Sunteți pe pagina 1din 8

Available online at www.sciencedirect.

com

MEAT
SCIENCE
Meat Science 80 (2008) 488–495
www.elsevier.com/locate/meatsci

A comparison of conventional and radio frequency tempering


of beef meats: Effects on product temperature distribution
K.W. Farag, J.G. Lyng *, D.J. Morgan, D.A. Cronin
UCD School of Agriculture, Food Science and Veterinary Medicine, Agriculture and Food Science Centre, College of Life Sciences, UCD Dublin,
Belfield, Dublin 4, Ireland

Received 11 November 2007; received in revised form 17 January 2008; accepted 31 January 2008

Abstract

This study aimed to develop radio frequency (RF) pilot-scale protocols for tempering beef meat blends (4 kg blocks) to achieve aver-
age temperatures between 2 and 5 °C. Post-tempering temperature distribution in these blocks was compared to products tempered
by conventional methods. The optimum RF power–time combination for tempering lean and 50:50 lean:fat mixtures to the target range
was 500 W for 11 min which produced respective means of 3.6 °C (s.d. 1.1) and 3.4 °C (s.d. 1.5). In contrast, 400 W for 11 min was
optimum for fat (mean 4.9 °C, s.d. 2.1). This study shows the principal advantages of RF over conventional tempering as an approx-
imate 30 fold tempering time reduction and a greater uniformity of end point temperature distribution under the conditions employed.
Furthermore, power consumption was reduced approximately ninefold with RF compared to conventional tempering. More uniform
temperature distribution was achieved in samples that were comminuted to a greater extent.
Ó 2008 Elsevier Ltd. All rights reserved.

Keywords: Radio frequency; Frozen beef; Tempering

1. Introduction that is more amenable to subsequent processing than deep


frozen meat (i.e., 18 to 20 °C). Tempering also reduces
Novel technologies that reduce process time and main- problems associated with thawing such as increased drip
tain or improve product quality are of considerable interest loss and bacterial growth (James & James, 2002).
to the food industry. One problematical area which is of Many methods can be used to temper and thaw prod-
particular relevance to the meat industry is the excessively ucts including placing the product in contact with air or
long time required in conventional heating to bring bulk water at controlled temperatures. Air tempering is the most
frozen meat to temperatures at which it can be processed. commonly used. In air tempering, the air in contact with
To defrost meat it may be thawed (temperature of the product transfers heat to the outer surface of the prod-
the coldest spot just above product freezing point) or tem- uct. This heat then transfers to the interior of the product
pered (temperature of the hottest spot just below product by conduction which is a relatively slow method of heat
freezing point). Of these two unit operations, tempering transfer. Industry is always interested in faster and more
is the more commonly used particularly for products which compact systems while yet maintaining the high quality
subsequently require further size reduction (e.g., grinding, demanded by consumers. In response to this demand,
bowl chopping). There are many reasons for this including methods such as vacuum heat (Lind & Hulthen, 1986),
the fact that it is faster than thawing, yet yields a product resistive (Yun, Lee, & Park, 1998), microwave and radio
frequency tempering have been evaluated and developed
though the uptake of these technologies by industry rela-
*
Corresponding author. Tel.: +353 1 7167710; fax: +353 1 7161149. tive to air tempering has been proportionally small to date.
E-mail address: james.lyng@ucd.ie (J.G. Lyng). Of the systems listed dielectric heating systems (i.e.,

0309-1740/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.meatsci.2008.01.015
K.W. Farag et al. / Meat Science 80 (2008) 488–495 489

microwave and radio frequency) would appear to have the The lean and fat were ground through a plate with 10 mm
most potential. In contrast to air tempering, these systems diameter holes using a mechanical mincer (Model No.
heat foods by dipole rotation and ionic depolarisation TS8E, Tritacarne, Omas, Italy). For the lean and fat,
mechanisms where in essence heat is generated within the the meats were bowl chopped with a Manica bowl chop-
product through the movement and friction of dipoles or per (Model No. CM22, Equipaimentos Carnicos,
ions in response to an alternating electric field (Buffler, Barcelona, Spain) for 120 s at knife speed one and 60 s at
1993). This will in theory allow for a relatively uniform knife speed two. A similar procedure was used for the
generation of heat within all parts of the product. 50:50 lean:fat blend after thoroughly hand mixing the
Relatively few studies have been published on the dielec- components.
tric heating of frozen foods. Thirty boxes (20  20  10 cm) were prepared using
In 1946, Cathcart and Parker successfully thawed small 0.3 cm thick freezer grade cardboard (Kepak, Clonee Co.
samples of egg, fruit, vegetables and fish from 17 °C to Meath, Ireland) and equal numbers were filled with lean,
+10 °C. Haddock fillets for example were defrosted to a fat or 50:50 mixture. This box size was chosen following
final temperature of 9.4 °C after 10 min of RF treatment preliminary studies with boxes of other dimensions, includ-
(14–17 MHz). Rapid RF thawing (36–40 MHz) of fish to ing 20  20  20 cm, 20  10  10 cm and 10  10 
give a temperature distribution of between 1 °C and 10 cm. Care was taken to exclude air during filling and then
+7 °C and lower losses than achieved with air and water the meat surfaces were covered with cellophane. Boxes
thawing was reported by Jason and Sanders (1962). Bengts- were frozen in a conventional air freezer at 20 °C for
son (1963) compared RF (35 MHz) and MW (2450 MHz) 48 h and subsequently maintained at this temperature until
defrosting of meat and fish and observed shorter treatment required. To evaluate the effect of comminution (Section
times for microwave but lower power penetration into fro- 3.1) on temperature distribution following RF tempering,
zen material. In 1966, Sanders attempted to thaw blocks of triplicate boxes of whole, minced (as above) and finely
beef and pork meat with RF (36–40 MHz), but the final comminuted lean (bowl chopped) were prepared.
temperature range achieved (2 °C to +19 °C) was very
high. More recently, Pizza, Pedrielli, Busetto, Bocchi, and 2.2. Tempering method
Spinelli (1997) investigated RF tempering of ham products
using a discontinuous RF heating method at 27.5 MHz and 2.2.1. Radiofrequency (RF) tempering
claimed to have successfully tempered meat from 20 °C to The RF oven used was a custom built 50 X system built
8 °C. Other than the work of Pizza et al. (1997), most of by C-Tech Innovation (Chester, UK) using a low power
the studies published focused on thawing rather than tem- (0.6 kW) RF generator (Model No. RFG 600-27, Coaxial
pering meat. Power System Ltd., Spectrum House, Finmere Road, East-
Because of the paucity of information of RF tempering bourne, East Sussex, UK) and a complementary automatic
of meat a key objective of the present study was to develop impedance matching network and controller (Model No.
a satisfactory protocol for RF tempering blocks of meat to AMN 600-27) at a frequency of 27.12 MHz. The boxed
a temperature ranging between 5 and 2 °C which is rep- meat was placed at the centre of the bottom electrode
resentative of target temperatures set for commercial meat (Fig. 1) and tempered using one of the time/power combi-
tempering. Neither the effect of composition (fat content in nations as summarised in Table 1.
particular) nor the degree of comminution of beef meat on
RF tempering have been adequately documented in the lit- 2.2.2. Conventional tempering
erature. Therefore, another objective of the study was to Conventional air tempering was performed on a labora-
investigate the temperature profile variation of different tory scale (400  400  330 mm) using a programmable
blends of beef meat subjected to tempering (lean beef, beef Binder incubator (Model No. KB 53, Binder GmbH, D-
fat and 50:50 mixture), in order to understand better the 78502 Tuttlingen, Germany). The time–temperature profile
impact of composition on RF tempering. RF tempering used for conventional tempering was developed based on
at 27.12 MHz was compared to conventional air tempering data supplied from industry contacts. To represent indus-
and also to MW tempering at 2450 MHz. The latter repre- try practice, a programme consisting of 20 stages each of
sents the upper extreme of electromagnetic radiation fre- different duration and temperature was developed
quency used in dielectric heating of food, thus giving a (Fig. 2). Individual meat boxes were placed inside the incu-
marked contrast to the RF frequency used in the current bator to ensure optimum distribution of air-flow around
study. the product.

2. Materials and methods 2.2.3. Microwave (MW) tempering


MW tempering was carried out using an Amana Com-
2.1. Meat and meat preparation mercial MW (Model No. RS591SS, Amana Refrigeration
Inc., PO Box 8901, IA 52204-0001, USA) at a frequency
Lean beef (95% visual lean) and beef fat was procured of 2450 MHz. MW lean beef samples were tempered at
from a local supplier (Kepak, Clonee, Co. Meath, Ireland). either 400 W for 14 min or 500 W for 13 min; different
490 K.W. Farag et al. / Meat Science 80 (2008) 488–495

20 cm
30 cm

Electrodes 20 cm
10 cm
20 cm
25 cm

46 cm
Block of
meat

Top electrode
1.4 cm
Block of meat

Bottom electrode

Fig. 1. Schematic diagram of (a) RF oven interior showing electrode positions and product and (b) side profile of top electrode and product.

Table 1 time–power combinations were examined in order to find


RF time–power combinations evaluated for tempering meat an optimized combination of the two.
Meat type Power (W) T1 (min) T2 (min) T3 (min) T4 (min)
Lean 400 9 10 11 – 2.3. Temperature measurement
500 8 9 10 11
600 7 8 9 10 2.3.1. Measurement of internal temperature
50:50 mixture 400 10 11 12 –
2.3.1.1. Jig construction and use. A thermocouple jig tech-
500 9 10 11 –
600 8 9 10 – nique was used to measure temperature from 25 locations
Fat 200 14 18 – – within the sample following tempering. However, due to
400 10 11 – – the solid nature of the frozen meat, it was necessary to drill
500 8 9 10 – holes at precise locations within samples prior to tempering
600 8 – – –
to ensure repeatable insertion of thermocouples. A fibre-
board template was prepared as illustrated in Fig. 3b. This
template had the same upper surface dimensions as those
of the meat blocks and, following their removal from the
20
freezer, it was immediately placed in position on the surface
of the meat. An electric drill mounted on a drill stand was
15 fitted with a 6 mm diameter drill bit (length 20 cm). Holes
(10 cm) deep were drilled into the samples at the locations
Temperature (˚C)

10 marked on the template to facilitate post-tempering inser-


tion of the temperature probes.
Endpoint temperature for each of the treatments was
5
evaluated for all blends tempered by inserting a purpose
built thermocouple jig into the product immediately post-
0
tempering. The jig was assembled from five wooden skew-
ers (diameter 3.5 mm) attached to a fibreboard holder
-5 (20  20  1 cm). Each skewer was 11 cm long and five
‘type K’ thermocouples (Radionics Ltd., Rialto, Dublin,
0 100 200 300 400 500 600
Ireland) were fixed to it to ensure the measuring tips were
Time (min)
located at 1, 3, 5, 7 and 9 cm depths from the upper surface
Fig. 2. Time–temperature profile of programmable incubator (—). of the meat (Fig. 3c).
K.W. Farag et al. / Meat Science 80 (2008) 488–495 491

3.3 cm 6.6 cm 6.6 cm 3.3 cm


2.5. Proximate analysis
3.3 cm
6.6 cm Moisture, protein, fat and ash content were determined
for the three types of meat (lean, fat and 50:50 mixture),
6.6 cm 20 cm using AOAC (1995) method numbers 950.46, 928.08,
Top 991.36 and 920.153, respectively. All samples were analysed
3.3 cm
in triplicate.
20 cm
1 cm 3. Results and discussion
3 cm
3.1. Effect of comminution level
5 cm
When meat was finely comminuted before tempering the
7 cm temperature distribution was more uniform compared to
minced or whole meat. As shown in Fig. 4, MnT were
9 cm 5.6 °C, 6.8 °C and 6.6 °C while DT were 3.6 °C,
11.7 °C and 9.8 °C for finely comminuted, minced and
Bottom whole meat, respectively. Therefore, in less homogenous
meat products (e.g., whole meat cuts), a wider range of
temperature distribution is more likely with greater poten-
tial for runaway heating. Localised fat and/or uneven
moisture distribution could in that case act as a trigger
for selective heating. In view of this fact, it was decided
20 cm

20 cm
1 2 2
Fig. 3. Illustration of (a) fibreboard template for drilling positions, (b) 0 0
2 -2
thermocouple jig and (c) position of thermocouples on wooden skewers. -2
3
Depth (cm)

4
5
-4
The jig was inserted for 2 min into the block at the end
6 -4
of the tempering process, while two time–temperature log- 7 0 -6
0 -2
gers (Model No. 1600 and 1250, Grant Instruments Ltd., 8 -2
Barrington, Cambridge CB2 5QZ, UK) were used to 9
record post-tempering temperatures at the 25 specified 6 8 10 12 14 16 18 20 22

points within the product at 6 s intervals from the time of Diagonal length (cm)
insertion. Maximum temperature (MxT), minimum tem-
1
perature (MnT), mean temperature ( xT) and the 2 -2 -2
MxTMnT temperature difference (DT) within the prod- 3
Depth (cm)

uct were evaluated for each data set (25 values per data 4
set) over the duration of the tempering process. 5 -4
-4 2 -2
6 -2
4
2.3.2. Surface temperature measurement 7 0 0
6 -6
8
An IRISYS, Universal Thermal Imager (Type: IRI 8
9
1011, Infrared Integrated Systems, Towcester Mill, Towc- 6 8 10 12 14 16 18 20 22
ester, Northants NN12 6AD, UK) was used to capture sur- Diagonal length (cm)
face temperature profiles of the meat thereby facilitating 1
the location of the hot and cold spots on the surface. 2 -3 -3
3
Depth (cm)

2.4. Measurement of power consumption 4


5 -4
-5 -4
6
A power meter (LEM Heme Analyst Model No. 2050,
7
LEM UK Ltd., Geneva court, Lancashire WN8 9QX, -3
8 -3
UK), was used to measure the power consumption in kilo- 9
watt-hour (kW h). All saved data were downloaded to a PC 6 8 10 12 14 16 18 20 22

for analysis using WinLog software (Great Britain & Ire- Diagonal length (cm)
land LEM UK Ltd., Geneva court, Lancashire WN8 Fig. 4. The effect of comminution level on end point temperature of
9QX, UK). tempered lean (a) whole, (b) minced and (c) finely comminuted.
492 K.W. Farag et al. / Meat Science 80 (2008) 488–495

to use finely comminuted meat for all subsequent temper-


ing experiments.

3.2. RF tempering

3.2.1. Effect of block dimensions


Preliminary experiments showed the block sizes 20 
10  10 cm and 10  10  10 cm to be unsatisfactory due
to tuning difficulties with the RF setup employed, particu-
larly with fat samples. The two larger sizes (20  20  20
and 20  20  10 cm) both proved satisfactory with regard
to tuning, but for reasons of economy the smaller size
(20  20  10 cm) was finally adopted.

3.2.2. Optimisation of RF tempering protocol


Fig. 5 provides an illustration of power–time combina-
tion effect on  xT, MnT and MxT of different beef blends
examined in the present study. In the case of lean, both
500 W/11 min and 600 W/9 min were found to be close to
the desired range of 2 to 5 °C, where MxT was 1.9
and 2.5 °C and MnT was 5.6 and 5.5 °C, respectively.
However, 500 W/11 min was favoured, as the percentage of
points in the block within the set range was higher than the
600 W/9 min combination. In case of 500 W for 11 min, 23
points out of 25 fell within the target range set where at
600 W/10 min only 21 points were within the set range.
For the 50:50 mixture, 500 W/11 min was closest to the tar-
get where MxT was 1.3 °C and MnT was 5.8 °C. Beef
fat was the most difficult of the three products tested to
bring to the target temperature range and the best that
could be achieved was 1.6 °C for MxT and 7.6 °C for
MnT using 400 W/11 min.
In Fig. 6 which shows temperature profiles of a diagonal
section in vertical plane, a distinctive pattern is evident in the
optimum RF treatments of the three blends with an increase
in contour lines between the different blends as the fat level
increases, i.e., lean < 50:50 mixture < fat. A greater number
of contour lines represent the increasing temperature distri-
bution observed for the respective samples.
Fig. 5. Effect of power/time treatment on mean, minimum and maximum
3.2.3. Effect of RF power and heating time temperature of RF tempered. (a) Lean, (b) 50:50 mixture and (c) fat.
As anticipated, the power level was an important factor
in RF tempering, i.e., for a given treatment time the higher
the power the higher end point temperature (EPT) In the current study it was observed that at lower or
attained. For example, at a specific point within lean, the higher power–time settings then the optimized ones already
temperature was 8.3, 6.7 and 5.7 °C, at 400, 500 and discussed (Section 3.2.2), cold (<5 °C) or hot (>2 °C)
600 W, respectively, after 6 min of RF tempering. While spots were a persistent feature with all samples studied.
there is no other published data on the effect of power on For example, at 400 W for 9 min MnT in lean was
RF tempering of frozen products, Houben, Schoenmakers, 7.3 °C, whereas, at 600 W for 10 min MxT in 50:50 mix-
van Putten, van Roon, and Krol (1991) observed an ture was 0.8 °C.
approximately linear relationship for temperature increase
as a function of electrode voltage, during continuous RF 3.2.4. Effect of composition
pasteurization of sausage emulsion. Duration was also Meat composition also had an influence on temperature
important in its effect on temperature. For example, at a distribution. In the current study the three beef formula-
specific point within blocks of lean following RF tempering tions (lean, beef fat and 50:50 mixture) differed significantly
at 600 W, temperatures were 9, 7.3 and 5.7 °C, respec- (P < 0.05) in fat, moisture, protein and ash levels, with the
tively, after 4, 5 and 6 min. greatest variation being observed in fat and moisture
K.W. Farag et al. / Meat Science 80 (2008) 488–495 493

Front right centre Back left Front right centre Back left
1 1
2 2 -3 -4
-3 -3 -4
3 3
Depth (cm)

Depth (cm)
4 4
-4
5 -4 5
6 6
7 7
-3
8 -3 8 -4 -4 -3
-3
9 9
6 8 10 12 14 16 18 20 22 6 8 10 12 14 16 18 20 22
Diagonal length (cm) Diagonal length (cm)

Front right centre Back left Front right centre Back left
1 1
-2 -2
2 -3 2 -3 -3
-3
3 3 -4 -4

Depth (cm)
Depth (cm)

4 4
-5 -4 -4
5 5
6 6
7 -2 7
-3
8 -3 8
-3 -3
-2 -4
9 9
6 8 10 12 14 16 18 20 22 6 8 10 12 14 16 18 20 22
Diagonal length (cm) Diagonal length (cm)

Front right centre Back left Front right centre Back left
1 1
-2 -3 -2 -1
2 -3 -4 2 -3 -2
-3
3 -4 -5 3 -4 -4
-5 -6
Depth (cm)

Depth (cm)

4 -6 4
5 -7 -7 5
6 -7 6 -5
7 -7 7
-6 -4
8 -6 8
-5 -5 -3 -4
-4 -4
9 9
6 8 10 12 14 16 18 20 22 6 8 10 12 14 16 18 20 22
Diagonal length (cm) Diagonal length (cm)

Fig. 6. End point temperature distribution profile of radiofrequency and conventional tempered meat beef blends (a) lean RF tempered, (b) 50:50 mixture
RF tempered, (c) fat RF tempered, (d) lean conventionally tempered, (e) 50:50 mixture conventionally tempered and (f) fat conventionally tempered.

(Table 2). Noticeable differences in temperature distribu- of 4.4, 6.5 and 9.4 °C. It was also noticed that the higher
tion were observed between these three beef blends. For the fat content in the product the greater the observed
example, in products heated at 600 W for 8 min, MnT val- DT (Fig. 6). This finding agrees with the observation of
ues were 6.4, 7.4 and 7.7 °C for lean, 50:50 mixture Sanders (1966) who reported a wider range of temperature
and fat blocks, respectively, with corresponding DT values in defrosted blocks of beef with higher fat level.
Bengtsson, Melin, Remi, and Söderlind (1963) stated
that the field strength in a fatty surface layer can be much
Table 2 higher than in lean (due to lower dielectric values for fat,
Proximate analysis of meat used in the study e.g., e0 and e00 values for lean meat at 35 MHz at 5 °C were
Product Moisture (%) Fat (%) Protein (%) Ash (%) 13 and 7.8, respectively, while the corresponding values for
Beef lean 67.8a 11.8c 17.9a 0.8a fat were 3.5 and 2.1, respectively). This results in a much
50:50 mixture 48.2b 36.1b 13.7b 0.6b higher power dissipation in fat and an increased tendency
Beef fat 26.3c 65.7a 7.8c 0.3c towards selective heating and a non-uniform temperature
SE 0.67 1.26 0.38 0.03
distribution. Such a trend can be seen by comparing tem-
a,b,c
Means in the same column with unlike letters are different (P < 0:05). perature distributions between Fig. 6a and c.
494 K.W. Farag et al. / Meat Science 80 (2008) 488–495

3.3. Conventional tempering 1 -1 0 -1


2 -2 -2
3 -3 -3
3.3.1. Tempering time

Depth (cm)
4 -4
Lean, 50:50 mixture and fat were tempered in 5 h -5 -5 -4
5
22 min, 5 h 48 min and 4 h 56 min, respectively, as shown 6
in Fig. 7, these times corresponding to the attainment of 7
an MnT within each block of 5 °C. Meanwhile, xT was 8 -3
-2 -4 -3
3.4, 3.1 and 3.1 °C while MxT was 2.2, 1.7 and 9
6 8 10 12 14 16 18 20 22
1.4 °C, respectively, for these blends.
Diagonal length (cm)

3.3.2. Effect of composition


An effect of composition on temperature distribution 40.0
was observed for conventionally tempered meat, with
higher fat content producing wider DT within the tempered 34.4
block. For instance, DT was 2.8, 3.3 and 6.4 °C, for lean,
50:50 mixture and fat, respectively. This is reflected in
Fig. 6 with a greater number of contour lines in the fat 28.8
compared to 50:50 mixture or lean.
50.4˚C
3.4. MW tempering 23.2

Fig. 8a shows the end point temperature distribution


profile of MW tempered lean beef at 400 W for 14 min. 17.6
While an  xT value of 2.5 °C fell well within the target
range, MxT was 1.2 °C suggesting a tendency to tempera- 12.0
ture runaway within the block. When an IR snapshot
was taken of the block surface evidence of overheating Fig. 8. MW tempering of lean at 400 W –14 min. (a) End point
temperature distribution of tempered block and (b) surface temperature
was unmistakable, with a hot spot of 50.4 °C being
using IR camera.
observed (Fig. 8b). However, it must be noted that such
temperatures were not detected by thermocouples inserted
1 cm below the surface. Thus, MW heating gives unsatis- 3.5. Overall comparison between the tempering methods
factory results which are in agreement with data of Bengts-
son (1963) who was unable to satisfactorily defrost meat or 3.5.1. Temperature profile
fish at 2450 MHz when the thickness was greater than Figs. 6 and 8 show the variations in temperature profiles
4 cm. It must be borne in mind that this frequency is gen- between the different tempering methods used. During con-
erally not encountered in industrial scale equipment, which ventional tempering of meats, the outer surfaces of the
uses a frequency of 896 (e.g., APV Baker Ltd.) or 915 MHz product heat first with heat subsequently transferring to
(e.g., Sairem). the colder interior mainly by conduction. By contrast,
dielectric technology (RF and MW) allows, in principle,
-2 all parts of the product to heat at the same rate depending
-4
on its homogeneity. However, in practice, the frequency of
the incident radiation limits the depth to which it can pen-
-6
etrate the product. RF is known to have a greater penetra-
tion depth than MW (Bengtsson, 1963). For instance, MW
Temperature (ºC)

-8

-10
penetration depth was 1.7 cm for beef lean similar to that
used in the present work, compared to RF which gave a
-12 value of 34.2 cm at 5 °C (Farag, Lyng, Morgan, & Cro-
-14 nin, 2008).
-16
3.5.2. Tempering time
-18 Fig. 7 shows a time–temperature profile of the beef meat
-20 blends. The time required to achieve conventional temper-
00:00:00 02:00:00 04:00:00 06:00:00 ing was 5 h 22 min, 5 h 48 min and 4 h 56 min, for lean,
Time (hr:min:sec) 50:50 mixture and fat, respectively. On the other hand,
Fig. 7. RF tempering times of lean (d), 50:50 mixture (j) and fat (N) RF tempering took 11 min for all three samples which rep-
versus conventional tempering times of lean (—), 50:50 mixture (—) and resented an approximately 30 fold reduction in tempering
fat (). time. This is in agreement with previous studies where
K.W. Farag et al. / Meat Science 80 (2008) 488–495 495

defrosting times were reduced from hours by conventional Measure (FIRM), directed by the Irish Department of
methods to minutes using RF technology (Cathcart, Par- Agriculture, Fisheries and Food.
ker, & Beattie 1947; Pizza et al., 1997).

3.5.3. Power consumption References


At the scale employed in the current study, conventional
tempering required 1.7 kW h to achieve satisfactory tem- Bengtsson, N. (1963). Electronic defrosting of meat and fish at 35 and
pering whereas RF required only 0.2 kW h to attain the 2450 MHz – a laboratory comparison. Food Technology, 17(10),
1309–1312.
same result, i.e., a ninefold reduction relative to the con-
Bengtsson, N. E., Melin, J., Remi, K., & Söderlind, S. (1963). Measure-
ventional method. ments of the dielectric properties of frozen and defrosted meat and fish
It must be noted that the RF oven used in this study was in the frequency range 10–200 MHz. Journal of the Science of Food and
made out of three components: (a) the RF generator, (b) an Agriculture, 14, 592–604.
impedance matching box and (c) a matching box control- Buffler, C. H. (1993). Dielectric properties of foods and microwave
materials. In Microwave cooking and processing (pp. 46–69). New
ler. The total power drawn was 0.2 kW h when the unit
York, USA: Van Nostrand Reinhold.
was operating at 600 W for 10 min (equivalent to Cathcart, W. H., & Parker, J. J. (1946). Defrosting frozen foods by high-
0.1 kW h rating for the RF generator with the balance of frequency heat. Food Research, 11, 341–344.
the power being consumed by the impedance matching Cathcart, W. H., Parker, J. J., & Beattie, H. G. (1947). The treatment of
box and controller). packed bread with high frequency heat. Food Technology, 1(2),
174–177.
Farag, K., Lyng, J. G., Morgan, D. J., & Cronin, D. A. (2008). Dielectric
4. Conclusion and thermophysical properties of different meat beef blends over a
temperature range of 18 to +10 °C. Meat Science, 79(4), 740–747.
RF tempering protocols were developed for defrosting Houben, J., Schoenmakers, L., van Putten, E., van Roon, P., & Krol, B.
comminuted beef meat blends, to an average temperature (1991). Radio-frequency pasteurization of sausage emulsions as a
continuous process. Journal of Microwave Power and Electromagnetic
in the range 2 to 5 °C, in a highly reproducible manner,
Energy, 26(4), 202–205.
and with an impressive 30 fold reduction in tempering time James, S. J., & James, C. (2002). Thawing and tempering. In Meat
when compared to conventional methods. Results also Refrigeration (pp. 159–187). Cambridge: Woodhead Publishing.
indicated that defrosting meat with RF heating can result Jason, A. C., & Sanders, H. R. (1962). Dielectric thawing of fish. Food
in similarly uniform end point temperature distribution Technology, 16(6), 101–112.
Lind, I., & Hulthen, B. (1986). Thawing of food in catering. Foodservice
profiles compared to conventional air tempering. In addi-
Research International, 4(2), 81–96.
tion, the RF process was more energy efficient at the scale Pizza, A., Pedrielli, R., Busetto, M., Bocchi, M., & Spinelli, R. (1997). Use
employed in this study. These findings will be of interest to of radiofrequencies in the meat processing industry. Effects on the
the meat industry. quality characteristics of meat and cooked meat products. Industria
Conserve, 72, 122–133.
Sanders, H. R. (1966). Dielectric thawing of meat and meat products.
Acknowledgments
Journal of Food Technology, 1, 183–192.
Yun, C. G., Lee, D. H., & Park, J. (1998). Ohmic thawing of a frozen
The authors wish to acknowledge the financial support meat chunk. Korean Journal of Food Science and Technology, 30(4),
of the Non-Commissioned Food Institutional Research 842–847.

S-ar putea să vă placă și