fi

151%?“

L: tr}...
1 .

1..
dink“?

ut..i.a!
{Wt-ii:
.451: 3 .: ...
é... Erufpimm
3.. .3
. hw‘hr‘ma.
hi I:
a... a
x

l:
7.3, it:
.‘ III";

94:11..
$23....
fink
av
«1152:12- 10 '
..111:Is!!sr tr:
12:1 $527: 7
1.1%.".
u...

. 1.
5.»:

‘ {all
l.‘ k... .. .
i, .3... .i 1“: ..h..l !. :3.

s .3. 13.3.“... . 12
...;:Eb:.¥$ ._ . ‘2‘...
:1: :23... 1.5.2.. .
01:11:!
.2:

. . .11:
i 13.5....
Av i 1 1‘3...
up" t1.‘ v“‘ll "I‘IK‘I ‘
\alsllvv:Ilr:\tl 41.9: I
‘o... n:\,\(u! f)...
twirling—Blilv ’0
i...is.l...,£! .{xttitlt
t.ll’g‘3V gt.lv’l
!. Ct .31.!“
3A I||||t 1

£31.13.
2:55.
grill... £29

 

a;
... A: :4

 

 

bl 2
E

1907

 

LIBRARY
i $3.55"?! State

v '

‘ University

 

 

 

This is to certify that the
dissertation entitled

QUALITY OF TART CHERRY NUTRACEUTICAL JUICE: A
COMPARISON OF JUICE PACKAGED IN FLEXIBLE POUCHES
AND BOTTLES MADE FROM GLASS, PET AND ALUMINUM

presented by

MARIA-PAZ GONZALEZ-MULET

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Packaging

 

 

 

Diana Twede
Major Professor’s Signature

May 5, 2009

MSU is an Affirmative Action/Equal Opportunity Employer

 

-.-.-- -V-l-O-Q-t-I-i-i-

 

PLACE IN RETURN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE DATE DUE

 

 

 

_ _. ..___ _.__ _.__.._-_—____. .4.

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KzlProj/AccaPreleIRCIDateDue.indd

 

 

QUALITY OF TART CHERRY NUTRACEUTICAL JUICE: A COMPARISON OF
JUICE PACKAGED IN FLEXIBLE POUCHES AND BOTTLES MADE FROM
GLASS, PET AND ALUMINUM
By

Maria-Paz Gonzalez-Mulet

A DISSERTATION
Submitted to
Michigan State University

In partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Packaging

2008

ABSTRACT
QUALITY OF TART CHERRY NUTRACEUTICAL JUICE: A COMPARISON OF
JUICE PACKAGED IN FLEXIBLE POUCHES AND BOTTLES MADE FROM
GLASS, PET AND ALUMINUM
By
Maria-Paz Gonzalez-Mulet

The main Objective of this research is to develop a protective, convenient
and consumer oriented product for the cherry industry: a pasteurized single-
strength tart cherry juice made from concentrate, which was hot filled in four
different commercially available packages with nitrogen flushed headspace:
composite stand-up pouches with a multilayer structure (Pet/Al/LDPE) and
bottles made from glass, aluminum and PET.

A more specific aim is to analyze the stability of anthocyanins and other
key properties over 12-months of storage under controlled ambient conditions
(23°C) and 50% RH, without light exposure. Gas liquid Chromatography-
mass spectrometry was used to measure the anthocyanin content over time.
Analyses were likewise conducted monthly for color, bacteria, yeast and mold,
solids and pH.

The results show that the tart Cherry juice can be pasteurized and packed
for 12 months in two different packages without suffering a dramatic anthocyanin
loss glass as expected and aluminum. Aluminum bottles performed nearly as
well as glass. With good gas and moisture barriers and good antioxidant
retention, aluminum bottles are becoming more widely available for processors,

and aluminum

Bottles are recloseable and easily recycled. One issue that did arise with the
trained sensory panelist was an Off-flavor that developed in the juice in aluminum
bottles after six months. The performance of the aluminum bottles was
dependent on the constitution of the inner lining which, if compromised, could
cause Chelating in the juice with the resulting off-flavor. The plastic bottles in the
test performed the least we" compared to glass and aluminum. The plastic
bottles showed a loss of 50 percentage of the antioxidants in the juice after six
months. Plastic is readily available to processors, but again, as generic
packages that are easily reclosed and are recyclable.

The performance of the stand up pouches was not evaluated through all
the 12 months since the packages failed during the fill stage of the project.

Results were favorable for the tart cherry juice industry. From the

consumers’ points of view results show a statistically significant preference for
the 12 month-stored juice packed in glass and aluminum bottles compared to

fresh juice.

DEDICATION

To my parents: Jorge and Mireille
To my husband: Luis
And my daughter Cloe
They have always believed in me.

AKNOWLEDGMENT

Special thanks to:
Dr. Diana Twede for accepting the Challenge,
Dr. Suzanne Thornsbury for her support,
Dr. Bruce Harte for his knowledge,
Dr. Janice Harte for the collaboration,
Dr. Soji Adeja for the inspiration,
Dr. Rafael Auras for the help,
Saint Gobain, Performance Packaging, ToyO Seikan and CCL Container for
providing the packages,
Food for Thoughts for packing the juice,
The Michigan Cherry Committee and the Cherry Marketing Institute for the
support and encouragement,
Sustainable Michigan Endowed Project (SMEP), Project Greeen, MSU Product
Center for Agriculture and Natural Resources and Michigan Agricultural
Experimentation Station (MAES) for the funding,

and to my family and friends who keep supporting me to reach my goals.

TABLE OF CONTENTS

TABLE OF CONTENTS ...................................................................................... vi
LIST OF FIGURES ............................................................................................... x
I. INTRODUCTION ........................................................................................... 1
ll. LITERATURE REVIEW ............................................................................... 13
2.1. Definition of juice ...................................................................................... 13
2.2. Tart cherry juice from concentrate ............................................................ 15
2.3. Functional foods ....................................................................................... 16
2.4. Anthocyanins ............................................................................................ 21
2.5. Shelf-life stability ...................................................................................... 28
2.5.1. Hot fill and product acidity .................................................................. 29
2.5.2. Color and anthocyanin stability .......................................................... 30
2.5.3. Flavor stability and food packaging interactions ................................ 33

2.6. Materials ................................................................................................... 37
2.6.1. Glass .................................................................................................. 38
2.6.2. Polyethylene terephthalate ................................................................ 41
2.6.3. Aluminum ........................................................................................... 44
2.6.4. Flexible packaging ............................................................................. 45

III. EXPERIMENTAL DESIGN AND METHODS ........................................... 48
3.1 Packages and Closures ............................................................................ 51
3.1.1. Glass bottles ...................................................................................... 51
3.1.2. Hot fill Polyethylene terephthalate (PET) bottles ................................ 52
3.1.3. Aluminum bottles ............................................................................... 54
3.1.4. Stand-up pouches .............................................................................. 56

3.2 Packaging technique and facilities ............................................................ 59
3.3. Tests ........................................................................................................ 61
3.3.1. Color .................................................................................................. 62
3.3.2. Solids ................................................................................................. 64
3.3.3. pH levels. ........................................................................................... 65
3.3.4. Microbial growth ................................................................................. 66
3.3.5. Antioxidants compounds .................................................................... 67
3.3.6. Sensory evaluation ............................................................................ 69
3.3.7. Statistical analysis ............................................................................. 72

IV. RESULTS AND DISCUSSION ................................................................ 74
4.1 Stand-up pouch test was compromised ................................................ 74
4.2. Changes in color ...................................................................................... 75
4.2.1. Changes in transparency (L* values) ................................................. 76
4.2.2. Changes in redness (a* values) ......................................................... 80
4.2.3. Changes in yellowness (b* values) .................................................... 83
4.2.4. Changes in chrominance ................................................................... 86

vi

4.2.5. Changes in hue .................................................................................. 89

4.3. Changes in Solids .................................................................................... 93
4.4. Changes in pH .......................................................................................... 96
4.5. Changes in microbial growth .................................................................... 98
4.6. Changes in anthocyanin compounds ....................................................... 99
4.6.1 Changes in cyanidin .......................................................................... 100
4.6.2. Changes in delphinidin ..................................................................... 104
4.6.3. Changes in peonidin 3-rutinoside .................................................... 108
4.6.4. Changes in pelargonidin 3-rutinoside .............................................. 109
4.6.5. Changes in petunidin 3-rutinoside ................................................... 111
4.6.6. Anthocyanin result summary ............................................................ 112

4.7. Trained panel results .............................................................................. 115
4.7.1. Color perception .............................................................................. 116
4.7.2. Aroma perception ............................................................................ 117
4.7.3. Mouth feeling perception .................................................................. 118
4.7.4. Sweetness perception ...................................................................... 120
4.7.5. Tartness perception ......................................................................... 121
4.7.6. Fresh cherry flavor perception ......................................................... 122
4.7.7. Off-flavor perception ........................................................................ 123
4.7.8. Overall flavor perception .................................................................. 124
4.7.9. Purchasing appropriateness perception ........................................... 125

4.8. Consumer preferences results ............................................................... 126
IV. CONCLUSIONS ..................................................................................... 128
V. FUTURE RESEARCH ............................................................................... 135
APPENDIX 1: ANOVA Tables (treatment analysis) ...................................... 136
APPENDIX 2: ANOVA tables (month analysis) ............................................. 143
APPENDIX 3: Consumer Consent Form and Trained panel survey ........... 152
APPENDIX 4: Stand-up pouch results .......................................................... 158
APPENDIX 5: Calibration curves ................................................................... 171
REFERENCES ................................................................................................. 173

vii

LIST OF TABLES

Table 1. Processed and fresh tart cherry output. ................................................. 4
Table 2. Price of processed tart cherry per pound in US. Dollars. ....................... 5
Table 3. Common juice designation. ................................................................... 14

Table 4. US. and Codex Alimentarius Brix values for juice from concentrate. .. 15
Table 5. Juice concentration systems. ................................................................ 16

Table 6. Estimated market size for functional foods for specific regions (US $

Millions) ............................................................................................................... 17
Table 7. Health claims for food packaging labeling approved by the FDA. ......... 18
Table 8. How do you perceive color? ................................................................ 117
Table 9. How do you perceive the aroma in the sample? ................................. 118
Table 10. How do perceive the mouth feel? ...................................................... 119
Table 11. How do you perceive the sweetness of the sample? ........................ 121
Table 12. How do you perceive tartness? ......................................................... 122
Table 13. How do you perceive the "fresh cherry flavor"? ................................ 123
Table 14. Do you perceive any off-flavor? ........................................................ 124
Table 15. How do you perceive the overall flavor? ........................................... 125

Table 16. Is this sample's quality appropriate for consumers to purchase the

dnnk? ................................................................................................................ 126
Table 17. Means of consumers’ preferences results ......................................... 127
Table 18. Ranking results ................................................................................. 127
Table 19. ANOVA full model (L* value) ............................................................. 137
Table 20. ANOVA fuII model analysis of a* values ............................................ 137
Table 21. ANOVA custom model analysis of a* values ..................................... 137

viii

Table 22. ANOVA full model analysis of b* values ........................................... 138

Table 23. ANOVA custom model analysis of chrominance. .............................. 138
Table 24. ANOVA custom model analysis of chrominance. .............................. 138
Table 25. ANOVA full model analysis of solids change. ................................... 139
Table 26. ANOVA custom model analysis of solids change. ............................ 139
Table 27. ANOVA full model analysis of pH ...................................................... 139
Table 28. ANOVA full model analysis for cyanidin 3-glucosylrutinoside changeao
Table 29. ANOVA full model analysis for cyanidin 3-rutinoside changes. ......... 140
Table 30. ANOVA full model analysis for delphinidin 3-glucosylrutinoside

changes. ........................................................................................................... 140

Table 31 .ANOVA full model analysis for delphinidin 3-rutinoside changes ....... 141
Table 32. ANOVA full model analysis for peonidin 3-rutinoside changes. ........ 141
Table 33. ANOVA full model analysis for pelargonidin 3-rutinoside changes. ..141

Table 34. ANOVA custom model analysis for pelargonidin 3- rutinoside changes.
.......................................................................................................................... 142

Table 35. ANOVA full model analysis for petunidin 3-rutinoside changes. ....... 142

Table 36. L* value month ANOVA analysis ...................................................... 144
Table 37. a* values month ANOVA analysis. ................................................... 144
Table 38. b values month ANOVA analysis. ..................................................... 145
Table 39. Chrominance month ANOVA analysis. ............................................. 145
Table 40. Hue values month ANOVA analysis .................................................. 146
Table 41. AE values month ANOVA analysis. .................................................. 146
Table 42. Change in solids month ANOVA analysis. ........................................ 147
Table 43. Change in pH values month ANOVA analysis. ................................. 147

Table 44. Changes cyanidin 3-glucosylrutinoside month ANOVA analysis ....... 148
Table 45. Changes cyanidin 3-rutinoside month ANOVA analysis. .................. 148

Table 46. Changes delphinidin 3-glucosylrutinoside month ANOVA analysis... 149

Table 47 Changes delphinidin 3-rutinoside month ANOVA analysis ................. 149
Table 48. Changes peonidin 3-rutinoside month ANOVA analysis. .................. 150
Table 49. Changes pelargonidin 3-rutinoside month ANOVA analysis. ........... 150
Table 50. Changes in petunidin 3-rutinoside ..................................................... 151

LIST OF FIGURES

Figure 1. Map of primary producing counties in Michigan, ................................... 1
Figure 2. Summary of steps needed when developing a new product ................. 8
Figure 3. Six common anthocyanins formed in plants, ........................................ 23

Figure 4. How anthocyanins can change from any stage to any stage with pH

change (cyanidin example), ................................................................................ 24
Figure 5. Free radical formation process. ........................................................... 27
Figure 6. Permeation, migration and absorption. ................................................ 35
Figure 7. Molecular arrangement of glass ........................................................... 39
Figure 8. Molecular structure of PET and condensation reaction. ...................... 43
Figure 9. Tart cherry juice shelf-life, conceptual model. ...................................... 49
Figure 10. Experimental model scheme .............................................................. 50
Figure 11. Glass bottle. ....................................................................................... 52
Figure 12. Multilayer PET bottle structure and oxygen scavenger process. ....... 53
Figure 13. The 8 oz PET bottle from Toyo Seikan. ............................................. 54
Figure 14. Aluminum bottle (dimensions in mm). ................................................ 56
Figure 15. Stand-up pouches multilayer structure ............................................... 57
Figure 16. The Stand-up pouch (dimensions in mm). ......................................... 58
Figure 17. Food for Thought, cookers. ................................................................ 59
Figure 18. Food For Thought, filler for bottles and jars. ...................................... 60
Figure 19. Food For Thought, cooling belt. ......................................................... 60
Figure 20. Sample preparation for color measurements. .................................... 62
Figure 21. Hunter Lab scale reference ................................................................ 63

xi

Figure 22. Refractometer used in the tests. ........................................................ 64

Figure 23. How a reading looks in the refractometer visor. ................................. 65
Figure 24. Digital pH meter and buffers for calibration. ....................................... 66
Figure 25. Counting plates with PDA medium and juice sample ......................... 67

Figure 26. Change in transparency (L* values) of tart cherry juice stored at room
temperature (23°C, 50%RH) for twelve months. ................................................. 78

Figure 27. Correlation between L* values and trained panel mouth feel of tart
cherry juice stored at room temperature (23°C, 50% RH) for twelve months ...... 80

Figure 28. Picture of the PET bottle delaminating with tart cherry juice stored at
room temperature (23°C, 50% Rh) after six months of storage. ......................... 82

Figure 29. Changes in a* values (redness) of tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage. ......................... 83

Figure 30. Changes in b* values (yellowness) of tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage. ......................... 85

Figure 31. Correlation between L*a*b* values of tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage. ......................... 86

Figure 32. Changes in chrominance of tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage. ......................... 88

Figure 33. Changes in hue of tart cherry juice stored at room temperature (23°C,
50% RH) during twelve months of storage .......................................................... 91

Figure 34. Color change (AE) of tart cherry juice stored at room temperature
(23°C, 50% RH) during twelve months of storage. ............................................. 93

Figure 35. Changes in solids of tart cherry juice stored at room temperature
(23°C, 50% RH) during twelve months of storage. ............................................. 95

Figure 36. pH changes of tart cherry juice stored at room temperature (23°C,
50% RH) during twelve months of storage .......................................................... 98

Figure 37. Changes in cyanidin 3-glucosylrutinoside in tart cherry juice stored at
room temperature (23°C, 50% RH) during twelve months of storage. .............. 101

Figure 38. Changes cyanidin 3-rutinoside in tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage ........................ 103

xii

Figure 39. Correlation between the two types of cyanidin in tart cherry juice
stored at room temperature (23°C, 50% RH) during twelve months of storage.104

Figure 40. Changes in delphinidin 3-glucosylrutinoside in tart cherry juice stored
at room temperature (23°C, 50% RH) during twelve months of storage. .......... 105

Figure 41. Changes in delphinidin 3-rutinoside in tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage. ....................... 107

Figure 42. Correlation between the two types of delphinidin in tart cherry juice
stored at room temperature (23°C, 50% RH) during twelve months of storage.108

Figure 43. Changes in peonidin 3-rutinoside in tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage. ....................... 109

Figure 44. Changes in pelargonidin 3-rutinoside in tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage. ....................... 111

Figure 45. Changes in petunidin 3-rutinoside in tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage. ....................... 112

Figure 46. Correlation between red and blue anthocyanins found in tart cherry
juice stored at room temperature (23°C, 50% RH) during twelve months of
storage. ............................................................................................................. 114

Figure 47. Accumulation of all anthocyanins in tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage. ....................... 115

Figure 48. Correlation between sweetness and tartness perceived by the trained
panel in tart cherry juice stored at room temperature (23°C, 50% RH) during

twelve months of storage. ................................................................................. 116
Figure 49. Path to commercialization ................................................................ 133
Figure 50. Changes in L* values. ...................................................................... 159
Figure 51. Changes in a* values. ...................................................................... 160
Figure 52. Changes in b* values. ...................................................................... 161
Figure 53 Changes in solids. ............................................................................ 162
Figure 54. Changes in pH. ................................................................................ 163
Figure 55. Changes in cyanidin 3-glucosylrutinoside. ....................................... 164

xiii

Figure 56.
Figure 57.
Figure 58.
Figure 59.
Figure 60.
Figure 61.
Figure 62.

Figure 63.

Changes in cyanidin 3-rutinoside. .................................................... 165

Changes in delphinidin 3-glucosylrutinoside. ................................... 166
Changes in delphinidin 3-rutinoside. ................................................ 167
Changes in peonidin 3-rutinoside. ................................................... 168
Changes in pelargonidin 3-rutinoside. ............................................. 169
Changes in petunidin 3-rutinoside. .................................................. 170
Step Calibration for cyanidin 3-glucosilrutinoside ............................ 172
Calibration curve for cyanidin 3-arabinoserutinoside ....................... 172

xiv

I. INTRODUCTION

Historically the United States (US) tart cherry supply was provided
primarily by growers located in areas bordering the Great Lakes. The Michigan
"fruit-belt” encompasses the state’s western coast where Lake Michigan creates
a micro-climate suitable for fruit production. Tart cherries are a very important
fruit crop to the Michigan economy. Currently, tart cherry production in Michigan
represents approximately seventy percent of the total tart cherry production of
the United States (Cherry Marketing Institute 2003). Primary producing counties
include Leelanau, Oceana, Grand Traverse, Antrim, and Mason, all located along
Michigan’s west coast (Figure 1). Tart cherry production decreased 30 percent
from 2005 to 2008 when the production went from 190 million tons in 2005 to 135

million tons in 2008 (USDA/NASS 2008).

 

  

 

1. Oceana

2. Mason

3. Grand Traverse
4. Leelanau

5. Antrim

 

 

 

L

Figure 1. Map of primary producing counties in Michigan,
Adapted from Geocities.com.

 

Among the primary producing states (Michigan, New York, Pennsylvania
and Wisconsin), Michigan has been the largest producer, and the price of
cherries is greatly determined by the Michigan output during each growing
season. For instance, prices go up in years when Michigan has a short crop,
while prices can go significantly down in years of bountiful harvests. The
economics behind this phenomenon are partially explained by the fact that
cherries only account for a small percentage of the overall cost of manufactured
food products that includes cherries as one of their many ingredients. Therefore,
buyers are willing to pay more per pound of cherries during short crop years as
this will not significantly affect the final cost of their products. On the other hand,
when output increases, prices fall because there is little demand for cherries
other than what is purchased for use in manufactured food products.
Furthermore, the knowledge of the economics during short crops following years
of large crops have created some reluctance in the manufacturing and retail
sector to invest in greater numbers of cherry-based products as shortages of
main ingredients can become a problem in satisfying the created demand for
new products (USDA 1996).

The industry’s economic behavior has led to government-issued Marketing
Orders. Marketing Orders can have different objectives in support of the industry.
Marketing orders exist in the US. for products such as milk, walnuts and certain
fruits and vegetables. These orders can be state-specific or federal. In the case

of the Tart Cherry Marketing Order, the organized cherry production and

 

 

marketing industry is federally authorized (USDA 2001) to harvest only a
percentage of the total production during large crops. The percentage harvested
is defined as utilized production, and it is calculated by the USDA each year. In
years when the estimated total harvest will be larger than the estimated utilized
production 20 percent or more of total production may remain unharvested. In
years with small crops all the cherries are harvested and utilized. Finally,
another factor influencing tart cherry supply and price are the carry-over stocks of
frozen or canned cherries from previous seasons.

More recently tart cherry supply has shifted to include greater global
participation (Zepp et al 1996). Countries such as Poland, Turkey and Hungary
are producing tart cherries as well, (Morello varieties principally) with similar
qualities to those produced in Michigan. These countries have increased their
world share with more global exports to satisfy the demand (T hornsbury 2005).
Increased global competition in tart cherry production has further increased price
instability since about twenty percent of the crop is sold to foreign buyers.

Bad weather in Michigan has also contributed to the strength of its
competitors by widening their market windows of opportunity. For example,
when the 2002 production was lost due to frost during the spring, Michigan
processors had to import from other states and countries, giving competitors the
opportunity to establish new and stronger market relationships with typical
Michigan buyers in state, and around the world (Rowley 2005).

Currently, Michigan tart cherries (Montmorency variety) are marketed

primarily in five different forms: (1) frozen, (2) canned, (3) dried, (4) fresh (in a

small window of three to four weeks in July), and (5) concentrated juice
(Michigan Cherry Committee 2003). Tart cherries are mostly processed, leaving
less than 0.3 % for the fresh market (Table 1). All five products are based on
minimum value added characteristics with no significant differentiation among

competitors.

Table 1. Processed and fresh tart cherry output.

 

 

 

 

Processed tart cherries Fresh tart cherries
Million pounds Million pounds
2005 207.5 0.5
2006 179.8 0.5
2007 192.5 0.5

 

 

 

 

 

Source: USDA/NASS, Non citrus and Nuts fruit report, 2007, preliminary summary.

Demand for frozen cherries depends on the form. Individually quick frozen
(IQF) fruit is mostly used for pies, pastries and fruit preserves. In processed form
the most stable quantity of IQF tart cherries has been produced since 1997. The
demand for the second form, “5+1” (fruit packed in sugar)‘, has declined partially
due to the lower quality of the fruit and to the large amount of sugar added.

Demand for canned tart cherries has also declined. This is due, at least in
part, to health concerns regarding the high sugar content, fewer people cooking
at home, and the shortage in production during the previous three years. The
canned sector is divided into two different packages: Water pack in institutional
#10 cans (105 ounces), and supermarket can size (14 to 16 ounces). The large
cans are mainly exported to Europe and Asia (Facer 2005). Table two shows the

price changes in processed cherries.

\

1 “5+1 products conversion factor is 33.33 pounds of fresh tart cherries for 5+1 fruit-to-sugar ratio
Calculated by multiplying finished weight by 1.11 to determine RPE”. (Cherry Marketing Institute
Statistical Handbook, 2003). RPE is the acronym for Raw Product Equivalent in the Code of
Regulations of the US. government.

Table 2. Price of processed tart cherry per pound in U.S. Dollars.

 

 

 

 

 

 

F Price of processed tart cherries $IIb
‘ Processed in
Million pounds 2005 2006 2007
from 2005-2007
Canned 51.0 0.240 0.160 0.260
Frozen 146.0 0.230 0.210 0.260
Other 10.5 0.141 0.153 0.161

 

 

 

 

 

 

Source: USDA/NASS, Non citrus and Nuts fruit report, 2007, preliminary summary.

Dried cherries and juice concentrate are the newest product forms.
Growth in sales for both has been limited by inconsistent quality standards,
packaging, labeling, and marketing strategies towards end-users. These
products represent a great opportunity for growth, given the right marketing mix
since they are new concepts that appear to have great potential.

Dried cherry consumption has remained stable because most sales are
made in bulk through the USDA school lunch program. Sales through the USDA
results in a very low commodity price because large amounts are ordered. The
price has declined, from $10.00 to $4.00 per pound during the last five years
(Nugent et al 2005).

Tart cherry juice is sold almost exclusively in concentrate form, and then
only in limited geographic areas and/or in large industrial-size quantities. This
segment has also declined, from 250 million gallons in 2001 to 160 million
gallons in 2003, and supply is influenced by low volume crops in 2002 and the
Marketing Order that influenced output in the subsequent years. The price of

d ried cherries and concentrate juice, represented in Table 2 as “others”. These

products had a more steady and regular price growth than the canned and frozen
categories, due principally to their novelty.

But product awareness is growing among consumers as the tart cherry’s
antioxidant properties have become better understood. Changes in consumer
life styles, the growth of new healthy foods, and alternatives to conventional
medicine have increased the pattern of demand for traditional tart cherry
products.

As a result, tart cherry processors are looking for new tart cherry value
added products to satisfy this growing potential demand. As in many other
agricultural industries, marketing and economic changes have increased
pressure on the Michigan fruit sector. First, demand for more traditional products
or commodities has lagged behind increases in output, which makes the demand
for tart cherries elastic; this means that grower prices rise sharply during years of
a small crop, and fall sharply in years with a large crop (USDA 1996). Increasing
long—run returns are necessary to cover additional risks for Michigan producers.

In light of these growing challenges, new strategies are needed to

increase the sustainability and profitability of agriculture in Michigan’s rural
economy (Nykiel 2003). Industry strategies to capitalize on consumer demand
trends must incorporate science-based decision-making tools (Day et al 1989;
Karipidis and Galanopoulos 2000). Developing new products for new segments
and markets, and creating value added for commodities are essential steps to
minimize risks for the industry. Best (2004) defines innovation and value added

D roducts to include every new product in the category with different presentation,

size, and concept that will suit the expectations of the consumers and potential
buyers.

When developing a new product many steps are required, in order to
minimize the launching risks. The industry needs to evaluate every single step
involved in developing a new product to make sure they will be able to deliver a
constantly high quality product. Adelaja (2004) presented at the Cherry
Marketing Institute a summary of steps to commercialization, adapted in Figure
2. It is very important to understand the raw material first, in order to add value
to the product and the supply chain. Then manufacturing processes must be
considered because they will transform the product into the desired value added
product. Without budgets, loans or cash, investment in equipment limits product
development. Simultaneously it is necessary to understand the market targets

and trends to indentify the correct market niche. These factors are inter-related

when developing a new product.

 

 

New Product
I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I I
| Raw Material J I Manufacturing | Marketing ‘ Finances |
Analysis
Availability Production Budgets
Quality Equipment Packaging Cash
Price Cost Others Loans
Seasonality Target Evaluation
Evaluation - Trends
4 Iii/gigging Evaluation
Availability A
Usages
Reclosability
Compatibility

 

 

 

Figure 2. Summary of steps needed when developing a new product
Adapted from Adelaja, 2004

 

 

 

 

 

 

 

This research focused on the packaging component of the process. It
looked for available materials in the area at a reasonable price, evaluating the
possible equipment the tart cherry industry has in Michigan without major
investments, minimizing the need for loans or expenditure of cash. This research
evaluated the compatibility of the package with the tart cherry juice to give
information to the tart cherry industry on how to add value to a new product:
Single strength tart cherry juice. Finally, and the uniqueness of this research was
to be a key link between academia and industry focusing on creating a project

that could be replicable at the industry level and leaving successfully the lab

benches under the academia settings. This link is important to establish and
reinforce both points of view in order to maximize the efforts and the cooperation
between both sectors.

Lack of consumer focus in the existing marketing mix of the tart cherry
industry has severely limited its long-run sustainability. In an industry that has
been primarily oriented towards bulk commodity sales, the packaging, supply
chain and logistical requirements will likely change substantially with a focus on
end—use purchasers (Hobbs 2002).

Packaging can be used to develop new presentations and sizes, and
minimize the risk of failure during the introduction phase in new and different
categories by satisfying specific consumer requirements. Therefore, packaging
has the potential to provide a valuable contribution to Michigan’s agricultural
strategies by moving away from commodity trade into more specialized markets
and value added products.

The industry is in the process of repositioning tart cherry concentrate from
the beverage commodity market to the consumer market for functional foods.
This is a strategy that the tart cherry industry has begun to explore based on
findings that cherries provide health benefits to human beings (Wang et al 1999).
One emerging area of interest is single-strength tart cherry juice and juice from
concentrate.

Several studies at Michigan State University and other institutes have
found that tart cherries have a high concentration of antioxidants and other

components that benefit those suffering from arthritis, chronic pain and cancer

(Chaovanalikit 2004; Raloff 1999). Most beneficial are the anthocyanins, the
pigments responsible for the red and blue colors of many fruits and vegetables,
which are found in high concentrations in tart cherries.

Repositioning tart cherry concentrate or juice poses a number of
challenges to the tart cherry industry. There is a lack of market information and
benchmark retail price-points. The supply chain and marketing efforts are
insufficient. Most of the packaging is poor, with a lack of quality control,
standards, tamper evidence and ease of use. Without a consumer focused
product, including the necessary packaging and marketing mix, the antioxidant
value that has been identified will not be transformed into sales value-added for
the industry.

There is a need to develop a packaged tart cherry juice product that is
shelf-stable. To preserve the health benefits, the juice needs to be protected
from oxidation which is exacerbated by UV light and from sorption of the
pigments and anthocyanins by the packaging material. The packaging material
should not interact with the juice, and it should preserve the juice’s health
benefits, flavor and perceived quality.

The goal of this research is to characterize the change in juice quality over
time in four different containers: bottles made from glass, aluminum and PET and
stand-up pouches. The factors to be evaluated are: anthocyanin content, color,
flavor, microbial activity, pH levels and Brix degrees. The objective is to identify

the best packaging material to retain the quality and health benefits of the tart

1O

cherry juice. The most common deteriorative mechanisms are oxidation and
flavor scalping.

This research will provide a benchmark for future reference and
development of other tart cherry products. Stability of the tart cherry juice, hot
filled, in bottles made from glass, PET and aluminum, as well as
PET/aluminum/PE pouches will be evaluated. Two working hypotheses will be
evaluated:

Null hypothesis: Not all the containers will perform equally.

Hypothesis 1: The four container systems will perform equally regarding
the stability of the juice.

Hypothesis 1: Juice packed in glass bottles will have higher quality than
PET bottles, stand-up pouches and aluminum bottles.

Two major benefits are expected from the selected package. (1) It will best
protect the juice's nutritional value, taste and health benefits. (2) It will protect the
product from inside and outside contamination and degradation.

Simultaneously, complementary consumer perception research on
different packaging forms (materials, shapes, sizes, labels and graphics) for the
juice has been conducted (Whaling 2007). Combined, the results of these
research projects will enable us to make recommendations about the best
package for tart cherry juice.

Industry and academia are not always following the same path and to
validate research efforts a link between the academia and the industry is

necessary. The uniqueness of this research is that created this necessary link to

11

benefit all the participants. Bringing science to the specific needs of the industry
considering their limitations and risks is a very viable and sustained way of

transferring knowledge and technology to the industry.

12

Il. LITERATURE REVIEW
In this section the relevant literature is reviewed to frame and develop the

experimental design for the tart cherry juice product development. It begins with
a definition of juice in order to create the correct juice proportion from
concentrate. This is followed by a discussion about functional foods and the
specific antioxidants present in the cherry juice (anthocyanins). The last two
sections of this chapter explain how to pack a shelf stable tart cherry juice and
describe the four commercially available packages chosen for this experiment.
2.1. Definition ofjuice

The FAO defines juice as the “extractable fluid content in the fruit cells of
tissues” (FAO 2005). Codex Alimentarius, defines juice as “unfermented but
fermentable liquid intended for direct consumption, obtained by mechanical
process from sound, ripe fruits preserved exclusively by physical means” (Codex
Alimentarius 1991). Juice can be clear (clarified) or turbid (puree); it can be
“single-strength” (direct consumption) or “from concentrate”.

“From concentrate” means a juice obtained by reconstituting concentrate
juice with drinkable water and pasteurizing it prior to packaging, as shown in
Table 3. The label must clearly show that the juice is made from concentrated
fruit juice. Sugars and acids can be added but must follow country-specific
regulations (FAO, 1992). The reconstitution levels differ by fruit, country and
institution. The Michigan tart cherry juice used for this research is reconstituted

and pasteurized juice made from frozen concentrate.

13

Table 3. Common juice designation.

 

 

Term Contents Properties
No additives, no
Pure juice 100% 100% juice adjustments and not from

concentrate

 

Fresh squeezed

Not pasteurized

Held refrigerated, short
shelf-life, food safety
concerns

 

Chilled, ready to serve

100% juice

Held refrigerated, made
from concentrate or
pasteurized juice

 

Not from concentrate

Single—strength

Pasteurized after extraction
and clarification

 

From concentrate

Made with concentrate juice

Reconstituted and
pasteurized

 

Single-strength, frozen after

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fresh frozen UnpasteurIzed extraction
Juice blend 100% juice A mixture of purejuices
. 100% fruit with pulp
Puree ContaIns pulp contents
Sugar, water and acid are
Nectar Pulpy or clear added can have 25 to 50%
juice *
. . . Needs dilution for
Nectar base Requrres reconstItutIon consumption
Juice drink Low in juice 10 to 20% juice *
Juice beverage Low in juice 10 to 20% juice *
Juice Cocktail Low injuice 10 to 20% juice *
o . .
Fruit + aid Lemonade, orangeade :vgtgr/‘Lju'ce + sugar and
. Fruit extracted by water
JUIce extract Water extract then con centrat e d*
Fruit punch Token juice 1% juice + natural flavors
Natural flavored Tokenjuice More than 1% juice

 

Source: FAO, 2005
'Differs in different countries.

A Brix degree (B°) is a measurement of the mass ratio of dissolved

sucrose to water in a liquid as defined by the U.S. Code of Federal Regulations

(21CFR 120.24 subpart B). The FDA requires higher levels of Brix degrees for

fruit juice than does Codex Alimentarius (Table 4), (Pollard 1990 and FAO 2005).

Tart cherry juice from concentrate must be in a range between 14 to 14.3

14

 

degrees Brix. Fourteen degrees Brix means that the juice has fourteen %
soluble solids.

Table 4. U.S. and Codex Alimentarius Brix values forjuice from
concentrate.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Name of the fruit Codex Alimentarius US. Brix Value
Brix Value
Apple 11.5 13.3
Apricot ' 11.7 14.3
Blackberry 10.0 10.0
Boysenberry 10.0 10.0
Chery 14 14.3
Guava 7.7 7.7
Loganberry 10.5 10.5
Mango 13.0 13.0
Nectarine 11.8 11.8
Papaya 11.5 11.5
Passion fruit 14.0 14.5
Peach 10.5 11.8
Pear 12.0 15.4
Pineapple 12.8 13.0
Plum 14.3 14.3

 

 

 

 

Source: Pollard 1990 and FAQ 2005
2.2. Tart cherry juice from concentrate

Fresh tart cherries have a very short shelf-life, and the fresh window for
U.S. supplies is only three to four weeks in July (after harvesting). To ensure a
year-long supply the cherries or juice concentrate must be frozen or canned.
Concentrated forms of juice have influenced the nature of the global juice
industry in product development, stability and transportation.

There are many different methods of concentration including open pan
boiling, vacuum evaporation, freeze concentration, reverse osmosis and electro-
dialysis. Each method has limitations regarding cost, quality of the output, energy
required to process and speed, as shown in Table 5 (FAO 2005). Most Michigan

tart cherry juice is concentrated via open pan boiling in the smaller operations

15

and vacuum evaporation in the larger facilities. Open pan boiling has the lowest

processing cost, but it has high energy consumption, a slow speed and the

quality of the output is lower. Vacuum vaporization provides a higher quality than

the open pan boiling system, and the energy consumption is very low, but the

equipment cost is high, and not all the processors have access to this

technology.

Table 5. Juice concentration systems.

 

 

 

 

 

 

 

Method Cost Quality Energy Speed Other

Open Pan Boiling Low Low High Slow

Vacuum High Medium Low n/a

Evaporation

Freeze High High n/a Slow Limited solids

Concentration

Reverse Osmosis High High n/a Slow Limited solids
and clear only

Electro-dialysis High high n/a Slow Limited solids
and clear only

 

 

 

 

 

 

Source: FAO, 2005

To maintain stability, concentrate is kept frozen to prevent quality loss and
Maillard browning.2 Normally this reaction occurs with rising temperatures and
in products with high viscosity. Results are a brown coloration of the product
(Davies and Labuza 2005).
2.3. Functional foods

“Let your food be your medicine, let your medicine to be your food”
(Hippocrates 400 BCE).

Functional foods do not have a universally accepted definition. Stephen
Defelice, founder and chairman of the Foundation for Innovation in Medicine

proposed the following definition: “Functional food is a food (or part of a food)

 

2 Maillard browning is a non-enzymatic browning process of simple sugars (carbonyl group) and
amino acids.

16

 

that provides medical or health benefits including the preventing and or treatment
of a disease” (Kalra 2003). The main markets for functional foods include Japan,
Western Europe and North America (McDonald 2004). It is difficult to quantify
the market potential for functional foods due to the complexity and diversity of the
products. Euromonitor (2008) reported that the US share of this market segment
has grown from US$ 5,786.1 million in 2002 to US$ 13533.7 million in 2007, as
shown in Table 6.

Table 6. Estimated market size for functional foods for specific regions (US
5 Millions).

 

2002 2003 2004 2005 2006 2007

 

NorthAmerica 5,786.1 6,692.9 8,058.8 10,482.4 12,142.2 13,533.7

 

 

 

Europe 89.1 144.5 200.4 254.8 326.0 422.3
Japan 6,247.5 6,377.0 7,235.4 7,239.5 7,177.1 7,344.8
World 15,364.6 17,262.1 20,269.5 23,996.3 26,616.2 28,833.6

 

 

 

 

 

 

 

 

 

Source: Euromonitor 2008

The Japanese market, for functional foods is the market with more rules
and standards for this category, and this market has taken the most steps toward
using food as preventive medicine. In 1991, Japan was the first country to
develop a regulatory system regarding functional foods. Here functional foods
are treated as a special category called: Foods for Specific Health Use
(FOSHU).3 This is a different category from vitamins and minerals.

Foods require a specific approval to carry the FOSHU claims (AFIC 2003).
FOSHU must be proven to improve human nutrition and health by researchers,

doctors and dietitians. FOSHU must guarantee balanced nourishment, and the

 

3 FOSHU are defined as: “foods in case of which specified effects contributing to maintain health
can be expected based on the available data concerning the relationship between the
foods/food's contents and health, as well as foods with permitted labeling which indicates the
consumer can expect certain health effects upon intake of these particular foods” (McDonald
2004,13)

17

active components must be scientifically confirmed in the product. The product

has to be consumed in a normal way and not in a medicine or pill form, and the

active component must come from a natural source (Nill 2001). Japan has a

large and diverse range of FOSHU foods. In 2006 they were more than five

hundred different products in this category (Functional Foods Japan 2006).

The European market is less structured and refined than the Japanese

market. The European market also has a greater focus on alternative medicine

and alternative sources of health. Germany has the largest market share

followed by Great Britain and France (McDonald 2004). The rest of the

European countries represent smaller markets for functional foods.

The U. S. is the largest single country market for functional foods, with a

market of $13,533] million in 2007 (Euromonitor 2008). But the U.S. market is

less demanding than Japan and the EU, and its regulatory system only supports

some food claims for preventing heart disease and cancer, as shown in Table 7.

Table 7. Health claims for food packaging labeling approved by the FDA.

 

 

 

Nutrient Disease
Selenium Cancer
Antioxidants (Vitamin C and E) Cancer

 

Nuts (specially Walnuts)

Heart disease

 

Omega-3 (fatty acids)

Coronary and Heart diseases

 

B Vitamins

Vascular disease

 

Phosphatidyl serine

Cognitive dysfunction and dementia

 

0.8 mg folic acid

Neural tube birth defects

 

 

 

 

Lutein Eye health

Soy protein Cancer

Green tea Cancer

Calcium Bone fractures, kidney/urinary stones,

menstrual disorders, various cancers
and hypertension

 

 

Lycopene from tomatoes

 

Cancer

 

Source:U.S. Food and Drug Administration 2003.

18

 

The U.S. does not yet have a legal definition of functional foods, and has
not yet approved any health claims regarding this food category. The market
growth for functional foods around the world is driven by demographics,
economics and social trends. The aging population, especially the baby-boom
generation, represents a large workforce that spends more in total on food than
any other generation and has the highest level of disposable income in history.
There is increasing interest in health and nutrition, and an increase in people who
desire to understand the link between them. There is an increase in the number
of educated people world-wide. Health costs have ballooned, and there has
been an explosion of preventive health and acceptance of “alternative
treatments" including more products with labels explaining health and
functionality (Scott Wolfe Management Inc. 2002).

Tart cherry juice can be targeted to different market segments due its
versatility and health benefits. The target market segments for functional foods
are baby-boomers4 and Generation-X5 principally, responding to lifestyle
changes, attitudes and activities developed over the past decades (Scott Wolfe
Inc. 2004). The Generation-X population is the key for the new product
development, because this group is beginning to develop product consumption
habits for themselves that they will pass on to their children who will then adopt

new products for their own consumption (Best 2004).

 

‘ Baby boomers are defined as a person who was born during the Post-World War II baby boom
between 1946 and the early 19603 (Statistics Canada 2006).

5 Generation-X is defined as people born from 1965 to around 1982 in many countries around the
world (Brown 1997).

19

But the functional foods category has some challenges that can
compromise growth in the ensuing years. Wade (2006) reports ten challenges
that the functional foods industry has to overcome to continue its growth: (1)
price, (2) lack of consumers awareness or demand, (3) taste/flavor quality, (4)
lack of category definition, (5) legal and regulatory environment, (6) lack of
marketing data, (7) inadequate distribution, (8) absence of long term opportunity,
(9) already too competitive and (10) the inadequate source of ingredients supply.

The lack of consumers' awareness of functional foods can be resolved in
part by a correct and positive labeling. Labeling is a way to show the consumer
the product benefits. It is part of nutrition education. Thus the labeling must
show the correct amount of the nutritional components in the product at the end
of the shelf-life to inform the consumer.

Claims regarding anthocyanins, such as those found in tart cherries, have
not yet been approved, and so labeling must be done carefully (McDonald 2004).
Some labeling concerns for tart cherries were raised by FDA since 2006 due to
health claims printed on the packages and websites. Processors had to change
their labels and web pages for the tart cherry products because they contained
non-approved health claims (Good Fruit Growers 2006).

Since there is no regulation against identifying the amount of nutritional
components like anthocyanins, it is important to understand their stability over
time so that labels can accurately reflect the amount. This can inform consumers
and avoid possible liability and legal issues with FDA. However, no literature has

been found on stability of anthocyanin content in functional beverages.

20

2.4. Anthocyanins
Anthocyanins are antioxidants, meaning that these compounds prevent

the degradation of human cells by inhibiting the initiation of free-radicals that
cause cell degradation (Cao et al 1993). They are the water-soluble pigments
that impart color to flowers, and other plant parts like fruits -- colors ranging from
violet and blue to most shades of red (Harborne 1967).

Anthocyanins are a subclass of flavonoids and belong to the group called
polyphenols. They can be found in the vacuole of the cell dissolved in the cell
sap. They have in the main molecular chain a CaCaCs skeleton that is typical of
flavonoids. Their main part is an aglycone that contains double bonds
responsible for the absorption of light around 500 nm making the pigment visible
to the human eye.

There are 22 different anthocyanins but only six are common in food.
These are pelargonidin, cyanidin, peonidin, delphinidin, malvidin and petunidin
and are represented in Figure 3 (Francis 1989). The most common are
pelargonidin, cyanidin and delphinidin. The main differences between these
aglycone are the number of hydroxyl groups at carbon 3, 4 and 7 in the A ring,
and methoxyl groups. In the B ring they change; pelargonidin has one hydroxyl
group at 4’, cyanidin has one hydroxyl group at 4’ and a second one at carbon 3’
and finally delphinidin has the hydroxyl groups at carbons 3’, 4’, and 5’ (Paul and
Palmer 1972). Products rich in cyanidin, pelargonidin or delphinidin are less
color stable than products containing petunidin or delphinidin because the
hydroxyl group is blocked, and sugar moiety influences stability, therefore this

mechanism still is not well understood and more research is being done (Von

21

Elbe and Schwartz 1996). Anthocyanins are greatly impacted by pH,
temperature and oxygen and in less proportion by degradative enzymes,
ascorbic acid, sulfur dioxide, metal ions and sugars (Von Elbe and Schwartz
1996)

Cyanidin is the most recurrent anthocyanin in woody plants. Cyanidin can
shift from the red to the blue of delphinidin with additional hydroxyl groups.
Methoxyl groups can replace the hydroxyl groups causing the color change.
Another factor that affects color stability is the sugar residue attached by glycosil
linkage to one or more hydroxyl groups (Paul and Palmer 1972). This means

that when the sugar starts to dissolve a color change occurs in the product.

22

 

 

FIILZW /. “.2... ‘
HO ’/ \ HO + f 77*— "I.
. As ,O .. ’.\ B m O / x, 2--..
\r/ ‘ Z w/ 1“ T / \I‘b / \‘g:\ /,-/i".:\, B {/5 OH
'i A C s ~ ~ 4 ‘xz‘i‘ ix {,I/
LIN -’€’A\ r”; C i i J
\59 '~\;»' CH ,i . .
, , 4/2 \\ PeonIdIn
OH PelargonIdIn ’/ OH
OCH3
, OH /
'/ ,r’:::/
’14 . :‘".~\ ‘/.’ \\\
+ f o: ,2 B OH
0, /Q B fl”“OH K2 g, If
\‘q; .2 \,\. \, ’x‘
' "3.5._-_.__--;"/ C = \\
C \. J. OCH3
’. . [JI/i/z \\
fly‘x 0“ f” ‘OH ..
’9' . MalvrdIn
DelphInIdIn
/ OCH3 / OH
/"' //
0+ // \\ / \,\
:5 eff/V“ B 4,} OH 0+ :\ ./.’{‘\\ B fit} ——-—- OH
Q" ' \:‘\"‘ "ff" -.\{§/' \\.\_\. {/I".
\‘- :1." l as, g ‘
“9'77 \O H OH 2:57” \ Cyanidin
PetunIdIn

 

 

 

Figure 3. Six common anthocyanins formed in plants,
Adapted from Francis 1989.

Anthocyanins can also suffer dramatic color changes in mediums when
the ion concentration or pH changes. In dilute acids, the oxygen in the rings
carries a positive charge, resulting in a structure called oxonimum or flavylium
ion. At pH less than 3.0 anthocyanin color shifts to more of a red color. If the
medium shifts to be less acidic, than the red color changes to purple (quinine

form), and in an alkaline solution it turns blue. At pH close to neutral (pH=7.0)

23

the violet base color is in equilibrium with a form of a pseudo base that many
times is colorless. Anthocyanins are more stable in the flavylium ion stage than

in the pseudo base where oxidation can start (Paul and Palmer 1972).

Cl- 0 OH
0+ 9-“ ‘ HCl 0 / _ =0
0 $0 \‘ OH
OH OH

OH Color Base
Flavylium ion Violet
Red
N32C03
0 0N8
0 __ =0
H
OH
Salt of Color Base
Blue

0 OH
- OH
(Du—.OH
O
OH

Pseudo Base

 

 

 

 

 

Figure 4. How anthocyanins can change from any stage to any stage with pH
change (cyanidin example),
Adapted from Paul and Palmer 1972.

Figure 4 shows the changes suffered by anthocyanins when the pH of the
solution changes. Not all anthocyanins will have such a dramatic change. Those
with 4 hydroxyl groups in the molecule and one unsubstituted hydroxyl at the 4’

position are more prone to change. Cyanidin has these four hydroxyls, and Is

24

one of the most susceptible anthocyanins to color change (Paul and Palmer
1972)

If products containing anthocyanins are packaged in metal packages like
cans or bottles, they should have a polymer coating because the anthocyanins
and acidity can cause pitting and perforation of the aluminum or steel. Seams or
imperfections in the coating plus the acidity of the product (such as juices) can
lead to corrosion. Corrosion occurs due to metal ions that have been dissolved
by the acid or by a depolarization mechanism where hydrogen is removed (Paul
and Palmer 1972).

Furthermore anthocyanins can also react with metals such as aluminum
and tin. Anthocyanins with more than two unsubstituted hydroxyl groups are
more prone to this reaction; red colors turn more purple or purple changes to
blue with a more stale hue making food unappealing. This reaction is called
chelating and can disassociate in a more acidic medium. Chelate is a chemical
compound in the form of a heterocyclic ring, containing a metal ion attached by
coordinate bonds to at least two nonmetal ions. This reaction is non-reversible.

Anthocyanins can react with temperature as well. This chemical reaction
is irreversible compared to the acidity related reaction previously described and
will create a browning of the pigments. The problem is not the loss of the
anthocyanins but a change in the molecule due to degradation. Degradation can
be caused by processing temperature, storage temperature, oxygen and UV
light. Degradation is promoted by high pH, oxygen in the head space, presence

of sugars and the presence of ascorbic acid (Paul and Palmer 1972).

25

When a molecule has at least one reactive oxygen specie (OHo), UV light
or oxygen can start a chain reaction creating free radicals that will lead to product
degradation (Burke and Nair 1986).6 When molecules have reactive oxygen
species such as hydroxyl (OHo), then peroxyl radicals (R000) and the
superoxide anion (020) are produced as a result of metabolic reactions (Wang et
al 1999). These reactions can be caused by a source of energy like heat or UV
light where the OH and double bonds in the anthocyanin molecule structure are
initiated. Once the reaction is initiated; it goes through a chain reaction that will
continue creating more free radicals. This chain reaction can be stopped by the
bonding of two radical intermediates resulting in a stable molecule or when the

molecules are consumed (Figure 5).

 

6 Similar reaction occurs with human cells that results in changes to DNA due to free
radical formation. These changes can lead cancer cells formation or mutations due to free radical
reactions (Burke and Nair 1986)

26

 

 

 

 

 

 

 

 

Cl-Cl+ energy —>ZCI- Initiation
-\
H H Ch in
H-C-H + Cl-—>H-¢ -+H-Cl Rezcflon
I'I H
>-
H H
H-¢ + CI-Cl+H-C—Cl+
CIH I'I
_/
\
2Cl- —_> Cl-CI
H H Termination
H-co + Cl- _. H-c -C|
H H >

ii'i l
_./

 

H '32
Eff—”m

 

 

 

Figure 5. Free radical formatio
Adapted from Reuch 1999.

Anthocyanins are also affected by anthocyanase, a browning enzyme, in

Many anthocyanins are found in cherries, and the most important are:

larger quantities of Cyanidin 3-gl

n process.

fresh fruits, that is activated when they are bruised. Enzymatic problems can be

eliminated by heating the fruit up to 80°C (Paul and Palmer 1972).

cyanidin, delphinidin, peonidin, pelargonidin and petunidin (Chandra et al 1992)
Tart cherries have a high Oxygen Radical Absorbance Capacity (ORAC) due to

ucoside, Cyanidin 3-rutinoside and Peonidin 3-

27

 

 

rutinoside. Concentration of these anthocyanins can be determined by high
pressure liquid chromatography (HPLC), mass spectrometry (MS) and gas
chromatography (GS) methods (Esti et al 2001; Kim et al 2005; Hong et al 1990;
Wang et al 1999 and Chandra et al 1992). i
2.5. Shelf-life stability

The shelf-life of a product is determined by the amount of time the product
remains acceptable to consumers, including taste, chemical, physical and
microbiological characteristics. Food Processing Technology(2008) defines
shelf-life as: “the period of time which a product can be stored, under specified
conditions, and remain in optimum condition and suitable for consumption”. For
a functional food, the product needs to retain its functional components over
time, especially when conformance to a labeled health claim is required.

Shelf-life is affected by the structure and composition of the product, the
environment outside of the package, and the interaction of the package and
product leading to chemical and physical reactions (Vercelino et al. 200; Ayhan
et al 2001). These reactions will lead to color and flavor changes, and could
even make a product unsafe. There are different methods to lengthen the shelf
life of fruit juice including heat, pressure, aseptic technology and electrical
conductivity. Each of these methods will have different degrees of advantages
and disadvantages. One of the most common methods is pasteurization and hot
filling. Hot fill packaging and a low product pH help to control the initial and end

of shelf-life microbial burden.

28

2.5.1. Hot fill and product acidity
Hot filling is a heating process in which the beverage is pasteurized at

temperature above 85° C (185° F) for a determined period of time and then
packaged while it is still hot. The main purpose of this process is to kill harmful
organisms such as bacteria, enzymes, viruses, protozoa, molds, and yeasts, and
create a shelf stable product that can be stored at ambient temperatures for
several months (Solberg et al 1990). This method was developed by the French
scientist Louis Pasteur in the nineteenth century and is used to achieve long
shelf-life for juices at low cost.

It is sufficient to inoculate viable vegetative forms of microorganisms but
not heat—resistant spores. If the manufacturing plant has good hygiene, hot fill is
suitable for non-carbonated beverages and juices.

The hot fill method is simpler than aseptic technology.7 The package
chosen must ensure integrity when exposed to high filling temperatures and
prevent post process contamination. It is necessary to choose packages with a
high barrier against gas penetration and aroma and sorption. Microorganisms
can be destroyed by heat, and each organism will have different resistance and
tolerance. Most significant are the Clostridium. botulinum (C. botulinum) spores,
which can be killed at 82° C (180° F). C. botulinum is the most poisonous and
resistant naturally occurring substance in the world, and if its death is ensured,

other organisms are also destroyed (Brin et al 1999; Hersom and Hulland 1980).

 

7 Aseptic packaging is the technology and process where food and beverages are first
sterilized for a short period of time and then filled and sealed under sterile conditions that include
the packaging containers, packaging devices and products to be packaged.

29

Hot fill temperatures must be high enough to reduce the microbial charge
to safe levels and to inactivate fruit enzymes that can produce browning and
fermentation. At the same time, temperatures have to be low enough to avoid
destroying the bonds in the anthocyanin molecules which would produce free
radicals, causing oxidation and changes in color and taste (Marquez et al 2003).

It is especially necessary for non-acidic foods to be heat processed. The
Danish biochemist Soren Peter Lauritz defined pH in 1909, as an inverse
logarithmic measure of hydrogen ion concentration (H).

pH = -log1o at.“ ; where aH+ is hydrogen ion activity.

The amount of acid present in food is measured on the pH scale
extending from 0 to 14. It is rare for foods to be in the alkaline range (a pH of 7
or above). Foods with a pH from 0 to 7 are acidic and can be divided into low
acid and high acid (Star 2004).

High acidity (with a pH below 4.5) is very important in food processing,
because C. botulinum cannot grow in this environment. Tart cherry juice is
considered to be high acid, with a pH of 3.92. It is essential to the stability of the
product to maintain the pH below 4.5. The high acidity will prevent the growth of
some bacteria, while heat will destroy most yeast and molds.

2.5.2. Color and anthocyanin stability

Anthocyanins are a very important part of the juice and a very important

attribute for product marketing. USDA and FDA regulations state that the

nutritional values on the label must be the minimum nutritional values in the

30

product through the duration of the shelf-life. Therefore, anthocyanin contents in
the juice must be evaluated over time.

The stability of color and anthocyanins depends on the pH, temperature,
process, anthocyanin concentration, physical-chemical properties, enzymes and
storage (Rein 2005 and de F reitas and Mateus 2006).

Anthocyanins and color are degraded by oxidation and elevated
temperatures. The combination of oxygen and elevated temperatures produces
the greatest color destruction and loss of health benefits (Nebeskey et al 1949;
Von Elbe and Schwartz 1996). A brown hue appears as the pigments degrade
and polymerize (Kearsley and Rodriguez 1981, Constela and Lozano 2005,
Marquez et al 2003, Turker and Erdogdu 2005). The heat process used for tart
cherry juice should be limited in time and temperature to prevent the browning
effect.

Enzymatic browning is another concern relative to the color and
anthocyanin stability. Enzymes react with other phenolic compounds present in
the food (corresponding quinones), which will then react with anthocyanins. This
results in degradation, giving a brown coloration to the product (Rein 2005).

Anthocyanins are more stable at low pH values, and will fade as pH is
increased. The majority of anthocyanins from plants are non-acylated and are
unstable when pH changes, especially cyanidin and its derivates (Stintzing et al
2002)

Turker and Erdogdu (2005) proposed that anthocyanins are more stable at

a pH level of 2 and concluded that the effect of pH change is more dramatic than

31

the temperature effect. Color loss is reversible (Scaman 2005; Jiang et al 2004;
Turker and Erdogdu 2005) and pigments can return to their original colors if pH
values fall. Jiang et al (2004) described how they manipulated the red color of
litchi fruit by modifying the pH. If the acidity level is the only change in the juice,
color can be manipulated by adding acid. Unfortunately for tart cherry juice
producers, pH is not the only change that the juice will undergo.

They are three major physical-chemical reactions that anthocyanins can
go through: (1) reaction of anthocyanins with small compounds, (2) condensation
between anthocyanins and flavanols mediated by aldehyde and (3) direct
condensation between anthocyanins (Freitas and Mateus 2006, Gomez-
Cordoves 2004). Pyruvic and phenolic acids and acetaldehyde, are some the
small compounds that can react with anthocyanins. Condensation of
anthocyanins and flavanols (tannins8 included) can be done directly or mediated
by acetaldehyde present in tart cherry juice. Tannins are mostly colorless;
therefore they can also range from yellow to light brown. When copolymerized
with anthocyanins, they can result in a bluer tint to the juice (Gomez-Cordovés
2004, Eglinton et al 2004)

The packaging material plays a very import role in protecting the
anthocyanins from oxidation. Many packaging materials have good gas and
good moisture barriers, especially colored glass and aluminum. Other packaging

processing techniques such as the controlling temperature and the eliminating

 

8 Tannins: An exact definition of tannins does not exist but, they are a very special
phenolic compounds that have the ability of combining with proteins and other polymers like
polysaccharides (Von Elbe and Schwartz 1996)

32

oxygen in the head-space also help to maintain the anthocyanins and color
stability of the product.
2.5.3. Flavor stability and food packaging interactions

Products can be monitored analytically to ensure safety for human
consumption. However, it is the final consumer who will decide whether or not to
purchase a product. If a product does not fulfill customer expectations it will not
be repurchased. Sensory evaluation is a science that more and more companies
are implementing to minimize the risks of developing new products and
reformulating mature products to better satisfy consumer needs and tastes
(Stone and Sidel 2004).

Flavor and taste are two very different concepts and they are many
different definitions. Therefore in food science two definitions of flavor are
accepted. The first definition is by Hall (1968 p. 352): “the sum of those
characteristics of any material taken in the mouth, perceived principally by the
senses of taste and smell and also by the general tactile and pain receptors in
the mouth as received and interpreted by the brain". The second definition of
flavor has been accepted by the Society of Flavor Chemists: “the sensation
caused by those properties of any substance taken into the mouth which
stimulates one or both of the senses of taste and smell and/or also the general
pain, tactical and temperature receptors in the mouth (Food Flavor 2007).

Taste is perceived when a chemical solution diluted in saliva or with any
other liquid is absorbed by the receptor in the taste buds. The tongue is the main

taste organ, but taste buds are also found in the hard and soft palate, in the

33

throat, cheeks and in the floor of the mouth. They all contribute to the taste
sensation. Four main tastes exist: sweet, salty, bitter and sour. The existence of
other tastes such as umami, astringency, alkaline and metallic have been also
described (Garden and Baird 2000).

Cherries have a number of volatile compounds that influence flavor and
aroma over time, including alcohols, aldehydes, esters and ketones (Matteis et
al 1997). Bauer-Christoph et al (2005), Mattheis et al (1997), and Girard and
Kopp (1998) agree that the most important volatile compounds in cherries are:
methanol, hexanol, benzaldehyde and acetaldehyde.

Benzaldehyde is responsible for much of the flavor strength, and
acetaldehyde gives the cherries their freshness and aroma. (Esti et al 2002,
Mattheis et al 1992). Benzaldehyde is also present in almonds and can be
recognized as the flavor of almond extract. Benzaldehyde has a benzene ring
with an aldehyde substituent. It is the simplest representative of the aromatic
aldehydes and one of the most used industrially in its family (European Chemical
Bureau 2006). Like all volatile compounds, benzaldehyde decreases over time,
but it is not as volatile as acetaldehyde. Amounts remaining in the product are
influenced by the packaging material, heat and time (Meheriuk et al 1995).

These compounds can be identified through gas chromatography analysis
(GC) as a quantitative method or with a trained panel for qualitative testing.

Packaging plays a very important role in protecting product flavor.
Packaging provides a barrier against oxygen, UV light and from the environment

in general (permeation). It prevents packaging compounds from migrating to the

34

products and prevents the absorption of flavor compounds from the product to

preserve flavor.

Interaction within the system refers to the different mass and energy

exchanges between the food, the packaging material and the external

environment that the package system to which it is exposed to (Hotchkiss 1997).

Permeation includes to two different mechanisms: diffusion of molecules through

the packaging material and absorption or adsorption from/into the internal or

external atmospheres. Migration refers to the release of compounds from the

material to the food product, which can be harmful for human health. Finally,

absorption or scalping refers to the transfer of molecules to the packaging

material which causes a loss in flavor, aroma and/or color. Figure 6 shows these

three mechanisms that can occur in a packaging-food system.

 

Environmen l
+—§-——

e
l

5‘ V}
g.

. .‘o“.- l ‘.__\‘ .
.~.. 1 _.. . -.

 

Packaging Material

Food or

: Permeation

2 ’ Migration

Absorption
Or
Scalping

 

 

 

 

Migrating

Result

 

 

Oxygen
Water Vapor

C02
Other gases

 

 

Monomers
Additives

 

Oxidation
Microbial growth
Mold Growth
Off flavor
Dehydration
Decarbonation

 

 

 

 

 

Aroma
Compounds
Fats
Organic Acids
Pigments
Anthocyanins

Off flavor
Safety issues

 

 

 

 

Loss aroma
intensity
Unbalanced flavor
Profile
Package damage

 

 

 

Figure 6. Permeation, migration and absorption.

Adapted from Nielsen and Jagerstad 1994.

35

 

 

The compatibility of the packaging material with a food or beverage is one
of the most important problems faced by industry when developing new products.
This mechanism can be extremely complex and depends on both the packaging
material and food molecular structure. Two of the most important parameters to
analyze when choosing a package are: (1) glass transition temperatures (T9) and
(2) polarity of the molecules in the food product and the packaging (van Williege
2002)

Glass transition temperature refers to the temperature where the materials
pass from a glassy or brittle stage to a rubbery stage (Wesselingh and Krishna
2000). Of the polymers used for bottles, polyethylene terephthalate (PET) and
polyethylene naphthalate (PEN) have a T9 above room temperature. They have
very stiff chains that reduce the diffusion coefficient for flavor molecules at low
temperatures since these molecules are harder to disrupt. On the other hand,
polyethylene (PE) and polypropylene (PP) have chains that are less stiff and
have a higher diffusion coefficient, and so volatile compounds can more easily be
absorbed by the polymer structure.

The electrical charge of the polymer molecules affects flavor and aroma
scalping. It is necessary to find packaging food systems that will prevent
scalping. Polyesters (PET and PEN for example) are polar and so they are less
attracted to non-polar volatiles or flavor compounds, therefore, they are not prone

to flavor or aroma scalping. On other hand, polyolefins like PE and PP are non-

36

polar and so they will promote the scalping mechanism that can shorten the shelf
length of products or minimize the overall quality (Sullivan 2005).
2.6. Materials

Euromonitor (2005) reports that, in the U.S., packaged food grew by 16%
from 2000 to 2005. The soft drink category, over the same period of time, grew
about 20%. Soft drink, or non-alcoholic beverages, is the category that includes
carbonated andnon-carbonated beverages. The non-carbonated beverage
category includes water, juices, artificial and natural flavored fruit beverages and
energy drinks (Hicks 1990 Nermark 2002).

Packaging for this category includes an array of possibilities including
glass, aluminum, steel, coated paperboard, different kinds of plastics and flexible
packages. Most common are glass, PET, aluminum, Tetra Pack® and flexible
packages that in the last ten years have been growing in market share
(Euromonitor 2005).

The most important beverage packaging trends for new product
development are convenience, recyclability and innovation. Convenience
includes portability, reclosability and individual portions. There is a demand for
recyclability or minimal waste production due to environmental issues and
consumer knowledge. Innovation and quality are part of the product image and
communication with the consumer (Getachew and Peterson 2005).

Euromonitor (2005) reports that the most common packages used for
functional foods in the European Union (EU) are first metal (51 %) and then

plastic. Plastic is first in the U.S. (73.61%) and Japan (52.55%), and metal is the

37

second choice. Glass is the third choice and flexible packaging is the fourth for
all three geographic regions. The package used least in the three regions for
functional foods is cartons. Cost, availability and recyclability are the big drivers
in the processors selection of packaging for functional foods.

For this project, packages were chosen to meet the consumer demand for
convenience, recyclability and innovation. Packages were also chosen to be
compatible with the tart cherry juice, with the manufacturing requirements,
market availability and commercial ability. Four packages were included: glass,
aluminum and PET bottles and stand-up pouches.

In general, glass is one of the most inert materials; it provides excellent
moisture and oxygen barrier, and scalping is not an issue. PET is a good gas
barrier, especially with the help of an oxygen scavenger layer that an trap the
incoming oxygen and soluble oxygen present in the juice. Aluminum bottles are
a new package for beverages and are also provide an excellent gas and
moisture barrier. These three packages can be recyclable and a recycling
stream is available in the U.S. as it is in other countries of the EU. The stand-up
pouch is not recyclable due to the different material layers; however, they do
produce less waste volume. Following is a review of the properties of each of

these materials.

2.6.1. Glass
Glass is one of the oldest materials used for containers although it was not

until the early 20th century that the glass container process was automated.

Glass bottles can be shaped by a blowing method or a pressing and blowing

38

method. Glass is a molten mixture of inorganic oxides: 73 % silica sand ($002),
15 % sodium oxide (N320), 10 % calcium oxide (CaO) and 2 % of other materials
like aluminum oxide (Ale3) and iron oxide (Fe203). ASTM C 162-03 defines
glass as a non-crystalline material obtained by a melting process. Glass has a
very irregular atomic arrangement as shown in Figure 7 (Masayuki and Asahara

2005)

 

 

 

 

 

Figure 7. Molecular arrangement of glass.
Adapted from Masayuki and Asahara 2005.

Commercial glass has a density of 2.5 g cm'3, one of the highest among
the different types of glasses, a coefficient of thermal expansion between 0 to
300° C of 92 x 107 °C'1, and a thermal conductivity that ranges between 10 to
11.2 Wcm"°C'1 when temperature changes from 10 to 100° C. It also has one of
the lowest thermal shock resistances among the different glass types (Bansal et
Doremus 1986).

Glass has been identified as the most chemically inert material available

for food packaging and has a very low coefficient of solubility of gases in its solid

39

state (Bansal and Doremus 1986). It is a great barrier to oxygen and moisture
due to its impermeability, absence of pores and homogeneity.

In its solid state, the molecules are bonded tightly together giving glass its
excellent barrier characteristics (Halloway 1973). Therefore the food product
stored inside the glass can be in direct contact with the internal wall without any
coafing.

Glass provides many options for labeling and marketing. Labeling options
include paper and plastic labels, with applied adhesive or self adhesive, as well
as plastic shrink labels. Special molds can be developed for unique and
innovative identification, although it is much more expensive to have to use a
custom mold compared to a stock bottle (Glass Packaging Institute 2006).

Glass bottles have five very practical operational properties: (1)
availability, (2) returnability, (3) ease of packaging, (4) ease of distribution and (5)
recyclability. They have four favorable market properties: (1) product recognition,
(2) container appeal, (3) flexibility and (4) environmental friendly (Kurylowsky
1996). Glass also offers better gas and moisture barrier properties, better
absorption barrier properties, and better UV protection than plastics, and thus
gives a longer stability for food products (Coutelieris and Kanavouras 2005).

There are also disadvantages of glass, most notably the high risk of
breakage during production, filling, storage, transportation, display and usage.
Glass is generally a high cost package due to the energy cost used to blow

bottles and transportation energy costs due to the heavy weight.

40

For over 100 years, glass was the most common package for beverage
containers (Nankivel 1997), although much of the beverage Industry has now
switched to PET and metal cans. U.S. glass industry shipments peaked in 1980
at 47 billion units, declining to 34.5 billion in 2004 due largely to substitution.
Glass containers now account for 4.5 % of all shipments of the packaging
industry (Impact Marketing Consultants Inc 2006). Glass has been reserved for
high-end products especially with the competition of other packaging materials
since the seventies (Hook and Heimlich 2006).

EPA reported that in 2003 Americans, generated 12.5 million tons of glass
that ended in the municipal waste stream with only 22% of that ending up in the
recycling stream.

2.6.2. Polyethylene terephthalate

Plastics are a newer packaging option and 29% of the 2005 production of
plastics was used for packaging (American Plastic Council 2006).

Polyethylene terephthalate (PET) became available in the 19703, and its use has
increased steadily. The plastic bottle industry volume was over 118 billion units
in 2004, of which PET soft drink bottles comprised the largest proportion at about
28% (33 billion units). PET bottles for water was the fastest growing segment: 23
billion units in 2004, up from only 3 billion in 1998, a growth rate of almost
35%lyear (Impact Marketing Consultants 2008). The majority of PET used for
packaging is for beverages (carbonated and non-carbonated), dressings, and

preserves.

41

PET belongs to the polyester family and is produced by a condensation
reaction of ethylene glycol and terephthalic acid or dimethyl terephthalate.
Condensation polymers are created using a two step mechanism. The first step
produces the amides or the esters as in PET. The second step is a much slower
mechanism and it is called polymerization when the polymer is created. When
condensation occurs in polymers a small amount of water or methanol is lost,
and the most common byproduct is water (Selke et al 2004). PET is generally
blow molded to produce bottles.

PET has many beneficial properties for packaging such as clarity,
inertness, light weight, strength (especially resistance to pressure), moisture
resistance, gas retention for carbonated beverages and good gas barrier.
Another advantage of PET is that the polarity of its molecules helps to repel
aroma and flavor compounds, although it is never as good as glass.

PET has a glass transition temperature (T9) of 80°C, and 22 to 25%
crystallinity (Reim 2005). Figure 8 shows the molecular structure of PET. It has
a benzene ring in the backbone which reduces the crystalline path and two

unsaturated bonds.

42

 

HO - CH2- CH2- OH

+
O O
HO—iIS- -'c':-
O O
O—iIZ- iIZ—O-CHz—CHZ + 2 n'1 molecules of

 

 

 

Figure 8. Molecular structure of PET and condensation reaction.
Adapted form Selke et al 2004. ,

PET has good oxygen and UV resistance. It is resistant to high
temperatures and can be coated or coextruded with different additives or oxygen
scavengers to improve the barrier to oxygen or moisture although this increases
the cost (Risch 1999; Ros-Chumillas et al 2006).

The disadvantages of PET, compared to other resins, is its proportionally
higher price, the two-step bottle production process, lack of rigidity, and bottle
deformation when the walls are too thin (Packaging Council of New Zealand
2004). PET bottles have different weights depending on the product
specifications: for 20 oz bottles, a bottle for water generally weighs between 15
and 19 g; for carbonated beverages the weight should fall between 23 and 27 g;
and for hot fill the range should be 33- 38 g to minimize the bottle distortion due

to high temperatures filling (Selke et al 2004).

43

Although PET is 100% recyclable (APC 2006), the use of recycled PET in
direct food contact has not yet been approved. PET can be recycled into product

forms other than food packages, such as carpets.

2.6.3. Aluminum
Aluminum is the most abundant metal and the third most abundant

element on earth. The first metallic aluminum was prepared in 1825 by Hans
Oersted. Aluminum is a white-silver metal with very low density (2.70 g/cm3). It
is non toxic and very machinable (Shakhashiri 2006)

Aluminum can be found In the packaging world in the form of foil or sheets
that are later transformed into containers. Aluminum foil has been available
commercially since 1910, and aluminum containers have been available since
1950.

Cans are the most common aluminum containers, coated with a polymeric
lining to be used as food grade. Aluminum cans are the second largest material
for can production (Second to steel). There has been a decline in aluminum can
use from a value of U.S. $4,823 million in 1997 to $ 1,677 million in 2004.
Aluminum remains the second largest material for can production (Impact
Marketing Consultants Inc 2006). Aluminum cans are mostly used for soft drinks
and beer (Impact Marketing Consultants Inc 2006).

Aluminum bottles were first used in Japan and are now used for
beverages world-wide. In 2004, a re-closable aluminum bottle was launched by
Alcoa and the Pittsburgh Brewing Co. (NBC News 2004). Bottles are formed by

the stamping method (CCL Containers 2005). Beer, sport drinks, energy drinks

44

and juices are now being packed in aluminum bottles. The aluminum provides a
very sleek, modern, state-of-the-art look.

No information has been published regarding the stability of products in
aluminum bottles. However, aluminum provides a good barrier against light, air
and moisture, and does not oxidize like steel, and so aluminum bottles are
expected to give good product stability. Unlike steel rust, which exposes the
surface to more oxidation, aluminum oxidation creates a white layer that will
protect the aluminum from further exposure. Food grade aluminum cans and
bottles are coated with various kinds of polymers.

Like PET, aluminum is light weight. It has an even higher recycling value
than PET, highest of all packaging materials (The Aluminum Association, Inc.
2001).

2.6.4. Flexible packaging

Multilayer flexible packaging has grown steadily for the past 20 years,
reaching 17% of the U.S. $127 billion packaging market in 2006 (Flexible
Packaging Association 2006). Flexible materials can be used to pack almost any
kind of product. Different combinations of polymers, as well as aluminum and/or
paper, are possible. The choice of materials depends on the intrinsic needs and
characteristics needed to ensure the protection of the food product. The
materials can be laminated, extrusion laminated or co-extruded.

Multilayer materials are not recyclable because they include a combination

of materials. Compared to rigid packaging, however, flexible materials can

45

reduce the volume and the weight of waste (Flexible Packaging Association
2006)

Six benefits of flexible packaging, compared to rigid packaging, are: (1)
cost savings, (2) less material used, (3) less space required through
transportation and storage, (4) reduced effects on the environment, (5) ease of
use and (6) efficiency (Risch 1999).

The main disadvantages are a lack of strength, which may requires an
additional of package to support the load when shipping, less convenience for
the consumer, and lack of recycling. However the addition of reclosable
polypropylene spouts has helped to increase sales in markets where flexible
packages compete with rigid packages (Selke et al 2004).

High density polyethylene (HDPE), low density polyethylene (LDPE) or
linear low density polyethylene (LLDPE) or another polyolefin is used inside of
flexible pouches for food contact and for heat sealing. Polyethylene (PE) is a
non-polar, linear thermoplastic with a density ranging from 0.910 to 0.940 g/cm3.
The melting temperature is between 128 and 138 °C (Fiadus and Tong 1997).
There are many methods to seal the inner polyethylene layer of pouches and
choosing one depends on the needs and the machinery available. Bar or
thermal sealing is the most common method for stand-up pouches. Thermal
sealing is based on two bars that heat press the materials together. Heat causes

the sealing materials to fuse. Some pressure and some dwell time9 are

 

9 Dwell time is define as the period during which a dynamic process remains halted in order that
another process may occur (ITS 2008)

46

necessary to develop a complete bond between the seals (Selke et al 2004).
Hot tack is the ability of the seal to resist strain while still molten.

Linear low density polyethylene (LLDPE) will be the material in contact
with the juice. Since LLDPE is non-polar, it is expected that it will attract volatiles
through the material structure causing absorption or scalping. Depending on the
levels of absorption, degradation may occur, including swelling and disruption of
the polymer molecular structure (Reim 2005).

Multi-layer flexible packaging materials can be laminated or co-extruded.
A barrier can be included between layers to provide a better or longer shelf-life
(Reim 2005). Most flexible packages for liquid have an aluminum foil barrier for
gasses, moisture and UV light, normally ranging from 0.2 to 0.4 mil (Selke 2004).

The outer layer is normally polyethylene terephthalate (PET) due to its
glossy properties, giving a more polished look to the flexible package, better gas
and moisture barrier and better canvas for printing purposes (Plastic Packaging

Innovation news 2005 and Packaging Digest 2004).

47

Ill. EXPERIMENTAL DESIGN AND METHODS

The major degradation mechanisms for tart cherry juice are:
- Microbial growth, which is limited by a hot filling high acid process,

- Oxidation, which is exacerbated by UV light, and

- Sorption of flavor, color and aroma by the packaging material.
The package and process must be selected to ensure that these are prevented.
With the above in mind, four packages were chosen: glass bottles, aluminum
bottles, PET bottles with oxygen scavenger and flexible stand-up pouches with
an aluminum foil barrier and central reclosable spout.

The methods used to analyze these degradation mechanisms are based
on color change, the pH and solids stability (Brix degrees), microbial growth, and
the stability of anthocyanins and flavor. This research is designed to record
these changes analytically and by sensory perceptions and consumer
preferences. Analytical and organoleptic tests were used to compare the stability
of the product in four different packages. During this research the oxidation due
to light (photo-oxidation) was not measured directly since the packages were
kept in closed cartons during the length of the experiment; but the effect of
oxidation was one of the parameters used to evaluate color and flavor due to the
singular brown tones and rusty flavor characteristic of this reaction. Figure 9

shows the conceptual model for this project.

48

 

 

Tart Cherry Juice Shelf —— Color

 

   
    
 

 

 

 

 

 

 

Oxidation pH
Brix
Photoxidation Packaging Microbial
a erIa ——- Growth

 

 

 

 

 

 

Anthocyanin

 

 

 

 

Figure 9. Tart cherry juice shelf-life, conceptual model.

49

The experiment was a series of tests for a 12 month stability study of hot-
filled tart cherry juice in the four package types. At one month intervals, the juice
was tested analytically for color, solids contents, acidity, microbial growth and
retention of anthocyanins compounds. Flavor, color, texture, and overall quality

were qualitatively analyzed with a trained panel as shown in Figure 10.

 

Tart Cherry Juice
Experimental design

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Packaging
Material
1 l l M, ' b, I l l
- Icro Ia .
Color pH BrIx Growth Anthocyanin Flavor
Colorimeter DH Refracto- Counting Mass Sensory
Mete meter Plates Spectrometry Evaluation

 

 

Every Month 3 samples per material

 

 

Trained panel + Consumer test

 

 

Statistical Analysis

 

 

 

 

 

 

Figure 10. Experimental model scheme.

50

3.1 Packages and Closures
Four different containers were chosen to evaluate their effect on product

quality: glass bottles, PET bottles, aluminum bottles and stand-up pouches with a
multi-layer structure. These containers were chosen because of their properties
and availability in the market place. They can be obtained easily by tart cherry
processors in different quantities. The packages for this research were
generously donated by Performing packaging (stand-up pouches), CCL
Containers (Aluminum bottles), Toyo Seikan (PET bottles) and St. Gobain (Glass
bottles.

3.1.1. Glass bottles

The clear glass bottles (flint) were donated by St. Gobain Inc. The bottle
mold code is 4356012B. These bottles are narrow neck and have a capacity of
12.75 fluid ounces. The specification sheet for these bottles details the following
dimensions: height 7.719 inches, diameter 2.492 inches and weighs 255.03 9.
Figure 10 shows the glass bottle used for the project and drawings of the
principal measurements.

The finish designation is 38-400 (continuous thread) and the closures are
metal screw caps with plastisol liners. These types of liners are used in hot fill
procedures, so the plastisol liner melts during processing and seals to the rim of
the bottle. Once the temperature cools, it solidifies creating a tight seal between
the container and the closure. The Standard Finish Dimension Tolerances gives
minimums and maximums for E (size of the finish without the tread), T (size of
the finish including the thread), H (height of the thread part of the finish), S

(distance between the first thread and the top of the bottle finish) and l (internal

51

size of the bottle finish). The 38—400 E has a maximum of 1.476 inches and a
minimum of 1.452 inches; T ranges from 1.382 to 1.358 inches; H from 0.418 to
0.388 inches; S from 0.61 to 0.31 inches; and the interior of the bottle finish (I) is
0.987. The GP|1° finish is 4009.

The metal cap weighs 5.04 9 making the total weight of the package
260.07g. Caps can be silver, gold, white and black, and in this study they are

gold (Figure 11).

 

 

 

7.719"

 

 

 

 

 

 

 

 

Figure 11. Glass bottle.

3.1.2. Hot fill Polyethylene terephthalate (PET) bottles
The PET bottles were donated by Toyo Seikan. The hot fill PET bottles

have a 5 layer structure as follows: 3 layers of PET sandwiched with 2 layers of

 

1° GPI stands for Glass Packaging Institute. They have create a volunteer set of standards that
allows manufacturers to be interchangeable and matching bottle finishes with caps.

52

Slrus 101, a Toyo Seikan patented oxygen scavenger material (composition
unknown and proprietary). Sirius 101 is an oxygen scavenger that provides a
high oxygen barrier by trapping oxygen from the outside and absorbing oxygen

present in the product as shown in Figure 12.

 

PET PET

III III
E)?

Outside F'_ 02

Food

 

 

 

 

 

 

 

 

 

 

 

 

L. _ L. H _ _
Oxygen Oxygen
Scavenger Scavenger

 

Figure 12. Multilayer PET bottle structure and oxygen scavenger process.

Common weights for hot fill PET bottles range from 33 to 38 grams. Since
the bottling temperatures will range from 82 to 95°C PET bottles will suffer
distortions because this temperature is above PET glass transition temperature
(Tg) (80°C). The packaging industry have found different methods to prevent this
PET shrinkage. One method is to “heat-set” bottles, increasing the crystallinity of
the bottle walls up to 30 % under stress to maintain the bottle shape. A second
method is to first heat the bottle pre—forms (10-15°C) and then blow at higher
temperatures (120-140°C) to form the bottles. The base of the bottle is kept cool

to prevent the formation of thermally induce crystals, and bottles are cooled with

53

circulating air for up to one minute to stop the crystallization process. A third
option is to include vacuum panel designs (varying from company to company) to
allow the filled bottle to compensate for pressure changes in the bottle through
the cooling process (Selke et al 2004). Lately the PET industry has found ways
to design “panel-free" hot fill bottles to promote a more aesthetically pleasing
bottle and ease the labeling process (Food Processing 2004 and Plastics
Technology 2005). The bottle walls have a thickness of 16 mil, and the bottle
weighs 33.43 grams meeting the requirements to prevent bottle shrinkage.

The finish is a 38-400 continuous tread with a tamper evident band that
breaks when opened. The finish measurements are the same as the glass
bottle. Figure 13 shows the PET bottle. The cap ls made from polypropylene

with teflon (PTEF) liners Produced by Toyo Seikan as well.

 

 

1

Figure 13. The 8 oz PET bottle from Toyo Seikan.

 

 

 

3.1.3. Aluminum bottles
The aluminum bottles were produced and donated by CCL containers.

The bottle’s capacity is 18 fl oz, and it is suitable for hot filled beverages. The

54

raw material is aluminum with an inside proprietary polymeric coating (HOBA
8280/2).

The shoulder profile is Sonic 66. The outside diameter is 66 mm, the
overall height is 206.0 mm. The weight of the empty aluminum bottle is 67.40 g,
the cap weight is 4.13 g and so the total weight is 71.53 g.

The bottles are made by stamping aluminum sheets into the shape of the
bottle. With this method there are no seams or welding necessary, therefore an
extra step is necessary to create a reclosable finish for the aluminum bottle. A
polyethylene lug sleeve is applied to the finish during the last spinning to create a
slug finish that will match a 38 snap slug shallow aluminum cap. This cap will
have a polyethylene liner to ensure a good seal when hot filled. Figure 14

shows the dimensions of the bottle and a picture of the container.

55

 

 

 

 

 

 

 

 

 

 

 

 

 

7.
34.
l——23~—I _
206.
V
CCL
Container 118'
Hermitage, PA.
(724) 981 —4420
5.0% F 4
I— 66. _t

 

 

 

Figure 14. Aluminum bottle (dimensions in mm).
3.1.4. Stand-up pouches

The flexible pouches were donated by Performance Packaging, and they
contain 7 fluid ounces, the smallest amount of any of the packages’ capacity.
The multilayer structure is formed from an inside layer of low linear density

polyethylene (LLDPE), an outside layer of polyethylene terephthalate (PET), and

56

an aluminum foil layer (Al) in between as is shown in Figure 15. The thickness of
the PET layer is 0.472 mil, the aluminum foil is 0.28 mil, and the LLDPE is 4 mil
making a total thickness of 4.752 mils. The weight of the structure is 11.70 g and
the cap is 1.20 9 making a total weight of 12.90 g. The stand up pouch has a

width of 100 mm and a height of 180 mm.

 

Aluminum Foil

PET 2

LDPE

 

 

 

 

 

Figure 15. Stand-up pouches multilayer structure.

The stand-up pouch has a centered Polyethylene (PE) cap and spout that
makes the flexible package reclosable, as is shown in Figure 16. The opening of
the spout is 9 mm.

Special fillers and cappers are necessary to fill this container, since
pouches are filled differently than bottles. The equipment varies from
manufacturer to manufacturer, though the most common type uses preformed

pouches that are received in a Z- stack. The packer/filler inserts the spout and

57

cap, and the pouches are filled from the bottom.11 The machine will make the

seal, separate the pouches, and then send them to the cooling conveyor.

 

 

  

 

 

 

 

 

 

20.
l HI pp
l 10' 1‘” Securitv band
1 Seal
3 ; //‘ Straw ”P"
10 l /
75 E ;
H
105
Printino 180

 

 

 

 

 

 

 

i 10 ___.__.__._____

 

Figure 16. The Stand-up pouch (dimensions in mm).

 

‘1 One manufacturer of equipment like this is Goglio in Italy.

58

3.2 Packaging technique and facilities
The juice was packed on March 24, 2006 by Food for Thought, a small

packing facility located in Honor, Michigan. Half of the cost was donated by the
company, and they agreed to pack the four different packages.

Food for Thought does not have the technology to pack the pouches.
Performance Packaging, sent the stand up pouches formed, sealed and with
spout and caps on. This is the reason why they were filled with the smallest
nozzle and capped by hand without nitrogen flush.

The company used three cookers with kettles from Savage Brothers in
Chicago, a filler with 2 nozzles from EPAK in LaPointe, IN, a SureKap spindle

capper from Winder, GA. and a cooling belt, as shown in Figures 17 to 19.

.. Il' E§ .. .

3 Cook

  

Figure 17. Food for Thought, cookers.

59

 

  

Figure 19. Food For Thought, cooling belt.

The tart cherry juice was reconstituted and heated to a pasteurization
temperature of 85°C or 185°F (Hicks 1990) before filling the different packages.
Hot fill was chosen over aseptic packaging, because it is the general practice for
juice production, especially for high acid juices (below 4.5 pH). The aseptic
packing equipment it is more expensive, and it is not available among the tart
cherry processors of Michigan.

The temperature was raised over a time period of 15 minutes (time
needed by the cookers at the plant to raise the temperature) and held for 30

seconds. This is sufficient to kill all the enzymes and potential pathogens (Paul

60

and Palmer 1972). The temperature selected was the lowest possible in order to
minimize the degradation of anthocyanins present in the tart cherry concentrate.
Bottles were heated with lamps for 10 minutes to prevent temperature shock and
breakage, and then filled. The headspace was flushed with nitrogen to displace
any oxygen, capped and cooled immediately.

One hundred packages were filled with the juice for each of the four types
of package. The bottles were closed with 15 lbs/in2 of torque with an automatic
capper. Manual torque was used to apply caps to the pouches since the existing
capper could not be adjusted for this kind of package.

3.3. Tests

After packing, the packages were stored at room temperature (23°C and
50%RH) in corrugated fiberboard shipping containers that shield from light
exposure. Every month, three replicates from each'treatment were evaluated
for:

- Color change

- Solids change measuring Brix° degrees

- pH change

- Microbial and pathogen growth

- Anthocyanins stability

- Flavor and taste

The samples were evaluated monthly for a twelve month period. The

different tests are described in the following section.

61

3.3.1. Color
The change in the color of the juice is related to the quantity of

anthocyanins, solids and pH. Color also indicates UV light degradation and
oxidation, when the juice has brown tints.

De Ancos et al (1999) recommended using a colorimeter to measure
Hunter L*a*b* values for juice. They proposed the use of cylindrical cups for the
test measurement.

As an adaptation for this research 15 ml Petri dishes were filled with 15 ml
of juice. They were closed and sealed with parafilm® and with black electrical
tape to prevent the penetration of light through the sides as shown in Figure 20.
A Hunter Lab Colorimeter 45/0 with Color Quest software from Hunter lab In
Road Reston, Virginia U.S. was used. The colorimeter was standardized using a
black and a white tiles to give the following results: L* value: 94.53, a* value: -
0.91 and b* value: 0.94. Samples were placed under the optical reader and
readings were taken by the colorimeter. Measurements were taken in triplicate

and then averaged for the analysis.

  

Figure 20. Sample preparation for color measurements.

62

The L* value is a measure of lightness varying from completely opaque
when the value is 0 and to completely transparent when the value reaches 100.
The a* value measures the redness and a negative value will measure
greenness. The b* value measures the yellowness and the negative value

blueness. Figure 21 gives a graphical reference of the 3 dimensions of color in

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

this scale.
L* =100
white
-a* = ‘\ +b* =
+a* =
_b* =
V
L* =0
Black

 

 

 

 

 

 

Figure 21. Hunter Lab scale reference.
Adapted from Hunter Lab 1999.

The hue value is the ratio of absorbance of fruit juice at 520 and 420 nm.
Lower values of hue represent oxidation and browning of the anthocyanins (Alper
et al 2005). The hue angle h is measured using the formula:

h= arc tan b*/a*.
Chrominance is the saturation of a specific color, and the formula is:
C: (3.2 + b* 2) vs

Finally, total color difference formula is:

63

AE= \/{(L*1—L*2)2 + (a*1-a*2)2 + (b*1-b*2)2} (Remon et al 2004).

3.3.2. Solids

Brix degrees (°Bx) are the amount of dissolved sugars to water mass ratio
of a liquid solution. If a liquid has 30 °Bx means that it has 30 grams of sugar
and 70 grams of water. Brix degrees were measured with an Aldrich Z281611
portable refractive index detector from Sigma-Aldrich in St. Louis, MO (Figure 21)
at room temperature following standard procedures (Serrano et al 2005; Righetto
et al 2005).

Three replicates were analyzed for each type of packaging. For each
sample of the juice, one reading was taken by observing the line where color

changes on the scale, as shown in Figure 23.

 

Figure 22. Refractometer used in the tests.

64

 

:19 Scale

—18
——17

. Reading taken
where color
changes

 

 

 

 

 

Figure 23. How a reading looks in the refractometer visor.
3.3.3. pH levels.

The pH measurement is used to determine the acidity of a product.
Acidity provides a general inhibition of pathogens (4.6 and lower pH) and the
data collected was used to assess the overall quality of the juice (Raso et all
1998).

The samples were tested using a probe, (KS701 ISFET) pH Tester with
Non-Glass Advance Electrode from Pulse Instruments from Bas Inc. in Tokyo,
Japan (Figure 24). The pH meter was calibrated each month before use with
three buffers: acid (4.0), neutral (7.0) and alkaline (10.0), as recommended by
the pH meter manufacturer and literature. Then the probe was submerged in

the tart cherry juice until a stable reading was obtained, and the stable reading

65

sign appeared in the pH meter panel. After every reading the probe was rinsed

with water.

‘ Juice sample

Digital pH meter it

10.0 pH .
Alkaline 7.0 pH
. Neutral

 

Figure 24. Digital pH meteranbduffers for calibration. ‘ 1
3.3.4. Microbial growth

The growth of yeast and mold was evaluated with general counting plates
containing potato dextrose agar (PDA). The juice was diluted with sterile distilled
water to 10", 10'2 and 103.12 Then 10 pl of the solutions were dispersed into the
medium in triplicate for each sample. After the dishes were stored at room
temperature (23 °C and 50% RH) for two weeks, the number of visible colonies

was counted (Figure 25).

 

‘2 10 pl of juice diluted 100 pl of distilled water gives 104,10 pl of 10'1 dilution with 100 pl of
distilled water gives 10:, finally 10 pl of 10'2 dilution with 100 pl of distilled water gives 10'3

66

   
 

,5:
Petri dishes with PDA
medium

and juice sample

Figure 25. Counting plates with PDA medium and juice sample.

3.3.5. Antioxidants compounds
There is general agreement among researchers to use high pressure

liquid chromatography (HPLC) techniques to determine anthocyanin content and
the intensity of flavor compounds in juices (Kim et al 2005; Bauer-Christoph et al
2005; Mattheis et al 1997; Girard and Kopp 1996; Serrano et al 2005 and
Meheriuk et al 1995). Typically C 18 columns are used with two solvent systems,
typically 99 % water as solvent A and 1% formic acid and acetonitrile mixture as

solvent B.

67

For this research the results obtained by HPLC” methods were not
consistent due to some calibration and technical problems with the equipment at
the time. This experiment was substituted by liquid chromatography/mass
spectrometry (LC/MS) at the Mass Spectrometry Laboratory in the Biochemical
Department at Michigan State University. Samples were evaluated by the
technicians of that department. A Waters LCT Premier XE mass spectrometer14
from Waters Corporation in Milford Massachusetts, U.S., with a Shimadzu HPLC
from Portland, OR, US was used. Data was acquired and managed with
MassLynx® MS software. A Novapack (Waters) C18 Column was used with a
flow rate of 0.8 ml min'1 at an injection rate of 20 pl at 25 °C. The mobile phase
was HCOOH (10%) as solvent A and ethanol (50%) as solvent B. Solvent A had

a gradient from 5 to 22 % in 40 minutes. (Esti et al. 2002)

 

‘3 The central problem with the current generation of HPLC detectors is that there is a broad
trade-off between sensitivity and specificity. Highly sensitive detectors, such as those utilizing
ultra-violet detection, tend only to be able to detect a certain range of compounds, while those
detectors that can detect a broad range of compounds, such as those utilizing refractive index
detection, and tend not to be very sensitive (Detectors 2008).

14 The LCT PremierTM XE mass spectrometer delivers high sensitivity, resolution, and exact
mass (Waters 2008).Features:

. High MS resolution for the selectivity needed to separate analyte spectra from isobaric
interferences and background chemical noise
High sensitivity for achieving very low detection limits
High linear dynamic range, which allows experiments to be carried out across a range of
concentration levels

. Exact mass MS measurements, which give elemental composition information that can
be used to identify analyze compounds; you can obtain exact mass on molecular and
fragment ions, simplifying the process of spectrum interpretation

0 Versatile ionization options that apply to a range of compound classes
Versatile software options for a variety of applications

68

3.3.6. Sensory evaluation
Sensory evaluation gained extra impetus during the 1940s and 19503

thanks to the US Army which tested different food products for the military.
Sensory evaluation was then an emerging science, and several other institutions
started developing new tests to obtain more accurate results to evaluate their
food products (Stone and Sidel 2004).

Various different sensory tests can be used with food and beverages such
as descriptive analysis or quantitative measurements using nominal, ordinal,
interval, or ratio scales (Meilgaard et al 1999). Tests can also vary depending
on the number of panelists and the number of samples to be tested. The data
can be qualitative, quantitative or both (van Oirschot and Tomlins 2005).

In sensory evaluation more than fifty subjects may be used when testing
for consumer preferences. A trained panel with six to twelve subjects who have
learned to recognize the specific product characteristics and deterioration
mechanisms is another approach. For this research both types were used:
consumer and trained panels.

The consumers’ preferences were evaluated at the end of the experiment.
The consumer’s test at the end of the experiment was used to establish ranking
and preferences for the juice stored after 12 months in the two better performing
packages (glass and aluminum) compared to freshly make tart cherry juice from
concentrate. These two better performing packages were those selected by the
trained panel based on their assessment of the juice at the end of the twelve
months of storage. The consumer’s test was done in the sensory lab at Michigan

State University in the department of Food Science and Human Nutrition. It was

69

done under controlled light and setting to minimize errors. One hundred
panelists participated. The panelists were recruited via flyers posted on the
Michigan Sate University campus and via e—mail. This panelists were people that
like to drink juice. The consumer’s test had two parts: Preferences and ranking.
In the preference part they had to evaluate the fresh juice and one of the stored
juices with a hedonic scale. The second part was to rank the two juices in order
to know which one was preferred.

For the trained panel sensory evaluation, a panel of 15 subjects was
identified in a screening test for this sensory evaluation (Stone and Sidel 2004;
Esti et al 2002; Romero et al 2008 and Ross and Weller 2008). Potential
participants were asked to identify different aromas such as coffee, vanilla,
allspice, nutmeg, cloves, lemon, onion and garlic. They were asked to identify
different degrees of sourness and sweetness by ranking samples in ascendant
order for each set of samples. The selection criteria were based on the
percentage of correct answers.

Subjects who scored over 60 percent correct were selected for the panel.
Once selected, panelists were trained to identity specific tart cherry juice
characteristics such as taste, color, texture, off flavor and acidity, as part of the
overall expected quality.

For the training sessions, panelists were introduced to different forms of
cherry flavor. For taste and aroma they tasted and smelled different products
with benzaldehyde (cherry flavor). As was mentioned in the literature review,

benzaldehyde is one of the most concentrated volatile compounds in tart cherry

70

juice. Benzaldehyde is responsible for the overall cherry/almond flavor and
aroma that many commercial products use to replicate or characterize cherry
flavor (e.g. Jell-o, cherry coke, maraschino cherries). Acetaldehyde, on the other
hand, is responsible for the fresh-like aroma and taste. During the training
sessions panelists were exposed to these two compounds in different products to
learn to characterize them and later to identify quantity changes in juice.

The panel training for descriptive analysis (DA) began in March 2006,
when juice was bottled. Four months later data started to be collected. The
moderator collected the panel’s impressions in addition to their written and
individual evaluations (Stone and Sidel 1999).

Panelists were presented, at each session, with four samples (glass,
aluminum and PET bottles and stand-up pouches) in transparent cups with a
three digit random code in random order for the first two sessions and three
sample cups (glass, aluminum and PET bottles) for the rest of the sessions.

The panel evaluated color, aroma, texture, flavor, off-flavor, overall quality
and quality adequate for consumption using semantic differential scales. These
were 15-point bipolar scales with words in the extremes that are antonyms:

-“too sweet” and “not too sweet” for sweetness

- “too tart” and “not too tart” for acidity

- “too dark” and “too light” for color

- “smooth” and “grainy” for texture,

- “fresh” and “not fresh” for fresh cherry juice aroma

- “light” and “strong” for off flavor good and bad for overall quality

71

A copy of the trained panel survey is attached in the Appendix. The
results were tabulated and analyzed using Quantitative Descriptive Analysis
(QDA). QDA was used with the subjects because it provides qualitative and
quantitative data and allows for the evaluation of different characteristics.
Additionally, the four different packages can be simultaneously tested in one
session. The QDA test recommends that 6 to 12 subjects be trained on the
specific properties of the product and the objectives of the project. Not all the
panelists were available for all the meetings and a minimum of six is acceptable
therefore there were always more than six trained panelists assisting to the
sessions and six of them where consistent through the research. For this test
language needed to be simple and easy to understand to translate the results to
the objectives of the project. With the use of a graphic rating scale, such as the
semantic differential scales, the qualitative results could be interpreted as
quantitative results (Stone and Sidel 1999).

3.3.7. Statistical analysis

The laboratory data was collected monthly for a twelve month period with
three replications of each treatment. Treatment refers to the different containers
analyzed in this study (glass, aluminum, PET bottles and stand-up pouches). It
was recorded and analyzed with SPSS statistics software. A two-way analysis of
variance (ANOVA) was used to identify the difference and interaction between
the different treatments and time at a ninety-five % confidence interval as

previous researches have done (Esti et al 2002; Mattheis et al 1997).

72

For the ANOVA analysis the full model was used. If interference between
treatment and month was found (significant statistical difference in the
treatment*month analysis), then a custom model was run to avoid the
interference. The full model can be defined by the following formula:

Design =lntercept + treatment + month+(Treatment *Month)+ cl + at + em + s(t.m)

The custom model is then defined as:
Design = Intercept + treatment+ month + 3| + at + em

The p-value obtained in the two factor ANOVA test without replication
determined the significance of the results and how much evidence was found
against the null hypothesis H1 (all packages behave equally). The significance
level 0: 0.05 means that is not statistically significant at a five % confidence
level.

Correlation was done between different sets of data to understand how

they relate to each other.

73

lV. RESULTS AND DISCUSSION
The juice packed in the four different packages was tested over the course

of twelve months, and the data was analyzed. This chapter reports results
regarding the changes in color, solids, acidity, microbial growth, antioxidant
content, and flavor of juice packed glass, aluminum and PET bottles. Results for
the stand-up pouch are presented in Appendix 5.

4.1 Stand-up pouch test was compromised

Unfortunately, after eight months the stand-up pouches were disqualified
from this research because microbial growth was found (yeast) and there was
significant reduction in Brix degrees, pH, and color. Two months earlier the
trained panel had identified an off-flavor, a decrease in color, sweetness, tartness
and aroma, and they did not think the quality was adequate for consumer
purchase.

It is possible that the microbial contamination had occurred during filling
and manually closing the pouches’spout. The plant where the juice was packed
only had the capacity to pack bottles, and it did not have the special machine
needed to pack stand-up pouches in a sterile manner. The special filler grips the
inverted pouches which already have the spout and cap applied, fills them from
the bottom, flushes the headspace with nitrogen, and then they are impulse
sealed along the bottom. Another possibility was that the pouches might be
contaminated during shipping and storage.

The reason for testing the stand up pouches was due to the novelty of the

container in the beverage market segment in the United States. This type of

74

container is very popular in other countries in Europe and Latin America and is
becoming more popular in the United States.

But in our case the pre-sealed stand-up pouches had to be filled through
the small spout, held and capped by hand with no standard torque application
and without flushing the headspace with nitrogen. When handling these
containers cross-contamination may have occurred. Since the caps were
screwed by hand, unequal closure torque may have been applied, and so the
seal quality may have been compromised. These are the most probable
reasons for the failure of the package, and so the decision was made to eliminate
them from the experiment and the statistical analysis.

For the statistical analysis it was necessary for all the variables to have
the same number of data points. The stand-up pouches only had eight months
of data and so they could not be evaluated in the statistical analysis with the
other containers. Therefore, results obtained for the stand-up pouches are
offered in Appendix 5.

Only data generated during twelve months from the glass, aluminum and
PET bottles were considered for analysis. These results are presented in the

following sections.

4.2. Changes in color
Color change in tart cherry juice can be due to: (1) oxidation, (2) change in

pH, (3) physico-chemical reactions and (4) material contact. In this research L*a*
b* values were measured monthly in triplicate and hue, chrominance and AE

were calculated. Then, all the results were analyzed with ANOVA.

75

A statistically significant difference between juice transparency (L* values),
b“ values and hue was not found among in the different packages or over time.
However the a* values and chrominance were significantly different. These
results are explained in detail in the following sections.

4.2.1. Changes in transparency (L* values)

The L* value measures the luminosity or transparency and the darkness of
the juice. Changes in transparency during the twelve months of storage were not
statistically significant (Figure 26).

A change in transparency was expected because change in solids and
polymerization of anthocyanins over time has been associated in other studies
with a loss of transparency in different juices. Cloudiness also has been found
due to the presence of precipitates (Von Elbe and Schwarts 1996; de Freitas and
Mateus 2006; Hernandez et al 1999; Gomez-Cordovés 2004; Eglinton et al
2004). Additionally, differences in mouth feel can be detected by consumers
(Paul and Palmer 1972). But the results of transparency testing were similar for
the juice in all of the three packages.

With the ANOVA full model, the treatment variable resulted in a p-value of
0.21 and the month*treatment variable resulted in a p-value of 0.53 (Table 10 in
Appendix 1). Since both values were higher than o=0.05, there was no statistical
significance for either variable. Therefore it was not necessary to run an ANOVA
custom model. The changes in the juice transparency were not found to be
significantly different between packages. There was, however a significant

difference between months for the first 10 months. The last two months were not

76

significantly different with a p-value of 0.164 (see Table 50 in Appendix 2).
Although the data showed a minor difference in transparency values, this
difference was not statistically significant enough to consider a difference
between time and treatment. Hypothesis 1: The container systems will perform
equally regarding the stability of the juice with respect to transparency was not
rejected, and it was concluded that the juice in the three packages is equally

transparent through the twelve months of storage.

77

The plotted statistical means of the L* values, for each of the three
containers, followed a similar parallel pattern (see Figure 26). After the first
month the mean L*value was 10. After the second month of storage the mean L*
values for all containers increased to 15, and for the next two months there was a
steep decline. Between month 4 and 9, transparency ranged between 7.5 and 5.

All these results were significant different between months.

 

 

‘—o—Glass
l
. —-— Aluminum

I+PET

 

 

 

 

 

1 , . . E m, E ,

Figure 26. Change in transparency (L* values) of tart cherry juice stored at
room temperature (23°C, 50%RH) for twelve months.

A possible explanation for these variations may be due to the anthocyanin
polymerization due to acetaldehyde. Gemez-Plaza et al 2004, described that a

decrease on monomeric anthocyanins in wine will not be reflected in the intensity

78

or saturation of the color, but in the transparency of the wines. When monomeric
anthocyanins polymerize due to reactions with pyruvic acid or acetaldehyde, they
create precipitation of solids and variations in transparency of the wine (Saucier
et al 2004). It has been reported that acetaldehyde and benzaldehyde are
present in cherries (Iapin variety) at 59.2 x 105 ion count and 72.9 x 105 ion count
respectively (Meheriuk et al 1994). Thus the presence of acetaldehyde may
have caused polymerization of the anthocyanins which could result in
precipitation of solids.

The presence of precipitate was also noticed in the sensory evaluation by
the trained panel and the consumers. The trained panel and consumers were
able to sense some particles at the bottom of the glass and detected a less
transparent juice for each of the three different packages after the sixth month.
There is positive correlation of 0.556 between the L* values and the question

about mouth feeling to the trained panel (see Figure 27).

79

 

 

 

[:LTEIuZ?“" 7”” l
7: flamed 339?};M3U1hieefl19l

 

L‘ and trained panel mouth feellng

 

 

 

’
Figure 27. Correlation between L* values and trained panel mouth feel of
tart cherry juice stored at room temperature (23°C, 50% RH) for twelve

months.

 

4.2.2. Changes in redness (a* values)

Any variation in the bright red color associated with tart cherries is
probably a result of oxidation or changes in acidity or solids. Changes in
oxidation are not reversible and can be measured by a constant and steady drop
in redness. The red color takes on a brown hue as the pigments degrade,
oxidize and polymerize (Kearsley and Rodriguez 1981; Constela and Lozano
2005; Marquez et al 2003; Turker and Erdogdu 2005).

On the other hand, color loss due to changes in acidity is reversible
(Scaman 2005; Jiang et al 2004; Turker and Erdogdu 2005). Pigment color can
change as pH values vary. Jiang et al (2004) described how the red color of

litchi fruit was manipulated by modifying the pH.

80

In this project a full ANOVA model was run to evaluate the a* values for
the month*treatment variable. This test resulted in a significant difference (p-
value below o=0.05) which indicated an interaction between the two variables.
Next a custom model was run to compare the two variables, month and
treatment. This custom ANOVA test resulted in a statistical significance for both
the month (Table 51 in Appendix 2) and treatment variables (Table 11 and 12 in
Appendix 1). These results suggest that at least one of the three packages
behaved differently throughout the twelve months of storage. Therefore
hypothesis 1 was rejected on the basis of a* values.

The statistical means of the red a* values for all of the three containers
followed a similar trend over time as shown in Figure 29. Between the first and
third months the overall mean values declined dramatically for the juice all three
containers. During months four and five, the mean values rose and then
declined between the sixth and seventh month. In the eighth month, the mean
values rose again to values similar to those in the sixth month. Between months
ten and twelve, there was yet another decline to the lowest value recorded in the
study.

Although the mean values forjuice in both the glass and the aluminum
bottle values were almost equal throughout the duration of the study, the results
for the PET bottles were significantly lower. Correlation results between a*
values and pH or Brix degrees have negative values of (043346) and (-0.26)

respectively.

81

These results indicate that the PET bottle is the package type behaving
differently over time (Figure 29). On direct observation, the inner layer of the
bottle was found to have a red tint, which is possibly explained by migration of
the pigment to the bottle walls. Additionally, the walls of the bottle were observed
to easily delaminate when bottles were crushed. The supplier's explanation is
that the oxygen scavenger may have been consumed over time, which could
cause the ties between layers to loosen and compromise the bottle structure, as

shown in Figure 28 (Personal communication 2006).

 

Figure 28. Picture of the PET bottle delaminating with tart cherry juice
stored at room temperature (23°C, 50% Rh) after six months of storage.

The up and down pattern is not correlated to changes in the pH and solids
in the juice (Figure 29). But, the changes in a* values are correlated (r=0.46)
with the color change perceived by the trained panel during the course of the

study.

82

35

 

 

33
31 7
29

27 ‘¥Glass

._._ Aluminum
25 .—A—-—7 7 PET?
23 7

21-

 

 

 

Figure 29. Changes in a* values (redness) of tart cherry juice stored at
room temperature (23°C, 50% RH) during twelve months of storage.

4.2.3. Changes in yellowness (b* values)

Oxidation would be evidenced by large amounts of yellow in the juice
(Scaman 2005; Jiang et al 2004; Turker and Erdogdu 2005). The b* value is a
measure of the amount of yellow or blue in the tart cherry juice. No significant
difference in mean b’ values was found in any of the three packages.

The full ANOVA model showed that there was not an interaction between
the month*treatment variable for the mean b* values represented by a p-value of

0.164 (Table 13 in Appendix 1). Additionally, there was no significant difference

83

between the mean a* values of the month*treatment variables for any of three
the packages. Therefore, the hypothesis 1 was not rejected on the basis of
yellowness.

The statistical means of the b* values for each of the three containers
followed a trend that was similar to the L* values (Figure 30). At the second
month the first increase in the mean b* values was observed. The mean b*
values decreased during the following two months. At the fifth month the b*
mean values rose to similar levels as the third month. Then the most significant
decrease was found during the seventh month. From the eighth to ninth months,
the mean values rose again to values similar to those in the fifth month. Between
months ten and twelve, there was yet another decline until the values reached
levels that were similar to those at the seventh month. But, throughout the
duration of the study the mean values for the juice in three packages were almost
equal. There was significant differences between months with the exception of
month 5 and 6 and month 11 and 12, where the p-values were 0.940 and 0.307

respectively (Table 52 in Appendix 2).

84

 

‘ -—;—Glass
' —-— Aluminum
‘ +PET

 

 

 

 

6 7
Month 7

Figure 30. Changes in b* values (yellowness) of tart cherry juice stored at
room temperature (23°C, 50% RH) during twelve months of storage.

These results indicate that there were no traces of yellowness or
oxidation. The juice was moving towards a more blue pigmentation and moving
away from the yellow scale, which would have indicated oxidation (Fulcrand et al
2004). When red anthocyanins like cyanidin, pelargonidin or peonidin are co-
pigmented with tannins they tend to turn more into the blue tones and this could
be an explanation of why the tart cherry juice was turning more into the blue
tones during the length of the study (de Freitas and Mateus 2006, Gomez-
Cordoves 2004 and Eglinton et all 2004). These results were confirmed by the
trained panel which identified that the juice had a more “wine-like", color or

looked more purple or blue tending to a burgundy color, at the end of the twelve

85

months of storage. The correlation between b* values and the color question to
the trained panel has a very weak positive relationship (r= 0.055).

In Figure 31 the positive relationship within these three parameters
(L*a*b*) is shown. The correlation results are: L* and a* values r= 0.946, b*and
a* values r= 0.832 and finally L*b* values= 0.783 showing a strong positive

relationship between these three parameters.

 

 

 

 

 

2,, _ i-“ ¥i§§f___
. l i I! 2.“...
j 1: 2° _ ' --..._. TT TT _ _._____-_ 'TTTTTT T'T ___ _ T __ _ ' T lobvalues,{
, 3: f 'lavaluesii
l " 15 2.- --_- if — w —-—~ *- ~- —~—~ ——~- ._____ -«-~--—-—- --—- 2.2-2-2; IALValEegl:
‘ i i j l t ° P I
10 — — t — — —— -~— I
i x ‘ i if I i I i i l
5 -_ _2 _E _.__-_ .__.--,.__ _ - 2 , 2_ I

0' .2
0 2 4 6 3 10 12 14

Months

 

L...__,_“_....__.s_._._. c..___.___. __.2

Figure 31. Correlation between L*a*b* values of tart cherry juice stored at
room temperature (23°C, 50% RH) during twelve months of storage.

4.2.4. Changes in chrominance

Chrominance is defined as the color attributes that include hue
(predominant color) and saturation (amounts of color) (PC Magazine 2008). In
other words chrominance is the judgment of the color compared to the brightness
of the surroundings (MacDougall 2002). In this test the amount of red was

evaluated to determine the development of any traces of brown hues as

86

evidence of oxidation. In this test, the PET package had a lower performance
than the glass and aluminum bottles.

After calculating the chrominance with the formula C= (a*2 + b* 2) V”,

an
ANOVA custom model was run, and both treatment and month variables were
found to be statistically significant different with p-values below o= 0.05.
Additionally, no interaction was detected between the month*treatment variable.
This significance meant that at least one of the containers was behaving
differently, resulting in rejection of hypothesis 1 for chrominance (Table 14 in
Appendix 1). ANOVA analysis was also run for the month variable determining
that almost all the month were significantly different with the exception of month
seven (p-value 0.97) and eleven (p-value 0.26) as shown in Table 30 in Appendix
2.

The statistical means of the chrominance values for each of the three
containers followed the same trend found for the L* and b* values as shown in
Figure 32. At the end of the second month the first increase in the chrominance
mean values was observed. The mean chrominance decreased during the
following two months and at the fifth month the chrominance values rose to
similar levels as those during the third month. Then the most significant
decrease was found during the seventh month. From eight to nine months, the
mean values rose again to values similar to those in the fifth month. Between

months ten and twelve, there was yet another decline until the valued reached a

level similar to month seven. Throughout the year, the mean values for the three

87

packages values, were parallel to each other, but, the PET bottle had a lower

performance.

 

 

800
700
600
500 g H 7 7
I—o—Glass

i-u- Aluminum,
:*— , 851,}

400 « , ~

Chrominance

3003 ,

 

200 ..

 

100 ,, 277272-7722-”

 

 

 

Figure 32. Changes in chrominance of tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage.

The zigzagging pattern the curves followed can also be explained by the
fact that the pH and solids in the juice were changing over time. However there
was no evidence of oxidation, since the mean chrominance values returned to
previous values at month five, eight and nine; and there was no consistent drop
in these same values (Figure 32). These results were confirmed by the
outcomes of pH, solids and anthocyanins tests and also by the trained panel.

Remén et al (2004) reported that cherries with CO; —enriched

atmospheres have higher chrominance values than the ones with not modified

88

atmosphere and that the chrominance numbers can go up during the storage
time. This might be a reason for the changes in chrominance.

There is a positive relationship between L* (r= 0.63), a*(r=0.5542), b* (r=
0.5195) values, hue (r= 0.4706) and AE (r= 0.2848), although, the relationship
still not really strong.

4.2.5. Changes in hue

Hue represents the saturation of a specific color. Saturation is a
measurement of colorfulness compared to its own brightness and purity. A pure
color (with high hue) has only one wave length and in high amounts. Only pure
colors can achieve purity; all other colors are a combination of them (Schanda
2007). It has been reported that the color intensity of pomegranate can increase
intensity during storage thanks to the anthocyanin activity (Artés et al 2002).
This research found that the tart cherry juice became darker because the red
color turned into a more blue-red but with similar levels of saturation and
chrominance. Fulcrand et al (2004) explained that anthocyanins which are co-
pigmented with help of acetaldehyde turn wine into more of a blue-red color than
the original red. Will and Dietrich (2006) find a similar reaction between
anthocyanins and aldehydes creating a more blue-red color plum juices, and
oxidized juices showed a distinct browning tendency. Freitas and Mateus (2006)
describe how the interaction between acetaldehyde and benzaldehyde were
responsible for the blue-red wine color. Bauer-Christoph et al (1997) reported
2.6 mg/100 ml of benzaldehyde in cherries, Mattheis et al (1997) explained how

benzaldehyde results from the hydrolysis of amygdalins contained in the cherry

89

pits. Finally, Girard and Kopp (1998) characterized volatile constituents in
cherries finding acetaldehyde and benzaldehyde present in cherries. Since
anthocyanins and acetaldehyde and benzaldehyde are present in the juice it can
explain the change from bright to a blue-red color. No browning was detected;
therefore it can be concluded that probably there no oxidation or very slow
oxidation rates were happening to the stored tart cherry juice.

In this project a full ANOVA model was run to evaluate the hue values for
the month*treatment variable. This test resulted in a significant difference which
indicated an interaction between the two variables. Then a custom model was
run between the two variables, month and treatment. This custom ANOVA test
resulted in no statistical significance for the treatment variables. There was not
enough evidence to reject the hypothesis 1 on the basis of hue.

There was no statistical significant difference been containers in the hue
measurements meaning that the juice in the three containers behaves similarly
and had similar hue results (Table 16 in Appendix 1). Freitas and Mateus (2006)
explained that the nature of the sugars (glucose, arabinos, rutinose) are also
important to influence the hue pigments. But, the packaging material is not an
influence to promote these changes since the three different materials have
similar behavior. Therefore the ANOVA results for the month variable were not
significantly different for months four (p-value 0.214), seven (p-value 0.141), ten
(p-value 0.293) and eleven (p-values 0.817) as shown in Table 31 in Appendix 2.

The statistical means of the resulting hue values for each of the three

containers (Figure 33) followed a trend similar to the L* and chrominance values.

90

At the second month the first increase in mean hue value was observed. The
mean hue values decreased during the following two months and at the fifth
month the mean hue values rose to levels similar to the third month. The most
significant decrease was found during the seventh month. From the eighth to
ninth months, the mean values rose again to values similar to those in the fifth
month. Between months ten and twelve, there was yet another decline until the
values reached a level similar to those at the seventh month. Throughout the
duration of the study, the mean values for the glass and aluminum bottles were

almost equal to each other (Figure 33).

l, 750 _ ..... .... m... .....-. . ...._ .. ......._ . ....._

 

650 L

550

I+Glass
E 450 —I— Aluminum
+, PET ,

350 E7

 

250 *

 

 

 

150 T

 

‘ Month
l . ,, ..

Figure 33. Changes in hue of tart cherryjuice stored at room temperature
(23°C, 50% RH) during twelve months of storage.

91

Two possible reasons for changes in hue are: changes in anthocyanins
sugar moieties and changes in acidity, both of which will be discussed later in
this chapter.

The trained panel and consumers also noticed that the color was darker
but no brown hues were identified. They were more inclined to describe the juice
as blue-red.

4.2.6. Color change (AE)

Color change (AE) is expressed as the correlates of lightness (L*values),
hue and chrominance (McDougall, 2002). Color change is a relationship
between these three components and gives a general idea of how the color has
changed through time. In all the data from L*a*b* values, hue and chrominance
it can be observed that the lower point in the data set was always at month
seven. In AE month seven represents the one of the highest amount of change
in the color of the juice (Figure 34). AE has a positive relationship with
correlation of L* (r=0.34), a*(r=0.266), b* (r= 0.198), Chrominance (r=0.28) and
finally with hue (r=0.22). The strongest correlation of the change is with regard of

the luminosity or transparency that suffers changes during the study.

92

 

 

* _:_ "as; *
‘ —-— Aluminum ,
- 1+.,,-I?ET- ‘

 

 

 

 

 

Figure 34. Color change (AE) of tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage.

4.3. Changes in Solids

Sugar moieties can change forms through time, and it has been found that
antioxidants play a very important role in changing soluble sugars. Sugars are
reorganized and attach to the antioxidants under different types of glycosils
(glucose, arabinose, rutinose) (Freitas and Mateus 2006). Hernandez et al (1999)
described how sugars in pomegranate juice changed during storage and
attributed these changes to the degradation in pectin. When pectin degrades

soluble sugars can increase. Von Elbe and Schwartz (1996) describe how

93

fructose, arabinose, lactose and sorbose at low concentrations are less stable
than glucose, sucrose and maltose, and they can change over time.

The soluble solids in the tart cherry juice changed throughout the twelve
months of storage, and results show a significant difference between containers
and time (Table 16 and 17 in Appendix 1 and Table 33 in Appendix 2).

With the full ANOVA model the variable month*treatment had a p-value
below o=0.05, meaning there was a significant difference and the hypothesis 1
was rejected on the basis of solids (Table 34 in Appendix 2). Then a custom
model was applied and the treatment was found to have a p-value below o=0.05,
and month of 0.193 (Table 35); again the null hypothesis was rejected on the
basis of solids.

The results from each container followed a similar trend, but a different
level, as shown in Figure 35. The juice in the PET bottles had a higher amount
of solids, in the range of 15.90 to 16.20 degrees Brix. The second highest was
the juice in aluminum bottles which had values between 15.30 and 15.60
degrees Brix. Finally the juice in the glass containers was between 15 and 15.30
degrees Brix. The three containers were in a range of .30 degrees Brix. The
largest decrease in Brix for the three containers was at the sixth month, but it

then rose back to the original levels at the ninth month (Figure 35).

94

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

,
BrixChanges
l
l
I 17.5
‘ 17 ‘,__ __, ____. f'?__,_.:P_ _.-P. — .i-m -————!— — —q'_‘TT ’T'T’;:"”‘-P"'
.. .. 1' -F
16.5 ~ —— 12"» -,_.-_ —-J~—~~’--T~~:-- r4
; . w ‘l‘ .. -II- VI -‘ -— _
' g 15 12 /i=-4- s _ .2 2V5 {—o—Glass
l g} EN 4 i\‘ ;_-— Aluminum
.— 1"
' .>< 155 - —- “It—'4“ w». =‘x—WJ‘MfiETI' +PET
C /
m L ., / \K _T‘” ‘
I—-1 :31
15 E---fi- r_.2_2 -22 T—IT -fi__-
14.5 .. 2 l - _- ___,‘H,__,.__ I — i? -- T —--: -_._,
fl 1 .. .L ._
14 . .. . " . .
123456789101112
Month

 

Figure 35. Changes in solids of tart cherry juice stored at room temperature
(23°C, 50% RH) during twelve months of storage.

The changes in solids overtime were similar to the changes in color which
were discussed above, and with the changes taking place with the anthocyanins,
which will be discussed later. However, there is negative statistical correlation
between the different color parameters and the changes in Brix degrees.

It could be concluded that sugars were suffering changes in their sugar
moieties, since they were going up and down in a zigzag - type curve (Paul and
Palmer 1972). Von Elbe and Schwartz (1996) described that cyanidin 3-

qucosyl-rutinoside is less stable and shorter life than cyanidin 3-rutinoside.

95

Freitas and Mateus (2006) in their description of physical-chemical properties
explain that anthocyanins can have self-association or co-pigmentation resulting
in different pH values and different sugars attached to the glycosil. The trained
panel confirmed these results. They found that the tart cherry juice was
becoming less tart and more sweet through time, but the correlation with the Brix
degree show a weak relationship between the variables (Brix degrees and

sweetness r= 0.169 and Brix degrees and tartness F 0.138).

4.4. Changes in pH

The pH levels of the tart cherry juiced also changed during the twelve
months of storage, but the tart cherry juice packed in the three containers
behaved similarly over time.

As with the other tests, the pH data was analyzed with a full ANOVA
model. The p-value for treatment*month was 0.170, greater than 0.05, and
provided no statistical evidence to reject the hypothesis 1 in the basis of pH
levels (Table 18 in Appendix 1). Therefore, the changes in pH levels were not
statistically different for the three different containers (Table 18 in Appendix 1).
For the month variable a consistent p-value for every month below o=0.05 was
found when running the ANOVA model (Table 34 in Appendix 2). This statistical
significance gives enough evidence to reject hypothesis 1 on the basis of the pH
levels.

The data on acidity shows that the containers were behaving similarly,
since the curves were very close to each other. But the changes over time were

more dramatic. At the end of three months of storage, the tart cherry juice

96

suffered the most dramatic decrease in pH. At six months the pH value
rebounded to almost the same values found at the beginning of the experiment,
and at nine months the pH increased to the highest level during the evaluation
time.

This trend was similar to the redness values and the solids curves
providing a clue to the relationship between the three characteristics in the
behavior of the tart cherry juice (Figure 36). Even though, the changes in the pH
levels are not significantly different, it cannot be ruled out that change in the pH
levels contributed to the changes in color. The relationship ofanthocyanins with
other components is very complex and difficult to describe in single parameters.
The changes in pH could benefit the co-pigmentation of anthocyanins and
change the colors (Will and Dietrich 2006, Freitas and Mateus 2006). Since
there is not a significant difference between the different containers it can be
inferred that the contact material plays no role in promoting the changes in pH

levels.

97

 

_. _. -_
j —o—- Glass '|
l —-— Aluminum l

 

 

 

 

 

 

 

 

 

 

 

 

Figure 36. pH changes of tart cherry juice stored at room temperature
(23°C, 50% RH) during twelve months of storage.

4.5. Changes in microbial growth
During the twelve months of storage no microbial growth occurred in any

of the three bottles. High acid foods like tart cherry juice inhibit mold and yeast
growth, and anaerobic environments like that in the sealed packages inhibit the
pathogen growth most likely to occur in tart cherry juice. Since the pH level was
always below 4.6 the tart cherry juice was considered a high acid food. The hot
fill process plus the nitrogen flush done before capping the bottles were good
methods for keeping the headspace oxygen-free. This not only created the
correct anaerobic environment where most of the aerobic bacteria could grow,

but also helped to prevent oxidation of the juice, as shown in the color evaluation.

98

The plates with glass, aluminum and PET bottle samples were always clean, with
no traces of microbial growth.

However, the plates with samples from the stand-up pouch had yeast and
mold beginning at the end of the six months. Probably this microbial growth
started before but only appeared at the end of the end of the sixth month of
storage. After eight months, stand-up pouch was suspended. This is the reason
that this container was not included in the analysis. As mentioned previously the
standing pouches could have been compromised by the lack of technology to
process this type of packages, the lack of ability to flush with nitrogen and the

possibility of being previously contaminated.

4.6. Changes in anthocyanin compounds

Anthocyanin content decreased in general during the twelve months of
storage, confirming findings by Freitas and Mateus (2006), Will and Dietrich
(2006) and Gemez-Cordovés (2004). The glass bottles retained the anthocyanin
content of the tart cherry juice best, followed by the aluminum and finally the PET
bottles. Changes in the anthocyanin content were correlated with other changes
in color, solids and pH noted previously.

The anthocyanins most present in fresh tart cherries were expected to be:
cyanidin 3-glucoside, cyanidin 3-rutinoside, peonidin 3-glucoside, peonidin 3-
rutinoside and pelargonidin 3-rutinoside with the largest concentration of cyanidin
3-rutinoside (Esti et al, 2002). Kim et al (2000) reported the presence of cyanidin

3-glucoside, cyanidin 3-rutinoside and peonidin 3-rutinoside in tart cherries.

99

Wang et al (1997) reported cyanidin and peonidin in tart cherries. Hong and
Wrolstad (1990) reported only the presence of cyanidin in cherries.

In this research the tart cherry juice was found to contain the following
anthocyanins: cyanidin 3-glucosylrutinoside, cyanidin 3-rutinoside, delphinidin 3-
glucosylrutinoside, delphinidin 3-rutinoside, peonidin 3-rutinoside, pelargonidin 3-
rutinoside and petunidin 3-rutinoside. Cyanidin 3-glucosylrutinoside was the one
found in the highest concentration in the tart cherry juice at the beginning of the
experiment.

Cyanidin is responsible for the crimson color, pelargonidin for the scarlet-
orange tones at low pH and the peonidin for the cherry red at low pH (Paul and
Palmer 1972). The following section analyzes the levels of these anthocyanins

present in the juice over time.

4.6.1 Changes in cyanidin

Cyanidin 3-glucosylrutinoside was the anthocyanin with the highest initial
concentration in the juice but the levels fell dramatically over time as reported by
Esti et al (2002); Kim et al (2000); Hong and Wrolstad (1990) and Wang et al
(1997). Cyanidin 3—glucosylrutinoside had a retention time (RT) at 7.90 minutes.
The glass bottles were found to preserve higher levels throughout the duration of
the project, followed by aluminum and finally by the PET bottles, as shown in
Figure 37.

There was an average decrease of 26 % for the three different packages
during the twelve months of the experiment. The p-values for treatment and

month alone were below o=0.05 for most of the entries suggesting that at least

100

one of the packages did not behave as the other two (Table 19 in Appendix 1
and Table 35 in Appendix 2). Although, cyanidin 3-glycosilrutinoside decline was
similar for all three containers, PET bottles retained the lowest amounts during

the twelve months of the experiment.

 

2500

2000

I —o—Glass
—-— Aluminu
+ PET

1500

Area

1000

500 ,

 

 

 

 

l

i O 1 2 3 4 5 6 7 8 9 10
Months

I MM,“ __,_ __,,, .2,

Figure 37. Changes in cyanidin 3-glucosylrutinoside in tart cherry juice
stored at room temperature (23°C, 50% RH) during twelve months of
storage.

On the other hand, as cyanidin 3-glucosylrutinoside decreased, cyanidin
3~rutinoside (RT 7.98 minutes) increased by 33 percent as shown in Figure 38,
which was seven percent more than the decrease In cyanidin 3-

glucosyrutinoside.

101

These changes can be explained by the theory that over time antioxidants
will change their glycosil attachments and change their attachment to different
sugar moieties soluble in the juice (Von Elbe and Schwartz 1996 and Freitas and
Mateus 2006). It can be assumed that the cyanidin lost the glucose moiety
attachment and retained the rutinose moiety.

The ANOVA analysis of the mean results for cyanidin 3-glucosylrutinoside
showed that there was no interaction between month and treatment with a p-
value of 0.190 (Table 20 Appendix 1). Therefore, treatment and month
separately had both p-values below 0: 0.05, which provides enough evidence to
reject the hypothesis 1, because the juice in PET had consistently the lowest
values of both kinds of cyanidin retention. The ANOVA analysis for each month
resulted in a significant difference between the second, third and fourth month.
All the other month came out with a p-value higher than 0.05 (Table 36 Appendix

2).

102

 

220

 

z—oi—Glassr
j—I— Aluminum.
+ PET ‘

Area

 

 

 

20

 

 

Figure 38. Changes cyanidin 3-rutinoside in tart cherry juice stored at room
temperature (23°C, 50% RH) during twelve months of storage..

Correlation analysis between the two types of cyanidin was done with
strong negative relationship between the two of them (r=-0.56), meaning that
when one was declining the other one was increasing, confirming the possibility

of changes in the sugar moieties described previously, and shown in Figure 39.

103

 

 

2500 __
2000+“; - ' 5 * - - - - 5 A ,- - -____-,-_--_-,,__-,

1500*”“7”f {TT T , TTi

1000*“""”“_“'”' ___-_______ g
l

50.21%: Lug—HT“

! .
0 2 4 6 1 2

 

 

 

:e- Tyanidin 3-rutfiioside I
g- Cyanidin 3-glucosyrutinoside;

 

 

 

 

Figure 39. Correlation between the two types of cyanidin in tart cherry
juice stored at room temperature (23°C, 50% RH) during twelve months of
storage.

4.6.2. Changes in delphinidin

For this compound the juice in all three containers behaved
similarly. Delphinidin has the same two glycosil attachments (3-glucosilrutinoside
and 3-rutinoside) as cyanidin, and had a similar behavior as well.

Delphinidin causes the blue-red tones in wines such as cabernet
sauvignon, creating a more burgundy hue. This is different from the bright
crimson caused by cyanidin. Contrarily to cyanidin, delphinidin 3-
glucosylrutinoside was found in lower quantities than delphinidin 3-rutinoside,
and while one decreased the other one increased in ways similar to the cyanidin
(Figures 38 and 39). In the case of this compound, with both of the glycosil
residues, there was no statistical significance between packages over time, and

so the null hypothesis cannot be rejected. The p-values for month*treatment

104

were 0.447, and 0.653 for the treatment alone, as shown in Table 21 in Appendix
1. The ANOVA results for the month variable show that there was no statistically
significant difference between months with p-values below a: 0.05 as shown in
Table 38 in Appendix 2.

Delphinidin 3-glucosylrutinoside increased by 68% during the first four
months but then had a drastic loss of forty eight percent at the fourth month.
From the sixth month it increased steadily to 46 % of the total value and stayed

at similar levels to those at which it started (Figure 40).

200

 

180
160 ,

140 -

. —0— Glass

120 ~‘ —-— Aluminum

Area (mg/100ml)

100

80

 

60*

 

4O . . , , , l

 

 

Figure 40. Changes in delphinidin 3-glucosylrutinoside in tart cherry juice
stored at room temperature (23°C, 50% RH) during twelve months of
storage.

105

Delphinidin 3-rutinoside, on the other hand, was the compound with higher
concentration in the delphinidin family, and its behavior was equally dramatic. In
the first five months the compound almost doubled. At the sixth month a
considerable drop was apparent when approximately 50 percent of the
compound decreased. At the seventh month the compound regained about 50
percent decreased previously, and at the end of the storage time with a total
decrease of 25 percent (Figure 41).

The results of the ANOVA full model analysis showed that there was no
statistical significance between treatment*month and treatment alone, showing
that for this compound the packaging material was not a reason for the changes
in concentration (Table 22 in Appendix 1). The p-values obtained for treatment
variable were 0.933 and 0.967 for the month*treatment (Table 22 in Appendix 1).
For the month variable the ANOVA gave as results a significant difference for the
first two months with p-values below a = 0.05. For the other months all the p-
values obtained were above 0.05 (Table 38 in Appendix 2).

The higher levels of delphinidin 3-rutinoside at the end of the storage time
can also explain why the color of the tart cherry juice changed to a more
burgundy (wine-like color) compared to the bright crimson color of the fresh juice.
Like cyanidin, delphinidin 3-glucosilrutinoside lost the glucose moiety and kept

the rutinose.

106

 

 

 

 

 

 

 

450» ....... 2-.:—AW. 2.2-7-2--. .

400
a Ila—"T alas; '

I l

2 350 ‘—-— Aluminum.‘
< l—t—PET

300 »

250

200

 

 

 

 

Month

 

 

 

Figure 41. Changes in delphinidin 3-rutinoside in tart cherry juice stored at
room temperature (23°C, 50% RH) during twelve months of storage.

There is a negative relationship between delphinidin 3—glucosylrutinoside
and delphinidin 3-rutinoside (r= -0.56) showing that when one decreased the

other one increased correlated (Figure 42).

107

 

 

100— __f _- ~ E--- —-—-—— ~——— __- _,‘_,__,-_ -* -

400 ,- 2 e A A E, 2.2 V- 2.-_- _- a -2-—2.-.- -- -2 2
350 +—— _r r---_____._-__r,__ _~_.~____-__._E. E
300 ~— A - w- “—2. — «H --_ .- s. -_ c --..-M., “2.22.2
250 «r-—-—- «I ._____-___k M.“ 2-2.2 , 2-12.2-2 2,..- -_
200 -. _. - .2... -- ..._. ._ _ ___ _ ._ __ ._._ ___ _w __, _ ,

1501.- 2 _ 2 2222—- —»— - v u#— ~— ———~-—7——-~~—5

50 E...__.._,_w ..,._.__.E ._...E..____. , _. ._.. __ __ ._ *._______-__._._____-

 

 

 

l
l
|
I
|
l
l

 

To Delphinidin 3-glucosilrutino3idfle .:
i, I Delphinidin 3-rutinoside ’

 

Figure 42. Correlation between the two types of delphinidin in tart cherry
juice stored at room temperature (23°C, 50% RH) during twelve months of

storage.

4.6.3. Changes in peonidin 3-rutinoside

Peonidin is a blue pigment that can be found in flowers like morning

glories and in blueberries. The juice in all three packages behaved similarly with

respect to peonidin 3-rutinoside. The outcome of the ANOVA full model analysis

found that there was no statistical significance between the different containers

(Table 23 in Appendix 1). The p-value for the treatment*month interaction was

1.000 and 0.860 for the treatment. The p-value for the month variable was all

superior to 0.05 meaning that there was not significant difference between

months (Table 39 in Appendix 2).

108

 

The peonidin values increased dramatically after the forth month. This
result correlates with the drop in redness (b* value) in the fourth month. In the
next month it oscillated between 100,000 and 125,000 mg/100 g, maintaining a

blue tint or a wine-type color for the juice (Figure 43).

 

1207*

100

—o— Glass
80 ~

Area

1 —-— Aluminum‘
+7 .- PET
60 ~ ~ ~

 

40

20

 

 

 

Figure 43. Changes in peonidin 3-rutinoside in tart cherry juice stored at
room temperature (23°C, 50% RH) during twelve months of storage.

4.6.4. Changes in pelargonidin 3-rutinoside

Glass bottles had better performance maintaining the pelargonidin

compound in the tart cherry juice, followed by aluminum bottles and finally the

109

PET bottles. Pelargonidin is responsible for the orange tones in the juice, similar
to red geraniums, red strawberries and tart cherries.

The changes in this compound were statistical significant. Treatment
(different containers) was also significantly different, since the month*treatment
interaction had a p-value of .001 in the full model (Table 24 in Appendix 1). In
the custom model both month and treatment had p-values below o= 0.05
implying that they are statistically different and that at least one container were
behaving differently to the others (Table 25 in Appendix 1). The p-values
obtained for the month variable were above 0.05 resulting as no significant
difference between months (Table 40 in Appendix 2).

The glass bottles best maintained pelargonidin in the tart cherry juice,
followed by aluminum bottles and the PET bottles shown in Figure 44. At the
end of the fifth month, the compound tripled for the juice in three containers and
oscillated between 125 mg/100ml and 175 mg/100ml until the end of the
experiment (Figure 44). Pelargonidin followed the same trend as the other

anthocyanins in that the compound values increased in the fifth month.

110

 

250 l

 

 

 

l
200
I
150 . . -
m ;' I—o—Glass
l
3?) —- Aluminum,
I .-.__PET .
100 , , ’ " ' - '
50
0 . . . i
0 1 2 3 4 5 6 7 8 9 10
Month

 

Figure 44. Changes in pelargonidin 3-rutinoside in tart cherry juice stored
at room temperature (23°C, 50% RH) during twelve months of storage.

4.6.5. Changes in petunidin 3-rutinoside

The three different containers behaved similarly for retaining the petunidin
compound. Petunidin 3-rutinoside, as pelargonidin 3-rutinoside, peonidin 3—
rutinoside and delphinidin 3-rutinoside, increased by one 100 percent in the fifth
month and stabilized at the eighth month (Figure 45). Petunidin provides a bluish
tint to the tart cherry juice, and its increase correlates with the color changes.
The juice turned to a more wine-like color due to the increase of the blue
pigments and the decrease in the red color.

This compound was not found to be affected by the type of package, since

the differences between packages was not statistically significant. The p-values

111

obtained with the ANOVA full model were 0.116 for month*treatment and 0.505

for treatment (Table 26 in Appendix 1).

 

 

WWW-m“; ' g. A,“ f.

Area

400 .er _. w __.____ N 7% i_i .i. ’ >_____k ____-__,,,v,_fi __ r-,v,, ___,J," -_ _‘_..

300 sh.- ._—__....._._~_-._ _ __7 _-._.._w __ w...__. w---“ _w _-_..-__._ fl.» _- ___-..

200 _ _ ..M__ ___ _i_ _ v- , i- i ,,._______.___._ -..__._ “___,I.’ ________f ”___-.e M, _

100 s — —* 7 ~ *-w ,.__ ---—-7 _.. —. ___ ..___ ___--- ._.--_. be ,.

 

 

 

i
l
l
l
l
i
i
l
l
l

500 P ~-_—~_ /' 7 i W H 1 ' _ . H ' " ”"5-I- Aluminumlj
r—a—PETfl "

 

Figure 45. Changes in petunidin 3-rutinoside in tart cherry juice stored at
room temperature (23°C, 50% RH) during twelve months of storage.

4.6.6. Anthocyanin result summary

As it can be observed all the decreases or increases in anthocyanin
content began to happen after two months of storage. This confirms Will and
Dietrich (2006) findings that anthocyanin changes occurred during storage and
more drastically after six months of storage. They describe that this change
might be caused by other reactions happening in the juice. They also explain
how new anthocyanins and anthocyanins attachments are created during aging

or storage due to polymerization.

112

Changes in color were noticeable, but very little traces of oxidation are
present since the anthocyanins were not eliminated or used up in free radical
reactions. In the color section it was discussed how the color of the juice
changed to a more blue or wine-like color, and this phenomenon is consistent
with the changes in anthocyanins. Cyanidin was responsible for the total
decrease of bright red color (considering cyanidin 3-glucosylrutinoside and
cyanidin 3-rutinoside) which decreased by seventy five "/0, retaining only twenty
five °/o of the original combined amount. Both types of delphinidin, peonidin 3-
rutinoside, pelargonidin 3-rutinoside and petunidin 3-rutinoside significantly
increased in the fifth month. Delphinidin combined increased by 39 percent;
peonidin 3-rutinoside and pelargonidin 3-rutinoside increased 500 percent.
Finally, petunidin was found to have a 200 percent increased.

Since the red-orange colored anthocyanins are less stable than blue-
purple ones, and both categories are present in tart cherry juice, this can explain
the shift from bright red from fresh juice to a blue-red color of the stored juice
(Von Elbe and Schwartz 1996). After averaging, red and blue anthocyanins they
were correlated with an r-value of 0.07 and as shown in Figure 46 when the red

decreased the blue increase.

113

 

 

1600

 

1400 .,._____, t _.,_ a -m- .. ------_ “___ «_WWNWE
1200 . A —- - - , f a A - a - a - A -- - ._-__—_-..-_-.z,---___-

1000 -»~ __.-,_g_ H-.. ---_V_-,_..-_ batman”? W ‘
5‘ l
m rut—"m: —— my
2 800 4- a -_,__ ~ _ _ - z 7 as _ --_. -__..,_.__-_,___,____,,__i j—o—Red pigments l.
< lti-Jueplgmentslt

l
l
l
l

l
l
400 «A «-——~_: ~—4 ~-- AH—Ws“ r '*‘ ”'/“"*‘"““"’“ """”“ ‘T '

/ i z
/ i i

 

 

 

__ ___—___._.______ _ __i

200 - --__ — —-— __e< / -— -
\ M

y *§

4

 

 

 

 

t __ ,fl ,_ ___g_____.____ .,__~_.________,__ - _. ___._ __ -.__ .___.._ .__ __ ___. A_l

Figure 46. Correlation between red and blue anthocyanins found in tart
cherry juice stored at room temperature (23°C, 50% RH) during twelve
months of storage.

Anthocyanins can polymerize or co-pigment in the presence of
acetaldehyde or benzaldehyde. Both volatile compounds are present in tart
cherry juice (Freitas and Mateus 2006, Gomez-Cordoves 2004) and they will
develop a blue-red tint in juices containing anthocyanins. This polymerization is
then supported by changes in the sugar moieties attached to the different
anthocyanins.

Not all of the anthocyanins analyzed had a significant difference related to

the packaging material. Only cyanidin and pelargonidin changes were correlated

114

by the different packaging materials used in the experiment. And, both of these
are responsible for the bright red tones. The other anthocyanins which are
responsible for the blue tones in the juice did not have an interaction with the
packaging materials.

The amount of total anthocyanins at zero, six and twelve months is presented in

Figure 47. There is no statistical difference between the different times.

 

 

3500
3000 , ,
2500 , ,

lid Months, ‘
‘I6 Months ,
,g 12 Months

2000

Area

1500 —»

1000 <

500* ,

 

 

 

Glass Aluminum PET

 

 

 

Figure 47. Accumulation of all anthocyanins in tart cherry juice stored at
room temperature (23°C, 50% RH) during twelve months of storage.

4.7. Trained panel results
The trained panel stated that there was not significant difference and that

the overall quality of the juice was good for consumption throughout the year.
The color was always perceived as good, as long as it was not compared with a
fresh sample. The aroma was still perceived as “fresh" at the end of the twelve
months of storage. The panel also agreed that the sweetness increased through
time and the tartness decreased. There is a positive correlation between the

changes in the sweetness and tartness perception with r= 0.44 (Figure 48).

 

Sweetness and tartness perceived by trained panel

1 2 _ _-_ -..-_ _.-,...._.....___._. _ z... ._.-______-~..

 

 

 

l
l
l
l
l
l
10 .- -_-.—.- —_———_--___—————_— _1 l
l
|
l
l
l

{fie—Trained panel Tamess in the juicec 5
l I Trained panel sweetness in the juice P1

 

 

 

 

Figure 48. Correlation between sweetness and tartness perceived by the
trained panel in tart cherry juice stored at room temperature (23°C, 50% RH)
during twelve months of storage.

The trained panel was asked at each session, in different order of
samples, about the same characteristics: color, aroma, mouth feel, sweetness,

tartness, cherry flavor and overall quality. A copy of the survey is attached in

Appendix 3.

4.7.1. Color perception
At the end of twelve months of storage, despite the changes in color, the

trained panel agreed that the juice color remained pleasant and not too dark.
First, the panel evaluated the color of the juice, where 0 was too light and 15 was
too dark. They were asked to evaluate the juice by color and to compare it with
the fresh juice on which they had been trained.

The p-value obtained for month and treatment were below the o=0.05,
meaning that there was a significant difference and enough evidence to reject

hypothesis 1. The three containers behaved differently, with aluminum being the

116

option that best preserved color, followed by glass and then PET bottles from the

trained panel observation (Table 8). Since this is a panel perception (qualitative

variable) it can be different from the analytical results that are exact

measurements. The analytical results showed that glass maintained better color

followed by aluminum and PET was the last one.

Table 8. Color perception by trained panel of tart cherry juice stored at

room temperature Q3°C, 50% RH) for twelve months.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B Std. Error t P-value Sig
August-06 2.3810 0.8120 2.9323 0.0041 yes
October-06 1 .0648 0.7623 1.3968 0.1652 no
November-06 -0.4278 0.7623 -0.561 1 0.5758 no
Januag—06 0.6429 0.8120 0.7917 0.4302 no
February-07 0.0000 . . .
Glass 1.9325 0.6076 3.1805 0.0019 yes
Aluminum 2.4950 0.6076 4.1062 0.0001 yes
PET 0.0000 0.6076 3.2436 0.0005 yes

 

 

4.7.2. Aroma perception

The trained panel found that the aroma of the tart cherry juice softened
over time. The second question regarded the aroma of the juice. The panelists
were asked to sniff the samples and to evaluate the aroma of the juice on a 15-
point scale, where 0 was too soft and 15 too strong.

The p-values for month and treatment were below the o=0.05 when
running an ANOVA full model analysis. Both numbers were smaller than a: 0.05
which provides enough evidence to reject hypothesis 1. At least one of the
packages behaved differently than the others.

Glass and aluminum had very

similar curve trends and numbers. But the PET bottles retained less of the

117

aroma compounds in the cherry juice and might be due to the polarity difference
between PET and the aroma compounds.

The aroma began in the strong range and remained there only decreasing
to a medium level at the end of storage for all 3 packages. Even after a year it
was still in the higher end of medium, and so the juice was not bland or without
any aroma (Table 9).

Table 9. Aroma perception by trained panel of tart cherry juice stored at
room temperature (23°C, 50% Rijor twelve months.

 

 

 

 

 

 

 

 

 

8 Std. Error t P-value Sig
Aug ust—06 3.2810 0.7653 2.9870 0.0003 yes
October-06 1 .0988 0.8764 1 .4569 0.0020 yes
November-06 0.3456 0.9654 0.4999 0.0499 yes
Januaq-OB 0.6429 0.7659 0.8760 0.0444 yes
February-07 0.0043 . .
Glass 1.8970 0.6076 3.4561 0.0017 yes
Aluminum 2.6543 0.6076 2.0987 0.0002 yes
PET 0.0000 0.6076 3.3323 0.0003 yes

 

 

 

 

 

 

 

 

4.7.3. Mouth feeling perception

The panelists found that at the end of the twelve month of storage that the
juice had some precipitate or sediments at the bottom of the glass similar to
those found in wine.

Feeling in the mouth was used to indicate the smoothness or graininess of
the juice. Clarified (juice with no pulp) tart cherry was used for all of the
experiments, therefore residues formed at the end of storage due to the changes
in antioxidants, pH and solids. The main reason for the sediment in the juice is

the polymerization of the antioxidants (Alper et al 2004).

118

There was no statistically significant evidence to reject hypothesis 1, since
the p-values for time and treatment were 0.692 and 0.874 respectively. These
numbers provide enough confidence to assume that all three packages behaved
similarly and that the observed changes were not different statistically.

The graininess trend for juice in the three containers followed a very
similar pattern with one exception: the graininess increased greatly in the
aluminum bottles during the ninth month of storage, only to drop below the others
by the end of the experiment. However, the increase in panel perception results
was not statistically significant (Table 10). Since these are perceptions from the
trained panel it might be the reason for the variation.

Table 10. Mouth feel perception by trained panel of tart cherry juice stored
at room temperature (23°C, 50% RH) for twelve months.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B Std. Error t P-value fig
August-06 0.1399 0.9952 0.1406 0.0490 yes
October-06 0.1319 0.9344 0.1412 0.0089 yes
November-06 -0.7292 0.9344 -0.7804 0.4368 no
January-06 -0.9435 0.9952 -0.9480 0.0340 yes
February-07 0 . . .
Glass -0.3187 0.7448 -0.4280 0.0007 yes
Aluminum -0.3500 0.7448 -0.4700 0.0007 Yes
PET 0.0000 0.7448 -0.4700 0.0007 .yes

 

 

It is interesting to observe that the values for the mouth feel always
remained at upper levels of graininess, and the juice was not considered smooth,
but not considered unpleasant to the taste, it was more referred as “body” like in

wine.

119

4.7.4. Sweetness perception

Predictably for tart juice, the trained panel found that the juice was in the
“not sweet” ranges. They also found that the sweetness levels changed through
time. At the end of the experiment, the juice was perceived as less sweet than
the fresh tart cherry juice.

One of the major constraints of tart cherries is the low degree of
sweetness, with which some consumers are not familiar. All major juices in the
American market tend to be sweet (some think overly sweet) to please
consumers. The juice used in this research was 100% tart cherry juice, with no
sweetener additive or sugar, which is why this question was critical.

When the ANOVA full model statistical analysis was run with the data
obtained from the trained panel, a p-value of 0.431 for treatment and 0.046 for
month were found. These results reveal that there was no perceptible difference
between juices packed in the different packaging materials, and that all three
packages maintained a similar pattern of behavior for the sweetness of the juice.

It can be observed that the behavior of the three packages appears close
to each other in range of sweetness of the scale but through time the perceived
sweetness of the juice decreased (Table 11) since it started in the scale, at ten
and ended at six.

These results can be correlated with the changes observed in the changes
of the solids and in the anthocyanins, especially for cyanidin and delphinidin
which changed their glycosil attachment after five months of storage and can

result in different perception than with fresh tart cherry juice.

120

Table 11. Sweetness perception by trained panel of tart cherry juice stored
at room temperature (23°C, 50% RH)

for twelve months.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B Std. Error t P-value SiL
August-06 2.0893 0.8269 2.5268 0.0129 yes
October-06 0.4491 0.7763 0.5785 0.5641 no
November-06 -0. 1620 0.7763 -0.2087 0.8350 no
January-06 1.1012 0.8269 1.3318 0.1856 no
February-07 0.0000 0.8269 1.3318 0.1856 no
Glass -0.7250 0.6188 -1.1717 0.2438 no
Aluminum -0.0187 0.6188 -0.0303 0.9759 no
PET 0.0000 0.6188 -0.0303 0.9759 no

 

 

4.7.5. Tartness perception
The trained panel found that the tart cherry juice became less tart by the

end of the experiment.

Similar to the sweetness of the juice, it was important to measure the
degree of tartness of the juice for reasons of consumer preferences. If the juice
was too tart consumers would not buy it nor find it pleasurable to drink. The right
balance between sweetness and tartness if desirable since this activates more
gustative cells in the tongue and mouth.

ANOVA results show that the treatment had a p-value of 0.439 and month
of 0.335. These values were not significantly different and there was not enough
evidence to reject the null hypothesis: all three packages behaved similarly with
regards to the tartness in the juice over time. The pH results were similar to
these findings, since the packaging materials behaved similarly and did not
interact with the tart cherry juice.

The mean curves plotted show a change in acidity in patterns that were

very similar for all three packages (Table 12), although the juice in glass started

121

with the higher levels of acidity and ended with the least tart samples. PET and
aluminum bottles behaved similarly and very close to each other.

Table 12. Tartness perception by trained panel of tart cherry juice stored at
room temperature (23°C, 50% RH) for twelve months.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B Std. Error t P-value Sig
August-06 2.0893 0.8269 2.5268 0.0129 yes
October-06 0.4491 0.7763 0.5785 0.5641 no
November-06 -0.1620 0.7763 -0.2087 0.8350 no
January-06 1.1012 0.8269 1.3318 0.0186 yes
February-07 0.0000 0.8269 1.3318 0.0186 yes
Glass -0.7250 0.6188 -1.1717 0.2438 no
Aluminum -0.0187 0.6188 -0.0303 0.9759 no
PET 0.0000 0.6188 -0.0303 0.9759 no

 

 

4.7.6. Fresh cherry flavor perception
The trained panel found that the juice lost some of its fresh cherry flavor.

Fresh cherry flavor is characterized by a benzaldehyde compound that is the
cherry or almond flavor most commonly used for artificial flavoring.
Benzaldehyde generally decreases and loses power over time and can be
transformed into acetaldehyde which will release some fermented notes to the
cherry flavor due to lower levels of oxidation.

The p-values obtained for treatment were 0.205 and 0.390 for month.
These differences were not statistically significant, and so the null hypothesis
cannot be rejected. All three containers behaved similarly in maintaining the
fresh cherry flavor, although aluminum was found to slightly better retain the

fresh cherry flavor, followed by glass and then the PET bottles (Table 13).

122

Table 13. "Fresh cherry flavor" perception by trained panel of tart cherry

juice stored at room temperature (23°C, 50% RH) for twelve months.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B Std. Error t P-value Sig
August-06 0.9259 0.8705 1.0637 0.2897 no
October-06 0.7407 0.8705 0.8510 0.3966 no
November-06 0.2857 0.9271 0.3082 0.7585 no
January-06 0.0000 0.8765 0.4567 0.6543 no
February-07 0.4938 0.6938 0.7117 0.4781 no
Glass 0.9563 0.6938 1.3783 0.1708 no
Aluminum 0.0000 0.3498 1.2450 0.1567 no
PET 0.0000 0.3498 1.7654 0.2345 no

 

 

4.7.7. Off-flavor perception

In general the trained panel did not found any strong off-flavor from any of
the containers. Off-flavor is always a very important parameter to evaluate when
developing a new product because it can be bad for human consumption or will
create a bad taste which consumers will reject. Glass, being an inert material,
was not expected to produce any off-flavor. However, the other two packages
(aluminum and PET) might produce an off-flavor or react with any chemical
component from the juice. Some panelists identified some different flavor more
often in aluminum and PET bottles than in glass bottles (Table 14).

Aluminum bottles have a polymer base coating that may be released and
give an off flavor to the juice. Also if there are pinholes in the polymeric coat, the
aluminum can Chelate with the acidity of the cherry juice and add a metallic off
flavor to the juice. Straws (2005) reported the interaction of aluminum containers

with polymeric liner with fruit juices to cause flavor loss and chelating.

123

For the PET bottles, an off flavor can come from different possibilities --
from sorption of any PET residues, from the oxygen scavenger, plasticizers or
adhesives.

Finally the off-flavor may have come from the changes in the flavor notes
of the cherry juice itself. The changes from benzaldehyde to acetaldehyde can
create an off-flavor as well.

Table 14. Off-flavor perception by trained panel of tart cherry juice stored at
room temperature 33°C, 50% RH) for twelve months.

 

 

 

 

 

 

 

 

Aug.06 Oct. 06 Nov.06 Jan.07 Feb.07
Yes No Yes No Yes No Yes No Yes No
Glass 2 5 2 6 2 6 3 3 3 4
Aluminum 2 5 2 6 5 3 3 3 5 2
PET 2 5 6 2 5 3 4 2 2 5
Stand-up Pouch 3 4 6 2
n= 7 8 8 6 7

 

 

 

 

 

 

 

 

 

 

 

 

 

4.7.8. Overall flavor perception
The trained panel found that the juice retained the overall expected flavor

and was pleasant to their palate throughout the year of testing.

The overall flavor is the total impression of the product and how it is
perceived by consumers. For this question the trained panelists evaluated the
quality of the juice.

An ANOVA full model analysis produced the following results for this
question: p-values were 0.164 for month and 0.174 for treatment. This numbers
were greater than a: 0.05, and so there is no statistical significance to reject the
null hypothesis.

All three packages behaved similarly and end results were above the

middle point in the scale (Table 15). One of the limitations is that the trained

124

panel was trained with the fresh cherry juice, and panelists were asked only to
report changes perceived in the product. They were not asked about their
preferences. Questions about preferences were posed in the consumer’s
preferences test, and results will be discussed in section 4.8.

Table 15. Overall flavor perception by trained panel of tart cherry juice

stored at room temperature (23°C, 50% RH) for twelve months.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

B Std. Error t P-value Sig_
August-06 0.8976 0.8234 1 .0798 0.3450 no
October-06 0.7890 0.8456 0.8976 0.3678 no
November-06 0.3456 0.9876 0.2897 0.8965 no
January-06 0.0000 . . . no
February-07 0.5432 0.7498 0.7654 0.5632 no
Glass 0.9654 0.7435 1.4567 0.2345 no
Aluminum 0.8754 0.4567 1.2435 0.3456 no
PET 0.0010 0.4567 1.8934 0.2367 no

 

 

4.7.9. Purchasing appropriateness perception

About half of the trained panel reported that the juice was still good
enough to purchase after twelve month of storage and half believed the opposite.
This last question addressed whether or not the trained panel considered that the
juice had retained enough characteristics for consumers to enjoy the juice.

Of course, it is important to remember that the panel had been trained with
the fresh cherry juice and had been evaluating the juice throughout the duration
of the project. Therefore this last question does not reflect general consumer’s
preferences or purchasing intentions. Further work is needed it to evaluate this
criteria, and other parameters of the marketing mix (product, place, price and

promotion) will influence the decisions to purchase or not.

125

Again for this question the panel tended to like the juice packed in the
aluminum bottles the best, with glass and PET following directly in that order.
The least preferred juice was packed in the PET bottle. Contrarily to the initial
concept that glass would be the least inert and would better protect the juice and
retain better quality, glass fell below the values for the aluminum bottle (Table
16).

Table 16. Is this sample’s of tart cherry juice stored at room temperature

(23°C, 50% RH) for twelve months, quality appropriate for consumers to
purchase the drink?

 

Aug.06 Oct. 06 Nov.06 Jan.07 Feb.07
Yes No Yes No Yes No Yes No Yes No

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Glass 5 2 7 1 8 0 5 1 4 3
Aluminum 5 2 7 1 6 2 6 0 5 2
PET 6 1 6 2 6 2 3 3 4 3
Standing Pouch 2 5 1 7

n= 7 8 8 6 7

 

4.8. Consumer preferences results
Surprisingly, the non-trained consumers preferred the juice stored for

twelve months over the fresh tart cherry juice. But, they preferred the color and
texture of the fresh tart cherry juice. Aroma, flavor, overall flavor and overall
acceptability were equal for both samples.

In a preference test, over one hundred participants compared fresh juice
to juice that had been stored for twelve months. Six major factors were
analyzed: appearance, aroma, flavor, texture, overall flavor and overall
acceptability.

The consumers’ data showed no statistical perceptible difference in
aroma, flavor, overall flavor and overall quality. The scores were very similar

(Table 17). Appearance and texture were perceived better for the fresh juice

126

 

than for the stored juice when compared directly, with p-values of 0.0007 and
0.0016 respectively. Consumers commented that they would not be disliked by
the color and texture of the juice that was stored for twelve months if they did not
compare it directly with the fresh juice.

There was no statistically significant difference in perceived aroma, flavor,
overall flavor and overall acceptability of the juice, and no strong difference
between fresh and stored juice (Table 18). Appearance and texture scores were
greater for the fresh juice; the difference with the twelve month old juice was
statistically significant at alpha level 0.05 (Table 17).

Finally, the consumer rankings show that the twelve month old juice was
preferred to the fresh juice, as shown in Table 8, with a statistically significant
difference. It seems that with all of the changes in the juice during the storage
time, flavors developed and aged in a way that seemed to be more pleasant to
consumers.

Table 17. Means of consumers’ preferences results of fresh tart cherry
'uice and twelve month old juice in glass and aluminum containers.

 

 

 

 

 

 

 

 

 

 

 

 

 

12 month old
storedtart 0:335; P-Value Sig

cherry junce
Appearance 7.29 7.72 0.0007 yes
Aroma 6.35 6.35 1 no
flavor 6.34 6.74 0.0746 no
Texture 5.9 6.6 0.0016 yes
Overall flavor 6.34 6.52 0.4457 no
Overall acceptability 7.05 7.02 0.8692 no

 

Table 18. Ranking results results of fresh tart cherry juice and twelve
month old juice in _glass and aluminum containers.

 

 

12 month old
stored tart Gigs" .3; P-Value Sig
cherry juice ryj
Banking 1.64 1.36 0.0051 yes

 

 

 

 

 

 

127

 

IV. CONCLUSIONS
Since human nutrition has received more emphasis, the stability of

functional food components, like the anthocyanins in tart cherry juice, are
becoming more and more important for consumers and for the industry to
understand. This is particularly true for food manufacturers who seek to identify
the level of such beneficial compounds on food package labels.

Packaging is a very important part of every new product, and it must be
understood from several perspectives to minimize the risks of launching or
keeping a product in the market. It will be very important to analyze packaging
from all the possible interactions: product/package, product-package/consumer,
product-package/distributor, product-package/retailer, product-
package/environment, and product-package/transport. These last two
interactions are becoming more important every day. This research aimed to
show how important it is to consider all of these points of view when developing a
new product. Tart cherry juice undergoes many changes during twelve months
of storage. Therefore, it is important to understand these changes in order to
choose the best package to commercialize the product.

It is important to conclude as well, that the link between academia and
industry was successfully attached since the results were well accepted by the
target group. The research provided the industry with clue elements to help them
minimize the risks when bottling tart cherry juice.

Not all the packages used provided the same result, even though each

was chosen using the same criteria. They are good barriers against humidity and

128

 

oxygen, and have good reclosability and commercial availability. But it has been
shown that the properties of the cherry juice will go through different slightly
patterns of change in each of the different containers.

Some of the changes observed in the juice were not related to the type of
container, since they took place in all of the packages at the same levels:
transparency, yellowness, hue, and the changes in the anthocyanins with a blue
tint such as petunidin, peonidin and delphinidin.

On the other hand, the changes in redness and the anthocyanins with red
tints (cyanidin and pelargonidin) were related more to the container in which the
juice was packed. Glass and aluminum were the containers that better kept the
juice characteristics. However, glass presents disadvantages since it is a
material that requires high energy for production and transport, and it breaks
easily.

In focus groups conducted it was found that consumers preferred glass
packages for premium juices or functional beverages in all of the tested age
ranges. Only the groups of mothers with children below twelve years old
preferred the PET bottles for their children’s juice. However, these same mothers
chose glass containers for their own juice since glass containers were perceived
to have better quality and well-being. In conclusion, consumers preferred
packages that were transparent and recyclable (Whaling 2007).

Processors do not like to use glass containers and are trying to shift to
other possible containers such as those used for milk, baby food, baby juices.

There are several reasons for this: (1) the transport costs and the carbon

129

 

footprint are very high, (2) glass breaks and production lines have to been shut
down for cleaning, (3) it is heavy to manipulate at the warehouses, (4) retailers
prefer juices to be packaged in other containers than glass.

PET bottles are widely used to pack juices because they have good
barrier, are transparent, are less heavy and transport costs can be reduced.
They are less breakable and molds are less expensive than for the glass bottles.
PET is easier to mold into proprietary shape; if generic bottles are used, the
marketing mix will not have the full benefit. It is much more effective to develop a
unique package as was done for Pom-Wonderful (pomegranate juice) in a glass
bottle shaped as two fruits, one on top of the other. Later the company switched
from the glass bottle to the same bottle made in PET. Other examples
companies like Simply Orange® are using clear PET packages for their gently
pasteurized juices.

The use of PET bottles is rapidly growing in the market. PET is desired
more by the beverage industry due to its light weight, ease of handling and lower
energy levels needed for its production and transport compared to glass. It is
also recyclable (where PET recycling operations exist) although FDA restricts the
use of recycled material in direct contact with food products (CFSAN/Office of
Food Additive Safety 2006).

However, this study shows that PET bottles are the least protective
package for tart cherry juice due to the migration, and the loss of antioxidants
and general characteristics over time. If the industry wants to use this kind of

package, one option is to acknowledge a shorter shelf-life (around 6 months) to

130

__AJb

avoid the major decay of the quality, anthocyanin content and taste properties of
the tart cherry juice. This is the approach that the soft drink industry has taken.

Aluminum bottles are a great barrier against light which could trigger
photo—oxidation; and the performance for preserving the antioxidants was similar
to glass. Aluminum bottles are becoming more popular and they are reclosable
and recyclable as well. Aluminum bottles provide a large printing area, and so
attractive graphics labels can be printed directed on the bottle. This could be a
good option for the industry, a middle point between glass and PET. One of the
disadvantages with aluminum bottles is the lack of transparency, a characteristic
that Whaling (2007) found critical for functional foods, since the majority of the
panelists participating in the focus groups agreed that transparency was one of
the major preferences for a functional juice. Transparency is also a sign of
quality.

It is unfortunate that the stand-up pouch tests could not be completed.
But Whaling’s focus groups considered them to reflect a very low quality product,
to the point that the juice inside was not considered to be natural (Whaling 2007).
But this perception may change over time, as graphic, shapes and applications
increase.

More and more products are packed by retailers in stand-up multi—layer
packages due to their versatility. Some of these products are tuna fish, crackers,
cookies, baby finger foods, candy, chocolates, etc. The flexible packaging
provides versatility since no molds are required. Transportation costs are also

low since the material comes in rolls, and the packages are light weight.

131

It can be also observed that pasteurization at low temperatures inhibits
pathogen growth and does destroy anthocyanins present in tart cherry juice.
With this type of gentle pasteurization the bottles of juice can be stored at room
temperature and in closed boxes for a period of twelve months without any
pathogen growth. It is important to not rely on a cold chain since it is difficult and
expensive to maintain through all of the supply chain. The packaging method of
hot fill and nitrogen flush before capping was successful in preventing the
microbial growth and the browning of oxidation.

Anthocyanins suffered losses in varying degrees, a fact that it is very
important to know the limits for labeling purposes. There is no authorized health
claim for antioxidants, but it is important to remember that labels can be a way to
communicate and educate consumers about the health benefits and strength of
the product. Since the front and information panel have very strict regulations of
contents (FDA CFR 21) the other panels can be used to inform and educate the
consumers. They can write how much antioxidant power there is in the bottle
(always using the end of the shelf life data to prevent liability), how it might help,
what is an antioxidant and benefits. Then consumers will be able to recognize
the different products with this power. Graphics and design need to be related to
the trend.

The anthocyanins that suffered the most were those that are red: cyanidin
and pelargonidin. These also apparently migrated into the PET bottle walls. The
blue anthocyanins, petunidin, peonidin and delphinidin, increased through time,

and there is no evidence that they were influenced by the type of container used.

132

 

 

 

The changes in sugars and pH, the co-pigmentation thanks to
acetaldehyde and benzaldehyde were linked to the changes in anthocyanins and
this resulted in color changes. There was no evidence of oxidation in the juice,
since the color only became darker but did not turn brown. The trained panel

never identified oxidation flavors in the juice during the different sessions.

New Product

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I I I
I Raw Material ll Manufacturing | Marketing ‘ Finances ‘
Analysis

Availability Production Budgets
Quality Equipment Packaging Cash
Price Cost Others Loans
Seasonality Target Evaluation
Evaluation - Trends

'32::ng Evaluation A

Availability L

Usages

Reclosability

Compatibility

 

 

 

 

 

 

 

 

 

 

Figure 49. Path to commercialization
Adapted from Adelaja 2004.

The tart cherry industry will have to make decisions about the future of
juice and packaging. It will continue to evaluate the product, packaging material
and processes. Tart cherries are considered as "super fruits" (fruits with
antioxidant power” but the cherry flavor did not make it on the top ten flavors of

2008 (Halliday 2007). The top flavors for super fruits were acai berry and

133

mangosteen. Maybe a possibility for the tart cherry industry will be to find blends
and alliances to find a market outlet for theirjuice.

The industry will have to evaluate the technical results presented in this
study and continue to evaluate consumer’s preferences. They will need to
determine if consumers are willing to pay higher price for their juice packaged in
glass or if they should switch to a PET bottle with a shorter shelf-life (around six
months). They might also rethink the aluminum possibility and more closely
evaluate consumer’s perception closer related to this innovative new packaging.
Flexible packaging will need more years to catch up with the perceptions of
quality.

These results have shown that the existing heat processing equipment
can be used without making great investments. Alternatively, aseptic packaging
technology may be adopted.

These results are a guide for the tart cherry industry to make decisions
about the implementation of value-added products. This research is a link to
support and consolidate future projects between Michigan tart cherry industry

and Michigan State University.

134

 

 

 

 

V. FUTURE RESEARCH

Since this research used tart cherry juice concentrate, it would be very
interesting to compare these findings to single strength juice with no previous
concentration. The limitations will be the seasonality of the tart cherry crop, since
the juicec will need to be packaged immediately after harvesting. Concentrating
the juice up to 65° Brix might have damaged the anthocyanin content and color.

They are many other packages that might be interesting to evaluate for
future research like biodegradable polymers and Tetra-packTM for example.
These packaging systems are not as readily available in the market and will
require further analysis or investment from the industry point of view.

Third, the industry now uses very basic heat processing equipment and
does not have plants with state of the art fillers. It does not have any aseptic
filling systems which are becoming more and more popular in the juice industry.
due that it minimizes the heat processes. It will be very interesting to analyze the
difference between the different possible packaging lines to see which produces
a better taste and anthocyanin outcome that could justify the investment in
equipment.

Fourth, It will be very interesting to see if the standing pouch could be a
viable packaging source since it produces less post-consumer waste and uses
are growing significantly around the world. The appropriate filling and closing
equipment is needed in order to evaluate this packaging possibility.

Finally, consumer and supply chain research could better identify target

markets and marketing strategies to better serve the tart cherry industry.

135

 

APPENDIX 1: ANOVA Tables (treatment analysis)

 

136

Table 19. ANOVA full model (L* value).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Type III Sum of
Source Sguares df Mean Sguare F Sig.
Corrected
Model 2011.226(a) 35 57464 14.782 .000
'“tercept 21106.924 1 21106.924 54294? .000
Month 1846.600 11 167.873 43.183 .000
Treatment 30.528 2 15.264 3.926 .021
Mm“ 134.099 22 6.095 1.568 .053
Treatment
Error 1119.601 288 3.888
Total 24237.751 324
Corrected Total 3130.827 323
Table 20. ANOVA full model analysis of a* values.
Type III Sum Mean
Source of Squares df Square F Sig.
C°rre°ted 3656.618(a) 35 104.475 25.847 .000
Model
'"tercept 221492.858 1 221492.858 5479851 .000
Month 2740.395 11 249.127 61.635 .000
Treatment 294.045 2 147.023 36.374 .000
M°“th * 622.178 22 28.281 6.997 .000
Treatment
Error 1164.090 288 4.042
Total 226313.566 324
Corrected Total 4820.708 323
Table 21. ANOVA custom model analysis of a* values.
Type III
Sum of Mean
Source Squares df Square F Sig.
C°rre°ted 3034.441 (a) 13 233.419 40.509 .000
Model
'“tercept 221492.858 1 221492.858 384395: .000
Month 2740.395 11 249.127 43.235 .000
Treatment 294.045 2 147.023 25.515 .000
Error 1786.267 310 5.762
Total 226313.566 324
0°"eC‘ed 4820.708 323
Total

 

 

 

 

 

 

 

 

137

Table 22. ANOVA full model analysis of b* values.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Type III Sum Mean
Source of Squares df Square F Sig.
C°rre°ted 4409.574(a) 35 125.988 12.040 .000
Model
'“tercep‘ 50416964 1 50416964 481733 .000
Month 4015.607 11 365.055 34.885 .000
Treatment 92.981 2 46.490 4.443 .013
¥°nth 300.986 22 13.681 1.307 .164
reatment
Error 3013.785 288 10.465
Total 57840.323 324
Corrected Total 7423.359 323
Table 23. ANOVA custom model analysis of chrominance.
Type III Sum of
Source Squares df Mean Square F Sig.
3‘;ngth 559305.631(a) 13 43023.510 12.380 .000
'“tercept 6675085845 1 6675085845 19207: .000
Treatment 39290.446 2 19645.223 5.653 .010
month 520015.185 11 47274.108 13.603 .000
Error 76456.05? 22 3475.275
Total 7310847533 36
C°rre°ted 635761.688 35
Total
Table 24. ANOVA custom model analysis of chrominance.
Type III Sum Mean
Source of Squares df Square F Sig.
Corrected
Model .169(a) 13 .013 16.894 .000
Intercept 6.922 1 6.922 9014.338 .000
month .168 11 .015 19.937 .000
Treatment .000 2 .000 .157 .855
Error .017 22 .001
Total 7.108 36
Corrected
Total .186 35

 

138

 

 

 

Table 25. ANOVA full model analysis of solids change.

 

 

 

 

 

 

Type III

Sum of Mean
Source Squares df Sware F Sig.
Corrected
Model 24.876(a) 35 .711 12.326 000
'“tercep‘ 25948.550 1 25948.550 45009184 000
treatment 16.470 2 8.235 142.817 000
month 1.750 11 .159 2.759 .005
treatment 6.655 22 .303 5.246 .000
month
Error 4.152 72 .058
Total 25977.578 108
Corrected Total 29.027 107

 

 

 

Table 26. ANOVA custom model analysis of solids change.

 

 

 

 

 

 

 

 

 

 

 

Type III
Sum of

Source Squares df Mean Square F Sig.
“0":ngth 18.220(a) 13 1.402 12.191 .000
Intercept 25948.550 1 25948.550 225701.429 .000
month 1.750 11 .159 1.384 .193
treatment 16.470 2 8.235 71.629 .000
Error 10.807 94 .115

Total 25977.578 108

Corrected

Total 29.027 107
Table 27. ANOVA full model analysis of pH.

Type III
Sum of Mean

Source Squares df Square F Sig.
323:?“ 2.162(a) 35 .062 64.894 .000
Intercept 1356.884 1 1356.884 1425520390 .000
Treatment .006 2 .003 2.989 .057
Month 2.128 11 .193 203.235 .000
Treatment *

Month .028 22 .001 1.352 .170
Error .069 72 .001

Total 1359.115 108

Corrected Total 2.230 107

 

 

 

 

 

 

 

139

 

Table 28. ANOVA full model analysis for cyanidin 3-glucosylrutinoside

 

 

changes.

Type III Sum of
Source Squares df Mean Square F Sig.
mg?“ 31960339.286(a) 32 998760.603 25.296 .000
Intercept 85703755135 1 85703755135 2170.6 .000
Month 28012231842 10 2801223184 70.949 .000
treatment 3189248900 2 1594624450 40.388 .000
Month *
treatment 758858.544 20 37942.927 0.961 .518
Error 2605831691 66 39482.298
Total 120269926111 99
Corrected Total 34566170977 98

 

 

 

 

 

Table 29. ANOVA full model analysis for cyanidin 3-rutinoside changes.

 

 

 

 

 

 

 

 

Type III Sum of
Source Squares df Mean Square F Sig.
Egg?“ 130969.959(a) 32 4092.811 11.861 .000
Intercept 1947098611 1 1947098611 5642.77 .000
Month 117947.058 1O 11794.706 34.182 .000
treatment 3812.406 2 1906.203 5.524 .006
Month *
treatment 9210.496 20 460.525 1.335 .190
Error 22773.999 66 345.061
Total 2100842570 99
Corrected Total 153743.959 98

 

Table 30. ANOVA full model analysis for delphinidin 3-glucosylrutinoside

 

 

changes.
Type III Sum of

Source Squares df Mean Square F Sig.
322:?“ 50575.788(a) 32 1580.493 3.744 .000
Intercept 1066412982 1 1066412982 2526.459 .000
Month 41555.218 10 4155.522 9.845 .000
treatment 362.214 2 181.107 .429 .653
Month *

treatment 8658.356 20 432.918 1.026 .447
Error 27858.461 66 422.098

Total 1144847230 99

Corrected Total 78434.249 98

 

 

 

 

 

 

 

140

 

 

 

Table 31.ANOVA full model analysis for delphinidin 3-rutinoside changes.

 

 

 

 

 

 

 

Type III Sum of
Source Squares df Mean Square F Sig.
323:?“ 309133.805(a) 32 9660.431 1.754 .028
Intercept 10397826147 1 10397826147 1887.365 .000
Month 255955.433 10 25595.543 4.646 .000
treatment 762.650 2 381.325 .069 .933
Month *
treatment 52415.722 20 2620.786 .476 .967
Error 363605.618 66 5509.176
Total 11070565570 99
Corrected Total 672739.423 98

 

 

 

Table 32. ANOVA full model analysis for peonidin 3-rutinoside changes.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Type III Sum of
Source Squares df Mean Square F g1.
322:?“ 240084.438(a) 32 7502.639 .851 .687
Intercept 505631.669 1 505631.669 57.352 .000
Month 220846.401 10 22084.640 2.505 .013
treatment 2658.660 2 1329.330 .151 .860
mm“ 16579.377 20 828.969 .094 1.000
treatment
Error 581870.715 66 8816.223
Total 1327586822 99
Corrected Total 821955.153 98
Table 33. ANOVA full model analysis for pelargonidin 3-rutinoside changes.
Type III Sum of
Source Squares df Mean Square F Sig.
l
32:?“ 192138.771(a) 32 6004.337 25.709 .000
Intercept 1208970118 1 1208970118 5176.508 .000
Month 163157.944 10 16315.794 69.860 .000
treatment 16271.743 2 8135.872 34.836 .000
Month *
treatment 12709.083 20 635.454 2.721 .001
Error 15414.258 66 233.549
Total 1416523147 99
Corrected Total 207553.029 98

 

141

 

 

 

Table 34. ANOVA custom model analysis for pelargonidin 3-rutinoside

 

 

 

 

 

 

 

 

 

 

 

changes.
Type III Sum of
Source Squares df Mean Square F Sig.
33$?“ 179429.687(a) 12 14952474 45.724 .000
Intercept 1208970118 1 1208970118 3696.980 .000
month 163157.944 10 16315.794 49.893 .000
treatment 16271.743 I 2 8135.872 24.879 .000
Error 28123.341 86 327.016 '
Total 1416523147 99
C°rre°‘ed 207553.029 98
Total
Table 35. ANOVA full model analysis for 0etunidin 3-rutinoside changes.
Type III Sum of Mean
Source Squares df Square F I Sig.
l
agggfted 2302240.290(a) 32 71945.009 5.246 I .000
'“tercept 39195897009 1 391958973 2858.110 I .000
Month 1875247971 10 187524.797 13.674 .000
Treatment 18935.498 2 9467.749 .690 .505
Month "
treatment 408056.821 20 . 20402.841 1.488 .116
Error 905118.885 66 13713.922
Total 42403256184 99
Corrected Total 3207359174 98

 

 

 

 

 

 

142

 

 

APPENDIX 2: ANOVA tables (month analysis)

143

Table 36. L* value month ANOVA analysis.

Std. 95% Confidence
Parameter B Error t Sig. Interval
Upper
Lower Bound Bound
Intercept 4.481 0.418 10.719 0.000 3.658 5.304
[Treatment=1] 0.748 0.274 2.734 0.007 0.210 1.287
[Treatment=2] 0.439 0.274 1.604 0.110 -0.100 0.977
[Treatment=3] 0.000 . . . . .
[Month=1] 5.285 0.547 9.656 0.000 4.208 6.362
[Month=2] 9.321 0.547 17.031 0.000 8.245 10.398
[Month=3] 4.563 0.547 8.337 0.000 3.486 5.640
[Month=4] 2.179 0.547 3.980 0.000 1.102 3.255
[Month=5] 3.650 0.547 6.669 0.000 2.573 4.727
[Month=6] 2.263 0.547 4.135 0.000 1.186 3.340
[Month=7] 1.301 0.547 2.377 0.018 0.224 2.378
[Month=8] 3.403 0.547 6.218 0.000 2.326 4.480
[Month=9] 3.910 0.547 7.143 0.000 2.833 4.987
[Month=10] 1.696 0.547 3.099 0.002 0.619 2.773
[Month=11] 0.763 0.547 1.394 0.164 -0.314 1.840
Table 37. a* values month ANOVA analysis.
Std. . 95% Confidence
c 3 Error t 39' Interval
Upper
Lower Bound Bound
Intercept 20.269 0.499 40.621 0.000 19.287 21.251
[Month=1] 9.851 0.653 15.079 0.000 8.566 11.137
[Month=2] 8.179 0.653 12.518 0.000 6.893 9.464
[Month=3] 5.649 0.653 8.646 0.000 4.363 6.934
[Month=4] 5.991 0.653 9.171 0.000 4.706 7.277
[Month=5] 6.424 0.653 9.832 0.000 5.138 7.709
[Month=6] 4.241 0.653 6.492 0.000 2.956 5.527
[Month=7] —0.449 0.653 -0.687 0.493 -1.734 0.837
[Month=8] 4.639 0.653 7.101 0.000 3.354 5.925
[Month=9] 4.017 0.653 6.148 0.000 2.731 5.302
[Month=10] 3.906 0.653 5.978 0.000 2.620 5.191
[Month=11] 1.914 0.653 2.930 0.004 0.629 3.200
[Month=12] 0.000 . .
[Treatment=1] 2.051 0.327 6.280 0.000 1.409 2.694
[Treatment=2] 1.989 0.327 6.089 0.000 1.346 2.632
[Treatment=3] 0.000

144

Table 38. b values month ANOVA analysis.

Parameter

Intercept
[Treatment=1 ]
[Treatment=2]
[T reatment=3j
[Month= 1]
[Month=2]
[Month=3]
[Month=4]
[Month=5]
[Month=6]
[Month=7]
[Month=8]
[Month=9]
[Month=10]
[Month=1 1]
[Month=12]

B

7.264
1.220
1.029
0.000
9.051
12.154
5.470
3.125
6.481
3.382
-0.067
5.049
5.806
2.166
0.911
0.000

Std.
Enor

0.680
0.445
0.445

0.890
0.890
0.890
0.890
0.890
0.890
0.890
0.890
0.890
0.890
0.890

t

10.687
2.741
2.313

10.170
13.657
6.146
3.511
7.283
3.800
-0.076
5.673
6.524
2.434
1.024

Sig.

0.000
0.006
0.021

0.000
0.000
0.000
0.001
0.000
0.000
0.940
0.000
0.000
0.016
0.307

Table 39. Chrominance month ANOVA analysis.

Parameter

Intercept
[Treatment=1 ]
[Treatment=2]
[Treatment=3]
[month= 1]
[month=2]
[month=3]
[month=4]
[month=5]
[month=6]
[month=7]
[month=8]
[month=9]
[month=10]
[month=1 1]
[month=12]

B

Std.
- Error

t

220.975
73.77376
65.68584

0
375.3896
381.31
201.6422
102.7445
232.7258
128.9227
-1.28813
163.6884
204.6694
112.4787
55.41583
0

36.7627
24.06684
24.06684

48.13367
48.13367
48.13367
48.1 3367
48.1 3367
48.13367
48.13367
48.13367
48.13367
48.13367
48.13367

6.010848
3.06537
2.72931

7.798898
7.921897
4.189212
2.134567

4.83499
2.678431
-0.02676
3.400705
4.252105
2.336799

1.15129

145

Sig.

4.75E-06
0.005665
0.012245

8.98E-08
6.94E-08

0.00038
0.044186
7.86E-05
0.013726
0.978891
0.002567
0.000326
0.028958
0.261972

95%

Confidence

Interval
Lower
Bound

5.927
0.344
0.153

7.300
10.403
3.718
1.374
4.730
1.631
-1.819
3.297
4.055
0.415
-0.840

95%

Confidence

Interval

Lower
Bound

144.73
23.86
15.77

275.56
281.48
101.81
2.92
132.90
29.09
-101.11
63.86
104.84
12.65
-44.40

Upper
Bound
£1602
£1095
‘I905

10.802
13.905
7.221
4.876
8.233
5.133
1.684
6.800
7.557
3.917
2.663

Upper

Bound
297.21
123.68
115.59

475.21
481.13
301.46
202.56
332.54
228.74

98.53
263.51
304.49
212.30
155.23

Table 40. Hue values month ANOVA analysis.

Parameter

Intercept
[month=1]
[month=2]
[month=3]
[month=4]
[month=5]
[month=6]
[month=7]
[month=8]
[month=9]
[month=10]
[month=1 1]
[month=12]

[Treatment=1]
[Treatment=2]
[Treatment=3]

8

0.353
0.142
0.240
0.104
0.029
0.123
0.058
0.035
0.106
0.139
0.024
0.005
0.000
0.006
0.000
0.000

Std.
Error

0.017
0.023
0.023
0.023
0.023
0.023
0.023
0.023
0.023
0.023
0.023
0.023

0.011
0.011

t

20.412
6.270
10.618
4.597
1.281
5.434
2.561
1.526
4.694
6.136
1.077
0.235

0.499
0.028

Sig.

0.000
0.000
0.000
0.000
0.214
0.000
0.018
0.141
0.000
0.000
0.293
0.817

0.623
0.978

Table 41. AE values month ANOVA analysis.

Parameter

Intercept

[Treatment=1 ]
[Treatment=2]
[T reatment=3j

[month=1]
[month=2]
[month=3]
[month=4]
[month=5]
[month=6]
[month=7]
[month=8]
[month=9]
[month=10]
[month=1 1]

B Std. Error
3.580391 1.4309503
-1.30512 0.9721393
-1.21905 0.9721393

0 .
2.901667 1.861505

5.915 1.861505
1.039333 1 .861505
1.333667 1.861505

1.349 1.861505
3.390333 1 .861505
5.079433 1.861505
-1.51847 1.861505
1.564333 1 .861505

0102 1.861505
0

t

2.502107
-1.34252
-1.25399

1.558775
3.177536

0.55833
0.716445
0.724682
1.821286

2.72867
-0.81572
0.840359
-0.05479

146

Sig.

0.021138

95% Confidence

Interval

0.317
0.095
0.193
0.057
-0.018
0.076
0.011
-0.012
0.059
0.092
-0.023
-0.042

-0.018
-0.023

95%

Confidence

Interval

Lower
Bound

0595480966

0.194468
0.224305

0.134734
0.004733
0.582817
0.481999

0.47704
0.083557
0.012939
0.424269
0.410641
0.956846

-3.33296517
-3.24690153

-0.98136468
2.031968649
-2.84369802
-2.54936468
-2.53403135
-0.49269802
1 .196401982
-5.40149802
-2.31869802
-3.985031 35

Upper
Bound
(1389
0189
(1287
0151
(1076
0170
(1105
(1081
0153
0186
(1071
(1052

0.029
0.024

Upper
Bound
6.565300852
0722728805
0808792441

6.784698018
9.798031351
4.922364684
5.216698018
5.232031351
7.273364684
8.962464684
2.364564684
5.447364684
3.781031351

Table 42. Change in solids month ANOVA analysis.

95%
Parameter B Std. Error t Sig. Confidence
Interval

Upper Lower Upper Lower Upper

Bound Bound Bound Bound Bound
Intercept 16.056 0.122 131.525 0.000 15.814 16.299
[month=1.00] -0.189 0.160 -1.182 0.240 -0.506 0.128
[month=2.00] -0.250 0.160 -1.564 0.121 -0.567 0.067
[month=3.00] -0.028 0.160 -0.174 0.862 -0.345 0.290
[month=4.00] 0.000 0.160 0.000 1.000 -0.317 0.317
[month=5.00] 0.083 0.160 0.521 0.603 -0.234 0.401
[month=6.00] 0.083 0.160 0.521 0.603 -0.234 0.401
[month=7.00] -0.167 0.160 —1.043 0.300 -0.484 0.151
[month=8.00] -0.250 0.160 -1.564 0.121 -0.567 0.067
[month=9.00] 0.056 0.160 0.348 0.729 -0.262 0.373
[month=1 0. 00] 0.111 0.160 0.695 0.489 -0.206 0.428
[month=11.00] -0.111 0.160 -0.695 0.489 -0.428 0.206
[month=1 2. 00] 0.000 . .
[treatment=1.00] -0.953 0.080 -11.922 0.000 -1.111 -0.794
[treatment=2. 00] -0.550 0.080 -6.882 0.000 -0.709 -0.391
[treatment=3.00] 0.000

Table 43. Change in pH values month ANOVA analysis.
Std. . 95% Confidence
Parameter B Error t Slg. Interval
Upper
Lower Bound Bound

Intercept 3.769074 0.011556 326.1469 0.000 3.74 3.79
[Month=1] -0.09444 0.015131 6.24185 0.000 -0.12 -0.06
[Month=2] 034333 0.015131 -22.691 0.000 -0.37 -0.31
[Month=3] -0.44889 0.015131 -29.6672 0.000 -0.47 -0.418
[Month=4] 043222 0.015131 -28.5656 0.000 —0.46 -0.40
[Month=5] -0.29778 0.015131 -19.6802 0.000 -0.32 -0.26
[Month=6] -0.16889 0.015131 -11.1619 0.000 -0.19 -0.13
[Month=7] -0.21556 0.015131 -14.2461 0.000 -0.24 -0.18
[Month=8] -0.18889 0.015131 -12.4837 0.000 -0.21 —0.158
[Month=9] 020333 0.015131 -13.4383 0.000 -0.23 -0.173
[Month=10] -0.17889 0.015131 -11.8228 0.000 -0.20 -0.1483
[Month=11] -1.2E-14 0.015131 -8.1E-13 1.000 0030042636 0030042636
[Month=12] 0 . . .
[T reatment=1] -0.01417 0.007565 -1.87256 0.064 0029187985 0000854651
[Treatment=2] -0.01639 0.007565 -2. 16629 0.033 0031410207 0.001367571
[Treatment=3] 0

147

Table 44. Changes cyanidin 3-glucosylrutinoside month ANOVA analysis.

Parameter

Intercept
[Month= 1]
[Month=2]
[Month=3]
[Month=4]
[Month=5]
[Month=6]
[Month=7]
[Month=8]
[Month=9]
[Month=10]
[Month=1 1]
[treatment=1 ]
[treatment=2]
[treatment=3]

B Std. Error

56.29848
1674.892
1117.511
1304.01
741.9682
921.511
498.7858
461.7108
176.7644
101.367
76.99833
0
437.651
255.0479
0

71 .6766
93.243177
93.243177
93.243177
93.243177
93.243177
93.243177
93.243177
93.243177
93.243177
93.243177

48.694661
48.694661

t

0.785451
17.96262
1 1.98491
13.98504
7.957346
9.882879
5.349301
4.951684
1.895736
1.087125

0.82578

8.987659
5.237697

Sig.

0.434
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.061
0.280
0.411

0.000
0.000

95%
Confidence
Interval

Lower

Bound

-86.1898804
1489.530477
932.1501439
1118648255
5566069216
7361496994
313.4244772
276.3494772
859685614
-83.9943006
-1 08.362967

340.8492179
158.2460967

Upper
Bound
198.7868501
1860253078
1302.872745
1489.370856
927.3295228
1 106872301
684.1470784
647.0720784
362.125745
2867283006
262.3596339

534.4527821
351.8496609

Table 45. Changes cyanidin 3-rutinoside month ANOVA analysis.

Parameter

Intercept
[Month=1]
[Month=2]
[Month=3]
[Month=4]
[Month=5]
[Month=6]
[Month=7]
[Month=8]
[Month=9]
[Month=10]
[Month=1 1]

B

158.435
-28.452
-80.754
~82.083
-114.828
7.888
-17.938
1.920
4.694
5.292
8.01 1
0.000

Std.
Error

10.725
15.167
15.167
15.167
15.167
15.167
15.167
15.167
15.167
15.167
15.167

t

14.773
-1.876
-5.324
-5.412
-7.571
0.520
-1.183
0.127
0.309
0.349
0.528

148

Sig.

0.000
0.065
0.000
0.000
0.000
0.605
0.241
0.900
0.758
0.728
0.599

95%
Confidence
Interval

Lower Bound
137.023
-58.734
-111.036
-—112.365
-145.1 10
-22.394
-48.220
-28.362
-25.588
-24.990
-22.271

Upper
Bound
179.848
1.830
-50.472
-51.801
-84.546
38.170
12.344
32.202
34.976
35.574
38.293

 

Table 46. Changes delphinidin 3-glucosylrutinoside month ANOVA

analysis.

Parameter

Intercept
[Month= 1]
[Month=2]
[Month=3]
[Month=4]
[Month=5]
[Month=6]
[Month=7]
[Month=8]
[Month=9]
[Month=10]
[Month=1 1]

B

93.24967
28.07733
21.50367
18.45367
49.58167
-13.1827
4.291667
-0.94933
-4.653
-0.62867
-1.651

0

Std.
Error

11.86167
16.77494
16.77494
16.77494
16.77494
16.77494
16.77494
16.77494
16.77494
16.77494
16.77494

7.861426
1 .673767
1 .281892
1 .100074
2.955699
-0.78585
0.255838
-0.05659
-0.27738
-0.03748
-0.09842

Sig.

0.000
0.099
0.204
0.275
0.004
0.435
0.799
0.955
0.782
0.970
0.922

95%
Confidence
Interval

Lower Bound

69.56707158
-5.41491383
-11.9885805
-15.0385805

16.0894195
4667491383
-29.2005805
-34.4415805
-38. 14524716
-34.12091383
-35. 14324716

Upper Bound

1 16.9322618

61 .5695805
54.99591383
51 .94591383
83.07391383

20.3095805
37.78391383
32.54291383
28.83924716

32.8635805
31 .84124716

 

Table 47 Changes delphinidin 3-rutinoside month ANOVA analysis.

Parameter

Intercept

[treatment= 1 ]
[treatment=2]
[treatment=3]

[Month= 1]
[Month=2]

[Month=3]
[Month=4]
[Month=5]
[Month=6]
[Month=7]
[Month=8]
[Month=9]
[Month=10]
[Month=1 1]

B

347.587
6.113827
0.481697

0

-124.486
-98.9176

-78.2947
-24.5059
24.29646
-39.8061
30.93599
-4.47806
18.99994
13.50696

0

Std. Error

25.203605
17.122478
17.122478

32.787048
32.787048

32.787048
32.787048
32.787048
32.787048
32.787048
32.787048
32.787048
32.787048

t

13.79116
0.357064
0.028132

-3.7968
-3.01697

-2.38798
-0.74743
0.741038
-1.21408
0.943543
-0.13658
0.579495

0.41 196

149

Sig.

0.000
0.722
0.978

0.000
0.003

0.019
0.457
0.461
0.228
0.348
0.892
0.564
0.681

95%
Confidence
Interval

Lower
Bound
297.4839065
-27.9245316
-33.5566619

-189.664439
-1 64.096062

-143.473151
~89.6843728
408820395
-104.984595
-34.2425061
-69.6565506
46.1785506
-51.6715395

Upper
Bound
397.6901218
40.15218617
34.52005587

59.30744942
33.73907164

13.1 1616053
40.67261724
89.47495058
25.37239502
96.1 1448391
6070043947
84.17843947
78.68545058

Table 48. Changes peonidin 3-rutinoside month ANOVA analysis.

Parameter

Intercept

[T reatment=1j
[T reatment=2j
[Treatment=3]
[month=1]
[month=2]
[month=3]
[month=4]
[month=5]
[month=6]
[month=7]
[month=8]
[month=9]
[month=10]
[month=1 1]

B

98.523
11.703
1.594
0.000
-87.938
-93.089
-92.263
-95.954
-1.933
-22.409
15.203
2.361
6.292
23.349
0.000

Std.
Error

30.229
20.536
20.536

39.324
39.324
39.324
39.324
39.324
39.324
39.324
39.324
39.324
39.324

t

3.259
0.570
0.078

-2.236

-2.367
-2.346
-2.440
-0.049
-0.570
0.387
0.060
0.160
0.594

Sig.

0.002
0.570
0.938

0.028
0.020
0.021
0.017
0.961
0.570
0.700
0.952
0.873
0.554

95% Confidence
Interval

Lower Bound
38.430
-29.122
-39.231

-166.112
-171 .263
-170.437
-174.128
-80.107
-100.583
-62.970
-75.813
-71.881
~54.825

Upper
Bound
158.615
52.528
42.419

-9.764
-14.915
-14.090
-17.780
76.241
55.765
93.377
80.534
84.466
101.523

Table 49. Changes pelargonidin 3-rutinoside month ANOVA analysis.

Parameter

Intercept
[month=1.00]
[month=2.00]
[month=3.00]
[month=4.00]
[month=5.00]
[month=600]
[month=7.00]
[month=8.00]
[month=9.00]
[month=1000]
[month=11.00]
[month=1200]

B

16.05648
-018889
-0.25
-0.02778
6.45E-15
0.083333
0.083333
-0.16667
-0.25
0.055556
0.111111
-0.11111
0

Std.
Error

0.122079
0.159839
0.159839
0.159839
0.159839
0.159839
0.159839
0.159839
0.159839
0.159839
0.159839
0.159839

t

131.5252
-1.18174
-1.56407
-0.17379
4.04E—14
0.521357
0.521357
-1.04271
-1.56407
0.347572
0.695143
069514

150

Sig.

0.000
0.240
0.121
0.862
1.000
0.603
0.603
0.300
0.121
0.729
0.489
0.489

95%
Confidence
Interval

Lower Bound
15.81409045
050625325
0567364361
0345142139
0317364361
0234031028
0234031028
0484031028
0567364361
0261808805
020625325
0428475472

Upper Bound
1629887252
0.128475472
0067364361
0289586583
0.317364361
0400697694
0.400697694
0.150697694
0.067364361
0.372919916
0.428475472

0.20625325

 

Table 50. Changes in petunidin 3-rutinoside.

Parameter

Intercept
[treatment=1 ]
[treatment=2]
[treatment=3]
[Month= 1]
[Month=2]
[Month=3]
[Month=4]
[Month=5]
[Month=6]
[Month=7]
[Month=8]
[Month=9]
[Month=10]
[Month=11]

B

702.6743
32.39176
7.606061

0
-317.277
-284.879
—223. 104
-1 77.309
42.63867
-135.224
66.87644
-1 0.0344
41.03589
42.62333

0

Std. Error

44.778163
30.420771
30.420771

58.251342
58.251342
58.251342
58.251342
58.251342
58.251342
58.251342
58.251342
58.251342
58.251342

t

15.69234
1.064791
0.250029

-5.44669
4.89051
-3.83003
-3.04386
0.731977
-2.32138
1.148067
-0.17226
0.704463
0.731714

151

Sig.

0.000
0.290
0.803

0.000
0.000
0.000
0.003
0.466
0.023
0.254
0.864
0.483
0.466

95%
Confidence
Interval

Lower
Bound
613.6582409
-28.0827344
528684314

433.077058
400678391

-338.90428
-293.108725
~73.161 1692
-251.023391
48.9233914

-125.83428

-74.763947
-73. 1765025

Upper Bound
791.6903248
9286624954
68.08055257

-201.4773863
4600787197
-107.3046086
6150905302
158.4385025
-1 9.42371 968
182.6762803
105.7653914
156.8357248
158.4231692

 

APPENDIX 3: Consumer Consent Form and Trained panel survey

152

 

Consent Form
“Tart Cherry Juice consumers testing’
Sample: Tart (sour) cherry juice

Before you decide to sign this consent form and continue to participate in our study,
please read carefully and thoroughly the reverse side form for the sample ingredients
and preparation information, purpose and procedure of this study, potential risks and
benefits from your participation, our assurance of your privacy, your right as human
subject in our study, etc.

If you have any questions during your reading this consent form, or during or after your
participation, please do not hesitate to contact the on-site sensory evaluation leader
and/or the principal investigator. Feel free to contact Dr. Suzanne Thornsbury, the
principal investigator of this study, via phone at (517) 432-5418 (211 Agriculture Hall,
Michigan State University, East Lansing MI 48823) or Maria-Paz Gonzalez at 517-353-
8628. You cacn also reachc us via email at thornsbu@msu.edu or gonza221@msu.edu
for any inquiry you might have due to your participation in our study.

In case you have questions or concerns about your role and rights as a research
participant, please feel free to contact Dr. Peter Vasilenko, Director of Human research
Protections by phone: (517) 355-2180, fax (517) 4324503, email: irb@ms.edu or regular
mail: 202 Olds Hall, East Lansing MI, 48824.

If you have read all the information we offer to you in this consent form and decide to
participate in our study an give us your valuable response to our questionnaire, you can
go ahead and sign this form now. Otherwise, you can stop here and feel free to
discontinue participation in our study without any penalty.

PLEASE NOTE UPON YOUR SIGNING THIS CONSENT FORM, YOU VOLUNTARLY
AGREE TO PARTICIPATE IN OUR STUDY. YOUR SIGNATURE INDICATES YOU
HAVE READ THE INFORMATION PROVIDED ABOVE AND THAT YOU HAVE AN
ADECQUATE OPPORTUNITY TO DISCUSS THE STUDY WITH THE PRINCIPAL
INVESTIGATOR AND HAVE ALL THE QUESTIONS ANSWERED TO YOUR
SATISFACTION. YOU WILL BE GIVEN A COPY OF THIS CONSENT FORM WITH
YOUR SIGNATURE FOR YOUR RECORDS UPON REQUEST

SIGNED: DATE
You indicate your voluntary agreement to participate by signing above.

 

153

Consent Form
“Tart Cherry Juice consumers testing’
Agricultural Economics Department and School of Packaging, Michigan State University.

Invitation to Participate:
You are invited to participate in this study that asses the quality attributes of tart
(sour) cherry juice from concentrate.

Purpose of this sttfl

This study is intended to study the quality of tart (sour) cherry juice from
concentrate and consumer acceptability. Appearance and flavor characteristics
of tart (sour) cheery juice from concentrate will be evaluated.

Procedure of this study:

Each participant will be presented with tart (sour) cherry juice from concentrate.
They will be asked to evaluate visually and flavor, score the attributes as
presented on the score sheet for each sample. Samples will be presented using
three digit random codes. Consumer marketing questions will also be asked.

Sample Ingregients and samplepreparation:

All the ingredients used in our samples are food-grad and FDA approved for
foods. The ingredients are: 100% tart cherry juice from concentrate, water and
no sugar added or chemical added.

Potential risks:

Because all ingredients we use in this study are food grade and FDA approved
for food applications, these samples pose no adverse health risk, provided the
subject has not been identified as being susceptible to an allergic reaction to the
previously listed sample ingredients. If you believe there is a potential of an
allergic reaction upon sniffing and tasting notify the on-site sensory evaluation
coordinator and/or principal investigator immediately. You will be released from
participating in this study. Please note if you are injured as a result of your
participation in this research project, Michigan State University will assist you in
obtaining emergency care. If you have insurance for medical care, your
insurance carrier will be billed in the ordinary manner. As with any medical
insurance, any costs that are not covers or in excess of whatever are paid by
your insurance, including deductibles, will be your responsibility. Financial
compensation for lost wages; disability, pain or discomfort are not available.
That does not mean that you are giving up any legal rights you may have. You
may contact Suzanne Thornsbury with any questions (517) 432-5418.

Potential benefits:

There are no benefits gained directly from your participation in this study.
However, your participation and response ill provide us valuable data, which can
be use to identify shelf life for tart cherry juice in different containers and help
keep these growers profitable.

154

 

Assurance of confidentialitm -

Any information obtained in connection with this study that could identify you will
be kept confidential by ensuring that all consent forms are securely stored. All
data collected and analyzed will be reported in an aggregate format that will not
permit associating subjects with specific responses or findings. Your privacy will
be protected to the maximum extent allowable by the law.

 

Withdrawal from this study:
Participation in this study is voluntary. Your decision to refuse participation or

discontinue participation during this study will be honored promptly and
unconditionally.

 

155

Trained panel survey

Name Date Code_

 

Please answer all the question for each of the 4 samples presented to you

Sample 358

1. How do you perceive color of the tart cherry juice?

[7"

 

 

 

a------—
---—-- d

 

N

0123456 89101112131415
Very good Very bad

 

2. How do you perceive the tart cherry juice aroma?

 

 

 

 

N

0123456 89101112131415
Very good Very bad

3. How do you perceive the mouth feel of the tart cherry juice?

 

 

 

---—---
-------
n------
o------
-----
—------
-------

 

a) +----
N

012345 89101112131415
smooth grainy

 

4. How do you perceive the sweetness in this sample?

-----fi

 

 

 

0123456 89101112131415
Not sweet enough Too sweet

V

156

5. How do you perceive the sample's tartness?

 

-----1
----'-"-q
-----———u
------d
-----.--d
-------1
——--—--1

 

I
I
I
l
I
I
I
l
1
I

 

-----‘

 

CD
\I

012345 89101112131415
Not tart enough Too tart

6. How do you perceive the fresh cherryjavor?

 

 

---—----1

 

 

A o-------1

0 2 3456 78 9101112131415
Very good Very bad

7. Do you perceive any off-flavor?
Yes No Which? Comments

 

 

 

8. How do you perceive the overall flavor?

 

 

------
------d
-------

-‘----d

 

 

01 .----..---

01234 6789101112131415
Very bad Very good

9. Is this sample’s quality appropriate for consumers to purchase the drink?

Yes No

 

Comments

 

 

 

157

APPENDIX 4: Stand-up pouch results

158

 

Color results

Figure 50. Changes in L* values.

 

 

 

 

 

I L*values I
I30
25 47777 77- 7477
I
20 2 , ., -, - 7777* 2-, ---.-_-¢,2,2,2,II
—+—Glass II
—I— Aluminum I?
15 7,’ . ~ ~ 2 -- - 2 7 - 2 -. ——A—PET I
-,*,:Star9-UPpoucm
\ /\ ’ ,1 fl
5 , -. ,, , - , 2 - - ‘2.
o . . . ‘
1 2 3 4 5 6 7 8 9 10 11 12

 

159

Figure 51. Changes in a* values.

 

 

a“ values

 

32

3O

28*

26

24,

22‘

201

 

 

 

 

160

 

_+_ 67...; 7
—|— Aluminum I
T+PET It

7 -o- - Stand-up pouch ‘

 

Figure 52. Changes in b* values.

 

 

 

 

 

 

 

b” values
30
25
20
I¥Gléss
15 I—I— Aluminum
+ PET .
L' f- - Stand-up pouch“
1O
5 .9
0
1 2 3 4 5 6 7 8 9 10 11 12
Month

161

Solids and PH results.

Figure 53 Changes in solids.

22 222222 222222 2222 -2 2 2 2 .2 2 2 22
l
|
I

 

Brix

‘ 17 22 222222222222222222 —-—~—--—J I

—I— Aluminum I
+PET I
- -o- - Stand-up pouch!I

 

 

 

i

 

 

13.5 .. »----—— 2 -—- 22 ——9 22 —.—- 2. 4 2.2. - -— 4—4— _

 

 

 

 

162

Figure 54. Changes in pH.

 

 

 

 

 

 

 

pH
4.2
4 2 2 2 2 2 2 2 2 2 2 2 2
3.8 —
1+" ’ Glass '
I_ .
3.6 l— Aluminum
I+PET I
‘- .0. - Stand-up pouch .
3.4
3.2
3 r
1 2 3 4 5 6 7 8 9 10 11 12

 

163

Anthocyanins results.

Figure 55. Changes in cyanidin 3-glucosylrutinoside.

 

 

 

 

 

 

 

 

Cyanidin 3-glucosylrutinnoside I

I

3000 I

I

I

2500~ I

I

I

2000 I I

:: ______+____fl _I

5 I——o—Glass II

a 1500 I—F Aluminum ‘

g II+PET II

E - -o- - Stand—up pouchIf

< T 22222 ’ A" 7 W I
1000 .

500 I

I

0 2 I

I

x. k '\ 's

YQQ be‘ 0° §\§\ 90" & Ce} ‘06 006 $06 I

o 6‘ Q 6‘ o k I

v. $0 00 04 6' 3’0 Q0 I,

90 V“ 00 I

164

Figure 56. Changes in cyanidin 3-rutinoside.

 

Cyanadin 3-rutinoside

 

 

 

 

 

300 m- 0.- u. ,
250 d , , 7 7 7, 7 7 , 7 ,, I, ”W
200 «
E ‘i—WO—élass 7
g 150 I—l— Aluminum
5 I+PET
g ' "' ' Stand-Up pouch
( r ,
100 7
50 ,
0
«2% Q?» 00 33* sv‘ 9} x “2} ‘ the! 2,6
Y9 ob ‘9 (P 50 ‘9
29¢ 69 9° Q ‘

 

165

Figure 57. Changes in delphinidin 3-glucosylrutinoside.

 

Area (mgl100ml)

 

200

180

160

140

120

100

Delphidin 3-glucosylrutinoside

 

 

 

 

 

166

—+—:Gliassfi
‘ —I— Aluminum
‘+PET

- .0. - Stand-up pouch

Figure 58. Changes in delphinidin 3-rutinoside.

 

 

 

 

 

 

 

 

 

 

 

 

 

167

Delphidin 3-rutinoside
450 I""‘"""‘"‘ ~
400 ~ , — , v ,_ ~%~#fl——————~4*+**~ ~
350+~**F*fi——~— —-_——‘—fi!-———~—~
300 f, a —e~ A—_ ,__ 1; A T I“
250 w—~fl-— - ' ' ~—‘<— —~
...I / / 3
E ‘ I \ I
a 200 0’“ “ I " J_ \ I .1. ““—
3 ‘ ’ 4.] l \
I, mm “II—___ ”___ _I
100,v_i::#_* ___ I w-w*‘-*ri
50 A _L, a 7- é are - ~~ w — -—-- # ~ ~
0 g ,_ A A ! i __ __ ,,- .7 .+ ## w 7 7
.L __
-50 I j T T l T T
° « x x
9 Q‘0 06} 40 000 so“ 50
9° \9 O Q

 

I—o— Glass
—I— Aluminum

I
I
I3; §£§1¢EBEPU§W

I

 

 

Figure 59. Changes in peonidin 3-rutinoside.

 

 

 

 

 

 

 

Peonidin 3-rutinoside
300
250
200
g 150 7 1‘ Glass ‘I
E I—l— Aluminum I
m
g I+PET ‘]
E 100 7 ‘- -o- -Stand-up pouch
< 7 7 7 7 7 7 7 I
50
0
-50 I I
-\ e x x ‘ K
Y3? 55‘ 0° 9* 90" sé 0° 0'0 0 I
x) 6‘ ‘9 g (0 0 K
Y‘ \0 0° 0“ 6‘ 3° ((6‘) I
‘on '5 0° I
I

168

Figure 60. Changes in pelargonidin 3-rutinoside.

 

 

 

 

 

 

 

Pelargonidin 3-rutinoside
250 -—~- -
200
E 150 —¢7—_Glass
% —I—- Aluminum
E +PET
E 100 I- + -Stand-up pouch‘
< ,- , ,
50
0 i
. «
Qs gs 00% 9* g so} oe‘ & ‘ 096 9‘
Y7 3 ‘3 6‘ \9 (5° 9 g
Y‘ \9' 0° 4 d" 3’5 27°
c9 ~99 0°

169

Figure 61. Changes in petunidin 3-rutinoside.

 

Petunidin 3-rutinoside
1000

 

90077 777

3007 7777777

 

  

 

 

170

g, 500 HI—l— Aluminum
E I+PET
E 400 . ‘.- -0- -Stand-up pouch
< ‘l
I
300 ,I
I
200 ,I
100 77777 I
I
0 I
x A ‘k g
Yg‘ ®% 30“) 3° 6)? ‘0 6)?) d“ é
“I of 06‘ of 696‘
9° V“

 

APPENDIX 5: Calibration curves

171

For cyanidin 3—glucosilrutinoside a step calibration is offered in

Figure 62. The lineal model is A= 15552.6 * Concentration

 

 

 

x
1500 — x
x
1000 —
500 —
l Co

I l l l
0 0.2 0.4 0.6 0.8 1

Figure 62. Step Calibration for cyanidin 3-glucosilrutinoside

 

 

 

For cyanidin 3-arabinorutinoside a calibration curve is offered in

Figure 63. The lineal model is A= 0.0125938 * Concentration + 0.00551505

 

 

 

1.2 —
X
1.0 —
0.7 —I X
0.5 A
0.2 -—
0 I I l l l CO
2 4 6 8 10

 

 

 

Figure 63. Calibration curve for cyanidin 3-arabinoserutinoside

172

REFERENCES

173

Adelaja, Soji. 2004. Jersey blues: a market peveloment initiative. Presentation
read at the Cherry Marketing Institute, 7 June, at East Lansing, Michigan.

Asian Food Information Center (AFIC). 2003. Dietary supplements & functional
foods: an overview, March 1, 2003.
<http://www.afic.org/FFA%20lssue%2017%20March°/o202003°/020-
%20dietary%20$upplements> (June 12, 2005).

Alper, N., Savas Bahceci, K. and Acar, J. 2005. Influence of processinggnd
pasteurization on color valpes and total phenolic compounds (momegranate
iuice. Journal of Food Processing and Preservation 29: 357-368.

Aluminum association, Inc. 2001. Aluminum bottles a rival pppularity of plastics.
<http://www.aluminum.org/AMfTemplate.cfm?Section=Home&TEMPLATE=ICM/
ContentDisplay.cfm&CONTENTID=16147> (June 23 2006).

American Chemical Council. 2005. Shifting wind plow U.S. plastic resins gown in
2005. <http://www.americanplasticscouncil.org/s_apc/docs/1800/1712.pdf >
(August 2, 2006).

American Plastic Council. 2005. PIPS year-end statistics for 2905.
<http://www.americanplasticscouncil.org/s_apc/bin.asp?ClD=296&DID=1678&D
OC=F|LE.PDF> (July 15, 2006).

Artés, F., Minguez, ML, and Hornero, D. 2002. Analyzing changes in fruit
pigments. In Color in food, edited by Douglas B. MacDougall. Boca Raton, FL:
Woodhead Publishing Limited 10: 248-282.

ASTM C 162. 2005. Standard terminologmmss and glass produficts. ASTM
International, West Conshohocken, PA. <www.astm.org> (June 3'd , 2006).

Bansal, NP. and RH. Doremus. 1986. Hanflmk of glass properties. London:
Academic Press Inc. Ltd.

Bauer—Christoph, C., Watcher, N., Christoph, A., RoBmann, and Adam, L. 2005.
Assiment of raw material and authentication of spirits by gas chromatography,
hydrogen-and carbon-isotope ratio measurement. European Food Research and
Technology 204 (6): 445-452.

 

Best, R.J. 2004. Market-Baseg management: Strategies for Growing Customer
Value and Profitabilim. 3rd edition. Upper Saddle River; Prentice Hall.

Brin, M.F., Lew, M.F., Adler, C.H., Comella, C.L., Factor, S.A., Jankovic, J.,
O'Brien, 0., Murray, J.J., Wallace, J.D., Willmer—Hulme, A., and Koller, M. 1999.
Safety and efficacy of NefiuroBloc (@tulinpm toxin type B) in type A-resistant
cervical dystonia. Neurology 53 (7): 1431-8.

174

Brown, B. L. 1997. New learping strategies forpqeneration-x. ERIC Digest No.
184, ERIC Clearinghouse on Adult Career and Vocational Education Columbus
OH, <http:/lwww.ericdigests.org/1998-1/x.htm> (July 15 2008).

Burke, BA. and Nair, MG. 1986. Phenylpropene, benzoic acmm flavanoiq
derivative from fruits of Jamaican Piper species. Phytochemestry 25: 1427-1430.

Cao, G., Alessio H., and Cutler, R. G. 1993. Oxygen-radical apsorpency
papacity assay for antioxidants. Free Radical, Biology and Medicine 14:303-311.

Carden, LA. and CS. Baird. 2000. Flavors. In Food Chemistry: principles and
applications. Ed. Genevieve L. Christian & J. Scott Smith. West Sacramento,
CA; Science Technology System 13: 201-214.

 

Goauda, M. 2005. Personal phone interview with Marc Gouda, CCL Containers
February 4, 2006.

Chandra, A., Nair, MG, and lezzoni, A. 1992. Evaluation and characterization
of anthocyanins pigments in tart cherries (prunus cerasus L.). Journal of
Agriculture and Food Chemistry 40:967-969.

Chandra, A., Rana, J. and Y, Li. 2001. Separationpigentificatiomggantification
and method validation of anthocyanins in botanical supplement raw materials by
HPLC and HPLC-MS. Journal of Agriculture and Food Chemistry 49: 3515-3521.

Chaovanalikit, A. and RE. Wrolstad. 2004. Total anthocyanins and total
phenolics of fresh and processed cherries and their antioxidant properties.
Journal of Food Sciences 69 (1): 67-72.

China Plastics and Rubber Journal. 2008. Status and technologypeveloment in
China's aseptic packaging market.
<http2/lpackaging.2456.com/eng/epub/details.asp?epubiid=5&id=10141> (July
16, 2008).

Codex Alimentarius. 2006. mart of the nineteen session of the ioint
EOE/Codex Alimentariu_s_gr0lm of experts on standardization on frpit iuices
Codex Alimentarius Commission, 19t session, July 1 -10 <
www.codexalimentan’us.net/download/report/71/cI00_01e.pdf > (August 23 2006)

Constenla, D.T. and J.E. Lozano. 2005. Effects ofpretreatments and_processing

conflions on the chemical. pfl/sical. micromolgqical and sensory characteristics
pf_garlic paste. Journal of Food Process Engineering 28: 313-329.

175

Coutelieris, FA, and A. Kanavouras. 2005. Experimental and theoretical
investigation ofpackecj olive oil: Development of a mlity indicator based on
mathematical medictions. Journal of Food Engineering 73 (1): 85-92.

Day E, Crask, M.R, Harris, C.E. Jr. 1989. Which marketingstrategies breec_l
success? Business. 39 (3): 32-41.

De Ancos, B., Cano, P., Hernandez, A. and Monreal, M. 1999. Effects of
microwave heating on pigment composition anp color of frth pprees. Journal of
Science and Food Agriculture 79: 663- 670.

Eglinton, J., Griesser, M., Henschke, P., Kwiatwoski, M., Parker, M. and
Herderich, M. 2004. Yeast mediated formation of pigmented polymers in red
wine. In Red wine color: revealingthe mysteries. Edited by. Andrew L.
Waterhouse and James A. Kennedy. Washington DC. American Chemical
Society (2): 7-21.

Enviromental Protection Agency (EPA). 2006. Municipal Solid Waste ,
<http://www.epa.gov/msw/glass.htm> (August 22, 2006).

Esti, M., Cinquanta, F., Sinesio, F., Moneta, E., and Di Matteo, M. 2002.
Physicochemicpl and sensory frpit characteristics of two sweet cherry cpltivars
after cool storage. Food Chemistry.76: 399-405.

Euromonitor. 2005. Packaged food database, North America.
<www.gmid.euromonitor.com/HitList.aspx> (August 22, 2006).

Euromonitor. 2008. Functional drinks database. North America, Asia Pacific and
Eastern Europe. <www.gmid.euromonitor.com/HitList.aspx> (August 28, 2008).

European Chemical Bureau (ECB). 2006. Chemicals. < www.ecb.jrc.it>
(September, 06 2006).

Facer, T. 2005. Canned Cherry Products. Paper read at North American Tart
Cherry Industry Conference, 15 March, at Grand Rapids, Michigan.

FAO. 2005. Principles anq Practices of Small — and Medium- Scale Fruit
Em_ce§§ipg.< http://wwwfao.org/Idocrep/OO5/y2515e/y2515e02.htm> (December
10,2005)

FAO. 2006. Fruit anglegetable processing.
<http://wwwfao.org/ldocrepN5030EN5030Eh.htm> (July 10, 2006).

Flexible Packaging Association. 2006. Industry Information
<http://www.flexpack.org/industry_information.htm> (July 10, 2006).

176

Food Flavor. 2007. Flavor definition. <http://food--
flavor.blogspot.com/2007/10ldefinition-of-favor.html visited>, (June 10 2008).

Food Processing Technology. 2008. Shelf-life.
<www.foodprocessingtechnology.com. (July 16 2008).

Francis, F.J. 1989. Food colorants: Anthocyanins. Food Science and Nutrition
28: 273-314.

Franz, R. and F. Welle. 2001. Recycled polyethylene terephthalate for direct
food contact applications: challenge test of an incline recycling_process. Food
Additives and Contaminants 18 (5): 502-511.

Freitas, V. and N. Mateus. 2006. Chemical transformation of anthocyanins
yielcmg a variety of colors. Enviromental Chemistry 4:174-183.

Fridaus, V. and PP. Tong. 1997. Polyethylene, linear and very low density.
Encyclgaedia of Packagi_ng Technology. Edited by Brody, AL and Marsh, K.S,
New York: Wiley and Sons, Inc.: 748-752.

Fulcrand, H., Atanasova, V., Salas, E., and Cheynier, V. 2004. The fate of
anthocyanins in wine: Are they determining factors. In R__ed wine color: revealing
the mysteries. Edited by Andrew L. Waterhouse and James A. Kennedy.
Washington DC: American Chemical Society (6): 68-88.

Functional foods Japan. 2006. Report. <http://www.functionalfoodsjapan.com>
(August 28 2008).

Getachew, A., and C. Peterson. 2005. New product development trendiin the
furuit sector. Working paper No. 1-102605, Michigan State University product
center for agricultural and Natural Resources, East Lansing Michigan.

Gianestras, J., Sarola, M., and Rogers, G. 2005. Frozen products market
segment analysis. Paper read at North American Tart Cherry Industry
Conference, 15 March, at Grand Rapids, Michigan.

Girard, B, and G. Kopp. 1996. Physicomecanical characteristics of selected
sweet cherry cultivars. Journal of Agriculture and Food chemistry 46: 471-476.

Glass Packaging Institute. 2006. Why glass? <www.gpi.org/why/> (July 24,
2006)

Goff, D. 1995. Dairy science and technology. University of Guelph.
<http://www.foodsci.uoguelph.ca/dairyedufrDT.html> (July 24, 2006).

177

 

 

Gomez-Cordovés, C. 2004. New pigments produced in red wines via different
enolological processes. In Red _wine color: eanflhe mysteries. Edited by
Andrew L. Waterhouse and James A. Kennedy. Washington DC: American
Chemical Society (7): 89-124.

Gemez-Plaza, E., Pérez—Prieto, L.J., Martinez-Cutillas, A., and Lepez-Roca, J.M.
(\2004. Color and phenolic compounds of oak matured wines as affected by
characteristics of the barrel. In Red wine color: exploring the mysteries. Edited
by Andrew L. Waterhouse and James A. Kennedy. Washington DC: American
Chemical Society (3): 22-34.

Good Fruit Growers 2006. Industry web site to address labeling problems.
<http://www.goodfruit.com/issues.php?article=186&issue=7> (February 2nd
2006).

Halliday, J. 2007. medruits flavours get ever more exotic. Food Navigator
Europe. < http://www.foodnavigator.com/FinanciaI-lndustry/Superfruit—flavours-
get-ever-more-exotic> (August 23 2008).

Halloway, D.G. 1973. The physical properties of glass. London: Wykeham
Publications Ltd.

Harborne, JB 1967. Cormarative piochemistryof the flavonoid_s. New York:
Academic.

Hall, R.L. 1968. Food flavors: penefits and problems. Food Technology
2221388.

Hemerson, AC, and ED. Hullard. 1980. Canned foods. Thermal Processing
and Microbiology. 7th edition. Edinburgh London and New York: Churchill and
Livingston.

Hernandez, F., Malgreko, P., Tomés-Barberan, F.A., Artés, F. 1999. Evolution
of iuice anthocyanins duridq ripening of new selectedpomeflnated (punica
granatummlones. European Food Research Technology 210(1):39—42.

Hicks, D. 1990. Production and packagi_ng of non-carbonated fruit iuices apd fruit
beverages. Glasgow: Blackie and Son Ltd.

Hippocrates. 400 BCE. Aphorisms. Classics Archive.
<http2/lclassics.mit.edu/Hippocrates/aphorisms.3.iii.html> (August 29 2008).

Hobbs, J.E. 2001. Developing supply chains for nutraceutical and functional

foods: Opportpnities and challenges. Université de I’Aval, Canada: Research
Series SR.01.05 (part 4).

178

Hong, V., and R. Wrolstad. 1990. Characterization of anLhocyanin-containing
colorants and fruit iuices py HPLC/photodiode array detection. Journal of
Agriculture and Food Chemistry 38: 698-708.

Hong, W., Cao G.,and Prior, P. 1990. Oxygen radical absorbing capacig of
anthocyanins. Journal of Agriculture Food Chemistry 45: 304-309.

Hook, P., and J.E. Heimlich. 2006. Packaging Histogy. Ohio State University
Fact Sheet <http://ohioline.osu.edu/cd-factl0133.html> (July 24, 2006).

Hotchkiss, J.H. 1997. Food-packagi_ng interactions inflpencing gual'dy and
safey. Food Additives and Contaminants 14: 601-607.

lbarz, A. J., Pagan, R., Panades, and Garza, S. 2005. Photochemical
destruction of color co_rnpounds in fruit igices. Journal of Food Engineering 69
(2): 155-160.

Impact marketing consultants Inc. 2006. Marketinpqguide to the U.S. pafckagipg
industm. Manchester Center, VT: Impact Marketing Consultants Inc.

International dairy foods association. 2005. PasteLization.
<http://www.idfa.org/facts/milk/pasteur.cfm> (July 24, 2006).

International Rm Information Copncil. 2004. F_unctiona| food;
<http:l/ific.org/nutrition/functional/index.cfm> (April 14, 2006).

Irish food packaging information service. 2005. Glass. <www.ucc.ie/c/PKglass>
(July 3, 2006).

Jensen, J. 2005. Tart cherry sales trend. Paper read at North American Tart
Cherry Industry Conference, 15 March, at Grand Rapids, Michigan.

Jiang, Y., Li, Y. and Li, J. 2004. Browningcontrolpshelf-life extension and
gu_a|ity maintenance of frogen litchi fruit by hydrochloric apid. Journal of Food
Engineering 63 (2): 147-151.

Kalra, E.K. 2003. Nptraceutical- Definition and Introduction. American
Association of Pharmaceutical Scientists 5 (3): 25.

Karipidis PI. and K. Galanopoulos. 2000. Food market value analysis: Product
Quality improvement. product of origin protection and timingpecisions in applp
market. Agricultural Economics Review 1 (1)249-56.

Kearsley, M.V. and N. Rodriguez. 1981. The stability and use of natgral colors in

food: Anthocyanins, B—carotene and riboflavin. Journal of Food Technology 16,
421-431.

179

Kim, D-O., Heo,H.J., Kim,Y.J., Yang, HS. and Lee, CY. 2005. Sweet and sour
cherry phenolics andmeiflrotective effects on nepronal cells. Journal of
Agriculture and Food Chemistry 53: 9921-9927.

Kurylowski, M. 1996. Glass Properties and Applications. Packaging in the Food
Industry. Leatherhead, UK: Pira International 1996:12.

La Cross, B., Raak, T. and Dorsing, T. 2005. dpice concentrates market
segment analysis. Paper read at North American Tart Cherry Industry
Conference, 15 March, at Grand Rapids, Michigan.

MacDougall, DB. 2002. Colour measurements of food: Principles and practice.
In Colour in food. Edited by Douglas B. MacDougall. Boca Raton FL: Woodhead
Publishing Limited (3): 33-63.

Marquez, C.A., Salcadori, V.O., Mascheroni, RH. and De Michelis, A. 2003.
@plication of transfer functions to the thermal processing of sweetgand sour
cherries preserves : lnflpence of particle and container siz_e§_. Food Science
Technology International 9 (2): 0069-8.

Masayuki, Y. and Y. Asahara. 2005. Glass properties. Cambridge, UK:
Cambridge University Press.

Mattheis, J.P., Buchanan, DA. and. Fellman, J.K. 1992. l_cLentification of head
space volatile commrnds from Bing sweet cherry fruit. Phytochemistry 31: 775-
777.

Mattheis, J.P., Buchanan, DA. and Fellman, J.K. 1997. Volatile constitpents of
bing cherry fruits following controlled_at_mosphere storage. Journal of Agriculture
and Food Chemistry 45: 212-216.

McDonald, SA. 2004. A comparative analysis of the regulatory framework
affecting functional fogs and functional food ingredient development 1051
commercialization in Canada, the United States. the European Union. Japan.
A__us_tralia and New Zealand. Agriculture and Agri-Food Canada.
<http://www4.agr.gc.ca/AAFC-AAC/display-

afficher.do?id=1 172765386339&lang=eng>(November 15, 2008)

Meilgaard, M., Civille, G.V. and Carr, ET. 1999. Sensory evalpgtion technidues.
Boca Raton, FL: CRC Press.

Meheriuk, M., Girard, B., MoyIs, L., Beveridge, H.J.T, McKenzie,D. Harrison, L.

J., Weintraub, S., and Hocking, R. 1995. Modified atmosphere packaging of
wins” sweet cherry. Food Research International 28 (3): 239-244.

180

Meroni, A. 2000. Active Packagjpg as an opmrturnitv to create package desigp
that reflects the commdnication,flnctional and logistical requirements.
Packaging Technology and Science 13: 243-248.

Michigan Cherry Committee. 2003. Tart Cherries cum statistics and market
analysis. Grand Rapids, MI: Michigan Cherry Committee.

Michigan Cherry Committee. 2005. The emerging natdgai foods: market for tart
cherry products. Grand Rapids, MI: Michigan Cherry Committee.

Nankivel, B. 1997. Glass properties and applications. Paper read at the
Conference proceedings of packaging in the food industry 10—11 September, at
Letterhead, UK: Pira International.

NBC News. 2004. Beer in aluminum bottles on the way.
<http://www.msnbc.msn.com/id/5810359/?GT1=4529> (January 13, 2006).

Nebesky, E.A., Esselenn W.B. Jr., McConnell,J.E.W., and Fellers CR. 1949.
Stability of color in fruit idices. Food Research 14: 261-274.

Nielsen, T.J., and. l.M. Jagerstad, 1994. Flavor scalping by food packagi_ng.
Trends in Food Science and Technology 5: 353-356.

Nill, K. 2001. gossary of biptechnology, Technique Publishing Co.
<www.biotechterms.org> (June 03, 2006).

Nugent, S., Bogard, D., and R. Atwater. 2005. _Dry cherry: Market segment
analysis. Paper read at North American Tart Cherry Industry Conference, 15
March, at Grand Rapids, Michigan.

Nutrition Business Journal. 2001. Functional foods rejort 2001.
<http://www.nutritionbusiness.com/>, (July 24, 2005).

Nykiel R.A. 200). Marketing your business: A guide to developing a strategic
plan. New York and London: Haworth press, Best Business Books.

Paul, P., and Palmer, H.H. 1972. RM theory and applications. Hoboken,
NJzJohn Wiley and Sons, Inc.

Packaging Council of New Zeeland. 2004. PET plastigackagLng.
<www.ronz.org.nz>, (July 27, 2006).

Packaging Digest. 2004. Four panel stand up pouch gets sporty for Hydarde.

<http://www.packagingdigest.com/toc-archive12004l20040201 .html, (August 22,
2008).>

181

PC Magazine. 2008. Definition of choma.
<http://www.pcmag.com/encyclopedia_term/0.2542,t=hue&i=44520,00.asp>
(July 25 2008).

Pindyck, RS, and BL. Rubinfeld. 2003. Microeconomics. Fifth edition. Upper
Saddle River: Prentice Hall.

Plastic Technology. 2005. Panel-less PET bottles win first commercial use.
<http:I/findarticles.com/p/articles/mi_hb6458lis_200512/ai_n25676335>, (August
28, 2008)

Plastic Packaging Innovation News. 2005. Soft opponent: development in
flexible packaging for drinks. Plastic Packaging Innovation News 1(19): 5.

Pollard, J. 1990. Lagislation for proddction. labeling and marketing. Glasgow:
Blackie and Son Ltd.

Raloff J. 1999. Arthritis ca‘rezgeyond tea and sympathy. Science News
155(16): 247.

Raso, J., Calderon, M.L., Gongora, M., Barbosa-Canovas, G.V., and Swanson,
B.G. 1998. Inactivation of msaccharomyces hailii in fruit iuices by heat, high
hydrostatic pressdre and pulse electric fielda. Journal of Food Science 83 (6):
1042-1043.

Reddy, AR. 1996. Genetic and molecular analysis of the anthocyanins
pigmentation in thejathway of rice. Paper read at the third international rice
genetics symposium, at Manila Philippines.

Reim, M. 2005. Copigmentation reaction and color stability of berrv
anthocyanins. PhD diss. University of Helsinki.

Remén. S., Ferrer, A., Lopez-Bueza, P., and Oria, R. 2004. Atmosphere
composition effect on bufirlat cherry color duridd coldstorapg. Journal of Science
Food Agriculture 22: 5142

Reuch, W. 1999. VirtJuaI textpook of organic chemist_ry.
<http://www.cem.msu.edu/~reuschNirthtJml/intro1.htm#contnt> (August 02
2006)

Righettlo, A.M., Netto, FM, and Carraro, F. 2005. Chemical composition and
antioxidant activity of iuices from mature and inmature acerola (malpighia
errfiginata DC). Food Science and Technology International 11(4): 315-321.

Risch, S.J. 1999. New development in packagi_ng materials. Food packaging
testi_ng methods and applications. Washington DC: American Chemical Society.

182

Romero del Castillo, R., Valero, J., Castanas, F. 2008. Training validation and
maintaining a panel to evaluate the textdre of day beans (Phaseolus VulgriSL
Journal of Sensory Studies 23 (3): 303-319.

Ros-Chumillas, M., Belisario, Y., Iguaz, A., and Lopez, A. 2006. Quality and
shelf-life of orange iuice aseptically packagad in PET bottles. Journal of Food
Engineering 79 (1): 234-242.

Ross, R. and K. Weller. 2008. Effect of serving temgrature on the sensory
attribuLtes of red and white wines. Journal of Sensory Studies 23 (3): 398-416.

Rowley, C. 2005. Proddction arfl trends & projections. Paper read at North
American Tart Cherry Industry Conference, 15 March, at Grand Rapids,
Michigan.

Sandgren, K. 1996. Material flow analysis - a case stddy in packagmg.
Nonrenewable Resources 5 (4)2235-247.

Saucier, C., Lopes, P., Mirabel, M., Guerra, C., and Glories, Y. 2004. Tannin-
anthocyanin interactions: Influence on wine color. In Red wine color: revealing
the mysteries. Edited by. Andrew L. Waterhouse and James A. Kennedy.
Washington DC. American Chemical Society (16) 265—273.

Scaman, C.H., Dou, J., Cliff, M.A., Yuksel, D., and King, MC. 2000. Principal
component similarity analysis for evaluation of wine Ludggperformance. Journal
of Sensory Studies 16: 287-300.

Scott Wolfe Management Industry. 2002. Agriculture and Agri-Food Cana_da.
<http:/Iwww4.agr.gc.ca/AAFC-AAC/display-
afflcher.do?id=1172236805144&Iang=eng> (November 15, 2008).

Schanda, J. 2007. Colorimetry: Understamlidg the CIE system. Hoboken,
New Jersey: VWey-interscience..

Selke, S.E.M., Culter, JD, and Hernandez, R. 2004. Plastics packaging:
Properties, Processing, Applications and Regulations. 2nd edition. Munich,
Germany: Hanser Publishers.

Serrano, M., Guillen, F., Martinez-Romero, D., Castillo, S., and Valero, D. 2005.
Chemical constflrents and antioxidant activity of sweet cherry at different
ripening stages. Journal of Agriculture and Food Chemistry 53: 2741-2745.

Statistics Canada. 2006. Canada’s populationfiby ax and sex.
<http://www.statcan.gc.ca/Dain/English/061026/d061026b.htm> (July 15 2008).

183

Skrede, G., Wrolstad, E.E., Lea, P., and Enersen, G. 1992. Color stability of
strawberry and black cdrrant syrdpg Journal Food Science 57:172-177.

Solberg, P., Castberg, H.B., and Osmundsen, J.|. 1990. Packaging systems for
fruit iuices Ed non-carbonated beverag_e_s_. International Journal of Food
Science and Technology 27 (1):99-110.

Steinka, l., Morawska, M., Turowska M., and Kukulowski, A. 2006. m
influence ofbiological factors on mpperties of some traditional and new polymers
used for fermented food_packaging. Journal of Food Engineering 77:771-775.

Stintzing F.C., Stintzing, A.C., Carle, B., Frei, B., and Worlstad, RE. 2002.
Color and antioxidant properties of cyanidin pigments. Journal of Agriculture and
Food Chemistry 50:. 6172-6181.

Stone, H., and J.L. Sidel. 2004. Sensory Evaluation Practices. Food_§cience
yd Technoldgylnternational Series. Redwood City, CA: Elsevier.

Sweeting, L. 1999. Acids lelephiles.
<http://pages.towson.edu/Iadon/orgrxs/reagent/acids.htm> (December 10 2006).

Thornsbury, S. 2005. International market isspes. dp_date on Poland. Paper
read at North American Tart Cherry Industry Conference, 15 March,_ Grand
Rapids, Michigan.

 

Timmer, C. P., Falcon, W., Pearson, S. 1983. Food_policy analysis. Baltimore,
MD: Johns Hopkins University Press for the World Bank.

Toyo Seikan. 2006. Personal communication discdssimossibilities of PET
bottle delamination. School of Packaging, Michigan State University (February15
2007)

Turker, N., Erdogdu, F. 2005. Effects of pH and tempera:ture of extraction
medium on effective diffusion coefficient of anthocyanins pigments of black carrot
@ucus carota var. I=._)_._ Journal of Food Engineering 76: 579-583.

 

U.S. Center for Food Safety and Applied Nutrition and Office of Food Additive
Safety. (2006, August). Use of recycled plastics in food packaging: Chemistry
considerations <http:l/www.cfsan.fda.govl~dms/opa2cg3b.htmI> (August 22
2008) ‘

U.S. Census Bureau. 2005. Glass Containersg005. U.S. Department of

Commerce. < http://www.census.gov/cir/wwwl327/m327g.htm|> (September 13,
2006)

184

U.S. Department of Agriculture Michigan Field Office, NASS. 2005. Michigan
Amiculture Statistics 2004-@05. Michigan Department of Agriculture 2004
Annual Report. < http://www.michigan.govlmda/0,1607,7-125-1572_2885_2888-
109498--,00.html> (September 18< 2006).

U.S. Department of Agriculture. 2001. Tart cherries grown in the state of
Michigan. Suspension of provisions under the federal Marketing Order for tart
cherries. 7 CF R Part 930. <http://www.ams.usda.gov/fv/mofinaldockets/FVO1—
930-5FIR.pdf> (August 6, 2006).

U.S. Food & Drug Administration. 2003. Guidance for indpstry and FDA: Interim
procedures for qualified health claims in the labeling of conventional human food
arfi human dietary sdpplements.

< http://www.cfsan.fda.gov/~dmas/hclmgui3.htlml> (July 24 2005).

U.S. Food and Drug Administration. 2004. Federal Food Drug and Cosmetic Act.
Chapter 5. <http://www.fda.gov/opacom/laws/fdcact/fdctoc.htm> (June 20,
2005)

Van Oirschot, Q.E.A., and Tomlins, KL 2005. Applyindanalytical sensory
evaluation techniqpes, which translates gpalitativeperceptions to numerical data
to research on development issues. Natural Resources Institute, University of
Greenwich, UK. <http://www.swan.ac.uk/cds/pdffiles/OIR'SCHOT.pdf> (June 15,
2006).

Van Williege, R.W.G. 2002. Effects of flavor absorption on foods and their
packaging materials. PhD diss., Wageningen University.

Von Elbe, J.H., and S. Schwartz. 1996. Colorants. In Food chemistry. Edited by
Owen R. Fennema. 3rd edition.New York: Marcel Dekker, Inc.

Wade, M. 2006. Defending functional foopp. Prepared Foods September 84-97.
Wang, H., Nair, M., Stasburg, G.M., Booren, A.M., and Gray, I. 1999.
Antioxidanmlyphenols from tart cherries (prunus cerasus). Journal Agriculture
Food Chemistry 47: 840-844.

Waters. 2008. LE premier mcifications. Waters Corporation
<http://www.waters.com/> (July 22, 2008).

Wesseling, J.A. and R. Krishna. 2000. MassLansfer in mdlticomponent
mixtures. Delf, Netherlands: Delf University Press.

Whaling, A. (2007). The effect of packagi_ng attrib_utes on consumer perception
of cherry iuice. Master’s thesis, Michigan State University.

185

Will, F. and H. Dietrich. 2006. Optimized processinmchnigpe for color — and
cloud-stable p_lt_Jm idices arfi stability of bioactive substances. European Food
Research and Technology 223: 419-425.

Zepp. G., Harwood, J., Harmond, C., and Somwaru, A. 1996. Tart cherries: An

economic assessment of the fea_s_ibility of providing multiple-peril crop insu_r_ance.
< http://www.rma.usda.gov/pilots/feasible/pdf/tcherry.pdf> (January 13 2006).

186

   

”'illilllilllllflllljilllillllllilll“