Food Chemistry 240 (2018) 11131120

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

The use of fuzzy logic to determine the concentration of betel leaf essential MARK
oil and its potency as a juice preservative
Suradeep Basak
Indian Institute of Technology Kharagpur, Agricultural and Food Engineering Department, Kharagpur 721302, West Bengal, India

A R T I C L E I N F O A B S T R A C T

Keywords: The present study was attempted to determine organoleptically acceptable concentration of betel leaf essential
Fuzzy logic oil (BLEO) in raw apple juice using fuzzy logic approach, and to evaluate the ecacy of the acceptable con-
Sensory evaluation centration in the juice under refrigerated storage. The presence of BLEO components in treated juice was con-
Essential oil rmed by FTIR spectroscopy. Based on similarity values, the acceptable concentration in the juice was found to
Apple juice
be 0.19 l/ml of BLEO. Total antioxidant capacity of untreated juice was found to be 16% less than treated juice
Shelf life
at the end of storage. The treated juice exceeded total aerobic plate count of 2 log10 (cfu/ml) on 15th day of
storage. Based on safe limits of microbial load, the shelf life of treated juice was extended by 6 days as compared
to untreated juice under refrigerated storage. BLEO contributes to green consumerism and its application as food
preservative will add value to the product.

1. Introduction
Raw apple juice (RAJ) is unltered, cloudy, high soluble solids, low
acid (pH < 4.5), unpasteurized non-alcoholic beverage that has
shorter shelf life of 23 weeks (Dock, 1999). Despite several foodborne
illness associated with unpasteurized apple juice, it has always been
preferred over thermally pasteurized apple juice by the consumers
(Parish, 1997). Thermal processing is one of the most common prac-
tices, but it has adverse eects on sensory, nutritional and functional
properties of food (Maas & Pagn, 2005; Raso & Barbosa-Cnovas,
2003).
The demand of minimally processed food has escalated with
emergence of green consumerism, which in turn has promoted the use
of naturally occurring antimicrobials (Juneja, Dwivedi, & Yan, 2012).
Essential oils are secondary metabolites of plants which can be ex-
tracted from herbs and spices. Antimicrobial properties of essential oils
can be attributed to the presence of bioactive phenolic compounds
(Tepe, Daferera, Sokmen, Sokmen, & Polissiou, 2005). Basak and Guha
(2015) have identied chavibetol, estragole, -cubebene, chavicol, and
caryophyllene as the major chemical compounds of betel leaf essential
oil (BLEO) of the cultivar Tamluk Mitha. And estragole of the ve BLEO
components has restricted use in the European Union, whereas all ve
chemical compounds of BLEO have GRAS status in US FDA.
Alfonzo et al. (2016) have investigated the biopreservative eect of
lemon essential oil micro-emulsion while improving safety and sensory
attributes of salted sardines. Use of essential oil is also limited due to its
ability to make undesirable alteration in sensory attributes of food
(Hyldgaard, Mygind, & Meyer, 2012). Therefore, detailed sensory eva-
luation of any product treated with essential oil forms the basis of its
marketability. Sensory evaluation is a scientic method to decide ac-
ceptance and rejection of food by the evaluator upon consumption
(Kemp, Hollowood, & Hort, 2009). Fuzzy logic is an important tool
which can draw an important conclusion regarding acceptance, rejec-
tion, ranking, strong and weak attributes of food using vague and im-
precise data in linguistic form lled in by single or multiple experts
(Zimmermann, 1991). Fuzzy sets provide the mathematical method
that represents the fuzziness of human expressions (Lazim & Suriani,
2009) using linguistic variables instead of numerical values
(Zadeh & Kacprzyk, 1999).
Comparison and choice is fundamental to consumers (Imm,
Lee, & Lee, 2011) so is the market risk of any food products. Alongside
organoleptic acceptability, the quality of any food products over sto-
rage period is determined by their physical and chemical character-
istics, microbiological and toxicological safety, and sensory attributes,
packaging and labelling (Molnr, 1995). From industrial point of view,
consumer acceptability of BLEO treated unpasteurized apple juice and
its potency in shelf life extension of the selected food product should be
studied. In view of the above research gap, this study was taken up to
conduct sensory analysis of raw apple juice treated with betel leaf es-
sential oil using fuzzy logic followed by evaluation of antimicrobial
ecacy of organoleptically acceptable concentration of BLEO in the
unpasteurized juice during refrigerated storage at 4 C.

E-mail address: suradeepbasak@gmail.com.

http://dx.doi.org/10.1016/j.foodchem.2017.08.047
Received 25 May 2017; Received in revised form 11 August 2017; Accepted 15 August 2017
Available online 16 August 2017
0308-8146/ 2017 Elsevier Ltd. All rights reserved.
S. Basak
2. Materials and methods
2.1. Essential oil
BLEO was extracted from fresh betel leaves of cv. Tamluk Mitha
(Guha, 2007), and the essential oil based microemulsion (BLEO-ME)
using Tween 20 (SRL Chem., India) as emulsier and water as the
continuous phase was formulated according to Basak and Guha
(2017a).
2.2. Preparation of raw apple juice
Apples (cv. Red delicious) at commercial maturity were purchased
and surface sterilization of the apples was performed according to
Buchanan, Edelson, Miller, and Sapers (1999). Briey, apples were
dipped in 2 l/ml sodium hypochlorite for 1 min, followed by rinsing
them in autoclaved distilled water for another 1 min. Each apple was
cut and made into pulpy juice using mixer grinder followed by ltration
using two layers of muslin cloth to obtain raw apple juice.
Based on the previous study by Basak and Guha (2015), 0.14, 0.19,
0.28, 0.37 and 0.56 l/ml of BLEO in the juice were selected for sensory
evaluation that corresponds to RAJ-S2, RAJ-S3, RAJ-S4, RAJ-S5 and
RAJ-S6, respectively. RAJ-S1 was served as untreated sample
throughout the evaluation process. A panel of 20 participants was se-
lected and trained to perform the sensory evaluation.
2.3. Fuzzy logic analysis
2.3.1. Sensory evaluation
All the selected panellists were trained to familiarize the quality
attributes of raw apple juice, sensory score sheets and method of
scoring the samples. Quality attributes of sensory analysis were colour,
avour, taste and mouthfeel. The fuzzy scale factors were Not sa-
tisfactory, Fair, Medium, Good and Excellent.
According to the explanation provided by Das (2005), ranking of
raw apple juice samples were done using triangular fuzzy membership
distribution function. Sensory scores of the juice samples obtained using
fuzzy scores given by the panellists were converted into triplets and
similarity analysis were performed in order to rank the samples. A
program was coded in Matlab 2015a (The Mathworks; McGarrity,
2008) to perform calculations involved in aforementioned steps. As
shown in Fig. S1 (supplementary materials), the distribution pattern of
5-point sensory scales comprises Not satisfactory/Not at all important,
(0, 0, 25), Fair/Somewhat important, (25, 25, 25), Medium/Im-
portant, (50, 25, 25), Good/Highly important, (75, 25, 25) and
Excellent/Extremely important, (100, 25, 0). The rst number in the
triplets denote coordinate of the abscissa where the value of the
membership function is 1, whereas the second and third number re-
presents distance to left and right of the rst number where the mem-
bership function is zero, respectively (Chakraborty, Das, & Das, 2011).
2.3.2. Triplets for sensory score of samples
Quality attributes of the samples in form of triplets were calculated
using sum of sensory scores, triplets associated with sensory scale and
number of panellists as given below:
Quality attribute of RAJ samples=[{no. of panellist (0 0 25)}+
{no. of panellist (25 25 25)} + {no. of panellist (50 25 25)}+
{no. of panellist (75 25 25)} + {no. of panellist (100 25 0)}]
/(Total no of panellist) (1)
Accordingly, values of triplets for colour, avour, taste and
mouthfeel of all samples were obtained using Eq. (1).
2.3.3. Triplets for panellists preference to importance of quality attributes
Preference of individual panellist to the importance of quality
Food Chemistry 240 (2018) 11131120
attributes of raw apple juice in general was calculated using sum of
sensory scores, triplets associated with sensory scales, and number of
panellists as given below:
Quality attributes of RAJ in general=[{no. of panellist (0 0 25)}
+ {no. of panellist (25 25 25)} + {no. of panellist (50 25 25)}
+ {no. of panellist (75 25 25)} + {no. of panellist (100 25 0)}]
/(Total no of panellist) (2)
Similarly, triplets for all quality attributes, viz. colour (QC), avour
(QF), taste (QT) and mouthfeel (QM) of RAJ in general were calculated
according to Eq. (2).
2.3.4. Triplet for relative weightage of quality attributes
Relative weightage of quality attributes were calculated so as to
determine the triplets for overall sensory scores of the juice using Eq.
(3):
Qsum = sum of first digit of triplets of QC ,QF ,QT and QM (3)
Triplet for relative weightage of colour (QCrel) attribute was cal-
culated using Eq. (4):
QC
QCrel =
Qsum (4)
Accordingly, relative weightage for all four quality attributes of raw
apple juice was calculated.
2.3.5. Triplets for overall sensory scores of the juice samples
The values of triplets for colour, avour, taste and mouthfeel of all
samples are tabulated in Table S1 (supplementary material). Similarly,
the triplets for sensory score of quality attributes of RAJ in general was
calculated based on the preferences of panel members (Table S2, sup-
plementary material). Overall sensory scores for every sample were
calculated according to Eq. (5):
SO1 = (S1 C QCrel) + (S1 F QFrel) + (S1 T QTrel) + (S1 M QMrel )
(5)
where, SO1 is the overall sensory score for RAJ-S1; S1C, S1F, S1T and
S1M represent the triplets corresponding to the colour, avour, taste
and mouthfeel of RAJ-S1; QCrel, QFrel, QTrel and QMrel represents the
triplets corresponding to the relative weightage of quality attributes of
RAJ in general. Multiplication of triplet (a, b, c) with (d, e, f) was
performed using the rule given in Eq. (6):
(a,b,c ) (d,e,f ) = [(a d ),(a e + d b),(a f + d c )] (6)
Using the above triplet multiplication rule, overall sensory score for
all RAJ samples were calculated.
2.3.6. Standard fuzzy scale and ranking of the samples
Standard fuzzy scale following triangular distribution pattern of 6-
point sensory scale is shown in Fig. S2 (supplementary material). The
linguistic expression of the standard fuzzy scale and values of mem-
bership functions for F1F6 are mentioned in Table S3 (supplementary
material).
2.3.7. Values of overall membership function of sensory scores on standard
fuzzy scales
Membership function of overall sensory scores of raw apple juice
samples was calculated on standard fuzzy scale. According to Fig. S3
(supplementary material), when the value of abscissa as a, the value of
membership function become 1, and when (a b) < abscissa < (a
+ c), the value of membership function becomes 0. For the given value
of x on abscissa, value of membership function Bx can be expressed as,
x (ab)
Bx = for (ab) < x < a
b
1114
S. Basak
(a + c )x
Bx = for a < x < (a + c )
c (7)
Bx = 0 for all other values of x
The value of membership function Bx at x = 0, 10, 20, 30, 40, 50,
60, 70, 80, 90 and 100 for every samples were determined.
2.3.8. Similarity values of the juice samples
After obtaining the membership function of each juice samples, si-
milarity values were compared with the corresponding values of
membership function of standard fuzzy scale. This is followed by cal-
culation of similarity values of samples under F1 (Not satisfactory), F2
(Fair), F3 (Satisfactory), F4 (Good), F5 (Very good), and F6 (Excellent)
were made using following equation:
FBT
Sm (F,B) =
maximum of ( F F Tand B BT) (8)
T
where, Sm is the similarity value for the sample; F B is the product of
matrix F with the transpose of matrix B; F FT is the product of matrix
F with the transpose of F; B BT is the product of matrix B with its
transpose. Therefore, for sample 1, Sm (F1, B1), Sm (F2, B1), Sm (F3,
B1), Sm (F4, B1), Sm (F5, B1) and Sm (F6, B1) were calculated using the
rules of matrix multiplication.
2.4. FTIR characterization
Fourier transform infra-red (FTIR) spectroscopic analysis of pure
BLEO, BLEO-ME (50 l/ml), raw apple juice (untreated and treated)
samples were loaded on KBr discs and signals were recorded from
wavenumber 4004000 cm1 at a resolution of 2 cm1 by a Thermo
Nexus-870 (USA).
2.5. Storage study
As suggested by SantAna et al., 2010, the polyethylene terephthalate
(PET) bottles and caps were washed and sanitized in a 0.05% and 0.01%
(v/v) of peracetic acid solution for 30 min, respectively followed by rinsing
with 70% ethanol and dried under laminar air ow chamber. Raw apple
juice was freshly prepared as mentioned earlier in Section 2.2. To every
sterile PET bottle, 100 ml of RAJ treated with organoleptically acceptable
concentration of BLEO was aseptically transferred. Untreated RAJ samples
were used as control sets alongside treated juice samples during entire
storage study under refrigerated (4 C) conditions.
PET bottles (untreated and treated) were randomly picked up for
microbial enumeration, determination of L, a and b values, pH, total
soluble solids (TSS) and total antioxidant capacity (TAOC) at every
3 days interval during storage.
2.5.1. Colour
The colour parameters of the juice was measured using a portable
Chromameter CR-400 (Konica Minolta Sensing, Inc., Osaka, Japan) in
the L (lightness), a (red-green) and b (yellow-blue) colour spaces at
constant lighting conditions after standardizing the instrument with
standard black and white plate. Total colour dierence (E) was cal-
culated to quantify the overall colour dierence of the juice samples
with respect to the reference sample. It was calculated using the fol-
lowing equation (Hirschler, 2012):
E = [(L)2 + (a )2 + (b )2]
where, L = (L1 L0); a = (a1 a0); and b = (b1 b0).
Subscript 0 and 1 depicts the colour value of reference sample (zero
day value of the sample) and the sample being analysed, respectively.
The browning index (BI) represents the purity of brown colour and it
was calculated according to Palou, Lpez-Malo, Barbosa-Cnovas,
Welti-Chanes, and Swanson (1999):
Food Chemistry 240 (2018) 11131120
[100(x 0.31)]
BI =
0.172
where x = (a + 1.75L)/(5.645L + a 3.012b )
Since, initial colour of raw apple juice was not monitored while
extraction, a normalized browning index (BI) was calculated as the
dierence between initial BI and the BI at a given time during storage
for untreated and treated juice samples.
2.5.2. pH and total soluble solids
The pH value of the juice was measured using the pH meter
(Toshcon Industries Pvt. Ltd., Ajmer, India). Hand held refractometer
(Erma, Tokyo, Japan) was used to measure total soluble solids (TSS) of
the juice.
2.5.3. Total antioxidant capacity
Total antioxidant capacity (TAOC) of the juice samples were mea-
sured according to Arabshahi-Delouee and Urooj (2007). Briey, 2.5 g
of RAJ was mixed with methanol in ratio 1:8 (v/v) and incubated for at
least 18 h at 4 C. After incubation, 100 l of sample was mixed with
1 ml of reagent solution (0.6 M sulphuric acid, 28 mM sodium phos-
phate, and 4 mM ammonium molybdate). The tubes were capped and
incubated at 95 C for 90 min. Samples were cooled at room tempera-
ture and absorbance was measured at 695 nm. TAOC of the samples
were expressed as equivalents of -tocopherol (extinction coecient,
= 4 103 M1 cm1).
2.5.4. Microbial counts
Microbial count in raw apple juice were enumerated by ten-fold
dilution of juice samples in phosphate buer (pH 7.4) and plated out in
duplicate on plate count agar (PCA; HiMedia, India) for bacteria and
potato dextrose agar (PDA; HiMedia, India) acidied with 10% tartaric
acid for yeasts and moulds (Mller, Noack, Greiner, Stahl, & Posten,
2014) growth.
3. Results and discussion
3.1. Overall sensory scores of raw apple juice samples
Sensory scores of raw apple juice samples as given by the panellists
are shown in Table S1 (supplementary materials) and sum of panellists
preferences to relative importance of quality attributes of raw apple
juice in general are shown in Table S2 (supplementary materials).
Triplets for overall sensory scores of all individual samples were ob-
tained as given below:
SO1 = (81.22, 50.59, 30.02)
SO2 = (61.33, 42.42, 31.47)
SO3 = (61.88, 42.92, 32.78)
SO4 = (43.58, 33.72, 32.17)
SO5 = (45.29, 33.29, 32.00)
SO6 = (37.67, 28.24, 30.62)
Values of overall membership function of sensory scores on standard
fuzzy scales
Using Eq. (7), the value of Bx at x = 0, 10, 20, 30, 40, 50, 60, 70, 80,
90 and 100 for individual juice samples was obtained as:
B1 = (0, 0, 0, 0.1854, 0.3830, 0.5810, 0.7783, 1.0405, 0.7074,
0.3744)
B2 = (0, 0.0255, 0.2613, 0.4970, 0.7328, 0.9686, 0.7246, 0.4068,
0.0890, 0)
B3 = (0, 0.0242, 0.2572, 0.4902, 0.72315, 0.9561, 0.7523, 0.4472,
0.1421, 0)
B4 = (0.0041, 0.3007, 0.5973, 0.8939, 0.8003, 0.4895, 0.1786, 0,
0, 0)
1115
S. Basak Food Chemistry 240 (2018) 11131120

Table 1 Table 2
Similarity values for raw apple juice under dierent scale factors. Similarity values of individual quality attribute of raw apple juice samples.

Sensory Scale RAJ-S1 RAJ-S2 RAJ-S3 RAJ-S4 RAJ-S5 RAJ-S6 Sensory Scale Colour Flavour Taste Mouthfeel

Not satisfactory, F1 0 0.0051 0.0048 0.0716 0.0565 0.1059 Sm value of the quality attribute of RAJ-S1
Fair, F2 0.0341 0.2141 0.2080 0.5388 0.4807 0.6297 Not satisfactory, F1 0.0128 0 0 0
Satisfactory, F3 0.3160 0.7379 0.7192 0.8951 0.8937 0.8084 Fair, F2 0.1912 0.0095 0.0130 0.0159
Good, F4 0.7502 0.9052 0.9065 0.4273 0.4777 0.2278 Satisfactory, F3 0.5526 0.1608 0.1783 0.1937
Very good, F5 0.8317 0.3433 0.3816 0.0357 0.0455 0 Good, F4 0.7519 0.4818 0.5034 0.5236
Excellent, F6 0.2680 0.0179 0.0281 0 0 0 Very good, F5 0.4406 0.7932 0.7938 0.7949
Excellent, F6 0.0738 0.4532 0.4346 0.4163
Bold values represents the maximum similarity value.
Sm value of the quality attribute of RAJ-S2
Not satisfactory, F1 0.0169 0.0174 0 0
B5 = (0, 0.2403, 0.5407, 0.8412, 0.8527, 0.5402, 0.2277, 0, 0, 0) Fair, F2 0.2237 0.2313 0.1341 0.0773
B6 = (0.0199, 0.3741, 0.7283, 0.9240, 0.5974, 0.2710, 0, 0, 0, 0) Satisfactory, F3 0.6114 0.6302 0.5045 0.3778
Good, F4 0.7532 0.7593 0.7743 0.7264
Very good, F5 0.3924 0.3722 0.5234 0.6477
3.2. Similarity values of the juice samples and their ranking Excellent, F6 0.0483 0.0385 0.1048 0.1723

Sm value of the quality attribute of RAJ-S3

Similarity (Sm) values were calculated using the values of mem- Not satisfactory, F1 0.0141 0.0135 0.0031 0
bership function of standard fuzzy scale and overall membership Fair, F2 0.2044 0.2066 0.1509 0.0771
Satisfactory, F3 0.5776 0.5932 0.5267 0.3664
function values of sensory scores according to Eq. (8). As shown in
Good, F4 0.7433 0.7575 0.7767 0.7033
Table 1, the Sm values of RAJ-S1 under not satisfactory category was 0, Very good, F5 0.4329 0.4138 0.4857 0.6638
while the same sample had Sm value of 0.0341, 0.3160, 0.7502, 0.8317 Excellent, F6 0.0825 0.0605 0.0828 0.1995
and 0.2680 under Fair, Satisfactory, Good, Very good and Excellent Sm value of the quality attribute of RAJ-S4
category, respectively. Hence, overall quality of RAJ-S1 can be con- Not satisfactory, F1 0.0215 0.0848 0.1389 0.0383
sidered as very good because the highest Sm value belongs to this Fair, F2 0.2491 0.4742 0.6902 0.3910
category. Interestingly, RAJ-S3 was found to have slightly higher Sm Satisfactory, F3 0.6405 0.8149 0.8356 0.8325
Good, F4 0.7364 0.5467 0.3243 0.6360
value of 0.9065 under good category as compared to Sm value of
Very good, F5 0.3605 0.1104 0.0215 0.1381
0.9052 for RAJ-S2 under same category. On the other hand, RAJ-S4, Excellent, F6 0.0396 0 0 0
RAJ-S5 and RAJ-S6 were placed under satisfactory category with Sm
Sm value of the quality attribute of RAJ-S5
value of 0.8951, 0.8937 and 0.8084, respectively. Based on Sm values, Not satisfactory, F1 0.0096 0.0520 0.0593 0.0480
the order of ranking of raw apple juice samples was found to be: Fair, F2 0.1761 0.4636 0.5298 0.4455
Satisfactory, F3 0.5361 0.8432 0.8787 0.8574
RAJS1 > RAJS3 > RAJS2 > RAJS4 > RAJS5 > RAJS6 Good, F4 0.7416 0.5272 0.4606 0.5641
Very good, F5 0.4647 0.0911 0.0483 0.0912
As far as general perception of raw apple juice is concerned it was
Excellent, F6 0.1006 0 0 0
found that Sm values of avour, mouthfeel and taste are highest under
highly important category, whereas colour has highest Sm value under Sm value of the quality attribute of RAJ-S6
Not satisfactory, F1 0.0242 0.1322 0.1802 0.0552
important category. The order of ranking of quality attributes of raw Fair, F2 0.2661 0.7309 0.8001 0.5071
apple juice in general as mentioned in Table S4 (supplementary mate- Satisfactory, F3 0.6652 0.8124 0.7923 0.8829
rials) was: Good, F4 0.7325 0.2370 0.2063 0.4631
Very good, F5 0.3293 0.0015 0 0.0452
Flavour> Mouthfeel > Taste>Colour Excellent, F6 0.0316 0 0 0

A comparable Sm values for avour, taste and mouthfeel of RAJ-S1
under very good category was observed, whereas colour had highest
Sm value under good category. This satised the ranking of quality
attributes of raw apple juice in general. Now, for BLEO treated raw
apple juice, RAJ-S2 and RAJ-S3 was found to have all four quality at-
tributes under good category, whereas quality attributes of remaining
samples were ranked accordingly to Sm values as given in Table 2.
RAJ-S3 was found to have a minor increment in Sm value (0.7767)
of taste as compared to RAJ-S2 (Sm value = 0.7743). On the contrary,
Sm values of remaining quality attributes viz. avour colour and
mouthfeel were found to be comparatively higher in RAJ-S2. It can also
be observed that Sm value of mouthfeel, avour and colour for RAJ-S3
under very good and excellent category are on the higher side as
compared to RAJ-S2 for the same. It can be suggested that increase in
concentration of BLEO by 0.05 l/ml increased the acceptability due to
tinge of BLEO essence in RAJ-S3 over RAJ-S2. It is evident that highest
concentration of BLEO has imparted a fair taste in RAJ-S6 as opposed
to satisfactory taste for RAJ-S4 and RAJ-S5. The bitter and pungent
taste at concentration greater than or equal to 0.28 l/ml of BLEO has
resulted in unacceptability of the juice. Therefore, raw apple juice
treated with any concentration less than or equal to 0.19 l/ml of BLEO
will be organoleptically acceptable.
Although overall quality of RAJ-S1 (untreated) sample was unan-
imously preferred by the panellists, however RAJ-S2 and RAJ-S3 fared
well among treated juice and was placed under good category. RAJ-S3
Bold values represents the maximum similarity value.
has additional benets of being green and clean food product despite
lower organoleptic acceptability due to presence of bioactive chemical
components in form of BLEO.
3.3. FTIR characterization
Fig. 1 depicts FTIR spectra of BLEO, BLEO based microemulsion,
treated and untreated raw apple juice. In general, BLEO exhibit char-
acteristic peaks at 3419 (hydroxyl group, H bonded eOH stretch), 3076
(vinylidene eCH stretch), 2954 (methyl eCH asymmetric stretch),
2929 and 2864 (methylene eCH asymmetric and symmetric stretch,
respectively), 1639 cm1 (alkenyl C]C stretch or conjugated ketone
group), 1614, 1595 and 1510 cm1 represents C]CeC aromatic ring
stretch, 1365 (phenol or tertiary alcohol eOH bend), 1267 and
1240 cm1 (aromatic ethers, arylOstretch), 1174, 1149, 1128, 1032,
993, 966 cm1 represents aromatic CeH in plane bend and 823, 762,
711 cm1 represents aromatic CeH out of plane bend, and nally 637
and 591 cm1 could be OeH out of plane bend (Fig. 1(a)). The ob-
tained functional groups and bond vibration are in agreement with
functional groups of chemical compounds of BLEO reported earlier by
Basak and Guha (2015). For BLEO based microemulsion (Fig. 1(b)),

1116
S. Basak
peak of H bonded eOH stretch shifted to 34193417 cm1 and peak at
3076 and 2954 cm1 of vinylidene eCH stretch and methyl eCH
asymmetric stretch was found to be missing in BLEO-ME. Also, peak at
2929 and 2864 for methylene eCH asymmetric and symmetric stretch
were shifted to 2925 and 2880 cm1, respectively. An additional peak
at 2104 cm1 was observed in BLEO-ME that could possibly represent
terminal alkyne groups present. FTIR spectra of treated and untreated
raw apple juice (Fig. 1(c) and (d), respectively) shared every char-
acteristic peaks except an additional peak at 1344 cm1 that suggested
the presence of primary or secondary or phenol or tertiary alcohol
(OeH in-plane bend or eOH bend) in juice treated with BLEO-ME.
Presence of the functional group in treated raw apple juice could be due
to reduced ingredient interaction of apple juice and essential oil com-
ponents of BLEO-ME that probably helped it retain its bioactivity
mostly unaltered. Previous report by Basak and Guha (2017b) sug-
gested no signicant alteration in antifungal potential of micro-
emulsied BLEO on growth of P. expansum in raw apple juice with re-
spect to in vitro growth medium. Ma, Davidson, and Zhong (2016) have
also reported to have similar antifungal potential of free and micro-
emulsied cinnamon bark essential oil against Salmonella enterica and
Escherichia coli O157:H7.
1117
Food Chemistry 240 (2018) 11131120
Fig. 1. FTIR spectra of (a) BLEO, (b) microemulsied BLEO (50 l/
ml), (c) BLEO treated raw apple juice (0.19 l/ml), and (d) untreated
raw apple juice.
3.4. Storage study
3.4.1. Colour
Total colour dierence (E) of untreated as well as treated juice
samples signicantly (P < 0.0001) increased from 0.42 0.06 and
0.28 0.02 to 1.86 0.09 and 1.35 0.08 during 3 and 15 days
after storage, respectively. Similarly, signicant (P < 0.0001) increase
in BI values from 0.67 0.02 and 0.32 0.06 to 3.2 0.11 and
2.1 0.11 was observed during 3 and 15 days after storage of un-
treated and treated juice samples, respectively.
The increase in browning can be due to oxidation of naturally occurring
phenolic components in apple by polyphenol oxidase that remained active
because the juice was not exposed to any thermal treatment prior to storage
(Ndiaye, Xu, & Wang, 2009). Perhaps, presence of polyphenols in BLEO
might have provided an added benet to the treated juice to have lower E
and BI values as compared to untreated juice, which has only indigenous
phenolic compounds of apple (Toms-Barbern & Espn, 2001). Other
reason for increase in BI value can be direct photo-oxidation of phenolic
compounds present in both untreated and treated juice samples (Manzocco,
Quarta, & Dri, 2009). Variation in E and BI values as given in Table 3,
suggested that concentration of BLEO selected in the present study does not
have much impact on prevention of browning, but it can slow down the
reaction in the treated juice as compared to untreated juice during storage.
S. Basak Food Chemistry 240 (2018) 11131120

Table 3 (500 M) free radicals was found to be 11.29 g/ml (unpublished data).
Total colour dierence (E*) and normalized browning index (BI) of raw apple juice The value of TAOC suggested high antioxidant potency of BLEO, which
during storage at 4 C.
was further elaborated by lower IC50 value of the essential oil. It is
DAS+ (E*)# (BI) evident that terpene, terpenoid and phenolic compounds of essential oil
of betel leaf alongside natural antioxidants in the juice might have
Untreated Treated Untreated Treated provided an enhanced electron donating capability to convert reactive
fg g gh free radicals into stable non-reactive products in treated raw apple juice
3 0.42 0.06 0.28 0.02 0.67 0.02 0.32h 0.06
6 0.86de 0.07 0.51eg 0.11 1.25ef 0.03 0.64gh 0.11 during storage.
9 1.33bc 0.03 0.70def 0.03 1.87cd 0.01 1.04fg 0.01
12 1.52ab 0.08 1.02cd 0.06 2.39b 0.10 1.55de 0.07 3.4.4. Microbial counts
15 1.86a 0.09 1.35bc 0.08 3.19a 0.11 2.10bc 0.11
The growth of microbial population in untreated and treated raw
apple juice during storage at 4 C is represented in Fig. 2. The initial
Values are mean S.E.
Means represented by the same letter in untreated and treated column of same parameter total aerobic plate count (TAPC) and yeast and mould (YM) count of
are not signicantly dierent according to ANOVA and Tukeys multiple comparison raw apple juice was enumerated to be 1.77 0.05 log10 (cfu/ml) and
tests. 2.23 0.04 log10 (cfu/ml), respectively. Initial microbial load in
+
DAS: Days after storage. freshly extracted apple juice can be supported by a few previous stu-
#
Values are compared with reference sample (zero day value of untreated sample).
dies. Among those studies, Beech (1958) has reported initial yeast po-
Values are based on dierence between initial browning index and browning index
on the observation. pulations of raw apple juice produced in a clean pilot plant and com-
mercial plant varied from 5.485.79 log10 (cfu/ml) and 5.836.11 log10
3.4.2. pH and total soluble solids (cfu/ml), respectively. Vasavada and Heperkan (2002) have also cited
During refrigerated storage of 15 days, no signicant variation was initial yeast levels of 3 log10 (cfu/ml) to 5 log10 (cfu/ml) for juices
observed in the pH of untreated (P = 0.338) as well as treated produced from healthy apples. Similarly, Miller and Keller (2005) have
(P = 0.614) raw apple juice samples. TSS of untreated juice decreased also reported TAPC and YM population of unsorted tree harvested fruit
signicantly (P = 0.002). However, the change in TSS of treated juice apple cider or juice ranged from 1.903.40 log10 (cfu/ml) and
was not signicant (P = 0.012) during storage (Table 4). The reduction 1.993.32 log10 (cfu/ml), respectively. They also reported that cider or
in TSS of untreated raw apple juice during refrigerated storage can be juice extracted from poorer quality, ground harvested apples have
correlated to the breakdown of polysaccharides into monosaccharide in substantially higher TAPC and YM count ranging from 4.195.43 log10
presence of spoilage microorganisms and utilization of the available (cfu/ml) and 3.845.23 log10 (cfu/ml), respectively. In the present
monosaccharide by the same for further growth and proliferation in the study, low initial bacterial load of the juice could be due to pre-treat-
untreated juice (Bhardwaj & Pandey, 2011). ment of the apples with sodium hypochlorite as suggested by Buchanan
et al. (1999). TAPC of untreated juice constantly increased during
storage and exceeded the permissible limit of 2 log10 (cfu/ml) count
3.4.3. Total antioxidant capacity (EFSA, 2005) on day 9, followed by TAPC of treated juice on day 15. On
As mentioned in Table 4, TAOC was found to have decreased from the contrary, initial microbial load of unpasteurized apple juice on PDA
10.12 0.06 to 2.46 0.15 mM -tocopherol/g of untreated juice plate was dominated by yeast population. However, lowest YM count of
sample, whereas TAOC in treated raw apple juice was reduced from untreated and treated juice was observed at day 6 with 1.98 log10 (cfu/
10.92 0.05 to 4.39 0.03 mM -tocopherol/g treated juice ml) and 1.63 log10 (cfu/ml), respectively. The dierence in YM count of
sample at 0 and 15 days after storage, respectively. As suggested by Van untreated and treated can be attributed to the fact that rate of inhibition
der Sluis, Dekker, Skrede, and Jongen (2002), the antioxidant activity using sub-lethal concentration ( MIC = 0.19 l/ml) of BLEO was slow
imparted by fresh raw apple juice is due to naturally occurring poly- due to heavy initial load of inherent microbial population in freshly
phenols, which provides only 10% of antioxidant activity of the fresh extracted unpasteurized apple juice. Therefore, low concentration of
apple, if the juice is obtained by pulping and pressing. No signicant the essential oil imparted lower toxicity on the inherent microorgan-
change in Brix, pH and concentrations of polyphenols in pasteurized isms in the juice samples that eventually led to constant increase in
apple juice during storage at 4 C was also observed by Van Der Sluis, TAPC and YM count at par with untreated juice.
Dekker, and Van Boekel (2005). On the contrary, present study deals According to the basic aspects of hurdle technology (Leistner,
with unpasteurized and unltered apple juice that had active oxidizing 2000), the sub-lethal concentration of BLEO in the present study can be
enzymes, which were perhaps responsible for degradation of TAOC of opted for eective preservation of the juice in combination with other
the juice during storage even at refrigerated condition. TAOC of pure preservation factors viz. mild temperature, pH, and water activity.
BLEO was found to be 362.9 1.24 mM -tocopherol/g of pure BLEO Previously, Tyagi, Gottardi, Malik, and Guerzoni (2013) have suggested
and the concentration of BLEO required to scavenge 50% of DPPH the preservation of a mixed fruit juice using combination of Mentha

Table 4
Comparison of pH, total soluble solids (TSS) and total antioxidant capacity (TAOC) of untreated and treated raw apple juice during refrigerated storage at 4 C.

DAS+ pH TSS (Brix) TAOC

(mM -tocopherol/g sample)

Untreated Treated Untreated Treated Untreated Treated

a a b
0 4.36 0.01 4.35 0.02 15.57 0.03 15.57 0.03 10.12 0.06 10.92a 0.05
3 4.35 0.02 4.35 0.02 15.50a 0.06 15.57a 0.03 6.96d 0.13 7.74c 0.08
6 4.33 0.02 4.34 0.02 15.47a 0.03 15.57a 0.03 5.74e 0.02 6.47d 0.23
9 4.33 0.02 4.33 0.01 15.47a 0.03 15.50a 0.06 4.86fg 0.03 5.26ef 0.17
12 4.32 0.02 4.33 0.01 14.37b 0.03 15.43a 0.03 3.14h 0.14 4.53g 0.04
15 4.31 0.02 4.32 0.01 14.27b 0.03 15.37a 0.03 2.46i 0.15 4.39g 0.03

Values are mean S.E.

Means represented by the same letter in untreated and treated column of same parameter are not signicantly dierent according to ANOVA and Tukeys multiple comparison tests.
+
DAS: Days after storage.

1118
S. Basak
piperita essential oil and mild thermal treatment had minimum altera-
tion in organoleptic properties viz. colour and avour. On the other
hand, present study showed that colour and avour are in same cate-
gory for all organoleptically rejected samples (RAJ-S4, RAJ-S5 and RAJ-
S6), but the taste of RAJ-S6 changed from remaining two rejected
samples. This study justied the need of detailed sensory analysis of
essential oil treated food products followed by its potential as food
preservative during storage along with proper threshold for organo-
leptically acceptable concentration.
4. Conclusion
The present study investigated the impact of BLEO on sensory at-
tributes of raw apple juice, and also its ecacy on the storability of the
juice. Fuzzy logic approach using similarity analysis provided an insight
into variation in consumer acceptability with respect to colour, avour,
taste and mouthfeel of raw apple juice under the inuence of essential
oil of betel leaf. Based on similarity values, the juice treated with
0.19 l/ml of BLEO was found to have highest organoleptic accept-
ability among all other BLEO treatments. The organoleptically accep-
table sub-lethal concentration of BLEO was eective to extend the shelf
life of raw apple juice by 6 days as compared to the untreated raw apple
juice samples under refrigerated conditions. Such extension of shelf life
was determined on the basis of safe limit of microbial load of the juice.
This study suggested the potency of BLEO as raw apple juice pre-
servative. As the sensory barrier restricted luxurious application of
BLEO in the juice, the future study on use of BLEO at sub-lethal con-
centration along with other hurdle techniques is necessary to achieve
longer shelf life so as to have economic feasibility at industrial scale.
Acknowledgement
The author is grateful to Indian Institute of Technology Kharagpur
for nancial assistance, infrastructure and facilities to conduct the re-
search. The author is also thankful to Prof. P. Guha, Agricultural and
Food Engineering Department, IIT Kharagpur for his supervision and
support.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.foodchem.2017.08.047.
References
Alfonzo, A., Martorana, A., Guarrasi, V., Barbera, M., Gaglio, R., Santulli, A., et al. (2016).
Eect of the lemon essential oils on the safety and sensory quality of salted sardines
(Sardina pilchardus Walbaum 1792). Food Control, 73, 12651274.
Arabshahi-Delouee, S., & Urooj, A. (2007). Antioxidant properties of various solvent
extracts of mulberry (Morus indica L.) leaves. Food Chemistry, 102(4), 12331240.
Basak, S., & Guha, P. (2015). Modelling the eect of essential oil of betel leaf (Piper betle
L.) on germination, growth, and apparent lag time of Penicillium expansum on semi-
synthetic media. International Journal of Food Microbiology, 215, 171178.
1119
Food Chemistry 240 (2018) 11131120
Fig. 2. Microbial count (a) TAPC and (b) YM count of un-
treated and BLEO treated (0.19 l/ml) unpasteurized raw
apple juice during storage at 4 C.
Basak, S., & Guha, P. (2017b). Use of predictive model to describe sporicidal and cell
viability ecacy of betel leaf (Piper betle L.) essential oil on Aspergillus avus and
Penicillium expansum and its antifungal activity in raw apple juice. LWT - Food Science
and Technology, 80, 510516.
Basak, S., & Guha, P. (2017a). Betel leaf (Piper betle L.) essential oil microemulsion:
Characterization and antifungal activity on growth, and apparent lag time of
Aspergillus avus in tomato paste. LWT - Food Science and Technology, 75, 616623.
Beech, F. W. (1958). The yeast ora of apple juices and ciders. Journal of Applied
Bacteriology, 21(2), 257266.
Bhardwaj, R. L., & Pandey, S. (2011). Juice blendsA way of utilization of under-utilized
fruits, vegetables, and spices: A review. Critical Reviews in Food Science and Nutrition,
51(6), 563570.
Buchanan, R. L., Edelson, S. G., Miller, R. L., & Sapers, G. M. (1999). Contamination of
intact apples after immersion in an aqueous environment containing Escherichia coli
O157:H7. Journal of Food Protection, 62(5), 444450.
Chakraborty, D., Das, S., & Das, H. (2011). Aggregation of sensory data using fuzzy logic
for sensory quality evaluation of food. Journal of Food Science and Technology, 50, 19.
Das, H. (2005). Sensory evaluation using fuzzy logic. Food processing operations analysis
(pp. 384402). New Delhi: Asian Books Private Limited.
Dock, L. L. (1999). Development of thermal and non-thermal preservation methods for pro-
ducing microbilogically safe apple cider. West Lafayette IN, USA: Purdue University.
EFSA (2005). Commission Regulation (EC) No 2073/2005 of 15th November 2005 on
microbiological criteria for foodstus. Ocial Journal of the European Union, L338,
126.
Guha, P. (2007). Extraction of essential oil: An appropriate rural technology for mini-
mizing wastage of surplus betel leaves. Agricultural Mechanization in Asia, Africa and
Latin America, 38(4), 4750.
Hirschler, R. (2012). Whiteness, yellowness and browning in food colorimetry. In J. L.
Caivano, & M. del P. Buera (Eds.), Color in food: Technological and psychophysical
aspects (pp. 93104). Boca Raton, USA: CRC Press.
Hyldgaard, M., Mygind, T., & Meyer, R. L. (2012). Essential oils in food preservation:
Mode of action, synergies, and interactions with food matrix components. Frontiers in
Microbiology, 3, 124.
Imm, B. Y., Lee, J. H., & Lee, S. H. (2011). Sensory quality index (SQI) for commercial
food products. Food Quality and Preference, 22(8), 748752.
Juneja, V. K., Dwivedi, H. P., & Yan, X. (2012). Novel natural food antimicrobials. Annual
Reviews in Food Science and Technology, 3, 381403.
Kemp, S., Hollowood, T., & Hort, J. (2009). Sensory evaluation: A practical handbook (1st
ed.). Chichester, United Kingdom: Wiley-Blackwell.
Lazim, M. A., & Suriani, M. (2009). Sensory evaluation of the selected coee products
using fuzzy approach. World Academy of Science, Engineering and Technology, 3(2),
717720.
Leistner, L. (2000). Basic aspects of food preservation by hurdle technology. International
Journal of Food Microbiology, 55, 181186.
Ma, Q., Davidson, P. M., & Zhong, Q. (2016). Antimicrobial properties of microemulsions
formulated with essential oils, soybean oil, and Tween 80. International Journal of
Food Microbiology, 226, 2025.
Maas, P., & Pagn, R. (2005). Microbial inactivation by new technologies of food pre-
servation. Journal of Applied Microbiology, 98(6), 13871399.
Manzocco, L., Quarta, B., & Dri, A. (2009). Polyphenoloxidase inactivation by light ex-
posure in model systems and apple derivatives. Innovative Food Science and Emerging
Technologies, 10(4), 506511.
McGarrity, S. (2008). Introduction to object-oriented programming in MATLAB.
Retrieved October 27, 2016, from < www.mathworks.com > .
Miller, A. J., & Keller, S. E. (2005). Microbiological safety of fresh citrus and apple juices.
In J. R. Gorny, A. E. Yousef, & G. M. Sapers (Eds.), Microbiology of fruits and vegetables
(pp. 211230). Boca Raton, Florida: CRC Press.
Molnr, P. J. (1995). A model for overall description of food quality. Food Quality and
Preference, 6(3), 185190.
Mller, A., Noack, L., Greiner, R., Stahl, M. R., & Posten, C. (2014). Eect of UV-C and
UV-B treatment on polyphenol oxidase activity and shelf life of apple and grape
juices. Innovative Food Science & Emerging Technologies, 26, 498504.
Ndiaye, C., Xu, S. Y., & Wang, Z. (2009). Steam blanching eect on polyphenoloxidase,
peroxidase and colour of mango (Mangifera indica L.) slices. Food Chemistry, 113(1),
9295.
Palou, E., Lpez-Malo, A., Barbosa-Cnovas, G. V., Welti-Chanes, J., & Swanson, B. G.
(1999). Polyphenoloxidase Activity and Color of Blanched and High Hydrostatic
S. Basak
Pressure Treated Banana Puree. Journal of Food Science, 64(1), 4245.
Parish, M. E. (1997). Public health and nonpasteurized fruit juices. Critical Reviews in
Microbiology, 23(2), 109119.
Raso, J., & Barbosa-Cnovas, G. V. (2003). Nonthermal preservation of foods using
combined processing techniques. Critical Reviews in Food Science and Nutrition, 43(3),
265285.
SantAna, A. S., Simas, R. C., Almeida, C. A. A., Cabral, E. C., Rauber, R. H., Mallmann, C.
A., et al. (2010). Inuence of package, type of apple juice and temperature on the
production of patulin by Byssochlamys nivea and Byssochlamys fulva. International
Journal of Food Microbiology, 142(12), 156163.
Tepe, B., Daferera, D., Sokmen, A., Sokmen, M., & Polissiou, M. (2005). Antimicrobial and
antioxidant activities of the essential oil and various extracts of Salvia tomentosa
Miller (Lamiaceae). Food Chemistry, 90(3), 333340.
Toms-Barbern, F. A., & Espn, J. C. (2001). Phenolic compounds and related enzymes as
determinants of quality in fruits and vegetables. Journal of the Science of Food and
Agriculture, 81(9), 853876.
1120
Food Chemistry 240 (2018) 11131120
Tyagi, A. K., Gottardi, D., Malik, A., & Guerzoni, M. E. (2013). Anti-yeast activity of
mentha oil and vapours through in vitro and in vivo (real fruit juices) assays. Food
Chemistry, 137(14), 108114.
Van der Sluis, A. A., Dekker, M., Skrede, G., & Jongen, W. M. F. (2002). Activity and
concentration of polyphenolic antioxidants in apple juice. 1. Eect of existing pro-
duction methods. Journal of Agricultural and Food Chemistry, 50(25), 72117219.
Van Der Sluis, A. A., Dekker, M., & Van Boekel, M. A. J. S. (2005). Activity and con-
centration of polyphenolic antioxidants in apple juice. 3. Stability during storage.
Journal of Agricultural and Food Chemistry, 53(4), 10731080.
Vasavada, P. C., & Heperkan, D. (2002). Non-thermal alternative processing technologies
for the control of spoilage bacteria in fruit juices and fruit-based drinks. Food Safety
Magazine, 10-13, 4647.
Zadeh, L. A., & Kacprzyk, J. (1999). Computing with words in information intelligent - part 1:
Foundations, part 2: Applications. Heidelberg and New York: Springer-Verlag.
Zimmermann, H. J. (1991). Fuzzy set theory and its applications (2nd ed.). Boston, MA:
Kluwer Academic Publishers.