CHLORINE DIOXIDE EFFICACY AGAINST BOTRYTIS CINEREA IN
STRAWBERRIES AND CHLORINE DIOXIDE ABSORPTION BY FRESH PRODUCE
By
Juliana Arango

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Packaging – Master of Science
2013

ABSTRACT
CHLORINE DIOXIDE EFFICACY AGAINST BOTRYTIS CINEREA IN
STRAWBERRIES, AND CHLORINE DIOXIDE ABSORPTION BY FRESH PRODUCE
By
Juliana Arango
Chlorine dioxide (ClO2) is an effective sanitizer and strong bactericide with a
broad spectrum disinfection capability. As a result of ClO2 gas treatments on fresh
produce, it is absorbed by the produce tissue and oxidized into different compounds
−

−

such as chlorate (ClO3 ) and chlorite (ClO2 ) ions upon contact.
Quantification of chlorine dioxide exposure and consumption by fresh produce is
critical for determining the target dose and assessing the residual species post
treatment as they could affect the safety and organoleptic profile of the treated fresh
produce. A new in situ approach to measure the total exposure to the produce as well
as consumption of ClO2 gas by produces was developed by using a system equipped
with a UV-vis spectrophotometer. The equipment was used to evaluate the effect of
ClO2 gas on Botrytis cinerea present in strawberries. The impact of high-concentrationshort-time and low-concentration-long-time of ClO2 gas exposure on the efficacy of
ClO2 on Botrytis cinerea as well as absorption of ClO2 by the strawberries was
evaluated. It was found that both exposure time and concentration were significant
factors in the ClO2 consumption by fresh produce and very fast consumption kinetics in
strawberry fruit were found.

ACKNOWLEDGEMENTS

My sincerest thanks to Dr. Rubino for her help, her patience and her time to listen
to the many questions and ideas I had throughout the execution of the project; and very
especially for her advice and suggestions that transcended the academic spheres and
reflected her true care for both my professional and personal development. To Dr.
Auras for explaining generously and thoroughly the many complex statistical challenges
that we faced, and for pushing me to achieve goals I thought impossible, for helping me
think differently and for always questioning the status quo. To Dr. Kijchavengkul for
giving me the pleasure of working with him, even if for just a couple of weeks; and for
sharing with me his ability in programming and electronics: you made it all seem very
easy and logical. Working on this project and writing this thesis has probably been one
of the most challenging activities that I have faced, but thanks to all of you it has
become a reality. I’d also like to thank the Department of Biosystems and Agricultural
Engineering from Michigan State University, and more specifically to Michael James for
always being in a good mood and for his readiness to help and propose ideas to do
things easier, and safer, at the pilot plant. A special thanks to Jerri Gillett and to her son
for their discipline and structure, for the many hours we spent inside the laboratory
running tests with strawberries and chlorine dioxide gas, for being open, honest and
always ready to help despite the difficulties that came along. I appreciate your help a lot
since without you I couldn’t have done it. To Fabianne Staschower for “initiating me” in
the chlorine dioxide world, for her tolerance with my basic chemistry questions, for
answering the phone on the weekends, and for always being in a good mood regardless

iii

of her workload. I would also like to thank the Center for Statistical Training &
Consulting for offering me the opportunity to access highly educated professionals in
the field like Juan David Munoz who guided me into the right the direction to analyze the
data.
Thank you also to my family: to my parents and my sisters for believing in me
and for supporting my decision to go back to school even though it meant that we had to
be separated for a while. Thank you for the Skype sessions, for the emails, for the calls
and for your visits to Michigan while I studied here. It was always very comforting to
know you were there watching out for my success.
My decision to come to Michigan State University was taken in junction with my
best friend and husband Nicolas Acevedo. Thank you for taking this step with me, thank
you for always being there, and for studying by my side until late at night during these
two years and for being such a role model for me with your discipline and perseverance.
This whole journey wouldn’t have been nearly as exciting or easy as it was thanks to
your company and enthusiasm!
Lastly, how not to thank both Brittany Burns and Alix Grabowski for welcoming
me into their city, for explaining to me so many different things about the American
culture, the weather, the celebrations, the mitten state, etc.; thanks for your friendship,
for your tolerance, for making me laugh hard; for listening to me, for your support and
advice; and for making this crazy journey a remarkable experience.

iv

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................ viii
LIST OF FIGURES......................................................................................................... ix
1.

INTRODUCTION ...................................................................................................... 1
BIBLIOGRAPHY ...................................................................................................... 5

2. LITERATURE REVIEW ............................................................................................ 9
2.1. Framework................................................................................................ 9
2.2. Food quality and safety ......................................................................... 10
2.2.1. Pathogenic and spoilage microorganisms associated with fresh
produce ................................................................................................ 13
2.2.1.1. Salmonella ...................................................................................... 13
2.2.1.2. Staphylococcus aureus ................................................................... 13
2.2.1.3. Esterichia Coli ................................................................................. 14
2.2.1.4. Listeria monocytogenes .................................................................. 14
2.2.1.5. Botrytis cinerea ............................................................................... 15
2.3. Microbial growth models....................................................................... 18
2.3.1. Primary models .................................................................................... 18
2.3.1.1. Gompertz Model.............................................................................. 19
2.3.1.2. Logistic function .............................................................................. 20
2.3.2. Secondary models................................................................................ 21
2.3.2.1. Ratkowski square-root model.......................................................... 21
2.3.2.2. Linear Arrhenius-Davey equation .................................................... 21
2.3.3. Tertiary models .................................................................................... 22
2.4. Strategies to prolong the shelf life and ensure safety of fresh
produce................................................................................................... 23
2.4.1. Modified atmosphere packaging (MAP) ............................................... 24
2.4.2. Irradiation ............................................................................................. 24
2.4.3. Antimicrobial packaging ....................................................................... 25
2.5. Common sanitizing agents ................................................................... 25
2.5.1. Chlorine ................................................................................................ 25
2.5.2. Hydrogen peroxide ............................................................................... 26
2.5.3. Ozone................................................................................................... 27
2.6. Chlorine Dioxide .................................................................................... 28
2.6.1. Physical and Chemical Properties ........................................................ 28
2.6.2. Antimicrobial Properties ....................................................................... 30
2.7. Methods of Generating Chlorine Dioxide Gas ..................................... 33
2.7.1. Constant gas flow generators............................................................... 34
2.7.2. Sachet .................................................................................................. 35
2.7.3. Liquid solution ...................................................................................... 36
v

2.8.

Absorption of ClO2 by fresh produce and available methods for its
quantification ......................................................................................... 37
2.8.1. Amperometric Titration Method ............................................................ 39
2.8.2. Colorimetric Method ............................................................................. 40
2.8.3. Ion chromatography ............................................................................. 40
2.8.4. UV spectroscopy .................................................................................. 41
2.9. Toxicity, Risks and Regulations of Chlorine Dioxide Gas ................. 42
BIBLIOGRAPHY .................................................................................................... 48
3. IN SITU QUANTIFICATION OF CHLORINE DIOXIDE CONSUMPTION BY
FRESH PRODUCE USING UV-VISIBLE SPECTROSCOPY ....................................... 57
3.1. Introduction ............................................................................................ 57
3.2. Materials and Methods .......................................................................... 60
3.2.1. The Equipment. .................................................................................... 60
3.2.2. Part I: Chlorine dioxide generation system ........................................... 62
3.2.3. Part II: Exposure chamber.................................................................... 65
3.2.4. Part III: Chlorine Dioxide’s detection system ........................................ 67
3.3. Gas exposure to fresh produce ............................................................ 67
3.4. Wavelength selection for ClO2 detection with Uv-vis
spectrophotometer ................................................................................ 68
3.5. Method for quantification of ClO2 consumption by fresh produce. .. 69
Application of the new system to quantify ClO2 consumption by
fresh produce ......................................................................................... 70
3.6.1. Impact of concentration and exposure on chlorine dioxide
consumption using Route 1 ................................................................. 70
3.6.2. ClO2 consumption by fresh produce at different concentrations using
Route 2 ................................................................................................ 71
3.7. Results and Discussion ........................................................................ 71
3.7.1. Wavelenght selection for ClO2 detection with Uv-vis
spectrophotometer ............................................................................... 71
3.8. Applications of the new system to quantify ClO2 consumption by
fresh produce ......................................................................................... 74
3.8.1. Impact of concentration and exposure time on chlorine dioxide
consumption using Route 1 ................................................................. 74
3.8.2. ClO2 dioxide consumption by fresh produce using Route 2 at different
concentrations. .................................................................................... 78
3.9. Conclusions ........................................................................................... 80
BIBLIOGRAPHY .................................................................................................... 81
3.6.

4. EVALUATION OF CHLORINE DIOXIDE AS AN ANTIMICROBIAL AGAINST
BOTRYTIS CINEREA IN CALIFORNIA STRAWBERRIES ......................................... 85
4.1. Introduction ............................................................................................ 85
4.2. Materials and Methods .......................................................................... 89
4.2.1. Selection of California strawberries. ..................................................... 89
vi

4.2.2.
4.2.3.
4.2.4.
4.2.5.
4.2.6.

Design of experiments.......................................................................... 89
Incubation tubs. .................................................................................... 92
Evaluating efficacy of ClO2 against the growth of Botrytis cinerea. ...... 93
Fruit quality assessment after treatment. ............................................. 93
Continuous ClO2 gas treatment. .......................................................... 94

4.2.7. Quantification of ClO2 absorption by fresh produce ............................. 96
4.2.8. Statistical Analysis................................................................................ 96
4.3. Results and Discussion ........................................................................ 97
4.3.1. Quality Assessment.............................................................................. 97
4.3.2. Quantification of ClO2 absorption by fresh produce. .......................... 100
4.3.3. Fungal growth. ................................................................................... 103
4.3.4. Quality and efficacy assessment of ClO2. .......................................... 104
4.4. Conclusions ......................................................................................... 115
BIBLIOGRAPHY .................................................................................................. 117
5. EXPLORATORY – PRELIMINARY DATA OF EFFECT OF SURFACE AREA AND
WEIGHT ON CLO2 CONSUMPTION BY FRESH PRODUCE ................................... 128
6. CONCLUSIONS AND FUTURE WORK ............................................................... 132
APPENDICES ............................................................................................................. 136
APPENDIX 1: Antimicrobial effectiveness of ClO2 to inactivate
microorganisms in fresh produce ...................................................... 137
APPENDIX 2: Amperometric titration for determination of ClO2 and ClO2 in
solution ................................................................................................. 139
APPENDIX 3: Determination of ClO2 solution using Thiosulfate Titration 141
APPENDIX 4: SAS CODES ............................................................................. 143
BIBLIOGRAPHY……………………………………………………………………………..162

vii

LIST OF TABLES

Table 1. Additional ClO2 physical and chemical properties (Windholz, Budavari et al.
1976; Keskinen and Annous 2011………....………………………………………………..30
Table 2. Mass Flow controllers set points and ClO2 generator concentration for all
treatments………………………………………………………………………………………64
Table 3. Chlorine dioxide’s absorption cross section (σ) at 4 different wavelengths
found experimentally and compared with reference values…………….………………...74
Table 4. Quantification of total chlorine dioxide absorbed by strawberries on each
treatment………………………………………………………………………………………..77
Table 5. First order reaction parameters of chlorine dioxide absorption using
Route 2…………………………………............................................................................80
Table 6. Summary of treatments 1, 2, and 3………………………………………………..90
Table 7. Summary of total chlorine dioxide absorption by California strawberries
after exposure to the different treatments considered in experiments 1, 2, and 3..…..100
Table 8. Repeated measures results for treatments in experiment 1………...………..107
Table 9. Repeated measures results for treatments in experiment 2……………….....109
Table 10. Gompertz model parameters for treatments in experiments 1, 2, and 3…...112
Table 11. Comparison of total chlorine dioxide absorption versus chlorites and
chlorine dioxide residues recovered after exposure…………........……………………..130
Table 12. Summary of antimicrobial effectiveness of ClO2 gas to inactivate
microorganisms in fresh produce………………………………………………………..…137

viii

LIST OF FIGURES

FIGURE 1. Mechanisms by which fresh produce becomes contaminated along the
production chain (Beuchat 1999)…………………………………………………………….11
FIGURE 2. Ripe strawberries infected with botrytis cinerea. ......................................... 16
FIGURE 3. Gompertz model with three parameters α: Asymptote, β: Growth rate,
and θ: inflection point. ................................................................................................... 19
FIGURE 4 - Free radical molecular structure of chlorine dioxide (Knapp and Battisti
2001) ............................................................................................................................. 29
FIGURE 5. Schematic diagram of a Minidox-M from ChlorDiSys ClO2 gas generation
system, adapted from (Czarneski and Lorcheim 2005) ................................................. 35
FIGURE 6. Schematic diagram of the three-component system for quantification of
total chlorine dioxide absorption by fresh produce ........................................................ 61
FIGURE 7. Picture of stainless steel platform holding strawberries inside the glass
chamber. ....................................................................................................................... 65
FIGURE 8. Picture of three legged stands holding strawberries inside the glass
chamber. ....................................................................................................................... 66
FIGURE 9. Spectra of chlorine dioxide at 1, 2.5, 5, 7.5 and 10 mg/L in a wavelength
range from 500 to 300 nm at 22° .................. ............................................................ 721
C
FIGURE 10. Absorbance (O.D) versus Concentration (mol/L) at 351±0.5, 360±0.5,
311±0.5 and 305±0.5 nm. ............................................................................................. 73
FIGURE 11. Absorbance versus time for treatment 1: concentration 0.22mg/L and
84 minutes of exposure ................................................................................................. 75
FIGURE 12. Chlorine dioxide concentration versus time for control and strawberries
after supplying the chamber with 3 mg/L concentration using Route 2. ........................ 79
FIGURE 13. Position and separation of strawberry fruits inside incubation tubs at
approximately 100% relative humidity. .......................................................................... 92
FIGURE 14. From left to right quality decay of calyx in strawberries before and after
chlorine dioxide treatments at 5 mg/L concentration and 7 minutes of exposure .......... 98

ix

FIGURE 15. Strawberries inside treatment chamber before exposure to chlorine
dioxide looking fresh, and with bright tissue color ....................................................... 999
FIGURE 16. Strawberries inside treatment chamber after exposure to 5 mg/L
chlorine dioxide gas for 7 minutes looking dull, color changed...................................... 99
FIGURE 17. Relation between percent of strawberries discolored versus total
chlorine dioxide absorption in mgClO2/g fruit .............Error! Bookmark not defined.101
FIGURE 18. Response Surface plot for ClO2 absorption by fresh produce in terms
of concentration and time ............................................................................................ 102
FIGURE 19. Repeated measures model over time for all treatments in experiment 1
.................................................................................................................................. 1055
FIGURE 20. Repeated measures model over time for all treatments in experiment 2
.................................................................................................................................. 1088
FIGURE 21. Repeated measures model over time for treated strawberries in
experiment 3 stored at 4 ° and 22 ° ............. ......................................................... 1111
C
C
FIGURE 22. Gompertz model fit for treatments in experiment 1. .............................. 1133
FIGURE 23. Gompertz model fit for treatments in experiment 2. .............................. 1133
FIGURE 24. Gompertz model fit for treatments in experiment 3 both at 4 ° and
C
22° .............................................. ............................................................................ 1144
C
2

FIGURE 25. Chlorine dioxide absorption versus weight [g] and surface area [in ] ... 1311

x

1. INTRODUCTION
An estimated of 9.4 million foodborne illnesses are caused by known pathogens
annually in the United States. The proportion of illness outbreaks associated to fresh
produce has increased in the last years (Bean and Griffin 1990), from <1% of all
reported outbreaks with known food vehicle in the 1970s to 6% in the 1990s (Sumathi,
Cindy et al. 2004), and the proportion of outbreak-associated cases accounted for by
fresh produce increased from <1% to 12% of illnesses in that same time period (Lynch,
Tauxe et al. 2009). Some of the reasons fresh produce associated outbreaks have
increased include higher awareness of fresh fruits and vegetables’ benefits on human
health such as improvement of life quality and prevention of chronic and cardiovascular
diseases (Bhagat 2010). Prove of this, is the increased per capita consumption of fresh
produce in America which has jumped from 91.6 to 121.1 kg, an increase of 32%
(Harris, Farber et al. 2003).
This issue brings up the question on how to prevent contamination of fresh
produce in a first place, and one of the answers is to develop and implement more
rigorous, effective sanitizing strategies along the food chain. Washing the produce with
sanitizing solutions is the most common method used during production (Harris, Farber
et al. 2003; Parish, Beuchat et al. 2003). After production, another sanitizing approach
includes the use of packaging systems that contain antimicrobial agents that reduce the
population of pathogens present in the produce and thus also prolong its shelf life.
These sanitizing agents could be added as a coating layer in the packaging material, as
a sachet that actively releases the agent inside the headspace until the reagents have

1

depleted, or by injecting different sanitizing agents inside the package to alter the gas
composition and help extend the produce’s shelf life (Appendini and Hotchkiss 2002).
One of the sanitizing agents that has caught the attention of researchers in the
last decades is chlorine dioxide (ClO2) thanks to its reported efficacy against a variety of
microorganisms in fruits and vegetables such as Listeria monocytogenes, Escherichia
coli O157:H7, and Salmonella among others (Han, Linton et al. 2000; Han, Floros et al.
2001; Han, Linton et al. 2001; Du, Han et al. 2002; Du, Han et al. 2003; Lee, Costello et
al. 2004). Although the great majority of researchers have studied the efficacy of ClO2
against pathogens, a few researchers have also investigated its effect against molds,
yeasts and other aerobes (Spotts and Peters 1980; Roberts and Reymond 1994).
Chlorine dioxide is a strong bactericide that has replaced Cl2 in several
applications at both the drinking water and food industry, because it has approximately
2.5 times the oxidation capacity of Cl2 (Benarde, Israel et al. 1965), and because it is
approximately 10 times more soluble above 11 ⁰C (United States Environmental
Protection Agency (EPA) 1999).
Chlorine dioxide could be used at different stages throughout the food chain
process. For example, it could be used during the processing of fresh produce as a
sanitizing solution for either fruits and vegetables itself, or for the tools and equipment to
prevent cross contamination. After processing, ClO2 could also be applied inside a
package in the gaseous form to extend the shelf life of fresh produce.

2

Therefore ClO2 can be used in the processing line as a washing solution for fruits
and vegetables, or can be applied inside a package in the gaseous form. There are two
ways ClO2 could be applied in the package, which are i) sachets that contain chemicals
that react to continuously generate and build up ClO2 gas concentration in the
headspace (Ellis, Cooksey et al. 2006), and ii) direct injection of ClO2 inside the
package by means of a commercially available gas generator using a reaction with
sodium chlorite and chlorine gas
In the gaseous form, ClO2 penetrates through irregularities and complex areas
more efficiently than the aqueous form, making ClO2 gas more effective than the
aqueous form (Du, Han et al. 2002; Lee, Costello et al. 2004).
However, there are some concerns when it comes to sanitation of fruits and
vegetables with ClO2 gas that need to be properly addressed. First of all, during
treatment (inside a package or during production), a good circulation of the gas must be
ensured so the gas can envelop the produce and effectively reach those sites where
microbes might be hiding (Staschower 2012). Improper circulation of the gas around the
produce will make the treatment less effective.
Another concern with the use of ClO2 gas as a sanitizer of fresh produce has to
do with its toxicology characteristics. ClO2 participates in a series of oxidative reactions
−

−

that result in by products such as (ClO3 ) and chlorite (ClO2 ) ions, which can at the
−

same time further transform into chloride (Cl ) (United States Environmental Protection

3

Agency (EPA) 1999; Gómez-López, Rajkovic et al. 2009). When produce surfaces
absorb ClO2 during treatments, it then oxidizes and breaks down upon contact
(Netramai 2011), leaving surface residues that could later cause the formation of
chlorinated byproducts. For these reasons, although regulation 21 CFR 173.300 allows
the use of ClO2 to disinfect fruits and vegetables, it requires a potable rinse step after
treatment in order to assure that there are no residues of concern for consumers’
consumption (Trinetta, Vaidya et al. 2011).
Some of the currently available methods to quantify surface residues on produce
after exposure to ClO2 are the colorimetric method, amperometric titration, and
chromatography. These methods have in common that i) they analyze a washing
solution after treating the produce with the gas, and ii) they require a post-treatment
technique in order to quantify the residues left on the produce’s surface.

In light of this discussion, the goal of this study is to develop a continuous
method to quantify ClO2 in situ: exposure concentration and consumption of ClO2 by
fresh produce.
To achieve this goal, two specific objectives were set and they are described
below:
To develop a new system and online method to quantify ClO2 exposure
concentration and consumption by fresh produce.
To assess the application of the new equipment by evaluating the ClO2 gas
efficacy against Botrytis cinerea on strawberries.

4

BIBLIOGRAPHY

5

BIBLIOGRAPHY

Appendini, P., Hotchkiss, J.H., (2002). Review of antimicrobial food packaging.
Innovative Food Science & Emerging Technologies 3(2), 113-126.
Bean, N.H., Griffin, P.M., (1990). Foodborne disease outbreaks in the United
States, 1973-1987: pathogens, vehicles, and trends. Journal of Food Protection 53.
Benarde, M.A., Israel, B.M., Olivieri, V.P., Granstrom, M.L., (1965). Efficiency of
chlorine dioxide as a bactericide. Applied Microbiology 13(5), 776-780.
Bhagat, A.R., (2010). Modeling Critical Factors to Optimize the Treatment of
Selected Fruits and Vegetables with Chlorine Dioxide Gas Using a Miniaturized
Insustrial-Size Tunnel System. Purdue University, West Lafayette, Indiana.
Du, J., Han, Y., Linton, R.H., (2002). Inactivation by chlorine dioxide gas (ClO2)
of Listeria monocytogenes spotted onto different apple surfaces. Food Microbiology
19(5), 481-490.
Du, J., Han, Y., Linton, R.H., (2003). Efficacy of chlorine dioxide gas in reducing
Escherichia coli O157:H7 on apple surfaces. Food Microbiology 20(5), 583-591.
Ellis, M., Cooksey, K., Dawson, P., Han, I., Vergano, P., (2006). Quality of Fresh
Chicken Breasts Using a Combination of Modified Atmosphere Packaging and Chlorine
Dioxide Sachets. Journal of Food Protection 69, 1991-1996.
Gómez-López, V.M., Rajkovic, A., Ragaert, P., Smigic, N., Devlieghere, F.,
(2009). Chlorine dioxide for minimally processed produce preservation: a review.
Trends in Food Science & Technology 20(1), 17-26.
Han, Y., Floros, J.D., Linton, R.H., Nielsen, S.S., Nelson, E., (2001a). Response
Surface Modeling for the Inactivation of Escherichia coli O157:H7 on Green Peppers
(Capsicum annuum L.) by Chlorine Dioxide Gas Treatments. Journal of Food Protection
64, 1128-1133.
Han, Y., Linton, R.H., Nielsen, S.S., Nelson, P.E., (2000). Inactivation of
Escherichia coli O157:H7 on surface-uninjured and -injured green pepper (Capsicum
annuum L.) by chlorine dioxide gas as demonstrated by confocal laser scanning
microscopy. Food Microbiology 17(6), 643-655.

6

Han, Y., Linton, R.H., Nielsen, S.S., Nelson, P.E., (2001b). Reduction of Listeria
monocytogenes on Green Peppers (Capsicum annuum L.) by Gaseous and Aqueous
Chlorine Dioxide and Water Washing and Its Growth at 7C. Journal of Food Protection
64, 1730-1738.
Harris, L.J., Farber, J.N., Beuchat, L.R., Parish, M.E., Suslow, T.V., Garrett, E.H.,
Busta, F.F., (2003). Outbreaks Associated with Fresh Produce: Incidence, Growth, and
Survival of Pathogens in Fresh and Fresh-Cut Produce. Comprehensive Reviews in
Food Science and Food Safety 2, 78-141.
Lee, S.-Y., Costello, M., Kang, D.-H., (2004). Efficacy of Chlorine Dioxide Gas as
a Sanitizer of Lettuce Leaves. Journal of Food Protection 67, 1371-1376.
Lynch, M., Tauxe, R., Hedberg, C., (2009). The growing burden of foodborne
outbreaks due to contaminated fresh produce: risks and opportunities. Epidemiology
and infection 137(3), 307.
Netramai, S., (2011). Utilization of chlorine dioxide gas in food packaging
application. Michigan State University, East Lansing, Michigan.
Parish, M.E., Beuchat, L.R., Suslow, T.V., Harris, L.J., Garrett, E.H., Farber, J.N.,
Busta, F.F., (2003). Methods to Reduce/Eliminate Pathogens from Fresh and Fresh-Cut
Produce. Comprehensive Reviews in Food Science and Food Safety 2, 161-173.
Roberts, R.G., Reymond, S.T., (1994). Chlorine dioxide for reduction of
postharvest pathogen inoculum during handling of tree fruits. Applied and environmental
microbiology 60(8), 2864-2868.
Spotts, R., Peters, B., (1980). Chlorine and chlorine dioxide for control of d’Anjou
pear decay. Plant Dis 64(1), 2.
Staschower, F., (2012). Residues from Chlorine Dioxide Gas Treatment,
Generated by Different Delivery Systems, on Fresh Produce. Michigan State University,
East Lansing, Michigan.
Sumathi, S., Cindy, R.F., Linda, C., Robert, V.T., (2004). Fresh Produce: A
Growing Cause of Outbreaks of Foodborne Illness in the United States, 1973 through
1997. Journal of Food Protection 67(10), 2342-2353.
Trinetta, V., Vaidya, N., Linton, R., Morgan, M., (2011). Evaluation of Chlorine
Dioxide Gas Residues on Selected Food Produce. Journal of Food Science 76(1), T11T15.

7

United States Environmental Protection Agency (EPA), (1999). Alternative
Disinfectants and Oxidants Guidance Manual: Chapter 4. Chlorine Dioxide, EPA 815-R99-014.

8

2. LITERATURE REVIEW
2.1. Framework
In recent years, an interest for healthier, fresher foods has become a common
denominator in consumers across age, sex and ethnic backgrounds. In (Sloan 2013)
article “Top 10 Food Trends”, listed healthier options in the second place out of the top
10 food trends in 2013.
Americans have increased their consumption of fruits and vegetables since 1979
in part due to an active promotion to include fruits and vegetables for a healthy diet.
From 1982 to 1997, the consumption of fresh fruits and vegetables per capita in the U.S
increased from 91.6 to 121.1 kg, an increase of 32% (Harris, Farber et al. 2003). During
the same period, organizations such as U.S. Department of Agriculture, the U.S.
Department of Health and Human Services, and the Centers for Disease Control and
Prevention (CDC) used resources to raise awareness and developed strategies to
combat obesity and other diseases (Plattner and Perez 2012). Research confirmed that
the consumption of minimally processed produce improves the quality of life and
prevents chronic and cardiovascular diseases (Bhagat 2010).
Minimally processed foods are those which do not generally contain
preservatives or antimicrobial agents and which rarely undergo any heat processing
prior to consumption (Seymour and Appleton 2001). Considering that the original
purpose of minimal processing was to minimize the thermal treatments that reduce food
quality, researchers were forced to develop non-conventional heat techniques to extend
the shelf life of food products (Balla and Farkas 2007).

9

However, not only does minimal processing not completely inactivate all
microorganisms present in the raw fruits and vegetables, but processing of raw
materials also promotes a faster physiological deterioration, which at the same time
cause microbial degradation. Therefore the microbiological safety during the shelf life of
minimally processed foods is of great importance (O’Beirne and Francis 2003; Balla and
Farkas 2007).
2.2. Food quality and safety
Foodborne disease outbreaks are a concern especially when it comes to
minimally processed foods like fresh fruits and vegetables, which have been identified
as vehicle for illnesses since they can carry both spoilage and pathogenic
microorganisms (Bhagat 2010). From harvest to shelf, fresh produce encounters many
different sites where possible contamination can occur including harvesting, processing,
distribution, and storage. Some examples of these potential sources of contamination
are suggested by Brackett (1999) as cross contamination with feces of domestic and
wild animals, contaminated irrigation water, use of untreated manure or sewage, lack of
sanitation in the field, irrigation and harvesting equipment and handlers at both the
retails and consumer levels (Brackett 1999). Figure 1, taken from Beuchat (1996) shows
the different mechanisms by which fresh produce becomes contaminated with
pathogenic microorganisms such as viruses, bacteria and parasites.

10

FIGURE 1. Mechanisms by which fresh produce becomes contaminated along
the production chain (Beuchat 1999).
During the period of 2009 to 2010, a total of 1,527 foodborne disease outbreaks
were reported resulting in 29,444 cases of illness, 1,184 hospitalizations, and 23 deaths
(Centers for Disease Control and Prevention 2013). CDC defines a foodborne disease
outbreak as the occurrence of two or more similar illnesses resulting from ingestion of a
common food. An estimated of 9.4 million foodborne illnesses are caused by known
pathogens annually in the United States. The proportion of illness outbreaks associated
to fresh produce has increased in the last years (Bean and Griffin 1990), from <1% of all
reported outbreaks with known food vehicle in the 1970s to 6% in the 1990s (Sumathi,
Cindy et al. 2004), and the proportion of outbreak-associated cases accounted for by
fresh produce increased from <1% to 12% of illnesses in that same time period (Lynch,
Tauxe et al. 2009). The possible causes behind the increment of fresh produce related
outbreaks are believed to go beyond the increased demand for these types of products,
and include the changes in agronomic, processing, preservation, packaging,

11

distribution, and marketing technologies on a global scale which have enabled the raw
fruit and vegetable industry to supply consumers with a wide range of high quality
produce from different countries year round, which have implemented practices and
technologies that may increase risk for human illness associated with pathogenic
microorganisms (Beuchat 2002).

However as mentioned before, fruits and vegetables are also susceptible of
contamination with spoilage type of microorganisms, which in most cases won’t hurt
humans health, but will cause food product losses and a reduction in the food
availability. According to Kantor et al., the annual edible food available to the U.S is
reduced by 27% due to food spoilage and waste at the retail and consumer level
(Kantor, Lipton et al. 1997). In a different study carried out between 2010 to 2011 by
The Swedish Institute for Food and Biotechnology (SIK) sponsored bythe Food and
Agriculture Organization of the United Nations (FAO), was suggested that roughly onethird of the food produced for human consumption is lost or wasted globally, which
amounts to about 1.3 billion tons per year (Gustavsson, Cederberg et al. 2011). In the
fruits and vegetables commodity group, both agriculture and processing are the two
major factors impacting total waste across all 7 regions studied (North America and
Oceania, Industrialized Asia, Subsahara Africa, North Africa, West & Central Asia,
South and Southeast Asia, and Latin America). The causes of food waste and losses
around the world are not limited to their contamination with spoilage microorganisms,
but leaving economic and political factors aside, this is one of the most relevant causes
of the problem especially in some regions of the world where there are inadequate

12

storage facilities and infrastructure and poor harvesting practices (Gustavsson,
Cederberg et al. 2011).
2.2.1. Pathogenic and spoilage microorganisms associated with fresh
produce
The top five pathogens causing domestically acquired foodborne illnesses have
been identified as Norovirus, Clostridium perfringens, Campylobacter spp., Salmonella
nontyphoidal, and Staphylococcus aureus (Centers for Disease Control and Prevention
2011). In addition to the latter two pathogens listed, Esterichia Coli and Listeria
monocytogenes are the most commonly found pathogens in fresh produce.
2.2.1.1.

Salmonella

The genus Salmonella has over 2700 serotypes. Animals and birds are the
natural reservoirs. Surveys of fresh produce have revealed the presence of several
Salmonella serotypes capable of causing human infection, though poultry and other
meat products, eggs and dairy products, are the most commonsources in salmonellosis
outbreaks. Fresh fruits and vegetables are implicated less frequently, although
outbreaks have been documented most notably in cantaloupe and sprouts. Laboratory
studies revealed that the pathogen can grow in damaged, chopped, or sliced tomatoes
(pH 4.1 – 4.5) stored at 20 to 30 ° (68 to 86 ° ) (Beuchat 1996; Harris, Farber et al.
C
F
2003). Produce items that were most frequently implicated in outbreaks of Salmonella
infection include salad, sprouts, and melons (Sumathi, Cindy et al. 2004).
2.2.1.2.

Staphylococcus aureus

13

According to the CDC, staphylococcus aureus (S. aureus ) is a type of bacteria
that most of the time does not cause any harm; however, sometimes it causes
infections which can be fatal such as bacteremia, pneumonia, endocarditis, and
osteomyelitis.
S. aureus has been detected on fresh produce and ready-to-eat vegetable
salads, and is known to be carried by food handlers. However, S. aureus does not
compete well with other microorganisms normally present on fresh produce, so incipient
spoilage caused by nonpathogenic micro flora would likely precede the development of
high populations of this pathogen (Harris, Farber et al. 2003).
2.2.1.3.

Esterichia Coli

Escherichia coli is a common cause of travelers' diarrhea. Fresh produce comes
in contact with this pathogen through cross-contamination from meat products, or
through direct contamination in the field with feces of wild or domestic animals (Bhagat
2010). Contaminated raw vegetables are thought to be a common source of the
pathogen which depending on the kind, may cause diarrhea, urinary tract infections,
respiratory illness and pneumonia, and other illnesses. (Beuchat 1996). Items frequently
implicated in outbreaks of Escherichia coli O157:H7 infection are lettuce and apple cider
(Sumathi, Cindy et al. 2004).
2.2.1.4.

Listeria monocytogenes

Listeria is a gram-positive foodborne intracellular human pathogen, which has
been associated with serious food-borne outbreaks and has been responsible of a
number of recalls of fresh produce (Bhagat 2010). Plants and plant parts used as salad

14

vegetables play a key role in disseminating the pathogen from natural habitats to the
human food supply (Beuchat 1996). L. monocytogenes is one of the most common
food-borne pathogens in ready to eat foods responsible for causing serious illness such
as septicemia and meningitis in humans where fatality rate is very high (up to 75% in
highly immunocompromised individuals) (Bhagat 2010).
On the other hand, spoilage bacteria are microorganisms that cause food to
deteriorate and develop unpleasant odors, tastes, and textures. These microorganisms
can cause fruits and vegetables to rot, get mushy or slimy, and to develop undesirable
color changes and odors. Each vegetable possesses a unique set of intrinsic factors
that can influence the survival and growth of different microorganisms such as the pH,
water activity and temperature (Beuchat 2002; Dagnas and Membre 2013).
Mold spoilage results from a biological process that includes having a product
contaminated with fungal spores which germinate and extend in the form of hyphae to
form a visible mycelium over time before the end of the product’s shelf life (Dantigny,
Guilmart et al. 2005). Optimum conditions at which mold growth is favored vary with the
type of microorganisms, but in general, food spoilage will occur at temperatures around
25⁰C and aw above 0.85 provided that food substances are contaminated with spores
and that these spores are able to germinate (Dantigny, Guilmart et al. 2005; Gougouli
and Koutsoumanis 2012)
A common fungus attacking all parts of the shoots of strawberries and
raspberries, causing the greatest losses on ripe or near-ripe fruit is botrytis cinerea.
2.2.1.5.

Botrytis cinerea

15

Botrytis cinerea occurs abundantly throughout the year as a saprophyte and
facultative parasite on a wide variety of plant materials in raspberry and strawberry
plantations (Jarvis 1962). Figure 2 show an example of symptoms of botrytis cinerea in
strawberries.
The majority of fruit infections are initiated from mycelium growing saprophytically
in contiguous plant material or foam spores germinating in solutions trapped between
the fruit surface (Jarvis 1962).
For these reasons, the infection of fruit with this fungus can be traced to the
infection of floral parts in the field (Powelson 1960) or by contact with contaminated
berries (Bhaskara Reddy, Belkacemi et al. 2000)The rot caused by botrytis cinerea
usually occurs at the stem end of the fruit.

FIGURE 2. Ripe strawberries infected with botrytis cinerea. For interpretation of
the references to color in this and all other figures, the reader is referred to the
electronic version of this thesis.
This was studied by Powelson et al. (1960), who demonstrated that the
proportion of symptomless strawberries at maturity was always higher in the group of
fruit where petals, stamens, and sepals were removed regardless of the environmental

16

conditions in the greenhouse (wet or dry) (Powelson 1960). Braun et al. (1988) found
that the receptivity of strawberry leaves to infection by botrytis cinerea depended on the
developmental stage of the leaf (Braun and Sutton 1988). This idea is in agreement with
Williamson et al. (2007), who described that botrytis cinerea is most destructive on
mature or senescent tissues of dicotyledonous hosts, even though it usually access
such tissues at a much earlier stage in crop development and remains quiescent for a
considerable period before rapidly rotting tissues when the environment is conducive
and the host physiology changes (Williamson, Tudzynski et al. 2007).
Therefore, evidence of fruit rot may happen long after harvest of apparently
healthy crops which deteriorate during transport and storage.
Botrytis cinerea has a variety of modes of attack, diverse hosts as inoculum
sources, and it can survive as mycelia and/or conidia or for extended periods as
sclerotia in crop debris (Williamson, Tudzynski et al. 2007).
For all these reasons, botrytis cinerea is difficult to control and normally one
single control measure won’t succeed.
One way science can help understand the interactions between the
microorganisms and hosts, or the mechanisms, dynamics and kinetics of bacteria or
fungal growth, is by employing predictive models for shelf life determination and to
predict risk of specific diseases.

17

2.3. Microbial growth models
Predictive modelling of filamentous fungal growth has not received the same
attention as that of bacterial development, maybe due to inherent complexities
associated with the quantification of fungal growth (Garcia, Ramos et al. 2009). These
complexities, or specificities described by Dantigny et al. include i) the reproductive
differences of bacteria and fungi which occur by fission due to their ability of dividing
therefore it is easier to enumerate, while in contrast fungal growth involves germination
and hyphal extension forming mycelium whose weight does not increase exponentially
except at early stages of growth, and the ii) factors affecting microorganism growth,
which for bacteria temperature plays the major role, while for mold growth water activity
is more important (Dantigny, Guilmart et al. 2005). However, the number of studies
focusing on fungal development has increased, and Dantigny et al. (2006) listed
standardizing techniques for the study of fungal germination after realizing that
techniques, definitions and methods for assessing mold germination varied greatly
between authors, making it hard to compare experimental data (Dantigny, Bensoussan
et al. 2006). Apart from other definitions, Dantigny et al. (2006) also mentioned two of
the primary germination models that are most commonly used by researchers: the
Gompertz and the logistic function models.
2.3.1. Primary models
Primary models are those which describe the growth of a population of
microorganisms over time from the basis of adopting strategies for food safety. Primary
microbial models can further be classified as germination, growth and inactivation
models (Garcia, Ramos et al. 2009). Growth models can be either empirical (set out
18

principally to describe) or mechanistic (attempt to give a description with understanding)
(López, Prieto et al. 2004).
2.3.1.1.

Gompertz Model

Increase of percentage of germinated spores over time can be modeled with the
Gompertz equation:

ܲ ൌ ߙ݁ ି௘

ሾഁሺ೟షഇሻሿ

(1)

Where P (%) is the percentage germination, α (%) is the asymptotic value of P
when time goes to infinite (t → ∞), β in inverse time units, is the growth rate or slope
term of the tangent line through the inflection point (θ), which in time units is the point at
which the concavity of the curve changes (Dantigny, Marín et al. 2007; Garcia, Ramos
et al. 2009).

FIGURE 3. Gompertz model with three parameters α: Asymptote, β: Growth rate,
and θ: inflection point.

19

2.3.1.2.

ܲൌ

Logistic function

௉೘ೌೣ

ଵା௘ ሾೖሺഓష೟ሻሿ

(2)

Where P (%) is the percentage germination, Pmax (%) is the asymptotic P value at
t → ∞, τ (in time units) is the inflection point where P equals half of Pmax, t is the time,
and k (in inverse time units) is related to the slope of the tangent line through the
inflection point (Dantigny, Guilmart et al. 2005; Dantigny, Marín et al. 2007).
Contrary to the Gompertz model, the logistic function is symmetric about the
point of inflection, for this reason the Gompertz model may offer greater flexibility than
the logistic (Gibson, Bratchell et al. 1987).
Gibson et al. (1987) used both Gompertz model and logistic equations to fit
microbial growth data of Clostridium botulinum type A in pasteurized pork slurry with a
reasonable good fit; therefore, it is possible then to estimate the growth parameters
with physical meaning (Gibson, Bratchell et al. 1987). Declerck et al. (2001) compared
the fit of the Schute, Gompertz and logistic models to fungal growth data and found that
all three fitted the data very well with very similar residuals (Declerck, D'or et al. 2001).
Dantigny et al. (2007) studied mold germination and compared the two models;
based on RMSE and the goodness of fit, they determined that both models were similar
and and provided equivalent outcomes (Dantigny, Marín et al. 2007). Similarly,
Gougouli et al. (2012) used the Gompertz equation to model the germination of
Penicilum expansum and Aspergillus niger spores on malt extract agar finding a
satisfactory description with acceptable goodness of fit for all parameters (Gougouli and
Koutsoumanis 2012).

20

Although both the Gompertz model and the logistic equation are widely accepted
and commonly used as means to predict fungal growth and development, there are
several, simple statistical methods to evaluate similarities and differences between
nonlinear models or to decide which one fits the data better. The main statistical
procedures for model comparison are residual analysis and tests for goodness of fit
(López, Prieto et al. 2004).
2.3.2. Secondary models
These type of models describe the influence of environmental factors such as
temperature, water activity, or pH on key parameters of the primary model (Dantigny,
Guilmart et al. 2005). Examples of secondary models are listed below
2.3.2.1.

Ratkowski square-root model

Originally developed for bacteria, the model is as follows:

ඥߤ௠௔௫ ൌ ܾሺܶ െ ܶ௠௜௡ ሻ൛1 െ ݁ ሾ௖ሺ்ି்೘ೌೣ ሻሿ ൟ

(3)

Where µ is the fungal growth rate, T is temperature, and b, c, Tmax and Tmin are
estimated constants (Garcia, Ramos et al. 2009).
2.3.2.2.

Linear Arrhenius-Davey equation

ln ߤ௠௔௫ ൌ ܽ଴ ൅

௔భ
்

൅

௔మ

்మ

(4)

Where µ is the fungal growth rate, T is absolute temperature (K), and a0, a1 and
a2 are constant to be estimated (Garcia, Ramos et al. 2009).

21

Other secondary models are also the Rosso cardinal model and the Gibson
model.
2.3.3. Tertiary models
Tertiary models could be defined as the integration of primary and secondary
models in a software using databases. Some tertiary model tools include the Pathogen
Modelling Program (PMP) developed by the USDA, which was designed as a research
and instructional tool for estimating the effects of multiple variables on the growth,
inactivation or survival of food borne pathogens (United States Department of
Agriculture (USDA) 2003), and the latest version of ComBase (www.combase.cc) that
includes a modelling tool that uses its database to generate growth or inactivation
curves (Garcia, Ramos et al. 2009).

In general, these models are used as tools that facilitate the decision process for
example as a quality assessment in a food manufacturing operation or product
development in order to determine the feasibility of a new formula, process or as a
strategy for the inactivation of microbial growth. The application of these models provide
an understanding of the inactivation mechanisms, the dynamics and the kinetics of
fungal growth, which as explained before, vary with the external environments and with
the type of crop, plant or fruit involved. All these insights bring to researchers and
companies a foundation to develop innovative strategies to inhibit microbial growth
while maintaining quality, freshness and safety. Some of such strategies that are
currently being used are described next.

22

2.4. Strategies to prolong the shelf life and ensure safety of fresh produce
As explained before in section 1.2, there are a number of different sources of
contamination for food which start early in the food chain process such as harvesting
and production. The human interaction at this stage is significant, and for that reason
the USDA proposed a plan in 2004 that aims to minimize foodborne illness associated
with fresh produce consumption (United States Food and Drug Administration (FDA)
2004) which accounts for processes equipment and personal involved in the food
production responsible for the control and elimination of microbial contamination of
produce intended for consumers.
Another approach being implemented by producers and distributors is the
HACCP (Hazard Analysis and Critical Control Point) which aims to prevent foodborne
diseases by identification, evaluation and control of key points in the fresh produce
supply chain (Netramai 2011). Simple practices such as education to growers about the
importance of hygiene habits, and good agricultural practices also add up towards the
collective goal of making sure fresh produce arrives in good quality and safe to
consumers.
During distribution and storage it is vital to maintain fresh produce under
adequate (cool) temperatures and relative humidity since they are the most important
factors affecting spoilage and growth of pathogenic microorganisms in fruits and
vegetables (Allende, Tomás-Barberán et al. 2006).
Other preservation techniques that are currently being implemented by fresh
produce industry include the use of antioxidants, irradiation, modified atmosphere

23

packaging, antimicrobial packaging and sanitizing agents among others (Parish,
Beuchat et al. 2003; Allende, Tomás-Barberán et al. 2006).
2.4.1. Modified atmosphere packaging (MAP)
The objective of a MAP technique applied to fresh produce is to extend its shelf
life by reducing the produce respiration without completely stopping it. Failure to allow
levels of oxygen inside the package could create opportunities for slower growing
pathogenic bacteria, and for that reason MAP techniques must be designed with care
(Allende, Tomás-Barberán et al. 2006). The packaged for a fresh produce is surrounded
by a different gas mixture other than air which could include gases such as CO2, N2 or
O2 depending on the application (Phillips 1996). Packaging materials intended to use
for this purpose must be designed with selective barrier properties that provide a
controlled oxygen transmission rate (OTR) and they should be selected according to the
specific produce’s respiration rate (Exama, Arul et al. 1993). For decades MAP
application has been successfully used in food products such as fish, meat, fruits and
vegetables for decades (Phillips 1996).
2.4.2. Irradiation
Although in some cases irradiation has affected the quality attributes of fresh
produce, irradiation with low-dose gamma is very effective in reducing bacterial,
parasitic, and protozoan pathogens in raw foods (Beuchat 1998; Rico, Martín-Diana et
al. 2007). The maximum dose approved by the FDA for fruits and vegetables is 1.0 kGy
irradiation (Rico, Martín-Diana et al. 2007). Ultraviolet light acts as an antimicrobial
agent due to DNA damage, and indirectly due to induction of resistance mechanisms in
24

different fruits and vegetables against pathogens (Allende, Tomás-Barberán et al. 2006;
Rico, Martín-Diana et al. 2007). UV consist of beams that accelerate the ions to interact
with microorganisms and inactivate them (El-Samahy, Youssef et al. 2000). Nonionizing, artificial ultraviolet-C (UV-C) radiation is extensively used in a broad range of
antimicrobial applications including water, air, food preparation surfaces and containers
(Allende, Tomás-Barberán et al. 2006).
2.4.3. Antimicrobial packaging
Antimicrobial packaging is another type of active packaging which objective is to
reduce, inhibit or retard the growth of microorganisms that may be present in the
package system (Appendini and Hotchkiss 2002). Antimicrobial packaging could take
several forms including i) addition of sachets/pads containing sanitizing agents into
packages, ii) incorporation of antimicrobial agents directly into polymers, iii) application
of surface coatings to polymers, or iv) use of polymers that are inherently antimicrobial
(Appendini and Hotchkiss 2002). Some of the sanitizing agents that are used either in
antimicrobial packaging or as a preservation technique during food processing are
described below.
2.5. Common sanitizing agents
2.5.1. Chlorine
Chlorine (Cl2) has been used for sanitation purposes in food processing for
several decades, and its perhaps the most widely used sanitizer in the food industry
(Parish, Beuchat et al. 2003). It can be used with water as a washing solution or as a
spray to sanitize produce surfaces including the food package (Bhagat 2010). The

25

mode of inactivation of Cl2 is through the bactericidal activity of hypochlorous acid
produced by the breakdown of sodium hypochlorite (Bhagat 2010). The Food and Drug
Administration (FDA) in the United States permits the use of Cl2 as a disinfectant in
wash, spray and fume waters in the raw fruit and vegetable industry (Seymour and
Appleton 2001).
The main disadvantages of Cl2 as a disinfectant are i) it can form chlorinated
compounds such as trihaloethanes which are considered carcinogenic (Parish, Beuchat
et al. 2003; Bhagat 2010) and that ii) its efficacy is affected by changes in pH,
temperature, presence of organic matter, light, air and metals (Parish, Beuchat et al.
2003). Cl2 is generally used in concentration between 50-200ppm for treatments of 1-2
minutes with reported reductions of less than 2 log CFU g-1 on vegetables and fruits
(Staschower 2012).
2.5.2. Hydrogen peroxide
Hydrogen peroxide (H2O2) possesses bactericidal and inhibitory activity due to
its properties as an oxidant agent, and due to its capacity to generate other cytotoxic
oxidizing species such as hydroxyl radicals (Parish, Beuchat et al. 2003). Its rapid
breakdown makes it a good sanitizing agent for food surfaces and packaging materials
in aseptic filling operations (Parish, Beuchat et al. 2003; Rico, Martín-Diana et al. 2007).
The main drawbacks of the use of H2O2 as sanitizer of fresh produce are that i) similarly
to Cl2, its effectiveness depends on temperature and pH, and that ii) treatments with
this agent caused significant quality loss on fresh produce (Beuchat 1998).
26

2.5.3. Ozone
Ozone (O3) is used in gaseous form or dissolved in water, it is a strong oxidizing
agent able to destroy microbial cells due in part to its strong penetrability in produce
surface (Rico, Martín-Diana et al. 2007). It is an effective treatment for drinking water
and will inactivate bacteria, fungi, viruses and protozoa (Parish, Beuchat et al. 2003).
Two main advantages of O3 are that i) it will decompose into non-toxic product, and ii)
its performance is not affected by pH (Rodgers, Cash et al. 2004; Rico, Martín-Diana et
al. 2007). Treatment with ozonated water showed an increase in shelf life of apples,
grapes, oranges, pears, raspberries, and strawberries by reducing microbial populations
and by oxidation of ethylene to retard ripening (Beuchat 1998). However, some of the
disadvantages of ozone for use as a sanitizing agent are that i) it can cause changes in
fruit quality, specifically color and texture (Daş, Gürakan et al. 2006) and furthermore ii)
due to its high corrosiveness and potential hazard, it has to be generated on site and
cannot be transported (Beuchat 1998).

The list of available sanitizers and strategies that aim to reduce human health
risks and food spoilage by preservation of fresh produce and other food products is
much longer. There is not one universal solution for the diversity of challenges that the
food industry encounters with different type of produce and microorganisms, affecting
different parts of the world where availability of resources, or technology or
infrastructure are different and unique. In many cases, a combination of two or more
preservation techniques is the best approach given that multiple points for control and
reduction of microorganism are necessary in order to minimize the possibilities for
27

outbreaks. However, there is one sanitizer agent that is being considered and assessed
by the scientific community as one promising option to disinfect fresh produce
efficiently. It is the case of chlorine dioxide.
2.6. Chlorine Dioxide
Discovered by Sir Humphrey Davy in the nineteenth century after acidifying
potassium chlorate with sulfuric acid, chlorine dioxide came a hundred years later to be
known as a powerful disinfectant and bleaching agent (Aieta and Berg 1986). Thanks to
its broad spectrum disinfection capability, chlorine dioxide has a wide variety of
applications (Knapp and Battisti 2001). The largest application of ClO2 is in the paper
industry as wood pulp bleaching where it is preferred over other types of oxidizing
agents thanks to its i) high solubility in water, ii) commercial availability of sodium
chlorite, and because iii) it does not compromise the mechanical integrity of the
bleached pulp (Kaczur and Cawlfield 2000). In other applications, chlorine dioxide is
used as a surface sanitizer of diverse kinds (Keskinen and Annous 2011), it is also used
in the drinking water treatment to produce safe potable water (Aieta and Berg 1986),
and most recently in the food and packaging industry for its usefulness as a treatment to
minimize microbial cross contamination and to extend the shelf life of products (Aieta
and Berg 1986).
2.6.1. Physical and Chemical Properties
Chlorine dioxide is a volatile, highly energetic molecule (United States
Environmental Protection Agency (EPA) 1999) which contains two atoms of chlorine
(Cl2) and one atom of oxygen (Knapp and Battisti 2001). In the gas form, ClO2 has a

28

yellow-green color with a pungent odor similar to chlorine and sodium hypochlorite;
while in the liquid form, ClO2 has a deep red color and is explosive at temperatures
above -40 ⁰C (Kaczur and Cawlfield 2000; Linton, Han et al. 2006). Two of the most
important physical characteristics of ClO2 are both that it exists almost exclusively as a
free radical monomer even when diluted in aqueous solutions, and its high solubility in
water where it forms a stable yellow greenish color solution provided that it is kept cool
in a closed container and in the dark (United States Environmental Protection Agency
(EPA) 1999). The free radical molecular structure of the gas is shown in Figure 3.

Cl
O

Cl
O

O

O

FIGURE 4 - Free radical molecular structure of chlorine dioxide (Knapp and
Battisti 2001)
As a strong oxidizing agent, ClO2 will participate in oxidation reduction reactions
-

-

that involve one electron transfer mechanism forming chlorite (ClO2 ), chloride (Cl ) and
-

chlorate (ClO3 ) (United States Environmental Protection Agency (EPA) 1999; Keskinen
and Annous 2011). The concentration at which these by-products are generated would
depend on the reaction pH and light conditions. For example, in aqueous solutions at
pH>10, chlorine dioxide will hydrolyze to form ClO3- and ClO2- ions, while in neutral or
near neutral solutions (4< pH <10) ClO2 will remain stable as a free radical (United
States Environmental Protection Agency (EPA) 2006). Some of these key reactions

29

include (United States Environmental Protection Agency (EPA) 1999; Qingdong,
Guangming et al. 2006):
−

ClO2 (aq) + e → ClO2
−

−

−

−

−

ClO2 + 2H2O + 4e → Cl + 4OH
-

−

−

ClO2 + 2OH → ClO3 + H2O + 2e
−

ClO2 + H2O → ClO3 + 2H+ + e
−

ClO2 + 4H+ + 4e → ClO3

−

−

−

Due to its free radical natural state, ClO2 will degrade with UV or fluorescent light
breaking the chlorine oxygen bond, forming O- and ClO- which are considered reactive
forms (Kaczur and Cawlfield 2000; Netramai 2010).
Table 1 provides a summary of additional ClO2 properties.

Table 1. Additional ClO2 physical and chemical properties (Windholz, Budavari et
al. 1976; Keskinen and Annous 2011)
CAS Registry Number
Molecular weight
Water solubility
Specific gravity
Critical temperature
Critical pressure

10049-04-4
67.46 g/mol
3.0 g/L at 25 C and 34 mmHg
1.642 g/mL at 0 C (liquid); 3.09 g/L (gas)
465 K
8621.6 kPa

2.6.2. Antimicrobial Properties
ClO2 disinfects by an oxidation mechanism without involving chlorination (United
States Environmental Protection Agency (EPA) 1999). Research has been carried out in

30

order to determine the bactericide, fungicide and viricide efficacy of ClO2 against
multiple microorganisms present in different types of produce. Appendix 1 shows a
summary of the main findings for fresh produce.
The attractiveness of ClO2 use as a biocide has increased to the extent that it
has replaced Cl2 in several applications at both the drinking water and food industry
because i) it has approximately 2.5 times the oxidation capacity of Cl2 (Benarde, Israel
et al. 1965), ii) it is approximately 10 times more soluble above 11 ⁰C (United States
Environmental Protection Agency (EPA) 1999), and iii) its efficacy is not affected by
changes in pH as it does not ionize in water (Benarde, Snow et al. 1967).
However, it is has also been found that ClO2 is less effective than ozone (United
States Environmental Protection Agency (EPA) 1999).
In general, although appendix 1 shows that ClO2 has been proved to be a
powerful sanitizing agent, its efficacy in reducing populations of bacteria, protozoa, or
virus will vary with factors such as produce, target microorganism, relative humidity
(United States Environmental Protection Agency (EPA) 1999) and temperature
(Benarde, Snow et al. 1967).
Another important factor that affects ClO2 efficacy is the state at which it is used
in a treatment; published data suggest that aqueous ClO2 is less effective than gaseous
when treating fruits and vegetables. For example, Han et al. explained that the greater
penetration ability of gases could be responsible for this phenomenon when they proved
the better efficiency of ClO2 gas versus aqueous solution in injured and uninjured green
31

peppers (Han, Linton et al. 2001). Singh et al. found that a treatment with a higher
concentration of aqueous ClO2 for 5 minutes produced a lower (1.55-log) reduction of
E. coli O157:H7 on lettuce than a treatment with a lower concentration of aqueous ClO2
for 30 seconds which resulted in a reduction of almost 5 times (5-log) when E. coli
ATCC 11229 was suspended in water. These results show that aqueous ClO2 is not as
effective at killing pathogens on vegetables such as lettuce as it is in killing pathogens in
aqueous suspensions (Singh, Singh et al. 2002; Lee, Costello et al. 2004). At the same
time, these results show that fruits or vegetables, just as the lettuce in the example
above, could provide protection or sites for microorganisms to hide, as they can be
attached to its surfaces in locations inaccessible to aqueous treatments such as
hydrophobic pockets, irregular folds, or underneath leaves or calyxes (Lee, Costello et
al. 2004; Staschower 2012).
Seo et al. also found that a liquid solution would be incapable of inactivating
microorganisms attached to the produce’s broken trichomes, cracks, stomata and cut
edges after finding many live E. coli O157:H7 in the stomata of lettuce after treatment
with a 20 mg/L aqueous solution (Seo and Frank 1999).
On the other hand, treatments with gaseous ClO2 proved to be efficacious in a
wide number of examples such as: Du et al (2002) who demonstrated that ClO2 gas
inactivated E. coli O157:H7 on apple surfaces, specifically on its calyx and cavities; Sy
et al. who found significant reductions of Salmonella populations on blueberries,
strawberries, and raspberries, as well as reductions of Salmonella, E. coli O157:H7, L.
monocytogenes, yeasts and molds on cabbage, carrots, apples, tomatoes and peaches

32

(Sy, McWatters et al. 2005; Sy, Murray et al. 2005); Han et al. who found that
treatments were highly effective in reducing E. coli O155:H7 and L. monocytogenes on
strawberry surfaces (Han, Selby et al. 2004); Lee et al. who showed efficacy of the gas
against E. coli O155:H7 on lettuce leaves (Lee, Costello et al. 2004); Du, Han, and
Linton who found gaseous ClO2 effective against both E. coli O155:H7 and L.
monocytogenes on apple surfaces (Du, Han et al. 2002; Du, Han et al. 2003); Han et al.
who found that treatments were effective in the inactivation of both E. coli O155:H7 and
L. monocytogenes on green pepper surfaces (Han, Floros et al. 2001; Han, Linton et al.
2001) and many others as shown in Appendix 1.
2.7. Methods of Generating Chlorine Dioxide Gas
Being highly unstable and explosive under pressure, ClO2 gas cannot be
compressed or stored commercially due to safety reasons (United States Environmental
Protection Agency (EPA) 1999). Thus, ClO2 gas cannot be shipped and must be
produced at the point of use (Keskinen and Annous 2011). Commercial generators of
ClO2 gas use sodium chlorite as the precursor feedstock chemical for the production of
ClO2 gas (Kaczur and Cawlfield 2000), although in the last two decades production of
ClO2 gas from sodium chlorate has been explored especially at water treatment plants
(United States Environmental Protection Agency (EPA) 1999). The principal generation
reactions that use sodium chlorite to produce ClO2 gas are listed below (United States
Environmental Protection Agency (EPA) 1999):

33

NaClO2 + Cl2 (gas) → 2 ClO2 (gas) + 2 NaCl
2 NaClO2 + HOCl → 2 ClO2 (gas) + NaCl + NaOH
5 NaClO2 + 4 HCl → 4 ClO2 (gas) + 5 NaCl + 2H2O
This set of reactions show how generators can differ even if the same precursors
are used.
2.7.1. Constant gas flow generators
The advantage of these type of generators is that they ensure the production of
ClO2 gas at constant flow rate and concentration. The generators are equipped with a
control system that allow the reactions to occur continuously as needed in order to
compensate for the depleted or consumed gas in the system. An example of a
commercial generator of this type is the system manufactured by ChloriDiSys Solutions
Inc. (Lebanon, NJ) which is shown in figure 5. Chlorine gas (2%) in nitrogen flows into
three sodium chlorite cartridges to generate ClO2 gas. Both the flow rate and the
concentration are controlled by a set of sensors and the integrated UV-VIS photometric
detection system which alert the system for more generation if needed (Czarneski and
Lorcheim 2005).

34

FIGURE 5. Schematic diagram of a Minidox-M from ChlorDiSys ClO2 gas
generation system, adapted from (Czarneski and Lorcheim 2005)
2.7.2. Sachet
This type of generation use a mix of solid precursors to initiate the reaction and
as a result the concentration is variable over time. The concentration starts to build up
until the target (maximum) concentration is reached where the concentration would start
to decrease due to the depletion of precursors. This way to generate ClO2 is done using
a sachet and it is considered a non-continuous process (Staschower 2012). An example
of a commercially available product are the Z-series technology developed by ICA
TriNova (Newnan, GA) where ClO2 gas is generated by mixing two dry solids where
the precursor could either be sodium chlorite or sodium chlorate, and activator which is

35

an acid (HCl, H2SO4, citric, acetic, etc.), following the following reaction (Linton, Han et
al. 2006; Staschower 2012):
−

+

−

−

+

4ClO2 + 4H → 2ClO2 + ClO3 + Cl + 2H + H2O
To ensure that the gas generated is released into the required atmosphere, the
components are mixed inside a permeable pouch which permits the gradual release of
gas either into water or air.
2.7.3. Liquid solution
Due to its volatile nature, a concentrated ClO2 solution will vaporize if it is
maintained inside a sealed chamber, creating a ClO2 atmosphere above the solution
(Mahovic, Bartz et al.). Every time the gas is consumed, more ClO2 will be generated
from the solution (Staschower 2012).

A commercial example of this type of ClO2 gas generation are the set of dry
mixtures manufactured by ICA TriNova (Newnan, GA) which similarly to the sachet
technology described above, are supposed to be mixed inside a permeable pouch and
then submerged inside deionized water for the gas to dissolve in it. In this case, NaClO2
is used as the precursor while a formulation of FeCl3 serves as an acid activator
(Mahovic, Bartz et al.)

36

2.8. Absorption of ClO2 by fresh produce and available methods for its
quantification
As previously discussed, ClO2 participates in a series of oxidative
−

−

reactions which result in its break down into chlorate (ClO3 ) and chlorite (ClO2 ) ions,
−

which can at the same time further transform into chloride (Cl ) (United States
Environmental Protection Agency (EPA) 1999; Gómez-López, Rajkovic et al. 2009).
This is a particularly important fact to consider when evaluating ClO2 gas as a sanitizer
of fruits and vegetables since the cell structures, pigments, and microflora existing on
produce surfaces will absorb ClO2 which will then oxidize and break down upon contact
(Netramai 2011), leaving surface residues that could later cause the formation of
chlorinated byproducts (Trinetta, Vaidya et al. 2011). A few researchers have
investigated the absorption of chlorine dioxide by fresh produce, and they have found
not only evidence of residues in the surface of the produce, but also other clues that
lead to the hypothesis that ClO2 and its residues actually permeate inside the tissue of
various produce types. For example, using the amperometric titration method, Han et al.
−

(2004) recovered residues of ClO2 and ClO2 on strawberries after 1 week of treatment
−

−

(Han, Selby et al. 2004); Trinetta (2011) found residues of ClO2, ClO2 , and ClO3 up
to two weeks after treatment with ClO2 in surfaces of hydroponic tomatoes, navel
oranges, apples, strawberries, lettuce, alfalfa sprouts and cantalopes (Trinetta, Vaidya
-

et al. 2011); Netramai (2011) found residues of ClO2 and ClO2 in lettuce after
37

treatments with ClO2 at different concentrations and exposure times (Netramai 2011);
-

and more recently Staschower (2012) recovered ClO2 and ClO2 on romaine lettuce
and cherry tomatoes after treatments ClO2 using sachet, concentrated solution and a
commercial gas generator (Staschower 2012). Staschower (2012) also recovered
significantly higher amounts of residues on the surface of lettuce than on tomatoes
which suggest that differences in the plant tissue and protective layers influence the
absorption of ClO2 by fresh produce. Another factor that may affect absorption are
bruises and cuts present in the produce being treated, which creates susceptible areas
for easier deterioration and biochemical reactions (Allende, Tomás-Barberán et al.
2006). This was proven by Netramai (2011) who evaluated the difference between ClO2
absorbed by shredded lettuce and whole lettuce, finding that shredded lettuce absorbed
approximately 10 times more than whole lettuce did. After analyzing absorption
amounts of washed versus not washed lettuce, she also found that the presence of
water did not significantly affect the absorption of ClO2. Finally, both Netramai (2011)
and Staschower (2012) found higher amounts of residues with increased ClO2
concentration and exposure time, which means that ClO2 absorption is affected by the
severity of the treatment (Netramai 2011; Staschower 2012).
Analytical methods currently available for the measurement of ClO2 gas
concentration are also used by researchers to measure ClO2 gas absorption by fresh

38

produce after dissolving it in water (Keskinen and Annous 2011). These methods are
described below.
2.8.1. Amperometric Titration Method
The Amperometric titration for quantification of ClO2 and its by-products is based
on the pH-dependent oxidation principle of potassium iodide (KI) to selectively
distinguish the various oxychlorine species from each other (Kaczur and Cawlfield
2000). The reaction of chlorine species with KI at various pH buffered conditions are
shown below (Aieta, Roberts et al. 1984):
−

−

Cl2 + 2I → I2 + 2Cl
−

2ClO2 + 2I → I2 + 2ClO2
−

pH 7, 2, <0.1
−

pH 7

+

−

2ClO2 + 10I + 8H → 5I2 + 2Cl + 4H2O
−

+

−

+

pH 2 <0.1

−

ClO2 + 4I- + 4H → 2I2 + Cl + 2H2O
−

−

pH 2, <0.1

ClO3 + 6I + 6H → 3I2 + Cl + 3H2O

pH<0.1

Common titrants for the determination of chlorine compounds are sodium
thiosulfate (Na2S2O3) and phenyl arsine oxide (C6H5AsO) which reactions with KI are
shown below (Kaczur and Cawlfield 2000):
+

C6H5AsO + I2 + 2H2O → C6H5As(OH)2 + 2H + 2I
+

2Na2S2O3 + I2 → Na2S4O6 + 2Na + 2I

−

−

These titrations can be performed manually using either a starch indicator for end
point or more accurately by amperometric methods. A practical method outlined for

39

titration with C6H5AsO is described by the standard method 4500-ClO2 C (Appendix 2)
where free chlorine, chloramines, chlorite and chlorine dioxide are measured
separately. The method for the titration with Na2S2O3 is outlined by ICA Trinova and it
is presented in Appendix 3.
2.8.2. Colorimetric Method
The colorimetric method is an analytical method commonly used for qualification
and sometimes quantification of compounds. It correlates color with concentration,
understanding that higher intensity will mean higher concentration of the specific
compound.
Although different colorimetric methods vary in the use of the indicator
(Greenberg, Clesceri et al. 1992), most of them follow the steps described in 4500-ClO2
D standard method which involves the reaction between ClO2 and N,N-diethyl-pphenylenediamine (DPD) to form an oxidized product that is measured at 550 nm
(Pepich, Dattilio et al. 2007). The method 4500-ClO2 D is popular, but subject to
interferences from other oxychlorine oxidizing species present (Kaczur and Cawlfield
2000). Another colorimetric method is the Environmental Protection Agency (EPA)
Method 327.0 which uses lissamine B and horseradish peroxidase which show fewer
interferences (Pepich, Dattilio et al. 2007).
2.8.3. Ion chromatography
The EPA describes the method in de document “Method 300.0 or 300.1 Determination of inorganic anions in drinking water by Ion Chromatography” by which
40

common inorganic anions and inorganic disinfection by-products are determined in
reagent water, surface water, ground water, and finished drinking water (Hautman,
Munch et al. 1997). The technique is able to determine multiple anions in a single
analysis and it involves eluding an IC-pack column (Dionex, Sunnyvale, CA.) with a
solution of 2 mM Na2CO3/0.75 mM NaHCO3 at a flow rate of 2 mL/min which is used as
the carrier (Trinetta, Vaidya et al. 2011). Ions present in the sample solutions are then
separated as they interact with the column content and detected at the exit with a
conductivity cell (Hautman, Munch et al. 1997). It is a very sensitive method for
measuring chlorite and chlorate by-products, with a detection limit of 0.01 mg/L for
chlorate and 2 mg/L for chloride (Trinetta, Vaidya et al. 2011).
2.8.4. UV spectroscopy
UV spectroscopy is a common method used to measure both aqueous and
gaseous ClO2. Due to its efficiency and accuracy, this method is commercially used on
process analyzers for the online control of ClO2 gas generation systems (Kaczur and
Cawlfield 2000). Due to its natural double-bond character, ClO2 absorbs UV light, and
so it can be detected (Robinson 2005). Researchers have studied ClO2 spectrum and
UV absorption at different wavelengths and temperatures. Kaczur et al (2000) found
that its maximum absorption happens near 360 nm wavelength and its molar extinction
−1

coefficient is more accurately valued at 1250 (M cm)

when using high resolution,

narrow bandwidth spectrophotometers (Kaczur and Cawlfield 2000). Wahner et al.
(1987) however, found that maximum absorption of ClO2 happened in the vicinity of 351

41

nm wavelength and reported values of absorption cross section in a wavelength region
of 240-480 nm at three different temperatures (Wahner, Tyndall et al. 1987). The values
reported by Wahner were in agreement with those reported by Kromminga (2003) who
used a Bruker IFS- 120 HR Fourier-transform spectrometer designed for operation in
the UV-Vis spectral range with a beamsplitter made of UV-Quartz as the detector
(Kromminga, Orphal et al. 2003).
The majority of the analytical methods mentioned above are approved by the US
Environmental protection Agency (EPA), the National Institute for Occupational Safety
and Health (NIOSH), Association of Official Analytical Chemists (AOAC) and the
American Public Health Association (APHA) (Agency for Toxic Substances and Disease
Registry Division of Toxicology and Envirommental Medicine 2004)
2.9. Toxicity, Risks and Regulations of Chlorine Dioxide Gas
−

Considering that disinfection by ClO2 is followed by the formation of ClO2 and
−

ClO3 , when evaluating the safety of ClO2 gas as disinfectant, the toxicity of both the
sanitizer and its by-products should be taken into consideration.
ClO2 gas is considered a hazard material because it is unstable and explosive at
concentrations above 10.1 KPa partial pressure in air. If involved in a fire, chlorine
dioxide would serve as source of oxygen thus allowing its propagation (Kaczur and
Cawlfield 2000). For these reasons, the gas cannot be compressed or stored
commercially (United States Environmental Protection Agency (EPA) 1999) and must
be produced exclusively at the point of use (Keskinen and Annous 2011). At the same
time, ClO2 and its by-products are considered a hazard because there is evidence to

42

prove that exposure to large amounts of these substances may provoke several health
problems to animals and humans.
According to the Agency for Toxic Substances and Disease Registry Division of
Toxicology and Environmental Medicine, after exposure to a hazardous substance, it is
important to consider the precautions to be taken and all the factors that would
categorize the degree of the hazard such as the type of substance, the exposure dose,
the exposure duration, the organs affected and the manner of exposure (e.g. if the
substance was inhaled, spilled, ingested, etc…), the age, sex, and health situation of
the exposed individual (Agency for Toxic Substances and Disease Registry Division of
Toxicology and Envirommental Medicine 2004).
Animal tests conducted in rats showed that after treatments with chlorite in
drinking water at concentrations of 100 mg/L and above resulted in decreased red blood
cell counts and hemoglobin concentrations after 30 and 60 days of exposure
(Heffernan, Guion et al. 1979). Abdel-Rahman et al. studied the effects of ClO2, ClO2

−

−

and ClO3 in both rats and chickens for periods of up to 9 months and found signs of
anemia especially in those exposed to drinking water with concentrations of 10 and 100
−

mg/L of ClO3 (Abdel-Rahman, Couri et al. 1979).
The Agency for Toxic Substances and Disease Registry Division of Toxicology
and Environmental Medicine describes that ClO2 gas acts primarily as a respiratory
tract and ocular irritant, producing eye, throat, nose, and lung irritation. However, the
−

greatest potential for human exposure to ClO2 and ClO2 is via oral exposure by
drinking water where if ingested in large amounts, can cause mouth, esophagus and

43

stomach irritation as well as it may result in increased levels of methemoglobin in the
blood which reduces the ability of oxygen to bind with hemoglobin (Heffernan, Guion et
al. 1979; Agency for Toxic Substances and Disease Registry Division of Toxicology and
Envirommental Medicine 2004).
−

−

In a review of toxicological effects of ClO2, ClO3 and ClO2 , Couri et al.(1982)
showed the adverse effects of the substances in hematologic parameters such as
alterations in erythrocyte morphology and osmotic fragility, as well as inhibition of DNA
synthesis in several organs (Couri, Abdel-Rahman et al. 1982).
Qingdong (2006) tested the effect of ClO2 and its by-products in water ingested
by rats at a concentration 120 times higher than for humans (553 mg/L of ClO2, ClO2

-

-

and ClO3 ) during 90 days. The study showed that the ingestion did not affect weight
gain, food utilization, indexes of blood and serum, liver/bodyweight, or
kidney/bodyweight (Qingdong, Guangming et al. 2006).
Other studies used humans as experimental subjects where volunteers ingested
water treated with 5 mg/L ClO2 (0.036 mg/kg/day) for 84 days and no adverse health
effects were detected (Condie 1986). Gomez-Lopez et al. didn’t find health effects in
populations living in areas where water was treated with ClO2 for 12 weeks (GómezLópez, Rajkovic et al. 2009).
As reported by The Agency for Toxic Substances and Disease Registry Division
of Toxicology and Environmental Medicine though, it is not likely that humans would be
-

exposed to critical levels of ClO2 or ClO2 in the drinking water, since an average

44

human male (70 Kg reference bodyweight) drinking 2 liters of water at a concentration
-

of 1 mgClO2 /L (minimum level allowed by EPA), would be consuming less than 1% of
what has been the lowest observed adverse-effect level (LOAEL) for repeated oral
exposure to chlorite.
Additional studies are necessary to evaluate the carcinogenic effect of ClO2, but
previous studies have shown no dermal cancer risk (United States Environmental
Protection Agency (EPA) 2006). Similarly, Condie et al.(1986), did not detect
development of any tumors after exposure to ClO2 in mice (Condie 1986). Thus, based
on the limited information available , the International Agency for Research on Cancer
determined that neither sodium chlorite or ClO2 are classifiable as to human
carcinogenicity (Gómez-López, Rajkovic et al. 2009).
For all these reasons, consumption and ingestion of ClO2 and its by-products
raise up concerns due to potential adverse health effects, and its use as a sanitizer is
regulated by different organizations. Examples of those regulations are listed below.
The liquid form of ClO2 was first registered as a surface disinfectant and sanitizer
by the EPA under the Federal Insecticide, Fungicide, and Rodenticide Act in 1967 for
use in livestock barns, bottling plants, food processing plants, and other manufacturing
and storage facilities (United States Environmental Protection Agency (EPA) 2007).
Then in 1988, the EPA registered ClO2 gas as a sterilizing agent for use in
manufacturing, laboratory equipment, environmental surfaces, tools, and clean rooms
(United States Environmental Protection Agency (EPA) 2007), and ever since, ClO2 gas

45

has been increasingly used by manufacturers and the pharmaceutical and medical
device industry as a sterilizing agent to clean rooms and pharmaceutical equipment
(Czarneski and Lorcheim 2005).
After collecting data from 29 water treatment plants using chlorine dioxide
treatment, the EPA determined maximum contaminant levels in treated drinking water
-

for chlorite ions as 0.8 mg/L and 1.0 mg/L for ClO2 and ClO2 respectively (United
States Environmental Protection Agency (EPA) 2006).
The expert committee from The Food and Agricultural Organization / World
Health Organization (FAO/WHO) determined an acceptable daily intake of 0.03 mg/kg
and 0.01 mg/kg of body weight for chlorite and chlorate respectively (Gómez-López,
Rajkovic et al. 2009).
The Food and Drug Administration (FDA) approved the use of ClO2 as an
antimicrobial agent in water used in both poultry processing and fruits and vegetables
washing in an amount not exceeding 3 ppm residual ClO2. Treatments of fruits and
vegetables with ClO2, must be followed by a potable water rinse, blanching , cooking or
canning (United States Food and Drug Administration (FDA) 2012).

In 2001, the FDA approved the incorporation of particles blended with in food
packaging grade of low density polyethylene (LDPE) film for the purpose of delivering a
controlled, time-released dose of ClO2 in the headspace of packaged meats, poultry
and seafood in order to extend their shelf life. The maximum allowed amount of the

46

2

additive was determined to be 17.5 micrograms chlorite/in of package film (United
States Food and Drug Administration (FDA) 2011).

47

BIBLIOGRAPHY

48

BIBLIOGRAPHY

Abdel-Rahman, M. S., D. Couri, et al. (1979). "Kinetics of Cl02 and effects of
Cl02, Cl02-, and Cl03- in drinking water on blood glutathione and hemolysis in rat and
chicken." Journal of environmental pathology and toxicology 3(1-2): 431-449.
Agency for Toxic Substances and Disease Registry Division of Toxicology and
Envirommental Medicine. (2004). "Toxicological Profile for Chlorine Dioxide and
Chlorite." Retrieved June 1st, 2013, from
http://www.atsdr.cdc.gov/toxprofiles/tp160.pdf.
Aieta, E. M. and J. M. Berg (1986). "A Review of Chlorine Dioxide in Drinking
Water Treatment." J. Am. Water Works Assoc. 78 (6): 62-72.
Aieta, E. M., P. V. Roberts, et al. (1984). "Determination of Chlorine Dioxide,
Chlorine, Chlorite, and Chlorate in Water." J. Am. Water Works Assoc. 76: 64-70.
Allende, A., F. A. Tomás-Barberán, et al. (2006). "Minimal processing for healthy
traditional foods." Trends in Food Science & Technology 17(9): 513-519.
Appendini, P. and J. H. Hotchkiss (2002). "Review of antimicrobial food
packaging." Innovative Food Science & Emerging Technologies 3(2): 113-126.
Balla, C. S. and J. Farkas (2007). Minimally Processed Fruits and Fruit Products
and their Microbiological Safety. Handbook of Fruits and Fruit Processing, Blackwell
Publishing: 115-128.
Bean, N. H. and P. M. Griffin (1990). "Foodborne disease outbreaks in the United
States, 1973-1987: pathogens, vehicles, and trends." Journal of Food Protection 53.
Benarde, M. A., B. M. Israel, et al. (1965). "Efficiency of chlorine dioxide as a
bactericide." Applied Microbiology 13(5): 776-780.
Benarde, M. A., W. B. Snow, et al. (1967). "Chlorine Dioxide Disinfection
Temperature Effects." Journal of Applied Microbiology 30(1): 159-167.
Beuchat, L. (1998). "Surface decontamination of fruits and vegetables eaten raw:
a review."
Beuchat, L. R. (1996). "Pathogenic Microorganisms Associated with Fresh
Produce." Journal of Food Protection 59(2): 204-216.

49

Beuchat, L. R. (1999). "Survival of Enterohemorrhagic Escherichia coli O157:H7
in Bovine Feces Applied to Lettuce and the Effectiveness of Chlorinated Water as a
Disinfectant." Journal of Food Protection 62: 845-849.
Beuchat, L. R. (2002). "Ecological factors influencing survival and growth of
human pathogens on raw fruits and vegetables." Microbes and Infection 4(4): 413-423.
Bhagat, A. R. (2010). Modeling Critical Factors to Optimize the Treatment of
Selected Fruits and Vegetables with Chlorine Dioxide Gas Using a Miniaturized
Insustrial-Size Tunnel System. Doctor of Philosophy, Purdue University.
Bhaskara Reddy, M. V., K. Belkacemi, et al. (2000). "Effect of pre-harvest
chitosan sprays on post-harvest infection by Botrytis cinerea and quality of strawberry
fruit." Postharvest Biology and Technology 20(1): 39-51.
Brackett, R. E. (1999). "Incidence, contributing factors, and control of bacterial
pathogens in produce." Postharvest Biology and Technology 15(3): 305-311.
Braun, P. and J. Sutton (1988). "Infection cycles and population dynamics of
Botrytis cinerea in strawberry leaves." Canadian Journal of Plant Pathology 10(2): 133141.
Centers for Disease Control and Prevention. (2011). "CDC Estimates of
Foodborne Illness in the United States." Retrieved 9 April, 2013, 2013, from
http://www.cdc.gov/foodborneburden/PDFs/FACTSHEET_A_FINDINGS_updated413.pdf.
Centers for Disease Control and Prevention. (2013). "Surveillance for Foodborne
Disease Outbreaks — United States, 2009–2010." Retrieved June 1st, 2013, from
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6203a1.htm?s_cid=mm6203a1_w.
Condie, L. W. (1986). "Toxicological Problems Associated With Chlorine
Dioxide." Journal (American Water Works Association) 78(6): 73-78.
Couri, D., M. S. Abdel-Rahman, et al. (1982). "Toxicological Effects of Chlorine
Dioxide, Chlorite and Chlorate." Environmental Health Perspectives 46: 13-17.
Czarneski, M. A. and P. Lorcheim (2005). "Isolator decontamination using
chlorine dioxide gas." Pharmaceutical technology: 124-133.
Dagnas, S. and J.-M. Membre (2013). "Predicting and Preventing Mold Spoilage
of Food Products." Journal of Food Protection 76(3): 538-551.

50

Dantigny, P., M. Bensoussan, et al. (2006). "Standardisation of methods for
assessing mould germination: A workshop report." International journal of food
microbiology 108(2): 286-291.
Dantigny, P., A. Guilmart, et al. (2005). "Basis of predictive mycology."
International journal of food microbiology 100(1-3): 187-196.
Dantigny, P., S. Marín, et al. (2007). "Mould germination: Data treatment and
modelling." International journal of food microbiology 114(1): 17-24.
Daş, E., G. C. Gürakan, et al. (2006). "Effect of controlled atmosphere storage,
modified atmosphere packaging and gaseous ozone treatment on the survival of
Salmonella Enteritidis on cherry tomatoes." Food Microbiology 23(5): 430-438.
Declerck, S., D. D'or, et al. (2001). "Modelling the sporulation dynamics of
arbuscular mycorrhizal fungi in monoxenic culture." Mycorrhiza 11(5): 225-230.
Du, J., Y. Han, et al. (2002). "Inactivation by chlorine dioxide gas (ClO2) of
Listeria monocytogenes spotted onto different apple surfaces." Food Microbiology 19(5):
481-490.
Du, J., Y. Han, et al. (2003). "Efficacy of chlorine dioxide gas in reducing
Escherichia coli O157:H7 on apple surfaces." Food Microbiology 20(5): 583-591.
El-Samahy, S., B. M. Youssef, et al. (2000). "Microbiological and chemical
properties of irradiated mango." Journal of food safety 20(3): 139-156.
Exama, A., J. Arul, et al. (1993). "Suitability of Plastic Films for Modified
Atmosphere Packaging of Fruits and Vegetables." Journal of Food Science 58(6): 13651370.
Garcia, D., A. J. Ramos, et al. (2009). "Predicting mycotoxins in foods: A review."
Food Microbiology 26(8): 757-769.
Gibson, A. M., N. Bratchell, et al. (1987). "The effect of sodium chloride and
temperature on the rate and extent of growth of Clostridium botulinum type A in
pasteurized pork slurry." Journal of Applied Microbiology 62(6): 479-490.
Gómez-López, V. M., A. Rajkovic, et al. (2009). "Chlorine dioxide for minimally
processed produce preservation: a review." Trends in Food Science & Technology
20(1): 17-26.

51

Gougouli, M. and K. P. Koutsoumanis (2012). "Modeling germination of fungal
spores at constant and fluctuating temperature conditions." International journal of food
microbiology 152(3): 153-161.
Greenberg, A. E., L. S. Clesceri, et al. (1992). Standard methods for the
examination of water and wastewater. Washington, D.C., American Public Health
Association: 4-56 - 54.81.
Gustavsson, J., C. Cederberg, et al. (2011). Global Food Losses and Food
Waste. International Congress SAVE FOOD!, Dusseldorf, Germany.
Han, Y., J. D. Floros, et al. (2001). "Response Surface Modeling for the
Inactivation of Escherichia coli O157:H7 on Green Peppers (Capsicum annuum L.) by
Chlorine Dioxide Gas Treatments." Journal of Food Protection 64: 1128-1133.
Han, Y., R. H. Linton, et al. (2001). "Reduction of Listeria monocytogenes on
Green Peppers (Capsicum annuum L.) by Gaseous and Aqueous Chlorine Dioxide and
Water Washing and Its Growth at 7C." Journal of Food Protection 64: 1730-1738.
Han, Y., T. L. Selby, et al. (2004). "Decontamination of Strawberries Using Batch
and Continuous Chlorine Dioxide Gas Treatments." Journal of Food Protection 67:
2450-2455.
Harris, L. J., J. N. Farber, et al. (2003). "Outbreaks Associated with Fresh
Produce: Incidence, Growth, and Survival of Pathogens in Fresh and Fresh-Cut
Produce." Comprehensive Reviews in Food Science and Food Safety 2: 78-141.
Hautman, D. P., D. J. Munch, et al. (1997). "US EPA Method 300.1,
Determination of Inorganic Anions in Drinking Water by Ion Chromatography,
EPA/600/R-98/118, NIST PB-98-169196INZ." (Revision 1.0).
Heffernan, W., C. Guion, et al. (1979). "Oxidative damage to the erythrocyte
induced by sodium chlorite, in vivo." Journal of environmental pathology and toxicology
2(6): 1487.
Jarvis, W. R. (1962). "The infection of strawberry and raspberry fruits by Botrytis
cinerea Fr." Annals of Applied Biology 50(3): 569-575.
Kaczur, J. J. and D. W. Cawlfield (2000). Chlorine Oxygen Acids and Salts,
Chlorous Acid, Chlorites, and Chlorine Dioxide. Kirk-Othmer Encyclopedia of Chemical
Technology. John Wiley and Son. New York. NY, John Wiley & Sons, Inc.: 1-26.

52

Kantor, L. S., K. Lipton, et al. (1997). "Estimating and addressing America’s food
losses." Food Review 20(1): 2-12.
Keskinen, L. A. and B. A. Annous (2011). Chlorine Dioxide (Gas). Nonthermal
Processing Technologies for Food. H. Q. Zhang, G. V. Barbosa-Cánovas, V. M.
Balasubramaniamet al. Oxford, UK, Wiley-Blackwell: 359 - 365.
Knapp, J. E. and D. L. Battisti (2001). Chlorine Dioxide. Disinfection, Sterilization,
and Preservation. S. S. Block. Philadelphia, Lippincott Williams & Wilkins: 215-227.
Kromminga, H., J. Orphal, et al. (2003). "New measurements of OClO absorption
cross-sections in the 325–435 nm region and their temperature dependence between
213 and 293 K." Journal of Photochemistry and Photobiology A: Chemistry 157(2–3):
149-160.
Lee, S.-Y., M. Costello, et al. (2004). "Efficacy of Chlorine Dioxide Gas as a
Sanitizer of Lettuce Leaves." Journal of Food Protection 67: 1371-1376.
Linton, R., Y. Han, et al. (2006). "Gas-/vapor-phase sanitation (decontamination)
treatments." Microbiology of fruits and vegetables: 401-435.
López, S., M. Prieto, et al. (2004). "Statistical evaluation of mathematical models
for microbial growth." International journal of food microbiology 96(3): 289-300.
Lynch, M., R. Tauxe, et al. (2009). "The growing burden of foodborne outbreaks
due to contaminated fresh produce: risks and opportunities." Epidemiology and infection
137(3): 307.
Mahovic, M., J. A. Bartz, et al. "Chlorine Dioxide Gas from an Aqueous Solution:
Reduction of Salmonella in Wounds on Tomato Fruit and Movement to Sinks in a
Treatment Chamber." Journal of Food Protection&#174; 72(5): 952-958.
Netramai, S. (2011). Utilization of chlorine dioxide gas in food packaging
application. Doctor of Philosophy Ph.D., Michigan State University.
O’Beirne, D. and G. Francis (2003). "Reducing the pathogen risk in MAPprepared produce." Novel food packaging techniques 12: 231-275.
Parish, M. E., L. R. Beuchat, et al. (2003). "Methods to Reduce/Eliminate
Pathogens from Fresh and Fresh-Cut Produce." Comprehensive Reviews in Food
Science and Food Safety 2: 161-173.

53

Pepich, B. V., T. A. Dattilio, et al. (2007). "An improved colorimetric method for
chlorine dioxide and chlorite ion in drinking water using lissamine green B and
horseradish peroxidase." Analytica Chimica Acta 596(1): 37-45.
Phillips, C. A. (1996). "Review: Modified Atmosphere Packaging and its effects
on the microbiological quality and safety of produce." International Journal of Food
Science & Technology 31(6): 463-479.
Plattner, K. and A. Perez. (2012). "United States Department of Agriculture
(USDA) Report: Fruit and Tree Nuts Outlook." from
http://www.ers.usda.gov/media/970458/fts354.pdf.
Powelson, R. (1960). "Initiation of strawberry fruit rot caused by Botrytis cinérea."
Phytopathology 50: 491-494.
Qingdong, X., Z. Guangming, et al. (2006). "Study on Subchronic Toxicity of
Chlorine Dioxide and By-Products in Water." Journal of Environmental Science and
Health, Part A: Toxic/Hazardous Substances and Environmental Engineering 41(7):
1347 - 1353.
Rico, D., A. B. Martín-Diana, et al. (2007). "Extending and measuring the quality
of fresh-cut fruit and vegetables: a review." Trends in Food Science & Technology
18(7): 373-386.
Robinson, J. W., Skelly Frame, Frame II (2005). Undergraduate Instrumental
Analysis. New York.
Rodgers, S. L., J. N. Cash, et al. (2004). "A Comparison of Different Chemical
Sanitizers for Inactivating Escherichia coli O157:H7 and Listeria monocytogenes in
Solution and on Apples, Lettuce, Strawberries, and Cantaloupe." Journal of Food
Protection 67(4): 721-731.
Seo, K. H. and J. F. Frank (1999). "Attachment of Escherichia coli O157:H7 to
Lettuce Leaf Surface and Bacterial Viability in Response to Chlorine Treatment as
Demonstrated by Using Confocal Scanning Laser Microscopy." Journal of Food
Protection 62: 3-9.
Seymour, I. J. and H. Appleton (2001). "Foodborne viruses and fresh produce."
Journal of Applied Microbiology 91(5): 759-773.
Singh, N., R. K. Singh, et al. (2002). "Effect of inoculation and washing methods
on the efficacy of different sanitizers against Escherichia coli O157:H7 on lettuce." Food
Microbiology 19(2-3): 183-193.

54

Sloan, A. E. (2013). "Top 10 Food Trends." from
http://sloantrend.com/attachments/article/9/2013-04IFTTop10Trends.pdf.
Staschower, F. (2012). Residues from Chlorine Dioxide Gas Treatment,
Generated by Different Delivery Systems, on Fresh Produce. Master of Science,
Michigan State University.
Sumathi, S., R. F. Cindy, et al. (2004). "Fresh Produce: A Growing Cause of
Outbreaks of Foodborne Illness in the United States, 1973 through 1997." Journal of
Food Protection 67(10): 2342-2353.
Sy, K. V., K. H. McWatters, et al. (2005). "Efficacy of Gaseous Chlorine Dioxide
as a Sanitizer for Killing Salmonella, Yeasts, and Molds on Blueberries, Strawberries,
and Raspberries." Journal of Food Protection 68: 1165-1175.
Sy, K. V., M. B. Murray, et al. (2005). "Evaluation of Gaseous Chlorine Dioxide as
a Sanitizer for Killing Salmonella, Escherichia coli O157:H7, Listeria monocytogenes,
and Yeasts and Molds on Fresh and Fresh-Cut Produce." Journal of Food Protection
68: 1176-1187.
Trinetta, V., N. Vaidya, et al. (2011). "Evaluation of Chlorine Dioxide Gas
Residues on Selected Food Produce." Journal of Food Science 76(1): T11-T15.
United States Department of Agriculture (USDA). (2003). "Pathogen modelling
program, version 7.0." Retrieved June 1st, 2013, from
http://ars.usda.gov/services/docs.htm?docid=6786.
United States Environmental Protection Agency (EPA) (1999). "Alternative
Disinfectants and Oxidants Guidance Manual: Chapter 4. Chlorine Dioxide, EPA 815-R99-014."
United States Environmental Protection Agency (EPA) (2006). Registration
eligibility decision for Chlorine Dioxide and Sodium Chlorite (Case Number 4023).
http://www.epa.gov/oppsrrd1/REDs/chlorine_dioxide_red.pdf.
United States Environmental Protection Agency (EPA). (2007). "Anthrax spore
decontamination using chlorine dioxide." Retrieved June 1st, 2013, from
http://www.epa.gov/pesticides/factsheets/chemicals/chlorinedioxidefactsheet.htm.
United States Food and Drug Administration (FDA). (2004). "Produce Safety
From Production to Consumption: 2004 Action Plan to Minimize Foodborne Illness
Associated with Fresh Produce Consumption." Retrieved June 1st, 2013, from

55

http://www.fda.gov/Food/FoodborneIllnessContaminants/BuyStoreServeSafeFood/ucm1
29487.htm.
United States Food and Drug Administration (FDA). (2011). "Agency Response
Letter GRAS Notice No. GRN 000062." Retrieved June 1st, 2013, from
http://www.fda.gov/Food/IngredientsPackagingLabeling/GRAS/NoticeInventory/ucm153
971.htm.`
United States Food and Drug Administration (FDA). (2012). "Secondary direct
food additives permitted in food for human consumption. Chlorine dioxide. CFR Title 21,
part 173.300." Retrieved 9 Apr 2013, 2013, from
http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfCFR/CFRSearch.cfm?fr=173.300.
Wahner, A., G. S. Tyndall, et al. (1987). "Absorption cross sections for symmetric
chlorine dioxide as a function of temperature in the wavelength range 240-480nm." The
Journal of Physical Chemistry 91(11): 2734-2738.
Williamson, B., B. Tudzynski, et al. (2007). "Botrytis cinerea: the cause of grey
mould disease." Molecular Plant Pathology 8(5): 561-580.
Windholz, M., S. Budavari, et al. (1976). The Merck index. An encyclopedia of
chemicals and drugs. Rahway, New Jersey, Merck & Co.

56

3. IN SITU QUANTIFICATION OF CHLORINE DIOXIDE CONSUMPTION BY FRESH
PRODUCE USING UV-VISIBLE SPECTROSCOPY
3.1. Introduction
Foodborne disease outbreaks originating from contamination of minimally
processed foods like fresh fruits and vegetables have been identified as a main vehicle
for illnesses because they can carry different microorganisms (Bhagat 2010). The
Centers for Disease Control and Prevention (CDC) estimate that at least 12 percent of
the foodborne outbreak associated illnesses in the 1990s were linked to fresh produce
items (United States Food and Drug Administration (FDA) 2004). Therefore, produce
quality assurance is critical to the effort of minimizing foodborne illnesses and diseases
linked to produce contamination. One of the current alternatives available to assure
quality and safety of fresh produce, is the use of sanitizers such as ozone, hydrogen
peroxide, chlorine or chlorine dioxide (ClO2) among others (Allende, Tomás-Barberán et
al. 2006). The use of ClO2 has become more attractive than chlorine as a disinfectant
due to its superior sanitizing efficacy (Akin, Hoff et al. 1982), and because it minimizes
changes in the odor and taste of the disinfected product (Couri, Abdel-Rahman et al.
1982). The efficacy of ClO2 against several microorganisms in different produce has
been extensively documented (Han, Guentert et al. 1999; Han, Floros et al. 2001; Han,
Linton et al. 2001; Du, Han et al. 2003; Han, Selby et al. 2004; Sy, McWatters et al.
2005; Mahmoud, Bhagat et al. 2007; Keskinen, Burke et al. 2009). In most cases,
gaseous ClO2 has proven to be more effective than aqueous ClO2 due to its higher
capacity to reach cavities and to penetrate produce with irregular surfaces, which

57

results in higher inactivation of microorganisms (Han, Linton et al. 2001; Du, Han et al.
2002; Lee, Costello et al. 2004). Moreover, the aqueous form of ClO2 leaves moisture
on the surface of the produce which promotes fungal growth (Trinetta, Vaidya et al.
2011).
To expand the use of ClO2 as a sanitizing agent for fresh produce, a thorough
analysis of the by-products that result upon the interaction between gaseous ClO2 and
the organic matter as well as its implications in human health are needed
(Nieuwenhuijsen, Toledano et al. 2000). ClO2 primary by-products include chlorite,
chlorate, chloride (United States Environmental Protection Agency (EPA) 1999). Some
organizations have established limits for the use of ClO2 for example, the U.S Food and
Drug Administration allowed the use of ClO2 not exceeding 3 mg/L in water as an
aqueous antimicrobial agent to sanitize fresh fruits and vegetables (United States Food
and Drug Administration (FDA) 2012), the EPA determined a maximum ingestion level
of 0.8 mg/L and 1.0 mg/L for ClO2 and chlorite respectively in treated drinking water
(United States Environmental Protection Agency (EPA) 2006), and the Food and
Agricultural Organization / World Health Organization (FAO/WHO) determined
acceptable daily intake values of body weight per day to 0.03 mg/kg and 0.01 mg/kg
chlorite and chlorate respectively (Gómez-López, Rajkovic et al. 2009).
Industrial applications of ClO2 oxidation technology exist in the paper processing
and water treatment industries. Therefore, several analytical methods have been
developed for quantification of ClO2 and its by-products in a solution in order to meet
58

the requirements of these industries. These methods include amperometric titration,
colorimetric method, ion and gas chromatography, and UV-vis spectroscopy (Aieta,
Roberts et al. 1984; Hautman, Munch et al. 1997; Kaczur and Cawlfield 2000; Trinetta,
Vaidya et al. 2011). Using some of these techniques, researchers have been able to
recover and quantify residues of ClO2 and by-product ions after exposing fresh produce
such as tomatoes, lettuce, alfalfa sprouts, and strawberries to gaseous ClO2 (Kim,
Marshall et al. 1999; Tsai, Huxsoll et al. 2001; Netramai 2011; Trinetta, Vaidya et al.
2011). However, these post-exposure quantification techniques mostly are able to
recover readily removed residual by-products left on the surface of the fresh produce by
analyzing a rinsed solution. Furthermore, these methodologies include several
preparation steps before the extracted solutions can be measured, which is not ideal for
-

-

reactive or unstable species such as ClO2, ClO2 , or ClO3 . Considerable levels of ClO2
and by-products were found on different produce even 14 days after exposure (Trinetta,
Vaidya et al. 2011) indicating initial washing is either inefficient or that ClO2 gas may
penetrate deeply into some fresh produce tissue, which may lead to erroneous
quantification of ClO2 and its by-products. Therefore, it is conceivable that consumers
may ingest unacceptably high quantities unreacted ClO2 and its by-products from
treated produce. Thus, the main goal of this study was to develop a new approach to
quantify total ClO2 gas consumption by fresh produce that would minimize sample
preparation. The specific objectives were to develop a system that accurately quantifies
ClO2 concentration while the fresh produce is being treated throughout a wide range of

59

concentrations and to apply the newly developed system to two specific applications by
determining the ClO2 consumption as function of exposure concentration, time and the
surface area, mass, and type of produce.
3.2. Materials and Methods
3.2.1. The Equipment.
Figure 6 shows the schematic of the developed equipment. The system can be
graphically divided into three main parts: (Part I) ClO2 gas generation of concentrations
between 0.01 to 5 mg/L at a constant flow rate of 500 mL/min; (Part II) exposure
chamber that was designed to promote a uniform ClO2 atmosphere for exposed
produce, and (Part III) a ClO2 detection system which can monitor and record ClO2
concentration throughout the desired treatments.

60

Part I

Part II

F

Part III

DAQ
MFC1
Computer

MFC1
Air Inlet

F
MFC3
F

Glass
Chamber

1

2

3

A

1

Spectrophotometer

2

3

Peristaltic Pump

MFC2
Gas Tank

ClO2 gas
generator

Pressure Regulator
Filter
Valve
Three way valve
A

Absorbance

Data Flow
Gas Flow
F

Flow Rate

DAQ Data Acquisition System
MFCi Mass Flow Controller

FIGURE 6. Schematic diagram of the three-component system for quantification of total chlorine dioxide absorption by
fresh produce

61

3.2.2. Part I: Chlorine dioxide generation system
This system includes a Minidox-M ClO2 gas generator from ClorDiSys Solutions
Inc (Lebannon, NY) where chlorine (Cl2) gas (2%) in nitrogen flows into three sodium
chlorite (NaClO2) cartridges to generate ClO2 gas, an airline with filter and pressure
regulator, and three Aalborg GFC (Models: 3NC-07-SS, 3AB-10-SS, and 3AB-06-SS)
mass flow controllers (Orangeburg, NY) with volumetric flow rates of 1 L/min (used for
ClO2 flow), 10 L/min (used for air flow) and 500 mL/min (used for diluting ClO2 gas in
air) respectively. The flow rates of the mass flow controllers were controlled from a
laptop computer using both a data acquisition (DAQ) device from National Instruments
(Austin, TX) which serves as interface, and an in-house written program using LabView
software from National Instruments (Austin, TX).
Filtered clean air at a pressure of 0.13 MPa (20 psi) flows into the first mass flow
controller (MFC1) and simultaneously, ClO2 gas generated by the Minidox-M at a
desired ClO2 concentration flows into the second mass flow controller (MFC2). Although
the ClO2 concentration generated by the Minidox-M is controlled by an integrated UVvis photometric detection system (Czarneski and Lorcheim 2005), its accuracy at
generating the desired concentration was evaluated and described in section 2.4. Both
gases (air and ClO2) are then combined into the third mass flow controller (MFC3),

62

which transports the gas mixture at the exposure ClO2 concentration into the
approximately 11 L glass chamber where the fresh produce is exposed.
At the beginning of each treatment, the chamber was filled with air that needs to
be displaced to build up the targeted ClO2 gas concentration inside. Introducing the
ClO2 gas into the chamber at constant flow rate and concentration could result in a lag
time to build the targeted ClO2 concentration. During this time period, the fresh produce
would be exposed to an undesirably low concentration level. So, a two-step process
was designed to speed up the filling of ClO2 gas and reach the target concentration in
the chamber during the first minute and a half for most treatments while keeping a
constant flow rate. The first step consisted on injecting ClO2 gas at a higher
concentration without air dilution gas to build up a concentration beyond the target level
by about 30 to 40%. In the second step, pure air flowed through MFC1 diluting the ClO2
until the target concentration was achieved. At the end of this two-step process, the
concentration which is monitored by UV-vis spectroscopy over time, reaches
equilibrium.
During the first step, both MFC2 and MFC3 are fully open while MFC1 is closed.
Table 2 shows a summary of the settings of the second step with its respective mass
flow controllers flow rate, and set points of the ClO2 generator for each treatment.
Treatment 0 (Purge) was used every time between treatments to clean the tubes,
chamber and spectrophotometer’s cuvette.

63

Table 2. Mass Flow controllers set points and ClO2 generator concentration for all treatments

Treatment

Concentration
[mg/L]

Injection
a
Time
[s]

Generator
[mg/L]

0

Purge

-

1

0.22

12

4.5

2

0.63

15

12.6

3
4
a
b
c

Duration
c
time
[min]

-

0.41
5

60
170

ClO2

b

MFC1
[mL/min]

MFC2
[mL/min]

MFC3
[mL/min]

1,000

0

500

84

9,000

500

500

7

9,500

500

500

9,000

1,000

500

1,500

500

500

-

17

4

7

20

Time duration of first step filling process described in section 2.1.1
Concentration at which ClO2 generator was manually set.
Duration of experiment

64

3.2.3. Part II: Exposure chamber
An approximately 11 L custom built glass chamber (Dow Chemical, Midland, MI)
of circular base with a diameter of 0.30m (12 in) and height of 0.15 m (6 in) was
equipped with four inlet/outlet ports, a glass lid and a PTFE O-ring; a metallic clamp
made to order by Clampco (Wadsworth, OH) provided a preferred means of sealing; a
stainless steel platform and a set of stainless steel three-legged stands were custom
made by the mechanical shop at the Department of Physics & Astronomy, MSU. Both
the stainless steel platform and stands were designed to hold the fresh produce and
facilitate the circulation of the gas around them ensuring that the whole surface of the
fresh produce was exposed to the ClO2 gas (Figures 7 and 8). Either device could be
used depending on the shape and size of the produce.

FIGURE 7. Picture of stainless steel platform holding strawberries inside the glass
chamber.

65

FIGURE 8. Picture of three legged stands holding strawberries inside the glass
chamber.
To determine if a high flow rate of 500 mL/min had detrimental effects on
produce such as excessive dehydration that may confound results from ClO2 exposure,
a small set of experiments were conducted to determine moisture loss. The tests were
run including exposing fresh produce to three different conditions: in presence of static
air, using air only at 500 mL/min flow rate for a duration of 1000 min, and using ClO2 at
500 mL/min flow rate and a concentration of 0.01 mg/L for the same duration to
determine if this maximum flow rate of MFC3 would not dry and damage the
strawberries, which was the most sensitive of the tested produce. The introduction of
ClO2 gas at 500 mL/min did not statistically affect the strawberry (less than 3% total
moisture loss in 1000 min). Also it was necessary to map the gas flow in the interior of
the chamber because some strawberries could be dried faster in front of the gas port.
Therefore, for the applications outlined in this research (e.g., California strawberries), a
66

flow rate of 500mL/min was appropriated as long as the fruit was placed at least one
inch away from the entrance port regardless of the support used (platform or three
legged stands).
3.2.4. Part III: Chlorine Dioxide’s detection system
The detection system consists of a Shimadzu UV 1800 spectrophotometer
(Shimadzu Scientific Instruments, Columbia, MD) and a 100 mm cylindrical quartz flow
cell with two windows (Part # 34-Q-100, Starna Cells, Atascadero, CA). As the mixed
gas exits the chamber, it is directed into the spectrophotometer for detection of ClO2
where the head space concentration is quantified and recorded as a function of time.
Length of pipes and tubes were kept at short as possible to minimize ClO2
interaction with system components and always covered with 1.27 cm (½ in) thick black
foam to reduce the effect of light that may cause decay of the gas during the
experiments. The glass chamber was also covered with a thick black cloth on every
trial.
3.3. Gas exposure to fresh produce
The system was set to provide two different ClO2 gas exposure profiles. In Route
1, a constant ClO2 concentration was flowed to the chamber throughout the treatment
after an initial high concentration of ClO2 injected to reach the target concentration. In
this case ports A and B of the three-way valve were open such that the system was
open to a well-ventilated atmosphere after the gas flowed through the
spectrophotometer. For Route 2, the system was designed to achieve a target ClO2

67

concentration inside the chamber and then stop ClO2 production while recirulating the
existing gas inside the chamber. This was accomplished by closing valve B and opening
valve C of the three-way valve, creating a closed-loop by switching the three-way to flow
from the spectrophotometer to the peristaltic pump and back into the chamber through
valve 3.
Route 1 simulates a disinfection process where a constant concentration is
steadily applied on fresh produce for a fixed period of time while Route 2 simulates the
kinetics inside a package that has been flushed with a known/target quantity of ClO2
gas, which is depleted over time as it interacts with the fresh produce and other
elements of the packaging environment.
3.4. Wavelength selection for ClO2 detection with Uv-vis spectrophotometer
Primarily, to ensure an accurate determination of the concentration inside the
chamber, it was critical to select the appropriate wavelength at which ClO2 would follow
a linear relationship of the Beer-Lambert law, described in Eq. (5), along the range of
desired concentrations,

‫ ܣ‬ൌ ߝܾܿ

(5)

where A is absorbance (O.D), ε is the molar absorptivity (M-1 cm-1), b is the path
length (cm), and c is concentration (mol/L).
Spectra of ClO2 were obtained by scanning known concentrations of ClO2 at 0,
1.0, 2.5, 5.0, 7.5 and 10 mg/L with the UV-vis spectrophotometer over a wavelength
range from 500 to 300 nm. Calibration curves were built at lambda maxima (λmax)

68

found from the spectra and at three additional wavelengths selected to compare the
experimental values of ClO2 absorption cross section (σ) and molar absorptivity (ε) with
values found in the literature and ultimately verify how accurate the Minidox-M ClO2
generation was.
3.5. Method for quantification of ClO2 consumption by fresh produce.

For each treatment listed in Table 2, curves of the concentration of ClO2 gas in
the empty chamber (control) as function of time were recorded and then compared with
the kinetic curves of the gas when the chamber contains fresh produce. The treatments
in the list were randomly chosen but ensuring combinations of low concentrations with
long exposure times, and high concentrations with short exposure times since it was
expected that ClO2 consumption by fresh produce exposed to a short treatment with a
low ClO2 concentration would be minimum, while exposure to a long treatment with a
high ClO2 concentration would probably damage the produce.
In order to quantify the total ClO2 consumption by the fresh produce, the area
under each curve was determined by numerical integration and the percent differences
between the paired experiments (control vs. full) were multiplied by the total amount of
ClO2 injected into the chamber using Eq.(6):

஺೐ ି஺೑

ܶ‫0݈ܥ ݈ܽݐ݋‬ଶ ‫ ݊݋݅ݐ݌ݎ݋ݏܾܣ‬ሾ݉݃ሿ ൌ ൬

69

஺೑

൰ ‫ݎ݂ כ ݐ כ ܿ כ‬

(6)

Ae is the numerical integral under the curve when the chamber is empty
(control), Af is the numerical integral under the curve when the chamber contains
produce, c is the ClO2 concentration of exposure (mg/L), t is the time of exposure (min)
and fr is the flow rate at which the gas enters the chamber (set at 500 mL/min).
Under the same conditions of temperature, hardware, exposure time and ClO2
concentration, the percentage difference between the two curves was attributed to the
ClO2 consumption by the fresh produce and not to degradation, oxidation or other
reactions as these phenomena were accounted for with the control.
3.6. Application of the new system to quantify ClO2 consumption by fresh
produce
A set of experiments were designed to evaluate and validate the new system by
quantifying total ClO2 consumption when the fresh produce was exposed to different
concentrations and exposure times using Route 1 and 2 settings.
3.6.1. Impact of concentration and exposure on chlorine dioxide
consumption using Route 1
With the system set for Route 1, the new method to quantify consumption
described in section 2.4 was applied using fresh California strawberries previously
inspected for bruises, physical damage, mold, firmness and overall appearance.
Strawberries were exposed to the different treatments shown in Table 2. Each treatment
was repeated 4 times and 18 strawberries were used in each replication. Recorded

70

curves of concentration over time, for both the empty and the full chamber, were used to
calculate ClO2’s consumption by the strawberries.
3.6.2. ClO2 consumption by fresh produce at different concentrations using
Route 2
As explained before, Route 2 simulates the behavior of ClO2 gas injection inside
a package, which depletes over time as it interacts with the interior environment of the
package, including produce. With the system set to Route 2, three different
concentrations: 1, 3 and 5 mg/L were generated both in the presence and in the
absence of the selected California strawberries. The gas was left to recirculate until the
concentration detected was negligible, and the curves were compared. Then, first order
exponential decay reaction kinetics for ClO2 consumption of the form A=Aoe(-kt) were
fitted and are proposed. These equations were corrected for losses in the system.
1.1. Statistical Analysis
All data were analyzed using SAS 9.3 (SAS Institute Inc., Cary, NC). Analyses of
variance were performed using ANOVA. Tukey’s HSD test was used to determine
differences (p≤0.05) among data groups.
3.7. Results and Discussion
3.7.1. Wavelenght selection for ClO2 detection with Uv-vis
spectrophotometer
Figure 9 shows a spectra of ClO2 for concentrations 1, 2.5, 5, 7.5 and 10 mg/L at
22 ° scanned at a wavelength range from 500 to 300 nm. Maximum ClO2 UV light
C

71

absorption (λmax) occured at approximately 360 nm for all concentrations, which is in
agreement with literature data (Wahner, Tyndall et al. 1987). However, at λmax
absorbance values overpass 1.5 a.u. for concentrations 5 mg/L and above suggesting
that using this peak for calculation of molar absorptivity could end up in errors in the
measurement due to very high light absorption by the analyte (low transmittance).

FIGURE 9. Spectra of chlorine dioxide at 1, 2.5, 5, 7.5 and 10 mg/L in a wavelength
range from 500 to 300 nm at 22°
C
Figure 9 also shows that lower wavelengths (range from 350 to 300 nm) provide
lower absorbance values across the different concentrations. For that reason, a plot of
absorbance versus concentration in units mol/L was constructed not only at λmax (360 ±
0.5 nm), but also at 305 ± 0.5, 311 ± 0.5 and 351 ± 0.5 nm (Figure 10). As expected, the

72

linearity of the Beer’s Law was compromised at λmax and at 351nm, while scanning
ClO2 at 305 and 311 nm provided a good linear fit for the entire range of concentrations
2

-1

(r ˃0.998), and their respective linear slopes were 8139.7 and 11004.58 M
respectively.

FIGURE 10. Absorbance (O.D) versus Concentration (mol/L) at 351±0.5, 360±0.5,
311±0.5 and 305±0.5 nm.
These values were used to calculate ClO2 absorption cross section at 22 ⁰C and
the results are presented in Table 3 along with reference values reported by Wahner et
al. 1987 (Wahner, Tyndall et al. 1987; Kromminga, Orphal et al. 2003) where it can be
observed that at both 311 and 305 nm, the percent difference was less than 4 and 0.5%
respectively, indicating a good agreement. For these reasons, the peak at 305 nm was

73

used to scan ClO2 throughout the experiments. Additional experiments were carried out
to determine the lower detection limit for ClO2 gas at this wavelength. Concentrations
below 0.01mg/L were at noise level.
Table 3. Chlorine dioxide’s absorption cross section (σ) at 4 different wavelengths found
experimentally and compared with reference values.
Wavelength
[nm]

Reference
σx10

-17

1
2

[cm ]

1.219
~360 (λmax)
~351
1.275
~311
0.435
~305
0.312
(Wahner, Tyndall et al. 1987)

Measured
-17
2
σ x 10
[cm ]
0.850
0.839
0.420
0.311

Percent
Difference,
%
-30.23
-34.23
-3.36
-0.34

3.8. Applications of the new system to quantify ClO2 consumption by fresh
produce
3.8.1. Impact of concentration and exposure time on chlorine dioxide
consumption using Route 1
Figure 11 shows an example of a plot of ClO2 headspace concentration versus
time for treatment 0.22 mg/L for 84 minutes. The blue solid lines represent the 4
replicates of the ClO2 concentration in the head space of the chamber while it was filled
with 18 strawberries, and the red dash line represents the concentration of ClO2 when
the chamber was empty (control). The control curve indicates that the concentration
peaks to maximum levels at the beginning of the treatment (step 1 described in section
2.1.1), and then few seconds later equilibrated (step 2 described in section 2.1.1). The
74

concentration equilibrates to the reach target concentration before 1.5 min except
treatment 4, which takes 3 min. The curves describing ClO2 concentration of the
headspace while the chamber was filled with fruit (blue curves) also showed a
maximum level at the beginning, sometimes peaking even higher than the control
curves (empty chamber) due to a higher apparent concentration resulting from the lower
volume inside the chamber taken up by the fresh produce itself. The final equilibrium
was slowly reached due to the constant interaction between the ClO2 and the
strawberries. Figure 11 also shows that with the exception of the initial system fill of
ClO2, the concentration in the headspace was lower when strawberries were present,
which is consistent with the consumption of ClO2 gas by organic matter.

FIGURE 11. Absorbance versus time for treatment 1: concentration 0.22mg/L and 84
minutes of exposure
75

Table 4 shows that the average total ClO2 absorbed by strawberries is
significantly different among treatments. A maximum absolute ClO2 consumption occurs
in treatment 1 (0.22 mg/L for 84 min), the treatment with lowest concentration and
highest exposure time. Treatment 4, having the highest concentration and lower time,
reaches almost 62% of what was absorbed on treatment 1 but in just a fraction of the
time. This implies that both factors (concentration and exposure time) contribute to ClO2
consumption by fresh produce.

76

Table 4. Quantification of total chlorine dioxide absorbed by strawberries on each treatment
Average total
Average ClO2
Percent
Exposure Total ClO2
Concentration
absorbed
ClO2
ClO2
Treatment
time
generated
-3
[mg/L]
absorbed
a
[x10 mgClO2/kg
[min]
[mg]
absorbed
[mg]
fruit]
1
0.22
84
9.24
4.18±1.15 (a) 11.30 ±2.69E-06
45.2 (a)
2
0.41
17
3.49
1.56±0.09 (b)
4.54 ±5.42E-07
44.7(a)
3
0.63
7
2.21
0.41±0.13 (c)
1.18 ±4.74E-07
18.8 (b)
7.67 ±8.98E-07
14.9 (b)
4
5
7
17.50
2.61±0.29 (d)
a
average total ClO2 absorbed divided by total ClO2 generated.
Note. Data of average total and absorbed ClO2 are given as mean ± standard deviation. Different letters indicate
significant differences among data in the column (P<0.05 adjusted - Tukey test).

77

Table 4 also shows that treatments 1 and 2 the longer experiments absorbed
around 45% of the injected ClO2 gas, and it was statistically different from treatments 3
and 4 which are the shorter treatments with less than 20% of the injected gas absorbed.
This effect is explained by the kinetics of achieving absorption equilibrium, since
treatments 3 and 4 are suspended before reaching this state, while treatments 1 and 2
are maintained at this steady state for most of the treatment.
3.8.2. ClO2 dioxide consumption by fresh produce using Route 2 at
different concentrations.
The decay of ClO2 inside the chamber when the gas circulated both in the presence
and absence (control) of the strawberries was recorded. For the control, ClO2 decayed
until the concentration was zero or almost zero after approximately 24, 33 and 10 h for
the concentrations 5, 3 and 1 mg/L, respectively. This decay could be due to baseline
system consumption or leaks. In the presence of strawberries, the ClO2 was absorbed
after approximately 11, 4 and 1 h after the chamber was supplied with the gas to reach
the target concentrations of 5, 3 and 1 mg/L, respectively. The difference was attributed
to the consumption of ClO2 by the strawberries. This implies a very fast ClO2
consumption kinetics in strawberry fruit. Figure 12 shows the fitted curves for the decay
of the gas both in presence of strawberries and the control for the concentration 3 mg/L.

78

FIGURE 12. Chlorine dioxide concentration versus time for control and strawberries
after supplying the chamber with 3 mg/L concentration using Route 2.
Table 5 shows the estimated parameters for the first order reaction equations fitted for
ClO2 consumption in the empty and full chamber. In this first order reaction equation
with everything else being equal, smaller value of k yield higher absorbance values
which in turn means that there is greater availability of ClO2 in the headspace, hence
smaller consumption of the gas. For concentrations of 3 and 5 mg/L, the reaction rate
were similar estimated at 1.14E-04 and 3.06E-04, while for the controls, the estimate of
the rate reaction parameter for the three curves (concentrations 5, 3 and 1 mg/L) was
significantly smaller at 3.81E-05 ± 3.31E-06.The order of magnitude difference between
the reaction rates for the control and the treatments are again explained by the
consumption of ClO2 by the strawberries

79

Table 5. First order reaction parameters of chlorine dioxide absorption using Route 2
Concentrations, mg/L
a
0, Control
Parameters
5
3
1
Ao
kx10
2

R

-4

4.14 ± 0.04
-6
1.14 ± 2.21x10

2.82 ± 0.01
-6
3.06 ± 1.14 x10

0.83 ± 0.01
-6
9.78 ± 8.50 x10

2.4 ± 1.62
-6
0.38 ± 3.31 x10

0.9364
0.9978
0.9974
0.995
Average and standard deviation calculated for control across concentrations

a

3.9. Conclusions
A novel method was developed to quantify in situ consumption of ClO2 using an UV-visible spectroscopy
equipment. The equipment could be used for a variety of applications, and two examples were described. This unique
online system configuration is accurate, easy to use and fast in determining ClO2 consumption by fresh produce. In the
future it could be correlated with other valid quantification methods such as amperometric titration to establish a
calibration curve to compare results in both systems. Other potential applications include the determination if surface area
or mass

80

BIBLIOGRAPHY

81

BIBLIOGRAPHY

Aieta, E.M., Roberts, P.V., Hernandez, M., (1984). Determination of Chlorine
Dioxide, Chlorine, Chlorite, and Chlorate in Water. J. Am. Water Works Assoc. 76, 6470.
Akin, E.W., Hoff, J.C., Lippy, E.C., (1982). Waterborne outbreak control: which
disinfectant? Environmental Health Perspectives 46, 7-12.
Allende, A., Tomás-Barberán, F.A., Gil, M.I., (2006). Minimal processing for
healthy traditional foods. Trends in Food Science & Technology 17(9), 513-519.
Bhagat, A.R., (2010). Modeling Critical Factors to Optimize the Treatment of
Selected Fruits and Vegetables with Chlorine Dioxide Gas Using a Miniaturized
Insustrial-Size Tunnel System. Purdue University, West Lafayette, Indiana.
Couri, D., Abdel-Rahman, M.S., Bull, R.J., (1982). Toxicological Effects of
Chlorine Dioxide, Chlorite and Chlorate. Environmental Health Perspectives 46, 13-17.
Czarneski, M.A., Lorcheim, P., (2005). Isolator decontamination using chlorine
dioxide gas. Pharmaceutical technology, 124-133.
Du, J., Han, Y., Linton, R.H., (2002). Inactivation by chlorine dioxide gas (ClO2)
of Listeria monocytogenes spotted onto different apple surfaces. Food Microbiology
19(5), 481-490.
Du, J., Han, Y., Linton, R.H., (2003). Efficacy of chlorine dioxide gas in reducing
Escherichia coli O157:H7 on apple surfaces. Food Microbiology 20(5), 583-591.
Gómez-López, V.M., Rajkovic, A., Ragaert, P., Smigic, N., Devlieghere, F.,
(2009). Chlorine dioxide for minimally processed produce preservation: a review.
Trends in Food Science & Technology 20(1), 17-26.
Han, Y., Floros, J.D., Linton, R.H., Nielsen, S.S., Nelson, E., (2001a). Response
Surface Modeling for the Inactivation of Escherichia coli O157:H7 on Green Peppers
(Capsicum annuum L.) by Chlorine Dioxide Gas Treatments. Journal of Food Protection
64, 1128-1133.
Han, Y., Guentert, A.M., Smith, R.S., Linton, R.H., Nelson, P.E., (1999). Efficacy
of chlorine dioxide gas as a sanitizer for tanks used for aseptic juice storage. Food
Microbiology 16(1), 53-61.

82

Han, Y., Linton, R.H., Nielsen, S.S., Nelson, P.E., (2001b). Reduction of Listeria
monocytogenes on Green Peppers (Capsicum annuum L.) by Gaseous and Aqueous
Chlorine Dioxide and Water Washing and Its Growth at 7C. Journal of Food Protection
64, 1730-1738.
Han, Y., Selby, T.L., Schultze, K.K., Nelson, P.E., Linton, R.H., (2004).
Decontamination of Strawberries Using Batch and Continuous Chlorine Dioxide Gas
Treatments. Journal of Food Protection 67, 2450-2455.
Hautman, D.P., Munch, D.J., Pfaff, J.D., (1997). US EPA Method 300.1,
Determination of Inorganic Anions in Drinking Water by Ion Chromatography,
EPA/600/R-98/118, NIST PB-98-169196INZ. (Revision 1.0).
Kaczur, J.J., Cawlfield, D.W., (2000). Chlorine Oxygen Acids and Salts, Chlorous
Acid, Chlorites, and Chlorine Dioxide, in: John Wiley and Son (Ed.), Kirk-Othmer
Encyclopedia of Chemical Technology, 5th Edition ed. John Wiley & Sons, Inc., New
York. NY, pp. 1-26.
Keskinen, L.A., Burke, A., Annous, B.A., (2009). Efficacy of chlorine, acidic
electrolyzed water and aqueous chlorine dioxide solutions to decontaminate Escherichia
coli O157:H7 from lettuce leaves. International journal of food microbiology 132(2-3),
134-140.
Kim, J., Marshall, M.R., Du, W., Otwell, W.S., Wei, C., (1999). Determination of
Chlorate and Chlorite and Mutagenicity of Seafood Treated with Aqueous Chlorine
Dioxide. Journal of Agricultural and Food Chemistry 47(9), 3586-3591.
Kromminga, H., Orphal, J., Spietz, P., Voigt, S., Burrows, J.P., (2003). New
measurements of OClO absorption cross-sections in the 325–435 nm region and their
temperature dependence between 213 and 293 K. Journal of Photochemistry and
Photobiology A: Chemistry 157(2–3), 149-160.
Lee, S.-Y., Costello, M., Kang, D.-H., (2004). Efficacy of Chlorine Dioxide Gas as
a Sanitizer of Lettuce Leaves. Journal of Food Protection 67, 1371-1376.
Mahmoud, B.S.M., Bhagat, A.R., Linton, R.H., (2007). Inactivation kinetics of
inoculated Escherichia coli O157:H7, Listeria monocytogenes and Salmonella enterica
on strawberries by chlorine dioxide gas. Food Microbiology 24(7–8), 736-744.
Netramai, S., (2011). Utilization of chlorine dioxide gas in food packaging
application. Michigan State University, East Lansing, Michigan.

83

Nieuwenhuijsen, M.J., Toledano, M.B., Elliott, P., (2000). Uptake of chlorination
disinfection by-products; a review and a discussion of its implications for exposure
assessment in epidemiological studies. Journal of exposure analysis and environmental
epidemiology 10(6 Pt 1), 586-599.
Sy, K.V., McWatters, K.H., Beuchat, L.R., (2005). Efficacy of Gaseous Chlorine
Dioxide as a Sanitizer for Killing Salmonella, Yeasts, and Molds on Blueberries,
Strawberries, and Raspberries. Journal of Food Protection 68, 1165-1175.
Trinetta, V., Vaidya, N., Linton, R., Morgan, M., (2011). Evaluation of Chlorine
Dioxide Gas Residues on Selected Food Produce. Journal of Food Science 76(1), T11T15.
Tsai, L.S., Huxsoll, C.C., Robertson, G., (2001). Prevention of Potato Spoilage
During Storage by Chlorine Dioxide. Journal of Food Science 66(3), 472-477.
United States Environmental Protection Agency (EPA), (1999). Alternative
Disinfectants and Oxidants Guidance Manual: Chapter 4. Chlorine Dioxide, EPA 815-R99-014.
United States Environmental Protection Agency (EPA), (2006). Registration
eligibility decision for Chlorine Dioxide and Sodium Chlorite (Case Number 4023),
http://www.epa.gov/oppsrrd1/REDs/chlorine_dioxide_red.pdf.
United States Food and Drug Administration (FDA), (2004). Produce Safety From
Production to Consumption: 2004 Action Plan to Minimize Foodborne Illness Associated
with Fresh Produce Consumption.
United States Food and Drug Administration (FDA), (2012). Secondary direct
food additives permitted in food for human consumption. Chlorine dioxide. CFR Title 21,
part 173.300.
Wahner, A., Tyndall, G.S., Ravishankara, A.R., (1987). Absorption cross sections
for symmetric chlorine dioxide as a function of temperature in the wavelength range
240-480nm. The Journal of Physical Chemistry 91(11), 2734-2738.

84

4. EVALUATION OF CHLORINE DIOXIDE AS AN ANTIMICROBIAL AGAINST
BOTRYTIS CINEREA IN CALIFORNIA STRAWBERRIES
4.1. Introduction
Food spoilage is both a sustainability and a commercial issue since as a result,
visible mold and undesirable odors lead to consumer rejection, which in turn causes
significant economic losses and food waste. According to Kantor et al. (1997), the
annual edible food available to the U.S is reduced by about 27% due to food spoilage
and waste at the retail and consumer level (Kantor, Lipton et al. 1997). This percentage
is especially significant in the context of a life cycle assessment (LCA) study for food
systems where input factors, including energy, water irrigation, packaging and
transportation, are considered (Heller and Keoleian 2000; Peano, Girgenti et al.
2012).This is such a relevant issue that organizations like the EPA and the USDA have
built their efforts to promote their initiatives “U.S. Food Waste Challenge” and the
“EPA's Food Recovery Challenge” where their objectives are to both raise awareness
about food waste management and to help entities of the U.S food chain to meet their
food-waste goals respectively. Within the food production chain, fruits and vegetables
may be contaminated at different stages since they are exposed to multiple sources of
bacteria, parasites and viruses such as insects, irrigation water or rain, manure based
fertilizers, manual handling by workers during the harvest and packaging process, food
processing facilities and transportation among others (Wei, Huang et al. 1995; Yuk,
Bartz et al. 2006). All these factors directly influence their mode of failure and the time
to reach the end of shelflife. Specifically, mold spoilage of food results from a biological
process that begins with contamination by fungal spores which germinate and extend

85

into hyphae to form a visible mycelium over time before the end of the product’s shelf
life (Dantigny, Guilmart et al. 2005). Mold growth is affected by factors such as water
activity (aw), pH and temperature, and is also affected by food constituents like proteins,
carbohydrates, lipids and organic acids (Garcia, Ramos et al. 2009; Dagnas and
Membre 2013). Optimum conditions for mold growth depends on the type of
microorganism, but in general, food spoilage will occur at temperatures above 25° and
C
aw above 0.85 as long as the food system is contaminated with spores and that these
spores are able to germinate (Dantigny, Guilmart et al. 2005; Gougouli and
Koutsoumanis 2012). Therefore, to improve quality of fruits and vegetables, and to help
prolong the shelf life of produce, it is necessary to both understand the behavior of
fungal growth and to identify methods that effectively reduce the population of
microorganisms. Researchers have proposed models to describe microorganism
population growth as a function of time, and to simulate and predict the behavior of
different types of fungi growth at different environmental conditions and at different
growth stages (germination and hyphae extension) such as the logistic, the Gompertz
and the Baranyi model among others (Declerck, D'or et al. 2001; López, Prieto et al.
2004; Dantigny, Bensoussan et al. 2006; Dantigny, Marín et al. 2007).
Understanding how molds behave and what factors influence their growth could,
among others, help identify suitable sanitation technologies that would effectively
reduce the fungi population that cause food spoilage. Chlorine dioxide (ClO2) gas for
example, has been extensively studied for its proven efficacy as a potent sanitizer and
disinfectant capable of reducing populations of pathogenic microorganisms that
compromise food safety (Reina, Fleming et al. ; Benarde, Snow et al. 1967; Costilow,

86

Uebersax et al. 1984; Abdul-Raouf, Beuchat et al. 1993; Han, Linton et al. 2000; Han,
Sherman et al. 2000; Du, Han et al. 2002; Netramai 2011), and while less research has
focused on studying its effect on spoilage molds, the few results found in the literature
are promising and worth exploring . Spotts et al. (1980) investigated the effect of ClO2
on fungi for the control of d’Anjou pear decay and found that ClO2 did not affect the
germination of Botrytis cinerea when concentrations lower than 5 mg/L were used even
when held for 10 minutes; in fact ClO2 was effective only when pears were treated with
10 mg/L concentration for 10 minutes (Spotts and Peters 1980). Roberts et al. (1994)
for example studied the percentage spore mortality of four fungal species after in vitro
exposure to different ClO2 concentrations and times and found that Botrytis cinerea was
one of the most resistant species, sensitive only to high concentrations of 3 µg/mL and
5 µg/mL within 1 minute of exposure (Roberts and Reymond 1994). This information
seems to suggest that i) contrary to what has been documented for pathogenic
microorganisms, it has been challenging to reduce populations of Botrytis cinerea using
treatments with low concentrations of ClO2; and that ii) despite Spotts et al. and Roberts
et al. agreement on the concentration of the effective treatment (5 mg/L), the difference
in exposure times (1 versus 10 minutes) implies that the biological subject where
Botrytis cinerea is present has an effect on the efficacy of ClO2 treatments. Specifically,
strawberries are a unique fruit with irregular shape, seed-studded surfaces, and high
organic acid content (Kallio, Hakala et al. 2000; Han, Selby et al. 2004) which shelf life
is short due to susceptibility to rot-causing pathogens and fast ripening due to high

87

respiration rates (Kim, Kim et al. 2010). Kaye et al. (2005) found that yeast and mold
populations (not specified what type of population) on strawberries were significantly
reduced when treated with gaseous ClO2 at concentrations of 8 mg/L for 120 minutes
(Sy, McWatters et al. 2005). More recently, Aday et al. (2010) found that treatments with
ClO2 on strawberries are capable of extending the quality of the fruit; and Vardar et al.
(2012) found that the incidence of decay of Botrytis cirenea on strawberries was
significantly reduced by treatments with ClO2 with little impact on sensorial
characteristics (Aday, Buyukcan et al. 2012; Vardar, Ilhan et al. 2012).
However, as a result of ClO2 treatments, it and its by-products may persist in the
fruit (United States Environmental Protection Agency (EPA) 1999) requiring knowledge
of the quantities of these toxic materials, particularly in the case of fruits and vegetables
because they are often consumed raw. Therefore a holistic approach to assess the
feasibility of use of ClO2 as a sanitizing technology to reduce Botrytis cinerea on
strawberries should focus on its efficacy, impact on quality, and ClO2 consumption by
the strawberries. For these reasons, the main goal of this study was to determine the
efficacy of ClO2 gas in the delay of onset of Botrytis cinerea, the most common
microorganism responsible for spoilage of strawberries, meanwhile quantifying the
concentration range that does not produce visual damage to strawberries. The specific
objectives of this work were to (i) expose strawberries to different ClO2 treatments to
determine its efficacy while maintaining the color and moisture content of the fruit, (ii) to

88

quantify total ClO2 absorption by the fruits and (iii) to correlate total ClO2 absorption with
concentration and exposure time.
4.2. Materials and Methods
4.2.1. Selection of California strawberries.
California strawberries purchased from a local supermarket store harvested
about 1-week prior to experiments and shipped in refrigerated trucks to Michigan. There
were differences between strawberry batches in terms of the suppliers and cultivars
based on size and shape of the berries. All strawberries were thoroughly inspected.
Only fruits similar in size, color and ripeness that were free of bruises and other damage
were chosen and were randomly assigned to treatments. Raw strawberries were
weighed and conditioned at 22 ± 2⁰C for an hour prior to treatments in order to avoid
condensation during treatments and to maintain a constant temperature inside the
treatment chamber.
4.2.2. Design of experiments.
A set of three completely randomized block designs (RCBD) (hereafter
experiment) was conducted in order to determine appropriate ClO2 sanitizing
treatments on California strawberries for delaying the growth of Botrytis cinerea without
affecting the visual quality of the fruit in terms of bleaching and dryness of the tissue or
calyx. Table 6 shows the different treatments in pairs of concentration and time of
exposure selected for each experiment. It is important to mention that the order of

89

treatments was randomized to account for variability as treatments were performed
sequentially.
Table 6. Summary of treatments 1, 2, and 3
Experiment 1
Concentration
[mg/L]

Exposure Time
[min]

0.01
0.01
0.04
0.08
0.22
0.63
1.26
5
5
Control

191
1000
21
1000
84
7
332
37
7

Replicates
1
1
1
1
2
1
2
1
1
10

No.
strawberries
per treatment
12
12
12
12
12
12
12
12
12
12

Experiment 2
Concentration
[mg/L]

Exposure Time
[min]

0.04
0.22
0.41
0.63
5
Control

21
84
17
7
7

Replicates
4
4
4
4
4
20

No.
strawberries
per treatment
18
18
18
18
18
360

Experiment 3
Incubation
No.
temperature
Replicates
strawberries
per treatment
[⁰C]
0.63
7
16
9
22 ± 2
Control
16
9
22 ± 2
0.63
7
16
9
4±2
Control
16
9
4±2
A
There is a set of control for each replicate of each treatment, therefore there are 4
replicates*5 treatments*18 strawberries per treatment. Note: Incubation temperature for
experiments 1 and 2 was 22 ± 2⁰C. Temperature numbers are reported as mean ±
standard deviation during the treatment
Concentration
[mg/L]

Exposure Time
[min]

90

The first experiment consisted of a 10-point response surface design covering
low and high ClO2 concentrations for short and long periods of time. Intentionally,
combinations of high concentration with long exposure times, and low concentrations
with short exposure times were left excluded because it was anticipated that in the first
case the quality of the produce would be significantly diminished (e.g., excessive
bleaching and dessication), and that in the second case the sanitizing effect of the gas
would be minimized.. Mold growth conditions were 22 ± 2⁰C and approximately 100%
relative humidity.
A second experimental design consisted of four of the conditions from the above
design plus an additional center point treatment of 0.41 mg ClO2/L for 17 minutes. In
this case, 4 replicates of each condition with 18 berries per replicate were used to
increase sample size and improve power level of the experiment A total of 360
strawberries were treated and 360 untreated berries were used as a control.. Mold
growth conditions were 22 ± 2⁰C and approximately 100% relative humidity.
A third experimental design, treatment 0.63 mg/L for 84 minutes was replicated
16 times and mold growth conditions were at 23 ° a nd 4 ° and approximately 100%
C
C
relative humidity. The additional temperature variable was added in order to both avoid
the growth of rhizopus in the strawberries and to delay the rate of fungal growth in both
treated and control samples in an attempt to better distinguish differences between the
treatments and onset of Botrytis growth. A total of 280 strawberries were selected and
treated in batches of 18 strawberries per replication. Another set of 280 strawberries
were selected as the control group.

91

4.2.3. Incubation tubs.
After each ClO2 treatment on experiments 1, 2, and 3, the fruits were assessed
by incubation tests consisting of arranging the fruit on 0.5-cm mesh screen in aluminum
pans (26 x 32 cm) making sure the fruit was equidistantly spaced and separated from
one another (see Figure 13). The tubs were filled with approximately 100 mL of water
below the screen level, and covered with plastic film to maintain a relative humidity
equal or close to 100%.

FIGURE 13. Position and separation of strawberry fruits inside incubation tubs at
approximately 100% relative humidity.

This approach is similar to that used by Schilder et al. (2012) in evaluation of
fungicide efficacy in fruit rot control in strawberries (16). Both mesh screens and

92

aluminum tubs used were properly sterilized by autoclaving prior to experiments. For all
experiments, the incubation tubs were analyzed for fungal growth which was rated daily
(see rating of fungal growth section).
4.2.4. Evaluating efficacy of ClO2 against the growth of Botrytis cinerea.
California strawberries for all experiments were rated individually by establishing
the percent area of each fruit that was infected on a daily basis. The continuous data
collected for experiments 1 and 2 ranged from 0 to 1, where 0 represents no signs of
mold and 1 represents 100% of the area covered with the fungus. Due to the size of
experiment 3, strawberries were rated every other day using a binary code where 0
represents no sign of mold, and 1 represents at least 10% of area covered with the
fungus (the amount of fungal growth that would be visible to a consumer and might lead
to rejection of the fruit – an economically important level of infection).. In other words,
the data collected in experiments 1 and 2 represents the percent growth of fungus while
data collected from experiment 3 represents the probability of a strawberry to be
infected. Rating was stopped when the majority of the strawberries reached
approximately 90% infection.
4.2.5. Fruit quality assessment after treatment.
Photos were taken before and after treatments in order to identify the location
and assess the magnitude of the discoloration or desiccation that could result after
exposure with ClO2. At the same time, pictures were taken daily during rating to record
the fungal growth process. A professional Canon EOS 40D camera equipped with a
zoom lens was placed and fixed on top of the exposure chamber at a distance of
93

approximately 70 cm from the platform on which the berries were placed. The glass
chamber was fixed to the table with Velcro pieces in order to avoid any movement
during exposure. Camera settings, contrast background color, amount of light, amount
of flash lights used and distances to the chamber were fixed and unchanged throughout
treatments to ensure that any changes in the photos from before and after ClO2
exposure were only due to the effect of treatments in strawberries.
4.2.6. Continuous ClO2 gas treatment.
The system for treating strawberries

was described

by Arango et al.

(2013)(Arango, Rubino et al. 2013). Figure 6 from previous chapter shows a schematic
diagram of the system developed for (i) the continuous ClO2 exposure and (ii) the
quantification of ClO2 absorption by fresh produce (e.g., strawberries). In summary, the
equipment could be divided into three main components. Component 1 consists of a
Minidox-M ClO2 gas generator from ClorDiSys Solutions Inc (Lebannon, NY) equipped
with a 98% nitrogen and 2% chlorine gas tank, an airtube with filter and pressure
regulator, and a set of three Aalborg GFC mass flow controllers (Orangeburg, NY). This
component continuously generates and dilutes ClO2 gas in air to reach gas
concentrations below 1 mg/L (minimum ClO2 concentration of the Minidox-M ClO2 gas
generator). The equipment is calibrated so that any ClO2 gas concentration from 0.01 to
5 mg/L could be produced. Component 2 consists of an 11 liter glass chamber with
three inlet/outlet ports, a glass lid, a polytetrafluoroethylene O-ring, a metallic clamp to

94

provide sealing and closure respectively and a stainless steel platform to hold the
produce inside the chamber. The chamber is connected to the mass flow controller 3
(MFC3 from Figure 6) from Component 1 and its main function is to hold the
strawberries and the gas during treatments. Component 3 consists of a Shimadzu UV
1800 spectrophotometer (Scientific Instruments, Columbia, MD) and a 100 mm
cylindrical quartz flow cell with two windows (Quartz spectrophotometer cell, cylindrical,
Starna Cells, Atascadero, CA). This component continuously detects the headspace
concentration inside the chamber before and during treatments upon the gas exit from
the chamber.
At the beginning of each treatment, the chamber has an air atmosphere part of
which was displaced in order for the concentration of ClO2 to build up. For this reason, a
two-step process was designed to speed up the chamber filling. The first step consisted
on injecting only ClO2 gas at a higher concentration than desired until the concentration
in the chamber headspace was past set point. In the second step, air started flowing
and diluting the ClO2 gas to achieve the desired concentration. At the end of the twostep process, the concentration entering the chamber and the concentration in the
chamber headspace equilibrated to the desired concentration and the UV-vis detector
recorded a steady value over time.

95

4.2.7. Quantification of ClO2 absorption by fresh produce

Total ClO2 absorption was calculated by equation (6), as discussed on previous
chapter, after recording curves of the headspace concentration over time for both the
empty and the full chamber (e.g., with strawberries).

Total ClO2 absorptionൌ

ቀ

஺೐ ି஺೑
஺೐

ቁ ‫ݎ݂ כ ݐ כ ܿ כ‬

(6)

where Af is the area under the curve with chamber containing produce (or full), Ae is the
area under the curve with the empty chamber, c is the ClO2 concentration of exposure
(mg/L), t is the time of exposure (min) and fr is the flow rate at which the gas enters the
chamber (always set at 500mL/min).
Under the same conditions of temperature, materials used, exposure time and
concentration, the percentage difference between the curves was attributed to the ClO2
absorption by fresh produce and not to degradation, oxidation or other reactions in the
system since they were already accounted while running the system with the empty
chamber (control).
4.2.8. Statistical Analysis.
To evaluate ClO2 efficacy, statistical analyses were performed using the
repeated measures methodology considering that the same strawberries were
subjected to quality assessment as a function of time. With the repeated measures
methodology, it was possible to compare all treatments with the control independently at

96

each day of rating. Simultaneously, a mathematical growth model was used to fit the
data: the non-linear Gompertz model as described by equation (1):

‫ ݕ‬ൌ ߙ݁ ି௘

షഁሺ೟షഇሻ

(1)

Where y is fungal growth, and α, β, and θ are estimates of parameters
asymptote, growth rate, and inflection point simultaneously (Dantigny, Marín et al. 2007;
Garcia, Ramos et al. 2009). To assess the effect of exposure time and concentration on
ClO2 absorption, a surface response analysis was conducted using JMP (SAS Institute
Inc., Cary, NC) and Matlab (The MathWorks, Inc., Natick, MA). All other statistical
analyses were done using SAS 9.3 (SAS Institute Inc., Cary, NC).
4.3. Results and Discussion
4.3.1. Quality Assessment.
Visual inspection of the exposed California strawberries showed that exposure to
either prolonged treatments such as 0.22 mg ClO2/L for 84 min, 0.01 and 0.08 mg
ClO2/L for 1000 min and 1.26 mg ClO2/L for 332 min, or exposure to high concentration
such as 5 mg ClO2/L for 7 and 37 min discolored strawberries by changing the red
pigment of the fruit from red to pale yellow as well as it dried and bleached the calyx
evidenced by a texture and color change from fresh green to an ashy yellow color
(Figure 14). Furthermore, the degree and area of bleaching increased as the
concentration and/or time increased.

97

FIGURE 14. From left to right quality decay of calyx in strawberries before and after
chlorine dioxide treatments at 5 mg/L concentration and 7 minutes of exposure
Figures 15 and 16 show and example of strawberries before and after exposure
to treatment 5 mg ClO2/L for 7 minutes, where the quality of the fruit was moderately
affected as i) every single treated strawberry had a lighter overall color, ii) some parts of
the tissue were more evidently discolored and damaged, and iii) every single calyx
showed evident color changes and dryness after exposure. On the other hand,
discoloration of strawberries exposed to treatments 0.63 mg ClO2/L for 7 minutes, 0.04
mg ClO2/L for 21 min and 0.01 mg ClO2/L for 191 min was minimum and almost
unnoticeable.

98

FIGURE 15. Strawberries inside treatment chamber before exposure to chlorine dioxide
looking fresh, and with bright tissue color

FIGURE 16. Strawberries inside treatment chamber after exposure to 5 mg/L chlorine
dioxide gas for 7 minutes looking dull, color changed.

99

4.3.2. Quantification of ClO2 absorption by fresh produce.

Total ClO2 absorption by strawberries was calculated for each treatment using
equation 1 and the results are shown in Table 7. Combining this information with the
results from the visual evaluation of the fruit, it can be demonstrated that the quality of
the fruit after exposure (discoloration or bleaching) is in agreement and positively
correlated to ClO2 absorption by the fruit as presented on Figure 17. The data fits an
2

exponential growth model with 2 parameters and r =0.89 shown in equation (7):

ܳ ൌ 100݁ ଴.ଵଵ஺

(7)

Where Q is the percentage of exposed strawberries presenting discoloration between 0
and 1, and A is the total ClO2 absorbed by the strawberries in each treatment.
Table 7. Summary of total chlorine dioxide absorption by California strawberries after
exposure to the different treatments considered in experiments 1, 2, and 3.

A

Percent Strawberries with
discoloration
[%]
0

A

100

Concentration
[mg/L]

Exposure
time
[min]

ClO2 absorption

0.01

191

1.83E-03

0.01

1000

10.9477

0.04

21

0.5483

0.08
0.22
0.63
1.26
5
0.41

1000
84
7
332
7
17

[mgClO2/g fruit]

A
A

73.6638
10.2231 ± 0.7831
1.3675 ± 0.1688
234.8402 ± 1.254
7.8815 ± 0.2618
4.5404 ± 0.1564

5
100
48.5
5
100
65
11

Note. Data in column 3 is given as mean ± standard error. A Treatments not replicated,
therefore only means are presented.

100

FIGURE 17. Relation between percent of strawberries discolored versus total chlorine
dioxide absorption in mgClO2/g fruit
The steep behavior of the graph shows a fast quality deterioration of the strawberries
with small changes in ClO2 absorption. This fact is of important concern because it
leaves a small window for treatments that could potentially be used without affecting the
fruit’s quality. In fact, according to this set of experiments, significant bleaching fraction
of strawberries occur at levels of absorption greater than at least 7.9 mgClO2/g of fruit,
and an acceptable limit for ClO2 absorption would be at least 4.5 mgClO2/g of fruit.
A mathematical model that predicts ClO2 absorption by fresh produce in terms of
concentration and exposure time was developed using a response surface method
(RSM). For the response variable, the analysis of variance indicated that the model was
101

significant (p<0.01) with significant linear and quadratic effects at a significance level
lower than 0.01. The prediction model is described by equation 7 and the respective plot
bounded by the limits of concentrations and exposure times considered in the set of
experiments is shown in figure 18.

‫ݏܾܣ‬஼௟ைమ ൌ 16.62 െ 41.27ܺଵ െ 44.24ܺଶ ൅ 45.34ܺଵ ܺଶ ൅

ଶ
ଶ
17.32ܺଵ ൅ 26.06ܺଶ

(3)

Where AbsClO2 is the response variable absorption of ClO2 in mg ClO2, X1 is the
logarithm of concentration in mg/L, and X2 is the logarithm of time in minutes.

FIGURE 18. Response Surface plot for ClO2 absorption by fresh produce in terms of
concentration and time

102

4.3.3. Fungal growth.
To simulate commercial conditions and fungal loads present on strawberries in
marketing channels as they reach retail stores and consumers, strawberries used in
these experiments were treated as they came from the store, and no inoculation, prewash or calyx removal was performed. Preliminary experiments had also indicated that
inoculation of strawberries with Botrytis cinerea made no difference in the amount of
mold growth that developed as most strawberries tend to be already infected in the field
prior to harvest ((Maas 1995)).
For these reasons, the variability in the degree of infection between strawberries
was high. Therefore, there were cases when treated or untreated strawberries would
remain intact after days of incubation in the aluminum tubs while other strawberries in
the same tray would be completely covered with fungal growth. Furthermore, even
though strawberries within experiments were all from the same lot and same producer,
strawberries between experiments were not, and so the variation in the rate of decay of
strawberries between experiments 1, 2 and 3 was also wide. Evidence of this is that
rating in experiment 1 lasted only for 4 days, after which the majority of strawberries
were rotted, while experiment 2 was suspended after 7 days of incubation, and rating of
strawberries from experiment 3 incubated at 22⁰C was suspended after 13 days.
Differences in cultivar susceptibility, harvest timing, and fungicide programs can
influence the incidence and rate of decay of strawberries by Botrytis cinerea ((Maas
1995; Wise, Gut et al. 2012; Schilder, Gillett et al. 2013)

103

4.3.4. Quality and efficacy assessment of ClO2.
To model repeated measures over time, the fungal growth data collected from
experimental designs 1, 2, and 3 was fitted to a first-order autoregressive model
because in this case, observations more adjacent to each other in time might be more
highly correlated with each other that those observations farther apart in time within
each strawberry. In experiment 1 it was found that both the effect of the treatments
(p<0.0150) and the interaction of treatment with rating day (p<0.0217) was significant.
From Table 8 and Figure 19, it can be observed that the efficacy of the ClO2 gas
against Botrytis cinerea was proven on treatments 5 mg ClO2/L for 7 minutes, 0.22
mgClO2/L for 84 minutes, 1.26 mgClO2/L for 332 minutes and 0.08 mgClO2/L for 1000
minutes, all which as of the second day post treatment showed significantly lower fungal
growth than the control. Other treatments such as 0.63 mg ClO2/L for 7 minutes and 5
mg ClO2/L for 37 minutes showed significant differences from the control but only in
some rating days.

104

FIGURE 19. Repeated measures model over time for all treatments in experiment 1.

Because the objective of these series of experiments was to find treatments
where both efficacy and quality were met, treatments 5 mgClO2/L for 7 minutes and
0.22 mgClO2/L for 84 minutes (both with moderate quality damage after exposure), and
treatments 0.04 mg ClO2/L for 21 minutes, and 0.63 mg ClO2/L for 7 minutes were
repeated in experiment 2. Although the last two treatments were not efficacious, they
were selected because they did not affect the quality of the strawberries after exposure
and replicating them with increased sample size to increase confidence in the efficacy
response was warranted (Figure 17). Additionally, as explained in the design of

105

experiments section, treatment 0.41 mg ClO2/L for 17 min was selected because it was
calculated with equation (7) that it would not damage the fruit after exposure (Q=16.36).
In experiment 2 it was found that the effect of the treatments was significant at
p<0.0585, while the interaction of treatment with rating day was significant at p<0.0101.
From Table 9 and Figure 20, it can be observed that treatments 0.63 mg ClO2/L for 7
minutes, 0.22 mg ClO2/L for 84 minutes and 5 mgClO2/L for 7 min are significant after
rating day 4 at 95% confidence level. These results showed that both criteria ClO2 gas
efficacy and quality of the strawberries after exposure were met with the treatment 0.63
mgClO2/L for 7 min, which was then run alone versus the control in experiment 3.
Further analysis assuming a power of 0.8 and the observed variation within strawberries
to test for significant differences at p<0.05, resulted in a required minimum sample size
of

144

strawberries

106

per

treatment.

Table 8. Repeated measures results for treatments in experiment 1.
Rating Day
2

Treatment
Control

1
9.78 ± 2.99

0.63 mg/L for 7 minutes

1.33 ± 10.39

5 mg/L for 7 minutes

1.00 ± 10.39

0.04 mg/L for 21 minutes

5.83 ± 10.39

5 mg/L for 37 minutes

2.83 ± 10.39

0.22 mg/L for 84 minutes

2.17 ± 10.39

0.01 mg/L for 191 minutes

0.92 ± 10.39

1.26 mg/L for 332 minutes

2.92 ± 7.35

0.08 mg/L for 1000 minutes

0.08 ± 10.39

0.01 mg/L for 1000 minutes

3.08 ± 10.39

A
A
A
A
A
A
A

A
A
A

34.77 ± 2.99

A

13.833 ± 10.39
4.83 ± 10.39

58.45 ± 2.99
A

B
A

24.25 ± 10.39

A

12.50 ± 10.39

B

10.39 ± 10.39

A

19.25 ± 10.39
11.58 ± 7.35

B
B

10.83 ± 10.39

A

24.33 ± 10.39

3

4
A
B

35.17 ± 10.39

B

19.92 ± 10.39

A

51.00 ± 10.39

B

35.25 ± 10.39

B

29.83 ± 10.39

A

43.33 ± 10.39
27.42 ± 7.35

B
B

21.75 ± 10.39

A

48.50 ± 10.39

78.79 ± 2.99

A
A

67.33 ± 10.39

B

43.42 ± 10.39

A

65.83 ± 10.39

B

55.08 ± 10.39

B

49.50 ± 10.39

A

68.92 ± 10.39
54.08 ± 7.35

B
B

41.42 ± 10.39

A

78.00 ± 10.39

Note. Data in columns are given as mean ± standard error. A,B Within columns, means followed by the same letter are not
significantly different (α=0.05).

107

FIGURE

20.

Repeated

measures

model

over

108

time

for

all

treatments

in

experiment

2

Table 9. Repeated measures results for treatments in experiment 2.
Rating Day
Treatment
1
2
3
4
Control
-2.84E-14
0.76 ±
5.90 ±
27.71 ±
A
A
A
A
± 3.80
3.80
3.80
3.80
0.63 mg/L for 7
-2.84E-14
0.46 ±
2.36 ±
15.78 ±
A
A
A
B
minutes
± 3.80
3.80
3.80
3.80
0.49 ±
3.99 ±
27.50 ±
0.41 mg/L for 17
-9.24E-14
A
A
A
A
minutes
3.80
3.80
3.80
± 3.80
0.04 mg/L for 21
-4.26E-14
0.15 ±
6.28 ±
29.93 ±
A
A
A
A
minutes
± 3.80
3.80
3.80
3.80
0.22 mg/L for 84
-5.68E-14
1.08 ±
4.85 ±
15.49 ±
A
A
A
B
minutes
± 3.80
3.80
3.80
3.80
5 mg/L for 7 minutes
-1.38E-12
0.49 ±
1.74 ±
11.14 ±
A
A
A
B
± 3.80
3.80
3.80
3.80

5
52.99 ±
A
3.80
29.85 ±
B
3.80
51.08 ±
A
3.80
49.89 ±
A
3.80
38.68 ±
B
3.80
38.49 ±
B
3.80

6
66.82 ±
A
3.80
52.08 ±
B
3.80
68.42 ±
A
3.80
66.49 ±
A
3.80
58.33 ±
A
3.80
60.13 ±
A
3.80

7
77.28 ±
A
3.80
64.00 ±
B
4.09
86.61 ±
A
3.80
81.39 ±
A
3.80
73.17 ±
C
3.80
73.28 ±
C
3.80

Note. Data in columns are given as mean ± standard error. A,B Within columns, means followed by the same letter are not
significantly different (α=0.05).

109

In experiment 3, 576 strawberries were divided into 4 groups: control at 4 °
C
incubation temperature, control at 22 ° incubation temperature, treated at 4 °
C
C
incubation temperature, and treated at 22 ° incuba tion temperature. The rating of
C
strawberries stored at 4 ° was suspended after 24 days, while the rating of those
C
stored at 22 ° was suspended after 13 days. Fungal growth in these strawberries took
C
almost twice as much than experiment 2, and 3 times as much than experiment 1,
which could be interpreted as resulting from an inherently more disease-resistant batch
of strawberries, or a significantly less contaminated one among other considerations.
Figure 21 shows the results of experiment 3 for both incubation temperatures. It was
found that the simple effects of rating day and temperature were significant while the
effect of the treatment 0.63 mgClO2/L for 7 minutes was not (p<0.3458). However, the
interaction of the treatment across rating days was significant at p<0.0529 and for
strawberries stored at 4 ° it was found that the treatment was significantly better than
C,
the control between days 17 through 20 with p<0.0586, and between 17 through 21 with
p<0.0702. To further understand how ClO2 was being effective at reducing the growth
of Botrytis cinerea in strawberries, the Gompertz model as described in equation (1) for
mold growth was employed.

110

FIGURE 21. Repeated measures model over time for treated strawberries in experiment
3 stored at 4 ° and 22 °
C
C
Figures 22, 23 and 24 show the plots for experiments 1, 2 and 3 fitted to the
Gompertz model, and Table 10 summarizes the results for the three parameters α, β,
and θ for all treatments. One of the advantages of using a parametric model such as the
Gompertz model to understand the effect of the treatments is that the parameters
provide physical meaning to the behavior of the fungal growth, and therefore they allow
for better understanding of the ClO2 gas efficacy. For example, from Table 10 shows
that treatments 0.63 mgClO2/L for 7 min, 5 mgClO2/L for 7 min, 1.26 mgClO2/L for 332
min, and 0.08 mgClO2/L for 1000 min from experiment 1, as well as treatments 0.63
mgClO2/L for 7 min, 5 mgClO2/L for 7 min, and 0.22 mgClO2/L for 84 min from
experiment

2,

show

the

lowest

111

β

and

highest

θ

values.

Table 10. Gompertz model parameters for treatments in experiments 1, 2, and 3.

Treatment
Control
0.63 mgClO2/L for 7 minutes
5 mgClO2/L for 7 minutes
0.04 mgClO2/L for 21 minutes
5 mgClO2/L for 37 minutes
0.22 mgClO2/L for 84 minutes
0.01 mgClO2/L for 191 minutes
1.26 mgClO2/L for 332 minutes
0.08 mgClO2/L for 1000 minutes
0.01 mgClO2/L for 1000 minutes
Treatment
Control
0.04 mgClO2/L for 21 minutes
0.22 mgClO2/L for 84 minutes
0.41 mgClO2/L for 21 minutes
0.63 mgClO2/L for 7 minutes
5 mgClO2/L for 7 minutes
Treatment
Control
0.63 mgClO2/L for 7 minutes
Treatment
Control
0.63 mgClO2/L for 7 minutes

Experiment 1
Asymptote (α)
Growth Rate (β)
1.05 ± 0.15
0.7 ± 15E-04
1.73 ± 3.41
0.5 ± 68E-04
1.11 ± 2.89
0.6 ± 11E-03
0.81 ± 0.34
0.9 ± 70E-04
0.88 ± 0.90
0.7 ± 81E-04
0.81 ± 0.52
0.7 ± 49E-04
1.05 ± 0.87
0.7 ± 68E-04
1.39 ± 2.08
0.5 ± 52E-04
1.02 ± 2.73
0.5 ± 96E-04
1.35 ± 1.21
0.6 ± 54E-04
Experiment 2
Asymptote (α)
Growth Rate (β)
0.82 ± 0.06
0.9 ± 17E-04
0.93 ± 0.10
0.7 ± 14E-04
0.97 ± 0.17
0.6 ± 14E-04
1.02 ± 0.11
0.7 ± 14E-04
1.13 ± 0.34
0.5 ± 16E-04
0.83 ± 0.08
0.9 ± 19E-04
Experiment 3 - Temp 4 °
C
Growth Rate (β)
Asymptote (α)
0.8 ± 0.04
0.4 ± 0.05
0.9 ± 0.08
0.3 ± 0.05
Experiment 3 - Temp 22 °
C
Asymptote (α)
Growth Rate (β)
0.87 ± 0.03
0.6 ± 0.07
0.9 ± 0.04
0.5 ± 0.08

112

Inflection Point (θ)
2.2 ± 25E-04
3.9 ± 0.04
3.9 ± 0.04
2.2 ± 64E-04
2.9 ± 0.02
3.0 ± 0.01
2.8 ± 0.01
3.9 ± 0.03
3.8 ± 0.05
3.0 ± 0.02
Inflection Point (θ)
4.1 ± 14E-04
4.3 ± 21E-04
4.9 ± 36E-04
4.5 ± 20E-04
5.5 ± 69E-04
4.7 ± 15E-04
Inflection Point (θ)
17.2 ± 0.20
18.3 ± 0.41
Inflection Point (θ)
5.3 ± 0.14
5.6 ± 0.20

FIGURE 22. Gompertz model fit for treatments in experiment 1.

FIGURE 23. Gompertz model fit for treatments in experiment 2.

113

FIGURE 24. Gompertz model fit for treatments in experiment 3 both at 4 ° and
C
22°
C
This indicates that the infection growth rate is slower for these treatments and
that as a result, the onset of Botrytis cinerea is delayed. Therefore, in cases where ClO2
gas was found to be effective, it acted as both an inhibitor of initial growth (delay of
onset) and as a regulator along time since the speed at which the disease grew was
significantly slower. These results are in good agreement with the findings from the
repeated measures modeling since the treatments that were selected for their better
efficacy are the same across experiments. Equally, the results from the Gompertz
model show that the treatment selected for experiment 3 behave similarly to the control
especially at incubation temperature 22 ⁰C, which means that ClO2 was not effective at
delaying the onset and growth of Botrytis cinerea on the strawberries.

114

One of the main differences in studies of effects of ClO2 on human-pathogenic
microbes on fruit compared to fungal decay of field-grown strawberries naturally
infected with Botrytis cinerea may be that the former are usually surface contaminants
or applied to the fruit surface by researchers, whereas Botrytis cinerea is a fungal plant
pathogen that infects fruit in the field prior to harvest, such that the fungus may already
be present within plant tissues by the time the ClO2 treatment is applied. This reduces
the capacity of ClO2 to reach and interfere with fungal growth. Furthermore, ClO2
treatment on the berries after shipping across the United States for least 5-7 days, even
though under refrigerated conditions, may be too late to achieve strong and consistent
efficacy of ClO2 as Botrytis cinerea may continue to grow, albeit slowly, during the
shipping process. In addition, regardless of our careful selection process, bruising of
berries during shipping was inevitable, thus increasing the susceptibility of fruit tissue to
fungal decay.
4.4. Conclusions
ClO2 absorption by fresh produce as a function of concentration and exposure
time, the information related to the description of their effects and interactions were
presented by a mathematical model using a response surface methodology design
(RSM. Experiments 1 and 2 showed the efficacy of ClO2 gas treatments on the onset
delay of Botrytis cinerea on strawberries though natural variability was high even when
statistical improvements were observed. However, when the treatment that met both
quality standards and efficacy against the mold was tested again for validation on the

115

final experiment, the results were inconclusive. The differences in the contaminated
proportion of strawberries across treatments, as well as the differences in fungal
resistance of strawberries across experiments are important sources of variation that
could have affected the results and its interpretation. Future work should focus on
minimizing sources of variation before treatments like for example removing the calix
from the fruit in order to ensure that all parts of the strawberry are exposed to the gas,
using strawberries of the same cultivar, field, and harvest period to ensure at least a
more homogeneous degree of contamination across samples. While sequential
application of treatments is unavoidable if a single ClO2 chamber is used for the
experiments, treating fresh strawberries immediately after harvest would eliminate
confounding factors that leave questions about the efficacy of the gas against Botrytis
cinerea.

116

BIBLIOGRAPHY

117

BIBLIOGRAPHY

Abdel-Rahman, M. S., D. Couri, et al. (1979). "Kinetics of Cl02 and effects of
Cl02, Cl02-, and Cl03- in drinking water on blood glutathione and hemolysis in rat and
chicken." Journal of environmental pathology and toxicology 3(1-2): 431-449.
Abdul-Raouf, U. M., L. R. Beuchat, et al. (1993). "Survival and growth of
Escherichia coli O157:H7 on salad vegetables." Appl. Environ. Microbiol. 59(7): 19992006.
Aday, M. S., M. B. Buyukcan, et al. (2012). "MAINTAINING THE QUALITY OF
STRAWBERRIES BY COMBINED EFFECT OF AQUEOUS CHLORINE DIOXIDE
WITH MODIFIED ATMOSPHERE PACKAGING." Journal of Food Processing and
Preservation: no-no.
Agency for Toxic Substances and Disease Registry Division of Toxicology and
Envirommental Medicine. (2004). "Toxicological Profile for Chlorine Dioxide and
Chlorite." Retrieved June 1st, 2013, from
http://www.atsdr.cdc.gov/toxprofiles/tp160.pdf.
Aieta, E. M. and J. M. Berg (1986). "A Review of Chlorine Dioxide in Drinking
Water Treatment." J. Am. Water Works Assoc. 78 (6): 62-72.
Aieta, E. M., P. V. Roberts, et al. (1984). "Determination of Chlorine Dioxide,
Chlorine, Chlorite, and Chlorate in Water." J. Am. Water Works Assoc. 76: 64-70.
Akin, E. W., J. C. Hoff, et al. (1982). "Waterborne outbreak control: which
disinfectant?" Environmental Health Perspectives 46: 7-12.
Allende, A., F. A. Tomás-Barberán, et al. (2006). "Minimal processing for healthy
traditional foods." Trends in Food Science & Technology 17(9): 513-519.
Appendini, P. and J. H. Hotchkiss (2002). "Review of antimicrobial food
packaging." Innovative Food Science & Emerging Technologies 3(2): 113-126.
Arango, J., M. Rubino, et al. (2013). Quantification of Total Chlorine Dioxide
Absorption by UV-visible Spectroscopy. Michigan, US, Michigan State University.
Balla, C. S. and J. Farkas (2007). Minimally Processed Fruits and Fruit Products
and their Microbiological Safety. Handbook of Fruits and Fruit Processing, Blackwell
Publishing: 115-128.

118

Bean, N. H. and P. M. Griffin (1990). "Foodborne disease outbreaks in the United
States, 1973-1987: pathogens, vehicles, and trends." Journal of Food Protection 53.
Benarde, M. A., B. M. Israel, et al. (1965). "Efficiency of chlorine dioxide as a
bactericide." Applied Microbiology 13(5): 776-780.
Benarde, M. A., W. B. Snow, et al. (1967). "Chlorine Dioxide Disinfection
Temperature Effects." Journal of Applied Microbiology 30(1): 159-167.
Beuchat, L. (1998). "Surface decontamination of fruits and vegetables eaten raw:
a review."
Beuchat, L. R. (1996). "Pathogenic Microorganisms Associated with Fresh
Produce." Journal of Food Protection 59(2): 204-216.
Beuchat, L. R. (1999). "Survival of Enterohemorrhagic Escherichia coli O157:H7
in Bovine Feces Applied to Lettuce and the Effectiveness of Chlorinated Water as a
Disinfectant." Journal of Food Protection 62: 845-849.
Beuchat, L. R. (2002). "Ecological factors influencing survival and growth of
human pathogens on raw fruits and vegetables." Microbes and Infection 4(4): 413-423.
Bhagat, A. R. (2010). Modeling Critical Factors to Optimize the Treatment of
Selected Fruits and Vegetables with Chlorine Dioxide Gas Using a Miniaturized
Insustrial-Size Tunnel System. Doctor of Philosophy, Purdue University.
Bhaskara Reddy, M. V., K. Belkacemi, et al. (2000). "Effect of pre-harvest
chitosan sprays on post-harvest infection by Botrytis cinerea and quality of strawberry
fruit." Postharvest Biology and Technology 20(1): 39-51.
Brackett, R. E. (1999). "Incidence, contributing factors, and control of bacterial
pathogens in produce." Postharvest Biology and Technology 15(3): 305-311.
Braun, P. and J. Sutton (1988). "Infection cycles and population dynamics of
Botrytis cinerea in strawberry leaves." Canadian Journal of Plant Pathology 10(2): 133141.
Centers for Disease Control and Prevention. (2011). "CDC Estimates of
Foodborne Illness in the United States." Retrieved 9 April, 2013, 2013, from
http://www.cdc.gov/foodborneburden/PDFs/FACTSHEET_A_FINDINGS_updated413.pdf.

119

Centers for Disease Control and Prevention. (2013). "Surveillance for Foodborne
Disease Outbreaks — United States, 2009–2010." Retrieved June 1st, 2013, from
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6203a1.htm?s_cid=mm6203a1_w.
Condie, L. W. (1986). "Toxicological Problems Associated With Chlorine
Dioxide." Journal (American Water Works Association) 78(6): 73-78.
Costilow, R. N., M. A. Uebersax, et al. (1984). "Use of Chlorine Dioxide for
Controlling Microorganisms During the Handling and Storage of Fresh Cucumbers."
Journal of Food Science 49(2): 396-401.
Couri, D., M. S. Abdel-Rahman, et al. (1982). "Toxicological Effects of Chlorine
Dioxide, Chlorite and Chlorate." Environmental Health Perspectives 46: 13-17.
Czarneski, M. A. and P. Lorcheim (2005). "Isolator decontamination using
chlorine dioxide gas." Pharmaceutical technology: 124-133.
Dagnas, S. and J.-M. Membre (2013). "Predicting and Preventing Mold Spoilage
of Food Products." Journal of Food Protection 76(3): 538-551.
Dantigny, P., M. Bensoussan, et al. (2006). "Standardisation of methods for
assessing mould germination: A workshop report." International journal of food
microbiology 108(2): 286-291.
Dantigny, P., A. Guilmart, et al. (2005). "Basis of predictive mycology."
International journal of food microbiology 100(1-3): 187-196.
Dantigny, P., S. Marín, et al. (2007). "Mould germination: Data treatment and
modelling." International journal of food microbiology 114(1): 17-24.
Daş, E., G. C. Gürakan, et al. (2006). "Effect of controlled atmosphere storage,
modified atmosphere packaging and gaseous ozone treatment on the survival of
Salmonella Enteritidis on cherry tomatoes." Food Microbiology 23(5): 430-438.
Declerck, S., D. D'or, et al. (2001). "Modelling the sporulation dynamics of
arbuscular mycorrhizal fungi in monoxenic culture." Mycorrhiza 11(5): 225-230.
Du, J., Y. Han, et al. (2002). "Inactivation by chlorine dioxide gas (ClO2) of
Listeria monocytogenes spotted onto different apple surfaces." Food Microbiology 19(5):
481-490.
Du, J., Y. Han, et al. (2003). "Efficacy of chlorine dioxide gas in reducing
Escherichia coli O157:H7 on apple surfaces." Food Microbiology 20(5): 583-591.
120

El-Samahy, S., B. M. Youssef, et al. (2000). "Microbiological and chemical
properties of irradiated mango." Journal of food safety 20(3): 139-156.
Ellis, M., K. Cooksey, et al. (2006). "Quality of Fresh Chicken Breasts Using a
Combination of Modified Atmosphere Packaging and Chlorine Dioxide Sachets."
Journal of Food Protection 69: 1991-1996.
Exama, A., J. Arul, et al. (1993). "Suitability of Plastic Films for Modified
Atmosphere Packaging of Fruits and Vegetables." Journal of Food Science 58(6): 13651370.
Garcia, D., A. J. Ramos, et al. (2009). "Predicting mycotoxins in foods: A review."
Food Microbiology 26(8): 757-769.
Gibson, A. M., N. Bratchell, et al. (1987). "The effect of sodium chloride and
temperature on the rate and extent of growth of Clostridium botulinum type A in
pasteurized pork slurry." Journal of Applied Microbiology 62(6): 479-490.
Gómez-López, V. M., A. Rajkovic, et al. (2009). "Chlorine dioxide for minimally
processed produce preservation: a review." Trends in Food Science & Technology
20(1): 17-26.
Gougouli, M. and K. P. Koutsoumanis (2012). "Modeling germination of fungal
spores at constant and fluctuating temperature conditions." International journal of food
microbiology 152(3): 153-161.
Greenberg, A. E., L. S. Clesceri, et al. (1992). Standard methods for the
examination of water and wastewater. Washington, D.C., American Public Health
Association: 4-56 - 54.81.
Gustavsson, J., C. Cederberg, et al. (2011). Global Food Losses and Food
Waste. International Congress SAVE FOOD!, Dusseldorf, Germany.
Han, Y., J. D. Floros, et al. (2001). "Response Surface Modeling for the
Inactivation of Escherichia coli O157:H7 on Green Peppers (Capsicum annuum L.) by
Chlorine Dioxide Gas Treatments." Journal of Food Protection 64: 1128-1133.
Han, Y., A. M. Guentert, et al. (1999). "Efficacy of chlorine dioxide gas as a
sanitizer for tanks used for aseptic juice storage." Food Microbiology 16(1): 53-61.
Han, Y., R. H. Linton, et al. (2000). "Inactivation of Escherichia coli O157:H7 on
surface-uninjured and -injured green pepper (Capsicum annuum L.) by chlorine dioxide

121

gas as demonstrated by confocal laser scanning microscopy." Food Microbiology 17(6):
643-655.
Han, Y., R. H. Linton, et al. (2001). "Reduction of Listeria monocytogenes on
Green Peppers (Capsicum annuum L.) by Gaseous and Aqueous Chlorine Dioxide and
Water Washing and Its Growth at 7C." Journal of Food Protection 64: 1730-1738.
Han, Y., T. L. Selby, et al. (2004). "Decontamination of Strawberries Using Batch
and Continuous Chlorine Dioxide Gas Treatments." Journal of Food Protection 67:
2450-2455.
Han, Y., D. M. Sherman, et al. (2000). "The effects of washing and chlorine
dioxide gas on survival and attachment of Escherichia coli O157: H7 to green pepper
surfaces." Food Microbiology 17(5): 521-533.
Harris, L. J., J. N. Farber, et al. (2003). "Outbreaks Associated with Fresh
Produce: Incidence, Growth, and Survival of Pathogens in Fresh and Fresh-Cut
Produce." Comprehensive Reviews in Food Science and Food Safety 2: 78-141.
Hautman, D. P., D. J. Munch, et al. (1997). "US EPA Method 300.1,
Determination of Inorganic Anions in Drinking Water by Ion Chromatography,
EPA/600/R-98/118, NIST PB-98-169196INZ." (Revision 1.0).
Heffernan, W., C. Guion, et al. (1979). "Oxidative damage to the erythrocyte
induced by sodium chlorite, in vivo." Journal of environmental pathology and toxicology
2(6): 1487.
Heller, M. C. and G. A. Keoleian (2000). Life cycle-based sustainability indicators
for assessment of the US food system, Center for Sustainable Systems, University of
Michigan.
Jarvis, W. R. (1962). "The infection of strawberry and raspberry fruits by Botrytis
cinerea Fr." Annals of Applied Biology 50(3): 569-575.
Kaczur, J. J. and D. W. Cawlfield (2000). Chlorine Oxygen Acids and Salts,
Chlorous Acid, Chlorites, and Chlorine Dioxide. Kirk-Othmer Encyclopedia of Chemical
Technology. John Wiley and Son. New York. NY, John Wiley & Sons, Inc.: 1-26.
Kallio, H., M. Hakala, et al. (2000). "Sugars and acids of strawberry varieties."
European Food Research and Technology 212(1): 81-85.
Kantor, L. S., K. Lipton, et al. (1997). "Estimating and addressing America’s food
losses." Food Review 20(1): 2-12.

122

Keskinen, L. A. and B. A. Annous (2011). Chlorine Dioxide (Gas). Nonthermal
Processing Technologies for Food. H. Q. Zhang, G. V. Barbosa-Cánovas, V. M.
Balasubramaniamet al. Oxford, UK, Wiley-Blackwell: 359 - 365.
Keskinen, L. A., A. Burke, et al. (2009). "Efficacy of chlorine, acidic electrolyzed
water and aqueous chlorine dioxide solutions to decontaminate Escherichia coli
O157:H7 from lettuce leaves." International Journal of Food Microbiology 132(2-3): 134140.

Kim, J., M. R. Marshall, et al. (1999). "Determination of Chlorate and Chlorite and
Mutagenicity of Seafood Treated with Aqueous Chlorine Dioxide." Journal of Agricultural
and Food Chemistry 47(9): 3586-3591.
Kim, J. Y., H. J. Kim, et al. (2010). "The effects of aqueous chlorine dioxide or
fumaric acid treatment combined with UV-C on postharvest quality of ‘Maehyang’
strawberries." Postharvest Biology and Technology 56(3): 254-256.
Knapp, J. E. and D. L. Battisti (2001). Chlorine Dioxide. Disinfection, Sterilization,
and Preservation. S. S. Block. Philadelphia, Lippincott Williams & Wilkins: 215-227.
Kromminga, H., J. Orphal, et al. (2003). "New measurements of OClO absorption
cross-sections in the 325–435 nm region and their temperature dependence between
213 and 293 K." Journal of Photochemistry and Photobiology A: Chemistry 157(2–3):
149-160.
Lee, S.-Y., M. Costello, et al. (2004). "Efficacy of Chlorine Dioxide Gas as a
Sanitizer of Lettuce Leaves." Journal of Food Protection 67: 1371-1376.
Linton, R., Y. Han, et al. (2006). "Gas-/vapor-phase sanitation (decontamination)
treatments." Microbiology of fruits and vegetables: 401-435.
López, S., M. Prieto, et al. (2004). "Statistical evaluation of mathematical models
for microbial growth." International journal of food microbiology 96(3): 289-300.
Lynch, M., R. Tauxe, et al. (2009). "The growing burden of foodborne outbreaks
due to contaminated fresh produce: risks and opportunities." Epidemiology and infection
137(3): 307.
Maas, J. (1995). Compendium of Strawberry Diseases. APS Press, St. Paul, MN.

123

Mahmoud, B. S. M., A. R. Bhagat, et al. (2007). "Inactivation kinetics of
inoculated Escherichia coli O157:H7, Listeria monocytogenes and Salmonella enterica
on strawberries by chlorine dioxide gas." Food Microbiology 24(7–8): 736-744.
Mahovic, M., J. A. Bartz, et al. "Chlorine Dioxide Gas from an Aqueous Solution:
Reduction of Salmonella in Wounds on Tomato Fruit and Movement to Sinks in a
Treatment Chamber." Journal of Food Protection&#174; 72(5): 952-958.
Netramai, S. (2011). Utilization of chlorine dioxide gas in food packaging
application. Doctor of Philosophy Ph.D., Michigan State University.
Nieuwenhuijsen, M. J., M. B. Toledano, et al. (2000). "Uptake of chlorination
disinfection by-products; a review and a discussion of its implications for exposure
assessment in epidemiological studies." Journal of exposure analysis and
environmental epidemiology 10(6 Pt 1): 586-599.
O’Beirne, D. and G. Francis (2003). "Reducing the pathogen risk in MAPprepared produce." Novel food packaging techniques 12: 231-275.
Parish, M. E., L. R. Beuchat, et al. (2003). "Methods to Reduce/Eliminate
Pathogens from Fresh and Fresh-Cut Produce." Comprehensive Reviews in Food
Science and Food Safety 2: 161-173.
Peano, C., V. Girgenti, et al. (2012). Strawberry Production Chain: Application of
a LCA model Acta Horticulturae. (ISHS)
Pepich, B. V., T. A. Dattilio, et al. (2007). "An improved colorimetric method for
chlorine dioxide and chlorite ion in drinking water using lissamine green B and
horseradish peroxidase." Analytica Chimica Acta 596(1): 37-45.
Phillips, C. A. (1996). "Review: Modified Atmosphere Packaging and its effects
on the microbiological quality and safety of produce." International Journal of Food
Science & Technology 31(6): 463-479.
Plattner, K. and A. Perez. (2012). "United States Department of Agriculture
(USDA) Report: Fruit and Tree Nuts Outlook." from
http://www.ers.usda.gov/media/970458/fts354.pdf.
Powelson, R. (1960). "Initiation of strawberry fruit rot caused by Botrytis cinérea."
Phytopathology 50: 491-494.
Qingdong, X., Z. Guangming, et al. (2006). "Study on Subchronic Toxicity of
Chlorine Dioxide and By-Products in Water." Journal of Environmental Science and

124

Health, Part A: Toxic/Hazardous Substances and Environmental Engineering 41(7):
1347 - 1353.
Reina, L. D., H. P. Fleming, et al. "Microbiological Control of Cucumber
Hydrocooling Water with Chlorine Dioxide." Journal of Food Protection 58(5): 541-546.
Rico, D., A. B. Martín-Diana, et al. (2007). "Extending and measuring the quality
of fresh-cut fruit and vegetables: a review." Trends in Food Science & Technology
18(7): 373-386.
Roberts, R. G. and S. T. Reymond (1994). "Chlorine dioxide for reduction of
postharvest pathogen inoculum during handling of tree fruits." Applied and
environmental microbiology 60(8): 2864-2868.
Robinson, J. W., Skelly Frame, Frame II (2005). Undergraduate Instrumental
Analysis. New York.
Rodgers, S. L., J. N. Cash, et al. (2004). "A Comparison of Different Chemical
Sanitizers for Inactivating Escherichia coli O157:H7 and Listeria monocytogenes in
Solution and on Apples, Lettuce, Strawberries, and Cantaloupe." Journal of Food
Protection 67(4): 721-731.
Schilder, A. M. C., J. M. Gillett, et al. (2013). "Evaluation of fungicides for control
of Botrytis fruit rot and leaf spot in strawberry, 2012." Plant Disease Management
Reports.
Seo, K. H. and J. F. Frank (1999). "Attachment of Escherichia coli O157:H7 to
Lettuce Leaf Surface and Bacterial Viability in Response to Chlorine Treatment as
Demonstrated by Using Confocal Scanning Laser Microscopy." Journal of Food
Protection 62: 3-9.
Seymour, I. J. and H. Appleton (2001). "Foodborne viruses and fresh produce."
Journal of Applied Microbiology 91(5): 759-773.
Singh, N., R. K. Singh, et al. (2002). "Effect of inoculation and washing methods
on the efficacy of different sanitizers against Escherichia coli O157:H7 on lettuce." Food
Microbiology 19(2-3): 183-193.
Sloan, A. E. (2013). "Top 10 Food Trends." from
http://sloantrend.com/attachments/article/9/2013-04IFTTop10Trends.pdf.
Spotts, R. and B. Peters (1980). "Chlorine and chlorine dioxide for control of
d’Anjou pear decay." Plant Dis 64(1): 2.

125

Staschower, F. (2012). Residues from Chlorine Dioxide Gas Treatment,
Generated by Different Delivery Systems, on Fresh Produce. Master of Science,
Michigan State University.
Sumathi, S., R. F. Cindy, et al. (2004). "Fresh Produce: A Growing Cause of
Outbreaks of Foodborne Illness in the United States, 1973 through 1997." Journal of
Food Protection 67(10): 2342-2353.
Sy, K. V., K. H. McWatters, et al. (2005). "Efficacy of Gaseous Chlorine Dioxide
as a Sanitizer for Killing Salmonella, Yeasts, and Molds on Blueberries, Strawberries,
and Raspberries." Journal of Food Protection 68: 1165-1175.
Sy, K. V., M. B. Murray, et al. (2005). "Evaluation of Gaseous Chlorine Dioxide as
a Sanitizer for Killing Salmonella, Escherichia coli O157:H7, Listeria monocytogenes,
and Yeasts and Molds on Fresh and Fresh-Cut Produce." Journal of Food Protection
68: 1176-1187.
Trinetta, V., N. Vaidya, et al. (2011). "Evaluation of Chlorine Dioxide Gas
Residues on Selected Food Produce." Journal of Food Science 76(1): T11-T15.
Tsai, L. S., C. C. Huxsoll, et al. (2001). "Prevention of Potato Spoilage During
Storage by Chlorine Dioxide." Journal of Food Science 66(3): 472-477.
United States Department of Agriculture (USDA). (2003). "Pathogen modelling
program, version 7.0." Retrieved June 1st, 2013, from
http://ars.usda.gov/services/docs.htm?docid=6786.
United States Environmental Protection Agency (EPA) (1999). "Alternative
Disinfectants and Oxidants Guidance Manual: Chapter 4. Chlorine Dioxide, EPA 815-R99-014."
United States Environmental Protection Agency (EPA) (2006). Registration
eligibility decision for Chlorine Dioxide and Sodium Chlorite (Case Number 4023).
http://www.epa.gov/oppsrrd1/REDs/chlorine_dioxide_red.pdf.
United States Environmental Protection Agency (EPA). (2007). "Anthrax spore
decontamination using chlorine dioxide." Retrieved June 1st, 2013, from
http://www.epa.gov/pesticides/factsheets/chemicals/chlorinedioxidefactsheet.htm.
United States Food and Drug Administration (FDA). (2004). "Produce Safety
From Production to Consumption: 2004 Action Plan to Minimize Foodborne Illness
Associated with Fresh Produce Consumption." Retrieved June 1st, 2013, from

126

http://www.fda.gov/Food/FoodborneIllnessContaminants/BuyStoreServeSafeFood/ucm1
29487.htm.
United States Food and Drug Administration (FDA). (2011). "Agency Response
Letter GRAS Notice No. GRN 000062." Retrieved June 1st, 2013, from
http://www.fda.gov/Food/IngredientsPackagingLabeling/GRAS/NoticeInventory/ucm153
971.htm.
United States Food and Drug Administration (FDA). (2012). "Secondary direct
food additives permitted in food for human consumption. Chlorine dioxide. CFR Title 21,
part 173.300." Retrieved 9 Apr 2013, 2013, from
http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfCFR/CFRSearch.cfm?fr=173.300.
Vardar, C., K. Ilhan, et al. (2012). "The application of various disinfectants by
fogging for decreasing postharvest diseases of strawberry." Postharvest Biology and
Technology 66(0): 30-34.
Wahner, A., G. S. Tyndall, et al. (1987). "Absorption cross sections for symmetric
chlorine dioxide as a function of temperature in the wavelength range 240-480nm." The
Journal of Physical Chemistry 91(11): 2734-2738.
Wei, C. I., T. S. Huang, et al. (1995). "Growth and Survival of Salmonella
montevideo on Tomatoes and Disinfection with Chlorinated Water." Journal of Food
Protection 58(8): 829-836.
Williamson, B., B. Tudzynski, et al. (2007). "Botrytis cinerea: the cause of grey
mould disease." Molecular Plant Pathology 8(5): 561-580.
Windholz, M., S. Budavari, et al. (1976). The Merck index. An encyclopedia of
chemicals and drugs. Rahway, New Jersey, Merck & Co.
Wise, J. C., L. J. Gut, et al. (2012). Michigan Fruit Management Guide 2013.
Extension Bulletin E-154. East Lansing, MI, Michigan State University.
Yuk, H.-G., J. A. Bartz, et al. (2006). "The Effectiveness of Sanitizer Treatments
in Inactivation of Salmonella spp. from Bell Pepper, Cucumber, and Strawberry."
Journal of Food Science 71(3): M95-M99.

127

5. EXPLORATORY – PRELIMINARY DATA OF EFFECT OF SURFACE AREA AND
WEIGHT ON ClO2 CONSUMPTION BY FRESH PRODUCE

This section shows the preliminary results found when attempting to determine
the correlation between total absorption of ClO2 (measured using new approach
developed) with surface area and weight of the produce, as well as with the surface
residues from the produce recovered through the amperometric titration described
previously (Netramai 2011; Staschower 2012). Romaine lettuce leaves were selected
because it was amenable to easy modification and measurement of surface area and
mass. For this study, the leaves were cut into rectangles of various sizes: 1.27 cm x
2.54 cm, 2.54 cm x 2.54 cm, 5.08 cm x 5.08 cm, and 2.54 cm x 12.7 cm corresponding
2

2

2

2

2

2

2

to 3.23 cm (or 0.5 in ), 6.45 cm (or 1 in ), 25.81 cm (or 4 in ) and 32.26 cm (or 5
2

inch ) surface area respectively. After weighting them, 18 pieces of each surface area
type were placed inside the chamber using the three-legged stands and exposed to a
treatment of 5 mg/L ClO2 concentration for 15 min. Total ClO2 absorption was
quantified in situ using the new method described above. As soon as each treatment
was carried out, the lettuce pieces were washed with 300 mL of distilled and deionized
water for 15 min and kept in the dark, after which the washing solution was titrated for
residual ClO2 and chlorite (ClO2 -) with phenylarsine oxide (C6H5AsO). The results from
both methods were statistically compared.

128

Table 11 shows total ClO2 absorption by the different romaine lettuce sizes
exposed to a concentration of 5 mg/L during 15 min. Absorption of ClO2 increases as
surface area and weight increase. Although Figure 25 shows that there is a positive
linear relationship between absorption with surface area and weight, the surface area
seems to be more strongly correlated to ClO2 absorption by the produce with a higher
2

2

coefficient of determination (R =0.78 versus R =0.58, p≤0.001) and narrower
confidence intervals. At the same time, the slope difference between the two factors
implies that with small surface area increments, the total absorption of ClO2 by the fresh
produce increases faster than with smaller changes in weight. Surface area showed a
greater impact on ClO2 absorption than the overall mass of the fresh produce.
The results of this trial were compared using the amperometric titration method.
The total mass of ClO2 absorbed by the product upon treatment with the gas could
either stay the form of ClO2 or it can freely participate in oxidation reactions and rapidly
−

−

break down to chlorate (ClO3 ) and chlorite (ClO2 ) ions, which could then be
−

converted into to chloride (Cl ). With the UV-vis method described in this document,
only ClO2 was measured as it becomes unavailable inside the chamber due to
absorption by the fresh produce. With the amperometric titration method used in this
−

trial, only ClO2 and ClO2 species could be recovered from the measured surface,
which in part explain the lower percent of mass recover as shown in Table 11. The

129

amperometric titration method does not measure directly the amount of ClO2 absorbed,
but it quantified products absorbed below the surface or bound to the strawberries since
they cannot be recovered through the successive washes that are required by the
method However, it can also be observed that there is an equivalent trend between both
approaches where total ClO2 absorption by produce is impacted by surface area rather
than by weight.

Table 11. Comparison of total chlorine dioxide absorption versus chlorites and chlorine
dioxide residues recovered after exposure
Surface
Area
2
[inch ]

Weight
[g]

Total ClO2
absorption
[mg]

Recovered
ClO2 & ClO2[mg]

%Recovery
[mg]

0.5
1
4
5

3.6 ± 0.30
4.0 ± 0.42
22 ± 0.14
30.82 ± 4.3

1.46 ± 0.54
2.34 ± 0.63
4.88 ± 0.19
5.63 ± 0.01

2.58E-02 ± 1.50E-03
2.24E-02 ± 1.84E-03
2.16E-01 ± 1.20E-02
2.96E-01 ± 2.88E-03

1.8%
1.0%
4.4%
5.3%

130

FIGURE 25. Chlorine dioxide absorption versus weight [g] and surface area [in2]

131

6. CONCLUSIONS AND FUTURE WORK
To be able to consider and fully assess the potential of ClO2 gas as an
antimicrobial agent for fresh produce, it is necessary to be able to monitor and
accurately measure both exposure of the ClO2 gas to the food product and absorption
or uptake of ClO2 by the food.

Quantifying and correlating exposure dose of ClO2 to

efficacy and absorption by the fresh produce is crucial for safety, sanitation and
packaging design purposes among others.
From the safety perspective prior to human ingestion, it is important to assess
and quantify surface by-product residuals as well as potentially ClO2 or other
chlorinated absorbed species by the fresh produce. Results from this assessment could
be transformed into input for the sanitation process, where it would be defined in a case
by case basis the maximum concentration levels and exposure times allowed in order to
keep absorption of ClO2 under safe limits. Finally, from the packaging design
perspective, understanding how much and how fast fresh produce absorbs or uptakes
ClO2 s during ClO2 gas treatments would provide a reference to determine gas releasespeed and concentration required in the headspace.
Also, considering how effective ClO2 gas has proven to be against pathogenic
microorganisms, it makes sense to also investigate its potential against spoilage
microorganisms.
This study provides a new approach to the quantification of ClO2 gas absorption
by fresh produce and the fungicidal potential of the gas against Botrytis cinerea on fresh
132

strawberries. The main goal of this work was to assess the potential use of ClO2 as a
sanitizer in strawberries considering ClO2 consumption by the fruit as well. This work
aimed to answer two specific questions stated in Chapter 1 and the key findings are
described here:
Development of a new system and online method to quantify ClO2 exposure
concentration and consumption by fresh produce.
The system built is robust, easy to use and fast in determining in
situ measurements of ClO2 gas concentration exposure to produce and
concentration of ClO2 consumed by fresh produce. Even though results
presented in this study are based on fresh strawberries, the equipment
allows for study of any other kind of fresh produce provided that it fits in
the current glass chamber.
The two routes on which the system could be operated provides a
good resource for research looking for answers when it comes to
understanding the mechanisms of inactivation by ClO2 of different
microorganism and the way fresh produce consume ClO2.
Some of the key findings with route 2 are the fast consumption
kinetics on strawberry fruit at all three tested concentrations (1, 3, and 5
mg/L).
Assessment on the application of the new equipment by evaluating ClO2 gas
efficacy against Botrytis cinerea on strawberries.
133

A total of three experiments were run to test the efficacy of the gas
against Botrytis cinerea in strawberries as well as the impact of the
treatments on visual quality. Despite the high variability found within
treatments, it was found that the treatment with ClO2 gas concentration
set at 0.63 mg/L and exposure time of 7 minutes was effective.
Although the treatment with ClO2 gas concentration set at 5 mg/L
and exposure time of 7 minutes was effective as well, it damaged the fruit
from the bleaching point of view.
From the assessment done to percent bleaching of strawberries as
a function of total ClO2 consumption by the fruit, it was found that the
visual quality of strawberries is highly sensitive to treatments with ClO2
gas, exposing very rapidly changes in color and texture on both the soft
red tissue and the calyx.
As a complement to this work, and considering the handling and operating
advantages of the new method developed, it would be interesting to balance a
stoichiometric formula and run comparative tests in order to correlate the ClO2
consumption results from this methodology with another current valid method used
to quantify surface residues on fresh produce. This would allow not only for a
validation of the method, but it would serve as a calibration to transform data from
one method to another.
Also, the tests performed on this study provide some preliminary insight from
where it seemed like surface area would be a more significant factor than weight

134

when correlation to total ClO2 absorption, but more experiments are needed in order
to establish this with confidence. Exploring deeper into the effects of surface area
and weight on ClO2 consumption by fresh produce would also help understand the
mechanism of inactivation of ClO2 gas: whether it acts only on the surface of the
produce or if it actually penetrates inside the fruit. Additionally, performing these type
of experiments in more than just one type of fresh produce could also provide
additional information as to whether those mechanisms of inactivation would differ
from produce to produce considering that some of them have softer, open cell
tissues (e.g. strawberries) while others are protected by a thicker skin (e.g
tomatoes).

135

APPENDICES

136

APPENDIX 1: Antimicrobial effectiveness of ClO2 to inactivate microorganisms in fresh produce

Table 12. Summary of antimicrobial effectiveness of ClO2 gas to inactivate microorganisms in fresh produce
MICROORGANISMS
Escherichia Coli
Escherichia Coli, Listeria monocytogenes, and
Salmonella
Escherichia Coli
Perservation Quality
Botrytis cinerea, Penicillium, alternaria
Escherichia Coli, Listeria monocytogenes, and
Salmonella
Listeria monocytogenes, Pseudomas aeruginosa.
Salmonella Typhimurim, Yersinia enterocolitica
Lactabacillus buchneri, Leuconostoc mesenteroides,
Eurotium spp.,
Penicilluim spp, Candida spp, and Saccharomyces
cerevisiae
Salmonella enterica spp
Escherichia Coli and Listeria monocytogenes
Escherichia Coli
Escherichia Coli
Listeria monocytogenes
Salmonella baildon
Escherichia Coli
Escherichia Coli, Listeria monocytogenes, and
Salmonella
Perservation Quality

PRODUCE
Green Peppers

AUTHOR
Y. Han, D.M. Sherman

Lettuce Leaves
Green Peppers
Green Bell Peppers
Grape

Sun-Young Lee
Y. Han, R.H. Linton
Du Jin-Hua
Ri ya Jin

Blueberries

Iuliano Popa

Blueberries

Vivian C.H. Wu

Juice
Roma tomatoes
Green Peppers
Iceberg Lettuce
Salad Vegetables
Lettuce
Shredded lettuce and diced tomatoes
Lettuce

Y Han, A.M. Guentert
V. Trinetta
Y. Han, R.H. Linton
C Diaz
U.M. Abdul-Raouf
Larry R. Beuchat
W.R. Weissinger
Larry R. Beuchat

Blueberries
Strawberries

Iuliano Popa
Mehment Seckin Aday

137

Table 12 (CONT’D)

Salmonella

Listeria Monocytogenes
Escherichia Coli
Salmonella enterica and Erwinia carotovora
Escherichia Coli
Escherichia Coli
Listeria Monocytogenes
Escherichia Coli
Listeria Monocytogenes
Escherichia Coli, Listeria monocytogenes, and
Salmonella
Escherichia Coli, Listeria monocytogenes, and
Salmonella
Escherichia Coli
Escherichia Coli, Listeria monocytogenes, and
Salmonella
Escherichia Coli
Listeria Monocytogenes
Escherichia Coli
Escherichia Coli

Salmonella, Yeast and Molds
Escherichia Coli and Listeria monocytogenes

Bell pepper, cucumber,
and Strawberry
Apples, lettuce,
Strawberries, and
cantaloupe
Apples
Tomatoes
Green Peppers
Apples
Apples
Green Peppers
Green Peppers
Lettuce Leaves
Cabage, carrot, and
iceberg lettuce
Lettuce Leaf
Raw Apples, Tomatoes,
and Lettuce
Lettuce Leaves
Iceberg Lettuce and
Romaine Lettuce
Tomato Fruits and
Lettuce Leaves
Apples
Blueberries,
Strawberries, and
Raspberries
Strawberries

138

Hyun-Gyun Yuk

Stephanie L. Rodgers
G.M. Sapers
S. Pao
Y. Han, J.D. Floros
J. Du
Jinhua Du
Y. Han, J.D. Floros
Y. Han, R.H. Linton
Sun-Young Lee
Kaye V. Sy
K.H. Seo
L.R. Beuchat
Lidsey A. Keskinen
Larry R. Beuchat
Suwimon Keeratipibul
G.M. Sapers

Kaye V. Sy
Y. Han

APPENDIX 2: Amperometric titration for determination of ClO2 and ClO2in solution

This procedure was outlined by Netramai (2010), it is a modification of the
original amperometric titration method for water and wastewater, 4500- ClO2 C
(Greenberg, Clesceri et al. 1992). The method is used to determine the residual ClO2
and ClO2 in a solution by dividing the sample in three parts in order to determine each
component separately.

Materials
•
•
•

60 mL sample cup for titrator
pH meter
Titrator equipped with electrode – Dual Ring PT (Mettler Tolledo, Columbus, OH)

Reagents

•
•
•
•
•
•

Deionized water
Phenylarsine oxide - PAO ((C6H5)AsO) – 0.00564N
Phosphate buffer solution (pH 7)
Potassium Iodide (KI) solution – 5% (weight)
Sodium Hydroxide (NaOH) – 6N, 0.02N and 0.002N
Sulfuric Acid (H2SO4) – 6N, 0.3N and 0.003N

Procedure

139

•
•

•

Sample 3 portions, 50 mL, of the sample solution (washed water) and place in
the titrator cup.
Portion 1 - determination of free available chlorine and chloramines
o Adjust the pH to ≥ 12 adding NaOH,
o Leave in dark for 10 min
o Correct the pH to 7 by adding H2SO4
o Add 1 mL of KI solution
o Titrate with PAO until the end point
o Record results as A
Portion 2 – determination of free available chlorine, chloramines, and 1/5 of
chlorine dioxide
o
o
o
o

Adjust the pH to 7 by adding phosphate buffer solution
Add 1 mL of KI solution
Titrate with PAO until the end point
Record results as B

•

Portion 3 – determination of free available chlorine, chloramines, chlorine dioxide
and chlorite
o Add 1 mL of KI solution
o Adjust the pH to ≤ 2 with H2SO4
o Leave in dark for 10 min
o Correct pH to 7 by adding NaOH
o Titrate with PAO until the end point
o Record result as C

•

Discard all solutions

Calculations
To calculate ClO2 in mg ClO2/L
ClO2 (mg/L) = 1.9(B - A)
To calculate ClO2- in mg Cl2/L
ClO2- (mgCl2/L) = 4A - 5B + C

140

APPENDIX 3: Determination of ClO2 solution using Thiosulfate Titration

Below is described the procedure to determine ClO2 concentration in solution
modified from ICA TriNova LLC (Staschower, 2012). The method described is for an
automatic titration.
Materials
•
•
•

60 mL sample cup for titrator
pH meter
Titrator with electrode – DMI 140 (Mettler Tolledo, Columbus, OH)
Solutions

•

Potassium iodide (KI) - 10% (weight)

•

Sodium Thiosulfate (Na2S2O3) – Certified Solution – 0.1N, 0.01N, and 0.001N

•

Sulfuric Acid (H2SO4) – 6N

Procedure

•

Sample a known amount of ClO2 solution and place in the sample cup (record as
Vs (mL))
Add KI solution if the volume is low

•

Titrate the solution (by hand) with Na2S2O3 until it is colorless

•
•

Correct the pH to 7 by adding H2SO4
Leave the solution in dark for 10 min

•
•
•

Titrate with Na2S2O3 until the end point (automatic titration)
Record result as Va
Discard the solution

•

141

Calculation

ܸ஺ ൈ ܰ ൈ 67500
݉݃
‫ܱ݈ܥ‬ଶ ቆ ൗ‫ܮ‬ቇ ൌ
4ൈܸ
௦
N= normality of Na2S2O

142

APPENDIX 4: SAS CODES
EXPERIMENT 1 – REPEATED MEASURES
** Importing dataset;
PROC IMPORT OUT= WORK.bot
DATAFILE= "C:\Users\Nico\Desktop\Statistical Analysis\trial1.csv"
DBMS=CSV REPLACE;
GETNAMES=YES;
DATAROW=2;
RUN;
data bot; set bot;
date=day; run;
proc print data=bot; run;
* checking samples;
proc freq data=bot;
tables day*trt;
run;
* extracting means per chamber;
proc means data=bot mean;
var rot;
class trt rep day;
ods output summary=bot2;
run;
proc print data=bot2; run;
data bot2; set bot2;
if trt=1 AND rep=4 AND day=7 then delete;
run;

** BASIC GLMM;
* cheking the basic structure in a LINEAR model;
proc mixed data=bot2;
class trt day rep;
model rot_Mean=trt|day/outp=jresiduals;
random rep*trt;
repeated day/subject=rep*trt type=cs;
run;
** Checking stat. assumptions;
*normality;
proc univariate data=jresiduals normal plot;
var resid ; histogram resid; qqplot resid; run;
* Variances;
proc sort data=jresiduals; by trt; run;

143

proc univariate data=jresiduals normal plot;
var resid; by trt; run;
proc sort data=jresiduals; by day; run;
proc univariate data=jresiduals normal plot;
var resid; by day; run;

** Adjusting a repeated structure;
* Cheking the structure in the VAR/COV matrix;
proc mixed data=bot2;
class trt day rep;
model rot_Mean=trt|day/ddfm=sat;
*random rep*trt;
repeated day/ subject=rep*trt type=cs;
run;
* -2 Res Log Likelihood 407.5
AICC (smaller is better) 411.7 ;

proc mixed data=bot2;
class trt day rep;
model rot_Mean=trt|day/ddfm=sat;
random rep*trt;
repeated day/subject=rep*trt type=ar(1);
run;
* -2 Res Log Likelihood 392.5
AICC (smaller is better) 396.7 ;
proc mixed data=bot2;
class trt day rep;
model rot_Mean=trt|day/ddfm=sat;
*random rep*trt;
repeated day/ subject=rep*trt type=un;
run;
* -2 Res Log Likelihood 381.6
AICC (smaller is better) 406.5;
proc mixed data=bot;
class trt day rep;
model rot_Mean=trt|day/ddfm=sat;
*random rep*temp*conc;
repeated day/subject=rep*trt type=sp(pow)(day);
run;
*-2 Res Log Likelihood 392.5
AICC (smaller is better) 396.7 ;

144

** Final model and reporting-listo;
proc mixed data=bot2;
class trt day rep;
model rot_Mean=trt|day/ddfm=sat;
random rep*trt;
repeated day/subject=rep*trt type=ar(1);
lsmeans trt*day/pdiff;
ods output lsmeans=jmeans diffs=jdiff;
run;
proc print data=jmeans; run;

*reporting:plotting;
symbol1 v=dot c='black' i=join;
symbol2 v=dot c='red' i=join;
symbol3 v=dot c='green' i=join;
symbol4 v=dot c='blue' i=join;
symbol5 v=dot c='orange' i=join;
symbol6 v=dot c='purple' i=join;
symbol7 v=dot c='brown' i=join;
symbol8 v=dot c='yellow' i=join;
symbol9 v=dot c='pink' i=join;
symbol10 v=dot c='ochre' i=join;
proc gplot data=jmeans;
plot estimate*day=trt;
run;
proc print data=jdiff;
where day=_day AND trt=trt AND probt<0.05; run;
proc glimmix data=bot2;
class trt rep day;
model rot_Mean=trt|day/ddfm=sat;
random day/subject=rep*trt type=ar(1);
lsmeans trt*day/slicediff=(trt day);
lsmeans trt*day/plot=meanplot(sliceby=trt join cl);;
run;

145

EXPERIMENT 1 - GOMPERTZ MODEL
** 1. Importing dataset;
PROC IMPORT OUT= WORK.bot
DATAFILE= "C:\Users\Nico\Desktop\Statistical Analysis\trial1.csv"
DBMS=CSV REPLACE;
GETNAMES=YES;
DATAROW=2;
RUN;
data bot; set bot;
date=day; run;
proc print data=bot; run;
** Plotting;
symbol1 v=dot c='black'; symbol2 v=dot c='red';
proc gplot data=bot;
where conc=0; plot rot*day=temp;
run;
proc gplot data=bot;
where conc=0.63; plot rot*day=temp;
run;
** 2. Fitting a Non-Linear function using proc nlin;
****************************************************;
* 2.1. Using Gompertz function;
proc sort data=bot; by trt; run;
proc nlin data=bot best=10 MAXITER=500 converge=0.05;
*where temp=4;
by trt;
parms alpha=60 to 100 by 0.1 beta=0 to 1 by 0.1 theta=0 to 5 by 0.1;
pa=alpha; pb=beta; pc=theta;
model rot = pa*2.71828**(-2.71828**(-pb*(day-pc)));
ods output ParameterEstimates=estimate1 ANOVA=ANOVA1 corrb=corrb1;
output out=mypred1 predicted=pred;
run;
proc print data=mypred1; run;
proc print data=estimate1; run;
proc print data=ANOVA1; run;
proc print data=corrb1; run;
* 2.2. Extracting estimated parameters;
proc sql;
select std(estimate) as sta into :sta from estimate1 where parameter='alpha';
select std(estimate) as stb into :stb from estimate1 where parameter='beta';
select std(estimate) as stc into :stc from estimate1 where parameter='theta';
select mean(alpha) as corrab into :corrab from corrb1 where parameter='beta';
select mean(alpha) as corrac into :corrac from corrb1 where parameter='theta';

146

select mean(beta) as corrbc into :corrbc from corrb1 where parameter='theta';
select mean(ms) as sse into :sse from anova1 where source='Error';
quit;
* 2.3. Extracting predicted values - Gompertz function;
proc nlmixed data=bot MAXITER=500;
where temp=4;
by conc;
bounds 0.5 < alpha < 1.5; bounds beta < 0; bounds theta < 0;
parms alpha=0.80 beta=-600 theta=-0.35;
pa=alpha; pb=beta; pc=theta;
pred= pa*2.71828**(-2.71828**(-pb*(day-pc)));
model rot~normal(pred,s2);
predict pred out=pred1;
run;
proc nlmixed data=bot MAXITER=500;
where temp=23;
by conc;
bounds 0.5 < alpha < 1.5; bounds beta < 0; bounds theta < 0;
parms alpha=0.80 beta=-20 theta=-0.35;
pa=alpha; pb=beta; pc=theta;
pred= pa*2.71828**(-2.71828**(-pb*(day-pc)));
model rot~normal(pred,s2);
predict pred out=pred2;
run;
* 2.4. plotting an overlayed graph;
*extracting data means;
data preds; set pred1 pred2;
if temp=4 AND conc=0 then trt='Tem4-Conc0';
if temp=4 AND conc=0.63 then trt='Tem4-C0.63';
if temp=23 AND conc=0 then trt='T23-Conc0';
if temp=23 AND conc=0.63 then trt='T23-C0.63';
run;
proc print data=preds; run;
proc sort data=preds; by trt; by day; run;
proc means data=preds nway chartype;
class trt day; var rot pred;
output out=mms1; run;
proc print data=mms1; run;
data means1; set mms1;
where _stat_='MEAN';
drop _TYPE_ _FREQ_ _STAT_;
run;
proc print data=means1; run;
*plotting;

147

symbol1 color=red interpol=none value=dot;
symbol2 color=black interpol=none value=dot;
symbol3 color=blue interpol=none value=dot;
symbol4 color=gray interpol=none value=dot;
symbol5 color=red width=4 interpol=spline2s value=none;
symbol6 color=black width=4 interpol=spline2s value=none;
symbol7 color=blue width=4 interpol=spline2s value=none;
symbol8 color=gray width=4 interpol=spline2s value=none;
axis1 label=('Percentage') order=(0 to 1 by 0.1) width=3;
axis2 label=('Model') order=(0 to 1 by 0.1) width=3;
legend1 label=('Mean Percentage');
legend2 label=('Model fitted');
proc sort data=means1; by trt; run;
proc gplot data=means1;
plot rot*day=trt/ vaxis=axis1 legend=legend1;
plot2 pred*day=trt/ vaxis=axis2 legend=legend2;
run;

148

EXPERIMENT 2 – REPEATED MEASURES
** Importing dataset;
PROC IMPORT OUT= WORK.bot
DATAFILE= "C:\SCC\13-03-27-Juliana Arango\trial2.csv"
DBMS=CSV REPLACE;
GETNAMES=YES;
DATAROW=2;
RUN;
data bot; set bot;
date=day; run;
proc print data=bot; run;
* checking samples;
proc freq data=bot;
tables day*trt;
run;
* extracting means per chamber;
proc means data=bot mean;
var rot;
class trt rep day;
ods output summary=bot2;
run;
proc print data=bot2; run;
data bot2; set bot2;
if trt=1 AND rep=4 AND day=7 then delete;
run;

** BASIC GLMM;
* cheking the basic structure in a LINEAR model;
proc mixed data=bot2;
class trt day rep;
model rot_Mean=trt|day/outp=jresiduals;
random rep*trt;
repeated day/subject=rep*trt type=cs;
run;
** Checking stat. assumptions;
*normality;
proc univariate data=jresiduals normal plot;
var resid ; histogram resid; qqplot resid; run;
* Variances;
proc sort data=jresiduals; by trt; run;
proc univariate data=jresiduals normal plot;
var resid; by trt; run;
proc sort data=jresiduals; by day; run;

149

proc univariate data=jresiduals normal plot;
var resid; by day; run;

** Adjusting a repeated structure;
* Cheking the structure in the VAR/COV matrix;
proc mixed data=bot2;
class trt day rep;
model rot_Mean=trt|day/ddfm=sat;
*random rep*trt;
repeated day/ subject=rep*trt type=cs;
run;
* -2 Res Log Likelihood
895.1
* BIC (smaller is better)
901.4;
proc mixed data=bot2;
class trt day rep;
model rot_Mean=trt|day/ddfm=sat;
random rep*trt;
repeated day/subject=rep*trt type=ar(1);
run;
* AIC (smaller is better)
844.6
* BIC (smaller is better)
847.0 ;
*tarea;
proc mixed data=bot2;
class trt day rep;
model rot_Mean=trt|day/ddfm=sat;
*random rep*trt;
repeated day/ subject=rep*trt type=un;
run;
proc mixed data=bot;
class temp conc day rep date;
model rot=temp|conc|day/ddfm=sat;
*random rep*temp*conc;
repeated day/subject=rep*temp*conc type=sp(pow) (date) group=conc*temp;
run;

** Final model and reporting-listo;
proc mixed data=bot2;
class trt day rep;
model rot_Mean=trt|day/ddfm=sat;
random rep*trt;
repeated day/subject=rep*trt type=ar(1);

150

lsmeans trt*day/pdiff;
ods output lsmeans=jmeans diffs=jdiff;
run;
proc print data=jmeans; run;

*reporting:plotting;
symbol1 v=dot c='black' i=join;
symbol2 v=dot c='red' i=join;
symbol3 v=dot c='red' i=join;
symbol4 v=dot c='red' i=join;
symbol5 v=dot c='red' i=join;
symbol6 v=dot c='red' i=join;
proc gplot data=jmeans;
plot estimate*day=trt;
run;
proc print data=jdiff;
where day=_day AND trt=trt AND probt<0.05; run;

151

EXPERIMENT 2 - GOMPERTZ MODEL
** 1. Importing dataset;
PROC IMPORT OUT= WORK.bot
DATAFILE= "C:\Users\Nico\Desktop\Statistical Analysis\jdlast\trial2.csv"
DBMS=CSV REPLACE;
GETNAMES=YES;
DATAROW=2;
RUN;
data bot; set bot;
date=day; run;
proc print data=bot; run;
** Plotting;
symbol1 v=dot c='black'; symbol2 v=dot c='red';
proc gplot data=bot;
where conc=0; plot rot*day=temp;
run;
proc gplot data=bot;
where conc=0.63; plot rot*day=temp;
run;
** 2. Fitting a Non-Linear function using proc nlin;
****************************************************;
* 2.1. Using Gompertz function;
proc sort data=bot; by conc; run;
proc nlin data=bot best=10 MAXITER=500 converge=0.05;
*where temp=4;
by conc;
parms alpha=70 to 150 by 0.1 beta=0 to 1 by 0.1 theta=0 to 7 by 0.1;
pa=alpha; pb=beta; pc=theta;
model rot = pa*2.71828**(-2.71828**(-pb*(day-pc)));
ods output ParameterEstimates=estimate1 ANOVA=ANOVA1 corrb=corrb1;
output out=mypred1 predicted=pred;
run;
proc print data=mypred1; run;
proc print data=estimate1; run;
proc print data=ANOVA1; run;
proc print data=corrb1; run;
* 2.2. Extracting estimated parameters;
proc sql;
select std(estimate) as sta into :sta from estimate1 where parameter='alpha';
select std(estimate) as stb into :stb from estimate1 where parameter='beta';
select std(estimate) as stc into :stc from estimate1 where parameter='theta';
select mean(alpha) as corrab into :corrab from corrb1 where parameter='beta';

152

select mean(alpha) as corrac into :corrac from corrb1 where parameter='theta';
select mean(beta) as corrbc into :corrbc from corrb1 where parameter='theta';
select mean(ms) as sse into :sse from anova1 where source='Error';
quit;
* 2.3. Extracting predicted values - Gompertz function;
proc nlmixed data=bot MAXITER=500;
where temp=4;
by conc;
bounds 0.5 < alpha < 1.5; bounds beta < 0; bounds theta < 0;
parms alpha=0.80 beta=-600 theta=-0.35;
pa=alpha; pb=beta; pc=theta;
pred= pa*2.71828**(pb*2.71828**(pc*day));
model rot~normal(pred,s2);
predict pred out=pred1;
run;
proc nlmixed data=bot MAXITER=500;
where temp=23;
by conc;
bounds 0.5 < alpha < 1.5; bounds beta < 0; bounds theta < 0;
parms alpha=0.80 beta=-20 theta=-0.35;
pa=alpha; pb=beta; pc=theta;
pred= pa*2.71828**(pb*2.71828**(pc*day));
model rot~normal(pred,s2);
predict pred out=pred2;
run;
* 2.4. plotting an overlayed graph;
*extracting data means;
data preds; set pred1 pred2;
if temp=4 AND conc=0 then trt='Tem4-Conc0';
if temp=4 AND conc=0.63 then trt='Tem4-C0.63';
if temp=23 AND conc=0 then trt='T23-Conc0';
if temp=23 AND conc=0.63 then trt='T23-C0.63';
run;
proc print data=preds; run;
proc sort data=preds; by trt; by day; run;
proc means data=preds nway chartype;
class trt day; var rot pred;
output out=mms1; run;
proc print data=mms1; run;
data means1; set mms1;
where _stat_='MEAN';
drop _TYPE_ _FREQ_ _STAT_;
run;
proc print data=means1; run;

153

*plotting;
symbol1 color=red interpol=none value=dot;
symbol2 color=black interpol=none value=dot;
symbol3 color=blue interpol=none value=dot;
symbol4 color=gray interpol=none value=dot;
symbol5 color=red width=4 interpol=spline2s value=none;
symbol6 color=black width=4 interpol=spline2s value=none;
symbol7 color=blue width=4 interpol=spline2s value=none;
symbol8 color=gray width=4 interpol=spline2s value=none;
axis1 label=('Percentage') order=(0 to 1 by 0.1) width=3;
axis2 label=('Model') order=(0 to 1 by 0.1) width=3;
legend1 label=('Mean Percentage');
legend2 label=('Model fitted');
proc sort data=means1; by trt; run;
proc gplot data=means1;
plot rot*day=trt/ vaxis=axis1 legend=legend1;
plot2 pred*day=trt/ vaxis=axis2 legend=legend2;
run;

154

EXPERIMENT 3 – REPEATED MEASURES

** Importing dataset;
PROC IMPORT OUT= WORK.bot
DATAFILE= "C:\Users\Nico\Desktop\Statistical Analysis\julianadata.csv"
DBMS=CSV REPLACE;
GETNAMES=YES;
DATAROW=2;
RUN;
data bot; set bot;
date=day; run;
proc print data=bot; run;
*************OLD COMPUTATION****************************;
** Organizing values;
data bot2; set bot;
if s1=2 then s1=1; if s2=2 then s2=1;
if s3=2 then s3=1; if s4=2 then s4=1;
if s5=2 then s5=1; if s6=2 then s6=1;
if s7=2 then s7=1; if s8=2 then s8=1;
if s9=2 then s9=1;
sum=s1+s2+s3+s4+s5+s6+s7+s8+s9;
mean=total/9;
event=0;
if total>0 then event=1;
drop s1 s2 s3 s4 s5 s6 s7 s8 s9;
total=9;
run;
proc print data=bot; run;
*******************END**********************************;
** Plotting;
symbol1 v=dot c='black';
symbol2 v=dot c='red';
proc gplot data=bot;
where conc=0;
plot rot*day=temp;
run;
proc gplot data=bot;
where conc=0.63;
plot rot*day=temp;
run;

** BASIC GLMM;

155

* cheking the basic structure in a LINEAR model;
proc mixed data=bot;
class temp conc day rep;
model rot=temp|conc|day/outp=jresiduals;
random rep*temp*conc;
repeated day/ subject=rep*temp*conc type=cs;
run;
** Checking stat. assumptions;
*normality;
proc univariate data=jresiduals normal plot;
var resid ; histogram resid; qqplot resid; run;
* Variances;
proc sort data=jresiduals; by temp; run;
proc univariate data=jresiduals normal plot;
var resid; by temp; run;
proc sort data=jresiduals; by conc; run;
proc univariate data=jresiduals normal plot;
var resid; by conc; run;
proc sort data=jresiduals; by day; run;
proc univariate data=jresiduals normal plot;
var resid; by day; run;
* Checking unequal variances acros treatments;
proc mixed data=bot;
class temp conc day rep;
model rot=temp|conc|day/ddfm=sat;
*random rep*temp*conc;
repeated day/subject=rep*temp*conc type=cs;
run;
*AIC (smaller is better) -1298.8
*BIC (smaller is better) -1294.5;
proc mixed data=bot;
class temp conc day rep;
model rot=temp|conc|day/ddfm=sat;
*random rep*temp*conc;
repeated day/subject=rep*temp*conc type=cs group=temp;
run;
*AIC (smaller is better) -1310.6
*BIC (smaller is better) -1302.0 ;
proc mixed data=bot;
class temp conc day rep;
model rot=temp|conc|day/ddfm=sat;
*random rep*temp*conc;

156

repeated day/subject=rep*temp*conc type=cs group=conc;
run;
*AIC (smaller is better) -1301.9
*BIC (smaller is better) -1293.2 ;
proc mixed data=bot;
class temp conc day rep;
model rot=temp|conc|day/ddfm=sat;
*random rep*temp*conc;
repeated day/subject=rep*temp*conc type=cs group=conc*temp;
run;
*AIC (smaller is better) -1312.3
*BIC (smaller is better) -1295.0 ;

** Adjusting a repeated structure;
* Cheking the structure in the VAR/COV matrix;
proc mixed data=bot;
class temp conc day rep;
model rot=temp|conc|day/ddfm=sat;
*random rep*temp*conc;
repeated day/subject=rep*temp*conc type=cs group=conc*temp;
run;
*AIC (smaller is better) -1312.3
*BIC (smaller is better) -1295.0 ;
proc mixed data=bot;
class temp conc day rep;
model rot=temp|conc|day/ddfm=sat;
random rep*temp*conc;
repeated day/subject=rep*temp*conc type=ar(1) group=conc*temp;
run;
*AIC (smaller is better) -2568.2
*BIC (smaller is better) -2550.9 ;
proc mixed data=bot;
class temp conc day rep date;
model rot=temp|conc|day/ddfm=sat;
*random rep*temp*conc;
repeated day/subject=rep*temp*conc type=sp(pow) (date) group=conc*temp;
run;
*AIC (smaller is better) -2568.2
*BIC (smaller is better) -2550.9;
** Final model and reporting;
proc mixed data=bot;
class temp conc day rep;

157

model rot=temp|conc|day/ddfm=sat;
random rep*temp*conc;
repeated day/subject=rep*temp*conc type=ar(1) group=conc*temp;
lsmeans conc*day*temp/pdiff; * alpha=0.1;
ods output lsmeans=jmeans diffs=jdiff;
run;
proc print data=jmeans; run;

*reporting:plotting;
proc sort data=jmeans; by temp; run;
symbol1 v=dot c='black' i=join;
symbol2 v=dot c='red' i=join;
proc gplot data=jmeans;
by temp;
plot estimate*day=conc;
run;
proc sort data=jdiff; by temp; run;
proc print data=jdiff;
where day=_day AND temp=_temp AND probt<0.1; run;
** Final model and reporting;
proc mixed data=bot;
class temp conc day rep;
model rot=temp|conc|day/ddfm=sat;
random rep*temp*conc;
*repeated day/subject=rep*temp*conc type=ar(1) group=conc*temp;
estimate 'con vs conc 17-20' conc 1 -1 temp*conc 1 -1 0 0 conc*day 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0.25 0.25 0.25 0.25 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -0.25 -0.25 -0.25 -0.25 0 0 0 0
temp*conc*day 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.25 0.25 0.25 0.25 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -0.25 -0.25 -0.25 -0.25 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0/e;
estimate 'con vs conc 17-21' conc 1 -1 temp*conc 1 -1 0 0 conc*day 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0.2 0.2 0.2 0.2 0.2 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -0.2 -0.2 -0.2 -0.2 -0.2 0 0 0
temp*conc*day 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.2 0.2 0.2 0.2 0.2 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -0.2 -0.2 -0.2 -0.2 -0.2 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0/e;
run;

158

EXPERIMENT 3 - GOMPERTZ MODEL
** 1. Importing dataset;
PROC IMPORT OUT= WORK.bot
DATAFILE= "C:\Users\Nico\Desktop\Statistical
Analysis\jdlast\julianadata.csv"
DBMS=CSV REPLACE;
GETNAMES=YES;
DATAROW=2;
RUN;
data bot; set bot;
date=day; run;
proc print data=bot; run;
** Plotting;
symbol1 v=dot c='black'; symbol2 v=dot c='red';
proc gplot data=bot;
where conc=0; plot rot*day=temp;
run;
proc gplot data=bot;
where conc=0.63; plot rot*day=temp;
run;
** 2. Fitting a Non-Linear function using proc nlin;
****************************************************;
* 2.1. Using Gompertz function;
proc sort data=bot; by conc; run;
proc nlin data=bot best=10 MAXITER=500 converge=0.05;
where temp=4;
by conc;
parms alpha=0.5 to 1.5 by 0.05 beta=0 to 1 by 0.1 theta=0 to 13 by 0.1;
pa=alpha; pb=beta; pc=theta;
model rot = pa*2.71828**(-2.71828**(-pb*(day-pc)));
ods output ParameterEstimates=estimate1 ANOVA=ANOVA1 corrb=corrb1;
output out=mypred1 predicted=pred;
run;
proc print data=mypred1; run;
proc print data=estimate1; run;
proc print data=ANOVA1; run;
proc print data=corrb1; run;
* 2.2. Extracting estimated parameters;
proc sql;
select std(estimate) as sta into :sta from estimate1 where parameter='alpha';
select std(estimate) as stb into :stb from estimate1 where parameter='beta';
select std(estimate) as stc into :stc from estimate1 where parameter='theta';

159

select mean(alpha) as corrab into :corrab from corrb1 where parameter='beta';
select mean(alpha) as corrac into :corrac from corrb1 where parameter='theta';
select mean(beta) as corrbc into :corrbc from corrb1 where parameter='theta';
select mean(ms) as sse into :sse from anova1 where source='Error';
quit;
* 2.3. Extracting predicted values - Gompertz function;
proc nlmixed data=bot MAXITER=500;
where temp=4;
by conc;
bounds 0.5 < alpha < 1.5; bounds beta < 0; bounds theta < 0;
parms alpha=0.80 beta=-600 theta=-0.35;
pa=alpha; pb=beta; pc=theta;
pred= pa*2.71828**(-2.71828**(-pb*(day-pc)));
model rot~normal(pred,s2);
predict pred out=pred1;
run;
proc nlmixed data=bot MAXITER=500;
where temp=23;
by conc;
bounds 0.5 < alpha < 1.5; bounds beta < 0; bounds theta < 0;
parms alpha=0.80 beta=-20 theta=-0.35;
pa=alpha; pb=beta; pc=theta;
pred= pa*2.71828**(-2.71828**(-pb*(day-pc)));
model rot~normal(pred,s2);
predict pred out=pred2;
run;
* 2.4. plotting an overlayed graph;
*extracting data means;
data preds; set pred1 pred2;
if temp=4 AND conc=0 then trt='Tem4-Conc0';
if temp=4 AND conc=0.63 then trt='Tem4-C0.63';
if temp=23 AND conc=0 then trt='T23-Conc0';
if temp=23 AND conc=0.63 then trt='T23-C0.63';
run;
proc print data=preds; run;
proc sort data=preds; by trt; by day; run;
proc means data=preds nway chartype;
class trt day; var rot pred;
output out=mms1; run;
proc print data=mms1; run;
data means1; set mms1;
where _stat_='MEAN';
drop _TYPE_ _FREQ_ _STAT_;
run;

160

proc print data=means1; run;
*plotting;
symbol1 color=red interpol=none value=dot;
symbol2 color=black interpol=none value=dot;
symbol3 color=blue interpol=none value=dot;
symbol4 color=gray interpol=none value=dot;
symbol5 color=red width=4 interpol=spline2s value=none;
symbol6 color=black width=4 interpol=spline2s value=none;
symbol7 color=blue width=4 interpol=spline2s value=none;
symbol8 color=gray width=4 interpol=spline2s value=none;
axis1 label=('Percentage') order=(0 to 1 by 0.1) width=3;
axis2 label=('Model') order=(0 to 1 by 0.1) width=3;
legend1 label=('Mean Percentage');
legend2 label=('Model fitted');
proc sort data=means1; by trt; run;
proc gplot data=means1;
plot rot*day=trt/ vaxis=axis1 legend=legend1;
plot2 pred*day=trt/ vaxis=axis2 legend=legend2;
run;

161

BIBLIOGRAPHY

162

BIBLIOGRAPHY

Abdel-Rahman, M. S., D. Couri, et al. (1979). "Kinetics of Cl02 and effects of
Cl02, Cl02-, and Cl03- in drinking water on blood glutathione and hemolysis in rat and
chicken." Journal of environmental pathology and toxicology 3(1-2): 431-449.
Abdul-Raouf, U. M., L. R. Beuchat, et al. (1993). "Survival and growth of
Escherichia coli O157:H7 on salad vegetables." Appl. Environ. Microbiol. 59(7): 19992006.
Aday, M. S., M. B. Buyukcan, et al. (2012). "MAINTAINING THE QUALITY OF
STRAWBERRIES BY COMBINED EFFECT OF AQUEOUS CHLORINE DIOXIDE
WITH MODIFIED ATMOSPHERE PACKAGING." Journal of Food Processing and
Preservation: no-no.
Agency for Toxic Substances and Disease Registry Division of Toxicology and
Envirommental Medicine. (2004). "Toxicological Profile for Chlorine Dioxide and
Chlorite." Retrieved June 1st, 2013, from
http://www.atsdr.cdc.gov/toxprofiles/tp160.pdf.
Aieta, E. M. and J. M. Berg (1986). "A Review of Chlorine Dioxide in Drinking
Water Treatment." J. Am. Water Works Assoc. 78 (6): 62-72.
Aieta, E. M., P. V. Roberts, et al. (1984). "Determination of Chlorine Dioxide,
Chlorine, Chlorite, and Chlorate in Water." J. Am. Water Works Assoc. 76: 64-70.
Akin, E. W., J. C. Hoff, et al. (1982). "Waterborne outbreak control: which
disinfectant?" Environmental Health Perspectives 46: 7-12.
Allende, A., F. A. Tomás-Barberán, et al. (2006). "Minimal processing for healthy
traditional foods." Trends in Food Science & Technology 17(9): 513-519.
Appendini, P. and J. H. Hotchkiss (2002). "Review of antimicrobial food
packaging." Innovative Food Science & Emerging Technologies 3(2): 113-126.
Arango, J., M. Rubino, et al. (2013). Quantification of Total Chlorine Dioxide
Absorption by UV-visible Spectroscopy. Michigan, US, Michigan State University.
Balla, C. S. and J. Farkas (2007). Minimally Processed Fruits and Fruit Products
and their Microbiological Safety. Handbook of Fruits and Fruit Processing, Blackwell
Publishing: 115-128.

163

Bean, N. H. and P. M. Griffin (1990). "Foodborne disease outbreaks in the United
States, 1973-1987: pathogens, vehicles, and trends." Journal of Food Protection 53.
Benarde, M. A., B. M. Israel, et al. (1965). "Efficiency of chlorine dioxide as a
bactericide." Applied Microbiology 13(5): 776-780.
Benarde, M. A., W. B. Snow, et al. (1967). "Chlorine Dioxide Disinfection
Temperature Effects." Journal of Applied Microbiology 30(1): 159-167.
Beuchat, L. (1998). "Surface decontamination of fruits and vegetables eaten raw:
a review."
Beuchat, L. R. (1996). "Pathogenic Microorganisms Associated with Fresh
Produce." Journal of Food Protection 59(2): 204-216.
Beuchat, L. R. (1999). "Survival of Enterohemorrhagic Escherichia coli O157:H7
in Bovine Feces Applied to Lettuce and the Effectiveness of Chlorinated Water as a
Disinfectant." Journal of Food Protection 62: 845-849.
Beuchat, L. R. (2002). "Ecological factors influencing survival and growth of
human pathogens on raw fruits and vegetables." Microbes and Infection 4(4): 413-423.
Bhagat, A. R. (2010). Modeling Critical Factors to Optimize the Treatment of
Selected Fruits and Vegetables with Chlorine Dioxide Gas Using a Miniaturized
Insustrial-Size Tunnel System. Doctor of Philosophy, Purdue University.
Bhaskara Reddy, M. V., K. Belkacemi, et al. (2000). "Effect of pre-harvest
chitosan sprays on post-harvest infection by Botrytis cinerea and quality of strawberry
fruit." Postharvest Biology and Technology 20(1): 39-51.
Brackett, R. E. (1999). "Incidence, contributing factors, and control of bacterial
pathogens in produce." Postharvest Biology and Technology 15(3): 305-311.
Braun, P. and J. Sutton (1988). "Infection cycles and population dynamics of
Botrytis cinerea in strawberry leaves." Canadian Journal of Plant Pathology 10(2): 133141.
Centers for Disease Control and Prevention. (2011). "CDC Estimates of
Foodborne Illness in the United States." Retrieved 9 April, 2013, 2013, from
http://www.cdc.gov/foodborneburden/PDFs/FACTSHEET_A_FINDINGS_updated413.pdf.

164

Centers for Disease Control and Prevention. (2013). "Surveillance for Foodborne
Disease Outbreaks — United States, 2009–2010." Retrieved June 1st, 2013, from
http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6203a1.htm?s_cid=mm6203a1_w.
Condie, L. W. (1986). "Toxicological Problems Associated With Chlorine
Dioxide." Journal (American Water Works Association) 78(6): 73-78.
Costilow, R. N., M. A. Uebersax, et al. (1984). "Use of Chlorine Dioxide for
Controlling Microorganisms During the Handling and Storage of Fresh Cucumbers."
Journal of Food Science 49(2): 396-401.
Couri, D., M. S. Abdel-Rahman, et al. (1982). "Toxicological Effects of Chlorine
Dioxide, Chlorite and Chlorate." Environmental Health Perspectives 46: 13-17.
Czarneski, M. A. and P. Lorcheim (2005). "Isolator decontamination using
chlorine dioxide gas." Pharmaceutical technology: 124-133.
Dagnas, S. and J.-M. Membre (2013). "Predicting and Preventing Mold Spoilage
of Food Products." Journal of Food Protection 76(3): 538-551.
Dantigny, P., M. Bensoussan, et al. (2006). "Standardisation of methods for
assessing mould germination: A workshop report." International journal of food
microbiology 108(2): 286-291.
Dantigny, P., A. Guilmart, et al. (2005). "Basis of predictive mycology."
International journal of food microbiology 100(1-3): 187-196.
Dantigny, P., S. Marín, et al. (2007). "Mould germination: Data treatment and
modelling." International journal of food microbiology 114(1): 17-24.
Daş, E., G. C. Gürakan, et al. (2006). "Effect of controlled atmosphere storage,
modified atmosphere packaging and gaseous ozone treatment on the survival of
Salmonella Enteritidis on cherry tomatoes." Food Microbiology 23(5): 430-438.
Declerck, S., D. D'or, et al. (2001). "Modelling the sporulation dynamics of
arbuscular mycorrhizal fungi in monoxenic culture." Mycorrhiza 11(5): 225-230.
Du, J., Y. Han, et al. (2002). "Inactivation by chlorine dioxide gas (ClO2) of
Listeria monocytogenes spotted onto different apple surfaces." Food Microbiology 19(5):
481-490.
Du, J., Y. Han, et al. (2003). "Efficacy of chlorine dioxide gas in reducing
Escherichia coli O157:H7 on apple surfaces." Food Microbiology 20(5): 583-591.
165

El-Samahy, S., B. M. Youssef, et al. (2000). "Microbiological and chemical
properties of irradiated mango." Journal of food safety 20(3): 139-156.
Ellis, M., K. Cooksey, et al. (2006). "Quality of Fresh Chicken Breasts Using a
Combination of Modified Atmosphere Packaging and Chlorine Dioxide Sachets."
Journal of Food Protection 69: 1991-1996.
Exama, A., J. Arul, et al. (1993). "Suitability of Plastic Films for Modified
Atmosphere Packaging of Fruits and Vegetables." Journal of Food Science 58(6): 13651370.
Garcia, D., A. J. Ramos, et al. (2009). "Predicting mycotoxins in foods: A review."
Food Microbiology 26(8): 757-769.
Gibson, A. M., N. Bratchell, et al. (1987). "The effect of sodium chloride and
temperature on the rate and extent of growth of Clostridium botulinum type A in
pasteurized pork slurry." Journal of Applied Microbiology 62(6): 479-490.
Gómez-López, V. M., A. Rajkovic, et al. (2009). "Chlorine dioxide for minimally
processed produce preservation: a review." Trends in Food Science & Technology
20(1): 17-26.
Gougouli, M. and K. P. Koutsoumanis (2012). "Modeling germination of fungal
spores at constant and fluctuating temperature conditions." International journal of food
microbiology 152(3): 153-161.
Greenberg, A. E., L. S. Clesceri, et al. (1992). Standard methods for the
examination of water and wastewater. Washington, D.C., American Public Health
Association: 4-56 - 54.81.
Gustavsson, J., C. Cederberg, et al. (2011). Global Food Losses and Food
Waste. International Congress SAVE FOOD!, Dusseldorf, Germany.
Han, Y., J. D. Floros, et al. (2001). "Response Surface Modeling for the
Inactivation of Escherichia coli O157:H7 on Green Peppers (Capsicum annuum L.) by
Chlorine Dioxide Gas Treatments." Journal of Food Protection 64: 1128-1133.
Han, Y., A. M. Guentert, et al. (1999). "Efficacy of chlorine dioxide gas as a
sanitizer for tanks used for aseptic juice storage." Food Microbiology 16(1): 53-61.
Han, Y., R. H. Linton, et al. (2000). "Inactivation of Escherichia coli O157:H7 on
surface-uninjured and -injured green pepper (Capsicum annuum L.) by chlorine dioxide

166

gas as demonstrated by confocal laser scanning microscopy." Food Microbiology 17(6):
643-655.
Han, Y., R. H. Linton, et al. (2001). "Reduction of Listeria monocytogenes on
Green Peppers (Capsicum annuum L.) by Gaseous and Aqueous Chlorine Dioxide and
Water Washing and Its Growth at 7C." Journal of Food Protection 64: 1730-1738.
Han, Y., T. L. Selby, et al. (2004). "Decontamination of Strawberries Using Batch
and Continuous Chlorine Dioxide Gas Treatments." Journal of Food Protection 67:
2450-2455.
Han, Y., D. M. Sherman, et al. (2000). "The effects of washing and chlorine
dioxide gas on survival and attachment of Escherichia coli O157: H7 to green pepper
surfaces." Food Microbiology 17(5): 521-533.
Harris, L. J., J. N. Farber, et al. (2003). "Outbreaks Associated with Fresh
Produce: Incidence, Growth, and Survival of Pathogens in Fresh and Fresh-Cut
Produce." Comprehensive Reviews in Food Science and Food Safety 2: 78-141.
Hautman, D. P., D. J. Munch, et al. (1997). "US EPA Method 300.1,
Determination of Inorganic Anions in Drinking Water by Ion Chromatography,
EPA/600/R-98/118, NIST PB-98-169196INZ." (Revision 1.0).
Heffernan, W., C. Guion, et al. (1979). "Oxidative damage to the erythrocyte
induced by sodium chlorite, in vivo." Journal of environmental pathology and toxicology
2(6): 1487.
Heller, M. C. and G. A. Keoleian (2000). Life cycle-based sustainability indicators
for assessment of the US food system, Center for Sustainable Systems, University of
Michigan.
Jarvis, W. R. (1962). "The infection of strawberry and raspberry fruits by Botrytis
cinerea Fr." Annals of Applied Biology 50(3): 569-575.
Kaczur, J. J. and D. W. Cawlfield (2000). Chlorine Oxygen Acids and Salts,
Chlorous Acid, Chlorites, and Chlorine Dioxide. Kirk-Othmer Encyclopedia of Chemical
Technology. John Wiley and Son. New York. NY, John Wiley & Sons, Inc.: 1-26.
Kallio, H., M. Hakala, et al. (2000). "Sugars and acids of strawberry varieties."
European Food Research and Technology 212(1): 81-85.
Kantor, L. S., K. Lipton, et al. (1997). "Estimating and addressing America’s food
losses." Food Review 20(1): 2-12.

167

Keskinen, L. A. and B. A. Annous (2011). Chlorine Dioxide (Gas). Nonthermal
Processing Technologies for Food. H. Q. Zhang, G. V. Barbosa-Cánovas, V. M.
Balasubramaniamet al. Oxford, UK, Wiley-Blackwell: 359 - 365.
Keskinen, L. A., A. Burke, et al. (2009). "Efficacy of chlorine, acidic electrolyzed
water and aqueous chlorine dioxide solutions to decontaminate Escherichia coli
O157:H7 from lettuce leaves." International Journal of Food Microbiology 132(2-3): 134140.

Kim, J., M. R. Marshall, et al. (1999). "Determination of Chlorate and Chlorite and
Mutagenicity of Seafood Treated with Aqueous Chlorine Dioxide." Journal of Agricultural
and Food Chemistry 47(9): 3586-3591.
Kim, J. Y., H. J. Kim, et al. (2010). "The effects of aqueous chlorine dioxide or
fumaric acid treatment combined with UV-C on postharvest quality of ‘Maehyang’
strawberries." Postharvest Biology and Technology 56(3): 254-256.
Knapp, J. E. and D. L. Battisti (2001). Chlorine Dioxide. Disinfection, Sterilization,
and Preservation. S. S. Block. Philadelphia, Lippincott Williams & Wilkins: 215-227.
Kromminga, H., J. Orphal, et al. (2003). "New measurements of OClO absorption
cross-sections in the 325–435 nm region and their temperature dependence between
213 and 293 K." Journal of Photochemistry and Photobiology A: Chemistry 157(2–3):
149-160.
Lee, S.-Y., M. Costello, et al. (2004). "Efficacy of Chlorine Dioxide Gas as a
Sanitizer of Lettuce Leaves." Journal of Food Protection 67: 1371-1376.
Linton, R., Y. Han, et al. (2006). "Gas-/vapor-phase sanitation (decontamination)
treatments." Microbiology of fruits and vegetables: 401-435.
López, S., M. Prieto, et al. (2004). "Statistical evaluation of mathematical models
for microbial growth." International journal of food microbiology 96(3): 289-300.
Lynch, M., R. Tauxe, et al. (2009). "The growing burden of foodborne outbreaks
due to contaminated fresh produce: risks and opportunities." Epidemiology and infection
137(3): 307.
Maas, J. (1995). Compendium of Strawberry Diseases. APS Press, St. Paul, MN.

168

Mahmoud, B. S. M., A. R. Bhagat, et al. (2007). "Inactivation kinetics of
inoculated Escherichia coli O157:H7, Listeria monocytogenes and Salmonella enterica
on strawberries by chlorine dioxide gas." Food Microbiology 24(7–8): 736-744.
Mahovic, M., J. A. Bartz, et al. "Chlorine Dioxide Gas from an Aqueous Solution:
Reduction of Salmonella in Wounds on Tomato Fruit and Movement to Sinks in a
Treatment Chamber." Journal of Food Protection&#174; 72(5): 952-958.
Netramai, S. (2011). Utilization of chlorine dioxide gas in food packaging
application. Doctor of Philosophy Ph.D., Michigan State University.
Nieuwenhuijsen, M. J., M. B. Toledano, et al. (2000). "Uptake of chlorination
disinfection by-products; a review and a discussion of its implications for exposure
assessment in epidemiological studies." Journal of exposure analysis and
environmental epidemiology 10(6 Pt 1): 586-599.
O’Beirne, D. and G. Francis (2003). "Reducing the pathogen risk in MAPprepared produce." Novel food packaging techniques 12: 231-275.
Parish, M. E., L. R. Beuchat, et al. (2003). "Methods to Reduce/Eliminate
Pathogens from Fresh and Fresh-Cut Produce." Comprehensive Reviews in Food
Science and Food Safety 2: 161-173.
Peano, C., V. Girgenti, et al. (2012). Strawberry Production Chain: Application of
a LCA model Acta Horticulturae. (ISHS)
Pepich, B. V., T. A. Dattilio, et al. (2007). "An improved colorimetric method for
chlorine dioxide and chlorite ion in drinking water using lissamine green B and
horseradish peroxidase." Analytica Chimica Acta 596(1): 37-45.
Phillips, C. A. (1996). "Review: Modified Atmosphere Packaging and its effects
on the microbiological quality and safety of produce." International Journal of Food
Science & Technology 31(6): 463-479.
Plattner, K. and A. Perez. (2012). "United States Department of Agriculture
(USDA) Report: Fruit and Tree Nuts Outlook." from
http://www.ers.usda.gov/media/970458/fts354.pdf.
Powelson, R. (1960). "Initiation of strawberry fruit rot caused by Botrytis cinérea."
Phytopathology 50: 491-494.
Qingdong, X., Z. Guangming, et al. (2006). "Study on Subchronic Toxicity of
Chlorine Dioxide and By-Products in Water." Journal of Environmental Science and
169

Health, Part A: Toxic/Hazardous Substances and Environmental Engineering 41(7):
1347 - 1353.
Reina, L. D., H. P. Fleming, et al. "Microbiological Control of Cucumber
Hydrocooling Water with Chlorine Dioxide." Journal of Food Protection 58(5): 541-546.
Rico, D., A. B. Martín-Diana, et al. (2007). "Extending and measuring the quality
of fresh-cut fruit and vegetables: a review." Trends in Food Science & Technology
18(7): 373-386.
Roberts, R. G. and S. T. Reymond (1994). "Chlorine dioxide for reduction of
postharvest pathogen inoculum during handling of tree fruits." Applied and
environmental microbiology 60(8): 2864-2868.
Robinson, J. W., Skelly Frame, Frame II (2005). Undergraduate Instrumental
Analysis. New York.
Rodgers, S. L., J. N. Cash, et al. (2004). "A Comparison of Different Chemical
Sanitizers for Inactivating Escherichia coli O157:H7 and Listeria monocytogenes in
Solution and on Apples, Lettuce, Strawberries, and Cantaloupe." Journal of Food
Protection 67(4): 721-731.
Schilder, A. M. C., J. M. Gillett, et al. (2013). "Evaluation of fungicides for control
of Botrytis fruit rot and leaf spot in strawberry, 2012." Plant Disease Management
Reports.
Seo, K. H. and J. F. Frank (1999). "Attachment of Escherichia coli O157:H7 to
Lettuce Leaf Surface and Bacterial Viability in Response to Chlorine Treatment as
Demonstrated by Using Confocal Scanning Laser Microscopy." Journal of Food
Protection 62: 3-9.
Seymour, I. J. and H. Appleton (2001). "Foodborne viruses and fresh produce."
Journal of Applied Microbiology 91(5): 759-773.
Singh, N., R. K. Singh, et al. (2002). "Effect of inoculation and washing methods
on the efficacy of different sanitizers against Escherichia coli O157:H7 on lettuce." Food
Microbiology 19(2-3): 183-193.
Sloan, A. E. (2013). "Top 10 Food Trends." from
http://sloantrend.com/attachments/article/9/2013-04IFTTop10Trends.pdf.
Spotts, R. and B. Peters (1980). "Chlorine and chlorine dioxide for control of
d’Anjou pear decay." Plant Dis 64(1): 2.

170

Staschower, F. (2012). Residues from Chlorine Dioxide Gas Treatment,
Generated by Different Delivery Systems, on Fresh Produce. Master of Science,
Michigan State University.
Sumathi, S., R. F. Cindy, et al. (2004). "Fresh Produce: A Growing Cause of
Outbreaks of Foodborne Illness in the United States, 1973 through 1997." Journal of
Food Protection 67(10): 2342-2353.
Sy, K. V., K. H. McWatters, et al. (2005). "Efficacy of Gaseous Chlorine Dioxide
as a Sanitizer for Killing Salmonella, Yeasts, and Molds on Blueberries, Strawberries,
and Raspberries." Journal of Food Protection 68: 1165-1175.
Sy, K. V., M. B. Murray, et al. (2005). "Evaluation of Gaseous Chlorine Dioxide as
a Sanitizer for Killing Salmonella, Escherichia coli O157:H7, Listeria monocytogenes,
and Yeasts and Molds on Fresh and Fresh-Cut Produce." Journal of Food Protection
68: 1176-1187.
Trinetta, V., N. Vaidya, et al. (2011). "Evaluation of Chlorine Dioxide Gas
Residues on Selected Food Produce." Journal of Food Science 76(1): T11-T15.
Tsai, L. S., C. C. Huxsoll, et al. (2001). "Prevention of Potato Spoilage During
Storage by Chlorine Dioxide." Journal of Food Science 66(3): 472-477.
United States Department of Agriculture (USDA). (2003). "Pathogen modelling
program, version 7.0." Retrieved June 1st, 2013, from
http://ars.usda.gov/services/docs.htm?docid=6786.
United States Environmental Protection Agency (EPA) (1999). "Alternative
Disinfectants and Oxidants Guidance Manual: Chapter 4. Chlorine Dioxide, EPA 815-R99-014."
United States Environmental Protection Agency (EPA) (2006). Registration
eligibility decision for Chlorine Dioxide and Sodium Chlorite (Case Number 4023).
http://www.epa.gov/oppsrrd1/REDs/chlorine_dioxide_red.pdf.
United States Environmental Protection Agency (EPA). (2007). "Anthrax spore
decontamination using chlorine dioxide." Retrieved June 1st, 2013, from
http://www.epa.gov/pesticides/factsheets/chemicals/chlorinedioxidefactsheet.htm.
United States Food and Drug Administration (FDA). (2004). "Produce Safety
From Production to Consumption: 2004 Action Plan to Minimize Foodborne Illness
Associated with Fresh Produce Consumption." Retrieved June 1st, 2013, from

171

http://www.fda.gov/Food/FoodborneIllnessContaminants/BuyStoreServeSafeFood/ucm1
29487.htm.
United States Food and Drug Administration (FDA). (2011). "Agency Response
Letter GRAS Notice No. GRN 000062." Retrieved June 1st, 2013, from
http://www.fda.gov/Food/IngredientsPackagingLabeling/GRAS/NoticeInventory/ucm153
971.htm.
United States Food and Drug Administration (FDA). (2012). "Secondary direct
food additives permitted in food for human consumption. Chlorine dioxide. CFR Title 21,
part 173.300." Retrieved 9 Apr 2013, 2013, from
http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfCFR/CFRSearch.cfm?fr=173.300.
Vardar, C., K. Ilhan, et al. (2012). "The application of various disinfectants by
fogging for decreasing postharvest diseases of strawberry." Postharvest Biology and
Technology 66(0): 30-34.
Wahner, A., G. S. Tyndall, et al. (1987). "Absorption cross sections for symmetric
chlorine dioxide as a function of temperature in the wavelength range 240-480nm." The
Journal of Physical Chemistry 91(11): 2734-2738.
Wei, C. I., T. S. Huang, et al. (1995). "Growth and Survival of Salmonella
montevideo on Tomatoes and Disinfection with Chlorinated Water." Journal of Food
Protection 58(8): 829-836.
Williamson, B., B. Tudzynski, et al. (2007). "Botrytis cinerea: the cause of grey
mould disease." Molecular Plant Pathology 8(5): 561-580.
Windholz, M., S. Budavari, et al. (1976). The Merck index. An encyclopedia of
chemicals and drugs. Rahway, New Jersey, Merck & Co.
Wise, J. C., L. J. Gut, et al. (2012). Michigan Fruit Management Guide 2013.
Extension Bulletin E-154. East Lansing, MI, Michigan State University.
Yuk, H.-G., J. A. Bartz, et al. (2006). "The Effectiveness of Sanitizer Treatments
in Inactivation of Salmonella spp. from Bell Pepper, Cucumber, and Strawberry."
Journal of Food Science 71(3): M95-M99.

172