RESIDUAL GAS EFFECTS ON HEAT TRANSFER IN OVERPRESSURE PROCESSING OF FLEXIBLE PACKAGES by SARA ELISABETH WEINTRAUB B.Sc, University of Alberta, 1982 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Food Science) We accept this paper as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April, 1986 © S a r a Weintraub, 1986 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree a t the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head o f my department o r by h i s or her r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n p e r m i s s i o n . Department o f Food Science The U n i v e r s i t y of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date A p r i l 9, 19 86 i i ABSTRACT Experiments have been performed to t e s t e f f e c t s of entrapped a i r on heating rate indices ( f h ) of t e s t pouches subjected to variable con-d i t i o n s i n overpressure r e t o r t processes. Teflon bricks with c e n t r a l l y located thermocouples were employed as a test material. Water was added within the pouch to provide vapor pressure which would simulate the water a c t i v i t y of a food. A i r was i n j e c t e d by means of a hypodermic syringe through a s i l i c o n e rubber droplet on the surface of the pouch which formed a septum. With repeated i n j e c t i o n s , a i r volumes were varied from 0 to 35 mL. Processing temperatures of 115, 120 and 125°C were studied under pressure l e v e l s corresponding to 65, 75 and 85% steam i n a steam/air mixture. Three r e t o r t s were employed: (1) an i n d u s t r i a l - s c a l e h o r i z o n t a l Lagarde steam/air r e t o r t , (2) a p i l o t - s c a l e v e r t i c a l p o s i t i v e flow steam/air r e t o r t and (3) the same p i l o t - s c a l e r e t o r t using superheated water with a i r overpressure. The heating rates of t e f l o n bricks without r e t o r t pouch packaging were also studied at the d i f f e r e n t temperatures and pressures; analysis of v a r i a n c e r e v e a l e d no s i g n i f i c a n t d i f f e r e n c e s (p>0.05) among conditions of temperature and overpressure l e v e l . There were, however, s i g n i f i c a n t differences among the r e t o r t systems used. In general, b r i c k s t e s t e d i n the Lagarde f o r c e d c i r c u l a t i o n steam/air r e t o r t demonstrated more rapid heating than i n the p i l o t - s c a l e r e t o r t with both media types. Relationships of heating rate indices as functions of included a i r volume consisted of e i t h e r one or two e s s e n t i a l l y l i n e a r segments. One segment formed a plateau region i n which there was no change i n f ^ with increasing volumes of included a i r . The second section, apparent with some conditions only, exhibited an increase i n f n as a function of increasing a i r volumes. A moving regression frame consisting of ten c o n s e c u t i v e p o i n t s was compared a g a i n s t a b a s e l i n e by c o v a r i a n c e a n a l y s i s . C r i t i c a l volumes of a i r were considered to be those volumes above which the f ^ vs. a i r volume function was s i g n i f i c a n t l y d i f f e r e n t (p<0.05) from the baseline. Covariance analysis was performed to com-pare s l o p e s of the h e a t i n g r a t e vs. a i r volume f u n c t i o n s with appreciable data i n the second section. Results of the experiments imply that with an increase i n over-pressure, larger amounts of a i r may be entrapped i n the pouch with heat transfer remaining unchanged. Air overpressure above 80 kPa allowed up to 35 mL of a i r to be included with no detrimental e f f e c t on heating rates. Increasing overpressure and temperature was shown to display decreasing slopes i n the second l i n e a r section. C r i t i c a l a i r volumes for water/air overpressure processes were s u b s t a n t i a l l y higher than those with steam/air mixtures. Steeper slopes were evident on the i n -creasing section with steam/air media i n d i c a t i n g a more severe d e t r i -mental e f f e c t of entrapped a i r on heat transfer. Studies with r e t o r t types demonstrated that processing i n the Lagarde r e t o r t allowed for larger amounts of a i r to be included without a f f e c t i n g heat transfer. Predictions of volumes of non-condensible gases which would p r e v a i l at r e t o r t conditions were made by determining expansion factors from id e a l gas r e l a t i o n s h i p s . Comparisons of slopes of the log f ^ vs. adjusted a i r function showed s i g n i f i c a n t d ifferences. Water/air processes i n the p o s i t i v e flow r e t o r t had the least slope, the Lagarde and p o s i t i v e flow steam/air processes had increasing slopes, r e s p e c t i v e l y ^ i v TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS i v LIST OF TABLES v i i LIST OF FIGURES i x NOMENCLATURE x ACKNOWLEDGEMENTS x i i INTRODUCTION 1 LITERATURE REVIEW 5 I Retort Pouch 5 1. Development of Pouch Material 5 2. Evacuation Techniques 6 3. Pouch Seals 9 II Critical Factors Affecting Processing 12 1. Residual Gases 13 a. Pressure-Volume Studies 13 b. E f f e c t on Heating Rate 17 c. Methods of Measurement 21 d. Other E f f e c t s of Residual Gases 23 2. Processing Media 25 3. Processing Conditions 29 A. Package Thickness 31 EXPERIMENTAL 33 I Sample Preparation 33 1. Test Bricks 33 2. Packaging 35 II Air Measurement 37 1. Non-destructive Measurement of A i r 37 2. Destructive Measurement of A i r 40 3. Combining Destructive and Non-destructive Measurements of A i r 42 III Processing Conditions 43 V Page IV Retorts 43 1. Lagarde Steam/Air 45 2. Vertical Positive Flow - Steam/Air 46 3. Vertical Positive Flow - Water/Air 48 V Data Acquisition 50 1. Temperature Profile Measurement 50 2. Analysis of Heating Rate Index (f^) 51 3. Data Treatment 52 RESULTS AND DISCUSSION 54 I Air Measurement 54 II Unpackaged Bricks 56 III Packaged Bricks: Included Air 61 IV Break-point Values 64 1. Determination of Break-point 64 2. Factors Affecting Break-point Volumes 72 a. Steam Percentage or Pressure Level 72 b. Temperature 74 c. Brick Thickness 75 d. Media Type: Steam/Air and Water/Air 76 e. Retorts: Lagarde and Positive Flow Steam/Air 77 3. Importance of Plateau Region 79 V Air Volumes Above the Break-point 80 1. Factors Affecting the Second Section 81 a. Brick Thickness 81 b. Steam Percentage/Pressure Level 83 c. Temperature 86 d. Media Type: Steam/Air and Water/Air 89 e. Retorts: Lagarde and Positive Flow Steam/Air 92 VI Prediction of Pouch Expansion 95 1. Expansion Factors 96 2. Comparison of Adjusted Volume Relationships 100 VII Other Process Parameters 106 v i Page CONCLUSIONS 107 LITERATURE CITED 109 APPENDICES 116 I Non-destructive a i r measurements 116 II Back c a l c u l a t i o n s f o r combination of destructive and non-destructive a i r measurements 118 I l i a Heating rate index vs. included a i r r e l a t i o n s h i p f o r thin bricks i n p o s i t i v e flow steam/air r e t o r t (a, 115°C; b, 120°C; c, 125°C) 119 I l l b Heating rate index vs. included a i r r e l a t i o n s h i p for thick bricks i n p o s i t i v e flow steam/air r e t o r t (a, 115°C; b, 120°C; c, 125°C) 120 IIIc Heating rate index vs. included a i r r e l a t i o n s h i p for th i n bricks i n p o s i t i v e flow water/air r e t o r t (a, 115°C; b, 120°C; c , 125°C) 121 H i d Heating rate index vs. included a i r r e l a t i o n s h i p f o r thick bricks i n p o s i t i v e flow water/air r e t o r t (a, 115°C; b, 120°C; c, 125°C) 122 v i i LIST OF TABLES Page Table 1. Thermophysical properties of t e f l o n . 34 Table 2. Summary of processing temperatures and pressures studied. 44 Table 3. Analysis of variance for heating rate indices of unpackaged t e f l o n b r i c k s . 57 Table 4. Duncan's multiple range te s t for heating rates of unpackaged bricks i n d i f f e r e n t r e t o r t s . 58 Table 5. C o e f f i c i e n t s of v a r i a t i o n of heating rate indi c e s of unpackaged b r i c k s . 60 Table 6. Test for s i g n i f i c a n c e of slope of the baseline 125°C-65% steam with Student-t t e s t . H 0: slope = 0. 66 Table 7. Student-t t e s t on f ^ of unpackaged bricks compared to f-^ values of baseline. H Q: = 67 Table 8. Break-point value of t h i n bricks determined by covariance analysis on a moving regression frame. 70 Table 9. Break-point value of thick bricks determined by covariance analysis on a moving regression frame. 71 Table 10. Covariance analysis r e s u l t s comparing baselines between r e t o r t systems tested. 78 Table 11. Covariance analysis on f ^ vs. included a i r , values above break-point; comparison of brick thickness, 115°C-85% steam (pressure l e v e l 3). 82 Table 12. Covariance analysis on f ^ vs. included a i r , values above break-point; comparison of pressure l e v e l , at 115°C. 84 Table 13. Covariance analysis on vs. included a i r , values above break-point; comparison of temperature at 85% steam pressure l e v e l 3). 87 Table 14. Covariance analysis on f ^ vs. included a i r , values above break-point; comparison of media type. 90 Table 15. Covariance analysis on f ^ vs. included a i r , values above break-point; comparison of steam/air r e t o r t s . 93 v i i i P a g e Table 16. T h e o r e t i c a l expansion factors f o r conditions studied. 99 Table 17. Covariance a n a l y s i s on log f ^ vs. adjusted a i r , a l l data; comparison of systems tested. 104 i x LIST OF FIGURES Page Figure 1. Apparatus for non-destructive measurement of a i r . 38 Figure 2. Apparatus for destructive measurement of a i r . 41 Figure 3. Heating rate index as a function of included a i r ; Lagarde r e t o r t , t h i n b r i c k s . (a, 115°C; b, 120°C; c, 125°C). 62 Figure 4. Heating rate index as a function of included a i r ; Lagarde r e t o r t thick b r i c k s . (a, 115°C; b, 120°C; c, 125°C). 63 Figure 5. Moving regression frame performed with covariance analysis to determine break-points. 69 Figure 6. Heating rate index vs. included a i r f o r comparison of pressure l e v e l ; t h i c k bricks processed at 115°C i n the Lagarde r e t o r t . 85 Figure 7. Heating rate index vs. included a i r f o r comparison of temperature; thick bricks processed at 85% steam t i n p o s i t i v e flow r e t o r t . 88 Figure 8. Heating rate index vs. included a i r for comparison of media type; thick bricks processed at 115°C and 85% steam (3). 91 Figure 9. Heating rate index vs. included a i r for comparison of steam/air r e t o r t s ; thick bricks processed at 115°C and 85% steam. 94 Figure 10. Heating rate index functions of adjusted a i r volumes for a l l conditions; t h i n b r i c k s . (a, Lagarde; b, P o s i t i v e Flow Steam/Air; c, P o s i t i v e Flow Water/Air). 101 Figure 11. Heating rate index functions of adjusted a i r volumes for a l l conditions; thick b r i c k s . (a, Lagarde; b, P o s i t i v e Flow Steam/Air; c, P o s i t i v e Flow Water/Air). 102 X NOMENCLATURE a half-length of t e f l o n brick (cm) A t h i r d non-destructive a i r reading, a f t e r processing (g) a w water a c t i v i t y b half-width of t e f l o n brick (cm) B buoyant force (N); second non-destructive a i r reading, before processing (g) c half-thickness of t e f l o n brick (cm) Cp s p e c i f i c heat capacity (cal/g C°) EF expansion factor f ^ heating rate index (min) Fo process l e t h a l i t y (min) g g r a v i t a t i o n a l force (N/kg); difference i n degrees between the r e t o r t temperature and product temperature (C°) H ^ height of water of each mL i n the buret (cm) H w height of water volume (cm) I f i r s t non-destructive a i r reading, i n i t i a l reading (g) j lag f a c t o r k thermal conductivity ( c a l / s cm C°) Mg weight of object (mg) Pj atmospheric pressure (kPa) P2 p a r t i a l p r e s s u r e of water at temperature T2 p l u s p a r t i a l pressure of enclosed a i r (kPa) P a p a r t i a l pressure of a i r enclosed (kPa) pressure when package i s i n a state of neutral buoyancy (kPa) Pj^ pressure of r e t o r t during processing (kPa) STP standard conditions of temperature and pressure x i T l i n i t i a l temperature (K) T 2 temperature of process (K) V l volume of a i r i n pouch at atmospheric conditions (mL) V2 volume of a i r i n pouch at processing conditions (mL) Vm measured volume of a i r (mL) V max maximum volume to f i l l the pouch (mL) V o volume of object at STP (mL) Vpch volume occupied by empty pouch (mL) V s volume of sample (brick and pouch) excluding a i r (mL) V sa volume of sample (brick and pouch) including a i r (mL) V t volume of object at r e t o r t conditions (mL) W h pressure of water l e v e l i n graduate (kPa) Wpch weight of an empty pouch (g) W s weight of sample (g) Wws weight of package suspended i n water (g) W j. ws+a weight of package and added a i r suspended i n water (g) a 2 thermal d i f f u s i v i t y (m / s ) ; l i n e a r thermal expansion factor B volume expansion Am differe n c e i n mass (g) At difference i n temperature (°C) Po density of object (g/mL) P p r density of product (g/mL) pw density of water (g/mL) x i i ACKNOWLEDGEMENTS The author i s si n c e r e l y g r a t e f u l to Dr. Marvin A. Tung f or his knowledgeable a d v i c e and encouragement d u r i n g the course of t h i s research project and review of the t h e s i s . She also wishes to thank the members of the research committee, Dr. W.D. Powrie, Dr. J . Vanderstoep, both of the Department of Food Science and Dr. K. V. Lo of the Department of Bio-Resource Engineering for t h e i r contribution to the project and constructive review of the th e s i s . Much appreciation i s extended to H.S. Ramaswamy and Trudi Smith for th e i r valuable expertise and assistance. The author i s g r a t e f u l to her parents, Jerome gnd Ann Weintraub for t h e i r support and understanding. A s p e c i a l thank you i s extended to Benoit Girard for his patience and encouragement. -1-INTRODUCTION Innovative food packaging has played an important r o l e i n providing consumers with the wide variety of food products that e x i s t s today. The research and development of alternate packaging technology i s a con-t i n u a l process. One recent development, a s h e l f - s t a b l e t h i n p r o f i l e f l e x i b l e package or r e t o r t pouch, was projected to reach $2 b i l l i o n i n s a l e s by 1980 ( P e t e r s , 1975) and earned the 1978 Food Technology I n d u s t r i a l Achievement Award (Mermelstein, 1978). Comparisons of qual-i t y and energy requirements place r e t o r t pouch products i n between t r a d i t i o n a l canned and frozen foods. Products i n pouches may have superior q u a l i t y as compared to canned goods and processing was reported to be less energy-intensive than freezing (Heintz, 1980). Si n c e r e t o r t pouches have been c o n s i d e r e d an a l t e r n a t i v e to canning, comparisons are t y p i c a l l y made between pouches and cans. Advantages of the f l e x i b l e package have been reported by many authors (Tuomy and Young, 1982; Cage and C l a r k , 1980; Mermelstein, 1976). Shorter processing times required for t h i n p r o f i l e packages r e s u l t i n improved taste, color, texture and retention of heat-sensitive vitamins (Davis et a l . , 1972). Preparation by the boil-in-bag technique i s simple and requires l i t t l e clean-up. The package i s e a s i l y opened and uses l e s s s t o r a g e and d i s p o s a l space. A major concern to food processors i s cost. Steffe et a l . (1980) reported that the t o t a l energy requirement f o r r e t o r t i n g pouches was 75% less than f o r cans. A 60% lower energy requirement when p r o c e s s i n g i n f l e x i b l e packages was r e p o r t e d by Mermelstein (1978). W i l l i a m s et a l . (1983) d e s c r i b e d reduced m a t e r i a l and storage c o s t s a s s o c i a t e d with r e t o r t pouch -2-packaging. However, i n i t i a l equipment costs were reported to be high. Although projections have been made and advantages c i t e d , the use of r e t o r t pouches has been l i m i t e d . European countries and Japan have had reasonable success i n t h e i r introduction. Production i n North America has been l i m i t e d to small i n d i v i d u a l companies. Presently, the industry targets for s p e c i a l t y goods or foods for a p a r t i c u l a r function. Peters (1985) described a l i n e of Smoky Canyon foods produced by Land o' Frost (Lansing, IL) aimed d i r e c t l y toward campers and outdoor enthus-i a s t s . Adams (1984) i n d i c a t e d t h a t t h e r e was a b r i g h t f u t u r e i n processing of seafood products i n r e t o r t pouches. Higher priced pro-ducts such as red snapper and blue crab meat for restaurant use could be competitive with frozen counterparts (Adams et a l . , 1983). For wide-spread consumer acceptance of the technology, t h e r e must be more exposure to products, perhaps by popular national brands i n r e t o r t pouches. Adoption of the technology by large companies has been slow. Kraft introduced an "A l a Carte" l i n e which was manufactured i n f a c i l i t i e s having a r e l a t i v e l y low capacity (Anon. 1982), but production was sus-pended i n 1985. Anon. (1982) reported that Nestle and Campbell Soup companies have shown an i n t e r e s t . H e s i t a t i o n towards l a r g e - s c a l e investment may be due to some technological problems such as slow l i n e speeds and the lack of regulations or concrete information regarding aspects of thermal processing. Introduction of materials and technical aspects have made the technology a r e a l i t y , but a d d i t i o n a l knowledge of c r i t i c a l f a c t o r s a f f e c t i n g r e t o r t pouch processing i s e s s e n t i a l . Beverly (1980) described c r i t i c a l factors as those which a f f e c t the -3-adequacy of the s t e r i l i z a t i o n process, product quality or the economics of the system. Some of the c r i t i c a l parameters are a consequence of the f l e x i b l e nature of r e t o r t pouches. When processing with a f l e x i b l e f i l m , there i s a tendency for the i n t e r n a l pressure within the package to equal the r e t o r t pressure. Only a small pressure d i f f e r e n t i a l can be maintained. Therefore, as the product temperature r i s e s , the i n t e r n a l pressure may increase and i f i t exceeds r e t o r t pressure to a large degree, the pouch may burst. This i s a p a r t i c u l a r problem during the early stages of the cooling process. A i r overpressure processes have been used to overcome t h i s problem. More information i s needed on the performance of these media types i n r e l a t i o n to t h e i r composition. Another c r i t i c a l factor of utmost importance i n thermal processing of pouches i s the release of entrapped r e s i d u a l gases from the product during heating. Moreover, there i s i n e v i t a b l y a c e r t a i n amount of a i r which remains within the pouch when i t i s sealed. The f l e x i b l e nature of the pouch allows for expansion of these gases which contributes to the problem of pouch volume increases which may have a detrimental e f f e c t on heating behavior. For r e t o r t pouch technology to reach i t s forecasted success, technical aspects must be advanced and c r i t i c a l parameters i n thermal processing must be i d e n t i f i e d and characterized. Heat penetration studies and process parameters derived may be used to design adequate heat processes. They also provide a means to e v a l -uate heating behavior of products subjected to various c r i t i c a l f a c t o r s . In a heat penetration t e s t , a time-temperature curve at a given point i n the product i s recorded. Changes i n the amount of thermal energy i n a food are produced by transmission of energy from the heating media -4-surrounding the object (Ball and Olson, 1957). Thermocouples and poten-tiometric devices are used to sense changes in heat energy. Pflug (1975) and Bee and Park (1978) described many types of equipment avail-able and techniques used to perform heat penetration tests. Data collected are organized by plotting the temperature difference between the retort and centerpoint (g) on a logarithmic scale as a function of time on an arithmetic scale. Stumbo (1973) discussed different procedures for plotting and evaluating thermal processing parameters. Computer programs have allowed for analysis of heat pene-tration data in a minimal time (Tung and Garland, 1978). An important process parameter, the heating rate index (f^) is the time required for the straight line portion of the heat penetration curve to traverse one log cycle. Heating rate indices have been used by many authors (Evans, 1977; Berry and Kohnhorst, 1983; Ramaswamy, 1983) to evaluate c r i t i c a l factors by comparing heating behavior of foods in various package forms. Because of the need to understand the effects of noncondensible gases within a f i l l e d retort pouch on the thermal sterilization process, a study was initiated using a model system. The objectives of this investigation were to: 1. Study the influence of noncondensible gases within a pouch on heating rate indices of materials contained by those pouches, 2. Characterize and compare this influence under different over-pressure processing conditions and retort systems, and 3. Determine limits at which residual gases would interfere with ex-pected heating behavior, as indicated by changes in the heating rate indices. -5-LITERATURE REVIEW I Retort Pouch 1. Development of Pouch Material Development of a pouch capable of withstanding thermal processing began with research i n t o adequate materials and s e a l i n g techniques i n the l a t e 1950's. Origins of many research projects l i e i n work done by Reynolds Metals Inc. (Richmond, VA), Continental F l e x i b l e Packaging (Chicago, IL) and the U.S. Army Natick R&D Command (Natick, MA). Both Reynolds and Continental manufactured packaging materials and have work-ed independently or i n conjunction with the Natick Laboratories to develop and evaluate s u i t a b l e materials for the production of s h e l f stable foods i n r e t o r t a b l e pouches. The U.S. Army has a vested i n t e r e s t i n r e t o r t pouches with hopes to completely replace the "meal-combat-i n d i v i d u a l " or C r a t i o n with a new "meal-ready-to-eat" using r e t o r t pouches and f a m i l i a r foods (Mermelstein, 1978). Early work by Nelson and Steinberg (1956) tested the s u i t a b i l i t y of polyester (50 and 75 um) and Trithene (50 and 100 um) for r e t o r t i n g foods. They concluded that general strength of the films was good but closure seals made with an impulse sealer must be improved. The pack-ages tested were clear laminates, which contributed to eye appeal, but created a problem with gas permeability (Hu et a l . , 1955; Ayoub et a l . , 1974). K e l l e r (1959) reported that laminates containing aluminum showed great promise because of excellent b a r r i e r properties as was shown by a polyester/aluminum f o i l / v i n y l laminate. Evaluations of more than 100 packaging materials, including many p l a s t i c s and combinations, were ca r r i e d out and i t was determined that aluminum f o i l was an e s s e n t i a l -6-component of the packaging m a t e r i a l (Rubinate, 1964). A l l r e t o r t pouches used i n North America have aluminum f o i l as the middle material i n the laminate. F o i l - f r e e pouches were produced i n Japan as reported by Ebben (1979) and by DRG Packaging i n England. Products i n these transparent packages had a shelf l i f e of s i x to eight weeks. T y p i c a l outer material was polyester, which provided p r i n t a b i l i t y and strength. Nylon was suggested but proved to be c o s t l y and was s e n s i t i v e to water (Goldfarb, 1970). The inner layer was commonly polypropylene or a m o d i f i e d p o l y o l e f i n s e r v i n g as a heat s e a l a b l e l a y e r and an i n e r t surface for food contact. Strength of the f l e x i b l e material has been tested primarily as a package for m i l i t a r y s e r v i c e . Agarwal and Kumta (1974) studied the puncture resistance of f l e x i b l e packages with d i f f e r e n t probes. Prelim-inary studies of rough handling evolved the conventional practice of enclosing the pouch i n a paperboard f o l d e r . Burke and Schultz (1972) performed rough handling treatment by v i b r a t i o n and drop te s t s f or comparison of pouches and cans. Their r e s u l t s showed no difference between the containers i n t h e i r c a p a b i l i t i e s of withstanding rough handling. However, cans resulted i n more damage to the outer shipping container. 2. Evacuation Techniques Removal of noncondensible gases from r e t o r t pouches i s standard p r a c t i c e . The s i g n i f i c a n c e of gas removal w i l l be discussed i n more d e t a i l i n a forthcoming section. E s s e n t i a l l y , i t i s desirable to remove as much gas as possible. Evacuation i s performed as part of a m u l t i -f u n c t i o n o p e r a t i o n ; u s u a l l y i n c o n j u n c t i o n with f i l l i n g and/or a -7-closure-sealing operation (Larapi, 1977). F i r s t generation vacuum s e a l -e r s used preformed pouches. An elementary attempt employed the "snorkel" technique, where a tube with a tapered end was inserted into a short unsealed section of the pouch. Goldfarb (1971) described a i r removal with t h i s technique. The pouch was purged and a vacuum was applied. When the pouch collapsed, the snorkel was withdrawn. Residual a i r contents were reported to be l e s s than 5 mL per pouch. This tech-nique presented problems by possible contamination of the s e a l area with tube withdrawal. Use of mechanical vacuum chambers was more widespread and s i m i l a r to systems i n use for luncheon meats (Lampi and Rubinate, 1973). Two systems commonly used were Swiss-Vac (Hamac-Hansella, Piscataway, NJ) and a d e s i g n by Koch M u l t i v a c (Kansas C i t y , MO). Horizontal conveyors which moved under a hood and closed to form a vacuum chamber, were c h a r a c t e r i s t i c of t h i s type. Morris (1981) report-ed use of a Swiss-Vac system. Pouches were placed manually onto pos-i t i o n s r e s u l t i n g i n slow production speeds, but a better vacuum was created. Systems are a v a i l a b l e implementing steam f l u s h to remove a i r from the headspace. An improvement i n pouch f i l l i n g speeds from 30 pouches/ min f o r mechanical vacuum to 60 pouches/min with steam f l u s h were reported by Przybyla (1984). Morris (1981) described use of a steam-vacuum Mitsubishi machine by Magic Pantry Foods Inc. (Hamilton, ON). Steam was i n j e c t e d immediately p r i o r to sealing to create a vacuum upon cooling. F i l l speeds were f a s t e r (rated at 50 pouches per minute) and although poorer a i r removal occurred, i t was well suited to more l i q u i d products. A development by FMC (Santa Clara, CA), the Steam Flow Pouch -8-F i l l e r and Sealer, was reported by Strasser (1979). This operation involved use of both mechanical vacuum and steam f l u s h i n g . Pouches were f i r s t flushed with steam, then moved to an a i r - f r e e environment to be f i l l e d . An a d d i t i o n a l j e t steaming or addition of i n e r t gas was i n t r o -duced j u s t p r i o r to s e a l i n g . B e v e r l y (1979) demonstrated the a i r removal e f f i c i e n c y of t h i s equipment. Removal of 40-50% of the o r i g i n a l a i r was found with steam f l u s h alone; furthermore, with the steam tunnel system, 93-95% of a i r was removed. Removal of a i r i n other systems was ensured by p h y s i c a l man-i p u l a t i o n of the pouch. Tsutsumi (1979b) described a non-vacuum method for packaging viscous l i q u i d s . A i r within the pouch was reduced to a minimum by stretching both sides of the pouch p r i o r to s e a l i n g . Normal operations of opening, f i l l i n g and sealing occurred. A modification to t h i s type of evacuation involved applying pressure with plates from both sides, creating a squeezing action to force gases out of the package. Narrow bars prevented the product from being pushed up into the seal area. Heid (1970) described a unique system designed with the intent of reducing r e s i d u a l headspace gases. The system, developed and patented i n the United States by FMC Corporation (Santa Clara, CA), d i f f e r e d from other processes i n that evacuation occurred during processing and the pouch was sealed a f t e r cooking. Two clamps held the unsealed portion of the pouch stretched to act as a one-way valve allowing gases to escape. The pouch was submerged i n water for heat processing and concomitant exhaustion of r e s i d u a l gases. Sealing was performed a f t e r the desired cook. A low l e v e l of r e s i d u a l a i r (1 mL) was reported, however, no i n d u s t r i a l a p p l i c a t i o n has been reported to date. -9-A second generation of equipment was developed to form, f i l l , remove a i r and s e a l the pouch. P r o d u c t i o n speeds of t h i s type of machinery were reported by Cage and Clark (1980) to be approaching 250 pouches per minute which would greatly increase production e f f i c i e n c y . Additional savings could r e s u l t from the use of r o l l stock rather than preformed pouches. Lopez (1981) claimed that although equipment costs were higher there would be s u b s t a n t i a l l y lower operational costs. At l e a s t two systems of t h i s type have been developed. Rexham's B a r t e l t Div. (Sarasota, FL) introduced a higher speed f o r m / f i l l / s e a l machine operating e l e c t r o n i c a l l y with a microprocessor (Anon., 1980). Another system, ho r i z o n t a l i n design, was introduced by Koch Multivac (Kansas Ci t y , MO). It was capable of 120 packages per minute, although r e a l -i s t i c a l l y 90 packages per minute were t y p i c a l (Anon., 1983). More advances i n the t e c h n i c a l aspects of r e t o r t pouch processing may r e s u l t i n lower production costs, making the technology more favorable. 3. Pouch Seals Seal i n t e g r i t y has long been considered as a v i t a l c r i t e r i o n for food containers. Thermally processed foods packaged i n f l e x i b l e pouches pose a p a r t i c u l a r problem due to stresses which develop during heating. A statement by Nelson and Steinberg (1956) that " i f the seal area could stand pressure s i m i l a r to the general surface, a major stumbling block to t h i s type of processing could be removed" i l l u s t r a t e d t h i s problem. With the development of new materials and modes of sealing, advances have occurred. A recent study by Roop et a l . (1983) reported that f a i l -ure of pouches du r i n g t e s t i n g was due to the m a t e r i a l s t r e n g t h -10-l i m i t a t i o n s rather than the seal formed. Closure seals for preformed pouches and a l l peripheral seals from r o l l stock were formed on the previously described f i l l e r / s e a l e r or f o r m / f i l l / s e a l equipment. A review of heat sealing methods by Brown and Keegan (1973) described techniques for many package types. T y p i c a l l y , for r e t o r t pouches, either the hot jaw or impulse seal was used. A hot jaw sealer consists of a bar or plate with an assembled heater on top and a silicone-rubber r e s i l i e n t jaw f i x e d underneath. A pressure source f o r c e s down the upper jaw to c r e a t e the s e a l . An impulse s e a l e r u t i l i z e s a tension wire supported at each end. Voltage applied to the clamps created a current flow through the high resistance wire which caused heating to form the s e a l . The contact surfaces were covered with a t e f l o n coated f i b r e g l a s s c l o t h . Lampi et a l . (1976) summarized c r i t e r i a for a high performance s e a l . Fusion of the inner layers was considered to be e s s e n t i a l . Assurance that a weld was formed could be tested by tensioning the seal to the point of f a i l u r e . Signs of delam-i n a t i o n would i n d i c a t e t h a t the f i l m s u r f a c e s were f u s i o n s e a l e d . S p e c i f i c a t i o n s of t e n s i l e strength required by the U.S.D.A. were report-ed by Shenkenberg (1975) to be 7 pounds per l i n e a r inch (12.3 N/cm). Pressurization-hold t e s t i n g required the a b i l i t y to withstand 138 kPa (gauge) for 30 seconds. Rubinate (1964) cautioned processors that seal strength was reduced at high temperature and recommended restrainment during processing to prevent seal rupture. In addition, there must be no signs of v i s i b l e aberrations. M i l i t a r y standards required a clean defect-free band across the closure, with a minimum width of 1/16 i n (1.6 mm) f o r an impulse s e a l and hot bar s e a l of 1/8 i n (3.2 mm) -11-according to M i l l e v i l l e (1981). V i s i b l e defects were thought to r e s u l t mainly from contamination i n the seal area or wrinkling of the pouch material. A s p e c i a l design i l l u s t r a t e d by Lampi et a l . (1976) used guards which were lowered i n t o the package opening during f i l l i n g to protect the upper areas. Improved f i l l e r nozzle design and high prec-i s i o n f i l l i n g pumps have helped reduce contamination (Tsutsumi, 1972). Reducing the o c currence of w r i n k l i n g has posed another c h a l l e n g e . Stretching of the closure area p r i o r to sealing has helped to minimize fold-over wrinkles. Despite these improvements, defects occur; however, methods e x i s t to detect them. V i s u a l examination i s a common p r a c t i c e . Use of 100% inspection was recommended rather than sampling a random proportion of the pouches ( M i l l e v i l l e and Badenhop, 1980). Tsutsumi (1974) reported 0.02% defects from a f i l l i n g operation with complete v i s u a l inspection. Lampi (1977) stated that as a rule of thumb, v i s u a l inspections under i d e a l conditions are 75% e f f e c t i v e . A system whereby c r i t e r i a for r e j e c t i o n were set, based on the severity of defects, was described by Lampi et a l . (1976). Small wrinkles or b l i s t e r s would be considered a c c e p t a b l e , however, occluded p a r t i c l e s or l a r g e w r i n k l e s must be rejected. There i s a need for development of non-destructive t e s t equipment for on-line inspection of sealed pouches. Infra-red scanning has been described as an e f f e c t i v e defect detector (Goldfarb, 1971; Lampi, 1977). The instrument works on transient-heat detection where a contaminated seal w i l l show surface temperature v a r i a t i o n s by impedance of the heat f l u x . D e t a i l s of the design were reported by Lampi (1977). A c a l i p e r -12-measurement was developed which detected i r r e g u l a r i t i e s i n seal t h i c k -ness due to contaminants. G a g l i a r d i et a l . (1984) designed a computer-aided video inspection system for q u a l i t y control of pouch sea l s . An a c q u i s i t i o n module stored an image of the product, then a computer analyzed the image to i s o l a t e defects and determine whether they would merit r e j e c t i o n . Mechanical implementation of the decision rejected or accepted the package on-line. This system required extensive program-ming on c r i t e r i a for defects and threshold l e v e l s of r e j e c t i o n . Stringent inspection and assurance that proper seals are formed free of contamination or defects, aids processing of r e t o r t pouches. If defects occur, weak spots may form i n the pouch material r e s u l t i n g i n f a i l u r e from stresses imposed by thermal processing. II C r i t i c a l Factors Affecting Processing As the technology for r e t o r t pouches continues to' develop, there i s an increasing need for e s t a b l i s h i n g processing and packaging s p e c i f i c -a t i o n s . C o n s i d e r a b l e work has been done s e t t i n g standards f o r evaluation of pouch seals and material i n t e g r i t y . A s i m i l a r e f f o r t towards i d e n t i f y i n g and c o n t r o l l i n g c r i t i c a l f a c t o r s during thermal p r o c e s s i n g has not been apparent (Badenhop and M i l l e v i l l e , 1980). I d e n t i f i c a t i o n of c r i t i c a l factors i s an ongoing process. For example, just recently Smith et a l . (1985) reported the r o l e of removal of con-densate from v e r t i c a l r e t o r t s as a c r i t i c a l parameter i n the w e l l established t r a d i t i o n of processing canned foods i n v e r t i c a l steam r e t o r t s . In order to set regulations governing the processing of low-acids foods i n r e t o r t pouches, studies evaluating c r i t i c a l parameters are c r u c i a l . -13-Factors are generally considered to be c r i t i c a l i f they a f f e c t the adequacy of the s t e r i l i z a t i o n process, product q u a l i t y or the economics of the system. Many factors must be considered and, due to the f l e x i b l e nature of the pouch, a d d i t i o n a l parameters e x i s t that are not present for r i g i d containers. Berry (1979) i d e n t i f i e d a number of c r i t i c a l f actors for r e t o r t pouch processing. Those factors pertaining to the product are: heating c h a r a c t e r i s t i c s , pouch s i z e , f i l l weight and residual gas entrapment. Factors contributed by r e t o r t design are racking configurations, the heating media employed, mode of c i r c u l a t i o n , come-up time and the effectiveness of the cool c y c l e . Many of these factors w i l l be dealt with i n more d e t a i l . 1. Residual Gases Although many c r i t i c a l factors i n pouch processing are s i m i l a r to those i n metal can processing, the e f f e c t s of entrapped gases d i f f e r considerably. In thermal processing of r i g i d cans only a small change i n t o t a l volume occurs. However, due to the f l e x i b l e nature of the pouch, a s i g n i f i c a n t change i n t o t a l volume may r e s u l t from thermal expansion of the s o l i d , l i q u i d s and, p a r t i c u l a r l y , entrapped gases. Lampi (1977) c i t e d reasons for removal of gases as primarily providing preclusion of pouch bursting during r e t o r t i n g , assurance of uniform and predictable heat transfer and product s t a b i l i t y . A dditional advantages of removal of a i r are easier detection of spoilage and easier cartoning or casing. a. Pressure-Volume Studies One of the f i r s t studies regarding a i r entrapped i n f l e x i b l e packages was performed by Wallenberg and J a r n h a l l (1957). Their r e s u l t s -14-showed that a maximum r a t i o between enclosed a i r and surface area of the package must not be exceeded i n order to prevent bursting. The maximum l e v e l d i f f e r e d depending on the conditions of s t e r i l i z a t i o n . When subjected to dry a i r there was no apparent bursting with a r a t i o of volume to surface of 2:1 (corresponding to 75% of maximum volume of the pouch) compared to a r a t i o of 3:1 (or 40% of maximum volume) r e s u l t i n g i n b u r s t i n g when a u t o c l a v e d w i t h steam. There was, however, the sug g e s t i o n t h a t m i c r o s c o p i c l e a k s o c c u r r e d i n the dry a i r t r i a l s , r e l i e v i n g i n t e r n a l pressures without bursting. I t was also evident from t h e i r studies that the resistance to bursting was dependent on the moisture content of the a i r enclosed i n the package. R e l a t i v e humidities of 100% created high pressures i n the order of 80.3 kPa with a s e a l e d package of co n s t a n t volume. Rubinate (1964) s t u d i e d the f e a s i b i l i t y of thermal processing using f l e x i b l e packages i n 100% steam and a water cook with a i r overpressure as used f o r glass containers. Results indicated that for use of pure steam as a heat transfer medium i t was e s s e n t i a l to remove as much a i r as possible. A maximum of 10 mL of r e s i d u a l gas i n a 155 g pouch permitted processing with a pure steam environment without bursting the pouch. Similar r e s u l t s were found by K e l l e r (1959), but although use of pure steam was possible i f enough r e s i d u a l gas was removed, a water cook with superimposed a i r pressure was preferred. The f l e x i b i l i t y of the fil m s and r e s u l t i n g expansion from entrapped gases d u r i n g h e a t i n g prompted s t u d i e s by Davis et a l . (1960) on pressure/volume r e l a t i o n s which e x i s t i n f l e x i b l e containers. A s p e c i a l apparatus was designed to measure the expansion i n volume of packages -15-containing water and a i r with increasing i n t e r n a l pressure. Internal pressures were found to develop due to an increase i n the vapor pressure of the water, an i n c r e a s e i n pres s u r e of a i r i n the headspace and expansion of the food product with i n c r e a s i n g temperature. These i n t e r n a l pressures were quantified using the i d e a l gas law, as follows: water at temperature T2, plus the p a r t i a l pressure of the enclosed a i r . In the experiment, the maximum volume of the bag was assumed to be 105 mL. Good agreement was found between the pressure calculated by the gas laws and the actual pressure created i n s i d e the pouch with various volumes of entrapped a i r . A maximum pressure d i f f e r e n t i a l of 2.7-23 kPa developed i n a pouch cooled with superimposed a i r pressure, and yet no f a i l u r e s were found. Data from i n t e r n a l pressures that e x i s t and know-ledge of strength of the pouch allow the c a l c u l a t i o n of optimum f i l l i n g l e v e l s of containers. The use of overriding a i r pressure to prevent expansion of r e t o r t pouches was described by Whitaker (1971). Ideal gas re l a t i o n s h i p s and thermal expansion of s o l i d and l i q u i d components i n a food system of peas and brine were used to ca l c u l a t e pressures i n the system. When no a i r was present, 227 g of product would expand an a d d i t i o n a l 11.3 mL when processed at 126°C with pure steam. A maximum allowable pouch (1) where P^V^ and T^ are pressure, volume and temperature (absolute) of the a i r before processing, T2 i s temperature of the heating process and V m a x , the maximum volume of the pouch. P2 i s the p a r t i a l pressure of -16-expansion of 35 mL was predetermined. Therefore, the volume at r e t o r t conditions did not exceed the allowable expansion. Consideration of the same system wi t h 5 mL of a i r was d e s c r i b e d . I f the product was e s s e n t i a l l y dry, the volume of a i r would decrease due to increased pressure of steam within the r e t o r t . However, when water was contained i n the system, Dalton's law of p a r t i a l pressures must be considered. As water i n the product reached r e t o r t temperature, the p a r t i a l pressure of vapor within the pouch was the same as r e t o r t pressure. P a r t i a l press-ure of a i r i n the system can be calculated assuming a maximum pouch volume of 35 mL and taking into account expansion of the contents. Calculated p a r t i a l pressures of included a i r were equivalent to the required overriding a i r pressures for prevention of pouch bursting. Yamano (1976) predicted pouch expansion i n order to determine the pressure needed to prevent pouch bursting. When the pouch was i n a state of equilibrium, the r e t o r t pressure was assumed to be equal to the i n t e r n a l pouch pressure. As with previously described experiments, the i n t e r n a l pressure was considered to be due to both water vapor and a i r pressure. From the i d e a l gas laws, an expansion r a t i o of the headspace and a t o t a l pouch volume were determined. Expansion r a t i o s of headspace gases at various steam/air r a t i o s were determined at d i f f e r e n t temp-eratures. A r e l a t i o n s h i p between p a r t i a l pressure of a i r within the r e t o r t was plotted against processing temperature f o r d i f f e r e n t head-space expansion r a t i o s . A maximum allowable expansion r a t i o of two for a 165 x 200 mm s i z e pouch containing 250 mL of food and a headspace volume of 50 mL was assumed. From t h i s , an expansion r a t i o of the headspace c o u l d be c a l c u l a t e d , and consequently a s a f e o v e r r i d i n g -17-pressure of 20-30 kPa was determined, b. E f f e c t on Heating Rate Besides the problem of pouch bursting, r e s i d u a l gases included i n pouches are of concern for safety reasons since a predictable heat t r a n s f e r i s r e q u i r e d . Nelson and S t e i n b e r g (1956) r e c o g n i z e d the p o t e n t i a l problem of reduced heat trans f e r due to expansion of gases i n f l e x i b l e pouches. Expansion of gases would r e s u l t i n an i n s u l a t i n g e f f e c t from poor heat conductivity of entrapped gases. They recommended that superimposed a i r pressure (at 220 kPa, gauge) held throughout the process would improve heat transfer because entrapped gases would not be allowed to expand. Results, however, were not quan t i f i e d . A study performed at Toyo Seikan Kaisha Ltd. (1973), a producer of r e t o r t a b l e pouch products i n Japan, described the i n s u l a t i n g e f f e c t of entrapped gases. Experiments were performed on pouches containing 180 g of a curry product processed at 120°C. The heating rate index, f^ , or negative r e c i p r o c a l slope of the s t r a i g h t l i n e portion of a heating curve and the time required f or a selected process l e t h a l i t y were evaluated. Results demonstrated that increasing headspace gas volumes from 0 to 15 mL caused an increase i n f ^ values from 6 to 7.2 minutes. With an a d d i t i o n a l 5 mL to a t o t a l of 20 mL, an f n value of 9.4 minutes was r e p o r t e d . Higher f n v a l u e s r e p r e s e n t slower heat p e n e t r a t i o n r a t e s . Longer p r o c e s s i n g times r e q u i r e d f o r adequate s t e r i l i z a t i o n were reported with larger amounts of a i r entrapped i n the package. A s i m i l a r r e s u l t was reported by Kopetz et a l . (1979) when i n s t i t u t i o n a l s i z e pouches were processed i n horizontal o r i e n t a t i o n . Increases i n process time up to 35% had been noticed when 150 mL of -18-r e s i d u a l gas remained within the pouch. S t r a t i f i c a t i o n and formation of an i n s u l a t i n g layer accounted for the increases seen. Predictions by Badenhop and M i l l e v i l l e (1980) that the i n s t i t u t i o n -a l s i z e r e t o r t pouch would replace the #10 can has become of i n t e r e s t . S i g n i f i c a n t cook time and cost advantages were reported. The low pro-f i l e of the pouch would improve product quality because les s severe heating was required. In a 12 x 15 i n (30.5 x 38.1 cm) pouch capable of holding 5 lb (2.3 kg), the cold spot i s 0.75 i n (19 mm) from the pouch surface. In contrast, the cold spot i n a number 10 can with the same capacity i s 3 i n (76 cm) from the can sides and ends. I t was also suggested that the i n s t i t u t i o n a l market may be easier to deal with since only a few buyers would be involved i n choosing the product. However, i n s t i t u t i o n a l s i z e pouches present a p a r t i c u l a r problem with regard to re s i d u a l gases, prompting research i n t h i s area. Huerta-Espinosa (1981) investigated the e f f e c t s of entrapped a i r content on heat penetration c h a r a c t e r i s t i c s of pears and green beans packed i n i n s t i t u t i o n a l s i z e r e t o r t pouches. Processing experiments were performed with horizontal r e t o r t racking o r i e n t a t i o n using both constraining ( f i r m l y sandwiched) and r e s t r a i n i n g methods. Required thermal processing times to reach an F 2 0 0 F ° ^ ^ m^-n ^ o r diced pears and an F Q value of 5 min f o r green beans were found to i n c r e a s e by approximately 15% f o r each 100 mL i n c r e a s e i n a i r content f o r constrained pouches. Increases were larger when pouches were placed i n restrained or unconstrained conditions. Similar r e s u l t s were found by Beverly et a l . (1980) from studies with i n s t i t u t i o n a l s i z e , 30.5 x 43.2 cm (12 x 17 in) r e t o r t pouches processed i n confined and unconfined -19-h o r i z o n t a l o r i e n t a t i o n . Water was contained i n the package with a i r volumes ranging from 0 to 250 mL and the required process times to achieve an F Q value of 6 min were determined. Results revealed that when pouches were unconfined there was a 20% increase i n required pro-cess time with 250 mL of a i r ; whereas, confined pouches displayed a 10% increase. The negative impact of entrapped gases a f f e c t s the product-i v i t y and economics of the system. I t appeared to be more d r a s t i c i f pouches were processed while unconfined on supporting racks. Another study by Berry and Kohnhorst (1983) focused on entrapped gases i n i n s t i t u t i o n a l s i z e r e t o r t pouches. Whole kernel corn and cream of celery soup were processed f o r 40 and 50 minutes, r e s p e c t i v e l y , i n pouches containing up to 250 mL of res i d u a l a i r . Heating rate indices of the cream of celery soup were found to increase from 33.5 min with no a i r present to a value of 57.3 min when 250 mL of a i r was included i n the pouch. Broken heating curves were displayed i n pouches containing whole k e r n e l corn i n b r i n e . The f ^ remained about the same with increasing a i r contents but f2 from the second s t r a i g h t l i n e portion of the heating curve demonstrated an increase of 67% when 250 mL of a i r was entrapped. Racking design was also shown to be a contributing factor to the e f f e c t s of entrapped gases i n studies by Evans (1977). A i r volumes of 0, 20 and 50 mL were tested i n r e t a i l s i z e pouches containing a 7% bentonite suspension. Pouches were tested when processed i n restrained (unconfined) or constrained positions at a i r overpressures of 34.5 and 69 kPa. Results showed that both a i r content and the method of racking were s i g n i f i c a n t . Increased f ^ values of 11% were found when 50 mL of -20-a i r was included i n a constrained pouch. Restrained pouches showed increases of 19% f o r pouches containing 50 mL of a i r compared to 0 mL. A i r overpressure was found not to be a s i g n i f i c a n t factor i n the l e v e l s tested. Ramaswaray (1983) investigated the e f f e c t s of packaging on s i l i c o n e rubber bricks thermally processed either i n ho r i z o n t a l or v e r t i c a l o r i e n t a t i o n under d i f f e r e n t processing conditions. Entrapped a i r (15-30 mL) i n pouches did not appear to influence heating rates i n the v e r t i c a l constrained o r i e n t a t i o n but f ^ values up to 260% higher were found when processing i n the ho r i z o n t a l unconstrained o r i e n t a t i o n at temperatures ranging from 105-120°C and steam/air r a t i o s above 65%. The i n s u l a t i n g behavior and ultimate decrease i n heat transfer rates has made i t necessary to provide guidelines regarding allowable volumes of a i r . Predictable heat trans f e r must be ensured for adequate processing of foods. Although government agencies have not imposed regulations, some good manufacturing pra c t i c e s p e c i f i c a t i o n s e x i s t . Tsutsumi (197A) suggested that 10 mL of entrapped a i r should be general-l y accepted. This l e v e l gave reasonable assurance of predictable heat transfer and prevented pouch bursting. In a report prepared by Carapden Food Preservation Research Association ( M i l l e v i l l e and Badenhop, 1980), a maximum a i r content a f t e r s e a l i n g of 2% of product volume was recommended for use i n England. Reviewers from the United States added a d d i t i o n a l comments, suggesting that t h i s l e v e l was too low and that a 10 mL s p e c i f i c a t i o n i n a s i n g l e serving s i z e pouch was more r e a l i s t i c . The current MRE program and r e t a i l processors have adopted a 10 mL allowable r e s i d u a l gas l e v e l . I t was suggested by both the English authors and American reviewers that measurement of r e s i d u a l gases should -21-be a routine i n process q u a l i t y c o n t r o l procedure. c. Methods of Measurement The most commonly used procedure to evaluate amounts of en-trapped gases i s a destructive t e s t . Advantages of t h i s technique are that i t i s simple to execute and equipment used i s r e a d i l y a v a i l a b l e . Shappee and Werkowski (1972) described i n d e t a i l the technique and apparatus used. An opening i n the package was made when i t was t o t a l l y immersed i n water and the a i r was allowed to escape and be c o l l e c t e d into a measuring c y l i n d e r , by dis p l a c i n g the water present. Since the water l e v e l below the c o l l e c t e d gas within the cy l i n d e r was usually above that of the surrounding tank, a correction was applied to the gas volume using Boyles law. Although t h i s t e s t was simple and standard, the destructive nature was a disadvantage when multiple samples were to be evaluated. Non-destructive a i r measurement would be useful i n t h i s respect and some procedures have been investigated. A non-destructive a i r measurement proposed by Shappee and Werkowski (1972) was derived from Archimedes' p r i n c i p l e and Boyles law. Method-ology involved weighing a package while suspended i n water and then reducing the pressure i n a surrounding vacuum chamber u n t i l the entrap-ped gases expanded, r e n d e r i n g the pouch i n t o a s t a t e of n e u t r a l buoyancy. At t h i s state, the pouch neither f l o a t e d to the top nor sank to the bottom. The following equation was used to c a l c u l a t e the volume of a i r i n the pouch at atmospheric pressure: PN(WWS> V = w w s ... (2) P1" PN -22-where Pj was the atmospheric pressure at the time of t e s t i n g , P j^ was the pressure when the package was i n a state of neutral buoyancy and Wwg was the weight i n water of the package at atmospheric pressure. Comparisons between destructive and non-destructive a i r measurements were performed and a l i n e a r r e l a t i o n s h i p was found with a c o e f f i c i e n t of determination of 0.996. However, actual a i r volumes were s l i g h t l y higher than c a l c -ulated volumes. This was a t t r i b u t e d to differences i n d a i l y atmospheric pressure and a d d i t i o n a l pressure due to the water head i n the non-destructive method apparatus having a compressive action on the package volume. Research by Yaraaguchi et a l . (1972) determined a close c o r r e l a t i o n between the same non-destructive measurement and t r a d i t i o n a l destructive methods. However, vaporization of dissolved gases i n foods appeared to a f f e c t vacuumization operations for accurate determination of pressure at neutral buoyancy. Ghosh and R i z v i (1982) reported r e s u l t s suggesting that non-destructive a i r measurements were e s s e n t i a l l y higher than found with destructive methodology. They a t t r i b u t e d d ifferences to d i f f i c u l t -i e s i n measuring a t r u e i n t e r n a l p r e s s u r e at the p o i n t of n e u t r a l buoyancy. Since the pouch was not i d e a l l y f l e x i b l e , the i n t e r n a l press-ure may not be equivalent to the environmental pressure. A multiple r e g r e s s i o n a n a l y s i s was used to d e s c r i b e the r e l a t i o n s h i p between i n t e r n a l pouch pressure and pressure of the system at neutral buoyancy. Correction factors were determined to obtain an improved estimate of r e s i d u a l a i r . Application of non-destructive techniques, based on the p r i n c i p l e of neutral buoyancy, i s l i m i t e d to packages which sink i n water. Con--23-s i d e r i n g the p o s s i b i l i t y of processing products such as bakery goods, a technique was developed by Gylys and R i z v i (1983) to non-destructively measure r e s i d u a l gases i n products which f l o a t i n water. The method involved placing a pouch i n a container f i t t e d with a buret on top. Water was added to a l e v e l i n the buret. Then the e n t i r e apparatus was placed i n a vacuum chamber. Two d i f f e r e n t vacuum l e v e l s were applied and volume differences i n water l e v e l s were recorded. Boyles law was used to c a l c u l a t e the amount of gas entrapped. R i z v i and Klemaszewski (1985) further investigated t h i s technique and measured actual i n t e r n a l pressures within the pouch. Equations were developed to c a l c u l a t e the amount of a i r included as corrected to atmospheric pressure. They also determined t h a t changes i n volume r e g i s t e r e d on the buret must be corrected f or expansion of dissolved a i r i n the water. Another tech-nique was described by Huerta-Espinosa (1981). Knowledge of product density and volume of the pouch was required. The t e s t was performed by weighing a pouch while suspended i n water, then weighing the pouch i n a i r . Archimedes' p r i n c i p l e was used to c a l c u l a t e the volume of entrap-ped gases. d. Other E f f e c t s of Residual Gases Entrapped gases, i n add i t i o n to causing deleterious e f f e c t s on heat t r a n s f e r , w i l l a f f e c t product s t a b i l i t y and q u a l i t y . These factors must be taken i n t o account when deciding upon r e s i d u a l gas l e v e l s for q u a l i t y c o n t r o l . Analysis of headspace gases i n a pouch containing hushed tomato gravy was performed by Yamaguchi et a l . (1972). Generally, gases i n the headspace a f t e r processing consisted of 89% 10% CO2, 0.6% O2 and 0.2% R^. Oxygen i n the headspace was of p a r t i c u l a r -24-concern because many oxidative d e t e r i o r a t i v e reactions e x i s t i n foods. Tsutsumi (1974) stated that about 10 mL of remaining a i r was generally accepted with regard to oxidation. One of the reactions occurring i n foods i s the peroxidation of l i p i d s leading to oxidative r a n c i d i t y . Aluminum i n the r e t o r t pouch material provides a good b a r r i e r to i n t r u s i o n of oxygen from the atmos-phere. Therefore, i t may be assumed that the t o t a l amount of oxygen i s f i n i t e and the extent of the reaction cannot exceed exhaustion (Karel, 1974). Other food components such as vitamins, pigments and some amino acids are also oxygen s e n s i t i v e . Studies by Toyo Seikan Kaisha Ltd. (1973) revealed that vitamin C retention j u s t a f t e r s t e r i l i z a t i o n depended on i n i t i a l a i r content i n the package. Occluded a i r volumes of 20 mL resulted i n 75.7% retention of vitamin C, compared to 83.7, 92.8 and 95.5% retained with 10, 5 and 0 mL of a i r , r e s p e c t i v e l y . Pereira (1980) investigated vitamin C retention i n cherry pie f i l l i n g thermally processed i n r e t o r t pouches packaged under high and low vacuum conditions. Although measurement of re s i d u a l gases was not performed, i t may be assumed that high vacuum conditions created an environment with lower a i r l e v e l s . Results i n d i -cated that higher i n t e r n a l vacuum p r i o r to sealing resulted i n a small but s i g n i f i c a n t increase i n vitamin C retention (measurements taken one week a f t e r processing). Ascorbic a c i d contents of French green beans thermally processed i n i n s t i t u t i o n a l pouches were evaluated by Huerta-Espinosa (1981). Observations of increased ascorbic a c i d content with increasing amounts of entrapped a i r were contrary to findings i n other studies and to the author's suggestion that r e s i d u a l gases be minimized. -25-The same author studied other indices of q u a l i t y s e n s i t i v e to entrapped a i r . Color measurements were performed on diced pears and green beans processed with variable a i r contents. Increased a i r contents were shown to have a s i g n i f i c a n t e f f e c t on color of diced pears; Hunter L and b values were reduced and a dramatically darker o v e r a l l appearance was detected. Darker co l o r s were explained by a reduction i n t o t a l phen-o l i c s , implying the formation of polymeric compounds responsible for browning. Hunter r e s u l t s for green beans showed increased lightness and decreased greenness. Loss of color pigment i n green beans was due to the loss of magnesium ions from the ch l o r o p h y l l molecule, thereby form-ing pheophytin which i s pale o l i v e i n c o l o r . Beverly et a l . (1980) reported oxidation which occured i n peaches a f t e r 8 months of storage. Samples with i n s u f f i c i e n t a i r removal showed severe browning. Studies have been performed comparing nutrient and q u a l i t y a t t r i b -utes of foods processed i n pouches compared to cans. Young (198A) found processing a luncheon-type ham product i n r e t o r t pouches retained 16 and 26% more thiamine than canned counterparts. Research evaluating color, f l a v o r and texture of peaches thermally processed demonstrated that q u a l i t y d e t e r i o r a t i o n during storage could be minimized by using r e t o r t pouches compared to cans (Potter et a l . , 1982). In recognizing the advantages of r e t o r t pouches, the problem of a i r i n c l u s i o n and i t s detrimental e f f e c t s must be considered. 2. Processing Media In the development of r e t o r t pouch processes, e f f o r t s were made to prevent problems associated with high i n t e r n a l pressures which occur during processing. Of p a r t i c u l a r concern were l a t t e r stages of the cook -26-cycle and during cooling because at these stages the i n t e r n a l pressure may exceed r e t o r t pressure. T y p i c a l l y , a i r overpressure i n a steam environment or overpressure with water may be employed to reduce press-ure d i f f e r e n t i a l s . Pure steam processes have been employed for heating, however, a i r overpressure must be introduced during the c o o l . Fundamental studies comparing media types were performed by Pflug (1964) and Pflug and Borrero (1967). Studies on e f f i c i e n c i e s of the recognized p o t e n t i a l media concluded that 100% steam was superior and presented fewer problems i n terms of temperature d i s t r i b u t i o n , come-up and heating rate of containers. The use of pure steam was reported by Tsutsumi (1979a) f o r high temperature-short time (HTST) and u l t r a high temperature (UHT) systems i n Japan, i n which temperatures of 135 and 150°C were employed. Since short cook times were associated with these systems, i n t e r n a l temperatures remained s u b s t a n t i a l l y lower than the r e t o r t temperature. A i r overpressure was e s s e n t i a l during cooling. Disadvantages of high temperature processes include the requirement of close monitoring of cook times to ensure desired l e t h a l i t y and the p o s s i b i l i t y of physical changes of some foods a f t e r prolonged storage. Ref r i g e r a t i o n i n the marketplace and future work i n combining temp-erature and water a c t i v i t y were suggested as remedies to the problem. Tung et a l . (1984a) studied heating rates at high temperatures. A mean f h value 8.3% higher for nylon and 5.4% higher for s i l i c o n e rubber were found i n bricks processed at 130 and 135°C compared to 105-120°C. Differences were a t t r i b u t e d to ei t h e r expansion of the t e s t brick or a reduced thermal d i f f u s i v i t y value at high temperatures. Processing of r e t o r t pouches i n conventional canning r e t o r t s was reported by Roop and -27-Nelson (1981). Successful processing of confined pouches at 115.6 C i n pure steam was achieved. Pflug (1964) reported that although water/air and steam/air were p o t e n t i a l l y more troublesome than processing with pure steam, they may be implemented for provision of overpressure conditions. These two types of media are common i n Japan, North America and Europe. Studies have been performed comparing media types, yet no conclusive recommend-a t i o n can be made because each medium type has been shown to have c e r t a i n advantages and disadvantages. Pflug and Borrero (1967) reported that, with s u f f i c i e n t c i r c u l a t i o n , steam/air r a t i o s of 75% or greater provided more uniform temperature d i s t r i b u t i o n than an equivalent water cook with 260 kPa t o t a l pressure. Advantages of steam/air processes were c i t e d by a r e t o r t company chief engineer as providing energy sav-ings and uniform heat d i s t r i b u t i o n (Anon, 1983). Tung and Smith (1980) described water immersion processes as requiring high water use and a great d e a l of energy. Surface heat t r a n s f e r was thought to be p o t e n t i a l l y lower with water cooks than with steam/air since no phase change was involved i n heat t r a n s f e r . In steam/air processes a high surface heat trans f e r e x i s t s when steam condenses on the package and releases the l a t e n t heat of vaporization. A comparison of steam/air and water/air was performed by Yamano (1976). Heating rates of 30 g pouches containing 25% bentonite cake were evaluated. I t was apparent that very long come-up times were associated with water processes. When the water r e t o r t reached temp-erature, f ^ values of 8.2 min were observed. Samples processed with steam/air exhibited a heating rate index of 10.2 min. The r e s u l t s -28-indicated that water may be an e f f e c t i v e medium, however, come-up times were" much fa s t e r when employing steam/air mixtures. Although i t was determined that use of steam/air was f e a s i b l e , and perhaps cheaper, many researchers have recommended water/air processes. Beverly (1979) advo-cated the use of hot water with overriding a i r pressure because of higher heat trans f e r c o e f f i c i e n t s and ease of c o n t r o l . O v e r a l l heat transfer c o e f f i c i e n t s of heating media were reported by both Beverly (1979) and Lopez (1981) based on t h e o r e t i c a l c a l c u l a t i o n s by Pflug (1964). C o e f f i c i e n t s f o r steam, water and 75% steam/ a i r mixtures were found to be 965, 595, and 497 W/m C r e s p e c t i v e l y . Overall heat transfer c o e f f i c i e n t s dropped s i g n i f i c a n t l y as the percentage of a i r i n the mixture increased. In terms of process times required with d i f f e r e n t heating media, Beverly (1980) reported a longer processing time i n steam/air than with water for a 280 g product to reach a target F Q of 6 min. Steam/air mixtures were reported by Wilson (1980) as being d i f f i c u l t to employ with the necessity of temperature and pressure c o n t r o l , along with adequate mixing. In water/air systems, although the r e t o r t response was slow, heat t r a n s f e r r a t e s to the c o n t a i n e r were g e n e r a l l y b e t t e r . Research by Kopetz et a l . (1979) i n d i c a t e d a 15-16% i n c r e a s e i n processing time with steam/air compared to water systems. For t h i s reason, they concluded that water/air was more e f f i c i e n t . Tsutsumi (1979b) reported that water processes were commonly used, p a r t i c u l a r l y for pouches containing large amounts of entrapped a i r , i n order to a t t a i n rapid and even heat penetration. He suggested avoiding the low steam/air r a t i o s which are necessary to provide adequate overpressure -29-f o r p r e v e n t i n g pouches from b u r s t i n g because of the p o t e n t i a l interference by a i r with heat t r a n s f e r . 3. Processing Conditions Optimum steam/air r a t i o s have been a focus of i n t e r e s t . There i s concern that processing with high a i r percentages w i l l cause deviations i n heat transfer conditions due to the low s p e c i f i c heat capacity of a i r and the absence of a condensation phase change as with steam. For example, the o v e r a l l heat transfer c o e f f i c i e n t of a i r i s 16.8 W/m C compared to 965 W/m C for pure steam. Increasing steam/air r a t i o s from 75 to 95% were r e p o r t e d by P f l u g and B o r r e r o (1967) to improve temperature uniformity and reduce the duration of temperature drops when a i r was introduced a f t e r come-up. An i n v e s t i g a t i o n by Pflug et a l . (1963) revealed that the f n of pouches heated at 240°F i n 90% steam did not d i f f e r from f ^ values for pouches processed with 75% steam at the same temperature. R e s u l t s by Yamano and Komatsu (1969) suggested steam/air mixtures over 70% were p r a c t i c a l f o r use. Heating rates and temperature d i s t r i b u t i o n were found to be almost independent of steam/air composition above 70%. These findings were further supported by work of Toyo Seikan Kaisha Ltd. (1973). Their r e s u l t s showed that decreasing steam/air r a t i o s from 100 to approximately 60 or 70% had l i t t l e e f f e c t on heating rates. However, reduction below 60% steam increased values of the heating rate index dramatically. Heat penetration studies by Tung et a l . (1984a) showed that f h values increased up to 11% by decreasing the steam content from 100 to 50% i n a p i l o t scale p o s i t i v e flow r e t o r t . Conversely, i n a p i l o t scale -30-commercial r e t o r t (Lagarde) steam content was found to be non-s i g n i f i c a n t over that range of composition. Properties of steam/air media were characterized by comparing surface heat trans f e r c o e f f i c i e n t s i n s t u d i e s by Tung et a l . (1984b) u s i n g methodology d e s c r i b e d by Ramaswamy et a l . (1983). Surface heat transfer c o e f f i c i e n t s i n both p i l o t scale r e t o r t s were found to increase i n an exponential fashion with increasing steam contents. C o e f f i c i e n t s i n the order of 2,000 and 12,000 W/m C were found for 50 and 100% steam, r e s p e c t i v e l y . Tung et a l . (1984b) pointed out that the surface heat transfer c o e f f i c i e n t gave a valuable estimate of media heating p o t e n t i a l , but the o v e r a l l heat transfer and heat rate indices would provide p r a c t i c a l information regarding heating of foodstuffs. Limited studies have examined e f f i c i e n c i e s of water/air cooks with variable overpressure conditions. Preliminary i n v e s t i g a t i o n s by Kopetz et a l . (1979) i n d i c a t e d t h a t o v e r r i d i n g a i r pres s u r e i n w a t e r / a i r systems helped to maintain heat penetration rates. Their l i m i t e d data showed a 6-12% reduction i n process time when overriding a i r pressure was increased from 69 to 172 kPa (gauge). Pereira (1980) studied the e f f e c t s of water with superimposed a i r pressure on processing time r e q u i r e d to reduce p e r o x i d a s e a c t i v i t y . A 2.85 D process with an experimentally derived z-value were evaluated f o r cherry pie f i l l i n g . Pressure was found to be a s i g n i f i c a n t f a c t o r . Generally, pouches pro-cessed under 103 kPa of superimposed a i r pressure required l e s s time f o r s t e r i l i z a t i o n than when processed with no overriding pressure, with some exceptions. Water c i r c u l a t i o n v e l o c i t y was found by Peterson and Adams (1983) to be an important factor when employing water/air processes. An -31-increase i n process time of 9.6% was found from high (6.9 L/s) to low (0.6 L/s) flow rates f o r 10% bentonite suspensions i n i n s t i t u t i o n s i z e r e t o r t pouches processed at 250°F with 69 kPa overpressure. V a r i a b l e temperature of the p r o c e s s i n g media was s t u d i e d by Ramaswamy (1983). Temperature was found not to be a s i g n i f i c a n t factor for the heating rate of s i l i c o n e rubber bricks processed i n a p o s i t i v e flow r e t o r t with steam/air media. However, nylon b r i c k s showed a s i g n i f i c a n t difference (p<0.05) i n the heating rate i n d i c e s . A 5.5% higher mean f ^ was found at 120°C (23.48 min) compared to a 105°C (22.25 min). No s i g n i f i c a n t differences (p>0.05) were found i n heating rates at variable temperatures with either s i l i c o n e rubber or nylon t e s t bricks when processing i n a forced c i r c u l a t i o n Lagarde p i l o t scale r e t o r t . 4. Package Thickness Retort pouch products derive t h e i r rapid heating c h a r a c t e r i s t i c s because of the t h i n p r o f i l e cross section contributing towards o v e r a l l b enefits. One fa c t o r a f f e c t i n g pouch thickness i s racking constraints. As previously discussed, confined racks w i l l not permit expansion of re s i d u a l gases and i t was possible to use 100% steam f o r pouch process-ing (Roop and Nelson, 1981). In confined pouches, the rack spacing establishes a maximum thickness that the package w i l l a t t a i n during processing. Improper confinement due to inadequate c i r c u l a t i o n of pro-cess media between layers may be detrimental to the heating rate of the product. The expansion of pouch contents ( p a r t i c u l a r l y r e s i d u a l gases) and f i l l weights are important factors when u t i l i z i n g unconfined racking -32-designs. Expansion of entrapped gases has been discussed previously. F i l l weights are c r i t i c a l because of ultimate differences in pouch thickness. Berry and Kornhorst (1983) found decreased heating rates (higher f^ values) with an increase in f i l l weight when processing cream of celery soup and kernel corn in brine. Effects were more pronounced for the soup, since natural convection currents were enhanced with large f i l l weights of corn i n brine. Beverly et a l . (1980) developed a relationship in order to determine f^ values for several pouch thick-nesses. For the product tested, f^ values for thicknesses of 0.59, 0.62, and 0.75 inches were calculated to be 8.A, 9.3 and 13.6 minutes, respectively. Heating rate indices can be compared for various thick-nesses by multiplying the f^ value by the square of the appropriate ratio of thickness (Ramaswamy, 1983). -33-EXPERIMENTAL I Sample Preparation 1. Test Bricks Experiments u t i l i z i n g model systems have proved valuable i n compar-ing factors a f f e c t i n g thermal processing of foodstuffs. A wide variety of materials has been implemented, f o r example, bentonite suspensions (Pflug, 1964; Peterson and Adams, 1983) and bricks of metal or nylon (Ramaswamy, 1983). Test bricks were fabricated from v i r g i n t e f l o n ( C a d i l l a c P l a s t i c s , Montreal, PQ) for use i n experimental phases. The bricks consisted of two t e f l o n slabs. One slab was channelled to allow for placement of a te f l o n - i n s u l a t e d 24 AWG copper/constantan thermo-couple with a fused end placed at the ce n t r a l l o c a t i o n . Each pair of slabs was cemented together with a f l e x i b l e p l a s t i c coating ( P l a s t i d i p , PDI Inc., Northwood, CA). The groove was f i l l e d i n s u f f i c i e n t l y with the material to prevent thermocouple movement and steam or water from entering. Four machine screws at 5.5 cm away from the centerpoint held the brick together. Additional screws were required around outer edges to ensure that the brick remained sandwiched together, since t e f l o n warped s l i g h t l y with repeated processing. Two brick thicknesses were studied and w i l l be referred to i n future discussions as thick or t h i n . F i n a l dimensions were: t h i c k b r i c k s , 2.1 cm t h i c k , 11.1 cm wide, 15.0 cm long; t h i n b r i c k s , 1.3 cm thi c k , 12.1 cm wide and 15.0 cm long. Using t e f l o n b r i c k s has many advantages. I t heats by conduction with a thermal d i f f u s i v i t y s i m i l a r to many food products. Table 1 l i s t s p h ysical properties as described by Mantell (1958) and a calculated thermal d i f f u s i v i t y which i s comparable to l i t e r a t u r e values as c i t e d by -34-TABLE 1 Thermophysical properties of t e f l o n . Properties of t e f l o n (Mantell, 1958) S p e c i f i c Gravity (p) 2.1 - 2.3 g/cm S p e c i f i c Volume (s.v.) 475.9 cm /kg Thermal Conductivity (k) 6 x 10" 4 c a l / s cm C° Thermal Expansion (TE) 10 _ 5/C° ( l i n e a r ) S p e c i f i c Heat Capacity (Cp) 0.25 cal/g C° Water Absorption (WA) 0 C a l c u l a t i o n of thermal d i f f u s i v i t y (Loncin and Merson, 1979) k 6 x 10~ 4 c a l / s cm C° a = p Cp 2.2 g cm • 0-25 cal/g C° = 1.09 x 10~ 7 m2/s -35-Tung et a l . (1984a). Another advantage i s that i t may be assumed there are no thermal degradations or permanent modifications of properties at the high temperatures used with r e t o r t processing. Importantly, t e f l o n bricks may be fabricated i n an i d e n t i c a l manner and be used for numerous t r i a l s . This provided consistent and repeatable measurements, thereby reducing v a r i a t i o n s due to sample c h a r a c t e r i s t i c s . 2. Packaging Retort pouches (American Can Co., Neenah, WI; 12 ym Polyester/9 ym Al f o i l / 7 6 ym Polypropylene) of 16.0 x 23.1 cm outside dimensions and 14.4 x 20.5 cm inside (taking seal into consideration) were used. A droplet (approximately 2.5 mL) of s i l i c o n e sealant (Dow Corning Corp., Midland, MI)) was applied i n a lower corner of the pouch and allowed to cure for at l e a s t eight hours. This functioned as a septum through which a i r was inje c t e d and with removal of the syringe needle, an a i r -t i g h t seal reformed. Pouches were f i t t e d with packaging glands (O.F. Ecklund, Cape Coral, FL). Use of extra gaskets sealing the s t u f f i n g box were nec-essary to prevent gases from escaping through the f i x t u r e . In order to ensure an a i r t i g h t s e a l , tightening was required on the screw f i t t i n g s u n t i l f i r s t signs of gasket deformation appeared. Pouches were prepared with about 25 cm of thermocouple wire extending from the pouch and equipped with connectors (Omega Engineering, Inc., Stamford, CT), since occasionally repackaging was required due to leakage of a i r i n t o the pouch. This allowed easy removal of the brick and f a c i l i t a t e d immediate repackaging into previously prepared pouches. A cotton b a l l was placed i n s i d e the pouch, d i r e c t l y below the septum to aid i n j e c t i o n . -36-Each brick was placed i n a pouch and connected to an extension thermocouple. Water (10 mL) was added to the pouch to provide vapor pressure simulating a food material. Since there was no absorption of water by t e f l o n , as indicated i n Table 1, water was assumed to be t o t a l -l y free and e x h i b i t i n g a vapor pressure s i m i l a r to foods with a water a c t i v i t y approaching 1.0. Samples were vacuum packaged to remove as much a i r as possible. For some t r i a l s , a Multivac (Model AG-5, Algau, W. Germany) equipped with a narrow bar impulse sealer was employed at the maximum vacuum s e t t i n g . To ensure an adequate seal f or r e t o r t processing, an a d d i t i o n -a l seal was formed using a Sentinel sealer (Packages Industries Group, Inc., Hyannis, MA) at operating conditions of: a i r l i n e pressure 375 kPa corresponding to a jaw pressure of 275 kPa and temperature s e t t i n g of 227°C. Volumes of a i r remaining a f t e r vacuum packaging were l e s s than 5.0 mL. In other t r i a l s , a mechanical vacuum Swiss-Vac machine located at Magic Pantry Foods Inc. (Hamilton, ON.) was used. I t was a produc-t i o n model for s o l i d f i l l foods with speeds rated at 30 pouches per minute (Morris, 1981). Pouch i n t e g r i t y was tested by allowing 3-8 h to elapse with no appearance of a i r within the pouch. Leakage could be detected v i s u a l l y since the pouch was packaged t i g h t l y against the b r i c k . I f signs of leakage were noticed, repackaging was necessary u n t i l an a i r t i g h t sample was ensured. -37-II Air Measurement 1. Non-destructive Measurement of Air A non-destructive measurement was useful in these experiments to determine air volumes within the pouch prior to processing. Shappee and Werkowski (1972), and Rizvi and Gylys (1983) described a method of non-destructive measurement as discussed earlier. Procedures involved determining the vacuum pressure within a bell jar necessary for the pouch to achieve neutral buoyancy in water. This method was attempted with results showing inconsistencies and lack of reproducibility. That approach was abandoned and a modified technique was employed to monitor addition of air and possible leakage of gases, into or out of the pouch. It was performed by weighing a vacuum packaged pouch while suspended in water. Figure 1 illustrates the apparatus used for this purpose. A scale equipped with a hook on the underside for weighing, mounted on a wooden base with a c i r c u l a r hole and placed over a water basin i s depicted. There was clearance between the wooden base and water basin. Pouches were suspended from the hook by use of a c l i p and t o t a l l y immersed in water. Water level was kept constant and the scale was zeroed with the c l i p on. Once each package was weighed, desired volumes of air were care-ful l y injected through the silicone septum. In each set of t r i a l s , target air volumes were administered. For 14 bricks of each thickness, Figure 1 . Apparatus for non-destructive measurement of a i r . -39-two bricks were l e f t unpackaged and three bricks were injected with volumes of 0, 10, 20 and 30 mL of a i r . Each brick was i d e n t i f i e d by a number, and the bricks were randomly assigned to the various t e s t con-d i t i o n s . After a i r addition, samples were weighed i n water again. Based on Archimedes' p r i n c i p l e , the difference i n weight was equal to the volume of a i r added from the equation: Wws ~ Wws+a = Pw Vs§ + Pw vl8 - P w V sg ...(4) Am = p wV! where Wwg i s weight of the sample i n water and W w g + a i s weight of the sample and added a i r i n water; p w the density of water; V s i s the volume of the sample; Vj i s the volume of a i r ; and g i s the g r a v i t a t i o n -a l force. Derivation of t h i s equation i s shown i n Appendix I. Assuming the density of water equal to 1.0 g/mL, for each mL of a i r added, the mass of the sample and a i r w i l l be reduced by 1.0 g. With t h i s assumption, addition of a i r and any leakage may be monitored. This technique was used only as a r e l a t i v e measure of a i r volumes. The measurements were performed i n three stages: 1. a f t e r vacuum packaging, 2. a f t e r i n j e c t i n g desired volumes of a i r , just p r i o r to processing, and 3. a f t e r processing, just p r i o r to destructive a i r measurement. The difference between measurement two and one, represents the amount of a i r added by i n j e c t i o n . Possible a i r leakage during processing may be calculated by the difference of measurements three and two. Any pouches -40-with leakage greater than 5 mL were omitted. 2. Destructive Measurement of A i r Conventional a i r measurement i s a procedure whereby the pouch i s opened under water and gases are c o l l e c t e d i n a volumetric vessel. Since pouch i n t e g r i t y i s l o s t and the a i r i s discharged from the pouch, such methods are generally termed "destructive". Shappee and Werkowski (1972) described the method used i n t h i s experiment. Figure 2 depicts equipment assembled to measure entrapped a i r . A funnel was attached to the mouth of a 50 mL buret to increase the area for entry of gases. As i l l u s t r a t e d , the buret was inverted and placed i n a water basin. Water was f i l l e d to the 50 mL measure by suction with an attached rubber tube. A constant column height above the water l e v e l was ensured by con-s i s t e n t l y immersing to a marking on the funnel. Testing was performed by c a r e f u l l y immersing the pouch preventing any accumulation of a i r bubbles on outside surfaces, then an opening was cut i n a corner of the pouch while d i r e c t l y under the funnel. Entrapped gases were allowed to escape and be c o l l e c t e d . Care was taken to prevent gases from escaping beyond the funnel area. Volumes of gas were measured by displacement of water. A c o r r e c t i o n f o r pressure d i f f e r e n t i a l due to water height was necessary using the r e l a t i o n s h i p : ( pl- Wh> V, P l m where Vj i s the volume of a i r at atmospheric pressure P^. The pressure due to the water l e v e l i n the buret (W^) was calculated from: Figure 2. Apparatus for destructive measurement of a i r . -42-W h - \ " 0.05) was found i n heating rates for variables of temperature and steam per-centage (pressure l e v e l ) . This i s consistent with the findings of Ramaswamy (1983) where s i m i l a r f a c t o r s were found n o n - s i g n i f i c a n t (p>0.05) for nylon bricks i n both v e r t i c a l p o s i t i v e flow and Lagarde r e t o r t s . This f i n d i n g demonstrates that a l l three pressure l e v e l s and temperatures have the same e f f i c i e n c y of heating on an unpackaged sample which excludes the p o t e n t i a l e f f e c t s of the pouch and r e s i d u a l gases. A s i g n i f i c a n t d i f f e r e n c e was seen i n the data from d i f f e r e n t r e t o r t s . Duncan's multiple range t e s t was performed and r e s u l t s are reported i n Table 4. In regard to t h i n bricks, heating rates i n the Lagarde r e t o r t were s i g n i f i c a n t l y faster (low f ^ values) than were evident with e i t h e r water/air or steam/air i n the p i l o t scale r e t o r t . A l l r e t o r t s were found to be s i g n i f i c a n t l y d i f f e r e n t from each other when processing thick b r i c k s . Highest mean f ^ values were found i n the p o s i t i v e flow r e t o r t employing steam/air. Water/air media i n the same r e t o r t yielded s l i g h t l y lower heating rate in d i c e s , and values found from processing i n the Lagarde system were lowest. I t i s evident from these r e s u l t s that processing of unpackaged bricks i n the Lagarde system may provide more e f f i c i e n t heat transfer than water/air or steam/air i n v e r t i c a l p o s i t i v e flow systems. -57-TABLE 3. Analysis of variance f or heating rate indices of unpackaged t e f l o n b r i c k s . Source of V a r i a t i o n Thin Thick df F-Ratio df F - r a t i o Retorts 2 6.46** 2 20.94** Temperature 2 1.57 ns 2 0.15 ns Steam %-Pressure l e v e l 2 1.55 ns 2 0.47 ns Interactions Retorts - Temperature 4 1.11 ns 4 0.07 ns Retorts - Steam 4 0.38 ns 4 0.71 ns Temperature - Steam 4 0.71 ns 4 1.04 ns Error 81 81 ns not s i g n i f i c a n t (p>0.05) ** s i g n i f i c a n t at p<0.01 -58-TABLE 4 Duncan's multiple range te s t f o r heating rates of unpackaged bricks i n d i f f e r e n t r e t o r t s . Thin Bricks Thick Bricks Retort f ^ (min) Duncan Test''' f ^ (min) Duncan Test'' mean (std. dev.) mean (std. dev.) Lagarde 6.08 (0.31) 17.78 (0.80) Steam/Air 6.42 (0.51) 19.24 (1.09) Water/Air 6.46 (0.65) 18.32 (0.83) Values with the same l e t t e r within a column are not s i g n i f i c a n t l y d i f f e r e n t (p>0.05) -59-P r e v i o u s s t u d i e s have found l a r g e v a r i a b i l i t y i n f ^ v a l u e s . Ramaswamy (1983) found c o e f f i c i e n t s of v a r i a t i o n f o r h e a t i n g r a t e indices of unpackaged bricks ranging from A.5-5.1% for nylon and 6.A% for s i l i c o n e rubber. Heating rate indices f o r pouches containing whole kernel corn i n brine with e s s e n t i a l l y no r e s i d u a l a i r were reported by Berry and Kohnhorst (1983) with c o e f f i c i e n t s of v a r i a t i o n of 9.9%. As indicated i n Table 5, s i m i l a r c o e f f i c i e n t s of v a r i a t i o n i n sets of data from each r e t o r t vary from 5-10% for thi n bricks and A.5-5.7% for thick b r i c k s . In contrast, experiments by Patino and H e i l (1985) on heat penetration with potatoes i n brine (no volumes of a i r reported) showed c o e f f i c i e n t s of v a r i a t i o n i n the order of 15.9-25.5%. A value of thermal d i f f u s i v i t y (a) may be calculated from measured heating rate indices and brick geometry by the r e l a t i o n s h i p reported by Olson and Jackson (19A2): whereby a, b and c are one- h a l f b r i c k t h i c k n e s s , l e n g t h and width r e s p e c t i v e l y . The average calculated thermal d i f f u s i v i t i e s were 8.82 x 10 m /s for thick and 1.02 x 10 m /s for thin bricks compared to —7 9 1.09 x 10 m /s calculated from thermophysical properties of t e f l o n i n Table 1. Differences may be due to the composition of t e f l o n from which the bricks were made or possible e f f e c t s of the p l a s t i c rubber compound used to secure thermocouples at brick center. a = .933 (11) -60-TABLE 5. C o e f f i c i e n t s of v a r i a t i o n f o r heating rate indices of unpackaged b r i c k s . Brick Thickness Retort C o e f f i c i e n t of V a r i a t i o n , % (std. dev./mean) x 100 Thin P o s i t i v e flow steam/air 8.0 P o s i t i v e flow water/air 10.0 Lagarde 5.2 Thick P o s i t i v e flow steam/air 5.7 P o s i t i v e flow water/air A.5 Lagarde 4.5 -61-III Packaged Bricks: Included Air Heating rate indices ( f ^ , min) were found for a l l t e s t packages containing a i r processed i n r e t o r t conditions as previously described. Figures 3 and 4 are examples of data c o l l e c t e d from t h i n and thick bricks processed i n the Lagarde r e t o r t . A l l other r e l a t i o n s h i p s are i l l u s t r a t e d i n Appendix III (a, b, c and d). By observation of these fi g u r e s , a general trend i n the data was evident. I t appears that some conditions, p a r t i c u l a r l y those of 65% steam (pressure l e v e l 1) showed no e f f e c t of included a i r volumes on heating rate i n d i c e s . These r e l a t i o n -s h i p s form e s s e n t i a l l y a h o r i z o n t a l l i n e . Data f o r some other conditions (usually 75% steam, pressure l e v e l 2) appeared to consist of two sections; one being a hori z o n t a l l i n e s i m i l a r to the r e l a t i o n s h i p described above and a second part demonstrating an increase i n f ^ as a i r volumes increase. This trend was apparent i n many conditions. Another type of r e l a t i o n s h i p was observed f o r conditions of 115°C and 85% steam (pressure l e v e l 3) with thick b r i c k s . There i s no apparent section of plateau and f ^ values increased at very low l e v e l s of included a i r . Therefore, the r e l a t i o n s h i p of f ^ as a function of a i r content can be described as containing either one or both sections of a two part model. One part e x i s t s as a plateau demonstrating that a i r volumes have no e f f e c t on f ^ ; the other part describes the changing e f f e c t of a i r v o l -umes by increasing f ^ values with increased a i r contents. These two d i s t i n c t behavior patterns may be explained by the nature of the model system used. A plateau may e x i s t when void areas around the brick periphery are f i l l e d with a i r , but an a i r layer has not formed on the bric k surface so heat transfer i s maintained. Heating rate indices - 6 2 -16 -15-14-13-12-11-10-» H 8-7-6-L e g e n d A 65% steom X 755! steom • 85% steom *x X X n x • A i X D A • x » D 4 _ c o 01 X 16-15-14-13-12-11-10-9 H 8-7 - ; to 6-16-15-14-13-12-11-10-s -8 -7-6-5-L e g e n d A 65% steom x 75% tliom D 85% steam • x 4 D X ^ A x x B x • • x * X A A A X A ^ A A L e g e n d A 65% stiam X 75% itiom • 85% steam A A A X A A A X A x 4 10 IS 20 25 Included A i r (mL) — i — 30 35 Figure 3. Heating rate index as a function of included air; Lagarde retort, thin bricks, (a, 115°C; b, 120°C; c, 125°C) -63-3 4 3 3 3 2 3 1 3 0 2 9 2 8 2 7 2 6 2 5 2 4 2 3 2 2 2 1 2 0 1 9 1 8 31 3 0 2 9 I " 2 E 2 5 2 4 2 3 C 2 2 S 2 1 1 2 0 1 9 1 8 x 0) "D C £ "5 3 1 3 0 -2 9 -2 8 -2 7 -2 6 2 5 2 4 -2 3 2 2 -2 1 2 0 1 9 -1 8 1 7 L e g e n d /} 653 steom x 75X sleom • 853 sleom D A O • O D • X x x x * X x x V A x L e g e n d A 653 steom x 753 steom D 85% iltom 0 D A x L e g e n d £ 653 steom X 753 steom • 853 steam D 8 X A K> 1 5 2 0 2 5 Included A i r (mL) i 3 0 3 5 Figure A. Heating rate index as a function of included a i r ; Lagarde retort, thick bricks, (a, 115°C; b, 120°C; c, 125°C) -64-begin to increase once the outer void locations have been f i l l e d and a layer of a i r begins to form over the brick face. Differences between conditions may be due to the volumes of a i r at which a t r a n s i t i o n from one part to the other occurs. In conditions described as a horizontal l i n e , the t r a n s i t i o n or break-point would occur at volumes of a i r larger than the maximum tested. Cases where f n increases at very low a i r v o l -umes demonstrate a very small or imperceptible plateau, so no apparent break-point e x i s t s . Due to the nature of the data, comparisons between variables were performed i n two ways. F i r s t , volumes of a i r at which a t r a n s i t i o n from the plateau region to an upward trend ( i f i t ex i s t s ) were determined. Secondly, comparisons were made between conditions where appreciable data existed i n the upward section or part two of the model. IV Break-point Values 1. Determination of Break-point In a l l r e t o r t s , the f ^ vs. included a i r r e l a t i o n s h i p s at 125°C and 65% steam or pressure l e v e l 1 exhibited a horizontal l i n e . This may be expected since these conditions represent the highest r e t o r t pressure tested. Due to f l e x i b i l i t y of the pouch, i t has been shown to respond to t h e o r e t i c a l pressure volume re l a t i o n s h i p s (Nelson et a l . , 1956). I t i s c l e a r from the data that conditions of 125°C and 65% steam (pressure l e v e l 1) l i m i t pouch expansion so no r e s t r i c t i o n on heat t r a n s f e r occurred i n the range of a i r volumes tested. A t e s t on s i g n i f i c a n c e of the regression was performed to s t a t -i s t i c a l l y determine i f a hori z o n t a l l i n e e x i s t s (Zar, 1974). A n u l l hypothesis of slope equal to zero was tested using a Student's t - t e s t . -65-Results of t h i s analysis i n Table 6 indicated that i n a l l r e t o r t s and for each thickness a l l l i n e s were e s s e n t i a l l y horizontal (p>0.05). Based on t h i s f i n d i n g , t h e h o r i z o n t a l l i n e of 125°C and 65% steam/pressure l e v e l 1 was used as a baseline for means of comparison. A Student's t - t e s t was also used to e s t a b l i s h i f a difference e x i s t s between packaged and unpackaged br i c k s . Since no r e l a t i o n s h i p e x i s t s between f n as a f u n c t i o n of i n c l u d e d a i r f o r the b a s e l i n e r e l a t i o n s h i p these values were compared against bare brick f n values. Results i n Table 7 show that the two means d i f f e r (p<0.05) i n f i v e out of s i x cases. No s i g n i f i c a n t difference was demonstrated i n the pos-i t i v e flow r e t o r t with t h i n b r i c k s . In a l l cases the mean f h was larger for packaged bricks than t h e i r unpackaged counterparts. Differences may be a t t r i b u t e d to an added resistance to heat transfer from the pouch and 10 mL of water added. Another possible reason may be that pouch edges extended outward and caused reduced media c i r c u l a t i o n between samples. Comparisons between regression l i n e s were performed using co-v a r i a n c e a n a l y s i s f o r a l l c o n d i t i o n s i n each r e t o r t a g a i n s t the respective baseline. Thick and t h i n bricks were analyzed separately because they had d i f f e r e n t magnitudes of baselines. Those conditions which displayed no s i g n i f i c a n t difference from the baseline i n either slope, l e v e l or o v e r a l l (p>0.05) were considered as h o r i z o n t a l l i n e s with no s t a t i s t i c a l evidence of a break-point. Conditions where a s i g n i f i c a n t difference was determined i n either slope or l e v e l (p<0.05) were considered to have an upward section ( i n a l l cases, e i t h e r slope or l e v e l were higher). -66-TABLE 6. Test f o r s i g n i f i c a n c e of slope of the baseline 125°C-65% steam with Student t - t e s t . H Q: slope =0. Brick Thickness Retort Slope (min/mL) t c a l c Thin Lagarde 0.012 0.0319 ns P o s i t i v e flow steam/air -0.0032 -0.0055 ns P o s i t i v e flow water/air 0.0089 0.0515 ns Thick Lagarde 0.0041 0.0055 ns P o s i t i v e flow steam/air 0.0056 0.0046 ns P o s i t i v e flow water/air 0.0009 0.0015 ns ns not s i g n i f i c a n t (p>0.05): accept H 0 -67-TABLE 7. Student t - t e s t on f h of unpackaged bricks compared to f h values of baseline. H Q: v i = y 2 mean f ^ , min Brick Retort unpackaged packaged t c a l c . Thin Lagarde 6.08 6.83 6.24** P o s i t i v e flow steam/air 6.42 6.71 1.91 ns P o s i t i v e flow water/air 6.46 6.95 3.97** Thick Lagarde 17.78 19.74 7.58** P o s i t i v e flow steam/air 19.24 19.49 2.56* P o s i t i v e flow water/air 18.32 20.12 6.18** ns not s i g n i f i c a n t (p>0.05) * s i g n i f i c a n t at p <0.05: r e j e c t H Q ** s i g n i f i c a n t at p <0.01: r e j e c t H Q -68-The procedure used for determining points of t r a n s i t i o n or break-points may be described as covariance analysis on a moving regression frame. This concept was previously described by Tung et a l . (1984a) for use i n a d i f f e r e n t a p p l i c a t i o n . I t was employed to determine the high-est a i r volume at which data show no s i g n i f i c a n t difference (p>0.05) i n ei t h e r slope or l e v e l from the baseline. The moving frame was performed by beginning with the l a s t ten consecutive data points of highest i n -cluded a i r volumes. Ten data points were chosen because that amount allowed for desired s e n s i t i v i t y of the t e s t , yet represented normal f l u c t u a t i o n i n the data. Covariance analysis was performed at each stage as the frame moved i n order of decreasing a i r contents u n t i l the l i n e showed no s i g n i f i c a n t difference, i n a l l respects, from the base-l i n e . Figure 5. schematically depicts how the analysis was performed. Frames i n each c o n d i t i o n were compared a g a i n s t the b a s e l i n e from respective r e t o r t s . The highest a i r volume of the f i r s t frame e x h i b i t -ing no s i g n i f i c a n t difference (p>0.05) represents the t r a n s i t i o n or breakpoint. The determined values are l i s t e d i n Tables 8 and 9. An a d d i t i o n a l comparison was performed on a l l data up to the break-point as i l l u s t r a t e d by the dashed l i n e i n Figure 5. In a l l cases, there was s t i l l no s i g n i f i c a n t difference (p>0.05) from the respective baselines. Air volumes up to the determined break-points were then considered to be the plateau region. As determined by the moving regression frame, i f a larger a i r volume was included, the r e l a t i o n s h i p became s i g n i f i c n t l y d i f f e r e n t from the baseline. These analyses were performed on a l l con-d i t i o n s with evidence of a break-point. The same method was repeated c o n s i s t e n t l y f o r a l l r e t o r t s , p r o c e s s i n g c o n d i t i o n s and b r i c k P o s i t i v e Flow S t e a m / A i r - T h i n B r i c k s B a s e l i n e C o n d i t i o n 115 C - 7 5 1 Steam A i r (mL) f. (min) n A i r (mL) f, (min) n 3.5 7.82 0.5 6.53 5.1 7.06 1.6 7.15 5-3 7-26 2.0 6.68 5.3 6.61 2.9 7.00 5.6 6.11 3.6 6.63 6.0 5-98 7.6 6.81 9.2 7.44 8.6 6.81 9-A 6.96 8.8 7.15 9.9 6.22 9. h 7.28 10.8 7.51 9.6 7.75 10.9 6.3A 10.2 6.88 18.3 6.28 15-0 8.87 18.4 6.43 16.3 8.69 18.5 6.50 16.5 o.Sk 18.8 6.07 17.3 7.35 19-3 6.13 20.5 9.37 20.7 6.13 21.1 9.31 20.8 6.79 22.0 9.9^ 21 .1 6.09 27.1 10.86 28.6 7.23 30.2 11.5^ 29.1 7.33 32.3 10.02 29.2 7.24 ns not s i g n i f i c a n t (p>0.05) * s i g n i f i c a n t a t p<0.05. ns o r d e r o f p e r f o r m a n c e F i g u r e 5- Moving r e g r e s s i o n frame p e r f o r m e d w i t h c o v a r i a n c e a n a l y s i s t o d e t e r m i n e b r e a k - p o i n t s . -70-TABLE 8. Break-point values of thi n bricks determined by covariance analysis on a moving regression frame. Retort Temperature Break-point Values (mL) °C 65% steam (1) 75% steam (2) 85% steam (3) Po s i t i v e flow steam/air 115 120 17.0 29.5 10.2 18.5 9.0 13.6 125 29.2 28.8 9.1 Po s i t i v e flow water/air 115 120 29.2 1 30.8 1 24.8 28.2 15.9 24.6 125 30.4 34.3 1 20.2 Lagarde 115 29.8 1 20.3 11.6 120 28.7 1 26.0 14.4 125 31.7 1 31.8 <5.02 ns to maximum a i r volume. sd to minimum a i r volume. -71-TABLE 9. Break-point values of thick bricks determined by covariance analysis on a moving regression frame. Break-point Values (mL) Retort Temperature °C 65% steam (1) 75% steam (2) 85% steam (3) Po s i t i v e flow steam/air 115 120 15.3 31.8 1 11.0 28.9 <5.02 9.3 125 34.5 27.3 <5.02 P o s i t i v e flow water/air 115 120 32.3 1 29.8 1 13.2 25.0 <5.02 9.3 125 31.8 35.0 1 13.3 Lagarde 115 31.6 1 16.4 <5.02 120 28.8 1 19.4 9.7 125 30.3 30.7 1 <5.02 ns to maximum a i r volume. sd to minimum a i r volume. -72-thicknesses. Since only one value of break-point may be determined, differences between values may not be s t a t i s t i c a l l y tested. Therefore, comparisons are made by differences i n magnitude of each break-point. 2. Factors A f f e c t i n g Break-Point Volumes a. Steam Percentage or Pressure Level Results shown i n Tables 8 and 9 may be studied from the stand-p o i n t of d i f f e r e n c e s i n steam percentages or p r e s s u r e l e v e l . By examination of values according to rows, the degree of overpressure may be compared for each variable of temperature, r e t o r t and brick t h i c k -ness. In a l l cases, break-point volumes of a i r increased as the degree of overpressure increased up to 65% steam/pressure l e v e l 1 i n t h i s study. I t i s apparent that at higher degrees of overpressure, larger a i r volumes may be entrapped with no interference i n heat t r a n s f e r . Results from bare bricks demonstrated no s i g n i f i c a n c e (p>0.05) i n degree of overpressures tested. Therefore, the differences observed may be a t t r i b u t e d to the influence of processing conditions on entrapped gases. As previously explained, the test samples consisted of r e c t -angular bricks packaged i n f l a t pouches. A i r entering the pouch could f i r s t accumulate around the t h i n edges of the bricks, then as volumes of a i r increase layers could form over the larger surfaces of the brick. Tables 8 and 9 ind i c a t e that almost a l l conditions with 65% steam/ pressure ( l e v e l 1) l i m i t expansion so that entrapped gases remained only i n void areas near the brick edges. Conditions of reduced overpressure allowed only small volumes of a i r to e x i s t i n peripheral brick areas because expansion during processing occurred to a larger degree. Above the break-point, a cushion of a i r may have accumulated between the large -73-faces of the brick and the pouch material, thereby impeding heat trans-f e r to the c e n t r a l l y located thermocouple. An example i n Table 8 of th i n bricks processed i n the Lagarde r e t o r t at 115°C displayed break-points of 29.8 mL (maximum a i r volume) at 65% steam, 20.3 mL for 75% and 11.6 mL for 85% steam. Thick bricks processed at 115°C and 85% steam ( l e v e l 3) showed no evidence of a plateau section i n a l l three r e t o r t s . Consequently, s u f f i c i e n t expansion may have occurred to i n t e r f e r e with heat transfer at small occluded a i r volumes (<5.0 mL) when processing with 30 kPa a i r overpressure. Results of t h i s study i n d i c a t e that when processing with lower steam r a t i o s (65% steam), heat transfer i n pouches with large amounts of a i r w i l l be maintained. Studies by Pflug and Borrero (1967) established that higher steam/air r a t i o s provided a more e f f e c t i v e medium because of increased enthalpy and heating e f f i c i e n c y . However, findings i n these experiments suggest that a higher degree of overpressure w i l l improve effectiveness of heating when large amounts of a i r are entrapped i n the package. This study makes no predictions below 65% steam ( l e v e l 1) as a l i m i t e x i s t s where e x c e s s i v e a i r o v e r p r e s s u r e w i l l reduce h e a t i n g e f f i c i e n c y i n the media. Most previous studies dealing with r e s i d u a l gas have focussed on either pressure-volume r e l a t i o n s to determine maximum allowable gas volumes to prevent pouch bursting (Whitaker, 1971) or the interference by entrapped a i r on heat transfer at a si n g l e processing condition. In a study by Evans (1977), heating rate indices were evaluated for pouches of variable a i r content processed at two l e v e l s of overpressure. His r e s u l t s showed an increase i n f ^ with high a i r volumes (50 mL), yet no -74-e f f e c t was contributed by overpressure even when pouches were restrained (as compared to constrained). Conditions studied were s i m i l a r to 120°C and 125°C at p r e s s u r e s c o r r e s p o n d i n g to 75% steam and 85% steam. Differences from t h i s study may be due to the f a c t that an overpressure representative of 65% steam was not tested and higher a i r volumes were evaluated. A study by Pereira (1980) evaluated various degrees of vacuum on pouches with two overpressure l e v e l s . The reported r e s u l t s were inconsistent; however, f or one thickness (thinnest package) with a low vacuum s e t t i n g (may assume high a i r ) a shorter s t e r i l i z a t i o n time was required when a higher degree of overpressure was used. This f i n d -i n g i s s i m i l a r to r e s u l t s i n t h i s study where higher o v e r p r e s s u r e processing l i m i t e d gas expansion and maintained heat t r a n s f e r , b. Temperature E f f e c t s of temperature may be compared by studying the tables i n columns f or each section. At conditions of 65 and 75% steam (press-ure l e v e l s 1 and 2), permissible a i r volumes increase with increasing temperature. In some cases, f o r example at 65% steam ( l e v e l 1), i t reached the maximum a i r l e v e l t e s t e d and remained as temperature increased. This trend may be explained by larger saturation vapor p r e s s u r e s at high temperatures. T h i s r e s u l t s i n a hi g h e r r e t o r t pressure s e t t i n g when employing the same steam percentage. The same p r i n c i p l e i s employed for UHT processing of f l e x i b l e pouches i n Japan (Tsutsumi, 1979a). Superheated steam at high temperatures provide high saturated vapor pressure to adequately process pouches i n pure steam (f o r heating only) while preventing pouch bursting. At conditions of 85% steam (pressure l e v e l 3) volumes deviate -75-from the pattern displaying lower break-point values at 125 C. A poss-i b l e explanation of t h i s occurrence i n t r i a l s performed i n the p i l o t scale r e t o r t may be due to low rotameter settings at conditions of 85% steam (pressure l e v e l 3). Although media flow rates were constant, the amounts of a i r passing through the system were lower. Media c i r c u l a t i o n may have been i n s u f f i c i e n t and r e s u l t e d i n poor heat t r a n s f e r . A s i m i l a r s i t u a t i o n , however, was found i n the Lagarde. Media flow rates are very high i n t h i s system due to the turbo fan c i r c u l a t i o n mechanism, c. Brick Thickness Although the two brick thicknesses have much d i f f e r e n t magni-tudes of b a s e l i n e f ^ v a l u e s , volumes of a i r which demonstrate no s i g n i f i c a n t difference from the baseline may be compared by studying values i n Table 8 and Table 9. In almost a l l cases where a break-point e x i s t s , the maximum a i r volume of the hori z o n t a l section i s higher f o r thin than f o r thick b r i c k s . For example, with water/air processing at pressure l e v e l s 2 and 3 at 115°C and l e v e l 3 at 120°C, there i s approx-imately a 10-15 mL difference i n volume of a i r forming the plateau region between thicknesses. This magnitude of difference occurs i n some other conditions, but not every case. Variations i n the volume of a i r at which no influence on heat-ing rate occurs may be a r e f l e c t i o n of the pouch s i z e i n r e l a t i o n to the brick volume. Since s i m i l a r pouches were used to package both brick s i z e s , there would be a difference i n the po t e n t i a l volume of space a v a i l a b l e around the b r i c k edges. Consequently, more a i r could accumulate i n surrounding spaces before r e s i d i n g over the brick face when t h i n bricks were packaged compared to pouches containing thick -76-br i c k s . d. Media Type: Steam/Air and Water/Air Modification of the v e r t i c a l p o s i t i v e flow r e t o r t to f a c i l -i t a t e water/air processing enabled studying d i f f e r e n t processing media within the same r e t o r t . In order to compare break-point values of con-d i t i o n s using the media types, a covariance analysis was performed on the baselines. No s i g n i f i c a n t d i f f e r e n c e (p>0.05) was found between baselines of water/air media and steam/air i n the v e r t i c a l p o s i t i v e flow r e t o r t (see Table 10) for either thick or t h i n b r i c k s . Although a s i g n i f i c a n t difference (p<0.05) was found between r e t o r t s for unpackaged thick bricks, i t i s not apparent when comparing baselines. This may be at t r i b u t a b l e to larger v a r i a b i l i t y i n data when a i r was included i n the package. Break-point values for these systems i n Tables 8 and 9 may be studied. I t was apparent that superheated water with a i r overpressure provided processing media which allowed larger volumes of a i r to be i n -cluded i n the sample pouches with no increase i n f ^ values. Data f o r th i n bricks r e f l e c t e d t h i s occurrence strongly; at 115°C for each over-pressure l e v e l , values for water/air processing ranged from 7-14 mL higher than f o r steam/air. A difference between media types was demon-strated by both brick thicknesses at 115°C and 65% steam (pressure l e v e l 1). When processing with steam/air a break-point was exhibited at 15.3 mL (thin) and 17.0 mL (thick) of included a i r , however, there was no evidence of a break-point with s i m i l a r processing conditions when employing water/air media. These studies indicated that when using water/air processes, -77-heating rate in d i c e s were affected to a lesser degree by included a i r than when steam/air mixtures were used. Improved heat transfer using superheated water with a i r overpressure may be due to an increase of pressure due to a hydrostatic head. A water depth of approximately 1 meter i n the v e s s e l would c o n t r i b u t e 10 kPa p r e s s u r e . Bare b r i c k analysis i n the case of thick bricks demonstrated that water/air media may be more e f f e c t i v e independent of a i r volumes. Other researchers have studied media types with s i m i l a r r e s u l t s . Yamano (1976) compared heating rates of pouches i n steam/air and water/air media. Although long come-up times were associated with water processes, when the media reached desired temperatures, heating rates were f a s t e r than with steam/ a i r mixtures. Pflug and Borrero (1967) recommended use of water/air over steam/air media based on s i m p l i c i t y of use and higher enthalpies associated with water/air mixtures per unit volume. Tsutsumi (1979b) stated that water/air processes were commonly used i n Japan when pouches contained larger amounts of included a i r . e. Retorts: Lagarde and P o s i t i v e Flow Steam/Air Steam/air media with s i m i l a r p r o c e s s i n g c o n d i t i o n s were u t i l i z e d i n both the Lagarde and v e r t i c a l p o s i t i v e flow systems. Analysis of covariance was performed on baseline conditions (125°C, 65% steam) of each r e t o r t . No s i g n i f i c a n t difference (p >0.05) was found between the r e l a t i o n s h i p s i n each brick thickness (Table 10). Although a s i g n i f i c a n t d i f f e r e n c e was found between r e t o r t s f o r both bri c k s un-packaged, t h i s was not apparent when comparing baselines. This may be at t r i b u t a b l e to la r g e r v a r i a b i l i t i e s i n data when a i r was included i n the packages. Values i n T a b l e s 8 and 9 may be compared f o r these TABLE 10 Covariance analysis r e s u l t s comparing baselines between r e t o r t systems tested. Brick Thickness Comparison F Ratio Test of Slope Test of Slope Overall Thin Pos. flow s/a 0.69 ns vs. Pos. flow w/a Pos. flow s/a 0.44 ns vs. Lagarde Pos. flow w/a 0.08 ns vs. Lagarde 3.51 ns 2.10 ns 0.37 ns 0.40 ns 0.21 ns 1.08 ns Thick Pos. flow s/a 3.51 ns vs. Pos. flow w/a Pos. flow s/a 1.57 ns vs. Lagarde Pos. flow w/a 0.31 ns vs. Lagarde 2.91 ns 3.30 ns 0.59 ns 1.08 ns 1.09 ns 0.54 ns ns - not s i g n i f i c a n t (p>0.05) -79-systems. Larger volumes of a i r comprise the plateau region f o r the Lagarde compared to the v e r t i c a l p o s i t i v e flow system. This occurrence was demonstrated i n both brick thicknesses. Results from conditions of 115°C and 65% steam exemplify the difference. S u f f i c i e n t conditions ex i s t i n the Lagarde r e t o r t so no interference i n heating rate was apparent at the maximum volumes of a i r added. However, a break-point was detected at 17.0 mL (t h i n bricks) or 15.3 mL (thick bricks) when processing with s i m i l a r r e t o r t conditions i n the v e r t i c a l p o s i t i v e flow r e t o r t . Other conditions exhibit the same trends. These r e s u l t s i n d i c a t e that systems used f o r processing may demonstrate an e f f e c t on the heating rate of products containing res-i d u a l gases. S i n c e s i m i l a r media compositions were u t i l i z e d , differences may be due to the effectiveness of c i r c u l a t i o n . In Lagarde r e t o r t s a h o r i z o n t a l a i r flow pattern i s ensured by c i r c u l a t i o n with a turbo fan causing media to flow p a r a l l e l across the brick surface. Media flow i n the v e r t i c a l p o s i t i v e flow r e t o r t i s i n an upward d i r e c -t i o n , perpendicular to the brick plane. Therefore, c i r c u l a t i o n may be hindered by bricks on the bottom l a y e r s . 3. Importance of Plateau Region Study of the data showed heating rate indices as a function of a i r entrapped i n packaged bricks to consist of either one or two sections. One part was e s s e n t i a l l y a hor i z o n t a l l i n e , the other exhibited an increase i n f h with increasing a i r contents. This two part behavior may be explained by the r i g i d nature of t e f l o n bricks forming void areas i n the brick periphery. Care must be taken i n extrapolating r e s u l t s to a -80-food system. A food product of s o f t e r and more p l i a b l e form would f i l l out to the edges and a i r would accumulate on top. Consequently, data from Tables 8 and 9 may not d i r e c t l y be extracted for implementation as q u a l i t y control s p e c i f i c a t i o n s (on the basis of heat transfer) without knowledge concerning the behavior of a i r with pouches containing foods. The r e s u l t s do suggest that included a i r l i m i t s may be set at higher l e v e l s and not contribute a l i m i t i n g factor to process times when high overpressures are used and adequate c i r c u l a t i o n i s ensured. Any changes i n q u a l i t y control l e v e l s must be tested on a food product, and other d e t e r i o r a t i v e implications of entrapped gases must be evaluated. Also i n regard to s p e c i f i c q u a l i t y control l e v e l s , conditions of 115°C and 85% steam ( l e v e l 3) may be a concern because a i r volumes of l e s s than 10 mL displayed an interference with heat t r a n s f e r . V A i r Volumes Above the Break-Point The previously described moving regression frame determined values of i n c l u d e d a i r at which a t r a n s i t i o n or b r e a k - p o i n t o c c u r r e d ( i f present) from a ho r i z o n t a l l i n e , to one portraying an increase i n heat-ing rate indices with higher a i r volumes. Work by previous researchers have shown a r e l a t i o n s h i p of t h i s nature where included a i r resulted i n an i n s u l a t i n g e f f e c t on heat t r a n s f e r . Evans (1977) reported increases i n f ^ values, Huerta-Espinosa (1981) described the i n s u l a t i n g e f f e c t by increased times required for s t e r i l i z a t i o n and Berry and Kohnhorst (1983) found a s i m i l a r behavior evaluated by decreasing s t e r i l i z a t i o n values or F Q . Experimental variables may be compared by covariance analysis on the data e x i s t i n g beyond the break-point describing the increasing -81-trend. A simple arithmetic l i n e a r r e l a t i o n s h i p was found to best des-cribe the data. A large standard error of estimate may be expected as large v a r i a t i o n s e x i s t i n f ^ values. Only some conditions may be com-pared since many r e l a t i o n s h i p s exhibited only h o r i z o n t a l sections and others showed increases at high a i r volumes, beyond which i n s u f f i c i e n t information was a v a i l a b l e f o r a r e l i a b l e regression l i n e . Therefore, comparisons of equations were focussed on those conditions with apprec-i a b l e data i n the sections above the break-point. Slopes of the l i n e s were parameters of i n t e r e s t since d i f f e r i n g ranges of a i r volumes were used and intercepts were i n e v i t a b l y d i f f e r e n t . 1. Factors A f f e c t i n g the Second Section a. Brick Thickness In the previous section i t was determined that t h i n bricks could accommodate larger volumes of a i r than thick bricks without imped-ing heat t r a n s f e r . The degree of increase beyond the plateau region was compared. Retort conditions of 115°C and 85% steam (pressure l e v e l 3) provided a large amount of data beyond the break-point, so the r e l a t i o n -s h i p s were used f o r comparison. As i n d i c a t e d i n Table 11, no s i g n i f i c a n t d i f f e r e n c e (p>0.05) was found between s l o p e s i n the r e l a t i o n s h i p of heating rate index as a function of a i r volume for t h i n and thick b r i c k s . I t may be assumed that once a c r i t i c a l a i r volume was reached and heating rates began to be affected, a d d i t i o n a l a i r volumes added to the pouch increased f ^ with the same severity for both brick thicknesses. Consequently, a cushion of a i r of s i m i l a r thickness was formed (past c r i t i c a l volumes) f o r each a d d i t i o n a l volume of a i r added. -82-TABLE 11 Covariance a n a l y s i s on vs. included a i r , values above break-point; comparison of brick thickness, 115°C-85% steam (pressure l e v e l 3) Retort Brick Slope Comparison Test of Slope Thickness (mL/min) F Ratio Lagarde Thin Thick 0.396 0.254 Thin vs. Thick 2.24 ns P o s i t i v e flow steam/air Thin Thick 0.091 0.104 Thin vs. Thick 1.33 ns P o s i t i v e flow water/air Thin Thick 0.192 0.261 Thin vs. Thick 0.21 ns ns not s i g n i f i c a n t (p>0.05) -83-I t was apparent i n Tables 8 and 9 that break-point volumes were larger for t h i n compared to t h i c k b r i c k s . Further comparisons of variables were performed on data of thick b r i c k s . b. Steam Percentage/Pressure Level Conditions of variable steam percentages (pressure l e v e l s ) were compared at 115°C for thick bricks i n a l l three systems. Although 115°C was used since i t contained the lowest break-point volumes of a i r , i n two systems a ho r i z o n t a l l i n e or plateau section only was displayed with 65% steam ( l e v e l 1). I f no b r e a k - p o i n t s were apparent, the r e l a t i o n s h i p s were not used for comparison. Results are reported i n Table 12. In one of the three systems (Lagarde), a s i g n i f i c a n t d i f f e r -ence (p<0.05) was found between slopes of 85 and 75% steam, the former having a larger slope than the l a t t e r . The r e l a t i o n s h i p s are i l l u s t r a t -ed i n Figure 6. This fi n d i n g supports evidence that higher degrees of overpressure l i m i t expansion of r e s i d u a l gases. The larger slope r e s u l t i n g from processing with 85% steam may be a t t r i b u t e d to a lower degree of overpressure allowing expansion of gases. Therefore, the i n s u l a t i o n e f f e c t was more severe as a d d i t i o n a l a i r was included. Comparisons of steam percentages i n other systems did not r e f l e c t these findings s t a t i s t i c a l l y since no differences (p>0.05) were found. However, slopes were largest f or 85% steam then decreased for 75 and 65% steam, r e s p e c t i v e l y . This indicated that higher degrees of over-pressure may have l i m i t e d expansion to a larger degree to r e s u l t i n a more gradual increase of heating rate. However, i t must be noted there was no s i g n i f i c a n t d i f f e r e n c e . The v e r t i c a l p o s i t i v e flow r e t o r t employing water processing with a i r overpresure also demonstrated no -84-TABLE 12 Covariance analysis on vs. included a i r , values above break-point; comparison of pressure l e v e l , at 115°C. % Steam/ Retort Pressure Slope Comparison Test of Slope Level (mL/min) F Ratio Lagarde 85 0.271 85 vs. 75 4.84* 75 0.540 Po s i t i v e flow 3 0.145 3 vs. 2 0.54 ns water/air 2 0.104 Po s i t i v e flow 85 0.260 85 vs. 75 1.14 ns steam/air 75 0.178 75 vs. 65 0.83 ns 65 0.088 85 vs. 65 4.02 ns ns * not s i g n i f i c a n t (p>0.05) s i g n i f i c a n t (p<0.05) c X 0) TJ cn c. Q) X 32-T 31-30 29 28 27 26 25 24 ° o x 22 21 20 19 18 17 • Legend • 65% steom A 75% steam X 85% steam X x x A A , A - " " A A • • A - ' A ' , - - - A - A' • -a • T 5 10 15 20 25 Included Air (mL) 30 35 i 00 U l I Figure 6. Heating rate index vs. included a i r for comparison of pressure l e v e l ; thick bricks processed at 115°C i n the Lagarde r e t o r t . -86-s i g n i f i c a n t difference (p>0.05) between slopes at d i f f e r e n t pressure l e v e l s . A s i m i l a r trend was evident since the slope at pressure l e v e l 3 was s l i g h t l y steeper than at pressure l e v e l 2 (corresponding to 85 and 75% steam, r e s p e c t i v e l y ) . Pressure l e v e l 1 exhibited a h o r i z o n t a l l i n e with no evidence of an upward trend, c. Temperature The e f f e c t s of temperature on the f ^ vs. included a i r function were compared at a constant degree of overpressure. Conditions at 85% steam (pressure l e v e l 3) were tested because more information was a v a i l -able for the second equation i n the model. Table 13 shows the r e s u l t s of the covariance a n a l y s i s . In the Lagarde and p o s i t i v e flow water/air r e t o r t s there was no s i g n i f i c a n t difference (p>0.05) between slopes of the l i n e s f o r a l l three temperatures. This would i n d i c a t e that although each temperature may allow for d i f f e r i n g volumes of a i r to be included without demonstrating an e f f e c t on heat transfer, the s e v e r i t y of i n -creases above the c r i t i c a l value showed no s i g n i f i c a n t difference. In the v e r t i c a l p o s i t i v e flow system employing steam/air mixtures a s i g n i f -icant difference (p<0.05) was found between slopes of the r e l a t i o n s h i p s for 115 and 125°C, the former being greater than the l a t t e r . The slope of 120°C was intermediate to 115°C and 125°C showing no s i g n i f i c a n c e (p>0.05) from e i t h e r value. Figure 7 depicts the r e l a t i o n s h i p s which e x h i b i t e d a s i g n i f i c a n t d i f f e r e n c e . A l a r g e r s l o p e at a lower temperature supported r e s u l t s found i n the previous section describing break-point values. Lower saturation vapor pressures associated with low temperatures could cause expansion of entrapped gases to a larger degree, r e s u l t i n g i n a more acute interference i n heat transfer as -87-TABLE 13 Covariance analysis on vs. included a i r , values above break-point; comparison of temperature at 85% steam (pressure l e v e l 3). Retort Temperature (°C) Slope (min/mL) Comparison Test of Slope F Ratio Lagarde P o s i t i v e flow water/air P o s i t i v e flow water/air 115 0.271 115 vs. 120 0.71 ns 120 0.209 120 vs. 125 0.09 ns 125 0.236 115 vs. 125 0.16 ns 115 0.145 115 vs. 120 0.07 ns 120 0.140 120 vs. 125 3.15 ns 125 0.056 115 vs. 125 3.37 ns 115 0.260 115 vs. 120 0.86 ns 120 0.198 120 vs. 125 0.27 ns 125 0.174 115 vs. 125 5.40 * ns not s i g n i f i c a n t (p>0.05) * s i g n i f i c a n t at p<0.05 33 c E. 32-31-30 29-28-27-Legend • 115 C • • • • 10 15 20 Included Air (mL) i 00 00 I 35 Figure 7. Heating rate indices vs. included air for comparison of temperature; thick bricks processed at 85% steam in positive flow retort. -89-displayed by the larger slope. d. Media Type: Steam/Air and Water/Air Comparisons between steam/air and water/air media i n the v e r t -i c a l p o s i t i v e flow r e t o r t system were made on equations describing data beyond c r i t i c a l values. Conditions of 75% steam (pressure l e v e l 2) and 85% steam (pressure l e v e l 3) at 115°C and 120°C were studied since there was substantial information for comparison. Conditions of 65% steam demonstrated a plateau section only f o r water/air processing; i n these cases no comparisons were made. In Table 14 are r e s u l t s from the t e s t of slopes by covariance a n a l y s i s . In one condition, 115°C and 85% steam, a s i g n i f i c a n t difference (p<0.05) was found between media types. Other conditions demonstrated no s i g n i f i c a n t difference (p>0.05). I t may be expected that 115°C and pressure l e v e l 3 would show differences to a larger degree since they represented r e t o r t conditions of the low-est pressure. Data from these r e l a t i o n s h i p s (Figure 8) provided the best l i n e s f or comparisons since there were no apparent plateaus. The slope for steam/air was s i g n i f i c a n t l y higher than for water/air which supports evidence found when comparing break-point values. A larger slope found with steam/air processing r e f l e c t s that steam/air mixtures may provide a l e s s e f f i c i e n t heating medium r e s u l t i n g i n greater degrad-ation of heat transfer with a d d i t i o n a l amounts included a i r . I t may be possible that hydrostatic pressure i n the water/air system contributed to l i m i t i n g the expansion of non-condensible gases, thereby r e s u l t i n g i n a lower slope. Other conditions tested demonstrated no s i g n i f i c a n t difference (p>0.05); however, i n a l l cases slopes were larger f o r steam/air pro--90-TABLE 14 Covariance a n a l y s i s on f h vs. included a i r , values above break-point; comparison of media type. Processing Conditions Media Type Slope Comparison Test of Slope (min/mL) F Ratio 115°C-75% steam (2) Steam/air Water/air 0.189 0.151 s/a vs. w/a 0.12 ns 115°C-85% steam (3) Steam/air Water/air 0.260 0.104 s/a vs. w/a 13.8 * 120°C-75% steam (2) Steam/air Water/air 0.179 0.050 s/a vs. w/a 0.15 ns 120°C-85% steam (3) Steam/air Water/air 0.198 0.140 s/a vs. w/a 1.07 ns ns not s i g n i f i c a n t (p>0.05) * s i g n i f i c a n t at p<0.05 33 32 31 30 F 28 c E 0) 26 TJ C 9S "5 cn "o X 23 20 17-• • Legend • • • • Steam/air D • ft A Water/air • • A A A , . A A A A A A A 0 10 15 20 25 Included Air (mL) 30 35 I Figure 8. Heating rate index vs. included air for comparison of media type; thick bricks processed at 115°C and 85% steam (3). -92-cesses. Although a trend was evident, the slopes were not s t a t i s t i c a l l y d i f f e r e n t . Combining t h i s information with r e s u l t s from c r i t i c a l v o l -umes which displayed no s i g n i f i c a n t difference from a baseline, i t may be assumed that water/air processes show an improvement over steam/air processes with regard to the e f f e c t s of entrapped gases on heat trans-f e r . e. Retort Design: Lagarde and P o s i t i v e Flow Steam/Air Similar to previous comparisons, there were appreciable data beyond the c r i t i c a l volume. Results from Table 15 indicated no s i g n i f i -cant difference (p>0.05) between slopes of the re l a t i o n s h i p s tested. Values of slopes revealed no consistencies as some values were larger i n the Lagarde and some i n the v e r t i c a l p o s i t i v e flow r e t o r t . Figure 9 i l l u s t r a t e s an example of these r e l a t i o n s h i p s . From r e s u l t s on break-point values of the plateau regions, i t was evident that larger volumes of a i r could be included without a f f e c t i n g heat transfer when processing i n the Lagarde r e t o r t . Forced c i r c u l a t i o n by the turbo fan and media flow p a r a l l e l to the brick surface may be responsible f o r t h i s d i f f e r -ence. I t may be expected that slopes of the increasing r e l a t i o n s h i p would be larger (more severe) f o r the v e r t i c a l p o s i t i v e flow system; however, t h i s was not the case. In some cases larger slopes were found when processing i n the Lagarde, which may be due to the absence of an a d d i t i o n a l spacer layer that was inherent i n racks used with the v e r t -i c a l p o s i t i v e flow r e t o r t . As a i r volumes increased beyond the c r i t i c a l point, rack clearance above the pouch would be reduced. In the Lagarde r e t o r t an adequate clearance was necessary to ensure c i r c u l a t i o n , where-as i n the p o s i t i v e flow r e t o r t a spacer layer ensured c i r c u l a t i o n even -93-TABLE 15 Covariance analysis on vs. included a i r , values above break-point; comparison of steam/air r e t o r t s . Processing Conditions Retort Slope Comparison Test of Slope (min/mL) F Ratio 115°C--75% steam Lagarde 0.053 Lag. vs. Pos. 1.45 ns Pos. flow 0.189 115°C--85% steam Lagarde 0.271 Lag. vs. Pos. 0.04 ns Pos. flow 0.261 120°C--75% steam Lagarde 0.138 Lag. vs. Pos. .007 ns Pos. flow 0.180 120°C--85% steam Lagarde 0.209 Lag. vs. Pos. .024 ns Pos. flow 0.198 ns not s i g n i f i c a n t (p>0.05) 35 i i i i l l l i 0 5 10 15 20 25 30 35 Included Air (mL) Figure 9 . Heating rate index vs. included a i r for comparison of steam/air r e t o r t s ; thick bricks processed at 115°C and 85% steam. -95-i f clearance space was diminished. Other researchers have determined racking configurations to be an important f a c t o r i n processing. Evans (1977) compared restrained and constrained racking designs by evaluating heating rate i n d i c e s of pouches containing r e s i d u a l gases. Constrain-ment of pouches reduced the extent of interference by entrapped gases on heat penetration. Ramaswamy (1983) studied v e r t i c a l and horizontal o r i e n t a t i o n with constrained and unconstrained rack types. Results showed that processing with unconstrained racking i n the hori z o n t a l o r i e n t a t i o n reduced heat transfer rates. From t h i s study, i t was e v i -dent that i f "unconstrained" racking was employed, clearance above the pouch may be reduced when large volumes of a i r are entrapped. To ensure proper c i r c u l a t i o n , adequate flow channels should be provided between sample l a y e r s . VI P r e d i c t i o n of Pouch Expansion Previous sections have dealt with the r e l a t i o n s h i p of f ^ as a func-t i o n of r e s i d u a l gases measured at conditions of standard temperature and pressure (STP). These r e l a t i o n s h i p s were valuable i n a p r a c t i c a l sense to determine heating rates of products since s i m i l a r a i r volume measurements are performed i n qu a l i t y control programs. However, v a r i -ables may be studied by predicting the volume of a i r which would p r e v a i l at r e t o r t conditions. Analysis of variance performed on bare bricks showed no s i g n i f i c a n t d i f f e r e n c e s i n v a r i a b l e s of temperature and pressure l e v e l . When a i r was included i n a pouch, differences were att r i b u t e d to an i n t e r a c t i o n between entrapped a i r and r e t o r t conditions of temperature and pressure. Predicted volumes of a i r may account f o r -96-any i n t e r a c t i o n so a l l processing conditions i n each system may be com-bined. A major study on pressure-volume r e l a t i o n s by Davis et a l . (1960) revealed that, due to f l e x i b l e nature of the pouch, i d e a l gas theories applied. Pressures within the pouch arose from vapor pressure of water i n food, a i r pressure i n the headspace or released from the product and thermal expansion of the food i t s e l f . Whitaker (1977) derived equations based on Boyle's law and Dalton's law of p a r t i a l pressures i n order to fi n d a t o t a l overriding pressure necessary to l i m i t expansion beyond a maximum volume (determined by confinement). A s p e c i f i c example was c i t e d where volume expansion of packaged peas i n brine was calculated when no a i r was entrapped and when a i r was included. Studies by Yamano (1976) u t i l i z e d the same concepts. With water and a i r contained i n a pouch, expansion was primarily caused by a i r pressure changes within the headspace. Expansion r a t i o s (expansion headspace equal to pouch expan-sion) were determined for d i f f e r e n t r e t o r t conditions. Assuming a maximum expansion r a t i o to preclude pouch bursting, a safe overriding a i r pressure of 20-30 kPa was determined. 1. Expansion Factors Pressure volume r e l a t i o n s h i p s were used to predict volumes of a i r e x i s t i n g at r e t o r t conditions. In a system of brick, water and a i r , expansion of a l l components must be quant i f i e d . Expansion of water was considered n e g l i g i b l e or decreasing, and since only 10 mL of water was added, any changes were small. Brick expansion may be estimated from the c h a r a c t e r i s t i c s of t e f l o n which has a l i n e a r thermal expansion factor (a of 10~~Vc°) as l i s t e d i n Table 1. Volume expansion (B) was -97-calculated by the following r e l a t i o n s h i p : V t = V Q (1 + 6At) ... (12) where V Q i s o r i g i n a l volume, V t i s volume at r e t o r t conditions, and At denotes temperature d i f f e r e n c e . Considering even the large brick and a temperature change of 100C°, volume expansion was i n s i g n i f i c a n t . Expan-sion due to entrapped a i r was therefore the most important component when considering the p o t e n t i a l e f f e c t on heating rate. The r e l a t i o n s h i p describing expansion r a t i o s volumes i s as follows: V2 P 1 T 2 Expansion Factor (EF) = = ... (13) V l T l p a where and T^ are room pressure and temperature ( i n absolute), T2 i s r e t o r t temperature ( i n absolute) and P a indicates pressure exhibited by a i r within the pouch. To t a l pressure within the pouch was assumed to be equal to r e t o r t pressure because of the f l e x i b l e nature of the f i l m . Since 10 mL of water was added to the pouch, vapor pressure w i l l con-t r i b u t e to the t o t a l i n t e r n a l p r e s s u r e . Dalton's law of p a r t i a l p r e s s u r e s was a p p l i e d to c a l c u l a t e p r e s s u r e due to a i r w i t h i n the system, P a, from the r e t o r t pressure minus saturation pressure of water at r e t o r t temperature. An expansion f a c t o r of ^2^1 ( t n e r a t i o of v o l -ume of a i r at r e t o r t conditions to volume of a i r measured) may be found. Expansion factors were calculated for a l l r e t o r t conditions tested. An example c a l c u l a t i o n at 115°C-75% steam i s as follows: -98-atmospheric pressure 100 kPa. re t o r t pressure - saturation vapor pressure at 115°C = 56.3 kPa. r e t o r t temperature = 388°K. room temperature = 298°K. E x p a n s i o n f a c t o r , 2.31 (14) This expansion factor i n d i c a t e s that for each mL of a i r added to the pouch i t w i l l expand 2.31 times when reaching r e t o r t conditions. Other values of expansion f a c t o r s for r e t o r t conditions studied are i n Table 16. Factors tabulated f o r each condition provide information aiding explanation of findings i n previous sections. The magnitude of expan-sion factors r e f l e c t s volumes of a i r included i n plateau regions. I t i s apparent i n Table 16 that a decrease i n overpressure from 65% steam to 85% steam r e s u l t s i n an increase i n expansion f a c t o r s . A d d i t i o n a l l y , a decrease i n expansion factor was calculated as temperatures increase (with a constant degree of overpressure). I t may be noted that the baseline r e l a t i o n s h i p of 125°C and 65% steam ( l e v e l 1) used for com-parisons to determine the breakpoint has an expansion factor close to one (1.07). As previously discussed, there was an increase of maximum volumes displaying no s i g n i f i c a n t difference from the baseline, as the degree of overpressure increased. S i m i l a r l y , increasing temperature at the same overpressure was characterized by larger plateau regions. Results showed that when expansion factors were small, large volumes of -99-TABLE 16 Theoretical expansion factors f o r conditions studied. Steam (%) Temperature, °C 65 75 85 115 1.43 2.31 4.36 120 1.25 1.99 3.76 125 1.07 1.73 3.26 -100-a i r could be included within the pouch without i n t e r f e r i n g with heat t r a n s f e r . Measured volumes of a i r may be adjusted by multiplying by respect-ive expansion factors to determine the corresponding volumes of a i r p r e v a i l i n g during processing. Retort temperature and pressure were set by c o n t r o l mechanisms as p r e v i o u s l y d e s c r i b e d ; however, i n many instances actual conditions may have s l i g h t deviations from the target. To correct f o r these differences, a l l volumes of a i r were adjusted based on expansion factors calculated from actual r e t o r t conditions. Figures 10 and 11 i l l u s t r a t e r e l a t i o n s h i p s between the log of heating rate indices and volume of a i r adjusted with expansion f a c t o r s (on an a r i t h -metic s c a l e ) . Uneven spacing of data points along the abscissa was due to differences i n expansion f a c t o r s . Factors for conditions of 65% steam ( l e v e l 1) varied from approximately 1.07 f o r 125°C to 1.43 at 115°C; therefore, values were spread out over lower adjusted volumes. Expansion factors of 1.73 (125°C) to 2.31 (115°C) were determined for 75% steam ( p r e s s u r e l e v e l 2) w i t h data ranging from low to medium adjusted volumes. Conditions of lower overpressure with pressure l e v e l 3 (85% equivalent) were represented by expansion factors ranging from 3.26 (125°C) to 4.36 (115°C) so data were spread across the e n t i r e abscissa from low to high adjusted a i r contents. 2. Comparison of Adjusted Volume Relationships A semi-logarithmic r e l a t i o n s h i p c l e a r l y indicated an increase i n f ^ with, i n c r e a s i n g a d j u s t e d volumes of a i r . The p l a t e a u s e c t i o n as demonstrated with included a i r volumes were no longer apparent because -101-1.1 i 110 -10 10 30 50 70 90 Ad jus ted A i r (mL) 130 Figure 10. Heating rate index functions of adjusted air volumes for a l l conditions; thin bricks, (a, Lagarde; b, Positive Flow Steam/Air; c, Positive Flow Water/Air). -102-16-, : Figure 11. Heating rate index functions of adjusted a i r volumes for a l l conditions; thick b r i c k s . (a, Lagarde; b, P o s i t i v e Flow Steam/Air; c, P o s i t i v e flow Water/Air). -103-those values remained at lower ranges of the adjusted volume abscissa. Covariance a n a l y s i s was performed to compare r e t o r t and media systems employed i n these experiments. Table 17 shows r e s u l t s of the analysis and slopes describing each r e l a t i o n s h i p . A s i g n i f i c a n t difference (p<0.01) was found between a l l three systems for both thicknesses. From values of the slopes and observation of Figures 10 and 11, r e s u l t s i n d i -c ate t h a t p r o c e s s i n g with steam/air media u t i l i z i n g the v e r t i c a l p o s i t i v e flow r e t o r t , displays a more severe e f f e c t of r e s i d u a l gases on heating rate than demonstrated by processing i n the Lagarde. Above a l l , water with a i r overpressure media was l e s s s e n s i t i v e to increasing a i r compared to both steam/air systems, as demonstrated by a s i g n i f i c a n t l y lower slope. The r e s u l t s were s i m i l a r f o r each brick thickness and concur with f i n d i n g s from unadjusted a i r volumes. Differences i n steam/ a i r systems may be a t t r i b u t a b l e to the turbo fan d r i v e n media c i r c u l a t i o n and flow patterns p a r a l l e l to pouch p o s i t i o n i n the Lagarde. This system provides more rapid heating than the v e r t i c a l p o s i t i v e flow r e t o r t . Higher heat transfer c o e f f i c i e n t s for water/air compared to steam/air may account f o r a more gr a d u a l i n c r e a s e i n l o g f ^ as a function of adjusted a i r volume. Another pl a u s i b l e explanation may be due to an increase i n pressure from hydrostatic forces i n the system. The r e s u l t s were contrary to findings by M i l l e v i l l e (1980) where a buoyant e f f e c t of the r e s i d u a l a i r bubble within the pouch was expected to reduce heat t r a n s f e r . However, observations of the buoyant e f f e c t of a i r i n s i d e the pouch were performed at atmospheric pressure. The adjusted a i r volume approach may prove useful i n predicting h e a t i n g r a t e i n d i c e s of pouches c o n t a i n i n g r e s i d u a l gases when -104-TABLE 17 Covariance analysis on log vs. adjusted a i r , a l l data; comparison of systems tested. Brick Retort System Slope (mL~^) Comparison Test of Slope Thickness x (10~ 3) F Ratio Thin Lagarde 2.32 Lag. vs. s/a 10.6 ** Po s i t i v e flow steam/air (s/a) 2.89 Lag. vs. w/a 99.2 ** Po s i t i v e flow water/air (w/a) 0.84 s/a vs. w/a 203 ** Thick Lagarde 1.18 Lag. vs. s/a 8.13 Po s i t i v e flow steam/air 1.46 Lag. vs. w/a 22.5 ** P o s i t i v e flow water/air 0.77 s/a vs. w/a 70.4 ** ** s i g n i f i c a n t at p<0.01 -105-processing within the range of conditions studied. An estimate of expanded a i r volumes that w i l l p r e v a i l at r e t o r t conditions can be c a l c -ulated and heating rates may be determined. The model system used, simulating a food with water a c t i v i t y ( a w ) equal to one, represents the most extreme case of pouch expansion. Many products have a s l i g h t l y lower water a c t i v i t y and w i l l e x h i b it a lower p a r t i a l pressure due to water vapor. Expansion factors may be calculated to r e f l e c t these dif f e r e n c e s . For example, an expansion factor when a w = 1.0 for 115°C and 85% steam i s 4.36; compared to a calculated f a c t o r of 3.40 for a food of a, = .95 at i d e n t i c a l r e t o r t conditions. Quite a sub s t a n t i a l w ^ difference e x i s t s , so water a c t i v i t y of the food processed must be con-sidered an important f a c t o r when evaluating e f f e c t s of re s i d u a l gases on heating behavior. Another p r a c t i c a l a p p l i c a t i o n of t h i s approach may a i d when pro-cessing products such as bakery foods or batters i n r e t o r t pouches. These goods require leavening and i n c l u s i o n of non-condensible gases (usually CO2) i s e s s e n t i a l f o r texture. Control of processing con-d i t i o n s i s important to achieve a balance between allowing the product to r i s e , but preventing the pouch from bursting. Leavening, s e t t i n g of structure and s t e r i l i z a t i o n are the three c r i t i c a l factors of concern. Andres and Duxbury (1972) developed d i f f e r e n t i a l pressure sensing and c o n t r o l l i n g methods to achieve proper product texture and maintain pouch seal i n t e g r i t y during r e t o r t processing. They did not, however, provide d e t a i l s of t h e i r developments. -106-VII Other Process Parameters As previously described, thermocouples were c e n t r a l l y located with-i n the b r i c k . When volumes of entrapped a i r form a l a y e r on the surface, the coldest heating point would s h i f t from the brick midplane. Temperature sensor placement errors were studied by Peterson and Adams (1985). Their r e s u l t s , based' on determined process times, show that p o s i t i o n of the thermocouple was not c r i t i c a l within a range of pos-i t i o n s around the center of the slab. I t was also evident that the p o s i t i o n error allowed without a f f e c t i n g process times was greater f o r thick slabs. Measurement of f h was not highly s e n s i t i v e to thermocouple l o c a t i o n within the sample, so t h i s was not a great concern i n experi-mentation. However, the lag factor j (a measure of the r a t i o of g value as estimated from a s t r a i g h t l i n e portion of the heating curve to actual g value at steam on), i s highly s e n s i t i v e to thermocouple l o c a t i o n . Berry and Kornhorst (1983) described coldest heating locations within pouches containing large amounts of a i r . I t was found that with more than 50 mL of a i r i n an i n s t i t u t i o n a l s i z e pouch, the slowest heating point measured 2/3 of the maximum pouch thickness. A s i m i l a r t e s t i n g must be performed for samples containing a i r i n t h i s experiment. Lag f a c t o r s can then be evaluated i n order to compare s t e r i l i z a t i o n values of processes or time required to achieve a desired s t e r i l i z a t i o n of pouches containing large amounts of r e s i d u a l gases. -107-OONCLUSIONS The r e l a t i o n s h i p of heating rate index as a function of included a i r may be described as consisting of e i t h e r one or two e s s e n t i a l l y l i n e a r segments. One segment formed a plateau region where there was no change i n f n with increasing volumes of included a i r . The second sec-t i o n , apparent i n some conditions only, exhibited an increase i n f ^ as a function of increasing a i r volumes. Increased overpressure l e v e l s and temperatures (with the exception of 125°C and 85% steam) allowed for larger volumes of a i r to be included i n the plateau region. A i r overpressure above 80 kPa allowed up to 35 mL to be included with no detrimental e f f e c t on heating rates. Con-versely, even small amounts of entrapped a i r (<5 mL) degraded heat transfer when processing with 30 kPa a i r overpressure. Increasing over-pressure and temperature were also shown to display decreasing slopes i n the second l i n e a r section where an increase i n f ^ i s seen as a function of increasing a i r volumes. Processing with water/air media allowed larger volumes of a i r to be included i n the plateau region compared to steam/air. Both media were evaluated i n a p o s i t i v e flow r e t o r t . . As well, steeper slopes of the second segment were demonstrated with steam/air media i n d i c a t i n g a more severe detrimental e f f e c t of entrapped a i r on heat t r a n s f e r . Studies with r e t o r t types demonstrated that when processing i n the Lagarde r e t o r t , larger volumes of a i r could be entrapped without a f f e c t i n g heat trans f e r than with the p o s i t i v e flow r e t o r t . No s i g n i f i c a n c e was seen i n slopes of the second section. Expansion factors for processing conditions were determined from -108-i d e a l gas laws. The factors were used to adjust a i r volumes for expan-sion that occurs at r e t o r t conditions. A l i n e a r r e l a t i o n s h i p was found between the logarithm of heating rate index and adjusted a i r volumes on an arithmetic a x i s . Comparisons of slopes showed s i g n i f i c a n t d i f f e r -ences; water/air processes i n the p o s i t i v e flow r e t o r t being of the le a s t slope, the Lagarde and p o s i t i v e flow steam/air processes had increasing slopes, r e s p e c t i v e l y . -109-LITERATDRE CITED Adams, J.P. 1984. Retort pouch uses for seafood products. Prepared Foods 153 (3): 90. Adams, J.P., Peterson, W.R. and O t w e l l , W.S. 1983. P r o c e s s i n g of seafood i n i n s t i t u t i o n a l - s i z e d r e t o r t pouches. Food Technol. 37 (4): 123. Agarwal, S.R. and Kumta, U.S. 1974. Puncture resistance of f l e x i b l e s , Modern Packaging 47 (10): 43. Andres, C. and Duxbury, D.D. 1972. Automatic l i n e produces thermally processed foods i n f l e x i b l e packages. Food Processing 33 (11): 58. Anon. 1980. Research and development associates f i n d a l l - o u t m i l i t a r y commitment to re t o r t a b l e pouch. Package Eng. 25 (5). Anon. 1982. What's happening with the r e t o r t pouch? Food Eng. 54 (4): 97. Anon. 1983. The r e t o r t pouch: how close? Food Eng. 55 (1): 83. Ayoub, J.A., Berkowitz, D., Kenyon, E.M. and Wadsworth, CK. 1974. Continuous microwave s t e r i l i z a t i o n of meat i n f l e x i b l e pouches. J . of Food S c i . 39 (2): 309. Badenhop, A.F. and M i l l e v i l l e , H.P. 1980. Special report: I n s t i t u t i o n a l s i z e r e t o r t pouches. Food Process. 41 (1): 82. B a l l , CO. and Olson, F.C.W. 1957. S t e r i l i z a t i o n i n Food Technology, 1st Ed., McGraw-Hill Book Company, Inc., New York, NY. Bee, G.R. and Park, D.K. 1978. Heat p e n e t r a t i o n measurement f o r thermal process design. Food Technol., 32 (6): 56. Berry, M.R., J r . 1979. The s t e r i l i z a t i o n of food i n pouches - c r i t i c a l parameters for s t i l l processing. Proceedings of the conference held i n Indianapolis, IN: Using Retort Pouches Worldwide - Focus on the Present with a Look to the Future. Sponsored by Food Sciences Inst. Purdue Univ., March 14-15, p.7. Berry, M.R., J r . and Kohnhorst, A.L. 1983. C r i t i c a l f a c t o r s f o r thermal processing of i n s t i t u t i o n a l pouches. J . of Food Prot. 46 (6): 487. Beverly, R.G. 1979. S t e r i l i z a t i o n methodology applied to pouches and tra y s . Proceedings of the conference held i n Indianapolis, IN: Using Retort Pouches Worldwide - Focus on the Present with a Look to the Future. Sponsored by Food Sciences Inst. Purdue Univ., March 14-15, p.102. -110-B e v e r l y , R.G. 1980. R e t o r t pouch i n the '80's. Food Eng. 52 (3):100. Beverly, R.G., Strasser, J . and Wright, B. 1980. C r i t i c a l f a c t o rs i n f i l l i n g and s t e r i l i z i n g of i n s t i t u t i o n a l pouches. Food Technol. 34 (9): 44. Brown, F. and Keegan, P.B. 1973. Heat s e a l i n g methods. Modern Packaging 46 (12): 314. Burke, P.T. and Schultz, G.L. 1972. The comparative performance of f l e x i b l e packages and metal cans. Technical Report, Army Natick Laboratories, Natick, MA. Cage, J.K. and Clark, W.L. 1980. Opportunities and constraints for f l e x i b l e packaging of foods. Food Technol. 34 (9): 28. Cruess, W.V. 1958. Commercial F r u i t and Vegetable Products, 4th Ed., McGraw-Hill Book Company, Inc., New York, NY. Davis, E.G., Karel, M. and Proctor, B.E. 1960. The pressure-volume r e l a t i o n i n f i l m packages during heat processing. Food Technol. 14 (3): 165. Davis, R.B., Long, F.E. and Robertson, W.F. 1972. E n g i n e e r i n g considerations i n r e t o r t processing of f l e x i b l e packages. Food Technol. 26(8): 65. Ebben, K.M. 1979. Producing pouches and t r a y s to s u i t p a r t i c u l a r needs. Proceedings of the conference held i n Indianapolis, IN: Using Retort Pouches Worldwide - Focus on the Present with a Look to the Future. Sponsored by Food Sciences Inst. Purdue Univ., March 14-15, p. 27. Evans, K.W. 1977. The e f f e c t of entrapped a i r on the rate of heat penetration i n s t e r i l i z e d f l e x i b l e pouches. Technical Memo No. 164, The Campden Food Preservation Research Assn., p.16. Ga g l i a r d i , G.R., S u l l i v a n , D. and Smith, N.F. 1984. Computer-aided video inspection. Food Technol. 38 (4): 53. Ghosh, A. and R i z v i , S.S.H. 1982. C o r r e c t i o n f a c t o r s f o r non-d e s t r u c t i v e measurement of r e s i d u a l volume of a i r i n r e t o r t pouches. J . of Food S c i . 47 (3): 969. Goldfarb, P.L. 1970. Pouch for low-acid foods. Part I, Modern Packag-ing 43 (12): 70. G o l d f a r b , P.L. 1971. Pouch f o r low - a c i d foods. P a r t I I , Modern Packaging 44 (1): 70. -111-Gylys, R.B. and R i z v i , S.S.H. 1983. A simple non-destructive method for determination of a i r content i n polymeric packages. J . Food S c i . 48 (3): 1560. Heid, J.L. 1970. Retort process minimizes r e s i d u a l headspace gases with 'breathable' pouch. Food Product Develop. 4(6): 82. Heintz, D.A. 1980. Marketing opportunities for the r e t o r t pouch. Food Technol. 34 (9): 32. Hu, R.H., Legault, Nelson R. and Steinberg, M.P. 1955. F e a s i b i l i t y of using p l a s t i c f i l m packages for heat processed foods. Food Tech-nol. 19 (9): 236. Huerta-Espinosa, V.M. 1981. S i g n i f i c a n c e of A i r Removal i n I n s t i t u t i o n a l Size Retortable Pouches, M.Sc. Thesis, Oregon State University, C o r v a l l i s , OR. Karel, M. 1974. Packaging protection for oxygen-sensitive products. Food Technol. 28 (8): 50. K e l l e r , R.G. 1959. F l e x i b l e packages for processed foods. Modern Packaging 33 (1): 145. Khazanie, R. 1979. Elementary S t a t i s t i c s i n a World of Applications, Goodyear Publishing Co., Inc., Santa Monica, CA. Kopetz, A.A., Prange, C A . and Flessner, R.J. 1979. The future i n our hands: c r i t i c a l f a c t o r s i n r e t o r t pouch thermal process assurance. A c t i v i t i e s Report of the R&D Associates 31 (2): 49. Lampi, R.A. 1977. F l e x i b l e packaging f o r thermoprocessed foods. Adv. Food Res. 23: 305. Lampi, R.A. and Rubinate, F.J. 1973. Performance and i n t e g r i t y of r e t o r t pouch sea l s . Food Technol. 30 (2): 38. Lampi, R.A., Schulz, G.L., C i a v a r i n i , T. and Burke, P.T. 1976. Thermo-processed foods i n f l e x i b l e packages: t r a n s i t i o n to production. Package Devel. 3 (4): 12. Le, C P 1978. UBC MFAV Analysis of Variance/covariance. Computing Centre, Un i v e r s i t y of B r i t i s h Columbia, Vancouver, BC. Loncin, M. and Merson, R.L. 1979. Food Engineering - P r i n c i p l e s and Selected Applications. Academic Press, New York, NY. Lopez, A. 1981. A Complete Course i n Canning. Book I. Basic Inform at i o n on Canning, 11th Ed., The Canning Trade Inc., Baltimore, MD. Mantell, C.L. (Ed.) 1958. Engineering Materials Handbook, 1st Ed., McGraw-Hill Book Co., New York, NY. -112-Mermelstein, N.H. 1976, An overview of the r e t o r t pouch i n the U.S. Food Technol. 30 (2): 28. Mermelstein, N.H. 1978. Retort pouch earns 1978 IFT food technology i n d u s t r i a l achievement award. Food Technol. 32 (6): 22. M i l l e v i l l e , H.P. 1980. Steam/air r e t o r t i n g of pouches: problems, s o l -utions. A c t i v i t i e s Report of the R&D Associates 32 (2): 45. M i l l e v i l l e , H.P. 1981. R e t o r t pouch p r o c e s s i n g advances. Food Processing 42 (1): 92. M i l l e v i l l e , H.P. and Badenhop, A.F. 1980. Guidelines on good manu-factu r i n g p r a c t i c e f o r s t e r i l i z a b l e f l e x i b l e packaging operations for low-acid foods. From Technical Manual No. 4, Campden Food P r e s e r v a t i o n Research A s s o c i a t i o n , Dept. of Food Science and Technol., Oregon State Univ., C o r v a l l i s , OR. Morris, C.E. 1981. Inside North America's new r e t o r t pouch plant. Food Eng. 53 (6): 56. Nelson, A.I., Hu, K.H. and Steinberg, M.P. 1956. Heat processible food f i l m s . Modern Packaging 29 (10): 173. Nelson, A.I. and Steinberg, M.P. 1956. Retorting foods i n p l a s t i c bags. Food Eng. 28 (1): 92. Olson, F.C.W. and Jackson, J.M. 1942. Heating curves, theory and p r a c t i c a l a p p l i c a t i o n . Ind. Eng. Chem. 34: 337. P a t i n o , H. and H e i l , J.R. 1985. A s t a t i s t i c a l approach to e r r o r analysis i n thermal process c a l c u l a t i o n s . J . of Food S c i . 50 (4): 1110. Perei r a , Vincente de Paula 1980. Processing of Cherry Pie F i l l i n g i n Retortable Pouches. Ph.D. Thesis, Michigan State Univ. Dept. of Food Science and Human N u t r i t i o n , East Lansing, MI. Peters, J.W., 1975. R e t a i l debut of r e t o r t pouch earns consumer accept-ance. Food Product Devel. 9 (2): 22. Peters, J.W. 1985. 'Niche marketing' targets r e t o r t pouch at campers. Packaging 30 (3): 31. Peterson, W.R. and Adams, J.P. 1983. Water v e l o c i t y e f f e c t on heat penetration parameters during i n s t i t u t i o n a l s i z e r e t o r t pouch pro-cessing. J . of Food S c i . 48 (2): 457. Pe t e r s o n , W.R. and Adams, J.P. 1985. Temperature sensor placement errors i n i n s t i t u t i o n a l s i z e r e t o r t pouches. J . of Food S c i . 50 (1): 192. -113-P f l u g , I . J . 1964. E v a l u a t i o n of Heating Media f o r Producing S h e l f Stable Food i n F l e x i b l e Packages. Phase I. F i n a l Report, Contract DA19-AMC-145(N). US Army Natick Laboratories, Natick, MA. Pflug, I . J . 1975. Procedures f o r Carrying Out a Heat Penetration Test and Analysis of the Resulting Data. Uni v e r s i t y of Minnesota Press, Minneapolis, MN. Pflug, I . J . , Boch, J.H. and Long, F.E. 1963. S t e r i l i z a t i o n of foods i n f l e x i b l e packages. Food Technol. 17 (9): 87. Pflug, I . J . and Borrero, C. 1967. Heating Media for Processing Foods i n F l e x i b l e Packages. Phase I I . Tech. Rep. 67-47-GP. U.S. Army Natick Laboratories, Natick, MA. Potter, K.M., Tung, M.A. and Kitson, J.A. 1982. Quality of processed peach s l i c e s stored i n f l e x i b l e pouches. Can. Inst. Food. S c i . Technol. J . 15 (2): 96. Przybyla, A. 1984. Shelf stable foods expertise l i e s i n r e t o r t pouch production. Prepared Foods 153 (5): 90. Ramaswamy, H.S., 1983. Heat Transfer Studies of Steam/air Mixtures f o r Food Processing i n Retort Pouches. Ph.D. Thesis, University of B r i t i s h Columbia, Vancouver, BC. Ramaswamy, H.S., Tung, M.A. and Stark, R. 1983. A method to measure surface heat transfer from steam/air mixtures i n batch r e t o r t s . J . of Food S c i . 48 (3): 900. Rees, J.A. 1979. S t e r i l i z a t i o n methodology a p p l i e d to pouches and trays. Proceedings of the conference held i n Indianapolis, IN: Using Retort Pouches Worldwide - Focus on the Present with a Look to the Future. Sponsored by Food Sciences Inst. Purdue Univ., March 14-15., p.88. R i z v i , S.S.H. and Gylys, R.B. 1983. Non-destructive method measures a i r i n pouches. Packaging 28 (9): 65. R i z v i , S.S.H. and Klemaszewski, J . L. 1985. Si m p l i f i e d t e s t measures amount of a i r i n packages. Packaging 30 (3): 72. Roop, R.A., Marks, J.S. and Nelson, P.E. 1983. E f f e c t of thermal pro-cessing and a g i t a t i o n i n 100% steam on the seal strength of r e t o r t -able pouches. J . of Food S c i . 48 (1): 250-252. Roop, R.A. and Nelson, P.E. 1981. Processing r e t o r t pouches i n con-ventional s t e r i l i z e r s . J . of Food S c i . 47 (1): 303. Rubinate, F.J. 1964. A new look i n packaging. Food Technol. 18 (11): 71. -114-Shappee, J . and Werkowski, S.J. 1972. Study of a Non-destructive Test for Determining the Volume of A i r i n F l e x i b l e Food Packages. Tech-n i c a l Report, Army Natick Laboratories, Natick, MA. Shenkenberg, D. 1975. USDA status on f l e x i b l e r e t o r t a b l e pouches for meat products. A c t i v i t i e s Report of the R&D Associates 27 (1): 157. Smith, T., Tung, M.A., Bennett, L. and Cumming, D.B. 1985. Importance of removing condensate from v e r t i c a l r e t o r t s . Food Technol., 39 (6): 53. Snedecor, G.W. 1965. S t a t i s t i c a l Methods, 5th ed., The Iowa State University Press, Ames, IA. S t e f f e , J.F., W i l l i a m s , J.R., Chinnan, M.S. and Black, J.R. 1980. Energy requirements and costs of r e t o r t pouch vs. can packaging systems. Food Technol. 34(9): 39. Strasser, J.H. 1979. Update on r e t o r t pouch f i l l e r s and s t e r i l i z e r s . A c t i v i t i e s Report of the R&D Associates 31 (2): 54. Stumbo, C.R. 1973. Therraobacteriology i n Food Processing. 2nd. Ed., Academic Press, New York, NY. Toyo Seikan Kaisha Ltd. 1973. Opening up tomorrow's markets. RP-F Guide, Tokyo, Japan. Tsutsumi Y. 1972. Retort pouch - i t ' s development and a p p l i c a t i o n to foodstuffs i n Japan. Japan P l a s t i c s 1:24. Tsutsumi Y. 1974. The growth of ret o r t a b l e pouch food packaging i n Japan. The 36th Annual National Packaging Forum of the Packaging I n s t i t u t e , Chicago, IL, October 7-9. Tsutsumi, Y. 1979a. New technology a p p l i e d to pouches and f u t u r e trends. Proceedings of the conference held i n Indianapolis, IN: Using Retort Pouches Worldwide - Focus on the Present with a Look to the Future. Sponsored by Food Sciences Inst. Purdue Univ., March 14-15, p.67-68. Tsutsumi, Y. 1979b. S t e r i l i z a t i o n methodology applied to pouches and trays. Proceedings of the conference held i n Indianapolis, IN: Using Retort Pouches Worldwide - Focus on the Present with a Look to the Future. Sponsored by Food Sciences Inst. Purdue Univ., March 14-15, p.86-87. Tung, M.A. and Garland, T.D. 1978. Computer c a l c u l a t i o n of thermal processes. J . Food S c i . 43 (2): 365. -115-Tung, M.A., Ramaswamy, H.S. and Papke, A.M. 1984a. Thermophysical Studies f o r Improved Food Processes. F i n a l Report. DSS F i l e No. 35SZ.01804-9-0001, prepared f or the Agriculture Canada PDR Program, Ottawa, ON. Tung, M.A., Ramaswamy, H.S., Smith, T. and Stark, R. 1984b. Surface heat trans f e r c o e f f i c i e n t s f o r steam/air mixtures i n two p i l o t scale r e t o r t s . J . Food S c i . , 49 (3): 939. Tung, M.A. and Smith, T. 1980. Innovations i n thermal processing, i n Processing 2000 Symposium, p.103, MacDonald College of Mc G i l l University, Montreal, PQ. Tuomy, J.M. and Young, R. 1982. Retort-pouch packaging of muscle foods fo r the armed forces. Food Technol. 36 (2): 68. Wallenberg, E. and J a r n h a l l , B. 1957. Heat s t e r i l i z a t i o n i n p l a s t i c s . Modern Packaging 31 (7): 165. Whitaker, W.C. 1971. Processing f l e x i b l e pouches. Modern Packaging 44 (2): 83. Williams, J.R., St e f f e , J.F. and Black, J.R. 1983. S e n s i t i v i t y of selected factors on costs of r e t o r t pouch packaging systems. Food Technol. 37 (4): 92. Wilson, D.C. 1980. The o r e t i c a l problems i n pouch processing. Proceed-i n g s of winter meeting of American S o c i e t y of A g r i c u l t u r a l Engineers, Chicago, IL. December 2-5. Yamaguchi, K., Komatsu, Y. and Kishimoto, A. 1972. S t e r i l i z a t i o n of foods i n f l e x i b l e packages, Part VI, Non-destructive method for determining r e s i d u a l a i r i n pouches. J . Food S c i . Technol. (Japan) 19 (7): 316. Yamano, Y. 1976. Studies of Thermal Processing of F l e x i b l e Food Pack-ages by Steam and A i r Retort. Ph.D. Thesis, Kyoto University, Kyoto, Japan. Yamano, Y. and Komatsu, Y. 1969. S t e r i l i z a t i o n of foods i n f l e x i b l e packages, P a r t I, Heat c h a r a c t e r i s t i c s of a p i l o t r e t o r t f o r s t e r i l i z a t i o n of f l e x i b l e packages. J . Food S c i . Technol. (Japan) 16 (3): 113. Young, K.E. 1984. Thiamine Degradation i n a Luncheon-type Ham Product Thermally Processed i n Retort Pouches and Cans. M.Sc. Thesis, University of B r i t i s h Columbia, Vancouver, BC. Zar, J.H. 1974. B i o s t a t i s t i c a l A n a l y s i s . P r e n c t i c e H a l l Inc., Englewood C l i f f s , NJ. -116-APPENDIX I Non-destructive A i r Measurement Archimedes P r i n c i p l e : B - Mg = pV Qg - p oV Qg = (p - p Q) V Qg where B = buoyant force (upward) Mg = weight of object (force downward) p = density of surrounding f l u i d p Q = density of object V Q = volume of object g = g r a v i t a t i o n a l f i e l d Applied to non-destructive a i r measurement Mg - B = p Q V Qg - p V Qg = (p Q-p) V Qg 1. Before a i r addition Wwb = M b § - B l = (Pb-Pw) V where Ww^ = weight of brick i n water M^ = mass of brick Bj = buoyant force (before a i r added) P^ = density of brick = volume of brick 2. After a i r addition Wwb +a " M b + a " B2 - (Pb+a " PW> < V V a > g where ^wb+a = weight of brick and a i r i n water M^ + a = mass of brick and a i r B2 = buoyant force ( a f t e r a i r added) P^ + a = density of brick and a i r V„ = volume of a i r a - 1 1 7 -Expand these equations. From Equation 1: Wwb = Pb V - P w V = M bg - p wV bg M b § = Wwb + Pw Vb§ Wwb +a - Pb +a ( W S " Pw ( V V a > 8 - M b + a 8 " Pw < V V a > 8 Mb+a§ = Wwb+a + Pw ( W S Assuming mass of a i r n e g l i g i b l e , then M b + a = Mb Wwb + PwVb § = Wwb+a + Pw ( Vb + Va> 8 From Equation 2: Rearrange to: Wwb - Wwb+a = P W ( V b + V a ) 8 - P wV bg = P wV bg + P wV ag - p wV bg Amg = p wV ag Am = p V w a Assuming density of water at 25°C = 1.0 g/mL, for each mL of a i r added, the mass of the brick and a i r w i l l be reduced by 1.0 g. -118-APPENDIX II Back Calculations f o r Combination of Destructive and Non-destructive A i r Measurements 1. Leakage of a i r during processing, mL: = (B-A) 1.0 mL/g 2 . Non-destructive volume, post processing, mL: = (I-A) 1.0 mL/g 3. I n i t i a l volume of a i r (before a i r addition), mL: = V - (I-A) 1.0 mL/g 4 . Projected weight of pouch containing no a i r , g: = A + V 1.0 g/mL where: V = corrected destructive a i r measurement (mL) I = f i r s t non-destructive a i r reading (g) (p r i o r to a i r addition) B = second non-destructive a i r reading (g) (before processing) A = t h i r d non-destructive a i r reading (g) ( a f t e r processing) -119-APPENDIX I l i a Heating rate index vs. included a i r r e l a t i o n s h i p for t h i n bricks i n pos i t i v e flow steam/air r e t o r t (a, 115°C; b, 120°C; c, 125°C). 0) c o ca u> "5 X u 13-12-11-10 9 8-7-6-5 Legend A 65% lltom X 75% sttom • 85% steom * 8 A 4 X I gA EL O X x * D A • X • X Legend A 65% steom X 75% steom • 85% steam x IX*£° O • X A A *x£> Legend A 65% steom X 75% steom • 85% steam D £ A • • A A • o o a X *X A 'AA A i D D A* * fi1 *AA X IS 20 Included A i r (mL) 25 i 30 35 -120-APPENDIX I l l b ing rate index vs. included a i r r e l a t i o n s h i p f o r thick bricks p o s i t i v e flow steam/air r e t o r t (a, 115°C; b, 120°C; c, 125°C). 32 30- | 29 28 26-| 25 24 23-22-21-20 19 18 H c x V TJ c o or O l _c "5 « x 31-30-29-28-27-26-25-24 23 22 21 20 19 18 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 4 Legend A 653 steom X 753 steom • 853 steom • O O • D • TT • ° oft x A x x D D A * x° * * A x A A A Legend A 653 steom x 753 steom • 853 steom D D • X • A a ( A A A , A A A A • A X Legend A 653 steom X 753 steom • 653 steom X • X X X X X A * A x 4 *A o o uo o x x x x X x A 15 20 25 Included A i r (mL) - i — 30 35 -121-APPENDIX IIIc ;ating rate index vs. included a i r r e l a t i o n s h i p for t h i n bricks p o s i t i v e flow water/air r e t o r t (a, 115°C; b, 120°C; c, 125°C). 6H x •o "5 cc . c "5 I 9 H 7H 6H ro H 7^ 6H Legend A Pr. lev. 1 X Pr. lav. 2 • Pr. lav. 3 Q Qb A A A A A " X 10 15 20 25 30 35 Included Air (mL) -122-APPENDIX H i d Heating rate index vs. included a i r r e l a t i o n s h i p f o r thick bricks p o s i t i v e flow water/air r e t o r t (a, 115°C; b, 120°C; c, 125°C). 28-27-26-25-24-23-22-21-20-1 9 -18-L e g e n d A Pr. l a v . 1 x Pr. l e v . 2 • Pr. l e v . 3 • O • O A • x A x A A ® X X X A A K A , X O' D D X . 27-26-25-24-L e g e n d A Pr. l e v . 1 X Pr. l e v . 2 • Pr. l e v . 3 •8 »H c £ 22-O OC CT, 2H _c "5 20- 15 20 Included A i r ( m L ) - i — 25 —r— 30 35