Professional Documents
Culture Documents
Encurtidos
Encurtidos
Encurtidos
Department of Food Science, University of Arkansas, 2650 North Young Ave., Fayetteville, AR 72703, United States
Facultad de Ciencias Exactas, Qumicas y Naturales, Universidad Nacional de Misiones, Flix de Azara 1552, 3300 Posadas, Misiones, Argentina
a r t i c l e
i n f o
Article history:
Received 21 March 2011
Received in revised form 7 July 2011
Accepted 15 July 2011
Available online 23 July 2011
Keywords:
Fermented cucumber
Pasteurization
Finite-difference method
High-acid food
Peroxidase
a b s t r a c t
In-container pasteurization of pickled cucumbers deactivates enzymes and spoilage microorganisms
with minimal organoleptic changes when packed in small jars. As jar volumes increase, pickle quality
is affected as a result of longer pasteurization times setting the practical volume limit to 3.8 L. Alternatively, this paper proposes out-of-container pasteurization followed by packaging in sanitized containers.
This study, including only the pasteurization step, was conducted with a discontinuous pasteurization
device using variable cucumber diameters, brine temperatures, and ows. Heat penetration data was
t to a mathematical model with the enzyme peroxidase as a parameter of pasteurization effectiveness.
Results showed that heat transfer was dominated by internal resistance, and according to the model,
enzyme deactivation depended on the cucumber diameter, and brine temperature, but not on the brine
ow. A 5-log reduction in peroxidase activity in the center of the cucumbers required 1.4 times longer
than the equivalent treatment in the whole pickle.
2011 Elsevier Ltd. All rights reserved.
1. Introduction
In-container pasteurization is the nal step in the production of
shelf-stable pickled cucumbers to extend their shelf-life. Heat pasteurization not only is effective in reducing the number of spoilage
microorganisms, but also in deactivating enzymes that participate
in quality deterioration, mainly due to tissue softening. Pasteurization, however, alters the organoleptic quality of pickles, so the
intensity of the treatment (time and temperature) must be minimized. Since pickled cucumbers have a pH lower that 4.6, they
are considered an acid food with no risk of Clostridium botulinum
growth. Accordingly, the pasteurization parameters currently used
by the pickling industry are used only for slowing the deterioration
caused by microorganisms and enzymes. The most common heat
treatment for pasteurization of pickled cucumbers in industry is
74 C (165F) for 15 min in the center of a cucumber located in
the center of the jar (Breidt et al., 2005).
As with any other in-container heat pasteurization or sterilization, the total heating duration to achieve the desired pasteurization units, or lethality, in the center of the jar depends on various
factors including container size. Since the external heat transfer
area of jars does not increase proportionally with the volume, the
length of heat treatments does not increase proportionally with
volumes. For instance, as the volume of the jar triples, the external
Corresponding author. Tel.: +1 479 575 4923; fax: +1 479 575 6936.
E-mail address: rmorawic@mail.uark.edu (R.O. Morawicki).
0260-8774/$ - see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jfoodeng.2011.07.020
290
R.O. Morawicki, M.E. Schmalko / Journal of Food Engineering 107 (2011) 289295
Nomenclature
Ac (k)
As (k)
Cp
D
Dc
Ds
DRc
DRs
hc
hs
kw
kT
MPE
Pr
Re
RMSE
Tc (k,t)
Tm
Ts(k,t)
Vc(k)
Vs(k)
z
Dt
Immediately after mixing, the absorbance at 370 nm was determined by a spectrophotometer (Beckman DU-520, Fullerton, CA)
and used as an indicator of the peroxidase activity (Sigma, 1995).
2.1.4. Determination of D- and z-value
The initial peroxidase activity of the cucumber extract was
determined and used as the maximum peroxidase activity. Then
aliquots of the extract (1 mL) were transferred to 2-mL Eppendorf
tubes and incubated in water baths set at 71.1, 73.9, 76.7, and
79.4 C (160, 165, 170, and 175F, respectively). Periodic samples
were taken and assayed for peroxidase activity as follows: every
15 min for extracts held at 71.1 C, every 10 min for extracts held
at 73.9 and 76.7 C, and every 5 min for extracts held at 79.4 C.
Incubation was continued until the peroxidase activity was reduced 2-logarithmic cycles or more compared to the starting activity. For each temperature, assays were conducted in triplicate on
different days and with different lots of cucumbers. The peroxidase
activity was then standardized by dividing the activity at every
sampling point by the initial activity. Standardized triplicates were
averaged and plotted in an xy scatter plot with a logarithmic scale
for enzyme activity and a regular scale for time. The data points
were then t with a linear regression line using the least-squares
criterion. D-value was calculated as the time required reducing
the peroxidase activity 1-logarithmic cycle.
The z-value for peroxidase deactivation was obtained by plotting D-values for each temperature in an xy scatter plot with a
logarithmic scale for the D-value and a regular scale for the temperature. A regression line was established as described for the
D-value determination) and the z-value was calculated as the increase in temperature required reducing the D-value by one logarithmic cycle.
2.2. Heat penetration during pasteurization
2.2.1. Materials
Commercial pickled cucumbers, Vlasic Kosher Dills packed in
1.3 and 2.36-L jars (Pinnacle Foods Corporation, Cherry Hill, NJ),
Gedney Dill Whole Pickles packed in 3.78-L jars (M.A. Gedney
Co., Chaska, MN), and Mt. Olive Kosher Dills packed 3.78-L jars
(Mt. Olive Pickle Co. Inc., Mt. Olive, NC), were purchased from a local grocer. From these jars, straight, undamaged pickles with
lengths between 0.08 and 0.125 m were selected, from which three
diameters were chosen (2.5 102 m, 3.3 102 m, and
4.0 102 m) and used for the pasteurization experiments.
2.2.2. Pasteurization device
The apparatus consisted of: (i) a perforated tray of 0.25 0.3 m
that held the cucumbers; (ii) a brine-collecting tray located below
R.O. Morawicki, M.E. Schmalko / Journal of Food Engineering 107 (2011) 289295
Brine
dispenser
Cucumber
291
4
2.3. Mathematical model
Flow
meter
3
2
Booster
pump
Heater/Pump
Brine
reservoir
the perforated tray; (iii) a 20-liter brine tank; (iv) a brine dispenser
on top of the cucumbers; (v) a heater/pump; (vi) a booster pump;
(vii) a volumetric variable area ow meter; and (viii) stainless steel
manually operated ball valves connected with silicone tubing
(Fig. 1). The brine dispenser was a 0.13-m-diameter perforated disk
with 20 holes (each with a diameter of 0.03 m) equally distributed
throughout the disk that dispensed the brine as a shower. The heater/pump combination was a VWR immersion circulator, model
1122S (VWR, West Chester, PA), that was used to heat the brine
to the desired temperature. The variable area ow meter was an
F-440 (Blue-White Industries, Ltd., Huntington Beach, CA) with a
range between 1.5 and 18 L/min (LPM). The booster pump was a
Wayne Reliant One Model 75982 (Wayne Water Systems, Harrison
OH).
2.2.3. Experimental design
The experimental design was a full factorial with three factors
(cucumber diameter, brine temperature, and brine ow) at three
levels. The levels were 2.5 102 m, 3.3 102 m, and
4.0 102 m for cucumber diameter, 76.7, 79.4, and 82.2 C (170,
175 and 180F) for brine temperature, and 6.4, 12.8, and
19.2 105 m3/s (1, 2, and 3 gal/min) for brine ow.
Each experimental run used three cucumbers that were placed
on the sample holding tray (as shown in Fig. 1) with two of them
connected via thermocouples to the data acquisition system.
Dt
hc Ac0 Tm Tc1; t
q Cp Vc1
kT Ac1
Tc1; t Tc2; t
1
DRc
Tc1; t 1 Tc1; t
Nodes k = 2 to n 1
Dt
TcK; t 1 TcK; t
q Cp VcK
"
#
Tck 1; t Tck; t
c kT Ack1
DRc
Tck; t Tck 1; t
kT AcK
DRc
Dt
TcN; t 1 TcN; t
q Cp VcN
kT AcN 1
TcN 1; t TcN; t
DRc
Dt
hs As0 Tm Ts1; t
q Cp Vs1
kT As1
Ts1; t Ts2; t
4
DRs
Ts1; t 1 Ts1; t
Nodes k = 2 to n 1
2.2.4. Temperature data acquisition
Temperature during heat treatments was acquired by two thermocouples located inside the cucumbers. One was placed as close
to the center as possible and the second one at the periphery just
under the skin. The thermocouples were Type T, model number
39138-T manufactured by Atkins (Cooper-Atkins Corporation,
Gainesville, FL), with bare tips, which were electrically insulated
Dt
Tsk; t 1 Tsk; t
q Cp VsK
"
#
Tsk 1; t Tsk; t
c kT Ask1
DRs
Tsk; t Tsk 1; t
kT Ask
DRs
Node k = n = 15 (at the center)
292
R.O. Morawicki, M.E. Schmalko / Journal of Food Engineering 107 (2011) 289295
Fig. 2. Cucumber adapted to the shape of a cylinder with hemispherical ends divided in zones and nodes.
(a)
(b)
Fig. 3. Decrease in the activity of peroxidase extracted from fresh cucumbers and
incubated at four different temperatures for various durations.
Dt
Tsn; t 1 Tsn; t
q Cp VsN
kT Asn 1
Tsn 1; t Tsn; t
DRs
MPE
n
100 X
ABST exp T pre
n 1
T exp
s
n
1X
T exp T pre 2
RMSE
n 1
The degree of heat treatment, indicated by the activity reduction of the enzyme peroxidase, was estimated by adding the heat
treatment in each node during a Dt interval (Eq. (9)):
log
n
C0 X
Dt
Vnode
D v alue Vtotal
C
1
(c)
Fig. 4. Experimental (exp) and calculated (calc) temperatures in the center of the
cucumber for (a) various temperatures, (b) various cucumber diameters, and (c)
various brine ow rates ( 105 m3/s) while maintaining the other two factors
constant.
293
R.O. Morawicki, M.E. Schmalko / Journal of Food Engineering 107 (2011) 289295
kw
hc 0:683
Re0:466 Pr1=3
Dc
10
kw
hs 0:6
Re0:5 Pr1=3
Ds
11
Peroxidase was used as a reference for pasteurization effectiveness because this enzyme is responsible for tissue softening in
cucumbers and other vegetables (Matsui et al., 2008; Polata
et al., 2009; Seyhan et al., 2002; Nicholas and Pug, 1962). The
activity of residual peroxidase for the temperatures studied, which
is expressed as a percentage of the activity at time zero, follows an
approximately linear pattern when plotted against the temperature (with coefcients of determination (R2) between 0.9782 and
0.9984) (Fig. 3). From the regression lines, D-values were calculated for each temperature and presented in Table 1. The z-value,
based on the D-values, was 8.0 0.36 C (mean standard error)
with an MPE of 4.1%. A z-value of 10 C has been reported previously (Nicholas and Pug, 1962). The discrepancy is likely due to
differences in cucumber varieties and methodology. An equation
was also developed to establish the dependence of D-value and
temperature (Eq. (12)), and this equation was used to estimate
the D-value used in Eq. (9).
Tck;t 71:1
D-value a log 3:653
8
Standard error
71.1
73.9
76.7
79.4
4323 738
2049 200
765 114
390 24
301
82
46
8
12
A similar expression of the relationship of D-value and temperature was used by Matsui et al. (2008) when working with deactivation of peroxidase in coconuts by microwave heating. Other
researchers use an Arrhenius model to describe the relationship
of enzyme activity with temperature, however, the model used
in Eq. (12) has been chosen due to its simplicity.
3.2. Application of the model
There was good agreement between the experimental data and
estimated values in the proposed model with a mean percent error
(MPE) of 4.89% and a root mean square error (RMSE) of 1.29 C (Ta-
Table 2
Experimental design, prediction errors of the model (MPR and RMSE), and treatment time to achieve a 5-logarithmic reduction in the activity of peroxidase (log C0/C = 5) for an
overall treatment of the whole cucumber (Time O) and for treatment at the center of the cucumber (Time C).
Experiment#
Temperature (C)
MPE
RMSE (C)
Time C (s)
Time O (s)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
Mean value
76.7
76.7
76.7
76.7
76.7
76.7
76.7
76.7
76.7
79.4
79.4
79.4
79.4
79.4
79.4
79.4
79.4
79.4
82.2
82.2
82.2
82.2
82.2
82.2
82.2
82.2
82.2
6.31
6.31
6.31
12.62
12.62
12.62
18.92
18.92
18.92
6.31
6.31
6.31
12.62
12.62
12.62
18.92
18.92
18.92
6.31
6.31
6.31
12.62
12.62
12.62
18.92
18.92
18.92
2.5
3.3
4.0
2.5
3.3
4.0
2.5
3.3
4.0
2.5
3.3
4.0
2.5
3.3
4.0
2.5
3.3
4.0
2.5
3.3
4.0
2.5
3.3
4.0
2.5
3.3
4.0
6.47
0.01
20.19
2.10
3.12
2.79
9.1
1.39
2.12
0.05
3.39
4.50
4.13
6.12
9.04
2.84
16.22
0.29
1.11
1.11
2.80
2.14
4.11
7.33
11.72
5.49
2.33
4.89
2.03
0.95
1.85
0.99
1.08
0.98
1.91
0.82
0.94
0.39
1.42
1.04
1.41
1.55
1.96
0.84
2.63
1.77
1.29
1.17
1.03
1.11
1.33
1.62
1.73
0.56
0.39
1.29
600
935
1134
632
951
1222
646
926
1170
600
885
1176
593
868
1186
596
860
1140
566
838
1125
564
832
1124
590
827
1120
878
420
570
626
429
588
694
437
568
649
513
538
620
392
626
502
394
497
607
564
362
470
360
465
563
390
461
560
514
R.O. Morawicki, M.E. Schmalko / Journal of Food Engineering 107 (2011) 289295
(a) 90
T media
80
Node 1
70
Temperature (oC)
ble 2). Fig. 4 shows experimental and calculated data for nine
experimental runs during the rst 600 s of heating. Each gure
shows the rise of temperature in the center of the cucumber with
time for one variable factor (brine temperature, ow, or diameter)
while maintaining the other two factors constant. It is important to
mention that initially the cucumbers were at room temperature
which resulted in a variation in the initial cucumber temperatures
of each experimental run. Also, the location of thermocouples was
different for each experiment and cucumber because of the difculty of placing them all at the same distance from the surface.
Fig. 4(a) shows how the temperature was inuenced by three
different brine temperatures (experimental and predicted data)
with a xed brine ow rate (6.31 105 m3/s) and a cucumber
diameter (3.3 102 m). The overlap between the different curves,
mainly in the rst stages of heat treatment, is likely due to the
small difference in the brine temperatures for the different experiments (2.8 C) as well as the different initial cucumber temperatures and the variable position of the thermocouples. Obviously,
higher temperatures would speed up heating, but they also would
increase quality deterioration on the surface. Fig. 4(b) shows the
inuence of diameter on the temperature evolution. As expected,
diameter has an important inuence on the heat penetration
curves. Fig. 4(c) shows the heating curves for three different brine
ows (1, 2, and 3 gal/min). Similarly to Fig. 4(a), no signicant difference in the heating rate was observed over the range of ows
tested.
60
Node 5
50
Node 10
40
30
Node 15
20
10
0
(b) 90
100
200
70
400
500
600
400
500
600
400
500
600
Node 1
60
Node 5
50
Node 10
40
30
Node 15
20
300
Time (s)
T media
80
Temperature (oC)
294
10
100
200
300
Time (s)
(c) 90
T media
80
Temperature (oC)
The time that it takes to produce a 5-log reduction of the peroxidase activity in the center (node 15) was estimated using the model for the different diameters, ows, and temperatures.
Subsequently, an analysis of variance was conducted, with time
as a dependent variable and the other factors (brine temperature,
diameter, and ow) as independent variables. According to the
analysis, the duration of heating depended on brine temperature
(p < 0.0395) and cucumber diameter (p < 0.0004) but not on the
brine ow (p < 0.9197). Fig. 5 shows the mean times and the 95%
condence intervals for three cucumber diameters and brine temperatures. As expected, the length of pasteurization time increased
with increasing cucumber diameters and decreasing temperatures.
The fact that brine ow had no inuence on heating was due to
a predominant internal heat of the cucumber, which is explained
by the Biot number. The calculated Biot for the three experimental
ows ranged from 5.6 to 11.9, and a Biot above 1 indicates that the
heat ux was dominated by the internal heat resistance (Kreith
and Bohn, 1997). Working with a different system, Fasina and
Fleming (2001) used convective coefcients between 500 and
6000 W/m2K and found that they do not have an inuence on
the internal temperature of the cucumbers. Therefore, the more
modest external heat transfer coefcients in this study, which varied from 212 to 511 W/m2K, are in agreement with the values reported by Fasina and Fleming (2001).
70
Node 1
60
50
Node 5
40
Node 10
30
20
Node 15
10
0
0
100
200
300
Time (s)
dependent on the diameter. For instance, to reach 50 C in the center of the cucumber with a 2.5 102 m diameter required 235 s
while the one with a 3.3 102 m diameter required 386 s and
the one with a 4 102 m -diameter required 600 s.
When developing the model, we assumed there was no difference between neighboring nodes of the cylinder and the semispherical ends (lack of a longitudinal temperature prole). To
check this assumption, we compared the temperatures between
the neighboring nodes during heating and found differences greater than 2 C only at the beginning of heating (<120 s). After 120 s,
the differences were less than 1 C, so the assumption was considered acceptable.
R.O. Morawicki, M.E. Schmalko / Journal of Food Engineering 107 (2011) 289295
Table 3
Values of log (C0/C) at selected nodes for an overall 5-logarithmic cycle reduction in
peroxidase activity in the whole cucumber from experiments 12, 13, and 14.
Experiment #
12
13
14
Node
1
10
15 (center)
10.66
12.38
13.41
3.01
2.32
1.72
1.01
0.54
0.28
0.65
0.30
0.14
295