH catalyst to oil, 1 wt%), and then the mixture was stirred by a magnetic stirrer for 2 h at room temperature (25 ± 1 °C) After the reaction, the mixture was neutralized with phosphoric acid (5 % v/v) and left to separate into two phases: the upper FAME phase and the lower glycerin phase Twenty lL of each solution was diluted in an appropriate volume of solvent and injected into the HPLC for the determination of phorbol and phorbol esters

In the case of the MeOH extract, 2 mL of the MeOH extract containing phorbol esters extracted from JCO was reacted with 0.4 mL of MeOH containing 0.008 g of KOH

as a catalyst In the case of the phorbol ester standard solution, 0.2 mL of MeOH containing 0.004 g of KOH was added to 1 mL of the standard solutions They were treated

in the same manner as the aforementioned transesterifica-tion reactransesterifica-tions

Results and discussion Phorbol and phorbol esters The PDA chromatogram of the MeOH extract from JCO at wavelengths ranging from 190 to 300 nm is shown in Fig.2 The UV absorption maximum at wavelengths ranging from 260 to 300 nm for a group of peaks that eluted with retention times of 35–40 min was coincident with those of phorbol esters The pattern of the peaks consist of six components was very similar to that reported

by Makkar et al (1997)

According to the results of the MRM LC/MS/MS analysis, a group of peaks eluted with retention times ranging from 35 to 40 min was shown to be a tigliane type

of phorbol (Vogg et al.1999)

The HPLC chromatogram of the MeOH phase extracted from JCO is shown in Fig.3a, and that of authentic sam-ples of phorbol and five phorbol esters are shown in Fig.3b The five phorbol esters eluted in wide range, from PDA (9.5 min) to PDD (66 min) Among them, TPA, which was used as external standard for quantification, had

a retention time of 58.7 min A group of phorbol esters

extracted from JCO eluted at retention times range from 52

to 56 min, as determined by monitoring at the 280 nm

wavelength of the UV region; however, the peaks eluted

just before the group of phorbol esters could not be

assigned as phorbol esters based on the UV spectra and

MRM analyses

To determine the concentration of phorbol esters in JCO, a 10 g of oil sample was extracted with 10 mL of MeOH, and the extraction was repeated three times After extraction, all extracts were combined After adjusting the volume with solvent, a 20 lL of aliquot was quantitatively analyzed using HPLC The concentration of phorbol esters was estimated from the total area of a group of phorbol

Fig 2 PDA chromatogram of

the MeOH extract from JCO at

wavelengths ranging from 190

to 300 nm

Fig 3 HPLC chromatogram of

MeOH extracts a from Jatropha

curcas oil and b authentic

phorbol ester standards Notes 1

phorbol; 2 PAA, phorbol 12,

13-diacetate; 3 PDBu, phorbol

12, 13-dibutyrate; 4 PDB,

phorbol 12, 13-dibenzoate; 5

TPA, phorbol 12-myristate

13-acetate; 6 PDD, phorbol 12,

13-didecanoate Each

concentration was ca 8.0 lg/

mL

ester components that had retention times ranging from 52

to 56 min in the chromatogram The quantification was

conducted by the external calibration method using TPA as

the standard material

The concentrations of phorbol esters contained in the

oils produced from J curcas seeds cultivated at three

dif-ferent areas in Vietnam were examined The results are

shown in Table1 Their concentrations ranged from 2 to

6 mg/g for phorbol esters and from 0.2 to 0.8 mg/g for

phorbol; the contents of phorbol and phorbol esters would

depend on the species of Jatropha, as well as the climatic

and geographic conditions in which the species were

cultivated

Partition coefficients

Ten g of JCO was extracted with 10 mL of MeOH The

partition coefficients KMeOH/Oil of the phorbol esters

between MeOH and JCO were estimated using the method

described in reference (Christian1986) The KMeOH/Oilwas

calculated using the following formula: KMeOH/Oil= (C1/

C2) - 1, where C1 is the concentration of a component

(mg/mL) of the first extraction and C2is that of the second

extraction The results are shown in Table2 These results

indicated that it was necessary to perform more than three

times extractions to attain more than 95 % efficiency

extraction of the PDD, because of its most hydrophobic

property of the phorbol esters tested (Wang et al.2000and

references therein) The most hydrophobic property of

PDD could be explained by its retention time on a

reversed-phase C18 column because it is the last compound

eluted as shown in Fig.3b, and its partition coefficient is

the lowest with value of 2.4 (Table2)

Transesterification Phorbol ester standards

A known amount of a TPA standard solution (80 mg/L) (Experiment 1) and a mixture of phorbol and five phorbol esters (PDA, PDBu, PDB, TPA, and PDD) standard solu-tions (each ca 8.0 mg/L) (Experiment 2) were reacted with MeOH in the presence of KOH as a catalyst After reacting, the mixtures were neutralized by phosphoric acid (5 % v/v), and aliquots of the products were analyzed by HPLC The HPLC chromatogram of the reaction products in Experiment 2 is shown in Fig.4b Five peaks of phorbol esters, as shown in Fig 3b, disappeared, and the intensity

of the phorbol peak with a retention time of 4.1 min increased The relationship between the concentrations of reactants and products in each experiment is shown in Table3 In Experiment 1, 13.9 mol of TPA was converted into 14.4 mol of phorbol (the molar ratio of phorbol to TPA was 1.04), and in Experiment 2, 9.0 mol of five phorbol esters and phorbol were converted into 9.3 mol of phorbol (the molar ratio of phorbol to phorbol esters was 1.03) For instance, 1 mol of phorbol and 2 mol of carboxylic acid methyl esters were produced from 1 mol of phorbol 12-myristate-13-acetate by the transesterification process (Eq 1) These molar ratios suggested that the reaction proceeded stoichiometrically, and thus, phorbol was pro-duced quantitatively These results suggest that phorbol esters were transesterified and completely converted into phorbol and the corresponding carboxylic acid methyl esters However, their methyl esters could not be identified because of their lower sensitivities in the HPLC analysis using the UV detector (280 nm)

Table 1 Concentrations of

phorbol and its esters contained

in JCO

Sample no Sources of location in Vietnam Phorbol esters (mg/g) Phorbol (mg/g)

Table 2 Partition coefficients of authentic phorbol esters

No Phorbol esters Partition coefficient K (CV %) No Phorbol esters Partition coefficient K (CV %)

PDA phorbol 12, 13-diacetate; PDBu phorbol 12, 13-dibutyrate; PDB 12, 13-dibenzoate; TPA phorbol 12-myristate 13-acetate; PDD phorbol 12, 13-didecanoate; PEs a group of phorbol esters extracted from JCO; CV coefficient of variation

MeOH extract from Jatropha curcas oil

The MeOH extract from JCO was reacted with a KOH/

MeOH solution, and the product was analyzed as described

in ‘‘Phorbol ester standards’’ section The HPLC

chro-matogram of the reaction products is shown in Fig.4a

Phorbol esters also disappeared, and the intensity of the

phorbol peak (4.1 min) increased in the same manner as

that of the authentic phorbol esters However, the molar

ratio of the phorbol in product to the total amount of

phorbol and phorbol esters was 0.62, which was lower than

that estimated stoichiometrically This was mainly caused

by the different sensitivities (per gram) (Dimitrijevic´ et al

1996) of each phorbol ester to TPA because the sensitivity

at 280 nm depends on the absorption coefficient of the

chromophore in the molecule and, as can be seen in the

HPLC chromatogram shown in Fig.3b, an approximately

three-fold higher sensitivity (per gram) of PDB was

esti-mated in comparison with that of TPA

As shown in Fig.4a, small three peaks in the range from

50 to 52 min near the phorbol ester peaks obtained by

analysis of transesterification products of MeOH extract are

detected The retention times and the pattern of these peaks

do not overlap with those of phorbol esters By GC-FID

analysis, these peaks are assigned as BDF produced from

JCO, of which contents are methyl palmitate (16.3 wt%),

methyl stearate (7.3 wt%), methyl oleate (35.7 wt%), and

methyl linoleate (36.7 wt%) A certain amount of JCO is

participated into MeOH phase during MeOH extraction;

therefore, the same components of FAME are produced in

the process of transesterification This result indicates that

small three peaks are not the products from phorbol esters

after transesterification

Jatropha curcas oil

In this section, the behaviors of phorbol esters contained in

the large matrix of JCO were examined in the process of

BDF production The transesterification of JCO was

conducted with a KOH/MeOH solution using a mechanical stirring method After the reaction was completed, the reaction products were neutralized by phosphoric acid, and then allowed to separate into the FAME and glycerin phases The FAME and glycerin phases were dissolved in THF and MeOH solvents, respectively, and an aliquot was analyzed by HPLC The chromatogram of the FAME and glycerin products was similar to that of the transesterifi-cation products of the MeOH extract (Fig 4a) The three main peaks of FAME in the glycerin phase were observed

in the range from 50 to 52 min The contents of phorbol esters were less than the detection level in both of the FAME and glycerin phases

The transesterification of JCO was further conducted using the co-solvent method with a co-solvent of acetone and THF (Thanh et al.2013; Luu et al.2014) The results were the same as those observed for the mechanical stirring method These results indicated that phorbol esters tained in the large matrix of JCO were completely con-verted into skeletal frame of phorbol and the corresponding carboxylic acid methyl esters After transesterification, the contents of phorbol in the crude BDF (BDF1) and glycerin phases are shown in Table4 Phorbol mostly participates into the glycerin phase (1.4–1.7 mg/g), but only small amount distributes into the FAME phase (0.0032 mg– 0.0046 mg/g) because of a polar property of phorbol Clean-up process

After transesterification, the reaction mixture was separated into the glycerin and BDF1 phases, and then a final product

of BDF (BDF2) was obtained by cleaning-up BDF1 with water to improve the BDF quality The distributions of phorbol in the FAME and aqueous phase were examined The results are shown in Table4 Phorbol, the content of which was 0.0037–0.0046 mg/g remained in FAME, was washed out with water and participated into the aqueous phase (0.0045–0.0064 mg/L) As a result, the level of phorbol in BDF2 was reduced to the non-detectable level

C

H3

OH

O HO

H C

H3 O

CH3

CH3 O

H

O

CH3

O C

H3

+ 2 CH 3 OH KOH H3C

OH

O HO

H C

H3 OH

CH3

CH3 OH

OH H H

O C

O

CH3 C

H3 catalyst

phorbol 12-myristate 13-acetate phorbol methyl acetate

methyl myristate

ð1Þ

As for the transesterification using acetone as a co-solvent,

it was impossible to determine the content of phorbol

because of overlapping with large peak of acetone

The toxic components of phorbol ester and phorbol

contained in BDF2 are less than the detection level, and the

BDF is safe to use, although further purification is needed

for production of the commercial product On the contrary, the wastewater, emitted from cleaning-up process con-taining not only toxic phorbol but also other chemicals such as solvents, alkali, and oily products, deteriorates the aqueous environmental quality when discharged without any treatment The fates of phorbol and phorbol esters in

Table 3 Comparison of the concentrations of phorbol and its esters before and after transesterification with KOH/MeOH

Reactant (conc.)1 Product (conc.)1 Reactant (conc.)1 Product (conc.)1 Reactant (conc.)1 Product (conc.)1

1 mg/L (mol/L 9 10-5), 2 The ratio of the molar concentration of phorbol in product to those of the reactants, PDA, phorbol 12, 13-diacetate; PDBu, phorbol 12, 13-dibutyrate; PDB, phorbol 12, 13-dibenzoate; PEs, a group of phorbol esters extracted from JCO; TPA, phorbol 12-myristate 13-acetate; PDD, phorbol 12, 13-didecanoate

Fig 4 HPLC chromatogram of

transesterification products

a from a MeOH extracts b from

phorbol ester standards

the process of wastewater treatment should be investigated

to estimate their impact to aqueous environment On the

other hand, in case of by-product of glycerin that is the

useful natural resource of medicines and cosmetics, the

detoxification of phorbol in glycerin obtained from JCO is

strongly required for avoiding the direct human health

effects before use

Conclusions

The behaviors of toxic phorbol esters during BDF

pro-duction were investigated A group of phorbol esters in

JCO in the experiment were assigned to be a tigliane type

The partition coefficient (KMeOH/oil) of these phorbol esters

was 2.4 Accordingly, it is necessary to perform at least

three MeOH extractions to remove more than 95 % of the

phorbol esters from the oil In the transesterification with

KOH/MeOH, BDF and glycerin, as a by-product, were

produced, and simultaneously, the phorbol esters were

converted into the skeletal frame of phorbol and the

cor-responding carboxylic acid methyl esters Notably, most of

phorbol partitioned into the glycerin phase The small

amount of phorbol residue in the BDF1 could be removed

by washing with water because of its high polarity These

results suggest that the BDF product produced by the

transesterification reaction followed by the purification

process is safe to use In case of by-product of glycerin

produced from JCO, the detoxification of phorbol is

strongly required for avoiding the direct human health

effects when it is used by cosmetics and medical products

Acknowledgments The authors thank the Japan Science and

Technology Agency (JST) and the Japan International Cooperation

Agency (JICA) for their support of the Science and Technology

Research Partnership for Sustainable Development (SATREPS)

pro-ject titled ‘‘Multi-Beneficial Measure for Mitigation of Climate

Change in Vietnam and Indochina Countries by the

Cultivation-Production-Utilization of Biomass Energy.’’

Compliance with ethical standards

Conflict of Interest The authors declare that they have no conflict

of interest.

References

Ahmed WA, Salimon J (2009) Phorbol ester as toxic constituents of tropical Jatropha curcas seed oil Eur J Sci Res 31:429–436 Berchmans HJ, Hirata S (2008) Biodiesel production from crude Jatropha curcas L seed oil with a high content of free fatty acids Bioresour Technol 99:1716–1721 doi: 10.1016/j.biortech 2007.03.051

Caetano N, Silva VM, Melo A, Martins A, Mata T (2014) Spent coffee grounds for biodiesel production and other applications Clean Technol Environ Policy 16:1423–1430 doi: 10.1007/ s10098-014-0773-0

Chakraborty R, Das S, Bhattacharjee S (2015) Optimization of biodiesel production from Indian mustard oil by biological tri-calcium phosphate catalyst derived from turkey bone ash Clean Technol Environ Policy 17:455–463 doi: 10.1007/s10098-014-0802-z

Christian GD (1986) Analytical chemistry Wiley, New York Chuah L, Aziz A, Yusup S, Bokhari A, Klemesˇ J, Abdullah M (2015) Performance and emission of diesel engine fuelled by waste cooking oil methyl ester derived from palm olein using hydrodynamic cavitation Clean Technol Environ Policy doi: 10.1007/s10098-015-0957-2

Delavari A, Halek F, Amini M (2015) Continuous biodiesel production in a helicoidal reactor using ultrasound-assisted transesterification reaction of waste cooking oil Clean Technol Environ Policy 17:273–279 doi: 10.1007/s10098-014-0790-z Devappa RK, Makkar HPS, Becker K (2010) Biodegradation of Jatropha curcas phorbol esters in soil J Sci Food Agric 90:2090–2097 doi: 10.1002/jsfa.4056

Dimitrijevic´ SM, Humer U, Shehadeh M, Ryves WJ, Hassan NM, Evans FJ (1996) Analysis and purification of phorbol esters using normal phase HPLC and photodiode-array detection J Pharm Biomed 15:393–401 doi: 10.1016/S0731-7085(96)01867-5 Freedman B, Pryde EH, Mounts TL (1984) Variables affecting the yields of fatty esters from transesterified vegetable oils J Am Oil Chem Soc 61:1638–1643 doi: 10.1007/bf02541649

Glaser J (2009) Carbon dioxide recycling Clean Technol Environ Policy 11:253–257 doi: 10.1007/s10098-009-0251-2

Goel G, Makkar HP, Francis G, Becker K (2007) Phorbol esters: structure, biological activity, and toxicity in animals Int J Toxicol 26:279–288

Gurusala N, Selvan V (2015) Effects of alumina nanoparticles in waste chicken fat biodiesel on the operating characteristics of a compression ignition engine Clean Technol Environ Policy 17:681–692 doi: 10.1007/s10098-014-0825-5

Haas W, Sterk H, Mittelbach M (2002) Novel 12-deoxy-16-hydrox-yphorbol diesters isolated from the seed oil of Jatropha curcas.

J Nat Prod 65:1434–1440 doi: 10.1021/np020060d Halek F, Delavari A, Kavousi-rahim A (2013) Production of biodiesel

as a renewable energy source from castor oil Clean Technol Environ Policy 15:1063–1068 doi: 10.1007/s10098-012-0570-6

Table 4 Concentrations of phorbol in the clean-up process

Transesterification method After reaction Clean-up with water

In BDF1 (mg/g) In glycerin (mg/g) In BDF2 (mg/g) In the water layer (mg/L)

BDF1 crude biodiesel fuel after phase separation; BDF2 a final biodiesel fuel after clean-up with water; ND not detected; –, impossible to determine because of the overlap with acetone solvents; THF, tetrahydrofuran; no solvent, without co-solvent

Iiyama K (2012) Jatropha: the Saviour or Satan? Japan International

Research Center for Agriculture Science, Tsukuba

Li C-Y, Devappa RK, Liu J-X, Lv J-M, Makkar HPS, Becker K

(2010) Toxicity of Jatropha curcas phorbol esters in mice Food

Chem Toxicol 48:620–625 doi: 10.1016/j.fct.2009.11.042

Luu PD, Truong HT, Luu BV, Pham LN, Imamura K, Takenaka N,

Maeda Y (2014) Production of biodiesel from Vietnamese

Jatropha curcas oil by a co-solvent method Bioresour Technol

173:309–316 doi: 10.1016/j.biortech.2014.09.114

Ma F, Hanna MA (1999) Biodiesel production: a review 1 Bioresour

Technol 70:1–15 doi: 10.1016/S0960-8524(99)00025-5

Maeda Y et al (2011) New technology for the production of biodiesel

fuel Green Chem 13:1124–1128 doi: 10.1039/c1gc15049a

Makkar HPS, Becker K, Sporer F, Wink M (1997) Studies on

nutritive potential and toxic constituents of different

prove-nances of Jatropha curcas J Agric Food Chem 45:3152–3157.

doi: 10.1021/jf970036j

Makkar H, Becker K, Schmook B (1998) Edible provenances of

Jatropha curcas from Quintana Roo state of Mexico and effect

of roasting on antinutrient and toxic factors in seeds Plant Food

Hum Nutr 52:31–36

Mentlein R (1986) The tumor promoter 12-O-tetradecanoyl phorbol

13-acetate and regulatory diacylglycerols are substrates for the

same carboxylesterase J Biol Chem 261:7816–7818

Parawira W (2010) Biodiesel production from Jatropha curcas: a

review Sci Res Essays 5:1796–1808

Roach JS, Devappa RK, Makkar HPS, Becker K (2012) Isolation,

stability and bioactivity of Jatropha curcas phorbol esters.

Fitoterapia 83:586–592 doi: 10.1016/j.fitote.2012.01.001

Shay EG (1993) Diesel fuel from vegetable oils: status and

opportunities Biomass Bioenerg 4:227–242 doi:

10.1016/0961-9534(93)90080-N

Siang CC (2009) Jatropha curcas: development of a new oil crop for biofuel East Asian Bureau of Economic Research

Silinsky EM, Searl TJ (2003) Phorbol esters and neurotransmitter release: more than just protein kinase C? Br J Pharmacol 138:1191–1201 doi: 10.1038/sj.bjp.0705213

Singh D, Ganesh A, Mahajani S (2015) Heterogeneous catalysis for biodiesel synthesis and valorization of glycerol Clean Technol Environ Policy 17:1103–1110 doi: 10.1007/s10098-014-0858-9 Thanh LT, Okitsu K, Sadanaga Y, Takenaka N, Maeda Y, Bandow H (2010a) A two-step continuous ultrasound assisted production of biodiesel fuel from waste cooking oils: a practical and econom-ical approach to produce high quality biodiesel fuel Bioresour Technol 101:5394–5401 doi: 10.1016/j.biortech.2010.02.060 Thanh LT, Okitsu K, Sadanaga Y, Takenaka N, Maeda Y, Bandow H (2010b) Ultrasound-assisted production of biodiesel fuel from vegetable oils in a small scale circulation process Bioresour Technol 101:639–645 doi: 10.1016/j.biortech.2009.08.050 Thanh LT, Okitsu K, Sadanaga Y, Takenaka N, Maeda Y, Bandow H (2013) A new co-solvent method for the green production of biodiesel fuel: optimization and practical application Fuel 103:742–748 doi: 10.1016/j.fuel.2012.09.029

Vogg G, Achatz S, Kettrup A, Sandermann H Jr (1999) Fast, sensitive and selective liquid chromatographic–tandem mass spectromet-ric determination of tumor-promoting diterpene esters J Chro-matogr A 855:563–573 doi: 10.1016/S0021-9673(99)00728-1 Wang QJ, Fang T-W, Fenick D, Garfield S, Bienfait B, Marquez VE, Blumberg PM (2000) The lipophilicity of phorbol esters as a critical factor in determining the pattern of translocation of protein kinase C d fused to green fluorescent protein J Biol Chem 275:12136–12146 doi: 10.1074/jbc.275.16.12136 Yuen-May C, Ah-Ngan M (2000) Plant power Chem Ind 530–534