Mgency
Industrial Environmental
Laboratory
Cincinnati OH 45268
EPA-600/2-79-193
October 1979
Research and Development
Intact or Unit-Kernel
Sweet Corn
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further deve10pment and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1.
2.
3.
4.
5.
6.
7
8.
9.
Environmental Health Effects Research
Environmental Protection Technology
Ecological Research
Environmental Monitoring
Socioeconomic Environmental Studies
Scientific and Technical Assessment Reports (STAR)
Interagency Energy-Environment Research and Development
"Special" Reports
Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technicallnforma-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-79-193
October 1979
INTACT OR UNIT-KERNEL SWEET CORN
by
G. H. Robertson, M. E. Lazar, D. F. Farkas,
J. M. Krochta, and J. S. Hudson
Western Regional Research Center
U.S. Department of Agriculture
Berkeley, California 94710
and
F. Pao, B. Terrell, and J. Farquhar
American Frozen Food Institute
McLean, Virginia 22101
Grant No. R-804597-01-1
Project Officer
H. W. Thompson
Industrial Pollution Control Division
Industrial Environmental Research Laboratory
Corvallis, Oregon 97330
This project was conducted
in cooperation with
American Frozen Food Institute
McLean, Virginia 22101
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
ii
-------
FOREWORD
When energy and material resources are extracted, processed, converted,
and used, the related pollutional impacts on our environment and even on our
health often require that new and increasingly more efficient pollution
control methods be used. The Industrial Environmental Research Laboratory-
Cincinnati (lERL-Ci) assists in developing and demonstrating new and improved
methodologies that will meet these needs both efficiently and economically.
This report discusses research work on a processing modification for
sweet corn. A small-scale system evaluated the methods of removing the
kernels of corn intact from the cob. This would significantly reduce the
quantity of organic material in the effluent wastewater and result in
greater yields. For further information on this project contact the Food and
Wood Products Branch, Industrial Pollution Control Division, Industrial
Environmental Research Laboratory-Cincinnati.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
iii
-------
ABSTRACT
Intact or unit kernels of sweet corn were substituted for conventional
or cut kernels in an attempt to reduce the copious amounts of liquid waste
produced during processing. A small-scale, simulated processing line was
set up to evaluate processing advantages and disadvantages, to establish
methods for producing intact kernels, and to generate samples by which the
product quality could be evaluated and compared to the conventionally pre-
pared product. Sweet corn varietal suitability for intact kernels was also
evaluated.
When compared to a conventional cutting process with a yield of 40 to
33 parts per 100 parts of corn in husk, the processing of intact kernels
resulted in corresponding waste reductions of 85% to 94% (based on chemical
oxygen demand per pound of product) and yield increases of 5% to 26%.
Sensory comparisons showed that regardless of the variety or method of
preservation, intact kernel samples received higher mean scores (hedonic
ratings) or were preferred by a larger percentage of the panel (paired pre-
ference rating) than the cut controls. In addition, intact kernels were
shown to be 14.5% higher in fiber content, 100% greater in fat content, and
5% to 16% higher in protein.
One method for producing intact kernels was preferred because the
kernels it produced yielded lower waste loads and were judged to be higher
in quality. However, the application of this method is probably subject to
varietal development of sweet corn cultivars with loose kernels. Processing
considerations would favor development of varieties with weak attachment
to the cob and strong adherence between adjacent kernels.
This report was submitted in fulfillment of Grant No. R-804597-01-1
by the American Frozen Food Institute under the sponsorship of the U.S.
Environmental Protection Agency. This report covers the period 15 July 1976
to 15 July 1978, and work completed as of 15 July 1978.
iv
-------
CONTENTS
Foreword iii
Abstract iv
Figures vi
Tables viii
Abbreviations x
Acknowledgments xi
1. Introduction 1
2. Conclusions 2
3. Recommendations 3
4. Experimental Materials and Procedures 5
5. Results and Discussion 19
References 52
Appendices
1. Waste Indices 53
2. Preliminary canning studies 54
-------
FIGURES
Number Page
1 Field map indicating location of 1976 plantings and estimated
direction of prevailing wind (p.w.) 5
2 First and second plantings for 1977 6
3 Third through seventh plantings for 1977 6
4 Schematic of "hole-saw" unit 8
5 Front and side view of element used to longitudinally
split ears of sweet corn 9
6 Apparatus used to split ears of corn 9
7 Schematic of textured-surface process for intact kernels . . 10
8 Schematic of smooth-surface process for intact kernels. . . 11
9 Smooth-surface frictional removal of sweet corn kernels. . . 12
10 Cross section illustrating corn split section position
with respect to frictional removal element 15
11 Cross section of a 45 x 3 kernel displacement test 17
12 Cross-section of a 90 x 1 or 90 x 2 kernel displacement
test 18
13 Liquid waste strength of cut and intact-kernel samples for
washing A and blanching B 20
14 Waste strength for cut-kernel samples 21
15 Kernel damage resulting from contact with frictional element. 24
16 Yield increase of intact over cut kernels for freshly pre-
pared kernels 25
17 Yield increase of intact over cut kernels for washed
kernels
vi
-------
18 Yield increase of intact over cut kernels for frozen
kernels 26
19 Ideal yield of intact sweet corn 26
20 Density flotation of sweet corn 31
21 The role of kernel position relative to friction surface
in successful kernel removal 32
22 Freshly prepared, unwashed samples of cut and intact
sweet corn 34
23 Frozen and thawed samples of cut and intact sweet corn
(var. Stylepak) 35
24 Frozen and thawed samples of cut and intact sweet corn
(var. Golden Happiness) 35
25 Frozen and thawed samples of cut and intact sweet corn
(var. Golden Jubilee) 36
26 Corn cross sections and individual kernels of (1 to r)
Golden Happiness, Golden Jubilee, and Stylepak 38
27 Adaptation of USDA interpretive guide illustrating pulled
kernels and cut kernels in canned and frozen corn .... 39
28 Kernel rupture during 45 x 3 (B) and 90 x 1 or 90 x 2
testing (A) 43
29 Rates of removal of sweet corn from Golden Happiness (A),
Golden Jubilee (C), and Stylepak (B) by friction with a
moving, smooth, neoprene surface 48
30 Schematic free-body diagram for displacement of sweet
corn kernels 49
vii
-------
Number
TABLES
1
Process for sweet-corn experiments
.....
.........
2
Washer and blancher liquid-effluent strengths.
.....
3
Contributions to effluent strength.
..........
4
Suspended solids collected during washing of cut and intact-SS
kernels (moisture-free basis). . . . . . . . . . . . .
5
Approach to complete or ideal kernel removal by smooth-surface
and textured-surface processes. . . . . . . . . . . . . . .
6
Ear location of kernels not detached by frictional technique. .
7
Mass recovery of sweet corn after each process step.
.....
8
Mean kernel weight distributions for freshly prepared
samples. ........ .... .........
9
Mean frozen kernel weight distributions. . . .
.....
10
Intact kernel yield comparison for textured and smooth
processes . . . . . . . . . . . . . . . . . .
. . . .
11
Predicted losses of intact kernels from intact-HS kernel
mixtures by flotation in NaCl solutions. . . . . . . .
12
Static coefficient of friction (ns) for clean and contaminated
neoprene surfaces. . . . . . . . . . . . . . . . . . . . . .
13
Mean incidence of glume tissues in freshly prepared intact-TS
kernel samples. ......................
14
Absolute change in distribution from washing intact kernels. .
15
Mean hedonic ratings for cut (normal) and intact-SS sweet
corn frozen in 1976 . . . . . . . . . . . . . . . . . . . .
16
Mean hedonic ratings for frozen and canned, cut and intact-SS
sweet corn prepared during 1977. . . . . . . . . . . . . . .
viii
Page
14
22
22
23
27
28
29
29
30
30
31
33
37
37
40
40
-------
17
18
19
20
21
22
23
24
A-1
A-2
A-3
Paired preference ratings for intact and cut sweet corn.
Fiber and total nitrogen in cut and intact sweet corn.
. . . .
Total amino acid content increase for intact sweet corn
relative to cut. . . . . . . . . . . . . . . . . . . . . .
Maximum total force per kernel measured during kernel
displacement. . . . . . . . . . . . . . . . . . .
.....
Sweet corn yield changes due to heat (5 min, 100°C steam)
preconditioning of ears to effect kernel loosening. ....
Thermal processing to alter strength of kernel attachment.
Maximum component forces measured during kernel displacement. .
Comparison of per-kernel force by actual 45 x 3 tests to force
by computed 45 x 3 tests based on component abscission and
interkernel frictional forces in 90 x 1 or 90 x 2 tests
Correlation of waste indices (I) with total organic carbon
measurement. . . . . . . .
. . . .
......
. . . .
Recommended stationary-retort processing conditions.
.....
Recommended agitated-retort processing conditions.
......
ix
41
41
42
43
44
45
47
51
53
54
55
-------
Analytical Measures
BOD
COD
N
SS
TOC
TS
SD
Kernel types
Intact - DC
Intact - HS
Intact - SS
Intact - TS
ABBREVIATIONS
-- Biological oxygen demand
-- Chemical oxygen demand
-- Force in newtons
-- Suspended solids
-- Total organic carbon
-- Total solids
-- Standard deviation
-- Intact sweet corn kernels produced by deep cutting
using commercial cutter.
-- Intact sweet corn kernels produced by extremely deep
cutting using experimental hole-saw cutter.
-- Intact sweet corn kernels produced by friction between
kernels and a moving smooth surface.
-- Intact sweet corn kernels produced by friction between
kernels and a moving textured surface.
Heat penetration parameters in canning
fc
fu
f2
Xbh
Time lag before the heating curve assumes a straight
line on semilog paper.
-- Time required for the straight line part of the
cooling curve to traverse one logarithmic cycle.
-- Time required for the straight line part of heat
penetration curve to traverse one logarithmic cycle.
-- Time required for the straight line portion of
the heat penetration curve following a break in the
initial heating curve to traverse one logarithmic
cycle.
-- Time from corrected zero time of the process to
the break in the heat penetration curve.
x
-------
ACKNOWLEDGMENTS
The authors wish to express their appreciation to Harold
Thompson, U.S. Environmental Protection Agency Project Officer, for his
guidance on this project. In addition, we would like to thank the follow-
ing personnel of the American Frozen Food Institute, who provided
Administrative assistance: Joanne Cox, Elaine Carter, Jean Bohannon and
Ray McHenry. We also wish to acknowledge the following persons and
organizations for their assistance. Professor Walton C. Galinat (University
of Massachussetts) and David Galinat who provided sources of Golden
Happiness seed. The University of Illinois Foundation (Champaign, Ill.)
provided Illini Xtra Sweet seed and Rogers Brothers Seed Co., Nampa, Idaho
provided Golden Jubilee seed. Dr. Charles Geise and Dr. William F. Hagan
(San Leandro, Calif.) and Ray Boone and Albert Erojo (Gilroy, Calif.) of
Del Monte Corporation, Agricultural Research assisted in the horticultural
aspects of the project. Don Corlett and Marty Fischler of of Del Monte
Corporation (Walnut Creek, Calif.) provided C-enamel cans. Jay Unverferth,
Christina Merlowe, Larry Lewis, and Richard Farrow of the National Food
Processors Association (Berkeley, Calif.) performed chemical analyses and
thermal studies, and prepared canned samples. Mike Haney (Rochester, Minn.)
and Joe Ohler and Jim Albrecht (Chicago, Ill.) of Libby, McNeill and Libby
provided assistance with and loan of corn processing machinery. Dave Pahl
of the Northwest Food Processors Association (Portland, Ore.) surveyed the
northwest canning and freezing industry regarding their practices and
interest. Bob Gallion (Yakima, Wash.) of Libby, McNeill, and Libby, Paul
Cover of the United Co. (Westminster, Md), Leslie Vadas and Jerry Smith of
Food Machinery Corporation (San Jose, Calif.) and Alvin Randall of the
Wisconsin Canners and Freezers Association (Madison, Wise.) provided special
assistance and encouragement of our objectives. Jos~ ~errotte (Brazil)
assisted in developing equipment prototypes, and Dante Guadagni (Western
Regional Research Center) supervised the sensory analyses.
xi
-------
SECTION 1
INTRODUCTION
Sweet corn processing effluent is a large contributing source of the
total liquid wasteload originating from vegetable processing in the United
States. Sweet corn is estimated to contribute 50 million kgs of BOD or
20% of the BOD from all vegetable processing sources (1). Much of this
loading occurs through process water contact with the cut surfaces of
individual corn kernels. These cut kernels are produced by forcing husked
ears of corn endwise against stationary or rotating knives. The washed
kernels are either canned or blanched and frozen as "whole kernel" corn
(2,3) or they are macerated and then canned in a mixture with cut kernels
as cream-style corn (4).
The substitution of intact or unit kernels was proposed earlier (5)
as a means of reducing the effluent loading. Intact kernels are separated
from the cob tissues at the natural abscission layer between kernel and
cob, do not have a cut surface, and completely contain the kernel juices
or "milk", thereby preventing transfer of these juices to the process water.
These authors found the effluent-reducing potential for manually produced
intact kernels to be about 80% for washing and blanching, and the yield
improvement potential to be approximately 20% on a per-kernel basis.
The research reported here describes the continuation of the evalua-
tion of intact-kernel sweet corn. It includes a detailed evaluation of the
wasteload characteristics of the different kernel styles, evaluation of
different methods for producing intact kernels, characterization of a
limited number of sweet corn varieties with respect to their suitability
for intact-kernel producing processes, and sensory comparison of canned
and frozen samples of cut and intact corn to assess consumer reaction.
1
-------
SECTION 2
CONCLUSIONS
The liquid wasteload from sweet corn processing that originates during
washing and blanching can be drastically reduced by substitution of intact
kernels for cut kernels. In theory, methods that effect detachment of
intact kernels without rupture or other damage to the kernel can virtually
eliminate the wasteload of processing waters.
The recovery of a substantial amount of additional corn solids is
another advantage of the intact kernel as opposed to the cut kernel.
Although solids recovery may differ in a commercial-scale device, this
factor is likely to continue as a strong incentive for intact-kernel prepara-
tion methods.
Results of controlled sensory tests are added incentives for the sub-
stitution of intact for cut kernels. Strong preferences were expressed for
intact frozen corn, and less strong, but significant preferences were indi-
cated for intact canned corn. Preferences were based on flavor and texture.
Intact corn was found to have more lipid material and slightly more fiber
than the cut comparison.
Several methods can successfully produce intact kernels. However,
when factors such as kernel quality and kernel separability are accounted
for, techniques that apply friction to kernels and cause initial kernel
separation at or near the natural separation or abscission zone appear to be
the most successful. These methods are subject to the requirement for new
materials handling methods when applied to split sections of ears and are
sensitive to ear position with respect to the friction surface. The quality
of the frictional contact greatly influences the quality and success of the
kernel detachment event.
Of two frictional methods applied, a technique using a smooth friction-
developing surface offered the advantages of continuous cleaning capability
and control over the friction surface quality.
Strength of kernel attachment to the cob plays an important role in the
Success of frictional removal methods. Kernels with very high strengths of
attachment can be difficult to detach before rupture of the kernel
occurs. Most varieties tested had kernel detachment strengths that enabled
successful kernel removal. However, since the detachment strength varied
over wide limits, varietal selection will be an important factor to the
successful production of intact kernels.
2
-------
SECTION 3
RECOMMENDATIONS
1.
Additional evaluation of the waste reduction potential of intact kernels
by monitoring the washing and blanching operations is unnecessary.
However, waste loads from the washing of small kernel samples can be
used to index the extent of damage originating from a given experimental
kernel-producing method.
2.
The wide differences in kernel attachment strength that were identified
in the testing of a very small population of sweet corn varieties indi-
cate that opportunities do exist for breeding selection for this
characteristic. A wider examination of the currently available varie-
ties should be conducted to identify currently suitable types and/or
identify genetic material for future breeding selection.
3.
There remains a need for a study of factors related to the ease of
kernel removal such as the development of pericarp strength and the
maturation of abscission tissues. Factors that can be imposed to
accelerate the kernel abscission and weaken the kernel attachment should
also be examined.
4.
Close contact between the breeding community and researchers involved
with machine invention, design, and development is strongly urged, since
the design of the kernel removal process and equipment depends on the
nature of the sweet corn. If a very loosely attached kernel can be
developed, methods such as the hole-saw technique should be reconsid-
ered, since cob tissue adhering to kernels might then be amenable to
removal by a tumbling or abrasive milling.
5.
Further development of the smooth belt process described herein should
be pursued. In particular, automatic methods for handling split ear
sections and for presentation of split sections to the frictional
surface should be investigated. Alternatively, methods for handling
sweet corn ears from which several rows have been removed should be
investigated. Such a development would enable estimation of equipment
size and production rates.
6.
As soon as the mechanical problems have been overcome, larger test
packs for consumer, focus panel, or market studies should be initiated
with an interested canner or freezer or a representative association.
Field testing with varieties and under conditioDs normally encountered
in corn producing areas should also be evaluated.
3
-------
7.
An ultimate goal of this work, a workable in-field or side-of-field
process is difficult to evaluate w1thout completion of recommended
research, but a preliminary design study might be a useful exercise.
8.
The extension of the unit kernel concept to the marketing of fresh corn
may be an interesting opportunity. Fresh, refrigerated, intact kernels
have keeping and eating qualities that should be evaluated relative to
corn-on-the-cob. Advantages would include greatly reduced shipping bulk
and transportation costs.
Kernel-detachment-testing recommendations.
9.
The ear-sample to ear-sample uniformity of the corn should be controlled.
This could be achieved by field selection and use of "prime" or first
ears only and by controlled maintenance of individual plants through a
technique such as drip irrigation.
10.
The genetic integrity of the corn should be maintained. This would be
achieved by homogenous plantings isolated from contaminating varieties.
11.
The component of interkernel friction and the related loosening by
adherence should be more effectively eliminated by applying a 90 x 1
test to ears from which alternate kernels have been cut away.
12.
The measurement procedure could be improved by measurement of both x-
and y- force components of force applied to kernels, by measurement of
the dimensions and location of the abscission, and by measurement of the
energy of abscission rather than-the maximum force required.
13.
The role of abscission altering agents other than heat should be
investigated. Chemical or biochemical agents which alter the abscission
process may reduce the net detachment strength. Indeed, the nature of
the abscission of kernels at this stage in their deveiopment has not
been investigated.
4
-------
SECTION 4
EXPERIMENTAL PROCEDURES
RAW MATER IALS
Sweet corn for these studies was grown during 1976 and 1977 on irri-
gated land in Gilroy, Calif., by subcontract to the Agricultural Research
Division of the Del Monte Corporation. The varieties Golden Jubilee and
Stylepak were grown as representative of commercial processing sweet corn.
The glumeless Golden Happiness and the sugary Illini Xtra Sweet were
also grown.
During both years, sweet corn was planted in the late spring. On each
date, at least two rows of each of two varieties were planted. The dates
and field arrangements are shown in Figure 1, 2 and 3. Seed was planted
at a rate to yield a 20-cm interplant and a 76-cm inter-row spacing. All
of the seed was pretreated to insure a high germination rate, but no pre-
cautions were taken to prevent insect or bird infestation of the growing
plants. Harvest of mature sweet corn began in August and continued until
early Oc tober.
PLANTING
DA TES
5.5-76
JUBILEE 121
SlYlEPAK 12)
JUBilEE 121
SIYlEPAK (21
JUBilEE 121
SIYlEPAK 12)
JUBilEE 12)
SIYlEPAK (2)
JUBilEE (2)
GOl DEN HAPPINESS (2)
JUBilEE (21
SIYlEPAK 121
JUBilEE 121
51YlEPAK 121
S-IB.76
5-28-76
6-B-76
6-15-76
6-22-76
6-29 -76
.
p~o m
VARIETY AND NUMBER OF ROWS
N~I'"
Figure 1.
Field map indicating location of 1976 plantings and estimated
direction of prevailing wind (p.w.).
5
-------
5-5.71
5-5.17
S.18~71
5-1B-77
VARIETY AND
NUMBER OF ROWS
JUBilEE (7)
5TYlEPAK (51
JUBilEE (2)
5TYlEPAK 12)
N~I'W
PLANTING
DATES
183 m
Figure 2.
First and second plantings for 1977.
PLANTING
DATES
5-27-77
N --i I'W
6-8.11 JUBILEE
(101
6.8.17 STYLEPAK
(8)
6-15.77 GOLDEN (~tPPINESS
t-
..,
~~
I'~
"'1<
~
1-(.-
~~
~
0....
-Po
~J'
5-21.71
6-15-77
5TYlEPAK
(61
JUBILEE
(41
6-29-77
STYlEPAK
(4)
EXPERIMENTAL V. VARIETIES (2)
STYlEPAK (21 V. VARIETIES (2)
IlliN. XTRA SWEET (1) STYlEPAK (2)
6-23-77
6-23-77
92 m
Figure 3.
Third through seventh planting for 1977.
6
-------
The fresh market variety Vanguard (Asgrow Seed Co) was available
before the processing season. This variety was grown, hydrocooled, and
iced by Alonzo Farms, Dixon, Calif. and obtained in Dixon or from a local
wholesale produce market.
PROCESSING METHODS
Sweet corn maturity was monitored as moisture content of machine-cut
kernels produced from a field sample of 10 or more "prime" or first ears.
Harvest of sweet corn at processing maturity of 74% to 70% moisture was
restricted to corn from one planting only. Sweet corn was harvested
manually, transported by truck to Albany. Calif., composited to eliminate
picker bias, divided into sublots of 16 to 130 kg and either husked in a
commercial husking machine (1976) or husked manually (1977). The ears
of each sublot were trimmed to exclude insect, bird, and microbial damage.
Sublots of trimmed corn then were delivered to the appropriate kernel
generation stations.
Cut Kernels
Kernels were produced using a commercial cutting machine (FMC
Corporation, Model 3AR). Cutting depths were adjusted to produce kernels
cut above the embryo (shallow cutting depth) and kernels cut midway
through the embryo or at the glume line (normal cutting depth) (5). During
normal plant operation a cutter of this type would have a low flowrate
stream of water directed over the blades. This stream continuously cleans
the cutter but combines with the kernel mass. The associated liquid waste
is eluted from the kernel mass during washing or blanching. This stream
was not used during the short duration process runs in this study. How-
ever, at the completion of a cutting test, the cutter was carefully
cleaned with water which was collected and measured. A sample was drawn
from this water for effluent characterization. Since this cutter cleaning
effluent was considered to be a normal part of the wash effluent in the
continuous use of this unit, it is so reported below.
Intact Kernels
Four methods were applied for the production of intact or unit kernels.
Two of these methods were based on the cutting principle: separation of the
corn kernel by impacting the tissue near the point of attachment to the cob
with a sharp, rapidly moving object. Two other methods were applied which
operated by applying forces to the top surface of the kernel and stressed
the kernel so that separation would occur at the natural abscission layer.
The first alternative, a cutting method, used a conventional cutting
machine set for its maximum or deepest cut. Cutter washing was conducted
as above. Kernels produced by this method are referred to as intact-DC
and the process is identified as the deep-cut process.
The second alternative, a cutting method, used a specially designed
unit to cut at depths greater than attainable by the conventional cutting
7
-------
machine and is shown in Figure 4. This unit or hole saw consisted of a
cylinder having a diameter less than or equal to the diameter of the circle
described by the locus of the kernel abscission layers. One end of this
cylinder was machined to provide self-cleaning teeth. The entire cylinder
was externally driven around its cylindrical axis by suitable mechanical
means. To produce intact kernels, an ear of corn was forced endwise into
the rotating teeth. The core of the ear passed through the cylinder
and exited at its far end; whereas, the kernels and "sawdust" dropped into
a receiving bin. Since corn ears with widely differing diameters were
tested, several sawing elements with suitable diameters were provided.
E
s
s
Figure 4. Schematic of "hole-saw" unit. Each ear of corn (C) is supported
on a guide (A) and forced into the cylindrical cutter (H). This cutter tube
is supported (S) and rotated (a) about its axis. Kernels are collected in a
pan (K) and stripped cobs leave the cutter tube at its exit (E).
Kernels produced by this unit are designated intact-HS and the process
is identified as the hole-saw process. The kernels and sawdust from the
hole-saw process were separated by screening. Yields reported below are for
freshly produced and screened kernels.
The third and fourth alternatives were each preceded by an ear dis-
assembly step which was designed to gain access to the kernels with a
minimal loss of yield to damaged kernels. This disassembly step was the
longitudinal splitting of each ear and was accomplished by forcing a narrow
wedge into each ear from one end through to the other end. The wedge
dimensions, 19-mm wide on the flat side with a 7.60 taper, were chosen so
so that the cob tissue only was affected during the splitting and kernels
were not cut or damaged. To split accurately, the wedge was fitted with a
5-mm diameter, 20-cm long, center-mounted guide pin (see Figure 5).
In operation, this pin was inserted into and through a O.64-cm diameter
8
-------
guide hole which had been drilled into the pithy core of each sweet corn
ear from butt to tip. Next. the wedge-opposite end of the supported ear
was restrained. and then pressure was applied to the wedge to force it
into and through the ear of corn (Figure 6).
c
D
e
~
Figure 5. Front and side view of
element used to longitudinally split
ears of sweet corn. Dimensions
shown are A. 19 mm; B. 5 mm; C.
200 mm; D. 19 mm; and 8. 7. 60 .
Figure 6. Apparatus used to
split ears of corn.
9
-------
Each of the third and fourth alternative methods produced intact
kernels by pressing ear split sections against a moving, friction develop-
ing surface. Here, an ear section was oriented so that the moving surface
in contact with the kernels was tangent to the ear radius, and kernels
were removed a row at a time.
The friction surface of the third alternative was a custom molded,
s.l-cm wide, 45 to 50 durometer, silicone rubber belt (General Electric
liquid molding compound type RTV-630 with 10% catalyst). The surface of
the belt was molded with integral, 0.9s-cm long, 0.32-cm diameter rubber
projections, which were arranged in rows parallel to the axis of rotation.
The center-to-center inter-projection spacing in each row and the center-
to-center inter-row spacing was 0.48 em. The projections in alternate
rows were staggered. Rows were grouped in threes with a 1.27-cm center-
to-center space between the projections of the closest rows of each group
of three. This spacing was provided to allow indexing of the rows
of kernels with the grouped rows of projections. This deeply textured
belt was mounted on a 11. I-em diameter, vertical wheel and driven at 60
rpm by a 370 watt (1/2 hp) motor.
Ears were positioned manually (Figure 7) at the 3 o'clock position
of the clockwise rotating wheel and pressed against the textured surface.
As each row of kernels was removed, the ear was rotated counterclockwise
around its long axis in crder to bring an undetached row of kernels into
contact with the surface. Detached kernels were collected in a pan placed
under the wheel. A 150° to 200°C air stream was directed across the con-
tact surface of the wheel in order to evaporate corn juices and maintain
friction with the kernels. The kernels produced by this method are
referred to as intact-TS and the process is identified as the textured-
surface process.
~c
p~
Figure 7.
Schematic of textured-surface process for intact kernels.
Shown are working wheel (E), air heater (H), sweet corn ear
(C), and collecting pan (P).
10
-------
The friction surface of the fourth alternative was a 10.8-cm wide,
3-mm thick, 58-durometer, food-grade neoprene belt. This belt was mounted
on two ll.l-cm diameter wheels as shown in Figure 8. The working wheel
(E) shown in the Figure 8 had a 0.64-cm thick overlying belt of 50-durometer
sil icon rubber. Two 0.15 gm/ cm3, closed cell, urethane foam "bumps" were
attached to this belt 1800 apart. These wedge-shaped bumps increased the
working radius to 1.5 cm over a circumferential distance of 7.0 cm. The
tapered bumps were designed to account for ear diameter differences of
rigidly supported ears presented to the surface by a mechanical feeder.
When ears were manually presented to the surface, the bumps served as an
index guide (as did the grouped projections of the textured surface) and
provided increased friction since they allowed for some conformation of
the belt surface around the kernel surface. Figure 9 illustrates the
presentation of a corn ear section to this device.
H
B
~C
w
p~
Figure 8.
Schematic of smooth-surface process for intact kernels.
Shown are drive wheel(D), working wheel (E), index bump (B),
friction surface (F), scrapers (S), water spray (W), hot-air
blower (H). and corn split (C).
11
-------
1r
Figure 9.
,,.,'
AI'
Smooth-surface frictional removal of sweet corn kernels.
1'2
J.i..
-------
Other features of this alternative are indicated in Figure 8 and were
incorporated to insure a high friction coefficient between kernels and
belt. For instance, scrapers were positioned against the belt to remove
corn juices, silk, and kernel fragments. A low-volume water spray was
directed against the belt between the scrapers to effect cleaning, and a
hot air stream from a blower dried the belt.
Wash water collected during process runs was weighed and sampled and
is included as a contributor in the total wash effluents described below.
Kernels produced by this method are referred to in the text as intact-
55 and the process is identified as the smooth-surface process.
Post-Preparation Processing
Corn kernels produced by each method were washed in water using a
pilot scale, shaker washer (A.B. McLauchlan Co., Inc.). Kernels entered
this washer through a II-rom square-opening, woven-wire screen to exclude
oversize extraneous material and exited from the washer over a 6-mm square-
opening, woven-wire screen to exclude undersize extraneous material. Water
flowed into the washer over a weir at the feed end, flowed co-currently
with the kernels, and exited from the washer through the undersize screen.
In addition, water was applied through two rows of fan-spray jets located
at the washer exit. The average ratio of water flow mass to kernel mass was
5.6 to 1. Waterflow was divided equally between the two rows of jets. The
first row of jets encountered by the kernels was pitched to impart forward
motion to the kernels to advance them out of the washing zone; whereas,
the second bank encountered was a perpendicular curtain over the undersize
screen and helped separate debris and kernel fragments from the desired
kernels. Kernels leaving the washer were collected in a perforated basket
and weighed.
All of the water used during washing was collected and sampled for
analysis. During 1977 this water was then filtered through a 32-mesh
screen to recover kernel embryos, silk, and undersize kernels which had
not been removed at the exit screen.
In some cases (1976), a brine flotation in a NaCl solution was applied
to eliminate excessive cob fragments.
Cleaned and floated kernels were blanched for 3.0 minutes on trays in
a continuous steam blancher operating at 99°C. Tray loadings of 8.1 kg/m2
were employed. Blancher effluent was collected during and for 10 to 15
minutes after blanching, and its volume measured and recorded. During 1976
blanched kernels were air cooled in forced air and then frozen on trays in
a cross-flow, air-blast freezer operating at -29°C and at 3.6 mps. During
1977 blanched kernels were either frozen as above or canned. Canning was
conducted the same day at the National Food Processors Association
(National Canners Association)/Tech S Corporation. Canned corn was retorted
in an agitated retort with a 5-rpm cooker speed (equivalent to 235 cans per
minute). The initial temperature of the corn was 38° to 49°C and each lot
13
-------
was processed for 20 minutes at 122°C and water cooled for 10 minutes.
C-enamel, 303 x 406 cans used in this study were filled with 0.34 kg of
corn and covered with a 4%-sucrose 2%-NaCl solution for a minimum 0.64-cm
interior headspace.
MATERIAL BALANCES, SAMPLE ANALYSES, AND PHYSICAL TESTS
The mass of each sublot of corn was measured before husking and after
each step indicated in Table 1. Weighed kernel samples were drawn at the
process steps indicated. A 70-90 gm sample was drawn for total solids
analysis (6) and only in 1976 a 130 to 200 gm sample was drawn for analysis
of kernel characteristics. Material balances were corrected for sample
weights, when appropriate. Liquid samples were drawn from the the cutter
effluent, the belt effluent (1977), the kernel washing effluent, and the
blancher condensate. All liquid and solid samples were stored at -29°C
before analysis. Thawed liquid samples were analyzed in 1976 by the
National Food Processors Association for TOC, COD, TS, SS, and BOD (8).
Liquid samples collected in 1977 were analyzed at the Western Regional
Research Center for TOC. Effluent data are reported below as units of
COD per 100 units of product sweet corn (pph). Conversion to other indices
can be made through use of factors described in Appendix 1.
TABLE 1. PROCESS FOR SWEET-CORN EXPERIMENTS
(1976) (1977) (1976)
Step description Weight Weight Kernel Effluent
recorded recorded sample sample
Harvest
Transport
Composite X X
Husk X
Trim X X
Generate Kernels X X X X
(Four methods)
Clean X X X X
(Two methods)
Steam blanch X X X
Air Cool X X
Freeze X X X
Subsamples of 70-100 g were drawn from stored bulk lots for analyses
of Kjeldahl nitrogen (9), crude fiber (10), crude fat (11), and for amino
acid composition on a Durrum Model D-500 Amino Acid Analyzer.
14
-------
Belt Friction Measurement
Drag forces were measured by applying a horizontal force to a 2.5-cm
diameter, 100 g, brass weight resting on a horizontal test surface. The
horizontal force was applied and measured with a maximum reading spring
gauge. Force was applied until the weight was set in motion. Data are
reported as static coefficient of friction, n , which equals the horizontal
force divided by the normal force. s
The surface was tested after cleaning with a detergent and water solu-
tion, rinsing, and drying; after application of a continuous water film on
the surface; and following application of a continuous film of freshly pre-
pared corn juice. In addition, the corn juice film was tested repeatedly as
it dried and finally when it had formed a hard, glassy crust. Each measure-
ment was repeated 10-20 times.
Measurement of Optimum Position of Ear Relative to Successful Kernel
Removal.
The belt device shown in Figures 8 and 9 was duplicated in a 2.5-cm
wide version and provided with a variable height table and a variable
position guide so that the split ears of corn presented to the working
wheel could be accurately positioned. For this experiment, each split
section was positioned as indicated in Figure 10 so that the point of con-
tact of the friction element at its greatest radiu~(AC) was identical
with the intersection of the interkernel divider (BC) and the arc based
on the radius from the ear center to the kernel tip (BE). For design
purposes the ear was assumed at 20 rows, and the ear radius (BE) at 2.54-cm.
Up to 12 ears were tested at each of several point-of-contact angles from
27° to 63°.
c
B
A
Figure 10.
Cross section illustrating corn split section position with
respect to frictional removal element. Shown are center of
rotation of removal element (A), initial point of contact (C),
center of ear section (B), direction of rotation (D), and
point on kernel surface (E) farthest from the center (B).
15
-------
Recovery on a per ear basis was measured by collecting and weighing
kernels obtained at each position. Damage was measured by mixing 100 g
of kernels produced at each position with 500 ml of water and stirring for
2.0 minutes. This mixture was separated on a 20 mesh screen, and the
liquid bottled and stored at O°C for subsequent thawing and TOC analysis.
A control sample of 9 ears was manually presented to the wheel so as to
obtain complete kernel removal with minimal damage, and these kernels were
weighed and extracted as above. Results are reported relative to the manual
experiment. Both Stylepak (20-22 rows) and Golden Jubilee (18-20 rows)
were employed in the testing.
Sensory Evaluations
Frozen samples were prepared by cooking for 2 minutes in boiling water
and draining. Canned samples were heated to boiling in their own liquid and
drained. The drained corn was kept warm in a bottom-heated, covered dish
during each of the panel sessions. Thirty to 40 g of each corn sample were
placed in coded 1 oz paper cups and served in pairs to the panel. Relative
acceptability of the freshly prepared and processed corn was estimated by
hedonically rating the paired samples on a 9 to 1 like - dislike scale (12).
Relative preference for cut or intact kernels was determined by paired
preference test. The hedonic results were analyzed by the paired t-test
and significance of the preference data was determined from tables (16).
The panelists for the hedonic and preference tests were a random mixture
of laboratory personnel about equally divided between male and female, and
ranging in age from 24 to 59 years.
All sensory tests were conducted in
maintained at a constant 24il°C and 50%
subdued lighting (7.5 watt green bulbs)
between the cut and intact kernels.
a room containing individual booths
RH. Each booth was equipped with
to minimize appearance differences
Measurement of Kernel Resistance to Abscission
Sweet corn ears were sampled from freshly harvested lots of corn or
from corn which had been refrigerated at 1°C for no more than two days.
Ears were husked, reselected for straight rows, and split longitudinally as
described above.
During 1976, split sections were tested directly or subjected to a
pre-test treatment. During 1977, split sections were tested after they
were stripped of kernel rows except for a single row of kernels along the
split-ear vertex with respect to its flat surface.
16
-------
1976 Testing This test, indicated as 45 x 3, is illustrated in
Figure 11. Each split ear tested was pressed manually against a probe
so that three kernels were removed simultaneously and so that the angle
between the applied stress and the kernel axis was 45°. The probe was
attached firmly to a force transducer (Daytronic 152A-10T). The signal
from the transducer was conditioned (Daytronic differential transformer
#61) and automatically rejorded. The probe surface was cushioned with a
5-mm layer of a 0.15-g/cm , closed cell, polyurethane foam.
Figure 11.
Cross section of a 45 x 3 kernel displacement
test showing transducer probe, P.
The maximum force during kernel detachment was recorded and a mean
computed for each ear. This mean did not include data from the row
adjacent to the split surface since the force computed for this row was
lower than for rows not adjacent to this surface.
Thermal treatments were applied to ear split sections. The detach-
ment force then was measured and compared to the detachment force measured
on the corresponding untreated half. Thermal treatments included exposure
to steam and full immersion in boiling water.
1977 Testing This test, indicated as 90 x 1 or 90 x 2 was applied
to a single row of kernels along the ear vertex as shown in Figure 12.
Split sections were mounted in a holder to increase rigidity of the piece
and the holder was attached firmly to the force transducer platform.
The platform was designed so that the ear flat surface was parallel to
and its long axis was perpendicular to the direction of measurement. The
transducer mounting platform was adjusted to position kernel(s) in front
of a slowly moving (77.3 cm/min) displacement hammer so that the face
of the hammer contacted the entire side of each kernel at and above the
glume line and displaced each kernel in a direction parallel to and towards
the transducer and perpendicular to the axes of ear and kernels. Further-
more, displacement hammer, kerne~ and transducer were in alignment.
17
-------
'-...p
~A
\y
J-.
z
Figure 12.
Cross section of a 90 x 1 or 90 x 2 kernel displacement
test showing kernels (K), holder (H), transducer platform
(p), transducer (T), and displacement hammer (DH).
The row on one split half of each ear was tested one kernel at
a time (90 x 1) and the row on the corresponding split was tested two
kernels at a time (90 x 2).
The maximum force during displacement of kernel(s) was recorded and
separate ear averages were computed for the single kernel displacement
and for the double kernel displacement. The ear averages excluded the
four kernels at the tip and at the butt to minimize end effects due to
kernel immaturity (tip and butt) and kernel distortion (butt end).
A force, FA' related to the butt or kernel abscission resistance to
detachment and a force, FF' related to frictional resistance between the
displaced kernel(s) and the adjacent undetached kernel were computed from
FA = F90 x 2 -F90 x 1
FF = 2F90 x 1 -F90 x 2
F90 x 1 is the average force for single kernel displacement (one abscission
and one friction component), and F90 x 2 is the average force for a two-
kernel displacement (two abscission and one frictional components).
Kernels separated during the 45 x 3 test or during the preparation
for 90 x 1 or 90 x 2 testing were collected, frozen to O°C, and later
thawed for measurement of moisture content and kernel dimensions.
18-
-------
SECTION 5
RESULTS AND DISCUSSION
The reasons for the substitution of intact kernels for cut kernels
were suggested in earlier work (5). This work elaborates on that initial
effort to determine the desirability of the substitution by using raw
materials and processing methods comparable to those currently in use
or which could be put in use in the processing industry and by using
these materials and processes on a scale which can more accurately model
a full-scale operation. Data and results for each of several methods
give waste-producing characteristics, yield characteristics and the
quality of the kernel mixture. At the same time, this work initiates
efforts to define the physical mechanisms by which intact kernels may be
produced.
WASTE GENERATION CHARACTERISTICS
Estimation of the Waste Reduction Potential for Intact Kernels
In order to estimate the waste reduction potential, wasteloads for
cut kernels were compared to estimates of practical and ideal minimum
waste loads for intact kernels. The estimate of the practical minimum
waste for intact kernels was made by correlating measured wasteloads from
washing or blanching with the percentage of washed or blanched kernels
which had cut surfaces or broken pericarp and then extrapolating to 0%
cut or broken kernels. The value of this intercept was taken as the
practical minimum. The wash wasteload versus broken or cut kernel corre-
lation is shown in Figure 13 and the value of the practical minimum was
0.37-pph COD. The practical minimum for blanching was 0.067-pph COD.
An ideal minimum wasteload, which corresponds to the waste strength of
carefully detached kernels, and a second practical minimum were estimated
in a separate experiment. In this experiment, the wash wasteload from
kernels that were mechanically detached from ear splits by the smooth-
surface process was compared to the wash wasteload from kernels that were
manually detached from the corresponding ear halves. A practical wash
wasteload of O.31-pph COD was obtained for the mechanically detached kernels
and compared favorably to the value of O.37-pph COD above. An ideal mini-
mum wash wasteload of O.13-pph COD was obtained for the manually detached
kernels.
19
-------
6
5
o
:I:
to-
e>
z
~~4
:;;~
Q.
~ '" 3
11'1 0
~o
~ 0 2
OU
-~
::I
o
o O/A
o o~
/0
o~:o
o - --
o 20 40 60 80 100
WEIGHT PERCENT OF KERNELS CUT OR BROKEN
B
...
Figure 13.
Liquid waste strength of cut and intact-kernel samples for
washing A and blanching B.
Cut-kernel waste strengths are shown in Figure 14 for washing freshly
prepared kernels. The fitted curve in this figure was interpreted as
constant for yields above 42% since this yield corresponds to the theoreti-
cal maximum yield for cut kernels. Furthermore, higher yields represent
inclusion of inedible "cob by" matter which does not contribute significantly
to the waste. The decline of specific waste loading with increasing yield
for the recovery range up to 42% was expected since the loading is propor-
tional to cut surface area: as the kernel is cut more deeply, the ratio
of cut surface to kernel mass decreases.
An estimate of the waste reduction potential for washing intact kernels
then was obtained from the practical and ideal minima for intact kernels
and from cut-kernel waste strengths determined for approximate industrial
conditions. Estimated "practical" waste reductions range from 85.2% for a
40% recovery to 94% at a 33% recovery. Estimated ideal reductions are 95%
and 98% for the same range. If industrial recoveries of 35% are use~ the
reductions predicted are 91% and 97% for the practical and ideal cases,
respectively-
20
-------
x
:r:
...
C)
z
...
'" x
...
VI-
.....c:
... D.
VI D. 4 X
4( ~
~ D
cC X
-0 3 X
~U
0-
::; X
:;t X
...
a
...
0 3S 40 45 50
YiElD OF CUT KERNElS
(% OF FRESH UNPROCESSED CORN)
Figure 14.
Waste strength for cut-kernel samples.
Waste From Alternative Kernel Producing Methods
The result of effluent measurements for kernels from the five kernel
producing methods are shown in Table 2. Based on these results, the
method generating the least waste was either the textured-surface or
smooth-surface process. Closely following was the hole-saw process; but
its waste load per mass of product is subject to revision to a higher value,
since not all of the kernel mass of frozen product would be suitable for
consumption. This revision also applied to the deep-cut samples. Further-
more, the waste from intact kernels produced by the deep-cut process more
closely resembled that from conventionally cut corn as would be expected
on the basis of the large percentage of cut kernels in the intact-DC
mixture.
Washing sweet corn kernels of any type resulted in a wasteload 10 to
20 times that produced by blanching. This observation reflects the order
in which the unit operations were performed, since the greatest leaching
would be expected to result from the first contact of kernels and water.
The portion of the wash wasteload attributed to the smooth-surface
belt washer is reported in Table 3. This contribution amounts to 14% to
17% of the total from the belt washer and kernel washer.
The wasteloading identified with variety in Table 3 increases in the
order Golden Happiness to Golden Jubilee to Stylepak for both cut and
intact samples. However, the percentage differences for washing intact
kernels of different variety, and especially the percentage differences
associated with the belt washer are larger than for the cut comparisons.
Varietal susceptibility to pericarp rupt~re during frictional kernel
detachment and the relative ease of kernel detachment may be factors
causing these differences.
21
-------
TABLE 2.
WASHER AND BLANCHER LIQUID EFFLUENT STRENGTHS
(Basis: 100 Mass Units of Frozen Product or pph)
Wash Blanch
Gross COD COD
recovery
of fresh
Kern(l kernel s
type a) percent pph SD pph SD
Cut (1976) 38. 3.5 0.9 0.15 0.04
Cut (1977) 34. 4.7 1.0 0.25 0.10
Intact-DC 46. 2. 1 0.4 0.16 0.04
Intact-HS(b) 46. 0.9 0.2 0.08 0.04
Intact-TS 41. 0.6 0.5 0.06 0.04
Intact-SS(c) 44. 0.6 0.2 0.04 0.01
(a) All variety averages
(b) Kernels screened before washing and blanching
(c) Wash waste includes contribution from belt washer,
See Table 3.
TABLE 3. CONTRIBUTIONS TO EFFLUENT STRENGTH
Belt washer effluent Kernel washer effluent
Kernel COD COD
Variety type pph SD pph SD
Golden Jubilee Intact SS 0.10 0.14 0.50 0.02
Stylepak Intact SS 0.13 0.10 0.70 0.10
Golden Happiness Intact SS 0.05 0.01 0.31 0.01
Golden Jubilee Cut 5.29 0.23
Stylepak Cut 5.60 0.18
Golden Happiness Cut 3.92 0.58
22
-------
Washing also was found to be a substantial source of filterable solid
waste. Suspended solids were separated from the kernel mass at the washer
feed screen as oversize solids and from the wash effluent as undersize
solids. As shown in Table 4, the oversize solids from washing cut or intact
kernels and the undersize solids from washing intact kernels were small
percentages of the unscreened corn kernels; however, the undersize solids
from washing cut kernels were a considerable fraction of the total.
TABLE 4.
SUSPENDED SOLIDS COLLECTED DURING WASHING OF CUT
AND INTACT-SS KERNELS(a)
Kernel Oversize Undersize
type solids solids
% of feed % of feed
Cut (1977) 0.7 6.6
Intact (1977) 0.8 0.8
(a)Calculation based on moisture-free solids in feed, oversize,
and undersize product streams.
The undersize solids from washing cut kernels included embryos; pieces
of pericarp; very shallowly cut kernels. which would not be considered
saleable; glumes; cob fragments; insects; and immature cut kernels originat-
ing at the tip of each ear. Undersize solids from washing intact kernels
included pericarp fragments and cob fragments. Oversize solids originating
from cut kernel samples included sections of cob broken from the ear during
cutting and husk leaves. Oversize solids from intact kernel samples
included strips of kernel rows attached to thin strips of cob tissue and,
especially in the case of Golden Happiness, groups of detached kernels which
were adhering to each other.
Waste Loading and Machine Variables (Smooth-Surface Process)
Waste loading, which results largely from kernel damage, was corre-
lated to the relative position of the ear split section and the moving
friction surface. This is shown in Figure 15 where TOC is used to index
kernel damage in kernel samples produced from rigidly supported ears of
sweet corn. At angles less than 30° so few kernels were removed that
effluent measurements were not made. Damage increased as the angle
increased above 30° because of the greater tendency to crush kernels before
detachment and to crush kernels against the cob after detachment.
23
-------
......
o
0-6
::to
.... 0
w::t
:!: ....
w 5
;( :!:
w u....
~Z
-------
Figure 16.
Figure 17.
70
~
- 60
~
U
«
~ 50
Z
Co: 40
o
~
~ 30
«
....
e5 20
~
VI 10
VI
«
~
24 26 28 30 32 34 36 38 40 42 44 46
CUT KERNEL YIELD (%)
Yield increase of intact over cut kernels for freshly prepared
kernels.
70
~
-60
~
U
«
~ 50
Z
.
Co: 40
o
~
~ 30
«
....
e5 20
~
~ 10
«
~
24 26 28 30 32 34 36 38 40 42 44 45
CUT-KERNEL YIELD (%)
Yield increase of intact over cut kernels for washed kernels.
25
-------
70
- ".
!:! 60 ,8
~ " 88\
8 8
8~8
~
u
« 50
~
Z
w 30
VI
«
~ 20
U
Z
- 10
VI
VI
«
~
Figure 18.
24 26 28 30 32 34 36 38 40 42 44 46
CUT KERNel YIELD (%)
Yield increase of intact over cut kernels for frozen kernels.
An experimental estimate of the maximum available edible material on
the cob was made by adding the mass recovered mechanically by either of the
intact frictional processes to the mass recovered by scraping the remaining
kernels from the ears. The result of these experiments is shown in
Figure 19 for Jubilee, Stylepak, and Golden Happiness. Ideal yields of 50%
at 70% moisture or 44% at 73% moisture are indicated by the fitted curve.
:00::
VI
::::I 52
Z
Z
- 50
Z
err:
o
u 48
...
o
~ 46
o
....
!!:! 44
>-
54
x
x
x
)(
x
x
100 73 72 71 70 69
INCREASING MATURITY (% MOISTURE)
Figure 19.
Ideal yield of intact sweet corn.
26
-------
One factor which affects the degree of success in approaching ideal or
complete removal is the variety of the sweet corn tested. As shown in
Table 5, the order of decreasing effectiveness of kernel removal is: Golden
Happiness, Stylepak, and Golden Jubilee. This order corresponds to the
order of decreasing strength of kernel attachment.
Another factor contributing to the success with which kernels are
removed is the position on the cob from which the kernels are detached.
Immature kernels at the tip and large, tightly-packed kernels at the
shank or butt were not easily removed and were frequently broken during
or before detachment. Qualitative observations of the extent of this
effect in each variety agree with the ranking of Table 5. The extent of
the effect was quantified during one test (Table 6) by cutting the
processed splits into approximate length fractions of 1/5 near the tip,
3/5 in the center and 1/5 at the butt. These results show that 79% of the
unrecovered kernels are located in the end fractions.
TABLE 5.
APPROACH TO COMPLETE OR IDEAL KERNEL REMOVAL BY
SMOOTH-SURFACE AND TEXTURED-SURFACE PROCESSES
Ratio of Average
actual ideal
to ideal recovery(a)
recovery(a)
Variety % SD % SD
Golden Happiness 94.5 2.3 48.5 3.9
Golden Jubilee 89.8 1.3 47.2 1.6
Stylepak 92.3 3.3 46.4 2.5
(a)Analysis of variance for actual recovery versus variety data yields F
of 4.76, which is significant at 3% level; while ideal recovery versus
variety yields F of 0.689, which is significant at 48% level.
27
-------
TABLE 6.
EAR LOCATION OF KERNELS NOT DETACHED BY FRICTIONAL TECHNIQUE
(STYLEPAK)
Position Approximate Weight fraction Weight fr ac tion
on linear of of undetached
ear fraction cob kernels
Tip 1/5 0.19 0.31
Midsection 3/5 0.52 0.21
Shank or butt 1/5 0.29 0.48
Yield From Alternative Processing Methods
Gross yields, which are yields which do not account for the quality of
the kernel mixture produced, were measured at each step in the process
sequence and are reported in Table 7. Data are shown for cut kernels and
for the three alternative methods tested during 1976. These tests compare
data for tests in which all four processes were applied to a given harvest
of corn. Every alternative process for intact kernels produced more gross
yield at each step than the conventional cut process. Furthermore, the
alternative processes may be ranked in order of increasing gross yield as
textured surface, deep cut, and hole saw. These data also illustrate the
greater resistance to losses by the intact-kernel processes.
These gross-yield data, however, are subject to interpretation since
the kernel mass produced by each method contains various amounts of
"defects" which detract from the uniformity and desireability of the
final product. An indication of these differences is shown in Tables 8
and 9 for freshly prepared and frozen kernels.
Clearly, the textured-surface process developed a kernel mass having
the most uniform character with the inclusion of the least amount of
unobjectional matter (kernels with attached cob, smashed kernels, and cob
fragments).
Moreover, if it is assumed that a clean separation can be made of the
desired intact, or intact plus cut kernels, from the undesired extraneous
matter; then potential, ideal, or net yield estimates can be made by
correcting the gross yields in Table 7 by the kernel analyses summarized
in Table 9. This correction changes the order of increasing yield to
hole saw, deep cut, and textured surface.
The yield of intact-TS and intact-SS kernels is shown in Table 10.
The freshly produced kernel yields for each process are approximately equal,
even though the weight of trimmed ears was somewhat greater for the smooth-
surface process. This difference was due principally to differences in
28
-------
husking procedure. Ears supplied to the textured-surface process were
mechanically husked and trimmed at the butt end and later trimmed at the
tip end. Ears for the smooth-surface process were manually husked and
trimmed. However, the difference in trimmed weights probably does not
affect the kernel yield since all of the increase in trimmed weight for the
textured-surface process may be accounted for by inclusion of tip and butt
ear mass from which kernels were not easily recoverable.
TABLE 7.
MASS RECOVERy(a) OF SWEET CORN AFTER EACH PROCESS STEP
Process
step
Cut
%
SD
Intact-DC
%
SD
Intact-HS
%
SD
Intact-TS
%
SD
Husked ears
Trimmed ears
Detached kernels
Washed kernels
Blanched kernels
Cooled kernels
Frozen kernels
(Gross)
Frozen kernels
(Ne t)
65.3 (2.6)
63.3 (2.1)
38. 1 (3. 1)
36.3 (3.8)
35. 1 (3.8)
33.1 (3.6)
32.0 (3.8)
31.0 (3.8)
66.5 (2.3)
63.8 (2.0)
46.3 (3.0)
47.0 (3.5)
44.2 (2.9)
41.6 (2.9)
39.6 (2.8)
34.9 (4.1)
66.2 (2.1)
63.7 (2.5)
46.4 (3.3)
49.9 (4.6)
46.8 (5.3)
44.0 (5.3)
42.8 (5.2)
34.0 (4.0)
66.3 (0.7)
63.5 (0.6)
40.9 (2.3)
43.2 (2.6)
42.8 (2.5)
40.4 (2.1)
38.8 (1.9)
37.7 (2.4)
(a)Basis 100 units of corn in husk at 71.8% moisture SD 1.3%.
TABLE 8. MEAN KERNEL WEIGHT DISTRIBUTIONS FOR FRESHLY PREPARED SAMPLES
Kernel Cut In tac t Intact kernels Smashed Cob
mixture kernel s kernel s with attached cob kernels fragments
% % % % %
Cut 80 12 0 3 6
Intact-DC 41 37 13(b) 4 4
Intact-HS(a) 22 52 23 (c) 1 2
Intact-TS 0 95 3 (d) 1 1
(a)Screened.
(b)100% of entry due to attached cupule.
(c)90% of entry due to attached cupule, remainder of entry due to
attached rachilla (16).
(d)54% of entry due to attached cupule, remainder to attached rachilla.
29
-------
TABLE 9.
MEAN FROZEN(a) KERNEL WEIGHT DISTRIBUTIONS
Kernel Cut Intact Intact kernels Smashed Cob
mixture kernels kernel s with attached cob kernel s fragments
% % % % %
Cut 86) 12 0 2 3
Intact-DC 38 41 15 2 4
Intact-HS 25 51 21 1 2
Intact-TS 1 95 2 1 1
(a)Actually measured on samples immediately after washing. No changes
in distribution occurred between working and freezing steps.
TABLE 10.
INTACT KERNEL YIELD COMPARISON FOR TEXTURED AND SMOOTH PROCESSES
Process step Textured surface yield Smooth surface yield
% SD % SD
Harve st 100 1 100 2
Husk & Trim 63.8(a) 2. 1 68.3(b) 2.6
Kernels 43.0 4.1 44.1 1.8
(a)Mechanical butting husker.
(b)Manual husk and light trim.
Yield Adjustment by Defect Removal (Density Separation)
One method for achieving the separation of desireable from undesire-
able intact kernels is by exploiting density differences. Summary data
for density flotation in NaCl solutions are shown in Figure 20. The
absolute position of these data along the abscissa was set by the results
of flotation applied to kernels produced in the yield/effluent experiments
30
-------
VI
... 100
...
Z 90
ao=
~ 80
Q 70
~ 60
c(
o 50
~ 40
Z 30
~ 20
ffi 10
A.
Figure 20.
4 6 8 10 12 16
INCREASING DENSITY (% NaCI)
Density flotation of sweet corn. Types are intact kernels (A),
intact kernels with attached rachilla (B), intact kernels with
attached cupule (C), cob fragments (D), and smashed kernels (E).
o
described above. The relative position of each curve in a family of
curves, as shown in the figure, was unchanged by variety, maturity, and
temperature. The curves describe a density decrease from the cut form to
the intact form, to intact forms with adhering rachilla, to intact forms
with rachilla and cupule. The absolute kernel density, hence, the positions
of the curve family, was found in preliminary tests to be sensitive to
kernel temperature, maturity, and variety so that the entire family of
curves would shift along the abscissa for each different condition.
The kernel density differences were not great, therefore, the removal
of one component of the mixture will necessarily be associated with partial
removal of other components. The predicted approximate effect of a density
flotation separation applied to hole-saw mixture to reduce kernels with
cupule and rachilla adhering is summarized in Table 11. Using this
table, the quality of a mixture equivalent to that produced initially by
the textured-surface process could be achieved by flotation of the hole-
saw kernels, but there would be an associated loss of more than 20% of
the desired kernels.
TABLE 11.
PREDICTED LOSSES OF INTACT KERNELS FROM INTACT-HS KERNEL
MIXTURES BY FLOTATION IN NaCl SOLUTIONS
Solution concentration
weight percent
Desired kernels with
rachilla or rachilla and
glume attached
we ight percent
Associated loss of
intact kernels
weight percent
10
10.5
11. 0
4.7
3.1
1.7
16
22
44
31
-------
This flotation loss would reduce the net yield from the hole-saw proc-
ess, and it would result in a larger value of waste per hundred pounds of
useable kernels produced since the effluent loading reported earlier for
hole-saw kernels was based on gross yields. The result of lower yield and
increased waste reduces the attractiveness of this technique relative to
the friction techniques.
Yield and Machine Variables
Kernel Position Relative to Friction Surface--
Yield of intact kernels was affected by the relative position of the
kernel and the friction developing surface during the kernel detachment
event. Yield, which is expressed here as the ratio of the mass of kernels
removed to the mass of kernels available in a given row, is shown in Figure
21 for various positions of the ear and detachment surface. This figure
indicated the ineffectiveness of small and large angles and suggested an
optimum angle between 36% and 54%. Since the waste increases with increas-
ing contact angle, an angle at the lower end of this range would be
desirable.
Different types of surfaces likely will have different levels of
sensitivity to position so that the conclusions reached in this section
apply only to the smooth-surface process.
~
~ 100
Z
- - - - - - - - _JL - - - -
~
....
z
ffi 80
"
x
w
...
'"
«
~ 60
>
o
u
....
'"
x
>-
'"
....
>
o 20
~
'"
...
o 40
~
...
....
z
'"
~ 0
18
27
36 "S
54
63
CONTACT ANGLE (II).
Figure 21.
The role of kernel position relative to friction surface
in successful kernel removal. Refer to Figure 10.
32
-------
Quality of Friction Surface--
The success of the techniques for kernel detachment which rely
on friction between the detaching element and the kernel surface depends
on maintaining a high coefficient of friction between the contact surfaces.
Friction is reduced if the juice of ruptured kernels coats either surface.
The magnitude of this effect in static testing is shown in Table
12 for the neoprene belt of the smooth-surface process. The clean and
dry surface provides the greatest friction, but is closely approximated
by a surface coated with a film of partially dried juice. In auxiliary
tests, mass production rates for partially dry surfaces and for clean, dry
surfaces were indistinguishable. The presence of either wet corn juice or
fully dried juice reduced the maximum friction by 88%.
TABLE 12.
STATIC COEFFICIENT OF FRICTION (~ ) FOR CLEAN AND CONTAMINATED
s
NEOPRENE SURF ACES
Surface Condition ~s SD
Clean and dry 3.3 0.4
Wa ter film 1.3 0.1
Fresh corn-juice film 0.4 O. 1
Partially dry corn-juice film 2.6 0.7
Fully dry corn- juice film 0.4 O. 1
The textured surface also benefitted from the drying of the surface.
When the surface was kept dry, flexure of the studded surface during kernel
detachment fractured the embrittled juice film and broke it away from the
surface. As an example of the effect, a 55% kernel recovery (husked ear
basis) was achieved using a wet, unheated textured surface. However, this
recovery was increased to 72% (husked ear basis) when the ear sections
were reprocessed on a dry, heated textured surface. It was not necessary
to continuously wash and dry the textured surface, since drying, the associ-
ated embrittlement of the juice film, and the flexure of the surface
resulted in the removal of this slippery, dry film.
33
---'"
-------
'I'I1II
KERNEL QUALITY CHARACTERISTICS
The substitution of intact kernels for cut kernels is highly desireable
from the standpoint of waste reduction and yield improvement. However, the
desireability of this substitution also depends on acceptable quality
attributes such as visual appearance, flavor, and texture. Intact kernels
are visually distinguishable from cut kernels. Furthermore, since the
intact kernel includes the kernel germ, whereas the cut kernel excludes
all or part of the germ, sensory response to each kernel type should also
be different.
Visual Evaluation of Sweet Corn Samples
Visual differences between samples of cut and intact kernels are
usually clearly evident. This can be seen in Figure 22 which illustrates
samples of corn freshly prepared by cutting and by the smooth-surface
process. Subsamples of frozen-and-thawed, cut and intact sweet corn corre-
sponding to the samples utilized in the 1976 sensory evaluation described
below are shown in Figure 23, 24, and 25. Corn in each cut/intact compari-
son came from the same planting, harvesting, and processing day.
~~ ...,
~
,
.........,
~
~..~
..
,-\'J
"'
~
..
~
't
, .....
'f/II'
Jk;/IJ.~
,.-4 ~ r
, . ...~ ..
~ ~~-J~'
'\ ~
~~..~
~~ -~
~"
)'
..
-
~
,
w
,
~
~
JIII&
y
.iIiL
~
Figure 22.
Freshly prepared, unwashed samples of cut and intact kernels
34
-------
.. ~ -7T..-11
~ 1 I . ~.I
.~~ ....~
I " , I j
~1..~~ .
---. '" ~.
..., .. ~..
.~
......
~\
~
,
...
Figur(~ 23.
Frozen and thawed samples of cut and intact sweet corn
(Var. Stylepak)
}~,~
.'-'
Figure 24.
Frozen and thawed samples of cut and intact sweet corn
(Var. Golden Happiness)
35
-------
-"II
Figure 25.
Frozen and thawed samples of cut and intact sweet corn
(Var. Golden Jubilee).
No formal attempt was made to compare the visual acceptability of the
two kernel types. This kind of evaluation would not have great meaning
since it depends strongly on the past experience of the observer. However,
visual differences were quantified within the group of kernels identi-
fied as intact kernels. A classification of kernel structural details was
made to assess varietal differences and the effects of washing on altering
these differences. The results are shown in Table 13 and Table 14 for
intact kernels only. Identified here were glume "wings" or glume tissues
which surround the lower third of the kernel (these are absent in Golden
Happiness, a glume-free or vestigial glume variety); base glumes, or very
short glumes which surround the base of individual kernels; and rachilla
tissue, the woody conducting tissue which connects to the kernel at the
abscission zone (17). Of these tissues, the glumes affect the visual
appearance somewhat, whereas the rachilla affects both the visual appearance
and the texture of corn. In the opinion of the authors, the glume-free,
rachilla-free kernel was the most attractive and the kernel with rachilla
tissue was the least attractive. Washing tends to improve the kernel
appearance by increasing the proportion of the sample described as glume
free.
36
-------
TABLE 13.
MEAN INCIDENCE OF GLUME TISSUES IN FRESHLY PREPARED
INTACT-TS KERNEL SAMPLES
Kernel Characteristics(a)
Glume- Gl ume "wing s" Gl um es Rac hill a
Variety free at base attached
% SD % SD % SD % SD
Golden Jubilee 3 0.5 7 7 88 3 2 0.5
Stylepak 16 10 1 2 82 8 1 1
Golden Happiness 69 15 0 0 30 14 1 1
(a)Mass percentage based on intact kernels only.
TABLE 14.
ABSOLUTE CHANGE IN DISTRIBUTION FROM WASHING INTACT KERNELS
Variety
Gl ume-
free
Rachilla
Glume "wings" Glumes at base attached
(Absolute change in mass percentage units)
Golden Jubilee
Stylepak
Golden Happiness
+3
+13
+5
-3
-1
o
o
-12
-6
o
o
+1
Visual appearance differences related to the presence or absence of
glumes can be seen in Figure 23 through 25. Also Figure 26 illustrates
differences in individual kernels and cobs for the principal varieties
studied. In this figure the Stylepak sample indicates a kernel with glume
wings (7 o'clock position), the Golden Happiness with neither side nor
base glumes, and the remaining kernels show glumes at the base.
37
-------
..
Figure 26.
Corn cross sections and individual kernels of (1 to r)
Golden Happiness. Golden Jubilee and Stylepak.
In the grading of sweet corn for commercial "cut" or "whole kernel"
packs. one factor which reduces the overall score for the product is the
presence of pulled kernels. Pulled kernels lack a cut surface (i.e.
an intact kernel) and their presence detracts from the sample uniformity.
Reference to the corrected official interpretive guide of the USDA. Produc-
tion and Marketing Administration and Inspection Division (Figure 27)
indicated that not all intact kernels are classifiable as pulled kernels
and that most kernels produced by the intact processes would not receive
this designation if graded according to the cut kernel standards. By and
large. however. these standards are probably inappropriate for intact
kernel samples.
38
-------
I ~w " ~ -
~ ~ ~ 'II'
1.
2.
3.
4.
5.
6.
Figure 27.
Adaptation of USDA interpretive guide illustrating pulled
kernels and cut kernels in canned and frozen corn. Kernels
captioned 1 and 2 are scorable as pulled kernels; whereas,
kernels 3 to 6 are not scorable as pulled kernels. When the
adhering cob material is very hard and affects the appearance
and eating quality, it is considered as a pulled kernel (13).
Sensory Evaluation of Sweet Corn Samples
The result of sensory evaluation of sweet corn frozen in 1977 is
shown in Table 15. In each case higher hedonic scores were obtained for
intact samples than for the control cut comparisons (a value of 7 indi-
cates the taste tester liked the sample very much; whereas, 9 indicates like
extremely, 5 neither like nor dislike, and 1 dislike extremely). Statistical
evaluation by the T-test supports the significance of these differences.
Data for hedonic ratings applied to frozen and canned samples
prepared during 1977 are shown in Table 16. Highest scores were obtained
for frozen intact corn. Moreover, the differences between intact and cut
kernel samples were greater in 1977 than in 1976 and tested with greater
significance. The shallower depth of cut in 1977 and the consequent exclu-
sion of the embryo from the cut sample may account for this result. Intact
ratings were also higher for canned comparisons but the differences between
canned-cut and canned-intact corn were smaller and tested with less signifi-
cance. The sal t and sugar used in the canning "brine" probably contributed
to the loss of a sensory difference.
39
-------
TABLE 15.
MEAN HEDONIC RATINGS FOR CUT (NORMAL) AND INTACT-SS SWEET
CORN FROZEN IN 1976
Probabil i ty
Kernel type of significance
as applied
Cut Intact to means
% Number Kernel
of moisture
Sample mean mean t-test judges %
Golden
Jubilee 6.6 7. 1 0.05 43 70.5
Stylepak 6.4 7.1 0.035 36 71.5
Golden
Happiness 6.5 7.3 0.02 40 73.2
TABLE 16.
MEAN HEDONIC RATINGS FOR FROZEN AND CANNED, CUT AND INTACT-SS
SWEET CORN PREPARED DURING 1977
Preservation Mean Ratings Kernel
Variety method cut intact Probabil ity T DF moisture
%
Golden Frozen 6.2 7.4 0.003 4.465 27 70.5
Jubilee Canned 5.9 6.4 0.1221 1.596 27
Stylepak Frozen 6.2 7.5 0.001 4.465 27 72.5
Canned 6.2 6.6 0.314 1 . 02 7 28
Golden Frozen 6.0 7.6 0.00001 7.864 28 73.0
Happiness Canned 6.1 6.8 0.017 2.536 27
Subsamples of the same 1977 lots, which were ranked hedonically above,
were also presented to panels in a paired preference test. The results of
this comparison are shown in Table 17; and, as expected, intact samples
received the preference. Moreover. stronger preference was expressed for
frozen than for canned samples. The most common reasons expressed by the
panelists for preference were flavor and texture.
40
-------
TAB LE 1 7 .
PAIRED PREFERENCE RATINGS FOR INTACT AND CUT SWEET CORN
Preserva- Percent preference N Probability Reason for
Variety tion method for intact preference
Golden Frozen 89 28 0.000027 Flavor and Texture
Jubilee Canned 64 28 0.184 " " "
Stylepak Frozen 86 29 0 . 0001 " " "
Canned 71 28 0.0356 " " "
Golden Frozen 86 28 0.00018 " " "
Happiness Canned 86 28 0.00018 " " "
Fiber, nitrogen, and lipid characteristics of intact and cut sweet
corn kernels are reported in Table 18. No differences in Kjeldahl
nitrogen were detected even though differences would be expected since
inclusion of the embryo should increase the kernel protein content. Crude
fiber differences between intact-SS and intact-TS kernels were small.
These fiber values also reflect the order of the increasing proportion
of kernels with attached cobby matter (Table 11).
TABLE 18.
FIBER AND TOTAL NITROGEN IN CUT AND INTACT SWEET CORN
Kjeldahl Crude Crude
Kernel style nitrogen fiber fat
we ig ht weight weight
% SD % SD % SD
Intact-TS 0.54 0.02 0.68 0.06
Intact-SS 0.53 0.01 0.74 0.18 0.98 .27
Intact-HS 0.54 0.03 0.87 0.19
Intact-DC 0.54 0.04 0.75 0.08
Cut (1976) 0.54 0.03 0.62 0.06
Cut (1977) 0.54 0.02 0.62 0.12 0.46 . 12
41
-------
The result of a comparison of the protein, which was indexed by the
total amino acid content, of intact to cut kernels is shown in Table 19.
The expected increase in protein for intact kernels is realized. Since the
data compare amino acid contents of samples of equal mass and since there
are fewer kernels per unit of mass in the intact sample, larger differences
would have been indicated if the comparison had been made on a per kernel
basis.
TABLE 19.
TOTAL AMINO ACID CONTENT INCREASE FOR INTACT SWEET
CORN RELATIVE TO CUT(a)
Percent increase in
amino acid for
intact kernel
Percent increase
in yield for
intact kernel
Variety
Golden Jubilee
Stylepak
Golden Happiness
15.9
15.6
5.5
27.6
28.1
34.8
(a)Based on equal masses of frozen samples from 1977 tests
KERNEL RESISTANCE TO DETACHMENT
Testing During 1976
The 45 x 3 test and the analysis of the resulting data assumed that
kernel detachment was largely resisted by the tissues making a physiologi-
cally functional connection at the base. Values of the maximum force
resisting kernel detachment calculated from 45 x 3 measurements are shown
in Table 20. Differences between these values are not large with the
exception of that between Vanguard and the others. Although qualitative
observations agree with the position of Vanguard in this grouping, they
do not agree with the position of Golden Happiness. Kernels of Golden
Happiness were markedly easier to remove than the varieties with which it
is closely grouped by this test.
The application of this test to individual kernel groups was subject
to two types of error. One error was kernel rupture before and during
detachment. The frequency of occurrence of rupture is indicated in
Figure 28. Kernel sensitivity to rupture was so great that at kernel
moistures higher than 76% the test could not be performed. Furthermore,
rupture ended measurements before detachment and the measured force
therefore underestimated the actual detachment force. Forces recorded
during rupture are not included in the average values reported here.
42
-------
TABLE 20.
MAXIMUM TOTAL FORCE PER KERNEL MEASURED DURING KERNEL
DISPLACEMENT
Rows Samples Moisture Force(a,b)
Variety per ear
% SD N SD
Golden Happiness 18 10 72.3 2.7 2.6 0.3
Golden Jubilee 18 65 71. 3 4.0 2.9 0.4
Stylepak 20 40 73.2 1.8 2.3 0.3
Vanguard 18 14 70.7 1.3 3.0 1.4
(a)Reported force is value measured by 45 x 3 test divided by 3.
(b) Analysis of variance of force versus variety yields F of 8.66 which is
significant at 0.003% level.
V!
....
V!
W
""w
....0:::
Z ~ 16
wQ..
~::>
e;lo:::
< Z 12
-A-
~C>
cZ
~j:: 8
0::)
....~
~ 0::: 4
U
0:::
w
Q..
A
20
B
o / I I I
100/80 78 76 72 70 68 66
INCREASING MATURITY (% M
-------
A second error source was the inability to achieve simultaneous
detachment of all three tested kernels. For instance, if the displace-
ment occurred sequentially; then, the maximum recorded force would be less
than would have been recorded by simultaneous displacement.
This test was useful for interpreting the effect of heat treatment
on decreasing the kernel attachment strength. For instance, early work by
these authors (5) using the 45 x 3 test applied to fresh market corn of
undetermined variety had shown that heat would substantially reduce the
force required for kernel removal. This effect, in itself, would be desir-
able for ease of kernel removal and would also reduce the liquid waste
during subsequent washing of the intact kernels by "setting" the starch (5).
This abscission altering effect was applied in tests using frictional
kernel removal described above as the textured-surface process and using
freshly harvested corn. However, yields shown in Table 21 for removal of
kernels from heat treated ears were lower than for untreated corn.
TABLE 21.
SWEET CORN YIELD CHANGES DUE TO HEAT (5 min, 100°C STEAM)
PRECONDITIONING OF EARS TO EFFECT KERNEL LOOSENING
Variety
Yield change
%
Condition of friction surface
SD
1.5
we t( a)
wet(a)
dry< b)
dry(b)
Golden Jubilee
Stylepak
Golden Jubilee
Stylepak
-11.2
-31.7
-8.1
-11.6
3.7
(a)Friction surface was not heated and became wet with corn juices.
(b)Friction surface was heated to maintain dry condition.
As a consequence of these unexpected results, new 45 x 3 tests were
applied to freshly harvested Golden Jubilee and Stylepak. The results
for this 45 x 3 testing are shown in Table 22 and indicate that in spite
of average reductions in the force resisting removal, no change and
occasional substantial increases in the detachment force were found. This
unexpected effect occurred twice as frequently in Stylepak as in Golden
Jubilee. The inconsistency of this effect is consistent with the results
of Table 21 and precludes the use of the heating method to alter the
kernel attachment strength of these varieties.
44
-------
TABLE 22.
THERMAL PROCESSING TO ALTER STRENGTH OF KERNEL ATTACHMENT(a)
Mean Range
Steam reduction of reduction Measuremeft~ with
Variety exposure of detachment detachment no change b or
time force force increase in D.F.
min % SD minimum maximum %
% %
Stylepak 1 13 6 4 16
2 10 6 2 16
4 8 6 -16 16 38
8 3 3 -14 19
12 15 15 -7 28
Golden 1 5 9 -6 15
Jubilee 2 13 4 8 19
4 11 17 -18 25 17
8 22 13 4 40
12 22 9 8 34
(a)45 x 3 Test.
(b)No change is defined here as change less than 5%.
Testing during 1977.
New qualitative observations during 1978 further challenged the
validity of the base-only abscission resistance assumption of the 45 x 3
test. Two observations were made in the course of attempts to identify
factors affecting the coefficient oi friction between kernels and the
kernel-removing friction surface, and each indicated the importance of
previously neglected interkernel friction in the separation. The first
observation was the discovery of the presence of microridges on the
surfaces of adjacent kernels. These ridges presumably interlock and pre-
vent a smooth, sliding separation. The second observation was the presence
of a thin, waxy coating on the kernel surface. The residue from the benzene
extraction of this coating was a very sticky substance which also prevented
a smooth, sliding separation. Consequently, the 45 x 3 test actually
measured three contributions from kernel bases and one from a side, and
the ranking and results of Table 20 were held suspect.
The 90 x 1 and the 90 x 2 tests were adopted in order to identify
the relative importance of abscission resistance at the kernel base and
of the interkernel frictional or side resistance between kernels in the
same row. The respective components were calculated from these tests.
45
-------
Values of both the abscission component and the interkernel friction
component shown in Table 23 indicate wide differences in each component.
Abscission values differing by nearly four-fold and interkernel friction
values differing by three-fold were observed.
Abscission values obtained here reflect the qualitative ease of kernel
removal indicated earlier as well as qualitative observations related to
Illini Xtra Sweet whose kernels are extremely difficult to detach.
Interkernel frictional resistance was so large in the Golden Happiness
samples that for a substantial number of individual tests, kernels adjacent
to the tested kernel(s) were also removed. This effect is n\werically tabu-
lated in Table 23 as adherence. In some instances, four or five additional
kernels were detached during the intentional detachment of one (or two)
kernel(s). Adherence detachment of entire kernel rows of this variety
could be effected by careful manual detachment of a single kernel if the
ear maturity approached that corresponding to 65% moisture. The high
values of interkernel friction in Golden Happiness may be due, in part, to
the absence of glume tissues since the absence of these tissues would
increase the area of contact between growth ridges and between wax-covered
surfaces.
The presence of strong interkernel attachment can be beneficial to
the mechanical process which will be applied eventually to the high speed
detachment of kernels. Two machine simplifications can be hypothesized
from the genetic increase of this interkernel attachment strength. For
one, alignment between kernel row and frictional separator will be less
critical, since only a few kernels in each row would need to be contacted.
For another, the machine itself will require less surface for contacting
kernels since, in principle, detachment forces would need only to be applied
at the ends or in the center and would need to be no wider than the width
of one or two kernels.
The occurrence of adherence injects an additional source of error into
this measurement method. For instance, during the intentional displacement
of a kernel (kernel 1), the friction with the adjacent kernel (kernel 2)
can reduce the "attachment" between kernel 2 and the next kernel (kernel 3).
Some disruption of the abscission strength of kernel 2 can also be expected.
Hence the test underestimates both components.
Kernel rupture during testing was more frequent (Figure 28) in the
90 x 1 or 90 x 2 testing than in the 45 x 3 testing. Fewer ruptures were
observed during manual testing (45 x 3) since the operator can often sense
or anticipate incipient rupture based on the visual appearance of the ear
and can therefore reduce the rate of displacement so that rupture is less
1 ikely.
46
-------
TABLE 23. MAXIMUM COMPONENT FORCES (NEWTONS (N» MEASURED DURING KERNEL DISPLACEMENT
Interkernel
NlDllber NlDllber Abscissi~n) fric tio() Frequency
Variety of of Moisture force c force d of
rows samples adhesion(b)
% SD N SD N SD
Golden Happiness 16 12 70.7 3.5 1.3 0.7 4.0 1.5 4.2
Golden Jubilee 18 13 70.2 2.1 3. 1 1.2 2. 1 1.7 0.3
Stylepak 20 22 71. 5 2.4 2.3 0.7 1.5 0.8 0.4
.jO- Vanguard 18 16 73.5 4.7 3.5 1.2 1.3 1.3 0.0
--.J
Illini Xtra Sweet 16(a) 8 72.3 3.3 5.5 2.3 2.3 2.0 0.8
(a)Average result for 3 samples of 18, 3 of 16 and 2 of 14.
(b)Frequency of adhesion is the nlDllber of occurrences of displacement caused by adherence
to a displaced kernel per row of kernel removal. A large number reflects strong
attachment to adjacent kernel and corresponds to a large friction force (or low
abscission force).
(c)Analysis of variance for abscission force versus variety yields F of 16.7 which is
significant at a level not greater than 0.000%.
(d)Analysis of variance for interkernel force versus variety yields an F of 7.9 which is
significant at the 0.003% level.
-------
Effect of Maturity
Changes in the component forces due to increasing maturity were not
detected in these data. Correlation of each component with percentage
moisture and with solids accumulated per kernel resulted in all-variety
average values for the coefficient of determination in the range of
O.ltoO.2.
However, maturity was found to playa role in the ease of detachment
when measured by rates of kernel production. As shown in Figure 29, sub-
stantial increases in kernel production rates were obtained within the
range of processing maturity investigated. The data expressed by lines
A (Golden Happiness) and B (Stylepak) connect points representing tests
applied to corn from successive harvests of the same planting of each
variety and have nearly equal slopes of 1.7 and 1.9 respectively. Golden
Jubilee (C) has a slope of 3.9 but compares results from different plantings.
On the basis of the base component data presented earlier these differences
do not appear to be due to changes in the abscission component since the
order of change shown in Figure 29 would be detectable by the test.
These changes with moisture may reflect the role of pericarp strengthen-
ing as the kernel ages.
If we assume that the rate of kernel production measured in Figure 29
is related to the ease of detachment, than the ratio of the rates should be
proportional to the inverse of the strength of the abscission component.
This comparison of rates is made for curves A and B only since the points
of C do not represent common plantings. The result of the rate comparison
of Figure 29 is approximately 1.8 and the result of the inverse force com-
parison is 1.7.
18
/
.!: 16
"-
e"
CI'J ~ 14
Cl'JUJ
«~
:2 ~ 12
...J
UJZ
Z 0 10
ffii=
~g
o 8
o
a:
Q. 6
P
/8. ,ti'C
tf
,
.
o L....;
100
7 J 72 71 70 69
INCREASING MATURITY ('7. MOISTURE)
68
Figure 29.
Rates of removal of sweet corn from Golden Happiness (A),
Golden Jubilee (C), and Sytlepak (B) by friction with a
moving, smooth neoprene surface.
48
-------
Correlation with Kernel Removal Effectiveness
The ease of kernel removal can also be correlated with the successful
removal of kernels during process tests. Data in Table 5 indicate that the
order of increasing success in kernel removal agrees with the order of
increasing ease of detachment; i.e. Golden Jubilee less than Stylepak less
than Golden Happiness.
Theoretical Analysis
Additional insight about kernel removal can be gained by static fail-
ure analysis of the kernel itself. In this analysis it was assumed that
the kernel is rigid. that stress developed in the kernel abscission layer
is proportional to the applied strain. and that the only forces acting on
the kernel are the applied force and reaction forces acting through bio-
logically connected tissues.
The free body diagram for a single kernel is shown in Figure 30.
Here F is the applied force. f is a fraction (0 to 1) multiplied by the
kernel length 1 which describes the distance from kernel base to the point
of application of the force. a is the length of the abscission zone (and
"b" its width); R is the reaction force. and m is the couple providing
x
a moment in reaction to the bending moment of F about the intersection of
x - x and the kernel abscission layer.
L
x
I
I
F
H
Figure 30.
Schematic free-body diagram for displacement of sweet corn
kernels.
49
-------
By analogy to the analysis of cantilever beams (14), we identify tissue
compression parallel to the fibers to the left of x - x and tension of
fibers to the right. Maximum stress occurs at the surface fibers farthest
from x - x and failure occurs by tension on the side of the kernel where
the force is applied.
Analysis of the free body diagram further assumes that a zero stress
occurs at the intersection of x - x and "a" and that the stress increases
linearly to the outermost fibers. This analysis yields a force, FD' which
is applied to initiate failure.
FD =
Q"d
6
. b . a2
f . 1
(1)
Here a is the stress for failure by tension.
applieg force, Fs' for shear failure is
By comparison, the resulting
F
s
= a
s
. b . a
(L)
where as is the stress for shear failure. Estimation of
is made by forming the ratio of FD to Fs. If this ratio
occurs by the mode corresponding to FD. The ratio is
the mode of failure
is <1 then failure
FD
F
s
ad
6.a
s
a
f . 1
(3)
and by substituting as/ad equal to 0.06 (15) as measured for wood (an
analagous conducting tissue) and a/f.1 equal to 0.1, the value of the ratio
FD/Fs is 0.3; hence, failure would be expected to occur sooner by bending
than by shearing. Observations during this testing program confirmed this
analytical result. (The reader should note that this force analysis neglects
the influence of a y-component of force, the effect of which will depend on
the position of the kernel surface at which the detaching force F is applied.
For instance, if F is applied to the right of x - x then the moment M will be
decreased and F required for detachment is increased. If the resultant is
applied to the left of x - x, then M is increased and F required for detach-
ment is decreased.)
Furthermore, on the basis of this analysis we can expect that varieties
of sweet corn with long kernels or small abscission layers will require less
force for detachment than varieties with short kernels and large abscission
layers provided the stress for tensile failure is identical. The varieties
tested did not differ greatly in length measured from crown to point of
abscission. Values of mean kernel length were 1.23 ~ 0.08 cm SD for Golden
Jubilee, 1. 23 + 0.05 cm SD for Stylepak, 1. 28 + 0.12 cm SD for Golden
Happiness, 1.19 + 0.08 cm SD for Vanguard, and-1.22 + 0.04 cm SD for Illini
Xtra sweet. Analysis of variance yielded an F of 2.24 which is significant
at the 7.5% level. The data may be ranked approximately by order of increas-
ing kernel length and decreasing abscission force with only Illini Xtra
Sweet out of order and with Jubilee and Stylepak approximately equal in the
length ranking. The geometry of the kernel abscission zone was not measured.
50
-------
The results of 45 x 3 and 90 x 1 - 90 x 2 tests may be resolved if an
accounting is made of two competing factors which would detract from the
comparison. The first factor relates to the point of application of the
force in each test. Since the 45 x 3 is applied at or near the kernel crown,
f = 0.9 to 1.0, and the 90 x 1 is applied centered at f = 0.67, a correction
factor of 2/3 is introduced in computing 45 x 3 values from 90 x 1 and
90 x 2 measurements. The second factor relates to the angle at which the
strain was applied. Since the x-component of the measured (45 x 3) which
causes displacement is related-fo the total applied force by sin (~/4). a
correction factor of (sin ~/4) or 1.4 is applied. The multiplied correc-
tion factors yield a factor of 0.94 to be applied to the 90 x 1 and 90 x 2
results.
When the calculation of the expected 45 x 3 result is made from measured
90 x 1 and 90 x 2 tests, the values of calculated per kernel forces (Table
24) compares favorably with the measured values with the exception of the
values for Golden Jubilee. At any rate, the calculated force would have been
expected to be higher than the measured force since the simultaneous displace-
ment of three kernels is difficult to achieve. Displacements occuring
sequentially would be expected to reduce the maximum observed values.
TABLE 24.
COMPARISON OF PER-KERNEL FORCE BY ACTUAL 45 x 3 TESTS TO
FORCE BY COMPUTED 45 x 3 TESTS BASED ON COMPONENT ABSCISSION
AND INTERKERNEL FRICTIONAL FORCES IN 90 x 1 OR 90 x 2 TESTS
Variety Measured 45 x 3 Total calc
-------
REFERENCES
1.
National Canners Association 1971. Liquid wastes from canning and
freezing fruits and vegetables. EPA Program Number 12060EDK,
p. 144.
Inglett, G. E. ed. 1970. Corn: Culture, Processing, Products. AVI
Publishing Inc., Westport, Conn.
Huelsen, W. A. 1954. Sweet Corn. Interscience Publishers, Inc.
New York, N.Y.
Lockwood, D. H. and Andres, C. 1977. Modified starch assures
stability, long shelf life, consistency of cream style corn.
Food Proc.: (6) 76-77.
Robertson, G. H., Lazar, M. E., Galinat, W. C., Farkas, D. F. and
Krochta, J. M. 1977. Unit operations for generation of intact or
unit kernels of sweet corn. J. Fd. Sci. 42:(5)1290-1303.
The Association of Official Agricultural Chemists. 1965. Official
Methods of Analysis, 10th ed. The Association, Washington, D.C.
p. 308.
National Canners Association. 1968. Laboratory Manual for Food Canners
and Processors. AVI Publishing Co, Inc., Westport, Connecticut.
Vol. 1, Chapters 8 and 9.
United States Environmental Protection Agency. 1971. Methods for
Chemical Analysis of Water and Wastes. EPA-16020-07/71.
Environmental Protection Agency, Washington, D.C.
The Association of Official Agricultural Chemists. 1975. Official
Methods of Analysis, 12th ed. The Association,
Washington, D.C. p 15. Method 2.049.
The Association of Official Agricultural Chemists. 1975 Ibid. P 137,
Method 7.054.
The Association of Official Agricultural Chemists. 1975 Ibid P 135,
Method 7.044.
Peryam, D. R. and Girardot, N. F. 1952. Advanced taste-test method.
Food Eng. 24: 58.
John, J. 1976. Private communication. United States Department of
Agricultural, Agricultural Market Service, USDA, 111 W. St. John
St, Suite # 416, San Jose, CA 95113.
Seeley, F. B. and Smith, J. O. 1956. Resistance of Materials. 4th ed.
John Wiley, New York, p 459.
Perry, R. H., (Ed.), 1967. Engineering Manual. 2nd ed. McGraw-Hill Book
Company. N.Y. p. 6-60.
Gacula, M. C., Moran, J. M., and Reaume, J. B. 1971. Use of the sign
test in sensory testing. Fd. Prod. Dev. ~(6), 98.
Galinat, W. C. 1979. On the usage of the terms pedicel and rachilla
in description of the cob, the female spikelet and the grain in
maize. Maize News Letter, in press.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
52
-------
APPENDIX 1.
WASTE INDICES
The measures of liquid waste strength applied to the wash and blanch
effluents encountered in this study were biological oxygen demand (BOD),
total organic carbon (TOC), chemical oxygen demand (COD), total solids
(TS), and suspended solids (SS). Since the application of all of these
measures is time consuming and costly, the data were analyzed to establish
correlations between the measures so that one or two simple tests could be
used for reporting purposes and for the analysis of additional experiments.
The result of these correlations is shown in Table AI. Here BOD,
COD, TS, and SS are correlated with TOC. Excellent correlations of BOD,
COD, and TS were obtained. Correlation of SS with TOC was not good but was
improved when the data were grouped according to their point of origin
from the blanching or washing stages. Since suspended solids do not
necessarily represent all of the effluent generated, the SS correlation with
TOC should not be expected to be strong. However, the strong correlation
between COD, BOD, and TOC was used to support the use of TOC as the
preferred measurement method during 1977. Data from 1977 are reported in
the text as COD as calculated using the regression formula.
TABLE AI.
CORRELATION OF WASTE INDICES (I) WITH TOTAL ORGANIC CARBON
MEASUREMENT. (I = A (TOC) + B).
Index Data base Regression Coefficient
year constants of determination
A B
BOD 1976 1.27 -46.9 0.980
COD 1976(a) 2.67 -12.8 0.997
TS 1976 2.24 71.7 0.992
TS 1977 2.52 -73.0 0.995
SS 1976 (b) 0.16 151 0.294
SS 1977 0.32 344 0.565
SS 1977(c) 0.311 -45.5 0.81
1977 (d) 0.704 -137 0.88
(a)Excludes samples with salt residuals
prior to blanch.
(b)See note (a). also excludes blanch.
(c)Blanch only.
(d)Wash only.
>0.5% from brine flotation
53
-------
APPENDIX 2.
PRELIMINARY CANNING STUDIES
Stationary Retort
Heat penetration studies were conducted on intact and cut corn kernels
packed in a 4%-sucrose 2%-NaCl solution in 303x406 tin cans. Data obtained
from the Technical Service Corporation, which provides this service regu-
larly to member canners, are shown in Table A2. In that a slightly shorter
cook is required, some advantage is indicated here for the recommended
processes for intact kernels. This effect reflects the differences in the
respective kernel matrices in the can. In the case of the cut corn, the
kernels tend to pack more closely together so that the brine does not
circulate freely to convect heat through the matrix; hence, the center is
heated more slowly. In the case of the intact kernels, the kernels pack
more loosely so that there is greater freedom of movement of the heated
solution within the can, and the center is more quickly heated.
TABLE A2.
RECOMMENDED STATIONARY-RETORT PROCESSING CONDITIONS(a)
Heat penetration curve Initial Time (min) at
parameters (min) temperature retort
Kernel j fh f2 Xbh temperature
type of °c 240°F 250 of
(116 °C) (I21°C)
cut 0.909 -6.6 -18.1 8.26 70 21 43 21
90 32 42 21
110 43 41 20
130 54 41 19
intac t -0.772 5.5 17.4 -7.26 70 21 40 19
90 32 39 18
110 43 39 17
130 54 38 17
(a)Reference 7.
54
-------
Agitated Retort
Heat penetration studies were also conducted in a laboratory agitated
retort and are reported in Table A3. Intact kernels required slightly
shorter processing times.
TABLE A3.
RECOMMENDED AGITATED-RETORT PROCESSING CONDITIONS(a)
Heat penetration curve
parameters (min) Time (min) at retort
Ke rnel j fh f2 fc Xbh Initial temperature of
Type Temp. 240°F 245°F 250 of
-
of °c (116°C)(118°C)(121°C)
cut 1.164 4.72 6.40 4.72 6.76 80 27 42 26 18
100 38 42 26 18
120 49 42 26 17
140 60 41 25 17
intact 1.309 3.60 4.67 3.60 2.45 80 27 41 25 17
100 38 41 25 17
120 49 41 25 16
140 60 40 24 16
(a)Reference 7.
55
-------
TECHNICAL REPOHT DATA
IPlease read JlIstruC'tiom 011 ,rhe re.p57 -14U 5467
-------
tnwironm«m»l Protection
Agency
Oferttar
0M 4S2M
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Offic*! Business
Penalty for Private Use. $300
Special Fourth-Class Rate
Book
Please make all necessary changes on the above label.
detach or copy, and return to the address in the upper
left-hand corner
II you do not wish to receive these reports CHECK HERE D,
detach or copy this cover and return to the address in the
upper left hand corner
EPA-600/2-79-193
------- |