Proquest Dissertations
INFORMATION TO USERS
This manuscript has been rapreducad from tha microfilm mastar. UMI films the text directly from the original or copy sutmittad. Thus, soma thesis and dissertation copies are in typewriter face, while others maybe from any type of computer printer.
Tha 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 t)leedthrough, substandard margins, and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI 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.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at tfie upper left-hand comer and continuing from left to right in equal sections with small overlaps.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6* x 9* black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order.
Bell & Howell Information and teaming 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 800-521-0600 IM
PACKAGING REQUIREMENTS FOR PULSED ELECTRIC FIELD
PROCESSED ORANGE JUICE
DISSERTATION
Presented in Partial Fulfillment of the Requirements for
the Degree Doctor of Philosophy in the Graduate
School of The Ohio State University
By
Zehra Ayhan, M.S.
*****
The Ohio State University
2000
Dissertation Committee;
Professor Q. Howard Zhang, Adviser Approved by
Professor Grady W. Chism
Professor Steven J. Schwartz
Professor David B. Min Adviser
Professor John Litchfield Food Science and Nutrition
Graduate Program UMI Number 9982519
UMI*
UMI Microform9982519 Copyright 2000 by Bell & Howell Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.
Bell & Howell Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Artxrr, Ml 48106-1346 ABSTRACT
Package forming parameters, seal integrity and package compatibility with foods are important issues in food packaging. These issues influence performance of packages.
The shelf life, quality and safety of foods heavily depend on the performance of food packages.
The thermoforming process was optimized to improve the quality and uniformity of semirigid food packages produced by a Benco aseptic packaging machine. Analysis of wall thickness data obtained for the thermoforming parameters (forming temperature, forming air pressure and heating time) used in this study showed that wall thickness was significantly (p< 0.05) affected by forming temperature, pressure and heating time.
Forming temperature was found to be the principle parameter influencing wall thickness distribution in a plug-assist thermoforming operation. The optimum operating conditions of the packaging machine for this thermoforming process are: 146-156°C for forming temperature, 2-4 bars for air-forming pressure and 74-97 seconds for heating time.
Seal integrity of plastic food packages was evaluated using non-contact, immersion type ultrasonic testing in pulse-echo mode using high frequency sound waves.
Ultrasonic signals, or echoes, reflected by the seal were used to develop A-scan and C- scan presentations. Discontinuities in the seal, short seal, non-bonded areas, imbedded foreign matter such as wire and Teflon in the seal, contaminated seal and abrasion resulted in reduced signal strengths. Pictures taken with optical microscopy demonstrated that signal amplitude was relatively high when the seal area was uniform.
Effects of packaging materials as well as pulsed electric field (PEF) processing and storage conditions on orange juice flavor, color and vitamin C were investigated using a pilot plant scale integrated PEF processing and glove box packaging system.
Single strength orange juice was treated with PEF at electric field strength of 35 kV/cm for 59ps and filled into sanitized bottles of glass, polyethylene terephthalate (PET), high density polyethylene (HDPE) and low density polyethylene (LDPE) in a sanitized glove box. Among plastic materials tested, PET was comparable to glass in terms of retention of flavor, color and vitamin C of PEF treated orange juice. PEF processed orange juice in glass and PET bottles had a shelf life of more than 16 weeks at 4°C storage. It is recommended to select compatible packaging material to maintain benefits of PEF processing during storage.
Ill ACKNOWLEDGMENTS
I would like to express my sincere appreciation to Dr. Q. Howard Zhang, my advisor, for his support, guidance and encouragement, which made possible the completion of this study. Special gratitude is also extended to Drs. David B. Min, Grady
W. Chism, Steven J. Schwartz and John Litchfield for serving on my dissertation committee and for valuable advises.
The authors acknowledge thermoforming project funding provided by US Army
Research, Development and Engineering Center. The Benco aseptic packaging machine was run by the PEF packaging group, Charles B. Streaker, Zehra Ayhan and Yifan Xu in the Food Science and Technology Department pilot plant at the Ohio State University
(OSU). Statistical consultation was provided by senior statistical consultant, Mohammed
Rahman, of University Technology Services at OSU. The Benco ASEPACK 2 aseptic packaging machine was made available as a cost-share to the Army project by Borden
Company.
The authors acknowledge seal inspection project funding provided by Combat
Ration Network for Technology Implementation (CORANET). Project participants included the Ohio State University (OSU), Edison Welding Institute (EWI), Packaging
Technology and Inspection (PTI), US Army Natick RD&E Center and Sterling Foods
IV Company. The authors wish to thank Mikail Kneller, Roger Spencer, John Hunt, and Eric
Cutlip for their technical assistance to the project.
We would like to acknowledge NSF CAPPS Center for operational fundings of orange juice project. PEF processing equipment is provided by a project funded by the
US Army Natick RD&E Center. Product is provided by member company, MinuteMaid.
The authors are thankful to Dr. C. Wang for operating the pilot plant PEF system, Mr. C.
B. Streaker for operating fluid handling system and Dr. H. W. Yeom for helping with product preparation. The authors are also thankful to Dr. S. Palaniappan of MinuteMaid for his technical advises.
I also would like to thank PEF group members who helped me when 1 needed and for their friendship.
Last but not least, special thanks must be credited to my parents, my sisters and brother for their love and faith during my absence from home. I like to extend my deepest appreciation to my friends Eser Tufekci, Latife Sahin and Gulnur Arabaci, for their love, continuous support and understanding during my stay in Columbus.
Appreciation is also extended to Mustafa Kemal University, Turkey for their sponsorship throughout my master and Ph.D. studies. VITA
June 12,1969 ...... Bom-Ankara, Turkey
1991 ...... B.S. Dept, of Food Science and Technology
University of Ankara
Ankara, Turkey
1996 ...... M.S. Dept, of Food Science and Nutrition
The Ohio State University
Columbus, OH
1994-1999 ...... Scholar of Mustafa Kemal University
Hatay, Turkey
1999-present ...... Graduate Research Associate
Dept, of Food Science and Nutrition
The Ohio State University
PUBLICATIONS
Research Publications
1. Zehra Ayhan, “Preservation of Fresh Cut Melons”. 1996. M.S. Thesis. The Ohio State University.
VI 2. Zehra Ayhan, Grady W. Chism and Edward R. Richter. 1998. The Shelf-life of Minimally Processed Fresh Cut Melons. J. Food Quality. 21:29-40.
3. Zehra Ayhan and Q. Howard Zhang. 2000. Wall thickness distribution in thermoformed food containers produced by a Benco aseptic packaging machine. J. Polymer Engineering and Science. 40 (1 ): 1-10.
Papers submitted:
I. Zehra Ayhan, Q. Howard Zhang, Bahram Farahbakhsh and Mikail Kneller. Inspection of seal integrity of food packages using ultrasound and pressure differential techniques. Submitted for publication in the Journal of Applied Engineering in Agriculture. (May 1999).
2. Zehra Ayhan, Charles B. Streaker and Q. Howard Zhang. Design, construction and validation of a sanitary glove box packaging system for product shelf life studies. Submitted for publication in the Journal o f Food Protection. (April 2000).
4. Zehra Ayhan, Q. Howard Zhang and David B. Min. Effects of pulsed electric field processing and storage on stability and quality of single strength orange juice. Submitted for publication in the Journal o f Food Science. (May 2000).
4. Zehra Ayhan, Hye Won Yeom, Q. Howard Zhang and David B. Min. Flavor, color and vitamin C retention of pulsed electric field processed orange juice in different packaging materials. Submitted for publication in the Journal o f Agricultural and Food Chemistry. (May 2000).
PRESENTATIONS
1. Zehra Ayhan, Grady W. Chism and Edward R. Richter. 1996. Preservation of Aesh cut melons. Institute of Food Technologists Annual Meeting, New Orleans, TX.
2. Zehra Ayhan and Q. Howard Zhang. 1998. Wall thickness distribution in thermoformed food containers produced by a Benco aseptic packaging machine. Research and Development Associates for Military Food & Packaging Systems, Inc. Fall Meeting, Boston, MA.
3. Zehra Ayhan and Q. Howard Zhang. 1999. Effects of thermoforming parameters on wall thickness uniformity. Institute of Food Technologists Annual Meeting, Chicago, IL.
vii 4. Zehra Ayhan, Q. Howard Zhang and Bahram Farahbakhsh. 1999. Inspection of seal integrity of food packages using ultrasound as non destructive technique. AIChE Fall Meeting, Dallas, TX.
5. Zehra Ayhan, Charles B. Streaker and Q. Howard Zhang. 2000. Design and validation of a glove box packaging system for extended shelf life studies. Institute of Food Technologists Annual Meeting, Dallas, TX.
6. Zehra Ayhan, Q. Howard Zhang and David B. Min. 2000. Effects of processing and packaging on orange juice flavor. Institute of Food Technologists Annual Meeting, Dallas, TX.
FIELDS OF STUDY
Major Field: Food Science and Nutrition
Specialization: Food Packaging
Vlll TABLE OF CONTENTS
Page
Abstract ...... ii
Acknowledgments ...... iv
Vita...... vi
List of Tables ...... xi
List of Figures...... xii
List of Symbols ...... xvi
Chapters:
I. Literature Review
Plastics in food packaging...... 1 Influence of thermoforming parameters on wall thickness uniformity of formed parts...... 3 Nondestructive techniques for inspection of seal integrity of food packages...... 10 Effects of processing and packaging on orange juice quality and shelf life...... 19 References...... 27
2. Influence of thermoforming parameters on wall thickness uniformity of formed cups produced by a Benco aseptic packaging machine
Abstract ...... 35 Introduction...... 37 Materials and Methods...... 40 Results and Discussion...... 45 Conclusions...... 49 References...... 50
IX 3. Inspection of seal integrity of food packages using ultrasound and new pressure differential techniques
Abstract ...... 66 Introduction...... 68 Materials and Methods...... 73 Results and Discussion...... 80 Conclusions...... 84 References...... 85
4. Design, construction and validation of a sanitary glovebox packaging system for product shelf life studies
Abstract...... I ll Introduction...... 113 Materials and Methods...... 115 Results and Discussion...... 119 Conclusions...... 120 References...... 121
5. Effects of pulsed electric field processing and storage on stability and quality of single strength orange juice
Abstract ...... 127 Introduction...... 128 Materials and Methods...... 130 Results and Discussion...... 135 Conclusions...... 139 References...... 140
6. Flavor, color and vitamin C retention of pulsed electricfield processed orange juice in different packaging materials
Abstract ...... 149 Introduction...... 151 Materials and Methods...... 154 Results and Discussion...... 159 Conclusions...... 165 References...... 166
7. Summary and recommendations...... 184
List of References...... 188 LIST OF TABLES
Table Page
2.1. Heating time calculations based on processing parameters...... 52
2.2. Effects of forming temperature on wall thickness of thermoformed containers at different locations...... 53
2.3. Effects of forming temperature on wall thickness of thermoformed containers on different sides...... 54
2.4. Effects of air forming pressure on wall thickness of thermoformed containers at different locations...... 55
2.5. Effects of air forming pressure on wall thickness of thermoformed containers on different sides...... 56
2.6. Effects of heating time on wall thickness of thermoformed containers at different locations...... 57
2.7. Effects of heating time on wall thickness of thermoformed containers on different sides...... 58
3.1. Peak amplitudes at short seal and different seal regions of semirigid containers...... 88
3.2. Peak amplitudes at different seal regions of semirigid containers with seal discontinuity...... 89
3.3. Peak amplitudes at disturbed and undisturbed seal regions of semirigid containers with wire embedded in the seal ...... 90
4.1. Absorbances, total plate count and yeast and mold (Y&M) counts of the nutrient broth incubated at 22“C for 2 weeks ...... 122
4.2. Absorbances, total plate count and yeast and mold (Y&M) counts of the nutrient broth incubated at 37“C for weeks ...... 123
5.1. Effects of PEF processing on orange juice flavor compounds...... 142 xi LIST OF FIGURES
Figure Page
2.1. Schematic diagram of a Benco Asepack aseptic packaging machine...... 59
2.2. Heater setting versus forming temperature graph...... 60
2.3. Calibration curve for infrared probe temperature measurement...... 61
2.4. Thickness measurement points of a thermoformed cup...... 62
2.5. Wall thickness distribution at different forming temperatures...... 63
2.6. Wall thickness distribution at different forming pressures...... 64
2.7. Wall thickness distribution at different heating times ...... 65
3.1. System components of non-contact immersion type ultrasonic system...... 91
3.2 A. Components of leak tester...... 92
3.23. Picture of leak test chamber ...... 93
3.3A. Partial C-scan image of a polymeric tray with seal contamination...... 94
3.33. A-scan presentation of undisturbed seal area of a polymeric tray...... 95
3.3C. A-scan presentation of seal contamination area of a polymeric tray...... 96
3.4. Partial C-scan image of a polymeric tray with seal abrasion ...... 97
3.5 A. C-scan presentation of a short sealed semirigid cup...... 98
3 .53. Microscopic image of 12 o’clock position of a short sealed semirigid cup 99
3.5C. Microscopic image of short seal area of a short sealed semirigid cup...... 100
3.6. C-scan image of a semirigid cup with seal discontinuiQr...... 101 xii 3.7A. Partial C-scan image of a semirigid cup with wire in the seal...... 102
3.7B. Microscopic image of undisturbed seal area of a semirigid cup ...... 103
3.7C. Microscopic image of seal area with wire of a semirigid cup...... 104
3.8. Partial C-scan image of a semirigid cup with teflon in the seal...... 105
3.9. Pressure change versus test time during leak testing of MRE pouches with oatmeal cookie (test time of 25 s, equalizing time of 20 s)...... 106
3.10. Pressure change versus test time during leak testing of MRE pouches with oatmeal cookie (test time of 25 s, equalizing time of 10 s)...... 107
3.11. Pressure change versus test time during leak testing of MRE pouches with oatmeal cookie (test time of 10 s, equalizing time of 5 s)...... 108
3.12. Pressure change versus test time during leak testing of MRE pouches with pound cake (test time of 25 s, equalizing time of 10 s)...... 109
3.13. Pressure change versus test time during leak testing of MRE pouches with pound cake (test time of 10 s, equalizing time of 5 s)...... 110
4.1. Schematic diagram of a glove box ...... 124
4.2. Flow chart of an integrated pulsed electric field processing and glove box packaging system...... 125
4.3. Comparison of the nutrient broth filled in chemically cleaned bottles with positive and negative controls stored at 22X for 14 days...... 126
5.1. Flow chart of an integrated pulsed electric fields processing and glove box packaging system...... 143
5.2. Retention (%) of PEF processed orange juice flavor compounds during 112 day storage at 4“C in glass...... 144
5.3. Retention (%) of PEF processed orange juice flavor compounds during 112 day storage at 22“C in glass...... 145
5.4. Effects of PEF processing on orange juice color...... 146
XIU 5.5. The stability of PEF processed orange juice color stored at 4“ and 22®C in glass...... 147
5.6. Color and overall flavor liking of PEF processed orange juice during 112 day storage at 4°C in glass...... 148
6.1. Flow chart of an integrated pulsed electric field processing and glove box packaging system ...... 169
6.2. D-limonene retention (%) in PEF processed orange juice during storage at 4“ and 22®C in different packages...... 170
6.3. Myrecene retention (%) in PEF processed orange juice during storage at 4® and 22®C in different packages...... 171
6.4. Pinene retention (%) in PEF processed orange juice during storage at 4® and 22®C in different packages...... 172
6.5. Octanal retention (%) in PEF processed orange juice during storage at 4® and 22®C in different packages...... 173
6.6. Decanal retention (%) in PEF processed orange juice during storage at 4® and 22®C in different packages...... 174
6.7. Ethyl butyrate retention (%) in PEF processed orange juice during storage at 4® and 22®C in different packages...... 175
6.8. Linalool retention (%) in PEF processed orange juice during storage at 4® and 22®C in different packages...... 176
6.9. Effects of cap liner (PE) on the sorption of d-limonene during storage at4®C...... 177
6.10. Comparison of orange juice color- L value- in glass and plastic bottles during storage at 4® and 22®C...... 178
6.11. Comparison of orange juice color- a value- in glass and plastic bottles during storage at 4® and 22®C...... 179
6.12. Comparison of orange juice color- b value- in glass and plastic bottles during storage at 4® and 22®C...... 180
XIV 6.13. Effects of packaging on mean hedonic responses for color intensity of PEF treated orange juice during 112 day storage at 4®C...... 181
6.14. Effects of packaging on mean hedonic responses for flavor intensity of PEF treated orange juice during 112 day storage at 4°C...... 182
6.15. Effects of packaging materials on the concentration of ascorbic acid in PEF treated orange juice during storage at 4“C...... 183
XV LIST OF SYMBOLS
D: Distance that base material travel from the entrance of the infrared heater up to the center of the forming unit
Dindex: Distance between lateral side of one container and lateral side of the following container
F: Flow rate of food product (L/h) tcycie: Each mold cycle time (4 seconds) plus delay time tdeimy: Time delay between each mold cycle th: Heating time of the base material right after the material enters infrared heater till the forming process
Vave: Average velocity for the film to travel (cm/s)
Vg: Volume of each container (200 ml)
N; North side of a cup (pull-tab as taken north)
S; South side of a cup
E: East side of a cup
W; West side of a cup
XVI CHAPTER 1
LITERATURE REVIEW
PLASTICS IN FOOD PACKAGING
Packaging has become an integral part of the processing, preservation, marketing and even the cooking of foods. Initially, packages served simply to contain products and to protect them from outside contamination. There has been tremendous growth in the development and design of new packages to serve specific needs. While there has been adaptation of many of the traditional packaging materials, metals, glass and paper, much of the growth has been in plastic or flexible packaging materials.
Plastics which were once perceived as undesirable by food processors and consumers are now often seen as the best form of packaging available. Nearly all types of food packaging use plastics as part of their construction.
Plastic packages are extremely versatile. Materials can be extruded into films, blow-molded into bottles, or layered onto each other for increased resistance to light, air, moisture, heat and package insults such as puncturing and tearing. Flexible packages are also more economical, easier to use and more convenient to store than traditional packaging materials. The new materials have allowed the development of food products that previously not feasible, and, in conjunction with new processing technologies, they offer greater convenience and increased or equivalent product quality.
The primary role of food packaging is to protect the product from the time and point of manufacture to the time and point of consumption. Three elements are critical if a package is to perform its role of protecting the product throughout its life. The package must provide:
1. Adequate moisture and oxygen barrier to prevent chemical and physical degradation of the product throughout its life.
2. Adequate physical strength to withstand the abuse encountered in thermal processing and throughout the distribution system.
3. Adequate package integrity to assure that product Is contained within the package throughout the product shelf-life and throughout the distribution system and also to prevent the ingress of microorganisms, oxygen, filth, or other environmental contaminants that could render the product unfit for consumption or which could simply reduce the quality of the product to a level less than that intended for consumption.
Forming of the package, sealing and selecting the proper package compatible with food are important factors that affect above listed three elements. Problems and needs in the areas of forming, sealing and selecting a package will be addressed by the following literature review. INFLUENCE OF THERMOFORMING PARAMETERS ON WALL
THICKESS UNIFORMITY OF FORMED PARTS
The thermoforming process is only one of many manufacturing methods that converts plastic resin material into numerous products. Yet in our modem life-style, we are coming to rely more and more on the benefits of thermoforming and make extensive use of plastic products produced by this process.
Thermoforming is the amalgamated description of the various thermoplastic sheet-forming techniques, such as vacuum forming, pressure forming, matched mold forming, and their combinations. All of these forming techniques require a premanufactured thermoplastic sheet, which is clamped, heated, and shaped into or over a mold. The softened material by heat can be formed into a mold by atmospheric pressure against a vacuum or slight air pressure. Upon contacting the cold mold the softened material freezes hence a thickness profile is obtained since the soft sheet contacts mold at different times. Products made by this process are generally finished after the trimming operation and are ready to be used (Rosenzweig et al., 1979; Rosen, 1971; Florian, 1987).
The thermoforming process offers fast forming and therefore lends itself to automation and long-term production runs. With its relatively fast molding cycles and comparatively inexpensive mold costs, the thermoforming process is often chosen as the most cost-effective manufacturing method over all the other processes (Florian, 1987).
However, nonuniform thickness and thinning at container base comers are major limitations of thermoforming (Aroujalian et al., 1997). Under all thermoforming conditions where pieces are shaped from a flat sheet or film, the surface area will become
3 larger and therefore the gauge thickness thinner. One of the decisive factors for this
thinning is the “draw ratio”, generally defined as the ratio of the maximum cavity depth
to the minimum span across the opening (Gruenwald, 1987). The deeper the draw, the
less uniform the wall thickness and barrier properties (Robertson, 1993).
The design of thermoformed parts differ 6om injection-molded parts in that the
local thermoformed part wall thickness is dependent on the allocation of material from a
contiguous sheet onto a single-sided mold surface and not upon the designer’s options in
mold fabrication. For vacuum thermoforming, the container wall thickness decreases in
proportion to the distance down into the mold (Throne, 1991).
To estimate the thinning which may occur, one should determine the area of the
sheet available for forming and divide by the area of the finally formed part, including
trim. In reality none of the flat areas will have the calculated thickness uniformly distributed because these walls are not formed at the same time but in sequence. From a
line of mold contact the sheet next touches only that part of the mold adjacent to it, and, only after further stretching and thinning, continues to form these sides until finally the
inside comers fill. These last-formed edges have the thinnest wall thickness.
Consequently, calculated wall thicknesses can be considered only as a starting point
(Gruenwald, 1987).
To find the thickness variations, the formed prototype part can cut into small pieces- or small discs can be pimched out of it- and their thickness determined with a micrometer. Other methods include using a translucent or transparent colored sheet, where the differences in color intensity correlate to the thinning of the sheet (Gruenwald,
1987).
4 Better wall thickness distribution is obtained using plug-assist or plug-assist vacuum thermoforming rather than using vacuum thermoforming. This is due to prestretching and better material reallocation in plug-assist and plug-assist vacuum thermoforming (Aroujalian et al., 1997). Prestretching enables material redistribution, with the objective being an increase in part wall thickness in load bearing regions.
Prestretching techniques have been devised to improve allocation across the shape
surface and to minimize thin comers. These techniques serve to move material differentially from those sheet areas that are not subjected to much stretching to those areas that are excessively stretched (Throne, 1996).
Typically, in the plug-assist thermoforming, the plastic sheet is stretched up to about 80% of the mold cavity with a plug and the process is then completed with vacuum. Similarly, in the plug-assist vacuum thermoforming the plastic sheet is stretched up to about 60% of the mold cavity with a plug before the process is completed with vacuum. Thermoforming of deep mold is more effective using plug-assist thermoforming than using the other techniques (Aroujalian et al., 1997; Chung and Lee,
1992).
In simple thermoforming, the local wall thickness at any point depends upon the relationship between the amount of material already on the wall and that remaining in the freely expanded hemisphere of material that has not touched the mold wall. Thus, it is typically the case that in a female mold, the sheet wall thickness is greater at the rim than at the bottom. Conversely, for a male mold, the sheet wall thickness is greater at the bottom than at the rim (Throne, 1991). The production of thermoformed containers made of coextruded materials has a unique limitation that is important to understand. Coextruded sheet materials provide exceptional barrier qualities and considerable ability to eliminate chemical migration.
The main reason for using coextruded layers in a sheet for making containers is that some of the individual layers possess these unique barrier properties. When these layers are combined into a multilayered coextruded sheet (or a laminated sheet), each layer brings its own barrier qualities to the combined sheet. In combination, the sheet layers complement each other and provide exceptional packaging medium that can substitute for most known metal and glass packaging materials (Florian, 1987).
Such coexruded barrier material is often formed into containers using the thermoforming process. The coextrusion method is the easiest one converting the barrier layers into a combined sheet form; each layer thickness is easily controllable. For the best economy, each layer is extruded at the minimum thickness, where its barrier qualities will remain at satisfactory levels. The original thickness of this sheet will be reduced in the process of thermoforming. The prestretching and the final forming can occasionally cause heavy reduction in the original gauges. This reduction of gauge can easily thin out a specific layer so that its original barrier qualities is lost (Florian, 1987).
In addition to this natural overall thickness reduction, thermoforming has another limitation with respect to coextruded sheets. The natural gauge reduction is easy to compensate for with additional increases in layer thickness. However, each layer in the combined sheet composition is produced from different materials. The various layers are used for their different barrier qualities; they may not have similar or even close softening temperature points. In addition, they may also have different flow characteristics and
6 varying levels of submissiveness to stretching or tear resistance. When coextruded barrier sheets are thermoformed into products for the packaging of any sensitive preparation, such as food or drugs, the results of thermoforming must undergo through testing and scrutiny. The true limitation in thermoforming of such containers is found in the inability of the process to guarantee even reduction in the forming of the individual layers. Some barrier material layers may reach softening and flowing viscosities before other layer components do, thus creating inconsistent layering. Material may ooze out from between the firmer layers, causing a loss of barrier quality in those areas. Any container that has lost barrier tendencies even over only portions of its surface should be considered useless. It is important to know that thermoforming can cause such diminishing qualities in a barrier layer and that a greater thickness may be needed to compensate for the thinning. It is also important to subject the formed parts to proper testing to ensure valid barrier properties (Throne, 1996).
The wall thickness of a thermoformed part generally has to lie between tolerances set by the designer. Techniques, such as air-slip forming, plug-assist, and billow snap- back can be employed to achieve these tolerances; however, it is stated that whatever technique is employed, the thickness is never uniform and the variation in thickness in combination with the tolerances on thickness must set a limit on the possible depth-to- width ratio of a thermoformed part. Factors other than technique that affect the variation in thickness are, therefore, of some interest. Temperature and its variation across the sheet during thermoforming are such factors, but an important factor, if temperature variation is negligible, is the flow property of the plastic: plastics with different flow properties should result in parts with different degrees of thickness variation when
formed with identical techniques (Lai and Holt, 1975).
It has been described by Schmidt and Cariey that “plastic memory” where blown
bubbles returned completely to their original flat-sheet form either by suddenly releasing
the forming pressure or by annealing spheroidal bubbles at a proper temperature. This
complete recovery has led them to conclude that elastic-like behavior can be ascribed to
the process of blowing soft plastic sheets. In such a process all the work deformation is stored with no viscous dissipation which may be true at high rates of deformation much shorter than characteristic relaxation times of the shaped sheet. In such systems the
forming temperature would have minor effect on the thickness distribution which was pointed out by Lai and Holt (1975).
Poller and Micheali (1992) studied the effects of plug and film temperatures on wall thickness. They reported that plug and film temperatures are the major influencing parameters on wall thickness distribution when a plug-assist thermoforming method was used. However, Lai and Holt (1975) showed that plastic film temperature in the range of
150 to 170“C for polymethyl methacrylate (PMMA) and 110 to 130°C for high impact polystyrene (HIPS) had no effect on the wall thickness distribution when the plastic sheets were formed into domes. On the other hand, forming temperature was pointed out as a prominent parameter affecting sagging and drawability of the plastic sheet
(Rosenzweig, 1979).
Lai and Holt (1975) also stated that there was a minimum temperature above which the wall thickness distribution was independent of film temperature in vacuum thermoforming of polystyrene, polyethylene, polyvinyl butyral and polyvinyl chloride into conical and truncated cone shapes. Apparently below this temperature, the plastic was not sufficiently plasticized to assume the mold contours.
Aroujalian et al. (1997) studied the influence of processing parameters namely film temperature, plug velocity and temperature on wall thickness distribution in plug- assist vacuum thermoformed fresh strawberry containers using HIPS. The study showed that wall location, plug temperature, plug velocity and their interactions influenced the wall thickness. However, film temperature did not have significant effect on the wall thickness and variation factor.
Shih (1991) reported that the forming pressure did not affect the thickness distribution as long as it was over the threshold value to obtain better resolution at amorphous polyethylene terephthalate (APET) cup comers using pressure bubble plug- assist forming technique.
Previous studies have shown that thermoforming process is specific for material, technique and the machine used. The thermoforming process needs to be optimized to provide the desired end use properties since the process has a key influence on the barrier and mechanical properties. The temperature distribution has to be found empirically via the setting of the individual infrared heaters for the process if industrial thermoforming machines have no temperature control system.
Wall thickness distribution has been chosen as important quality index to evaluate the processing. The influence of the film temperature has been particularly decisive for an optimum wall thickness. However, it must be considered that the degree of crystallization and the orientation which have also key influence on the properties of formed part are also determined by the temperature. The temperature and other
9 processing parameters such as forming pressure, heating time and plug temperature need to be investigated for optimum wall thickness.
NONDESTRUCTIVE TECHNIQUES FOR INSPECTION OF SEAL INTEGRITY
OF FOOD PACKAGES
Package integrity is a measure of a package’s ability to keep the product in and to keep potential contaminants out (Guazzo, 1994). Adequate package integrity to assure that the product is contained within the package throughout the product shelf-life and throughout the distribution system and also to prevent the ingress of microorganisms, oxygen, filth or other environmental contaminants that could render the product unfit for consumption or which could simply reduce the quality of the product to a level less than that intended for consumption (Bourque, 1995).
The purpose of inspecting and testing flexible and semirigid containers is to ensure that the hermetic condition of the container has not been compromised. The degree of a container or seal defect may be classified by the impact the defect has on the hermetic condition. Defects are classified as critical, major and minor defect. Critical defect is a defect providing evidence that the container has lost its hermetic condition or evidence that there has been microbial growth in the container. Major defect is a defect that results in a container that does not show visible signs of having lost in hermetic condition, but the defect is of such magnitude that the container may have lost its hermetic condition. Minor defect is a defect that has no adverse effect on the hermetic condition of the container (Gavin and Weddig, 1995). These defects defined are mainly
10 caused by sealing equipment and improper sealing parameters which can compromise package integrity and performance.
The food industry has conventionally relied upon destructive methods such as bubble test, electrolytic test, dye penetration test, microbial challenge test, burst and seal strength test for assuring package integrity. While destructive testing is an indicator of packaging equipment performance, it does not provide the means of isolating and rejecting defective packages. Furthermore, destructive testing is expensive due to loss of product, package and time. It has been indicated that aseptic packaging operations may lose over 2% of their total production volume due to destructive testing (Floros and
Gnanasekharan, 1992).
Nondestructive methods of package integrity evaluation offer the advantage of
100% in-line testing and highly desirable. Nondestructive 100% in-line inspection would ensure product safety and enable rapid detection of deviations in processing and packaging operations. The highly competitive food industry has a need for application of nondestructive testing for aseptic packages and other forms of packaging such as cans, retort pouches, bottles and etc. (Floros and Gnanasekharan, 1992).
Currently available nondestructive testing methods can be classified on the basis of operating principle as follows:
1. Optical methods
2. Ultrasonic detection methods
3. Pressure difference methods
4. Other methods
11 1. Optical Methods
1.1. Visual Methods
Visual testing involves inspecting the seals for absence of voids, wrinkles or pleats, checking the seal alignment, and searching for product contaminated seals or delamination of packaging materials. Visual inspection is generally a gross leak detection technique with a sensitivity of 10'^ Pa m^/s. This method is time consuming, expensive and its monotonous nature induces operator fatigue (Floros and
Gnanasekharan, 1992).
Machine inspection is possible and practical but has inherent limitations in the ability of the sensors used to detect the types of hidden defects that can provide an entry pathway for pathogenic microorganisms (Morris et al., 1998).
1.2. Machine Vision
Machine vision is the use of digital imagery for measurement, inspection, and control (Shaw et al., 1995). Machine vision has been successfully applied to 100% inspection of package seals at full production rate. Where 100% inspection is not feasible for any reason, either monitoring of the fill process or statistically significant sub-sampling may suffice.
Machine-vision types of sensors to detect defects have limitations for the materials that are opaque or have obstructive overprinting (Morris et al., 1998) though it has been demonstrated that star burst defects can be visualized in foil packages
(Blakistone and Harper, 1995).
Shaw et al. (1995) reported that an online system was developed by Jones and
Griner in 1994, detecting seal flaws in the foil covering of rigid food containers. They
12 categorized the defects as follows: thin seal area, wrinkles in the foil in the seal area, channels all the way through the seal area, and contaminants pressed under-neath the seal
area. They found that flaws tended to have significant gradient vectors parallel to the seal
area, whereas, illumination artifacts tended to have gradient vectors perpendicular to the
seal area. The system was not developed beyond the prototype stage, but the results were
promising for future developments of machine vision defect detection.
Testing package integrity optically is an automated, and hence more rapid, form of visual inspection and can be described as computer-aided video inspection (Floros and
Gnanasekharan, 1992). The use of a video camera in combination with image processing
techniques have been reported as potential methods of detecting seal defects (Gagliardi et
al., 1984; Wagner, 1983).
A video-based image processing system can be used to automatically inspect
package seal integrity. Each seal material requires a specific lighting solution. It is
difficult to inspect package seals that are translucent or partially transparent. When using
reflective material, great care must be taken to avoid misleading results. Experimentation
with IR, UV, and visible light has shown that visible light is the best for defining seal
area (Gibson, 1995).
Video imaging applications are limited by contrast sensitivity problems and the
need for characterization of all possible defects. Detection of cracks and crevices is
difficult using conventional image processing equipment due to low contrast around the
seal area. A special lighting technique and filters have been used to minimize this
problem and effectively enhance low contrast images (Floros and Gnanasekharan, 1992).
13 2. Ultrasonic Detection Methods
These techniques have been widely used in many scientific and industrial areas.
These techniques are commonly used as diagnostic imaging tools to detect defects in steel and other metals. Wide-ranging application methods include measurement of anatomical structure and material properties, and non-destructive evaluation of influence of processing variables on fabricated products (Ozguler et al., 1998; Szilard, 1987).
In food industry, applicability of ultrasonic techniques to quality control of both fresh and processed foodstuff has also been studied. The stability of reconstituted orange juice, the skin texture of oranges, cracks In tomatoes and defects in husked sweetcom have been investigated with ultrasonics. Ultrasonic measurements have been employed to detect microbial spoilage in aseptically packed milk products through various packaging materials. Ahvenainen et ai. (1989a) reported that ultrasound images of milk products were digitalized and the numerical values of images correlated very well with the bacterial counts. Ahvenainen et al. (1989b) also studied the detection of microbial spoilage of various packaged foods such as ice cream, tomato sauce and pea soup using ultrasonic imaging.
Another application in the food industry is for non-destructive measurements of the thickness of eggshells (Gould, 1972). The speed of transmission of ultrasound pulses through fats and oils has been also studied in estimating their solid/liquid ratios (Miles et al., 1985).
Tests of a noncontacting acoustic technology indicated that it may have the ability to detect fill level and differentiate between containers that have vacuum (seal) and those that do not (leaker) (Rodriguez, 1995).
14 Acoustic methods are based on the use of high frequency ultrasonic waves at low power levels; hence, it is non-intrusive. A composite material with good bonding between its various layers can be excited by an ultrasonic pulse, and it resonates with a specific frequency. In principle, when some degree of delamination is present in the composite material, its resonant frequency is altered. This forms a basis for ultrasonic detection of seal integrity (Gnanasekharan and Floros, 1994).
A preliminary study done by Morris et al. (1998) showed that acoustic imaging can nondestructively image micrometer-scale seal discontinuity. It was indicated that a
10pm channel type discontinuity was detected using a SLAM (Scanning Laser Acoustic
Microscopy) with a 20pm resolution. However, a 10pm channel defect could not be measured with useful accuracy. Conventional ultrasound systems such in the case for
SLAM require water as a coupling medium. In SLAM, ultrasonic waves are transmitted through water and then measured by vibrometer. According to Morris et al. (1998), the on-line implementation of a SLAM based method seems impractical since the equipment is cumbersome and expensive.
A recent study by Ozguler et al. (1998) used ultrasonic imaging to detect micro leaks and seal contamination in flexible food packages by the pulse-echo technique. The results of this study showed that 17.3-MHz pulse-echo Backscattered Amplitude Integral
(BAI)-imaging detected deliberate channel defects in the range of 9.5-15pm and all simulated food strand inclusion defects in both types of plastic and foil containing retortable pouches.
15 3. Pressure Difference Methods
Pressure difference methods have been used to test cups, trays and pouches for leak detection. Packages to be tested are inserted into a specifically designed chamber that seals around them and test is conducted (Stauffer, 1992).
When there is a pressure differential across the wall of a package, a possible leak will cause a test gas such as air or nitrogen to flow in or out of the package leak. An observed gas flow is an indication of a leak. There are two common methods used for detecting gas flow. The first method is based on measuring pressure changes using a very sensitive pressure sensor and the second method is measuring deflections of the package wall caused by the gas flow using a proximity sensor (Stauffer, 1988; Floros and
Gnanasekharan, 1992).
MRE pouches formed using horizontal-form-fill-seal (HFFS) equipment were inspected for leaks and weak seals on-line non-destructively. The leak testing consisted of a unit placed immediately after the sealing station for detecting channel leaks and testing seal strength. According to Yam (1995), the system is capable of detecting 50 |un diameter channel leaks in filled pouches within a few seconds.
The accuracy and reliability of testing depends on the method used for detection.
Stauffer (1988) pointed out that the simple two step method, pressurizing and taking readings of simple pressure or vacuum drop, has often failed to detect major leaks. A stabilizing period before testing the packages in the three step approach was suggested to achieve reliable readings (Stauffer, 1988).
16 4. Other Methods
4.1. X-Ray
X-ray systems have so far not been able to demonstrate the ability to find an unfilled leak with any useful resolution due to the lack of differential in density between a whole seal and two unsealed but adjacent pieces of material (Morris et al., 1998). X ray will detect a 1 pm drop of water in the seal area but will not detect 1 pm channel leak because it cannot detect voids unless there is something in them (Blakistone and Harper,
1995).
4.2. Infrared Scanning
Lampi et al. (1976) described a non-destructive seal-defect-detection method that uses infrared radiometric scanning of heated seal surfaces. In this technique, the heat source and the detector are stationary, while the pouch seal area is passed between them at up to 15 cm per sec. In principle, defects in the seal area should impede heat flow sufficiently so that the detector can measure the temperature drop. A prototype machine to scan seals and reject defective ones has been constructed and proven feasible and reliable, but its cost is high.
The inhared scanner is technically applicable for detecting seal defects at speeds at least up to 6 inches of seal per second. Its pragmatic applicability depends on three points: (1) effectiveness of measures to eliminate concern over seal contamination, (2) definition of defects and establishment of acceptance/rejection criteria, and (3) justification of its relatively high cost (Lampi, 1977).
17 4.3. Testing of Aseptic Packages
There are three commonly used destructive tests that provide information on container integrity: (1) Teardown, (2) Electrolytic test, and (3) Dye test. However, the ideal test for a package would be a machine, which could detect untight or unsterile packages without opening the package. There are two instruments to perform such functions. The first is an instrument which continuously scans the package profile as they pass by on the conveyor. If the package leaks, then air entering the package should result in a change, which can be picked up by the scanner. This is similar to dud detectors used by the canning industry. However, with aseptic packages, the profile scanners are not effective because the package defect usually does not immediately change the package profile. The second instrument, the Valio Electester, can detect unsterile packages after they have been incubated for 7 days. The instrument senses difference in rotational inertia caused by the viscosity changes in the product. This instrument is applicable only to products which undergo a viscosity change when they spoil and requires a 7 day incubation followed by testing of each individual package
(Sizer, 1983).
Methods to nondestructively test seal integrity of rigid containers such as cans or glass jars led to confidence by both processors and consumers in these packaging forms.
However, there is not a way to evaluate semirigid and flexible containers. As more shelf- stable foods are packaged in heat sealable-containers, a reliable method to nondestructively test these containers will be needed by both food processors and regulatory agencies. A scientific examination of possible nondestructive techniques should lead to an improved control of heat-seals and better container integrity.
18 EFFECTS OF PROCESSING AND PACKAGING ON ORANGE JUICE
QUALITY AND SHELF LIFE
Orange juice is the most popular juice in the United States, comprising 60% of the total juice market revenue (Chen et al., 1993). Orange juice is popular not only because of its high vitamin C content but also its unique and delicate citrus flavor and balanced taste. However, this delicate fresh flavor of orange juice is easily changed by heat treatment during processing or storage (Nijssen, 1991). The juice undergoes compositional changes that invariably cause alteration in the original flavor and aroma and also vitamin C content of the hesh juice.
Thermal processing is the most common technology to inactivate microorganisms and enzymes in orange juice. Unfortunately, it also reduces nutritional and flavor qualities and produces undesirable ofif-flavor compounds (Tatum et al., 1975; Ekasari et al., 1986; Nijssen, 1991). Orange juice is susceptible to degradation not only by heat but also by microorganisms, enzymes, oxygen and light during processing and storage
(Graumlich et al., 1986; Trammell et al., 1986; Sadler et al., 1992). That’s why, citrus industry has been exploring innovative processing with minimal heat treatment to increase markets by improving nutritional and flavor qualities (Sadler et al., 1992).
Pulsed electric field (PEP) processing has been shown as an alternative nonthermal processing to minimize the destruction of flavor compounds and vitamin C content while it is inactivating microorganisms and enzymes (Duim and Peralman, 1987;
Mertens and Knorr, 1992; Zhang et al., 1995; Yeom, 2000). Unlike heat processing, PEP
19 did not cause significant change in Vitamin C (Yeom, 2000). However, it is necessary to
select a proper packaging material that is compatible with food so that benefits of PEF
processing can be maintained during storage time.
Packaging material selection as well as processing influences quality of foods during storage due to the absorption of flavor compounds by packaging materials or permeation through them and degradation of flavor, color and nutrients by oxygen transmitted through the package. Once the package is sealed, the shelf life and nutrient composition of the orange juice are greatly influenced by the barrier properties of the package, the interactions of the juice with the package and the environment during storage (Sizer et al., 1988).
Paperboard cartons are the most popular packaging materials of orange juice in the United States. Usually low-density polyethylene (LDPE) is the inner layer of paperboard cartons because of ease of sealing. However, many researchers have reported high losses of d-limonene and other aroma compounds fiom citrus juices in contact with
LDPE (Sheung, 1995; Halek and Meyers, 1989; Kwapong and Hotchkiss, 1987; Sadler and Braddock, 1991). Kwapong and Hotchkiss (1987) and Hotchkiss (1987) showed that the consequence of this absorption significantly affect sensory quality in model systems.
Mannheim et al. (1987) stated that d-limonene absorption shorted the shelf life of orange juice. Moshonas and Shaw (1989) claimed that absorption contributed to flavor changes, which were sensorially detectable in orange juice.
Orange flavor is a very complex mixture with more than two hundred volatile constituents (Shaw, 1991; Maarse and Visscher, 1989). Flavor components only occupy
0.02% total weight of orange juice. O f the 0.02 % volatile orange flavor components, 75-
20 98% are hydrocarbons (mainly d-limonene), 0.6-1.7% are aldehydes, 1% are esters, 1% are ketones and 1-5% are alcohols (Sizer et al., 1988; Shaw, 1991; Chen et al., 1993). Of these major flavor constituents, ethyl butyrate, acetaldehyde, citral, a-pinene, d-limonene and octanal were reported major contributors of orange flavor (Shaw, 1991; Ahmed et al., 1978).
In the hydrocarbon group, d-iimonene and a-pinene are believed to be the most important flavors in orange juice. Although d-limonene is the second abundant volatile component in orange juice after ethanol (Shaw, 1991), it was not found the most important flavor in orange juice (Sizer et al., 1988) but it acts as a carrier for other oil- soluble orange flavor contributors, a-pinene is another terpene hydrocarbon that is one of the most important contributors of orange flavor.
Octanal and decanal, two straight chain aldehydes are generally known as important compounds contributing orange flavor. Ethyl butyrate is a major ester component considered “top note' huit flavor in orange juice. Since it is a very important contributor to desirable orange flavor, a decrease in ethyl butyrate resulted in decrease in flavor quality in processed orange juice products (Nisperos-Carriedo and Shaw, 1990).
Sheung (1995) studied the sorption of d-limonene, a-pinene, ethyl butyrate, and octanal compounds in orange juice by low density polyethylene (LDPE), polyethylene terephthalate (PET), polyvinylidene chloride (PVdC) and ethylene vinyl alcohol (EVOH) copolymers in a model system. Sheung reported that d-limonene and a-pinene were more sorbed by LDPE and EVOH than by PET and PVdC packaging materials. Ethyl butyrate and octanal sorption did not differ between the four packaging materials.
21 Paik (1991) reported that polyester sorbed much less limonene and ethyl butyrate
than LDPE. Imai (1990) showed that the sorption of limonene in orange juice by co
polyester was significantly lower than EVOH and LDPE.
The study was done with orange juice packaged in LDPE/foil/paper/polyethylene
laminate cartons and glass containers. After about 2.5 months storage at 25°C, an
experienced taste panel detected a significant (p<0.05) difference between orange juices
in cartons and glass containers. Analysis of the d-limonene (one of the major components
of the essential oils in citrus juices) content showed that it had decreased from 70 to
40ppm in the cartons within 35 days. The limonene had been absorbed (scalped) by the
polyethylene surface of the package in contact with orange juice. As well, ascorbic acid
degradation and consequent browning was accelerated due to contact with polyethylene
film. Thus the shelf life Is determined by flavor changes to the juices as a result of
scalping of the flavor components by the package (Mannheim et al., 1987).
Mathlouthi (1994) reported that products such as spring water, milk and fruit juices frequently had a taint after being stored in bottles of LDPE or LDPE-lined cartons,
probably caused by volatile compounds like saturated imsaturated hydrocarbons,
aromatic hydrocarbons and aromatic hydrocarbons with an imsaturated side chain. Some
of them (C3 and Q-alkyl benzenes) were described as having ‘intense plastic off-flavor’.
A study comparing regular LDPE-coated paperboard cartons with cartons made
from a special board with a double thickness of LDPE on the inside and an aluminum foil
lined board found that the CO 2 content dropped to 25% or less of its original
concentration in the first two types o f containers after 13 days of storage, while it was
maintained in the foil-lined cartons (Robertson, 1993). 22 Polyolefîns can develop a wax-like odor if the package is overheated. Fatty acid amide, used as a slip additive, can also cause odor problems when it is stored too long or under conditions where the compound can oxidize. Pouches made from paper/foil/LDPE laminates were incubated at 60°C for 20 min and then analyzed by Gas Chromatography
(GC). Three major components were identified as acetaldehyde, allyl alcohol and acrolein. When odorous pouches were compared with nonodorous pouches, a direct correlation between odor, acetaldehyde and allyl alcohol levels was obtained. Those compounds were considered to be thermal oxidative decomposition products of polyethylene (Briston, 1974),
The sorption of dairy flavor compounds (aldehydes and methyl ketones) by LDPE films has been investigated quantitatively in an attempt to assist aseptic processors to select appropriate packaging materials for maximum flavor stability. Headspace analysis of UHT processed milk packaged in aseptic cartons revealed a loss of higher molecular weight flavor compounds after 12 weeks storage, due to interaction between the LDPE packaging material and the proteins (Robertson, 1993).
Overall polarity of flavor compound and packaging material, crystallinity of polymer, the ratio of surface area to volume and thickness of packaging materials were reported to be major factors which affect the flavor sorption (Landois-Garza and
Hotchkiss, 1987).
Isolation of Flavor Compounds by Solid Phase Micro Extraction (SPME)
SPME has been used in pharmaceutical, waste material, drug test in urine, pesticides and recently in food products. Solid phase micro extraction was first
23 developed for extracting volatile compounds from waste material (Arthur and Pawliszyn,
1990; Arthur et al., 1992).
This method is based on the adsorption of volatile compounds onto the binding site of coated fiber. An SPME unit consists of a holder, a fused silica fiber, which is coated with a layer of stationary phase such as nonpolar polymethylsiloxane, polar polyacrylate or mixture of both compounds. When a SPME fiber tip exposed in the headspace of a sealed sample vial, an equilibrium partition process occurs between the sample and the stationary phase (Zhang and Pawliszyn, 1994). The equilibrium partition of volatiles between the headspace of juice and the SPME stationary phase mainly depends on the incubation time, temperature, sample volume and sample concentration.
The volatile compounds adsorbed in the SPME fiber tip are desorbed in the inlet of GC, and then separated by a capillary column. Compared to other commonly used extraction methods, the SPME is solvent free, rapid, simple, inexpensive and with good repeatability which make this method suitable for flavor analysis.
The coating material may be selective toward some groups of compounds. Chin et al. (1996) reported that the polyacyrlate-coated fiber was more effective than the polymethylsiloxane nonpolar fiber for short chain fatty acid analysis.
Heating and agitation increase the release of analytes into headspace that facilitates headspace sampling. Zhang and Pawliszyn (1994) studied the absorption mechanism of SPME fiber. They observed that static headspace with an agitated aqueous phase provided more rapid binding of volatiles onto coated fiber than that from a static aqueous phase. It is reported by Zhang et al. (1994) that rate of SPME headspace extraction is affected by rate of mass transfer of analyte to the surface of sample matrix,
24 rate of desorption of analytes from the sample surface, convective transport of the analyte to the extraction fiber as well as rate of absorption for analyte onto the fiber coating.
Proper agitation and increasing the extraction temperature facilitate the mass diffusion.
In recent years, several researchers reported the applications of this method for food flavor analysis. Yang and Harper (1995) used SPME to determine possible off- flavor compounds in whey protein concentrates. They found that the species of volatile compounds absorbed on polyacyrlate SPME fiber was four times more than that with solvent extraction method. Yang and Peppard (1994) employed this technique to analyze flavor in ground coffee, a fruit juice beverage, and a butter flavor in vegetable oil.
Penton (1996) determined flavor volatiles in a fruit beverage with automated SPME.
Jia et al. (1997) optimized SPME conditions for headspace flavor compounds of orange juice. They reported that the equilibrium of flavor compounds between SPME coating and headspace required 30 minutes at 40°C or 20 minutes at 60°C . The amount of orange flavor compounds adsorbed by SPME coating decreased as the orange juice temperature increased from 25°C to 80“C.
Separation of Flavor Compounds by Gas chromatography (GC)
Gas chromatography is the most important and common separation technique to analyze individual compounds in a complex food matrix product. In order to better quantitatively and qualitatively analyze as many volatile compound as possible, sample preparation such as isolation and concentration techniques are needed prior to gas chromatographic analysis (Moshonas and Shaw, 1994; Nawar and Fagerson, 1962;
Sheung, 1995).
25 Solvent extraction is the first means to isolate volatile compounds. The major
disadvantage of solvent extraction before GC analysis is that it is too time consuming. It
usually needs several successive extraction steps and takes at least one day for sample
preparation. The low recovery and lack of reproducibility of this method are huge
drawbacks of quantitative analysis of flavor compounds (Sheung, 1995; Wolford et al.,
1963).
SPME coupled with GC requires neither solvent extraction and purification steps,
nor a complicated purge-and-trap apparatus. It is expected that the SPME-GC will
become an alternative or complementary method to other flavor analysis techniques.
In summary, PEF, an emerging nonthermal processing technology, inactivates
microorganisms in foods without the significant adverse effects on the flavor, taste and
nutrients caused by conventional thermal processing. Because PEF processing is controlled at ambient temperature for a very short treatment time of microseconds, it
provides fresh-like foods with safety and extended shelf life. However, keeping this
fresh-like flavor and nutritional value of PEF processed food during storage depends on
packaging materials and methods. Many researchers mentioned the advantage of PEF for retention of fresh flavor and nutrients (Sale and Hamilton, 1967; Dunn and Peralman,
1987; Mertens and Knorr, 1992; Castro et al., 1993; Zhang et al., 1994; Zhang et al.,
1995; Qin et al., 1996; Sharma et al., 1998; Qiu et al , 1998; Jia et al., 1999). However, no literature was found to quantitatively examine the effects of packaging materials on retention of food flavor, color and nutrients of PEF processed foods.
26 REFERENCES
Ahmed, E.M., Dennison, R.A. and Shaw, P.E. 1978. Effect of selected oil and essence volatile components on flavor quality of pumpout orange juice. J. Agric. Food Chem. 26:368-372.
Ahvenian, R., Wirtanen, G. and Manninen, M. 1989a. Ultrasound Imaging-A non destructive method for monitoring the microbiological quality of aseptically-packed milk products. Lebensm. Wiss. U. Technol. 22: 382-386.
Ahvenian, R., Wirtanen, 0. and Manninen, M. 1989b. Ultrasound Imaging-A non destructive method for monitoring the microbiological quality of aseptically packed foodstuffs. Lebensm. Wiss. U. Technol. 22:273-278.
Arndt, G.W. 1992. Examination of flexible and semirigid food containers for integrity. In FDA Bacteriological Analytical Manual, ed. AGAC, Arlington, VA.
Arnold, S.I. 1984. Thermoforming. In Packaging Alternatives for Food Processor. pp.61-62. NFPA. Washington, D.C.
Aroujalian, A., Ngadi, M.O. and Emond, J-P. 1997. Wall thickness distribution in plug- assist vacuum formed strawberry containers. Poly. Eng. Sci. 37: 178-182.
Arthur, C.L. and Pawliszyn, J. 1990. Solid phase microexraction with thermal desorption using hised silica optical fibers. Anal. Chem. 62:2145-2148.
Arthur, C.L., Killam, L.M., Bucholz, K.D. and Pawlis:qrn, J. 1992. Automation and optimization of solid-phase microextraction. Anal. Chem. 35:1962-1966.
Baner, A.L., Kalyankar, V. and Shoun, L.H. 1991. Aroma sorption evaluation of aseptic packaging. J. Food Sci. 56(4): 1051-1054.
Blakistone, B. and Harper, C. 1995. New developments in seal integrity testing. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone, and C.L. Harper, Eds. pp. 1-9. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Bourque, R.A. 1995. Importance of package integrity. In Plastic Package Integrity Testing Assuring Seal Quality. BA Blaldstone, and C.L. Harper, Eds pp. 11-14. Institute of Packaging Professionals and Food Processors Institute. Washington, D.C.
Briston, J.H. 1974. Plastics in contact with food. Food Trade Press Ltd., London.
27 Castro, A.J., Barbosa-Canovas, G.V. and Swanson, B.G. 1993. Microbial inactivation of foods by pulsed electric fields. J. Food Process. Preser. 17:47-73.
Chen, C.S., Shaw, P.E. and Parish, M.E. 1993. Fruit juice processing technology: orange and tangerine juices, pp. 110-156. Agscience, Inc., Florida.
Chin, H.W., Bernhard, R.A. and Rosenber, M. 1996. Solid phase microextraction for cheese volatile compounds analysis. J. Food Sci. 61(6): 1118-1122,1128.
Chung, W.R. and. Lee, T.W. 1992. Thermoforming simulation of a plastic cup. SPE ANTEC Tech. Papers. 38:117-121.
Dunn, J.E. and Peralman, J.S. 1987. Methods and apparatus for extending the shelf life of fluid food products. US Patent 4,695,472.
Ekasari, I., Jongen, M.F. and Pilnik, W. 1986. Use of a bacterial mutagenicity assay as a rapid method for the detection of early stage of maillard reactions in orange juices. Food Chem. 21:125-131.
Floros, J. D. and Gnanasekharan, V. 1992. Principles, technology and applications of destructive and nondestructive package integrity testing. In Advances in Aseptic Processing Technologies. R.K Singh and P.E Nelson, Eds. pp. 157-188. Elsevier Applied Science, NY.
Florian, J. 1987. Practical Thermoforming and Applications, Marcel Dekker Inc., NY.
Frados, J. 1976. Thermoforming plastic film and sheet. In Plastic Engineering Handbook, 4“* ed, pp. 273-325. Van Nonstrand Reinhold, NY.
Gagliardi, G.R., Sullivam, D. and Smith, N.F. 1984. Computer aided video inspection. Food Technol. 38(4): 53-59.
Garrett, K.M. 1988. Laboratory testing procedures: heat sealed package. In Plastic Package Integrity: Successfully Meeting the Challenge. NFPA. Food Processors Institute. Washington, D.C.
Gavin, A. and Weddig, L.M. 1995. Closures for semirigid and flexible containers. In Canned Foods: Principles o f Thermal Process Control, Acidification and Container Closure Evaluation. 6* ed. pp 151-166. Food Processors Institute. Washington, D.C.
Giacin, J R. 1995. Factors affecting permeation, sorption, and migration processes in package product systems. In Foods and Packaging Materials-Chemical Interactions. P. Ackerman, M. Jagerstad and T. Ohlsson, Eds. The Royal Society of Chemistry. Thomas Graham House, Science Park, Cambridge.
28 Gibson, T.G. 1995. A system for inspection of package seal integrity. SPIE. 2597:37-39.
Gnanasekharan, V. and Floros, J.D. 1994. Package integrity evaluation: criteria for selecting a method. Pack. Tech. Eng. 3(6): 44-46,48.
Gould, R.W. 1972. Non-destructive egg shell thickness measurements using ultrasonic energy. Poultry Sci. 51: 1460-1461.
Graumlich, T.R., Marcy, J.E. and Adams, J.P. 1986. Aseptically packaged orange juice and concentrate: a review of the influence of processing and packaging conditions on quality. J. Agric. Food Chem. 34: 402-405.
Gruenwald, G. 1987. Thermoforming-A Plastic Processing Guide, Technomic Inc., Lancaster, Pennsylavania.
Guazzo, D.M. 1994. Package integrity testing. In Parenteral Quality Control. M.J. Akers Ed. (2"“* ed), pp. 247-298. Marcel Dekker, Inc., New York.
Halek, G.W. and Meyers, M.A. 1989. Comparative sorption of citrus flavor compounds by low density polyethylene. Pack. Tech. Sci. 2:141.
Harper, C.L., Blakistone, B.A., Litchfield, J.B. and. Morris, S.A. 1995. Developments in food packaging integrity testing. Trends in Food Sci. Tech. 6(10): 336-340.
Hirose, K., Harte, B.R., Giacin, J R., Miltz, J. and Stine, C. 1988. Sorption of d-limonene by sealant films and effect on mechanical properties. Ch 3 In Food and Packaging Interactions. H.J. Hotchkiss, Ed. ACS Symp. Ser. 365. ACS. Washington, D.C.
Hotchkiss, J.H. 1987. Comparative sorption of aromatic flavors by plastics used in food packaging. Activities Report of the R&D Associates 39:65.
Hotchkiss, J. H. 1995. Overview on chemical interactions between food and packaging materials. In Foods and Packaging Materials-Chemical Interactions. P. Ackerman, M. Jagerstad and T. Ohlsson, Eds. The Royal Society of Chemistry. Thomas Graham House, Science Park, Cambridge.
Hriciga, A.L. and. Stadelman, D.J 1988. Flavor-polymer interactions: coffee aroma alteration. Ch 5 In Food and Packaging Interactions. H.J. Hotchkiss, Ed. ACS Symp. Ser. 365. ACS. Washington, D.C.
Imai, T., Harte, B.R. and Giacin, J R. 1990. Partition distribution of aroma volatiles firom orange juice into selected polymeric sealent films. J. Food Sci. 55(1):158-16I.
Jarman, D., Farahbakhsh, B. and Hendg, R. 1994. Ultrasonic inspection of seal integrity of bond lines in sealed containers. U.S. patent.5.372.042.
29 Jia, M.Y., Zhang, H.Q. and Min, D.B. 1998. Optimization of solid-phase microextraction analysis for headspace flavor compounds of orange juice. J. Agric. Food. Chem. 46:2744- 2747.
Jia, M., Zhang, Q.H. and Min, D.B. 1999. Pulsed electric field processing effects on flavor compounds and microorganisms of orange juice. Food Chem. 65:445-451.
King, P. 1995. The federal government’s role: regulating plastic container integrity. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone, and C.L. Harper, Eds. pp. 15-21. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Kinsler, L.E., Frey, A.R., Coppens, A.B and. Sanders, J.V 1982. Fundamentals of Acoustics, John Wiley & Sons, NY.
Kwapong, O.Y. and Hotchkiss, J.H. 1987. Comparative sorption of aroma compounds by polyethylene and ionomer food-contact plastics. J. Food Sci. 52(3): 761-763, 785.
Lai, M.O. and Holt, D.L. 1975. Thickness variation in the thermoforming of poIy(methyl methacrylate) and high-impact polystyrene sheets. J. Appl. Polym. Sci. 19:1805-1814.
Lampi, R.A., Schulz, G.L., Ciavarini, T. and Burke, P.T. 1976. Performance and integrity of retort pouch seals. Food Tech. 30(2): 38-48.
Lampi, R.A. 1977. Flexible packaging for thermoprocessed foods. In Advances in Food Research. C O. Chichester, E. M. Mrak and O F. Stewart. Eds. Academic Press, Inc., New York.
Landois-Garza, J. and Hotchkiss, J.H. 1987. Plastic packaging can cause aroma sorption. Food Eng. 4:39-42.
Maarse, H. and Visscher, C.A. 1989. Volatile compounds in food: Qualitative and quantitative data. Marcel Dekker, Inc., NY.
Mannheim, C H., Miltz, J. and Letzter, A. 1987. Interaction between polyetylene laminated cartons and aseptically packed citrus juices. J. Food Sci. 52(3): 737-740.
Marcy, J.E. 1995. Integrity testing and biotest procedures for heat-sealed containers. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone and C.L. Harper, Eds. pp. 35-47. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Mathlouthi, M. 1994. Food Packaging and Preservation. Blackei Academic and Professional. Glaslow, UK.
30 Mertens, B. and Knorr, D. 1992. Developments of nonthermal processes for food preservation. Food Tech. 46:124-133.
Miles, C.A., Fursey, G.A.J and. Jones, R.C.D. 1985. Ultrasonic estimation of solid/liquid ratios in fats, oils and adipose tissue. J. Sci. of Food Agric. 36:215-228.
Miltz, J. and Mannheim, C.H. 1986. The effects of polyethylene contact surface on the shelf life of food products. In Food Product Package Compatibility. Gary J.I., Harte, B.R. and Miltz, J, Eds. Technomic Publish. Co., Inc. Lancaster.
Morris, S.A., Ozguler, A. and O’Brien, W.D. 1998. New sensors help improve heat-seai microleak detection. Pack. Tech. Eng. 7(7): 44-49.
Moshonas, M.G. and Shaw, P.E. 1989. Flavor evaluation and volatile flavor constituents of stored aseptically packaged orange juice. J. Food Sci. 54(1):82.
Moshonas, M.G. and Shaw, P.E. 1994. Quantitative determination of 46 volatile constituents in fresh, unpasteurized orange juice using dynamic headspace gas chromatography. J. Agric. Food Chem. 42:1525-1528.
Nawar, W.W. and Fagerson, I.S. 1962. Direct gas chromatographic analysis as an objective method of flavor measurement. Food Tech. 11:107-109.
Nielsen, T.J., Jagerstad, I.M., Oste, RE. and Wesslen, B.O. 1992. Comparative absorption of low molecular aroma compounds into commonly used food packaging polymer films. J. Food. Sci. 57(2):490-492.
Nijssen, L.M. 1991. Off flavors. Ch 19 In Volatile Compounds in Foods and Beverages H. Maarse, Ed. Marcel Dekker, Inc., NY.
Nisperos,-Carriedo, M.O. and Shaw, P.E. 1990. Comparison of volatile flavor components in fresh and processed orange juices. J. Agric. Food Chem. 38:1048-1052.
Ozguler, A., Morris, S.A. and O’Brien, W.D. 1998. Ultrasonic imaging of micro-leaks and seal contamination in flexible food packages by the pulse-echo technique. J. Food Sci. 63(4): 673-678.
Paik, J.S. 1991. Flavor scalping phenomenon extensively examined in recent study. Food Eng. April: 64-66.
Penton, Z. 1996. Flavor volatiles in a fruit beverage with automated SPME. Food Test. Anal. June: 16-18.
31 Pieper, G., Borgudd, L., Ackerman, P. and Fellers, P. 1992. Absorption of aroma volatiles of orange juice into laminated carton packages did not affect sensory quality. J. Food Sci. 57(6): 1408-1411.
Poller, S. and Michaeli, W. 1992. Film temperatures determine the wall thickness of thermoformed parts. SPE ANTEC Tech. Papers. 38: 104-108.
Purohit, K.S. 1995. Biotesting of plastic containers. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone and C.L. Harper, Eds. pp. 67-77. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Qin, B.L., Pothakamury, U.R., Barbosa-Canovas, G.V. and Swanson, B.G. 1996. Nonthermal pasteurization of liquid foods using high-intensity pulsed electric fields. Crit. Rev. Food Sci. Nutr. 36(6): 603-627.
Qiu, X., Sharma, S., Tuhela, L., Jia, M. and Zhang, Q.H. 1998. An integrated PEF pilot plant for continuous nonthermal pasteurization of fresh orange juice. Trans. ASAE. 41(4): 1069-1074.
Risch, S.J. 1991. Introduction. In Food and Packaging Interactions-U. S.J. Risch and J.H. Hotchkiss, Eds. ACS Symp. Ser. 473. ACS. Washington, D.C.
Robertson, G.L. 1993. Food Packaging Principles and Practice. Marcel Dekker, Inc., NY.
Rodriguez, J.G. 1995. Noncontacting acoustic ultrasonic signature analysis development. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone and C.L. Harper, Eds. pp. 107-111. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Rosen, S.L. 1971. Fundamental Principles of Polymeric Materials, A Wiley Interscience, NY.
Rosen, S.R. 1992. Practical application of thinwall thermoforming molds from prototype to production on web fed inline thermoforming machines. SPE ANTEC Tech. Papers. 38: 518-521.
Rosenzweig, N., Narkis, M. and Tadmor, Z. 1979. Wall thickness distribution in thermoforming. Polym. Eng. Sci. 19:946-951.
Sadler, G.D. and Braddock, R.J. 1991. Absorption of citrus flavor volatiles by low density polyethylene. J. Food Sci. 56(1): 35.
Sadler, G.D., Parish, M E. and Wicker, L. 1992. Microbial, enzymatic and chemical changes during storage of fresh and processed orange juice. J. Food Sci. 57:1187-1191.
32 Sale, A.J.H. and Hamilton, W.A. 1967. Effects of high electric fields on microorganisms. Killing of bacteria and yeasts. Biochem. Biophys. Acta. 148: 781-788.
SAS Institute Inc. 1989. SAS/STAT User’s Guide, Version 6, Volume 2, Cary, NC.
Sharma, S.K., Zhang, Q.H. and Chism, G.W. 1998. Development of a protein fortified fruit beverage and its quality when processed with pulsed electric field treatment. J. Food Qual. 21:459-473.
Shaw, P.E. 1991. Volatile compounds in foods and beverages: Fruits II. pp. 305-327. Marcel Dekker, Inc., NY.
Shaw, E.E., Litchfield, J.B., Reid, J.F. and Harper, C.L. 1995. Detection of defects in plastic laminated foil pouches: machine vision sensing. In Plastic Package Integrity Testing Assuring Seal Quality. Blakistone, B.A. and Harper, C.L, Eds. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Sheung, S.K. 1995. Sorption of orange juice flavor compounds into polymeric packaging materials. Master thesis. The Ohio-State University, Columbus.
Shih, W.K. 1991. Thermoforming of amorphous PET. SPE ANTEC Tech. Papers. 37: 584-587.
Sizer, C.E. 1983. Methods for Container Testing. Capitalizing on Aseptic. NFPA. The Food Processors Institute. Washington, D.C.
Sizer, C.E., Waugh, P.L., Edstam, S. and Ackermann, P. 1988. Maintaining flavor and nutrient quality of aseptic orange juice. Food Tech. June: 152-159.
Stauffer, T. 1988. Non-destructive in line detection of leaks in food and beverage packages-an analysis of methods. J. Pack. Tech. 2(4): 147-149.
Stauffer, T. 1992. On-line leak and seal integrity testing of plastic containers. Pack. Tech. Eng. 1(1): 30-37.
Szilard, J. 1982. Ultrasonic Testing-Nonconventional Testing Techniques. John Wiley & Sons, NY.
Tatum, J.H., Steven, N. and Roberte, B. 1975. Degradation products formed in canned single strength orange juice during storage. J. Food Sci. 40: 707-709.
Throne, J.L 1991a. Guidelines for thermoforming part wall thickness. Polym. Plast. Technol. Eng. 30 (7): 685-700.
33 Throne, J.L. 1991b. Plug-assist thermoforming-a new design protocol for rectangular parts. Polym. Plast. Technol. Eng. 30 (7): 701-721.
Throne, J.L. 1996. Technology of Thermoforming, Hanser/Gardner Publications, Inc., Cincinnati, OH.
Toledo, R.T. and Chapman, J.R. 1973. Aseptic packaging in rigid plastic containers. J. Food Tech. 27; 68-76.
Trammell, D.J., Dalsis, D.E. and Malone, C.T. 1986. Effects of oxygen on taste, ascorbic acid loss and browning for HTST pasteurized single strength orange juice. J. Food Sci. 51:1021-1023.
Wagner, J. N. 1983. Inspecting the impossible. Food Eng. 55(6), 79-92.
Wolford, R., Attaway, J., Alberding, 0 . and Atkins, C. 1963. Analysis of the flavor and aroma constituents of Florida orange juices by gas chromatography. J. Food Sci. 28:320- 328.
Yang, X. and Peppard, T. 1994. Solid phase microextraction for flavor analysis, J. Agric. FoodChem. 12: 1925-1930.
Yam, K.L. 1995. Pressure differential techniques for package integrity inspection. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone and C.L. Harper, Eds. pp. 137-145. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Yeom, H.W. 2000. The effects of PEF on the quality of foods. Ph D. dissertation. The Ohio State University, Columbus.
Zhang, Z. and Pawliszyn, J. 1994. Headspace solidphase microextraction. Anal. Chem. 65: 1843-1852.
Zhang, Z., Yang, M. and Pawliszyn. 1994. Solid phase microextraction. Anal. Chem. 66:(17): 844A-853A.
Zhang, Q., Monsalve-Gonzaiez, A., Qin, B.L., Barbosa-Canovas, G.V. and Swanson, B.G. 1994. Inactivation of Saccharomyces cerevisiae in apple juice by square-wave and exponential-decay pulsed electric fields. J. Food Process Eng.. 17:469^78.
Zhang, Q.H., Qin, B.L., Barbosa-Canovas, G.V. and Swanson, B.G. 1995. Inactivation of £. coli for food pasteurization by high intensity short duration pulsed electric fields. J. Food Process Preserv. 19:103-118.
34 CHAPTER 2
INFLUENCE OF THERMOFORMING PARAMETERS ON WALL THICKNESS
UNIFORMITY OF FORMED CUPS PRODUCED BY A BENCO ASEPTIC
PACKAGING MACHINE
ABSTRACT
Effects of process parameters such as forming temperature, forming air pressure and heating time on wall thickness distribution in plug-assist thermoformed food containers using multilayered material were investigated. Multilayered rollstock-base material formed into containers by thermoforming process using a Benco aseptic packaging machine. Forming temperatures in the range of 131-170°C, air-forming pressures of 2, 3, 3.5 and 4 bars, and heating times of 66, 74, 84, 97 and 114 seconds were used in the thermoforming process. Analysis of wall thickness data obtained for the thermoforming parameters used in this study showed that wall thickness was significantly affected by forming temperature, pressure and heating time at 0.05 significance level.
Besides the processing parameters, wall location, container side, and their interactions significantly affected wall thickness. Forming temperature was found to be the principle parameter influencing wall thickness distribution in a plug-assist thermoforming
35 operation. The optimum operating conditions of the packaging machine for the thermoforming process are: 146-156°C for forming temperature, 2-4 bars for air-forming pressure and 74-97 seconds for heating time.
36 INTRODUCTION
Thermoforming is one of the most frequently used thermoplastic sheet-forming techniques in food packaging due to ease of production, low cost, high speed and high performance. In a thermoforming process, a premanufactured thermoplastic sheet is clamped in place and heated to its softening temperature (Rosenzweig, 1979; Florian,
1987; Toledo and Chapman, 1973). This softened material is formed into a chilled female mold or male mold, by applying atmospheric pressure against a vacuum (vacuum forming) or against a slight positive air pressure (pressure forming) (Rose, 1971; Frados,
1976; Rosenzweig, 1979; Florian, 1987; Aroujalian, 1997; Chung and Lee, 1992). Upon contacting the cold mold, the softened material freezes and is held against mold surface until the sheet becomes rigid (Rosenzweig, 1979; Throne, 1991a).
Thermoforming has some major limitations. Non-uniform wall thickness and thinning at container base are most significant ones (Aroujalian, 1997; Throne, 1996). A thickness profile is obtained during cooling since the soft sheet does not contact the mold surface at the same time but in sequence (Rosenzweig, 1979; Gruenwald, 1987). The edges formed last have the thinnest wall thickness. Thus, it is typical to have greater wall thickness at the rim than at the bottom for a female mold. However, for a male mold the wall thickness is greater at the bottom than at the rim (Gruenwald, 1987; Throne, 1991b).
Layer reductions in coextruded barrier materials and some loss of detail transfer from mold to the formed part are other limitations of thermoforming process (Florian, 1987).
For more accurate and uniform wall thickness distribution, techniques such as air- slip forming, plug-assist billow snap-back and others can be employed (Rosenzweig,
37 1979; Lai and Holt, 1975; Shih, 1991). Rosen (1992) reported that uniform wall
thickness could be produced using a female mold and a plug-assist thermoforming
technique. Prestretching enables polymer material redistribution with the objective being
an increase in part wall thickness in load bearing regions (Throne, 1996). However, Lai
and Holt (1975) stated that in general whatever technique is used for thermoforming, the
thickness is never uniform. Therefore, factors other than technique that affect the
thickness distribution are of interest.
Since wall thickness distribution has a key influence on the properties of the
formed part (mechanical and barrier), it is important to know the effects of processing
parameters on the wall thickness distribution during thermoforming (Poller and Michaeli,
1992). Poller and Micheali (1992) studied the effects of plug and film temperatures on
wall thickness. They reported that plug and Him temperatures are the major influencing
parameters on wall thickness distribution when a plug-assist thermoforming method was
used. However, Lai and Holt (1975) showed that plastic film temperature in the range of
150 to 170“C for polymethyl methacrylate (PMMA) and 110 to 130®C for high impact
polystyrene (HIPS) had no effect on the wall thickness distribution when the plastic sheets were formed into domes. On the other hand, forming temperature was pointed out as a prominent parameter affecting sagging and drawability of the plastic sheet
(Rosenzweig, 1979).
Aroujalian et al. (1997) studied the influence of processing parameters namely
film temperature, plug velocity and temperature on wall thickness distribution in plug- assist vacuum thermoformed fresh strawberry containers using HIPS. The study showed that wall location, plug temperature, plug velocity and their interactions influenced the
38 wall thickness. However, film temperature did not have significant effect on the wall thickness and variation factor.
Lai and Holt (1975) also stated that there was a minimum temperature above
which the wall thickness distribution was independent of film temperature in vacuum thermoforming of polystyrene, polyethylene, polyvinyl butyral and polyvinyl chloride
into conical and truncated cone shapes.
Shih (1991) reported that the forming pressure did not affect the thickness distribution as long as it was over the threshold value to obtain better resolution at amorphous polyethylene terephthalate (APET) cup comers using pressure bubble plug- assist forming technique.
Since there are controversial results about the effect of forming temperature on wall thickness distribution and lack of literature about the effects of other processing parameters on wall thickness uniformity especially using multilayered barrier materials, there is a need for further research. Most of the thermoforming studies have been done with single layered material commonly PVC and PS (Arnold, 1984). This study was conducted to determine the influence of processing parameters such as forming temperature, pressure and heating time in plug-assist thermoforming of multilayered material using a commercial scale Benco ASEPACK 2 aseptic packaging machine. The optimum operating conditions of the Benco aseptic packaging machine, which works with form-fill-seai principle, were also studied.
39 MATERIALS AND METHODS
Materials and Properties
The rollstock multilayered base material (Allistra Plastic Packaging Company,
Muncie, IN) consisted of high impact polystyrene (HIPS) as outer layer, low density polyethylene (LDPE) as inner layer and polyvinyledene chloride (PVDC or SARAN) in between as barrier layer, was used in this study. The material is a five layer composite of
HIPS/ adhesive/ Saran/ adhesive/ LDPE with the target thickness of
1.0414/0.0254/0.0914/0.0254/0.2642 mm (.041/ .001/ .0036/ .001/ .0104 inches) respectively for each layer. The total material thickness is 1.4478 mm (0.057 inch). The thermoforming temperature suggested by Allistra Plastic Packaging Company was 170°C on the HIPS side and 155“C on LDPE side.
E quipm ent
The Benco ASEPACK 2 aseptic packaging machine (Piacenza, Italy) in Figure
2.1 was used to form 200 ml containers. The machine consists of thermoforming, aseptic filling, heat sealing and cutting units. A thermoforming unit involves an infrared oven, a pair of stainless steel female mold with aluminum bottom insertion and a pair of stainless steel upper mold with plastic plugs in for clamping and pressurizing.
Rollstock-base material is sterilized in hydrogen peroxide tank which is then heated in infrared oven. Softened plastic sheet is thermoformed by forming unit which is explained in details in methodology section. The formed containers joined in a web move to the filling station, where aseptic product is filled. The filled containers advance
40 to the sealing unit, where presterilized lid stock having an inner surface coated with a heat-sealable material (polyethylene) is positioned over the web. The sealed containers move towards the cutting unit where cups are removed from the web. Thermoforming, filling and sealing all occur in an aseptic zone. Figure 2.1 illustrates this process.
A Dickson Model IR 550 infrared thermometer (Addison, IL) was used to measure the surface temperature of the film on the HIPS side. The thermometer is capable of measuring the surface temperature with the working distance of 7 inches to 10 feet. The temperature range of the probe is -46®C to +538°C and the accuracy is ±1% of full scale at 25"C ambient. The thermometer has 1.0 s average response time with the repeatability of ±1% of reading or ±1 digit.
A Mitutoyo Model 5C 721 Digimatic micrometer (Tokyo, Japan) was used to measure the thickness of formed containers. The micrometer measures in the range of 0-
25 mm (0-1 inch) with 0.001mm (0.00005-inch) sensitivity.
Methodology
The plastic sheet from the rollstock-base was heated, single-sided from the top, in the infrared oven until it became soft and then moved towards the forming unit. The heated web was clamped between the female mold and the upper mold. Plastic plugs descended, stretching the web into the mold. Forming was completed with compressed air without vacuum. Formed cups were cooled by the chilled mold (21°C), which was released as the plugs retracted and the molds moved down. The film then advanced one index at a time as the cups moved toward the filling station.
The infrared thermometer probe was placed under the plastic sheet facing the
HIPS side and the temperature was recorded manually immediately before the forming 41 process. The Benco has direct control for lateral and central oven heaters in a percent scale, rather than a temperature scale. Increasing the percentage of heater setting increased the film surface temperature. Therefore, the desired temperatures of film were provided with different settings of central and lateral heaters (Figure 2.2). The lateral heaters were always set up extra S% higher than the central heaters to compensate the heat deficiency at the edges.
A standard calibration curve in Figure 2.3 was obtained by plotting oven temperature versus infrared probe-measured temperature by placing plastic strips into an oven (Thermolyne, Dubuque, Iowa) at preset temperatures in order to calibrate the probe temperature readings. The probe-measured temperature was then converted to oven temperature at which plastic surface temperature should be using the calibration curve
(R^ = .9927, y = 1.0683x + 16.21). This temperature of film bottom surface calculated by regression equation was defined as forming temperature throughout this paper. The forming temperatures used in this study were 131, 140, 146, 152, 156, 165 and 170°C which were obtained using different heater settings. Forming pressure of 2.75 bars and heating time of 74 s were used.
Forming pressure was directly controlled with an air-forming pressure valve. The pressures of 2, 3, 3.5 and 4 bars were used in this study. The temperature (156°C) and heating time (80 s) were fixed for these experiments.
Heating times of 66, 74, 84, 97 and 114 s were used at fixed temperature (152"C) and pressure (2.75 bars). Fixed temperature was obtained using a heater setting versus forming temperature graph. A mold cycle time to form a cup was 4 seconds. Different delay times (tdeiay) for which the machine had direct control were considered between
42 mold cycles to obtain different heating times. Total cycle time (tcycie) was the time of each mold cycle (4 s) plus tdeiay between each mold cycle. First, total tcycie was calculated based on matching flow rate, and then desired delay time was determined (tdeiay = tcycie - 4 s) so that the same volume (200 ml) can be filled with different delay times. Heating time (th) was calculated based on average velocity and the distance that base material traveled from the entrance of the infrared oven up to the center of the forming unit. Calculated heating times and other processing variables are presented in Table 2.1 based on the following equations;
tcycie (s) = Vc (ml) X 2 X 3600 fs) (Vc = 200 ml) (I ) F(L/h) X 1000 (ml/L)
Vave (cm/s) = Dindex (cm) / tcyde (s) (D |„dex = 10.6 Cm) (2)
th (s) = D (cm)/ Vave (cm/s) (D = 71.5 cm) (3)
Where tcyde is mold cycle time (4 seconds) plus tdeiay which is the time delay between each mold cycle, Vc is the volume of each container (200 ml), F is the flow rate of food product (L/h), Vave is the average velocity for the film travel (cm/s), Dindex is the distance between lateral side of one container and lateral side of the following container, th is the heating time of the base material right afrer the material enters infrared heater till the forming process, D is the distance that base material travel from the entrance of the infrared heater up to the center of the forming unit.
After the thermoforming process, 10 containers were randomly selected from each processing condition for thickness measurement and visual analysis. Wall thickness was measured at 0, 10, 20, 30, 40 and SO mm distances from the rim on different sides
43 named S, N, E, W (Figure 2.4) using the Digimatic micrometer. Measurements were taken at 240 data points from 10 containers for each given processing condition. The effects of location, side, processing parameters (temperature, pressure and time) on wall thickness and their interactions were analyzed with 3 way ANOVA (two within subjects one between groups) using the SAS statistical package (SAS, 1989). Mean comparisons were done by the Tukey test at 5% significance level in order to determine if there is a significant difference between locations, sides and processing parameters.
44 RESULTS AND DISCUSSION
Analysis of wall thickness data showed that forming temperature, wall location, sides and their interaction affected wall thickness of thermoformed containers significantly (p< 0.05). In general, increasing the forming temperature in the range of
131-170°C decreased the wall thickness significantly at locations of 0, 10, 20 and 30 mm from the cup rim (Figure 2.5). The film first contacts with the upper part of the mold cavity, where it cools and can no longer be flowed. The cooling process does not take place instantaneously and the film stretches as it cools. With higher forming temperatures stretching continues for a longer time, resulting in decreased film thickness at the locations close to rim as stated by Poller and Micheali (1992). Our results showed similar trend. With increased forming temperatures (146-1 TO^C) container wall thickness uniformity also improved. At 13rC, the wall thickness at the top (0 mm) and base (50 mm) were 1.105 and .286 mm respectively, whereas at 170“C the wall thickness at corresponding locations were .533 and .403 mm respectively. Considering all locations, there was no significant difference in wall thickness at the temperatures of 131 and
140°C. However, the decrease in wall thickness became significant beyond 146 to
165°C. On the other hand, there was no significant difference observed in overall mean wall thickness when forming temperature was in the range of 146-165°C. Considering the statement that wall thickness is independent of film temperature above its softening point (Aroujalian, 1997; Lai and Holt, 1975), the temperature firom 146 to 165°C could be the softening range for the material used.
45 Table 2.2 represents mean wall thickness at each location and temperature. This table provides multiple statistical comparisons between locations for a given temperature or between temperatures at a given location.
As well as location, side also had a significant effect on wall thickness distribution (Table 2.3). There was no significant difference in wall thickness between
North (pull-tab location) and West sides at all temperatures. There was also no significant difference observed between South and East sides. However, the difference between two groups N, W and S, E became significant. Increasing the temperature did not decrease the wall thickness for a given side when temperature was raised from 146 to 170°C.
According to the feeding direction of the sheet, the N side was heated by the central heaters and S side by lateral heaters. Since the wall thickness at S and E were significantly higher than the N and W at all temperatures used in this study, it indicates non-uniform heating throughout the base material. Although lateral heaters were always set 5% higher than the central heaters, the central part of the material received more heat.
This caused non-uniform heating and further non-uniform thickness at different sides of the container. As indicated by Poler and Micheali (1992), the hotter part of the film was stretched to a greater extent than the cooler areas.
Although uniformity improves at higher forming temperatures (146-170®C), temperatures beyond 165 and 170"C caused very thin container wall and some visual problems were observed such as delamination. Thus, the temperature range of the Benco aseptic packaging machine for the multilayered material used in this study is suggested between 146 and 1S6°C for better results in terms of uniformity and physical quality of containers. 46 Pressure affected wall thickness significantly as temperature did (Figure 2.6 and
Table 2.4). However, the effect of pressure on wall thickness distribution was not found as important as temperature effect. Increasing the pressure from 2 to 3 bars did not result in any significant change on wall thickness distribution at the locations of 10, 20, 30, 30,
40 and SO mm. Increasing the forming pressure beyond 3 bar resulted in significant decrease in mean wall thickness only at the locations of 0, 10, 20 mm. There was also no significant difference between 3.5 and 4 bar at all locations from top (0 mm) to the base
(50 mm). As it was stated by Shih (1991), the forming pressure does not affect the thickness distribution much as long as it is over the threshold value.
Mean wall thickness at sides of N and W was significantly lower than the sides of
S and E at all forming pressures (Table 2.5). There was no interaction observed between pressure and side. Therefore, increasing the pressure did not result in a significant change of wall thickness at the corresponding side, thus supporting that non-uniform wall thickness at these sides being the effect of non-uniform heating rather than pressure changes.
The operating range of the packaging machine for pressure was between 0 and 6 bars. However, the pressures between 2-4 could be used in this study, 2 being the lowest critical forming pressure. The machine did not allow the use of pressure beyond 4 bars.
Thus, forming pressure for the machine was suggested in the range of 2-4 bars.
Heating time showed a significant effect on the wall thickness at certain locations close to the top of containers (Figure 2.7 and Table 2.6). Increasing the heating time in the range of 66-114 seconds resulted in increased wall thickness at the locations of 0, 10,
20 mm. The reason wall thickness increased by increasing the heating time at the
47 corresponding locations is that temperature of top surface of plastic is higher than that at the bottom surface. The shorter the heating time the hotter the top surface increasing the temperature difference between the top and the bottom. The hotter the top surface, it becomes less viscous, stretches more during forming, increasing the uniformity of wall thickness. During the experiments, bottom surface temperature was maintained the same
(152“C) for all heating times. There was no significant difference obser\'ed in wall thickness when the heating time was raised from 66 to 74 seconds. Increase in wall thickness became significant after 84 seconds. However, increasing the heating time further beyond 97 seconds did not cause any significant difference in wall thickness at almost all locations except that the location close to rim (0 mm). There was no significant difference in wall thickness from 30 to 50 mm at all heating times.
Mean wall thickness at sides of N and W was again significantly lower than the sides of S and E at all heating times (Table 2.7). There was no interaction observed between side and time making the wall thickness at corresponding sides independent of heating time.
The desired forming temperature (1S2"C) was obtained with the combination of the heater setting and the heating time. It was necessary to have higher heater setting with shorter heating time to ascertain the desired forming temperature. It was found that a low heater setting with longer heating time caused higher difference in average wall thickness from the top, close to rim, to the bottom, close to base.
48 CONCLUSIONS
Since wall thickness has a direct influence on the strength and permeation properties of food containers, the optimum processing parameters were identified to obtain a uniform wall thickness. Wall thickness of the thermoformed food containers produced by the Benco aseptic packaging machine was significantly affected by forming temperature, forming pressure and heating time (p^ 0.05). In addition to these processing variables, wall location and side of container played a significant role on wall thickness distribution. Among these factors, forming temperature was found to be the major influencing parameter of wall thickness distribution in a plug-assist thermoforming operation using a multilayered plastic material.
49 REFERENCES
Arnold, S.L 1984. Thermoforming. In Packaging Alternatives for Food Processors. pp.61-6., NFPA. Washington, D.C.
Aroujalian, A., Ngadi, M.O. and Emond, J-P. 1997. Wall thickness distribution in plug- assist vacuum formed strawberry containers. Poly. Eng. Sci. 37:178-182.
Chung, W.R. and Lee, T.W. 1992. Thermoforming simulation of a plastic cup. SPE ANTEC Tech. Papers. 38: 117-121.
Florian, J. 1987. Practical Thermoforming and Applications, Marcel Dekker Inc., NY.
Frados, J. 1976. Thermoforming plastic film and sheet. In Plastic Engineering Handbook. 4“* ed, pp. 273-325. Van Nonstrand Reinhold, NY.
Gruenwald, G. 1987. Thermoforming-A Plastic Processing Guide, Technomic Inc., Lancaster, Pennsylavania.
Lai, M.O. and Holt, D.L. 1975. Thickness variation in the thermoforming of poly(methyl methacrylate) and high-impact polystyrene sheets. J. Appl. Polym. Sci. 19: 1805-1814.
Poller, S. and Michaeli, W. 1992. Film temperatures determine the wall thickness of thermoformed parts. SPE ANTEC Tech. Papers. 38: 104-108.
Rosen, S.L. 1971. Fundamental Principles of Polymeric Materials, A Wiley Interscience, NY.
Rosen, S.R. 1992. Practical application of thinwall thermoforming molds firom prototype to production on web fed inline thermoforming machines. SPE ANTEC Tech. Papers. 38: 518-521.
Rosenzweig, N., Narkis, M. and Tadmor, Z. 1979. Wall thickness distribution in thermoforming. Polym. Eng. Sci. 19:946-951.
SAS Institute Inc. 1989. SAS/STAT User’s Guide, Version 6, Volume 2, Cary, NC.
Shih, W.K. 1991. Thermoforming of amorphous PET. SPE ANTEC Tech. Papers. 37: 584-587.
Throne, J.L 1991a. Guidelines for thermoforming part wall thickness. Polym. Plast. Technol. Eng. 30 (7): 685-700.
so Throne, J.L. 1991b. Plug-assist thermofonning-a new design protocol tor rectangular parts. Polym. Plast. Technol. Eng. 30 (7); 701-721.
Throne, J.L. 1996. Technology of Thermoforming. Hanser/Gardner Publications, Inc., Cincinnati, OH.
Toledo, R.T. and Chapman, J.R. 1973. Aseptic packaging in rigid plastic containers. J. Food Tech. 27: 68-76.
51 F (L/h) tcvcle(s) ldcl«v (s) Vave (cm/s) th(s)
85 16.9 12.9 .6272 114
100 14.4 10.4 .7361 97
N 115 12.5 8.5 .8466 84
130 11.0 7.0 .9636 74
145 9.9 5.9 1.0707 66
Table 2.1. Heating time calculations based on processing parameters Mean Thickness (n=40) at Different Forming Temperatures
Location (mm) 131 °C 140°C 146“C 152°C 156°C 165“C 170“C
0 1.105a .974g 737h 653k 651k .574t 533i
10 .811b 733h .5161 .4631 .4601 .397uv .345jy
20 437cl 406cnu .346js .324jmw .336jos 309orwx 265ef
30 260df .23 Id 241de 247de 269efp .287fpx 282fpx % 40 251de .224d 259df .303moprq 313jnw .338jqw .343jy
50 .286finr 272efm .334jmsw 386nsuv .367suqy 419cv 403cv
Mean values with the same letter indicate no significant difference at 0.05 significance level
Table 2.2. Effects o f forming temperature on wall thickness of thermoformed containers at different locations Mean Thickness (n=60) at Diflerent Forming Temperatures
Side n r c I40“C 146"C 152“C 156“C 165“C 170“C
N .417ag .357dh .341 fhl 333h 342hm .349hm .323h
S .640bc .570e 464g 461g •452g •427gi .396dgk
E ,622c .564e •452g .453g 442gi .415gi .386dglm
W .420ag .403adfij .365dfh .338h .362hjkm .359fhkm 343hm
Mean values with the same letter indicate no significant difference at 0.05 significance level
Table 2. 3. Effects of forming temperature on wall thickness of thermoformed containers at different sides Mean Thickness (n=40) at Different Pressures
Location (mm) 2 bar 3 bar 3.5 bar 4 bar
0 .725a •763g .65 Ih 634h
10 531b .529b .4501 .441 ip
20 .367cj .366cj 340jkl 328kn
30 .285do 275do .277do .2580
40 .327ek .301 den 330eln 305dn
50 .383fc 390cm .410finp 386cm
Mean values with the same letter indicate no significant difference at 0.05 significance level
Table 2.4. Effects of air forming pressure on wall thickness of thermoformed containers at different locations Mean Thickness (n= 60) at Different Pressures
Side 2 bar 3 bar 3.5 bar 4 bar
N 360a .358a .352a 329a
S .494bd 496bd .458cde .448e
E .523b .531b .464de .449e
W .367a .364a 365a .342a
Mean values with the same letter indicate no significant difference at 0.05 significance level
Table 2.5. Effects of air forming pressure on wall thickness of thermoformed containers at different sides Mean Thickness (n=40) at Different Heating Times
Location (mm) 66 s 74 s 84 s 97 s 114 s
0 .673a .701ag 728g 812jg .839p
10 .467b 498bh .508h 582k 605k
20 .331cf .345fr .3470 .377el .388eo
30 256d .257d 258di 265dm 270dm
40 307ci .303cin .289imn 274dn .276din
50 .387eo .393e 380eo .361flo 366elr
Mean values with the same letter indicate no significant difference at 0.05 significance level
Table 2.6. Effects of heating time on wall thickness of thermofonned containers at different locations Mean Thickness (n=60) at Different Heating Times
Sides 66 s 74 s 84 s 97 s 114 s
N .339a 338a .341a 364af .369af
S 467b 494bcd ,482bcd .517cde .519ce
E .471 be 486bcd .SOlbcde .523de 549e
% W .337a .346af .348af .377af .392f
Mean values with the same letter indicate no significant difference at 0.05 significance level
Table 2.7. Effects of heating time on wall thickness of thermoformed containers at different sides VO
Figure 2.1. Schematic diagram of a Benco Asepack aseptic packaging machine: 1. Plastic material roll, 2. Hydrogen peroxide tank, 3 & 16. Mechanical dryers, 4 & 17. Pneumatic dryers, 7. Infrared heater, 8. Forming die, 9. Adjustment nut for plugs, 10. Filling unit, 13. Lidding material, 15. Hydrogen peroxide tank, 20. Sealing unit with vacuum chamber, 21. Sealing oven, 26. Cutting unit, 30. Waste material cutting unit, 32. Protective shields for sterile tunnel. 1 8 0
170 y = 1.4629X + 25.333 R^ = 0.9599 e 160 ïI ... 140 I 130
120 7580 8570 90 95 100
Heater setting (C%+L%)
Figure 2.2. Heater setting versus forming temperature graph
60 200
190 y = 1.0683x4-16.291 R^ = 0.9927 ^I 180 K 170 I I O 160
150
140 110 120 130 140 150 160 170 180
Infrared probe-measured temperature (°C)
Figure 2.3. Calibration curve for infrared probe temperature measurement
61 N
Top View
4
81 mm
0 mm
10 mm
20 mm Side View 52 mm 30 mm
40 mm
72 mm
Figure 2.4. Thickness measurement points of a thermoformed cup 62 1.2
131C I40C I 146C I52C 156C 0.8 165C s I70C
0 0.6 'ë
1 0.4
0.2
0 0 10 20 30 40 50
Location (mm)
Figure 2.5. Wall thickness distribution at different forming temperatures
63 2 bars 3 bars 3.5 bars 4 bars I
0 10 20 30 40 50
Location (mm)
Figure 2.6. Wall thickness distribution at different forming pressures
64 66s 0.9 74s 0.8 84s
0.7 14s I 0.6 I 0.5 0.4
0.3
0.2
0.1
0 0 10 20 30 40 50
Location (mm)
Figure 2.7. Wall thickness distribution at different heating times
65 CHAPTERS
INSPECTION OF SEAL INTEGRITY OF FOOD PACKAGES USING
ULTRASOUND AND PRESSURE DIFFERENTIAL TECHNIQUES
ABSTRACT
To prevent spoilage of content, food packages undergo incubation and inspection.
Over the years many inspection techniques are used with varying degree of success.
With this study, two techniques, ultrasound and pressure differential technique, were assessed and the effectiveness of these techniques was quantified. Non-contact, immersion type ultrasonic testing in pulse/echo mode using high frequency sound waves was evaluated for use in defect detection in the seal area of semi-rigid cups and polymeric trays. At disparities within the seal, sound waves generated by 20 MHz ultrasonic transducer were reflected back to the receiver. Received ultrasonic signals, or echoes, were used to develop A-scan and C-scan presentations. Discontinuities in the seal, short seal, non-bonded areas, imbedded foreign matter such as wire and teflon in the seal, contaminated seal and abrasion were detected using ultrasound based on reduced signal strengths.
66 Integrity of meais-ready-to*eat (MRE) pouches was inspected using a new pressure (vacuum) differential technique. This technique is based on a three-step approach. To evaluate the system air was linked into a test chamber through a calibrated needle valve simulating a leak at different leak rates. A simulated leak generated values higher than reference, and dependant upon the leak rate, was accepted or rejected by the leak tester. The technique proved to be effective, although, short-sealed non-vacuum packed poundcake pouches failed during subsequent leak testing.
67 INTRODUCTION
Destructive methods are commonly used to inspect the seal integrity of food packages. However, destructive test procedures incur significant cost due to loss of time, product and package material. This has resulted in an increased demand for reliable,
100%, on-line, non-destructive testing methods.
Current test methods for determining a loss of seal integrity and strength in flexible plastic pouches and semirigid containers with heat sealed lids include destructive test methods such as biotest, dye penetration test, electrolytic test, burst testing, tear/tensile test, trace gas sniffers, and pressure change devices (Morris et al., 1998;
Harper et al., 1995; Sizer, 1995; King, 1995; Arndt, 1992; Guazzo, 1994; Gavin and
Weddig, 1995; Gnanasekharan and Floros, 1994). These destructive methods are developed to be indicators of seal integrity if they are based on a statistically significant number of samples. However, they cannot be used reliably to indicate a loss in seal quality (Harper et al., 1995; Ozguler et al., 1998).
Existing destructive methods are inadequate for indicating microleaks, as regularly timed samplings may not indicate randomly occurring defects. These methods are very costly (Harper et al.; 1995; Marcy, 1995) and often time consuming (Garrett,
1988). Another drawback of most destructive testing methods is their inability to provide information related to leak size (Gnanasekharan and Floros, 1994).
Currently, on-line human visual inspection is being used to evaluate seal integrity, which imposes the limitations of both operator skill variability and human ocular resolution (»50pm). It is also costly, having been estimated at $10,000 per million
68 packages (Harper et al., 1995; Marcy, 1995). Human visual inspection suffers from the predictable effects of fatigue and error. Visual examination allows for the detection of obvious defects such as misaligned seals, burnt and contaminated seals, wrinkles, delaminations, and leakers (Marcy, 1995; Gavin and Weddig, 1995).
There is clearly a need for increased research efforts to develop 100%, on-line, non-destructive integrity testing systems to assure that every package is hermetically sealed (Bourque, 1995). A tool for nondestructive inspection system must be appropriately validated to be reliable, repeatable, and robust for sealed products (Purohit,
1995). According to Harper et al. (1995), research in this area is needed in two directions: first, the determination of threshold dimensions for defects; second, the development of an on-line non-destructive package integrity evaluation system. Non-destructive methods of package integrity evaluation should offer the advantage of 100% on-line inspection.
Implementation of such methods would ensure product safety and enable rapid detection of deviations in processing and packaging operations (Floros and Gnanasekharan, 1992).
One of the promising methods of non-destructive inspection of packages is the ultrasonic technique.
Ultrasonic Detection Methods
These techniques have been widely used in many scientific and industrial areas.
These techniques are commonly used as diagnostic imaging tools to detect defects in steel and other metals. Wide ranging application methods include measurement of anatomical structure and material properties, and non-destructive evaluation of influence of processing variables on fabricated products (Ozguler et al., 1998; Szilard, 1987).
69 In food industry, applicability of ultrasonic techniques to quality control of both fresh and processed foodstuff has also been studied. The stability of reconstituted orange juice, the skin texture of oranges, cracks in tomatoes and defects in husked sweetcom have been investigated with ultrasonics. Ultrasonic measurements have been employed to detect microbial spoilage in aseptically packed milk products through various packaging materials. Ahvenainen et al. (1989a) reported that ultrasound images of milk products were digitalized and the numerical values of images correlated very well with the bacterial counts. Ahvenainen et al. (1989b) also studied the detection of microbial spoilage of various packaged foods such as ice cream, tomato sauce and pea soup using ultrasonic imaging.
Another application in the food industry is for non-destructive measurements of the thickness of eggshells (Gould, 1972). The speed of transmission of ultrasound pulses through fats and oils has been also studied in estimating their solid/liquid ratios
(Miles et al., 1985).
Tests of a noncontacting acoustic technology indicated that it may have the ability to detect fill level and differentiate between containers that have vacuum (seal) and those that do not (leaker) (Rodriguez, 1995).
Acoustic methods are based on the use of high frequency ultrasonic waves at low power levels; hence, it is non-intrusive. A composite material with good bonding between its various layers can be excited by an ultrasonic pulse, and it resonates with a specific frequency. In principle, when some degree of delamination is present in the composite material, its resonant frequency is altered. This forms a basis for ultrasonic detection of seal integrity (Gnanasekharan and Floros, 1994).
70 A preliminary study done by Morris et al. (1998) showed that acoustic imaging
can nondestructively image micrometer-scale seal discontinuity. It was indicated that a
10pm channel type discontinuity was detected using a SLAM (Scanning Laser Acoustic
Microscopy) with a 20pm resolution. However, a 10pm channel defect could not be
measured with useful accuracy. Conventional ultrasound systems such in the case for
SLAM require water as a coupling medium. In SLAM, ultrasonic waves are transmitted
through water and then measured by vibrometer. According to Morris et al. (1998), the
on-line implementation of a SLAM based method seems impractical since the
equipment is cumbersome and expensive.
A recent study by Ozguler et al. (1998) used ultrasonic imaging to detect micro
leaks and seal contamination in flexible food packages by the pulse-echo technique. The
results of this study showed that 17.3-MHz pulse-echo Backscattered Amplitude Integral
(BAl)-imaging detected deliberate chatmel defects in the range of 9.5-15pm and all
simulated food strand inclusion defects in both types of plastic and foil containing
retortable pouches.
Pressure Differential Methods
Pressure difference methods have been used to test cups, trays and pouches for leak detection. Packages to be tested are inserted into a specifically designed chamber that seals around them and test is conducted (Stauffer, 1992).
When there is a pressure differential across the wall of a package, a possible leak will cause a test gas such as air or nitrogen to flow in or out of the package leak. An observed gas flow is an indication of a leak. There are two common methods used for detecting gas flow. The first method is based on measuring pressure changes using a
71 very sensitive pressure sensor and the second method is measuring deflections of the package wall caused by the gas flow using a proximity sensor (Stauffer, 1988; Floros and
Gnanasekharan, 1992).
MRE pouches formed using horizontal-form-flll-seal (HFFS) equipment were inspected for leaks and weak seals on-line non-destructively. The system consists of two separate units. The leak testing consisted of a unit placed immediately after the sealing station for detecting channel leaks and testing seal strength. According to Yam (1995), the system is capable of detecting 50 pm diameter channel leaks in filled pouches within a few seconds.
The accuracy and reliability of testing depends on the method used for detection.
Stauffer (1988) pointed out that the simple two step method, pressurizing and taking readings of simple pressure or vacuum drop, has often failed to detect major leaks. A stabilizing period before testing the packages in the three step approach was suggested to achieve reliable readings (Stauffer, 1988).
In this study a technical feasibility of using ultrasound in pulse- echo mode and a new pressure differential technique is reported for the inspection of seal integrity of three food packages. Ultrasound was used to detect physical defects of two package types: semi-rigid plastic cups and polymeric trays, both with heat-sealed laminated lids. MRE pouches were inspected for leaks using the pressure differential technique developed at
Packaging Technologies and Inspection (PTI, Tuckahoe, NY).
72 MATERIALS AND METHODS
Basic Components of Ultrasonic System
System components of the ultrasonic system used in this study are illustrated in
Figure 3.1. A general-purpose digitally controlled 35 MHz ultrasonic pulser/receiver
(Model Sonix DPR 35-S, Springfield, VA) controlled by a host personal computer was used with a 20 MHz ultrasonic transducer (Model Panametrics, Crescent St Waltham,
MA) which was a spherically focused, V* inch (6.35 mm) diameter. The ultrasonic pulse produced was propagated into the test material by the transducer. The same transducer received the reflected echo and converted it into an electrical impulse. The signal was then displayed on a digitizing oscilloscope.
A motorized three axis (x-y-z) scanning bridge capable of accurately positioning the transducer along any of the three axes was used to scan the seal. In a normal scan, the bridge traversed a rectangular area by moving along on one axis and indexing along the other axis.
Materials used for Ultrasonic Testing
Semirigid plastic containers with heat-sealed lid formed by a Benco aseptic packaging machine (Piacenza, Italy) were used in this study. Containers were thermoformed using roll-stock multilayered material (Allistra Plastic Packaging
Company, Muncie, IN) consisting of high impact polystyrene as outer layer, polyvinyledene chloride as barrier layer, and low density polyethylene as inner layer. The cups were filled and heat-sealed to a laminated lidding material (Allistra Plastic
Packaging Company, Muncie, IN) consisting of nylon as outer layer, aluminum foil as
73 barrier layer, and low density polyethylene as inner layer. Defective samples were prepared. Teflon and copper wire were placed over the container and taped prior to the filling and sealing operations. A group of containers was overfilled at the filling station to create incomplete seal. Short sealed containers were created by off cutting unit.
Polymeric trays supplied by US Army Natick RD&E Center with contaminated seal and abrasion on the seal were also tested. The tray structure has a multilayer base material (polypropylene, recycled polypropylene (regrind), ethylene vinyl alcohol, regrind and polypropylene) thermoformed immediately after sheet forming and heat- sealed using laminated lid material (from inside to outside polyolefin, aluminum foil, nylon, polyester).
Procedure used for Ultrasonic Testing
Both the transducer and the package to be tested were totally immersed in water.
The sound beam was directed onto the material by the transducer. The high frequency pulser-receiver operating in pulse/echo mode controlled by the PC was used to produce pulses which excited the spherically focused ultrasonic transducer. The puiser generated an electrical pulse, which was transmitted to the ultrasonic transducer. The transducer converted electrical excitation to an ultrasonic pulse. This ultrasonic pulse, generated by the piezoelectric type transducer, was propagated into the test material. At interfaces such as the front face and the back surface or internal defects, sound waves were reflected back to the same transducer where they were received. Received echo signals were amplified and then displayed on a digitizing oscilloscope.
The ultrasonic echoes were used to develop visual presentations such as A-scan and C-scan. An A-scan is x-y display of ultrasonic intensity as a function of time of travel
74 of ultrasonic pulse. In the A scan presentation, the horizontal base line (x) on the graph indicates elapsed time (from left to right) for the ultrasonic wave to travel through the material, and is translated in terms of material depth. The vertical axis (y) represents the
intensity of ultrasonic energy as the signal amplitude of the transmitted or reflected beams. The first peak on the A-scan graph represents a signal reflected from the front surface of the package. The second higher peak is the signal at disparities within the seal.
By scanning the transmitting/receiving transducers, the whole object or part of it can be mapped and the information can be displayed as C-scan. C-scan is more like a plan view from the above and provides depth information by using a color or a gray scale code.
The C-scan is color coded in decibels (dB) or in 16 linear increments and displayed on the color monitor.
In this study, multiple signals were recorded from the “good” or “undisturbed” region of the seal, and the peak from the “defective” or “disturbed” region was compared with the signal from the good seal area based on the reduced signal strength. To determine the correlation between ultrasonic response and the actual seal condition, various seal regions of a semirigid container with different ultrasonic signal strength were sectioned for optical microscopic imaging for seal characterization.
Calculations of Water Path and Material Depth
Water path is the distance between container front surface and the transducer, or the distance between transducer initial pulse and surface reflection in front surface follower gate (displayed in |xs and inch or mm on A-scan presentation). Water path was calculated based on the following formula where velocity of sound in the water was
0.0584 In/ps (1.483 mm/ps at 20®C)) (Kinsler et al., 1982).
75 Water path (In) = Velocity o f sound in water fin/psi X Time for sound to travel in water (usi 2
Material depth or thickness of material at the rim which is the distance between the front surface (first crossing point in the front surface follower gate) and the peak amplitude, was calculated using the following formula;
Depth (In)= Primary material velocity (In/ps) X Time for sound to travel in the material (us) 2
Velocity of primary material was calculated as 0.1176 In/ps by the following formula;
V = VI LI +V2 L2 (L=L1+L2) L
V: Velocity of sound in combined material or velocity of primary material
L: Thickness of combined material; 0.0630 In
VI: Velocity of sound in plastic base material: 0.1115 In/ps (2.83 mm/ps) (measured using Sonatest Masterscan Model 330)
LI : Thickness of plastic material at rim: 0.06 In
V2: Velocity of sound in lidding material: 0.24 In/ps (measured using Sonatest
Masterscan Model 330)
L2: Thickness of lidding material: 0.003 In
Procedure for Optical Microscopic Imaging
Areas of the food container to be sectioned were marked as closely as possible based on ultrasonic results. The food container was cut with a saw (Struers Accutom,
Westlake, OH) leaving the least deformation in the material edge. The samples were marked and then placed on a piece of acetate, which had adhesive on one side. The
76 samples stuck to the acetate to keep them horn floating in the liquid mounting material.
The edge to be ground and polished stuck to the adhesive.
A plastic-mounting ring was then placed on the adhesive side of acetate so that it surrounded the specimens. Liquid epoxy (epoxy resin, HQ) (Struers, Westlake, OH) was then mixed in a stirring cup in a ratio of 2cc catalyst (epoxy hardener or epoxy cold mounting resin which contains Triethylene tetramine) (Struers, Westlake, OH) to 15cc of epoxy. The mixture was gently but thoroughly stirred so that a minimum of air bubbles were trapped in the solution. The liquid epoxy was then carefully poured into the plastic- mounting ring until all the specimens were completely immersed.
The mounting ring was then placed in a vacuum chamber (Buehler, Lake Bluff,
IL) and the air was evacuated. This caused any entrapped air to be drawn out of the liquid epoxy. The vacuum was turned off and the solution allowed to set for about 24 hours or until it was well hardened. The epoxy mount was removed from the plastic- mounting ring, and the specimens were identified. The surface of the mount was ground using 320 grit paper and then progressively finer grits as 800, 1200, 2400, and 4000 grit until all desired surfaces were finely ground and flat. The surface was lightly polished
(Polisher-Grinder, Buehler, Lake Bluff, IL) using 0.05 alumina. The surface was examined under a light microscope for any voids or separation of the surfaces. The samples were photographed at SOX or any magnification available up to lOOOX.
Basic Components of Leak Tester
Components of leak tester (Model PTI FB-400, Tuckahoe, NY) are presented in
Figure 3.2A. A chamber in Figure 3.2B sealed with a gasket around the rubber was used to seal an MRE pouch and hold the vacuum inside during testing. This chamber was
77 flexible so that different size MRE pouches could be tested using the same chamber.
Flexibility of chamber is important to insure atmospheric pressure inside the package during the test. The chamber was connected to the air tubing and closing cylinder. Clean, dry shop air was hooked-up to the main air regulator. The filling valve was closed when the vacuum in the chamber reached the fill level. Afier the test was completed, a venting valve was opened to vent the chamber and release the seal. As soon as the vacuum was released from the chamber, the top part of the chamber was moved up and the package was removed from the bottom part of the chamber. The leak detector was then ready for another test. The tester was controlled by direct logic 205 PLC through PTI LT Windows
95 PC software. All data was automatically collected to Excel spreadsheet. Test vacuum ranged from -100 mbar to -900 mbar, with an option of -999 mbar with a deep vacuum pump. Readings were obtained using a vacuum transducer with accuracy of 0.5%.
Materials used for Leak Testing
MRE (Meals-Ready to Eat) pouches were provided by Sterling Foods Company
(San Antonio, TX). Different sizes of MRE pouches packed with oatmeal cookie, cracker and poundcake were tested. Oatmeal and cracker pouches were vacuum packed.
Pouches were heat-sealed in all four sides using laminated material, thick polyester as outer layer, polypropylene as inner layer and aluminum foil in between as barrier layer.
Procedure used for Leak Testing
Each individual pouch was enclosed in the flexible chamber and the test was started at predetermined conditions. In the first step (filling step) a vacuum was applied to a pre-set fill level (900 mbar) which was the level of vacuum in the chamber at the end of the fill time. The package was allowed to stabilize during a pre-selected time period,
78 called second step or equalizing step. Differential pressure was recorded during the final
step (testing step). Well sealed, defect free pouches were tested first, thus enabling the system to recognize good samples and to calculate reference values for filling rate (P/t),
starting pressure and differential pressure. Starting pressure was the absolute chamber pressure at the beginning of the test stage, which was normally lower than the filling pressure. Differential pressure or delta P was drop in the chamber pressure from the
moment test starts to the test time. Mean values, minimum and maximum readings were recorded based on total number of pouches and then system calculated the limit or
reference values based on the following formulas;
Fill Rate Reference = Average - (Sensitivity * Standard Deviation)
Start P Reference = Average - (Sensitivity * Standard Deviation)
Test Reference Point = Average + Sensitivity * (Standard Deviation + St Dev Avg)/2 where sensitivity is the number of standard deviation to be offset from the average.
Air was introduced into the test chamber through a calibrated needle valve simulating a leak at different leak rates. If a simulated leak generated values higher than the reference, it is rejected by the leak tester depending on the leak rate. If fill rate was less than the fill reference, the package was considered to have a large leak and the test was aborted. In the case where actual start pressure was less than the start pressure reference, the leak was classified as medium leak. If leak was detected during last step where delta P was higher than test reference point, the package was considered a small leak.
79 RESULTS AND DISCUSSION
Ultrasonic Test
To examine the applicability and sensitivity of the ultrasonic technique, a controlled laboratory study was conducted using samples with known seal defects such teflon and wire, and seal conditions such as short seal, contamination by food particles, abrasion and spillage due to overflll.
Seal contamination was defined as foreign matter in the seal area, such as, but not limited to, water, grease or food and indicated as one of the major defects for semirigid containers with heat sealed lids (Gavin and Weddig, 1995). Partial C-scan image of a polymeric tray with the defect of seal contamination by food particles is shown in Figure
3.3A. The color on the C-scan changed with the intensity of the ultrasonic signal received. In general, the black color in Figure 3.3 A depicted the strongest portion of the seal or good seal followed by gray and white. Good seal area or black color region had
100% signal amplitude in full screen height (FSH% on y-axis) which is seen by A-scan presentation in Figure 3.3B. A-scan presentation in Figure 3.3C showed that this signal amplitude dropped to 7.1% at seal contamination which was observed as white color on
C-scan image. The gray area on C scan had a signal amplitude of approximately 50%.
The study done by Ozguler et al. (1998) showed that 17.3 MHz pulse-echo technique was able to detect inclusion defects using 20-60pm tendons as the simulated food particles such as meat fiber that might be caught in the seal region of a package.
Figure 3.4 represents the partial C-scan image of another polymeric tray with an abrasion on the seal area. Good seal area or black colored region had 69.3% signal
80 amplitude which dropped to 18.9% where the abrasion was inspected. The abrasion was perceived as white color on C-scan image. Abrasion is another major or minor defect depending on severity associated with semirigid containers and defined as a scratch partially through the surface layers of the package caused by mechanical, rubbing or scuffing.
Five semirigid cups with short seal were scanned and all were readily detected with ultrasonic imaging. Amplitudes of signals at clockwise positions and short seal area were presented in Table 3.1. A representative C-scan image of a cup with short seal was chosen and presented in Figure 3.SA. Short seal was observed between 3 o’clock and 6 o’clock positions (Figure 3.5A) where loss in signal amplitude was observed relative to other positions such as 9, 12, 3 and 6 o’clock positions. Microscopic images of 12 o’clock and short sealed area associated with the represented cup in Figure 3.SA were shown in Figures 3.SB and 3.SC respectively. Unfavorable geometry in Figure 3.SC explained why there was a loss in strength of ultrasonic signal compared to 12 o’clock seal area which was quite uniform. Short seal or seal width variation is also one of the major defects associated with semirigid containers that can interfere with hermetic condition.
Semirigid containers were overfilled with water and heat sealed at 360°F.
Overfilled containers were not completely sealed due to spillage and all leaked between 3 o’clock and 6 o’clock positions where loss of signal amplitude was observed. C-scan image in Figure 3.6 showed the seal became incomplete between 3 o’clock and 6 o’clock positions. Signal amplitude at 3 o’clock seal area was 49.6%, which dropped to 13.4% at incomplete seal area. Incomplete seal can be defined as a portion of the seal that has a
81 lack of adhesion between lid and body. This defect is categorized as critical defect which provides evidence that the container has lost its hermetic condition and considered a potential public health problem. Table 3.2 represented signal amplitudes at different positions of overfilled containers (n=5).
Results showed that copper wire with diameter of 0.005 inch in the seal area was detectable by the ultrasonic imaging. A gap was observed on C-scan image where wire was implanted (Figure 3.7A). Signal amplitude of good seal area, 48.8%, dropped to
16.5% where wire was implanted in the seal. Optical microscopic pictures were taken from undisturbed (good) and disturbed positions were presented in Figures 3.78 and 3.7C respectively, supporting that ultrasonic signal amplitude was relatively high when seal area was uniform. Signal amplitudes of the undisturbed and disturbed seal areas of containers (n=9) with wire in the seal area were given in Table 3.3. Results showed that approximately 50% signal drop was observed when wire was imbedded in the seal compared to the reference signal at undisturbed regions of seal area.
A partial C-scan image of a semirigid cup with Teflon strand in the seal can be seen in Figure 3.8. Teflon as imbedded foreign matter in the seal area was not as readily detectable as wire. Among 10 containers. Teflon was detected in the seal area of only three containers. Figure 3.8 was the best representation where Teflon was detected.
Signal amplitudes of undisturbed and disturbed seal areas were obtained as 43.3% and
16.5% respectively. Jarman et al. (1994) applied transmission ultrasonic technique to detect contaminants such as Teflon, silicon grease and meat fiber within the seal region of a flexible packaging material and showed that these defects were detectable using ultrasonics.
82 Pressure DilTerential Test
MRE pouches were inspected for leak using the PTI leak tester. Results of oatmeal cookie pouches were presented in Figures 3.9, 3.10 and 3.11. Vacuum-packed oatmeal cookie pouches were tested at different equalizing and testing times. A minimum leak rate of 0.27 cc/min became detectable when the equalizing and testing time were 20 s and 25 s respectively (Figure 3.9). However, when the equalizing time was decreased to 10 s with the same testing time (25 s), minimum detectable leak rate was 0.51 cc/min
(Figure 3.10). With equalizing and test times reduced to 5 s and 10 s respectively, the minimum leak rate detected was 1.53 cc/min (Figure 3.11). Leak testing of cracker pouches gave similar results. Thus, a longer equalizing and testing steps allow for detection of a smaller leak rate and increased system sensitivity.
Pressure difference for the non-vacuum packed pouches were recorded higher than pressure difference for the vacuum packed pouches during testing stage (Figures
3.12 and 3.13).
83 CONCLUSIONS
In summary, the results of this laboratory work confirmed that ultrasonic is capable of detecting the defects in the seal and is a non-destructive tool to characterize
the heat seal layer. This study provides a basis for interpretation of ultrasonic signals and
seal defects. However, the technique used is not suited to packaging production lines
because it requires immersing the package in a fiuid bath to achieve ultrasonic coupling.
The leak testing technique proved to be effective, although, short-sealed non
vacuum packed poundcake pouches failed during subsequent leak testing. The leak testing using vacuum method can be suggested for packages having a certain amount of
headspace or residual gas so that gas can fiow out of the package since the pressure inside the package (at about 1 atmosphere) is higher than in the vacuum chamber.
84 REFERENCES
Ahvenian, R., Wirtanen, G. and Manninen, M. 1989a. Ultrasound imaging-a non destructive method for monitoring the microbiological quality of asepticaily-packed milk products. Lebensm. Wiss. U. Technol. 22:382-386.
Ahvenian, R., Wirtanen, G. and Manninen, M. 1989b. Ultrasound Imaging-A non destructive method for monitoring the microbiological quality of asepticaily packed foodstuffs. Lebensm. Wiss. U. Technol. 22:273-278.
Arndt, G.W. 1992. Examination of flexible and semirigid food containers for integrity. In FDA Bacteriological Analytical Manual, ed. AOAC, Arlington, VA.
Blakistone, B. and Harper, C. 1995. New developments in seal integrity testing. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone, and C.L. Harper, Eds. pp. 1-9. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Bourque, R.A. 1995. Importance of package integrity. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone, and C.L. Harper, Eds. pp. 11-14. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Floros, J.D. and Gnanasekharan, V. 1992. Principles, technology and applications of destructive and nondestructive package integrity testing. In Advances in Aseptic Processing Technologies. R.K Singh and P.E Nelson, Eds. pp. 157-188. Elsevier Applied Science, NY.
Garrett, K.M. 1988. Laboratory testing procedures: heat sealed package. In Plastic Package Integrity: Succes^lly Meeting the Challenge. NFPA. Food Processors Institute. Washington, D.C.
Gavin, A. and Weddig, L.M. 1995. Closures for semirigid and flexible containers. In Canned Foods: Principles of Thermal Process Control, Acidification and Container Closure Evaluation. 6'*' ed. pp. 151-166. Food Processors Institute. Washington, D.C.
Gnanasekharan, V. and Floros, J.D. 1994. Package integrity evaluation: criteria for selecting a method. Pack. Tech. Eng. 3(6): 44-46,48.
Gould, R.W. 1972. Non-destructive egg shell thickness measurements using ultrasonic energy. Poultry Sci. 51: 1460-1461.
Guazzo, DM . 1994. Package integrity testing. In Parenteral Quality Control. M.J. Akers, Ed. (2"^ ed), pp. 247-298. Marcel DeUcer, Inc., NY.
8S Harper, C.L., Blakistone, B.A., Litchfield, J.B. and Morris, S.A. 1995. Developments in food packaging integrity testing. Trends in Food Sci. Tech. 6(10): 336-340.
Jarman, D., Farahbakhsh, B. and Herzig, R. 1994. Ultrasonic inspection of seal integrity of bond lines in sealed containers. U.S. Patent.5.372.042.
King, P. 1995. The federal government’s role: regulating plastic container integrity. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone, and C.L. Harper, Eds. pp. 15-21. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Kinsler, L.E., Frey, A.R., Coppens, A.B. and Sanders, J.V. 1982. Fundamentals of Acoustics, John Wiley and Sons, NY.
Marcy, J.E. 1995. Integrity testing and biotest procedures for heat-sealed containers. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone and C.L. Harper, Eds. pp. 35-47. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Miles, C.A., Fursey, G.A.J. and Jones, R.C.D. 1985. Ultrasonic estimation of solid/liquid ratios in fats, oils and adipose tissue. J. Sci. of Food Agric. 36: 215-228.
Morris, S.A., Ozguler, A. and O’Brien, W.D. 1998. New sensors help improve heat-seal microleak detection. Pack. Tech. Eng. 7(7): 44-49.
Ozguler, A., Morris, S.A. and O’Brien, W.D. 1998. Ultrasonic imaging of micro-leaks and seal contamination in flexible food packages by the pulse-echo technique. J. Food Sci. 63(4): 673-678.
Purohit, K.S. 1995. Biotesting of plastic containers. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone and C.L. Harper, Eds. pp. 67-77. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Rodriguez, J.G. 1995. Noncontacting acoustic ultrasonic signature analysis development. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone and C.L. Harper, Eds. pp. 107-1II. Institute of Packaging Professionals and Food Processors. Washington, D.C.
Sizer, C.E. 1983. Methods for Container Testing. Capitalizing on Aseptic. NFPA. The Food Processors Institute. Washington, D.C.
Stauffer, T. 1988. Non-destructive in line detection of leaks in food and beverage packages-an analysis of methods. J. Pack. Tech. 2(4): 147-149.
86 StaufTer, T. 1992. On-line leak and seal integrity testing of plastic containers. Packaging Tech. Eng. 1(1): 30-37.
Szilard, J. 1982. Ultrasonic Testing-Nonconventional Testing Techniques. John Wiley and Sons, NY.
Yam, K.L. 1995. Pressure differential techniques for package integrity inspection. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone and C.L. Harper, Eds. pp. 137-145. Institute of Packaging Professionals and Food Processors. Washington, D.C.
87 Peak Amplitudes (%Full Screen Height) at Different Positions
9 o’clock 12 o’clock 3 o’clock 6 o’clock Short seal cup 1 16.5 29.9 4.7 8.7 5.8 cup 2 12.6 20.5 20.5 37.0 7.9 cup 3 26.0 58.3 44.1 23.6 15.0 cup 4 15.0 22.0 34.6 37.0 11.8 cup 5 13.4 19.7 17.3 40.9 5.1
Mean 16.7ab 30.1a 24.2ab 29.4a 9.1a
* 9 o'clock represents the pull tab position
Mean values with different letter indicate significant difference at 0.1 significance level.
Table 3.1. Peak amplitudes at short seal and different seal regions of semirigid containers Peak Amplitudes (%Full Screen Height) at Different Seal Regions
cup 1 13.4 20.5 55.1 31.0 6.3 cup 2 34.6 57.5 26.8 15.7 8.3 cup 3 19.7 52.8 30.7 49.6 13.4 cup 4 32.3 55.9 29.1 63.0 12.6 cup 5 41.7 52.8 15.7 41.7 9.9
Mean 283ab 47.9a 31.Sab 40.2a 10.1b
* 9 o'clock represents the pull tab position
Mean values with different letter indicate significant difference at 0.05 significance level.
Table 3.2. Peak amplitudes at different seal regions of semirigid containers with seal discontinuity Peak Amplitudes (% Full Screen Height)
Undisturbed seal Disturbed seal (with wire)
Cup 1 49.2 22.4
Cup 2 48.5 29.9
Cup 3 48.8 16.5
Cup 4 51.6 28.3
Cup 5 49.2 31.5
Cup 6 46.5 29.1 g Cup 7 58.5 28.3
Cup 8 49.6 33.1
Cup 9 51.6 25.2
Mean 493m 27.4b
Mean values with different letter indicate significant difference at 0.05 significance level.
Table 3.3. Peak amplitudes at disturbed and undisturbed seal regions of semirigid containers with wire embedded in the seal Z-Axis Manipulator Computer Printer Pulser/Receiver Immersion Motor Tank Transducer
Package
Figure 3.1. System components of non-contact immersion type ultrasonic system s
Figure 3.2 A. Components of leak tester Gasket seal
Rubber
Figure 3.2B. Picture of leak test chamber natioii
f Undisturbed Seal 4 ^
Figure 3.3 A. Partial C-scan image of a polymeric tray with seal contamination s
i
Figure 3.3B. A-scan presentation of undisturbed seal area of a polymeric tray Figure 3.3C. A-scan presentation of seal contamination area of a polymeric tray Undisturbed Seal
} Abrasion
Figure 3.4. Partial C-scan image of a polymeric tray with seal abrasion 12 o’clock
9 o’clock 3 o’clock 9>
g 6 o’clock Short seal
Figure 3.SA. C-scan presentation of a short sealed semirigid cup Nylon (outer layer of lid)
A1 foil LDPE (inner layer of lid) LDPE (inner layer of base) PVDC
HIPS (outer layer of base)
Figure 3.5B. Microscopic image of 12 o’clock position of a short sealed semirigid cup (SOX) Short seal area
S
Figure 3.SC. Microscopic image of short seal area of a short sealed semirigid cup (SOX) 3 o clock poa ition
Disconti nuity
Figure 3.6. C-scan image of a semirigid cup with seal discontinuity s
Figure 3.7A. Partial C-scan image of a semirigid cup with wire in the seal Figure 3.7B. Microscopic image of undisturbed seal area of a semirigid cup (SOX) Wire in the seal
g
Figure 3.7C, Microscopic image of seal area with wire of a semirigid cup (50X) s %
Figure 3. 8. Partial C-scan image of a semirigid cup with teflon in the seal 1000
900
800
700 I I I 8 600 Pk Avg Mn 500 1 Max 2 400 Um s 300 0.27cc/tnn O.SIcc/nin 200
100 4.03cc/rnn
10.21cc/nnin
0.00 2.50 5.00 7.50 10.00 12.50 15.00 17.50 20.00 22.50 25.00
Test Time, second
Figure 3.9. Pressure change versus test time during leak testing of MRE pouches with oatmeal cookie
(test time of 25 s, equalizing time of 20 s) 1500 1400 1300 1200 1100 1000 "5 Î 900 800 700 1 600 500 s 400 0.27cc/mm 0.51cc/nm 300 0.73ccmm 200 1.53cc/mn 100 4.03cc/tnn
0 11 11 I 11 I 11 I T I I I r r ■! T I 1 I I 1 1 I 11 I il I T TM I I 1 1 M r I I I 1 1 1 I 1 1 t I 1 I I ri 0.00 2.50 5.00 7.50 10.00 12.50 15.00 17.50 20.00 22.50 25.00
Test Time, second
Figure 3.10. Pressure change versus test time during leak testing of MRE pouches with oatmeal cookie
(test time of 25 s, equalizing time of 10 s) 1000
900
800
700
I 600 k k 500 3 Awg S 400 Mn Max 300 Um 0.73cc/min 200 1.53cc/nnin 4.03cc/iiin 100 10.21cc/rri
1.00 2.00 3.00 4.00 5.00 6.000.00 7.00 8.00 9.00 10.00
Test Time, second
Figure 3.11. Pressure change versus test time during leak testing of MRE pouches with oatmeal cookie
(test time of 10 s, equalizing time of 5 s) î Avg Mn I Max Um 0.22cc/min O.SIcc/min 1.53cc/inn 4.03cc/nin 10.21cc/in
I I I I I ! I I I I I # I I ...... 10.00 12.50 15.00 17.50 20.00 22.50 25.00
Test Time, second
Figure 3.12. Pressure change versus test time during leak testing of MRE pouches with pound cake
(test time of 25 s, equalizing time of 10 s) Awg MNn Max Um large leak large leak large leak large leak large leak large leak large leak large leak large leak
0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00
Test Time, second
Figure 3.13. Pressure change versus test time during leak testing of MRE pouches with pound cake
(test time of 10 s, equalizing time of 5 s) CHAPTER 4
DESIGN, CONSTRUCTION AND VALIDATION OF A SANITARY GLOVE
BOX PACKAGING SYSTEM FOR PRODUCT SHELF LIFE STUDIES
ABSTRACT
A glove box has been designed and constructed as part of an integrated pilot plant scale pulsed electric field processing and packaging system to facilitate studies of product shelf life with selected packaging materials. The glove box was sprayed with 35% hydrogen peroxide and exposed to germicidal UV light prior to packaging. A HEPA air filter provided bacteria-free air and a positive glove box pressure during filling. The filling assembly in the glove box was part of a sanitary fluid-handling system, which was sterilized with a pressurized hot water cycle (105°C, 30 minutes). Non-selective nutrient broth, pH adjusted to 3.7 to mimic high acid foods, was sterilized using the fluid handling system, and filled into presanitized bottles inside the glove box. These bottles were dipped into a 3% hydrogen peroxide bath and rinsed with sterile water prior to filling.
Negative and positive controls were included in the experiment. All bottles were incubated at 22°C and 37°C for two weeks and checked for microbial growth by measuring optical densi^ at 600 nm using a spectrophotometer and by plating on plate
111 count agar and potato dextrose agar for total aerobic and, yeast and mold counts, respectively. No turbidity or microbial growth was observed in the media filled in sanitized bottles using the sanitized glove box or negative controls at 22°C and 37°C after two weeks. Growth was evident in the positive control.
112 INTRODUCTION
A sanitary fluid handling system integrated with a pilot plant scale pulsed electric field (PEF) processing unit and a Benco aseptic packaging machine at The Ohio State
University offers flexibility of processing using heat, PEF, or a combination of both for fluid and particulate foods for extended product shelf-life (Ruhlman, 1999; Streaker,
1999; Yeom, 2000). However, the Benco aseptic packaging machine that uses the form- fill-seal principle can form only one type of base packaging material at a time. The forming and sealing operations are limited to packaging materials that can be thermoformed and heat sealed (Ayhan and Zhang, 2000). A glove box was constructed as part of an integrated pilot plant scale PEF processing and packaging system to fill preformed bottles of various packaging materials to evaluate product/package interactions during shelf life.
Wadsworth and Basette (1984) constructed a laboratory scale indirect ultra high temperature (UHT) system to process milk and collected processed milk with a sanitized glove box. A combination of chemical sanitizer and UV light was used to kill any surface microorganisms in the glove box before filling. They claimed that milk processed with this system remained sterile for at least 4 months at both 7° and 32°C storage temperatures.
Hydrogen peroxide has been used as a sterilant for food packaging materials for a number of years (Robertson, 1993). Ultraviolet at a wavelength of 2S4 nm has been used to control surface and airborne microorganisms in laboratories, cleanrooms, meat coolers and other facilities (Maunder, 1977). The lethal effect of hydrogen peroxide, ultraviolet
113 irradiation or combination of both on microorganisms including heat resistant spores has
been shown by many researchers (Bayliss and Waits, 1979; Bayliss and Waits, 1980;
Bayliss and Waits, 1982; Stannard et al., 1983; Stannard and Wood, 1983).
The objective of this study was to construct a glove box packaging system
integrated with pilot plant scale pulsed electric field or heat processing and validate processing, filling and packaging equipment for high acid food products.
This system offers flexibility of using various packaging materials with different sizes and shapes. The system will provide better understanding o f quality changes that occur during storage of PEF or heat-treated products in different packaging materials and aid in the selection of compatible packaging materials that match the required shelf life.
114 MATERIALS AND METHODS
Materials
Glass, polyethylene terepthalate (PET), and high density polyethylene (HDPE) with polypropylene (PP) caps were purchased from General Bottle Supply Co. (Los
Angeles, CA). Low density polyethylene (LDPE) bottles with PP caps were purchased from Consolidated Plastic Co. (Twinsburg, OH). The size of all bottles was 500 ml.
Nonselective nutrient broth (DIFCO, Detroit, MI) was used to validate the system. Plate count agar (PCA) and potato dextrose agar (PDA) (DIFCO, Detroit, MI) were used for surface plating. A germicidal UV light and pair of butyl gloves were purchased from
Cole Parmer (Vernon Hills, IL). A HEPA air filter was purchased from Fisher Scientific
(Pittsburgh, PA).
Design and Construction
A glove box consisted of a gas-tight box fitted with a window, a pair of gloves, a transfer chamber, a germicidal UV lamp, a fluorescent lamp and a HEPA air filter as illustrated by Figure 4.1. The glove box was made of stainless steel. The front window was made of glass. The inner and outer transfer chamber doors were made of plexiglass.
The transfer chamber with inner and outer doors allow easy transfer of preformed bottles between the pilot plant and the box, while maintaining an established sanitary glove box environment. A pair of butyl gloves resistant to chemicals was attached to glove ports at the front side of the box with stainless steel clamps. The glove ports and the glass were sealed with a gasket to provide low leak interior atmosphere. Control switches for the fluorescent and UV lamps were conveniently located on the control panel.
IIS The germicidal UV lamp (UV-C, 254 nm) with intensity of 76 pW/cm^ was attached to the ceiling of the glove box to control surface and airborne microorganisms in the glove box prior to filling. The HEPA air filter system with 0,3 pm pore size and
1600 cm^ filtration area was attached between a compressed air cylinder and the glove box to provide a positive pressure of clean air in the glove box. A stainless steel filling valve and pipeline were installed in the glove box and connected to the fluid-handling system (FHS). The fluid handling system has flexibility of connecting either to the glove box or to the aseptic packaging machine (Figure 4.2).
Chemical Cleaning Procedure
Glass and plastic bottles and their caps were dipped into 3% hydrogen peroxide bath and rinsed with sterile water in a sanitized chemical hood. An exhaust fan was kept on to remove hydrogen peroxide fiunes during washing. Chemically cleaned bottles were capped and leA inside the hood overnight. The concentration of residual hydrogen peroxide in the containers was determined by preliminary trials using a Chemetrics hydrogen peroxide residue test kit (CHEMetrics Inc., Claverton, VA).
The day before processing, the bottles and glove box were sanitized in the following steps. The filling unit assembly was removed from the glove box. The glove box was cleaned with soap and sprayed with 35% hydrogen peroxide and wiped out. The filling unit assembly was wrapped with aluminum foil and autoclaved at 121°C for 30 minutes separately and then placed inside the cleaned glove box. The UV light was kept on overnight to control surface and airborne microorganisms inside the glove box. The glove box front window and the outer transfer chamber door were covered with cardboard to prevent eye contact with the UV light.
116 Sterilization Procedure
The entire fluid handling system and the pipeline were hot water sterilized at
10S°C for 30 minutes. The end of the pipe attached to the filling unit inside the glove box was capped until sterilization was completed. The filling valve was kept open during hot water sterilization. After the sterilization was completed, the filling valve was closed, the end cap was removed and the autoclaved filling unit was attached to the filling valve using the gloves of the glove box.
Sixty liters of non-selective nutrient broth was prepared in the product tank of the fluid handling system and the pH was adjusted to 3.7 using citric acid. This solution was pumped through the entire fluid handling system, including the pipeline in the glove box for the sterilization-in-place (SIP) process at 10S°C for 30 minutes. The sterilization sequences were 35 minutes of preheating, 30 minutes of sterilizing at 105“C, and 20 minutes of cooling prior to filling. The system was pressurized at 110 kPa (16 psi), and the product flow rate was 98L/hr. The UV light was turned off and the HEPA filtered air was turned on to provide positive air pressure of 14-28 kPa (2-4 psi) inside the box during filling to avoid drawing contaminants inside.
Glove Box Filling Procedure
Chemically cleaned and tightly capped bottles were transferred to the glove box using the transfer chamber. Only one of the two doors of the transfer chamber was open at a given time. The bottles were opened inside the glove box, and filled with pH adjusted sterile media at the filling temperature of 29°C. Filled bottles were capped and transported out of the glove box using the transfer chamber.
117 After completion of the processing and filling, cleaning in place (CIP) procedure took place to clean the FHS and pipeline using chemicals (Streaker, 2000). The glove box was then manually cleaned.
Validation Procedure
Chemically cleaned bottles filled with the sterile media in the glove box were compared to positive and negative controls prepared. As negative control, the same media (from the tank, prior to sterilizing with the SIP) was filled in glass bottles and autoclaved at 121°C for 30 minutes. Empty glass bottles were autoclaved and filled with the sterile media (from the SIP of the FHS) in the glove box as another negative control.
As positive control, some of the chemically cleaned bottles filled with the media in the glove box were inoculated with natural flora of orange juice. Representative colonies were selected from PCA and PDA petri dishes plated with orange juice. All bottles were incubated at 22®C and 37“C for 2 weeks and checked for microbial growth by measuring optical density at 600 nm using Spectronic Genesys 5 spectrophotometer (Milton Roy,
Rochester, NY) and by plating on PCA and PDA for total aerobic and yeast and mold counts, respectively. The negative control, which was autoclaved at 121° C for 30 minutes, was used as a blank for optical density readings. PCA and PDA plates were incubated at 30°C for 48h and 22°C for 5 days, respectively, before enumeration. The entire experiment was duplicated.
118 RESULTS AND DISCUSSION
The concentration and amount of H2O2 necessary for sanitation of the bottles were determined by preliminary study. H 2O2 residue should be below 0.1 ppm per container right before filling with food product that is regulated by FDA (Robertson, 1993).
Dipping into 3% H2O2 bath and rinsing with sterile water provided the residue less than 0.1 ppm and this procedure was followed for sanitizing bottles before filling.
Results are presented in Tables 4.1 and 4.2 for the nutrient broth incubated at
22°C and 37°C for 2 weeks. The media filled in chemically cleaned or autoclaved bottles had very low absorbances (0.000-0.005) at 600 nm. There was no growth observed on
PCA and PDA plates of the nutrient broth filled in chemically cleaned or autoclaved bottles. Negative controls had no growth, as expected. Growth was evident in the positive control which had absorbance of 0.221 with total plate counts of 6.56 log
CFU/ml and, yeast and mold counts of 6.32 log CFU/ml at 22®C for 2 weeks. The difference in turbidity was visible between positive control and the media in chemically cleaned bottles (Figure 4.3).
Very similar results were observed after 2 week incubation at 37“C. The positive control had total plate count of 5.68 log CFU/ml when absorbance was 0.122. PCA and
PDA counts at 37°C were lower than that of at 22°C, which is optimum temperature of yeast and molds, major flora of orange juice. Yeom (2000) reported that major microorganisms in the orange juice were yeast.
119 CONCLUSIONS
In summary, the results showed that the media filled in chemically cleaned bottles using a sanitized glove box did not have turbidity or microbial growth at incubation temperatures of 22°C and 37°(C for two weeks. Non-turbid samples with no growth validated our cleaning and filling procedures for the glove box packaging system.
Single strength orange juice treated with pulsed electric fields was filled into various packaging materials using this glove box packaging system to evaluate shelf life and product package interactions during storage. Our results showed that PEF processed orange juice using this system was microbiologically stable for 112 days of storage at
4°C.
120 REFERENCES
Ayhan, Z. and Zhang, Q.H. 2000. Wall thickness distribution in thermofonned food containers produced by a Benco aseptic packaging machine. Poly. Eng. Sci. 40(1): 1-10.
Bayliss, C E. and Waites, W.M. 1979. The combined effect of hydrogen peroxide and ultraviolet radiation on bacterial spores. J. Applied. Bacteriology. 47:263-269.
Bayliss, G.E. and Waites, W.M. 1980. The effect of hydrogen peroxide and ultraviolet radiation on non-sporing bacteria. J. Applied. Bacteriology. 48:417-422.
Bayliss, G.E. and Waites, W.M. 1982. Effects of simultaneous high intensity ultraviolet irradiation and hydrogen peroxide on bacterial spores. Food. Tech. 17:467-470.
Maunder, D.T. 1977. Possible use of ultraviolet sterilization of containers for aseptic packaging. Food Tech. 31(4): 36-37.
Robertson, G.L. 1993. Food Packaging Principles and Practice. Marcel Dekker, NY.
Ruhlman, K.T. 1999. Product examination and reformulation for pulsed electric field processing. Thesis. The Ohio State University, Columbus.
Stannard, G.J., Abbiss, J.S. and Wood, J.M. 1983. Combined treatment with hydrogen peroxide and ultra-violet irradiation to reduce microbial contamination levels in pre formed food packaging cartons. J. Food Protec. 46(12): 1060-1064.
Stannard, C.J. and Wood, J.M. 1983. Measurement of residual hydrogen peroxide in preformed food cartons decontaminated with hydrogen peroxide and ultraviolet irradiation. J. Food Protect. 46(12): 1074-1077.
Streaker, C.B. 1999. Design, construction, and validation of a pilot plant fluid handling system for pulsed electric field processing and aseptic packaging of foods. Thesis. The Ohio State University, Columbus.
Wadsworth, K.D. and Basette, R. 1985. Laboratory scale system to process ultrahigh- temperature milk. J. Food Prot. 48 (6): 530-531.
Yeom, H.W. 2000. The effects of pulsed electric field on the quality of foods. Ph D. dissertation. The Ohio State University, Columbus.
121 Sample Absorbance Total Plate Count Y&M identification (600 nm) (Log CFU/ml) (Log CFU/ml)
Negative control' 0.000 <1 <1
Negative control^ 0.000 <1 <1
Positive control ^ 0.221 6.56 6.32
Samples'* Glass 0.001 <1 <1
PET 0.000 <1 <1
HDPE 0.003 <1 LDPE 0.002 <1 'Negative control: Media autoclaved at 12l°C for 30 minutes ^Media filled in autoclaved bottles in glove box ^Positive Control: Media inoculated with natural flora of orange juice ^Media filled in chemically cleaned bottles of various materials in glove box The coimts represented as <1 log CPU Est/ml had no colonies 6om the lowest dilution which was 10*'. Table 4.1. Absorbances, total plate count, and yeast and mold (Y&M) counts of the nutrient broth incubated at 22°C for 2 weeks. 122 Sample Absorbance Total Plate Count Y&M identification (600 nm) (Log CFU/ml) (Log CFU/ml) Negative control' 0.000 < l <1 Negative control^ 0.002 <1 <1 Positive control^ 0.122 5.68 5.16 Samples'* Glass 0.004 <1 <1 PET 0.002 <1 HDPE 0.004 <1 <1 LDPE 0.005 <1 <1 ‘Negative control: Media autoclaved at 12TC for 30 minutes ^Media filled in autoclaved bottles in glove box ^Positive Control: Media inoculated with natural flora of orange juice ^Media filled in chemically cleaned bottles in glove box The counts represented as <1 log CPU Est/ml had no colonies fi:om the lowest dilution which was 10’*. Table 4.2. Absorbances, total plate count, and yeast and mold (Y&M) counts of the nutrient broth incubated at 37°C for 2 weeks. 123 Product inlet Product outlet Control panel ^ ^ HEPA air filter Air gas c^inder Fluorescent light Filling valve < !u Transfer chamber Glove port Butyl gloves Figure 4.1. Schematic diagram of a glove box Product in Product CIP Tank Tank Chilled Water Hot Water Pump PEF Generator CIP loop Flow Meter PEF Hold Treatment Cooling Tube Heating Chambers Glove Box or Cooling Backpressure Benco Aseptic Packaging Tap Water Product out Figure 4.2. Flow chart of an integrated pulsed electric field processing and glove box packaging system. 125 Positive Control Negative Control Autoclaved Glass Chemically Cleaned Figure 4.3. Comparison of the nutrient broth filled in chemically cleaned bottles with positive and negative controls stored at 22*’C for 14 days 126 CHAPTERS EFFECTS OF PULSED ELECTRIC FIELD PROCESSING AND STORAGE ON STABILITY AND QUALITY OF SINGLE STRENGTH ORANGE JUICE ABSTRACT Microorganisms, flavor and color of pulsed electric field (PEF) treated orange juice were investigated during storage of 112 days at 4® and 22°C. Single strength orange juice was treated with PEF at electric field strength of 35 kV/cm for 59|is and filled in glass bottles using a sanitary glove box. PEF treated orange juice was microbiologically stable for more than 112 days at 4® and 22°C. Hydrocarbon flavor compounds, d-limonene, a-pinene, myrecene and vaiencene, had significantly (p <0.05) increased while octanal, decanal, ethyl butyrate and linalool had no change after the PEF processing. Hydrocarbon flavor compounds did not significantly change over the storage time at both storage temperatures. Meanwhile significant loss of octanal, decanal and ethyl butyrate was observed within 14 days at 4®C. However, degradation of the compounds did not have significant effect on the sensory quality. PEF treated orange juice retained its color for 112 days at 4®C. 127 INTRODUCTION Orange juice is the most popular juice in the United States, comprising 60% of the total juice market revenue (Chen et al., 1993; Graumlich et al., 1986; Marcy et al., 1989; Ting, 1980). Orange juice is popular not only because of its high vitamin C content but also its unique and delicate citrus flavor and balanced taste. However, this delicate fresh flavor of orange juice is easily changed by heat treatment during processing or storage (Graumlich et al., 1986; Trammell et al., 1986; Sizer et al., 1988; Sadler at al., 1992). The juice undergoes compositional changes that invariably cause alteration in the original flavor and aroma (Nisperos-Carriedo and Shaw, 1990). Pulsed Electric Field, an emerging nonthermal processing technique, has been shown as an alternative processing to inactivate microorganisms in foods without the significant adverse effects on the flavor, taste and nutrients caused by conventional thermal processing (Castro et al., 1993; Zhang et al., 1994; Qin et al., 1996; Sale and Hamilton, 1967; Dunn and Peralman, 1987; Mertens and Knorr, 1992; Zhang et al., 1995). Thermal processing is the most common technique to inactivate microorganisms and enzymes in orange juice. However, it also reduces nutritional and flavor qualities, and produces imdesirable off-flavor compounds (Tatum et al., 1975; Ekasari et al., 1986; Nijssen, 1991). Because PEF processing is controlled at ambient temperature for a very short treatment time of microseconds, it provides fresh-like foods with safety and extended shelf life. Jia et al. (1999) reported that the average losses of flavor compounds in orange juice processed by 240,480 ps PEF and heat at 90"C for 1 minute process were 3%, 9% 128 and 22%, respectively. However, they mentioned that this flavor loss was due to vacuum degassing in the PEF process using laboratory scale PEF unit. PEF processing caused less protein dénaturation and lower vitamin C loss in protein fortified orange juice comparing to heat processing (Sharma et al., 1998). Retention of vitamin C of PEF processed orange juice using a pilot plant scale PEF system integrated with aseptic packaging was reported by Qiu et al. (1998). However, there are no reports about the effects of PEF on fresh orange juice flavor compounds and flavor stability of PEF processed orange juice during storage at pilot plant scale. There is a need for studies at pilot plant or near-commercial scale to provide better understanding of PEF effects on orange juice flavor quality and stability. The objectives of this study were to evaluate flavor and color changes due to PEF processing and also to investigate the stability of flavor and color of PEF processed orange juice during storage of 112 days at 4° and 22"C using a sanitary fluid handling system in an integrated pilot plant scale PEF processing and glove box packaging system. 129 MATERIALS AND METHODS Materials Single strength Valencia orange juice was provided by Minute Maid (Houston, Texas) in 208L drum as frozen and kept in a -25°C freezer until processing. Standard flavor compounds of d-limonene, a-pinene, vaiencene, myrecene, octanal, decanal, ethyl butyrate and linalool were purchased from Aldrich Chemical (Milwaukee, WI). A solid phase microextraction (SPME) flber coated with 100 pm polymethylsiloxane, 10 ml serum bottles. Teflon coated rubber septa and aluminum caps were purchased from Supelco (Bellefonte, PA). Total plate count agar was purchased from Difco Laboratories (Detroit, MI). Glass bottles (500 ml) with polypropylene caps were purchased from General Bottles Supply Co. (Los Angeles, CA). Preparation and Processing of Orange Juice Freshly squeezed single strength Valencia orange juice, quickly frozen in 208 L drum was stored in a -25°C freezer until processing. The frozen juice was thawed at refrigeration temperature for twelve days prior to processing. Orange juice was processed using a sanitary fluid handling system in an integrated pilot plant scale PEF processing and glove box packaging system (Figure 5.1). The entire fluid handling system was sterilized by a sterilization-in-place (SIP) process at 105°C for 30 minutes. Single strength orange juice was treated with PEF at electric field strength of 35 kV/cm for 59ps. PEF treated orange juice was filled into sanitized bottles in a presanitized glove box, minimizing the amount of air headspace to 1%. Glass bottles 130 were sanitized by dipping into 3% hydrogen peroxide bath and rinsing with autoclaved water. The glove box was first sprayed with 35% hydrogen peroxide and exposed to germicidal UV light (Cole Parmer, Vernon Hills, IL) (UV-C, 254 nm) with the intensity of 76 pW/cm^ for overnight before processing. A HEPA air filter (Fisher Scientific, Pittsburgh, PA) system with 0.3 pm pore size and 1600 cm^ filtration area was installed to provide bacteria-free air and a positive pressure during filling processing in the glove box. Microbial Analysis PEF treated juice was tested for pathogens. Salmonella, Listeria monocytogenes and E. coli 0157;H7, one day after the treatment by Silliker Laboratories (Columbus, OH). Total aerobic plate count was determined using total plate count agar (PCA) before and right after processing, and during storage at 4° and 22*C. For each dilution prepared, duplicate samples were plated using surface plating method. PCA plates were incubated at 30°C for 48 hr before enumeration. Flavor Measurement Selected flavor compounds in the headspace of orange juice or standard compound solutions were analyzed by a combination of solid phase microextraction (SPME) and gas chromatography (GC). One ml orange juice or standard solution was transferred into a 10 ml serum bottle having a magnetic stirring bar (3x 10 mm). The sample bottle was sealed with a Teflon septum and aluminum cap. The SPME fiber coated with 100 pm polymethylsiloxane was inserted into the headspace of orange juice sample bottle, which was magnetically stirred and heated at 60°C for 20 minutes in water 131 bath to maintain the flavor compounds equilibrium between the headspace and the SPME coating. The SPME fiber was removed &om the sample bottle and inserted into the 0.75 mm i d. splitless glass liner of GC injection port, and held for 2 minutes at 220°C to desorb the flavor compounds adsorbed on the SPME coating. The desorbed flavor compounds were separated by a HP 5890 gas chromatography (Hewlett-Packard, Wilmington, DE) equipped with a HP-5 capillary column of 0.53 mm internal diameter x 30 m coated with 2.65 pm of 5% phenyl substituted methylpolysiloxane and a flame ionization detector (FID). The GC oven was programmed from 60°C to 120°C at 10 "C/minute, held for 1 minute; ramped to 140®C at 4 "C/minute, held for 1 minute; ramped to 200"C at 20“C/minute and then held for another 5 minutes at 200®C. The temperature of FID was set to 250®C. Ultra high purity nitrogen was used as the carrier gas with the inlet pressure of 80 psi. The flavor compounds were identified by comparing the retention times of gas chromatographic peaks with those of standard compounds under the identical experimental conditions. The peak areas or FID responses, which are highly proportional to the flavor concentration, were used to evaluate PEF processed juice versus fresh. During storage, flavor compounds were evaluated based on percentage of relative retention. GC peak area found at zero time was taken 100% as initial value. D-limonene, myrecene, a-pinene, valencene, octanal, decanal, ethyl butyrate, and linalool were tested in orange juice at 4° and 22**C. Samples were taken at 0,2, 7,14,28, 56, and 112 days for analysis. At each sampling date, duplicate samples (bottles) were taken from each storage temperature for flavor analysis by SPME-GC. 132 Color Measurement Color was measured using a HunterLab Ultrascan colorimeter (Hunter Associates Laboratory, Inc.Reston, VA). The values for L, a, and b were recorded to evaluate color in fresh and PEF treated orange juice and also during storage. The parameter L is a measure of brightness/whiteness and it ranges from 0 to 100 (100=white; 0=black); a is an indicator of redness and it varies from -a to +a (-a=green, a= red) and the parameter b is a measure of yellowness and it varies from -b to +b (-b=blue, +b= yellow). Duplicate samples were taken for color measurements at 0, 7, 14, 28, 56 and 112 days of storage at both storage temperatures. Sensory Analysis The sensory evaluation was based on a nine-point hedonic scale; 1= dislike extremely; 5=neither like nor dislike; 9=like extremely, to determine degree of liking for color and overall flavor of PEF processed Juice using an experienced panel of 12. Randomly coded samples were served at ~13°C. The PEF processed orange juices stored only at 4°C were evaluated by the panel at 2,7,14,28,56 and 112 days of storage. Data Analysis The entire experiment was duplicated as replication. Data were analyzed statistically using SPSS statistical package (SPSS, 1999). Two sample t test at a ^ 0.05 was used to compare fresh juice with PEF processed juice in terms of color and flavor. Analysis of variance (two way ANOVA) was carried out to determine if storage temperature and time had significant effects on the percent flavor retention and color. Tukey’s specific comparison test was used to determine which means were significantly 133 difTerent when a difference was found. All tests for significance were made at the o< 0.05. Sensory data were analyzed using one way ANOVA to determine effects of storage time. Specific differences were determined by the Tukey test at a 5% level of significance. 134 RESULTS AND DISCUSSION Previous study showed that maximum microbial inactivation of orange juice was achieved with electric fields of 35kV/cm for 59 ^is of total treatment time (Yeom, 2000). The PEF treated orange juice using these processing conditions was reported negative for the pathogens. Salmonella, Listeria monocytogenes and E. coli 0157:H7. Fresh orange juice had total plate count of 2.34 log CFU/ml. The total aerobic plate count was less than 1 log CFU Est/ml right after the PEF treatment and also during storage of 112 days at 4°C and 22“C in glass bottles. Effect of PEF on orange juice flavor compounds tested is listed in Table 5.1. The PEF processing did not result in significant (p^0.05) flavor loss of d-limonene, valencene, myrecene and a-pinene. In contrast, these hydrocarbon compounds were significantly (p^0.05) increased after the processing. D-limonene and myrecene in PEF treated juice increased 18% compared to fresh orange juice according to gas chromatography. Valencene and a-pinene increased 32% and 29% respectively. However, the increase in these compounds are most likely not due to the formation of these compounds by PEF. These hydrophobic flavor compounds tended to reside inside the pulp. The availability of these compoimds to headspace gas chromatography analyses will be increased after the PEF processing due to increase in cell permeability exposing the compounds from the pulp to aqueous phase. This observation suggests that orange juice flavor compounds which are generally nonpolar in nature may be easily extracted in the PEF treated orange juice. There was no significant (p <0.05) effect of the PEF 135 treatment on octanal, decanal, linaool and ethyl butyrate. Jia et al. (1999) reported that PEF processing for 240 ps resulted in 2.8% loss of limonene, 0.8% loss of a-pinene and 5.1% loss of ethyl butyrate, however, they stated that the loss was mainly due to vacuum degassing in the PEF process. No loss was reported for octanal and decanal by the same researchers. Retention (%) of PEF processed orange juice flavor compounds during 112 day storage at 4°C and 22°C in glass was shown in Figures 5.2 and 5.3 No significant (p <0.05) effects of storage time and temperature were observed on the relative retention (%) of d-limonene, myrecene and a-pinene. This indicated that the hydrocarbon flavor compounds tested in the PEF treated orange juice stayed stabile for 112 days in glass bottles at both storage temperatures of 4® and 22®C. Ackerman and Wartenberg (1986) also reported that no significant change was observed in the retention of d-limonene of aseptically processed orange juice stored for 7 weeks at 10®C in glass bottles. However, for aldehydes, octanal and decanal, both storage time and temperature played significant (p ^0.05) role on the retention (%). There was a significant reduction in % retention of octanal within 7 days at 4®C and 2 days at 22“C. The difference between 4®C and 22®C became significant after 2 day storage. For retention of decanal effects of storage time and temperature were more severe than octanal. There was significant decline in the % retention of decanal within 2 day storage at 4®C and 0 day at 22®C. As shown by Figures 5.2 and 5.3, significant (p ^0.05) reduction in the % retention of octanal and decanal occurred in the first few days of storage and stayed more stable at 4°C comparing to storage at 22®C for the remainder of the storage time. As it 136 was stated by Ackerman and Wartenberg (1986), octanal and decanal retention of aseptically processed orange juice in glass were reduced by about 35% and 70% respectively within 14 days of storage due to flavor degradation. As shown by Figures 5.2 and 5.3 significant (p ^0.05) reduction was observed in the retention of ethyl butyrate of PEF treated orange juice stored for 112 days at 4° and 22°C in glass bottles. No significant effect of storage time was found at 4°C; however, increasing storage temperature to 22°C resulted in significant loss of linalool after 2 day storage. During 112 day of storage, there were 29 and 9 % loss of ethyl butyrate and linalool at 4“C respectively. At 22®C, 35% loss for ethyl butyrate and 23% loss for linalool were observed at 112 days. Color, L, a and b, of fresh versus PEF processed orange juice were presented in Figure 5.4. There was a significant (p <0.05) difference observed between fresh and PEF processed orange juice for L, a and b values. PEF processed orange juice had significantly (p <0.05) higher L and b, and lower a values than fresh orange juice. PEF processing did not alter the juice color. However, negative effects of heat processing on color has been stated (Sizer et al., 1988). The stability of PEF processed orange juice color was shown in Figure 5.5. No significant (p <0.05) effects of storage time was observed on L, a and b values at 4°C. However, storage time played significant role on color at 22*’C. This indicated that color of PEF processed juice was stable during 112 day of storage at refrigeration temperature; however, increasing the storage temperature to 22°iC resulted in significant (p <0.05) decrease in L and b values afrer 28 days of storage. As it was stated by Sizer et al. (1988) 137 storage temperature remains as the single most important factor in achieving satisfactory shelf life and quality. The self life of PEF treated juice could be extended if we deaerated the juice since oxygen has detrimental effects on the retention of color and flavor. Mean hedonic scores for PEF processed orange juice color and overall flavor over the storage time at 4°C were presented by Figine S.6. There was no significant (p <0.05) change in color and flavor liking of PEF processed juice during storage of 112 days at 4°C. Significant (p <0.05) reduction in the retention of octanal, decanal, ethyl butyrate and linalool during storage did not have significant (p <0.05) effect on the perceived sensory attributes. Pieper et al. (1992) also reported that absorption of up to 50% of d- limonene and small amounts of aldehydes and alcohols had no significant effect on sensory quality of orange juice. 138 CONCLUSIONS PEF processed orange juice was microbiologically stable (<1 log CPU Est/ml) during storage of 112 days at both 4° and 22°C in glass bottles. Hydrocarbon flavor compounds tested were more stable than octanal, decanal and ethyl butyrate during storage. Significant reduction in the retention of octanal, decanal and ethyl butyrate within 2 weeks attributed to dissolved oxygen in the juice since the product was not deaerated prior to the processing. However, the loss of these flavor compounds had no significant effect on the perceived sensory attributes of orange juice at 4°C for 112 days. The PEF processed orange juice had a shelf life of more than 16 weeks at 4°C which is 6- 8 weeks for aseptically processed juice in cardboard packages at the market. Storage temperature, time and oxygen in the product remain as the most important factors in achieving satisfactory shelf life and quality of PEF processed orange juice when packaging material is inert. 139 REFERENCES Ackerman, P.W. and Wartenberg, E.W. 1986. Shelf-life of citrus juices: A comparison between different packages. Report of 19“* Symposium of International Federation of Fruit Juice Producers. Den Haah. pp. 143. Chen, C.S., Shaw, P.E. and Parish, M.E. 1993. Orange and tangerine juices in &uit juice processing technology, pp.l 10-156. Agscience, Inc., Florida. Castro, A.J., Barbosa-Canovas, G.V. and Swanson, E.G. 1993. Microbial inactivation of foods by pulsed electric fields. J. Food Process. Preser. 17:47-73. Dunn, J.E. and Peralman, J.S. 1987. Methods and apparatus for extending the shelf life of fluid food products. US Patent 4,695,472. Ekasari, 1., Jongen, M.F. and Pilnik, W. 1986. Use of a bacterial mutagenicity assay as a rapid method for detection of early stage of maillard reactions in orange juices. Food Chem.21: 125-131. Graumlich, T.R., Marcy, J.E. and Adams, J.P. 1986. Aseptically packaged orange juice and concentrate: a review of the influence of processing and packaging conditions on quality. J. Agric. Food Chem. 34:402-405. Jia, M., Zhang, Q.H. and Min, D.B. 1999. Pulsed electric field processing effects on flavor compounds and microorganisms of orange juice. Food Chem. 65:445-451. Marcy, J.E., Hansen, A.P. and Graumlich, T.R. 1989. Effects of storage temperature on the stability of aseptically packaged concentrated orange juice and concentrated orange drink. J. Food Sci. 54: 227-230. Mertens, B. and Knorr, D. 1992. Developments of nonthermai processes for food preservation. Food Tech. 46:124-133. Nijssen, L. M. 1991. Off-flavors. Volatile Compounds. Ch. 19 in Foods and Beverages. H. Maarse, Ed. Marcel Dekker, NY. Nisperos-Carriedo, M.O. and Shaw, P.E. 1990. Comparison of volatile components in 6esh and processed orange juices. J. Agric. Food Chem. 38:1048-1052. Pieper, G., Borgudd, L., Ackermann, P. and Fellers, P. 1992. Absorption of aroma volatiles of orange juice into laminaied carton did not affect sensory quality. J. Food Sci. 57:1408-1411. 140 Qin, B.L., Pothakamury, U.R., Barbosa-Canovas, G.V. and Swanson, B.G. 1996. Nonthermai pasteurization of liquid foods using high-intensity pulsed electric fields. Critical Reviews in Food Science and Nutrition. 36 (6): 603-627. Qiu, X., Sharma, S., Tuhela, L., Jia, M. and Zhang, Q.H. 1998. An integrated PEF pilot plant for continuous nonthermai pasteurization of fresh orange juice. Trans. ASAE. 41(4): 1069-1074. Sadler, G.D., Parish, M.E. and Wicker, L. 1992. Microbial, enzymatic and chemical changes during storage of fresh and processed orange juice. J. Food Sci. 57:1187-1191. Sale, A.J.H. and Hamilton, W.A. 1967. Effects of high electric fields on microorganisms. Killing of bacteria and yeasts. Biochem. Biophys. Acta. 148: 781-788. Sharma, S.K., Zhang, Q.H. and Chism, G.W. 1998. Development of a protein fortified fruit beverage and its quality when processed with pulsed electric field treatment. J. Food Qual. 21:459-473. Sizer, C.E., Waugh, P.L., Edstam, S., and Ackerman, P. 1988. Maintaining flavor and nutrient quality of aseptic orange juice. Food Tech. 42: 152-159. SPSS Inc. 1999. SPSS base 10.0 user’s guide. SPSS Inc., Chicago, IL. Tatum, J.H., Steven, N., and Roberte, B. 1975. Degradation products formed in canned single strength orange juice during storage. J. Food Sci. 40: 707-709. Ting, S.V. 1980. Nutrients and nutrition of citrus fruits. Ch. 1 In Citrus Nutrition and Quality, S. Nagy and J.A. Attaway, Eds., pp.3-24. American Chemical Society. Washington, DC. Trammell, D.J., Dalsis, D.E. and Sastry, S.K. 1997. High voltage pulsed electric field treatment chambers for the preservation of liquid food products. US Patent 5,690,978. Zhang, Q., Monsalve-Gonzalez, A., Qin, B.L., Barbosa-Canovas, G.V. and Swanson, B.G. 1994. Inactivation of Saccharomyces cerevisiae in apple juice by square-wave and exponential-decay pulsed electric fields. Food Process Eng. 17:469-478. Zhang, Q.H., Qin, B.L., Barbosa-Canovas, G.V. and Swanson, B.G. 1995. Inactivation of E. coli for food pasteurization by high intensity short duration pulsed electric fields. J. Food Process Preserv. 19:103-118. 141 Mean FID response Flavor compound Fresh PEF processed D-limonene 6.7 X 10’ a 7.9 X 10’ b Valencene 2.1x10'^ a 2.8 X 10® b Myrecene 1.1 X lO S 1.3xlO®b a-Pinene 1.8 X 10^ a 2.3 X 10® b Linalool 2.8 X 10® a 3.0 X 10® a Octanal 1.7x10® a 1.9 X 10® a Decanal 3.9 X 10® a 4.5 X 10® a Ethyl butyrate 2.6x 10^‘a 2.9 X 10" a Mean values (n=S) in the same row that are not preceded by the same letter indicate significant difference (p < 0.05) for a given flavor compound. Table 5.1. Effects of PEF processing on orange juice flavor compounds 142 Product in Product CIP Tank Tank H ot WaterChilled Water Hot WaterChilled Pump PEF CIP loop ; Generator Flow Meter PEF Hold Treatment CoolingTube Heating Chambers Cooling Backpressure Glove box Tap Water Product out Figure 5.1. Flow chart of an integrated pulsed electric field processing and glove box packaging system. 143 B 60 - I i 1 40 (S ’Limonene • Myrecene • Pinene 30 ’Octanal ' Decanal ■Ethyl butyrate 20 ’Linalool 10 0 J 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Fifüre 5.2. Retention (%) of PEF processed orange juice flavor compounds during 112 day storage at 4°C in glass 144 100 § 1 Limonene Myrecene Pinene OctanalDecanal Ethyl butyrate Linalool 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 5.3. Retention (%) of PEF processed orange juice flavor compounds during 112 day storage at 22°C in glass 145 60.00 1 □ Fresh □ PEF 50.00 40.00 ? 30.00 20.00 0.00 L value b value a value Figure 5.4. Effects of PEF processing on orange juice color 146 60.00 50.00 •Lvalue at 4C ■L value at 22C 40.00 •b value at 4C •b value at 22C I> 30.00 ■a value at 4C •a vaiueat22C xT - f 20.00 10.00 I D 0.00 J r 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 5.5. The stability of PEF processed orange juice color stored at 4°and22°Cinglass 147 9 8 3 I 7 1 6 o 5 Color I 4 Overall flavor I 3 2 1 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 5.6. Color and overall flavor liking of PEF processed orange juice during 112 day storage at 4°C in glass 148 CHAPTER 6 FLAVOR, COLOR AND VITAMIN C RETENTION OF PULSED ELECTRIC FIELD PROCESSED ORANGE JUICE IN DIFFERENT PACKAGING MATERIALS ABSTRACT Effects of packaging materials, storage temperature and time on the stability of pulsed electric field processed (PEF) orange juice were investigated. Single strength orange juice was treated with PEF at electric field strength of 35 kV/cm for 59ps an integrated pilot plant scale PEF processing and glove box packaging system. The retention of eight orange juice aroma compounds, color and vitamin C in glass, polyethylene terepthalate (PET), high density polyethylene (HDPE) and low density polyethylene (LDPE) were evaluated at 4° and 22°C for 112 days. Packaging material had significant effect (p< 0.05) on the retention of orange juice aroma compounds, color and vitamin C. Hydrocarbon flavor compounds and linalool were more stable than aldehydes and ester compounds tested during storage. A trained panel distinguished the difference in flavor intensity in LDPE bottles after 56 day storage at 4®C. The PEF treated orange juice had higher a and lower b color values in HDPE and LDPE 149 comparing to those in glass and PET at 4° and 22°C (p < 0.05). The concentration of ascorbic acid in glass and PET bottles was significantly higher than those of HDPE and LDPE bottles at 4°C (p £ 0.05). PEF treated orange juice had a shelf life of more than 16 weeks in glass and PET bottles at 4®C. 150 INTRODUCTION Citrus industry has been exploring nonthermai processing methods with minimal heat treatment to increase markets by improving flavor and nutritional qualities. Pulsed electric field (PEF), an emerging processing technology, has been shown as an alternative nonthermai processing to inactivate microorganisms in foods without significant adverse effects on the flavor, taste and nutrients caused by conventional thermal processing. Because PEF processing is controlled at ambient temperature for a very short treatment time of microseconds, it provides fresh-like foods with safety and extended shelf life (Mertens and Knorr, 1992; Castro et al., 1993; Zhang et al., 1994; Zhang et al. 1995; Qin et al., 1996). However, keeping this firesh-like flavor and nutritional value of PEF processed food during storage may depend on packaging materials, methods and storage conditions. Packaging material selection as well as processing influences quality of foods during storage due to the absorption of flavor compounds by packaging materials or permeation through them, and degradation of flavor, color and nutrients by oxygen transmitted through package. Paperboard cartons and plastic containers are commonly used packaging materials of orange juice with low-density polyethylene (LDPE) as product contact layer (Pieper et al., 1992). Interaction between these materials and food has been an increasing area of research. Many researchers have reported high losses of d-limonene and other aroma compounds fiom citrus juices in contact with LDPE (Pieper et al., 1992; Sheung, 1995; Halek and Meyers, 1989; Kwapong and Hotchkiss, 1987; Sadler and Braddock, 1991). 151 Kwapong and Hotchkiss (1987) and Hotchkiss (1987) showed that the consequence of this absorption significantly affect sensory quality in model systems. Mannheim et al. (1987) stated that d-limonene absorption shortened the shelf life of orange juice. Moshonas and Shaw (1989) claimed that absorption contributed to flavor changes, which were sensorially detectable in orange juice. However, Pieper et al. (1992) reported that experienced sensory panel did not distinguish between orange juice stored in glass bottles and polyethylene laminated cartons, which absorbed 50% of d-limonene. Sizer et al. (1988) stated that the predominant factor in the change of flavor during the storage of orange juice in aseptic cartons is not absorption, but chemical degradation of flavoring components and development of off-flavor components from the degradation products. They also mentioned that control of storage temperature remains as the single most important factor in delaying flavor loss and achieving satisfactory shelf life and quality. Oxygen in the juice, in the headspace and permeating through packaging material should be minimized to avoid detrimental effects on the retention of vitamin C, color and flavor (Sizer et al., 1988). Using a nonthermai processing such as PEF with the selection of compatible material has potentials to increase the shelf life of orange juice while keeping fresh-like flavor and nutrients. Many researchers mentioned the advantage of PEF for retention of fresh flavor and nutrients (Castro et al., 1993; Zhang et al., 1994; Qin et al., 1996; Sale and Hamilton, 1967; Dunn and Peralman, 1987; Mertens and Knorr, 1992; Zhang et al., 1995; Sharma et al., 1998; Qiu et al.,1998; Jia et al., 1999). However, no literature was found in quantitative examination of the effects of packaging materials on retention of food flavor and nutrients of PEF processed foods. 152 The objective of our investigation was to determine the effects of packaging material, storage temperature and time on flavor, color and nutrient quality of PEF processed single strength orange juice. A further objective was to evaluate whether changes in flavor and color influence on the sensory quality of orange juice under commercial refngeration (4“C). 153 MATERIALS AND METHODS Materials Single strength orange juice was provided by Minute Maid (Houston, Texas) as frozen and kept in freezer at -2S°C until processing. Standard flavor compounds of d- limonene, a-pinene, myrecene, octanal, decanal, ethyl butyrate and linalool were purchased from Aldrich Chemical (Milwaukee, WI). A solid phase microextraction (SPME) fiber coated with 100 pm polymethylsiloxane, 10 ml serum bottles, Teflon coated rubber septa and aluminum caps were purchased from Supelco (Bellefonte, PA). Total plate count agar was purchased from Difco Laboratories (Detroit, MI). Glass, PET and HDPE bottles (500 ml) with 28 mm polypropylene caps were purchased from General Bottles Supply Co. (Los Angeles, CA). LDPE bottles (500 ml) were purchased from Consolidated Plastic Co. (Twinsburg, OH). Preparation and Processing of Orange Juice Freshly squeezed single strength orange juice from Valencia oranges, quickly frozen in 208 L drum was stored in a -25°C freezer until processing. The frozen juice was thawed at refrigeration temperature for twelve days prior to processing. Orange juice was processed using a sanitary fluid handling system in an integrated pilot plant scale PEF processing and glove box packaging system (Figure 6.1). The entire fluid handling system was sterilized by a sterilization-in-place (SIP) process at 105°C for 30 minutes. Single strength orange juice was treated with PEF with electric field strength of 35 kV/cm for 59|is. PEF treated orange juice was filled into sanitized bottles in a presanitized glove box, minimizing the amount of headspace to 1%. The 154 bottles were presterilized by dipping into 3% hydrogen peroxide bath and rinsing with autoclaved water. The concentration of residual hydrogen peroxide inside the bottles was determined using a Chemetrics hydrogen peroxide residue test kits (CHEMetrics, Inc., Claverton, VA). The glove box was first sprayed with 35% hydrogen peroxide and exposed to germicidal UV light (UV-C, 254 nm) (Cole Parmer, Vernon Hills, IL) with the intensity of 76 ^W/cm^ for overnight before processing. A HEPA air filter system (Fisher Scientific, Pittsburgh, PA) with 0.3 pm pore size and 1600 cm^ filtration area was installed to provide positive pressure of bacteria-fiee air in the glove box. Microbial Analysis PEF treated juice was tested for pathogens. Salmonella, Listeria monocytogenes and E. coli 0157;H7, one day afier the treatment by Silliker Laboratories (Columbus, OH). Total aerobic plate count was determined using total plate count agar (PCA) before and right afier processing, and during storage at 4° and 22°C. For each dilution prepared, duplicate samples were plated using surface plating method. PCA plates were incubated at 30“C for 48 hr before enumeration. Flavor Analysis Selected flavor compounds in the headspace of orange juice or standard compound solutions were analyzed by a combination of solid phase microextraction (SPME) and gas chromatography (GC). One ml orange juice or standard solution was transferred into a 10 ml serum bottle having a magnetic stirring bar (3x 10 mm). The sample bottle was sealed with a Teflon septum and aluminum cap. The SPME fiber coated with 100 pm polymethylsiloxane was inserted into the headspace of orange juice sample bottle, which was magnetically stirred and heated at 60"C for 20 minutes in water 155 bath to maintain the flavor compounds equilibrium between the headspace and the SPME coating. The SPME fiber was removed from the sample bottle and inserted into the 0.75 mm i.d. splitless glass liner of GC injection port, and held for 2 minutes at 220°C to desorb the flavor compounds adsorbed on the SPME coating. The desorbed flavor compounds were separated by a HP 5890 gas chromatography (Hewlett-Packard, Wilmington, DE) equipped with a HP-5 capillary column of 0.53 mm internal diameter x 30 m coated with 2.65 pm of 5% phenyl substituted methylpolysiloxane and a flame ionization detector (FID). The GC oven was programmed from 60 ®C to 120 ®C at 10 °C/minute, held for 1 minute; ramped to I40°C at 4 "C/minute, held for 1 minute; ramped to 200“C at 20®C/minute and then held for another 5 minutes at 200°C. FID was set to 250“C. Ultra high purity nitrogen was used as the carrier gas with the inlet pressure of 80 psi. The flavor compounds were identified by comparing the retention times of gas chromatographic peaks with those of standard compounds under the identical experimental conditions. Flavor compounds were evaluated based on percentage of relative retention. The GC peak area found at zero time in glass was taken 100% as initial value. D-limonene, myrecene, a-pinene, octanal, decanal, ethyl butyrate, and linalool were tested in orange juice at 4® and 22®C. Samples were taken at 0, 2, 7, 14, 28, 56, and 112 days for analysis. At each sampling date, duplicate samples (bottles) were taken 6om each storage temperature for flavor analysis by SPME-GC. 156 Color Measurement Color was measured using a HunterLab Ultrascan colorimeter (Hunter Associates Laboratory, Inc.Reston, VA). The values for L, a, and b were recorded to evaluate color changes of PEF treated orange juice during storage. The parameter L is a measure of brighmess/whiteness and it ranges from 0 to 100 (I00=white; 0=black); a is an indicator of redness and it varies from -a to +a (-a=green, a= red) and the parameter b is a measure of yellowness and it varies from -b to +b (-b=blue, +b= yellow). Duplicate samples were taken for color measurements at 0, 7, 14, 28, 56 and 112 days of storage at both storage temperatures. Sensory Analysis The sensory evaluation was based on a nine-point hedonic scale: 1= light or orange color; 9= dark or brown color, to determine color intensity. Flavor intensity was also determined on hedonic scale: 1= none; 9= strong orange flavor. Randomly coded samples were served at ~13®C. The PEF processed orange juices stored only at 4®C were evaluated by a trained panel of 12 at 2, 7, 14, 28, 56 and 112 days of storage. The panel was asked to evaluate color before they do flavor. For flavor evaluation, they were instructed to take a few sips of the juice, swirl in their mouth for few seconds before swallowing. Vitamin C Analysis Vitamin C content in the orange juice was measured using a high performance liquid chromatography (HPLC) system (Howard et ai., 1987). A Hewlett-Packard liquid chromatograph (Wilmington, DE) equipped with an autosampler and a detector at 254 nm was used. The HPLC chromatograph peak area was calculated using a Hewlett- 157 Packard integrator (HP3396 Series II). A reverse-phase C-18 column (5 pm particle size, 4.6 mm diameter, 250 mm length, Hewlett-Packard) along with a Hewlett-Packard C-18 guard column was used to separate the vitamin C using methanol and acidified water (10:90, v/v) as a mobile phase. The water was acidified with phosphoric acid (0.01%, v/v). The mobile phase was filtered using a 0.45 pm membrane filter (Micron Separations Inc., Westboro, MA) and de-gassed using helium gas before passing through the column at a flow rate of 1.0 mL/min. A standard calibration curve was obtained by using L-ascorbic acid (Sigma Chemical Co., St. Louis, MO) in the concentrations ranging firom 20 to 100 mg/100 mL. The orange juice was centrifuged at 12,535 x g for 10 minutes in a Beckman Microfuge E (Beckman Instruments Inc., Palo Alto, CA) to remove pulp and coarse cloud particles. A 10 pL of the supernatant was injected into the column using the HPLC autosampler. The reproducibility of six time analyses per each orange juice sample, based on the relative standard deviation, was found to be within 5% for vitamin C. PEF processed orange juice stored at 4*C was taken for vitamin C analyses at 0, 2, 7, 14, 28, 56 and 112 days. Data Analysis The entire experiment was duplicated. Data were analyzed using SPSS statistical package (SPSS, 1999). The effects of packaging materials, storage temperature and time on flavor and color retention were determined by Analysis of Variance (3 way ANOVA). The effects of packaging materials and storage time on vitamin C retention and sensory attributes were analyzed using two way ANOVA. Mean comparisons for specific differences were done by the Tukey test at a < 0.05. IS8 RESULTS AND DISCUSSION Microbial Stability Optimum PEF processing conditions, electric fields of 3SkV/cm and 59 ps of total treatment time, were determined by preliminary study to achieve maximum microbial inactivation for orange juice. Electric field strength and total treatment time have been reported as major factors of PEF processing to determine microbial inactivation (Jeyamkondan et al., 1999). The PEF treated orange juice using these processing conditions was reported negative for the pathogens.Salmonella, Listeria monocytogenes and E. coli 0157:H7. Fresh orange juice had total plate count of 2.34 log CFU/ml. The total aerobic plate count was less than 1 log CFU Est/ml right after the PEF treatment and during storage. The PEF treatment prevented the growth of microorganisms at 4° and 22“C for 112 days. Effects of Packaging Materials on the Retention of Flavor Compounds Relative retention (%) of flavor compounds in PEF processed orange juice stored at 4°C and 22®C for 112 days were presented by Figures 6.2-6.S. Data analysis showed that packaging material, storage temperature and time had significant (p < 0.05) effects on the retention of all flavor compounds tested. There was no significant (p ^ 0.05) difference observed between glass and PET in the retention of d-limonene at 4“ and 22°C for 112 days (Figure 6.2). However, significant absorption of d-limonene occurred within two weeks into HDPE and LDPE at both temperatures. A similar absorption behavior was observed for myrecene and a- pinene during storage due to similar polarity and solubility of the hydrocarbon type 159 aroma compounds (Figures 6.3 and 6.4). As reported by Durr et al. (1981) a distinct loss of d-limonene in orange juice in polyethylene lined cartons occurred the first two weeks of storage and then reached a steady state level, which remained constant for the remainder of the storage period. Many researchers have reported high losses of d- limonene and other aroma compounds from citrus juices in contact with LDPE (Sheung, 1995; Halek and Meyers, 1989; Kwapong and Hotckiss, 1987; Sadler and Braddock, 1991). It was reported that the consequence of this absorption significantly affected sensory quality in model systems (Kwapong and Hotchkiss, 1987; Hotchkiss, 1987). Mannheim et al. (1987) stated that absorption of d-limonene shorted the shelf life of orange juice. However, Ackerman and Wartenberg (1986) reported that absorption of orange juice components in polyethylene was not a significant factor in the retention of desirable flavor. Degradation of ethylbutyrate, neral, geranial, and other aldehydes was the most significant effect on the orange juice flavor (Ackerman and Wartenberg, 1986). Octanal, decanal (Figure 6.5 and 6.6 respectively) and ethyl butyrate (Figure 6.7) were retained less than that of the hydrocarbon flavor compounds but the pattern was similar. The loss of aldehydes and ethyl butyrate was the highest in HDPE and LDPE bottles especially at 22°C. The aldehydes and ester compounds tested were also significantly reduced in the glass bottles during within a few weeks of storage and stayed stabile for the remainder of the storage. Since glass bottles have known as inert, the loss of aldehydes and ester compounds was related to chemical degradation as it was stated by Ackerman and Wartenberg (1986). However, this loss was more pronounced in HDPE and LDPE bottles than glass and PET especially 22°iC possibly due to acceleration of flavor degradation by oxygen transmission through the polyethylene packages. Sizer et 160 al. (1988) stated that the predominant factor in the change of flavor during the storage of orange juice in an aseptic cartons is not absorption but chemical degradation of flavoring components and development of off flavor components from the degradation products. Like hydrocarbon flavor compounds tested, linalool was more stable in glass and PET than HDPE and LDPE at both temperatures (Figure 6.8). Increasing the temperature significantly reduced the retention of linalool in all packaging materials tested. Influence of cap liner on the flavor loss was investigated using PP caps with polyethylene liner and without liner. There was no significant (p < 0.05) effect of the liner on d-limonene loss during storage of PEF processed orange juice in glass bottles at 4°C (Figure 6.9). Effects of Packaging Materials on the Retention of Color Comparisons of L, a and b values in glass and plastics tested during storage time of 112 days at 4® and 22“C were shown by Figures 6.10, 6.11 and 6.12 respectively. Statistical analysis of color data showed data L, a and b values were significantly affected by packaging material, storage temperature and time (p ^ 0.05). Specific comparisons by the Tukey test showed that there was no significant (p < 0.05) difference observed in L value or brightness of orange juice packed in glass, PET, HDPE and LDPE at 4®C. However, L value significantly reduced in HDPE and LDPE bottles comparing to glass and PET after 28 day storage at 22®C due to acceleration of oxygen transmission through polyethylene packages at higher temperatures. The PEF treated orange juice had higher a and lower b values in HDPE and LDPE than that of glass and PET at 4® and 22®C storage (p ^ 0.05). Darkening or browning in orange juice 161 color was visually observed in HDPE and LDPE bottles stored at 22°C after 28 days. There was no significant (p ^ 0.05) difference between PET and glass in terms of L, a and b values during storage of 112 days at 4° and 22°C. Detrimental changes in color of orange juice are primarily caused by nonenzymatic browning (Klim and Nagy, 1988). Temperature and oxygen are the most important factors to control nonenzymatic browning. Effects of Packaging Materials on Sensory Quality Hedonic scores for color and flavor of PEF treated orange juice in the packages tested during 112 day storage at 4®C were presented by Figures 6.13 and 6.14. There was no significant (p< 0.05) difference in perceived color intensity between any of the samples even though PEF treated orange juice had higher a and lower b values in HDPE and LDPE comparing to those in glass and PET at 4°C. Sensory evaluation showed a significant difference in flavor intensity after 56 day storage between PEF treated juices packed in LDPE and in other materials. However, experienced sensory panel did not find difference in perceived overall orange juice flavor in glass, PET and HDPE bottles during storage of 112 days. Mannheim et al. (1987) reported that there was a significant difference determined by experienced tasters between juices packed in glass and cartons stored at ambient temperatures. However, another study done by Pieper et al. (1992) showed that experienced sensory panel did not distinguish between orange juice stored in glass and polyethylene laminated cartons. Effects of Packaging Materials on the Retention of Vitamin C Effects of packaging materials on the concentration of ascorbic acid in PEF treated orange juice during storage at 4"C were shown in Figure 6.15. The concentration 162 of ascorbic acid in glass bottle was significantly higher than that of PET, HDPE and LDPE bottles during storage at 4°C (p < 0.05). Ascorbic acid retention has been used as one measure of shelf life for chilled orange juices (Shaw, 1992). Berry et al. (1971) monitored the ascorbic acid level in single strength orange juice with storage and found that shelf life in plastic bottles was considerably shorter than glass bottles. Bissett and Berry (1975) reported that glass bottles showed the best retention of ascorbic acid compared to polyethylene, polystyrene and cardboard containers. Vitamin and flavor are destabilized in containers that are permeable to atmospheric oxygen (Hendrix and Redd, 1995). Marshall et al. (1986) reported that permeability of oxygen by soft-pack containers is the most critical factor in the shelf stability of aseptic processed juices. A relatively short shelf life of 28-42 days of chilled orange juice was due to the permeability to oxygen of packaging materials used (Varsel, 1980). Permeation of oxygen in polymeric packages is known to limit effectiveness for aseptic products at ambient temperature (Braddock, 1999). Polyethylene is known to be not a good barrier to gases (Robertson, 1993). It explains that significant reduction of ascorbic acid in HDPE and LDPE bottles compared to glass and PET bottles. Marshall et al. (1986) also reported greater reduction in ascorbic acid levels with higher concentration of air in the headspace. Johnson and Toledo (1975) found the presence of oxygen or especially hydrogen peroxide (the sterilent used in aseptic packaging) to be detrimental to shelf life. Hydrogen peroxide residue was less than 0.1 ppm per bottle prior to filling. The presence of oxygen in the juice and headspace gases above juice plays a role in the shelf life of chilled juice products (Shaw, 1992). The 163 amount of headspace and dissolved oxygen in the product should be kept to a minimum since ascorbic acid destruction and nonenzymatic browning of aseptically packaged orange juice are accelerated by oxygen (Graumlich et al., 1987). Assuming all bottles tested had the same amount of dissolved oxygen and headspace of 1%, the significant reduction in vitamin C retention in plastic bottles was attributed to oxygen transmission. 164 CONCLUSIONS In summary, the retention of all flavor compounds tested, vitamin C and color was significantly higher in glass and PET than in HDPE and LDPE. Increasing the storage temperature from 4° to 22°C had adverse effect on the flavor and color retention. The loss of aldehydes and ester compounds was more notable than that of hydrocarbons and alcohol in all packages including glass within the first few weeks of storage. However, this flavor loss was more pronounced in HDPE and LDPE bottles. The difference of the retention of orange juice flavor compounds, vitamin C and color in different packaging materials could be explained by: (1) absorption of flavor compounds into polymeric packaging materials tested; (2) acceleration of ascorbic acid, color and flavor degradation due to initial oxygen concentration and transmission of oxygen through the plastic packages (3) increase in browning, absorption and degradation of flavor compounds by increasing storage temperature. The PEF treated orange juice had a shelf life of more than 16 weeks in glass and PET bottles at 4°C. It is necessary to select compatible packaging material with orange juice to maintain the benefits of PEF processing during storage. 165 REFERENCES Berry, R. E., Bissett, O. W. and Veldhuis, M. K. 1971. Vitamin C retention in orange juice as related to container type. Citrus Ind. 52: 12-13. Bissett, O. W. and Berry, R. E. 1975. Ascorbic acid retention in orange juice as related to container type. J. Food Sci. 40: 178-180. Braddock, RJ. 1999. Handbook of citrus by-products and processing technology, pp 53- 83, John Wiley & Sons, Inc., NY. Castro, A.J., Barbosa-Canovas, G.V. and Swanson, B.C. 1993. Microbial inactivation of foods by pulsed electric field. J. Food Process. Preser. 17:47-73. Dunn, J.E. and Peralman, J.S. 1987. Methods and apparatus for extending the shelf life of fluid food products. US Patent 4,695,472. Durr, P., Schobinger, U. and Waldrogel, R. 1981. Aroma quality of orange juice after filling and storage in soft packages and glass bottles. Lebens-Verpackung. 20:91. Graumlich, T.R., Marcy, J.E. and Adams, J.P. 1986. Aseptically packaged orange juice and concentrate; A review of the influence of processing and packaging conditions on quality. J. Agric. Food Chem. 34:402-405. Halek, G.W. and Meyers, M.A. 1989. Comparative sorption of citrus flavor compounds by low density polyethylene. Pack. Tech. Sci. 2: 141. Hendrix, C. M.; and Redd, J. B. 1995. Chemistry and technology of citrus juices and by products. Ch. 2 In Production and Packaging ofNon-Carbonated Fruit Juices and Fruit Beverages; Ashurst, P. R., Eds. pp. 53-87. Chapman & Hall. London, UK. Hotckiss, J.H. 1987. Comparative sorption of aromatic flavors by plastics used in food packaging. Activities Report of the R&D Associates 39:65. Howard, R.C., Peterson, T., Kastl, P. R. 1987. High performance liquid chromatographic determination of ascorbic acid in human tears. J. Chromatog. 414:434-439. Jeyomkondan, S., Jayas, D.S. and Hailey, R.A. 1999. Pulsed electric field processing of foods. J. Food Prot. 62(9): 1088-1096. Jia, M., Zhang, Q.H. and Min, D.B. 1999. Pulsed electric field processing effects on flavor compounds and microorganisms of orange juice. Food Chem. 65:445-451. 166 Johnson, R. L. and Toledo, R. T. 1975. Storage stability of 55“brix orange juice concentrate aseptically packed in plastic and glass containers. J. Food Sci. 40:433-434. Klim, M and Nagy, S. 1988. An improved method to determine nonenzymatic browning in citrus juices. J. Agric. Food Chem. 36:1271-1274. Kwapong, O.Y. and Hotchkiss, J.H. 1987. Comparative sorption of aroma compounds by polyethylene and ionomer food contact plastics. J. Food. Sci. 52(3):761. Mannheim, C H., Miltz, J., Letzter, A. 1987. Interaction between polyethylene laminated cartons and aseptically packed citrus juices. J. Food Sci. 52(3):737. Marshall, M.; Nagy, S.; Rouseff, R. L. 1986. Factors impacting on the quality of stored citrus fruit beverages. In The Shelf Life o f Foods and Beverages, Charalambous, G., Ed. pp. 237-254. Elsevier, NY. Mertens, B. and Knorr, D. 1992. Developments of nonthermal processes for food preservation. Food Tech. 46: 124-133. Moshonas, M.G. and Shaw, P.E. 1989. Flavor evaluation and volatile flavor constituents of stored aseptically packaged orange juice. J. Food Sci. 54(1 ):82. Pieper, G., Borgudd, L., Ackermann, P. and Fellers, P. 1992. Absorption of aroma volatiles of orange juice into laminated carton did not affect sensory quality. J. Food Sci. 57: 1408-1411. Qin, B.L., Pothakamury, U.R., Barbosa-Canovas, G.V. and Swanson, B.G. 1996. Nonthermal pasteurization of liquid foods using high-intensity pulsed electric fields. Critical Reviews in Food Science and Nutrition. 36(6): 603-627. Qiu, X , Sharma, S., Tuhela, L., Jia, M. and Zhang, Q.H. 1998. An integrated PEF pilot plant for continuous nonthermal pasteurization of fresh orange juice. Trans. ASAE. 41(4): 1069-1074. Robertson, G. L. 1993. Structure and related properties of plastic polymers. Ch. 2 In Food Packaging, pp. 9-62. Marcel Decker, Inc., NY. Sadler, GT). and Braddock, RJ. 1991. Absorption of citrus flavor volatiles by low density polyethylene. J. Food Sci. 56(1): 35. Sale, A. J.H. and Hamilton, WA. 1967. Effects of high electric fields on microorganisms. Killing of bacteria and yeasts. Biochem. Biophys. Acta. 148:781-788. 167 Sharma, S.K., Zhang, Q.H. and Chism, G.W. 1998. Development of a protein fortified fruit beverage and its quality when processed with pulsed electric field treatment. J. Food Qual. 21:459-473. Shaw, P. E. 1992. Shelf life and aging of citrus juices drinks and related soft drinks. In Quality Control Manual for Citrus Processing Redd, Plants', J. B., Shaw, P. E., Hendrix Jr., C. M., Hendrix, D. L., Eds. pp. 173-199. Agscience, Inc., Aubumdale, FL. Sheung, K. 1995. Sorption of orange juice flavor compounds into polymeric packaging materials. Thesis. The Ohio State University, Columbus. Sizer, C.E., Waugh, P.L., Edstam S., Ackerman, P. 1988. Maintaining flavor and nutrient quality of aseptic orange juice. Food Tech. 42:152-159. SPSS Inc. 1999. SPSS base 10.0 user’s guide. SPSS Inc., Chicago, IL. Varsel, C. 1980. Citrus juice processing as related to quality and nutrition. In Citrus Nutrition and Quality. Nagy, S., Attaway, J. A., Eds, pp. 225-271. American Chemical Society, Washington, DC. Yeom, H. W. 2000. The effects of pulsed electric field on the quality of foods. Ph.D. dissertation. The Ohio State University, Columbus. Zhang, Q., Monsalve-Gonzalez, A., Qin, B.L., Barbosa-Canovas, G.V. and Swanson, B.G. 1994. Inactivation of Saccharomyces cerevisiae in apple juice by square-wave and exponential-decay pulsed electric fields. J. Food Process Eng. 17:469-478. Zhang, Q.H., Qin, B.L., Barbosa-Canovas, G.V. and Swanson, B.G. 1995. Inactivation of E. coli for food pasteurization by high intensity short duration pulsed electric fields. J. Food Process Preserv. 19:103-118. 168 Product in Product CIP Tank Tank Hot WaterChilled Water Hot WaterChilled Pump PEF Generator CIP loop Flow Meter PEF Hold Treatment Cooling Tube Heating Chambers Cooling Backpressure Glove box Tap Water Product out Figure 6.1. Flow chart of an integrated pulsed electric field processing and glove box packaging system. 169 100 90 80 e 70 1 60 î 50 40 30 Glass at 4C PETat4C HDPE at 4C LDPE at 4C 20 Glass at 22C PETat22C HDPE at 22C LDPE at 22C 10 0 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.2. D-limonene retention (%) in PEF processed orange juice during storage at 4° and 22°C in different packages 170 100 90 80 i 60 M 50 40 30 Glass at 4C PETat4C HDPE at 4C LDPE at 4C 20 Glass at 22C LDPEat22C 10 0 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.3. Myrecene retention (%) in PEF processed orange juice during storage at 4° and 22°C in different packages 171 § 1 î CL U 40 I 30 ‘Glass at4C •PETat4C •HDPE at 4C •LDPE at 4C 20 ■Glass at22C •PETat22C •HDPEat22C •LDPEat22C 10 I "I I II 1 — I I I I I 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.4. Pinene retention (%) in PEF processed orange juice during storage at 4° and 22°C in different packages 172 100 ♦ Glass at 4C PETat4C HDPE at 4C -M-LDPEat4C Glass at 22C PET at 22C -+-HDPEat22C LDPE at 22C 60 • 73 50 - o 40 S, 30 ■ 20 ■ 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.5. Octanal retention (%) in PEF processed orange juice during storage at 4° and 22"C in different packages 173 100 Glass at 4C FETat4C HDPE at 4C -M -LD PE Glass at 22C PETat22C HDPE at 22C LDPE at 22C I ■a - I 40 I a 20 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.6. Decanal retention (%) in PEF processed orange juice during storage at 4° and 22°C in different packages 174 ^ 70 I I !■ 1 20 ■Glass at 4C ■PETat4C •HDPE at 4C ■LDPE at 4C ■Glass at 22C •PETat22C ■HDPE at 22C -e -L D P E at 22C 10 - I I I I 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.7. Ethyl butyrate retention (%) in PEF processed orange juice during storage at 4" and 22°C in different packages 175 6I I i Glass at 4C ■PETat4C HDPE at 4C ■LDPE at 4C •Glass at 22C ■PETat22C HDPE at 22C LDPE at 22C 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.8. Linalool retention (%) in PEF processed orange juice during storage at 4° and 22°C in different packages 176 100 90 ■ § 70 i 60 cap w/Iiner 40 - cap w/o liner 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.9. Efifects of cap liner (PE) on the sorption of d-Limonene during storage at 4®C 177 60.00 50.00 40.00 3 1 30.00 20.00 - 'Glass at 4C # PET at 4C HDPE at 4C —#^LD PE at 4C 10.00 ■Glass at 22C -» -P E T a t2 2 C —I— HDPE at 22C LDPE at 22C 0.00 I I 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.10. Comparison of orange juice color-L value- in glass and plastic bottles during storage at 4° and 22°C 178 6 .0 0 5.00 4.00 ■a 3.00 - 2.00 - Glass at 4C PETat4C HDPE at 4C LDPE at 4C 1.00 ■ Glass at 22C PET at 22C HDPE at 22C LDPEat22C 0.00 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.11. Comparison of orange juice color-a value- in glass and plastic bottles during storage at 4" and 22°C 179 25.00 1 20.00 - 15.00 I 10.00 ■ Glass at 4C PETat4C HDPE at 4C 5.00 - at 4CGlass at 22CLDPE PETat22C HDPEat22C LDPEat22C 0.00 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.12. Comparison of orange juice color-b value- in glass and plastic bottles during storage at 4° and 22°C 180 9 8 f 7 6 i 5 I 4 3 Glass PET HDPE -M-LDPE 2 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.13. Effects of packaging on mean hedonic responses for color intensity of PEF treated orange juice during 112 day storage at 4“C 181 9 8 I 7 I 6 a 5 I 4 i 3 Glass PET HOPE -M-LDPE 2 1 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.14. Efifects of packaging on mean hedonic responses for flavor intensity ofPEF treated orange juice during 112 day storage at 4“C 182 60 55 50 I 45 ! 40 Iu 35 € 30 Glass o 25 PET § 20 HOPE I 15 10 5 0 0 10 20 30 40 50 60 70 80 90 100 110 120 Storage time (days) Figure 6.15. Efifects of packaging materials on the concentration of ascorbic acid in PEP treated orange juice during storage at 4°C (From Yeom, 2000) 183 CHAPTER? SUMMARY AND RECOMMENDATIONS The primary role of food packaging is to protect the product from the time and point of manufacture to the time and point of consumption. There are three important elements in packaging that can affect package performance. The first element is forming. Forming techniques and processing parameters and the machines used have a key influence on package performance such as barrier properties and mechanical strength. Thus, forming process needs to be optimized to provide the desired end use properties. Thermoforming process was optimized to improve the quality and uniformity of semirigid containers produced by the Benco aseptic packaging machine. Wall thickness distribution was chosen as quality index to optimize processing parameters of thermoforming temperature, air-forming pressure and heating time for the multilayered base material. It is important to determine optimum thermoforming parameters especially for multilayered material so that barrier layer can remain at satisfactory level after natural overall thickness reduction due to thermoforming. Better wall thickness was obtained at the forming temperature of 146-IS6°C, the forming pressure of 2-4 bars and the heating time of 74-97 seconds. Forming temperature was found to be the principle parameter influencing wall thickness distribution in this plug assisted thermoforming process. 184 In the thermoforming study, the temperature distribution was found empirically via the setting of the individual infiared heaters and using external infrared thermometers to measure the film temperature during forming at only thermoforming station. Incorporation of row sensors could be recommended between the heating station and thermoforming station to provide temperature distribution over the surface of the sheet. This could provide more control over the forming process. Plug velocity and temperature have also been reported important parameters of uniform wall thickness distribution. Plastic plugs of the forming unit of the Benco could be replaced by metal plugs. Heating the metal plugs will decrease temperature difference between the plug and the plastic film, providing more uniform wall thickness distribution. Sealing is the second element in packaging since integrity of a package in long term is provided by good seal. Package seal has to be uniform, strong enough and free of any type of defects that compromise package integrity. There is a need for a precise nondestructive technique to determine seal integrity. Ultrasound has been shown to be the promising technique for inspection of seal integrity among alternative nondestructive techniques. Seal integrity of food packages such as semirigid cups produced by the Benco aseptic packaging machines were tested using non-contact ultrasonic system to determine the feasibility of ultrasound in seal integrity testing. Alternative packages such as polymeric trays were also tested. The results of this laboratory work confirmed that ultrasonic is capable of detecting the defects in the seal and is a non-destructive tool to characterize the heat seal layer. This study provides a basis for interpretation of ultrasonic signals and seal defects. However, the technique used is not suited to 185 packaging production lines because it requires immersing the package in a fluid bath to achieve ultrasonic coupling. Optical microscopic pictures supported that detection by ultrasound was based on reduced signal strengths or amplitudes at altered seal. Integrity of meals-ready-to-eat (MRE) pouches was inspected using a new pressure (vacuum) differential technique. This technique is based on a three-step approach. To evaluate the system air was linked into a test chamber through a calibrated needle valve simulating a leak at different leak rates. A simulated leak generated values higher than reference, and dependant upon the leak rate, was accepted or rejected by the leak tester. The technique proved to be effective, although, short-sealed non-vacuum packed poundcake pouches failed during subsequent leak testing. The vacuum method can be suggested for packages having a certain amount of headspace or residual gas so that gas can flow out of the package since the pressure inside the package (at about 1 atmosphere) is higher than in the vacuum chamber. Even if these two elements meet with expected criteria, package has to be compatible with food, which is the third important issue in assuring package performance. Possible interactions between food, package and environment have to be searched before food and package are brought together to assure food-package compatibility. T he recent trend on the part of the food industry to replace traditional glass and metal containers with plastic ones has focused attention on the interactions between aromas or flavors and the polymers used in these new structures since the food flavor profile is in extremely small quantity, even a small loss leads to imbalance of flavor profile, shortening the shelf-life of the food. 186 Effects o f packaging materials as well as pulsed electric field (PEF) processing and storage conditions on orange juice flavor, color and vitamin C were investigated using a pilot plant scale integrated PEF processing and glove box packaging system. Single strength orange juice was treated with PEF at electric field strength of 35 kV/cm for 59ps and filled into sanitized bottles of glass, PET, HDPE and LDPE in a sanitized glove box. Among plastic materials tested, PET was comparable to glass in terms of retention of flavor, color and vitamin C of PEF treated orange juice. PEF processed orange juice in glass and PET bottles had a shelf life of more than 16 weeks at 4°C storage. It is recommended to select compatible packaging material to maintain benefits of PEF processing during storage. For future studies, it could be recommended to use PET as a base material instead of our existing laminated material (LDPE/PVDC/HIPS) to be formed by the Benco aseptic packaging machine since PET is thermoformable and heat sealable. 187 LIST OF REFERENCES Ackerman, P.W. and Wartenberg, E.W. 1986. Shelf-life of citrus juices: A comparison between different packages. Report of 19*** Symposium of International Federation of Fruit Juice Producers. Den Haah, pp. 143. Ahmed, E.M., Dennison, R.A. and Shaw, P.E. 1978. Effect of selected oil and essence volatile components on flavor quality of pumpout orange juice. J. Agric. Food Chem. 26:368-372. Ahvenian, R., Wirtanen, G. and Manninen, M. 1989a. Ultrasound Imaging-A non destructive method for monitoring the microbiological quality of aseptically-packed milk products. Lebensm. Wiss. U. Technol. 22:382-386. Ahvenian, R., Wirtanen, G. and Manninen, M. 1989b. Ultrasound Imaging-A non destructive method for monitoring the microbiological quality of aseptically packed foodstuffs. Lebensm. Wiss. U. Technol. 22:273-278. Arndt, G.W. 1992. Examination of flexible and semirigid food containers for integrity. In FDA Bacteriological Analytical Manual. 7*** ed. AGAC, Arlington, VA. Arnold, S.I. 1984. Thermoforming. In Packaging Alternatives for Food Processors. pp.61-62. NFPA. Washington, D C. Aroujalian, A.,. Ngadi, M.O and Emond, J-P. 1997. Wall thickness distribution in plug- assist vacuum formed strawberry containers. Poly. Eng. Sci. 37:178-182. Arthur, C.L. and PawUszyn, J. 1990. Solid phase microexraction with thermal desorption using fused silica optical fibers. Anal. Chem. 62:2145-2148. Arthur, C.L., Killam, L.M., Bucholz, K.D. and PawUszyn, J. 1992. Automation and optimization of solid-phase microextraction. Anal. Chem. 35:1962-1966. Ayhan, Z. and Zhang, Q.H. 2000. Wall thickness distribution in thermoformed food containers produced by a Benco aseptic packaging machine. Poly. Eng. Sci. 40(1): 1-10. Baner, A.L., Kalyankar, V. and Shoun, L.H. 1991. Aroma sorption evaluation of aseptic packaging. J. Food Sci. 56(4): 1051-1054. 188 Bayliss, C E. and Waites, W.M. 1979. The combined efifect of hydrogen peroxide and ultraviolet radiation on bacterial spores. J. Applied. Bacteriology. 47:263-269. Bayliss, C.E. and Waites, W.M. 1980. The effect of hydrogen peroxide and ultraviolet radiation on non-sporing bacteria. J. Applied. Bacteriology. 48; 417-422. Bayliss, G.E. and Waites, W.M. 1982. Effects of simultaneous high intensity ultraviolet irradiation and hydrogen peroxide on bacterial spores. Food. Tech. 17:467-470. Berry, R. E., Bissett, O. W., Veldhuis, M. K. 1971. Vitamin C retention in orange juice as related to container type. Citrus Ind. 52:12-13. Bissett, O. W., Berry, R. E. 1975. Ascorbic acid retention in orange juice as related to container type. J. Food Sci. 40: 178-180. Blakistone, B. and Harper, C. 1995. New developments in seal integrity testing. In Plastic Package Integrity Testing Assuring Seal Quality, B.A Blakistone, and C.L. Harper, Eds. pp. 1-9. Institute of Packaging Professionals and Food Processors. Washington, D.C. Bourque, R.A. 1995. Importance of package integrity. In Plastic Package Integrity Testing Assuring Seal Quality, B.A Blakistone, and C.L. Harper, Eds. pp. 11-14. Institute of Packaging Professionals and Food Processors. Washington, D.C. Braddock, R.J. 1999. Handbook of citrus by-products and processing technology, pp 53- 83, John Wiley & Sons, Inc., NY. Briston, J.H. 1974. Plastics in contact with food. Food Trade Press Ltd., London. Castro, A.J., Barbosa-Canovas, G.V. and Swanson, B.C. 1993. Microbial inactivation of foods by pulsed electric fields. J. Food Process. Preser. 17:47-73. Chen, C.S., Shaw, P.E. and Parish, M.E. 1993. Fruit juice processing technology: orange and tangerine juices, pp. 110-156. Agscience, Inc., Florida. Chin, H.W., Bernhard, R.A. and Rosenber, M. 1996. Solid phase microextraction for cheese volatile compounds analysis. J. Food Sci. 61(6):1118-1122,1128. Chung, W.R. and. Lee, T.W. 1992. Thermoforming simulation of a plastic cup. SPE ANTEC Tech. Papers. 38:117-121. Dunn, J.E. and Peralman, J.S. 1987. Methods and apparatus for extending the shelf life of fluid food products. US Patent 4,695,472. 189 Durr, P., Schobinger, U. and Waldrogel, R. 1981. Aroma quality of orange juice after filling and storage in soft packages and glass bottles. Lebens-Verpackung. 20:91. Ekasari, I., Jongen, M.F. and Pilnik, W. 1986. Use of a bacterial mutagenicity assay as a rapid method for the detection of early stage of maillard reactions in orange juices. Food Chem. 21:125-131. Florian, J. 1987. Practical Thermoforming and Applications. Marcel Dekker Inc., NY. Floros, J.D. and Gnanasekharan, V. 1992. Principles, technology and applications of destructive and nondestructive package integrity testing. In Advances in Aseptic Processing Technologies, R.K Singh and P.E Nelson, Eds. pp. 157-188. Elsevier Applied Science, NY. Frados, J. 1976. Thermoforming plastic film and sheet. In Plastic Engineering Handbook, 4*** ed, pp. 273-325.Van Nonstrand Reinhold, NY. Gagliardi, G.R., Sullivam, D. and Smith, N.F. 1984. Computer aided video inspection. Food Technol. 38(4): 53-59. Garrett, K.M. 1988. Laboratory testing procedures: heat sealed package. In Plastic Package Integrity: Successjully Meeting the Challenge. NFPA. Food Processors Institute. Washington, D.C. Gavin, A. and Weddig, L.M. 1995. Closures for semirigid and flexible containers. In Canned Foods: Principles of Thermal Process Control, Acidification and Container Closure Evaluation. 6* ed. pp. 151-166. Food Processors Institute. Washington, D.C. Giacin, J R. 1995. Factors affecting permeation, sorption, and migration processes in package product systems. In Foods and Packaging Materials-Chemical Interactions. P. Ackerman, M. Jagerstad and T. Ohlsson, Eds. The Royal Society of Chemistry. Thomas Graham House, Science Park, Cambridge. Gibson, T.G. 1995. A system for inspection of package seal integrity. SPIE. 2597:37-39. Gnanasekharan, V. and Floros, J.D. 1994. Package integrity evaluation: criteria for selecting a method. Pack. Tech. Eng. 3(6): 44-46,48. Gould, R.W. 1972. Non-destructive egg shell thickness measurements using ultrasonic energy. Poultry Science. 51:1460-1461. Graumlich, T.R., Marcy, J.E. and Adams, J.P. 1986. Aseptically packaged orange juice and concentrate: a review of the influence of processing and packaging conditions on quality. J. Agric. Food Chem. 34:402-405. 190 Gruenwald, G. 1987. Thermofonning-A Plastic Processing Guide, Technomic Inc., Lancaster, Pennsylavania. Guazzo, D.M. 1994. Package integrity testing. In Parenteral Quality Control. M.J. Akers Ed. (2"** ed), pp. 247-298. Marcel Dekker, Inc., NY. Halek, G.W. and Meyers, M.A. 1989. Comparative sorption of citrus flavor compounds by low density polyethylene. Pack. Tech. Sci. 2:141. Harper, C.L., Blakistone, B.A., Litchfield, J.B. and Morris, S.A. 1995. Developments in food packaging integrity testing. Trend. Food Sci. Tech. 6(10): 336-340. Hendrix, C. M., Redd, J. B. 1995. Chemistry and technology of citrus juices and by products. Ch. 2 In Production and Packaging o f Non-Carbonated Fruit Juices and Fruit Beverages, Ashurst, P. R., Ed. pp 53-87. Chapman & Hall: London, UK. Hirose, K., Harte, B.R., Giacin, J.R, Miltz, J. and Stine, C. 1988. Sorption of d-limonene by sealant films and effect on mechanical properties. In Food and Packaging Interactions. H.J. Hotchkiss, Ed. ACS Symp. Ser. 365. ACS. Washington, D.C. Hotchkiss, J.H. 1987. Comparative sorption of aromatic flavors by plastics used in food packaging. Activities Report of the R&D Associates. 39:65. Hotchkiss, J. H. 1995. Overview on chemical interactions between food and packaging materials. In Foods and Packaging Materials-Chemical Interactions. P. Ackerman, M. Jagerstad and T. Ohlsson, Eds. The Royal Society of Chemistry. Thomas Graham House, Science Park, Cambridge. Howard, R.C., Peterson, T., Kastl, P. R. 1987.High performance liquid chromatographic determination of ascorbic acid in human tears. J. Chromatog. 414:434-439. Hriciga, A.L. and Stadelman, D.J. 1988. Flavor-polymer interactions: coffee aroma alteration. In Food and Packaging Interactions. H.J. Hotchkiss, Ed. ACS Symp. Ser. 365. ACS. Washington, D.C. Imai, T., Harte, B.R. and Giacin, J R. 1990. Partition distribution of aroma volatiles fi'om orange juice into selected polymeric sealant films. J. Food Sci. 55(1):158-161. Jarman, D., Farahbakhsh, B. and Herzig, R. 1994. Ultrasonic inspection of seal integrity of bond lines in sealed containers. U.S. Patent.5.372.042. Jeyomkondan, S., Jayas, D.S. and Hailey, R.A. 1999. Pulsed electric field processing of foods. J. Food Prot. 62(9): 1088-1096. 191 Jia, M.Y., Zhang, H.Q. and Min, D.B. 1998. Optimization of solid-phase microextraction analysis for headspace flavor compounds of orange juice. J. Agric. Food. Chem. 46:2744- 2747. Jia, M., Zhang, Q.H. and Min, D.B. 1999. Pulsed electric field processing effects on flavor compounds and microorganisms of orange juice. Food Chem. 65:445-451. Johnson, R. L., Toledo, R. T. 1975. Storage stability of 55° brix orange juice concentrate aseptically packed in plastic and glass containers. J. Food Sci. 40: 433-434. King, P. 1995. The federal government's role: regulating plastic container integrity. In Plastic Package Integrity Testing Assuring Seal Quality, B.A Blakistone, and C.L. Harper, Eds. pp. 15-21. Institute o f Packaging Professionals and Food Processors. Washington, D.C. Kinsler, L.E., Frey, A.R., Coppens, A.B. and Sanders, J.V. 1982. Fundamentals of Acoustics, John Wiley & Sons, NY. Klim, M. and Nagy, S. 1988. An improved method to determine nonenzymatic browning in citrus juices. J. Agric. Food Chem. 36: 1271-1274. Kwapong, O.Y. and Hotchkiss, J.H. 1987. Comparative sorption of aroma compounds by polyethylene and ionomer food-contact plastics. J. Food Sci. 52(3): 761-763, 785. Lai, M.O. and Holt, D.L. 1975. Thickness variation in the thermoforming of poly(methyl methacrylate) and high-impact polystyrene sheets. J. Appl. Polym. Sci. 19: 1805-1814. Lampi, R.A., Schulz, G.L., Ciavarini, T. and Burke, P.T. 1976. Performance and integrity of retort pouch seals. Food Tech. 30(2): 38-48. Lampi, R.A. 1977. Flexible packaging for thermoprocessed foods. In Advances in Food Research. C O. Chichester, E. M. Mrak and G.F. Stewart, Eds. Academic Press, Inc., NY. Landois-Garza, J. and Hotchkiss, J.H. 1987. Plastic packaging can cause aroma sorption. Food Eng. 4:39-42. Maarse, H. and Visscher, C.A. 1989. Volatile compounds in food: qualitative and quantitative data. Marcel Dekker, Inc., NY. Mannheim, C H., Miltz, J. and Letzter, A. 1987. Interaction between polyetylene laminated cartons and aseptically packed citrus juices. J. Food Sci. 52(3): 737-740. Marcy, J.E., Hansen, A.P. and Graumlich, T.R. 1989. Effects of storage temperature on the stability of aseptically packaged concentrated orange juice and concentrated orange drink. J. Food Sci. 54: 227-230. 192 Marcy, J E. 1995. Integrity testing and biotest procedures for beat-sealed containers. In Plastic Package Integrity Testing Assuring Seal Quality, B.A Blakistone and C.L. Harper, Eds. pp. 35-47. Institute of Packaging Professionals and Food Processors. Washington, D.C. Marshall, M., Nagy, S., and Rouseff, R. L. 1986. Factors impacting on the quality of stored citrus fruit beverages. In The Shelf Life o f Foods and Beverages, Charalambous, G., Ed. pp. 237-254. Elsevier: NY. Mathlouthi, M. 1994. Food Packaging and Preservation. Blackei Academic & Professional. Glaslow, UK. Maunder, D.T. 1977. Possible use of ultraviolet sterilization of containers for aseptic packaging. Food Tech. 31(4): 36-37. Mertens, B. and Knorr, D. 1992. Developments of nonthermal processes for food preservation. Food Tech. 46: 124-133. Miles, C.A., Fursey, G.A.J. and Jones, R.C.D. 1985. Ultrasonic estimation of solid/liquid ratios in fats, oils and adipose tissue. J. Sci. Food. Agric. 36:215-228. Miltz, J. and Mannheim, C.H. 1986. The effects of polyethylene contact surface on the shelf life of food products. In Food Product Package Compatibility, Gary J.I., Harte, B.R. and Miltz, J., Eds. Technomic Publish. Inc., Lancaster. Morris, S.A., Ozguler, A. and O’Brien, W.D. 1998. New sensors help improve heat-seal microleak detection. Pack. Tech. Eng. 7(7): 44-49. Moshonas, M.G. and Shaw, P.E. 1989. Flavor evaluation and volatile flavor constituents of stored aseptically packaged orange juice. J. Food Sci. 54(1 ):82. Moshonas, M.G. and P.E, Shaw. 1994. Quantitative determination of 46 volatile constituents in fresh, unpasteurized orange juice using dynamic headspace gas chromatography. J. Agric. Food Chem. 42:1525-1528. Nawar, W.W. and I S. Fagerson. 1962. Direct gas chromatographic analysis as an objective method of flavor measurement. Food Tech. 11:107-109. Nielsen, T.J., Jagerstad, I.M., Oste, R.E. and Wesslen, B.C. 1992. Comparative absorption of low molecular aroma compounds into commonly used food packaging polymer films. J. Food. Sci.57(2):490-492. Nijssen, L.M. 1991. Off flavors. In Volatile Compounds in Foods and Beverages (Chapter 19). H. Maarse, Ed. Marcel Dekker, Inc., NY. 193 Nisperos-Carriedo, M.O. and Shaw, P.E. 1990. Comparison of volatile flavor components in fresh and processed orange juices. J. Agric. Food Chem. 38:1048-1052. Ozguler, A., Morris, S.A. and O’Brien, W.D. 1998. Ultrasonic imaging of micro-leaks and seal contamination in flexible food packages by the pulse-echo technique. J. Food Sci. 63(4): 673-678. Paik, J.S. 1991. Flavor scalping phenomenon extensively examined in recent study. Food Eng. April: 64-66. Penton, Z. 1996. Flavor volatiles in a fruit beverage with automated SPME. Food Test. Anal. June: 16-18. Pieper, G., Borgudd, L., Ackerman, P. and Fellers, P. 1992. Absorption of aroma volatiles of orange Juice into laminated carton packages did not affect sensory quality. J. Food Sci. 57(6): 1408-1411. Poller, S. and Michaeli, W. 1992. Film temperatures determine the wall thickness of thermoformed parts. SPE ANTEC Tech. Papers. 38: 104-108. Purohit, K.S. 1995. Biotesting of plastic containers. In Plastic Package Integrity Testing Assuring Seal Quality, B.A Blakistone and C.L. Harper, Eds. pp. 67-77. Institute of Packaging Professionals and Food Processors. Washington, D.C. Qin, B.L., Pothakamury, U.R., Barbosa-Canovas, G.V. and Swanson, B.G. 1996. Nonthermal pasteurization of liquid foods using high-intensity pulsed electric fields. Critical Reviews in Food Science and Nutrition. 36(6): 603-627. Qiu, X , Sharma, S., Tuhela, L., Jia, M. and Zhang, Q.H. 1998. An integrated PEF pilot plant for continuous nonthermal pasteurization of fresh orange juice. Trans. ASAE. 41(4): 1069-1074. Risch, S.J. 1991. Introduction. In Food and Packaging Interactions. II. S.J. Risch and J.H. Hotchkiss, Eds. ACS Symp. Ser. 473. Washington, D.C. Robertson, G.L. 1993. Food Packaging Principles and Practice. Marcel Dekker, Inc., NY. Rodriguez, J.G. 1995. Noncontacting acoustic ultrasonic signature analysis development. In Plastic Package Integrity Testing Assuring Seal Quality, B A Blakistone and C.L. Harper, Eds. pp. 107-111. Institute of Packaging Professionals and Food Processors. Washington, D.C. Rosen, S.L. 1971. Fundamental Principles of Polymeric Materials, A Wiley Interscience, NY. 194 Rosen, S.R. 1992. Practical application of thinwall thermoforming molds from prototype to production on web fed inline thermoforming machines. SPE ANTEC Tech. Papers. 38: 518-521. Rosenzweig, N., Narkis, M. and Tadmor, Z. 1979. Wall thickness distribution in thermoforming. Polym. Eng. Sci. 19:946-951. Ruhlman, FC.T. 1999. Product examination and reformulation for pulsed electric field processing. Thesis. The Ohio State University, Columbus. SAS Institute Inc. 1989. SAS/STAT User’s Guide, Version 6, Volume 2, Cary, NC. Sadler, G.D. and Braddock, R.J. 1991. Absorption of citrus flavor volatiles by low density polyethylene. J. Food Sci. 56(1): 35. Sadler, G.D., Parish, M.E. and Wicker, L. 1992. Microbial, enzymatic and chemical changes during storage of fresh and processed orange juice. J. Food Sci. 57: 1187-1191. Sale, A.J.H. and Hamilton, W.A. 1967. Effects of high electric fields on microorganisms. Killing of bacteria and yeasts. Biochem. Biophys. Acta. 148: 781-788. Sharma, S.K., Zhang, Q.H. and Chism, G.W. 1998. Development of a protein fortified finit beverage and its quality when processed with pulsed electric field treatment. J. Food Qual. 21:459-473. Shaw, P.E. 1991. Volatile compounds in foods and beverages: Fruits II. pp. 305-327. Marcel Dekker, Inc., NY. Shaw, P. E. 1992. Shelf life and aging of citrus juices, juice drinks and related soft drinks. In Quality Control Manual for Citrus Processing Redd, Plants, J. B., Shaw, P. E., Hendrix Jr., C. M., Hendrix, D. L., Eds. pp. 173-199. Agscience, Inc., Aubumdale, FL. Shaw, E.E., Litchfield, J.B., Reid, J.F. and Harper, C.L. 1995. Detection of defects in plastic laminated foil pouches: machine vision sensing. In Plastic Package Integrity Testing Assuring Seal Quality. Blakistone, B.A. and Harper, C.L., Eds. Institute of Packaging Professionals and Food Processors. Washington, D.C. Sheung, S.K. 1995. Sorption of orange juice flavor compounds into polymeric packaging materials. Thesis. The Ohio-State University, Columbus. Shih, WJC. 1991. Thermoforming of amorphous PET. SPE ANTEC Tech. Papers. 37: 584-587. Sizer, C.E. 1983. Methods for Container Testing. Capitalizing on Aseptic. NFPA. The Food Processors Institute, Washington, D.C. 195 Sizer, C E., Waugh, P.L., Edstam, S. and Ackermann, P. 1988. Maintaining flavor and nutrient quality of aseptic orange juice. Food Tech. June: 152-159. Stannard, C.J., Abbiss, J.S. and Wood, J.M. 1983. Combined treatment with hydrogen peroxide and ultra-violet irradiation to reduce microbial contamination levels in pre formed food packaging cartons. J. Food Protec. 46(12): 1060-1064. Stannard, C.J. and Wood, J.M. 1983. Measurement of residual hydrogen peroxide in preformed food cartons decontaminated with hydrogen peroxide and ultraviolet irradiation. J. Food Protect. 46(12):1074-1077. Stauffer, T. 1988. Non-destructive in line detection of leaks in food and beverage packages-an analysis of methods. J. Pack. Tech. 2(4): 147-149. Stauffer, T. 1992. On-line leak and seal integrity testing of plastic containers. Pack. Tech. Eng. 1(1): 30-37. Streaker, C.B. 1999. Design, construction, and validation of a pilot plant fluid handling system for pulsed electric field processing and aseptic packaging of foods. Thesis. The Ohio State University, Columbus. Szilard, J. 1982. Ultrasonic Testing-Nonconventional Testing Techniques. John Wiley & Sons, NY. Tatum, J.H., Steven, N., and Roberte, B. 1975. Degradation products formed in canned single-strength orange juice during storage. J. Food Sci. 40: 707-709. Throne, J.L I99la. Guidelines for thermoforming part wall thickness. Polym. Plast. Tech. Eng. 30 (7): 685-700. Throne, J.L. 1991b. Plug-assist thermoforming-a new design protocol for rectangular parts. Polym. Plast. Tech. Eng. 30 (7): 701-721. Throne, J.L. 1996. Technology of Thermoforming, Hanser/Gardner Publications, Inc., Cincinnati, OH. Ting, S.V. 1980. Nutrients and nutrition of citrus fruits. Ch. 1 In Citrus Nutrition and Quality, S. Nagy and J.A. Attaway, Eds. pp. 3-24. American Chemical Society. Washington, DC. Toledo, R.T. and Chapman, J.R. 1973. Aseptic packaging in rigid plastic containers. J. Food Tech. 27:68-76. 196 Trammell, D.J., Dalsis, D.E. and Malone, C.T. 1986. Effects of oxygen on taste, ascorbic acid loss and browning for HTST pasteurized single strength orange juice. J. Food Sci. 51: 1021-1023. Trammell, D.J., Dalsis, DÆ. and Sastr>', S.K. 1997. High voltage pulsed electric field treatment chambers for the preservation of liquid food products. US Patent 5,690,978 Varsel, C. 1980. Citrus juice processing as related to quality and nutrition. In Citrus Nutrition and Quality, Nagy, S., Attaway, J. A., Eds. pp. 225-271. American Chemical Society,Washington, DC. Wagner, J. N. 1983. Inspecting the imposible. Food Eng., 55(6), 79-92. Wadsworth, K.D. and Basette, R. 1985. Laboratory scale system to process uitrahigh- temperature milk. J. Food Prot. 48 (6): 530-531. Wolford, R., Attaway, J., Alberding, G. and Atkins, C. 1963. Analysis of the flavor and aroma constituents of Florida orange juices by gas chromatography. J. Food Sci. 28:320- 328. Yam, K.L. 1995. Pressure differential techniques for package integrity inspection. In Plastic Package Integrity Testing Assuring Seal Quality. B.A Blakistone and C.L. Harper, Eds. pp. 137-145. Institute of Packaging Professionals and Food Processors. Washington, D.C. Yang, X. and Peppard, T. 1994. Solid phase microextraction for flavor analysis. J. Agric. Food Chem. 12: 1925-1930. Yeom, H.W. 2000. The effects of pulsed electric field on the quality of foods. Ph.D. dissertation. The Ohio State University, Columbus. Zhang, Z. and Pawliszyn, J. 1994. Headspace solidphase microextraction. Anal. Chem. 65: 1843-1852. Zhang, Z., Yang, M. and Pawliszyn, J. 1994. Solid phase microextraction. Anal. Chem. 66 (17): 844A-853A. Zhang, Q., Monsalve-Gonzalez, A., Qin, B.L., Barbosa-Canovas, G.V. and Swanson, B.G. 1994. Inactivation of Saccharomyces cerevisiae in apple juice by square-wave and exponential-decay pulsed electric fields. Food Process Eng. 17:469-478. Zhang, Q.H., Qin, B.L., Barbosa-Canovas, G.V. and Swanson, B.G. 1995. Inactivation of E. coli for food pasteurization by high intensif short duration pulsed electric fields. J. Food Process Preserv. 19:103-118. 197