Heat Transfer during Blanching and Hydrocooling

of Broccoli Florets
scar Hernandez-Calderon, Jorge Zazueta-Niebla, Roberto Gutierrez-Dorado,
Rosalina Iribe-Salazar, Jose Caro-Corrales, O
Marco Carrazco-Escalante, and Yessica Vazquez-Lopez

Abstract: The objective of this work was to simulate heat transfer during blanching (90 C) and hydrocooling (5 C)
of broccoli florets (Brassica oleracea L. Italica) and to evaluate the impact of these processes on the physicochemical and
nutrimental quality properties. Thermophysical properties (thermal conductivity [line heat source], specific heat capacity
[differential scanning calorimetry], and bulk density [volume displacement]) of stem and inflorescence were measured as
a function of temperature (5, 10, 20, 40, 60, and 80 C). The activation energy and the frequency factor (Arrhenius
model) of these thermophysical properties were calculated. A 3-dimensional finite element model was developed to
predict the temperature history at different points inside the product. Comparison of the theoretical and experimental
E: Food Engineering &
Materials Science

temperature histories was carried out. Quality parameters (firmness, total color difference, and vitamin C content) and
peroxidase activity were measured. The satisfactory validation of the finite element model allows the prediction of
temperature histories and profiles under different process conditions, which could lead to an eventual optimization aimed
to minimize the nutritional and sensorial losses in broccoli florets.

Keywords: blanching, broccoli, heat transfer simulation

Practical Application: Temperature histories inside broccoli florets can be estimated during blanching and hydrocooling
at different process conditions. This could be useful to minimize nutritional and sensorial losses during blanching and
cooling of this vegetable.

Introduction ter in determining the evolution of the process is the convective

The blanching and subsequent hydrocooling of broccoli florets
have a beneficial effect on the physicochemical and nutritional
properties, which can improve food quality. One of the main
purposes of blanching is inactivation of the enzymes that have an
adverse effect on food quality, providing a substantial prolonged
shelf life under refrigerated storage conditions (Barrett and others
2000). Broccoli is distinguished by the presence of numer-
ous bioactive substances with health-promoting properties, like
glucosinolates, phenolic compounds, and vitamin C (Domnguez-
Perles and others 2010; Rodrguez-Hernandez and others 2012).
Peroxidase (POD) is the most thermally resistant enzyme in veg-
etables, and therefore, it is generally used as a biological indicator
of the blanching process efficiency (Dosz and Jeffery 2013).
Thermophysical properties depend strongly on temperature and
composition of the food (Espinoza-Guevara and others 2010) and
are essential for designing any food engineering processes, specif-
ically for transformations that include heat transfer such as drying,
pasteurization, sterilization, blanching, and freezing. They are also
important for the prediction and control of various changes oc-
curring in food during thermal processing and storage (Mahapatra
and others 2011).
Accurate modeling of heat transfer to foodstuffs requires precise
knowledge of the thermophysical properties of the material and
the boundary conditions. The most important external parame-
MS 20150885 Submitted 5/26/2015, Accepted 9/14/2015. Authors are with predict temperature profiles and histories inside complex irregu-
Posgrado en Ciencia y Tecnologa de Alimentos, Univ. Autonoma de Sinaloa, Apdo. larly shaped foods, which will help to achieve optimal conditions
Postal 1354, C.P. 80000, Culiacan, Sinaloa, Mexico. Direct inquiries to author
Caro-Corrales (E-mail: josecaro@uas.edu.mx).
heat transfer coefficient. Once the thermophysical properties and
the convective coefficient have been determined they can be in-
putted to a finite element model to predict temperature histories
(Cronin and others 2010). Validation of these predictions will then
be accomplished by comparison with experimental results. Finite
element modeling has been used for successfully simulating several
processing operations and specifically heat transfer in a variety of
foodstuffs such as cheese (Caro-Corrales and others 2010), meat
(Cronin and others 2008), vegetables (Martens and others 2001),
and apples (Kursat and others 2011). A great variety of studies for
simulating 3-dimensional (3D) objects have been carried out using
axis symmetry in 2D regions for nonsteady-state processes that
accurately represent the 3D geometries. Finite element analysis
can be used when the thermophysical properties are spatial and
temperature dependent, the food has an irregular geometry, and
the boundary conditions are no linear. There is a high potential
for the use of finite element modeling during food processing
to optimize food quality in terms of texture and nutrient reten-
tion. Broccoli florets have a complex geometry and in literature
no reports were found on building a 3D finite element model
to simulate heat transfer in nonsteady-state for this vegetable.
An adequate simulation of the heat transfer inside the broccoli
can provide information to estimate the vitamin C retention and
the residual enzyme activity from a heat penetration analysis us-
ing the corresponding decimal reduction times and z values.
Therefore, it is convenient to simulate heat transfer processes to
that minimize nutritional and sensorial loss. The objective of this
work was to simulate heat transfer during blanching (90 C) and

C 2015 Institute of Food Technologists

 R

E2774 Journal of Food Science r Vol. 80, Nr. 12, 2015 doi: 10.1111/1750-3841.13109
Further reproduction without permission is prohibited
Heat transfer of broccoli florets . . .

hydrocooling (5 C) in broccoli florets (Brassica oleracea L. Italica) Activation energy

and to evaluate the impact of these processes on physicochemical
and nutrimental quality properties.
Materials and Methods
Mature broccoli (Brassica oleracea L. Italica) with beads well de-
veloped and firm head was selected for uniformity of size, color,
and absence of defects. Thermophysical properties (thermal con-
ductivity [k], specific heat capacity [Cp ], and bulk density []) of
stem and inflorescence were measured as a function of temperature
(5, 10, 20, 40, 60, and 80 C). Moisture content (AOAC 2012) of where K represents the thermophysical property of stem or inflo-
stem and inflorescence was 92.3% and 88.5%, respectively.
Thermal conductivity
The activation energy was calculated to determine which part
of the broccoli floret is more sensible to the temperature change.
A high activation energy implies that the thermophysical property
depends strongly on temperature. The activation energy (Ea ) and
the frequency factor (A) were calculated from linearization of
Arrhenius equation.
Kt p = Ae RT
Ea
(4)
tp
rescence (k, C , and ) measured at 5, 10, 20, 40, 60, and 80 C,
p
R is the universal gas constant (8.314 J/mol/K), and T is absolute
temperature (K).

E: Food Engineering &

Thermal conductivity was measured with the line heat source or

Materials Science
probe method (Espinoza-Guevara and others 2010). A probe was
inserted inside the sample and temperature change was measured
Convective heat transfer coefficient
(Caro-Corrales and others 2010). Thermal conductivity (k) was For blanching and hydrocooling, the convective heat transfer
obtained by regression analysis between temperature (T) and the coefficient (h) was evaluated using the lumped capacity method. A
natural log of time (t): broccoli floret of 60 mm length was manufactured from aluminum.
A T type thermocouple was inserted in the stem center at a
  depth of 6 mm to obtain the experimental temperature histories
QL t
T T1 = ln (1) using a data acquisition system (DAQ, OMB-DAQ-56; Omega
4k t1 Engineering, Stamford, Conn., U.S.A.). The equation for the
lumped capacity method is:
where QL = I 2 R/L is heat input per unit length of line source
(W/m), I is the supplied current intensity (A), R is the probe T T hA t
resistance (), and L is the probe length (m). Coordinates (ln t1 , = e CP V (5)
T1 ) correspond to the instant that the graph of temperature against T0 T
natural log of time starts to be linear.
where T0 is the initial temperature of the aluminum floret
(20 C), T is bulk water temperature (90 C for blanching and
Specific heat capacity 5 C for hydrocooling), A and V are superficial area and volume of
A differential scanning calorimeter (DSC, TA Instruments 2920, the aluminum floret, which were measured using grid paper and
New Castle, Del., U.S.A.) was used to measure specific heat ca- the liquid displacement method; and C are bulk density and
p
pacity (Cp ). The sample size (m) was about 15 mg and the rate specific heat capacity for aluminum. The convective heat transfer
of temperature scanning (dT/dt) was 20 C/min (Caro-Corrales coefficient (h) was obtained from the slope (hA/[C V]) through
p
and others 2002). Specific heat capacity was measured from 5 to a regression analysis between ln ([T T ]/[T T ]) against time
 0 
90 C. The enthalpy change, H is given on the DSC curve (Caro-Corrales and others 2002).
of heat flow against temperature by the difference in heat flow
between the baseline and test material curve. The specific heat
capacity can be obtained from: Thermal center
Broccoli florets of 60 mm length were used to determine the
E H thermal center. Three different critical zones were selected for this
Cp = (2) purpose: zone I included thin stems near inflorescence, zones II
dT/dt m
and III were located in the central axis of the stem at 8 and 25 mm
where the calibration constant, E was obtained using a calibrating from its base. A T type thermocouple was inserted in each zone and
reference (sapphire) at the temperature of interest. temperature was recorded using a data acquisition module (OMB-
DAQ-56, Omega Engineering, Inc., Stamford, CT, U.S.A.). From
temperature histories for each zone, the slowest heating/cooling
Bulk density
zone for blanching/hydrocooling indicated the thermal center.
The liquid displacement method was applied to measure bulk
density (), using toluene as the liquid medium (Tocci and
Mascheroni 2008). Heat transfer modeling
Finite element modeling (ANSYS software, ANSYS Inc., re-
Thermal diffusivity lease 14.0, Irvine, CA, U.S.A.) was used for heat transfer simulation
Based on its definition, thermal diffusivity () was calculated inside a 3D broccoli floret. Transient heat transfer in a rectangular
from: coordinate system is governed by Fouriers equation:

k T
= (3) CP = T (KT) (6)
C p t

Vol. 80, Nr. 12, 2015 r Journal of Food Science E2775

 Heat transfer of broccoli florets . . .
where T = ( x y z
) and K is the conductivity matrix. The the latter absorbs at 520 nm (Durust and others 1997). Results were
initial and boundary conditions employed: expressed in mg ascorbic acid/100 g of fresh sample.
T (x, y, z, 0) = T0

T 
K = h (TS T )
n   form a brown product known as tetraguaiacol. Changes in ab-
where n is the outward normal unit vector,  the surface bound-
ary, and TS surface temperature. For simulating conduction heat
transfer, thermophysical properties both of stem and inflorescence
as a function of temperature were used. Two different convective
heat transfer coefficients (h), from the lumped capacity method,
for blanching and hydrocooling, were input. Initial temperature
(T0 ) and water temperature (T ) were set at 20 and 90 C for
blanching, and at 90 and 5 C for hydrocooling. Elements were
E: Food Engineering &
Materials Science
selected from the tetrahedral thermal solid family of ANSYS with
ten nodes. The sparse direct solver was used and the time step
size was 3 s. Temperature at the target nodes was obtained from
the corresponding coordinates. For thin stems near inflorescence
(at 35.2 mm), and stem (at 8 and 25 mm), all measured from
the base, coordinates were (0, 0.0352 m, 0.0145 m), (0, 0.008 m,
0.025 m), and (0, 0.025 m, 0.025 m), respectively. For validating
the model, broccoli florets were blanched (90 C) and hydrocooled
(5 C), and then the temperature histories corresponding to the
three critical zones were compared to the finite element solutions.
Once the finite element model was validated, broccoli florets
were blanched (90 C, 3 min) and hydrocooled (5 C) for mea-
suring physicochemical and nutrimental quality properties, before
and after processing. As the product quality depends on cooling
rate and storage time, then, after blanching, florets were also cooled
at room temperature, or frozen with liquid nitrogen, or frozen and
stored for 1 month.
Firmness
As a texture indicator, firmness of broccoli florets was mea-
sured using a digital penetrometer (Chatillon DFE 100, Ametek
TCI division, Largo, Fl., U.S.A.) with an 11-mm-diameter probe
(Ayon-Reyna and others 2015). The surface of the bottom of
each stem was penetrated (5 mm depth) with a constant speed of
50 mm/min (Marangoni and others 1995). Results were reported
as the maximum compression force to penetrate the tissue, ex-
pressed in Newtons (N).
Total color difference
A Minolta colorimeter (model CR-200; Minolta Co. Ltd.,
Osaka, Japan) was used to measure color at the center of the
stem according to Ayon-Reyna and others (2015) based on the
CIELAB color parameters L , a , and b . Total color difference
(E) was calculated using the equation:
 2  2  2 1/2
E = L L f + a a f + b b f (8)
where Lf , af , bf and L , a , b are color parameters for fresh and
after treatment samples, respectively.
Vitamin C
A spectrophotometer (Thermo Electron Corp Genesys 10 UV,
Genesys 10-S, Madison, Wis., U.S.A.) was used to measure vi-
tamin C content through the reaction of ascorbic acid with 2,6-
dichlorophenolindophenol (DCPI). Vitamin C reduces DCPI and
E2776 Journal of Food Science r Vol. 80, Nr. 12, 2015
Peroxidase activity
(7) Activity of peroxidase (POD) enzyme was measured based on
guaiacol oxidation in presence of hydrogen peroxide (H2 O2 ) to
sorbance were determined at 470 nm and 25 C for 3 min at
10 s intervals using a spectrophotometer (Thermo Electron Corp
Genesys 10 UV, Genesys 10-S, Madison, Wis., U.S.A.; Zhang and
others 2005). Volumetric activity (aV ) in U/mL was calculated
with:
 A 
VT
aV = t (9)
l VE xtr
where A/t is the slope (min1 ) of the absorbance against time
graph, is the extinction coefficient [25.5/mM/cm; Thongsook
and Barrett (2005)], VT is the total reaction volume (3.1 mL),
VExtr is the volume of enzymatic extract (0.15 mL), and l is the
cell thickness (1 cm).
Experimental design
For thermophysical properties (k, , Cp , and ), a completely
randomized design was used. Factors were temperature (5, 10, 20,
40, 60, and 80 C) and type of material (stem and inflorescence)
with 5 replicates. For quality parameters, broccoli florets were
blanched at 90 C for 3 min (Gormley and Tansey 2011) and the
factor was the type of treatment [Fresh; Blanching + Freezing
with liquid N2 (B + F); Blanching + Hydrocooling at 5 C
(B + HC 5 C); Blanching + Cooling at room temperature (B +
C 25 C); and Blanching + Thawing after one month of storage
(B + T 1 Mo)] with 3 replicates. The latter treatment (B + T
1 Mo) was previously frozen with liquid N2 . Fishers test was used
to compare means ( = 0.05).
Results and Discussion
Thermophysical properties
Thermal conductivity and specific heat capacity for stem and
inflorescence as a function of temperature are shown in Figure 1.
For both materials, thermal conductivity increased linearly with
temperature. Thermal conductivity ranged from 0.561 to 0.731
W/m K and from 0.326 to 0.687 W/m K for stem and inflores-
cence, respectively. Predictive model of this property for stem was
kstem = 0.0024T + 0.5482 with k in W/m K, T in C, and a de-
termination coefficient (R2 ) of 0.994, rmse = 6.79 103 W/m
K, and for inflorescence kinflorescence = 0.0050T + 0.2802 with R2
of 0.975, rmse = 2.99 102 W/m K, which indicate a high lin-
ear dependence in both materials. Thermal conductivity for stem
of broccoli florets was higher than that for inflorescence, which
means that stem has a higher capacity for transferring energy than
inflorescence, as a result of higher moisture content in stem. This
behavior is similar to that reported by Espinoza-Guevara and oth-
ers (2010). As temperature was increased, specific heat capacity
increased linearly from 3216 to 4813 J/kg K for stem and from
2795 to 4664 J/kg K for inflorescence. Predictive model of spe-
cific heat capacity for stem was C p stem = 22.0T + 3130 with Cp
in J/kg K, T in C, and R2 of 0.990, rmse = 82.1 J/kg K, and
for inflorescence C p inflorescence = 24.8T + 2670 with R2 of 0.986,
rmse = 109 J/kg K. Specific heat capacity for stem was higher,
which indicates a higher capacity for storing energy; this behavior
Heat transfer of broccoli florets . . .

can be attributable to its higher moisture content. For both ma- Table 1Activation energy (Ea ) and frequency factor (A) of ther-
terials, an adequate fitness was obtained. The trend in increasing mal conductivity (k), specific heat capacity (Cp ), and thermal
diffusivity () for stem and inflorescence.
specific heat capacity agrees with reports from Espinoza-Guevara
and others (2010). Part of Frequency
Bulk density () did not change (p > 0.05) at the studied lev- broccoli Property factor Ea (kJ/mol) R2
els of temperature. Mean bulk density was 1031 and 940 kg/m3 k (W/m K) 2.12 3.08 0.994
for stem and inflorescence, respectively. Tocci and Mascheroni Stem Cp (kJ/kg K) 23.0 4.54 0.991
(2008) reported a similar behavior when evaluating bulk density (m2 /s) 8.08108 1.73 0.951
in Kiwi from 0 to 25 C. Thermal diffusivity for stem ranged from k (W/m K) 11.0 8.20 0.986
Inflorescence Cp (kJ/kg K) 29.7 5.45 0.987
1.43 107 to 1.70 107 m2 /s and for inflorescence from 1.23 (m2 /s) 3.71107 2.55 0.931
107 to 1.62 107 m2 /s. The predictive model of this property
for stem was stem = 3.21 1010 T + 1.71 107 with in Units for the frequency factor are the same as for the correspondent thermophysical
property.
m2 /s, T in C, and R2 of 0.939, rmse = 3.06 109 m2 /s, and for
inflorescence inflorescence = 4.34 1010 T + 1.213 107 with
R2 of 0.947, rmse = 3.84 109 m2 /s. Thermal diffusivity was ture range, energy propagates faster than it is stored when heat is

E: Food Engineering &

higher for stem, which indicates that for the studied tempera- transferred in nonsteady state.

Materials Science
Activation energy
The activation energy (Ea ) and the frequency factor (A) for
thermophysical properties that had a temperature dependence (k,
0.8
Cp , and ) in both materials are shown in Table 1. The higher the
A 0.731
absolute value of Ea , the stronger the dependence of the thermo-
0.692
0.7 0.647 physical property with temperature. Therefore, thermal conduc-
Thermal Conductivity (W/m K)

tivity and specific heat capacity for inflorescence (Ea of 8.20 and
0.596 0.687
0.574 5.45 kJ/mol, respectively) were the properties exhibited signifi-
0.6 0.561
cant temperature dependence. These results can be confirmed as
the regression coefficients of temperature (slope) in the predictive
0.5 0.534 models for inflorescence (0.0050 for k and 24.8 for Cp ) are higher
than those for stem (0.0024 for k and 22.0 for Cp ). From a general
0.4 0.452 point of view, thermophysical properties of inflorescence change
faster with temperature than those of stem, as indicated by their
0.375 higher Ea ; therefore, properties of inflorescence are more sensible
0.344
0.3 to the temperature change. In addition, the negative sign of acti-
0.326
Stem Inflorescence vation energy for thermal diffusivity of stem (Ea = 1.73 kJ/mol)
0.2 indicates that this property decreases as temperature increases. This
0 10 20 30 40 50 60 70 80 could be because, in the studied temperature range, the specific
heat capacity increment was higher than that in thermal conduc-
Temperature (C)
tivity.

6000
Heat transfer modeling
For blanching and hydrocooling processes, the convective heat
B transfer coefficient (h) was 1011 46 W/m2 /K (Bi = 0.013) and
5500
Specific heat capacity (J/kg K)

4,813 468 48 W/m2 /K (Bi = 0.006), respectively. The lower coeffi-

5000 cient for hydrocooling can be attributed to a higher thickness of
4,535
the thermal boundary layer. The finite element geometry model
4500 4,104
4,664 for a broccoli floret is shown in Figure 2. The meshed 3D volume
4000 resulted in 100900 elements and 169957 nodes with an element
3,470 size less than 1 mm. The size of these elements could provide a
3,385 4,046
3,216
3500 closer approximation to the dynamic behavior of heat transfer in-
3,512 side the broccoli, which will be verified through comparison with
3000 experimental results. From results of thermophysical properties as
3,235
2,993 a function of temperature, the convective heat transfer coefficient,
2500 2,795
Stem Inflorescence and the finite element model, different process conditions can be
2000 assayed to get a better retention of nutritional quality and a proper
0 10 20 30 40 50 60 70 80 inactivation of enzymes that affect sensorial quality in broccoli.
Temperature histories simulated through finite element analysis
Temperature (C)
and experimental results are shown in Figure 3. They correspond
to the three different critical zones: thin stems near inflorescence
Figure 1Thermal conductivity (A) and specific heat capacity (B) for
stem and inflorescence of broccoli florets. (For thermal conductivity,
and central axial portions of the stem at 8 and 25 mm from its base.
LSD = 0.043 W/m K and for specific heat capacity, LSD = 467 J/kg The high determination coefficients (R  0.997) indicate that
2

K, 5 replicates, = 0.05). temperature histories obtained through finite element analysis fit
properly to experimental values. The thermal center was located

Vol. 80, Nr. 12, 2015 r Journal of Food Science E2777

 Heat transfer of broccoli florets . . .

at the central axial portion of the stem at 25 mm from its base;
therefore, this is the region that takes longer to reach the operating
temperature. The highest temperature differences between the
thermal center and thin stems were 44.8 C at 33 s and 53.1 C
at 561 s of the process. These temperature differences are due to
the irregular geometry of the broccoli and have a special role on
the more thermosensible zones causing the surface quality may be
affected. Figure 4 displays snapshots of the temperature distribution
at different process times inside the broccoli floret. They confirm
the thermal center is located at the central axial portion of the
stem at 25 mm from its base and that thin stems near inflorescence
reach faster the operating temperature.

Figure 2Finite element geometry model in 3D

of a broccoli floret. Squares represent the
thermocouple locations for each selected zone.
E: Food Engineering &
Materials Science

100 Figure 3Simulated temperature histories of

broccoli florets using finite element analysis (FEA).
90 Thin stems (experimental)
Thin stems near inflorescence and central axial
portions of the stem at 8 and 25 mm from its base
80 8 mm (experimental)
Temperature (C)

70
25 mm (experimental)

60
Thin stems FEA, R2 = 0.997
50
8 mm FEA, R2 = 0.997
40
25 mm FEA, R2 = 0.998
30

20

10

0
0 200 400 600 800 1000 1200
Time (s)

E2778 Journal of Food Science r Vol. 80, Nr. 12, 2015

Heat transfer of broccoli florets . . .
Quality parameters
Firmness, total color difference, and vitamin C content for broc-
coli florets are shown in Figure 5. Firmness ranged from 57.1 to
30.4 N. The highest compression force (57.1 N) corresponded
to the control (fresh); this value is slightly higher than that re-
ported by Fernandez-Leon and others (2012). They determined
firmness in two different cultivars of broccoli and values ranged
from 44.8 to 50.2 N; this difference can be explained due to the
authors measured firmness in the compact head of broccoli and
not in the stem of each floret, and they are structurally different.
There was no difference in firmness of broccoli florets for fresh,
blanching plus freezing with liquid N2 (B + F), and blanching
plus hydrocooling at 5 C (B + HC 5 C). This implies that
Figure 4Snapshot of ANSYS model temperature distribution for broccoli florets. Numbers in legends indicate temperature (C).
regardless of whether blanched broccoli is frozen with liquid N2
or cooled in water at 5 C, the firmness shows no change. The
smallest value (30.4 N) corresponded to blanching plus thawing
after one month of storage (B + T 1Mo); this represents a 53%
decrease compared to the control. Firmness is one of the most
affected quality indexes during blanching, freezing, and storage
of frozen vegetables. Olivera and others (2008) reported a reduc-
tion in firmness (>80%) when studying the impact of storage for
8 months in frozen Brussels sprouts previously blanched by dif-
ferent treatments. This decrease can be explained because changes
associated with blanching include loss of turgor in cells due to
thermal destruction of membrane integrity. Furthermore, the ef-
fects after freezing are remarkable in high moisture and thin cell
wall vegetables (Fuchigami and others 1995). The loss of textural
E: Food Engineering &
Materials Science
Vol. 80, Nr. 12, 2015 r Journal of Food Science E2779
 Heat transfer of broccoli florets . . .
quality can be due to mechanical damage from ice crystals during
freezing.
Total color difference (Figure 5b) ranged from 4.6 to 9.3 with
respect to fresh. The lowest E corresponded to blanching plus
freezing with liquid N2 (B + F), whereas the largest one to blanch-
ing plus thawing after 1 month of storage (B + T 1 Mo). This
result agrees with that reported by Patras and others (2011); they
analyzed the total color difference in broccoli, carrots, and green
beans at different days of storage, and E increased with storage
time. Although, freezing is an effective method of food preser-
vation, some deterioration in the quality of frozen food during
storage is caused. There was no difference in E of broccoli flo-
rets for blanching plus freezing with liquid N2 (B + F), blanching
plus hydrocooling at 5 C (B + HC 5 C), and blanching plus
E: Food Engineering &
Materials Science
Figure 5(A) Firmness, (B) total color difference, and (C) vitamin C content
of broccoli florets. For color, fresh product was considered as the reference
(E = 0). For firmness, E, and vitamin C, LSD was 8.18 N, 1.3, and Therefore, the finite element modeling allows a closer approxima-
10.5 mg/100 g of fresh sample, 3 replicates, = 0.05.
E2780 Journal of Food Science r Vol. 80, Nr. 12, 2015
cooling at room temperature (B + C 25 C). Independently of
whether blanched broccoli is frozen with liquid N2 or hydro-
cooled at 5 C or cooled at room temperature, the E showed
no alteration. In all treatments, the a color parameter increased
respect to control, which indicates an increase in green color of
vegetables. This phenomenon can be attributed to the removal of
air between cells, altering the reflective properties of food surface.
This result coincides with reports by Fernandez-Leon and others
(2012) and Olivera and others (2008).
Vitamin C content (Figure 5c) ranged from 99 to 153 mg/100 g
fresh sample (from 11.1 to 17.1 mg/g d.s.). The highest vitamin C
Content was found in the fresh broccoli and the lowest value cor-
responded to blanching plus cooling at room temperature (B +
C 25 C), which represents a 35% degradation of vitamin C with
respect to control (fresh). This loss of vitamin C can be due to the
slow cooling rate to remove heat inside the product; thereby caus-
ing degradation, since vitamin C is a very thermolabile molecule
that can be easily degraded by chemical and enzymatic oxidation
during processing (Galgano and others 2007; Munyaka and oth-
ers 2010). This result agrees with reports by Howard and others
(1999) who found a 32% decrease of vitamin C in steam blanched
broccoli. Koh and others (2009) reported a wide variation in vi-
tamin C content when analyzing 80 commercial samples of fresh
broccoli; levels were between 57.4 and 131.4 mg/100 g fresh
sample. Alike, Martinez-Hernandez and others (2013) reported
170 mg/100 g fresh sample for vitamin C in fresh broccoli. For
blanching plus freezing with liquid N2 (B + F), vitamin C content
was 135 mg/100 g fresh sample. For blanching plus thawing after
one month of storage (B + T 1 Mo), 113 mg/100 g fresh sample
were obtained, representing a 26% degradation compared to con-
trol. This result is consistent with a study conducted by Albrecht
and others (1991) who reported a loss of vitamin C ranging from
2% to 48% in six different broccoli cultivars stored at 2 C for
21 days. Similarly, Howard and others (1999) observed a loss of
vitamin C of 13% and 48%, in broccoli after 3 weeks of storage at
4 C. Although there was a reduction in vitamin C for blanching
plus thawing after 1 month of storage (B + T 1 Mo) compared
to the control, there was not difference between this treatment
and blanching plus hydrocooling at 5 C (B + HC 5 C), proba-
bly because storage at low temperatures stabilized and maintained
vitamin C, thereby avoiding subsequent degradation.
Activity of peroxidase (POD) in fresh broccoli was 0.313 U/mL,
whereas in the other treatments enzyme activity was not detected.
This value is similar to that reported by Morales-Blancas and others
(2002), who obtained an activity of 0.328 U/mL in fresh broccoli
florets. Peroxidase activity was reduced to 100% during blanching
and no enzyme regeneration was observed after one month of
storage. This result indicates that blanching of broccoli florets at
90 C for 3 min is suitable for peroxidase inactivation.
Conclusions
The high determination coefficients (R2 > 0.997) indicated
that temperature histories estimated through finite element mod-
eling were properly fitted to the experimental results. This im-
plies that thermophysical properties in function of temperature of
stem and inflorescence and the convective heat transfer coefficient
were adequately evaluated; and the three-dimensional model
structure was suitably built for estimating temperatures in function
of time inside broccoli florets during blanching and hydrocooling.
tion to the dynamic behavior of heat transfer inside the vegetable.
The thermal center was located at the central axial portion of
Heat transfer of broccoli florets . . .
the stem at 25 mm from its base. For the treatment of blanching
plus cooling at room temperature, the most affected quality pa-
rameter was vitamin C. Although, for the treatment of blanching
plus thawing after 1 month of storage, the most affected quality
parameters were firmness and total color difference. This indi-
cates that storage time is an important factor to be considered if
these parameters are the most relevant for food quality. During the
applied treatments, peroxidase enzyme was properly inactivated,
which indicates that the blanching process (90 C, 3 min) was
sufficient to maintain the product without regeneration of this
enzyme after one month of storage. Blanching and hydrocooling
of broccoli have a beneficial effect on the physicochemical and
nutrimental properties leading to a better retention of product
quality. With results from this study, temperature profiles and his-
tories can be predicted at process conditions differing from those
used in this work (blanching at 90 C, hydrocooling at 5 C, both
for 3 min). A heat penetration study can be performed to op-
timize the blanching and hydrocooling processes to achieve the
inactivation of peroxidase and simultaneously a higher retention
of vitamin C.
Acknowledgments
This research was supported by Programa de Fomento y
Apoyo a Proyectos de Investigacion (PROFAPI-2012/034), Univ.
Autonoma de Sinaloa.
References
Albrecht JA, Schafer HW, Zottola EA. 1991. Relationship of total sulfur to initial and retained
ascorbic acid in selected cruciferous and noncruciferous vegetables. J Food Sci 55:1813.
AOAC. 2012. Official Methods of Analysis, Association of Official Analytical Chemists,
Washington, DC. Official Method 934.06.
Ayon-Reyna LE, Tamayo-Limon R, Cardenas-Torres F, Lopez-Lopez ME, Lopez-Angulo G, Patras A, Tiwari BK, Bruton NP. 2011. Influence of blanching and low temperature preservation
Lopez-Moreno HS, Lopez-Cervantes J, Lopez-Valenzuela JA, Vega-Garca MO. 2015. Ef-
fectiveness of hydrothermal-calcium chloride treatment and chitosan on quality retention and
microbial growth during storage of fresh-cut papaya. J Food Sci 80(3):594601.
Barrett DM, Garca EL, Russell GF, Ramirez E, Shirazi A. 2000. Blanch time and cultivar effects
on quality of frozen and stored corn and broccoli. J Food Sci 65:53440.
Caro-Corrales J, Cronin K, Gao X, Cregan V. 2010. Heat transfer analysis of cheese cooling
incorporating uncertainty in temperature measurement locations: Application to the industrial
process. J Food Eng 99:159165.
Caro-Corrales J, Cronin K, Abodayeh K, Gutierrez-Lopez G, Ordorica-Falomir C. 2002. Anal-
ysis of random variability in biscuit cooling. J Food Eng 54:14756.
Cronin K, Caro-Corrales J, Gao J. 2010. Heat transfer analysis of cheese cooling incorporating
uncertainty in temperature measurement locations: model development and validation. J Food
Eng. 99:175-183.
Cronin K, Caro-Corrales J, Tobin J, Kerry J. 2008. Impingement cooking of meat products:
effect of variability on final temperature. Food Sci Techol Int 14:24150.
Domnguez-Perles R, Martnez-Ballesta MC, Carvajal M, Garca-Viguera C, Moreno DA.
2010. Broccoli-derived by-productsa promising source of bioactive ingredients. J Food Sci
75:C383C392.
Dosz EB, Jeffery EH. 2013. Modifying the processing and handling of frozen broccoli for
increased sulforaphane formation. J Food Sci 78:H145963.
Durust N, Sumergen D, Durust Y. 1997. Ascorbic acid and element contents of foods of Trabzon
(Turkey). J Agr Food Chem 45:20857.
Espinoza-Guevara R, Caro-Corrales J, Ordorica-Falomir C, Zazueta-Morales J, Vega-Garcia
M. 2010. Thermophysical properties of pulp and rind of papaya cv. Maradol. Int J Food Prop
13:6574.
Fernandez-Leon MF, Fernandez-Leon AM, Lozano M, Ayuso MC, Gonzalez-Gomez D. 2012.
Identification, quantification and comparison of the principal bioactive compounds and ex-
ternal quality parameters of two broccoli cultivars. J Funct Foods 4:46573.
Fuchigami M, Hyakumoto N, Miyazaki K. 1995. Texture and pectic composition differences
in raw, cooked and frozen-thawed Chinese cabbages due to leaf position. J Food Sci 60:
1536.
Galgano F, Favati F, Caruso M, Pietrafesa A, Natella S. 2007. The influence of processing
and preservation on the retention of health-promoting compounds in broccoli. J Food Sci
72:S1305.
Howard LA, Wong AD, Perry AK, Klein BP. 1999. -Carotene and ascorbic acid retention in
fresh and processed vegetables. J Food Sci 64:92936.
Koh E, Wimalasiri KMS, Chassy AW, Mitchell AE. 2009. Content of ascorbic acid, quercetin,
E: Food Engineering &
kaempferol and total phenolics in commercial broccoli. J Food Compos Anal 22:637
Materials Science
43.
Kursat CE, Rennie AEW, Akinci I. 2011. Deformation behaviour simulation of an apple under
drop case by finite element method. J Food Eng 104:2938.
Mahapatra A K, Lan Y, Harris DL. 2011. Influence of moisture content and temperature on
thermal conductivity and thermal diffusivity of rice flours. Int J Food Prop 14:67583.
Marangoni AG, Jackman RL, Stanley DW. 1995. Chilling associated softening of tomato fruits
is related to increased PME activity. J Food Sci 60:127781.
Martens M, Scheerlinck N, De Belie N, De Baerdemaeker J. 2001. Numerical model for the
combined simulation of heat transfer and enzyme inactivation kinetics in cylindrical vegetables.
J Food Eng. 47:18593.
Martinez-Hernandez GB, Artes-Hernandez F, Gomez PA, Artes F. 2013. Induced channges
in bioactive compounds of kailan-hybrid broccoli after innovative processing and storage.
J Funct Foods 5:13343.
Morales-Blancas EF, Chandia VE, Cisneros-Zevallos L. 2002. Thermal inactivation kinetics of
peroxidase and lipoxigenase from broccoli, green asparagus and carrots. J Food Sci 67:146
54.
Munyaka AW, Makule EE, Oey I, Loey AV, Hendrickx M. 2010. Thermal stability of l-ascorbic
acid and ascorbic acid oxidase in broccoli (Brassica oleracea var. italica). J Food Sci 75:C33640.
Olivera DF, Vina SZ, Marani CM, Ferreyra RM, Mugridge A, Chaves AR, Mascheroni RH.
2008. Effect of blanching on the quality of Brussels sprouts (Brassica oleracea L. gemmifera DC)
after frozen storage. J Food Eng 84:14855.
strategies on antioxidant activity and phytochemical content of carrots, green beans and
broccoli. Food Sci Technol-LEB 44:299306.
Rodrguez-Hernandez MC, Moreno DA, Carvajal M, Garca-Viguera C, Martnez-Ballesta
MC. 2012. Natural antioxidants in purple sprouting broccoli under mediterranean climate.
J Food Sci 77:C105863.
Gormley R, Tansey F. 2011. Sous vide and cook-chill processing. In: Sun D-W, editor. Hand-
book of Food Safety Engineering. Oxford, UK: Wiley-Blackwell. p 46896.
Thongsook T, Barrett DM. 2005. Purification and partial characterization of broccoli (Brassica
oleracea Val. Italica) Peroxidases. J Agr Food Chem 53:3206321.
Tocci AM, Mascheroni RH. 2008. Some thermal properties of fresh and osmotically dehydrated
Kiwifruit above and below the initial freezing temperature. J Food Eng 88:2027.
Zhang Z, Pang X, Xuewu D, Ji Z, Jiang Y. 2005. Role of peroxidase in anthocyanin degradation
in litchi fruit pericarp. Food Chem 90(1):4752.
Vol. 80, Nr. 12, 2015 r Journal of Food Science E2781
Copyright of Journal of Food Science is the property of Wiley-Blackwell and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.