This document summarizes research on using the Terminalia belerica plant for various biotechnological applications. Key points: - The plant's seeds can be used to produce a new edible oil, comprising 37% of the dry kernel weight. Analysis found the oil contains fatty acids similar to olive oil. - Byproducts of oil extraction include an oilcake containing high levels of nitrogen (8.34%) and proteins (60% digestible) suitable for use as biofertilizer. - Other byproducts include tannins extracted from the fruit pulp, suitable for use in leather/medicine, and antioxidants like gallic acid in the seed coat with potential to preserve vegetable oils.

1. Biotechnological applications in agriculture: A new source of edible
oil and production of biofertilizer and antioxidant from its by-products
D. Bera b
, D. Lahiri b
, Antonella De Leonardis a
, K.B. De c
, A. Nag c,*
a
Department of Agricultural, Food, Environmental and Microbiological Science and Technologies (DiSTAAM), University of Molise, via De
Sanctus, 86100 Campo basso, Italy
b
Rural Development Centre, Indian Institute of Technology, Kharagpur, India
c
Natural Product Laboratories, Chemistry Department, Indian Institute of Technology, Kharagpur 721 302, West Bengal, India
Received 14 September 2005; received in revised form 28 November 2006; accepted 29 November 2006
Available online 21 January 2007
Abstract
Terminalia belerica Roxb (Combretaceae) known as bahera, found abundant in tropical Asia, is a source of new edible oil (37% by dry
weight of kernel), biofertilizer, tannin and antioxidant. The oilcake contains high amount of nitrogen (8.34%). On biochemical evaluation
form the oil cake it is evident that about 60% NaCl extractable protein is digestible which can be converted into biofertilizer or some
useful fodder. The extractable high quality of tannin present in fruit pulp can be used in the leather industry and herbal medicines.
The different processes for the extraction of tannin have been discussed. The maximum tannin was extracted at 135 °C over 12 h with
shaking. The seed coat contains high amount of gallic acid (3.2 mg/ml) which showed good antioxidant properties on different vegetable
oils.
Ó 2007 Elsevier Ltd. All rights reserved.
Keywords: Bahera oil; Antioxidant; Biofertilizer; Tannin
1. Introduction
The scarcity of edible oils in India (annually require-
ments 5 million tones approximately) and other Asian
countries has jeopardized the economy to a great extent.
At present, production of oilseeds in some Asian countries
cannot meet the demand. A breakthrough is required to
find a new source of edible oil to meet this demand.
Terminalia belerica Roxb (Combretaceae) locally known
as bahera is one such abundantly available oil bearing fruit
in tropical Asia. Generally in India the fruit extract is used
against myocardial necrosis or hypoglycemic activity (Kar,
Choudhury, & Bandyopadhyay, 2003).
Bahera plant can tolerate moderate drought and heavy
rainfall. Bahera plant is able to mature in 6–8 years and
yields about 500 kg of raw fruits annually. Some informa-
tion (Chopra, Mayer, & Chopra, 1976; Nag & De, 1995)
regarding composition of oil from the seed kernel has been
reported. But in this paper, we have envisaged new uses of
bahera seeds as edible oil and production of biofertilizer,
tannin and antioxidant from its by-products.
2. Materials and methods
Bahera fruits (120; total weight 570 g), just after harvest
were washed with dilute potassium permanganate and cop-
per sulfate solution (approximately 1% (w/v)) each to
reduce natural fungal growth. The fruits were then dried
in the sun (final weight 500 g). The seed kernels were col-
lected by cracking the hard nut of bahera seed. The seed
kernels (59 g) were dried at 70–80 °C. After drying, seed
kernels were ground into powder by a ball mill. The weight
of the dried powder was about 51 g. The oil was extracted
from the dried and powdered meal with hexane (bp 40–
60 °C) seven times, using a soxhlet apparatus. The oil
0260-8774/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jfoodeng.2006.11.034
*
Corresponding author.
E-mail address: ahinnag@chem.iitkgp.ernet.in (A. Nag).
www.elsevier.com/locate/jfoodeng
Journal of Food Engineering 81 (2007) 688–692

2. was recovered from hexane after distilling hexane in a
nitrogen atmosphere. The oil was obtained (37%) by weight
of dry kernel of the seed (Bera, Lahiri, & Nag, 2004).
2.1. Determination of fatty acids composition of oil
The composition of oil was determined by a GC capil-
lary column, Supelcowax 10 (Bellefonte, PA, USA) of
30 m  0.32 mm, film 0.5 lm connected to a PC in action
with the Chrome-Card 1.2 software (Thermoquest Instru-
ment, Rodano, MI, Italy). Experimental conditions to
determine fatty acids were as follows: carrier gas He at
50 kPa; split injection system with a splitting ratio 1:40;
FID detector system; injector and detector temperatures
250 °C; oven temperature 240 °C; injected quantity 1 ll;
fatty acids were converted to esters by cold methylation
with a solution of KOH (2 N) in methanol.
Saponification, iodine and acetyl values of the oil were
determined as discussed in our previous paper (Nag &
De, 1995).
2.2. Determination of digestibility of the protein
Defatted oil cake (0.5 g) and pepsin (0.2 g) were mixed
with 50 ml of 0.1 (N) HCl in a 100 ml conical flask plugged
with cotton. The mixture was kept in incubator for 48 h at
37 °C followed by the addition of 5.5 ml of 10 (N) NaOH
solution. Trypsin (0.2 g) was added in the mixture and diges-
tion was continued for 48 h. At the end of tryptic digestion,
the mixture was filtered and aliquots were taken for determi-
nation of total nitrogen. A similar control was performed
where only the enzymes were used. Changes in the solubility
of the protein in oil cake was determined by extracting oil
cake (5 g) with 100 ml of 10% NaCl solution and the mixture
was kept for 12 h in a refrigerator. It was centrifuged and the
total nitrogen in aliquot of the extract was determined
against standard bovine serum albumen. The nitrogen pre-
cipitated by trichlroroacetic acid was also determined in ali-
quots of the original NaCl extract (Table 3) (Nag, 2006).
2.3. Extraction of tannin and determination of its structure
Hydrolysable tannin was obtained from the seed coat
dust of Bahera. The tannin has been first separated from
Bahera seed by solvent extraction using water, ethanol
and methanol sequentially under shaking condition at dif-
ferent temperatures and time following a method reported
elsewhere (Hagerman & Butler, 1980).
In a 100 ml conical flask bahera seed coat dust (5 g) and
50 ml distilled water were taken and kept in an incubator at
30 °C. Tannin in the solution was detected by Follin–Cio-
calteu method with and without shaking at different inter-
vals of time at 30 °C. Similarly at different time and
temperature, extracted tannin value was calculated. At
the end, tannin value (Fig. 1) was measured by changing
all three parameters, i.e., soaking time, shaking time and
temperature and is noted in Table 5.
This extracted tannin was dried in nitrogen environment
and then subjected to anhydrous methanol at room tem-
perature 25 °C. This methanol solution was collected, dried
and acetylated by using pyridine and acetic anhydride (1:2
volume). This acetylated product was separated by thin
layer chromatography. Separated fractions were used for,
FTIR (Thermo Nicollet FTIR Spectrophotometer Nexus
870 Model using KBr disk), NMR (1 H 200 MHz) (Bruker
200 MHz/52 MM) analysis and element (C,H,N) detection
(Perkin Elmer Instrument 2400 Series II CHNS/O).
2.4. Determination of phenol in extract using UV and HPLC
About 10 g of dried bahera seed coat powder with 50 ml
of methanol–water 80:20 (v/v) was placed in sonicator for
15 min. Phenols were recovered by three replicate extrac-
tions with ethyl acetate (20 ml each time). The solvent
was removed by rotavapor; dry residue was dissolved in
25 ml of distilled water. Bahera extract was filtered by
Whatman No. 1 filter paper. Total Phenols were deter-
mined by the Folin–Ciocalteu spectrophotometer method
taking gallic acid as standard and expressed in mg/ml of
gallic acid.
Phenolic composition was analysed by HPLC at the fol-
lowing conditions: Instrument Model ProStar 230 (Varian,
Musgrave, AUS) with an UV–visible spectrophotometer,
equipped with a Luna 5u phenyl–hexyl 250 Â 4.6 mm col-
umn from Phenomenex (USA). Mobile phase A: methanol;
mobile phase B: 2% acetic acid; eluent flow, 1 ml min-1;
inject quantity, 20 ll; fixed wavelength, 280 nm. Elution
program: A(%)/B(%): 0 min 95/5; 10 min 75/25; 20 min
50/50; 30 min 100/0; 40 min 5/95.
2.5. Evaluation of antioxidant effect
Antioxidant effect of the phenolic extract was compared
with that of control 100 ppm of gallic acid in different vari-
eties of oils (Flax seed, Cod liver and lard). The character-
istics of the three oils are shown in Table 4.
2 4 6 8 10 12
0.0
0.5
1.0
1.5
2.0
T
S
SS
SST
Amt.ofTannin(gm)
Observations
Fig. 1. Extraction of tannin with changing all the parameters (tempera-
ture T, soaking S, soaking and shaking SS and soaking, shaking and
temperature SST.
D. Bera et al. / Journal of Food Engineering 81 (2007) 688–692 689

3. The dose of phenol added was expressed in mg kgÀ1
.
The doses added were ranged between 50 and 400 mg kgÀ1
.
Antioxidant effects were evaluated by Rancimat
Methrom Instrument (AG, Herisau, Switzerland) MOD
730 under 120 °C temperature and 20 L hÀ1
air flow. Pro-
tection factor is calculated as ratio between induction time
of oil with and without the antioxidant.
3. Results and discussion
From the experimental results, the composition of fatty
acids was as follows, palmitic acid 18.25%, stearic acid
8.20%, oleic acid 50.20%, linoleic 10.8% and others 6.6%
(Table 1). The absence of hydroxyl and other such objec-
tionable groups in fatty acids were also noted. The fatty
acids composition of bahera oil has also been compared
with olive oil (Table 1).
One very interesting point to note here is that the oil has
only about 10% of the constituent (stearic) saturated fatty
acid. The iodine value is fairly high and the free fatty acid is
a little higher than olive oil.
The nitrogen value (8.24%) of oil cake was determined
(Hilditch, 1979) by Kjeldhal’s method. Total potassium
(0.44%) and phosphorous (0.19%) in oil cake were deter-
mined by flame photometry and colorimetric instrument.
From Table 2, we found that no oil cakes generally used
as biofertilizer have such higher amount of nitrogen value.
On biochemical evaluation it is also evident about 60%
NaCl extractable protein (Table 3) in cake is digestible.
As a trial to use oil cakes (50 g) as biofertilizer, we applied
to winter vegetable such as Spinach grown in pots. It has
been observed that plants in the pots which contained
bahera oil cakes had healthy and bushy plants with low
level of incidence of insects attack than the plants grown
without oilcakes.
In Fig. 1, the amount of tannin extracted against the
temperature (°C) has been shown. It has been observed
that at a particular temperature (80 °C), maximum
amount of tannin was extracted and with increase in tem-
perature, there was no changed on the amount of tannin
extraction. It has also been observed that when the three
parameters, i.e., soaking time, shaking time and temperature
changes simultaneously then yield of tannin is maximum
(Table 5).
3.1. Element detection
Element detection is performed using elemental analyzer
Perkin Elmer Instrument 2400 Series II CHNS/O analyzer.
C = 40.29%, H2 = 3.59%, N2 = 1.70% and other 54.42%.
3.2. Functional group detection
The compound mainly contains phenolic –OH, car-
bonyl, ester, unsaturation groups confirmed by FeCl3 test,
DNP test, Phenolphthalein test and Bayer test. Further it
was confirmed from IR analysis. IR- (200–400 cmÀ1
KBr)
3396 (phenolic –OH: inter molecular H bond) 2935 (aro-
matic C–H str) 1749 (carbonyl@CO), 1608 (aromatic
C@Cstr), 1240 (C–O str).
Table 1
Comparison of chemical properties between bahera and olive oil
Sl. no. Chemical properties Bahera oil Olive oil
1. Free fatty acid 1.71–1.9 0.3–1
2. Saponification value 180 185–196
3. Acetyl value <0.4 Nil
4. Iodine value 80 79–81
5. Palmitic acid 18.2 7.5
6. Stearic acid 8.2 2.5
7. Oleic acid 50.2 75.5
8. Linoleic acid 10.8 6.6
9. Others 6.6 7.6
Table 2
Chemical composition of some oil cake for use as biofertilizer
Sl. no. Oil cake N2 P2O5 K2O
1. Coconut (Cocos nucifera) 6.50 1.30 1.10
2. Rapeseed (Brassica napus) 4.80 2.0 1.30
3. Neem (Azadirachta indica) 5.20 1.10 1.30
4. Mahua (Madhuca indica) 2.50 0.80 1.90
5. Jojoba (Simmondsia chinensis) 5.0 1.70 1.90
6. Bahera (Terminalia belerica Roxb) 8.24 0.19 0.42
Table 3
Biochemical estimation of oil cake
Sl.
no.
Particulars % of nitrogen per 100 g
defatted oil cake
1. Total nitrogen N2 8.24
2. 10% NaCl extractable protein N2 2.853
3. Protein N2 precipitated with TCA 0.118
4. Protein N2 left after pepsin and
trypsin treatment
1.214
5. Digestibility protein 1.634
Table 4
Oil and lard samples used in the Rancimat tests
Flax seed oil Refined olive oil Lard
Acidity (% as oleic acid) 1.0 0.1 0.5
Peroxides values (meqv O2 kgÀ1
) 6.0 3.0 2.1
C12:0 0.1 – 0.1
C14:0 5.66 10.2 1.5
C16:0 0.2 1.1 26.1
C16:1n7 – – 2.1
C16:3n4 – 0.1 –
C17:0 0.1 – 0.3
C17:1 – 2.1 0.2
C18:0 4.1 74.1 15.8
C18:.1 23.4 11.2 41.6
C18:2n6 18.8 0.6 10.7
C18:3n3 45.9 – –
C20:0 0.2 0.3 –
C20:1n11 0.3 0.3 0.2
C22:0 0.2 0.1 –
C24:0 0.2 – –
690 D. Bera et al. / Journal of Food Engineering 81 (2007) 688–692

4. The 1
H NMR of the acetylated product showed two
one-proton singlet signals at 7.81 (8H, s) and 7.70 (2H,
s). These signals suggested the presence of aromatic ring
and it is likely to be galloyl group. The observation of ali-
phatic proton signals at d5.7 (1H, d, J = 2.6 Hz), d5.3 (3H,
m), d4.3 (1H, m) indicates the presence of hexapyranose
structure i.e., D-glucose unit. The presence of signals at
d2.28 (2H, d, J = 4.6 Hz) indicates (CH2–) group present
attached with the hexapyranose ring and d2.15 (9H, s)
d2.15 (9H, s), d2.0 (27H, s), indicates the proton signals
of the acetyl group.
From the functional group detection, IR, NMR analy-
sis of the plant extracted tannin it is confirmed that it
bears –OH group and also it is attached with the aromatic
ring.
The methanolic extract of bahera seed coat extract con-
tains 3.2 mg/ml of gallic acid as indicated by HPLC exper-
iment at 280 nm. The bahera extract which has been
subjected to the test of antioxidation properties of three
oils by rancimate test (Fig. 2). In all samples, gallic acid
was more effective than the bahera extract which may be
due to the bahera extract contains other compounds than
gallic acid which may interfere with the protection prob-
lems. It has been also found that protection factor in the
case of lard was higher than the flaxseed and refined oil.
It may be concluded that protection factor was more effec-
tive on saturated fatty acid esters than unsaturated fatty
acid esters.
4. Conclusion
Considering the high oil content (37%) by weight of
the dry kernel of the seed and the nitrogen value
(8.24%) of oil cake, Bahera fruit appears very promising
for several commercial exploitation and can be consid-
ered as ‘‘Olive of India.” Its by-products can be utilized
for production of biofertilzer, tannin and antioxidant
compounds.
References
Bera, D., Lahiri, D., & Nag, A. (2004). Kinetic studies on bleaching of
edible oil using charred sawdust on a new adsorbent. Journal of Food
Engineering, 65, 33–36.
Chopra, R. N., Mayer, S. L., & Chopra, I. C. (1976). The glossary of
indian medicinal plants (1st ed.). New Delhi: CSIR.
Table 5
Extraction of tannin with changing all the parameters (temperature T, soaking S, soaking and shaking SS and soaking, shaking and temperature SST)
No. of
observations
Extraction of tannin with
soaking time (°C)
Extraction of tannin with
soaking and shaking time
(°C)
Extraction of tannin with
temperature (°C)
Extraction of tannin by changing the three
parameters
Soaking
time (h)
Shaking
time (h)
Amount
of tannin
(g)
Soaking and
shaking time
(h)
Amount
of tannin
(g)
Temperature
(°C)
Amount
of tannin
(g)
Soaking
time (h)
Shaking
time (h)
Temperature
(°C)
Amount
of tannin
(g)
1 1 0 0.30 1 0.35 25 0.3 1 1 25 0.65
2 2 0 0.35 2 0.68 30 0.40 2 2 30 1.10
3 3 0 0.45 3 0.97 35 0.60 3 3 35 1.44
4 4 0 0.55 4 1.16 40 0.68 4 4 40 1.53
5 5 0 0.70 5 1.28 45 0.74 5 5 45 1.62
6 6 0 0.92 6 1.37 50 0.77 6 6 50 1.75
7 7 0 1.09 7 1.40 55 0.81 7 7 55 1.80
8 8 0 1.15 8 1.44 60 0.82 8 8 60 1.85
9 9 0 1.18 9 1.49 65 0.82 9 9 65 1.87
10 10 0 1.19 10 1.56 70 0.82 10 10 70 1.90
11 11 0 1.2 11 1.58 75 0.83 11 11 75 2.00
12 12 0 1.2 12 1.60 80 0.84 12 12 80 2.00
0.00
0.50
1.00
1.50
2.00
2.50
3.00
3.50
ITincreasing
0
10
20
30
40
50
PF
IT increasing (h) 0.2 2.9 2.6
PF 1.1 5.3 41.1
Flaxseed oil (100˚C )
Refined olive oil
(120˚C)
Lard (120˚C)
Fig. 2. Increasing of induction time (rancimat test) and protection factor
in relation to the control by 100 ppm of gallic acid.
PF ¼
ITtest
ITcontrol
IT increasing of 100 ppm ¼ ITtest À ITcontrol:
D. Bera et al. / Journal of Food Engineering 81 (2007) 688–692 691

5. Hagerman & Butler, J. (1980). Extraction and purification of
Sorghum Tannin. Journal of Agricultural and Food Chemistry,
28, 947–952.
Hilditch, T. P. (1979). The chemical constitution of natural fats. New York:
John Willey and sons.
Kar, A., Choudhury, B. K., & Bandyopadhyay, N. G. (2003). Compar-
ative evaluation of hypoglycemic activity of some Indian medicinal
plants in alloxan diabetic rats. Journal of Ethnopharmacology, 84,
105–108.
Nag, A. (2006). Analytical techniques in agriculture, biotechnology and
environmental engineering. New Delhi: Prentice-Hall of India Private
Limited, pp. 58–74.
Nag, A., & De, K. B. (1995). In search of new edible oil. Journal of
Agricultural and Food Chemistry, 43, 902–903.
692 D. Bera et al. / Journal of Food Engineering 81 (2007) 688–692