108

CHAPTER 5

TWO-STEP ACID ALKALI CATALYZED TRANSESTERIFICATION OF

C. pentandra SEED OIL FOR BIODIESEL PRODUCTION

5.1 INTRODUCTION

Environmental concerns of fossil fuel depletion and fluctuating oil price has

intensified the search for alternate fuel from renewable resources. Vegetable oil and

animal fats are found to be the best alternate energy source that can be used directly

in the existing engine. Their direct use is limited due to two main reasons, that is

high viscosity and low volatility (Knothe and Steidley 2005). Transesterification

technique has been widely used to reduce the viscosity of oils and fats.

Transesterification is nothing but displacement of alcohol from an ester by another

alcohol (Srivastava and Prasad 2000). The advantages of biodiesel as diesel fuel is

its portability, ready availability, renewability, higher combustion efficiency, lower

sulfur and aromatic content, higher cetane number and higher biodegradability

(Demirbas 2009a).

The use of edible vegetable oils and animal fats for biodiesel production has

received great concern because they compete with food materials (Kalam et al

2008). The demand for vegetable oils for food has increased tremendously in recent

years. It is impossible to justify the use of these oils for fuel purposes such as

biodiesel production. Moreover, these oils could be more expensive to use as fuel

(Demirbas 2009b). The uses of non-edible plant oil sources are keeping competition

with food edible oil for biodiesel feed stock. Hence, the contribution of non-edible

oil from C. pentandra will be significant source for biodiesel production.

109

5.1.1 Description of the Plant

C. pentandra is a tall, deciduous tree bearing short, sharp prickles all along

the trunk, branches and supported by pronounced buttresses at the base. It has a

light crown and is leafless for a long period. Leaves are alternate with slender green

petioles. There are usually 5 leaflets in a mature form. Great quantities of flowers

are in lateral clusters near the ends of the twigs. Fruit is leathery, ellipsoid,

pendulous capsule, 10-30 cm long and usually tapering at both ends. White, pale

yellow or grey floss originates from the inside wall of the fruit. Seed capsules split

open along 5 lines. Each capsule releases 120-175 seeds rounded black seeds

embedded in a mass of grey woolly hairs. Seeds are in dark brown color. The

generic name comes from a local South American word. The specific name,

‘pentandra’, is Latin for ‘five-stemmed’ from the Greek word ‘penta’ (five) and

‘andron’ (male).

5.1.2 Economic Benefits

The pressed cake is cattle feed containing about 26% protein. Sheep, goats

and cattle relish the foliage. The fiber from the inner wall of the fruit is unique in

that it combines springiness and resilience to make it ideal for stuffing pillows,

mattresses and cushions, life jackets and lifeboats. It is an excellent material for

insulating iceboxes, refrigerators, cold-storage plants, offices, theatres and

aeroplanes.

C. pentandra seed contains 20 to 25% non-drying oil, and is used as a

lubricant, in soap manufacturing and in cooking. Medicine: Compressed fresh

leaves are used against dizziness, decoction of the boiled roots is used to treat

oedema, gum is eaten to relieve stomach upset, tender shoot decoction is a

contraceptive and leaf infusion is taken orally against cough and hoarse throat.

110

5.1.3 Origin of Plant

It is believed that this tree originated in Central America. It has been

cultivated for a long time and can be found pantropically between 16 degrees north

and 16 degrees south.

It is native to India, Indonesia and United States of America. In Australia,

Cambodia, Eritrea, Ethiopia, Gambia, Ghana, Kenya, South Africa, Tanzania,

Thailand, Uganda and Zanziba are the region where it grows exotically In India, it

is found usually in southern parts of India. C. pentandra is grown around villages

and temples in Tamil Nadu, India, as an ornamental tree.

At present the C. pentandra oil has only limited application and the natural

production of seeds remain under utilized. Literature shows that no work has been

established so far on the production of biodiesel using C. pentandra oil. In this

present investigation, C. pentandra oil has been used as a potential source for

biodiesel production. The reaction conditions have been investigated to optimize the

process variables that lead to higher yield of biodiesel and to develop a simple

kinetic model for extraction process.

5.2 METHODOLOGY

Pods of C. pentandra were collected from local villages near Chennai, Tamil

Nadu, India during the month of July. Sample was identified as C. pentandra and

authenticated at Centre for Advanced Studies in Botany, University of Madras,

Chennai, Tamil Nadu, India.

5.2.1 Extraction

The C. pentandra pods were disrupted and seeds were removed manually

from the fiber. The collected seeds were dried under sun, ground to powder, passed

through 60 mesh and then the seed powder was dried at 105 C until a constant

111

weight was obtained. The C. pentandra seed powder was mixed with one forth

weight of diatomaceous earth for better solvent flow through sample. The mixture

was packed inside a thimble and extracted as prescribed in Chapter 3. The oil yields

obtained was expressed in terms of weight percentage of the samples.

5.2.2 Oil Characterization

The acid, saponification and iodine values were determined by titrimetry

(Sadasivam and Manickam 2004). Water content was determined using a KF

titrator. The unsaponifiable fractions of the extracted oils were analyed in duplicate

and the results are presented as mean values (Leon-Camacho et al 2004).

5.2.3 Biodiesel Production and Characterization

The transesterification reaction was carried out in a system as described in

Chapter 3. The effect of different parameters like catalyst concentration, methanol

to oil molar ratio, reaction temperature and reaction time were optimized. The

stirring rate of 600 rpm was kept constant throughout the process to get sufficient

mixing.

5.2.3.1 Acid catalyzed pre-esterification process

The alkali catalyzed reaction is reported to be very sensitive to the content of

FFAs, which should not exceed a certain limit to avoid deactivation of catalyst by

formation of soaps and emulsion (Van Gerpen 2005). Therefore, FFAs were first

converted to respective esters in a pretreatment process with methanol using an acid

catalyst (H2SO4). It was reviewed from the literature and found that the product

having acid value < 2 mg KOH g-1 is used for alkali catalyzed reaction (Sharma et

al 2008).

The acid catalyzed esterification is a pretreatment process employed to

decrease the acid value of the feedstock below 2 mg KOH g-1. Based on the results

112

of Chongkhong and Tongurai (2007) esterification reaction was performed by

employing methanol to oil ratio as 8:1 at 65°C with 1.834 wt% H2SO4 as a catalyst.

The FFA level of the mixture was checked at different time intervals. When the

required FFA level was reached, the mixture was cooled to room temperature and

transferred to a separating funnel without agitation, leading to separation of two

phases. Finally the acid value of the product separated at the bottom was

determined.

5.2.3.2 Alkali-catalyzed transesterification process

Alkali-catalyzed transesterification is the most effective in the

transesterification processes and is used in the commercial production of biodiesel.

Even at ambient temperature, the alkali-catalyzed reaction proceeds rapidly usually

reaching 95% conversion. It is noted that the parameters like catalyst concentration,

methanol to oil molar ratio, reaction temperature and reaction time play an

important role in production of biodiesel (Pilar et al 2004). The effect on varying

these parameters such as catalyst concentration (0.25, 0.50, 0.75, 1.0 and 1.25

wt%), methanol to oil molar ratio (3:1, 6:1, 9:1 and 12:1), reaction temperature (45,

50, 55, 60 and 65°C), reaction time (15, 30, 45 and 60 min) on the biodiesel yield

was studied.

The H1NMR spectra of bio-diesel were recorded. as per Knothe and Kenar

(2004), Gelbard et al (1995). Error bars shown in the figures represent the standard

deviations of experiments that had been done in triplicates.

To study the fuel properties, two 200 mL batches of biodiesel were produced

at optimized condition. The obtained dried methyl ester was properly stored in an

airtight brown glass container for characterization studies. Biodiesel fuel properties

were determined by ASTM test methods (ASTM 1998) and compared with ASTM

D6751 standards.

113

5.3 RESULTS AND DISCUSSION

5.3.1 Effect of Different Solvents on Extraction of Oil

The selection of the solvent system for oil extraction from C. pentandra seed

is an important factor. Solvent selection for extraction of oil at the initial step would

allow cost-effective for fuel production without further expense required for the

purification of the product. The solvent chosen should have good extraction

capacity and low viscosity to enhance the free circulation. An efficient extraction

requires the penetration of solvent into the seed and to match the polarity of the

targeted compounds. An organic solvent has a higher solubility with oil, this solvent

system was used further to degrade the cell walls of the seed and to dissolve the oil

to enhance the oil yield.

The percent oil yield values for the different solvents at 60°C are shown in

Table 5.1 for oil extraction. The solvent required for extraction was selected on the

basis of oil yield and umsaponifiable matter content. Higher amount of

unsaponifiable matter requires intensified pre-treatment for oil to be used for

biodiesel production. The oil yields peaked for THF at 27.2 wt% with 3.56 wt%.

This higher amount of unsaponifiable matter was undesirable. Methanol extract

yields was poor due high polar in nature and having high percentage of

unsaponifiable matter. Solvent hexane yields high oil (26.4 wt%) with less amount

of unsaponifiable matter (1.98 wt%). Hence, it was chosen for as a solvent for

extraction.

114

Table 5.1 Effect of different solvents on extraction of C. pentandra oil

Solvents Yield (wt%) Unsaponifiable matter wt%

Hexane 26.4 1.98

Petroleum ether 27 2.81

Tetra hydro furan 27.2 3.56

Methanol 14 5

Chloroform 20.4 2.9

5.3.2 Effect of Solvent Ratios on Oil Extraction

The effect of seed to hexane weight ratios on the oil extraction is shown in Figure

5.1. The experiments were studied under batch condition at 250 rpm, 65°C for 2 h

in a temperature controlled shaker. The influence of seed to hexane ratio from 1:4 to

1:12 on oil extraction was studied. As the seed to solvent ratio increased from 1:4 to

1:10, the oil yield was found to be increased from 11.9 wt% to 26.1 wt%. The trend

was continued with increase in seed to hexane ratio up to 1:10. Further increase

above 1:10 did not show much improvement in the oil extraction. Therefore the

ratio of 1:10 was found to be an optimum ratio for the further study.

0

5

10

15

20

25

30

1:4 1:6 1:8 1:10 1:12 Seed to Hexane weight ratio

Figure 5.1 Effect of oil yield on weight of seed to solvent different ratio

115

5.3.3 Kinetic and Thermodynamic Studies on Oil Extraction

Extraction was performed in batch mode at different time intervals 20, 40, 60

80 and 100 min. The percentage oil yields at various temperatures are given in

Table 5.2. From the analysis of the data, the oil yield was found to be increased

with increase in extraction time. The yield was also analyzed with respect to the

extraction time at constant temperature ranging from 30 to 60°C.

Table 5.2 Percent oil yield at various extraction temperature

Temperature (°C) Time (min) 30 40 50 60

20 8.92 10.62 12.43 14.6

40 9.58 11.49 13.52 15.92

60 10.29 12.46 14.72 17.39

80 11.06 13.46 15.97 18.97

100 11.89 14.6 17.4 20.7

Using the values in Table 5.2 and applying the differential method, plots of

ln Y versus ln (dY/dt) at different temperatures with optimum conditions were

established. A first-order kinetic model was fitted well with average regression

coefficient (R2) value obtained as 0.936 (Figure 5.2). The reaction rate constants

and the order of the reaction were determined using the intercept and slope of the

liner plot (Table 5.3).

116

Figure 5.2 Plot of ln (dY/dt) versus ln Y at different temperatures ranged from 30

to 60°C for extraction of oil

Table 5.3 Values of the reaction rate constants at different temperature

Temperature (°C) k min-1 R2 value

30 2.9557*10-3 0.9996

40 3.284*10-3 0.9575

50 3.7975*10-3 0.9646

60 4.0533*10-3 0.9949

117

5.3.3.1 Calculation of activation energy

The rate constant k increases with increasing temperature, and this trend is

shown in Table 4.3. The changes can be described by the Arrhenius equation

(Levenspiel 2003). A plot of ln k versus 1/T (Figure 5.3) gives a straight line whose

slope represents the activation energy of extraction ( Ea/R) and whose intercept is

the Arrhenius constant (ln A). Thus, the activation energy and the Arrhenius

constant were calculated as Ea= 9.1803 kJ mol-1 and A =0.147 s-1, respectively.

y = -1104.2x - 2.1793 R² = 0.9852

-5.85

-5.8

-5.75

-5.7

-5.65

-5.6

-5.55

-5.5

-5.45

0.00295 0.003 0.00305 0.0031 0.00315 0.0032 0.00325 0.0033 0.00335 T-1 K

Figure 5.3 Plot of ln k versus 1/T for oil extraction

5.3.3.2 Calculation of activation thermodynamic parameters

The activation thermodynamic parameters were calculated according to the

transition state theory (Topallar and Gecgel 2000). The activation entropy ( S ), the

activation enthalpy H and the activation free energy or Gibb's energy G at

different temperatures were shown in Table 5.4 for each temperature.

118

Table 5.4 Activation thermodynamic parameters at different temperatures

T (Kelvin) H (kJ mol-1 ) S (J mol-1K-1 ) G (kJ mol-1 )303 6.66 -260.98 85.73

313 6.57 -261.25 88.34

323 6.49 -261.51 90.95

333 6.41 -261.76 93.56

5.3.3.3 Calculation of thermodynamic parameters

Thermodynamic parameters ( H, S and G) for the extraction of . C.

pentandra oil using hexane as solvent was estimated. A plot of ln YT vs. 1/T

(Figure 5.4) at 100 min, gives a straight line whose slope represents the enthalpy

change of extraction, - H. Thus, the enthalpy change was calculated to be H =

0.016 kJ mol-1 for oil extraction. The H value obtained was indicating the physical

and chemical nature of the C. pentandra oil extraction process. Other

thermodynamic parameters ( S and G) and the equilibrium constant values for oil

extraction at 100 min are given in Table 4.5 for each temperature.

Figure 5.4 Plot of ln YT versus 1/T for oil extraction

119

According to these results from Table 5.5, the positive value of enthalpy

indicates that the process is endothermic and requires energy during process. In

addition, the negative value of G ( G < 0) at 60°C indicates that there is a

decrease in the free energy. The extraction process using hexane at 60°C is

spontaneous process. The system initially consists of the seeds and hexane, whereas

the oil molecules are extracted from the C. pentandra seeds during the extraction

process and therefore, the entropy of the mixture increases in the course of the

extraction, which is the positive value of entropy change ( S > 0) at 60°C indicates

that the process is irreversible.

Table 5.5 Thermodynamic parameters ( S and G) and equilibrium

constants at different temperatures

T, Kelvin K S, J mol-1K-1 G, kJ mol-1

308 0.8194 5.71 -0.53

318 1.2372 8.32 -0.55

328 1.9333 11.58 -0.57

338 3.6315 17.65 -0.58

5.3.4 Oil Properties and Characteristics of C. pentandra oil

Extracted C. pentandra oil was pale yellow in colour. The oil yield of the

seed was found to contain a mean value of 26.4% (w/w). The physical and chemical

properties of C. pentandra oil are given in Table 5.6. The oil yields were

appreciably higher than soybeans (18 to 22%) and lower than many other oily

vegetables such as palm (40%), rape seed (41%) and sunflower (40%) (Lidefelt

2007). The saponification value of 195 clearly suggests that the oil consist mainly

of medium-chain fatty acids (i.e. C16 and C18). Its initial acid value was

determined as 28.71 mg KOH g-1 corresponding to a FFA level of 14.35%.

Unsaponifiable matter present in the extracted oil was 1.98%, which includes

120

tocopherols, sterols, triterpenic alcohols, hydrocarbons, aliphatic alcohols and

waxes (Lidefelt 2007). The iodine value of the oil is 101.1 which was quite high

and it lies within the values of semi-drying.

Table 5.6 Physical and chemical properties of C. pentandra oil

Properties Values Density 921 kg m-3

Viscosity at 40°C 29.32 mm2 s-1

Water content 0.05 wt%

Acid value 28.71 mg KOH g-1

FFA 14.35% as oleic acid

Saponification value 195

Iodine value 101.9

Average molecular weight of oil 848 g mol-1

Unsaponifiables 1.98 wt%

5.3.4.1 Fatty acid profile

The fatty acid compositions of the C. pentandra oil are given in Table 5.7,

which shows the principal fatty acid profile of linoleic (35.11%), oleic (29.69%),

palmitic (23.20%) and stearic (5.68%) acids. The level of total monounsaturated,

polyunsaturated and saturated fatty acid of the extracted oil was 52.89%, 35.11 %

and 9.44% respectively (Figure 5.5). The high content of unsaturated fatty acid

makes it a stable liquid at room temperature.

121

Figure 5.5 Chromatograph of C. pentandra oil fatty acid

Table 5.7 Fatty acid distribution in C. pentandra oil

Fatty acid Composition wt% Myristic 0.11

Palmitic 23.20

Stearic 5.68

Oleic 29.69

Linoleic 35.11

Arachidic 1.89

Behenic 0.25

Lignoceric 1.51

122

5.3.5 Acid Catalyzed Esterification Pretreatment

C. pentandra oil shows a high FFA content. It was found that the

pretreatment stage considerably reduced the acid value from 28.71 to below 2 mg

KOH g-1. The effect of acid value on reaction time is shown in Figure 5.6. The

experimental results suggested that the acid catalysis esterification occur in three

stages i.e. fast, slow and stationary. In the fast stage, the rate of the reaction was fast

for which the acid value drop down from 28.71 to 3.6 in 40 min and the reduction in

acid value was 87.47%. After 40 min the rate of the reaction slows down up to 60

min, the reduction in acid value was 94.15%. In the stationary stage the

esterification reached equilibrium after 60 min. The acid value was observed less

than 2 mg KOH g-1 at 50 min, but it was desirable to keep a longer reaction time of

60 min to get a much lesser target of 1.68 mg KOH g-1.

0

5

10

15

20

25

30

0 10 20 30 40 50 60 70

Time(mins)

Figure 5.6 Effect of reaction time on acid catalyzed esterification

123

5.3.6 Alkali Catalyzed Transesterification Process

5.3.6.1 Effect of catalyst

In general, from the literature it was observed that alkali catalyst

concentration of less than or equal to 1% is required for a successful conversion of

oils and fats to FAME depending on the type of oil used (Hanh et al 2007, Felizardo

et al 2006, Tomasevic and Siler-Marinkovic 2003). According to the results in

Figure 5.7, the reaction yield was found to be low for the catalyst concentration

0.25 to 0.75%. This is due to the insufficient amount of KOH. It is observed that 1.0

% KOH is optimal enough to get higher conversion of 79%. The addition of excess

amount of catalyst results in the formation of soap, which decreases the yield by

giving rise to emulsification of biodiesel and glycerol phase.

0

20

40

60

80

100

0 0.25 0.5 0.75 1 1.25

Catalyst (wt%)

Figure 5.7 Effect of catalyst on biodiesel production at methanol to oil ratio 6:1, 55°C

reaction temperature, 30 min reaction time and 600 rpm stirrer speed

124

5.3.6.2 Effect of methanol to oil molar ratio

The molar ratio of alcohol to oil is one of the most important variable

influencing the conversion into esters. Although the stoichiometric molar ratio of

methanol to triglyceride for transesterification is 3:1, higher molar ratios are used to

enhance biodiesel conversion (Noureddini et al 1998). The effect of methanol to oil

molar ratio was studied in the range of 3, 6, 9 and 12. Figure 5.8 depicts the effect

of methanol to oil molar ratio in the yield of biodiesel. It has been observed that the

yield of the process increases with increase in molar ratio. The optimum conversion

of 86% was obtained at methanol to oil ratio of 9:1 at constant reaction temperature

of 55°C and reaction time of 30 min with catalyst concentration of 1% (Figure 5.8).

With further increase in molar ratio the yield remains more or less same. The excess

methanol was removed in downstream process.

30

40

50

60

70

80

90

0 3 6 9 12

Methanol oil ratio

Figure 5.8 Effect of methanol to oil molar ratio on biodiesel production at

1% of catalyst, 55°C reaction temperature, 30 min reaction time and

600 rpm stirrer speed

125

5.3.6.3 Effect of reaction temperature

Reaction temperature was found to significantly affect the yield of biodiesel.

At minimum reaction duration (30 min), when the transesterification reaction was

carried out at 45°C, the conversion of biodiesel was only 49%. However, the

conversion reaches the maximum at 93 %, when the reaction temperature was

increased to 65°C. In another words, reaction temperature has a more significant

effect on the conversion at higher reaction temperature than lower temperature.

Figure 5.9 shows the effect of reaction temperature on biodiesel conversion. At

lower reaction temperature, there is insufficient energy to promote extensive

collisions among reactant particles. However, at higher reaction temperature, the

possibility of collision among reactant particles became greater and easily obtains

the necessary activation energy (Al-Widyan and Al-Shyoukh 2002).

40

50

60

70

80

90

100

40 45 50 55 60 65Temparture °C

Figure 5.9 Effect of temperature on biodiesel production at 1% of catalyst,

methanol to oil molar ratio 9:1, at 30 min reaction time and 600 rpm

stirrer speed

5.3.6.4 Effect of reaction time

Figure 5.10 shows the effect of reaction time on biodiesel conversion. The

dependency of reaction time was studied at different time intervals ranging from 15

126

to 60 min. The optimum conversion of 99.5% was observed at 45 min as shown in

Figure 5.11. The increase in reaction temperature increases the reaction rate and

reduces the reaction time as reported by Antonlin et al (2002). Thus if longer

contact timeis provided, most of the reactants will collide and reacts to give higher

yield of biodiesel (Yee et al 2011).

50

60

70

80

90

100

110

10 20 30 40 50 60

Time (mins)

Figure 5.10 Effect of time on biodiesel production at 1% of catalyst, methanol to oil

ratio 9:1, 65°C reaction temperature and 600 rpm stirrer speed

Figure 5.11 H1NMR spectrum of obtained biodiesel at 99.5% conversion

127

5.3.7 Fuel Properties of C. pentandra Biodiesel

Pretreatment followed by alkali-catalyzed transesterification was

successfully used to produce high quality biodiesel. The fuel properties of final

product biodiesel obtained from C. pentandra oil were determined with the aid of

standard methods and were summarized in Table 5.8. The biodiesel properties were

similar to those of ASTM D6751 standard. The exhibited properties also indicate

that the downstream process was adequate.

Table 5.8 Fuel specification of C. pentandra oil compared with ASTM

D6751 standard

Properties Units Test methods Limits C. pentandra oil Biodiesel

Specific gravity --- ASTM D4052 --- 0.876 Flash point oC ASTM D93 130 Min 169 Cloud point oC ASTM D2500 Report 1

Viscosity@40oC mm2 s-1 ASTM D445 1.9-6 4.17

Acid number mg KOH g-1 ASTM D664 0.05 Max 0.036

Carbon residue wt% ASTM D4530 0.05 Max 0.042 Water &

sediments vol% ASTM D2709 0.05 Max 0.031

Copper strip corrosion --- ASTM D130 Number 3

Max 1a

Sulphated ash wt% ASTM D874 0.02 Max 0.01 Phosphorous

content wt% ASTM D4951 0.001 Max 0.0008

Na & K combined Ppm EN 14538 5 Max 4.2 Ca& Mg combined Ppm EN 14538 5 Max 2

Cetane Number --- ASTM D613 47 Min 47

128

5.4 CONCLUSION

This study elaborates the utilization of underutilized, non-edible C.

pentandra oil as a source to produce fuel quality biodiesel. A two-stage process

through acid pre-treatment and alkali catalyzed transesterification for the production

was discussed. The effects of different parameters and the kinetics of the alkaline

catalysts transesterification were studied. The potential use of C. pentandra oil

appears to be promising and have possible applications in the biodiesel production.

The production of C. pentandra biodiesel could be an added value to an

underutilized agricultural product.