Fuel 116 (2014) 794–802

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Extraction of oil from Crotalaria Juncea seeds in a modified Soxhlet

apparatus: Physical and chemical characterization of a prospective
bio-fuel
Ratna Dutta, Ujjaini Sarkar ⇑, Alakananda Mukherjee
Department of Chemical Engineering, Jadavpur University, Kolkata 700032, India

h i g h l i g h t s

 A bio oil is extracted from Crotalaria Juncea (or Sunn Hemp) seeds.
 A modified Soxhlet apparatus is used for the extraction.
 The oil contains high amount of nitrogen and low sulphur.
 The energy content of the bio oil is 34.128 MJ/kg.
 The oil contains oxygenated species, low molecular weight fatty acids.

a r t i c l e i n f o a b s t r a c t

Article history: Oil extraction from crushed Crotalaria Juncea seeds is done by standard solvent extraction method using
Received 1 March 2013 a modified Soxhlet apparatus. Parameters such as size fractions of the crushed seeds, extraction time,
Received in revised form 15 August 2013 aging of seeds, choice of solvents and use of co-solvents along with the primary solvent are optimized
Accepted 20 August 2013
in order to enhance the yield of bio oil production. The bio-oil is characterized using Fourier Transform
Available online 10 September 2013
Infrared Spectroscopy (FT-IR) and Gas Chromatography Mass Spectrometry. Further, some tests for the
elemental analysis for Carbon, Hydrogen, Nitrogen and Sulphur, moisture content, residual carbon and
Keywords:
dynamic viscosity are executed. Some of the basic fuel properties such as specific gravity, iodine number,
Crotalaria Juncea
Bio oil
saponification value, calorific value, flash point, fire point and kinematic viscosity are investigated to
Co-solvents characterize fuel quality of the bio-oil. Though the energy content of the Sunn Hemp oil (34.128 MJ/
Basic fuel properties kg) is slightly lower than that of gasoline (47 MJ/kg) and diesel fuel (44.8 MJ/kg), other fuel properties
GC/MS indicate that there is a great prospect for using this bio oil as a fuel for a diesel engine after some
modifications.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction
Global emissions [1] of CO2 and other harmful gases such as CO,
SOx, NOx and particulates generated by fossil fuel combustion have
led to serious greenhouse effects, acid rain, remarkable human
health problem and deterioration of the global environmental con-
dition as a whole. The price of crude petroleum has gone up to its
highest level. Therefore identification of environment friendly and
renewable sources of alternative energy [2] is one of the thrust
areas of the scientific community.
Biodiesel from bio oil is proposed to serve a lot to mitigate this
problem to a good extent. Previously, biofuel used to be produced
from soybeans, canola oil, animal fat, palm oil, corn oil, waste
⇑ Corresponding author. Tel./fax: +91 33 2414 6378.
E-mail addresses: usarkar@chemical.jdvu.ac.in, abhi_nandan47@rediffmail.com
(U. Sarkar).
cooking oil and jatropha oil [3–6], etc. However, attempts are
now made towards biofuel production from non-food crops, micro-
algae [7,8] (i.e. phytoplankton), macro algae (i.e. sea wood) etc. Re-
cent studies indicate that the aromatic content and type, sulphur
content, extraction temperature, and density are important factors
for emission control [9]. Biomass based agricultural oils have a
great potential towards production of cleaner bio-diesel substitute
[10,11]. Amongst the oil bearing crops already identified, only sun
flower, safflower, soybean, cotton seed, rapeseed and peanut oils
are considered as potential alternative fuels for diesel engines.
These bio-oils are better than diesel fuel in terms of sulphur con-
tent, flash point, aromatic content and biodegradability [12].
Bio oil from Crotalaria Juncea (Sunn Hemp) seed is a promising
new source for production of biofuel. The crop grows best in trop-
ical and subtropical climate. Sunn Hemp is an important bast fibre,
used in green manuring as well as a green fodder crop of India. It is
mainly cultivated in Uttar Pradesh, Madhya Pradesh, Orissa, Tamil

0016-2361/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.fuel.2013.08.056
 R. Dutta et al. / Fuel 116 (2014) 794–802 795

Nadu, Rajasthan and Maharashtra. Sunn Hemp seeds contain mois- Table 1
ture (8.6%), crude protein (34.6%), Fat (4.3%), Starch (41.1%), Fibre Improved varities of Sunn Hemp seeds. (Source: Central Research Institute for Jute
and Allied fibres, Indian Council of Agricultural Reasearch, Barrackpore Kolkata:
content (8.1%), ash (3.3%), oil (12.6%). The oil contains saturated 700120, India).
fatty acids (20.3%), Linolenic acid (4.6%), Linoleic acid (46.8%) and
Oleic acid (28.3%). The endosperm of Sunn Hemp seeds contains Name of the variety Yield (q/ha) Area recommended

thin walled cells, enriched in oil. The average oil productivity of K-12 Yellow 9–12 Sunn Hemp growing belt
Sunn Hemp is 63.6 - 280.0 kg/ha [13] as compared to that of jatro- SH-4 (Sailesh) 15 Sunn Hemp growing belt
SUN 053 (Swastika) 9–12 Sunn Hemp growing belt
pha curcas oil, which is 900 kg/ha [14]. The green biomass poten- Chindwara NA Madhya Pradesh
tial, dry matter accumulation and nitrogen contribution of
Crotalaria Juncea has been reported to be higher than traditionally
grown Sesbania aculeate [15]. The 60 days old crop accumulates
about 170 kg N, 20 kg P and 130 kg K/ha [16]. There is no published 2.1.2. Raw seeds
work on the biofuel production from Sunn Hemp seed oil yet. The K-12 Yellow variety of the Sunn Hemp seeds is used for this
Solvent extraction is one of the traditional techniques of research work. Sunn Hemp seeds, used in our experiments, are ob-
extracting vegetable oil from oilseeds. In this method, pretreated tained from CRIJAF, Barrackpore, Kolkata: 700120, India. The dry
(if necessary) oil seeds are put in contact with a suitable solvent, seed is packed and supplied in a jute bag. It is stored in cold, dry
in its pure form, for extracting the oil from the solid matrix to place. Sunn Hemp seeds are firstly cleaned by separating the other
the liquid phase. Solvent extraction technique [17–19] is one of flora adhered to them, dried and stored carefully in a dry jute sack
the cheapest and most efficient processes, applied to produce oil at ambient temperature. Dried seeds are milled using a domestic
from seeds. Jojoba oil, soybean oil, palm oil, jatropha oil and many kitchen mill (Make: Bajaj, India; Model: Twister Mixer-Grinder;
other oils are produced by this method. Power consumption: 750 W; Speed: 18,000 rpm). Crushed seeds
The main objective of this research work is to produce a bio-oil are stored at 10 °C. Finally, crushed Sunn Hemp seeds are meshed
from Sunn Hemp seeds with the help of a conventional Soxhlet to obtain a grain size of less than 2 mm. The screening data of Sunn
apparatus. The yield of the production is optimized with respect Hemp seeds is given in Table 2.
to size fractions of the crushed seeds, extraction time, aging of
seeds, choice of solvents and use of co-solvents along with the pri- 2.1.3. Solvent
mary solvent. Later this bio-oil is physico-chemically characterized Though Hexane is the most popular solvent for oil extraction
using Fourier Transform Infrared Spectroscopy (FT-IR) for the exis- from seeds, other solvents are also tried to avoid the toxicological,
tence of surface functional groups and Gas Chromatography Mass environmental and safety issues related to its use. Four other dif-
Spectrometry for the identification of the saturated fatty acids. Fur- ferent organic solvents like Ethanol, 2-Propanol, Chloroform, Tolu-
ther, elemental analysis for CarbonÓ, Hyhrogen (H), Nitrogen (N) ene and Hexane [Make: MERCK EMPARTA; Grade: ACS; Purity
and Sulphur (S) is carried out along with the tests for moisture con- (GC): P95%] are purchased from local vendor for using in this re-
tent, residual carbon and dynamic viscosity. Some of the basic fuel search project. The physical properties of the solvents are indicated
properties like specific gravity, iodine number, saponification va- in Table 3.
lue, calorific value, flash point, fire point and kinematic viscosity
are determined to characterize fuel quality of the bio-oil.
2.1.4. Apparatus
In the present study, a batch extractor (Soxhlet Extractor) with
round-bottom flask of capacity 1 lt is used. The cylindrical Soxhlet
extractor (Capacity: 700 ml) is placed onto the round-bottom flask
2. Experimental
containing the extraction solvent. Top of the Soxhlet is equipped
with a bulb-condenser, operated on cooling water. The solvent is
2.1. Materials
heated to reflux. The solvent vapour travels up a distillation arm
and floods into the chamber housing the thimble packed with solid
2.1.1. Sunn Hemp
seeds. The condenser ensures a zero loss of any solvent vapour. The
Sunn Hemp is generally considered to have originated in India,
vapour cools and drips back into the chamber. When the Soxhlet
where it has been cultivated since prehistoric times. However, this
chamber is full, the chamber is automatically emptied through a
is now widely grown throughout the tropics and subtropics. In
siphoning action along a special side arm, with the solvent running
northern India, it is grown in summer and in rainy season while
back, down into flask. This cycle may be allowed to repeat many
in the southern parts of country, where the winter is not so pro-
times until extraction is complete.
nounced, it is grown in rabi season. It is one of the oldest known
fibres in the Indo-Pakistan subcontinent, as mentioned in ancient
Sanskrit literature. Basically, Sunn Hemp is an important bast fibre 2.2. Bio-oil production
used in green manuring and as a green fodder crop of India. Fibre
obtained from Sunn Hemp is dull white in color, strong and dura- In earlier days, hydraulic pressing is usually used to extract oil
ble. It is mainly cultivated in Uttar Pradesh, Madhya Pradesh, Oris- from an oil-containing seed. But, in recent days Soxhlet extractors
sa, Tamilnadu, Rajasthan and Maharashtra. The fibre contains are employed for leaching the oil from any seed.
78.3% a-cellulose, 3.6% hemicelluloses and 4% lignin. The proper-
ties of the fibre makes it ideal for making specialty paper like used
Table 2
in currency notes, tissue papers, etc., but it is not used for textile
Screening data of crushed Sunn Hemp seeds (using B.S. standard sieves).
purposes. Different improved varieties of Sunn Hemp are available,
[as per brochure published by Director, Central Research Institute Mesh Weight Weight Clear opening, Average
size retained (g) fraction (xi) DPi (m) diameter, DP (m)
for Jute and Allied Fibres (CRIJAF), Kolkata – 700 120, India] as
listed in Table 1. The seeds of Sunn Hemp are secondary sources 8 350 0.467 0.002
18 250 0.333 0.00085
of income. It contains 30–35% protein and can be used for making
36 100 0.133 0.00042 0.00129
adhesives for plywood industries. Sunn Hemp seeds are sometimes 52 50 0.067 0.0003
used as herbal medicine.
796 R. Dutta et al. / Fuel 116 (2014) 794–802
Table 3
Physical properties of various solvents used.
Solvents Boiling point (°C) Density (kg/l)
Ethanol 79 0.789
Isopropanol 82 0.785
Chloroform 61 1.498
Hexane 69 0.655
Toluene 111 0.867
Fig. 1. Schematic diagram of a modified Soxhlet extractor with a wire-meshed
cylidrical sample holder put inside a conventional Soxhlet extractor.
2.2.1. Modified Soxhelet apparatus
Solvent extraction in a modified Soxhlet apparatus is done to
extract the Sunn Hemp oil from its seeds. This particular Soxhlet
apparatus (see Fig. 1) consists of a glass extractor, fitted in between
a round bottom flask at the bottom and a bulb condenser at the
top. Inside the glass thimble holder, solid matrix of seeds is placed
within a cylindrical packed bed of wire-mesh. The round-bottom
distillation flask initially contained an extracting solvent and it is
heated up. As the solvent vapour goes up to the condenser, it con-
denses and accumulates inside the extractor. Here, the solvent
comes in contact with the seeds and oil is leached out of the seeds.
When the condensate moves down through the bed of seeds, mass
transfer takes place. However, major amount of mass transfer of oil
from the seeds to solvent occurs when the accumulated solvent
moves up purely due to the hydrostatic pressure head. So, surface
area offered by the bed and the seed-solvent contact time are the
two major factors for the yield of the oil production. For each
run, crushed Sunn Hemp seed is used because it is found that the
whole seeds are unable to produce oil by the solvent extraction
procedure. Generally, ground, naturally dried seeds (moisture
content of 8.6%) are screened using B.S. standard sieves. The aver-
age seed diameter is 1.29  103 m. Instead of placing the seeds
Refractive index (25 °C) Odour Cost of the solvent (INR/1 L)
1.3616 Distinct alcoholic smell 800
1.3772 Distinct alcoholic smell 215
1.4459 Pleasant, sweet odour 363
1.3723 Mild, gasoline like smell 1243
1.4941 Sweet pungent odor 280
Fig. 2. Variation of Sunn Hemp oil recovery with different extraction time.
right into a thimble, a cylindrical wire-mesh (length: 21.5 cm;
diameter: 4 cm and one end closed) is used to hold the seeds.
2.2.2. Choice of solvents
In order to find out the best solvent that would give the maxi-
mum yield, extraction is carried out for predetermined period of
time (2.5 h roughly) using a variety of solvents, for a fixed weight
of seed (150 g) and fixed volume of solvent (700 ml) in the three-
necked distillation flask. The solvents used are hexane, isopropa-
nol, chloroform, ethanol and toluene. Leaching is carried out at
the boiling point of each solvent until a clear liquid is obtained
with an indication of complete leaching of the bio oil. At the end
of a batch run, the packed bed of seeds is removed and afterwards
the oil–solvent mixture is distilled further by simple batch distilla-
tion technique followed by separation of the mixture using a rotary
evaporator apparatus (Make: BUCHI; Model: Rotavapor R-3). The
attached vacuum pump with the rotary evaporator is used to
remove the solvent completely. The extractions are (see Fig. 2) per-
formed in triplicate and their mean values and standard deviations
are calculated. The percentage yield is calculated on the basis of
the following equation with W1 being the amount of seed taken
in gm and W2 is the amount of oil produced in g.
% Oil yield ¼ ðW 2 =W 1 Þ  100:
The leaching process depends mainly on the chemical nature of the
solvent used and type of solute which will be extracted from the so-
lid matrix. Solubility of oil in various solvents is not the same. All
the other conditions like bed height, apparatus capacity, etc. are
kept constant. Five types of organic solvents with different polarity
indices and with a boiling range of 60–115 °C are used for this test.
Recovery of oil from the seeds with different solvents is done for
different time periods to check the extent of oil extracted.
 R. Dutta et al. / Fuel 116 (2014) 794–802
2.3. Analysis of bio-oil
2.3.1. FTIR analysis
FTIR can be utilized to identify some of the functional groups
present in a solid, liquid or gaseous sample. In the present study,
the functional groups of the oil sample are analyzed by using Fou-
rier transform infrared spectroscopy (FTIR) (Make: Shimadzu;
Model: IR Affinity-1). The absorption frequency spectra are re-
corded and plotted as transmittance versus wave number. The
standard IR spectra of hydrocarbons are used to identify the func-
tional groups of the bio oil.
2.3.2. GC/MS analysis
Fatty acid composition of the oil is determined by GC/MS anal-
ysis. A Gas Chromatograph (Make: Thermo Scientific; Model: Trace
GC Ultra; Model No.: 320080111) with TR-Wax MS column (Make:
Agilent), 30 m long, 0.25 mm ID, with film thickness of 0.25 lm is
used. The GC is equipped with EI (electron impact) detector (Make:
Varian, Model: MS 211912, Polaris Q). Helium gas of 99.99% purity
is used as carrier gas with 1 mL/min flow rate and 10 ml s1 linear
velocity. The split ratio is kept at 1:20. Polyethylene glycol coating
is present as binding material of the column and polysiloxane
derivative is present as a high temperature resistant material. Be-
fore charging the oil, the sample is purified by filtration using syr-
inge filter (Make: MILLEX; Model: GV; Pore Size: 0.22 lM). 1 ll of
the sample is injected. The initial temperature of the oven is 80 °C
for a hold-up time of 1 min. In the first ramp, the oven is heated
with a rate of 20 °C min1 to reach 150 °C temperatures and is kept
for 15 min. In the second ramp, the oven temperature reached
240 °C with same heating rate and a holding time of 8 min. The
ultimate temperature of the oven is set at 260 °C. Normally,
70 eV electron impact ionization (EI) mass spectra are collected
from the runs. Data analysis is performed using GC/MSD CHEM-
STATION software. The GC/MS instrument conditions are given in
Table 4.
2.3.3. Physiochemical properties
2.3.3.1. Elemental analysis. The elementary analysis of Sunn Hemp
oil (as well as Sunn Hemp seed) is conducted within a CHNS ana-
lyzer (Make: Elementar, Model: Vario MICRO cube). The samples
are weighed in a tin or silver vessel and loaded in the analyzer.
Each sample is individually flushed with helium (inert gas) to re-
move any atmospheric nitrogen, resulting in a zero blank sampling
process. The gas mixture, after catalytic combustion, is separated
in its components on a column by temperature programmed
desorption (TPD) and subsequently fed into a thermal conductivity
detector (TCD). An on-line PC computes the concentrations of the
Table 4
Operating conditions for the GC/MS (temperature program is followed to analyze
fatty acid methyl esters, FAME). Column: TR wax column 30 m long, 0.25 mm ID, with
film thickness of 0.25 lm.
GC
Carrier gas Helium
Operation mode Constant pressure
Inlet temperature (°C) 80 °C
Oven temperature (°C) 260 °C
Temperature program 80–140–240–260 °C
Injection volume 1 ll
Split ratio 1:20
Run time 32 min
MS
Transfer line temperature (°C) 240 °C
Source temperature (°C) 220 °C
Solvent delay 2 min
Scan mass range 45–500
797
elements from the detector signal and the sample weight, on the
basis of the stored calibration curves.
2.3.3.2. Moisture content. The moisture content of the oil is mea-
sured by using infrared moisture analyzer (Make: Mettler Toledo,
Model: MJ33). The temperature range which is available in this
model is 50–160 °C and minimum sample weight required is
0.5 g. A sample of weight 0.567 g is taken. The temperature of dry-
ing is set at 74 °C and the time required for drying is 2 min 16 s.
2.3.3.3. Conradson carbon residue (CCR) test. The CCR test (as per
ASTMD 189) is performed to determine the amount of residual car-
bon left when the oil is evaporated under a specified condition.
2.3.4. Basic fuel properties
The properties that are tested include specific gravity, iodine
number, saponification number, Gross Calorific value (GCV), flash
point, fire point, pour point, kinematic viscosity and sulphur con-
tent. All the properties are evaluated in accordance with the ASTM
and AOAC standard procedures as shown in Table 5. All the exper-
imental data are statistically analyzed using general linear model.
2.3.4.1. Specific gravity. The density is measured [20] by using stan-
dard pycnometer method.
2.3.4.2. Iodine number. Iodine number is a measure of the degree of
unsaturation of the fuel. The unsaturation in the fatty acid chain is
the main source of thermal instability and reason for causing car-
bon deposits due to burning. The sample is dissolved in solvent and
is mixed with Wijs solution (0.1 mol/l iodine monochloride in gla-
cial acetic acid). After certain time, potassium iodide and water are
added and the liberated iodine is titrated with a sodium thio-sul-
phate solution where starch is used as indicator. Results are then
expressed [21] in grams of iodine absorbed by 1 g of sample.
Iodine value ¼ ½12:69ðB  SÞðN=WÞ
where B is the volume of standard sodium thiosulphate solution re-
quired for the blank in ml. S is volume of standard sodium thiosul-
phate solution required for the sample in ml. N is normality of the
standard sodium thiosulphate solution (0.1 N), W is the weight of
the sample in g.
2.3.4.3. Saponification value. Saponification value is used as an indi-
cator of fatty acid chain length. The value is simply given as a mea-
surement of the ml of KOH required to complete the hydrolysis of
one gram of oil. The high saponification value is associated with
corrosion problems to the critical parts of a diesel engine. The
range of saponification values for various types of bio oils is 160–
190 mg KOH/g. Samples are dissolved in a solvent (1:1 ethanol
and ether) and a defined amount of alcoholic potassium hydroxide
is added. After refluxing the mixture for 30 min, the residual potas-
sium hydroxide is titrated with a hydrochloric acid solution using
Table 5
Standard test methods for the determination of physical, chemical and fuel related
properties.
Physical and chemical property Standard methods
Density ASTM D4052-91
Iodine value AOAC CD1-25
Saponification value AOAC CD3-25
Gross calorific value (MJ/kg) ASTM D3286
Net calorific value (MJ/kg) Calculation
Sulphur content (wt%) ASTM D 7039
Flash point (°C) ASTM D93
Fire point (°C) ASTM D93
Kinematic viscosity at 40 °C ASTM D445
798 R. Dutta et al. / Fuel 116 (2014) 794–802

Table 6
Variation of the global and real yields (%) of Sunn Hemp seed oil for use of various extraction solvents (amount of seed taken: 52 g, amount of solvent taken: 400 ml, bed height:
10.2 cm).

Extraction solvent Oil produced (with fatty acids) (g) Oil produced (without fatty acids) (g) Global yield (%) Real yield (%)
n-Hexanea [ISO-Standard No. 659] 1.2 1.199 2.3077 2.307
Chloroform 5.2 5.198 10 9.999
Toluene 3.7 3.699 7.1154 7.114
2-Propanol 5.9 5.899 11.3461 11.346
Ethanol 2.4 2.399 4.6154 4.615
a
International Organization for Standardization, oilseeds – determination of hexane extract (or light petroleum extract), called ‘Oil Content,’ ISO, Genève, Standard No. 659
(1988).

7.5
Table 7 7
Effect of extraction time on yield of Sunn Hemp seed oil produced (solvent: 2- 6.5 Aged Seed
propanol, amount of seed taken: 52.9 g, amount of solvent taken: 400 ml and bed 6
height: 10.2 cm). Fresh seed
5.5
5
Extraction time (h) % Oil extracted (g oil/g Sunn Hemp seed taken)
4.5
1 5.9 4
2 6.7 3.5
3.5 11.24 3
4.5 12.9 2.5
2
1.5
1
phenolphthalein as indicator. Results [22] are then expressed in 0.5
milligrams of potassium hydroxide consumed for the saponifica- Toluene n-Hexane 2-Propanol Choloroform

tion of a gram of the sample.

Fig. 3. Effect of aging of Sunn Hemp seeds in the yield during the recovery of Sunn
Hemp seed oil (amount of seed taken in bed: 150.5 g, bed height: 19.5 cm, volume
2.3.4.4. Calorific value. The number of units of energy produced by of solvent taken: 700 ml, duration of extraction: 1.5 h).
the combustion of a unit mass of a fuel oil is termed as calorific va-
lue. The calorific value of fuel can be expressed as follows:
Table 8
Effect of solvent (2-propanol) to co-solvent (ethanol) ratio on Sunn Hemp seed oil
2.3.4.5. Higher heating values at constant volume (dry basis). Com- recovery [real yield of the bio-oil = 11.346% using isopropanol (solvent) without using
bustion products consist of oxygen, nitrogen, carbon dioxide, sul- any co-solvent).
phur dioxide and liquid water. Condensing the water vapour
Solvent:Co solvent (by volume) Oil extracted (g oil/g seed) (wt%)
increases the amount of energy recovered from the oil. The water
1:1 3.564
vapour can arise from two sources, namely the moisture content
5:1 1.149
of the oil and the formation of water from the hydrogen in the oil. 7:1 2.43
10:1 4.067
2.3.4.6. Low heating value at constant pressure (dry basis). The low 11.67:1 3.11
heating value in wet basis is perhaps the most practical measure 13:1 2.72
20:1 2.22
of energy content. The higher heating value or gross calorific value
(HHV or GCV) is measured by bomb calorimeter. The bio oils are
usually directly measured for their high heating values (HHV), preheating of bio-oils is useful to reduce their viscosities and to im-
the LHV can then be calculated using the following equation [23]: prove pumping and atomization.
LHV ¼ HHV  218:3  H%ðwt%Þ ðkJ=kgÞ

3. Results and discussion

2.3.4.7. Flash point, fire point. The flash point and fire point temper-
ature of a liquid fuel is defined as the lowest and highest temper-
atures for the vapour above the liquid to be ignited when exposed
to a flame. The flash points of bio-oils are usually measured with
closed-cup testers and vary in the range of 40–70 °C or above
100 °C, depending on the contents of the light organic volatiles. It
is usually difficult to measure the flash points at 70–100 °C, due
to vigorous evaporation of water, which will inhibit the ignition
of the fuel vapour. The flash point and fire point of the sample is
determined with an open cup tester. Fuels having high flash points,
have lower volatilities, causes delayed ignition, arising difficulties
during the startup of the engine.
2.3.4.8. Kinematic viscosity. The kinematic viscosity is measured
using a glass capillary viscometer. Its variations, in the tempera-
ture range of 40–100 °C are recorded. A water bath is used to con-
trol the measuring temperature with stability within ±0.1 °C.
Generally, bio-oils have wide range of viscosities. Moderate
3.1. Yield of oil extraction
3.1.1. Effect of use of different solvents
Table 6 indicates the variation of the Global1 and Real yields of
Sunn Hemp seed oil using various extraction solvents. The average
Global (11.3461%) and the Real yields (11.346%) are both found to
be the highest for 2-propanol, a polar solvent. This result shows that
Sunn Hemp oil yield is better with polar solvents (like chloroform,
2-propanol, and ethanol) as compared to those with non-polar sol-
vents like hexane and toluene. Isopropanol is the best solvent in this
research since its cost is relatively low and it leached the highest
percentage of oil, while chloroform gave comparable percentage of
oil yield with isopropanol, but its cost is relatively high. Toluene
has a high boiling point so it needs more heat to form vapour in
any distillation recovery process compared to other solvents. Ethanol
1
Real Yield = Global Yield – (% of all fatty acids per gram of extract).
 R. Dutta et al. / Fuel 116 (2014) 794–802 799

Fig. 4. FTIR analysis of Sunn Hemp seed oil for the determination of various functional groups present.

Table 9
FTIR based functional groups of the bio oil.

Frequency range (cm1) Frequency (cm1) Group Class of compound

2980–2800 2924, 2854 CAH stretching Alkanes
1850–1650 1743 C@O stretching Ketones, carboxilic acids, aldehydes, esters
1650–1580 1604 C@C stretching Alkenes
1550–1490 1494 ANO2 stretching, NAH bending, Aromatic C@C stretching Nitrogeneous compounds
1470–1350 1462, 1369 CAH bending Alkanes
1300–950 1161, 1082 CAO Stretching, OAH bending Alchols, ethers
915–650 725, 694 CAH in-plane benching Aromatic compounds
Less than 605 600 Various inorganic compounds

Fig. 5. Total GC based chromatogram of Sunn Hemp seed oil.

leached the least amount of oil and it is quite expensive. Hexane is by the Bligh and Dyer method (see Appendix-I). A standard
not very suitable for the recovery of oil probably because of its procedure is followed in this method for the extraction of oil and
non-polar character. Overall, isopropanol is considered as the best the yield is found to be 2.224 (%) in comparison to standard Soxhlet
solvent for this work. The yield of Sunn Hemp oil is also found out extraction with the combination of the same solvents used in this
800 R. Dutta et al. / Fuel 116 (2014) 794–802

Table 10 3.2. FTIR analysis

Elemental composition of Sunn Hemp seed and the Sunn Hemp seed oil.

Element Sunn Hemp seed Sunn Hemp seed oil The FTIR analysis is presented in Fig. 4. The functional groups
Carbon (wt%) 39.68 77.3 identified from FTIR spectrum of the bio oil are shown in Table 9.
Hydrogen (wt%) 6.467 10.414 Band 1 (2494, 2854 cm1) is mainly due to the alkanes. Band 2
Nitrogen (wt%) 6.19 1.48
(1743 cm1) is mainly due to un-conjugated alkyl aldehydes and
Sulphur (wt%) 0.170 0.0
Oxygen (wt%) 47.493 10.806 alkyl esters.
H/C molar ratio 1.955 1.616
O/C molar ratio 0.897 0.104

3.3. Fatty acid composition of Sunn Hemp seed oil

method in the same proportion (7.547%). Bligh and Dyer method
showed much lower yields due to the losses during the separation
of the two phases, namely aqueous layer (methanol–water) and
the organic layer (chloroform).
3.1.2. Effect of extraction time
Extraction time is varied using the most suitable solvent (i.e. 2-
propanol), and corresponding variations in yields are presented in
Table 7 and Fig. 2. The equilibrium extraction time is approxi-
mately 2.5–3 h.
The total ion chromatogram from GC/MS analysis is illustrated
in Fig. 5. Based on GC analysis, only two fatty acids are identified
and quantified. Surprisingly, both fatty acids are saturated. The
fatty acids are Hexadecanoic acid and 14-methyl Heptadecanoic
(details are given below). Oil with polyunsaturated fatty acids
with more than three double bonds appears to exhibit inferior
oxidative and thermal stability. Hence, it can be concluded that
the oil used in this study has excellent oxidative and thermal
stability.

Hexadecanoic Acid, methyl ester

C17H34O2, Palmitic Acid

CAS # 112-39-0, Entry # 35245

RT: 20.82 mins

14, Methyl Heptadecanoic Acid,

methyl ester C19H38O2,

CAS # 57247-45-0, Entry # 35407

RT: 23.54 mins

3.1.3. Effect of aging of seed 3.4. Physiochemical properties

The effect of aging of the seeds in the yield of oil produced is
also tested for the same variety of seeds (refer Fig. 3).
3.1.4. Effect of use of a co-solvent and varying solvent to co-solvent
ratio
After selection of the most suitable solvent, the effect of use of
co solvents, on the performance of the extraction process, is
studied. Use of a second solvent in the extraction process may
increase the yield of extraction in two ways; acting in parallel
with the main solvent or acting against the main solvent.
Experimental results show that ethanol is a suitable co-solvent
for this purpose with respect to the recovery of oil. This is
because the polarity index of ethanol (5.2) is more than that of
2-propanol (3.9) and these solvents are similar in chemical nat-
ure, having very close values of boiling points and almost the
same functional groups. Various ratios of solvent to co-solvent
are taken and results are presented in Table 8. As the ratio
increases, initially the wt% of oil recovered increases, except for
an equal volume ratio. However, the yield decreases on reaching
an optimum ratio of 10:1.
3.4.1. Elemental composition
The elemental compositions of Sunn Hemp seed and the bio oil
are presented in Table 10. The bio oil has higher contents of carbon
and hydrogen than the seed, but lower contents of nitrogen and
sulphur. It is important to note that the nitrogen content of the
bio oil is much higher than that derived from Sunn Hemp seed
(usually below 2%) [11]. High nitrogen content is responsible for
higher nitrogen oxides emissions during the combustions of bio
oil [17]. However, the nitrogen content of the bio oil is feedstock
dependent and difficult to reduce.
3.4.2. Moisture content
The measured water content of the bio oil is 2.12 wt%. Though
water content of the bio oil decreases from its feedstock (water
content Sunn Hemp seed: 8.6 wt%), but it is taken under consider-
ation before using the bio oil as fuel. The presence of water has
both negative and positive effects on the storage and utilization
of bio oils. It lowers heating values, and may cause phase separa-
tion of bio oils. Moreover, it increases ignition delay, and reduces
combustion rates. On the other hand, it helps to reduce viscosity
and facilitate atomization. Furthermore, it is beneficial for reducing
 R. Dutta et al. / Fuel 116 (2014) 794–802 801

Table 11 metering of Sunn Hemp seed oil results in the delivery of greater
Basic fuel properties of the bio oil, as compared to jatropha curcas Oil. mass of fuel. Since, vegetable oil has lower energy content on both
Fuel property Value (Sunn Hemp) Value (jatropha curcas) volumetric and mass basis, the actual energy delivered is less than
Density (kg/l) 0.9895 0.90317 diesel fuels.
Iodine value 92.24 ± 0.08 103.62 ± 0.07 The tolerable range of iodine value for vegetable oil is 105–130. In
Saponification value 210.375 ± 0.59 193.55 ± 0.61 an internal combustion engine where high temperature is very com-
Gross calorific value (MJ/kg) 34.128 39.77 mon, the polymerization process can get accelerated and the engine
Net calorific value (MJ/kg) 31.855
Sulphur content (wt%) Less than 0.1% 0.04
can quickly become gummed up with the polymerized fatty acid
Flash point (°C) 95 °C 235 °C methyl esters (FAME) leading to the formation of deposits. Iodine
Fire point (°C) 100 °C number is a measure of the degree of unsaturation in fuel. Unsatura-
C/H (wt%) 7.42 13.11 tion in fatty acid chains is the main source of instability in bio oils. If
Kinematic viscosity (mm2/s) two or more double bonds are present, they suffer from mutually
At 25 °C 51.91 activating effect. To increase stability, addition of additives may be
At 40 °C 24.51 41.51
suggested. Iodine number greater than 115 is not acceptable [25].
At 60 °C 13.62
At 80 °C 8.58 Sunn Hemp oil has very low iodine number than other vegetable
At 100 °C 5.87 based bio oils which refers better oxidation stability.
The saponification value of Sunn Hemp oil (210.375 mg KOH/g)
is high as compared to any other bio oil. This value reflects the
pollutant emissions during combustion. Water lowers and presence of lower molecular weight fatty acids in the glycosides.
balances the temperature profiles evenly in ignition chambers that The higher calorific value of the Sunn Hemp oil (34.128 MJ/kg)
are important for suppressing NOx formation. among the bio oils family as compared to GCVs of other bio oils
(GCVricehusk = 16.5 MJ/kg [14]; GCVstd diesel = 44.4 MJ/kg) makes it
3.4.3. CCR value attractive to use it as fuel. Still it requires up-gradation for a tech-
CCR value of the bio oil is 47%, which is much larger as com- no-economical feasible conversion to bio diesel etc.
pared to any petroleum fuel. CCR is a measure of the tendency of The high flash and fire point temperatures of the bio oil indicate
a fuel to form carbon deposits during combustion and indicates comparatively less fire hazard on storage. Compared to other veg-
the relative coke forming tendencies of heavy oil. Carbon-rich fuels etable oil, it has very low flash point and fire point which indicates
are more difficult to burn and have specific combustion character- high fire hazards under normal condition.
istics which lead to the formation of soot and carbon deposits. The viscosity of the bio oil is relatively high as compared to a
Since carbon deposits are a major source of abrasive wear, the diesel fuel. Generally, kinematic viscosity for vegetable oil ranges
CCR value is an important parameter for a diesel engine. from 30 cSt to 40 cSt at 40 °C. Sunn Hemp oil has lower kinematic
viscosity (24.51 cSt at 40 °C) as compared to other vegetable oils. A
diesel engine requires a viscosity of 2–4.5 cSt viscosity.
3.5. Basic fuel properties
4. Conclusion
Basic fuel properties of the bio-oil are shown Table 11. A
summary of the physical and chemical properties of Sunn Hemp The bio oil extracted from Sunn Hemp seeds is very much sim-
oil and other seed oils used for biodiesel production are listed in ilar to other bio oils in chemical composition and basic fuel prop-
Table 12 [24–27]. erties. It contains high amount of nitrogen and some amount of
Density affects the breakup of fuels injected in engine cylinders. inorganics which are feedstock dependent. Overall evaluation of
As density increases, compression ratio and amount of fuel the experimental data gathered from various experiments (carried
injected, in terms of mass (for the same fuel volume of input) out till date) yields the following conclusion:
increase. Specific gravity directly affects injection timing and injec-
tion spray patterns. Heating value of the oil and the engine exhaust  The energy content of the bio oil is 34.128 MJ/kg which is
emissions are also functions of density. Higher density fuels de- slightly lower than that of gasoline (47 MJ/kg), diesel fuel
crease atomization and mixing, and are associated with higher par- (44.8 MJ/kg) or petroleum (42 MJ/kg), but higher than that of
ticulate matter and NOx emissions than the low density fuels. coal (32–37 MJ/kg).
Density increases with the number of carbon atoms and the degree  The H/C ratios obtained from the Sunn Hemp oil show that the
of saturation. Density of Sunn Hemp seed oil is found to be fractions are quite similar to some of the currently utilized
0.9895 kg/l. Although the density of Sunn Hemp oil is similar to that transport fuels.
of various other vegetable oils, but it is found higher than that found  FTIR analysis shows that the Sunn Hemp oil is highly dominant
in the Indian regulation (0.82–0.86 kg/l) [28]. Therefore, volumetric with oxygenated species.

Table 12
Properties of Sunn Hemp oil and other vegetable oils.

Vegetable oil Kinematic viscosity at 40 °C Energy content (MJ/ Flash point Density Sulphur Iodine Saponification number
(mm2/s) kg) (°C) (kg/l) (wt%) number (mg KOH/g)
Corn 34.9 39.5 277 0.9095 0.01 122.6 187–195
Cottton seed 33.5 39.5 234 0.9148 0.01 105.7 189–198
Peanut 39.6 39.8 271 0.9026 0.01 NA 187–196
Rapeseed 37.0 39.7 246 0.9115 0.01 130 168–181
Sesame 35.5 39.3 260 0.9133 0.01 106.6 187–195
Soyabean 32.6 39.6 254 0.9138 0.01 112.5 189–195
Sunflower 33.9 39.6 274 0.9161 0.01 125.5 188–194
Tobacco oil 27.7 39.4 220 0.9175 0.006 135.0 193
Sunn Hemp 24.51 34.128 95 0.9895 60.01 92.24 210.375
oil
802 R. Dutta et al. / Fuel 116 (2014) 794–802
 Studies reveal that the sulphur content of this bio oil is much
less, thereby making it environment friendly, producing less
pollutant and less corrosion.
 GC–MS analysis of the oil indicates the presence of low molec-
ular weight fatty acids with no unsaturation.
 Since, water content and CCR value of the Sunn Hemp oil is very
high, so upgrading of this bio-oil is necessary to improve its
quality before charging to diesel engine.
Sunn Hemp oil may become economically competitive in addi-
tion to being environmentally desirable. Additional research in this
area is therefore recommended as follows:
 Distillation and thermo gravimetric analysis (TGA) of the Sunn
Hemp oil to confirm the contents of low-boiling and high-boil-
ing compounds. This analysis will help us to characterize the
evaporation, decomposition and combustion kinetics.
 Determination of the thermal stability of the Sunn Hemp oil by
observing the change of viscosity and water content of the oil
with temperature.
 Determination of pour point, cloud point temperature and sur-
face tension of the Sunn Hemp oil are necessary to categorize it
as fuel.
 Determination of fuel properties of the fuel blends.
 Engine tests of these blends performed in the short and long
term,
 Techno-economic analysis for using it as a fuel.
Acknowledgments
The authors thank the National Test house, Kolkata, India;
Jadavpur University, Kolkata, India and Bose Institute, Kolkata, In-
dia for their support during some of the tests.
Appendix Bligh. and Dyer extraction of total lipids
It is a rapid method for total lipid (fat + oil + phospho-
lipid + wax + steroids + other compounds) extraction and
purification.
Standard. procedure
1. 100 g of Sample and 80 g water are to be homogenized with
100 ml of chloroform and 200 ml of methanol (monophasic
system).
2. Solution is to be re-homogenized with an additional 100 ml of
chloroform, followed by 100 ml of distilled water.
3. Solution is to be filtered under suction; the final biphasic sys-
tem is to be allowed to separate into two layers.
4. Upper layer, aqueous phase (methanol + wáter) will contain
some polar species like acyl-CoAs, should be discarded.
5. Lower organic layer (contains TAGs, membrane lipids and other
neutral lipids) is to be transferred into a new container and then
homogenized using 100 ml chloroform.
6. Lipid content is then to be determined gravimetrically after
evaporating the solvent using a rotary evaporator.
7. The binary solvent systems successfully used in the Bligh and
Dyer method have been for total lipid extraction [29,30],
because they are more polar organic solvents. In contrast to
more nonpolar solvents, like petroleum benzene or hexane,
these binary systems are also able to extract more polar compo-
nents, such as phospholipids [31].
Bligh and Dyer method Soxhlet method
Oil yield (wt.%) 2.224 7.547
References
[1] Bilgen S, Keles S, Kaygusuz A, Sari A, Kaygusuz K. Global warming and
renewable energy sources for sustainable development: a case study in
Turkey. Renew Sustain Energ Rev 2008;12:372–96.
[2] Pahl G. Biodiesel-growing a new energy economy. 2nd ed. White River
Junction Vermont: Chelsea Green Publishing Company; 2008.
[3] Zhang y, Zdube MA, Mclean DD, Kates M. Biodiesel production from waste
cooking oil: economic assessment and sensitivity analysis. Bioresour Technol
1990;3:229–40.
[4] Mustafa B. Potential alternatives to edible oils for biodiesel production – a
review of current work. Energ convers Manage 2011;52:1479–92.
[5] Morshed M, Ferdous K, Khan MR, Mazumder MSI, Islam MA, Uddin MT. Rubber
seed oil as a potential source for biodiesel production in Bangladesh. Fuel
2011;90:2981–6.
[6] Nabi MN, Hoque SMN, Uddin MS. Prospect of Jatropha curcas and Pithraj
cultivation in Bangladesh. J Eng Technol 2009;7(1):41–54.
[7] Abubakar LU, Mutie AM, Kenya EU, Muhoho A. Characterization of algae oil
(oilgae) and its potential as biofuel in Kenya. J Appl Hytotechnol Environ Sanit
2012;4:147–53.
[8] Chisti Y. Biodiesel from microalgae. Biotechnol Adv 2007;25:294–306.
[9] Kuljuka T, Savela K, Peltonen K, Rantanen L, Mikonnen S. Effect of fuel
reformulation on diesel particulate emissions—application of DNA adduct test.
SAE paper 982650; 1998.
[10] Marshall W, Leon G, Howell S. Engine exhaust emissions evaluations of a
Cummins L10E when fueled with biodiesel blend. SAE paper 952363; 1995.
[11] Schmidt K, Gepren VP. The effect of biodiesel fuel composition on diesel
combustion and emissions. SAE paper 961086; 1996.
[12] Goering E, Schwab W, Daugherty J, Pryde H, Heakin J. Fuel properties of eleven
vegetable oils. Trans ASAE 1982;25:1472–83.
[13] USDA-NRCS. ‘Tropic Sun’ Sunn Hemp release brochure. Hoolehua Plant
Materials Center, Molokai, HI. <http://www.plant-materials.nrcs.usda.gov/
pubs/hipmcrb8433.pdf>; 2009 [accessed 31.08.12].
[14] Akbar E, Yaakob Z, Kamarudin SK, Ismail M, Salimon J. Characteristic and
composition of Jatropha Curcas oil seed from Malaysia and its potential as
biodiesel feedstock. Eur J Scient Res 2009;29(3):396–403. ISSN 1450-216X.
[15] Aulakh MS, Khera TS, Doran JW, Singh K, Singh B. Yields and nitrogen
dynamics in a rice–wheat system using green manure and inorganic fertilizer.
Soil Sci Soc Am J 2000;64(5):1867–76.
[16] Kolar JS, Grewal HS, Singh B. Nitrogen substitution and higher productivity of a
rice–wheat cropping system through green manuring. Trop Agric (Trinidad)
1993;70(4):301–4.
[17] Luque de Castro MD, Priego-Capote F. Soxhlet extraction: past and present
panacea. J Chromatogr A 2010;1217:2383–9.
[18] Carrin ME, Crapiste GH. Mathematical modeling of vegetable oil-solvent
extraction in a multistage horinzontal extractor. Food 2008;85:418–25.
[19] Abu-Arabi MK, Allawzi MA, Al-Zoubi HS, Tamimi A. Extraction of jojoba oil by
pressing and leaching. Chem Eng J 2000;76:61–5.
[20] AOAC 920.212. Specific gravity (apparent) of oils, pycnometer method; 1984.
p. 72.
[21] AOAC 920. 159. Iodine absorption number of oils and fats; 1984. p. 76.
[22] AOAC 920.160; Saponification number of oils and fats; 1984. p. 78.
[23] Demirbas A. Fuel properties and calculation of higher heating values of
vegetable oils. Fuel 1998;77:1117–20.
[24] Goering E, Schwab W, Daugherty J, Pryde H, Heakin J. Fuel properties of eleven
vegetable oils. Trans ASAE 1982;25:1472–83.
[25] Giannelos PN, Zannikos F, Stournas S, Lois E, Anastopoulos G. Tobacco seed oil
as an alternative diesel fuel: physical and chemical properties. Ind Crops Prod
2002;16:1–9.
[26] Giannelos PN, Sxizas S, Lois E, Zannikos F, Anastopoulos G. Physical, chemical
and fuel related properties of tomato seed oil for evaluating its direct use in
diesel engines. Ind Crops Prod 2005;22:193–9.
[27] Oasmaa A, Leppamaki E, Koponen P, Levander J, Tapola E. Physical
characterization of biomass-based pyrolysis liquids. Espoo Technical
Research Centre of Finland: Application of standard fuel oil analyses; 1997.
[28] jIS 1460:1995. Specifications, fuel regulation: India, available from <http://
www.dieselnet.com/standards/in/fuel.php> ;2000.
[29] Eder K, Reichlmary-Lais AM, Kirchgessner M. Studies on the extraction of
phospholipids from erythrocyte membranes in the rat. Clin Chim Acta
1993;219:93–104.
[30] Marmer WN, Maxwell RJ. Dry column method for the quantitative extraction
and simultaneous class separation of lipids from muscle tissue. Lipids
1981;16:365–71.
[31] Maxwell RJ. Determination of total lipid and lipid sub-classes in meat and
meat products. J Assoc Off Anal Chem 1987;70:74–7.