close

Вход

Забыли?

вход по аккаунту

?

Thermal processing of hermetically packaged low -acid foods using microwave -circulated water combination (MCWC) heating technology

код для вставкиСкачать
THERMAL PROCESSING OF HERMETICALLY PACKAGED LOW-ACID FOODS
USING MICROWAVE-CIRCULATED WATER COMBINATION (MCWC)
HEATING TECHNOLOGY
By
DONGSHENG GUAN
A dissertation submitted in partial fulfillment of
the requirements for the degree of
DOCTOR OF PHILOSOPHY
WASHINGTON STATE UNIVERSITY
Program in Engineering Science
DECEMBER 2003
© Copyright by DONGSHENG GUAN, 2003
All Rights Reserved
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI N um ber: 3133153
Copyright 2003 by
Guan, Dongsheng
All rights reserved.
INFORMATION TO USERS
The quality of this reproduction is dependent upon the quality of the copy
submitted. Broken or indistinct print, colored or poor quality illustrations and
photographs, print bleed-through, substandard margins, and improper
alignment can adversely affect reproduction.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if unauthorized
copyright material had to be removed, a note will indicate the deletion.
®
UMI
UMI Microform 3133153
Copyright 2004 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
© Copyright by DONGSHENG GUAN, 2003
All Mights Reserved
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To the Faculty o f Washington State University:
The members o f the Committee appointed to examine the dissertation of
DONGSHENG GUAN find it satisfactory and recommend that it be accepted.
Chair
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENT
I would like to acknowledge the great support from my major advisor, Dr. Juming
Tang, for his guidance, motivation and encouragement throughout the course o f my study
and research at Washington State University. Without him, I would never have been able to
face the challenges from the world of microwaves. I appreciated very much my doctoral
committee members, Dr. Denny Davis, Dr. Barry Swanson and Dr. Dong-Hyun Kang, for
their support, advice and guidance during the course of my research and in the preparation of
this dissertation. Special thanks go to Dr. Stephanie Clark, Ms. Virginia Plotka and Mr. Peter
Gray for assistance in different phases of my research.
I would like to thank my former fellow graduate students Drs. Yifen Wang and Ming
Lau for their support, assistance and encouragement. I would like to thank the microwave
sterilization project members, Drs. Fang Liu, Vyacheslav Komarov and Surya Pathak and
fellow graduate student Ram Pandit for their assistance to the final part of my dissertation. I
am thankful to Mr. Frank Younce, Mr. Wayne Dewitt and Mr. Vincent Himsl for their
technical assistances.
This project was financially supported by the US Army Soldier Center, Natick,
Massachusetts, with additional support from Kraft Foods, Glenview, Illinois, and from
Washington State University Agricultural Research Center.
Their support is greatly
appreciated. I thank the contribution of expertise from other government, institution and
industrial partners. The inputs of Dr. Tom C. S. Yang (US Army Natick labs), Dr. Keith Ito
and Mr. Bradley Shafer (National Food Processors Association, NFPA) and Mr. Evan Turek
(Kraft foods) were of great value in guiding this pursuit.
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Finally, I would like to thank the supports from my family, especially my wife,
Mingfaua and my son, Steven. Without the encouragement and understanding from them, I
could not have completed my study.
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
THERMAL PROCESSING OF HERMETICALLY PACKAGED LOW-ACID FOODS
USING MICROWAVE-CIRCULATED WATER COMBINATION (MCWC)
HEATING TECHNOLOGY
Abstract
by Dongsheng Guan, PhJD.
Washington State University
December 2003
Chair: Juming Tang
The feasibility o f sterilizing packaged foods using a 915 MHz Microwave-Circulated
Water Combination (MCWC) heating technology was studied with a pilot-scale microwave
heating system. This system combines microwave energy with hot water (123~125°C) and
can heat food products up to 121.1°C or above. The products are held at high temperature to
reach desired degrees of sterilization (F0 value). The MCWC heating system shortens the
overall processing time and provides microbiologically safe products with acceptable sensory
quality.
For microbial challenge studies, thermal resistances of PA 3679 spores in a macaroni
and cheese entree were determined using Thermal Death Time (TDT) mini-retorts.
Inoculated pack studies by Count-reduction and End-point Methods were conducted. The
results indicated that the MCWC heating system can process products with designed degrees
o f sterilization (Fq value). Sensory qualities of macaroni and cheese entree processed by
MCWC heating system were compared with freshly cooked macaroni and cheese entree.
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The MCWC heat sterilized entrees received higher scores than a popular boxed macaroni and
cheese entree.
To better understand the interaction between foods and microwaves, dielectric
properties o f a mashed potato entree were measured using open-ended coaxial-line probe
technique over frequencies between 1 MHz and 1800 MHz and temperatures between 20 and
120°C. Influence of moisture and salt on dielectric properties was considered. Descriptive
equations at 27 MHz and 915 MHz were developed; and penetration depths were calculated.
The results indicated that the dielectric constant and loss factor of mashed potato decreased
with frequency. The dielectric constant increased with temperature at 27 MHz, changed little
at 40 MHz and decreased with temperature at 433, 915 and 1800 MHz. Dielectric loss factor
■#
increased with temperature. Salt increased the dielectric loss factor, especially at 27 and 40
MHz. Power penetration depth decreased with temperature and frequency.
Heating patterns o f the mashed potato entree processed by the newly-developed 915
MHz single mode MCWC heating system were determined using an infrared thermal image
method.
Computer simulation results using the Finite-Difference Time-domain (FDTD)
method were compared with those patterns experimentally obtained. The results indicated
that heating patterns of mashed potatoes were repeatable and predictable.
The study demonstrates that 915 MHz MCWC heating system has potential to
sterilize food products while maintaining a relatively good sensory quality; the newlydeveloped 915 MHz single mode MCWC heating system can provide predictable and
repeatable heating pattern.
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
Acknowledgements.......................................................................................................... ill
Abstract.................................
v
Table of Contents................................................................................................................. vii
List of Tables
........ siii
List of Figures ...................................... Beasaaaetaeeeee&oooeoeeaaoooffieseseastsosssaeossB'eseaeaaes..................... XV
Dedication
.....
xix
Dissertation Outline ......................................................................................................... 1
Chapter 1. Thermal Processing of Hermetically Packaged Low-Acid Fi
Introduction..
........
Thermal Processing o f Low Acid Foods
3
.....
4
Evaluation of Thermal Processing....................................
....6
Microbiological Consideration............................
.....7
F-value and c-value
......
...7
High Temperature Short Time (HTST) Processing
..........
................9
Challenges for Conventional Thermal Processes .....
Affecting Factors
..10
......
10
Location of Cold Spots
......
..............10
Challenges from Engineering Aspects...............................................................................11
Optimization of Thermal Process
.......... ........................11
References..........................
......12
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C hapter 2. Microwave Heating Fundamentals and Applications
introduction.......
.....
...................................................................................14
Electromagnetic Radiation (EMR) and Spectrum .............................................................15
Generation of Microwaves Using a Magnetron.....................................
.....17
Transmission of Microwave Powering Using Waveguides...............................................19
Waveguides
...........
...19
Transmission of Microwaves and Energy.....
Propagation M odes
......
........20
.......
................................21
Energy Injection and Removal to and from Waveguides ..................................................23
Design and Construction of the Waveguides...
......
.......26
Absorption o f Microwave Energy................
26
Dielectric Properties of Materials.................................................
...26
.....
.29
Affecting Factors for Dielectric Properties of Foods....
Microwave Heat Generation and Absorption........
Penetration Depth
.......
.....
...33
Power Measurement at Microwave Frequencies
Power Measurement and Methods
.....
...35
.....
....35
Microwave Power M eters
.....
Temperature Measurement in Microwave Heating
Temperature Scales
.31
.....
37
..42
......
........42
Measurement Principles o f Temperature Sensors.............................
42
Characteristics o f Fiber-optic Temperature Sensors....
.....
.....49
Use of Microwaves in Packaged Food Sterilization
.....
.....51
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Challenges in Development of Microwave Sterilizaion Technology....
.......54
Final Remarks.............................
References..........
C hapter
3.
..................57
.....
Microbiological
59
Validation
of
M icrowave-Circulated
Water
Combination (MCWC) Heating Technology by Inoculated Pack Studies
.....
Abstract
Introduction......
..........................65
......
Materials and Methods......
66
...............
................67
Preparation of Macaroni and Cheese Entree........................
..................68
Heat Resistance of PA 3679 in Neutral Phosphate Buffer (pH 7.0)........
68
Heat Resistance o f PA 3679 in Macaroni and Cheese Entree (pH 5.7) ........
69
MCWC Heating System
.....
.......70
Temperature Measurement during MCWC Heating.......
Inoculated Pack Tests
Count-reduction Method..
.71
.....
Package Integrity and Sealing of Products
MCWC Heating Process Procedures
.....
...72
.....
...73
......
....75
........
...75
End-point Method..................
75
Results and Discussion
.......
.....76
Thermal Resistance of the PA 3679 Spores.......................................................................76
MCWC Processing and Integrity of Packages...................
Inoculated Pack Studies..
......
.......77
...................80
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Count-reduction Method
80
End-point Method.............................................
......80
Conclusions........................................................................................................................82
Acknowledgements
.......
......82
References...................
...83
C hapter 4. Sensory Evaluation of M acaroni and Cheese Entrees Processed with a
M icrowave-Circulated W ater Combination (MCWC) Heating Technology
Abstract...
......
...86
.....87
Introduction................................................................................
Materials and Methods.....................
.................................88
Preparation of Macaroni and Cheese Entree forProcessing............. ..................................88
Evaluated Samples..........................
................89
MCWC Heating System .....
...90
MCWC and Conventional Heating Procedure...
.....
....91
Sensory Evaluation Studies................................................................. ........................... ......93
Statiscal Analysis..
.....
Results and Discussion
......95
......
......95
Effect of MCWC and Conventional RetortHeating on Processing Time..........................95
......91
Effect of Product Formulation on Heating History...
Sensory Evaluation
Trained Panel
.....
....................................99
......
....................................99
Consumer Panel................
............................100
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Conclusions...
..............
.103
Acknowledgements..............
.......................................................................................104
References.
.......
...........104
C hapter 5. Dielectric Properties of M ashed Potatoes Relevant to Microwave and
Radio Frequency Pasteurization and Sterilization Processes
Abstract...
................
107
Introduction..............
...108
Materials and Methods.....................................................
............110
Mashed Potato Composition Determination and Preparation..............
110
Dielectric Properties Measurement System.......................
112
Measurement of Dielectric Properties ..............
113
Descriptive Equations Development and Calculation of Power Penetration D epth
.......
...114
Results and Discussion
.....
...........115
Effect of Frequency and Temperature
Effect of Moisture Content
............
......121
.............
Descriptive Equations and Power Penetration Depth
..122
........
......
..126
...130
Acknowledgements
References
115
..........
Effect o f Salt Content (added NaCl)
Conclusions
.
........
............130
.......
........131
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 6. Heating Characteristics of M ashed Potatoes in a 915 MHz Single Mode
Microwave-Circulated Water Combination (MCWC) Heating System
Abstract...................
Introduction...........
135
.....
.136
Materials and Methods.....................................................................................................137
Preparation of Mashed Potato
......
.........137
Dielectric Properties Measurement
MCWC Heating System
..........
...............138
........
138
Temperature Measurement and Data Logging
........
......139
Power Calibration of Microwave Generating System...............
MCWC Heating Procedure and Infrared Thermal Images
.140
.......
.141
FDTD Simulation Method and Quick Wave (QW-3D) Software Simulation................ .142
Results and Discussion
........
147
Dielectric Properties of Mashed Potato at 915 M H z...............
.147
Infrared Thermal Images
..147
.........
Simulation Results .........
Discussion....
....154
......
Conclusions..
160
........................
Acknowledgements
161
.....
......161
References....................
161
Conclusions..............
...164
Future W ork..........................
167
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Chapter
3.
Microbiological
Validation
of
Microwave-Circulated
W ater
Combination (MCWC) Heating Technology by Inoculated Pack Studies
Table 1: Processing procedures for three processing levels ........... ............. ...............
74
Table 2: Thermal resistance of PA 3679 in phosphate bufferand macaroni cheese entrees
.......
...76
Table 3: Result from count-reduction method ....
........80
....
Table 4: Results from end-point method
.......................81
C hapter 4. Sensory Evaluation of M acaroni and Cheese Entrees Processed with a
M icrowave-Circulated W ater Combination (MCWC) Heating Technology
Table 1: Recipes for macaroni and cheese entree prepared for sensory evaluation
90
Table 2: Mean scores of selected macaroni and cheese entree quality attributes judged by
a trained panel (N=10)
......
.98
Table 3. Mean acceptability scores of macaroni and cheese entrees by an untrained
consumer panel (N =115)
.......
..101
Chapter 5. Dielectric Properties of M ashed Potatoes as Relevant to Microwave and
Radio Frequency Pasteurization and Sterilization Processes
Table 1. Mean ± standard deviation (three replicates) of dielectric properties for mashed
potatoes with different moisture content (%, w.b.), no salt added...................................116
xiii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2. Mean ± standard deviation (three replicates) of dielectric properties for mashed
potatoes with different salt levels (mass ratio), moisture content: 85.9% (w.b.)......... 123-4
Table 3. Descriptive equations for the dielectric properties of mashed potatoes ...........127
Table 4. Power penetration depth (mm) for mashed potatoes at different moisture content
(w.b.), no salt added)
.........
............128
Table 5. Power penetration depth (mm) for mashed potatoes (moisture content: 85.9%,
w.b.) at different salt contents and frequencies......................................................... .........129
C hapter 6. Heating Characteristics of M ashed Potatoes in a 915 MHz MicrowaveCirculated W ater Combination (MCWC) Heating System)
Table 1. Dielectric properties (dielectric constant/loss factor) of mashed potatoes vs.
moisture content and added salt at 915 MHz
Table 2.
.....
..147
Field distribution (Max/min ratio) from simulation for products at three
temperatures (2% added salt) ..........................................................................................158
Table 3. Field distribution (Max/min ratio) from simulation for products with four added
salt concentrations (60°C )
.....
...................158
xiv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
C hapter 1. T hermal Processing of Hermetically Packaged Low-Acid Foods
Figure 1. Classification of foods by pH scale....
................
..................... 5
C hapter 2. Microwave Heating Fundamentals and Applications
Figure 1. Electromagnetic waves propagate in the form of sinusoidal waves
............15
Figure 2. Electromagnetic spectrum-a family of waves (Frequency: Hz; Wavelength: m)
................
..17
Figure 3. Magnetrons for 2450 MHz (Matsushita Electric Industrial CO. Ltd.) and 915
MHz (Microdry Model IV-5)............
...............19
Figure 4. Different waveguides: planar, rectangular and circular (Left to Right) ........20
Figure 5. View of cross section of a rectangular waveguide with a TE iq mode......... .....22
Figure 6. Waveguide terminated with a funnel-shaped horn. Left: conical hom; Right:
pyramidal hom
.......
Figure 7. Power measurement using directional coupler
.......
..25
....37
Figure 8. Measurement setup using a flow-type microwave calorimeter (water load) to
calibrate generator output pow er
.....
..................................38
Figure 9. Water loads for microwave power measurement at the frequencies of 2450
MHz (Left, RF Technologies Corporation, Lewiston, Maine) and 915 MHz (Right,
Ferrite Components, Inc., Hudson, NH) at W SU
.......39
Figure 10. Total internal reflection requires all light strikes the boundary at an angle
greater than the critical angle.............................................................................................44
xv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 11. White-light sensing method is used to determine the cavity length (redrawn
by the author and originally from www.fiso.com) ............................................................47
Figure 12.
Increases in the crystal's temperature shifts its transmission spectrum to
higher wavelengths (Temperature unit: °C, from Alexandre, 2001)....................... .......... 48
Chapter
3.
Microbiological
Validation
of
Microwave-Circulated
W ater
Combination (MCWC) Heating Technology by Inoculated Pack Studies
Figure 1. A typical temperature-time heating history for the MCWC heating process
(equivalent to a processing procedure with degree of sterilization o f 2.4).......
Figure 2. Visual observation of package integrity of typical food tray....
...78
..........79
C hapter 4, Sensory Evaluation of M acaroni and Cheese Entrees Processed with a
M icrowave-Circulated W ater Combination (MCWC) Heating Technology
Figure 1. Schematic diagram of the microwave and circulated water control systems.... 92
Figure 2. Typical temperature history of macaroni cheese and circulated water during
..96
MCWC and conventional retort heating..................
Figure 3. Representative temperature profiles of two macaroni cheese entrees during
MCWC treatment.....
.....
...........................97
Chapter 5. Dielectric Properties of M ashed Potatoes as Relevant to Microwave and
Radio Frequency Pasteurization and Sterilization Processes
Figure 1. Open-ended coaxial probe dielectric measurement system at WSU...............112
Figure 2. Change o f dielectric constant of mashed potato (moisture content: 81.6%, w.b.;
xvi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
no added NaCl) with frequency at six temperatures
.... ............... .......... .................117
Figure 3. Change o f dielectric loss factor of mashed potatoes (moisture content: 81.6%,
w.b.; no added NaCl) with frequency at six temperatures...............................................117
Figure 4. Change of dielectric constant of mashed potatoes (moisture content: 85.9%,
w.b.; no added NaCl) with temperature at five frequencies
.....
..118
Figure 5. Change of dielectric loss factor of mashed potatoes (moisture content: 85.9%,
w.b.; no added salt) with temperature at five frequencies ............
120
Figure 6. Loss factor o f mashed potatoes as affected by moisture content (120°C)......126
Figure 7. Added salt and the dielectric loss factor of mashed potatoes (moisture content:
85.9%) at 27 MHz
......
125
Figure 8. Added salt and the dielectric loss factor of mashed potatoes (moisture content:
85.9%) at 915 MHz..........
...125
Chapter 6. Heating Characteristics of Mashed Potatoes in a 915 MHz Single Mode
Microwave and Circulated Water Combination (MCWC) System
Figure 1. Schematic diagram of MCWC heating system at WSU....
Figure 2. Typical heating profile for the MCWC heating system
........
.........
Figure 3. Quickwave simulation set up of the MCWC heating system ....
139
141
145
Figure 4. Thermal image of heated mashed potatoes (initial temperature: 25°C; moisture
content (w.b.): 85.9% ) ......
...149
Figure 5. Thermal images of heated mashed potatoes (initial temperature: 25°C; moisture
content (w.b.): 85.9%; 0.5% added salt)...............
.........150
Figure 6. Thermal images o f heated mashed potatoes (initial temperature: 25°C; moisture
xvii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
content (w.b.): 85.9%; 1% added salt; Left tray)....................
..............................151
Figure 7. Thermal images of heated mashed potatoes (initial temperature:25°C; moisture
content (w.b.): 85.9%; 1% added salt; right tray)................................. ...........................152
Figure 8. Thermal image of heated mashed potatoes (initial temperature: 25°C; moisture
content (w.b.): 85.9%; 2% added salt)..
......
.153
Figure 9. Thermal image o f heated mashed potatoes using 80°C hot water only (initial
temperature: 25°C; moisture content (w.b.): 85.9%; 2% added salt)
......
..154
Figure 10. Power density distribution within the mashed potatoes (initial temperature:
25°C; moisture content (w.b.): 85.9% and added salt: 1.0%).........
............156
Figure 11. Simulated dominant electric field (Ey) distribution within the mashed potato
(initial temperature: 25°C; moisture content (w.b.): 85.9% and added salt: 1%)............ 157
Figure 12. Simulated power density distribution for the mashed potatoes at Middle layer
(initial temperature: 25°C; moisture content (w.b.): 85.9%) .....
xviii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
159
DEDICATION
This dissertation is dedicated to my parents (Mrs. Shuzhen JIA and Mr. Fuhai GUAN), and
my wife (Mrs. Minghua CHENG) and son (Steven D. GUAN)
xix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DISSERTATION OUTLINE
This dissertation is organized into six chapters.
The first and second chapters are
review articles regarding thermal processing of hermetically packaged low-acid foods and
the fundamentals of microwave heating and its application.
Chapter 3 presents the
microbiological validation o f Microwave-Circulated Water Combination (MCWC) heating
technology by inoculated pack studies. The microbiological safety of the processed foods
using the MCWC heating system was demonstrated through inoculated packs. Chapter 4
presents the superior sensory quality of the MCWC heating processed macaroni and cheese
entrees. Chapter 5 presents the determination of the dielectric properties of mashed potato
entrees as affected by frequency, temperature, moisture content and salt (added NaCl)
content. Chapter 6 describes the heating patterns of mashed potato entrees with different
moisture content and salt (added NaCl) content. The results from corresponding computer
simulation were compared with those results obtained experimentally. A conclusion was
made following Chapter 6 and the future work is recommended for the industrialization of
MCWC heating technology in food sterilization field.
Since the dissertation is composed of published and completed manuscripts, the format
of each article follows the style of related journals.
A list of the published chapters and chapters accepted for publication as of September
1, 2003 includes:
Chapter 3
Guan D, Gray P, Kang DH, Tang J, Shafer B, Ito K, Younce F and Yang TCS. 2003.
Microbiological validation of microwave-circulated water combination heating technology
by inoculated pack studies. J. Food Sci. 68 (4): 1428-1432.
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 4
Guan D, Plotka VCF, Clark S and Tang J. 2002. Sensory evaluation of microwave
treated macaroni and cheese. J. Food Process. Preserv. 26: 307-322.
C hapter 5
Guan D, Cheng M, Wang Y and Tang J. 2003. Dielectric properties of mashed potatoes
relevant to microwave and radio frequency pasteurization and sterilization Processes, (in
review).
C hapter 6 was presented at the 2003 ASAE Annual International Meeting, Riviera
Hotel and Convention Center, Las Vegas, Nevada, USA, 27- 30 M y 2003.
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1
THERMAL PROCESSING OF HERMETICALLY PACKAGED LOW-ACID FOODS
INTRODUCTION
Heating processes have been widely used in the food processing industry.
For
conventional heating using steam or hot water, the thermal energy incident on the surface of
a material by conduction, convection, radiation or combinations must be conducted to the
interior of the material in order to heat the entire product. The driving force to transport
thermal energy within the product is the temperature gradient among different parts of the
material. Properties of the product such as specific heat, thermal conductivity, density and
product geometry determine the heat transfer rates. Reliance on temperature gradients in
food products often lead to uneven heating such as overheated surfaces or edges, relatively
long processing times, and undesirable quality.
The heating process in the canning industry is often referred to as thermal processing.
A major task in developing a thermal process is to determine the time-temperature history at
the cold spot(s) in packaged foods and to design a procedure to ensure the microbiological
safety for the processed products with acceptable quality (Holdsworth 1997). Through the
use o f thermal energy, the microorganisms causing both food spoilage and poisoning are
inactivated so that the processed foods with extended shelf-life can then be supplied to
consumers (Fennema 1975; Ramesh 1999).
In conventional thermal processing operations, heat is typically supplied by
condensing steam or pressurized hot water through conduction and convection from the
heating medium to the products. A typical thermal process for sterilizing food products can
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
be divided into three phases:
•
The come-up phase:
In this stage, the packaged product is heated to a desired
temperature at the locations where the product is heated at the slowest rate (also
called cold spots).
•
The holding phase: In this stage, the product’s temperature is held at the desired
temperature until the desired lethality or degree of sterilization value is obtained.
•
The cooling phase: In this stage, the heated product is cooled so that the packaged
foods can be taken out at atmosphere pressure.
An ideal thermal process would consist of instantaneous come-up and cooling times;
all regions o f the product would simultaneously reach the identical temperatures. However,
in reality transport of thermal energy driven by temperature gradients takes time. Many
factors can affect the come-up and cooling times such as the initial temperature of the
product, the mass and type of materials, the geometry of the package, and the processing
conditions. For these reasons, the temperatures throughout the product do not reach identical
temperatures at the same time.
THERMAL PROCESSING OF LOW ACID FOODS
The acidity or pH value of a food is generally used to determine its processing
requirements. Based on pH values, most foods can be divided into two categories, as shown
in Figure 1 (FDA 21 CFR 113).
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Acid foods
0
4.6
4
Low acid foods
6 - 8
10
12
14
Neutral (7.0)
Figure 1: Classification of foods by pH scale
Acid foods are foods that have a natural pH of 4.6 or below. "Natural pH" means the
pH prior to processing. Acidified foods are low acid foods to which acid(s) or acid food(s)
are added. After equilibrium, the products have a water activity (aw) greater than 0.85 and a
pH of 4.6 or below (FDA 21CFR 110). The acid foods and acidified foods include most
fruits, pickles, sauerkraut, jams, jellies, marmalades and fruit butters. The FDA regulation of
21CFR110 applies to the processing of acid foods. It is recommended that foods preserved
by acidity have a pH o f 4.6 or less with proper sanitation and acidification by adding lemon
juice, citric acid, or vinegar. Below 4.6, production of deadly toxins by the organism that
causes botulism is inhibited. To prevent possible spoilage, processors usually heat acidified
foods to 88°C (180°F) and package them hot, killing yeast, mold spores, and vegetative
bacteria from the products and the containers. The product is then placed in cardboard cases
to cool down and prevent "stack bum". In some cases, where certain products should not be
heated, more acid should be added, or chemical preservatives should be used, or both. For
example, sodium benzoate and potassium sorbate are the preservatives commonly used
together to take advantage o f their combined effects. A pH well below 4.6 is formulated to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
provide a safety factor for the acid and acidified foods; and most acidified foods have a pH of
less than 4.0. But proper sanitation and care in manufacturing are required; and the highest
standards of cleanliness and product protection must be adhered to.
Low acid foods are defined as "any foods, other than alcoholic beverages, with a
finished equilibrium pH value greater than 4.6 and a water activity greater than 0.85"
(21CFR 113). They include red meats, seafood, poultry, milk and all fresh vegetables except
for most tomatoes, tomato products and figs.
Low acid foods contain too little acidity to prevent the growth of Cl. botulinum
bacteria. To preserve the low acid foods, they need to be thermally processed, namely, by
receiving adequate heat to destroy the targeted microorganisms, Cl. Botulinum spores. This
is often referred to as a commercial sterilization process. The related processing procedures
are regulated under the low acid foods regulations (21CFR 113). Naturally acid foods and
fermented foods along with jams, jellies, preserves, certain dressings and sauces are exempt
from the provisions o f 21CFR 113. However, if the food contains a mixture of acid and low
acid foods, the regulation applies.
EVALUATION OF THERMAL PROCESSING
Because o f the non-uniformity of conventional heating, the temperature history at the
cold spots inside the products is the most important information used to evaluate the degree
of processing. For example, to ensure the overall microbiological safety of processed foods,
the product material near the wall are usually over-processed (Harlfinger 1992), even though
the processed products will bear little resemblance to the fresh product (Fryer 1997). The
following main consideration summarizes the ways to evaluate a common thermal process.
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(1) Microbiological consideration: Pathogens may produce toxins and infect humans
directly. They must be completely destroyed to ensure the safety of shelf stable foods. In the
field of commercial sterilization of foods, Cl. botulinum is chosen as the target organism
(Stumbo 1973). With knowledge of the Dm.i value for CL botulinum being 0.3 min and a zvalue of 10°C, a 12 D process, also called minimum process or a 'botulinum cook’, is
suggested to establish. The 12 D process aims at a microbial survival rate of one spore in
1012 or less and is adopted as a minimum standard by the canning industry in United States
and United Kingdom; it is also widely used in the other regions of the world.
(2) F-value and c-value: F-value was originally developed as the thermal processing
unit by the National Canners’ Association (Bigelow 1920), and subsequently by others (Ball
and Olson 1957). By measuring the temperature profile at the cold spots for the packaged
products, the cumulative lethality delivered by a thermal process is calculated by the
following relationship:
‘
T~T„
where F0 is the net lethality (min) for the process with a reference temperature T0 of
121.1°C; z value is 10°C for Cl. botulinum and varies with the targeted organisms; and t is
the processing time (min).
The validation of a thermal process can be accomplished through inoculated pack
tests using a surrogate with known Dm., value and z-value to estimate the total integrated Fvalue for the whole container.
Inoculated pack studies are very time consuming and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
expensive. In order to evaluate a certain process properly, the thermal resistance (D-value
and z-value) o f the targeted bacteria in the food product must be correctly measured: and
probability o f survival of microorganisms throughout the whole package, not just at the
center, is considered (Stumbo 1973).
The heat penetration test must be carried out to
determine the heating patterns or temperature distributions throughout the packages under the
particular heating conditions. Then a certain number of targeted bacteria with known thermal
resistance are either evenly distributed or placed at certain locations inside the products. The
inoculated packs are processed with specified procedures; and the survivors of the bacteria
are detected to check that the heating procedures have the sterilization effect as expected.
The concept o f cooking value, or c-value, was developed by FMC Inc., San Jose, CA
(Mansfield 1962). The c-value describes the degree of a treatment on the heat sensitive
factors of interest. The c-value is only an accompanying index for the adequacy of the
process. Definitions of c-value and the heat sensitivity of flavor, color, texture and various
nutrients may vary. The general reference temperature is 100°C and the cook value c (min)
can be given by:
ZzISL
C = 10 z dt
o
/
J
(2)
where t is total processing time; T is the instantaneous temperature of the material; and T0 is
the reference temperature, usually at 100°C; and z describes the incremental change of the
rate of quality degradation in response to a change in temperature and is commonly assigned
to be 33°C (Holdsworth 1997). A large cook value indicates a severe heat abuse of the
products.
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
HIGH TEMPERATURE SHORT TIME (HTST) PROCESSING
High-Temperature, Short-Time (HTST) processing strategies have been developed in
the canning industry, which takes the advantage of difference in activation energy, and
therefore, z-values for microbial inactivation and for thermal degradation of products
qualities.
For example, the z-values for cooking and nutrient degradation are generally
around 25-45°C, much higher than those for the inactivation of bacteria, which are 7-12°C
(Lund 1975 and 1977). As the temperature increases, the degree of cooking, or c-value,
decreases when the sterilizing value remains constant, reducing the adverse effect of the
thermal process on food quality. HTST processes also have the advantage of increasing the
throughput of a canning plant because of the relatively short processing time.
An ideal HTST processing causes less thermal damage to the product than does
conventional thermal processing. HTST processing has been well developed for liquid foods
to reduce adverse thermal degradation while still ensuring food safety. However, this only
applies if there is no heat transfer lag in the product, and consequently does not apply to
larger size containers filled with solid or semi-solid foods. The low thermal conductivity of
foods slows heat transfer from the heated surface to the cold interior (Holdsowth 1997).
Because o f slow heat transfer in solid and semi-solid foods, high temperature processing (>
121.1°C for medium temperature) has also been reported to have an adverse effect on the
nutrient retention for conduction heating products (Stumbo 1973; Fryer 1997; Ramesh 1999).
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHALLENGES FOR CONVENTIONAL THERMAL PROCESSES
(1) Affecting factors: Many factors can affect heat penetration and thermal process
severity, including retort temperature profile, process time, heat transfer medium, container
agitation, product consistency, initial temperature, initial spore load, thermo-physical
properties o f the food product, acidity (pH), additives, container materials, and container
shape dimensions (Holdsowth 1997).
The rates of food quality degradation and
microbiological lethality increase exponentially with temperature. All these can influence
the establishment of process schedules required to ensure the safety of all thermally
processed packaged foods.
(2) Location o f cold spots: To ensure the adequacy of a thermal process, determining
the heating pattern as well as the cold spot is paramount. But all the factors mentioned in the
above subsection also affect the locations that experience relatively slow heating rate,
referred to as the cold spots. Furthermore, the cold spot locations may move under different
circumstances. For example, the place close to the geometrical center of a food package is
usually the slowest heating point during conventional thermal processing; but this is not
necessarily true during cooling phase (Hurwicz and Tischer 1955). The cold spot was also
reported to shift during the processing for liquid foods or foods that undergo phase change
such as starch gelatinization (Datta and Teixerira 1988). With the help of model systems,
heat transfer in containers can be studied to obtain thermal sensor correction factors and
demonstrate the various modes of heat transfer (Robertson and Miller 1984).
A total
integrated or mass-averaged value (referred to as Fs) for the entire contents of a container
had been introduced to compensate for those problems arising from the unpredictability of
the cold spots (Stumbo 1973).
In practice, determining the location for cold spots is a
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
challenging step, however.
(3) Challenges from engineering aspects: Many factors can influence the design,
operation and choice o f sterilizing system for packaged foods. These factors include, but are
not limited to, the product characteristics, the nature and form of the packaging materials,
and the product throughout. For example, since a plastic container can not withstand the
differential pressures imposed by steam heating; pressurized hot water processing and steamair mixtures were developed.
To increase the heat transfer coefficient affected by the
velocity of the hot water and the way it impacts the containers, adequate pumping, consistent
change of can orientation and agitation in the retorting system must be considered (Peterson
and Admas 1983). To place pouches or semi-rigid plastic containers in the retort, rigid
support systems are used to ensure a strict geometrical arrangement.
(4) Optimization of thermal process: Optimization of the balance between food
safety and the loss of certain qualities such as vitamins, protein and colors has always been
one of the major challenges for food thermal processors (Lund 1975). Microbiological safety
is always the priority for thermal processing. However, how to minimize the quality losses
under the conditions of food safety is still an interesting yet challenging topic that requires
further research.
The challenges for every thermal process are unique. A reasonable way to deal with
these challenges is by having a good understanding of the specific thermal process, the
processing requirement, and careful plan as well as execution.
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCE:
Ball CO and Olson FCW. 1957. Sterilizaiton in food technology-theory, practice and
calculations. McGraw-Hill. New York.
Beigelow WD, Bohart GS, Richardson AC and Ball CO. 1920. Heat penetration in
processing canned food. Bulletin NO. 16 L. National Canners’ Association. Washington, DC.
Datta AK and Teixerira AA. 1988. Numerically predicted transient temperature and
velocity profiles during natural convection heating in canned liquid foods. J. Food Sci. Vol.
53. No. 1: 191-195.
FDA. 1977. Code o f Federal Regulations. Title 21. Food and Drugs. Food and Drug
Administration, Government Services Administration. Washington, DC.
Fennema OR. 1975. Introduction to food preservation. In Principles o f Food Science,
Part II - Physical Principles o f Food Preservation, M. Karel, O.R. Fennema and D.B. Lund,
editors, pg. 1-7. Marcel Dekker, Inc., New York, NY.
Fryer PJ. 1997. Thermal treatment of foods. In Chemical Engineering fo r the Food
Industry, P.J. Fryer, D.L. Pyle and C.D. Rielly, editors, pg. 331-382, Blackie Academic and
Professional, London, England.
Harlfinger L. 1992. Microwave sterilization. Food Technology 46(12):57-60.
Holdsworth SD. 1997. Thermal Processing o f Packaged Foods. Blackie Academic
and Professional. London. UK.
Hurwicz H and Tischer RG. 1952. Heat processing of beef. I. A theoretical
consideration o f the distribution of temperature with time and in space during processing.
Food Res. 17: 380-392.
Lund DB. 1975. Heat transfer in foods. In Principles o f Food Science, Part II -
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Physical Principles o f Food Preservation. M. Karel, O.R. Fennema and D.B. Lund, editors,
pg. 1-7. Marcel Dekker, Inc. New York.
Lund DB. 1977. Maximizing nutrition retention. Food Technol. 30:71-78
Mansfield T. 1962. High temperature short time sterilization. Proc. 1st Int. Congress
Food Sci. Technol. No. 4: 311-316.
Peterson WR and Admas JP. 1983. Water velocity and effect on heat penetration
parameters during Institutional size retort pouch processing. J. Food Sci. 48: 457-459, 464.
Ramesh MN. 1999. Food preservation by heat treatment. In Handbook o f Food
Preservation, M.S. Rahman (editor). Page: 95-172. Marcel Dekker, Inc. New York.
Robertson GL and Miller SL. 1984. Uncertainties associated with the estimation of
the F0 values in cans which heat by conduction. Food Technol. UK. 19: 623-630.
Stumbo CR. 1973. Thermobacteriology in Food Processing, 2nd ed. Academic Press.
London.
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2
MICROWAVE HEATING FUNDAMENTALS AND APPLICATIONS
INTRODUCTION
Microwave heating refers to the use of electromagnetic waves at frequencies between
0.3 GHz and 300 GHz to generate heat in a dielectric material (Metaxas 1996; Metaxas and
Meredith 1988; Roussy and Pearce 1995; Decareau 1985). The heating of a material with
microwaves results from complex interaction between the electromagnetic field and the
dielectric materials. When electromagnetic fields are alternated and applied to dielectric
material, particles within the dielectric material can be displaced from their equilibrium
positions; this phenomenon is known as polarization.
There are two different types of
polarizations: induced polarization and orientation polarization. Induced polarization takes
place when electrons are displaced from their nuclei (electronic polarization), or when atomic
nuclei themselves are displaced because of an unequal distribution of charge in the molecule
formation (atomic polarization). Orientation polarization occurs in dielectric materials that
have permanent dipoles, such as water, which tend to reorient in response to an external
electric field (Curtis 1999). In microwave applications where high frequencies are used,
orientation polarization is dominant; the strong inter-atomic bonds hinder the dipole rotation,
causing volumetric heat generation.
Microwave heating has several advantages when compared with conventional
heating.
Microwaves can provide volumetric and accelerated heating when directly
coupled to a material.
Microwave heating can potentially reduce surface temperature,
decrease thermal gradients, and improve product quality and uniformity by shortening
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
processing time. Microwave heating is also relatively easy to operate and control. This
chapter presents information regarding the fundamentals of microwave heating, microwave
power and temperature measurement and microwave heating applications in food
industries, particularly in the area of food sterilization.
ELECTROM AGNETIC RADIATION (EMR) AND SPECTRUM
Electromagnetic radiation is the result of mutually induced or coupled oscillating
electric and magnetic fields and propagates in the form of sinusoidal waves in the space
(Figure 1).
E field
y
H field
Figure 1: Electromagnetic waves propagate in the form of sinusoidal waves
According to the theory of electromagnetism developed by James C. Maxwell around
1873 (Ulaby 1999), the circulation of the magnetic field strength around a closed contour is
equal to the net current passing through the surface enclosed by the contour, namely, a
current is surrounded by a magnetic field (Ampere’s law); the circulation of the electric field
strength around a closed contour is determined by the rate of change of the magnetic flux
thorough the surface enclosed by the contour (Faraday’s law); net magnetic flux out of a
region must be zero and net electric flux is related to the charges contained within the region
(Gauss’ law). Maxwell’s theory can be given in the form of differential equations:
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
VxH =J H
dt
(Ampere’s law)
1)
V x E = - ~ (Faraday’s law)
dt
(2)
V mB = 0 (Gauss’ magnetic law)
(3)
V • D = p v (Gauss’ electric law)
(4)
where E (Volt/m) and D (C m2) are electric field vectors interrelated by D = s E ,with s being
the
electricpermittivity of the material (Faraday/m); B (Wb/m2) and H (Ampere/m) are
magnetic field vectors interrelated by B = p H , with p being the magnetic permeability of the
material (Henry/m); p v is the electric charge density per unit volume (C/m3); and J is the
current density per unit area (A/m2). These equations hold in any material, including free
space (vacuum) and at any spatial location. Because of the bilateral coupling between the
electric and magnetic field quantities, the electromagnetic waves are generated and can
propagate in the space.
Microwaves commonly used for heating are a part of the Electromagnetic Spectrum,
a continuum of all electromagnetic waves arranged according to frequency and wavelength
(Figure 2).
This spectrum includes not only microwave (only a small portion of this
spectrum) but also radio, infrared, visible light, ultraviolet, x and gamma waves.
electromagnetic energy passes through space at the speed of light (3 x 108m/sec).
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
All
915 - 2450 MHz
Visible light (red-blue),Ultra-violet
/ '
(4 x 10-7-10-8 )
Wavelength
mo3
i
Radio
w aves
10s
io-3-?x io -7 7* 10-7- 4 x l# 7
Micro­ waves Infrared
m
108
1010
IQ’8 - 10-12
X-rays
IQ'1
Gamma rays
MINI
1013 1015 io 17
io20
1022
to25
Figure 2: Electromagnetic spectrum-a family of waves (Frequency: Hz; Wavelength: m)
GENERATION OF MICROWAVES USING A MAGNETRON
Microwaves are generated by a magnetron or a klystron. But most industrial and
domestic microwave ovens are using magnetrons to generate microwaves. A magnetron is
basically a vacuum tube in which electron current is controlled by magnetic fields. The first
magnetron was designed by Albert W. Hull of General Electric Co. (NJ, USA) in 1921; later
the magnetron became an efficient source to generate microwaves (www.luminet.net).
A magnetron generally functions as a self-contained microwave oscillator that
converts electrical energy to microwave radiation and produces the high-power output. The
major parts that make up the magnetron are listed below (Meredith 1998):
(1)
The diode vacuum tube, a semi conductor having two terminals and allowing
the current to flow in one direction only;
(2)
The cathode constructed of a high-emission material, also called filament or
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
heater;
(3)
Coaxial anode, a hollow cylinder of iron from which an even number of anode
vanes extend inward, also called plate;
(4)
Permanent ring magnets mounted onto the two sides of the vacuum tube;
(5)
Resonance cavities, the open trapezoidal shaped areas between each of the
vanes that serve as tuned circuits and determine the output frequency of the
tube;
(6)
Antenna, a probe or loop that is connected to the anode and extends into one
of the tuned cavities.
When coupled to the waveguide, an antenna can
transmit the microwave energy to the outside.
To generate microwaves, a transformer first changes the incoming alternating current
(AC) line voltage (110V or 480 V) to the required higher levels (the order of thousand Volts):,
a capacitor, in combination with a diode (together called a rectifier), converts the AC current
to direct current (DC). When the direct current with a high potential difference (the order of
thousand Volts) runs through the cathode located in the center, electrons are repelled and
emitted from the heated cathode, and then attracted by the positively charged anode
surrounding the cathode. Because of the permanent magnets, the electrons take a circular
path instead of a straight one (Lorentz's force law.). As the electrons pass by resonating
cavities, a continuous pulsating magnetic field, or electromagnetic radiation is generated.
The frequency o f the electromagnetic wave is determined by the time it takes for the
electrons to travel from the cathode toward the anode and back again. When the highvelocity electrons return to strike the cathode, most of the output power is provided during
the emission of the large number of electrons.
Through an antenna and waveguides,
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
microwave energy is transferred to a load of interest. The magnetron can be cooled by air
(e.g., a domestic 2450 MHz microwave oven), circulated water (e.g., a 915 MHz industrial
microwave operating system), or combination during operation.
Figure 3: Magnetrons for 2450 MHz (M atsushita Electric Industrial CO. Ltd.,
the antenna is on the top) and 915 MHz (Microdry Model IV-5, the antenna is in the
launcher)
TRANSMISSION OF MICROWAVE POW ER USING WAVEGUIDES
Microwave transmission is based on the propagation of electromagnetic fields due to
its high frequency. This Is different from the conventional circuit theory that is based on
voltages and currents. For example, two-wire transmission lines are used in conventional
circuits, however, at high frequencies, transmission lines become lossy and inefficient due to
the skin effect, so coaxial lines and waveguides are used Instead to transfer electromagnetic
energy.
Waveguides: Waveguides are essentially coaxial lines without center conductors;
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
they are constructed from conductive materials.
Waveguides can be planar, rectangular,
circular, etc. in shape (Figure 4).
Figure 4: Different waveguides: planar, rectangular and circular (left to right).
Waveguides have several advantages over two-wire and coaxial transmission lines
(Balanis 1989; Roussy and others 1995; Ulaby 1999).
The large surface area of the
waveguides greatly reduces copper losses (or F'R losses); the dielectric losses, caused by the
heating of insulation between the conductors, are also much lower in waveguides than in
two-wire or coaxial transmission lines; the waveguides can handle more power than coaxial
lines of the same size.
However, waveguides are only practical for use at microwave
frequencies because of the requirements of physical dimensions, which will be discussed
later; waveguides are also relatively difficult to install.
Transmission o f microwaves and energy:
According to the theory of Poynting
vector (Ulaby 1999), electromagnetic energy transmitted through waveguides consists of an
electric field (E-field) and a magnetic field (H-field) that are at right angles (90 degrees) to
each other and at right angles to the direction of propagation. The average power transmitted
by any electromagnetic wave is given by:
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
S „ = ^ « 4 e x H-]
(S)
where Savis the average power (W/m2); E is the electric field intensity (Volt/m) and H* is
the magnetic field intensity (Ampere/m).
Two boundary conditions also have to be satisfied for energy propagation in a
waveguide (Balanis 1989; Ulaby 1999):
(1) The electric field at the surface of a conductor must be perpendicular to the
conductor.
(2) A varying magnetic field must form closed loops in parallel with the
conductors and be perpendicular to the electric field.
In reality both E field and H field always exist at the same time in a waveguide; if a
system satisfies one o f these boundary conditions, it must also satisfy the other since neither
field can exist alone.
Propagation modes:
A magnetron can give rise to a number of possible field
distributions; each type of field arrangement represents a mode when conveying microwaves
from the magnetron to the load.
Three major modes exist: transverse electromagnetic
(TEM); transverse electric (TE) and transverse magnetic (TM) (Balanis 1989).
(1)
TEM mode: the field components are all perpendicular to the direction of
propagation.
(2)
TE mode: the entire electric field is in the transverse plane, perpendicular to
the direction o f propagation.
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(3)
TM mode: the entire magnetic field is in the transverse plane, perpendicular to
the direction of propagation.
A rectangular waveguide cannot support a TEM mode since TEM waves have
transverse variations, which cannot exist inside a region bounded by a single conductor. The
modes of field patterns are described by using subscripts.
For example, in rectangular
waveguides, a TEmn mode indicates the numbers of half-wave patterns in the "a" dimension
(width) and "b" dimension (height) are m and n, respectively (Figure 5).
b
E-field
a
Figure 5: View of cross section of a rectangular waveguide with a TEjo mode.
A given waveguide can only operate above a certain frequency, called cutoff
frequency (fc). Below the cutoff frequency, the wavefronts will be reflected back and forth
across the guide and set up standing waves; therefore no energy will be transmitted along the
waveguide. TEio mode is the most commonly used field pattern for a rectangular waveguide
(Figure 5) and referred to as the dominant mode. In this case, the "a" dimension of the
waveguide must be kept near the minimum allowable value to ensure that only the dominant
mode will exist. Theoretically, the wavelength at the cutoff frequency is approximately two
times of the width o f the waveguide. In practice, this dimension is usually 0.7k, where X is
the wavelength in free space (www.tupb.com).
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Waveguides may also be designed to operate in a mode other than the dominant
mode. O f the possible operation modes available for a given waveguide, the dominant mode
has the lowest cutoff frequency.
Energy injection and removal to and from waveguides:
Special devices are
necessary when energy needs to be put into and removed from a waveguide at its two ends.
Three devices commonly used are probes, loops and slots (also called apertures, or windows)
(www.tpub.com). The placement of the three devices in the waveguide, size, and shape can
critically impact the related input/output efficiency. Sometimes high resistive loads (dummy
loads) can be placed at the end of a waveguide to absorb the energy from microwaves.
(I) Probes:
The small probe inserted into a waveguide acts as a quarter-wave
antenna. When supplied with microwave energy, current flows in the probe and an E field is
established.
The E lines in turn detach themselves from the probe.
For a rectangular
waveguide, the probe is usually placed at the center of the "a" wall, parallel to the "b" wall,
and one quarter-wavelength from the shorter end of the waveguide. This location has the
largest E field for a dominant mode; therefore, the energy transfer (coupling) is at its
maximum. A probe’s length can be adjusted to alter the degree of energy transfer. For
example, a probe can be moved out of the center of the E field, or be shielded to reduce the
amount of energy transfer when a lesser degree of energy transfer, called loose coupling, is
desirable. A probe’s size and shape determine its frequency, bandwidth, and power-handling
capability. When increasing its diameter, a probe’s bandwidth increases; and its powerhandling capability is affected by the surface areas of the probe. The removal of energy from
the other end of the waveguide can be accomplished by using the same type of probes in a
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reverse manner from the energy injection process.
(2) Loops:
waveguide.
Loops put energy into a waveguide by setting up an H field in the
The inserted small loop carrying a high current builds up a magnetic field
around itself. The magnetic field in turn expands to fit the waveguide. If the frequency of
the current in the loop is within the bandwidth of the waveguide, energy will be transferred
into the waveguide. For most efficient coupling, the loop is usually inserted at a location
with the greatest magnetic field strength. When less efficient coupling is required, a loop can
be rotated or moved until it encircles a smaller number of H lines. The diameter of loop
determines its power-handling capability; and the size of a loop wires can modify its
bandwidth. A similar loop can be used to remove energy from the waveguide when an H
field is present.
(3) Slots: Slots or apertures are used for very loose (inefficient) coupling. When the
energy enters through a small slot in the waveguide, the E field expands into the waveguide
by first crossing the slot and then the interior of the waveguide. If the size of a slot is
properly proportioned to the frequency of the energy, energy reflections are minimized
during injection or removal.
(4) Dummy loads: A dummy load is an end piece of a waveguide that is either filled
with a graphite and sand mixture or water. Sometimes, a high-resistance rod is placed at the
center of the E field in the end of the waveguide, too. The dummy loads are required to
match the characteristic impedance of the waveguide. When the fields enter the dummy
loads, a current flow is induced and microwave energy is dissipated as heat. The termination
of a waveguide can reflect the energy by permanently welding a metal plate at the end of the
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
waveguide.
Leaving the ends open is definitely not the most efficient way to inject and remove
energy in a waveguide because of formation of electromagnetic fields around the end of the
waveguide. These fields cause an impedance mismatch, developing standing waves and
leading to power loss. In fact the impedance of a waveguide does not always match the
impedance of the load; and any abrupt change in impedance results in reflection and standing
waves, thus decreasing the efficiency of the waveguides. When the change in impedance at
the end o f a waveguide is gradual, for example, by terminating the waveguide with a funnelshaped horn, little standing waves are formed, thus minimizing non-uniform heating (Figure
6).
Figure 6: Waveguide is terminated with a funnel-shaped horn. Left: Conical
horn; Right: pyramidal horn.
On the other hand, major applications of microwave energy in food industries are
characterized by the absorption of microwave energy coming from the magnetron and the
waveguide. Most waveguides are terminated in cavities, in which the materials of interest
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
are placed and heated. At Washington State University, a hom-shape waveguide was newly
developed and terminated into a cavity. The repeatability and predictability of its heating
pattern were studied using model foods (mashed potatoes).
Design and construction o f the waveguides: Waveguides are mainly hollow metal
pipes, whose designs are determined by the frequency and power level of the electromagnetic
energy they will carry. The size, shape, and dielectric material of a waveguide must be
constant throughout its length for energy to move from one end to the other without
reflections. Any abrupt change in its size or shape can cause reflections and a loss in overall
efficiency. If a change is necessary, the bends, twists, and joints of the waveguides must
meet certain conditions to prevent reflections.
An entire waveguide system cannot normally be molded into one piece; instead, the
waveguide must be constructed in sections and connected with joints.
The waveguide
sections can be taken apart for maintenance and repair. Energy losses in a waveguide system
are mainly from improperly connected joints, damaged inner surfaces and the moisture
content change from the air in the waveguide. A properly constructed waveguide has a low
loss ratio unless the waveguide is damaged from physical error, internal arcing or other
reasons.
ABSORPTION OF MICROWAVE ENERGY
Dielectric properties o f materials'. Dielectrics have the capacity to store energy in an
electric field through the polarization of their molecules and atoms in the presence of an
imposed electric field (Nelson 1990). The dielectric properties of a material can affect the
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
behavior o f electromagnetic waves passing through them.
The parameter used to quantify the response of a material to an applied electric field
is referred to as complex relative permittivity sr, which is given as below:
I
Sr
t!
-je r
(6)
Here er the real part of the complex permittivity, often referred to as the dielectric
constant or relative permittivity, describes the ability of a material to store energy in response
to the applied electric field. The imaginary part of the complex permittivity, sr", commonly
referred to as the loss factor, describes the ability of a material to dissipate energy in
response to an applied electric field (Mudgett 1985).
Complex relative permeability (jur) is a third parameter to describe energy storage and
dissipation of a material in an applied magnetic field.
Pr = p r ’ - j p r ”
(7)
where jjr \ the real part o f the complex permeability, also referred to as the relative dielectric
constant, describes the ability of a material to store energy in response to an applied
magnetic field; and p r ”, the imaginary part of the complex permeability, or the relative loss
factor, describes the ability of a material to dissipate energy in response to an applied
magnetic field (von Hippel 1954).
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The dielectric properties of a material, Sr and fir. are relative values with respect to
free space (vacuum) to describe the behavior of a material in an electromagnetic field. The
complete expression for the permittivity and permeability of a material are given as follows:
s = sr £0 Farads!meter
(8)
H = fir /loHenrys!meter
(9)
where ji0 and e0 are the permeability and permittivity of the free space and given by
s0 = 8.854187817 x 1O'12Farads/metex
(10)
p.0 = 4ti x IQ'7 HenrysImeter
(11)
Most foods do not have a significant interaction with the magnetic field, having a
relative permeability o f fir = 1 - jO (Metaxas and others 1993.). Since the magnetic field
does not contribute much to the generation of thermal energy in microwave heating, the
relative magnetic permeability is not considered when discussing the dielectric properties of
food materials.
The electrical conductivity of a material (&), the third electromagnetic constitute
parameter for the material medium, can sometimes be used to express the imaginary part of
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the complex relative permittivity with a relationship given by:
6}£q
2}jf£Q
where a (or 2nf) is the radian frequency (radian/sec) and / i s the temporal frequency (Hz).
In electromagnetics, a dielectric material can often be described by the following four
characteristics: homogeneity, isotropy, dispersion and linearity. They illustrate the change of
a material’s dielectric properties with respect to location or from point to point
(homogeneity), direction (isotropy), and frequency (dispersion), respectively.
If the
magnitude o f the induced polarization field is directly proportional to the magnitude of the
electric field, the material is said to be linear (Ulaby 1999).
For most food products
subjected to microwave heating, they are seldom entirely homogeneous.
Anisotropy is
usually found in crystalline or striated materials and is not found in most foods. Since the
dielectric constants of most food materials change with frequency and the materials can
propagate different frequency components at different speeds, dispersion occurs and
common food materials cannot be assumed to be non-dispersive. Neither the phenomenon of
dispersion nor the non-linearity significantly impacts microwave heating. All the model food
materials in this study are assumed to be homogeneous, linear, isotropic and dispersive.
Affecting factors for dielectric properties o f foods: Extensive experimental data for
the dielectric properties of various foods have been reported (Calay and others 1995) as well
as the influence of the affecting factors, which mainly include frequency of the
electromagnetic waves, temperature, water content and salt content (Engelder and others
1991; Galema 1997). A brief review of the influencing factors on dielectric properties is
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
provided below. ■
(1) Frequency".
Dielectric properties over a broad range of frequencies can be
measured at each temperature using modem network analyzers and impedance analyzers.
Frequency dependence of the dielectric properties of most materials is mainly the result of
molecular polarization, an important phenomenon arising from the orientation with the
imposed electric field (Nelson 1978). The frequency dependence might help determine an
optimal working frequency for dielectric heating.
(2) Temperature:
Reflecting the dielectric relaxation processes, temperature
dependence of dielectric properties is complex since more than one dispersion mechanism
such as bound water, free water and ionic dispersions can be involved. Determination of a
material’s dielectric properties must, therefore, include measurements over a complete range
of temperatures of interest.
The dielectric constant and dielectric loss factor for polar
materials could either increase or decrease with temperature (Nelson and others 1990). For
example, dielectric constants of red delicious apples (Malus domestica Borkh) with moisture
content greater than 70% were reported to decrease with temperature in the microwave
frequency range (Feng and others 2002); the dielectric constant of pre-gelatinized bread
increased with temperature from 20 to 65 °C , then became nearly constant from 60 to 95 °C
(Goedeken and others 1997).
(3) Moisture content".
Moisture content affects the dielectric properties of
hygroscopic materials. The degree of the effect strongly depends upon the form of water in
the materials. The majority of water in foods with moisture content at or over a critical value
o f 35-40% is in a free form in the capillaries (Sun and others 1995; Tran and others 1987);
and free water is the dominant component governing the overall dielectric behavior of foods
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Sun and others 1995). Three distinct ranges of dielectric behavior governed by moisture
content are discussed at microwave frequencies (Mudgett and others 1980), namely, below
10% moisture content bound water contributes to low dielectric activity of materials; when
increasing the moisture level to intermediate level (10-35%), the availability of loosely
bound or free water cause the dielectric constant and loss factor to increase rapidly with the
contribution from ionization of bound salts; above 35%, the dielectric loss factor shows little
dependence on moisture content.
(4)
Salt content: Salts or dissolved ions generally cause a reduction in polarization of
water and a decrease in the overall dielectric constant by binding water. Salts also increased
the dielectric loss factor above that of pure water because of electrophoretic migration
(Mudgett 1985).
These influences are directly related to the nuclear charge effect that
depends on the size and charge of dissolved ions (Bircan and others 1998). Galema (1997)
reported that the presence of electrolyte (NaCl) did not seem to influence the dielectric
constant greatly while it did have a marked effect on the dielectric loss factor.
Microwave heat generation and absorption: Microwave heating is widely used both
at home and in industries for cooking, defrosting and drying. Microwave heating actually
belongs to the category of dielectric heating, that is, by directly coupling a lossy dielectric
material with a high frequency electromagnetic radiation. In the United States, 2450 MHz
microwaves are used for home microwave ovens and industrial systems; 915 MHz
microwaves are allocated for industrial use.
Another popular form of dielectric heating
operates between 6 and 40 MHz (e.g., 27 MHz); this is known as radio frequency (RF)
heating. RF heating is beyond this study and will not be discussed here.
The origin of microwave heating is the result of the ability o f the electric field to
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
polarize the charges in the material and the inability of this polarization to follow extremely
rapid reversals of the electric field. The charge particles are redistributed under the influence
o f the externally applied electric field and form conducting paths, particularly in mixtures of
heterogeneous materials. Coupled with these polarization effects, a dielectric can be heated
through direct conduction effects.
Characterized by volumetric heating, the thermal energy is generated throughout the
volume of a material in microwave heating. Energy from an alternating electromagnetic field
can be dissipated by currents in the electron cloud (conventional ionic conductivity) and in
losses incurred in the reorientation of dipoles at high frequencies (dielectric loss). But it is
difficult to separate the contributions from ionic conduction and dipolar relaxation for most
complex materials.
In general, the average power dissipation per unit volume (Pav) in a dielectric
subjected to an electromagnetic field is given by:
Pav = 2KS0fe"E 2 Watts/meter3
It can be rewritten using the parameter of conductivity as:
Pav = crE2 Watts/meter3
(14)
where/ i s the frequency (Hz); a is the equivalent conductivity (Siemens/meter); and E is the
electric field intensity inside the product (Volts/meter) (Meredith 1998).
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
When incident on the surface of food materials, microwaves are either reflected or
pass through the materials.
During propagation inside the foods, a fraction of the
microwaves is absorbed and converted into heat.
The rest of the microwave energy is
transmitted through the material to reach the other interface between the products and
ambient medium.
Penetration depth:
During transmission inside the product, the strength of the
microwaves decreases because of energy absorption.
The degree of reduction is
characterized by the penetration depth (dp) of a material, also known as the skin depth. This
parameter describes the distance from the surface of the product to the location where the
electric field intensity o f an incident electromagnetic wave diminishes to 1/e (37%) of its
original magnitude at the surface (Ramo and others 1984).
The penetration depth (dp) of the electric field intensity is given by:
d p = ----------------- ‘-----------------
(15)
where c is the speed of light in vacuum and often given as 3.0 x 108 m/s; f is the frequency of
the electromagnetic waves (Hz); e'r and s ”
r are the dielectric constant and dielectric loss
factor of the materials being heated (dimensionless). According to the power absorption
(Equation 13), the power dissipation is reduced by a factor of 1/e2, or about 13.5% of its
original value at the surface within one skin depth.
Another concept for the penetration depth is given from the viewpoint of power
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
absorption: power penetration depth (Dp). The power penetration depth is defined as the
depth from the incident surface to the location where the power intensity o f an incident
electromagnetic wave diminishes to 1/e (37%) of its original amplitude at the surface. The
power penetration depth (Dp) is given by:
(16)
f
1+
s." \
2
-1
where c is the speed of light in vacuum and often given as 3.0 x 108 m /s ;/is the frequency of
the electromagnetic waves (Hz); e'r and e"r are the dielectric constant and dielectric loss
factor of the materials being heated (Buffler 1993). It is obvious that power penetration
depth is one half o f the skin depth in the same case.
Both the power penetration depth and skin depth describe the transmission ability of
microwaves when incident on products. This ability is dependent upon the frequency of the
waves and the dielectric properties of the material. According to Equations 15 and 16, the
two penetration depths are inversely proportional to the working frequency (f); this can help
us choose the size o f the package at the frequency of interest.
Generally packages of
microwave treated products are smaller than those treated by RF heating devices. However,
the penetration depth does not vary exactly as l / f since the dielectric properties vary with
frequency. The influence of dielectric properties is complex. For example, for salty foods or
products with high moisture content, the penetration depth can be very small because of the
relatively large dielectric loss factor. The temperature effect can be very significant when the
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
dielectric loss factor of products increases with temperature, causing the phenomenon of
“thermal runaway heating”. In other words, the part of a product with the higher temperature
absorbs more heat than the lower temperature part when the temperature increases,
aggravating the non-uniformity of microwave heating. Strategically selecting different initial
temperatures for products to be heated at various locations, particularly for different food
materials packaged in multiple compartment trays, can be an effective way to minimize this
“runaway” heating (Meredith 1998). Heat conduction from location with relatively high
temperature to the location with low temperature within the foods helps to improve heating
uniformity.
This leveling effect can be efficient when using relatively low microwave
power, or when the microwave system works in an on-off pulsing mode (Morris 1991).
P O W E R M E A S U R E M E N T A T M IC R O W A V E F R E Q U E N C IE S
Power measurement and methods: Power, defined as the quantity of energy
dissipated or stored per unit time (Watts, or Joules/sec), can determine the overall heating
phenomenon and efficiency and is one of the most fundamental parameters in dielectric
heating.
In DC circuits, the power P is easily measured and calculated by
V2
P = — = I 2R
R
(17)
where V is the voltage (Volts) across a resistance R (Ohms) and I is the current (Amperes)
through the same resistance (Wilcox and others 1970).
In an AC circuit containing inductance or capacitance, power measurement becomes
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
complicated because o f the voltage and current are out of phase. Power is calculated by
p = VI x cos <9
(18)
where 0 is the phase angle difference between V and I. The magnitudes of V and I are the
RMS values (Wilcox and others 1970). The power measurement in a DC circuit can be
generally carried out by using a voltmeter (V) and an ammeter (A); and the active power in
AC circuits can be measured by a dynamometer.
Power measurement, almost indispensable for the use of microwave in food
industries, presents more of a challenge. The instruments used to measure voltage or current
can cause de-tuning of the microwave circuit or coaxial transmission line, thereby making
measurements inaccurate (www.azdtechnology.com). However, the power that causes the
component heating is the same at different source frequency (DC, 60 Hz AC, or
RF/microwave); voltage and current vary with the length of a lossless coaxial cable while
power is not a function of length. It is still possible to make accurate power measurement for
microwave generation systems and microwave circuits such as amplifiers, oscillators and
filters.
Two types o f methods are commonly used for measuring microwave power:
directional and terminating. Directional method means the measurement o f the forward and
reflected power individually, using a directional coupler inserted into the microwave
transmission lines or waveguides (Figure 7). For accurate measurement at higher power
levels, the loss at the point o f insertion should be minimized. A directional power meter can
remain connected for continuous monitoring (www.microlease.com).
The terminating
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
method means the power meters absorb the energy signal at the termination point and
indicate the power
by
measuring the heat dissipated by the microwave power.
Power
measurement using terminating power meters is accomplished through the connection of a
load to the applied microwave system.
Directional coupler meter
Detector
Detector
Reflected power
Forward power
Source
Directional coupler
Termination
Figure 7: Power measurement using directional coupler
Microwave power meters: A microwave power meter usually consists of a power
sensor, the temperature o f which is raised by the conversion of microwave power into heat
energy and is used to calculate the final power. In general, there are two types of thermally
sensitive power sensors: calorimetric and bolometric. A calorimetric power sensor has a
temperature-sensitive indicator (e.g., a thermocouple) external to the thermal element itself.
Calorimetric power meters can be categorized into flow (liquid) and static (dry) (Aditya and
others 2001). The flow-type calorimeter uses a liquid such as water to carry heat away from
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the load in a controlled maimer.
The dissipation of power in the load is obtained by
measuring the temperature rise of the liquid which flows in or around the load. A basic
measurement setup for a flow-type microwave calorimeter is shown in Figure 8. The static
calorimeter measures the electromagnetic field generated in a thermopile placed between a
reference level and an active one dissipating input power. Two water loads for microwave
power measurement at 2450 and 915 MHz microwaves are also shown in Figure 9.
*
Outlet
Oul
Circulator
=zO=
Waveguide
!„
Water load
Inlet
Microwave generator
Figure 8: M easurem ent setup using a flow-type microwave calorim eter (water load) to
calibrate the generator power output.
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 9: W ater loads for microwave power measurement at the frequencies of 2450
MHz (Left, RF Technologies Corporation, Lewiston, Maine) and 915 MHz (Right, from
Ferrite Components, Inc., Hudson, NH) at WSU.
A major concern for calorimeter measurement is whether there is a linear relationship
between temperature rise and the actual power. Sometimes, thermal inertia occurs because
o f the lag between the application of microwave power and the parameter readings.
A
microwave calorimeter is often used to measure high powers for standards and calibration
purposes (Aditya and others 2001).
Bolometer means any device whose electrical resistance changes with power
absorption (Legowski, 1999). The two most common types o f bolometers are baretters and
thermistors.
A baretter has a thin metallic (platinum) wire with a positive temperature
coefficient of resistance; while a thermistor has a semiconducting material with negative
temperature coefficient o f resistance.
They can be easily mounted in microwave
39
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
transmission lines or waveguides. Baretters are more delicate than thermistors; they are used
only for very low power (up to a few milliwatts). Because of the small overload allowed and
fragility, power sensors using barretters are no longer manufactured (Legowski 1999). A
low-power thermistor sensor can measure medium and high power levels with proper
attenuators, but they are not typical power sensors now. Calorimeters, bolometers and other
sensors based on microwave heating effect usually can be independently calibrated, but they
can measure only the active (average) power.
The active (or average) power in microwave power measurements is averaged over
many periods o f the signal and is defined as:
\
PtQ+nT\
(19)
where t0 is the initial time, I) is the period of the lowest frequency component of v(t) and i(t);
v(t) and i(t) are the alternating voltage {Volts) and current (Amperes), respectively (Aditya S
and others 2001). Since the averaging time is typically in the range from several hundredths
of a second to a few seconds and is much greater than Th it is not essential to integrate over
an integer number of periods of v(t) and i(t) (Bucci and others 1999; Legowski 1999).
Schottky barrier diodes are also used to measure microwave power while not using
the heating effect o f the applied power as an indicator (www.microlease.com).
RF/microwave diode detectors depend on the non-linear diode junction to generate a DC
voltage when a microwave signal is applied. Because of the small amount of power (in the
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
order of 1 W) the Schottky diodes deal with, directional couplers or attenuators have to be
used together to measure higher power levels.
The Schottky diodes also have limited
measurement accuracy and application because they are calibrated by using reference power
meters or pulse power standards.
The measured power can be expressed as absolute power in watts (W) or as relative
power in dBm. The definition of dBmis given by:
dBm =10tog,0(p)
(20)
where the active power P is expressed in milliwatts.
Three kinds o f circuits can affect the readouts from power meters: the circuit from the
microwave part; the circuit that produces DC or low frequency power equal to the microwave
power dissipated in the power sensor; and the measurement circuit of the DC or low
frequency power (Aditya and others 2001). The errors normally encountered in microwave
power measurements are due to mismatch loss, RF loss, and substitution error (Hewlett
Packard 1972). The mismatch loss can be minimized by matching the load and the source
through the insertion o f attenuation to reduce the reflections. The RF loss and substitution
error are both associated with the calibration factor. Errors in the overall power measuring
system can be compensated by using a thermistor with a calibration figure of about 0.9 to 1.
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TEMPERATURE MEASUREMENT IN MICROWAVE HEATING
Temperature scales: All temperature scales are based on some easily reproducible
states, for example, the freezing and boiling point for the Celsius scale. A mixture of ice and
water is said to be 0 °C if the mixture is in equilibrium with air saturated with vapor at 1 atm
pressure (the freezing point); while a mixture of liquid water and water vapor without air is
100°C if the mixture is in equilibrium at 1 atm pressure (the boiling point) (Qengel and
others 2002). Temperature is one of the most important indicators in the food processing
industries.
As a standard scientific index, temperature has always been quantitatively
recorded by different kinds o f thermometers or sensors. By carefully monitoring the material
or ambient temperatures, the interactions among various materials can be well controlled; the
QA/QC procedures can be accurately documented; the process protocols can be optimized
and validated.
Measurement principles o f temperature sensors:
The U.S. Food and Drug
Administration (FDA) require that Mercury-in-Glass (MIG) thermometers be used as the
standard temperature reference in canning and aseptic processing systems to sterilize foods
(FDA, CFR 113).
In practice, electronic sensors are used widely for temperature
measurement, including resistance temperature detectors (RTDs), Integrated Circuits (IC)
sensors, thermistors, and thermocouples.
The first three types consist of either metal
materials or semi-conducting materials as the sensing elements, whose resistance changes
with temperature are well understood and repeatable.
Thermocouples have no sensing
elements. Instead, a thermocouple incorporates two electrical conductors made of different
materials and connects them at one end, which is used to measure temperatures and is called
the measurement junction', the other end (reference junction) is used to connect the
conductors to the measurement unit.
When the two junctions of a thermocouple are at
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
different temperatures, a millivolt potential is created in the conductors.
The type of
thermocouple used, the millivolt potential magnitude, and the reference junction temperature
allow users to determine the temperature at the measurement junction.
At present, thermocouples are playing a major role in the temperature sensing field
(Desmarais and others 2001).
Thermocouples calibrated by Mercury-In-Glass (MIG)
thermometers have been widely accepted to monitor processing temperatures. However, the
metal parts from the electronic sensors can cause electromagnetic interference (EMI) when
these sensors are used in a high electromagnetic field environment. To minimize the effect
of the EMI, fiber optic temperature sensors are developed and used to measure the
temperature during dielectric heating (microwave and RF).
The optical fibers are made of non-conducting glass or plastics; and the fiber’s
immunity to EMI is attributed to the fact that photons, instead of electrons, are used as its
signal propagation element. During the transmission of light or energy through the optical
fiber, the principle of total internal reflection pertains (Balanis 1989). That is, all the light
striking a boundary between two media is totally reflected (Figure 10). The total internal
reflection principle applies as long as the following two conditions are satisfied (Ulaby 1999;
Biala 2001).
(1)
The angle o f incidence is larger than the critical angle for the particular
combination of materials, for instance, core and the cladding for the optical
fiber.
(2)
The index o f refraction of the fiber is larger than that of the surrounding
medium. That is, the light is from a denser medium and incident on a less
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
dense medium. For a fiber optical cable, it requires the cladding material be
less dense than the core material.
C la d d i n g
\
\
A
\
Figure 10: Total internal reflection requires that all light strike the boundary at
an angle greater than the critical angle.
Most fiber-optic temperature sensors available on the market today share similar
characteristics: great versatility, high accuracy, and large dynamic range. But the principles
behind differ from one to another (Alexandre 2001; www.fiso.com; Stokes and others 2002).
Following is a brief summary for three types of fiber-optic temperature sensors.
(1)
Fluoroptic Thermometry (FOT) method: Both Luxtron Corp (Santa Clara, CA,
USA) and Ipitek Corp (Carlsbad, CA, USA) offer fiber optic temperature sensors based on
this technology.
At their sensor tips, certain temperature-sensitive materials such as
phosphor are placed. At the near end, the probe attaches to the measurement unit with a
standard fiber-optic connector.
At the far end, the probe tip is either embedded in the
material to be measured or placed on the surface. The excitation and emission light are
separated in the unit to enhance sensitivity.
By using the low power source inside the
instrument, the temperature sensitive material is excited via the optical fiber; and the
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
resulting luminescence travels back to a detector. After the excitation source is pulsed, the
decaying fluorescence intensities (I) are sampled at two accurately separated times, tj and
to monitor lifetime. The yielded signals are given by
h
/(/,) = / / ' | r ) + C
(21)
h
/((! ) = / 8e " ® + C
(22)
where I(t) is the decaying fluorescence intensities at different time; x(T) is temperaturedependent fluorescence lifetime; C is offset due to background crosstalk, ambient light.
When C = 0 , T can be solved by taking the ratio:
R=y ,)
i(h)
(23)
Rephrase Equation (23), we get
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 '
=—
inR
(24)
where At — t, - t2, the time interval between the first and the second readings of emission
intensity.
Fluorescent lifetime, an intrinsic property of the phosphor material (YsAlsG^Cr3*,
Cr:YAG), mainly depends on phosphor temperature, except for quite negligible pressure
dependence. By getting T, the temperature is found from a look-up table established through
prior calibrations o f the probe.
Only the ratio of correlated readings and the accurate
determination of a time interval are entailed for a temperature measurement process; and all
other instrumental and environmental variables automatically cancel out. The small thermal
mass of the phosphor element at the fiber tip enable rapid and accurate temperature
measurement.
(2)
Cavity length method: FISO Technologies Inc. (Sainte-Foy, Quebec, Canada)
fabricates Fabry-Perot fiber-optic temperature sensors, whose gauges are designed based on
a Fabry-Perot interferometer (FPI). The FPI basically consists of two mirrors facing each
other with a distance called the cavity length. The success of FPI technology is mainly
dependent upon the precise and reliable measurement of the Fabry-Perot cavity length
through the unique white-light cross-correlator (U.S patents 5392117 and 5202939), which
can measure the absolute Fabry-Perot cavity length with linearity and consistency.
The white-light sensing method is illustrated in Figure 11. Light from a broadband
source is launched into one arm of a 2 x 2 coupler and directed toward the Fabry-Perot
gauge. The light signal is wavelength-modulated by the gauge and is reflected back toward
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the readout Fiber-Optic Sensor Instrument, focused on a line, transmitted through the crosscorrelator, and detected by a linear CCD array. The cross-correlator functions as a spatiallydistributed Fabry-Perot cavity whose length varies along the lateral position. Facing the
CCD array, each pixel is associated with a predefined Fabry-Perot-like cavity length. Thus,
this device works like an optical cross-correlator with a spatially varying cavity length.
Micro capillary
Fused welding
Fabry-Perot
Gauge
Multimode optical
Semi-reflecting mirrors
Optical fiber
Connector
Reflected light
Readout instrument
Incident light
2*2 Coupler
Light Source
Incident light
Lens
Cross-correlation function
Linear CCD a nay
WW
White-light cross-correlator
FPI Modulated Reflected Light
Figure 11: W hite-light sensing method is used to determine the cavity length
(redrawn by the author and originally from www.fiso.com).
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(3)
Crystal method: Nortech Fibronic Inc. (Quebec, PQ, Canada) used to offer fiber­
optic temperature sensors that take advantage of the light absorption/transmission properties
o f gallium arsenide (GaAs) coated at the end of the sensor. The transmission spectrum of the
semi-conducting crystal, or the light that is not absorbed, shifts to higher wavelengths when
its temperature increases (Figure 12). At any given temperature, transmission jumps from
essentially 0% to 100% at a specific wavelength. This jump is called the absorption shift.
The relationship between the temperature and the specific wavelength at which the
absorption shift takes place is very predictable.
Transmission spectrum
T=-40
T=120
T=40
Wavelength
Figure 12: Increases in the crystal's temperature shifts its transmission
spectrum to higher wavelengths (Temperature unit: °C, from Alexandre, 2001)
The absorption shift occurs because of the variation in the semiconductor’s energy
band gap. An energy band gap means the energy required to bump the electrons in the
material from a relaxed, steady state into an excited state. When entering the crystal, only
those photons carrying enough energy to get an electron across the gap are absorbed;
otherwise, they will be transmitted. The shorter a photon’s wavelength, the more energy it
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
carries. Since the band gap narrows as the crystal’s temperature increases, and less energy is
needed to jump the gap, photons with less and less energy (longer and longer wavelengths)
are absorbed by the band, moving the absorption shift to longer wavelengths. The more
energy in the form of heat that enters the crystal, the narrower the band gap becomes;
consequently, measuring the position of the absorption shift gives a measure o f the crystal’s
temperature.
The GaAs crystal’s universal and constant response requires no calibration for a
probe. When measuring temperature, the sensor must be in contact with the material of
interest. The response of the crystal to temperature increases with the contact intimacy.
Characteristics o f fiber-optic temperature sensors: Many fiber optic instruments are
available on the market today as alternatives to metal probes to measure temperature in
microwave heating application. The common advantages and disadvantages for most fiber­
optic temperature sensors are summarized as follows:
Advantages:
(1) Fiber-optic probes have high chemical resistance and need no metal wires or other
metal parts.
(2) Measuring accuracy is not affected in the presence of severe EMI.
(3) Response time is small, for example, around 1.5-1.8 sec.
(4) Measurement has an accuracy of ±1°C or better with a resolution of 0.1 °C.
(5) The probe can either be placed in surface contact with the material of interest or
immersed in the sample vial.
(6) Fiber-optic cable is flexible and can be produced to any length (up to 5 km).
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(7) It Is possible to disconnect the gauge or shut the system down while keeping track
o f the actual value because of Its absolute measurement and no loss of reference.
(8) Perfect linear response Is achieved through absolute measurement.
(9) They are insensitive to light loss due to fiber bending, cable length, or light source
fluctuations. Versatility o f the technique allows the user to perform various types
o f measurements.
Disadvantages:
(1) Fiber optic temperature probes are very fragile and expensive compared with
thermocouples and the other sensors.
(2) It is not easy to place them in pressure vessels, particularly for microwave
sterilization processing in our study where water is being circulated during the
measurement.
(3) Glass fibers tend to break and cannot withstand sharp bending, stretching,
extreme vibration, pulling, and other harsh treatment.
(4) Though a few broken strands in a bundle are generally not noticeable, there will
be a proportionate loss of signal strength when large numbers are severed.
Fiber-optic temperature sensors from FISO Technologies Inc were used to measure
temperature for most of the research work in this microwave heating study.
All of the
sensor’s benefits and drawbacks have been experienced throughout the microwave heating
studies. Efforts have also been made to use thermocouples to monitor microwave processing
temperature.
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
USE OF MICROWAVES IN PACKAGED FOODS STERILIZATION
Microwave heating started as a by-product of the radar technology during World War
II (Meredith 1998). Due to the speed and efficiency o f volumetric heating, microwave ovens
using 2450 MHz electromagnetic waves have been widely used at home. Microwave energy
using 915 (or 896) MHz and 2450 MHz electromagnetic waves has also been efficiently used
by the food industry for thawing, continuous baking, and vacuum drying, tempering, pre­
cooking, blanching and final drying.
On the contrary, using microwave energy for sterilization of foods is not popular.
Only two commercial systems worldwide have been reported recently for sterilizing food
products using microwave energy (FDA 2000).
Similar to a conventional sterilization
process, a microwave sterilization process mainly consists of preheating, heating, holding,
and cooling stages, all o f which are completed under pressure (Harlfinger 1992; Guan and
others 2002 and 2003).
Equilibration of the product following heating is sometimes
proposed to level the temperature distribution and improve uniformity (Fakhouri and others
1993; Ramaswamy and others 1992).
The major advantage of microwave sterilization
technology is its relatively short come-up time, which helps retain the organoleptic qualities
and justify the preference of using microwave processing instead of conventional thermal
processing. The destruction of microorganisms in foods sterilized by microwave energy is
mainly from the thermal effect (Goldblith and others 1967; Rosen 1972).
That is, the
microbial inactivation kinetics of microwave sterilization is very similar to that of
conventional thermal processing; and the time-temperature histories at the coldest locations
determine the microbial safety of the process.
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Critical process factors involved in microwave sterilization technology are mostly
from three major aspects; all the critical factors mentioned can alter the spatial distribution of
microwave absorption, heating rate and time-temperature history.
1.
Food composition:
food composition has a much greater impact on
microwave processing than in conventional processing mostly because of the influence of
moisture and salt contents on product’s dielectric properties. The food composition changes
the product’s thermal properties including specific heat, density, thermal conductivity and
resultant heating uniformity.
2.
Processing condition and system design: These include microwave frequency,
power level, oven dimension, power level, cycling, and presence of hot water or air around
the food, etc.)
Processing and equipment design affect both the magnitude and spatial
variation o f the power absorption in the products. Like domestic microwave ovens, mode
stirrers and turntables improve microwave heating uniformity; placement of foods inside the
oven affects the magnitude and uniformity of power absorption.
3.
Package: The packaging materials used for microwave sterilization need to be
microwave transparent and have high melting points. Contrary to commercial sterilization
processing where metal containers offer minimum thermal resistance, metallic components
(e.g., aluminum foil and susceptors) can dramatically influence the heating rates and patterns
inside packaged foods.
Metals are sometimes used to redistribute microwave energy to
increase heating uniformity.
Three ways have been reported in research to monitor a microwave sterilization
process:
direct
temperature
measurement
'
using
fiber-optic
temperature
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
probes;
microbiological validation using classical surrogate/indicator microorganisms such as Cl.
sporogenes (Guan and others 2003); and chemical time-temperature integrators such as the
intrinsic chemical markers (Kim and others 1993; Lau and others 1999).
In industry, implementation of a microwave sterilization process varies significantly
among the manufacturers (Kenyon and others 1970; Mudgett and others 1982; Decareau
1985). Due to the nature o f the microwave sterilization process, it is complex to determine
and control process deviations.
During implementation, product trays must be precisely
placed in the pressurized tunnels. Each package receives a pre-calculated, spatially varying
microwave power profile optimized previously. Both infrared surface temperature and top
surface swelling due to internal steam generation during heating are monitored for each tray,
using infrared cameras and distance tracers, respectively.
At precise locations in the
package, maxi-thermometers are sometimes placed to measure maximum temperatures as
visual control. Inside the microwave tunnel, endoscopes are installed to observe the heating
process (FDA 2000).
Power settings for individual magnetrons are stored over time;
DataTrace metallic temperature data-loggers are applied inside the package to monitor the
time-temperature history, however, use of such metallic data-loggers requires careful
considerations and interpretations (Harlfinger 1992). A food product package is rejected by
an automated control system at the end of the cooling system based on available information
as listed above. It is true that the control system can be programmed for each individual
product, but extensive experiments are generally required to validate the effectiveness and
reliability of a microwave sterilization process.
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHALLENGES FOR MICROWAVE STERILIZATION TECHNOLOGY
In theory, food products can be sterilized by using microwave energy. Through the
direct interaction between microwaves and foods that are hermetically sealed in
microwaveable packages, microwave heating offers the potential to overcome the limitation
of heat transfer from the heating medium to the food during conventional heating (Burfoot
1988; DeCareau 1985; Harlfinger 1992, Meredith 1998). Rapid volumetric heating occurring
from microwaves in properly designed systems can reduce the heating severity and
processing time to achieve the required lethality for thermal processes.
Microwave
processing can potentially improve product quality (Giese 1992; Stenstrom 1974; O ’Meara
and others 1977; Ohlsson 1987 and 1991).
Much research has been conducted to
demonstrate the potential application of microwaves in the field of food sterilization (Ayoub
and others 1974; Muddget and others 1982).
But applications of microwave heating for food sterilization in the processing
industry are much less common than home applications. Historically, this low use was due
to the lack of basic information on food dielectric properties, lack of ways to measure the
temperature during processing, and high cost of equipment and electricity. Generally the
food processing industry has been reluctant to make expensive investments in a technology
that has not been proven thoroughly reliable in large-scale or long term use (Mudgett 1989).
At present, dielectric properties can be readily and extensively measured with commercially
available systems such as impedance analyzer or network analyzer; and fiber-optic
temperature sensors can be used to measure the temperature without interference between
foods and the applied electromagnetic fields. More reliable magnetrons have been developed
and ferrite circulators are available to protect generating tubes; microwave equipment has a
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
much longer operating life than ever. The cost for microwave equipment has been steadily
reduced over the years.
It is promising for microwave heating application in food
sterilization.
Major challenges still delay this application. Firstly, because o f the reflection and
refraction phenomena of microwaves at the interface between the food and the surrounding
air, discontinuity of the electric field components occurs, resulting in non-uniform
distribution o f the microwave energy within the food products (Ayoub 1974; DeCareau 1985;
Ruello 1987; Schiffinann 1990; Stanford 1990; Keefer and others 1992).
This non­
uniformity within the food causes severe comer and edge heating (Ohlsson 1991).
To
minimize non-uniform heating effects, many techniques have been developed; experiments
using water immersion during microwave heating have been conducted (Stenstrom 1972;
Ohlsson 1992; Lau 2000; Guan and others 2002 and 2003).
For example, studies have
indicated that a 915 MHz MCWC heating system can provide a relatively uniform heat
distribution within food products packaged both in pouches and trays, and the loss of color
attributes in certain food products are minimized (Lau 2000). But neither microbiological
safety nor corresponding sensory attributes of the microwave processed foods have been
systematically studied.
Secondly, it is difficult to provide and predict desired heating patterns in foods, which
in turn make it challenging to locate critical temperature measuring points in the processed
foods. The intrinsic chemical marker technology developed at US Army Natick Solider
Center by Kim and others (1993) holds promise for non-invasively assessing heating
uniformity within foods in microwaves sterilization systems (Kim and others 1996, Prakash
and others 1997). However, the relationship between the marker yield and the sterilization
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
value at the same location within foods is still unclear. There might be alternative way to
determine the temperature distribution within a package during microwave heating, such as
preliminary inoculated pack studies and computer simulations. Inoculated pack studies rely
on the amount of bacteria survivors to predict the heating pattern, and it is time consuming
and expensive. Using proper computer simulation software can be helpful in determining the
heating pattern under certain cases, but there always exists discrepancy between reality and
the results o f simulation work, mainly due to the limitation of the software used, the complex
nature of the microwave heating and the complicated microwave heating system. Inserting
temperature probes at a number of locations might help to determine the region that
experiences the minimum heat treatment. However, only a relatively clear heating pattern,
not an exact location of the cold spot under specific heating conditions, can be determined
after all these efforts.
Thirdly, energy efficiency and on-line controls present a challenge for the
industrialization of microwave sterilization technology.
It is infeasible to strive for the
highest power density and shortest process time. Even though a very high power-dissipation
density can be achieved with modest applied voltage stress, this may still result in voltage
breakdown (Merdith 1998).
Besides the temperature parameter, other critical factors or
indexes should be identified and be used to monitor the microwave sterilization process. All
this information relies on a relatively stable microwave heating system, which does not exist
in the US until lately.
Fourthly, there are other challenging topics not covered here, some of which may be
important. These include design of packages for microwave heating and optimization of the
process conditions for food products packaged in single- or multi-components during
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MCWC heating.
In conclusion, microwave heating application in the food sterilization field is very
promising, even though it is premature to predict when the food sterilization processing
systems using microwave energy can be industrialized.
FINAL REMARKS
In Chapters 1 and 2, the fundamental information regarding the thermally processing
o f hermetically packaged low-acid foods, microwave heating, and the challenges for its
industrial applications in food sterilization are introduced. Basic principles are also reviewed
to provide a general summary microwave sterilization technology. Based on the background
information reviewed and specific requirements from the microwave sterilization project at
WSU, research results over the last 3.5 years are given in the following chapters. Formats of
manuscript vary to fit those required for publication.
Objectives: The overall objective of this Ph.D. study is to address technical
processing issues related to Microwave-Circulated Water Combination (MCWC) heating
technology being developed at Washington State University. The sub-objectives are listed
below:
(1) To evaluate the microbiological safety of MCWC processed entrees, for example,
macaroni and cheese product: This research helps to evaluate the potential of using MCWC
heating technology in food sterilization.
Procedures for measuring thermal resistance of
targeted microorganisms (PA 3679) and inoculated pack studies provide groundwork for
later studies in this area.
(2) To evaluate the sensory attributes of the MCWC processed entree (macaroni and
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
cheese) foods compared with controls (freshly cooked): This research further demonstrates
the potential for pilot-scale MCWC heating technology to produce high quality shelf-stable
food products.
(3) To better understand the interactions between electromagnetic waves and food
materials: In this part of research dielectric properties of mashed potatoes were measured
and effects o f frequency, temperature, salt and moisture content on these properties were
studied. Dependencies on the affecting factors are explained and predicative equations are
developed;
(4) To investigate the heating pattem(s) of mashed potato through experimental tests
and through computer simulation work: The dielectric properties were used with QW-3D
computer simulation software to help explain performance of the MCWC heating system and
locate the slowest heating region during the process. Results from this part of work can be
used to guide further inoculated pack studies.
Significance: This research can provide convincing evidence that the MicrowaveCirculation Water Combination (MCWC) heating technology can provide predictable and
relatively uniform heating for selected packaged foods.
Reliable process protocols
accompanied by a microbiological process validation procedure can support the filing for
FDA approval of this novel thermal processing technology. It is also anticipated that the
results will speed the development and commercialization of Microwave-Circulated Water
Combination (MCWC) heating technology in the United States. This technology can provide
food manufacturers a new and effective way to produce shelf-stable foods with less quality
degradation, shorter thermal process times, easier process control and less environmental
impact.
The technology can be of great benefit to US companies and certainly provide
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
opportunities for new military and retail product development that is not possible with
conventional thermal processes.
REFERENCES:
Aditya S and Arokiaswami A. 2001. Microwave measurement instrumentation.
Webster I (ed.). Wiley Encyclopedia of Electrical and Electronics Engineering Online. John
Wiley & Sons, Inc. USA.
Alexandre I. 2001. Fiber-optic temperature measurement. Sensors. 18 (5).
Ayoub JA, Berkowitz D, Kenyon EM and Wadworth CK. 1974. Continuous
microwave sterilization of meat in flexible pouches. J. Food Sci. 39:309- 313.
Balanis. CA. 1989. Advanced engineering electromagnetics. John Wiley & Sons, Inc.
NY. USA.
Biala N. 2001. An introduction to fiber-optic sensors. Sensors. 18 (12).
Bircan C and Barringer SA. 1998. Salt-starch interactions as evidenced by viscosity
and dielectric property measurement. J. Food Sci. 63 (6): 983-986.
Bucci C, Landi C and Capua CD. 1999. Microwave power measurement. Wiley
Encyclopedia of Electrical and Electronics Engineering Online. Webster J(ed.). John Wiley
& Sons, Inc.
Buffler CR. 1993. Microwave cooking and processing. Ch. 5. In Microwave cooking
and processing. Van Nostrand Relnhold. NY. USA.
Burfoot D, Griffin WJ and James SJ. 1988. Microwave pasteurization of prepared
meals. J. Food Engineering. 8:145-156.
Calay RK, Newborough M, Probert D and Calay PS. 1995. Predicative equations for
the dielectric properties o f foods. Int’l J Food Sci. and Technol. 29:699-713.
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C^engel YA and Boles MA. 2002. Thermodynamics: an engineering approach. 4th
edition. Mc-Graw Hill Companies, Inc. NY. USA.
Curtis JM Jr. 1999. Experimental verification for microwave processing of materials
in a single mode rectangular resonant cavity. Master thesis. Virginia Polytechnic Institute and
State University. USA.
DeCareau RV. 1985. Microwaves in the Food Processing Industry. Academic Press,
Inc. NY. USA.
Desmarais R and Breueraeis J. 2001. How to select and use the right temperature
sensor. Sensors. 18 (1).
Engelder DS and Buffer CR. 1991. Measuring dielectric properties of food products
at microwave frequencies. Microwave World. 12: 6-15.
Fakhouri MO. and Ramaswamy HS. 1993. Temperature uniformity of microwave
heated foods as influenced byproduct type and composition. Food Res. Int. 26:89-95.
FDA. 1977. Code of Federal Regulations. Title 21. Food and Drugs. Food and Drug
Administration, Government Services Administration. Washington, DC.
FDA. 2000. Microwave and radio frequency processing. From Kinetics o f microbial
inactivation fo r alternative food processing technologies. Center for Food Safety and
Applied Nutrition.
Feng H, Tang J and Cavalieri RP. 2002. Dielectric properties of dehydrated apples as
affected by moisture and temperature. Trans. Am. Soc. Agric. Eng. 45: 129-135.
Galema SA. 1997. Microwave Chemistry. Chem. Soc. Reviews 29(3): 233-238.
Giese J. 1992. Advances in microwave food processing. Food Technol. 46:118-123.
Goedeken DL, Tong CH and Virtanen AJ. 1997. Dielectric properties of a
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pregelatinized bread system at 2450 MHz as a function of temperature, moisture, salt and
specific volume. J. Food Sci. 62: 145-149.
Goldbiith SA and Wang DIC. 1967. Effect of microwaves on Escherichia coli and
Bacillus subtilis. Appl. Microbiol. 15(6): 1371-1375.
Guan D, Plotka VCF, Clark S and Tang J. 2002. Sensory evaluation of microwave
treated macaroni and cheese. J. Food Process. Preserv. 26: 307-322.
Guan D, Gray P, Kang DH, Tang J, Shafer B, Ito K, Younce F and Yang TCS. 2003a.
Microbiological validation of microwave-circulated water combination heating technology
by inoculated pack studies. J. Food Sci. 68 (4): 1428-1432.
Harlfinger L. 1992. Microwave sterilization. Food Technol. 46(12):57-60.
Hewlett Packard power meter 432A operating and service manual. 1972. 1501 Page
Hill Road. Pal alto, CA. USA.
Kenyon EM, Westcott DE, LaCasse P and Gould J. 1971. A system for continuous
processing of food pouches using microwave energy. J. Food Sci. 36(2):289-293.
Kim HJ and Taub LA. 1993. Intrinsic chemical markers for Aseptic processing of
particulate foods. Food Technol. 47(l):91-97, 99.
Kim HJ, Taub IA, Choi YM and Prakash A. 1996. Principles and applications of
chemical markers of sterility in high-temperature-short-time processing of particulate foods.
In Chemical Markers for Processed and Stored Foods, Ch. 6, T.C. Lee and H J. Kim, eds.,
pg. 54-69. American Chemical Society, Washington, D.C.
Kuang W and Nelson SO. 1998. Low-frequency dielectric properties of biological
tissues: A review with some new insights. Trans. Am. Soc. Agric. Eng. 41(1): 173-184.
Lau MH, Tang J, Taub IA, Yang TCS, Edwards CG and Younce FL. 1998.
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Microwave heating uniformity of food during 915 MHz microwave sterilization process.
Proceedings o f the 33rd Microwave Power Symposium. 78-81
Lau MH. 2000. Microwave heating uniformity in model food systems. From
Microwave pasteurization and sterilization offood products. Ph.D. Dissertation. Washington
State University. USA.
Legowski SF. 1999. Power meters. Wiley Encyclopedia of Electrical and Electronics
Engineering Online. Webster J (ed.). John Wiley & Sons, Inc.
Meredith R. 1998. Engineers’ handbook of industrial microwave heating. The
Institution o f Electrical Engineers. London. UK.
Metaxas AC and Meredith RJ. 1993. Industrial Microwave Heating. Peter Peregrinus
Ltd. London. UK.
Morris CE.1991. World’s 1st microwave sterilization system. Food Engineering Int’l.
6: 63-66.
Mudgett RE, Goldblith SA, Wang DIC and Westphal W.B. 1980. Dielectric behavior
of semi-solid food at low, intermediate and high moisture contents. J. Microwave Power.
15(1): 27-36.
Mudgett RE and Schwartzberg HG. 1982. Microwave food processing: Pasteurization
and sterilization: A review. AIChE Symposium Series. 78(218): 1-11
Mudgett RE. 1982. Microwave properties of foods in microwave processing. Food
Technology 36:109-115.
Mudgett RE. 1985. Dielectric properties of foods. In: Microwaves in the Food
Processing Industry (R. V. Decareau, author). Academic Press, New York, NY.
Nelson SO. 1978. Electrical properties of grain and other food materials. J. Food
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Process. Preserv. 2: 137-154.
Nelson SO and Kraszewski AW. 1990. Dielectric properties of materials and
measuring techniques. Drying Technol. 8(5): 1123-1142.
Ohlsson T. 1987. Sterilization of foods by microwaves. Paper presentation at
International Seminar on New Trends in Aseptic Processing and Packaging o f Food stuffs.
Munich, October 22-23.
Ohlsson T. 1991. Microwave heating uniformity. Paper presented at the AICHE
Conference on Food Engineering, Chicago. March 11-12.
Ohlsson T. 1992. Development and evaluation of a microwave sterilization process
for plastic pouches. Paper presented at the 8th World Conference of Food Science and
Technology, Toronto, September 29- October 4.
O ’Meara JP, Farkas DF and Wadsworth CK. 1977. Flexible pouch sterilization using
combined microwave-hot water hold simulator. Contact No. (PN) DRXNM 77-120, U.S.
Army Natick Research & Development Laboratories, Natick, MA 01760.
Prakash A, Kim HJ and Taub IA. 1997. Assessment of microwave sterilization of
foods using intrinsic chemical markers. J. of Microwave Power and Electromagnetic Energy
32(l):50-57.
Ramaswamy HS and Pilletwill T. 1992. Temperature distribution in microwaveheated food models. J. Food Quality. 15(6):435-448
Ramo S, Whinnery JR and Van Duzer T. 1984. Fields and waves in communication
electronics. John Wiley & Sons, Inc.
Rosen CG. 1972. Effects of microwaves on food and related materials. Food Technol.
26(7):36-40, 55
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Roussy G and Pearce IA. 1995. Foundations and industrial applications o f microwave
and radio frequency fields: physical and chemical processes. John Wiley & Sons, England.
Ruello JH. 1987. Seafood and microwaves: some preliminary observations. Food
Technol.in Australia. 39, 527-530.
ScMffinann RF. 1990. Problems in standardizing microwave over performance.
Microwave World. ll(3):2G-24.
Stanford M. 1990. Microwave oven characterization and implications for food safety
in product development. Microwave World. 11(3): 7-9.
Stenstrom LA. 1972. Taming microwaves for solid food sterilization. Paper 7.4
presented at the IMPI symposium. Ottawa, Canada. May 24-26.
Stenstrom LA. 1974. US Patents 3,809,845; 3,814,889.
Stokes J and Palmer G. 2002.A fiber-optic temperature sensor. Sensors. 19 (8).
Sun E, Datta A and Lobo S. 1995. Compostion-based prediction of dielectric
properties of foods. J. Microwave Power and Electromagnetic Energy. 30 (4): 205-212.
Tran VN and Stuchly SS. 1987. Dielectric properties of beef, beef liver, chicken and
salmon at frequencies from 100 to 2500 MHz. J. Microwave Power. 22: 29-33.
Ulaby FT. 1999. Fundamentals of applied electromagnetics. Prentice Hall, Inc. NJ.
USA.
U.S patents 5392117 and 5202939.
von Hippel AR. 1954. Dielectric material and applications. John Wiely & Sons. NY.
USA.
Wilcox G and Hesselberth C. 1970. Electricity for engineering technology. Ally and
Bacon, Inc. Boston. USA. Page 50, 220.
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3
MICROBIOLOGICAL VALIDATION OF MICROWAVE-CIRCULATED
WATER COMBINATION HEATING TECHNOLOGY BY INOCULATED PACK
STUDIES
Dongsheng Guan3, Peter Grayb, Dong-Hyun Kangb, Juming Tang3*, Bradley Shaferc
Keith Itoc, Frank Younceab and Tom C.S. Yangd
aDepartments o f Biological Systems Engineering and bFood Science and Human
Nutrition, Washington State University, Pullman, WA, 99164
cNational Food Processors Association (NFPA), Dublin, CA, 94568
dUS Army Soldier and Biological Chemical Command, Natick, MA, 01760
ABSTRACT
A 915 MHz Microwave-Circulated Water Combination (MCWC) heating technology
was validated for a macaroni and cheese product using inoculated pack studies. Prior to the
tests, heat resistances of a Cl. sporogenes (PA 3679) spore crop were determined in neutral
phosphate buffer and macaroni and cheese product. Trays of macaroni and cheese products
were subjected to three processing levels: target processing (Fo - 2.4), under-target
processing (Fo = 1.2) and over-target processing (Fo = 4.8). The inoculated packs were
evaluated by Count-reduction method and End-point method. The microbial results showed
that microbial destruction resulting from MCWC heating technology matched with the
calculated degree of sterilization (Fo value). This study suggests that the MCWC heating
technology has potential for sterilizing packaged foods.
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTRODUCTION
Microwave heating refers to the use of electromagnetic waves of frequencies between
300 MHz to 300 GHz to generate heat in a material (Metaxas and Meredith 1993). Research
has been conducted to use microwave heating for food pasteurization and sterilization
(Ayoub and others 1974; Mudgett 1982).
These studies took advantage of volumetric
heating resulting from the direct interaction between microwaves and foods to reduce process
times (Ohlsson 1978).
It is generally believed that the destruction of microorganisms during microwave
heating is due to thermal effect (Fujikawa and others 1992).
But achieving heating
uniformity remains a major challenge in the research and development o f microwave heating
technologies. This non-uniform heating arises from the discontinuous dielectric properties
between foods and the surrounding air (Ramaswamy and Pillet-Will 1992) as well as the
difference in the dielectric properties of different food constituents (RyynSnen and Ohlsson
1996).
Several researchers have made use of a water immersion technique (Stenstrom 1972;
Ohlsson 1992; Lau and others 1998) and 915 MHz microwaves (Lau and others 1998) to
minimize
non-uniform
microwave heating.
Ohlsson
(1987)
demonstrated
good
bacteriological safety of several microwave-sterilized products with a pilot scale 2450 MHz
microwave-water immersion processing unit. The products exhibited sensory qualities that
were superior to conventionally processed foods.
A pilot scale 915 MHz Microwave-
Circulated Water Combination (MCWC) heating system was developed at Washington State
University (WSU, Pullman, WA, USA) which demonstrated a relatively uniform heat
distribution within certain food products packaged in pouches and trays (Lau, 2000).
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
However, the microbiological safety of 915 MHz microwave-processed foods was not
validated.
Another challenge in developing a microwave sterilization process is how to monitor
the processing history to ensure the microbiological safety of the processed foods. Fiber
optic sensors have been used to measure temperatures in microwave heating.
They are
particularly suited for use in high temperature short-time processes for the following reasons:
(1) Fiber optic temperature probes do not interfere with microwave fields; (2) The probe
sizes can be as small as 0.8 mm in diameter, resulting in short response times (from 0.05 s to
0.2 s in most foods); and (3) They provide accuracy comparable to thermocouples in a
normal heating medium (Fiso Technologies, Inc., Quebec, Canada). At WSU, fiber optic
sensors have been used to monitor the temperature of products during microwave heating
processes and the data were used to calculate degrees of sterilization (Fo values). But there
was no evidence to support the hypothesis that these calculated Fo values truly reflect the real
lethality of the processes. Further efforts were required to confirm the reliability of using
fiber optic sensors to monitor MCWC heating processes with a pilot scale test unit.
The objectives o f this study were to determine the sterilization effect of the MCWC
heating test system by utilizing inoculated pack studies and to determine the practicability of
using fiber optic temperature sensors for the MCWC heating processes.
MATERIALS AND METHODS
Preparation o f Macaroni and Cheese Products: Margarine and 2 % fat milk were
purchased from local grocery stores. Box-type noodles and cheese powder were supplied by
Kraft Foods (Glenview, IL). To prepare the macaroni and cheese samples, 166.1 g of dry
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
noodles were precooked in 1.5 L boiling water and stirred periodically for 6 minutes, then
drained and cooled immediately with tap water. Cheese sauce was prepared separately by
blending melted margarine, milk, and cheese powder together at 50°C. Both the partially
cooked noodles and sauce were prepared in a hygienic lab kitchen and were packaged
immediately after preparation.
Heat Resistance o f PA 3679 in Phosphate Buffer (pH 7.0): Costridium sporogenes
(PA 3679, NFPA NO. SC 218) spores were obtained from the Center for Technical
Assistance o f the National Food Processors Association (NFPA, Dublin, CA, USA).
Thermal death time (TDT) tests were conducted at three different temperatures, 115.6°C,
118.3°C and 121.1°C using a thermoresistometer at NFPA, Dublin, CA. During the tests
(115.6°C: 3, 5, 8, 15, 25 min; 118.3°C: 2.5, 3, 5, 8, 15 min and 121.1°C: 1, 2, 3, 5, 7 min),
sample cups containing 0.01 ml of the diluted spore suspension (1.0 x 106 spores/ml, 1/15M
phosphate buffer) were placed in the carrier boats. After predetermined exposure times to
saturated steam of constant temperature, sample cups were moved out of the pressurized
heating chamber and fell directly into tubes containing culture media at ambient temperature.
Sterile Vaspar (~6 oz. paraffin in 2 liters Vaseline) was used to overlay the culture to provide
anaerobic conditions in the tubes, which were then incubated at 30° C for 3 weeks. Positive
growth was indicated by gas production and confirmed by characteristic odor and
microscopic examination.
The D-values of PA 3679 spores were calculated using Equation (1) (Stumbo
1973):
D = ---------- -------------logjo a - iogio b
(1)
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
where U is the heating time; a is the initial spores count per sample multiplied by the number
of replicates; and b (referred to as most probable number of spores surviving the timetemperature relationship to which the samples were subjected) was calculated by Equation
(2):
b = xxn
(2)
Here, n is the total number o f replicates and x (referred to as most probable number of
spores surviving per replicate sample) was calculated by Equation (3):
x = 2.3026 £ogw (n/q)
(3)
where q is the number of sterile samples as evidenced by lack of growth in subculture
medium.
Heat Resistance o f PA 3679 in Macaroni and Cheese Products: Thermal Death
Time (TDT) mini-retorts (NFPA, Dublin, CA) were used to determine the heat resistance of
PA 3679 spores in macaroni and cheese product. The six retort units were connected directly
to a steam line with an automatic temperature controller. Mercury thermometers for the TDT
retorts were graduated to within 0.28°C (0.5°F).
The general description of the
thermoresistometer and TDT mini-retorts was given by Townsend (Townsend and others
1956).
Macaroni and cheese product was prepared as described previously and made into a
puree with a blender. 15.0 grams of homogeneous and thick puree was weighed into a TDT
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
can (208 x 006, or 6.4 cm in diameter x 0.95 cm in height).
The desired number of
organisms, 104 spores in 0.1 ml de-ionized water, was pipetted into the center o f each can. In
total, 180 vacuum-sealed cans were heated at three temperatures: 115.6°C, 118.3°C and
121.1°C. After being heated for five time intervals selected for each temperature (115.6°C:
3, 5, 8, 15, 25 min; 118.3°C: 2.5, 3, 5, 8, 15 min and 121.1°C: 1, 2, 3, 5, 7 min), the TDT
cans were incubated at 30°C for three months. Another five cans were used as controls, three
of which were heat shocked (100°C and 4 min) and two untreated. Positive growth was
indicated by gas production and confirmed by characteristic odor and microscopic
examination. Non-swollen cans and non-swollen cans in the next highest process level were
also examined.
The D-values at the three temperatures were calculated using Equation (1) and the
corresponding z-values were obtained by plotting D-values on semi-log papers.
M CW C Heating System: The 915 MHz MCWC heating system consisted of three
major components: (1) a 5 kW 915 MHz microwave generating system (Microdry Model IV5 Industrial Microwave Generator, Microdry Incorporated, Crestwood, KY) and a multimode
cavity (121.3 cm wide x 121.3 cm long x 151.1 cm high); (2) a pressurized microwave
heating vessel; and (3) a water circulation heating and cooling system.
The 915 MHz microwave system was equipped with a circulator to protect the
microwave generator from heat damage caused by reflected power. A directional coupler
with appropriate sensors was used to measure forward and reflected powers. The output
microwave power was calibrated and stabilized at 1.0 kW by regulating anode current to the
magnetron.
The pressurized microwave-heating vessel allowed treatment of a single meal tray at
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
selected time intervals under an over-pressure.
The sidewall of the vessel was made of
cylindrical aluminum tube (23.0 cm in diameter and 5.0 cm in height). Its top and bottom
plates were made o f Tempalux material (Ultem Polyetherimide Resin, Lenni, PA, USA) that
had a high melting temperature (above 150°C) and was transparent to microwaves. An over­
pressure (34 ~ 35 psig) was provided by compressed air in a surge tank and used within the
vessel to maintain the integrity o f the food package. Fittings were provided to allow for
temperature measurements and for concurrent circulation of pressurized water.
In the circulated-water control system, circulation water was maintained at the desired
temperature by two plate heat exchangers and used to heat and cool the food package during
MCWC processing.
The exchangers were heated and cooled with steam and tap water,
respectively. A Think & Do™ computer program (Entivity, Ann Arbor, MI) was used to
control the modulating valves of the exchangers. The flow rate of the circulated water was
9.5 L/min.
Temperature Measurement during M CW C Processing:
To measure the sample
temperature using optical fiber sensors, a 3.0 cm-long polyimide tubing (OD: 0.075 inch or
0.1905 cm; ID: 0.0710 inch or 0.18034 cm; thickness: 0.00200 inch or 0.00508 cm, ColeParmer, IL, USA) was sealed at one end using silicone sealant (Dow Coming®, Dow
Coming Corp., Midland, MI, USA). The tube was inserted through a hole in the side of the
tray such that the sealed tip was located in the center of a tray (10.0 cm wide x 14.0 cm long
x 2.5 cm deep x 0.3 cm thickness, Polypropylene and EVOH trays, Rexam™ Union, MO)
before packaging. Two pieces of rubber (Diameter: ~1 cm, thickness: 1/32 inch or 0.7938
cm, McMaster-CARR Supply Company, CA, USA) adhered the tubing to both sides of the
tray wall using silicone sealant, keeping it from shifting in the sample.
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In this study, a fiber optic sensor was inserted into the tubing to take the product
temperature, which was at the center of the tray. Due to the inherent non-uniformity in
microwave heating, the center was not necessarily the coldest spot(s) in the tray.
Not
knowing exactly the location o f cold spot(s) for the product packaged in the tray, we used the
center as a reference location to measure the temperatures. Another reason for this selection
is a chemical marker technique was used to understand the heating pattern of this MCWC
heating technology for the same product packaged in pouches.
The pouch had similar
dimensions as the tray being used (Lau, 2000). Although no direct relationship between the
marker yield and the heat received was shown, it indicated that the least marker yield was
obtained at the center of the pouches.
The F0 values were calculated by Equation (4) based on the temperature histories for
different process procedures (Lopez, 1987).
(4)
o
where F0Z is the degree o f sterilization (F0 value, unit: min.) at 121.1°C for a certain z value;
T is the actual temperature o f the product (°C); Tr is the reference temperature (121.TC); the
z-value in this test is 6.78°C, which is obtained from the above heat resistance tests and t is
heating time (min.).
During the MCWC process, the circulating water temperature for in-line heating and
cooling as well as the product temperature at the center of the tray was displayed and
recorded every 6 seconds. The degrees of sterilization (F0 values) were also shown on the
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
screen instantly.
Package Integrity and Sealing o f Products: Package integrity, critical to product
stability, was visually observed after processing and during incubation. The products were
sealed under vacuum to make rapid microwave heating and cooling possible. Nitrogen flush
was applied during sealing and the overpressure was regulated throughout the process.
The sealing prototype unit, customize built by Rexam Containers (Model N o.l,
Rexam™, Union, MO), consisted of a heating mechanism in an enclosed chamber. A pump
and a nitrogen tank were connected to the chamber for vacuum seal and subsequent gas
flushing. A metal “nest” holder secured trays containing the product in the sealing chamber.
The holder was aligned with a thermostatically controlled heat-sealing head driven by a
pneumatic cylinder.
A control panel displayed the operation parameters including seal
pressure (psig), sealing head temperature (°F), chamber vacuum (inches o f Mercury) and seal
duration (seconds).
Before sealing the product, the partially cooked noodles (98.0 g) were placed into the
tray in the laboratory kitchen and the inoculated liquid sauce (102.0 g) was poured onto the
noodles. The tray filled with products was flushed with nitrogen and heat sealed with a 0.1
mm lid stock (Polypropylene/EVOH laminated) under vacuum (14 inches of Mercury, or
58.69 kPa). A food tray prior to MCWC heat treatment is shown in Figure 3-a.
M CW C Heating Process Procedures: MCWC heating processes were similar to
conventional steam or pressurized hot water retorting processes. It included four stages:
preheating, combined heating, holding, and cooling. The product in the vessel was first
preheated to 75°C with circulating hot water at 100°C. The combined heating started when
the microwave power (1.0 kW) was turned on and the circulation water was set at 120°C.
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The holding stage began by maintaining the circulated water at 120°C while the microwave
was tamed off. After the desired holding period, the tray was cooled using circulating water
at 80°C (2 min.) under pressure, then by 20°C tap water at ambient pressure.
Before testing with inoculated packs, the product tray inserted with fiber optic
temperature sensors at the center was treated with the MCWC heating system. The lethality
of the process was determined using Equation (4). The procedures, as controlled by initial
temperature of the product (around 45°C), the time for preheating, heating time and cooling,
were repeated for the inoculated pack without the inserted tubing.
Three processing procedures (Table 1) were selected in this study aiming at three
levels of degree o f sterilization (Fg value). “Target” processing procedure was designed to
eliminate the inoculated PA 3679 spore population (1.1 xlO6 spore /' 200g, D12U - 0.40 min)
with a sterilization value (log reduction value) of 6.04, corresponding to a degree of
sterilization (F0 value) o f 2.4 min. “Target” processing was equivalent to an 8D process for
Cl. botulinum (assuming a D12U of 0.3 min). Over-target processing was designed to destroy
the inoculated PA 3679 spores completely, aiming at a sterilization value (or log reduction
value) o f 12.1 or an F0 value of 4.8; under-target processing procedure was selected to allow
certain amount of spores to survive after processing and was designed to have a sterilization
value of 3 or an F0 value o f 1.2.
Table 1: Processing Procedures for Three Processing Levels (Unit: min)
Processing
Designed Degree
Preheating
Combination Heating Holding
Cooling
levels
o f Sterilization (F0 Time
Time
Time
Time
______________ val u e ) ________________________________________ _____________________________
Under1.2
3.8
2.8 "
'
0
6.0
Proper2.4
3.8
3.2
0
6.0
Over
4.8
3.8
3.2
1.2
6.0
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Inoculated Pack Tests: The same batch o f PA 3679 spore crop was re-enumerated at
WSU right before preparing inoculated cheese sauce.
The spore suspension was heat-
shocked at 80°C for 10 min, cooled in crashed ice-water, tenfold serially diluted in 0.2 %
peptone water and spread-plating 100 pi onto duplicate SFP agar plates. Counts taken after
48 hours incubation at 37°C indicated the initial concentration of the spore suspension. The
diluted spore suspension containing approximately 1.1 x 106 viable spores/ml was added into
the liquid cheese sauce, targeting to give the inoculated level o f 1.1 x 106 spores/2OOg.
The survival of PA 3679 spores in MCWC-processed macaroni and cheese product
were analyzed by the following two methods:
(1) Cmmt-reduction Method: In this method, the log reduction (sterilization value,
or SV) of PA 3679 spores in the processed products was determined by counting the
survivors after incubation. All the MCWC processed macaroni and cheese products (200g)
from one tray were divided into two 100 g portions. They were homogenized with a Seward
400 Circulator Stomacher (Seward, Ltd., London, UK) in 200 ml sterile 0.2 % peptone water
at 260 rpm (xg) for 2 min. Four 2.5 ml portions of homogenate from each portion were pourplated with Clostridium-selective SFP Agar Base (Difco, Detroit, MI) and incubated in
Anaerobic Gas Pack Systems (BBL) at 37°C. Colony counts were recorded after 48 hours
incubation. Colony counts from both 100 g portions were added and expressed as viable
CFU/tray. If no viable spores were detected in a processed tray, the survival numbers of PA
3679 spores in the tray was recorded as below the detection limit (30 CFU/tray). Three trays
subjected to each MCWC heating process level were evaluated.
(2) End-point Method: Because of the inherent detection limit (>30 CFU/tray) o f the
count-reduction method, the end-point method was used to further confirm the lethality of the
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
processed products.
In this method, ten trays processed under each process level were
incubated at 37°C for three months to check the survival of PA 3679 spores. The trays were
checked every 2-3 days during incubation. Bulged trays were indicative of viable PA 3679
spores, which were further confirmed by the presence of the characteristic putrefactive odor.
Non-swollen frays and non-swollen trays in the next highest process level were also
examined. Trays that showed no signs o f bulging after three months were scored as having
zero viable spores.
RESULTS AND DISCUSSION
Thermal Resistance o f the PA 3679 Spores: The D-values at 121.1°C and z-value for
spores in neutral phosphate buffer (pH = 7.0) and macaroni and cheese product (pH = 5.7)
are listed in Table 2.
Table 2: Therm al Resistance of PA 3679 in Phosphate Buffer and Macaroni Cheese
Phosphate
Buffer
Macaroni Cheese
Phosphate
Buffer *
1.07
9.92
0.40
6.78
1.06
9.33
pH Range ** Cl.
(pH>4.5)
botulinum
** (Type A,
____________ B)
D]2] j
z-vaiue (°C)
0.10-1.5
7.78-10
0.1-0.20
7.78-10
*From Nordsiden, 1978
**In Low Acid and Semi-acid Foods, from Stumbo, 1973
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
At 121.1°C, the D value (1.06 min) and z-value (9.33°C) of the spore crop in
phosphate buffer were very close to literature values (Nordsiden and others 1978). But the
D-value (0.40) in macaroni and cheese product was at the lower end of the value range (0.10
~ 1.5 min) for low acid foods (pH > 4.5) and the related z-value (6.78°C) was off the
corresponding range (7.78 ~ 10°C) (Jay 2000).
Many factors affect the heat resistance of bacteria: inherent genetic and
environmental factors during the growth of bacteria, heating of the bacterial suspension, the
pH value of the medium (Santos and others 1993) and salt and fat/lipid content (Molin and
Snygg 1967). Inherent resistance varies not only with species but also with different strains
o f the same species. Different strain of the same species grown in the same medium and
heated in the same menstruum might show widely different resistance (Stumbo 1973).
On the other hand, lowering pH of a medium or increasing salt content typically
reduces the thermal resistance of spores (Stumbo 1973). In this study, certain ingredients in
the macaroni and cheese products obviously decreased the heat resistance of PA 3679 spores.
However, it is not clear which ingredients led to this reduction of heat resistance.
M CW C Processing and Integrity o f Packaging: Figure 1 shows a typical MCWC
heating history with a degree of sterilization (Fo value) of 2.4. In microwave processing,
cooling, rather than heating, is said to set the process speed limit (Stenstrom 1970).
According to Figure 1, the cooling time takes 6 minutes, about one half of the total
processing time.
technology.
This should be considered in the industrialization of MCWC heating
Package Integrity was visually examined after the MCWC processing and
incubation period (90 days and 37°C). It appeared that the tray wall was slightly softened
upon removal from the process vessel after processing. The package expanded slightly, with
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
stretching o f the material. But the package integrity was maintained during the microwave
sterilization process and over the three-month storage period (Figure 2-top and middle). The
seals held well in trays that were under-processed as shown in Figure 2-bottom.
140
120
100
80
60
Circulated w ater
40
Macroni
and c h e e se
20
0
2
4
6
8
10
12
Time (min)
Figure 1- A typical temperature-time heating history for the MCWC heating
process (equivalent to a processing procedure with degree of sterilization of 2.4)
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 2- Visual observation of package integrity of typical food trays
Note:
Top-Food tray prior to MCWCV heat treatment
Middle-Target-processed tray after three-month incubation at 37°C
Bottom-Under-target processed tray stored for 3 months at 37°C
79
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Inoculated Pack Studies: The initial concentration of the spore suspension was 1.0
x IQ8 spores/ml. The results from the inoculated pack studies were summarized in Table 3
and 4.
(1) Count-reduction Method: The number of surviving spores in processed food
trays was counted after incubation; the process values equivalent to log reduction (PA 3679)
were shown in Table 3. No viable spores were detected in macaroni and cheese product from
trays subjected to target and over-target processing. The surviving numbers of PA 3679
spores in these trays were recorded as below the detection limit (30 CFU/tray).
The
corresponding log reduction values (4.56) of the detection limit were given as the actual
sterilization values for these processes.
The actual sterilization value from the trays
subjected to under-processing was slightly more than the targeted one.
Table 3: Result from Count-reduction M ethod
Inoculated levels
Process levels
*Designed Degree
**Designed
Actual sterilization
(Spores / 200g)
o f sterilization (F0
sterilization value
value (SV)
____________ __________ Value)_____________ (SV)_________________________________
TW O5
Under1.2
3.0
3.3
!.1*1G6
Proper2.4
6.04
>4.56
1.1*106
Over4.8
12.1
>4.56
* Fo value = SV x D n u (Unit: min)
** Equivalent Logic Reduction for PA 3679
(2) End-point Method: All 10 controls subjected to no heat treatment swelled within
two weeks due to gas production caused by the growth of PA 3679. Gas production also
occurred in the 10 trays subjected to under-target processing. The 20 (10 x 2) inoculated
trays subjected to target and over-target processing showed no evidence of gas production
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and lacked characteristic odor.
These results suggest that these processing levels could
destroy the targeted PA 3679 spores adequately, which agreed with the designed degree of
sterilization (Table 4).
Table 4: Results from End-point M ethod
Inoculated levels
(Spores / 200g)
Process levels
Lino6
Control
UnderProper
Over-
1.1*106
1.1*106
1.1*106
Designed
Sterilization
value (SV)*
N/A
3 .0
6.04
12.1
Number of
processed trays
Number o f
positive trays**
10
10
10
10
10
10
0
0
*Equivalent Log Reduction
** Indicated by gas production and characteristic odor, storage period: 3 months
As mentioned earlier, one of the major concerns in developing microwave
sterilization processes is how to monitor the non-uniform microwave heating. The designed
degrees o f sterilization
(Fq
values) in this study were based on the temperature history
obtained through fiber optical sensors inserted at the center of the trays, whereas the
microbial inoculation test indicated the possible biological safety for the whole tray. The fact
that the results of end-point studies matched the calculated degree of sterilization (Fo value)
suggests the practicability of using fiber optic sensors to monitor the microwave sterilization
process for this 915 MHz MCWC heating test system. However, future studies are needed to
determine the real location of the coldest spots for various kinds of products.
We are
designing a 915 MHz MCWC heating system that could process more than one tray at a time.
One of the trays can be used to monitor the heating processes along with other trays that
contain no sensors.
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C O N C L U S IO N S
The inoculated packs studies suggest that microbial destruction by a pilot-scale
Microwave-Circulated Water Combination (MCWC) heating system matched with designed
degrees o f sterilization (Fo value). This study suggests that the MCWC heating technology
has a potential in sterilizing packaged foods and it is practical to use fiber optic sensors to
measure the temperature for the 915 MHz MCWC heating test unit.
ACKNOW LEDGEM ENTS
The authors acknowledge the financial support provided by the U.S. Army Natick
Soldier Support Center and Kraft Foods. We also thank Mr. Evan Turek and Dr. Ming H.
Lau of Kraft Foods, Glenview, IL, for their technical advice and assistance.
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
Ayoub JA, Berkowitz D, Kenyon EM and Wadworth CK. 1974. Continuous
microwave sterilization of meat in flexible pouches. J. Food Sci. 39: 309-313.
Fujikawa H, Ushioda H and Kudo Y. 1992. Kinetics of Escherichia coli destruction
by microwave irradiation. Appl Environ Microbiol. 58: 695-698.
Jay, JM. 2000. Modem Food Microbiology. 6th ed. Chapman & Hall. New York.
USA.
Lau MH, Tang J, Taub IA, Yang TCS, Edwards CG and Younce FL. 1998.
Microwave heating uniformity of foods during 915 MHz microwave sterilization process.
Proceedings o f 33rd Microwave Power Symposium. P: 78-81.
Lau MH. 2000. Microwave Pasteurization and Sterilization of Food Products, Ph.D.
Thesis. Washington State University, Pullman, WA.
Lopez A. 1987. A Completed Course in Canning and Related Process. 12th ed. Book
II. Packaging; Aseptic Processing; Ingredients. The Canning Trade Inc. Baltimore, Maryland.
P: 13-16.
Metaxas AC and Meredith RJ. 1993. Industrial Microwave Heating. Peter Peregrinus
Ltd., London. UK.
Molin N and Snygg BG. 1967. Effect of liquid materials on heat resistance of
bacterial spores. J Appl. Microbiol. 15: 1422-1426.
Mudgett RE. 1982. Microwave properties of foods in microwave processing. Food
Technol 36: 109-115.
Nordsiden KL, Thompson DR, W olf ID and Zottola EA. 1978. Home canning of
food: effect of a higher process temperature (250 °F) on the safety of low-acid foods. J Food
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Scl. 43: 1734.
OMsson T. 1978. Temperature distribution of microwave faeating-spheres and
cylinders. J Microwave Power. 13: 303-309.
OMsson T. 1987. Sterilization of foods by microwaves. Paper presentation at
International Seminar on New Trends in Aseptic Processing and Packaging of Food stuffs.
Munich. Oct. 22-23.
OMsson T. 1992. Development and evaluation of a microwave sterilization process
for plastic pouches. Paper presented at the 8th World Conference of Food Science and
Technology, Toronto, September 29- October 4.
Ramaswamy HS and Pillet-Will T. 1992. Temperature distribution in microwaveheated food models. J Food Quality 15: 435-448.
Ryynanen S and OMsson T. 1996. Microwave heating uniformity of ready meals as
affected by placement, composition and geometry. J Food Sci. 61 (3): 620-624.
Santos MBS, Zarzo JT, Santamarta AA and Toran MJP. 1993. Citric acid lowers heat
resistance o f Clostridium sporogenes PA 3679 in HTST white asparagus puree. International
J. Food Sci. and Technol. 28: 603-610.
Stenstrom LA. 1970. The Golden Book. Presentation of a new food system. ALFALAVAL.
Stenstrom LA. 1972. Taming microwave for solid food sterilization. Paper 74
presented at the International Microwave Power Institute symposium, Ottawa, Canada, May
24-26.
Stumbo CR. 1973. Thermobacteriology in Food Processing, 2nd ed. Academic Press.
London.
84
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Townsend CT, Somers II, Lamb FC and Olson NA. 1956. A Laboratory Manual for
the Canning Industry. 2ed. National Corners Association.
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4
S E N S O R Y E V A L U A T IO N O F M IC R O W A V E T R E A T E D M A C R O N I A N D C H E E S E
ENTREE
Dongsheng Guan1,Virginia C. F. Plotka2, Stephanie Clark2, Juming Tang1*
’Department of Biological Systems Engineering and 2Food Science and Human Nutrition
Washington State University, Pullman, WA 99164-6376
A BSTRA CT
A pilot-scale 915 MHz Microwave-Circulation Water Combination (MCWC) heating
system was used to treat macaroni and cheese entrees prepared according to recipes selected
to minimize treatment effects on sensory quality.
Modifications to the traditional
commercial boxed macaroni and cheese recipe included the selection o f a noodle better
suited for applications requiring prolonged heat treatments and the addition of twice the
amount of cheese sauce to optimize heating uniformity.
The MCWC heating system
provided desired sterility (with a Fq value of 7 minutes) using one fourth of the time required
for conventional retort methods.
Descriptive analysis was used to identify the quality
attributes most significantly affected by MCWC processing. Formulation changes, such as
noodle type and amount of cheese sauce present, affected the overall quality of MCWC
treated macaroni and cheese entrees. Durum semolina noodles were superior to box-type
noodles (a blend o f durum and common wheat) in applications involving microwave heat
treatment.
A consumer panel rated microwave treated macaroni and cheese as being
acceptable when compared to freshly cooked controls.
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTRODUCTION
Intensive research has been conducted on the use of microwave energy in food
processing applications since the Second World War (Meredith 1998). The unique electrical
volumetric heating obtained through the interaction between dielectric materials and
microwave radiation has been efficiently used in the food industry for thawing, continuous
baking, and vacuum drying. It is also used for pasteurizing and, to a lesser extent, sterilizing
prepackaged foods. A major challenge for food processors to effectively use microwave
technology in food sterilization continues to be the non-uniform distribution of the
microwave energy within the food products (Ayoub 1974; Ruello 1987; Schiffinann 1990;
Stanford 1990; Keefer and Ball 1992). Discontinuous dielectric properties between food and
the surrounding air cause reflection and refraction phenomena of the microwaves at the
interface, resulting in non-uniform electric field distribution within the food and severe
comer and edge heating (Ohlsson 1990).
Water immersion technique (Stenstrom 1972; Ohlsson 1992; Lau 1998) and 915 MHz
microwaves (Lau 1998) have been explored to reduce non-uniform heating. Studies using a
chemical marker method have revealed that a 915 MHz Microwave-Circulation Water
Combination (MCWC) heating system provides a relatively uniform heat distribution within
food products packaged both in pouches and trays (Lau 2000). Our earlier experiments with
macaroni and cheese and scrambled eggs suggested that MCWC treatment reduced color
degradation in those products through a relatively brief heat exposure compared with
conventional retort treatments (Lau 2000).
The objectives of this study were to select macaroni and cheese formulations suitable
for treatment by MCWC heating and to compare the sensory quality of products processed by
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MCWC with freshly cooked macaroni and cheese.
MATERIALS AND METHODS
Preparation of Macaroni and Cheese for Processing: Whole milk, margarine, and
elbow-shaped 100% durum semolina noodles were purchased from a local grocery store.
Box-type noodles' (made from a combination of common and durum wheat) and cheese
powder mixtures were supplied by Kraft Foods (Glenview, IL). The cheese powder mix
contained some o f the following ingredients: dairy product solids, flour, whey, salt, Cheddar
cheese, butter, buttermilk, modified food starch, sodium phosphate, lactic acid, natural flavor,
color (Yellow 5 and 6) and disodium phosphate.
To prepare the macaroni and cheese
samples, 166.1 g of dry noodles were placed in 1.5 L o f boiling water and stirred
occasionally during the selected cooking times. The box-type noodles were boiled for either
8.5 minutes (fully cooked) or 4 minutes (pre-cooking time prior to MCWC treatment) and the
durum semolina noodles were either fully cooked for 12 minutes or pre-cooked for 6
minutes. The noodles to be processed by MCWC were under-cooked to reduce the adverse
effect of overheating on product quality, especially texture, during the thermal treatment.
After completion of the selected cooking time, the noodles were drained and immediately
cooled with tap water.
The cheese sauces were prepared separately by blending melted
margarine, milk, and cheese powder mix together. Preliminary testing enabled formulations
and sauce quantities for subsequent tests to be established. Three levels of cheese sauce,
which led to minimal edge burning and appropriate hydration, were selected for MCWC
treatment and sensory tests. The recipes used to prepare macaroni and cheese samples for
sensory evaluation testing are shown in Table 1. To prepare product to be thermally treated,
88
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the cheese sauce and the boiled noodles were mixed into a tray and vacuum packed with
nitrogen flushing. The dimensions and thickness of the tray (Polypropylene and EVOH,
Rexam™, Union, MO) used for packaging were 10.0 cm x 14.0 cm x 2.5 cm and 0.3 mm,
respectively.
Evaluated Samples: A popular commercial macaroni and cheese product was
selected for analysis (Table 1).
The boxed product (sample 4) was freshly prepared
according to packaging instructions. Preliminary experiments (data not included) indicated
that the boxed product noodle integrity did not hold up (i.e. original cylindrical shape became
flattened and deformed) under MCWC heating conditions, so an elbow-shaped 100% durum
semolina noodle was selected for the remaining treatments (samples 1, 2, and 3). Sample 1
was also freshly prepared and identical to sample 4 except for noodle type. Samples 2 and 3
(labeled 2X) contained twice as much cheese mix and milk as samples 1 and 4 (labeled IX).
Sample 3 differed from sample 2 in that it contained additional hydration water and was
MCWC treated.
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE 1: RECIPES FOR MACARONI AND CHEESE ENTREES PREPARED FOR
SENSORY EVALUATION
Sample No.
Treatment
Preparation Method
Recipe*
Boiling
Time
IX Semolina
Freshly prepared
62.0 g milk
12 min
39.5 g cheese mix
2X Semolina
Freshly prepared
124.0 gm ilk
12 min
79.0 g cheese mix
2X-MW
Semolina
Microwave
treated
and heated in trayd
124.0 g milk
79.0 g cheese mix
88.0
g
6 min°
hydration
water
IX Boxed
Freshly prepared
62.0 g milk
8.5 min
39.5 g cheese mix
NOTE:
aAll macaroni and cheese product contains 166.1 g of noodles (either semolina or
box-type) and 36.9 g of margarine.
bBoiling time required to cook noodles thoroughly.
cNoodle boiling time prior to MCWC treatment (pre-cooking time).
dTray was heated to 60°C in a water bath.
MCWC Heating System: The 915 MHz MCWC heating system consisted of three
major components: 1) a multiple-mode pressurized microwave heating vessel; 2) a 5 kW 915
MHz microwave generating system (Microdry Model IV-5 Industrial Microwave Generator,
Microdry Incorporated, Crestwood, KY); and 3) a circulation water heating and cooling
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
system (Figure 1). The pressurized microwave heating vessel allowed treatment of a single
meal tray at a time under certain pressure. The side wall of the pressurized microwave
heating vessel was made of a section of cylindrical aluminum tube (23 cm in diameter and 5
cm tail).
The top and bottom plates were made of Tempalux materials (Ultem
Polyetherimide Resin, Lenni, PA). The two plates had a high melting temperature (above
150°C) and were transparent to microwaves.
An appropriate pressure (30-35 psig) was
maintained within the vessel to ensure the integrity of the food during processing. Fittings
were provided to allow for continuous pressure and temperature measurements and for
concurrent circulation o f water for heating and cooling. The 915 MHz microwave system
was equipped with a circulator to protect the microwave generator from heating damage
caused by reflected power, and a directional coupler with appropriate sensors to measure
forward and reflected power. The output microwave power could be adjusted between 0.2
and 5 kW by regulating anode current to the magnetron. The temperature of the circulating
water for in-line heating or cooling of the food package was measured with a computer
program (Think & Do
Tfvf
, Gumming, GA). This program controlled modulating valves for
two plate heat exchangers heated and cooled with steam and tap water, respectively.
MCWC and Conventional - Heating Procedures: The major advantage of
microwave energy is its fast heating due to the direct coupling o f microwave energy into
dielectric materials to generate heat. But in many cases, adequate processing time is needed
to reduce target microorganism load, to obtain a desired F0 value, or to allow certain
chemical reactions such as browning to occur (Schiffinann 1986 and 1997). Product heating
can be controlled by the appropriate selection of microwave power. In this study, 0.5 kW
microwave power was used to allow a relatively short process time (5 min) and a certain
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
degree of thermal conduction to reduce possible uneven heating.
Directional
Couplers
Circulator
915 MHz
Microwave
Generator
Stirrer
Water Load
Pressurized Vessel
Optical Fiber
W ater
m
P ro d u c t
______
Control Systems
Water in
Water Out
A
FIG. 1. SCHEMATIC DIAGRAM O F T H E MICROWAVE AND CIRCULATION-
WATER C O N T R O L S Y S T E M S
Since smaller temperature differences between the starting and the desired
temperature (127°C) resulted in lower consumption of microwave energy, all products were
preheated with circulated water from ambient temperature to 75°C prior to starting the
microwave generating system.
After the product reached the desired temperature, the
microwave system was turned off, followed by a 30 sec holding period. Tap water at 20°C
was circulated to cool the product until the center o f the sample reached 80°C, at which point
the product could be reintroduced to atmospheric pressure without damaging the package.
The product was removed from the pressurized microwave heating vessel and immediately
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
placed into an ice-water bath.
For conventional retort heating, the product was packaged in the same trays used for
the MWCW process and placed in the pressurized microwave heating vessel, but only
circulating water at 121°C was used to treat the sample under the same over-pressure used for
the MWCW process. The processing time was around 40 minutes to provide a F0 value of 7.
Both the circulation water temperature and the cold spot temperature o f the product
were measured using optical fiber sensors (Fiso Technologies, Sainte-Foy, Canada). For the
MCWC treatment, the location of the cold spot, 2.3 cm off the geometric center of the tray,
was determined by a chemical marker technique used in preliminary experiments (Lau 2000),
while the cold spot during conventional retort heating is known to be located at the geometric
center of the tray (Lopez 1987). By monitoring the product temperature through the data
acquisition program, an integrated sterilizing value (F0) using the General Method was
calculated (Lopez 1987):
F , J \ l t i T-T')lzdt
o
(I)
where F0 is the sterilizing value(min.), T is the actual temperature of the product(°C), Tr is
the reference temperature 121.1°C, the z-value is 10°C, and t is heating time(min.).
Sensory Evaluation Studies: MCWC treated macaroni and cheese samples were
evaluated and compared to freshly made product by both a trained panel (n=10) and an
acceptance panel (n=l 15) within two days of being processed. Participation in this study was
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
voluntary and judges were compensated for their time. All panelists signed a Washington
State University (WSU) human subject approval form prior to the evaluation and completed
a questionnaire regarding frequency of consumption, buying interests and demographic
information. Four different treatments (one MCWC treated and three freshly made controls,
Table 1) were evaluated by both panels in duplicate. Tasting sessions were spread over three
different days to prevent errors due to sensory fatigue. Consumer panelists were encouraged
but not required to participate in each of the sessions; trained panelists attended all three
sessions. The macaroni and cheese samples were placed in clear glass containers that were
held for no more than 30 minutes in a 65°C water bath. Freshly prepared macaroni and
cheese was cooked approximately every thirty minutes and placed in identical jars for
serving.
All evaluations took place in individual booths equipped with controlled
incandescent lighting.
Panelists were encouraged to drink water in between samples to
cleanse their palates. The samples were labeled with randomly generated three-digit codes
and presented one at a time using a randomized complete block design with repeated
measures. Some imbalance may have occurred due to the fact that not all samples were
tested on each day by every person.
Descriptive analysis (Stone and Sidel 1992) was used to describe the qualitative and
quantitative characteristics of the selected macaroni and cheese treatments.
Judges with
previous experience in sensory evaluation were chosen among WSU employees and students.
They were trained in six sessions to discuss and select a list of descriptors to characterize the
appearance, aroma, flavor and texture attributes of the macaroni and cheese samples. They
also defined a standard evaluation procedure using a 9-point scale. Performance evaluation
was initiated after the first three training sessions had been completed to identify problems
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
among individual panelists or misunderstandings in the use of the descriptive terminology.
Faculty, staff and student volunteers with or without previous experience in sensory
evaluation participated in one or more of three consumer-panel sessions. Panelists were
asked to indicate their perception of appearance, aroma, flavor, texture and overall
acceptance for each o f the samples tested using a 9-point hedonic scale. Space for comments
was provided on the evaluation ballots.
Statistical Analysis: Mean scores were analyzed with an analysis of variance
(ANOVA) based upon a randomized complete block (panelists) design with two repeated
measures for each panelist. Trained and untrained panelists were the units of replication and
panelists were treated as random effects.
Differences were considered significant when
resultant p-values were <0.05 (SAS/STAT®, 1989). When the ANOVA was significant, the
Least Significant Difference (LSD) method was used to separate treatment means.
RESULTS AND DISCUSSION
Effect of MCWC and Conventional Retort Heating on Processing Time: Figure 2
shows the temperature profiles o f 2X-sauce macaroni and cheese entrees during MCWC and
conventional retort heating conditions. The corresponding integrated sterilizing values (F0)
were very close to 7 min. Under MCWC heating, the product reached 129.1°C within 8 min,
while it took more than 35 min for the same product to reach 119.1°C under conventional
retort heating. Preliminary test indicated the shorter processing time of the MCWC treatment
led to considerable reductions in color degradation and loss of texture and flavor quality.
The difference in quality obtained with the two heating methods was extreme. Upon visual
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
inspection, the researchers agreed that the poor quality obtained by conventional retort
heating made it unnecessary to include this sample in the sensory evaluation protocol.
Compared to the microwave processed product, the retorted macaroni and cheese was darker
throughout the tray, had severely burnt edges and appeared extremely dry. For this reason,
only MCWC treated samples were tested and compared with freshly cooked macaroni and
cheese controls.
120
oo 100
£
3
£0)
Q.
E
©
H
0
10
5
15
20
25
30
35
40
45
Time(min)
—o— circulation w ater terrperature during MCWC heating
— h r—
product terrperature during MCWC heating
—
circulation water during conventional retort heating
—jfc— product terrperature during conventional retort heating
FIG. 2. TYPICAL TEMPERATURE HISTORY OF MACARONI & CHEESE AND
CIRCULATED WATER DURING MCWC AND CONVENTIONAL RETORT
HEATING
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Effect of Product Formulation on Heating History: Figure 3 shows representative
temperature profiles o f two macaroni and cheese entrees treated by MCWC heating at a fixed
microwave power of 0.5 kW. During preliminary experiments, additional cheese sauce or
cheese powder mix and thickeners or stabilizers, such as instant starches and gums, were
added to the macaroni and cheese formula for MCWC treatment. In microwave heating, the
ability o f materials to absorb electromagnetic energy depends to a great extent on the
dielectric properties o f the product being heated (Buffier and Stanford 1991; Thostenson and
Chou 1999), which mainly depend on moisture and salt contents (Mudget 1986). The change
in formulation did affect the moisture and salt content of the product, however, due to the
relatively small size of the tray, the change in dielectric properties was compensated for by
the heating provided by the circulation water.
As shown in Figure 3, the change of
formulation had no significant effect on the MCWC heating history at similar Fg levels, but
its impact on the sensory quality of the final product was considerable (Table 2).
120
o
100
80
60
40
20
0
2
4
6
8
10
12
14
16
Time (min)
—O— circulation water temperature during 2X sample processing
—□— 2X sample temperature
■A
•
'
circulation water temperature during 1Xsample processing
1X sample temperature
FIG. 3. REPRESENTATIVE TEM PERATURE PROFILES OF TWO MACARONI
AND CHEESE ENTREES DURING MCWC TREATMENT
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE 2. MEAN SCORES OF SELECTED MACARONI AND CHEESE QUALITY
ATTRIBUTES JUDGED BY A TRAINED PANEL (N=10)
Flavor
Aroma
Texture
Attribute
IX
2X
2X-MW
IX-boxed
Noodle Integrity
6.20bc
6.60c
5.69b
4.15s
Adhesiveness
2.90a
2.60a
4.62b
4.20b
Shine
5.55°
6.8Gd
3.17a
4.30b
Color
4.47b
4.50b
5.20°
3.30a
Amt. of Sauce
4.30a
6.40b
3.50a
3.60a
Smoothness of Sauce
5.75b
6.55°
1.83a
3.85b
Cheddar Flavor
3.85b
5.85°
5.15°
2.45a
Buttery
5.85
5.20
4.63
4.65
Starchy Flavor
4.05ab
3.05a
3.79ab
5.25b
Off-Flavor
1.37
1.50
1.85
1.33
Cheddar Aroma
5.00b
5.40b
4.38ab
Starch Aroma
4.10ab
3.90a
4.87ab
5.25b
Off-Aroma
1.20ab
o
©
Category
Appearance
1.48ab
1.70b
Firmness
6.85°
6.10°
4.60b
2.80a
Chewiness
6.40°
5.60°
4.40b
3.20a
Stickiness
4.60b
3 25a
4.72b
4.65b
Moistness
4.55a
7.30b
3.88a
4.70a
Graininess
3.65b
2.60a
4.65°
3.60b
Mouthfeel/coating
2.75b
1.85a
2.67ab
2.80b
■
Note: a, b, c, d: Different superscripts indicate significant di Terences exist across rows.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.30a
Sensory Evaluation: (1) Trained Panel: Mean scores for selected aroma,
appearance, flavor and texture attributes given by a trained panel are presented on Table 2.
The scores were significantly different for most of the attributes evaluated. The amount of
cheese sauce in each recipe had a direct impact on shine, amount of sauce and smoothness of
sauce, as indicated by the significantly higher scores given to the freshly cooked 2XSemolina sample compared to the two IX samples for those attributes. In addition, when
more free cheese sauce was present in the final product, better noodle integrity was noted,
and vice versa.
Likewise, noodle type had an effect on noodle integrity and on color.
Specifically, the semolina samples scored consistently higher for noodle integrity and were
darker than the boxed product.
The appearance attributes most affected by MCWC
processing were noodle integrity (initial cylindrical shape became flattened), adhesiveness
(noodles tended to stick to each other), shine (product appeared more dull), color (became
darker), and smoothness of sauce (appeared somewhat curdled). The amount of free cheese
sauce present was lower in the MCWC treated sample than in the rest of the samples due to
greater water absorption during the extra processing time.
As expected, Cheddar flavor was highest for the freshly cooked 2X and the 2X-MW
treatments since the amount of cheese powder mix in the sauce was doubled. The trained
panel also identified starch flavor as being more pronounced in the boxed product, but no
significant differences between treatments were found for the attributes buttery and offflavor.
The MCWC process had a negative impact on aroma, since the 2X-MW samples
received a lower Cheddar aroma score than the IX-Semolina samples in spite of containing
twice as much cheese as the IX-Boxed sample. Both the starch and off-aroma attributes
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
were more intense in the MCWC treated sample than in the freshly cooked semolina
samples, but not as high as in the boxed product. Since the composition of the cheese sauce
in the boxed and the IX-Semolina products was identical, and significantly lower flavor and
aroma scores were associated with the boxed product compared to IX-Semolina, either the
type of noodle used or adverse noodle-sauce interactions explain the differences.
Noodle type and amount o f cheese sauce present also influenced texture attributes.
The 2X freshly cooked semolina sample received the highest moistness scores, while
stickiness and graininess were significantly lower than for the other three products. Noodle
type affected firmness and chewiness scores, which were highest for the freshly cooked
semolina samples, followed by the microwave treated semolina, and lowest for the boxed
product. As illustrated by the texture results, the main effect of the MCWC treatment on
texture was a decrease in firmness and chewiness, and an increase in graininess felt on the
roof of the mouth while chewing.
(2) Consumer Panel: Table 3 shows the mean accep tab ility scores given to the
selected macaroni and cheese entrees by a consumer panel. On average, consumer responses
fell within the range of 5.0 (neither like nor dislike) to about 7.0 {like moderately). Overall
acceptance scores for the different macaroni and cheese treatments were like moderately for
2X-Semolina, tike slightly for iX-Semolina and 2X-MW Semolina, and neither tike nor
dislike for I X Boxed. Highest scores were consistently given to the 2X-Semolina sample for
all attributes.
The appearance of the IX-Semolina and 2X-MW Semolina samples was rated
significantly higher than the boxed product, but not as high as 2X-Semoiina, suggesting a
consumer preference for the elbow noodle, which was plumper than the boxed noodle, and
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
for extra cheese sauce. The appearance attributes that might have contributed the most to
higher consumer acceptability for the 2Z-Semolina sample-and a lower acceptability for the
boxed product are noodle integrity and amount o f sauce.
TABLE 3. Mean acceptability scores8 of macaroni and cheese samples by an untrained
consumer panel (N=115)
2X-MW
IX Semolina
2X Semolina
(sample 1)
(sample 2)
IX Boxed
Semolina
Attribute
(sample 4)
(sample 3)
Appearance
5.65°
6.73d
5.81°
5.08b
Aroma
6.41°
6.96d
5.24b
6.01°
Flavor
5.03b
5.90°
5.74°
4.95b
Texture
6.67°
7.38d
6.22b
6.15b
Overall
5.86s5
6.76°
5.69b
5.42b
a: l=highly unacceptable; 9=highly acceptable
b. c, d: Different superscripts indicate significant differences exist across rows.
A comparison o f consumer and trained data shows a direct correlation between
consumer flavor acceptability scores and the trained attribute Cheddar flavor. Low starch
flavor also correlated well with flavor acceptability. Consumer acceptability scores for the
freshly prepared and MCWC treated products (2X and 2X-MW Semolina, respectively) did
not differ significantly, indicating that MCWC processing did not adversely affect flavor. In
terms of aroma, a correlation between amount of cheese present and consumer acceptability
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
was not observed, since the 2X-MW Semolina product received the lowest aroma
acceptability score o f the four samples studied, in spite of containing additional cheese. The
effect of specific aroma attributes identified by the trained panel (Cheddar, starch- and offaroma) on consumer acceptability was not clear. As shown on Table 2, the freshly cooked
boxed product received the lowest Cheddar aroma and highest starch- and off-aroma scores.
However, consumer aroma acceptability scores for the microwave treated macaroni and
cheese were significantly lower than for the boxed product, suggesting there might have been
a change in aroma during MCWC processing that was not detected by the trained panel.
Alternatively, the consumer panelists may have detected an aroma attribute that did not fit
into any of the aroma descriptors developed by the trained panel.
The higher texture scores given to the two freshly cooked semolina noodle samples
(2X and I X Semolina) than to the boxed product also indicate a preference for the firmer
semolina noodle over the noodle typically used in boxed macaroni and cheese. Likewise,
higher texture scores for the 2X over the I X semolina samples further confirm a consumer
preference for extra cheese sauce.
The additional cooking time required for MCWC
treatment had an adverse effect on texture quality, as shown by the significantly different
texture scores for the 2X and 2X-MW semolina samples (Table 3).
But in spite of the
microwave treatment, the 2X-MW and the freshly cooked boxed products received similar
texture scores, demonstrating the superiority of durum wheat over common wheat for noodle
manufacture in applications involving severe heat treatment, such as in a ready-to-eat, shelfstable macaroni and cheese. Firmness and chewiness, which scored significantly higher for
I X and 2X freshly cooked Semolina, were the predominant attributes responsible for the
higher consumer acceptability scores received by the macaroni and cheese made with elbow-
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
shaped semolina noodles.
The consumer population sampled was composed of 87-93% who consume macaroni
and cheese (mainly boxed-type) at least once per month, suggesting familiarity with the
product presented. Therefore, it is safe to extrapolate these data to macaroni and cheese
consumers at large. No correlation between frequency of consumption and buying interest
was observed. When questioned after tasting the macaroni and cheese samples, the majority
(73%) of the respondents indicated they were at least somewhat likely to buy a ready-to-eat,
shelf-stable macaroni and cheese, with 9% who were very likely to do so.
CONCLUSIONS
The MCWC heating system treated macaroni and cheese entrees within one fourth of
time required by conventional retort methods to attain the same sterility (Fq value with 7
minutes). Changes in formulation, such as amount of cheese sauce and noodle type, affected
the quality and acceptability of MCWC treated macaroni and cheese entrees. Although the
MCWC treated product received lower mean consumer acceptability scores than freshly
cooked products made with the same noodle type, the MCWC entrees received higher scores
than a popular boxed macaroni and cheese.
Combined with consumers’ indication of
willingness to purchase shelf-stable macaroni and cheese, these findings show the potential
for success of this process.
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
The authors would like to acknowledge the financial support provided by the U.S.
Aarmy Soldier Support Center and Kraft Foods. We would also like To thank Dr. Marc
Evans for statistical guidance, and Frank Younce of WSU, Evan Turek and Dr. Ming H. Lau
of Kraft Foods, Glenview, IL, for their technical advice and assistance. The students and
university employees, especially Dr. Lloyd Luedecke, Xiaoming Liu, and Lisa Pitka, were
appreciated for their help during the tasting sessions.
REFERENCES
Ayoub, J.A., Berkowitz, D., Kenyon, E.M. and Wadworth, C.K. 1974. Continuous
microwave sterilization of meat in flexible pouches. J. Food Sci. 59,309-313.
Buffler, C.R. and Standford, M.A. 1991. Effects of dielectric and thermal properties
on microwave heating of foods. Microwave World 12(4), 15-23.
Keefer, R.M. and Ball, M.D. 1992. Improving the final quality of microwavable
foods. Microwave World. 15(2),14-21.
Lau, M.H., Tang, J., Taub, LA., Yang, T.C.S., Younce, F.L. 1998. Microwave heating
uniformity of foods during 915 MHz microwave sterilization process. Proceedings of the 33rd
Microwave Power Symposium, 78-81.
Lau, M.H. 2000. Ph.D. dissertation. Washington State University. 133.
Lopez, A, 1987. A completed course in canning and related processes. Book II.
Packaging; Aseptic processing; Ingredients. 12th Ed. pp 26, 56. The Canning Trade Inc.
Baltimore, Maryland.
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Mudget, R.E. 1986. Microwave properties and heating characteristics o f foods. Food
Technol. 40(6), 84-93, 98.
Ohlsson, T.1990. Controlling heating uniformity - The key to successful microwave
products. European Food and Drink Review. 2, 7-11.
Ohlsson, T. 1992. Development and evaluation of microwave sterilization process for
plastic pouches. Paper presented at the AICHE Conference on Food Engineering, Chicago,
March 11-12.
Meredith, R. 1998. Engineers’ handbook of industrial microwave heating. The
Institution o f Electrical Engineers. London, United Kingdom.
Ruello, J. H. 1987. Seafood and microwaves: some preliminary observations. Food
Technol.in Australia. 39, 527-530.
SAS Institute 1989. SAS/STAT® Statistical Software and User's Guide, ver. 6, 4th
Ed. Vol. 2. SAS Institute, Inc., Carey, NC.
Schifftnann, R.F. 1986. Food development for microwave processing. Food Technol.
40(6) 94-98.
Schiffinann, R.F. 1990. Problems in standardizing microwave over performance.
Microwave World. 11 (3),20-24.
Schifftnann, R.F. 1997. Microwave pasteurization of foods: an overview. Prepared
Foods. 4, 111-115.
Stanford, M. 1990. Microwave oven characterization and implications for food safety
in product development. Microwave World. 11(3), 7-9.
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Stenstrom, L.A. 1972. Taming microwaves for solid food sterilization. Paper 7.4
presented at the IMPI symposium. Ottawa, Canada. May 24-26.
Stone, H. and Sidel, J.L. 1992. Descriptive analysis. In Sensory Evaluation Practices,
Second Ed. (S.L. Taylor, ed.) pp. 202-242, Academic Press, Inc., San Diego.
Thostenson, E.T. and Chou, T.W. 1999. Microwave processing: fundamentals and
applications. Composites: Part A 30, 1055-1071.
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5
DIELECTRIC PROPERTIES OF MASHED POTATOES RELEVANT TO
MICROWAVE AND RADIO FREQUENCY PASTEURIZATION AND
STERILIZATION PROCESSES
Guan D, Cheng M, Wang Y and Tang* J
Department of Biological Systems Engineering
Washington State University,
ABSTRACT
Dielectric properties of mashed potatoes relevant to microwave and radio frequency
(RF) pasteurization and sterilization processes were measured using the open-ended coaxialline probe method over frequency range from 1 to 1800 MHz, and temperature range from 20
to 120 °C. Four moisture levels (81.6%, 84.7%, 85.9% and 87.8%) and three salt levels
(0.5%, 1.0% and 2.0%, added NaCl) were considered.
Dielectric constant and loss factor decreased with frequency.
Dielectric constant
increased with temperature at 27 MHz. But it changed little at 40 MHz and decreased with
temperature at 433, 915 and 1800 MHz. Dielectric loss factor increased with temperatures,
especially at 27 and 40 MHz, suggesting the necessity to provide uniform electromagnetic
field to prevent thermal runaway heating for RF sterilization heating system. Dielectric loss
factor also increased with added salt content.
Power penetration depth decreased with
temperature and frequency. Descriptive equations were developed at 27 and 915 MHz to
relate the dielectric properties of mashed potato to frequency, moisture and salt contents.
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTRODUCTION
Experimental studies on microwave and radio frequency (RF) heating o f foods started
from the forties and the sixties in the twentieth century, respectively (Kenyon and others
1971; Cathcart and others 1947). Because of the direct interaction between food materials
and electromagnetic energy, dielectric heating at the microwave and radio frequencies can be
more uniform than conventional heating (Datta and Hu 1992; Wang and others 2003a).
Several factors can still cause possible uneven temperature distribution in microwave
processed foods, such as variations in electric fields, placements of foods, and localized
absorption of the electromagnetic energy by heterogeneous constituents (Berek and
Wickersheim 1988).
The non-uniform dielectric heating can result in the survival of
microorganisms at the less heated part of the foods (Schiffinann 1990; Stanford 1990) and
thus compromise the microbiological safety of the processed foods. Many techniques have
been used in research and application to improve the uniformity of dielectric heating. These
techniques include rotating and oscillating foods, providing an absorbing medium (water) to
surround products (Stenstrom 1974; Guan and others 2002, 2003; Wang and others 2003a),
cycling power (Morris 1991), and varying frequency (Bows 1999) and phase (Bows and
others 1999).
To further develop a better uniform heating process using microwave (mostly at 915
and 2450 MHz) and RF heating (e.g., at 27 and 40 MHz), knowledge regarding the dielectric
properties (dielectric constant s’ and dielectric loss factor e” ) o f food materials is particularly
important. The dielectric properties determine how foods react to an external electric field
(Kuang and Nelson 1998). The dielectric constant (s’) reflects the ability o f a material to
store electromagnetic energy.
The dielectric loss factor (e” ) measures the ability of a
material to dissipate electrical energy as heat. Factors that affect dielectric properties of
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
materials include frequency of the electromagnetic waves, temperature, water content and
salt content (Engelder and Buffler 1991; Galema 1997). Extensive experimental data for the
dielectric properties of various foods have been reported (Calay and others 1995; Yagmaee
and Durance 2002). However, reliable values of dielectric properties for foods above 100°C
are scarce.
Ohlsson and Bengtsson (1975) measured 16 industrially prepared foods at
temperature from 40 to 140°C at 450, 900 and 2800 MHz and stated that the dielectric
properties data could not be extrapolated from lower temperatures up to sterilization region.
The effect o f compositions on the dielectric properties of food materials was not mentioned
for a single food in the above study, however. Sipahioglu and Barringer (2003) developed
the predictive equations for the dielectric properties of fruits and vegetables in the
temperature range from 5 to 130°C. But the measurements and predictions are only available
at the frequency o f 2450 MHz. The other three literatures mentioning the measurement of
dielectric properties at high temperature up to 130°C are for macaroni and cheese products,
macaroni noodles as well as whey protein gel products at microwave and radio frequency
(Nelson and others 2000, 2001; Wang and others 2003b).
Generally, there is no single
equation that can predict dielectric properties for complex foods over broad temperature and
frequency ranges used for microwaves and RF heating. Obtaining reliable data of dielectric
properties of various foods for specific purpose such as food pasteurization and sterilization
thus become indispensable.
At Washington State University (WSU, Pullman, WA, USA), a pilot-scale 915 MHz
Microwave-Circulated Water Combination (MCWC) heating system and a pilot-scale 27
MHz radio frequency (RF) sterilization system were developed for food pasteurization and
sterilization research. Both systems can shorten the processing time and help retain natural
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
texture and flavor o f macaroni and cheese products (Guan and others 2002; Wang and others
2003a). Based on suggestions from industrial partners, mashed potato was chosen as one of
the model food materials to develop microwave and RF sterilization process protocols for
packaged foods and to evaluate the effect of the processes on the processed product’s quality
and microbiological safety.
As a popular food item, mashed potatoes are relatively
homogeneous and easy to formulate, making itself a benchmark reference for future product
development in industrial microwave and RF pasteurization and sterilization processes. It is,
therefore, desirable to determine the dielectric properties of mashed potatoes at microwave
and radio frequencies as a function of moisture and salt contents, particularly over the
pasteurization and sterilizations temperature range (80-120°C). The available information is
essential for understanding the interaction between the mashed potato products and
electromagnetic fields during microwave or RF heating processes.
It can also help the
computer simulation work for guiding the design of industrial microwave or RF sterilization
systems and assisting future product development to enhance the process stability and
heating uniformity.
The objectives o f this study were to (1) measure dielectric properties of mashed
potato as influenced by frequency, temperature, moisture content and salt content; (2)
develop descriptive equations at 27 MHz and 915 MHz; (3) calculate penetration depths and
investigate their potential effect on dielectric heating.
MATERIALS AND METHODS
Mashed potato composition determination and preparation:
Instant mashed
potato flakes were purchased from Oregon/Washington Potato Company (Boardman, OR,
USA). The mean values and standard deviations of the compositions of the potato flakes
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
were calculated from three replicates. The moisture content was measured by AOAC method
(925.10, 1990), total starch content was measured by AACC method (76.13, 1983) using the
Megazyme kits (Megazyme International Ireland Ltd., Wicklow, Ireland), and ash content by
AACC method (30.25, 1983). Protein content was determined through the FP528 LECO
instrument (LECO Corporation, St. Joseph, MI, USA), for which EDTA was used as a
control, while Oxygen and Helium were used as carrier gases.
The combustion furnace
temperature was 851.84 °C and the system pressure was 793.12 mmHg. The moisture, ash,
protein and total starch contents of the mashed potato flakes are 8.25±0.04, 5.64±0.02,
11.64±0.10 and 68.6±1.90 (%, d.b.), respectively.
The potato flake supplier recommended a water/potato ratio of 5.5:1 (mass) to
formulate mashed potato with desired texture. Four mass ratios of de-ionized water/potato
(4:1, 5:1, 5.5:1 and 6.5:1) were considered in this study; and the corresponding moisture
contents are 81.6, 84.7, 85.9 and 87.8.(%, w.b.), respectively. Beyond the above range, the
samples were either runny or too coarse to prepare mashed potatoes. The tested moisture
range seemed to be relatively small, but it represents a typical range from the perspective of
preparation of mashed potato product that is acceptable for consumers’ consumption.
Information of dielectric properties of mashed potatoes in this moisture range would help
industrial processors to evaluate the possible influence of the material’s moisture deviation
when microwave and RF heating are applied to food pasteurization and sterilization. The
effect o f salt (NaCl) was investigated by adding predetermined amounts of salt into the
mashed potato flakes (water/potato mass ratio: 5.5:1) to obtain samples with three different
salt contents (0.5%, 1% and 2%). Mashed potatoes prepared from flakes were stored at 4°C
until use within two days.
Ill
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Dielectric properties measurement system:
An open-ended coaxial-line probe
technique was used to measure dielectric properties of mashed potatoes because this method
did not require any particular sample shape and offered broad-brand measurement.
A
schematic diagram o f the measurement system is shown in Figure 1.
S S S lS ll
liisilliiL
Coaxial cable
Agilent 4291B Impedance analyzer
Pressure proof test cell
IEEE-488 (GPIB) bus
High temperature hoses
f
Computer
Oil bath
Figure 1-Open-ended coaxial probe dielectric measurement system at WSU
It consisted of an RF impedance analyzer with a calibration kit (429IB, Agilent
Technologies, Palo Alto, CA), a test cell (Inner diameter: 20 mm and height: 94 mm,
stainless steel) custom-built at WSU, a high-temperature coaxial cable, a dielectric probe kit
(85070B, Hewlett Packard Corp., Santa Clara, CA) and an oil bath equipped with a
programmable circulator (Model 1157, VWR Science Products, West Chester, PA). The
112
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
system can make measurement over a frequency range from 1 to 1800 MHz and temperature
range from 10 to 130°C.
Customized software (DMS 85070, Innovative Measurement
Solutions Inc., Santa Clara, CA) was used to control the impedance analyzer and log the data.
The 429IB calibration kit including four standards (open, a short, a 500 load and a low-loss
capacitor) was used to calibrate the impedance analyzer. The 85070B dielectric probe kit
including short circuit (a gold-plated precision shorting block), open and a known load (de­
ionized water at 25°C) was used to calibrate the test probe.
After calibration, mashed potatoes were placed into the test cell. The dielectric probe
was sealed into the loaded cell and kept in contact with samples through pressure from a
stainless steel spring and a stainless steel piston during measurement.
A type T rigid
thermocouple probe (Diameter: 1.02 mm) connecting with a digital handheld thermometer
(Bamant Company, Barrington, IL, USA) was placed into the sample’s center to measure
temperature, which was maintained through the circulation of a mixture of ethylene glycol
and de-ionized water (9:1, v/v) being heated in an oil bath. O-rings were placed between the
sample holder and the cover plates to prevent any moisture loss as steam.
A detailed
description of the test cell is given by Wang and others (2003b).
Measurement of dielectric properties: Dielectric properties of each sample were
measured at 200 discrete frequencies between 1 MHz and 1800 MHz at 20, 40, 60, 80, 100
and 120°C, respectively. When measurements under one temperature were completed over
the whole frequency range, the sample was heated to the next temperature level in 12-15
minutes. Dielectric properties for mashed potato were measured in triplicate. Mean values
and standard deviations were calculated. The data for the four Industrial, Scientific and
Medical (ISM) frequencies (27, 40, 433 and 915 MHz) were reported, 27 and 40 MHz
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
allocated for RF heating applications; 433 and 915 MHz allocated for microwave heating
applications, according to the European Radiocommunications Committee (ERC) and
Federal Communication Commission (Metaxas and others 1993).
The dielectric data at
frequency o f 1800 MHz are also reported because 1800 MHz is the upper limit of the
Impedance Analyzer.
Descriptive equations development and calculation of power penetration depth:
To develop descriptive equations at 27 and 915 MHz, the response variables including
dielectric constant and loss factor were fit through regression analysis (Minitab, Minitab Inc.,
State College, PA, USA) using the least squares technique.
The equations and the
predicators included in the equations had a significance of p< 0.001. The goodness of fit was
assessed from the adjusted coefficient o f determination ( R a2dj) of the equation.
Power penetration depth is one of the essential dielectric processing parameters. It is
defined as the distance that an incident electromagnetic wave can penetrate beneath the
surface of a material as the power decreases to 1/e (e = 2.718, roughly 37%) of its power at
the surface.
The power penetration depths at 27 and 915 MHz were calculated by the
following equation (Buffler 1993):
(i)
.
2-Si4<
I
2
I ( s rr"s'
-1
Jl +
\Z
r
)
»
where c is the speed for light in free space (3 x 108 m/s), f is the temporal frequency, s'r is the
dielectric constant and sr is the dielectric loss factor. The calculated power penetration
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
depth can further indicate the effect of dielectric properties changes on the dielectric heating
process and help to select the appropriate thickness of the mashed potato to be used.
RESULTS AND DISCUSSION
Effect of frequency and temperature:
The measured values of the dielectric
properties of mashed potatoes with four moisture content levels at the five frequencies (27,
40, 433, 915 and 1800 MHz) are given in Table 1. At all moisture levels, the changes of the
dielectric properties of mashed potato with respect to temperature and frequency have similar
trends. Namely, both the dielectric constant and the loss factor decreased with frequency.
This agrees with the results reported by Pace and others (1968). Figures 2 and 3 showed the
dielectric properties o f mashed potatoes with a moisture content o f 81.6% (w.b.).
The molecular polarization arising from the orientation with an imposed electric field
is an important phenomenon contributing to the frequency dependence of the dielectric
properties o f most materials (Nelson 1978). Electric conduction and the other polarization
mechanisms such as dipole, electronic, atomic and Maxwell-Wanger all contribute to the
dielectric properties o f foods and the ionic conductivity plays a major role at lower
frequencies (e.g., below 200 MHz) (Metaxas and others 1993). When food materials have
moisture contents o f 35-40% or above, the majority of water in foods is in free form and
supposed to be the dominant component governing the overall dielectric behavior of foods
(Tran and others 1987; Sun and others 1995).
As shown in Figures 2 and 3, both the
dielectric constant and loss factor of mashed potato increased sharply with reduced frequency
and with increasing temperature, indicating that the dispersion mechanism is dominated by
ionic conductivity in free water. Figure 2 and Table 1 also show that the dielectric constant
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1-Mean ± standard deviation (three replicates) of dielectric properties for
mashed potatoes with different moisture content (%, w.b.), no salt added
Sample
81.6%
T (°C )
20
£'
27MHz
87.5±1.0
e"
328.2±11.5
40
£'
60
£'
80
£"
£'
100
£'
120
£’
£"
£"
84.7%
20
£'
£"
40
£'
60
£"
£'
80
E"
£'
£"
100
£ ’
120
£'
£"
£"
85.9%
20
£'
£"
40
£'
60
£'
£"
80
£'
£"
£"
87.8%
100
£'
120
£"
Z
20
40
£a
£'
e"
£'
60
e"
£'
£"
80
£'
£"
100
£'
£"
120
£'
£"
89.111.2
459.0116.1
90.712.1
605.0119.2
94.013.1
783.4115.4
100.714.3
1001.8118.4
113.217.1
1340.0160.3
88.613.7
297.5+11.9
89.115.7
409.8122.3
89.917.3
541.0131.9
91.3+9.2
689.7147.5
96.1110.6
895.9143.5
102.8112.1
1153.8149.2
89.2+7.2
276.6129.9
89.818.1
382.9142.4
90.518.0
513.6142.0
92.519.7
660.1163.9
96.8+9.9
839.9161.8
102.1110.5
1060.1+60.6
88.215.5
277.4123.2
89.014.9
380.9131.4
89.315.7
497.7144.4
91.9+5.3
638.4161.7
98.817.9
846.3177.8
104.918.0
1066.01105.0
40 MHz
80.412.4
224.417.3
78.3+1.3
312.8111.2
79.210.6
410.4112.4
79.310.7
530.319.6
82.511.9
678.1+11.0
88.614.1
906.3+42.4
82.411.4
203.617.0
81.212.7
279.6113.6
79.513.6
367.9119.5
78.414.3
468.4129.2
79.517.7
607.5125.8
81.6110.1
782.4129.1
82.2+4.9
188.8120.8
81.015.0
260.3129.5
79.414.2
348.2128.7
78.514.4
446.9143.3
79.013.6
567.9142.0
80.013.2
716.6+41.1
81.314.4
189.3114.3
79.8+3.2
259.2119.9
77.7+3.4
337.9128.2
77.3+2.3
432.7+39.8
79.4+3.0
573.1+70.4
80.912.7
721.8168.2
433 MHz
60.715.3
33.114.5
59.712.7
38.613.9
57.812.4
46.214.2
56.211,5
56.112.7
55.011.3
68.611.4
54.611.5
88.512.9
68.710.5
26.613.3
65.510.3
32.913.4
62.3+0.2
40.113.2
58.910.4
48.913.8
56.111.7
62.715.5
53.6+2.6
78.815.8
68.1+2.1
32.113.2
67.0+1.6
32.615.0
63.711.5
39.914.7
60.812.0
47.715.3
58.211.0
58.315.5
55.410.4
71.714.8
68.7+6.6
27.115.8
66.714.3
29.510.9
62.7+4.2
36.610.6
60.312.3
43.3+2.9
58.512.0
55.9+5.9
56.111.7
69.8+5.5
915MHz
1800MHz
58.412.8
56.417.6
22.3+3.6
15.810.5
56.8+3.4
57.3+1.0
23.713.4
15.610.8
55.3+2.5
55.510.6
26.1+3.7
16.111.0
53.612.0
53.011.0
30.713.2
17.810.9
52.811.7
■ 51.510.9
35.9+2.1
20.210.6
52.511.0
50.411.3
44.4+1.2
24.410.6
66.310.3
63.710.1
18.9+3.8
15.712.1
63.4+0.0
61.410.1
20.613.6
14.812.0
60.410.4
58.6+0.2
22.813.0
14.911.7
57.010.6
55.410.4
26.4+3.1
16.011.8
53.712.1
53.011.1
33.815.7
18.912.5
51.1+3.1
50.9+1.4
41.516.0
22.512.6
64.211.4
65.8+1.9
27.113.0
16.310.7
65.7+0.7
63.710.6
22.814.5
14.4+1.5
62.510.8
60.710.6
25.214.2
14.711.3
59.911.5
58.011.0
27.614.0
15.411.5
57.310.7
55.510.5
32.014,4
17.511.6
54.510.4
52.810.1
38.113.8
20.111.3
66.018.6
65.614.9
14.811.4
20.318.1
65.516.0
63.714.5
12.910.4
18.613.1
61.6+5.7
60.513.8
21.212.8
•13.010.3
59.613.0
57.712.5
22.6+0.9
13.510.7
57.712.7
55.812.1
28.212.6
15.911.4
55.3+2.3
53.611.8
34.912.3
19.011.2
116
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28Q
240
40°C
200
100“C
160
120°C
120
100
1000
10000
Frequency (MHz)
Figure 2-Change of dielectric constant of mashed potatoes (moisture content:
81.6%, w.b.; no added NaCl) with frequency at 6 temperatures
3500
3000
2500
2000
1500
1000
500
100
1000
10000
Frequency (MHz)
Figure 3-Change of dielectric loss factor of mashed potatoes (moisture content:
81.6%, w.b.; no added NaCl) with frequency at 6 temperatures
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
decreased with frequency at each temperature. But below a certain frequency, the higher the
temperature, the larger the dielectric constant; above that frequency, the dielectric constant is
declining with increasing temperature.
A similar phenomenon was observed for whey
protein gel products with moisture content of 74% (w.b.) and was explained as the dielectric
relaxation shifting from the low frequency to higher frequencies as temperature increased
(Nelson and others 2001).
Temperature dependence reflects dielectric relaxation processes operating under
certain conditions; the dielectric constant and dielectric loss factor for polar materials can
either increase or decrease with temperature (Nelson and others 1990). It seems that there is
no significant difference among the dielectric constant data for mashed potatoes with a
moisture content o f 85.9% (w.b.) (Figure 4 and Table 1).
120
100
-
80 ■£
60 -
40 MHz
■*— 915MHz
433MHz
20
-
1800MHz
100
120
Temperature (°C)
Figure 4-Change of dielectric constant of mashed potatoes (moisture content:
85.9%, w.b.; no added.NaCl) with temperature at 5 frequencies
118
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
However, the dielectric constant gently increased when temperature increased from
20 to 120 °C at 27 MHz; it changed little at 40 MHz and decreased with temperature at the
other observed frequencies including 433, 915 and 1800 MHz (Table 1). This is similar to
the results from Wang and others (2002b) who reported that dielectric constants of whey
protein gels, liquid whey protein mixture, cooked macaroni noodles and cheese sauce
increased in the RF range (27 MHz), but decreased in the microwave range (915 MHz and
1800 MHz).
Feng and others (2002) also reported that the dielectric constants of red
delicious apples (Malus domestica Borkh) decreased with temperature in the microwave
frequency range (915 and 1800 MHz) when the moisture content was greater than 70%.
Figure 4 also indicates that certain dielectric constants at 27 MHz and 40 MHz were greater
than 80. The relatively higher readings of the dielectric constant were attributed to the ions
at the face of the probe (Wang and others 2003b). Sheen and others (1999) attributed this
phenomenon to poorly conditioned calibration at lower frequencies. Dielectric constants of
certain biological tissues and agricultural products greater than 80 over a frequency range
from 1 to 200 MHz were also reported by Stuchly and others (1980).
The measured values of the loss factor as functions of temperature and frequency are
shown in Figure 5 and Table 1. The loss factor for mashed potatoes increased sharply when
the temperature increased from 20 to 120°C at 27 and 40 MHz. But this sharp increase was
not obvious at 433, 915 and 1800 MHz. For example, the loss factors of mashed potatoes at
27 MHz increased from 276.6 to 1060.1 (relative increase: 280%); while the corresponding
change at 915 MHz was from 27.1 to 38.1 (relative increase: 40%). That is, the loss factor
has much less temperature dependence at 915 MHz microwave frequency than at 27 MHz,
the RF frequency. This can affect the rate of energy absorption with temperature under a
119
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
given electric field inside the product, based on the following relationship:
pav - 2 7 te J s rE 2
W atts/m 2
(2)
where f is the frequency (Hz), e0 is the permittivity o f the free space (8.854 x 10"12 F/m), s"r is
the dielectric loss factor and E is the electric field intensity inside the product (Volt/m)
(Metaxas and others 1993). It suggested that there is a larger tendency for mashed potatoes
to experience thermal runaway heating at 27 MHz than at 915 MHz. Indeed, it is extremely
critical to provide uniform electric field strength for RF pasteurization and sterilization
applicators to minimize the thermal runaway heating and ensure system stability.
1200
1000
-
27MHz
40 MHz
- x - 915MHz
433MHz
800 -
1800MHz
600 400 -
200 ft
100
120
Temperature (°C)
Figure 5-Change of dielectric loss factor of mashed potatoes (moisture content:
85.9%, w.b.; no added salt) with temperature at 5 frequencies
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Effect of moisture content: In general, the dielectric constant changed little within
the moisture content range at a given temperature and frequency; but increasing moisture
content reduced the loss factor at all measuring conditions (Table 1). Figure 6 is a typical
example regarding the effect of moisture content on the loss factor under different
frequencies and shows that the effect is less significant at microwave frequency (433, 915
and 1800 MHz) than that at RF range (27 and 40 MHz).
1400
1200
-
1000
-
800 40 MHz
27MHz
433 MHz
•*— 915MHz
■ *- 1800MHz
200
-
0.81
0.82
0.83
0.84
0.85
0.86
0.87
0.88
0.89
Mositure content
Figure 6-Loss factor of mashed potatoes as affected by moisture content (120°C)
The effect o f moisture content on the dielectric properties of hygroscopic materials
strongly depends upon the form of water in the materials.
Mudgett and others (1980)
reported the dielectric behavior at microwave frequencies was governed by three distinct
moisture content ranges. Namely, below 10% moisture content, the bound water, contributes
121
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to the low dielectric activity of materials; when increasing the moisture level to intermediate
level (10-35%), the availability of loosely bound or free water cause the dielectric constant
and loss factor to increase rapidly with the contribution from ionization of bound salts; above
35%, the dielectric loss factor shows little dependence on moisture content. Nelson (1978)
stated that it was possible for the dielectric loss factor to decrease with moisture content. In
the test range of this study, the influence of reduced moisture content to increased loss factor
may be contributed to the increase in ionic conductivity.
Effect of salt content (added NaCl): Measured values of the dielectric properties of
the salted mashed potatoes (moisture content 85.9%; added salt levels: 0.5%, 1.0% and 2.0%)
and the non-salted sample (moisture content: 85.9%) are given in Table 2. Similar trends for
the changes with respect to temperature and frequency are observed for both the salted and
non-salted mashed potatoes (Figure 2, 3 and Table 2). According to Table 2, added salt did
not significantly influence the dielectric constant any frequency including 27 MHz and 915
MHz, however, addition of salt sharply increased the corresponding loss factor. Both the
absolute increases and relative increases in loss factor were larger at 27 MHz than at 915
MHz. For example, at 27 MHz, the absolute increase in loss factor between 1% salted
mashed potato and non-salted samples is 436.7 at 20°C, representing a relative increase of
157.9%.
The absolute increase is 2065.1 at 120°C, representing a relative increase of
194.8%. At 915 MHz, the absolute and relative increases are 1.2 and 0.04% at 20°C; the
corresponding values are 57.1 and 149.6% at 120°C. An increase in the amount of added salt
from 0.5% to 2.0% also enhanced the loss factors for a given temperature and frequency, as
shown in Figures 7 and 8 for the frequencies of 27 MHz and 915 MHz, respectively.
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2- Mean ± standard deviation (three replicates) of dielectric properties for
mashed potatoes with different salt levels (mass ratio), moisture content: 85.9% (w.b.)
27MHz
T (°C)
0%
20
40
60
80
100
120
0.5%
20
40
60
80
100
120
40 MHz
433 MHz
915MHz
1800MHz
£'
89.2±7.2
82.2±4.9
68.1±2.1
64.2±1.4
65.8±1.9
S"
276.6±29.9
188.8±20.8
32.1 ±3.2
27.1±3.0
16.3±0.7
t'
89.8±8.1
81.0±5.0
67.0±1.6
65.7±0.7
63.7±0.6
t"
382.9±42.4
260.3±29.5
32.6±5.0
22.8±4.5
14.4±1.5
£'
90.5±8.0
79.4±4.2
63.7±1.5
62.5±0.7
60.7±0.6
£"
513.6+42.0
348.2+28.7
39.9±4.7
25.2±4.2
14.7±1.3
S'
92.5±9.7
78.5+4.4
60.8±2.0
59.9±1.5
58.0±1.1
E"
660.1+63.9
446.9±43.3
47.7±5.3
27.6±4.0
15.4±1.8
S'
96.8±9.9
79.0+3.6
58.2±1.0
57.3±0.7
55.5±0.5
Z"
839.9±61.8
567.9+42.0
58.3±5.5
32.0±4.4
17.4±1.6
S'
102.1+10.5
78.0±3.2
55.4±0.4
54.5±0.4
52.8±0.1
s"
1060.1±60.6
716.6±41.1
71.7±4.8
38.1±3.8
20.1 ±1.3
s'
88.2+1.7
81.9±2.6
68.4±2.6
66.0±2.5
64.3+3.5
s"
583.0±38.2
393.4+25.4
41.2±2.7
24.0±1.9
17.5±0.1
£'
88.5±1.2
79.9±2.2
64.7±3.0
62.5+2.9
61.1±4.2
s"
817.1±59.4
550.3±39.3
54.8±4.0
29.4±2.4
18.4±0.8
s'
90.1+0.2
79.1±1.2
61.8±2.5
59.6±2.5
58.6+3.5
s"
1097.1 ±64.3
738.5+42.6
71.5±4.2
36.7±2.5
21.4±0.6
s'
92.2±0.5
78.4±0.8
58.9±2.2
56.6±2.4
55.6±4.0
s"
1416.9±80.4
952.9±53.1
90.7±5.0
45.3±2.8
25.4±0.9
s'
95.4±1.6
78.7±1.9
56.5±2.2
54.3±2.2
52.9±3.2
t"
1782.5±92.9
1197.8±61.7
112.8±5.8
55.5±3.1
30.4±1.7
€
102.8±1.1
81.5±1.0
54.6±0.5
52.3±0.8
53.0±0.9
€"
2238.3+42.5
1507.3±28.5
140.8±2.8
68.5±1.7
36.0+2.1
(Continued)
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Continued)
27MHz
T (° C )
1.0%
20
£'
78.2±6.1
e" 713.3+112.7
40
60
120
* 2 .0%
20
40
60
80
100
915MHz
1800MHz
71.415.4
56.9+5,0
55.114.3
53.515,0
480.5175.3
49.4+6.8
28.413.0
19.411.4
79.4+5.0
70.4+4.1
54.713.4
52.812.8
51.7+3.5
£"
999.6±118.9
672.3+79.8
66.3+7.2
35.613.1
21.5+1.2
£'
79.9±3.0
68.612.6
51.5+2.2
49.411,8
48.312,9
878.2162.7
84.415.7
43.412.5
24.7+1.1
67.5+0.3
48.010.1
46.110.4
.45.2+1.0
1084.5+20.0
102.8+0.3
51.710.5
28.7+0.4
£'
81.6±0.4
£" 1614.4±27.1
100
433 MHz
£'
£" 1306.7±93.8
80
40 MHz
£'
94.1+2.7
74.3+1.6
48.610.6
46.7+0.2
46.311.7
£"
2225.4±48.9
1498.1132.7
140.512.8
69.310.9
37.5+0.5
£'
112.2±5.8
84.313.9
50.8+1.3
48.711.0
48.812.9
£"
3125.2±74.6
2104.4+50.5
195.6+4.6
95.211.8
50.7+1.2
£'
100.6±0.8
87.910.3
65.210.3
62.410.4
59.4+0.4
£"
1405.3±22.9
945.5+15.6
95.712.7
52.4+2.2
32.4+1.5
£'
104.5+0.4
87.910.4
62.411.2
59.710.9
57.9+0.9
£"
1905.7±73.3
1280.0149.1
125.015.8
65.0+3.7
37.111.4
£'
109.411.0
88.011.6
59.0+2.4
56.0+2.0
55.8+0.2
£ ”
2524.4156.0
1699.3137.7
162.214.9
82.2+3.4
45.111.5
£'
120.2+0.0
91.711.1
56.7+2.7
55.111.8
54.8+0.5
£"
3352.1+124.2
2257.1183.8
212.219.3
105.3+5.7
56.112.6
£'
133.510.9
97.211.6
54.513.0
53.0+1.8
53.612.5
£"
4197.4154.3
2822.136.6
263.815.2
129.014.1
69.914.2
*At 120°C the dielectric data for mashed potatoes were beyond the measurement range.
124
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4500
4000
0.50%
3500
1%
3000
2%
2500
2000
1500
1000
500
100
120
Temperature (°C)
Figure 7- Added salt and the dielectric loss factor of mashed potatoes (moisture
content: 85.9%) at 27 MHz
4500
4000
0.50%
3500
1%
3000
2%
2500
2000
1500
1000 ------500
100
120
Temperature (°C)
Figure 8- Added salt and the dielectric loss factor of mashed potatoes (moisture
content: 85.9%) at 915 MHz
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In general, salt or dissolved ions caused a reduction in polarization o f water and a
decrease in the overall dielectric constant by binding water.
Salts also increased the
dielectric loss factor above that of pure water because of electrophoretic migration (Mudgett
1985). The above influences were directly related to the nuclear charge effect that depends
on the size and charge of dissolved ions (Bircan and others 1998). Galema (1997) reported
that the presence o f electrolyte (NaCl) did not seem to influence the dielectric constant
greatly while it did have a marked effect on the dielectric loss factor, which was observed in
this study.
Descriptive equations and power penetration depth:
The related descriptive
equation for dielectric properties of mashed potatoes as affected by temperature, moisture
content and salt content (ash content plus added salt content) are shown in Table 3 at the
frequencies of 27 and 915 MHz. Each of the equations has ani?^. equal to or above 0.90
except the one for the dielectric constant at 915 MHz. The calculated data differed from the
measured values by less than 20%.
especially at 20°C.
Discrepancy occurred mostly at lower temperature,
The equations indicate that both temperature and salt content play
important roles in the dielectric properties of mashed potatoes.
126
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3-Descriptive equations for the dielectric properties of mashed potato
27 MHz
D ielectric constant e ’
s ’ = 721 - 196 S + 0.00269 T 2 - 0.0113 T*M + 0.131 T*S + 14.9 S2
( R ; dJ
Dielectric loss factor e”
=0.90)
s” = 443 - 96.0 T + 0.0807 T 2 + 16.4 T*S
( < = 0.98)
915 MHz
Dielectric constant e’
e’ =
205 +1.26 M - 74.1 S - 0.00127 T*M + 5.42 S2
( ^ ■ = 0.79)
Dielectric loss factor e”
e” « 273 - 1.70 T- 79.2 S + 0.00310 T 2- 0.0122 T*M + 0.440 T*S + 6.29 S2
(K a
= «•»)
Note: T = temperature (°C), M = moisture content (%, w.b.),
S = total salt content (%, d.b.) =added salt (%, d.b.) + ash content (%, d.b.)
Tables 4 and 5 list the calculated power penetration depth for mashed potatoes of four
moisture levels (no added salt) and four salt (added) levels (moisture content: 85.9%, w.b.).
The power penetration depth generally decreased with increasing temperature and frequency.
No significant change appeared in penetration depth caused by moisture content (Table 4).
Increasing salt concentration reduced the power penetration depth (Table 5). Salt content
changed the dielectric properties and resultant power penetration depth of electromagnetic
waves at 27 MHz and 915 MHz in the mashed potatoes, while moisture content affected
little. This indicated that the salt content instead of moisture content should be carefully
monitored during the preparation of mashed potatoes for sterilization using RF or microwave
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
heating. Microwave heating is advisable for packages with relatively smaller dimensions
whereas RF heating can be applied for packages and trays with large institutional sizes.
Table 4-Power penetration depth (mm) for mashed potatoes at different moisture
content (w .b .), n o salt added
Moisture content
(81.7%)
20 °C
40 °C
60 °C
80 °C
100 °C
120 °G
27MHz
40 MHz
433MHz
915MHz
78.7
64.3
54.8
47.4
41.5
35.6
67.2
54.0
45.9
39.5
34.4
29.4
26.8
23.1
19.4
16.2
13.6
11.1
17.9
16.9
15.3
12.9
11.1
9.2
Moisture content
(84.7%)
20 °C
40 °C
27MHz
40 MHz
433MHz
915MHz
1800MHz
84.0
68.8
58.4
50.9
44.1
38.5
72.1
58.3
49.0
42.4
36.6
31.8
35.0
27.9
22.7
18.6
14.7
12.1
22.7
20.4
18.1
15.3
11.8
9.6
13.6
14.2
13.7
12.5
10.4
8.6
Moisture content
(85.9%)
20 °C
40 °C
60 °C
80 °C
100 °C
120 °C
27MHz
40 MHz
433MHz
915MHz
1800MHz
88.1
71.8
60.2
52.2
45.7
40.3
75.9
61.0
50.6
43.6
38.0
33.3
29.1
28.4
23.0
19.2
15.9
13.1
18.8
18.8
16.7
15.0
12.8
10.6
13.3
14.8
14.1
13.2
11.5
9.8
Moisture content
(87.8%)
20 °C
27MHz
40 MHz
433MHz
915MHz
1800MHz
87.8
71.9
61.3
53.2
45.6
75.6
61.0
51.5
44.3
37.8
33.2
34.4
31.3
24.8
20.9
16.5
13.5
21.1
23.0
19.6
18.2
14.4
11.6
14.6
16.5
15.9
15.0
12.6
10.4
60 °C
80 °C
100 °C
120 °C
40 °C
60 °C
80 °C
100 °C
120 °C
40.2
1800MHz
. 12.9
13.0
12.4
11.0
9.6
7.9
128
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5-Power penetration depth (mm) for mashed potatoes (moisture content
85.9%, w.b.) at different salt contents and frequencies
0% added salt
20 °C
27MHz
88.1
71.8
60.2
52.2
45.7
40.3
40 MHz
75.9
61.0
50.6
43.6
38.0
33.3
433MHz
915MHz
29.1
28.4
23.0
19.2
15.9
13.1
18.8
18.8
16.7
15.0
12.8
10.6
0.5% added salt
20 X
40 °C
60 °C
80 °C
100 °C
120 X
27MHz
55.8
46.2
39.3
34.3
30.4
27.0
40 MHz
47.2
38.7
32.8
28.5
25.2
22.3
433MHz
23.1
17.4
13.6
11.1
9.3
7.9
915MHz
17.9
14.4
11.4
9.3
7.6
6.3
1800MHz
12.3
11.4
9.6
8.0
6.6
5.6
1% added salt
20 °C
40 X
60 X
80 X
100 X
120 X
27MHz
49.5
41.1
35.7
31.9
27.1
22.8
40 MHz
41.5
34.3
29.6
26.4
22.4
18.8
433MHz
18.1
13.9
11.3
9.6
7.8
6.3
915MHz
14.1
11.2
9.1
7.7
6.1
4.8
1800MHz
10.2
9.0
7.7
6.5
5.1
4.0
2% added salt
20 X
40 X
60 X
80 X
100 X
27MHz
34.6
29.4
25.4
22.0
19.6
40 MHz
28.8
24.4
21.0
18.1
16.2
433MHz
11.0
8.9
7.3
6.1
5.3
915MHz
8.4
6.9
5.6
4.6
4.0
1800MHz
6.5
5.7
4.7
3.9
3.2
40 °C
80 °C
80 X
100 °C
120 °C
1800MHz
13.3
14.8
14.1
13.2
11.5
9.8
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C O N C L U S IO N S
The dielectric constant (s’) and loss factor (s” ) decreased with Increasing frequency.
Dielectric constant gently increased with temperature at 27 MHz, but it changed little at 40
MHz and decreased with temperature at microwave frequencies (e.g., 433, 915 and 1800
MHz).
Dielectric loss factor increased with temperatures.
Addition of salt in mashed
potatoes increased the dielectric loss factor, especially at the RF range, but dielectric constant
changed little. The penetration depth decreased with the temperature and frequencies; it was
less dependent on moisture content (between 81.6 and 87.8%) than added salt content (0.5%
to 2%). The moisture contents in the tested range did not significantly affect the dielectric
properties of mashed potatoes. Salt content instead of moisture content should be carefully
monitored during the preparation of mashed potatoes for sterilization using RF or microwave
energy.
ACKNOWLEDGEMENTS
The authors would like to acknowledge the financial support provided by the U.S.
Army Soldier Support Center (Natick, MA, USA) and Kraft Foods (Glenview, IL, USA).
We would also like to thank Oregon/Washington Potato Company (Boardman, OR, USA) for
the donation o f mashed potato flakes and Dr. Byung-Kee Balk for the composition
measurement o f potato flakes in his lab at Washington State University.
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
AACC. 1983. Approved Methods of the Association. American Association of Cereal
Chemists.
AO AC. 1990. Official Methods of Analysis. 15 ed. Washington DC: Association of
Official Analytical Chemists.
Bircan C and Barringer SA. 1998. Salt-starch interactions as evidenced by viscosity
and dielectric property measurement. J. Food Sci. 63 (6): 983-986.
Berek HE and Wickersheim KA. 1988. Measuring temperatures in microwaveable
packages. J. Packaging Technol. 2: 164-168.
Bows JR. 1999. Variable frequency microwave heating of foods. I Microwave Power
and Electromagnetic Energy. 34 (4): 227-238.
Bows JR, Patrick ML, Janes R, Metaxas A and Dibben D. 1999. Microwave phase
control heating. Int’l J. Food Sci. and Technol. 34: 295-304.
Buffler CR (Ed.). 1993. Microwave cooking and processing: 47-68. New York: Van
Nostrand Reinhold.
Calay RK, Newborough M, Probert D and Calay PS. 1995. Predicative equations for
the dielectric properties o f foods. Int’l J Food Sci. and Technol. 29:699-713.
Cathcart WH, Parker JJ and Beattie HG. 1947. The treatment of packaged bread with
high frequency heat. Food Technol. 1: 174-177.
Datta AK and Hu W. 1992. Optimization of quality in microwave heating. Food
Technol. 46(12):53-56.
Engelder DS and Buffer CR.1991. Measuring dielectric properties of food products at
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
microwave frequencies. Microwave World. 12: 6-15.
Feng H, Tang J and Cavalieri RP. 2002. Dielectric properties of dehydrated apples as
affected by moisture and temperature. Trans. Am. Soc. Agric. Eng. 45: 129-135.
Galema SA. 1997. Microwave Chemistry. Chem. Soc. Reviews 29(3): 233-238.
Guan D, Plotka VCF, Clark S and Tang J. 2002. Sensory evaluation of microwave
treated macaroni and cheese. J. Food Process. Preserv. 26: 307-322.
Guan D, Gray P, Kang DH, Tang J, Shafer B, Ito K, Younce F and Yang TCS. 2003
Microbiological validation of microwave-circulated water combination heating technology
by inoculated pack studies. J. Food Sci. 68 (4):1428-1432.
Kenyon EM, Westcott DE, Casse PL and Gould JW. 1971. A system for continuous
thermal processing of food pouches using microwave energy. J. Food Sci. 36: 289-293.
Kuang W and Nelson SO. 1998. Low-frequency dielectric properties of biological
tissues: A review with some new insights. Trans. Am. Soc. Agric. Eng. 41(1): 173-184.
Megazyme. 1998. Amylose/amylopectin assay kit: for the measurement of the
amylose and amyiopectin contents of starch. Megazyme International Ireland Ltd, Co.
Wicklow, Ireland.
Metaxas AC and Meredith RJ. 1993. Industrial Microwave Heating. London. UK.
Peter Peregrinus Ltd.
Morris CE.1991. World’s 1st microwave sterilization system. Food Engineering Int’l.
6: 63-66.
Mudgett RE, Goldblith SA, Wang DIC and Westphal W.B. 1980. Dielectric behavior
of semi-solid food at low, intermediate and high moisture contents. J. Microwave Power.
132
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15(1): 27-36.
Mudgett RE. 1985. Dielectric properties of foods. In: Microwaves in the Food
Processing Industry (R. Y. Decareau, author). Academic Press, New York, NY
Nelson SO. 1978. Electrical properties of grain and other food materials. J. Food
Process. Preserv. 2: 137-154.
Nelson SO and Kraszewski AW. 1990. Dielectric properties of materials and
measuring techniques. Drying Technol. 8(5): 1123-1142.
Nelson SO and Bartley PG. 2000. Measuring frequency- and temperature-dependent
dielectric properties of food materials. Paper of Am. Soc. Agric. Eng. (00-6096).
Nelson SO and Bartley PG. 2001. Frequency and temperature dependence of the
dielectric properties o f some food materials. Paper of Am. Soc. Agric. Eng. (01-6144).
Ohlsson T and Bengtsson NE. 1975. Dielectric food data for microwave sterilization
processing. J. Microwave Power: 10(1): 93-108.
Pace WE, Westphal W.B., Golw.b.lith SA and Van Dyke D. 1968. Dielectric
properties of potato and chips. J. Food Sci. 33: 37-42.
Schifimann RF. 1990. Problems in standardizing microwave oven performance.
Microwave World. ll(3):20-24.
Sheen NI and Woodhead DVL 1999. An open-ended coaxial probe for broad-band
permittivity measurement o f agricultural product. J. Agric. Eng. Res. 74: 193-202
Sipaghioglu O and Barringer SA. 2003. Dielectric properties o f vegetables and fruits
as a function o f temperature, ash and moisture content. J. Food Sci. 68 (1): 234-239.
133
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Stanford M. 1990. Microwave oven characterization and implications for food safety
in product development. Microwave World 11(3):7-9.
Stenstrom LA. 1974. US Patents 3,809,845; 3,814,889.
Stuchly MA and Stuchiy SS. 1980. Dielectric properties of biological substancetabluiated. J. Microwave power, 15 (1): 19-26.
Sun E, Datta A and Lobo S. 1995. Compostion-based prediction of dielectric
properties o f foods. J. Microwave Power and Electromagnetic Energy. 30 (4): 205-212.
Tran VN and Stuchly SS. 1987. Dielectric properties of beef, beef liver, chicken and
salmon at frequencies from 100 to 2500 MHz. I. Microwave Power. 22: 29-33.
Wang Y, Wig TD, Tang J and Hallberg LM. 2003a. Sterilization of food stuffs using
radio frequency heating. J. Food Sci. 68(2): 539-544.
Wang Y, Wig TD, Tang J and Hallberg LM. 2003b. Dielectric properties of foods
related to RF and microwave pasteurization and sterilization. J. Food Engr. 57: 257-268.
Yagmaee P and Durance TD. 2002. Predicative equations for dielectric properties of
NaCl, D-sorbitol and sucrose solutions and surimi at 2450 MHz. J. Food Sci. 67(6): 22072211 .
134
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 6
HEATING CHARACTERISTICS OF MASHED POTATOES IN A 915 MHZ
SINGLE-MODE MICROWAVE-CIRCULATED WATER COMBINATION (MCWC)
SYSTEM
Guan D, Liu F, Tang J*, Komarov V, Pandit R, Younce F and Patfaak S
Department of Biological Systems Engineering
Washington State University
ABSTRACT
A pilot scale 915 MHz Microwave-Circulated Water Combination (MCWC) Heating
system was used to process hermetically packaged mashed potatoes with different moisture
content and salt content. The heating patterns for mashed potatoes were obtained by an
infrared thermal camera. Similar heating situations were simulated by a Quick Wave 3D
software package using Finite-Difference Time-domain (FDTD) method.
Agreements
between experimental results and modeling ones were observed. The results indicated that
heating patterns of mashed potatoes using the 915 MHz MCWC heating system were
repeatable and could be predicated through simulation work. This study helps to determine
the locations o f cold spot(s) in the package and to justify the way of monitoring the MCWC
heating process for its future industrialization.
135
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTRODUCTION
Microwaves have been used widely in the field of food processing including drying,
defrosting, pasteurizing and sterilizing of perishable foods (Decareau 1985). Because of the
direct interaction between microwaves and food products, microwave volumetric heating can
overcome the slow heat transfer between heating media and packaged foods during
conventional heating (OMsson 1978).
However, achieving heating uniformity remains a
major challenge in developing microwave heating technologies, especially for food
sterilization, in which the product has to reach a desired temperature (e.g., 121 °C) and held at
the elevated temperature for an adequate period of time to ensure the microbiological safety
of the processed foods.
Many techniques have been used to improve the uniformity of microwave heating,
including rotating foods in a microwave cavity (O’Meara and others 1977); using absorbing
medium such as water (Guan and others 2002) around products (Stenstrom 1974; OMsson
1991; Lau 2000; Guan and others 2003a) to reduce the reflection and refraction of
microwaves; cycling microwave power (Morris 1991) and changing microwave frequency
and phase (Bows 1999; Bows and others 1999). But few published reports mentioned the
repeatability of the related heating patterns and the way to monitor the processes; both are
particularly important for microwave heating o f foods in sterilization applications.
For more than ten years, computation of electromagnetic energy distribution in
microwave ovens and loads has been used to understand the heating pattern of foods in
various types of microwave applicators. Computer modeling can reduce the cost of building
and test different design options in searching for ideal designs.
At Washington State University, a pilot scale 915 MHz Microwave-Circulated Water
136
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Combination (MCWC) heating system was developed to thermally process hermetically
packaged food products. The frequency of 915 MHz (ISM band) was chosen because it
provides longer penetration depth in foods than the frequency of 2450 MHz. In this system,
a WR975 waveguide was terminated in an oversized hom-shaped waveguide section which is
in tum connected to a rectangular cavity so that the hom-shape rectangular waveguide could
be used to heat food packages having dimensions larger than that of the WR 975 waveguide.
Mashed potatoes were chosen as a model food to study the heating pattern because it was
relatively homogenous and easy to formulate.
The objectives of this study were to experimentally evaluate the heating pattern in
packaged mashed potatoes in WSU MCWC heating system using infrared thermal imaging;
and to compare experimentally determined heating pattern with that obtained from
simulation work using Quick Wave 3D software (Celuch and others 2000) based on FiniteDifference Time-Domain (FDTD) method. This study will help researchers to understand
the heating phenomenon of this MCWC heating system and to justify a way of monitoring
the heating procedure for future development of commercial food sterilization system.
MATERIALS AND METHODS
Preparation of mashed potatoes: Instant mashed potatoes flakes were obtained from
Oregon Potatoes Company (Boardman, OR, USA).
The supplier recommended a
water/potatoes ratio o f 5.5:1 (mass) to prepare mashed potatoes from flakes. Three ratios of
de-ionized water/potatoes (4:1, 5.5:1 and 6.5:1) were considered in this study. The samples
were either runny or too coarse beyond the above range.
Predetermined amount of salt
(NaCl) was added to the mashed potatoes to obtain samples with three different salt contents
137
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(0.5%, 1% and 2%, added by weight) at the ratio of 5.5:1 (water/potatoes).
Dielectric properties measurement: An open-ended coaxial-line probe technique
was used to measure the dielectric properties of mashed potatoes samples. This technique
required no particular shape for sample. Detailed description of the measurement system and
procedure are given by Wang and others (2003) and Guan and others (2003b). The system
can make measurement over frequency range from 1 MHz to 1800 MHz and temperature
range from 10 to 130°C. Samples of mashed potatoes with different moisture content and
salt content were measured at 200 discrete frequencies between 1 MHz and 1800 MHz at 20,
40, 60, 80, 100 and 120°C, respectively. Mean values and standard deviations from triplicate
were calculated. The dielectric data at 915 MHz used for computer simulation are reported.
MCWC heating system: The MCWC heating system included a 5 kW 915 MHz
microwave generating system (Microdry Model IV-5 Industrial Microwave Generator,
Microdry Incorporated, Crestwood, KY) with a WR 975 waveguide (Dimension: 247.7 x
123.8 mm). The microwaves from the generator were split into two parts via a Tee-shape
waveguide, from which two WR 975 waveguides were connected. The microwaves in turn
were fed into hom-shape waveguides in the cavity. Hot water up to 130°C and tap water
could be circulated to and from surge tanks in the cavity. A schematic of the system is
shown in Figure 1. The temperature of hot water was maintained through modulating valves
of a plate heat exchanger and surge tanks controlled by a Think & Do™ computer program
(Entivity, Ann Arbor, MI).
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
K) Circulated w ater
v control system
Fl
915 MHz m icrowave |______\ — 1 '
generating system '
i/ - ■
—
\
Cavity
\
Waveguide
Figure 1: Schematic diagram of MCWC Heating System at WSU
Inside the cavity a microwave transparent customized supporter was placed to hold
two trays o f mashed potatoes. To prepare the potatoes package, 200g (±2.0g) of mashed
potatoes were sealed into a microwave transparent tray (10.0 cm wide x 14.0 cm long x 2.5
cm deep x 0.3 cm thickness, Polypropylene and EVOH trays, Rexam™ Union, MO) under
vacuum (14 inches o f Mercury, or 58.69 kPa). The tray had a melting temperature greater
than 150°C.
A plastic tiding material (polypropylene and nylon, W olff Walsrode, Burr
Ridge, IL) was sealed upon the tray using a Rexam™ vacuum try sealer (Rexam Containers,
Union, MO). Detailed descriptions of sealing machine and sealing conditions are given by
Guan, et al. (2003a).
Temperature measurement and data logging:
Fiber optic sensors (FISO
Technologies Inc., Sainte-Foy, Quebec, Canada) were inserted at the center of each tray
139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
through polyimide tubing (OD: 0.075 inch or 0.1905 cm; ID: 0.0710 inch or 0.18034 cm;
thickness: 0.00200 inch or 0.00508 cm, Cole-Parmer, IL, USA) to measure the temperature
of mashed potatoes during processes. The tube was inserted through a hole in the side o f the
tray such that the sealed tip was located in the center of a tray. Detailed description is given
by Guan and others. (2003a). Due to the inherent non-uniformity in microwave heating, the
center was not necessarily the coldest spot(s) during heating and was only used as a reference
location.
Other critical data recorded every one second were time (s), ambient water
temperature (°C), flow rate of circulated water (L/min), water gauge pressure inside the
cavity (psig), actual forward power (kW) and reflected power (kW).
Power calibration of microwave generating system: After setting up the MCWC
heating system, the microwave generating system was turned on and adjusted to a desired
power level based on the power meters on the front panel. The microwave power generator
and the power meters on the front panel were calibrated through the water load method.
Namely, a customized water load was connected with the WR975 waveguide at its one end.
The generator, circulator and waveguide were terminated with circulated water, and the
temperature change before and after the heating was used to measure the actual power level.
In this study, the inlet of the load was connected with tap water and the heated water from the
outlet was stocked to calculate its volume flow rate (L/min).
Two calibrated T-type
thermocouples (diameter: 1.02 mm) connecting with a digital handheld thermometer
(Bamant Company, Barrington, IL, USA) were inserted to measure water temperatures at the
inlet and outlet. The actual power o f the generator was calculated by:
Q = c x m x At
( 1)
140
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
where Q is the actual output power of the microwave generator (kW), c is the specific heat of
the water (kW/ (kg °C)),
m is the mass flow rate of water (kg/s) and At, normally around
30°C, is the temperature rise (°C) during measurement.
MCWC heating procedure and Infrared thermal imaging: Before each test, the
MCWC heating system as well as the tray support was warmed for 5 min by circulating 80°C
water. After that, two trays of mashed potato having the same moisture content and salt
content were placed onto the tray support. The cavity was closed and filled with 80°C water.
The filling lasted for 40 sec. The microwave generator was turned on at the desired power
level (1.6 kW) for 70 sec, while the 80°C water was circulated at a flow rate of 33.4 L/min.
Water in the chamber was drained immediately within 90 sec when the microwave generator
was turned off. A typical heating profile used to determine the heating pattern is given in
Figure 2.
90 -i
—o— Oven pressure (psig)
= 80--.
8
f
g 701
m
%
£
«s
—o— oven water-inlet
temperature (°C)
—a— Forward power-top
—*— Reflected power-top
re
£L
—*— Forward powerbottom
20
Reflected powerbottom
-
— i— Product temperature
(°C)-Right tray
mmmmmmmrmMm- -■
reAO
110
-10 J
130
150
Time (sec)
Figure 2: Typical heating profile for the MCWC heating system
141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The sealed plastic lid was cut off immediately after the trays were taken out o f the
cavity.
TTyf
Thermal images were taken using an infrared camera (ThermaCAM ' , FLIR
systems, La Mirada, CA, USA).
The emissivity was calibrated based on the measuring
distance, ambient temperature and humidity and set before tests.
FDTD simulation method and Quick Wave (QW) software simulation:
The
Finite-Difference Time-Domain method developed by Yee (1966) is a useful tool for
scattering problems in electromagnetics.
The FDTD method is based upon Maxwell’s
equations, given as follows:
V x E = —ft
m.
dt
(2)
dE
dt
VxH = o i +e—
(3)
These expressions can be rearranged and separated into their orthogonal components
(for a rectangular Cartesian coordinate system):
dHx _ 1 dEy ■ SEZ
dt
jj,
dz
8EX _ 1
dt
e
dy
dz
dHy
dE,
dE,\
dt
A dx
'j
dz
dEy_
dt
e
dz
(4)
dy
(5)
(6)
(7)
dx
142
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
dff,
dt
1 ' dE,x
fi ^ dy
dEz = 1 dHy
dt
s dx
dEy
dx j
dHx
dy
(8)
-dE,
(9)
According to Yee (1966) and Sadiku (1992), a point in a grid established in the
solution space can be defined as follows:
(10)
(ij,k) = (/Ax, j Ay,khz)
And any function of space and time is defined by:
F n(/, j , k) - F (iS, j S , kS, nAt)
(11)
where 8 = Ax = Ay = Az are the increments in space, and At is the increment in time.
Central finite difference approximations for the space derivatives and the time
derivatives are given by:
dF n{i,j,k)
dx
d F “(i,j,k)
dx
F n{i + l / 2 , j , k ) - F n( i - l f 2 , j , k ) ,
+ 0( 8*)
F
!
1
n—
2{i,j,k)~F- 2{i,j,k) ,
-+ 0 { A t1)
At
143
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
( 12)
(13)
A means of evaluating E and H is obtained at alternating half-steps in time by
applying the above two equations to the six orthogonal Maxwell equations.
The FDTD
method increases the number of cells in a much lower rate than other methods including the
Finite Element Method and Method of Moments (Sadiku 1992).
QW3D computer software package is a convenient, commercial product.
The
simulation file as a form of User-defined objects (UDO’s) was developed and the resultant
system set-up similar to the pilot plant MCWC heating is shown in Figure 3. During the
simulation, the exciting field was a TE10 mode and the excitation waveform was in the format
o f pulse of spectrum (0.915 ± 0.05 GHz). The circulated water was maintained at 80°C (e =
60.04 + j 1.54) and the product was set at 60°C, the temperature at which most of the thermal
images were taken.
The simulation work was also done when the temperature of the
products were set at 40 and 80°C when the added salt levels were 2% to observe the effect of
product temperature on the field distribution.
144
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 3: Quickwave simulation set up of the MCWC Heating System
The computer software started simulation after the build-up o f the MCWC heating
system. For the purpose o f simulation, the cell sizes were defined to be 5x5><5 (mm) in the
air and 5x5x3 (mm) in the water medium, respectively.
The dimension o f the cell was
chosen according to the “rule o f thumb” for a standard FDTD method, where it is
recommended that at least 10 cells per wavelength in the medium with permittivity e be
chosen (Marcysiak and Gwarek 1994). That is,
Ace// <
(14)
10 f4 e ’
145
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
where c is velocity o f lig h t,/is frequency of the wave and e is the permittivity values o f the
medium. For example, the cell size should be less than 33 mm at 915 MHz in free space. In
this study, the cell sizes were 5x5><5 (mm) in the air and 5x5x3 (mm) in the water medium,
respectively.
In this study, the QW3D software (version 2.1) was ran on a Dell® DIMENSION
8250 desktop computer (CPU: 2.8 GHz Pentium IV; RAM: 1.0 GB) using the Microsoft®
Windows® XP Professional (Version 2002, Service Pack 1) operating system. Generally the
simulation work became stable after 30,000 cycles, according to one of the general criteria
for terminating the simulation. Namely, the amplitudes of the power density or dominant
electrical field (Ey-component) are reduced to negligible values during the simulation. When
the stability o f the simulation was observed, the current amplitudes for electromagnetic field
distribution and absorbed power distribution were accumulated on the top, middle and
bottom layers for at least 5,000 cycles, respectively. The corresponding patterns for power
density and electromagnetic field distribution are reported to indicate the simulated heating
pattern.
There are certain limitations for this simulation work, for example, (1) the
temperature of food samples and water could not change during the simulation; (2) heat
transfer between the water and food sample were not considered; (3) simulation results were
not affected by the absolute value of microwave output power; and (4) the simulation
software functioned as a “Black box” in that the user knows little about its implementation of
the FDTD method and solution method.
146
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
RESULTS AND DISCUSSION
Dielectric properties of mashed potatoes at 915 MHz: Generally, the loss factor of
mashed potatoes at 915 MHz increased with the temperature at all moisture and salt levels,
while the dielectric constant decreased with temperature (Table 1). At a given temperature,
the loss factor decreases with moisture content; however, it increases with salt content.
There is no clear trend for the change of dielectric constant with moisture content or with salt
content. A detailed description about the change of dielectric properties with the previously
mentioned affecting factors (temperature, moisture and salt content) is given by Guan and
others (2003b).
Table 1: Dielectric properties (Dielectric constant/Loss factor) of mashed potatoes vs.
m oisture content and added salt levels at 915 MHz (extracted from C hapter 5)
Temperature\Moisture
81.70%
85.90%
87.80%
40°C
56.8/23.7
65.7/22.8
65.9/18.6
60°C
55.3/26.1
62.5/25.2
61.6/21.2
80°C
53.6/30.7
59.9/27.6
59.6/22.6
T emperature\Salt
0.50%
1%
2%
40°C
62.5/29.4
52.8/35.6
59.7/65.0
60°C
59.6/36.7
49.4/43.4
56.8/82.2
80°C
56.6/45.3
46.1/51.7
55.1/105.3
Infrared thermal images:
The thermal images for the top surfaces of different
mashed potatoes are provided in Figures 4-8. The corresponding 3D views are also shown.
147
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Obvious heating patterns are shown throughout the figures, i.e., the temperatures were higher
both in the center and on the edge of the tray, between which lay the lower temperature parts.
The heating pattern has a “cap” shape for both trays, and this heating characteristic was
repeatable for products with various moisture and salt contents. In Figures 6 and 7, nine
parallel lines were drawn on the tray from top to bottom and the related temperature profiles
were presented. Both figures showed that temperature differences existed in the top layer.
Thermal images o f mashed potatoes heated only by 80°C hot water for 110 sec (Figure 9) are
given as a comparison; in these there existed no obvious “cap” shape.
This might be
explained by the lack of microwave heating effect when using hot water only. For hot water
heating, heat transfer takes place because of the presence of temperature gradients among
different locations; and the location closest to the geometric center usually received the heat
at the slowest rate. However, the characteristically volumetric microwave heating might
increase the temperatures close to the geometric center at a higher rate than that from hot
water heating. The combination heating effect from hot water (or conventional) heating and
microwave heating might lead to the formation of the typical “cap” heating pattern.
148
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4-B: Right tray (Thermal image)
4-A: Left tray (Thermal image)
4-C: Left tray (3D view)
4-D: Right tray (3D view)
Figure 4: Thermal image of heated mashed potatoes (initial temperature: 25°C;
moisture content (w .l.): 85.9%, no added salt, microwave heating time: 7§ sec)
149
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5-A: Left tray (Thermal image)
5-B: Right tray (Thermal image)
5-C: Left tray (3D view)
5-D: Right tray (3D view)
Figure 5: Therm al images of heated mashed potatoes (initial temperature: 25°C;
moisture content (w.b.): 85.9%; 0.5% added salt, microwave heating time: 70 sec)
150
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Figure 6: Thermal images of heated mashed potatoes (initial temperature: 25°C;
moisture content (w.b.): 85.9%; 1% added salt; Left tray, microwave heating:
70
6-B: Left tray (3D view)
6-A: Left tray (thermal image)
— —
6-C: Temperature curve on the nine lines
Label
Min
Max
Max - Min
Avg
Stdev
LI01
45.9
56.7
10.8
49.9
3
LI02
44.2
55.6
11.5
47.9
3.3
LI03
46.1
56.6
10.5
49.2
2.1
LI04
47.9
54.8
6.9
51.3
2.2
LI05
48.4
57.4
9
53.2
2.7
LI06
48.4
60.9
12.5
55
3.8
LI07
46.1
55.2
9.1
51.1
2.6
LI08
42.9
51.6
8.8
45.9
2.5
LI09
43.4
52.2
8.8
45.7
2.4
6-D: Analysis of the temperature curves
151
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
sec)
Figure 7 : Thermal images of heated mashed potatoes (initial temperature: 25°C;
moisture content (w.b.): 85.9%; 1% added salt; right tray, microwave heating: 70 sec)
7-A: Right tray (thermal image)
7-B: Right tray (3D view)
a;
§11
7-C: Temperature curve on the nine lines
Label
Min
Max
Max - Min
Avg
Stdev
LI01
50.6
55.8
5.2
52.4
1.3
LI02
46
54
8
49.1
2.1
LI03
45.9
55.2
9.3
50.4
2.3
LI04
46.9
55.1
8.1
51.2
2.2
LI05
48.3
56.4
8.2
52.4
2.2
LI06
49.1
57
7.9
52.9
2.4
LI07
46.4
53.5
7.1
49
1.8
LI08
43.9
54.2
10.4
46.7
3.1
LI09
47.2
54.4
7.2
49.8
2.2
7-D: Analysis of the temperature curves (°C)
152
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8-A: Left tray (Thermal image)
8-B: Right tray (Thermal image)
8-C: Left tray (3D view)
8-B: Right tray (3D view)
Figure 8: Thermal image of heated mashed potatoes (initial temperature: 25°C;
moisture content (w.b.): 85.9%; 2% added salt; microwave heating time: 70 sec)
153
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9-A: Left tray (Thermal image)
9-B: Right tray (Thermal image)
9-C: Left tray (3D view)
9-D: Right tray (3D view)
Figure 9: Thermal images of heated mashed potatoes using 80°C hot water only (initial
temperature: 25°C; moisture content (w.b.): 85.9%; 2% added salt; heating time: 110
sec)
Simulation results: Simulation results for mashed potatoes with different moisture
and salt contents are all comparable to heating patterns obtained experimentally by an
154
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
infrared thermal camera. Representative results for mashed potatoes (1% added salt) are
given in the form of power density distributions on the top, middle and bottom layers; Figure
10 shows these distributions where red indicates a relatively high density and green indicates
a low density. According to Figure 10, the power density distribution on the top is very
similar to that at the bottom, but the power density distribution in the middle layer is lower
than those on the top and bottom layers. This is in agreement with related electric field
distributions (Figure 11). The low electric field distribution and resultant power density
distribution in the middle layer might be due to absorption of microwave power during the
penetration.
Tables 2 and 3 present the corresponding maximum/minimum ratios under
different simulation conditions. None of the tables indicate a significant influence from the
product temperature or salt levels on the field distribution on the same layer. However,
increased salt contents did reduce the power density in the middle layer (Figure 12), which
matched the decreased o f penetration depth with added salt (Guan and others, 2003b).
155
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10-A: Top layer
10-B: Middle layer
10-C: Bottom layer
Figure 10: Power density distribution within the mashed potatoes (simulation
temperature: 60°C; moisture content (w.b.): 85.9% and added salt: 1.0%)
156
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11-A: Top layer
11-B: Middle layer
11-C. Bottom layer
Figure 11: Simulated dom inant electric field (Ey) distribution within the mashed
potatoes (Temperature: 60°C; moisture content (w.b.): 85.9%; added salt: 1%)
157
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2: Field distribution (max/min ratio) from simulation for products at 3
temperatures (2% added salt)
Temperature (°C)
80
Right
Left
Right
Left
Right
1.7
1.6
1.6
1.6
1.5
1.7
pd
2.9
3.2
3.0
3.1
3.0
3.1
Ey
1.2
1.3
1.3
1.4
1.3
1.7
pd
1.4
1.1
4.6
4.5
3.9
3.8
Ey
1.7
1.6
1.7
1.6
1.6
1.7
pd
2.9
3.2
3.0
3.0
2.7
3.1
Location
(Layer/Tray)/Parameter
Top
Ey
Middle
Bottom
40
60
Note: pd means power dens
Table 3: Field distribution (max/min ratio) from simulation for mashed potatoes with 4
added salt levels (60 °C)
Added salt level
Location
0%
0.5%
2%
1%
Left
Right
Left
Right
Left
Right
Left
Right
E,
1.5
1.8
1.6
1.7
1.6
1.7
1.6
1.6
pd
2.6
2.9
2.1
2.3
2.1
2.3
3.0
3.1
Ey
1.6
1.6
1.4
1.4
1.7
1.7
1.4
1.3
pd
2.5
2.5
3.5
2.5
2.7
2.8
4.6
4.5
Ey
1.8
1.7
1.6
1.6
1.6
1.6
1.7
1.6
pd
2.1
2.1
2.3
2.3
2.5
2.4
3.0
3.0
(Layer/Tray)/Parameter
T° P
Middle
Bottom
■
158
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12-B: 0.5% added salt
12-D: 2% added salt
Figure 12: Simulated power density distribution for the mashed potatoes at Middle
layer (initial temperature: 6§°C; moisture content (w.b.): 85.9%)
159
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Discussion: In this study, infrared thermal images of mashed potatoeses were
obtained to document the heating pattern of mashed potatoes for the pilot-scale microwavecirculated water combination heating system. However, there exists possible gap between
thermal images and actual processing results since time elapsed when removing products
from the cavity. For example, relatively low temperature (80°C) for the circulated water was
proposed to heat the food samples and to obtain the related heating patterns in this study; but
the heating patterns obtained at low temperature may not match with those obtained at higher
temperature (e.g., 121.1°C. Furthermore, the product was immersed in relatively static water
during simulation; in real processing, the water is being circulated at certain flow rate
instead.
The coupling effect from microwave heating and conventional hot water heating was
not considered, since the software program has no built-in ways to analyze the coupling
effect.
For example, as the foods heat up non-uniformly, its dielectric properties vary
spatially and change with time. At lower temperature, the material is less lossy, allowing
microwave to penetrate more. As the materials heat up, the dielectric loss factor increases.
Most of microwave energy will be absorbed at locations near the surface, moving the
location o f the highest temperature close to the surface.
Further effort needs to take to improve the simulation work so that the results from
simulation will help to understand the real microwave-circulated water combination heating
technology.
160
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C O N C L U S IO N S
Heating patterns of mashed potatoes with different moisture . and salt contents
processed by the pilot-scale 915 MHz Microwave-Circulated Water Combination heating
system were repeatable when the circulated water was held at 80°C. Related simulation
results showed that the heating patterns from the above heating system could be predicated
approximately by computer simulation work, such as that using Quickwave-3D commercial
software. Repeatable and predictable heating patterns obtained for this microwave circulated
water combination heating system may be used to monitor temperature profiles and ensure
suitability of the system’s application in the field of food sterilization.
ACKNOW LEDGEM ENTS
The authors acknowledge the financial support provided by the U.S. Army Natick
Soldier Support Center and Kraft Foods. We also thank Mr. Evan Turek and Dr. Ming H.
Lau o f Kraft Foods, Glenview, IL, for their technical advice and assistance.
REFERENCES
Bows JR. 1999. Variable frequency microwave heating of foods. J Microwave Power
and Electromagnetic Energy. 34 (4): 227-238.
Bows JR, Patrick ML, Janes R, Metaxas A and Dibben D. 1999. Microwave phase
control heating. Int’l I. Food Sci. and Technol. 34: 295-304.
Celuch M and Gwarek WK. 2000. Advanced features of FDTD modeling for
microwave power applications. Presented at the 35th Microwave Power Symposium.
Montreal. Canada.
161
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Decareau R. 1985. Microwaves in the Food Processing Industry, Academic Press.
New York.
Guan D, Plotka VCF, Clark S and Tang I. 2002. Sensory evaluation of microwave
treated macaroni and cheese. I. Food Process. Preserv. 26: 307-322.
Guan D, Gray P, Kang DH, Tang I, Shafer B, Ito K, Younce F and Yang TCS. 2003a.
Microbiological validation of microwave-circulated water combination heating technology
by inoculated pack studies. J. Food Sci. 68 (4): 1428-1432.
Guan D, Cheng M, Wang Y and Tang I. 2003b. Dielectric properties of mashed
potatoes as relevant to microwave and radio frequency pasteurization and sterilization
processes. (Submitted)
Lau MH. 2000. Microwave Pasteurization and Sterilization of Food Products. Ph.D.
Thesis. Washington State University, Pullman, WA.
Marcysiak MC and Gwarek WK. 1994. Higher-order modeling of media interfaces
for enhanced FDTD method analysis of microwave circuits. Proceedings of 24th European
Microwave Conference. Cannes. Page 1530-1533.
Morris CE.1991. World’s 1st microwave sterilization system. Food Engineering Int’l.
6: 63-66.
Mudgett RE. 1982. Microwave properties of foods in microwave processing. Food
Technol 36: 109-115.
Ohlsson T. 1978. Temperature distribution of microwave heating-spheres and
cylinders. J Microwave Power.13: 303-309.
Ohlsson T. 1991. Microwave heating uniformity. Paper presented at the AICHE
162
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Conference on Food Engineering, Chicago, March 11-12.
O’Meara IP, Farkas DF and Wadsworth CK. 1977. Flexible pouch sterilization using
combined microwave-hot water hold simulator. Contract No. (PN)DRXNM 77-120, U.S.
Army Natick Research & Development Laboratories, Natick, MA 01760 .
Sadiku MN. 1992. Numerical techniques in electromagnetics. CRC Press. Boca
Raton.
Stenstrom LA. 1974. US Patents 3,809,845; 3,814,889.
Pathak S, Liu F and Tang J. 2002. Finite difference time domain (FDTD) modeling of
a rectangular horn waveguide for microwave sterilization of foods at 915 MHz. Presented at
2002 annual ASAE meeting. Chicago, Illinois. USA. July 28-July 31.
Wang Y, Wig TD, Tang J and Hallberg LM. 2003. Dielectric properties of foods
related to RF and microwave pasteurization and sterilization. J. Food Engr. 57: 257-268.
Yee, KS. 1966. Numerical solution of initial boundary value problems involving
Maxwell’s equations in isotropic media. IEEE Trans. Antennas Propagat. Vol. AP-14: 302307.
163
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C O N C L U S IO N S
This work addresses feasibility of sterilizing packaged foods using 915 MHz
Microwave-Circulated Water Combination (MCWC) heating technology through a pilotscale system. This system combines microwave energy with circulated water (123~125°C)
and can heat food products up to 121.1 °C or above.
The products are held at high
temperature to reach desired degrees of sterilization (F0 value). The MCWC heating system
provides microbiologically safe products with good sensory quality. The newly-developed
915 MHz single mode MCWC heating system can provide predictable and repeatable heating
patterns.
The study is divided into three major phases. In Phase One, it is proposed that the food
products processed by MCWC heating system were microbiologically safe. In Phase Two, it
is proposed that the food products processed by MCWC heating system could be of
acceptable sensory quality. In Phase Three, it is proposed that the newly-developed 915
MHz single mode MCWC heating system could provide repeatable and predictable heating
patterns.
For Phase One, the idea was validated through inoculated pack studies (Chapter 3).
Thermal resistances o f PA 3679 spores in macaroni and cheese entree were determined using
Thermal Death Time (TDT) mini-retorts. Results indicated that microbial destruction by the
pilot-scale 915 MHz MCWC heating system matched with designed degrees of sterilization
(Fq value). This study showed that MCWC heating technology had potential in sterilizing
packaged foods.
For Phase Two, the idea was validated through sensory evaluation tests on macaroni
and cheese entree processed by MCWC heating system.
MCWC heating system treated
macaroni and cheese entrees within one fourth of time required by conventional retort
164
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
methods to attain same sterility (Fq value: 6-7 minutes). Changes in formulation, such as
amount of cheese sauce and noodle type, affected sensory quality and acceptability of
MCWC entrees.
For Phase Three, the idea was validated both from computer simulation work using
commercial software and experiments using infrared thermal image method (Chapter 6). The
simulation work was completed using 3D-Quickwave commercial software through the
Finite-Difference Time-domain (FDTD) method.
For the purpose of simulation work,
dielectric properties of mashed potato entree were measured using open-ended coaxial-line
probe method over frequencies between 1 MHz and 1800 MHz and temperatures between 20
and 120°C. Influence o f frequency, temperature, moisture content and salt (added NaCl)
content moisture and salt were considered during the measurement (Chapter 5). Results
indicated that dielectric constant and loss factor decreased with frequency.
Dielectric
constant increased with temperature at 27 MHz and changed little at 40 MHz and decreased
with temperature at 433, 915 and 1800 MHz.
Dielectric loss factor increased with
temperatures. Addition o f salt increased the dielectric loss factor, especially at 27 and 40
MHz, but dielectric constant changed little. Penetration depth decreased with temperature
and frequencies; it was less dependent on moisture content (between 81.6 and 87.8%) than
added salt content (0.5% to 2%). Moisture content in the tested range did not significantly
affect the dielectric properties of mashed potato. Salt content instead of moisture content
should be carefully monitored when preparing mashed potato samples for sterilization using
RF or microwave energy. The heating patterns were repeatable when setting circulated water
at 80°C.
Computer simulation results were compared with those results obtained
experimentally.
Both results indicated that heating patterns of mashed potato were
165
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
repeatable and predictable, making industrial application of the system in food sterilization
possible.
The study demonstrates that 915 MHz MCWC heating system has potential to
sterilize food products while maintaining a relatively good sensory quality; the newlydeveloped 915 MHz single mode MCWC heating system can provide predictable and
repeatable heating pattern.
166
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FUTURE W ORK
Success in commercial microwave sterilization technology are promising, however,
several factors delay this progress in the United States because of the complexity of this
technology and its high costs.
Future research needs to be focused on the following areas:
1.
Development of a reliable and practical method to determinethe
heating
patterns in food packages;
2.
Development of a simple and reliable way to control microwave sterilization
processing and determine the critical process points;
3.
Development of food products suitable to be processed by microwave
sterilization technology and study on the effects of food formulation on
heating patterns;
4.
Development o f suitable packaging materials;
5.
Better understanding of the effects of equipment design factors on the heating
uniformity;
6.
Better understanding of the process deviation and the way to correct the
deviation real-time;
7.
Economics study.
167
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
18 823 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа