<%BANNER%>

Modeling Headspace Pressure in Retortable Packages

Permanent Link: http://ufdc.ufl.edu/UFE0045117/00001

Material Information

Title: Modeling Headspace Pressure in Retortable Packages
Physical Description: 1 online resource (55 p.)
Language: english
Creator: Paluri, Sravanti
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: canned -- flexible -- headspace -- modeling -- package -- pressure -- processing -- thermal
Chemical Engineering -- Dissertations, Academic -- UF
Genre: Chemical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Flexible packages, a new attraction in the market for shelf-stable canned foods, are more susceptible to damage during thermal processing for sterilization. In order to maintain the integrity of the package, it is important to keep the pressure differential across the package walls at a minimum. In this study,experiments were carried out to predict internal pressure in the headspace of a rigid pressure-tight module that withstood variations in temperatures and pressures. A stainless steel module was custom-made with a reseal-able lid,provisions for thermocouples, an inlet valve from which to pull vacuum, and a dial gauge to measure headspace pressure. The module was instrumented so as to measure and record the temperature profile at two different locations, one near the top and one near the bottom of the module, while undergoing thermal processing in a retort. For each set of experiments, three food systems were used to fill the module; pure distilled water, saline solution (5%) and sucrose solution (10%). A mathematical model was used to predict the headspace pressure profile in response to two experimentally measured temperature profiles (near the top and bottom). Results from these studies were compared with results from earlier work to account for the error between experimental and predicted headspace pressure profiles found in that work. This will lead to better control and knowledge of over pressures required during thermal processing to minimize differential pressure, thus avoiding damage to the flexible packaging and maintaining the quality and safety of food. Results showed that differences between predicted pressures from top and bottom temperatures and differences between measured and predicted pressures found in the previous work followed similar patterns and were of the same order of magnitude. These findings supported the hypothesis that disagreement between measured and predicted pressures found in the previous work was likely due to the reason that the temperatures used to predict pressures were measured at the bottom of the module, while the measured pressure was generated in response to headspace temperature at the top of the module.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Sravanti Paluri.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Svoronos, Spyros.
Local: Co-adviser: Teixeira, Arthur A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0045117:00001

Permanent Link: http://ufdc.ufl.edu/UFE0045117/00001

Material Information

Title: Modeling Headspace Pressure in Retortable Packages
Physical Description: 1 online resource (55 p.)
Language: english
Creator: Paluri, Sravanti
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: canned -- flexible -- headspace -- modeling -- package -- pressure -- processing -- thermal
Chemical Engineering -- Dissertations, Academic -- UF
Genre: Chemical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Flexible packages, a new attraction in the market for shelf-stable canned foods, are more susceptible to damage during thermal processing for sterilization. In order to maintain the integrity of the package, it is important to keep the pressure differential across the package walls at a minimum. In this study,experiments were carried out to predict internal pressure in the headspace of a rigid pressure-tight module that withstood variations in temperatures and pressures. A stainless steel module was custom-made with a reseal-able lid,provisions for thermocouples, an inlet valve from which to pull vacuum, and a dial gauge to measure headspace pressure. The module was instrumented so as to measure and record the temperature profile at two different locations, one near the top and one near the bottom of the module, while undergoing thermal processing in a retort. For each set of experiments, three food systems were used to fill the module; pure distilled water, saline solution (5%) and sucrose solution (10%). A mathematical model was used to predict the headspace pressure profile in response to two experimentally measured temperature profiles (near the top and bottom). Results from these studies were compared with results from earlier work to account for the error between experimental and predicted headspace pressure profiles found in that work. This will lead to better control and knowledge of over pressures required during thermal processing to minimize differential pressure, thus avoiding damage to the flexible packaging and maintaining the quality and safety of food. Results showed that differences between predicted pressures from top and bottom temperatures and differences between measured and predicted pressures found in the previous work followed similar patterns and were of the same order of magnitude. These findings supported the hypothesis that disagreement between measured and predicted pressures found in the previous work was likely due to the reason that the temperatures used to predict pressures were measured at the bottom of the module, while the measured pressure was generated in response to headspace temperature at the top of the module.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Sravanti Paluri.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Svoronos, Spyros.
Local: Co-adviser: Teixeira, Arthur A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0045117:00001


This item has the following downloads:


Full Text

PAGE 1

1 MODELING HEADSPACE PRESSURE IN RETORTABLE PACKAGES By SRAVANTI PALURI A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Sravanti Paluri

PAGE 3

3 To both my grandfathers Dr. Paluri Bhimeswara Rao and Dr. Yerrapragada V enkata Giri Satyanarayana Murti

PAGE 4

4 ACKNOWLEDGMENTS I am thankful to Dr. Arthur Teixeira, Professor, Department of Agricultural and Biological Engineering, University of Florida, for his guidance and for giving me an opportunity to apply my chemical engineering background to challenging new areas in food engineering. I am also extreme ly grateful to Dr. S pyros S voronos, Professor, Department of Chemical Engineering, University of Florida, for his guidance and for helping me to take all the necessary courses to better this p roject. I thank Dr. Bruce Welt and Dr. Damodaran for their helpf ul inputs on the project. I am very grateful to Dr. Weihua Yang, Department of Food Science and Human Nutrition, University of Florida for providing me desk space and access to the Food Science and Human Nutrition pilot plant. I thank Brandon Doan, undergr aduate student, Department of Agricultural and Biological Engineering, for designing, and, fabricating the pressure tight module for the experimental studies. I wish to thank my friends at Food Science, Cheryl, Akshay, Braulio, Bhaskar, and, Tara for their tremendous contribution and support both morally and academically towards the completion of this project. I am also grateful to my friends at Agricultural Engineering, Diana, Sandra, and, Marc for their help and support. I am thankful to my family of clo se friends, Lekha, Ravi, Karishma, Monica, and, Devanshi for always believing in me, encouraging me, and inspiring me to realize my dreams. Lastly and most importantly, I am eternally indebted to my parents, Nataraj and Satya valli and my sister, Sravya fo r all the joy on my ha ppy days and the strength in my bad.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 13 Background and Justification ................................ ................................ .................. 13 Previous Work ................................ ................................ ................................ ........ 14 Objectives ................................ ................................ ................................ ............... 16 2 METHODS AND PROCEDURES ................................ ................................ ........... 19 Task 1 ................................ ................................ ................................ ..................... 19 Design of Module ................................ ................................ ............................. 19 Fabrication and Instrumentation of Module ................................ ...................... 19 Testing the Module ................................ ................................ ........................... 20 Task 2 ................................ ................................ ................................ ..................... 20 Experimental Set up ................................ ................................ ......................... 21 Measuring and Recording Data ................................ ................................ ........ 22 Experimental Procedure ................................ ................................ ................... 22 Retort operation ................................ ................................ ......................... 22 Data acquisition ................................ ................................ ......................... 22 Thermocouple calibration ................................ ................................ ........... 23 Design and Execution of Experiments ................................ .............................. 23 Distilled water ................................ ................................ ............................. 23 Saline solution (5%) ................................ ................................ ................... 24 Sucrose solution (10%) ................................ ................................ .............. 24 Task 3 ................................ ................................ ................................ ..................... 24 Model for Distilled Water ................................ ................................ .................. 24 Model for Saline Solution (5%) ................................ ................................ ......... 25 Model for Sucrose Solution (10%) ................................ ................................ .... 26 Task 4 ................................ ................................ ................................ ..................... 27 3 RESULTS ................................ ................................ ................................ ............... 32 Distilled Water ................................ ................................ ................................ ......... 32

PAGE 6

6 Saline Soluti on (5%) ................................ ................................ ............................... 34 Sucrose Solution (10%) ................................ ................................ .......................... 35 4 CONCLUSIONS ................................ ................................ ................................ ..... 51 5 SUGGESTI ONS FOR FUTURE STUDY ................................ ................................ 53 LIST OF REFERENCES ................................ ................................ ............................... 54 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 55

PAGE 7

7 LIST OF TABLES Table pa ge 2 1 Initial product and process conditions for retort experiments with rigid pressure tight module ................................ ................................ ......................... 28

PAGE 8

8 LIST OF FIGURE S Figure page 1 1 Comparison of measured and predicted internal pressure profiles for distilled water by Ghai et al. (2011) ................................ ................................ ................. 17 1 2 Comparison of measured and predicted internal pressure profiles for saline solution (5%) by Ghai et al. (2011) ................................ ................................ ..... 17 1 3 Comparison of measured and predicted internal pressure profiles for sucrose solutio n (10%) by Ghai et al. (2011) ................................ ................................ ... 18 2 1 Stainless steel module used to record temperatures during thermal processing in the retort ................................ ................................ ................................ ......... 29 2 2 Experimental set up for the thermal processing runs ................................ ............ 29 2 3 Still steam retort at the Food Science and Human Nutrition pilot plant, University of Florida ................................ ................................ ............................ 30 2 4 Omega 56 data acquisition system used to measure temperatures inside the module ................................ ................................ ................................ ................ 30 2 5 Calibration of thermocouples used to measure temperatur es during thermal processing runs ................................ ................................ ................................ .. 31 3 1 Measured temperatures versus time at top and bottom of pressure tight module during retort process for distilled water by Paluri ................................ ... 36 3 2 Top and bottom temperature profiles along with predicted pressure profiles in response to top and bottom temperature profiles for distilled water by Paluri .... 37 3 3 Predicted and experimentally measured pressure profiles by Ghai et al. (2011) for distilled water ................................ ................................ ................................ 38 3 4 Predicted pressure profiles from top and bottom temperature s by Paluri and predicted and measured pressure profiles by Ghai et al. (2011) for distilled water ................................ ................................ ................................ ................... 39 3 5 Comparison of difference in pressure profiles predicted from the top and bottom temp eratures (Delta Paluri) with differences between predicted and measured profiles reported by Ghai et al. (2011) (Delta Ghai) for distilled water ................................ ................................ ................................ ................... 40 3 6 Measured temperatures versus time at t op and bottom of pressure tight module during retort process for saline solution (5%) by Paluri .......................... 41

PAGE 9

9 3 7 Top and bottom temperature profiles along with predicted pressure profiles in response to top and bottom temperature profiles for saline solution (5%) by Paluri ................................ ................................ ................................ .................. 42 3 8 Predicted and experimentally measured pressure profiles by Ghai et al. (2011) for saline solution (5%) ................................ ................................ ....................... 43 3 9 Predicted pressure profiles from top and bottom temperatures by Paluri and predicted and measured pressure profiles by Ghai et al. (2011) for saline solution (5%) ................................ ................................ ................................ ....... 44 3 10 Comparison of difference in pressure profiles predicted from the top and bottom temperatures (Delta Paluri) with differences between predicted and measured profiles reported by Ghai et al. (2011) (Delta Ghai) for saline solution (5%) ................................ ................................ ................................ ....... 45 3 11 Measured temperatures versus time at top and bottom of pressure tight module during retort process for sucrose solution (10%) by Paluri ..................... 46 3 12 Top and bottom temperature profiles along with predicted pressure profiles in response to top and bottom temperature profiles for sucrose solution (10%) by Paluri ................................ ................................ ................................ ............. 47 3 13 Predicted and experimentally measured pressure profiles by Ghai et al. (2011) for sucrose solution (10%) ................................ ................................ .................. 48 3 14 Predicted pressure profiles from top and bottom temperatures by Paluri and predicted and measured pressure profiles by Ghai et al. (2011) for sucrose solution (10%) ................................ ................................ ................................ ..... 49 3 15 Comparison of difference in pressure profiles predicted from the top and bottom temperatures (Delta Paluri) with differences between predicted and measured profiles reported by Ghai et al. (2011) (Delta Ghai) for sucrose solution (10%) ................................ ................................ ................................ ..... 50

PAGE 10

10 LIST OF ABBREVIATIONS CRT Constant Retort Temperature DAQ Data Acquisition System EHEDG European Hygiene Engineering and Design Group FDA Food and Drug Administration MIG Mercury in Glass IAPWS International Association for the Properties of Water and Steam USDA United States Department of Agricul ture VRT Variable Retort Temperature

PAGE 11

11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science MODELING HEADSPACE PRESSURE IN RETORTABLE PACKAGES By Sravanti Paluri December 2012 Chair: Spyros A. Svoronos Cochair: Arthur A. Teixeira Major: Chemical Engineering Flexible packages, a new attraction in the market for shelf stable canned foods are more susceptible to damage during thermal p rocess ing for sterilization. I n order to maintain the integrity of the package, it is important to keep the pressure differential across the package walls at a minimum In this study, experiments were carried out to predict internal pressure in the headspa ce of a rigid pressure tight module that withstood variations in temperatures and pressures A stainless steel module was custom made with a reseal able lid, provisions for thermocouples, an inlet valve from which to pull vacuum, and a dial gauge to measur e headspace pressure. The module was instrumented so as to measure and record the temperature profile at two different locations, one near the top and one near the bottom of the module, while undergoing thermal processing in a retort. For each set of exper iments, t hree food syste ms were used to fill the module; pure distilled water, saline solution (5%) and sucrose solution (10%) A m athematical m odel was used to predict the headspace pressure profile in response to two experimental ly measured temperature p rofiles (near the top and bottom) Results from these studies were compared with results from earlier work to

PAGE 12

12 account for the error between experimental and predic ted headspace pressure profiles found in that work. This will lead to better control and know ledge of overpressures required during the rmal processing to minimize differential pressure, thus avoiding damage to the flexible packaging and maintaining the quality and safety of food. Results showed that differences between predicted pressures from top and bottom temperatures and differences between measured and predicted pressures found in the previous work followed similar patterns and were of the same order of magnitude. These findings supported the hypothesis that disagreement between measured and p redicted pressures found in the previous work was likely due to the reason that the temperatures used to predict pressures were measured at the bottom of the module, while the measured pressure was generated in response to head space temperature at the top of the module.

PAGE 13

13 CHAPTER 1 INTRODUCTION Background and Justification The technology of packaging food in cans, jars and other rigid containers now has to compete greatly with shelf stable foods in flexible packages. Flexible packages offer the advantages of lower storage space and lighter weight as compared to rigid glass or metal containers which are hard and heavy They may be in the form of wraps, bags, pouches and sacks. An important attribute of flexible packaging is its ability to form thinner, lighte r, more compact packages (Selke et al., 2001). Products in flexible packages require significantly less heating time to achieve commercial sterility. Thus, such products are fresher, superior in quality and require lower processing cost (Brody, 2002). Whil e being processed under high temperature and high pressure conditions in the retort, the flexible package may suffer damage to its walls and seal. Due to its delicate nature, careful control of overriding air pressure is needed during retort processing Du ring thermal processing in the retort, internal pressure is b uilt up in the headspace of sealed packages. This internal pressure is caused by int ernal water vapor and entrapped gases expanding in response to increasing temperature. At high temperatures, th e solubility of gases in the solution decreases and they are released into the headspace of the package. The entrapped gases in the headspace consist primarily of the released gases and residual air after vacuum. This internal pressure may be greater than the saturation pressure of the steam in the retort. If the internal pressure is greater, the package may swell or burst, and if it is lesser, the package will indent or collapse. The increasing internal pressure in a flexible package must be

PAGE 14

14 continuously c ounterbalanced by providing external pressure This external pressure can be provided by the introduction of compressed air along with saturated steam in the retort. The delicate nature of a flexible package undergoing thermal processing requires precise c ontrol of overriding air pressures, and is of critical importance. In order to keep the pressure differential across the package at a minimum, internal pressures must be known or estimated. In most cases, a trial and error approach is followed to determine optimum overriding air pressures. Mathematical modeling of internal headspace pressure could reduce this element of uncertainty as well as time and cost, thus, becoming a useful tool to predict optimal overriding air pressures during thermal processing. T he quest for such models has been the subject of much previous work. Previous Work Chen et al. (2004) developed a means to choose a reasonable temperature profile for processing, called multiple ramp variable retort temperature control (VRT). Alternately, a constant retort temperature (CRT) approach by Salgado et al. (2011) or C hen et al. (2002) could be adapted to save process times and preserve quality of food. Various methods to control retort temperatures are real time data acquisition, on line correcti on factors, mathematical heat transfer models described in the work by Simpson et al. (2007) and Teixeira et al. (1997). However, these studies do not address the damage caused on the package wall and seal due to internal pressures built up in the headspac e of the package. Pascal et al. (2002) addressed the effects of the pressure differential on seal strength and sensitivity of package wall to changes in headspace volume. But,

PAGE 15

15 predictions of pressures inside the package that raised concerns of seal strengt h and package wall sensitivity were not made. Ghai et al. (2011) attempted to develop a mathematical model to predict internal pressure in the headspace of a rigid pressure tight module during thermal proc essing. This model predict ed internal headspace pre ssure in response to the temperature of the product under specified initial and final conditions. The equation of the I nternational Association for the Properties of Water and Steam (IAPWS) was us ed to estimate vapor pressure of water and was coupled with The Ideal Gas Law served as basis for estimating headspace pressure caused by expansion of non condensable gases as a function of temperature and specified initial and boundary conditions. Results from this wo rk revealed some disagreement between predicted and measured pressures during retort come up time as can be seen in figure s 1 1, 1 2, 1 3 for distilled water, saline solution (5%) and sucrose solution (10%). Moreover in the c ase of saline solution (5%), e xperimentally measured pressure profile appeared as a ramp of increasing pressure over time. This was attrib uted to the likelihood that A luminum in the construction material of the module reacted with salt inside the module producing Hydrogen gas This gas increased internal pressure over time, and could not be accounted for in the mathematical model. One hypothesis proposed in our study was that the disagreement between the measured and predicted pressures by Ghai et al. (2011) in the case of saline soluti on could be due to the material (Aluminum) from which the module was fabricated.

PAGE 16

16 A second hypothesis was t hat the disagreement during retort come up time was caused by the positioning of the temperature sensor ( bottom of the module ) while pressure buildup was driven by temperature in the headspace at the top of the module. Objectives The objectives of this work were to do the following: 1. Arrange for design, fabrication and testing of a new pressure tight module with which to measure and record internal temp erature profiles at the top and bottom of the module while it underwent thermal processing in a retort. The new module was fabricated to the same shape and dimensions as that used by Ghai et al. (2011) but with Stainless Steel 2. Measure and record tempera tures at two different locations inside the module (top and bottom) during thermal processing runs in the retort for each of three food systems; pure distilled water, saline solution (5%), and, sucrose solution (10%). 3. Predict internal pressure profiles in response to the top and bottom temperature profiles using the mathematical model developed by Ghai et al. (2011). 4. Compare the difference in pressure profiles predicted from the top and bottom temperature profiles with differences between predicted and meas ured profiles reported by Ghai et al. (2011)

PAGE 17

17 Figure 1 1. Comparison of measured and predicted internal pressure profiles for distilled water by Ghai et al. (2011) Figure 1 2. Comparison of measured and predicted internal pressure profiles for s aline solution (5%) by Ghai et al. (2011)

PAGE 18

18 Figure 1 3. Comparison of measured and predicted internal pressure profiles for sucrose solution (10%) by Ghai et al. (2011)

PAGE 19

19 CHAPTER 2 METHODS AND PROCEDURES To meet the above objectives, a scope of work was u ndertaken which consisted of the following four tasks: Task 1 The first task was to design, fabricate and test a rigid pressure tight module for measuring and recording internal temperature data during thermal processing. Design of Module A pressure tight cylindrical module was fabricated in the Agricultural and Biological Engineering Department machine shop at the University of Florida. The engineering design of the module was undertaken with the assistance of an undergraduate engineering student as a seni or design project. The material of construction was chosen to be stainless steel (food grade) to avoid chemical reactions with the contents at elevated temperatures, and to comply with United States ng standards and with those of the European Hygiene Engineering and Design Group (EHEDG). The module was developed to be pressure tight and had a standard factor of safety of 3 or greater. This ensured that the module would have to be pressurized greater t han 3 times its nominal operating value in order for it to fail. Fabrication and Instrumentation of Module The module (figure 2 1) consisted of two separate pieces, one cylindrical body and a separate circular lid piece. The capacity of the module was 430 ml. The cylindrical body had an inner diameter of 7.1 cm, a height of 10.9 cm and a wall thickness of 0.4 cm. It also contained two holes in the side wall, one near the top and th e other near the

PAGE 20

20 bottom to accommodate thermocouples. Two female ends of K ty pe thermocouples were inserted horizontally through the holes in the body of the module to reach the center line and permanently sealed. One of these was 0.6 cm from the top of the body and the other was 1.0 cm from the bottom, leaving the two thermocouple s 8.6 cm apart. The bottom cylinder also had a flange with an outer diameter of 10.0 cm and a wall thickness of 0.4 cm, to accommodate bolts on the lid for a pressure tight seal. The lid was fabricated to fit over the body with an O ring gasket. It had six holes drilled symmetrically along the outer diameter. Six bolts secured the lid to the body in a pressure tight manner. The lid also had two holes in the center to accommodate a dial gauge to measure internal pressure and an inlet valve through which to d raw vacuum. An Ashcroft pressure gauge was installed to easily check for the pressure tight nature of the lid. The vacuum inlet had a stainless steel ball valve leading to a temporary connection with a tube to pull vacuum into the module. Testing the Modu le The module was first tested on the bench top to ensure proper functionality and no leakage. It was filled with distilled water, sealed under vacuum and kept on the bench for over two hours to check for a constant pressure reading and no loss in vacuum. It was necessary to check that the module was pressure tight and indicated a constant value of pressure without slowly increasing to atmospheric pressure. Upon confirmation of its leak proof nature, the module was declared fit for experimental runs. Task 2 The second task was to measure and record internal temperatures during triplicate pilot plant retort runs for each of the three aqueous solutions, pure distilled water or saline solution (5%) or sugar solution (10%), for each set of experiments.

PAGE 21

21 Experimen tal S et up The experimental set up consisted of the module, vacuum pump and steam retort figure 2 2) Module. A constant headspace of 8mm was marked on the inside wall of the module and the aqueous solution was filled up to this point for each run. The vol ume of solution in each run was 410ml. Prior to each experimental run, the solution was weighed and measured for exact volume. The lid was bolted on to the module with the six bolts. Vacuum pump The ball valve for drawing vacuum was opened and vacuum was drawn with a 1/3 horse power Emerson vacuum pump to 15 inches Hg (~0.5 atm or 50.8 kPa). The valve was closed when desired vacuum level was reached. Steam retort The module was placed inside a pilot scale still cook steam retort (figure 2 3) in the Food Science and Human Nutritio n Department Pilot Plant, at University of Florida. The retort was equipped with a pressure tight packing gland on the side wall to accommodate lead cables for thermocouples to pass through the retort. It was also equipped with a Mercury in Glass (MIG) thermometer and a pressure gauge to enable external monitoring of temperatures and pressures as per FDA regulations. There was a drain at the bottom right corner for draining condensate during come up and water during cooling period. The lid of the retort was secured by bolting down six fastening lugs in diagonally opposite pairs. A bleeder on the top of the lid was kept open throughout the operation to ensure a constant flow of steam as per safe operational procedures of thermal proc essing given by FDA. The retort contained an inlet for cooling water which was sprayed inside the retort through a sparger under the lid, as well as an inlet for compressed air. The steam inlet was at the bottom of the retort

PAGE 22

22 through spreaders, and the ste am flow control valve was automatically controlled by a pneumatic temperature recorder/controller mounted on the wall behind the retort. The process temperature chosen for retort operation was set on the recorder. The recorder controlled the amount of stea m under pressure and automatically kept the retort at the constant set temperature till the set point was manually changed. Measuring and Recording D ata A total of three lead cables passed through the packing gland on the retort, two for thermocouples insi de the can and one for the thermocouple in the retort. Inside the retort, the male ends of two thermocouples were inserted into the female ends fitted on the module and one thermocouple was hung on a fixture inside the retort to measure and record retort t emperatures. Outside the retort, the thermocouple lead wires were connected to an Omega 56 data acquisition system (DAQ) ( figure 2 4) which in turn was connected to a computer containing the necessary software to capture and record temperature data in rea l time. Experimental P rocedure Retort o peration After placing the filled and sealed module inside the retort, all thermocouple lead wires were connected to their respective ports on the DAQ Omega 56 before closing the retort. The thermal processing test w as run for 10 minutes at 121 C (250 F). After which, steam was shut off and cooling cycle was begun. Data acquisition Once the retort was ready for processing, the Omega 56 DAQ was initialized to acquire temperature data in real time. A graph of temperatu re versus time opened and real time reading of temperatures were observed during the process to ensure that

PAGE 23

23 desired process conditions were being achieved. At the end of cooling, the DAQ was stopped and data files were saved with appropriate batch numbers indicating the type of product and the date of the run. Thermocouple calibration Prior to executing the experiments, it was necessary to calibrate all thermocouples. To do this, an experiment was run with all connections described above. But, the module wa s placed in the retort, open and empty, to record the saturated steam conditions in the retort. The three thermocouples, two inside the module and one hung on the interior wall of the retort were exposed to the steam in the retort. The experiment was run w ith steam, and data were collected to note temperature readings for all thermocouples during the come up, holding and cooling periods. The readings and errors (if any) were recorded to calibrate the thermocouples in order to record the same temperat ures at the same time intervals (figure 2 5) Design and Execution of Experiments Three experimental replicate runs on each aqueous solution food system were carried out, with initial conditions (table 2 1 ) for each system as follows: Distilled water Distilled wa ter was used first, being the simplest food system, and as the baseline for experimental runs. In a clean beaker, about 500ml distilled water was brought to boil to evaporate dissolved gases. The water was allowed to cool down and 410ml was collected in a beaker. Exact volume, weight and initial temperature of the water were measured and recorded before being added to the module.

PAGE 24

24 Saline solution (5%) Table salt (food grade) from a local supermarket was used for making the saline solution. Considering a pre decided volume of solution as 410ml, calculations were made to determine the amount of salt to be dissolved to keep the initial concentration as 5% salt solution. Excess amount of distilled water was boiled to evaporate dissolved gases and cooled. The requ ired amount was poured into a beaker and the calculated amount of salt was added to it and dissolved. Exact volume, weight and initial temperature of the solution were measured and recorded before being added to the module. Sucrose solution (10%) Powdered sugar (food grade) from a local supermarket was used for making the sucrose solution (10%) in a similar manner as described above for saline solution. Task 3 Th e third task was to predict internal pressure profiles in response to the recorded temperature p rofiles using the mathematical model developed by Ghai et al. (2011) for each of the three above aqueous solutions. This model was designed to predict the internal headspace pressure as a sum of the pressures exerted by the vapor and the gaseous phases: The model is described below: Model for Distilled W ater The internal pressure of plain distilled water was the sum of the pressure exerted by water vapor and the pressure exerted by the gas phase.

PAGE 25

25 The vapor pressure of pure water was a function of real time temperature and was calculated using the expression taken from the International Association for the Properties of Water and Steam (IAPWS) adopted in 1995: Log e (P w / 22.064*10 6 ) = 647.096/T *( 7.85951783*v + 1.8 4408259*v 1.5 11.7866497*v 3 + 22.6807411*v 3.5 15.9618719*v 4 + 1.80122502*v 7.5 ) with temperature (T) in Kelvin and pressure (P w )in Pascals and, v = 1 T/647.096, where the value 647.096 K is the triple point of water. The pressure exerted by the gaseous phase for a reversible adiabatic (Isentropic) process with a specific heat ratio of 1.35 as a function of the initial pressure, and the init ial and final temperatures is : P 2 = P 1 (T 2 / T 1 ) 3.86 Thus, the total predicted pressure profile of pure distilled water was the addition of the pressures exerted by the vapor and the gaseous phases: Model for S aline Solution (5%) The mathematical model for predicting the inter nal pressure of saline solution was the sum of pressure exerted by the vapors of the solut ion, as a function of real time temperature; and pressure exerted by the gas phase as a function of initial and final (real time) temperatures. The pressure exerted by water vapors from the solution as a function of temperature was calculated by applying R volatile solute is equal to the vapor pressure of the pure solvent at that temperature multiplied by its mole fraction. In equation form, this reads:

PAGE 26

26 P = x solv P solv where P solv is the vapor pressure of the pure solvent at a particular temperature, and x solv is the mole fraction of the pure solvent in the solution The mole fraction of water was calculated in the saline solution and was multiplied by the vapor pressure data of pure water calculated at a particular temperature using the IAPWS formulation to get the final vapor pressure Since, the presence of a non volatile solute had no effect on the pressure exerted by the gaseous phase; it was calcu lated in a way similar to distilled water. Pressures exerted by the vapor and the gaseous phases were then added to get the final predicted internal pressure profile of the saline solution (5%) as a model food system. Model for Sucrose Solution (10%) The total internal pressure of the sucrose solution, as a food product system, was calculated using a method similar to that of saline solution, but changing the mole fraction of the non volatile solute in the solution. With the model, headspace pressure pr ofiles were predicted in response to internal temperature profiles Two pressure profiles were obtained for each fo od system, one calculated with temperature profile of the thermocouple at the top of the module and one at the bottom. Plots were made for th eoretical pressure profiles, two plots for each food system for comparison studies in the next task.

PAGE 27

27 Task 4 The final task was to compare predicted profiles of headspace pressure obtained in response to temperatures at the top with those from the bottom of the module, while it was being thermally processed. The two pressure profiles obtained were plotted on the same graph to observe the difference in values of predicted pressures at any given time during come up time for each solution. Particular attention was given to observing the pressure differences during initial come up time and comparing them with the differences reported by Ghai et al. (2011) during the same process. This task was carried out by following these three steps: 1. Predicted pressure profil es were plotted in r esponse to top and bottom temperatures. The pressure differences were determined at points in time during the come up period. 2. The data obtained by Ghai et a l. (2011) were plotted to show the differences between predicted and experimenta lly measured pressures they obtained. 3. The above two sets of pressure profiles differences were compared to determine if these differences followed a similar pattern, and could offer explanation for differences between measured and predicted pressures in th e previous work by Ghai et al. (2011).

PAGE 28

28 Table 2 1. Initial product and process conditions for retort experiments with rigid pressure tight module Product Volume (milliliters) Weight (grams) Headspace (millimeters) Vacuum drawn in the headspace (kiloPasca ls) Initial Temperature ( degree Celsius) Distilled Water 410 410 8 50.8 28 Saline solution (5%) 410 410 8 50.8 25 Sucrose solution (10%) 410 410 8 50.8 25

PAGE 29

29 Figure 2 1. Stainless steel module used to record temperatures during thermal proce ssing in the retort Figure 2 2. Experimental set up for the thermal processing runs

PAGE 30

30 Figure 2 3. Still steam retort at the Food Science and Human Nutrition pilot plant, University of Florida Figure 2 4. Omega 56 data acquisition system used t o measure temperatures inside the module

PAGE 31

31 Figure 2 5. Calibration of thermocouples used to measure temperatures during thermal processing runs

PAGE 32

32 CHAPTER 3 RESULTS Distilled Water Profiles of experimentally measured temperatures versus time were plotted in figure 3 1. Starting at an initial temperature of about 28 C in the solution, both temperatures increased steadily in the first eight minutes of the process, signifying the come up period, leveling out at a constant temperature at ten minutes. After they leveled, the retort was kept at this constant temperature for ten more minutes, called the holding period. When the holding period ended, the steam was shut off and cooling water was introduced to commence the cooling period. During the first ten minutes of the process, the come up period, both temperatures followed different trajectories, the temperature at the top consistently higher than the temperature at the bottom. The temperature at the top (black line) showed a steep increase fro m 29 C to 87 C after two minutes whereas the bottom temperature (grey line) only increased to 55 C during the same time interval. Note that four minutes into the process, top temperature had rocketed to 118 C, but, the bottom temperature was slower to rise reaching a value of only 103 C at four minutes. Finally, at ten minutes, temperatures at the top and bottom leveled off and stayed constant throughout the holding period. The following discussion focuses strictly on the first ten minutes (come up time). F igure 3 2 shows both top and bottom temperatures and predicted pressure profiles in response to those top and bottom temperature profiles versus time, for the system of pure distilled wat er. Predicted pressures in kilo Pascals were plotted on the primary ve rtical axis along with temperatures at the top and bottom from figure 3 1on the secondary vertical axis. The module was filled, sealed and a vacuum of 15 inch Hg

PAGE 33

33 was drawn, which corresponded to 0.5 atm or 7.3 psi or 50.8 kPa. For convenience, pressures pr edicted from top and bottom temperatures are denoted by pressure from top temperature (black line) and pressure from bottom temperature (grey line), respectively. At four minutes, the temperature at the top was about 118 C and the corresponding predicted p ressure was 326 kPa. Simultaneously, the bottom temperature, lesser than the top, was 103 C which produced a predicted pressure value of only 232 kPa. The difference between predicted pressures at a time instant of four minutes was 94 kPa when the differen ce in top and bottom temperatures was about 15 C. As compared to pressure from top temperature, the pressure from bottom temperature took longer to achieve constant values since the bottom temperature rose slower than the top. This variation in increasing pattern of predicted pressures from to p and bottom temperatures was due to the different temperature curves measured at the top and bottom of the module, as observed in figure 3 1. Figure 3 3 shows predicted and experimentally measured pressure profiles fo r distilled water by Ghai et al. (2011). Pressures were predicted by applying the mathematical model developed by Ghai et al. (2011) to their experimentally measured temperatures measured at the bottom of the module. The module they used was sealed at abou t 34.5 kPa of vacuum (lower than the 50.8 kPa used in present work). After the first minute, both measured pressure (black line) and predicted pressure (grey line), exhibited different curves, the former sharper than the latter. Measured pressure rocketed to 174 kPa at four minutes. Simultaneously, predicted pressure ascended to only 147 kPa. The significant difference in predicted and measured pressure profiles

PAGE 34

34 during the come up time in the work by Ghai et al. (2011) presented an opportunity of improvemen t in the current experimental work. Predicted pressure profiles from top and bottom temperatures by Paluri and predicted and measured pressure profiles by Ghai et al. (2011) are shown in figure 3 4. Differences between the values of predicted pressures fro m top and bottom temperatures by Paluri, and differences between measured and predicted pressures by Ghai et al. (2011) were denoted as Delta Paluri and Delta Ghai, respectively. The differences, Delta Paluri and Delta Ghai, were of the same order of magni tude. For example, at four minutes, Delta Paluri was 94 kPa and Delta Ghai was 27 kPa. Fi gure 3 5 shows the pattern of these pressure differences, Delta Paluri and Delta Ghai, over time. The delta pressure profiles, Delta Paluri and Delta Ghai, showed a si milar trend in their patterns. Differences in their actual values were attributed to the fact that retort operating and process conditions used by Ghai et al. (2011) differed from those used in this study. Saline Solution (5%) Similar to the case of distil led water, five figures were plotted for data analysis. Profiles of experimentally measured temperatures versus time were plotted in figure 3 6. The following discussion focuses strictly on the first ten minutes (come up time). Figure 3 7 shows both top an d bottom temperatures and predicted pressure profiles in response to those top and bottom temperature profiles versus time, for the system of saline solution (5%). At four minutes, the temperature at the top was about 106 C and the corresponding predicted pressure was 250 kPa. Simultaneously, the bottom temperature, lesser than the top, was 91 C which produced a predicted pressure value of only 181 kPa. The difference between predicted pressures at a time instant of four

PAGE 35

35 minutes was 68 kPa when the differen ce in top and bottom temperatures was about 15 C. As compared to pressure from top temperature, the pressure from bottom temperature took longer to achieve constant values since the bottom temperature rose slower than the top. This variation in increasing pattern of predicted pressures from top and bottom temperatures was a due to the different temperature curves measured at the top and bottom of the module, as observed in figure 3 6. F igure 3 8shows predicted and experimentally measured pressure profiles f or distilled water by Ghai et al. (2011). Measured pressure rocketed to 173 kPa at four minutes while predicted pressure ascended to only 141 kPa. Predicted pressure profiles from top and bottom temperatures by Paluri and predicted and measured pressure pr ofiles by Ghai et al. (2011) for saline solution (5%) are shown in figure 3 9. Differences between the values of predicted pressures from top and bottom temperatures by Paluri, and differences between measured and predicted pressures by Ghai et al. (2011) were denoted as Delta Paluri and Delta Ghai, respectively. The differences, Delta Paluri and Delta Ghai, were of the same order of magnitude. For example, at four minutes, Delta Paluri was 68 kPa and Delta Ghai was 32 kPa. Figure 3 10 shows the pattern of these pressure differences, Delta Paluri and Delta Ghai, over time. Similar to the case of distilled water, the delta pressure profiles, Delta Paluri and Delta Ghai, showed a similar trend in their patterns. Sucrose Solution (10%) Figure s 3 11, 3 12, 3 13 3 14, 3 15 contain plots for sucrose solution (10%) corresponding to the plots for distilled water and saline solution (5%). Similar analysis was made for all three aqueous solutions.

PAGE 36

36 Figure 3 1. Measured temperatures versus time at top and bottom of pressure tight module during retort process for distilled water by Paluri

PAGE 37

37 Figure 3 2. Top and bottom temperature profiles along with predicted pressure profiles in response to top and bottom temperature profiles for distilled water by Paluri

PAGE 38

38 Fig ure 3 3. Predicted and experimentally measured pressure profiles by Ghai et al. (2011) for distilled water

PAGE 39

39 Figure 3 4. Predicted pressure profiles from top and bottom temperatures by Paluri and predicted and measured pressure profiles by Ghai et al. (2011) for distilled water

PAGE 40

40 Figure 3 5. Comparison of difference in pressure profiles predicted from the top and bottom temperatures (Delta Paluri) with differences between predicted and measured profiles reported by Ghai et al. (2011) (Delta Ghai) for distilled water

PAGE 41

41 Figure 3 6. Measured temperatures versus time at top and bottom of pressure tight module during retort process for saline solution (5%) by Paluri

PAGE 42

42 Figure 3 7. Top and bottom temperature profiles along with predicted pressure profi les in response to top and bottom temperature profiles for saline solution (5%) by Paluri

PAGE 43

43 Figure 3 8. Predicted and experimentally measured pressure profiles by Ghai et al. (2011) for saline solution (5%)

PAGE 44

44 Figure 3 9. Predicted pressure profile s from top and bottom temperatures by Paluri and predicted and measured pressure profiles by Ghai et al. (2011) for saline solution (5%)

PAGE 45

45 Figure 3 10. Comparison of difference in pressure profiles predicted from the top and bottom temperatures (Delta Pa luri) with differences between predicted and measured profiles reported by Ghai et al. (2011) (Delta Ghai) for saline solution (5%)

PAGE 46

46 Figure 3 11. Measured temperatures versus time at top and bottom of pressure tight module during retort process for su crose solution (10%) by Paluri

PAGE 47

47 Figure 3 12. Top and bottom temperature profiles along with predicted pressure profiles in response to top and bottom temperature profiles for sucrose solution (10%) by Paluri

PAGE 48

48 Figure 3 13. Predicted and experimen tally measured pressure profiles by Ghai et al. (2011) for sucrose solution (10%)

PAGE 49

49 Figure 3 14. Predicted pressure profiles from top and bottom temperatures by Paluri and predicted and measured pressure profiles by Ghai et al. (2011) for sucrose sol ution (10%)

PAGE 50

50 Figure 3 15. Comparison of difference in pressure profiles predicted from the top and bottom temperatures (Delta Paluri) with differences between predicted and measured profiles reported by Ghai et al. (2011) (Delta Ghai) for sucrose solu tion (10%)

PAGE 51

51 CHAPTER 4 CONCLUSIONS The conclusions of this study are: The stainless steel module used for thermal processing runs in the retort avoided reactions with saline solution (5%). A white precipitate, Aluminum Chloride, which was originally formed in the studies by Ghai et al. (2011) when Aluminum of their module reacted with saline solution (5%), was absent in this study. Temperature profiles obtained from experimental measurements near the top and bottom of the module exhibited a difference in the ir heating patterns during come up time. The solution near the bottom of the module heated more slowly than that near the top, giving a temperature difference of about 15 C at the same instance in time. Pressure profiles predicted from top and bottom tempe rature profiles, using the mathematical model by Ghai et al. (2011), gave different values at the same instance in time during the come up period. Ideally, predictions of pressure in the headspace of the module must be similar. However, due to convectional currents and heating characteristics of the filled food system, temperatures recorded at top and bottom of the module gave two values significantly different from each other. Therefore, owing to the fact that the bottom of the module heated slower than th e top, the pressure profile predicted using bottom temperature was consistently lower than that predicted using top temperature. The above results showed that differences between predicted pressures from top and bottom temperatures by Paluri and difference s between measured and predicted pressures by Ghai et al. (2011) followed similar patterns and were of the same order of magnitude. Disagreement between measured and predicted pressures shown by Ghai

PAGE 52

52 et al. (2011) was likely due to the reason that the temp eratures used to predict pressures were measured at the bottom of the module, while the measured pressure was generated in response to headspace temperature at the top of the module.

PAGE 53

53 CHAPTER 5 SUGGESTIONS FOR FUTURE STUDY In further studies, the following suggestions could be adapted: The wireless pressure probe from studies by Ghai et al. (2011) could be used for measuring pressures in real time. The measured pressures could be compared with predicted pressures from the top and bottom temperatures in the m odule to determine the optimum location of the thermocouple. Accurate predictions of pressures could be made from mathematical model by Ghai et al. (2011) using temperatures measured at this optimum location. Unfortunately, there was no access to such a pr obe in present studies. It is also proposed that in follow up work, more headspace could be maintained in the rigid pressure tight module to be able to insert a pressure probe in the headspace and clearly monitor the internal pressure build up due to expan sion of gases in the ample headspace. lower predicted pressure during holding time in case of saline solution and sucrose solutions. This equation may be used to calculate the wa ter activity for non ideal systems like saline and sucrose solutions. Since the water activity of these solutions is factor for predicting gas pressures. Alternately, R for predicting internal pressure by Ghai et al. (2011).

PAGE 54

54 LIST OF REFERENCES Brody AL. 2002. Food Canning in the 21 st Century. Foo d Technology 56(3): 75 78 Chen CR, Ramaswamy HS. 2004. Modeling and optimization of constant retort temperature thermal processing using coupled neural networks and genetic algorithms. Journal of Food Process Engineering 25(5): 351 379. Chen CR, Ramaswamy HS. 2004. Multiple ramp variable retort temperature control for optimal thermal processing. Trans IChemE, Part C, Food and Bioproducts Processing, 82(c1): 78 88. Ghai G, Teixeira AA, Welt BA. 2011. Measuring and predicting head space pressure during retorting of thermally processed foods Journal of Food Science 76(3): e298 e308. Pascall MA. 2002. Ev aluation of a laboratory scale pressure differential (force/decay) system for non destructive leak detection for flexible and semi rigid packaging. Packaging Technology and Science 15: 197 208. Salgado D, Torres JA, Welti Chanes J, Velazquez G. 2011. Effec t of input data variability on estimations of the equivalent constant temperature time for microbial inactivation by HTST and retort thermal processing. Journal of Food Science 76(6): e495 e502. Selke SE, Hernandez RJ. 2001. P ackaging : Polymers in Flexible Packaging Encyclopedia of Materials Science and Technology, 2 nd Ed. Pergamon. Simpson R, Teixeira A, Almonacid S. 2007. Advances with intelligent on line retort control and automation in thermal processing of canned foods. Food Control 18(7): 821 833. Teixeira AA, Tucker GS. 1997. On line retort control in thermal sterilization of canned foods. Food Control 8(1): 13 20.

PAGE 55

55 BIOGRAPHICAL SKETCH Sravanti Paluri graduat ed from the University of Mumbai in 2011 with a ngineering. In Mumbai, she was a part of a yearlong project in food and biochemical engineering which inspired her to pursue food e ngineering in higher studies. She moved to the University of Florida Gainesville in the fall of 2011 to obtain a Master of Science degree in c hemical e ngineering with a minor from the Department of A gricultural and Biological E ngineering. She will graduate in December 2012 with a thesis and speciali zation in food e ngineering and want s to take up further research work in this area. With academic and hands on exp erience in rheology, mechanics, sterilization, science and processing of food, her next step is to become a doctoral candidate at a good unive rsity in the United States. She believes her extensive research work, short projects, familiarity with H azard Analysis and Critical Control Point s system and Better Process Control School certification by the University of California, Davis make her an in novative food engineer.