LWT - Food Science and Technology 79 (2017) 119e125

Contents lists available at ScienceDirect

LWT - Food Science and Technology

journal homepage: www.elsevier.com/locate/lwt

Phase transitions during high pressure treatment of frozen carrot juice

and inuence on Escherichia coli inactivation
Songming Zhu a, b, Chunfang Wang a, b, Hosahalli S. Ramaswamy c, Yong Yu a, b, *
a
College of Biosystems Engineering and Food Science, Zhejiang University, 866 Yuhangtang Road, Hangzhou 310058, China
b
Key Laboratory of Equipment and Informatization in Environment Controlled Agriculture, Ministry of Agriculture, 866 Yuhangtang Road, Hangzhou
310058, China
c
Department of Food Science, McGill University, St-Anne-de-Bellevue, QC H9X 3V9, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Inuence of high pressure (HP) treatment (200e400 MPa; 0e10 min) on phase transition behavior of
Received 2 August 2016 frozen carrot juice and the resulting inuence on the inactivation kinetics of Escherichia coli ATCC 25922
Received in revised form were evaluated. Experiments were carried out in a specially designed container to prevent heat exchange
1 December 2016
from the environment, except for the compression heating and decompression cooling. Solid to solid and
Accepted 8 January 2017
Available online 9 January 2017
solid to liquid transitions were recognized during HP treatment. Transition to Ice-III was observed from
the temperature-pressure proles when the application pressure was >350 MPa. Inactivation of E. coli in
frozen carrot juice followed the rst order kinetics with D values between 2.62 and 2.12 min at 300
Keywords:
High pressure
e400 MPa pressure levels, much shorter than those observed in unfrozen carrot juice. The combination
Frozen carrot juice of frozen state, phase transition status and pressure level likely contributed to the better inactivation of
Phase transition E. coli in frozen carrot juice.
2017 Elsevier Ltd. All rights reserved.

1. Introduction pressure and elevated temperatures have synergist acceleration of

microbial destruction and enzyme inactivation kinetics
High pressure (HP) processing is a promising alternative to (Ramaswamy, Riahi, & Idziak, 2003). Temperature effect on pres-
thermal processing for extending the shelf-life and improving the sure destruction kinetics of microbial spores has also been shown
safety of food products (Bari, Ukuku, Mori, Kawamoto, & to be signicant when the temperatures are elevated and
Yamamoto, 2007; Picouet, Sa
rraga, Cofa

n, Belletti, & Dolors, 2015; approached the lethal levels (Shao, Ramaswamy, & Zhu, 2007). HP
Ramaswamy, 2011; Vervoort et al., 2012). Studies have conrmed effects on destruction kinetics have also been recognized to depend
that HP processing can effectively inactivate vegetative cells of on product properties such as pH, soluble solid concentrations,
pathogenic microorganisms (Ramaswamy, Zaman, & Smith, 2008) composition (fat, protein and carbohydrate content) (Ramaswamy,
without changing the sensory and nutritional properties of the food Jin, & Zhu, 2009). In some cases, protection effect of pressure on
(Dede, Alpas, & Bayindirli, 2007; Oey, Lille, Loey, & Hendrickx, thermal destruction kinetics at elevated temperature levels in the
2008; Picouet et al., 2015). HP processing does not affect the low lethal range has been reported (Shao et al., 2007). This has been
molecular and covalently bound food compounds, such as vitamins, reported to be due to the increase in temperature and pressure
avoring agents, etc., and hence HP processing can provide fresh- causing opposite effects on volume expansion; with temperature
like carrot juice with better sensory properties for long refriger- contributing to volume increase while pressure reversing that ef-
ated storage (Picouet et al., 2015). fect. For similar reasons, some studies have shown that at sub-
HP effects on microorganisms depend on several factors, but the lethal levels, the pressure destruction effects are elevated at
effects of pressure level and time have been studied most often. A lower temperatures (Su et al., 2014).
number of studies have demonstrated that the combination of There are far fewer studies in HP processing research that use
the combination of pressure and refrigerated conditions. Research
related to the use of subzero (frozen) temperatures on inactivation
* Corresponding author. College of Biosystems Engineering and Food Science, kinetics are even more scarce (Luscher, Balasa, Frohling, Ananta, &
Zhejiang University, 866 Yuhangtang Road, Hangzhou 310058, China. Knorr, 2004; Picart, Dumay, Guiraud, & Cheftel, 2005). Su et al.
E-mail address: [email protected] (Y. Yu).

http://dx.doi.org/10.1016/j.lwt.2017.01.022
0023-6438/ 2017 Elsevier Ltd. All rights reserved.
120 S. Zhu et al. / LWT - Food Science and Technology 79 (2017) 119e125

(2014) reported a combination or interaction effect of pressure and forming units (CFU)/mL.
subzero temperature for the inactivation of microorganisms in
food. Phase transitions of Ice-I/Ice-III by pressurizing frozen sys- 2.2. Sample preparation
tems above 200 MPa was shown to be responsible for bacterial
destruction (Ferna
ndez et al., 2007). However, there is little infor- Fresh carrots were purchased from Wal-Mart store near Zhe-
mation on the evaluation of temperature proles and phase tran- jiang University (Hangzhou, China). The carrots were peeled and
sition in frozen samples under HP processing conditions, especially squeezed with a juice extractor (JYZ-B550, Joyoung Co., Ltd.,
using a real food matrix. Hangzhou, China), and the juice was then centrifuged at 3200g for
Fruit and vegetables are important components of a healthy 10 min to remove the suspended solids. The supernatant was
diet. Their daily consumption in adequate quantities could help ltered through a 0.23-mm-pore-diameter lter. The clear carrot
prevent several major diseases (Picouet et al., 2016), such as heart juice had a pH of 6.5 0.3 (FE 20, Mettler-Toledo, Shanghai, China).
problem, cancer, diabetes and obesity, as well as the prevention and The juice was subjected to a thermal treatment, in a thermostati-
alleviation of several micronutrient deciencies. Among common cally controlled water bath (DKS-224, Zhongxin Medical Instru-
fruits and vegetables, carrots are high in bers, carotenoids, vita- ment Co., Ltd., Jiaxing, China) at 65
C for 30 min. Microbial
mins C and E, and phenolics such as p-coumaric, chlorogenic, and enumeration on BHI agar in heat treated samples returned negative
caffeic acids (Alasalvar, Grigor, Zhang, Quantick, & Shahidi, 2001). counts.
However, the low-acid condition in carrot juice is conducive to the The prepared carrot juice was transferred aseptically into sterile
growth of pathogenic microorganisms (Patterson, McKay, Connolly, bags (110 mm  185 mm, BagLight PolySilk, Interscience, Paris,
& Linton, 2012). So storage of raw, unprocessed carrot juice may France) (about 150 mL for each bag) and sealed using a heat sealer
also lead to microbiological safety problems and shortened shelf- (FS-300, Yongkang Teli Packing Machinery Co., Ltd., China). The
life. Therefore, extending the shelf-life using mild processing bags were frozen stored in a freezer (BC/BD-103HA, Haier, China)
technologies that minimally affect the sensory and nutritional at 20
C for subsequent handling.
properties will be of interest for the food industry. Van Opstal, Carrot juice was fully thawed at 4
C for ~4 h prior to inoculation
Vanmuysen, Wuytack, Masschalck, and Michiels (2005) reported of E. coli culture. After thawing, carrot juice was aseptically trans-
that pressure inactivation of Escherichia coli MG1655 was signi- ferred to a sterile beaker. The previously prepared E. coli pellet was
cantly lower in carrot juice than in buffer (150e600 MPa, 5e45
C). re-suspended in 20 mL carrot juice and then added to the rest of
Pilavtepe-elik, Buzrul, Alpas, and Bozog lu (2009) reported that carrot juice in a beaker. After stirring at 500 rpm (RCT BS25, IKA,
carrot juice even had a protective effect on E. coli. Staufen, Germany) for 1 min, the inoculated carrot juice was
Therefore, the objective of this study was to evaluate the phase aseptically transferred to a sterile 2.0 mL cryogenic vial (430 659,
transition behavior of frozen carrot juice during HP treatment and Corning Inc., USA), up to the brim and closed with sterile screw cap
to extend the work of Su et al. (2014) on E. coli inactivation kinetics leaving no headspace to avoid possible cracking during HP
in frozen carrot juice. A specially designed container was used to treatment.
isolate the sample and to achieve the ice phase transition under HP
at room temperatures. This helped to overcome a serious practical 2.3. Insulated container
problem of maintaining the entire pressure chamber at subzero
temperatures. Maintaining subzero temperatures in commercial HP The cryogenic vials were individually vacuum packed in two
vessels is not only impractical but also be very expensive and en- layers of polyethylene bags and divided into two batches. One batch
ergy intensive. Inactivation kinetic of E. coli was used for comparing of the cryogenic vials were then loaded into a specially designed
the HP treatment effectiveness in frozen vs unfrozen carrot juice. plastic container consisting of a compressed sponge as previous
study used (Su et al., 2014) and slightly modied, as shown in Fig. 1.
2. Materials and methods A K-type thermocouple (OMEGA Engineering, Stamford, CT, USA)
was installed into the container for temperature recording using a
2.1. Escherichia coli strain and culture preparation data logger (34970A, Agilent Technologies GMBH, Germany). Four
cryogenic vials were placed in one container with one of them used
The culture of E. coli ATCC 25922 (CGMCC 1.2385) was obtained for recording sample temperature during HP treatment. The plastic
from China General Microbiological Culture Collection Center container was lled water to completely soak the sponge and
(CGMCC, Beijing, China). A fresh culture was prepared every 2 frozen at 20
C for 24 h, and were secured by vacuum-packing in a
weeks to ensure their viability. To prepare the inoculation stock, exible thermo-stable PA/PE pouches (30 cm  42 cm) prior to HP
several loops of isolated colonies of stock culture were transferred treatment (details shown in Fig. 1). Samples of E. coli inoculated
to 50 mL sterile nutrient broth (Sinopharm Chemical Reagent Co., carrot juice frozen stored at 20
C, prepared in a similar manner
Ltd. Shanghai, China) in 100 mL Erlenmeyer asks and incubated at but without HP treatment, were used as control. The second batch
37
C for 24 h with agitation (150 rpm). Several loops of the incu- was not frozen but stored at 4
C and treated likewise for providing
bated broth were then transferred to another 50 mL sterile nutrient HP inactivation data under unfrozen conditions.
broth and incubated at 37
C for 24 h incubation. Following this,
30 mL incubated broth was aseptically transferred into a 50 mL 2.4. High pressure equipment
sterilized centrifuge tube and centrifuged at 3200g for 5 min at
20
C (5810R, Eppendorf AG, Germany). The cell pellet obtained was HP treatments were carried out in a laboratory-scale HP
re-suspended in 20 mL nutrient broth and enumerated using the equipment (UHPF-750, Baotou Kefa High Pressure Technology Co.,
pouring plate method on a brain heart infusion agar plate (BHIA) Ltd., China) with a maximum chamber capacity of 5 L. The high
(Hangzhou Tianhe Microorganism Reagent Co., Ltd., Hangzhou, pressure unit was connected to a data logger for temperature and
China), and incubated at 37
C for 48 h and counted. The E coli pressure (current signal) recoding during HP treatment. Water was
selective media was not used in this study since a pure culture of used as the pressure-transmitting medium. The pressure vessel
E. coli was used and in order to obtain counts of both surviving and was maintained at room temperature (~20
C) and with some
injured cells (Ramaswamy et al., 2003). The initial population of added ice if the temperature was above 20
C before pressure
E. coli in the inoculated culture stock was approximately 108 colony treatment. The near room temperature setup was used so that the
 S. Zhu et al. / LWT - Food Science and Technology 79 (2017) 119e125 121

Fig. 1. Schematic cross-section and details of the insulated container for holding test samples (the red color material is the sponge covering the vials). (For interpretation of the
references to colour in this gure legend, the reader is referred to the web version of this article.)

process can be replicated in other commercial HP equipment using 2.7. Kinetic analysis
the special attachment. The pressure come-up rate was about
200 MPa/min and the depressurization time was less than 10 s. The pressure inactivation kinetics of E. coli during pressure
Data were recorded at one second interval. holding time was analyzed based on a rst order reaction rate (Shao
et al., 2007) indicating a logarithmic order of death, and expressed
as:
2.5. HP treatment  
Nt t
log 10  (1)
The frozen samples carrot juice were subjected to HP treatment N0 D
at selected pressure levels (200, 250, 300, 350 and 400 MPa) for
selected treatment times (0, 1, 3, 5, 7.5 and 10 min holding time). where Nt is the number of E. coli survivors (CFU mL1) after HP hold
Unfrozen samples were only treated at 300, 350, and 400 MPa for time treatment time of t (min), and N0 is the initial number of E. coli
holding times 0, 1, 3, 5, 7.5 and 10 min. The pressure holding times (CFU mL1) in frozen samples at 20
C without HP treatment. D
mentioned did not include the pressure come-up time (1e2 min) value (min) is the decimal reduction time of E. coli which implies
and pressure release time (~10 s). After HP treatment of frozen treatment holding time in minutes at constant pressure for 90%
samples, the special plastic container holding the test vials were destruction of the existing microbial population. D values were
observed to make sure water soaked sponge remained largely in obtained from the linear regression slope of log10(Nt/N0) versus t as
the frozen state (melting extending no more than 7 mm on the shown in Eq. (2) or time taken to traverse one logarithmic cycle:
sides) and then immersed in running water (room temperature) for
2 h to complete thawing. Treated sample vials (thawed vials of D 1=slope (2)
frozen and vials of unfrozen) were held at 4
C (maximum 4 h) prior For pressure resistance (pressure sensitivity) analysis, Zp values
to microbial enumeration. Due to adiabatic heating, the tempera- were used (Shao et al., 2007). Zp value represented the pressure
ture of the pressure transmission medium (water) is expected to range between which the D value change by a decimal factor and
increase about 3
C per 100 MPa on an average. However, because can be determined as the negative reciprocal of the slope of log10
of addition of some ice, the water temperature during the treat- (D) versus pressure curve.
ment remained between 23 and 30
C. Statistical analysis was performed using one-way analysis of
variance (ANOVA). Duncan's test (P < 0.05) was applied to compare
the differences in average values by using SPSS 20.0 (Statistical
2.6. Enumeration of E. coli
Package for the Social Sciences, SPSS Inc., USA).

The sample vials were aseptically opened and the carrot juice
was aseptically transferred to a sterile tube, followed by a serial 3. Results and discussion
dilution in 1 g/L peptone water (Sinopharm Chemical Reagent Co.,
Ltd, Shanghai, China). Surviving cells of E. coli were enumerated on 3.1. Temperature and phase transition in frozen carrot juice during
BHIA using pour plate method. Colonies of survived E. coli were HP treatment
counted after aerobic incubation at 37
C for 48 h. Initial counts of
E. coli in carrot juice were obtained likewise from untreated control Fig. 2 illustrates the transient temperatures in frozen carrot juice
samples both at 4 and -20
C to differentiate the effect of freezing during HP treatment for a holding time of 1 min at different pres-
on microbial reduction. sure levels (Fig. 2a, c, e, g, i) and temperature-pressure proles of
122 S. Zhu et al. / LWT - Food Science and Technology 79 (2017) 119e125

Fig. 2. Temperature changes with pressure of frozen carrot juice samples during HP treatment holding for 1 min under different pressures (a, c, e, g, i) and temperature-pressure
proles superimposed on the phase diagram of water (b, d, f, h, j).

frozen carrot juice superimposed on the phase diagram of water was 2 min and shorter times were required proportionally at other
(Bridgman, 1912) (Fig. 2b, d, f, h, j), respectively. Pressure increased pressure levels. The interval t2-t3 was 60 s in all plots. During
in a stepwise fashion and the duration of t1-t2 depended on the decompression, the pressure decreased (t3-t4) rapidly within 10 s.
nal pressure. During pressure treatment 400 MPa, come up time Table 1 shows the temperatures achieved during HP treatment of
 S. Zhu et al. / LWT - Food Science and Technology 79 (2017) 119e125 123

Table 1 always higher than the sample container, it is likely that thawing
Temperatures achieved at the end of pressure come up time and holding time continued (endothermic reaction) in surrounding sponge space in
(n 3).
the container during the holding phase while keeping the sample
Pressure (MPa) Temperature (
C) to stay at lower temperature. When the pressure was released, the
End of pressure come up End of holding 1 min phase transition from Ice-III to Ice-I and then to liquid water
a resulted at 350 and 400 MPa (exothermic reaction). At pressures
200 20.1 1.1 19.2 1.7
250 22.6 1.3 21.7 1.5 lower than 350 MPa, phase transition of metastable Ice-I to Ice-I
300 27.5 0.3 26.6 0.0 and water seemed to occur (exothermic reaction).
350 18.2 0.2 18.2 0.2 Super cooling was caused during the decompression, so partial
400 17.6 0.8 17.7 0.9
pressure shift freezing also occurred (exothermic reaction). The
a
Mean values standard deviation. sample temperature decrease as a result of expansion cooling was
only observed at 350 and 400 MPa during decompression. During
depressurization, the increase or decrease of sample temperature
frozen carrot juice. It can be observed that the temperature was depended on the combined effect of expansion cooling, the amount
kept well below the conventional freezing point during HP freezing, of melted water converted to Ice-I (an exothermic reaction) and the
The intensier used for generating high pressure was a batch type amount of Ice-III converted to Ice-I (also an exothermic reaction)
unit which generated a stepwise ladder-like pressurization build (Van Buggenhout, Messagie, Van der Plancken, & Hendrickx, 2006).
up. Hence the ladder-like pressure and temperature change was At higher pressures (above ~220 MPa), a solid-solid transition
visible during the compression. Carrot juice samples contained in could be induced as the theoretical Ice-I and Ice-III phase transition
the special container were taken out from a freezer at 20
C. The line of pure water was exceeded upon compression (Van
temperature rose slightly during the preparation, vacuum packing, Buggenhout et al., 2007). Similar phase transition of Ice-I to Ice-III
connecting of thermocouples, installation of sample container into has been reported at pressures above 250 MPa (Su et al., 2014;
the pressure chamber and the quick lling of water into the Van Buggenhout et al., 2007), but the sudden changes in pressure
chamber. During compression, the temperature still rose at rst, as and temperature were not similar. Again this might due to the
the heat of compression and the heat input from pressure chamber differences in the media used: microbial culture (335 MPa, 35
exceeded the heat uptake of melting ice crystals (endothermic re- to 18
C), carrot (280 MPa, 36 to 30
C) and carrot juice
action) in the sponge surrounding the vials. Partial thawing (349 MPa, 35 to 19
C). Ferna
ndez et al. (2007) reported that
occurred and the melting of ice crystals increased in the sponge. phase diagram for beef meat showed the same phase transition
However, the heat of compression and the heat input from pressure regions as water, but shifted to lower temperature and different
chamber were insufcient to fully melt ice in the sponge and hence pressure values.
help to protect the frozen carrot juice in the test vials. The tem-
perature decreased upon further compression due to lowering of
3.2. High pressure inactivation of E. coli in unfrozen carrot juice
the freezing point and approached the melting line of carrot juice
(water melting line shown in Fig. 2). Heat of compression and latent
The average initial concentration of E. coli in carrot juice before
heat in this phase were in equilibrium, and solid-liquid transition
freezing was 7.8  107 CFU/mL. After frozen storage for 24 h, E. coli
was initiated. When the temperature was 10
C or 5
C at the
count in carrot juice was reduced slightly to 7.65  107 CFU/mL. The
beginning (Fig. 2c, g), the temperature could decrease even during
initial concentration of E. coli in carrot juice samples varied be-
compression and approached the melting line. Because the tem-
tween 107 and 108 CFU/mL and around were normalized to
perature was still lower than the freezing point at the process
107 CFU/mL (by dividing all counts by their actual initial count and
pressure, the results indicated that the special container with water
then multiplying by 107). This facilitates better presentation of
soaked frozen sponge had enough heat sink and insulation to
survival curves for comparison purposes with all curves starting
protect the sample from heat gain and hold them under frozen
from the same point. The normalization process only results in a
state.
small intercept shift and will not affect the slope, and therefore D
According to the temperature-pressure proles superimposed
on the phase diagram of water in Fig. 2 (b, d, f, h, j), the metastable
Ice-I region was positioned above 220 MPa. This pressure was little
higher than 207 MPa reported by Su et al. (2014) for E. coli sus-
pension. On the other hand, the phase transition point of pure
water is 210 MPa. Hence there existed some small differences
which were probably due to differences in the nature of the support
medium (water vs carrot juice). In Fig. 2 (g and i), a sudden increase
in sample temperature (35 to 19
C) was observed during
compression (highlighted in a gray circle). A sudden pressure
decrease (349e334 MPa) was also observed at 400 MPa (high-
lighted in a dark gray circle in Fig. 2i). These results indicated the
transition to Ice-III, including metastable Ice-I to Ice-III (endo-
thermic reaction) and water re-crystallization to Ice-III (exothermic
reaction). Because Ice-III (1.14 g/cm3) has a higher density than
water (1 g/cm3) and Ice-I (0.92 g/cm3), the volume change con-
tributes to a change in pressure. The temperature increase might be
caused by the exothermic transformation of some liquid water to
Ice-III (Van Buggenhout, Grauwet, Van Loey, & Hendrickx, 2007).
After compression, the temperatures at 200, 250, 300, 350 and
400 MPa remained at 19, 21, 27, 18 and 18
C, respectively,
Fig. 3. E. coli survivors in unfrozen carrot juice samples after different high pressure
with a slight increase. Since the surrounding temperature was (300 MPa (B), 350 MPa (), 400 MPa ()) and different holding time.
124 S. Zhu et al. / LWT - Food Science and Technology 79 (2017) 119e125

values remain the same.

Fig. 3 shows the survivor curves of E. coli in the unfrozen carrot
juice, and Table 2 shows the computed kinetic parameters of E. coli
for HP treatment. The extent of microbial inactivation at 300 and
350 MPa during the 10 min holding time was small resulting in
calculated D values of 28.5 and 9.2 min, respectively. It improved a
little better at 400 MPa, with a calculated D value of 5.3 min. The
range of holding times (0e10 min) was low but kept the same as for
HP treatment of frozen samples to provide baseline data for
comparison.
The extent of HP inactivation of E. coli in unfrozen carrot juice
were similar to those reported by other researchers (Bari et al.,
2007; Patterson et al., 2012; Pilavtepe-elik et al., 2009; Van
Opstal et al., 2005). D values ranging from 38 to 2 min were re-
ported by Van Opstal et al. (2005) for E. coli MG1655 inactivation
after HP treatment at 300e400 MPa in the temperature range
23e40
C. Su et al. (2014) reported slightly lower D values for the
same strain of E. coli suspended in water. A carrot juice matrix was
used in this study and the use of real food matrix could have been
the reason for the observed higher values. Others have also re- Fig. 4. E. coli survivors in frozen carrot juice samples after different high pressure
ported that cells in food matrix are more resistant than in buffer (200 MPa (,), 250 MPa ()300 MPa (B), 350 MPa (), 400 MPa ()) and different
solutions (Alpas, Lee, Bozoglu, & Kaletun, 2003; Pilavtepe-elik holding time.
et al., 2009).

3.3. High pressure inactivation of E. coli in frozen carrot juice

Table 3
Fig. 4 shows the logarithmic survivor curves of E. coli in frozen Inactivation kinetics of E. coli in frozen carrot juice under high pressure.
carrot juice subjected to HP treatment at 200e400 MPa for
Pressure (MPa) D value (min) R2Adj Zp value (MPa) R2Adj
0e10 min holding time. The E. coli inactivation in frozen samples
a
also followed the rst order rate kinetics during the pressure 200 4.03 0.35 0.963 613 118 0.866
250 3.97 0.39 0.954
holding time (R2 > 0.95). Table 3 shows the D values and other
300 2.62 0.19 0.974
kinetic parameters of E. coli in frozen carrot juice treated by HP. 350 2.21 0.10 0.991
The reduction trends in E. coli in frozen carrot juice were similar 400 2.12 0.08 0.994
to those in unfrozen carrot juice samples - survivors decreased with a
SE - standard error.
increasing pressure and treating time, only by a greater margin
(both rst order rate). The survival lines were steeper at higher
pressures indicating that the associated D values decreased at
higher pressures. inactivation in carrot juice.
The E. coli count reduction at 200 MPa after pressure treatment From the D value data, the computed pressure sensitivity
for 0 and 10 min were 0.42 and 3.00 log10 CFU/mL, while they parameter (Zp value) was 613 MPa much higher than the 196 MPa
increased to 2.00 and 6.80 log10 CFU/mL at 400 MPa, respectively. reported by Su et al. (2014). It is clear that the pressure sensitivity
The count reduction was far more in frozen carrot juice than in was affected signicantly when carrot juice was treated in the
unfrozen samples. For example, after 10 min treatment at 400 MPa, frozen state. The nature of the phase transition, pressure and
the reduction of E. coli count in unfrozen samples was 1.87 sample state could be the reasons for this observed remarkable
log10 CFU/mL, but 6.80 log10 CFU/mL in frozen samples. The difference in the computed Zp value. Although the D value data
calculated D value (5.32 min) of E. coli at 400 MPa in unfrozen were combined for the purpose of computation of Zp value, it can be
samples was longer than the D values in frozen samples obtained seen from Fig. 3 that the HP inactivation rate of E. coli at the two
even at much lower pressures of 200 and 250 MPa (4.03 and lower pressure levels (200, 250 MPa) were distinctly different from
3.97 min, respectively. The D value of frozen carrot juice at 400 MPa those at the two higher levels (300e400 MPa). At 200 and 250 MPa
(2.12 min) found in this study was close to Van Opstal et al. (2005) the associated D values were 4.03 and 3.97 min, respectively, while
reported value of 2.0 min of E. coli MG1655 at 400 MPa, 40
C in at 300, 350 and 400 MPa they were 2.62, 2.21 and 2.12 min,
freshly extracted carrot juice; but they found the D values to be respectively, indicating a break in the continuity pattern. Therefore,
longer at lower temperatures (5e30
C). The D value of frozen although the inactivation rate increased with pressure level
carrot juice at 300 MPa (2.62 min) was consistent with Su et al. (consistent with general understanding), a pattern difference was
(2014) of frozen E. coli suspension (2.59 min). The frozen state also apparent. This is probably due to the re-crystallization of Ice-III
therefore contributed to a remarkable enhancement of E. coli and phase transition from metastable Ice-I to Ice-III during
compression and Ice-III to Ice-I and/or water during decompres-
sion, when the pressure was above 250 MPa. It has been reported
Table 2
Inactivation kinetics of E. coli in unfrozen carrot juice under high pressure. earlier that phase transitions may lead to mechanical disintegration
of cells (Edebo & Hede
n, 1960) and the inactivation would be
Pressure (MPa) D value (min) R2Adj Zp value (MPa) R2Adj
caused by a mechanical effect related to the Ice-I to Ice-III phase
300 28.53 5.27 a
0.850 139 28 0.922 transition (Ferna
ndez et al., 2007; Luscher et al., 2004). The crystal
350 9.20 1.03 0.940 formation and phase transition therefore could play important
400 5.32 1.06 0.827
roles in the HP in inactivation on E. coli in frozen medium.
a
Standard error (SE).
 S. Zhu et al. / LWT - Food Science and Technology 79 (2017) 119e125 125

Table 4 49(49), 1410e1416.

Pulse effect of high pressure on E. coli in frozen carrot juice. Mean Alpas, H., Lee, J., Bozoglu, F., & Kaletun, G. (2003). Evaluation of high hydrostatic
values standard deviation (n 3). pressure sensitivity of Staphylococcus aureus and Escherichia coli O157:H7 by
differential scanning calorimetry. International Journal of Food Microbiology,
Pressure (MPa) Reduction (log10 CFU/mL) 87(3), 229e237.
Bari, M. L., Ukuku, D. O., Mori, M., Kawamoto, S., & Yamamoto, K. (2007). Effect of
200 0.42 0.11a
high-pressure treatment on survival of Escherichia coli O157:H7 population in
250 0.53 0.05a tomato juice. Journal of Food Agriculture & Environment, 5(1), 111e115.
300 0.85 0.09b Basak, S., & Ramaswamy, H. S. (2001). Pulsed high pressure inactivation of pectin
350 1.41 0.35c methyl esterase in single strength and concentrated orange juices. Canadian
400 2.00 0.24d Biosystems Engineering, 43(3), 25e29.
Bridgman, P. W. (1912). Water in the liquid and ve solid forms under pressure, Pro-
Different letters (a, b, c, d) indicate signicant difference (p < 0.05) be-
c.Am. Acad. Arts Sci., XLVIl(13), 441e558.
tween treatments.
Dede, S., Alpas, H., & Bayindirli, A. (2007). High hydrostatic pressure treatment and
storage of carrot and tomato juices: Antioxidant activity and microbial safety.
Journal of the Science of Food and Agriculture, 87(5), 773e782.
3.4. Pressure-pulse inactivation of E. coli in frozen carrot juice Edebo, L., & Hede
n, C. G. (1960). Disruption of frozen bacteria as a consequence of
changes in the crystal structure of ice. Journal of Biochemistry, Microbiology and
Technological Engineering, 2(1), 113e120.
The inactivation effect of pressure pulse on E. coli in frozen Ferna
ndez, P. P., Sanz, P. D., Molina-Garca, A. D., Otero, L., Guignon, B., &
carrot juice is shown in Table 4. For unfrozen samples, the pressure Vaudagna, S. R. (2007). Conventional freezing plus high pressureelow tem-
perature treatment: Physical properties, microbial quality and storage stability
pulse had almost no effect at any of the pressure levels from 200 to
of beef meat. Meat Science, 77(4), 616e625.
400 MPa (data not shown). However, the pressure pulse effect was Luscher, C., Balasa, A., Fro hling, A., Ananta, E., & Knorr, D. (2004). Effect of high-
clearly noticeable in frozen carrot juice samples which was low at pressure-induced Ice I-to-Ice III phase transitions on inactivation of Listeria
200 and 250 MPa, but signicantly increased to 0.85, 1.41 and 2.00 innocua in frozen suspension. Applied & Environmental Microbiology, 70(7),
4021e4029.
logarithmic cycle reductions at 300, 350 and 400 MPa, respectively. Oey, I., Lille, M., Loey, A. V., & Hendrickx, M. (2008). Effect of high-pressure pro-
It was consistent with the D values pattern between 200-250 MPa cessing on colour, texture and avour of fruit- and vegetable-based food
and 300e400 MPa discussed earlier. Therefore these differences products: A review. Trends in Food Science & Technology, 19(6), 320e328.
Patterson, M. F., McKay, A. M., Connolly, M., & Linton, M. (2012). The effect of high
could also be ascribed to the frozen state of carrot juice and phase hydrostatic pressure on the microbiological quality and safety of carrot juice
transition during HP, especially the Ice-I/Ice-III transition above during refrigerated storage. Food Microbiology, 30(1), 205e212.
300 MPa. Bari et al. (2007) reported that repeating a 10 min high Picart, L., Dumay, E., Guiraud, J. P., & Cheftel, J. C. (2005). Combined high pressure-
subzero temperature processing of smoked salmon mince: Phase transition
pressure treatment cycle with a 5 s pause could achieve extra phenomena and inactivation of Listeria innocua. Journal of Food Engineering,
reduction. It is possible that repeating HP pulse treatment likewise 68(1), 43e56.
would get more reduction as demonstrated in some previous Picouet, P. A., Hurtado, A., Jofre
, A., Ban ~ on, S., Ros, J., & Dolors Gua rdia, M. (2016).
Effects of thermal and high-pressure treatments on the microbiological,
studies with multiple pressure pulse inactivation (Basak &
nutritional and sensory quality of a multi-fruit smoothie. Food and Bioprocess
Ramaswamy, 2001). Technology, 9(7), 1219e1232.

rraga, C., Cof

Picouet, P. A., Sa an, S., Belletti, N., & Dolors Gua rdia, M. (2015). Effects of
thermal and high-pressure treatments on the carotene content, microbiological
4. Conclusions
safety and sensory properties of acidied and of non-acidied carrot juice. LWT
- Food Science and Technology, 62(1), 920e926.
The transient time temperature data obtained under pressure Pilavtepe-elik, M., Buzrul, S., Alpas, H., & Bozog lu, F. (2009). Development of a new
processing conditions involving frozen carrot juice revealed that mathematical model for inactivation of Escherichia coli O157:H7 and Staphylo-
coccus aureus by high hydrostatic pressure in carrot juice and peptone water.
the ice structure could be predominantly of type Ice-III or Ice-I. Journal of Food Engineering, 90(3), 388e394.
Solid-solid and solid-liquid transitions were apparent during Ramaswamy, H. S. (2011). High pressure sterilization of foods. In J. M. M., Simpson
compression and decompression. E. coli inactivation in frozen R. V., Welti-Chanes J., Bermudez-Aguirre D. and Barbosa-Canovas G. Aguilera,
et al. (Eds.), Food engineering interfaces, food engineering series (pp. 341e351). #
carrot juice followed the rst order rate kinetics after accounting Springer Science & Business Media, LLC. http://dx.doi.org/10.1007/978-1-4419-
for a pulse effect inactivation. The high pressure inactivation of 7475-4_15.
E. coli in frozen samples was far better than in unfrozen samples, Ramaswamy, H. S., Jin, H., & Zhu, S. (2009). Effects of fat, casein and lactose on high-
pressure destruction of Escherichia coli K12 (ATCC-29055) in milk. Food and
and was enhanced by pressure level and holding time. The phase Bioproducts Processing, 87(C1), 1e6.
transition, pressure level and sample state contributed to the better Ramaswamy, H. S., Riahi, E., & Idziak, E. (2003). High-pressure destruction kinetics
inactivation of E. coli in frozen carrot juice, especially the Ice-I/Ice- of E. coli (29055) in apple juice. Journal of Food Science, 68, 1750e1756.
Ramaswamy, H. S., Zaman, S. U., & Smith, J. P. (2008). High pressure destruction
III transitions at pressure 350 MPa. The high inactivation rate at 
kinetics of Escherichia coli (O157:H7) and Listeria monocytogenes (Scott A) in a
350 MPa in frozen state indicates a good potential for commercial sh slurry. Journal of Food Engineering, 87(1), 99e106.
application. Further, the special insulator attachment provides Shao, Y., Ramaswamy, H. S., & Zhu, S. (2007). High pressure destruction kinetics of
spoilage and pathogenic bacteria in raw milk cheese. Journal of Food Process
opportunity for accomplishing HP inactivation in the frozen state
Engineering, 30(3), 357e374.
using HP equipment operating at room temperatures. However, Su, G., Yu, Y., Ramaswamy, H. S., Hu, F., Xu, M., & Zhu, S. (2014). Kinetics of
further studies are needed to evaluate the stability and the inu- Escherichia coli inactivation in frozen aqueous suspensions by high pressure and
ence on the food quality and properties and the use of these pro- its application to frozen chicken meat. Journal of Food Engineering, 142, 23e30.
Van Buggenhout, S., Grauwet, T., Van Loey, A., & Hendrickx, M. (2007). Effect of
cesses might be limited due to the frozen state. high-pressure induced ice I/ice III-transition on the texture and microstructure
of fresh and pretreated carrots and strawberries. Food Research International,
Acknowledgements 40(10), 1276e1285.
Van Buggenhout, S., Messagie, I., Van der Plancken, I., & Hendrickx, M. (2006). In-
uence of high-pressureelow-temperature treatments on fruit and vegetable
This work was supported by the Key Program of Natural Science quality related enzymes. European Food Research and Technology, 223(4),
Foundation of Zhejiang Province (Grant No. LZ14C200002) and the 475e485.
Van Opstal, I., Vanmuysen, S. C. M., Wuytack, E. Y., Masschalck, B., & Michiels, C. W.
National Natural Science Foundation of China (Grant No. 31171779). (2005). Inactivation of Escherichia coli by high hydrostatic pressure at different
temperatures in buffer and carrot juice. International Journal of Food Microbi-
References ology, 98(2), 179e191.
Vervoort, L., Plancken, I. V. D., Grauwet, T., Verlinde, P., Matser, A., Hendrickx, M.,
et al. (2012). Thermal versus high pressure processing of carrots: A comparative
Alasalvar, C., Grigor, J. M., Zhang, D., Quantick, P. C., & Shahidi, F. (2001). Comparison
pilot-scale study on equivalent basis. Innovative Food Science & Emerging
of volatiles, phenolics, sugars, antioxidant vitamins, and sensory quality of
Technologies, 15(1), 1e13.
different colored carrot varieties. Journal of Agricultural & Food Chemistry,