close

Вход

Забыли?

вход по аккаунту

?

229

код для вставкиСкачать
Journal of the Science of Food and Agriculture
J Sci Food Agric 79:979±986 (1999)
Chemical and physical characteristics of
different barley samples
Annica AM Andersson,1* Cajsa Elfverson,2 Roger Andersson,1 Sigurd Regnér2 and
Per Åman1
1
Swedish University of Agricultural Sciences, Department of Food Science, PO Box 7051, S-750 07 Uppsala, Sweden
Swedish University of Agricultural Sciences, Department of Agricultural Engineering, PO Box 7033, S-750 07 Uppsala, Sweden
2
Abstract: Eight different barleys, including covered and naked samples containing low, normal and
high amylose starches as well as a sample with low starch and high b-glucan content and a malting
barley, were examined both from a chemical and physical perspective. In the chemical characterisation of the samples, analyses of nutrients were performed, while in the physical characterisation,
weight of individual kernels, sieving fractions, bulk density, terminal velocity by pneumatic
classi®cation and parameters from a Single Kernel Characterization System (SKCS) were analysed.
Results were evaluated by principal component analysis (PCA). The major trend found was that hull
content and endosperm composition varied independently of each other. Constituents found mainly in
the hull were positively correlated with each other, but negatively with bulk density and the 95%
quantile of terminal velocity. There was a positive correlation between average value and standard
deviation of grain mass, fat and starchy endosperm components such as extractable dietary ®bre
residues and b-glucan. The latter variables were negatively correlated to b-glucan extractability.
# 1999 Society of Chemical Industry
Keywords: barley; grain characteristics; PCA
INTRODUCTION
The increasing demand for high-quality barley as feed,
food and industrial raw material has led to the
development of many different cultivars of barley.
Barley is rich in spontaneous and induced mutants
that have been used throughout the world in different
barley breeding programmes.1Mutant kernel traits
that affect end-use quality are, for example, high
protein content, high lysine content,2 low and high bglucan content,3 waxy endosperm4 and high amylose
starch.5
Barley differs greatly in chemical characteristics, due
to genotype and environment and the interaction
between the two. Wide ranges in chemical composition of barley have been reported.6±8 In normal
covered barley, starch is the major constituent
accounting for about 600 g kgÿ1 of dry matter,
followed by total dietary ®bre and protein with about
200 and 110 g kgÿ1, respectively. Fat, ash and low
molecular weight sugars are minor components constituting about 30, 20 and 40 g kgÿ1 of dry matter,
respectively.6 b-Glucan and arabinoxylan are important dietary ®bre constituents. The content of b-glucan
has been reported to vary between 30 and 70 g kgÿ1
and the content of arabinoxylan between 40 and
70 g kgÿ1.4,8,9
Most physical characteristics of barley, as well as of
other grains and seeds, are normally distributed.10
Although the difference between individual grains in a
bulk sample may vary considerably, usually only the
mean values are measured. The sample is thereafter
treated as if it was homogenous. 1000 Kernel weight is
often used to determine differences in grain mean
mass between samples or varieties and has been
reported to vary between 36 and 54 g.11 In many cases
far less than 1000 grains are weighed, leading to an
increase of the 95% con®dence interval of the mean.12
Sieving is a method used either to clean, sort or
grade a sample. Sieve openings with different shapes
can be used with different sieving results. When using
sieves with oblong openings the least dimension, ie the
thickness of a barley grain, is the separating factor. The
mean thickness of barley grains has been reported to
vary between 2.5 and 2.9 mm.11 Bulk density is
in¯uenced by, for example, moisture content, grain
shape, surface characteristics and protein content.13
The presence of a hull in¯uences both grain shape,
surface characteristics and packing ef®ciency of the
* Correspondence to: Annica AM Andersson, Swedish University of Agricultural Sciences, Department of Food Science, PO Box 7051, S-750
07 Uppsala, Sweden
Contract/grant sponsor: Swedish Farmers Foundation for Agricultural Research
(Received 19 February 1998; accepted 13 November 1998)
# 1999 Society of Chemical Industry. J Sci Food Agric 0022±5142/99/$17.50
979
AAM Andersson et al
Cultivar
Table 1. The different barley cultivars and their
characteristics
Covered
Golf
High amylose Glacier
SW 906129
Karin
Normal starch, 2-rowed
High amylose starch, 6-rowed
High amylopectin (waxy) starch, 2-rowed
Normal starch, 6-rowed, malting quality
Naked
SW 8775
Hashonucier
Bz 489-30
Prowashonupana
Normal starch, 2-rowed
High amylose starch, 2-rowed
High amylopectin (waxy) starch, 2-rowed
High amylopectin (waxy) starch, low starch, 2-rowed
grains. Naked grains such as wheat and rye generally
have a higher bulk density than covered barley and
oats;11 95% hull-less barley has been reported to have
signi®cantly higher bulk density than 50% hull-less
barley.14
Terminal velocity is equivalent to the velocity of a
vertical air stream which ensures an equilibrium
between the weight of a grain and the force of the air
stream. It is in¯uenced by different properties of the
grain, eg shape, ratio between weight and crosssectional area and surface texture.11,15 The terminal
velocity of irregular shaped grains cannot be calculated
theoretically with suf®cient accuracy and it is therefore
necessary to determine it experimentally. A mean
value for terminal velocity of barley grains is reported
to be 7.6 msÿ1 compared to 8.5 msÿ1 for wheat.10
In this study we have examined samples of eight
different barley cultivars, of which most were new and
needed to be studied in greater detail. They were also
chosen to represent a wide range of different properties
in order to cover a signi®cant part of the variation
existing in barley. To our knowledge only a few studies
including both chemical and physical data have been
performed before,16,17 but none of these included as
many different analyses as this investigation. The aim
of this work was to study the variation and to ®nd
correlation between the chemical and physical characteristics of the eight different barley samples.
EXPERIMENTAL
Material
The barley (Hordeum vulgare L) cultivars Golf, High
amylose Glacier, SW 906129, Karin, SW 8775,
Hashonucier and Bz 489-30 were obtained from
SvaloÈf Weibulls AB, Sweden, while Prowashonupana
was obtained from ConAgra Inc, USA. The barley
cultivars and their characteristics, according to the
supplying companies, are presented in Table 1. The
naked cultivars Hashonucier and Prowashonupana
contained some grains with the hull remaining after
threshing (30% and about 10%, respectively) and were
therefore treated in an oat dehuller before analysis.
After dehulling the content of partly hulled grains was
10% and 0%, respectively.
980
Characteristics
Chemical analyses
Prior to analysis, representative grain samples (100 g)
were ground in a Tecator cyclone mill to pass a 0.5 mm
screen. All chemical analyses are reported on a dry
matter basis as an average of duplicate analyses. Dry
matter content of the ground grains was determined by
oven-drying at 105 °C for 5 h. Starch was determined
enzymically18 while ash and crude protein (N 6.25)
were analysed according to standard methods.19
Crude fat was extracted with petroleum ether in a
Tecator Soxtec System HT (Tecator AB, Sweden)
after acid hydrolysis with 3 M HCl.20 Dietary ®bre,
de®ned as the sum of non-starch polysaccharide
residues, amylase-resistant starch and Klason lignin,
was analysed essentially according to Theander et al,21
but with some modi®cations to separate the extractable and unextractable fractions.22
Total and unextractable b-glucan were determined
Ê man and Graham.9 The
enzymically according to A
extractability of b-glucan was calculated as the
difference between total and unextractable b-glucan,
divided by total b-glucan. Content of cellulose was
calculated as the difference between total glucose
residues in the dietary ®bre analysis and total b-glucan,
and content of arabinoxylan as the sum of arabinose
and xylose residues.
Analyses of physical characteristics
A representative sample of about 10 000 grains was
taken from each barley. The grains were weighed by
use of an automatic sorter23 which weighed each grain
individually (resolution setting of the balance 0.5 mg).
A personal computer controlled the process, recorded
the mass of the grains and determined the distribution
of mass of the individual grains for each sample. Three
representative samples of about 100 g of each barley
were sieved for 3 min in triplicates in a `Sortimat'
laboratory sieve (Pfeuffer, Tiefenstockheim, Germany) with oblong openings (2.8, 2.5 and 2.2 mm).
The bulk density was measured with a 1 l split cylinder
chondrometer (Hartner, Albstadt-Ebinger, Germany)
on two representative samples for each barley. The
moisture content was determined on unground grains
by oven-drying at 130 °C for 20 h.24
The terminal velocity was measured in a pneumatic
J Sci Food Agric 79:979±986 (1999)
Chemical and physical characteristics of barley
Cultivar
Table 2. Chemical composition of the barley
cultivars (g kgÿ1 dry matter)
Ash
Fat
Protein
Starch
Dietary ®bre
Covered
Golf
High amylose Glacier
SW 906129
Karin
23
25
25
26
22
35
34
22
87
104
105
92
638
521
566
632
189
238
198
198
Naked
SW 8775
Hashonucier
Bz 489-30
Prowashonupana
23
21
20
21
28
32
29
62
113
131
113
181
644
597
619
239
136
171
135
345
classi®er (constructed and built at the Department of
Agricultural Engineering) in which the velocity of the
air stream could be adjusted. A representative sample
of about 150 g of each barley was classi®ed. The
fraction of the sample which had lower terminal
velocity than the actual velocity of the air stream was
carried by the air stream to a cyclone where the grains
were separated from the air and passed to a settling
container. Each fraction was weighed and a cumulative frequency of the distribution of terminal velocity
was calculated.
Two samples of 300 grains of each barley, except
High amylose Glacier, whose grains were too large,
were analysed in a Single Kernel Characterization
System, SKCS 4100 (Perten Instruments North
America Inc, Reno, USA). In this system, mass,
diameter, hardness and moisture content of the
individual grains were measured and thereafter a mean
value was calculated for each sample. The system is
developed for wheat and no adjustments were made
for measurements of barley.
Statistical analyses
In order to study the variation in chemical composition
and physical characteristics and to ®nd correlation
between different variables, principal component
analysis (PCA, SIRIUS Pattern Recognition system
A/S, Bergen, Norway) was performed. The following
variables were included: content of starch; ash; crude
protein; crude fat; total dietary ®bre; extractable and
unextractable dietary ®bre; extractable and unextractable dietary ®bre polysaccharide residues; Klason
lignin and total, extractable and unextractable bglucan, as well as extractability of b-glucan; average
value and standard deviation (SD) of grain mass;
skewness of the distribution of grain mass; sieve
distribution; bulk density; 5%, 50% and 95% quantiles of terminal velocity and mass; diameter and
hardness of grains from the SKCS analysis.
RESULTS
Chemical analyses
Gross chemical composition of the eight barley
samples is shown in Table 2. Starch was generally
the major constituent (521±644 g kgÿ1), followed by
J Sci Food Agric 79:979±986 (1999)
total dietary ®bre (135±238 g kgÿ1) and protein (87±
131 g kgÿ1). In Prowashonupana the main component
was total dietary ®bre (345 g kgÿ1), followed by starch
(239 g kgÿ1) and protein (181 g kgÿ1). Fat and ash were
only minor constituents in all samples (22±62 g kgÿ1
and 20±26 g kgÿ1, respectively).
With the exception of Prowashonupana, the naked
barley samples were lower in ash and dietary ®bre and
higher in fat, protein and starch, compared to the
covered barley samples. The highest content of ash
was found in Karin, while the highest content of total
dietary ®bre and fat was found in High amylose
Glacier. Both high amylose barley samples were higher
in total dietary ®bre than the normal and waxy barley
samples. The content of protein was highest in
Hashonucier, while the content of starch was highest
in SW 8775. The lowest content of starch was found in
the high amylose barley samples. Prowashonupana
was a deviant, with a very low content of starch, and
consequently higher contents of ash, fat, protein and
dietary ®bre than any of the other barley samples.
The content of unextractable dietary ®bre was about
2.5±3.5 times higher than the content of extractable
dietary ®bre for the covered barley samples, and 1.5±2
times higher for the naked samples (Table 3). High
amylose barley samples were higher in both extractable
and unextractable dietary ®bre, compared to normal
and waxy samples, Prowashonupana excepted. Karin
had a remarkably low content of extractable dietary
®bre, especially glucose residues, compared to the
other barley samples. Glucose was the dominating
residue of both extractable and unextractable dietary
®bre in all samples. Arabinose and xylose residues
were also found in relatively large amounts. The
calculated content of cellulose and arabinoxylan was
higher in covered than in naked barley samples,
Prowashonupana excepted. The content of total and
unextractable b-glucan was higher in the waxy and the
high amylose barley samples than in the normal
samples, with the highest content found in Prowashonupana. The lowest content of b-glucan was found
in Karin. The extractability of b-glucan was lower in
the high amylose barley samples than in the normal
and the waxy samples, but lowest in Prowashonupana
at only 8%. Rhamnose and fucose residues were only
found in trace amounts in all samples, while mannose,
981
AAM Andersson et al
Table 3. Content of extractable and unextractable dietary fibre polysaccharide residues, Klason lignin and total and unextractable b-glucan (g kgÿ1 dry matter)
Covered
Naked
Constituent
Golf
High amylose
Glacier
SW 906129
Karin
SW 8775
Hashonucier
Bz 489-30
Extractable
Arabinose
Xylose
Mannose
Galactose
Glucose
Uronic acid
Total a
2.4
3.2
0.7
0.7
32
1.5
40
4.0
5.6
1.4
0.8
49
1.7
63
3.3
4.6
0.9
0.8
46
2.1
58
2.6
3.2
0.8
0.7
21
1.1
29
3.5
4.9
1.0
1.1
32
1.5
44
4.5
6.6
1.4
0.8
48
1.7
63
2.8
3.6
0.9
0.7
37
1.9
46
7.8
13
3.7
1.8
123
2.4
152
21
50
3.6
2.0
55
2.9
15
149
23
57
6.7
2.3
67
3.4
15
175
22
45
3.7
2.0
50
3.1
14
140
23
55
6.7
2.1
49
3.5
17
169
17
27
3.9
1.6
33
1.9
7.4
92
18
28
4.7
1.5
42
1.9
11
107
18
24
4.2
1.8
33
1.7
6.9
89
38
61
10
2.9
67
3.3
10
193
69
43
61
30
28
13
46
22
74
48
56
26
149
137
38.2
47
90
51.1
35
75
52.7
42
83
51.3
19
52
35.1
16
57
53.3
14
48
8.1
41
120
Unextractable
Arabinose
Xylose
Mannose
Galactose
Glucose
Uronic acid
Klason lignin
Total a
Total b-glucan
47
Water
25
unextractable bglucan
46.1
Extractability% b
40
Cellulose c
Arabinoxylan d
77
Prowashonupana
a
Including also traces (<0.1 g kgÿ1) of rhamnose and fucose.
(Total b-glucan ± unextractable b-glucan)/total b-glucan.
c
Glucose ± total b-glucan.
d
Arabinose ‡ xylose.
b
galactose and uronic acid residues as well as Klason
lignin were found in small amounts. The content of
Klason lignin was higher in the covered than in the
naked barley samples.
Physical characteristics
The distributions of mass of the individual grains are
presented in Fig 1. High amylose Glacier had the
highest average grain mass and Prowashonupana the
lowest. High amylose Glacier and Hashonucier had
the highest standard deviation (SD) of the mass of the
individual grains and Prowashonupana the lowest
(Table 4). The naked barley samples, with the
exception of Prowashonupana, had signi®cantly higher bulk density than the covered ones (P < 0.0001)
(Table 4). SKCS hardness index varied between 57.5
(Bz 489-30) and 97.4 (Prowashonupana) (Table 4).
All barley samples were classi®ed as hard grains,
according to this analysis. High amylose Glacier and
SW 906129 had the largest proportion of grains with
least dimension larger than 2.8 mm, and Prowashonupana the smallest. All barley samples had only a few
grains with least dimension smaller than 2.2 mm
except Karin and Prowashonupana, which had 6.3%
and 95.6%, respectively, in that fraction (Table 5).
The terminal velocity varied quite considerably (Fig
982
2). The two barley samples with lowest mean mass,
Karin and Prowashonupana, also had the lowest
terminal velocity. However, High amylose Glacier
with the highest mean mass had almost the same
distribution as Prowashonupana. The naked barley
samples, except of Prowashonupana, had the highest
terminal velocity and SW 8775 the highest of them all.
Figure 1. Cumulative frequency of mass of the individual grains: *,
Prowashonupana; &, Karin; ~, Bz 489-30; x SW 906129; —,
Hashonucier; ---, Golf; *, SW 8775; &&, High amylose Glacier.
J Sci Food Agric 79:979±986 (1999)
Chemical and physical characteristics of barley
Automatic sorter
grain mass mg a
Cultivar
Table 4. Grain mass determined with the automatic
sorter, bulk density and SKCS hardness index
Bulk density g lÿ1a
SKCS hardness index
45.4
52.4
9.4
11.8
732
661
69.4
not analysed
43.2
33.5
8.3
8.3
731
716
70.6
73.4
Naked
SW 8775
Hashonucier
Bz 489-30
Prowashonupana
46.4
44.6
40.3
30.5
9.6
11.2
9.5
5.8
787
799
796
684
60.1
81.6
57.5
97.4
a
The moisture content of all samples was between 105 and 110 g kgÿ1.
PCA was used to visualise the variation in the material
and to ®nd correlation between different variables.
Prowashonupana was excluded after a ®rst analysis,
since it was too different from the other barley samples
to make further analysis meaningful (result not
shown). PCA generated two signi®cant principal
components (PC1 and PC2), explaining 39% and
29% of the variance, respectively (Figs 3 (a) and (b)).
A score plot of the relationship between PC1 and PC2
showed that the barley samples could be separated into
two groups, one with the naked barley samples with
positive scores for PC1, and one with the covered
barley samples with negative scores for PC1 (Fig 3
(a)).
The loading plot showed the relationship between
different variables (Fig 3 (b)). All variables included in
the PCA except average values and SD for hardness,
SD for diameter, average values and SD for mass
determined by SKCS, F1, F2, unextractable mannose
residues, extractable uronic acid residues, extractable
galactose residues, extractable b-glucan and extractability of b-glucan were explained to more than 60%
by PC1 and PC2. The plot showed that the
constituents mainly present in the hull (Klason lignin,
ash, total dietary ®bre, unextractable dietary ®bre and
unextractable glucose, galactose, arabinose, xylose and
Cultivar
Covered
Golf
High amylose Glacier
SW 906129
Karin
J Sci Food Agric 79:979±986 (1999)
SD
Covered
Golf
High amylose
Glacier
SW 906129
Karin
Principal component analysis
Table 5. Relative distribution (weight%)
of sieve fractions
Average
Naked
SW 8775
Hashonucier
Bz 489-30
Prowashonupana
uronic acid residues) were found in the upper left
square. All these variables, except total dietary ®bre
and unextractable glucose residues, which are also
typical constituents of the starchy endosperm, were
clustered together and opposite to the variables
describing the 95% quantile of terminal velocity and
bulk density. Endosperm cell wall components (total,
unextractable and extractable b-glucan, extractable
dietary ®bre and extractable glucose, mannose, arabinose, xylose and uronic acid residues) were found in
the upper right square, together with fat content and
the average value and SD for grain mass. The 5%
quantile of terminal velocity and the skewness of the
distribution of grain mass were found opposite to each
other close to PC1, while starch and F4 were found
opposite each other close to PC2.
Objects that appear close together, like Bz 489-30
and SW 8775, have similar values for many of the
variables studied. Objects with large scores for one PC
have large values for variables with large positive
loadings for the same PC, and low values for variables
with large negative loadings. That is, High amylose
Glacier had a large F4 fraction and high content of hull
components, while it had a low starch content. Karin,
SW 8775 and Bz 489-30, on the other hand, had a
small F4 fraction, a low content of hull components
and a high starch content. Hashonucier had a high
F1
<2.2 mm
F2
>2.2 mm, <2.5 mm
F3
>2.5 mm, <2.8 mm
F4
>2.8 mm
1.1
0.2
0.5
6.3
16.0
6.4
10.2
34.2
53.7
39.0
39.9
42.9
29.3
54.4
49.5
16.6
1.2
2.9
2.8
95.6
14.7
32.9
28.6
2.6
54.8
48.5
60.0
0.6
29.2
15.7
8.6
0.3
983
AAM Andersson et al
Figure 2. Cumulative frequency of terminal velocity: &, Karin; ,
Prowashonupana; &, High amylose Glacier; x, SW 906129; ---, Golf; ~, Bz
489-30; —, Hashonucier; *, SW 8775.
protein content. Both high amylose barley samples had
a high content of b-glucan, while the solubility was
low.
DISCUSSION
Chemical components
The lower content of ash and dietary ®bre, and higher
content of fat, protein and starch in the naked barley
samples is due the absence of a hull. The hull usually
constitutes about 100±130 g kgÿ1 of the grain dry
weight25 and consists mainly of cellulose, hemicellulose (xylans), lignin and a smaller quantity of
protein.26 The higher content of protein observed in
naked barley samples has been reported before, as well
as the lower content of dietary ®bre.8,27 The level of
dietary ®bre in naked barley samples, however, has
been shown to be about 25% higher than in pearled
covered barley samples.28 The presence of hull in the
covered barley samples probably also accounts for the
larger proportion of unextractable ®bre/extractable
®bre in these samples compared to naked samples.
The waxy and high amylose barley samples had a
higher total dietary ®bre and b-glucan content than the
samples with normal starch, which is in agreement
with earlier reports.4,8,29 High contents of b-glucan
have been shown to be bene®cial for health, due to its
ability to lower serum cholesterol and blood glucose
levels.30,31 The very high content and low extractability of b-glucan in the Prowashonupana sample may be
of special interest in this respect, which has also been
shown by Newman et al 32 and Liljeberg et al.33 When
using barley for malt, the content of b-glucan should
be low. A high content of b-glucan may lead to poorer
quality of the beer because of diminished rate of wort
®ltration, haze formation and reduced extraction
ef®ciency.34 Karin is thus a suitable cultivar for
malting, since the content of b-glucan was low
compared to the other barley samples.
Physical characteristics
For terminal velocity and bulk density, which are both
in¯uenced by several properties, there was a clear
984
Figure 3. (a) Score plot and (b) loading plot of the first two principal
components from analysis of six different barley samples. DF, uDF and
eDF = total, unextractable and extractable dietary fiber; uAra, uXyl, uMan,
uGal, uGlc and uUA = unextractable arabinose, xylose, mannose,
galactose, glucose and uronic acids residues; KL = Klason lignin; eAra,
eXyl, eMan, eGal, eGlc and eUA = extractable arabinose, xylose, mannose,
galactose, glucose and uronic acid residues; bG, ubG and ebG = total,
unextractable and extractable b-glucan; extrbG = extractability of b-glucan;
F1–F4 = sieve fractions 1–4; BD = bulk density; S5, S50 and S95 = 5%, 50%
and 95% quantile of terminal velocity; Avg G and SD G = average value and
standard deviation of grain mass; Gskw = skewness of grain mass; Avg Wt
and SD Wt = average value and standard deviation of weight measured by
Single Kernel Characterization System; Avg Dia and SD Dia = average
value and standard deviation of diameter measured by Single Kernel
Characterization System.
difference between the covered and the naked barley
samples. The naked samples had both higher terminal
velocity and bulk density, with the exception of
J Sci Food Agric 79:979±986 (1999)
Chemical and physical characteristics of barley
Prowashonupana which has an abnormal shape leading to low values for these parameters. Wheat, which
resembles naked barley, also has both high terminal
velocity and bulk density.11 The 1000-kernel weight
and the mean thickness of the barley cultivars in this
study were generally in the same range as those
reported by Mohsenin.11 Karin was, however, lighter
and Prowashonupana was both lighter and had a lower
mean thickness. Naked barley has been reported to be
a soft grain, when measuring hardness with a
Brabender micro hardness tester.35 In our study, all
cultivars, both covered and naked, were classi®ed as
hard according to the SKCS method. The discrepancy
in results may be explained by differences in the
materials and methods used and that no adjustments
were made, or that our study included cultivars of
barley not used or not available in the ®rst study.
Principal component analysis
This study included samples with very different
characteristics, covering a wide range of the variation
in barley available for various applications. The
numerous analyses made on these samples provided
a good opportunity to study correlation between
different chemical and physical variables. This was
done with PCA, which showed large systematic
variation between samples with a few major trends.
Samples varied in different ways, for example,
Hashonucier had a high content of protein and a low
content of starch, while SW 8775 had a high content of
both protein and starch but a lower content of dietary
®bre. The major trend found was that content of hull
and composition of endosperm were independent of
each other, which is shown by the two arrows in Fig
3(b). The content of starch, protein and endosperm
cell walls was thus not affected by the content of hull
components. Generally, the content of hull was
explained by PC1 and the composition of endosperm
by PC2.
That the average grain mass from the automatic
sorter and from the SKCS analysis are not found close
together in the loading plot is dif®cult to explain. One
possible reason is that the SKCS only weighs 300
grains and that the standard error of the mean mass is
too large to make it possible to detect differences in
mean mass between samples.12 The SKCS is also
developed for wheat and no adjustments were made
before measuring the barley samples.
The variance for the variables close to the origin
(average hardness, extractable galactose residues and
SD for diameter, weight and hardness of the grains)
was minimally explained by PC1 or PC2, due to their
low correlation to other variables. The hull constituents were negatively correlated to bulk density, as well
as the 95% quantile of terminal velocity. This
correlation may be explained by the fact that the
presence of a hull leads to differences in, for example,
density, grain shape and surface characteristics.11,13,15
Starch content and fraction F4 were found to be
negatively correlated for the barley samples examined
J Sci Food Agric 79:979±986 (1999)
in this study. This means that the barley samples with a
high proportion of large grains (High amylose Glacier
and SW 906129) had a lower content of starch than
the other barley samples, which may be due to a
restricted starch synthesis in these cultivars. There was
a positive correlation between endosperm cell wall
constituents, such as extractable dietary ®bre residues
and b-glucan, and fat and average value and SD of
grain mass. These variables were negatively correlated
to extractability of b-glucan. This negative correlation
is mainly caused by the low extractability of b-glucan
in the high amylose barley samples.
In summary, we found a large variation between the
barley samples and interesting correlations between
chemical and physical characteristics. Conclusions
regarding groups or genotypes of barley should,
however, be drawn with caution because of the limited
number of samples in this study. That physical
characteristics of the individual grains in a bulk sample
are usually normally distributed is well known.10 It
would be interesting to investigate if the chemical
characteristics are distributed in the same way.
Fractionation using one or several physical characteristics may be useful to separate a bulk sample of
cereals into two or more fractions with different
properties of value for the end-users.
ACKNOWLEDGEMENTS
This work was partly ®nanced by the Swedish Farmers
Foundation for Agricultural Research. The authors
also wish to thank BjoÈrn Larsson, Perten Instruments
AB, Huddinge, Sweden, who performed the SKCS
analysis.
REFERENCES
1 Nilan RA and Ullrich SE, Barley: Taxonomy, origin, distribution, production, genetics and breeding, in Barley Chemistry
and Technology, Ed by MacGregor AW and Bhatty RS,
American Association of Cereal Chemists Inc, St Paul, pp 1±
29 (1993).
2 Ullrich SE, Kleinhofs A, Coon CN and Nilan RA, Breeding for
improved protein in barley, in Cereal Grain Protein Improvement. STI/PUB/664, International Atomic Agency, Vienna, pp
93±104 (1994).
3 Aastrup A, Selection and characterization of low b-glucan
mutants from barley. Carlsberg Res Commun 48:307±316
(1983).
4 Ullrich SE, Clancy JA, Eslick RF and Lance RCM, b-Glucan
content and viscosity of extracts from waxy barley. J Cereal Sci
4:279±285 (1986).
5 Merritt NR, Development of starch and other components in
normal and high amylose barley. J Inst Brewing 75:156±164
(1969).
Ê man P, Hesselman K and Tilly A-C, The variation in chemical
6A
composition of Swedish barleys. J Cereal Sci 3:73±77 (1985).
Ê man P and Newman CW, Chemical composition of some
7A
different types of barley grown in Montana, USA. J Cereal Sci
4:133±141 (1986).
Ê man P,
8 Oscarsson M, Andersson R, Salomonsson A-C and A
Chemical composition of barley samples focusing on dietary
®bre components. J Cereal Sci 24:161±170 (1996).
Ê man P and Graham H, Analysis of total and insoluble mixed9A
985
AAM Andersson et al
10
11
12
13
14
15
16
17
18
19
20
21
22
linked (1±3),(1±4)-b-D-glucans in barley and oats. J Agric Food
Chem 35:704±709 (1987).
Sitkei G, Mechanics of Agricultural Materials, Elsevier/AkadeÂmiai
KiadoÂ, Amsterdam/Budapest (1986).
Mohsenin NN, Physical Properties of Plant and Animal Materials,
Gordon and Breach Science Publishers, New York (1986).
RegneÂr S, Kernel mass related properties of cereal grains, Report
194, Dissertation, Department of Agricultural Engineering,
Swedish University of Agricultural Sciences, Uppsala, Sweden
(1995).
Yamazaki WT and Briggle LW, Components of test weight in
soft wheat. Crop Sci 9:457±459 (1969).
Rameshbabu M, Jayas DS, Muir WE, White NDG and Mills JT,
Bulk and handling properties of hulless barley. Can Agricult
Eng 38:31±35 (1996).
Grochowicz J, Machines for Cleaning and Sorting of Seeds, Foreign
Scienti®c Publications Department, Warsaw (1980).
éverland M, Heintzman KB, Newman CW, Newman RK and
Ullrich SE, Chemical composition and physical characteristics
of proanthocyanidin-free and normal barley isotypes. J Cereal
Sci 20:85±91 (1994).
Zhang WJ, Campbell LD and Stothers SC, An investigation of
the feasibility of predicting nitrogen-corrected true metabolizable energy (TMEn) content in barley from chemical
composition and physical characteristics. Can J Anim Sci
74:355±360 (1994).
Ê man P, Westerlund E and Theander O, Determination of
A
starch using a thermostable alfa-amylase. Methods Carbohydr
Chem 10:111±115 (1994).
AOAC, Of®cial Methods of Analysis, 14th edn, Association of
Of®cial Analytical Chemists, Washington DC, USA (1984).
Anonymous, Determination of crude oils and fats. Of®cial
Journal of European Community L 15:995±997 (1984).
Ê man P, Westerlund E, Andersson R and
Theander O, A
Pettersson D, Total dietary ®bre, determined as neutral sugar
residues, uronic acid residues and Klason lignin, gas chromatographic-colorimetric-gravimetric method (Uppsala method),
AOAC method 994.13, in Of®cial Methods of Analysis, 16th
edn, Suppl 1, Association of Of®cial Analytical Chemists,
Washington, USA (1995).
Ê man
Andersson AAM, Merker A, Nilsson P, Sùrensen H and A
P, Chemical composition of the potential new oil crops
Barbarea vulgaris, Barbarea verna and Lepidium campestre. J
Sci Food Agric 79:179±186 (1999).
986
23 RegneÂr S, Apparatus for determination of kernel weight
distribution in cereal grains and for sorting kernels by weight,
Report 166, Department of Agricultural Engineering, Swedish
University of Agricultural Sciences, Uppsala, Sweden (1993).
24 ASAE, Moisture measurement ± grain and seeds, in ASAE
Standard, American Society of Agricultural Engineers, pp 351
(1991).
25 Bhatty RS, Berdahl JD and Christison GI, Chemical composition and digestible energy of barley. Can J Anim Sci 55:759±
764 (1975).
26 Palmer GH and Bathgate GN, Malting and brewing. III.
Anatomy of the barley grain, in Advances in Cereal Science
and Technology, Ed by Pomeranz Y, American Association of
Cereal Chemists Inc, St Paul, Minnesota, pp 238±245 (1976).
27 Edney MJ, Tkachuk R and MacGregor AW, Nutrient composition of the hull-less barley cultivar, Condor. J Sci Food Agric
60:451±456 (1992).
28 Heen A, Eide OR and Frùlich W, Hull-less barley, agronomical
traits and nutritional quality, in International Barley Genetic
Symposium 6th, Helsingborg, Sweden, pp 446±448 (1991).
29 Xue Q, Newman RK, Newman CW and McGuire CF, Effects of
waxy gene on b-glucan, dietary ®ber content and viscosity, in
International Barley Genetic Symposium 6th, Helsinborg, Sweden, pp 458±460 (1991).
30 Newman RK, Newman CW and Graham H, The hypocholesterolemic function of barley b-glucans. Cereal Foods World
34:883±886 (1989).
31 Granfeldt Y, Liljeberg H, Drews A, Newman R and BjoÈrck I,
Glucose and insulin responses to barley products: In¯uence of
food structure and amylose-amylopectin ratio. Am J Clin Nutr
59:1075±1082 (1994).
32 Newman RK, Klopfenstein CF, Newman CW, Guritno N and
Hofer PJ, Comparison of the cholesterol-lowering properties
of whole barley, oat bran and wheat red dog in chicks and rats.
Cereal Chem 69:240±244 (1992).
33 Liljeberg HGM, Granfeldt YE and BjoÈrck IME, Products based
on a high ®bre barley genotype, but not on common barley or
oats, lower postprandial glucose and insulin response in
healthy humans. J Nutr 126:458±466 (1996).
34 Bamforth CW, Cambridge prize lecture. Biochemical approches
to beer quality. J Inst Brewing 91:154±160 (1985).
35 Bhatty RS, The potential of hull-less barley - a review. Cereal
Chemistry 63:97±103 (1986).
J Sci Food Agric 79:979±986 (1999)
Документ
Категория
Без категории
Просмотров
4
Размер файла
113 Кб
Теги
229
1/--страниц
Пожаловаться на содержимое документа