Content uploaded by Göktürk Öztürk
Author content
All content in this area was uploaded by Göktürk Öztürk on Dec 09, 2018
Content may be subject to copyright.
CHARACTERIZATION OF GRAPE MOLASSES/SESAME PASTE/
HONEY BLENDS: MULTIPLE RESPONSE OPTIMIZATION
OF SOME PHYSICOCHEMICAL, BIOACTIVE, VISCOELASTIC
AND SENSORY PROPERTIES
SAFA KARAMAN
1,5
, MUSTAFA TAHSIN YILMAZ
2
, GOKTURK OZTURK
3
, FERHAT YUKSEL
1,4
,
€
OMER SAID TOKER
2
and MAHMUT DOGAN
1
1
Engineering Faculty, Food Engineering Department, Erciyes University, Kayseri 38039, Turkey
2
Chemical and Metallurgical Engineering Faculty, Food Engineering Department, Yıldız Technical University, Istanbul 34210, Turkey
3
Food Technology Department, Kaman Vocational High School, Ahi Evran University, Kırs¸ehir 40300, Turkey
4
Engineering and Natural Sciences Faculty, Food Engineering Department, Gumushane University, 29100 Gumushane Turkey
5
Corresponding author.
TEL: 190 352 207 6666-32754;
FAX: 190 352 4375784;
EMAIL: skaraman@erciyes.edu.tr
Received for Publication January 28, 2016
Accepted for Publication April 12, 2016
doi:10.1111/jfpe.12406
ABSTRACT
In this study, blends of grape molasses, sesame paste and honey were prepared at
different ratios according to the mixture design and some physicochemical,
rheological, bioactive and sensory properties of final blends were studied to create
the optimum formulation for the final product. The results showed that the
mixture components had significant change on the studied parameters (P<0.05).
Sugar composition of samples changed significantly by the addition of grape
molasses and honey which resulted an increase in fructose and glucose. Grape
molasses increase in the blend provided a significant increment in total phenolic
content and antiradical activity. The blends had viscoelastic character and the
mixtures containing higher honey showed more viscous behavior. The highest
complex viscosity values were recorded for the blends of sesame paste and honey.
Regression models having quite high determination of coefficients (R
2
>0.71)
were constructed for the studied parameters. Multiple response optimization
results showed that the most preferred blend should contain 34.66% of grape
molasses, 34.11% of sesame paste and 31.23% of honey.
PRACTICAL APPLICATIONS
Grape molasses, honey and sesame paste are commonly consumed food products
which are rich in energy and many functional components. The people generally
consume them in their daily diets especially in their breakfasts. To combine their
functionality and sensory properties, an optimization procedure was applied and
the mostly desired blend was determined. Blending affected the rheological,
textural and bioactive properties of the sample significantly. The results of the
study are very important for food industry.
INTRODUCTION
Grape molasses, sesame paste and honey are the foods con-
sumed in daily diet especially in the breakfasts by many peo-
ple in the world because of their special properties. Grape
molasses is a concentrated grape juice obtained by boiling of
crushed grapes up to 70–80% soluble dry matter
1-3
.Grape
molasses is a good energy source because of its quite high
sugar level and it is also a functional food due to its compo-
sitions like mineral, phenolics, organic acids, etc. (Seng€
ul
et al. 2005; Akbulut et al. 2008; Karaman and Kayacier
2011). Sesame paste is pressed and pasted sesame seeds
which are ground, dehulled and dry roasted. It is a popular
food product in Turkey and other Eastern countries and this
Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc. 1
Journal of Food Process Engineering ISSN 1745–4530
product is rich in proteins (17–27%), carbohydrates
(6.4–21%), lipids (54–65%) and dietary fiber (9.3%) in
addition to many functional components. It was reported
that it had high antioxidant activity because of its high poly-
unsaturated fatty acid and it showed ability to reduce the
cholesterol (Abu-Jdayil et al. 2002; Abu-Jdayil 2004; Ghare-
hyakheh et al. 2014). Honey is a natural sweet product pro-
duced by bees from the nectar of plants and it is very
important energy source because it is a concentrated sugar
solution containing organic acids and some amino acids, as
well as certain macro and microelements, and many biologi-
cally active compounds (Juszczak and Fortuna 2006; Ahmed
et al. 2007; Arslan et al.2008;Karamanet al. 2011). In gen-
eral, people prefer to consume these types of products as
blended and molasses/sesame paste blends are very popular
food mixture and it is started to produce industrially.
Because of the increase in demand for the consumption of
blends of these food products, many studies have been con-
ducted by the researchers to characterize the many different
properties of the blended samples. Alpaslan and Hayta
(2002) investigated the rheological and sensory properties of
molasses/sesame paste (tahin) blends containing different
levels of molasses and reported that the addition of molasses
to the tahin changed the physicochemical, rheological and
sensory properties of final product significantly. Similarly,
Arslan et al. (2008) investigated the rheological properties of
tahin/molasses blends at different ratios and concluded that
the blends showed pseudoplastic behavior. Akbulut et al.
(2012) performed a study investigating the rheological, some
physicochemical and sensory properties of sesame paste and
honey blends and reported that the blends of sesame paste
and honey showed non-Newtonian shear thinning behavior
and affected the many parameters by the blending. As a
result, several studies have carried out to see the effect of
blending of molasses/sesame paste and sesame paste/honey,
but there has been no report on bioactive, viscoelastic and
sensory properties of grape molasses/sesame paste/honey
blends (GSHB). The aim of the current study was to investi-
gate the effect of GSHB on some physicochemical, rheologi-
cal, bioactive and sensory properties of the final blended
product by using mixture design modeling approach and
optimize the most preferred formulation which could be
produced using grape molasses, sesame paste and honey to
combine their desired sensory and functional characteristics.
MATERIALS AND METHODS
Materials
Sesame paste (Koska Food Co, Bursa, Turkey), flower honey
(Balparmak Food Co., _
Istanbul Turkey) and grape molasses
(Koska Food Co, Bursa, Turkey) were commercially pur-
chased from a local supermarket in Kayseri, Turkey.
Preparation of Blends
To prepare the blends, the samples were mixed according to
the design given in Table 1. For this purpose, the samples
were weighed in a same beaker and mixed for 5 min using a
stirrer at 600 rpm (IKA, RW 20DZM, IKA-works Inc., NC)
at room temperature. After the preparation of the blends,
the analyses were performed immediately and accordingly.
Physicochemical Analysis
An automatic colorimeter (Konica Minolta, model CM-5,
Mississauga, ON, Canada) was used for the measurement of
color parameters recorded as L,aand b. Water activity values
were measured at 258C using an Aqualab water activity (a
w
)
meter (Decagon, Pullman, WA).
The pH values of each blend solutions 10% (w/v) in dis-
tilled water at 258C were measured with a pH meter (WTW-
Inolab, Weilheim, Germany). An automatic refractometer
(Reichert AR 700) was used for the determination of brix
values at 208C. Dry matter contents were determined by con-
ventional drying method according to the described method
in AOAC (2000). For the ash content, the samples were
incinerated at 6258C in a muffle oven (Protherm, Ankara,
Turkey). All analyses were performed as triplicate.
Determination of Sugar Composition
Major sugar amounts (fructose, glucose and saccharose) of
the blends were determined using High Pressure Liquid
Chromatography (HPLC) according to the procedure
described by Jahanbin et al. (2012). One gram of sample was
dissolved with 9 mL of distilled water and the mixture was
TABLE 1. COMPONENTS OF GRAPE MOLASSES/SESAME PASTE/
HONEY BLENDS ACCORDING TO SIMPLEX LATTICE MIXTURE DESIGN
Blends
Coded values
Uncoded values
(Ingredient proportions)
X
1
X
2
X
3
Grape
molasses
(%)
Sesame
paste (%)
Honey
(%)
1 1.00 0.00 0.00 100 0 0
2 0.00 1.00 0.00 0 100 0
3 0.00 0.00 1.00 0 0 100
4 0.50 0.50 0.00 50 50 0
5 0.00 0.50 0.50 0 50 50
6 0.50 0.00 0.50 50 0 50
7 0.25 0.75 0.00 25 75 0
8 0.00 0.25 0.75 0 25 75
9 0.75 0.00 0.25 75 0 25
10 0.75 0.25 0.00 75 25 0
11 0.00 0.75 0.25 0 75 25
12 0.25 0.00 0.75 25 0 75
13 0.50 0.25 0.25 50 25 25
14 0.25 0.50 0.25 25 50 25
15 0.25 0.25 0.50 25 25 50
GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS S. KARAMAN ET AL.
2Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc.
shaked to be sure the all sugars dissolved effectively. Then,
the samples were mixed with 1 mL of Carrez I and 1 mL of
Carrez II and the samples were centrifuged at 5,500 3gfor
5 min. Thereafter, the supernatant was filtered using a 0.45
mm syringe filters. The filtrate was injected to the HPLC
(Agilent 1100) system equipped with a refractive index
detector. An Agilent Zorbax carbohydrate analysis column
(5 lmand4.6mm3150 mm) was used and the analysis
conditions were set as following: mobile phase, 80% acetoni-
trile and 20% water; flow rate, 1.4 mL/min; injection vol-
ume, 20 lL and the column temperature was set to be 258C.
Sugars were identified according to their retention times by
comparing with sugar standards. The sugar concentrations
of the samples were calculated using the prepared calibration
curve of the each sugar.
Determination of Bioactive Properties
Total Phenolic Content. A modified method described
by Karaman et al. (2014) was used for the determination of
total phenolic content (TPC). For this purpose, roughly one
g of sample was weighed and 9 mL of distilled water was
added to obtain the extract. One hour shaking was per-
formed for the extraction and finally, the samples were cen-
trifuged a 5,500 3g for 5 min. The supernatant was filtered
using 0.45 lm syringe filter. Then, 0.2 mL of the extract was
mixed with 1.8 mL of distilled water in a tube and 1 mL of
diluted Folin–Ciocelteau’s phenol reagent (1:10 with distilled
water) was added into a tube followed by mixing with a vor-
tex for a while. At the end, 2 mL of Na
2
CO
3
(2%, w/v) was
added to the tubes and they were incubated for 2 h in a dark
place at room temperature. At the end of the incubation, the
absorbance of the samples was recorded at 760 nm using a
spectrophotometer (8453E UV-Vis, Spectroscopy System,
Agilent). TPCs of the mixed samples were calculated as mg
of gallic acid equivalents (GAE) per kg sample.
Antiradical Activity. Antiradical activity (AA) analysis
was performed using 2,2-diphenyl-1-picrylhydrazyl (DPPH)
free radical according to the method of Karaman et al.
(2014). Three point nine milliliter of DPPH solution
(0.1 mM in methanol) was incorporated into the tubes con-
taining 0.1 mL of extract and then the tubes were mixed
using vortex. Then, the samples were incubated for 30 min
in a dark place at 258C. Absorbance of the samples was
recorded at 517 nm using a spectrophotometer (8453E UV-
Vis, Spectroscopy System, Agilent) at the end of incubation.
AA of the blend samples was expressed as % inhibition using
the following equation:
% Inhibition5½ðAbsorbance of control2Absorbance
of sampleÞ=Absorbance of controlðÞ3100
(1)
Rheological Measurements
In order to determine viscoelastic properties of samples, a
stress/strain controlled and peltier temperature control unit
equipped rheometer (Mars III, Karlsruhe, HAAKE, Ger-
many) was used. To perform the rheological measurements,
plate-plate geometry was (plate diameter 35 mm, and gap
size 0.5 mm) used. Prior to frequency sweep test application,
the linear viscoelastic region (LVR) of the blend samples was
determined using stress sweep test. In LVR test, variation of
dynamic mechanical spectra of samples versus increased
stress was characterized. LVR of the blend samples was deter-
mined over a stress range of 0.1–10 Pa at constant frequency
(1 Hz). In oscillatory frequency sweep test, dynamic mechan-
ical spectra of the mixed samples were evaluated in the fre-
quencyrangeof0.1–10Hzatconstantstress(0.2Pa,within
the range of LVR) at constant temperature (258C). A sinusoi-
dal stress or strain with an increasing frequency was applied
to the samples and the elastic modulus G0,theviscousmodu-
lus G00, complex modulus G* and complex viscosity g*and
loss tangent were calculated as a function of frequency.
Sensory Analysis
The sensory analysis of the blend samples was performed
based on the protocols described before
16
.Basically,50gof
each blend was presented and served at certain intervals in
randomly coded glass beakers of 100-mL capacity. Sensory
evaluation was performed in a room with appropriate tem-
perature (258C) in open sitting. Sensory analyses of the sam-
ples were carried out by fifteen selected staff and graduate
students of Food Engineering Department at Erciyes Univer-
sity, comprised of 10 females and 5 males. Each panelist was
trainedbeforeevaluationinordertofamiliarizewiththesen-
sory analysis, samples and methodology. All coded blend sam-
ples were evaluated for color, oiliness, spreadability, firmness,
adhesiveness, mouth coating, taste and overall acceptance
properties in a scale ranging from 1 to 9 points where 1
reflected a very low in terms of disliking and 9 a very high
score in terms of liking. Panelists evaluated all (15) samples in
three sessions (five at each session) consecutively in same day.
Experimental Design and Optimization
Simplex Lattice Mixture Design. In the present study,
the simplex lattice mixture design (SLMD) was used to eval-
uate the effect of grape molasses (x
1
), sesame paste (x
2
)and
honey (x
3
) on some physicochemical, compositional, bioac-
tive, viscoelastic and sensory properties of the blended
S. KARAMAN ET AL.GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS
Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc. 3
samples. Component proportions in the blends were
expressed as fractions of the mixture with a sum
(X
1
1X
2
1X
3
) of one. These three factors; namely, grape
molasses, sesame paste and honey (processing components),
levels and experimental design in terms of coded and
uncoded as 15 combinations are presented in Table 1.
Multiple linear regression analysis approach was used in
the modeling. The following second-order polynomial equa-
tion of function x
i
was fitted for each factor assessed at each
experimental point.
Y5X
3
i51
bixi1X
3
i51
i<jX
3
j5i11
bijxixj
5b1x11b2x21b3x31b12x1x21b13 x1x31b23x2x3
(2)
where Ywas the estimated mixture response; b
1
,b
2
,b
3,
b
12
,
b
13
and b
23
were linear and interaction terms, respectively, pro-
duced for the prediction models of processing components.
Predictive models were constructed to evaluate the effect of
mixture components (grape molasses, sesame paste and honey)
on the characterized properties of blended samples. The best
fitting models were determined using multiple linear regres-
sions with backward elimination regression (BER) wherewith
insignificant factors and interactions were eliminated from the
regression models and only the variables having significant
effect at P<0.01, P<0.05 and P<0.1 levels were selected for
the model construction using BER procedure.
Multiple Response Optimization. In industrial appli-
cations, optimization should be synchronously performed
for all the responses involved since all responses are correla-
tively changed. In other words, it is not possible to think a
response would change alone; namely independent of other
responses. Moreover, a competition occurs between these
responses in many cases; namely, improving one response
may lead another response to deteriorate, further complicat-
ing the situation. In order to overcome this problem, multi-
ple responses are solved through use of a desirability
function which combines all the responses into one mea-
surement (Yilmaz et al. 2011). Therefore, in this study, mul-
tiple response optimization (MRO) procedure that is
applied to find the operating conditions, xproviding the
“most desirable” response values was followed. After each
response variable was calculated, desirability values were
combined into a single desirability index, D.Forthispur-
pose, each response was transformed in a dimensionless
function. This is called partial desirability function, d
i
which
reflects the desirable ranges for each response ranging from
zero to one (least to most desirable, respectively). It is possi-
ble to calculate the weighted geometric mean of nindividual
desirability functions (all transformed responses) (Eq. (3))
by definition of the partial desirability functions (Eq. (3)).
The simultaneous objective function is a geometric mean of
all transformed responses (Lewis et al. 1949; Myers and
Montgomery 1995):
D5d1p13d2p23d3p33...3dnpi
ðÞ
1=Ppi
5Y
n
i51
dipi
"#
1=Ppi(3)
where p
i
was the weighting of the i
th
term, and normalized
in order that Pn
i51pi51. By weighting of partial desirability
functions, it is possible to enable the optimization process to
take the relative importance of each response into considera-
tion. Allowing the examination of the form of the desirabil-
ity function, it is permitted to find the region where the
function is close to 1 and determine the compromised opti-
mum conditions.
Statistical Analysis
Design-Expert version 7.0 (Stat-Ease Inc., Minneapolis) and
JMP version 9.0.2 (SAS Institute, Inc., Cary, NC) were used
for the computational work including designation of experi-
mental points, randomization and fitting of the second-
order polynomial models as well as optimization. Analysis of
variance was performed using the JMP version 5.0.1 (SAS
Institute, Inc.). Least Significant Differences test was used to
determine the significant differences at (P<0.05) between
blends for each parameter.
RESULTS AND DISCUSSION
Physico-chemical Properties of Blends
Physico-chemical properties of GSHB are presented in Table
2. L,aand bvalues were found to be 7.85, 20.008 and 1.168
for grape molasses; 45.11, 3.12 and 13.88 for sesame paste
and 85.09, 1.308 and 39.79 for honey samples, respectively.
Slight differences were observed between the results reported
by Toker et al. (2013) and our findings, which might have
resulted from concentration and types of pigment com-
pounds such as anthocyanins present in molasses. L,aand b
values of the blends changed between 12.41 and 41.74, 2.678
and 7.514 and 5.400 and 15.98, respectively, highlighting
that as expected concentration of the mixture components
significantly influenced the color values of the blends
(P<0.05). Effect of concentration of each mixture compo-
nent and their interactions on color values of GSHB was
indicated by the predicted model equations presented in
Tabl e 3 . A s i s se e n R
2
values calculated for L,aand bvalues
were 0.93, 0.85 and 0.89, respectively, indicating that the
GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS S. KARAMAN ET AL.
4Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc.
generated models could be used to predict the color values
of the blends depending on grape molasses, sesame paste
and honey concentration. As understood from the equa-
tions, the linear terms in all predicted models were found to
be significant (P<0.05). Similar results attributed to con-
centration of mixture components and color values were
reported by Akbulut et al. (2012) who determined that addi-
tion of honey in sesame paste/honey blends increased the L,
aand bvalues. In addition, Alpaslan and Hayta (2002)
determined that the brightness value decreased with addition
of grape molasses in sesame paste/grape molasses blends.
When conceiving the importance of color in preferability of
the products, color of the blend could be adjusted depending
on the consumer group by using predicted model equations.
Regarding the other physico-chemical properties; namely,
pH, a
w
, ash, dry matter and brix values, they were observed
to be affected by addition of each component since their
properties were found to be different from each other. pH
and a
w
values were determined as 5.73 and 0.70 for grape
molasses, 6.50 and 0.45 for sesame paste and 4.22 and 0.44
for honey, respectively. From magnitudes of the regression
coefficients of each component (Table 3), it is clear that ses-
ame paste and grape molasses had the greatest effect to
increase pH and a
w
values of the blends, respectively. As
known pH and a
w
values play an important role in microbial
stability of the product; therefore, these values could be
standardized regarding storage conditions of the blends by
changing concentration of mixture compounds. Ash, dry
matter content and brix values of grape molasses, sesame
paste, honey and their mixtures were found to between
0.136 and 3.282%, 73.5 and 95.7% and 68.0 and 81.9,
respectively. Model equations established for those values
and their R
2
values are also presented in Table 3. R
2
values
were found as 0.99, 0.98 and 0.86 for ash, dry matter content
and brix value, respectively. All linear terms were significant
in the models. Regression coefficients in the models asserted
that as sesame paste had the greatest increasing effect on ash
and dry matter content and regarding brix value, honey had
the greatest effect, which was expected when those properties
of the grape molasses, sesame paste and honey samples as
mentioned above were taken into consideration. As it is well
known that determination of calorie value of such products
is substantial during product development step; therefore,
those established models can assist calorie calculations of the
blends.
Fructose, glucose and sucrose contents of the grape molas-
ses, sesame paste and honey and their blends prepared
according to formulation presented in Table 1 also tabulated
in Table 2. Sugar contents of the sesame paste were found to
be very low. Fructose, glucose and sucrose concentrations of
the grape molasses were determined as 22.1, 26.8 and 0.37%,
and those for honey was 30.3, 27.9 and 1.64, respectively.
Blend formulation markedly influenced the sugar composi-
tion as expected and fructose, glucose and sucrose concen-
trations of the blends were found to between 7.04 and 31.1,
9.26 and 32.1 and 0.13 and 2.40%, respectively (Table 2).
Model equations were established for sugar contents and R
2
values were stated in Table 3. The sugar levels of the samples
prepared according to the mixture levels could be calculated
using mass balance. In this study, all mixtures were exposed
to sugar analysis to understand the interactive reactions lev-
els between reducing sugars in grape molasses or honey and
proteins in sesame paste. For that reason, regression models
were constructed to predict the sugar levels of the sample
TABLE 2. MEAN VALUES FOR PHYSICO-CHEMICAL PROPERTIES OF GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS
Blends
Physical properties Chemical properties Sugar composition
La bpH a
w
Ash (%)
Dry
matter
(%) Brix
Fructose
content
(%)
Glucose
content
(%)
Sucrose
content
(%)
1 7.852m 20.008l 1.168k 5.73j 0.700a 1.447fg 73.5k 72.4h 22.1f 26.8c 0.37fg
2 45.11b 3.120i 13.88f 6.50a 0.446g 3.282a 99.6a 68.0fg 0.03o 1.86i 0.38fg
3 85.09a 1.308k 39.79a 4.22n 0.440g 0.136k 85.7g 81.7a 30.3b 27.9b 1.64b
4 23.60k 6.056d 10.27i 6.04f 0.632cd 2.321cd 82.3h 77.3c 12.3l 15.7g 0.30g
5 40.30d 5.298f 15.98b 6.23d 0.531f 1.760e 92.6c 73.8fg 15.9j 15.1g 0.87e
6 23.74k 7.404a 12.01h 5.31l 0.630d 0.747ij 79.7i 77.4c 31.1a 32.1a 1.52c
7 31.00h 5.406f 13.82f 6.25c 0.567e 2.874b 92. 8c 73.2gh 7.04n 10.0h 0.41f
8 37.75e 2.678j 13.67g 6.05f 0.537f 0.973hi 91.1d 81.9a 23.1e 22.2e 1.70b
9 12.41l 3.880g 5.400j 5.53k 0.670b 1.135h 75.3j 74.5e 25.6d 28.2b 0.90e
10 27.76j 7.514a 13.59g 5.99g 0.660bc 1.824e 79.8i 74.2f 19.1i 23.3d 0.37fg
11 41.74c 3.414h 13.92f 6.46b 0.364h 2.575c 95.7b 73.6fg 9.40m 9.26h 0.39fg
12 29.95i 6.630c 14.57d 4.96m 0.583e 0.511j 82.7h 78.7b 27.6c 27.7b 2.40a
13 31.47g 7.166b 15.17c 5.88h 0.630d 1.565ef 82.8h 76.4d 19.8h 23.2d 1.26d
14 33.95f 5.328f 14.36e 6.09e 0.584e 2.122d 89.7e 78.5b 14.2k 16.9f 0.13h
15 33.95f 5.796e 15.28c 5.78i 0.593e 1.214gh 86.7f 77.9bc 21.8g 23.0de 1.15d
a–p
Different superscript lowercase letters show differences between the rows (mixtures) (P<0.05).
S. KARAMAN ET AL.GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS
Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc. 5
TABLE 3. EFFECT OF EACH MIXTURE COMPONENT
a
AND THEIR INTERACTIONS ON PARAMETERS OF GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS AS PREDICTED BY MODEL EQUATIONS
Parameters (Y) Predicted model equations
b
R
2
Physicochemical properties
LLog
10
(Y)50.88X
1
11.68X
2
11.82X
3
10.81X
1
X
2
20.61X
2
X
3
11.87X
1
X
2
(X
1
2X
2
) 0.934
aY520.10X
1
14.03X
2
12.40X
3
119.50X
1
X
2
122.50X
1
X
3
122.40X
1
X
2
(X
1
2X
2
) 0.853
b
†
Log
10
(Y)50.15X
1
11.09X
2
11.38X
3
12.16X
1
X
2
11.03X
1
X
3
12.53X
1
X
2
(X
1
2X
2
) 0.893
pH Y55.71X
1
16.47X
2
14.31X
3
11.01X
1
X
3
13.79X
2
X
3
23.48X
2
X
3
(X
2
2X
3
) 0.982
a
w
Y50.68X
1
10.44X
2
10.44X
3
10.27X
1
X
2
10.28X
1
X
3
10.19X
2
X
3
20.84X
2
X
3
(X
2
2X
3
) 0.949
Ash
†
Y51.45 X
1
13.30 X
2
10.14X
3
20.14X
1
X
2
10.23X
2
X
3
20.73X
1
X
2
(X
1
2X
2
) 0.993
Dry matter Log
10
(Y)51.86X
1
12.00X
2
11.94X
3
0.981
Brix Log
10
(Y)51.86X
1
11.84X
2
11.92X
3
10.14X
1
X
2
0.861
Sugar composition
Fructose content
†
Log
10
(Y)51.36X
1
21.41X
2
11.49X
3
15.13X
1
X
2
10.19X
1
X
3
15.35X
2
X
3
210.55X
1
X
2
X
3
24.68X
1
X
2
(X
1
2X
2
)15.22X
2
X
3
(X
2
2X
3
) 0.982
Glucose content
†
Log
10
(Y)51.45X
1
10.30X
2
11.46X
3
11.40X
1
X
2
11.27X
2
X
3
21.06X
1
X
2
(X
1
2X
2
)11.01X
2
X
3
(X
2
2X
3
) 0.990
Sucrose content
†
Y50.34X
1
10.16X
2
11.84X
3
12.21X
1
X
3
0.820
Bioactive properties
Total phenolic content Log
10
(Y)53.25X
1
12.61X
2
12.49X
3
10.60X
1
X
2
10.74X
1
X
3
10.56X
2
X
3
20.39 X
1
X
2
(X
1
2X
2
)20.39X
1
X
3
(X
1
2X
3
)10.31 X2X3(X22X3) 0.993
Antiradical activity
†
Log
10
(Y)51.38X
1
10.67X
2
10.78X
3
20.16X
1
X
2
21.05X
1
X
3
22.51562X
2
X
3
112.50X
1
X
2
X
3
15.98X
2
X
3
(X
2
2X
3
) 0.839
Viscoelastic properties
G0
†
Log
10
(Y)52.88X
1
12.22X
2
11.53X
3
13.60X
1
X
2
16.63X
2
X
3
210.6X
1
X
2
(X
1
2X
2
)19.66X
2
X
3
(X
2
2X
3
) 0.826
G00
†
Log
10
(Y)52.51X
1
12.02X
2
12.02X
3
15.18X
1
X
2
10.17X
1
X
3
16.68X
2
X
3
219.71X
1
X
2
X
3
212.63X
1
X
2
(X
1
2X
2
)17.52X
2
X
3
(X
2
2X
3
) 0.912
g*
†
Log
10
(Y)51.93X
1
11.41X
2
11.25X
3
15.34X
1
X
2
10.94X
1
X
3
17.17X
2
X
3
221.77X
1
X
2
X
3
211.21X
1
X
2
(X
1
2X
2
)19.35X
2
X
3
(X
2
2X
3
) 0.903
G*
†
Log
10
(Y)52.72X
1
12.20X
2
12.05X
3
15.34X
1
X
2
10.94X
1
X
3
17.17X
2
X
3
221.77X
1
X
2
X
3
211.20X
1
X
2
(X
1
2X
2
)19.35X
2
X
3
(X
2
2X
3
) 0.902
tan d
†
Y50.59X
1
11.15X
2
18.70X
3
20.40X
1
X
2
– 18.32X
1
X
3
217.06X
2
X
3
134.22X
1
X
2
X
3
119.60X
1
X
3
(X
1
2X
3
)18.85X
2
X
3
(X
2
2X
3
) 0.958
Sensory properties
Color Y58.05X
1
17.80X
2
17.93X
3
210.60X
1
X
2
26.38X
1
X
3
28.36X
2
X
3
174.0X
1
X
2
X
3
18.70X
1
X
2
(X
1
2X
2
) 0.893
Oiliness Y58.23X
1
17.80X
2
16.47X
3
28.28X
1
X
2
210.2X
1
X
3
24.25X
2
X
3
156.4X
1
X
2
X
3
28.53X
1
X
3
(X
1
2X
3
) 0.901
Spreadability Y55.03X
1
16.45X
2
17.58X
3
22.28X
1
X
2
25.85X
1
X
3
27.78X
2
X
3
181.2X
1
X
2
X
3
117.7X
1
X
3
(X
1
2X
3
) 0.822
Firmness Y56.55X
1
16.32X
2
17.72X
3
14.56X
1
X
2
29.70X
1
X
3
0.804
Adhesiveness Y55.71X
1
14.40X
2
17.58X
3
12.51X
1
X
2
24.86X
1
X
3
22.54X
2
X
3
154.8X
1
X
2
X
3
0.851
Mouth coating Y55.88X
1
16.29X
2
17.44X
3
10.66X
1
X
2
25.97X
1
X
3
22.25X
2
X
3
150.73X
1
X
2
X
3
0.712
Taste Y58.30X
1
16.73X
2
18.43X
3
24.07X
1
X
2
28.27X
1
X
3
25.18X
2
X
3
167.3X
1
X
2
X
3
0.728
Overall acceptance Y57.67X
1
16.82X
2
18.28X
3
22.41X
1
X
2
28.54X
1
X
3
26.47X
2
X
3
141.3X
1
X
2
X
3
110.13X
1
X
2
(X
1
2X
2
) 0.812
a
X
1
,X
2
and X
3
were the mixture components; grape molasses, sesame paste and honey, respectively.
b
By applying BER “backward elimination regression” procedure, non-significant interactions were removed from the equations. Only the variables significant at P<0.01, P<0.05 and P<0.1 levels
were selected for the predicted model construction.
†
These were parameters whose ratios of maximum response to minimum one were greater than 10, which means that a transformation was required. Therefore, a logarithmic transformation,
Y5log
10
(y1k) was performed for these parameters.
GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS S. KARAMAN ET AL.
6Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc.
without performing an analysis. R
2
values calculated for
fructose, glucose and sucrose content were found to be 0.98,
0.99 and 0.82, respectively, implying that sugar contents of
the GSHB could be satisfactorily predicted depending on
grape molasses, sesame paste and honey concentration found
in the formulation. All of the linear factors significantly
affected the sugar contents as understood from equations.
As expected sesame paste had the lowest effect on the change
in concentration of all sugar types and honey had the great-
est effect. Generally, sugar contents were increased by honey
and grape molasses in GSHB (Table 2), sesame paste which
contain less sugar than the others decreased fructose
(Tables 2 and 3). Chemically, honey (80–85%) and molasses
(%70–72) comprise sugar (Seng€
ul et al. 2005; Akbulut et al.
2012; Tornuk et al. 2013). As is seen, generally sugar compo-
sition level increased with the addition of grape molasses
and honey in blends, which should be taken into considera-
tion during product optimization.
Bioactive Properties of Blends
Table 4 shows the TPC and AA values of both sole and blend
samples. For the mixture components, grape molasses
showed the highest TPC (1,751 mg GAE/kg sample) com-
pared to sesame paste (410.2 mg GAE/kg sample) and honey
(310.3 mg GAE/kg sample). For the blend samples, the high-
est TPC level was in the sample (R9) prepared by the mix-
ture of grape molasses and honey at the ratio of 75:25 (w/w)
while the lowest value was in the sample (R8) by the mixture
of sesame paste and honey at the ratio of 75:25 (w/w). Figure
1a shows the change in the TPC of the samples depending
on the level of mixture component in the prepared blends. It
is clear from the figure that TPC increased toward to the ver-
tex of the grape molasses and decreased to the vertex of the
honey. According to table, 100% grape molasses is inclusive
blend which involves the highest results of TPC. Sesame
paste and grape molasses mixtures at the level of 50:50 (w/
w) showed higher TPC compared to blend of honey : grape
molasses at the level of 50:50 (w/w) because ternary plot
shows higher increment toward to the edge of sesame paste
and grape molasses blends. Statistically significant differences
were determined in terms of their TPC among the blends
(P<0.05). AA values of samples were also tabulated in Table
4 and it was seen that there was a significant positive correla-
tion between TPC and AA (r50.853, P<0.05). The highest
AA value (18.91%) was determined in grape molasses sample
while the lowest (0.306%) was in the sample having the low-
est TPC among the blend samples. Ozturk et al.(2014)
reported that there was a relationship between TPC and rad-
ical scavenging activity. Figure 1b shows the ternary plot of
the change in the AA values of samples. As similar to the
change in TPC depending on the mixture component, AA
significantly increased toward to the vertex of the grape
molasses and decreased toward to the edge of sesame paste
and honey. Table 3 shows the predicted regression equations
showing the effect of concentration of each mixture compo-
nent and their interactions on bioactivity values of blend
samples. High determination of coefficients showed that the
constructed models for TPC and AA could be used to pre-
dict the bioactivity values of the blends depending on grape
molasses, sesame paste and honey concentration (Table 3).
As understood from the equations, the linear terms of all
TABLE 4. MEAN VALUES FOR BIOACTIVE AND VISCOELASTIC PROPERTIES OF GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS
Blends
Bioactive properties Viscoelastic properties
Total phenolic
content (mg
GAE/kg sample)
Antiradical
activity
(% inhibition) G0(Pa) G00 (Pa) h*(Pa s) G* (Pa) tan d
1 1751a 18.91a 359.4cd 258.2c 70.55e 443.3e 0.747cd
2 410.2i 4.883g 74.87d 70.61c 16.42e 103.21e 1.023cd
3 310.3j 9.942d 12.49d 108.0c 17.31e 108.75e 8.965a
4 1232c 9.601d 1658cd 1159c 322.1d 2023d 0.703cd
5 528.7h 2.286h 2342cd 2786c 579.3c 3640c 1.190cd
6 1122d 8.974de 389.1cd 188.1c 68.85e 432.6e 0.488cd
7 815.1f 7.545ef 38870b 51820a 10375a 65195a 1.387bc
8 379.7i 0.306i 264.2cd 525.0c 97.47de 612.3de 2.225b
9 1462b 16.00b 406.8cd 294.9c 82.97de 521.2de 0.909cd
10 1436b 15.15b 545.0cd 271.1c 96.85de 608.7de 0.498cd
11 497.6h 4.516g 48460a 16899b 8390.0b 52715b 0.352d
12 725.3g 2.163h 147.4d 123.2c 30.72e 193.0e 0.910cd
13 1166d 11.98c 713.0cd 365.6c 127.5de 801.2de 0.514cd
14 886.3e 8.517def 3006c 2433c 616.6c 3874c 0.810cd
15 805.5f 6.745f 295.7cd 328.8c 70.68e 444.1e 1.144cd
a–i
Different superscript lowercase letters show differences between the rows (mixtures) (P<0.05).
S. KARAMAN ET AL.GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS
Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc. 7
mixture components in the predicted models for bioactivity
properties were found to be significant (P<0.05). As a con-
clusion, the mixture component type used in the prepared
blend has significant effect on the bioactive activity of the
final product. It was obvious from these results that grape
molasses had a major effect on phenolic content and AA.
The reason for this effect is undoubtedly their contents;
according to literature; grape molasses contains high
amounts of sugar, mineral and organic acid in addition to
sesame paste is rich source in dietary fiber, niacin, calcium,
iron, phosphorous, thiamin and sesamol (Ust€
un and Tosun
1997; Demir€
oz€
uet al.2002;Seng
€
ul et al. 2005) and honey
contains organic acids (gluconic acid, acetic acid, etc.), vita-
mins (ascorbic acid) and phenolic substances such as flavo-
noids and carotenoids (Habibi-Najafi and Alaei 2006;
Gharehyakheh et al. 2014).
Viscoelastic Properties of Blends
Loss modulus, storage modulus, complex viscosity, complex
modulus and tangent delta values were determined for the
each sample and the viscoelastic properties of samples were
shown in Table 3. Storage modulus and loss modulus values
of all samples were determined as a function of frequency
and generally, an increase in the frequency increased the
storage and loss modulus values of the samples (Fig. 2).
Grape molasses and sesame paste exhibited elastic behavior,
showing that G0value was higher than G00 values. However,
honey samples exhibited viscous behavior. It was found that
G00 was significantly (P<0.05) higher compared to G0value
in honey (R3) and the blends rich in honey. Kayacier et al.
(2014) and Yilmaz et al. (2014) reported that the loss modu-
lus values of honey were tremendously higher than storage
modulus, showing that honey is weak gel like liquid product
FIG. 1. TERNARY CONTOUR PLOTS SHOWING THE EFFECTS OF PROCESSING COMPONENTS ON BIOACTIVE PARAMETERS G0,G00 AND g*GRAPE
MOLASSES/SESAME PASTE/HONEY BLENDS
GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS S. KARAMAN ET AL.
8Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc.
because of its Newtonian behavior. Storage modulus values
of samples at 1 Hz ranged between 12.49 and 48,460 Pa. The
lowest storage modulus value was in the honey sample (R3)
while the highest was in blend sample prepared by sesame
paste : honey at the level of 75:25 (w/w). The differences
between the all storage modulus values of samples were
determined to be significant statistically (P<0.05) As can be
seen clearly from the Fig. 1, elastic modulus increased tre-
mendously toward to the vertex of honey and there was a
decrease toward to the edge of grape molasses-honey. Loss
FIG. 2. G0(STORAGE MODULUS) AND G00 (LOSS MODULUS) VALUES OF GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS AS A FUNCTION OF
ANGULAR FREQUENCY (x). R1–R15, THE EXPERIMENTAL BLENDS (RUNS) FROM 1 TO 15
S. KARAMAN ET AL.GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS
Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc. 9
modulus values of the samples also changed depending on
the mixture component significantly (P<0.05) and the low-
est loss modulus value (108 Pa) was in the sample of sesame
paste. The highest value (16,899 Pa) was recorded for the
blend prepared with sesame paste and honey (R11). The vis-
cous behavior of honey samples systematically increased as
function of angular frequency. Domination viscous behavior
of samples indicated weak particle–particle interactions and
there was no network formation in honey samples. Complex
viscosity values of samples were also affected by the mixture
component significantly (P<0.05). Addition of honey,
grape molasses and sesame paste to the blends increased vis-
cosity of the final products (Table 3). Specially, addition of
the sesame paste in blends increased viscosity tremendously
(Fig. 2e). In the blends coded as R7–R11 containing %75
sesame paste was present, complex viscosity was the highest.
However, the lowest complex viscosity was observed in the
blends coded as R3–R12 containing 75–100% honey (Table
3). These results indicated that sesame paste was the compo-
nent having the highest increasing effect on the consistency
of the blends samples. Furthermore, binary interactions of
sesame paste had more effect on increasing viscosity in the
blends. This effect can be attributed to interactions between
protein-carbohydrates in sesame paste blends. Effect of con-
centration of each mixture component and their interac-
tions on viscoelastic parameters was indicated by the
predicted model equations presented in Table 3. As is seen
R
2
values calculated for G0,G00,G*, g*andtandvalues were
0.83, 0.91, 0.90, 0.90 and 0.96, respectively, indicating that
the constructed models could be used to predict the
dynamic mechanical properties of the blends depending on
grape molasses, sesame paste and honey concentration
(Table 3). It is clear from the equations that the linear terms
in all regression models were found to be significant
(P<0.05). It was reported that the viscosity of sesame paste
known as tahin increased with the increase of honey addi-
tion significantly (P<0.05). Honey is rich in sugar which is
responsible for the viscosity of honey, whereas sesame paste
is basically structured oil and protein
13
. The increase in the
viscosity is of course is also related to solid content because
the higher solid content generally cause increment in the
viscosity because of molecular movements and interfacial
film formation (Bhattacharya et al. 1992; Maskan and
G€
o
g€
us¸ 2000; Alpaslan and Hayta 2002). The other visco-
elastic parameters namely complex modulus and tangent
delta were also significantly affected by the mixture compo-
nents (P<0.05).
Sensory Properties of Blends
As is known, during product formulation, optimization of
sensory analysis of the product is escapable since it deter-
mines acceptance and rejection of the product according to
consumers’ response. All sensory properties of the GSHB
were generally affected by concentration of each mixture
component concentration (Table 5); therefore, those results
should be considered depending on the consumer require-
ment during formulation step. Established model equations
and R
2
values are also tabulated in Table 3 where findings
indicated that dependent parameters (color, oiliness, spread-
ability, firmness, adhesiveness, mouth coating taste and over-
all acceptance) can be victoriously predicted using generated
models. Due to a dilution effect of sesame paste on the
sweetness of grape molasses and honey, sesame paste may
increase the sensory properties of blends
13
. The adhesiveness
TABLE 5. MEAN VALUES FOR SENSORY PROPERTIES OF GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS
Blends
Sensory properties Viscoelastic properties
Color Oiliness Spreadability Firmness Adhesiveness
Mouth
coating Taste
Overall
acceptance
1 8.2a 8.2a 5.3cde 6.5bcdef 5.6bcde 5.8cd 8.5a 7.9ab
2 8.0ab 8.0ab 6.9abc 6.2cdef 4.8de 6.7abc 7.1bcdefg 7.0abcd
3 7.9abc 6.4cd 7.4a 7.7ab 7.5a 7.2abc 8.2abc 8.1a
4 6.1ef 6.3cd 6.4abcd 8.0a 6.4abcd 7.2abc 7.5abcde 7.6ab
5 5.6fg 6.0de 5.0de 6.6bcdef 5.1cde 5.9bcd 5.9gh 5.7ef
6 6.7bcdef 4.9e 5.3cde 4.0g 5.6bcde 4.9d 6.7defg 5.8def
7 4.5g 5.9de 3.1f 6.6bcdef 4.1e 5.1d 5.2h 5.0f
8 6.6cdef 6.2cd 6.3abcd 7.5ab 6.6abc 7.0abc 7.3abcdef 7.0abcd
9 6.7bcdef 4.8e 4.3ef 5.7ef 5.1cde 5.1d 6.2fgh 6.1cdef
10 6.3ef 6.5cd 5.9abcde 6.9abcde 5.9abcd 6.1abcd 7.0cdefg 7.3abc
11 6.2ef 6.8bcd 4.8def 7.0abcd 4.8de 6.4abcd 6.3efgh 6.0def
12 6.5def 5.8de 5.6bcde 5.4f 6.2abcd 6.3abcd 7.0cdefg 6.7bcde
13 7.8abcd 7.3abc 7.3ab 6.0def 7.2ab 7.1abc 7.8abcd 7.5ab
14 7.2abcde 6.6cd 7.0abc 7.3abc 7.1ab 7.4ab 8.3ab 6.7bcde
15 7.8abcd 6.8bcd 7.4a 6.7bcde 7.0ab 7.5a 8.2abc 7.0abcd
a–g
Different superscript lowercase letters show differences between the rows (mixtures) (P<0.05).
GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS S. KARAMAN ET AL.
10 Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc.
and spreadability value of blends increased by the addition
of grape molasses to sesame paste. Akbulut et al. (2012)
obtained similar findings about spreadability. Also the
addition of grape molasses to sesame paste increased color
and oiliness values of blends. On the contrary, according to
Alpaslan and Hayta (2002) oiliness had been decreased
with addition of grape molasses to sesame paste. The high-
est oiliness value must be 100% sesame paste in blends. But
in this study, it can be seen that the grape molasses has
highest oilness score (Table 5). Honey indicated more vis-
cous characteristic which influenced more mouth coating
than the others and addition of honey in blends indicated
more mouth-coating properties (Table 5). Overall accep-
tance of the GSHB was significantly affected by all of the
linear factors in the following order from greatest effect
to lowest effect: honey, sesame paste and grape molasses.
Figure 3 shows the ternary contour plots for color, spread-
ability, mouth coating, taste and overall acceptance
parameters.
Multiple Response Optimization
Foods are very complex products; therefore, during the opti-
mization of formulation, many factors should be taken into
consideration meanwhile since there are many factors deter-
mining quality of the product. Accordingly, in the present
study, MROs were performed to optimize formulation of
GSHB. Three different optimization criteria were deter-
mined. One of them was bioactive properties: the blend
including 30.28% of sesame paste and 69.72% of honey had
the highest TPC and AA (Table 6). Regarding viscoelastic
properties, the sample containing 34.01% of grape molasses,
4.89% of sesame paste and 61.10% of honey had the highest
viscoelastic parameters’ (G0,G00,g*,G* and tan d)value.The
FIG. 3. TERNARY CONTOUR
PLOTS SHOWING THE
EFFECTS OF PROCESSING
COMPONENTS ON SENSORY
PROPERTIES OF GRAPE
MOLASSES/SESAME PASTE/
HONEY BLENDS
S. KARAMAN ET AL.GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS
Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc. 11
last criterion was selected as sensory properties of the sam-
ples and the blend comprised of 57.31% of grape molasses
and 42.69% of honey had the highest sensory scores.
CONCLUSIONS
Honey, sesame paste and grape molasses are widely con-
sumed in breakfast. Blending of them provides advantage in
many aspects. By this way, it is possible to fabricate products
with functional characteristics, sweet and unique aroma,
which can motive people of all ages to consume this novel
natural product. Concentration of honey, sesame paste and
grape molasses plays an important role in determining qual-
ity characteristics of the blend; namely, chemical, bioactive,
rheological and sensory properties. Therefore, optimization
of product formulation is vital. In this study, SLMD was
accomplishedly performed to observe change in those qual-
ity parameters as a function of honey, sesame paste and
grape molasses concentration. Almost all parameters were
significantly affected by concentration of them. MRO tech-
nique was conducted to simultaneously optimize quality
parameters of the blend, which is very important for the
food industry in many aspects. By means of mixture design,
it is possible to produce blend depending on intended use
(consumer group) and storage condition considering quality
parameters of the final product. In addition, this natural and
functional blend can be used in formulation of different
food products such as biscuits, cake, chocolates, etc.
REFERENCES
ABU-JDAYIL, B. 2004. Flow properties of sweetened
sesame paste (halawa tehineh). Eur. Food Res. Technol. 219,
265–272.
ABU-JDAYIL, B., AL-MALAH, K. and ASOUD, H. 2002.
Rheological characterization of milled sesame (tehineh). Food
Hydrocoll. 16,55–61.
AHMED, J., PRABHU, S.T., RAGHAVAN, G.S.V. and NGADI,
M. 2007. Physico-chemical, rheological, calorimetric and
dielectric behavior of selected Indian honey. J. Food Eng. 79,
1207–1213.
ALPASLAN, M. and HAYTA, M. 2002. Rheological and sensory
properties of pekmez (grape molasses)/tahin (sesame paste)
blends. J. Food Eng. 54,89–93.
AKBULUT, M., C¸OKLAR, H. and €
OZEN, G. 2008. Rheological
characteristics of Juniperus drupacea fruit juice (pekmez)
concentrated by boiling. Food Sci. Technol. Int. 14,321–328.
AKBULUT, M., SARICOBAN, C. and OZCAN, M.M. 2012.
Determination of rheological behavior, emulsion stability,
color, and sensory of sesame pastes (Tahin) blended with pine
honey. Food Bioprocess Technol. 5, 1832–1839.
AOAC. 2000. Official Methods of Analysis,17thEd.,Assoc.ofOffi-
cial Analytical Chemists, Arlington, VA.
TABLE 6. MULTIPLE RESPONSE OPTIMIZATION PROCEDURE APPLIED TO FIND OPTIMUM VALUES FOR BIOACTIVE, VISCOELASTIC AND SENSORY
PARAMETERS OF GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS
Multiple response optimization
Minimization Maximization
Parameters
Minimum
levels
Grape
molasses
(%)
Sesame
paste (%)
Honey
(%) Desirability
Maximum
levels
Grape
molasses
(%)
Sesame
paste (%)
Honey
(%) Desirability
Bioactive properties
Total phenolic content* 445.8 0.00 30.28 69.72 0.941 1,739 100 0.00 0.00 0.996
Antiradical activity
†
0.72 19.1
Viscoelastic properties
G0(Pa) 597.1 34.01 4.89 61.10 1.000 19,739 14.27 85.73 0.00 0.342
G00 (Pa) 2,442 32,491
g* (Pa s) 183.8 4273
G* (Pa) 1,155 26,847
tan d0.321 1.289
Sensory properties
Color 6.4 57.31 0.00 42.69 0.718 7.8 34.66 34.11 31.23 0.83
Oiliness 4.7 7.1
Spreadability 4.7 7.6
Firmness 4.7 6.3
Adhesiveness 5.3 7.4
Mouth coating 5.1 7.6
Taste 6.3 8.4
Overall acceptance 5.8 7.2
*Total phenolic content values were expressed as mg GAE/kg sample.
†
Antiradical activity values were expressed as % inhibition.
GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS S. KARAMAN ET AL.
12 Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc.
ARSLAN, E., YENER., M.E. and ESIN, A., 2008. Rheological char-
acterization of tahin/pekmez (sesame paste/concentrated grape
juice) blends. J. Food Eng. 69,167–172.
BHATTACHARYA, S., BHAT, K.K. and RAGHUVER, K.G. 1992.
Rheology of bengal gram Cicer arietinum flour suspensions. J.
Food Eng. 17,83–96.
DEMIR €
OZ€
U, B., S€
OKMEN, M., UC¸AK, A., YILMAZ, H. and
G€
ULDEREN, S. 2002. Variation of copper, iron, and zinc levels
in pekmez products. Bull. Environ. Contam. Toxicol. 69,
330–334.
GHAREHYAKHEH, S., AMIRI, M. and SAVAREH, S.J. 2014.
Rheological behavior of sesame paste/honey blend with differ-
ent concentration of honey. Eur. J. Exp. Biol. 4, 53–57.
HABIBI-NAJAFI, M.B. and ALAEI, Z. 2006. Rheological proper-
ties of date syrup/sesame paste blend. World J. Dairy Food Sci.
1, 01–05.
JAHANBIN, K., MOINI, S., GOHARI, A.R., EMAM-DJOMEH,
Z. and MASI, P. 2012. Isolation, purification and characteriza-
tion of a new gum from Acanthophyllum bracteatum roots.
Food Hydrocoll. 27,14–21.
JUSZCZAK, L. and FORTUNA, T. 2006. Rheology of selected Pol-
ish honeys. J. Food Eng. 75,43–49.
KARAMAN, S. and KAYACIER, A. 2011. Effect of temperature on
rheological characteristics of molasses: Modeling of apparent
viscosity using Adaptive Neuro Fuzzy Inference System
(ANFIS). LWT- Food Sci. Technol. 44, 1717–1725.
KARAMAN, S., TOKER, O.S., C¸AM, M., HAYTA, M., DO
GAN,
M. and KAYACIER, A. 2014. Bioactive and physicochemical
properties of persimmon as affected by drying methods. Drying
Te c h n o l. 32,258–267.
KARAMAN, S., YILMAZ, M.T. and KAYACIER, A. 2011. Simplex
lattice mixture design approach on the rheological behavior of
glucomannan based salep-honey drink mixtures: An optimiza-
tion study based on the sensory properties. Food Hydrocoll. 25,
1319–1326.
KAYACIER, A. and KARAMAN, S. 2008. Rheological and some
physicochemical characteristics of selected Turkish Honeys. J.
Te x t u re S t u d i e s 39,17–27.
KAYACIER, A., YUKSEL, F. and KARAMAN, S. 2014. Dynamic
mechanical spectra of selected Turkish honeys: Construction of
predictive models for complex viscosity using two different
nonlinear modeling techniques. Int. J. Food Prop. 17,93–110.
LEWIS, V.M., ESSELEN, W.B. and FELLERS, C.R. 1949. Nonen-
zymatic browning of foodstuffs. Nitrogen free carboxylic acids
in the browning reaction. Ind. Eng. Chem. 41,2591.
MASKAN, M. and G €
O
G€
US¸, F. 2000. Effect of sugar on the rheo-
logical properties of sunflower oil-water emulsion. J. Food Eng.
43,173–177.
MYERS, R.H. and MONTGOMERY, D.C. 1995. Response Surface
Methodology. Process and Product Optimization Using Designed
Experiments,p.705.Wiley,NewYork.
OZTURK, G., DOGAN, M. and TOKER, O.S. 2014. Physicochem-
ical, functional and sensory properties of mellorine enriched
with different vegetable juices and TOPSIS approach to deter-
mine optimum juice concentration. Food Biosci. 7, 45–55.
SENG€
UL, M., ERTUGAY, M.F. and SENG€
UL, M. 2005. Rheologi-
cal, physical and chemical characteristics of mulberry pekmez.
Food Control 16,73–76.
TOKER, O.S., DOGAN, M., ERS€
OZ, N.B. and YILMAZ, M.T.
2013. Optimization of the content of 5-hydroxymethylfurfural
(HMF) formed in some molasses types: HPLC-DAD analysis to
determine effect of different storage time and temperature lev-
els. Ind. Crops Prod. 50,137–144.
TORNUK, F., KARAMAN, S., OZTURK, I., TOKER, O.S.,
TASTEMUR, B., SAGDIC, O., DOGAN, M. and KAYACIER, A.
2013. Quality characterization of artisanal and retail Turkish
blossom honeys: Determination of physicochemical, microbio-
logical, bioactive properties and aroma profile. Ind. Crops
Prod. 46,124–131.
UST€
UN, N.S. and TOSUN, I. 1997. Pekmezlerin biles¸imi. Gida
22,417–423.(InTurkish).
YILMAZ,M.T.,KARAMAN,S.,CANKURT,H.,KAYACIER,A.and
SAGDIC, O. 2011. Steady and dynamic oscillatory shear rheolog-
ical properties of ketchup–processed cheese mixtures: Effect of
temperature and concentration. J. Food Eng. 103, 197–210.
YILMAZ, M.T., TATLISU, N.B., TOKER, O.S., KARAMAN, S.,
DERTLI, E., SAGDIC, O. and ARICI, M. 2014. Steady, dynamic
and creep rheological analysis as a novel approach to detect
honey adulteration by fructose and saccharose syrups: Correla-
tions with HPLC-RID results. Food Res. Int. 64, 634–646.
S. KARAMAN ET AL.GRAPE MOLASSES/SESAME PASTE/HONEY BLENDS
Journal of Food Process Engineering 00 (2016) 00–00 V
C2016 Wiley Periodicals, Inc. 13
