- ”91%?“ I 333933;“. ,. V33.
‘33 \3 “W“ 31331-3 \‘

.II‘EQILhIII-h: 63% 3(1)?” 331‘ .
33‘ I .333 "' W. r .

.... I

.3‘3 8-} web-r." W’Vf

33313.7."
'7} -I '_ I‘..3.3..v:::3||3u
‘ 333003333 \' “in. ‘I3I
“331‘?- . 0%
{‘3

I' ‘~‘ I,“
3.3: 2335.5: 2?:
N' 2’33"“:

. I ‘3:
. 'SJ'SI3'\

'33:“: }‘0

33.1
, _ "Eli?

4‘<4~|o ,.

V---._

n
' . JH'
.
|
{II P. ‘
3 3.. (1".3/

fl

£33.,3‘I3I333333

“.3313”
r. I
W 313% 3%»
.3VJ'
c-IIA.% ‘2 {'3' ‘3":-3:.‘)‘3".7 3‘3 31“}
I:-*?*'?“I»:I~,:f3v 3.. 33.333 . .. ‘ ‘3.“ 2 3
I 4 Kil“

I
J: I 'I'u

ll

‘ Y |‘

If? II‘»I"".‘..-l
‘ ' ‘3". AI'I‘u

33'.

.”.-v 33".) '3
$.A‘INH"";‘ 352,1.“
'. 'l|'

I

T31: :‘I‘JIJ' I‘ '
' ‘ I
' "1‘ \I |

' \

II.

\
I II‘H".
‘7‘. “qu”:

3‘

3' . ‘|4 ‘
I
:"II “I. 3 h ‘

1‘ 923;
:33

#Il d“...
x'ng‘

. ' ;\-3;3
IV‘::|II ‘3.“
I» a
«#3.
133315;»
'03“ it?
' $3} I‘m/5)“. fly 3 3 {2'35
-‘:‘. III'. "' . lorv“. II:\\$\‘:‘|::‘:\I" I _
I 1333333339393 Irw'dl‘fl 33 3: “1-" 3
”1931:?“ I}; ‘.I v.3 .: 3

a?“ ’ II\'. “II I

3'35}. . _ _
33“ 3133?; 3335'“ .'¢ """"\
PA,‘ 3“

 

3“ IVY-4,".
Ii.“ {125313 33.3»: I.. 3353‘:
TI 31w .. '3; gift

14'

 

try “'31,.

mum-3'3"

 

 

 

 

 

 

llll ll'ellllllllllllllll‘lllllll L

3 1293 00823 6121

 

 

This is to certify that the

thesis entitled

PROCESSING OF GREEN BEANS IN
RETORTABLE POUCHES

presented by
Vicente de Pau1a Pereira
has been accepted towards fulfillment

of the requirements for

M. S . degree in FOOd Science

 

Major professor (—
new

0-7639

I

-c‘r
\. '_p‘ 9 a in

PROCESSING OF GREEN BEANS IN

RETORTABLE POUCHES

BY

Vicente de Paula Pereira

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

MASTER OF SCIENCE
Department of Food Science

1978

4‘ I
- ' 1‘ l
b ‘» 1.x“.

ABSTRACT

PROCESSING OF GREEN BEANS IN
RETORTABLE poucans
by

Vicente de Paula Pereira

Green beans were thermally processed in 7 x 10-
inch pouches and in no. 303 cans. Quality comparisons
were then carried out, using texture, color, and thiamine
retention as criteria of quality.

Heat penetration studies were carried out using
copper-constantan thermocouples, held at the middle plane
of the pouch by steel wire spring. Lethality F0 was
calculated by the Improved General Method.

Processing time requirements for different pouch
treatments in water at 250°F were found to vary from 10.24
to 12.91 minutes to give an F0 3 3.5 with 0.997 confidence
coefficient. In order to give equal heat treatment, 303
x 406 cans were processed in steam at 250°F for 12 minutes.

Green beans processed in pouches were firmer in
texture and higher in thiamine value than canned beans
that had received the same sterilization equivalent, but
no statistically significant differences in color were

found.

ACKNOWLEDGEMENT

The author very sincerely wishes to express
his thanks to Dr. T. Wishnetsky, his advisor, for
encouragement, patient guidance in his graduate study
program and for his valuable suggestions and assistance
in the preparation of this thesis.

Also, to his guidance committee members,
Dr. P. Markakis, Dr. L. E. Dawson and Dr. S. W. Gyeszly
his sincere thanks for their assistance during the prepa-
ration of this work.

Special thanks are given to H. E. Strassheim
from the Continental Diversified Industries Flexible
Packaging Division for his contribution of pouch material

for this research.

ii

TABLE OF CONTENTS

LIST OF TABLES O O O O O O O O O O O O O 0
INTRODUCTION 0 O O O O O C O O C O O O O 0
LITERATURE REVIEW . . . . . . . . . . . . .

Retortable Pouch Material . . . . . . .
sealing O O O O O O O O O O O O O O O 0

Residual Gases Counterpressure Relationship

Temperature Measurement . . . . . . . .
Process Determination . . . . . . . . .
Processing Effect on Color . . . . . . .
Thiamine Stability During Processing . .
Texture Changing During Processing . . .

EXPERIIdENTA-L O O O O O O O O O O O O O O 0

Determination of Process Conditions . .
Green Beans . . . . . . . . . . . . . .
Experimental Design . . . . . . . . . .
Quality Evaluation Methods . . . . . . .
Thiamine . . . . . . . . . . . . . . . .
Extraction Procedures . . . . . . . .
Analytical System for Thiamine
Determination . . . . . . . . . . .

Texture . . . . . . . . . . . . . . .
Procedure Used . . . . . . . . . . .
Processed green beans . . . . . . .

Raw Product . . . . . . . . . . . .
Calculations . . . . . . . . . . .

COlor O O I O O I O O 0 O O O I O O O 0
Moisture Determination . . . . . . . . .

RESULTS AND DISCUSSION . . . . . . . . . .

Heat Penetration Study . . . . . . . . .
Effect of Superimposed Air Pressure on

Time of Processing . . . . . . .
Effect of Pouch Thickness on Time of
Process lng O O O O O 0 O O O O 0

Effect of Brine on the Processing Time

Page

. 21
. 24
. 27
. 28

O 28

. 30
. 30
. 30
. 3O
. 30

. 31
. 31

. 33

TABLE OF CONTENTS (cont'd)

Page

Quality Evaluation . . . . . . . . . . . . . . . . . 38
Thiamine Retention . . . . . . . . . . . . . . . 38

Effect of pouch thickness on
thiamine retention . . . . . . . . . . . . . 38

Effect of brine addition on

thiamine retention . . . . . . . . . . . . . 38

Effect of superimposed air on
thiamine retention . . . . . . . . . . . . . 42

Texture O O O O O O O O O O O O O O O O O O 0 O O 45
C01or O O O O O O O I O O O O O O O O O O O O O 0 50
Effect of pouch thickness on color . . . . . . 52
Interaction of thickness versus brine
addition on color . . . . . . . . . . . . . . 52
Interaction of superimposed-air pressure
versus brine addition on color . . . . . . . 57
SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . 58
Texture O O O I O O O O O O O O O O O O O O O O O 59
Thiamine . . . . . . . . . . . . . . . . . . . . 59
COlor O O O O O O O O O O O O O 0 O O O O O O O O 59
REFERENCES 0 O O O O O O O O O O O O O O O O O O O O O 61
APPENDIX A O O O 0 O O O O O O O O O O O O O O O O O O 66

APPENDIX B . . . . . . . . . . . . . . . . . . . . . . 70

Table

1.

10.

11.

12.

LIST OF TABLES

Treatment notation with levels used
for each factor . . . . . . . . . . .

Processing time in minutes for test
pouches and cans . . . . . . . . . .

Process lethality (F0) calculated by
the Improved General Method, using
center temperature at 0.5 minutes
intervals, and assuming Z = 18°F . .

Time of processing required for pouches
in water at 250°F to give a F0 3 3.5
minutes 0 O O O O O O O O O O O O O O

Thiamine content in ug/gram (dry weight
basis) in processed green beans from
pouches and 303 cans . . . . . . . .

Analysis of variance of data in Table 17

Effect of brine addition on thiamine
retention in pg/gram in green beans

processed in pouches in water at 250°F

Change in thiamine content in green beans

during processing . . . . . . . . . .

Effect of superimposed air pressure (P)

on thiamine retention ug/gram in green
beags processed in pouches in water at

250 F . . C O O O O O O O O O O O O 0

Texture of processed green beans measured

by the shear press and converted to
pounds force/gram - . . . . . . . . .

Analysis of variance of data in Table 18

Examination of the thickness versus brine
addition interaction for the data from

Table 18 . . . . . . . . . . . . . .

Page

25

26

34

35

39
4O

41

43

44

46

47

49

LIST OF TABLES (cont'd)

Table Page

13. Color measurement of processed green beans
with Hunter Lab D25 Color and Color-
Difference Meter. Hunter Lab Standard
Cell No D33 C20-60 Data transformed to
-a/b . . . . . . . . . . . . . . . . . . . . 51

14. Analysis of variance of data in Table . . . . 53
15. Examination of thickness versus brine

addition interaction for data in Table 19 . 54
16. Examination of the brine addition versus

superimposed air pressure interaction
for data in Table 19 - - - - . . . - . . . . 55

17. Thiamine concentration in ug/gram (dry

weight basis) of processed green beans

in pouches . . . . . . . . . . . . . . . . . 70
18. Texture in pound force/gram of green

beans processed in pouches . . . . . . . . . 71
19. Color ratio -a/b of green beans processed

in pouches . . . . . . . . . . . . . . . . . 72

INTRODUCTION

The use of a flexible polymeric-foil laminate
pouch to contain and maintain the wholesomeness of thermo-
processed foods has been proved to be technically and com-
mercially feasible over 16 years of research and over nine
years of commercial experience in Japan, Europe, and more
than two years in Canada.

Flexibly packaged, thermoprocessed, food items
may be compared with items packed in the common three-piece
steel plate or drawn aluminum sanitary can. In each
instance, heat sterilization and the air tightness of the
container assure sterility up to the point of use. How-
ever, the outward appearance of the flat pouch and the con-
veniencecnfreheating by immersion of the unopened pouch in
hot water invite a comparison with frozen "boil-in-bag"
entree items.

Reduction in processing time of 30 to 50 per cent,
may be achieved for equivalent sterilization treatments
because the pouch has a thinner profile than the rigid
can. Improved product quality is possible as the product
around the periphery receives less overcooking.

Although overall production costs of retortable
pouch products enclosed in paperboard cartons are currently

slightly higher than those of canned and frozen foods,

distribution and storage costs are basically lower for
the pouched products. For these and other reasons, the
retort pouch is a rapidly growing commercial reality.
Commercially, at this time, most foods in retor-
table pouches are entrees or sauce items. Only in Britain
are there a number of flexibly-packaged vegetable products,
and many of these are processed in semi-rigid containers.
Further work is required to explore the potential for

vegetable products in retortable pouches.

LITERATURE REVIEW

Retortable Pouch Material

 

Retort pouch wall is formed by three layers
bonded by adhesive; an outer layer of polyester for
strength, a middle layer of aluminum foil, as moisture,
light, and gas barrier and an inner layer of polyolefin
(high density modified polyethylene) or polypropylen
ethylen copolymer which comes in contact with the food
material and has heat sealing properties. In order to
bond these three layers together an appropriate adhesive
system should be used. The adhesive used must keep the
layers together and not migrate through the inner layer
to the product. Toluene diisocyanate (TDI), one of the
reactants necessary for the adhesive, was found to be
extracted by the food-simulating solvents. This extract
revealed traces (0.3 to 3 parts per billion) of toluene
diisocyanate. As described by Goldfarb (1973) a
polyester-isocyanate is used to form the adhesive. First
the polyester polymer is formed by the reaction of a
dihydric alcohol with a dibasic acid. Thus the polymer
is mixed with a coreactant diisocyanate (TDI sometimes
called catalytic), and heated. The condensation product
is an adhesive. This adhesive is specially applicable to

retort pouches, since it has a strong affinity for foil

and the isocyanate component reacts with surface hydroxyl
and carboxyl groups of the polyester and available carboxyl
groups of corona-treated polyolefins. Toluene diisocyanate
although an irritant, is quite nontoxic, Lampi (1977)

(LD of 5.8 mg/kg body weight).

50
Beyond the inherent low extractive property of
retortable pouch films, suppliers and researchers (Duxbury
‘2; al., 1970; Torpe and Atherton, 1972) recognized from a
practical production aspect that quality assurance sur-

veilance is mandatory on roll stock and for preformed
pouches to guarantee both freedom from odor and removal
of solvent.

Film specifications of current suppliers (Duxbury
gt 31., 1970), include retortability or sterilization
resistance requirements. In Japan, cited by Lampi (1977),
products are being commercially packed in a film described
as 12-micron polyester/Q-micron aluminum-foil/15 micron
oriented Nylon 6/50-micron polypropylen and processed at
temperatures up to 275°F for times ranging from 2.7 to
9 minutes.

The pouches discussed thus far have been retorted
at temperatures in the range from 2400 to 250°F. There
are recent film developments that permit processing in
the 2750 to 293°F range. Studies reported by Komatisu
and Yamaguchi cited by Lampi (1977) show that vegetables

retain favorable color, and seafoods such as crab, shrimp,

and Whitefish, are good candidates.

'Sealing

Sealing of pouch refers to forming of pouches from
the roll stock, regardless of whether the food packing
facility uses preformed pouches or roll stock with one
side pouch fabrication. It also includes forming of the
two side seals and a single bottom seal and the final
sealing of the closure after filling and air removal. The
reliability of the closure seal is directly affected by
the ability of the filler operations, to leave the seal
surface free of product contamination.

Until now there is not a definition of good seal
and standard method to evaluate the performance of seal.
For the retort pouch, this standard has been the sanitary
can and its historically documented satisfactory perfor-
mance. Burke and Schultz(l972) compared the resistance of
metal can and retort pouch to a rough-handling cycle repre-
sentative of the military distribution system. They found
a 2% or less leaker after biotesting for both metal can
and pouches.

Four different tests have been performed by dif—
ferent investigators, Brown and Keegan (1973); Duxbury e:
31. (1970); Pflug and Loong (1966); R.W.P. Flexible
Packaging (1974); Shenkenberg (1975), in order to evaluate
the pouch seal: (1) Fusion: Fusion is a requirement that

is met when the opposing seal surfaces form a dual weld.

Such a weld is characterized by the inability to distin-
guish visually the inner opposing seal surface at the
inner seal junction after seal tensioning. (2) Internal
burst test: Pressure and time to burst at that pressure
is recorded. (3) Tensile test: This is currently
measured dynamically on Instron Universal Testing Machine
or similar equipment. (4) Visual examination: To detect
the absence or presence of fusion, heat creep, significant
wrinkles, occluded matter in the seal area. Lampi gt 31.
(1976) in their studies on retorting and storage effect
on seal strength reported that seal strength was lower
after retorting than before retorting and this effect was
produced by temperature effect and not by the kind of
product in the pouch. No differences were found in seal

strength during storage.

Residual Gases Counterpressure Relationship

 

With the exception of bakery products where the
dominant residual component is CO2 from the leavening re-
action it is generally recognized that residual gases in
the retort pouches should be kept as low as possible to
preclude chances of rupture during a retort cook cycle.
Davis 35 31. (1960) devised apparatus and studied the pouch-
retort pressure relationship during processing. They
enumerated four causes of internal pressure increase as:

(1) increase in the vapor pressure of the water in the

processed food with increasing temperature; (2) increase

in the pressure of air in the headspace with increasing
temperature; (3) release of additional air from the pro-
duct, due to a decrease in gas solubility with increasing
in temperature; and (4) thermal expansion of the food
itself. Yamano and Komatsu (1969) detailed the relation-
ships among internal pressures, residual gas volumes,
process temperatures, headspace expansion ratios and steam-
air ratios. Their studies permitted calculations of
required counterpressures to prevent failure. Rubinate
(1964) in his studies concluded that an air pressure of
3 to 10 psig above the equilibrium pressure for water or
steam at the process temperature over the total period of
the cook is desirable; the reason is pouch restraint to
prevent agitation and movement that could rupture the seals.
Aside from assurance of package integrity, super-
imposed air pressures, according to Keller (1959) and
Nelson and Steinberg (1956), should be maintained to improve
heat transfer. Japanese investigators cited by Lampi (1977)
support this contention in relation to residual gases and
offer data to indicate that the head space gas can also
interfere with sterilization. Increasing head space gas
volumes from 0 to 15 cc resulted in changes in fh values
from 6 to 7.2, and sterilization times at 250°F from 6.7
to 7.1 minutes. When 20 cc of occluded air was incorporated
into the product (but no head space gas, per se) the fh

increased to 9.4 and the sterilization time to 12.5 minutes.

Temperature Measurement

 

Temperature measurement for determining heat pene-
tration rates generally have been by conventional copper-
constantan thermocouples for preliminary heat penetration
experiments and batch-production systems.

Where product characteristics, pouch-holding
techniques, and pouch placement within the retort permitted,
metal packing glands such as those used for conventional
cans, Pflug gt gt. (1963) have been successfully employed
to introduce wire and thermocouples into the pouch. Pflug
gt gt. (1963) described a procedure whereby 30-gage thermo-
couple lead wire was introduced through the seal area.

The section of wire that passed through the seal area was
stripped of all insulation, cleaned with solvent, and
coated with lacquer that is heat sealed to the innermost
layer of the laminate. Torp and Atherton (1972) intro-
duced thin, lacquered cooper—constantan wires through
small holes made in the laminate aboutCL75 inch above the
bottom seal and sealed with a small amount of adhesive.
Once the lead wire has been introduced into the package,
positioning can be controlled by several means. Pflug gt
gt, (1963) used a plastic saddle with the metal packing
gland to orient the thermocouple tip to the geometric
center of the pouch for liquid or semiliquid food under
investigation. Pflug (1964) also reported using a piece

of wood placed diagonally across the pouch with the

thermocouple stapled onto the center. Torpe and Atherton
(1972) used a rigid nylon spacer. With recipe packs, such
as sweet and sour pork and beef stroganoff, the tip of the
spacer-positioned thermocouple was secured in the center

of a piece of meat of known size. Dewey Redesign (1975)

. described a disposable thermocouple/pouch arrangement where
the lead wire enters through the pouch wall through an
epoxy sealant and the thermocouple junction is suspended

in the center of a conical coil spring.

When continuous retort is used, a different system
should be used for heat penetration measurement. Goldfarb
(1970, 1971) described and illustrated a temperature tele-
metry system in which a transistor sensor is inserted into
the pouch through a packing gland similar to those used
for conventional wire systems. The thermistor lead is
connected to a battery-powered transmitter. The pouch
transmitter combination, as it passes through the continuous
retort, sends radio-frequency signals that are converted
as temperature in a digital telemetry receiver. This
system has been successfully used for temperature measure-
ments on vegetables packaged in retort pouches using a

continuous retort as reported by Lampi (1977).

Process Determination

 

ThermOprocessing conditions have been generally
determined for pouches by methods proved satisfactory over

decades of use for metal cans. Since the microbial history

10

and growth environment (foods) are identical, the same
concepts related to microbial survival have been fully
accepted for direct applicability to retort pouches:

a. Fo - lethality in terms of minutes at 250°F required
to destroy the specified spoilage organism in a specific
medium assuming "Z" is equal to 18°F.

b. D - time in minutes to accomplish a 90% reduction in
number of organisms (or spores).

c. Z - shape (Slope) of the thermal death-time curve in
degrees Fahrenheit required for the curve to transverse
.one log cycle (D versus oF).

Large-scale production tests such as those
reported by Duxbury gt gt. (1970); Goldfarb (1970) indi-
cated that F0 values suitable for commercially canned
products are generally adequate for retort pouches. Lampi
(1977) recommended that thermal-process parameters, as
described by Ball and Olsen, can be established and should
be confirmed using inoculated packs, but calculation tech-
niques reduce time and permit translation of results from
one set of conditions to another. These methods basically
have consisted of integrating the desired lethality charac-
teristics into the heat penetration curves of cans. Pflug
gt gt. (1963); Thorne (1976); and Tung (1975), working
with can and pouches using different products, found
that center of retort pouches heat significantly faster

than cans of equal net weight and concluded that processing

11

time could be reduced from 30 to 44 percent. Similar com-
parison have been cited by Lampi (1977).

Heat penetration curves for cylindrical cans have
been calculated and plotted on semilog paper and single or
dual (broken) straight lines, can be obtained.

The time, in minutes, required for the straight
line portion of the heat penetration curves\to traverse
one log cycle is termed fh’ It represents the slope of
the curve. Another parameter used is j which is often
referred to the heating lag factor. It is a factor which,
when multiplied by the difference between retort tempera-
ture and initial food temperature, locates the intersection
of the extension of the straight line portion of the heat
penetration curve and a vertical line, representing the
beginning of process or zero time, Stumbo (1973).

Pflug gt gt. (1963) calculated the fh value for a
0.7Sinch slab and found that it agreed closely with experi-
mental heating data, indicating that a calculated fh can be
useful in checking on experimental data for pouches.

Pflug (1964) and Pflug and Borrero (1967) in their
extensive studies to compare various heating media (100%
steam, water and steam~air) for applicability to commercial
retort and to define specific heating characteristics of

each medium using f j, and F0 values as criteria for

h,

their comparison, concluded that usually f and j were

h
adequate parameters to describe heat transfer in the

12

laboratory. In their studies withacommercially sized re-
tort, however, although they found that the three heating
media gave predictable and reproducible results, they noted
that some of the semilogarithmic plots were straight lines,
some broke one time, and some actually were curves. This
was especially true of water cooks, which exhibited a slow
retort come-up time, with product temperature lagging 200
to 30°F behind water temperature. For those nonlinear
heating situations, they recommend evaluations and design
of thermal processes based on the general method.

Goldfarb (1970), Herndon (1971),in their studies,
showed that predicted sterilization values for inoculated
test packs would be closer to actual test values if the
pOpulation distribution of the slope indices from a sample
of heat-rise curves were used instead of the traditional
slowest or mean single value of slope index in the steri-
lization calculations. Herndon (1971), to confirm his
postulations, checked the accuracy of the proposed method
with 240 gm retort pouches of whole kernel corn (180 gm of
corn and 60 gram brine), first in a bath laboratory retort
and then in a production prototype continuous retort
(Hydrolok) using steam-air as the heat medium. The actual
spoilage results from the inoculated test pack series on
whole kernel corn closely correlated with the computer-
calculated, predicted values, based upon population dis-

tribution of the heat-rise curves.

13

Yamano gt gt. (1975), reporting on studies using
steam-air mixtures, found no appreciable differences among
the heating rate parameters (fh) for two types of film
(12-micron polyester/50 micron polyethylene and lZ—micron
polyester/9 micron foil/50 micron polyethylene) and among
processing temperatures from 105°C to 120°C.

As a whole, the reported experience indicated that
heat penetration into retort pouches could usually be des-
cribed by fh and j and standard mathematical formulas for
process calculations could be used. Frequent exceptions
can occur (when the heat penetration curve shows no straight
section) and then the general method becomes applicable.
All researchers cautioned, and in his review, Lampi (1977)
agrees, that calculations should be used only as means for
assisting process calculations studies and not as a sub-
stitute for full process determination. Lampi (1977)
pointed out that the significance of residual gas should
not be discounted. Such dual approaches to process deter-
mination have been documented by Yamano gt gt. (1969a)
with Chinese meat and soft spaghetti and for continuous
steam air retort (Hydrolok) by Goldfarb (1970). Lampi
(1977) cited investigations conducted by Schmidt and
Robertson with inoculated-pack and heat penetration studies
of vegetable products in which they concluded that: Heat
penetration studies are a valid means for establishing

processes, and that F0 values at cold point equal

14

to those of cans would be sufficient to assure commercial
sterility. Manson and Stumbo (1970) present a technique
that evaluates the effects of thermal processes of con-
duction heating foods in rectangular-shaped containers.
Bacterial lethality and the retention of any vulnerable
factor that is degraded logarithmically upon subjection
to moist heat may be determined by that digital computer

program.

Processing Effect gt Color

 

Tremendous effort has been aimed at stabilizing
the color of green vegetables during processing, and/or
storage. Heat processing, by far the most widely used
method of preservation, causes very undesirable degradation
of color in green vegetables. The major reason for this
change is due to the formation of pheophytins from chloro-
phylls. Because of the commercial importance of heat
processed green vegetables, stabilization of chlorophylls
to maintain the green color has been the object of con-
siderable research. The chlorophyll may be altered in many
ways. In food processing, the most common alteration is
the replacement of magnesium by hydrogen in the molecule
and the consequent formation of the dull, olive-brown
pheophytins. Chlorophyllides may be formed by the removal
of the phytol chain. These compounds are green in color
and have essentially the same spectral properties as chloro-

phylls, except they are water soluble, while the chlorOphylls

15

are ether soluble. If the magnesium in the Chlorophyllides
is replaced by hydrogen, the corresponding pheOphorbides
are formed which have the same color and spectral proper-
ties as the pheophytins, but are water soluble, Clydesdale
and Francis (1970a). MacKinney and Joslyn (1940, 1941),
Joslyn and MacKinney (1938), sought to define the kinetics
of the conversion of chlorophyll to pheophytin. They

used solution of chlorophyll in aqueous acetone containing
various amounts of acid. The reaction rate was determined
at several temperatures between 0 and 51°C. Their results
indicated that the reaction was first order with respect
to both acid and chlorophyll. A number of methods have
been deveIOped to inhibit the degradation of chlorophyll
in canned green vegetables during heat sterilization by
reducing the concentration of acid in the tissue. On

the basis of this principle several patents have been
issued. The "Blair Process" (Blair, 1940 and 1940a) is
probably the best, and the most successful alkalizing
agent found was MgCO3, Gupte (1964).

Still another approach, has been the use of High-
Temperature Short-Time (HTST) processing. This method
produces a very attractive product immediately after
processing but the beneficial effects are quickly lost
upon storage, Tom and Francis (1962), Luh gt gt. (1964).
Gupte and Francis (1964) combined HTST processing in con-

junction with MgCO3. This produced a very attractive

16

processed product, but the storage problem was not over-
come. Clydesdale and Francis (1970), Gold and Weckel

(1959) studying pigment degradation and methods of measure-
ment found that the percentage of pigment chlorophyll lost
correlates very well with certain color functions. The
correlation coefficient, r, between the value -(a/b), a
function of hue, and the percent of chlorophyll degradated
was very high. The correlation coefficient, r, was -0.992
and highly significant as reported by Clydesdale and Francis
(1970).

Clydesdale and Francis (1970) pointed out that
instrumental color functions can be used to trace the per-
centage loss of chlorophyll, the pigment responsible for
green color. Chemical pigment analyses are a great deal

more difficult and time consuming than color measurements.

Thiamine Stability During Processing

 

The effect of heat and certain environmental
factors on thiamine stability has been the subject of
several comprehensive investigations. A number of food
processing techniques have been involved which show promise
of improving the quality of heat-processed foods. Many of
these are based on achieving sterilization by high tempera—
ture-short time processing or by speeding up the rate of
heating (thereby reducing the time required for steriliza-
tion) by means of agitation of container during processing.

FEaster,Tbmpkins and Ives (1947) compared the influence of

17

high temperature-short time with conventional processing
methods on thiamine retention, in peas, carrots, green lima
beans, green bean purees and cream style yellow corn.

They showed that the high temperature-short time canning
procedures resulted in increased thiamine retention over
that obtained with conventional processing times and tem-
peratures. It was found that increasing the rate of heat
penetration through agitation of the cans during processing
reduced the amount of thiamine loss by one-third.

Feliciotti and Esselen (1957), studying the
thiamine degradation in buffered solutions and several
foods, concluded that the thermal destruction of thiamine
in foods may be dependent on the interrelationship of pH
and the relative proportions of the free and combined forms
of the vitamin. The data for carrots, green beans, peas
and pork showed similar thermal destruction characteristics
for thiamine.

Several methods of analysis have been used for
thiamine determination. Recently,because of the increased
need for routine analysis of vitamins in a variety of food
products,semi-automated and automated methods have been
developed.

Kirk (1974) described a semi-automated method for
thiamine determination in milk with application to other
selected foods.

Roy and Conetta (1976),using the Autoanalyser II

18

analytical system, showed that chemical reactions occur
in continuously flowing, air-segmented streams (air bubbles
segment the flowing streams to maintain sample integrity
and prevent cross mixing of sample) and every step of
analysis is automatic from aspiration of sample, to detec-
tion, and finally to readout on a recorder and/or digital
printer. Gregory and Kirk (1977) applied several methods
of sample preparation for fluorometric analysis in order
to maximize the removal of interfering compounds in green
bean samples. They concluded that preparation chromato-
graphy using Decalso ion-exchanger columns provided lower
apparent thiamine values than either direct analysis or
pre-extraction with isobutanol of the green bean sample
extract. They suggested that, for other foods, a prelimi-
nary testing of the effect of extract purification on each
product be assayed to ensure accurary of routine chemical
analysis.

Pelletier and Madere (1977) also described an auto-
mated method for thiamine and riboflavin determination
in various foods and they recommended the addition of
ethylene glycol monomethyl ether prior to separation of
insoluble matter from certain fruits to prevent loss of

those vitamins.

Texture Change During Processtgg
Texture of canned green beans can be affected by

handling, blanching and retorting operations. Van Buren

19

gt gt. (1960) concluded that the firmness of green beans
and their resistance to sloughing and splitting can be
greatly influenced by blanching treatment, by the time
lapse between blanching and filling, and by the presence
of calcium salts. Sistrunk and Cain (1960), working with
pole snap beans, concluded that blanching at temperatures
between 1700 and 180°F for times of 1.5 to 5 minutes were
the optimum for pole snap beans. Adjustments in the tem-
perature of blanch or the time of blanch within this tem-
perature range should be sufficient for controlling the
sloughing and softening in pole snap beans for canning.

The pectic substances and pectin methylesterase
(PME) are believed to be involved in changes occurring
during the blanching and retorting. According to Van Buren
gt gt. (1962) PME catalyses the removal of methyl groups
from pectic substances, allowing calcium ions to react
with exposed carboxyl groups. The insoluble calcium
pectinates and pectates remain in the intracellular regions
serving as cellular binding substances.

The specific effects of blanching and holding
treatments and calcium additives have been determined
largely on varieties of green beans common to the East and
West cost of the United States, Kaezmarzyk gt gt. (1963).

Resistance to shear of green beans has been
measured by the shear-press, Kramer gt gt. (1951); Sistrunk

gt gt. (1960) and Kaezmarzyk gt gt. (1963). A sample of

20

150 grams of drained canned green beans is carefully put
in a standard cell and the maximum-force for shearing the

beans is recorded for each sample, and converted to pounds

force per gram of product.

EXPERIMENTAL

Determination gt Process Conditions

 

 

Process conditions were established through heat
penetration studies of green beans in pouches. The use
of a "goop" (Silicone glue - GE2562-01DP) enabled a
hermetic seal to be obtained at the point of entry of the
thermocouple that was required to monitor product tempera—
ture at the center of the package. 24-gauge, copper-
constantan thermocouples covered with teflon were used in
these experiments. The 7 by 10 inch flexible packages
were constructed of 0.5 mil polyester (Mylar), 0.35 mil
Al foil, 3 mil C-79 polyolefin (Continental Can Company).

In order to avoid the break of vacuum in the pouch
and/or leak of brine or liquid from the product throughout
the thermocouple cover, about 1.5 inches of the thermo-
couple cover was removed at a point about 6 i 1 inches
from the thermocouple junction. The 6-inch length of lead
wire was passed through a hole made for the purpose in the
exact center of the pouch wall. Silicone glue was used to
attach the l l/2—inch, coverless section of thermocouple
wire to the outer wall of the pouch at the point of entry
of the wire into the pouch, at the same time sealing the
hole with the silicone.

The extra six inches of thermocouple wire was

21

22

needed so that placement of the thermocouple junction at

the center of a piece of green bean pod could be accomp-

lished outside of rather than within the pouch during the
pouch-filling operation.

The thermocouple junction was placed in the green
bean pod, and was then secured in the pouch's middle plane
by positioning it in the center of a helical steel spring
of a diameter of 0.75 in and 1.0 in, for pouches of 0.75
in and 1.0 in of thickness respectively, by 1.75 to 2 in
long.

The rigid steel coil maintained package thickness
of 0.75 and 1.0 in respectively and ensured that the thermo-
couple measured at point of greatest temperature lag.

Metal carriers of two different thicknesses were
employed in order to keep the maximum desired thickness
during the filling, sealing and retorting operation.

The pouches were filled with 8.0 and 9.5 ounces
of green beans for 0.75 and 1.0 inch of thickness respec-
tively. To some pouches a hot brine was added to the
pouches as the experimental design.

The pouches were sealed in a Multivac vacuum seal
machine. Type AGW Serial No. 969, with vacuum dial set
to 2, which gives a vacuum of 0.95 kg/cmz, and impulse
seal dial set to 8.

Thermal processing was performed in a 45 cm diameter

vertical retort using a water cook with superimposed air

23

pressure. The retort was equipped with a water ball over-
flow valve with a capacity of 50 gallons/minute.

Steam, water and superimposed air inlet located
at the bottom of the retort allowed a uniform distribution
of heating and cooling during the process.

An automatic temperature and pressure control
equipment from Foxboro was used in this experiment which
gives a temperature 2500 : 10F and pressure in the range
30 i 2 and 20 i 1 psig. Cooling water was 15°C.

Pouches were placed vertically in the open shelf
of a retort rack.

The retort rack was constructed of galvanized iron
measuring 23 cm high, 29 cm wide, and 29 cm long.
Between pouches a 0.25 inch apperture allowed a free cir-
culation of heating and cooling medium. A wire screen
was clipped over the pouch rack to prevent pouches flotation.

"Dummy" pouches were used in order to maintain a
full rack load at all times. Thermocouples were used to
monitor the retort temperature throughout each process
run. The thermocouples were connected to a Honeywell strip
chart recorder, Model No. K152X89-C-II-III-16.

Before the heat penetration study, the pouches
were placed in water bath at 122°F (50°C) for 10 minutes
in order to balance and equalize the initial temperature
for the heat penetration study.

Water in the retort was preheated to 122°C. The

24

loaded rack was immersed, the retort closed and brought
up to the process temperature of 250°F in 5.00 t .20
minutes. The initial product temperature was 122°F (50°C).
A superimposed air pressure of 20 or 30 psig was maintained
throughout the heating and cooling cycles, according to
each treatment.

Test pouches, according to treatments in Table 2
were processed at three different times in order to make
a regression analysis of the data and predict the time of
processing (period of time from steam on to steam off)
which gives a F0 _>_ 3.5 with 0.997(i : 38) of confidence
coefficient.

The process time for still cooked control cans
was 12 minutes in steam at 250°F with water cooling

system.

Green Beans

 

Green beans of unknown variety were bought at the
Farmer Market in Lansing during the period from October 20,
1977 to January 20, 1978.

The raw product was washed and snipped to remove
ends. It was then cut into pieces 1 inch long. The indi-
vidual portions for each container were weighed and blanched
separately by immersion in water at 180°F for 1.5 minutes,
then cooled by immersion in cold water and packed.

In the case of brine-containing samples (Table l)

the amount of brine was equal to 30% of the weight of the

25

Table 1. Treatment notation with the levels used for

each factor.

Treatment notationl Factor

T (inches) P (psig) B (ml)

A tl pl bl 0.75 20 -
B tl p1 b2 0.75 20 70
C tl p2 bl 0.75 30 -
D tl p2 b2 0.75 30 70
E t2 pl bl 1.0 20 -
F t2 p1 b2 1.0 20 80
G t2 p2 bl 1.0 30 -
H t2 p2 b2 1.0 30 80
I 303 x 406 can

1t1, t2 correspond to levels of thickness factor T.

pl, p2 correspond to levels of superimposed air pressure
factor P.

b b2 correspond to levels of brine addition factor B.

1!

26

Table 2. Time of processing1 in minutes for test pouches

and cans.

Trials

Treatments notation2 l 2 3
A tlplbl 8 10 12
B tlple 9 11 13
C tlpzb1 9 11 13
D tlpzb2 9 11 13
E tzplbl 10 12 14
F tzplbz 11 13 15
G t2p2bl '11 13 15
H tzpzb2 11 13 15
I 303 can 12 -- --

1 Time of processing corresponds to period of time from

steam on to steam off.
2

t1, t2 correspond to levels of thickness of factor T.
p1: p2 correspond to levels of superimposed air pressure
of factor P.

b1' b2 correspond to levels of brine addition of factor B.

27

beans.

Experimental Design

 

A 23 factorial completely randomized design was

used in this experiment, with three factors at two different
. levels, with ten repetitions for each treatment. Two
notations will be used to label the treatments. One, a
simple notation will be used for general comparison and a
second one which gives factor levels will be used for
specific comparison of main effect and interaction of
factors.
Factors:
Thickness of the pouch which corresponds to the
factor T with the following levels:
t + 0.75 inch
+ 1.00 inch
The superimposed air pressure in the retort which
corresponds to the factor P with the following levels:
pl + 20 psig (which corresponds to 4 psig above
the pressure corresponding to the
water temperature in the retort)
p2 + 30 psig (which corresponds to 14 psig above
the pressure corresponding to the
water temperature in the retort)
Brine addition which corresponds to the factor
B with the following levels:

bl + no brine addition

28

b2 + an amount of brine, which corresponds
to 30% of the weight of the beans was
added; 70 m1 of 2% brine was added to
treatments B and D and 80 ml of 2%

brine to treatments F and H.

Qualitnyvaluation Methods

 

After heat penetration studies batches of pouches
from each treatment were processed in order to evaluate
the pouch and 303 products. Those evaluations were con-
ducted within the three days after having been processed.

After being processed at 250°F container contents
were evaluated for thiamine, color, texture and moisture

by the following methods:

Thiamine

The procedure employed for thiamine determination
was the alkaline potassium ferricyanide oxidation of
thiamine to thiocrome which is measured fluorometrically,
using a Technicon AutoAnalyzer modules, Sampler II, 30
samples/hr; pump I, with modified plattered Technicon
manifold No. ll6-D207 (Kirk, 1974); and fluorometer with
360 nm primary filter, and 436 nm secondary filter, and

recorder.

Extraction Procedures

 

After draining the processed green beans, a sample

of about 14 grams weighed to the 4th decimal point,

29

(macerated) ground, was washed with 50 m1 of(L1N HCl
solution and transferred to a 250 ml flask, autoclaved
for 30 minutes at 15 psig, and cooled to room temperature.
After that the pH was adjusted to the range 4.0 to 4.5
with 2N Sodium Acetate, 4 ml 10% clarase solution added
and incubated at 45 to 50°C for 3.5 hours. After refrige-
rating the samples overnight at 5°C they were diluted to
100 ml withlLlN HCl solution. Using a Whatman #42 filter
paper sample was filtered and 15 ml of the filtrate was
mixed with 30 ml of isobutanol for pigment extraction
prior to assay.

The recovery standard was prepared by adding 20 pg
thiamine to duplicate samples of each treatment prior to

extraction.

Analytical System for Thiamine Determination

 

Initially, water was pumped through all tubes and
recorder base line adjusted to 5, using sample opening of
2 and reference opening of l. Reagent lines were placed
in appropriate reagents and system was purged 20-30 minutes
before establishing the baseline reading of 5 with reagents.
High standard is aspirated into the system and its maximum
response is adjusted to 95 using a full scale recorder
control. The sample test is then placed in water and the
system is allowed to return to baseline before standards
and samples are analyzed. The standard should be assayed

after 10 to 15 samples to monitor any possible recorder

30

drift. The blanks for each sample were determined by
analyzing samples and standard substituting 15% sodium

hydroxide for potassium ferricyanide oxidizer, Kirk (1974).

Texture
The texture of raw product and processed product
was determined by Kramer Shear press unit, using a stan-
dard cell.
The Kramer Shear press unit with a stroke length
of 8,84 cm, a time of 17 seconds and rate of speed of

0.52cm/second lb. ring set to 3000 pounds.

Procedure Used

 

Processed green beans. A drained sample of 150

 

grams was carefully placed in the cup of the standard cell,
The range sensitivity dial was set at 5 in order to get a
good reading in the middle third of the chart paper.

Raw product. A 50 gram sample of raw product was

 

placed in the cup of standard cell, the range sensitivity
dial set to 50 in order to get a good reading in the middle
third of the chart paper.

Calculations. The texture was expressed in pounds

 

force/gram of product. By using the peak-height, the
texture of the sample was calculated by the following
formula:

pgak height x range sensitivity

100 ’ 100
weight Ef’sampIe (g)

 

1b ring x

 

lb force/gram =

 

31

where: 1b ring = 3000 lb
peak height = read on the chart paper
range sensitivity = 5 for processed green beans
50 raw product
weight of sample = 150 grams for processed green
beans

50 grams for raw product

c.2152.

The color of the raw and processed green beans was
determined by Hunter Lab Model 25 Color and Color Difference
Meter using a Hunter Lab Standard Cell No. D33G-2060.

The samples of 50 grams of beans were put in a
glass dish and two readings were taken from each sample.

The first one with the dish in normal position, and the
second one with the dish rotated 900 on the horizontal

plane.

Moisture Determination

 

Samples of the product were cut in 0.5 cm portions
of about ten grams and dried at 100 : 2°C for five hours,
and were then reweighed for moisture calculation. This
indirect method of moisture determination was checked at
different times for constant weight. The chosen five-
hour drying time was considered to be adequate since addi-
tional drying time (with just a few samples) was found to

give no further weight loss.

32

RESULTS AND DISCUSSION

Heat Penetration Study

 

For each test pouch, process lethality (F0) was
calculated by the Improved General Method (Schlutz and
Olson, 1940) based on measurements of center temperature
at 0.5-minute intervals and a computer program to calcu-
late the lethal rates (assuming Z value of 18°F) and total
the interval lethalities. In the heat penetration study,
24 process lethalities were obtained, which constituted
eight replications of each of the three process times
that were used for a given treatment. A regression analy-
sis procedure was applied to determine the process time
which would assure an individual F0 3 3.5 minutes with a
0.997 (E 1 38) confidence coefficient. With an initial
product temperature of 122°F (50°C), the time of processing
for all treatments considered in this experiment are given
in Table 2.

The calculation of lethality was done using the

method recommended by Schultz and Olson, 1940, where

-1 T - 250
Z

Fahrenheit and Z is the slope of thermal death time curve.

L = log , T is the temperature in degrees

Data, average and standard deviations are presented in
Table 3.

After F0 values were calculated at three different

33

times with eight replications for each treatment (Table 3),
a regression analysis was applied to the data, regression
equations were determined and a predicted processing time
to give a F0 3 3.5 minutes with 0.997 (E i 38) confidence

coefficient was determined (Table 4).

Effect gt Superimposed Air Pressure gt

 

Time of Processing

 

Pouches, unlike cans, do not have a rigid shape.
During heat processing, some expansion of gases and of
product occurs. As a consequence, pressure inside the
pouch is increased. The use of superimposed air pressure
counterbalances this effect and brings the pouch wall in
closer contact with the food, thus improving the heat
transfer rate, and consequently achieving a reduction in
time of processing required for sterilization. On the
other hand, the use of higher pressure results in more
air entering the retort and this results in more air
bubbles coming into contact with the pouch walls, thus
reducing the heat transfer rate from the medium to the
product because air has a smaller heat transfer coefficient

than does water.

Effect 9f Pouch Thickness n Time f Processing

 

Quality retention during thermal processing becomes
increasingly difficult with increases in container size.

Shape of the container can also be important with respect

34

 

 

Table 3. Process lethality (F0) calculated by Improved
General Method, using center temperature at
0.5 minute intervals, and assuming Z = 18°F
Treat- Time Replications —l q
ment min 2 3 4 5 6 7 8 X I
A 8 1.69 1.79 1.26 1.35 1.55 1.83 1.35 1.36 1.55 0.21
tlplbl 10 3.94 3.69 4.41 3.75 3.70 4.00 3.59 3.47 3.82 0.29
12 5.85 5.57 5.19 5.50 5.81 6.36 5.91 5.83 5.75 0.34
B 9 3.07 3.45 2.65 3.38 2.93 3.12 3.07 3.06 3.09 0.25
tlplb2 11 4.91 5.15 5.10 4.99 4.68 5.39 5.71 5.22 5.14 0.31
13 6.86 - 6.58 6.63 7.17 6.61 7.02 7.17 6.86 0.26
C 9 2.55 2.73 2.29 2.23 1.97 1.66 2.47 1.88 2.22 0.36
tlpzb1 11 4.74 4.46 4.48 3.39 4.62 4.13 4.40 4.80 4.38 0.45
13 6.21 6.35 6.62 6.21 7.53 7.27 7.41 7.31 6.86 0.57
D 9 2.71 2.56 2.83 2.12 2.15 1.71 3.00 2.59 2.46 0.43
tlpzb2 11 4.41 4.11 4.32 4.41 4.65 3.38 4.36 3.60 4.19 0.45
13 7.46 6.97 6.52 5.91 6.64 6.12 5.69 6.65 6.62 0.54
E 10 2.07 2.07 3.10 1.93 1.46 1.94 2.10 1.79 2.06 0.47
t2p1b1 12 4.56 4.45 3.41 3.79 4.14 4.14 3.37 3.71 3.95 0.45
14 6.32 6.33 6.14 6.04 7.04 6.71 6.31 — 6.41 0.35
F 11 3.52 3.70 4.19 5.51 3.75 5.43 3.52 4.73 4.29 0.83
t2p1b2 13 7.01 7.43 5.65 6.69 5.98 5.85 5.25 6.92 6.35 0.77
15 7.59 8.43 6.93 8.00 8.02 8.49 7.76 7.18 7.80 0.55
G 11 2.51 3.70 2.96 3.62 3.20 2.12 2.80 2.46 2.93 0.56
t2p2b1 13 4.55 4.54 5.58 5.34 4.46 4.44 - - 4.82 0.51
15 7.55 7.44 7.10 7.21 9.25 7.94 7.93 7.18 7.58 0.42
H 11 2.99 2.36 3.29 2.79 3.65 4.14 3.66 3.06 3.21 0.54
t2p2b2 13 5.41 4.70 4.89 4.47 5.05 5.64 5.35 5.62 5.14 0.43
15 7.10 6.97 7.90 7.56 7.63 6.16 7.35 8.00 7.46 0.38
1i - mean

(1
0-

standard deviation

Table 4.

Treatment

A

B
C
D

'11

Time of processing1

at 250°F to give a F

notation
tlplbl
tlple
tlpzbl
tlpzbz
tzplbl
t291b2
tzpzbl
tzpzbl
303 can

35

required for pouches in water

.2 3.5 minutes.

0
Superimposed
Thickness air pressure Brine

(in) (psig) (9)
0.75 20 -
0.75 20 70
0.75 30 -
0.75 30 70
1.00 20 -
1.00 20 80
1.00 30 -
1.00 30 80

lCorrespond to period of time

Processed in steam at 2500F

from steam on

to steam off

Time
(min)

10.57
10.24
11.25
11.52
12.52
12.56
12.91
12.65
12.00

36

to quality loss during the sterilization process. The
flat shape of the pouches and the fact that any desired
thickness of product may be packed in them for thermal
processing, gives them a distinct advantage over cans and
jars. As the distance from the coldest point to the
surface of the container is reduced, the time of processing
can be reduced, consequently increasing nutrient retention
and minimizing other undesirable changes in food charac-
teristics like color, flavor, texture, etc.

The reduction in pouch thickness from 1.0 inch to
0.75 inches allowed the following reductions in time of
processing, as shown in Table 4: At 20 and 30 psig of
superimposed air pressure, the reductions in time of pro-
cessing for pouches without brine were 1.95 and 1.66
minutes,respectively. The reductions in time of processing
for the pouches with brine addition were 2.32 and 1.13

minutes at 20 and 30 psig,respective1y.

Effect gt Brine gt the Processing Time

 

 

Heat penetration is faster by convection than by
conduction. When a liquid is mixed with pieces of solid
material, both convection and conduction will occur. The
relative ratio of convection to conduction depends on
several variables (density of liquid and of solid product,
shape and size of pieces, shape of the container, ratio
of liquid to solids pieces).

The brine addition did not greatly alter the time

37

of processing. This small effect can be explained by the
small amount of brine added to each pouch in proportion
to the amount of beans and secondly, by the fact that the
temperature of the thermocouple junction inserted was
measured within the green bean pod, where conduction
heating would be the major controlling factor.

The water cook with superimposed air pressure of
20 and 30 psig maintained throughout the heating and cooling
cycles of the retort pouch processes was found effective
in preventing vibration during the initial period of
heating. The high external pressure also ensured that the
pouch maintained contact with the product. Also, the
high vacuum used before sealing the pouch made it very
tight to the product, maximizing the heat transfer from

medium to the product.

Quality Evaluation

 

Thiamine Retention

 

Table 5 shows the thiamine concentration in ug/gram
of product (dry weight basis) for each treatment. Mean
values followed by the same letter did not differ statisti-
cally at the 1% level, as calculated by Duncan's Multiple
Range Test.

Table 5 results show all packed green beans, with
exception of treatments D and H, to be significantly
higher (at 1% level) in thiamine content than conventionally

canned beans (Treatment I).

Effect gt pouch thickness gt thiamine retention

 

 

Analysis of variance (Table 6). indicated that
thiamine retention was unaffected by any increase in pouch
thickness from 0.75 to 1.0 inches. The addition of brine,
on the other hand, did result in a statistically signifi-
cant difference in thiamine retention. Variation in super-
imposed air pressure also produced a statistically signifi-

cant change in thiamine retention.

Effect gt brine addition gg thiamine retention

 

 

The addition of brine to the pouch decreased the
thiamine retention in the product as shown in Table 7.

Some thiamine leaked from the product into the fluid,

38

39

.umoa
mmcmm maafiuasz m.:moc:0 ofiu 50 Hw>mH Naum 5HHmuwumfiumum umwmeu you 00 umuumH 08mm m:u :uH3.mamwz .ammz I x

03.0
53.0
03.0
03.0
05.0
00.0
03.0
5H.H

03.0

8000

H

00.00
00.33
05.03
00.00
53.30

00.03

H3.0

00.3

05.0

00.3

H0.0

00.0

33.3

50.3

03.0

0H

03.0
00.3
50.0
00.3
30.0
00.3
00.3
00.3

H0.3

mammn c0000 00mmmoonm ca

03.0 H0.3
00.3 00.0
H0.3 00.3
50.3 00.3
05.0 50.0
00.3 00.3
50.3 05.3
00.0 03.3
H3.0 0H.0
0 5
mmumofiaawm

coaumw>mv wumvawum I m

0

l.N

ucmEmHSmmmE mcfiemwsu How 0cHHQEmm mnemwn «mo wmcfimuv vfisvaq

03.3
H0.3
03.3
00.3
00.0
00.0
00.0
50.0
03.0

0

50.3

00.3

00.3

05.0

H0.0

00.0

00.3

00.0

00.0

0

00.3

00.3

05.3

03.0

H5.0

H0.3

00.3

00.0

00.0

3

mcmu 000 0cm monosom
Amflmmn #:0003 hupv Emu0\0: ca gamucoo wcwEmwne

00.3

30.3

00.3

H0.0

05.0

00.3

00.0

HH.0

00.0

0

00.0
30.0
05.3
H0.0
H5.3
03.3
H3.0
00.0
00.0

N

00.0
00.0
00.0
00.3
00.0
00.0
55.0
HH.5
30.0

H

H

amo 000 H

Newame

Hamamu

NeHcNu

Heeamu

Nemaeu

Homaeu

NEHaHu

fleece“

ucmEumwuH

.0 manna

Table 6. Analysis of variance of data in Table 17

 

dfl

Thickness (T) 1
Brine (B) 1
Superimposed—
air pressure (P) 1
Interactions

TB 1

TP 1

BP 1

TBP 1
Error 72
Total 79

*

1df - degrees of freedom

2SS - square sum

3MS - mean square

4 F MS
MS (error)

 

40

582

0.005
5.266

7.248

0.331
0.008
0.007

0.006

28.602
41.447

M83

0.005
5.266

7.248

0.331
0.008
0.007
0.006
0.397

F4

0.015

**
13.256

**
18.245

0.833
0.020
0.018

0.015

*
Indicates significance at 0.01 level of probability

41

Table 7. Effect gf brine addition on thiamine retention
ug/gram in green beans processed in pouches in
water at 250 F

Treatment Superimposed-air pressure Difference Reduction

 

T P b1 b2 ug/g %

tl pl 5.59 4.96 0.63 11.27
tl p2 4.96 4.32 0.64 12.90
t2 pl 5.45 5.07 0.38 6.97
t2 p2 4.88 4.50 0.38 7.79

1Average of 10 analyses, ug/g (dry weight basis)

42

reducing the thiamine content in the product. Values for
the thiamine content of the raw, processed beans and
liquid from each treatment are given in Table 8.
Considering the fact that packing liquid for
vegetables is not normally consumed, reduction in the
packing liquid added to the product will reduce the
thiamine loss, and the loss of other soluble substances
from the product. This loss can be avoided if the liquid

is consumed or used to make up other types of foods.

Effect gt superimposed air pressure gg thiamine retention

 

 

Although superimposed air pressure made no
difference in the processing time, it had an undesirable
effect on thiamine retention. Table 9 shows the effect
of superimposed air pressure on thiamine retention in
processed green beans in pouches in water at 250°F. It
seems that thiamine degradation is accelerated by higher
pressure during the processing. The data show that a
change in superimposed air pressure from 20 to 30 psig
gave a reduction of about 11% in thiamine.

No explanation was found in the literature rela-
tive to the effect of superimposed air pressure during

processing on thiamine retention.

43

Table 8.. Change in thiamine content in green beans dur-
ing processing

Drained liquid Total

 

 

 

Thiamine Content1 from containers Thiamine
Treatment Raw Processed Thiamine Thiamine Total loss in
notation Product Beans2 Retention3 content4 liquid liquid

Z ug/g s we

tlplb1 5.93 5.59 94.27 0.32 21.0 6.72
tlplb2 5.93 4.96 83.64, 0.24 64.5 15.48
tlplbz 5.93 4.96 83.64 0.33 19.0 6.27
tlpzb2 5.93 4.32 72.85 0.23 72.0 16.65
tzplb1 5.93 5.45 91.91 0.32 23.5 7.50
tzplb2 5.93 5.07 85.50 0.32 77.3 24.73
tlpzbl 5.93 4.88 82.29 0.31 22.5 6.98
t2p2b1 5.93 4.50 75.89 0.24 84.2 20.25
303 can 5.93 3.98 67.00 0.14 215.8 30.21

 

ug/gram of product (dry weight basis) — average of ten determinations
Liquid drained off before measurement
It doesn't include thiamine content of drained liquid

Average of five determinations

44

 

Table 9. Effect of superimposed—air pressure (P) on

thiamine retention ug/gram in green beans

processed in pouches in water at 250°F
Treatment Superimposed-air pressure Difference Reduction
T B P1 P2 09/9 %
tl bl 5.59 4.96 0.63 11.27
tl b2 4.96 4.32 0.64 12.90
t2 bl 5.45 4.88 0.57 10.46
t2 b2 5.07 4.50 0.57 11.24

1Average of ten analyses, ug/g (dry weight basis)

45

Texture

The data, means and standard deviations of ten
determinations of texture from each treatment of pouches
and 303 cans are shown in Table 10.

The significance of differences among the means
given in TableIUJwas determined by the analysis of variance
using Duncan's Multiple Range Test. Means followed by the
same letter do not differ statistically from one another
at the 1% level.

TableIUJresults show all packed green beans, with
the exception of treatments D and G, to have a signifi-
cantly firmer texture (at the 1% level) than conventionally-
canned beans (Treatment I), averaging 0.65 pounds of force/
gram as compared with 0.39 pounds of force/gram for the
can-packed product. Treatments D and F also exhibited
higher shear force values than I (0.51 and 0.47, respec-
tively) but the differences (from 0.39) were insufficient
for 1% statistical significance.

In order to evaluate the main effect of each
factor in this factorial experimental design as well as
first and second order interactions, analysis of variance
was done and the results are given in Table 11.

Only pouch thickness main effect had a statis-

tically significant effect on green bean texture. The

cOHumH>mw vumvcmum t m .
.ume 0

mmnmm mHaHuHsz m.amu::0 0:0 00 Hw>mH NH um hHHmUHumHumum umMMHv 00: 00 HmuuwH 05mm 050 nuHB mammz .cme I x

46

..N
ucmampammme 000000 000 wmanuv 0H=0HAH

00.0 000.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 000 000 0
00.0 00000.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 N00000 0
00.0 0000.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 000000 0
00.0 00000.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 N00000 0
00.0 00000.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 000000 0
00.0 00000.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 N00000 0
00.0 000.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 000000 0
00.0 00000.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 N00000 0
00.0 0000.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.0 000000 0
mm. NM. 00000 00 0 0 0 0 0 0 m N 0 000000000

kuoe mmumoHHamm acmaummuH

.Emum\00000 mccsom ou

vmuum>coo 0cm mmmum ummnm may 03 wwusmmmfi mammn c0000 00mmmooum mo musuxme .0H mHnma

H

Table 11. Analysis of variance of data in Table 18

 

df1
Thickness (T) 1
Brine (B) l
Superimposed-
air pressure (P) 1
Interactions
TB 1
TP 1
BP 1
TBP 1
Error 72
Total 79

47

882

0.135
0.019

0.004

0.464

0.003

0.0002

0.245
0.806
1.545

M83

0.135

0.019

0.004

0.464
0.003
0.000
0.245

0.011

**
12.27

1.73

0.36

**

42.18
0.270
0.017

**
22.27

**
Indicates significance at 0.01 level of probability

ldf - degrees of freedom
2SS - Square Sum

3M8 - Mean Square

4 MS

F:

 

MS (error)

48

green beans processed in pouches with 0.75 inches of
thickness proved firmer than beans in 1.0-inch thick
pouches. This result can be explained by the difference
in time of processing, the thinner pouches having been
processed for a shorter time than the thicker ones.

A first order interaction between thickness and
brine addition was observed (Table 11 and Table 12),
indicating that the significant reduction in shear-force
value attributable to package thickness occured only in
those samples to which brine had not been added. The
statistically significant TB interaction also revealed
the existence of a softening effect of brine addition,
but only among 0.75-inch-thick samples.

A statistically-significant seCond order inter-
action (TBP) was also noted (Table 11), indicating that
pressure variation influenced the effect of package
thickness and brine addition on beans texture. The
significantly firmer texture of treatment H beans, as
compared with treatment D (TableJJD, was opposite in
direction to the textural change noted in the experiment
as a whole (with respect to package thickness effect) and
is indicative of the TBP interaction just referred to.

Within the range over which tested variables were
studied, the use of a 0.75-inch-thick package to which no
brine is added plus a superimposed air pressure during
processing of 30 psig appears to provide Optimum conditions

for maximum firmness of final product.

Table 12. Examination of the TB

49

data from Table 18.~

 

 

Thickness = T
Brine = B t1 t2 t1 - t2
bl 14.89 11.22 + 3.67
b2 10.20 10.62 - 0.42
bl - b2 4.69 0.60
(b t - b t )2
l 1 2 l

B within t SS =

B within t SS =

T within b 88 =

T within b SS =

*

 

2 x r x p

(14.89 - 10.20)2

 

2'x 107x 2

2
(blt2 - bltzl

 

2 x r x p

(11.22 - 10.62)2

 

2 x 10 x 2

2
(blt1 - blt2)

 

’2Ix r x p

(14.89 - 11.22)2

 

*2’x I0 x‘2_
2

(bzt1 - b2t2)
z—x r x p—_

(10.20 - 10.60)2

 

2 x 10 x 2

interaction for the

**
= 0.550

= 0.009

**
= 0.337

0.004

*
Indicates significance at 0.01 level of probability

221.0

The results of color measurements were analyzed
and will be discussed in terms of the -a/b ratio since the
conversion of chlorophyll to pheophytin is very well
correlated with the ratio -a/b (Clydesdale, 1970a). Also,
Hunter values L, a, b obtained were found to be very well
correlated with visual rankings.

A plus value for g indicates redness; a minus
value, greenness. A plus value for g indicates yellowness;
a minus value, blueness. L measures lightness. The ratio
-a/b, average and standard deviation values are in Table 13.

The significance of differences among the means
given in Table 13 was determined by the analysis of vari-
ance using the Duncan's Multiple Range Test. Means with
the same letter do not differ statistically at the 1% level.

The overall effect of treatments on color was
small. There was little difference between green beans
processed in pouches and 303 cans. The chlorophyll pig-
ment is very sensitive to heat treatment and the reduction
in time of processing for pouches did not make a big
difference in the color of the processed green beans.

Data in Table 13 show a big variation among
replicate readings, which gave large standard deviations.
This variability may have resulted from variations of

color in the part of the pod cut, and some differences

50

51

.GOHumH>ww unavamum0
. .ummH 00:00
mHnaanz cmuasa 0:0 00 Hm>mH NH 000 um AHHmoHumHumum umMMHv uoc 00 umuumH wEMm wnu :uHs mcmwx
.ucmamunmmua whomwn «mo uoawmuv stvHHM

3~o.0 mmo.o 000.0 050.0 530.0 330.0 500.0 000.0 000.0 mm
0300.0: 0500.0: n_5000: 0000.0: 0000.0: 0000.0: 0~00.0: 000~.0: 0050.0: «x
0500.: 000.0: 000.0: 5N0.o: 050.0: 000.0: 000.0: 0N3.~: ~05.0: 00000
300.0: 500.0: 000.0: 000.0: 000.0+ 000.0: 0Ho.0: 050.0: 000.0: 00
000.0: 0N0.o: 050.0: 000.0: 000.0: No:.0: 000.0: 500.0: 000.0: 0
000.0: «No.0: 300.0: 000.0: 000.0: 000.0: 000.0: 050.0: 000.0: 0
500.0: 000.0: 500.0: 0~N.o: 000.0: 000.0: 300.0: 500.0: 050.0: 5
«00.0: 000.0: 350.0: 500.0: 0N3.o: ~0o.o: 000.0: 00~.0: 000.0: 0
500.0: 030.0: 000.0: 000.0: 500.0: «00.0: 050.0: 500.0: 000.0: 0
000.0: ~00.o: 0N0.0: 000.0: 000.0: 000.0: 00H.o: 00N.o: 500.0: 3
~00.0: H~0.0: 300.0: 000.0: 000.0: 550.0: 000.0: 050.0: 050.0: 0
500.0: 000.0: 050.0: «00.: 000.0: 350.0: 000.0: 03~.0: 500.0: N
000.0: 000.0: 300.0: ~o~.o: 050.0: mo~.o: 000.0: 03~.o: 000.0: H
000 000 ~0N0~0 00~0~0 00H0~0 000000 000000 HaweHu 000000 000000 000000
0 0 0 0 m 0 0 0 3 :00000
aOHuwuoz unmaummue
.n\m: cu cmEuowmcmuu mama
.00:0~O 000 .oz HHmo wumocmum QMHumucsm .kumz moqmumMMHo HoHou 0cm
uoHou 000 anuwucsm nuH3 mcmon c0000 00mmwooum mo ucmsmnsmmma uoHoo .0H mHnma

H

52

in thickness of the layer of green beans on the plate
during the readings. The two readings for each sample
cut down somewhat on this variability, but in order to
reduce it further, it might be a better idea to grind
the samples before the reading.

All the factors considered in this experiment
T, P and B, had statistically significant effects upon
the change in color of the green beans during heat

processing in pouches.

Effect of Pouch Thickness on Color

 

Higher average values of the ratio -a/b, which
means greener color of beans, were obtained with thinner
pouches, with brine addition, processed at lower super-
imposed air pressure (tlplb2)° Lower average values of
-a/b were obtained with higher levels of all factors con-
sidered (tzpzbz) (Table 13%

The two first order interactions were determined
between thickness x brine (TB) and brine x superimposed
air pressure (BP) (Table 14).

These interactions were studied and Tables 15
and 16 summarize data for interactions TB and BP respec-

tively.

Interaction of Thickness Versus Brine Addition (TB).

 

Reduction in pouch thickness had a significantly

beneficial effect (i.e., higher -a/b value) on color

Table 14.

Thickness (T)
Brine (B)

Superimposed-
, air pressure (P)

Interactions

 

TB

TP

BP

TBP

Error

Total

df

1

72

79

53

SS2

0.067
0.070

0.035

0.0818
0.0134
0.0355
0.0005
0.1690
0.4720

MS3

0.0670
0.0700

0.0350

0.0818
0.0134
0.0355
0.0005
0.0024

Analysis of variance of data in Table 19.

F4

2 **
28.54

**
29.82

**
14.91

**
34.85
*
5.71
**
15.12
0.21

*
Indicates significance at 0.05 level of probability

ti:
Indicates significance at 0.01 level of probability

df -

NH

SS -

3M8 -

4 F =

square sum

mean square

 

MS (error)

degrees of freedom

54

 

 

Table 15. Examination of the TB interaction for data from
Table 19.
Brine = B Thickness = T
tl . . t2 t1 - t2
b1 -l.270 -1.393 +0.123
b2 -3.734 -l.296 -2.348
b1 - b2 +2.464 -0.097
2
(bltl - bltz)

T within blSS =

 

2 x r x p

[-1.270 - (-1.393)]2
2’x410 x 2

 

= 0.0004

2
(b t - b t )
. . 2 1 2 2
T Within bZSS 2—k r x p

 

[-3.734 - (1.296)]2

**
2 x 10 x 2* ‘ 0°149

 

2

r x p

B within t SS

 

2
[-1.270 - (-3.734)] _ **
*2 x 10 x 2 ‘ 0'152

2
(t b - t2b2)

B within t SS =
r x p
[-1.395 - (-l.296)]2=

02 x 10 x 2 0.0002

 

**
Indicates significance at 0.01 level of probability

55

 

 

Table 16. Examination of BP interaction for data from
Table 19.
Brine = B Superimposed-air pressure P
pl ' 92‘ pl 7 Pz
b1 -l.327 -l.336 +0.009
b2 -3.356 -l.674 -l.686
bl - b2 +2.029 +0.338

2
(b p - b p )

P within b SS 1 1 l 2
l 2 x r x t

 

[-l.327 - (1.336)]2
27x 10 x 2

2

 

= 0.0002

(b p - b p )
. . 2 l 2 2
P Wlthln bZSS 2 x r x t

 

[-3.356 O (-l.674)]2
2 x 10 x 2

 

**
0.071

2
(plbl ' plbz)
2'x t

 

B within plSS

[-l.327 - (-3.356)]2
2 x 10 x 2

)2

**
= 0.103

 

(szl ’ pzbz
p2SS - 2 x r x t

 

B within

[-l.336 - (1.674)]2

2 x 10 x“: 0'0003

 

**
Indicates significance at 0.01 level of probability

56

retention for treatments processed with added brine.
Brine addition produced statistically significant color
benefits only in the treatments involving the thinner
pouches (t1).

Differences in the amount of air trapped within
the package may provide the explanation for the measured
differences in color degradation of the various treatments.
It was observed during these tests that beans could not
be packed as tightly in the thicker pouches as in the
thinner. This, apparently, resulted from the tendency of
a greater proportion of beans in the thicker pouch to
position themselves horizontally rather than vertically,
thereby creating more air space (voids) in the package.
This observation has no supporting data but, if correct,
would mean that the thicker pouches contained a higher
proportion of air space to liquid, thereby making them
correspond more closely in this respect to the treatments
without brine than did the thinner, brine-packed samples.
Thus, if brine addition "protected" the color, the effect
would theoretically be greatest in samples having the
highest prOportion of brine. The actual results reported
in Tables 15 and 19 are in keeping with this line of
reasoning, since beans in thinner pouches with brine added
(t1b2) had better color than thicker pouches with brine
added (tzbz) and were also better in color than either

thickness of pouch without brine added (tlb and tzb

1 1"

57

Interaction 9£_Superimposed Air Pressure

 

Versus Brine Addition (PB)

Overall, results of these tests suggest that low
superimposed air pressure, some brine addition, and a
thin-profile pouch will reduce the color change in pouch-
packed green beans that are processed under conditions

comparable to those used in these experiments.

SUMMARY AND CONCLUS IONS

The effects of pouch thickness, brine addition
and superimposed air pressure on processing time require-
ments and product quality have been evaluated. Thiamine
content, color, and texture were the parameters used to
evaluate quality of product.

Process lethality (F0) was calculated by the
Improved General Method (Schultz and Olson, 1940) at three
different processing times. Regression analysis procedures
were applied to predict the time of processing needed to
assure an F0 3 3.5 minutes with 0.997 confidence coefficient.

The processing time varied from 10.24 to 12.91
minutes with a come-up time to process temperature of
250°F of 5.0 i 0.2 minutes.

The process time for 303 x 406 cans in steam at
2500F in order to give the same Fo value was 12 minutes,
with a come-up time of 1.0 minute.

Overall comparison of beans processed in pouches
with those processed in cans showed statistically signi-
ficant superiority for the former with respect to thiamine
retention and texture, but no statistically significant
difference in color.

Studying the effect of each factor (thickness,

brine addition, and superimposed air pressure) on pouch

58

59

product quality, the following general conclusions can be
drawn:

Texture - within the range over which tested
variables were studied, the use of a 0.75-inch thick
package to which no brine is added and a superimposed
air pressure during processing of 30 psig appears to
provide optimum conditions for maximum textural firmness.

Thiamine - within the range over which tested
variables were studied, package thickness did not affect
thiamine retention. However, thiamine retention was
significantly improved by use of 20 psig superimposed air
pressure during processing and by packing the product
without brine.

gglgr - overall, results of these tests suggest
that lower superimposed air pressure, the addition of
some brine, and the use of thinner-profile pouches all
reduce color change in pouch-packed green beans. Varia-
bility of replicate color readings might have been reduced
by grinding the product before measurement.

The long come-up time during processing was
unavoidable because of process equipment limitations.
Reduced come-up time can be achieved by using a retort
with high-capacity heating and circulation of hot water.
Overall process time could thereby also be substantially
reduced.

For overall product quality, the best combination

60

of pouch thickness, brine addition and superimposed air
pressure may depend on the relative importance:of each
quality factor (color, texture, and thiamine retention)
since optimum processing conditions are not identical for

each quality factor.

LIST OF REFERENCES

REFERENCES

A.O.A.C. 1975. "Official Methods of Analysis", 12th ed.
pp. 823-824. Assn. Offic. Anal. Chem. Washington,
D.C.

Blair, J.S., 1940. Color stabilization of green vege-
tables. U.S. Patent No. 2,186,003.

Blair, J.S., 1940a. Color stabilization of green vege-
tables. U.S. Patent No. 2,189,774.

Brown, F. and Keegan, P.B., 1973. Heat sealing methods.
Modern Packaging 46(12), 314, 317, and 321—327.

Burke, P. T. and Schultz, G. L., 1972. "The comparative
performance of flexible packages and metal cans".
Tech. Report 73-62 U.P. U.S. Army Natick Labora-
tories, Natick, Massachusetts.

Clydesdale, F.M., Fleischman, D.L. and Francis, F.J.,
1970. Maintenance of color in processed green
vegetables. Food Product Development 4(5), 127-138.

Clydesdale, F.M., and Francis, F.J., 1970a. Color measure-
ment of foods: Green vegetables. Food Product
Development 4(12), 51-54, 56 and 96.

Davis, E.G., Karel, M. and Proctor, B.E., 1960. The
pressure volume relation in film packages during
heat-processing. Food Technol. 14(3), 165-169.

Dewey Redesign, 1975. "Retort pouch T/C". Dwg No.-186
Dewey Redesign, Eagle River, Wisconsin.

Duxbury, D.D., Sams, P.F., Howeler, W.F., Gee, J.A. and
Miller, W.N., 1970. "Reliability of flexible
packaging for thermoprocessed foods under produc—
tion conditions, Phase I Feasibility". Tech Rep.
72—77 G.P. U.S. Army Natick Laboratories, Natick,
Massachusetts.

Feaster, J.F., Tompkins, M.D. and Ives, M., 1947. Influence

of processing technique on vitamins in canned corn.
Canner, 104(11), 16-18.

61

62

Feliciotti, E. and Esselen, W.B., 1957. Thermal destruc-

tion rates of thiamine in pureed meats and vege-
tables. Food Technol. 11(2), 77-84.

Fold, J.H. and Weckel, K.G., 1959. Degradation of chloro-
phyll to pheophytin during sterilization of canned
green peas by heat. Food Technol. 13(5), 281-286.

Goldfarb, P.L., 1970. Pouch for low-acid foods I. Modern
Packaging 43(12), 70-76.

Goldfarb, P.L., 1971. Pouch for low-acid foods. II. Modern
Packaging 44(1), 70-76.

Goldfarb, P.L., 1973. Flexible packaging for heat processed
foods. Packaging and product evaluation. Packaging
Inst., St. Louis, Missouri.

Gregory, J.F. and Kirk, J.R., 1977. Comparison of chemi-
cal and biological methods for determination of
thiamine in foods. Michigan Agricultural Experiment
Station Journal Article No. 8160, May 1977.

Gupte, S.M. and Francis, F.J., 1964. Effect of pH adjust-
ment and high temperature short-time processing on
color and pigment retention in spinach puree. Food
Technol. 18, 1645-1658.

Herndon, D.H., 1974. POpulation distribution of heat raise
curves as a significant variable in heat steriliza-
tion process calculations. J. Food Sci. 36, 299-305.

Joslyn, M.A. and MacKinney, G., 1938. The rate of conver-
sion of chlorophyll to pheophytin. J. Am. Chem. Soc.
60, 1132-1136.

Kaezmarzyk, L.M., Fennoma, O. and Powrie, D.W., 1963.
Changes produced in Wisconsin green snap beans by
blanching. Food Technol. 17(7), 123-126.

Keller, R.G., 1959. "Flexible packages for processed foods".
QMFCIAF Rep. 31-59. Quarter Master Food and Con-
tainer. Institute for the Army Forces, Chicago, IL.

Kirk, J.R., 1974. Automated method for the analysis of
thiamine in milk, with application to other selected
foods. Journal of AOAC 57(5), 1081-1084.

Kramer, A., Burkhardt, G.T. and Rogers, H.P.J., 1951. The
shear-press an instrument for measuring quality of
foods. Canner, 112(3), 34-36.

 

63

Lampi, R.A., 1977. Flexible packaging for thermoprocessed
foods. Advances in Food Research 22, 304-377.

Lampi, R.A., Schultz, G.L., Ciavarini, T. and Burke, P.T.,
1976. Performance and integrity of retortable
pouch seals. Food Technol. 30(2), 38-48.

Luh, B.S., Leonard, S., Simonine, M. and Villareal, F.,
1964. Aseptic canning of foods VII, strained lima
beans. Food Technol. 18, 363-366.

MacKinney, G. and Joslyn, M., 1940. The conversion of
chlorophyll to pheophytin. J. Am. Chem. Soc. 62,
231-232.

MacKinney, G. and Joslyn, M., 1941. Chlorophyll-pheophytin.
Temperature coefficient of the rate of pheophytin
formation. J. Am. Chem. Soc. 63, 2530-2531.

Manson, J.E., Zahradnik, J.W. and Stumbo, C.R., 1970.
Evaluation of lethality and nutrient retentions
of conduction-heating foods in rectangular con-
tainers. Food Technol. 24(10), 109-113.

Nelson, A.I. and Steinberg, M.P., 1956. Retorting foods
in plastic bags. Food Engineering 28(1), 92-93.

Neter, J. and Wasserman, W., 1974. Applied Linear Statis-
tical Models. Richard Dirwin Inc., Homewood, IL
60430, 1-92.

Ordway, G.B. and Schultz, G.L., 1972. Inspects Package
Seals. Food Engineering 44(2), 64-65.

Pelletier, O. and Madere, R., 1977. Automated determina-
tion of thiamine and riboflavin in various foods.
Journal of the AOAC 60(1), 140-146.

Pflug, I.J., 1964. "EvolutiOn of heating media for pro-
ducing shelf stable foods in flexible packages
Phase I." Final Rep. Contract DAl9-AMC-l45 (N)
U.S. Army Natick Laboratories, Natick, Massachusetts.

Pflug, I.J. and Barrero, C., 1967. "Heating media for
processing foods in flexible packages Phase II".
Tech. Rep. 67-47 G.P., U.S. Army Natick Labora-
tories, Natick, Massachusetts.

Pflug, I.J., Bock, J.H. and Long, F.E., 1963. Steriliza-
tion of food in flexible packages. Food Technol.
17(9). 87-92.

 

64

Pflug, I.J. and Long, F.E., 1966. Static load tests
evaluate flexible packaging materials at elevated
temperatures. Package Engineering 11(5), 91-95.

Roy, R.B. and Conetta, A., 1976. Automated analysis of
water-soluble vitamins in food. Food Technol.
30(10), 94, 96, 98, 100, 103-104.

Rubinate, F.J., 1964. Army's "Obstacle Courses" yield a
new look in food packaging. Food Technol. 18(11),
73-74.

R.W.P. Flexible Packaging, 1974. Outline specification
for sterility MFEP". Bristol, England.

Schulz, O.T. and Olson, F.C.W., 1940. Thermal processing
of canned foods in tin containers. III. Recent
improvements in the general method of thermal
process calculation--a special coordinate paper and
methods of converting initial and retort temperature.
Food Reserach 5, 399-407.

Sistrunk, W.A. and Cain, R.F., 1960. Chemical and physical
changes in green beans during preparation and
processing. Food Technol. 14(7), 357-362.

Shenkemberg, D., 1975. USDA Status on flexible retortable
pouches for meat products. Act. Rep. Res. Dev.
Assoc. Mil. Food Packaging System 27(1), 157-160.

Steel, R.G.D. and Torrie, J.H., 1960. Principles and
procedures of statistics. McGraw-Hill Book Co.,
Inc., New York. 107-108, 194-230.

Stumbo, C.R., 1949. Thermobacteriology as applied to food
processing. Advances in Food Research 2, 47-115.

Stumbo, C.R., 1973. Thermobacteriology of Food Processing.
Academic Press, New York. 329 pp.

Tan, C.J. and Francis, F.J., 1962. Effect of processing
temperature on pigments and color of spinach.
J. Food Sci. 27, 232-241.

Thorne, S., 1976. Retortable pouches. Nutrition and Food
Science 45, 2-4.

Torpe, R.H. and Atherton, D., 1972. "Sterilized foods in
flexible packages". Technical Bull. No. 21 Fruit
and Vegetable Preservation Research Association.
Chipping Campden, Gloucestershire, England.

 

65

Tung, M.A., Garland, M.R. and Campbell, W.E., 1975.

Quality comparison of cream-style corn processed
in rigid and flexible containers. Can. Inst. Food
Sci. Technol. Journal 8(4), 211-216.

Tung, M.A., Garland, M.R., Maurer, A.R., 1977. High

quality, heat processed vegetable products prepared
in flexible pouches. Food Product Development
11(7), 112, 117 and 121.

Van Buren, J.P., Moyer, J.C. and Robinson, W.B., 1962.

Pectin methylesterase in snap beans. J. Food Sci.
27, 291-294.

Van Buren, J.P., Moyer, J.C., Wilson, D.E., Robinson, W.B.

Yamano '

Yamano,

Yamano,

Zoueil,

and Hand, D.B., 1960. Influence of blanching con-
ditions on sloughing, splitting, and firmness of
canned snap beans. Food Technol. 14, 223-236.

Y., Ejire, K., Endo, T. and Sendo, M., 1975. Heat
penetration into flexible packages, with food

simulated materials. J. Food Sci. and Technol.,
Japan 25(5), 199-204.

Y. and Komatizu, Y., 1969. Sterilization of foods
in flexible packages, Part I. Heat characteristics
of a pilot retort for steriliZation of flexible
packages. J. Food Sci. and Technol., Japan 16(3),
113-118.

Y., Komatizu, Y. and Ikegami, Y., 1969a. Sterili-
zation of foods in flexible packages, Part II.
Thermal characteristics and storage stability of
various film-wrapped foodstuffs. J. Food Sci. and
Technol., Japan 16(3), 119-123.

M.E. and Esselen, W.B., 1959. Thermal destruction
rates and regeneration of peroxidase in green beans
and turnips. Food Research 24, 119-133.

APPENDIX A

APPENDIX

Statistical Analysis
Two distinct statistical analyses were performed
in this work. The first one was done to evaluate the
general effect of the pouch treatments and the 303 can
treatment on quality of processed green beans.
The second one was performed in order to evaluate
the main effect factors and their interactions on quality

of the processed green beans.

General Analysis of Variance

 

k = 9 treatments n = 10 observations
1. Mean of observation in the ith sample (i = 1,2,...K).
_ nl
.1037 2 xi.
1 j=1 3
2. Standard deviation of observation in the ith sample.
n1 2 —2 1/2
S1 ‘ [(.E X ij ‘ “1X1) /n1 ' 1]
3—1
3. 001: Z Xf
j=l 3
4. Total sum of squares
k E 21 2
_ n1 2 ‘._ ._ Xij’
TSS — Z 2 xi. - 1-1 3-1
i=1 j=1 3 k
2 n
j=1 1

66

10.

11.

12.

67

Treatment sum of squares

n1 2 k

( Z X..) ( Z Z X..)
j=1 13 =
1 n1 - k

TrSS =

 

 

"MW

Error sum of squares
ESS = TSS - TrSS
Treatment degree of freedom

dfl = K - 1

Error degree of freedom

k
df2 = .E n. - K
1—

Total degree of freedom

Treatment mean square

TrSS

TrMS = df

 

I-‘

Error mean square

EMS = ———

The F ratio

TrMS
EMS

 

F = (with degrees of freedom dfl, dfo)

68

Analysis of variance for pouch treatments

Step I
Compute -
Correction term = C
Total SS = TSS
Treatment SS = TrSS
Error SS = TSS - TrSS
Step II

The treatment sum of squares is partitioned into

components attributable to main effects and interactions.

 

 

Factors -
T - Thickness levels t1, t2,...,ti
P - Superimposed air pressure-levels
pl, pzpooo’pj
B - Brine addition-levels bl,b2...,bk
r - replication 10
>3<pj>2
SS(P) = rtb - C
2(ti)2
_ i _
SS(T) — rpb C
20,92
k
SS(B) rpt - C
2
Z (t.p.)
ij 1 3
SS(TP) = ' - C - SS(T) - SS(P)
rb
z (t.bk)2
i k 1
SS(TB) = ' - C - SS(T) - SS(B)

 

rP

SS(BP)

SS(TPB)

69

 

 

 

 

 

 

 

2: (bp.)2
103"”
rt - C - SS(B) - SS(P)
Z (t.p.bk)2
ijklj
' ' r - C - SS(T) - SS(P) - SS(B)
SS(TP) - SS(TB) - SS(PB)
Linear Regression Analysis
2x1 Zyl
Zx.y. -
1 1 n
b = 2
2x2 - (2X1)
i n
Zy. [X
a: l—b—
n n
Ex - 3:— Zyi
iyi n
r: 2 2
ZX-) (237.)
2 l 2 2 l 2
:2x.- 11/ [2y- 11/

APPENDIX B

70

Table 17. Thiamine concentration in ug/gram1 (dry weight
basis) of processed green beans in pouches

 

 

g S. Imposed air Brine = B Total
Thickness T pressure 3 P b b b + b
1 2 l 2
p1 55.90 tlplb1 49.58 tlplb2 105.48 tlp1

 

 

 

 

 

p2 49.60 tlpzb1 43.20 tlp2b2 92.80 tlp2
Total‘pl-i-p2 105.50 tlb1 92.78 tlb2 198.28 tl
p1 54.47 tzplb1 50.68 tzplb2 105.15 tzp1
t:2
p2 48.79 t2p2b1 44.96 t2p2b2 93.75 t2p2
Totalpl-l-p2 103.26 tzb1 95.64 t2b2 198.90 t2
Total pl 110.37 plb1 100.26 bzp1 210.63 p1
t1 + t2 p2 98.39 pr1 88.16 p2b2 186.55 p2
Total p1+p2 208.76 b1 188.42 b2 397.18 C

 

1Each value is a total of ten replicates of each treatment. Data from
Table 5

71

Texture in pounds force/gram1 of green beans

 

 

 

 

 

 

 

Table 18.
processed in pouches.
Thickness Superimposed Brine = B Total
- T air pressure - P b1 b2 b1 + b2
pl 7.21 tlplbl 6.11 tlplb2 13.32 tlp1
t1 .
p2 7.68 tlpzbl 5.11 tlpzb2 12.79 tlp2
Total pl+p2 14.89 tlb1 11.22 tlb2 26.11 tl
p1 5.49 tzplbl 5.93 tzplb2 11.42 tzp1
t2
p2 4.71 tzpzb1 6.69 tzpzb2 11.40 tzp2
Total p1+p2 10.20 tzb1 12.62 tzb2 22.82 t2
Total t1+t p1 10.70 plb1 12.04 plb2 24.74 p1
p2 12.39 pr1 11.80 pr2 24.19 p2
Total p1+p2 25.09 bl 23.84 b2 48.93 G

 

1
Each value is a total of ten replicates of each treatment.

Data came

from Table 10.

72

Color ratio -a/bl of green beans processed in

 

 

 

 

 

 

 

Table 19.
pouches.
Superimposed
Thickness air Brine = Total
= T pressure = P b1 b2 b1 + b2
p1 -0.752 tlplbl -2.429 tlplb2 -3.181 tlp1
t1
p2 -O.518 tlp2b1 -1.305 tlpzb2 -1.823 tlp2
Total p1+p2 -1.270 tlb1 -3.734 tlb2 -5.004 tl
p1 -O.575 t2p1b1 -O.927 tzplb2 -l.502 tzpl
t‘2
p2 -0.818 t2p2b1 -0.369 tzpzb2 -1.187 tzp2
Total p1+p2 -l.393 th1 -1.296 th2 -2.689 t2
Total pl -l.327 plb1 -3.356 plb2 -4.683 p1
t1 + t2 p2 -1.336 pzbl -1.674 p2b2 -3.010 p2
Total p1+p2 -2.663 bl -5.03O b2 -7.693 G
1Each value is a total of ten replicates of each treatment. Data came

from Table 13.

"1111111011@1111)?