Full text of DOI:10.1002/j.2050-0416.2003.tb00142.x

)WXMQEXMRK &EVPI] 'LEVEGXIV JSV 7LSGLY 9WMRK E
7MRKPI /IVRIP 'LEVEGXIVM^EXMSR 7]WXIQ ꢀ7/'7ꢁ
A. Iwami1,2, Y. Kajiwara1 and T. Omori1
tion processes have been reproduced on a small scale,
including a pearling examination, a digestion examination
J. Inst. Brew. 109(2), 129–134, 2003
of barley koji and a fermentation examination4,5,8. In the
The SKCS (Single Kernel Characterization System) is a device
for measuring the physical properties of whole grain. Using the
pearling examination, about 70% pearling and the related
yield of pearled barley are considered to be important. In
SKCS, the physical properties of barley produced in Australia in
the digestion examination of barley koji, the amount of
2001 were examined in relation to their suitability for shochu
citric acid generated by Aspergillus kawachii and the di-
production (shochu is a Japanese clear distilled beverage with an
gestiveness of the koji are evaluated from the amount of
citric acid present. A specific level is required to prevent
moromi (mash) from rotting. Digestiveness, an indicator
of conversion power (a result of multiplying the sacchari-
fication power by the digestion power) is important. The
higher a conversion power the barley material exhibits, the
better it is considered to be. Because barley koji is equiva-
lent to malt in the whiskey production process and is used
as a source of various enzymes, the digestion examination
of barley koji is deemed as an important test allowing one
to examine fermentation suitability in advance.
However, current methods are inefficient because they
are time consuming. In addition, it has been difficult to
perform the above-mentioned methods in Australia, be-
cause of problems relating to equipment and experience.
Since evaluation methods were inefficient, it was neces-
sary to establish a routine evaluation method/indicator for
barley that would be suitable for shochu production.
Generally evaluating the material characteristics of
whole grain, i.e. the chemical properties such as raw pro-
tein content, starch content, raw fat content and minerals;
and the physical ones such as specific gravity and particle-
size distribution, is considered important. This report fo-
cuses on whole grain characteristics using barley pro-
duced in Australia in 2001 and a device called a SKCS
(Single Kernel Characterization System)1,3,6,7. This allowed
the measurement of hardness, weight, diameter and water
content of whole grain and has been adopted primarily for
the quality classification of wheat. Here, we report the re-
sults of an evaluation by first identifying the physical prop-
erties of the whole grain, i.e., the solid-state properties of
whole grain, pearled barley properties obtained by con-
ventional pearling examination, and fermentation perform-
ance as indicated by the digestion properties of barley koji.
alcohol content of ~25%). The hardness of whole grain showed a
positive correlation with pearling, but negative correlations with
both broken kernel ratio and conversion power. Softer endo-
sperms facilitated crushing and fermentation (i.e. fermentation
potential). From these correlations, quality characteristics of
barley for shochu could be deduced. The measurement of barley
hardness with the SKCS is important in the evaluation of barley
for shochu production and the hardness of the whole grain could
be used as one indicator in the evaluation of barley. The proce-
dure can be carried out locally and rapidly.
Key words: Barley, hardness, shochu, ‘Single Kernel Character-
ization System’, SKCS.
-2863(9'8-32
Two-rowed barley is generally used as a raw material
for barley shochu. Shochu is a Japanese clear distilled bev-
erage with an alcohol content of ~25%. Currently, in Ja-
pan, most of the raw barley for shochu is imported from
Australia as two-rowed barley for food use. The annual
import volume is approximately 160–180 thousand tons.
The quality control of two-rowed barley in Australia is
performed in some states similar to that of barley for sho-
chu. However, it is basically an evaluation for malt pro-
duction and no appraisal is performed by the shochu manu-
facturers. Therefore, it cannot be said that a quality classi-
fication of two-rowed barley suitable for barley shochu is
in place.
The basic method of shochu manufacturing involves
barley pearled approximately 70% from whole grain (un-
polished barley) as the raw material, The production pro-
cess consists of four stages: material treatment (imbibition
and steam boiling), koji stage, fermentation, and distilla-
tion (Fig. 1). The quality evaluation of barley for shochu
must be performed on both pearling and fermentation char-
acteristics. In the conventional methods of evaluating the
pearling and fermentation characteristics, actual produc-
1%8)6-%07 %2( 1)8,3(7
1 SANWA SHURUI Co. Ltd, 2231-1 Yamamoto, Usa-shi Oita 879-
0495, Japan
2 Corresponding author. E-mail: iwami-a@kokuzo.co.jp
8IWX WEQTPIW
Samples (139 batches) of barley produced in Australia
in 2001 were tested, after storing in a cooler at 4°C for
one month, to perform moisture control. Specifications are
given in Table I.

Steamed barley
Barley koji
Whole grain
Koji mould
Pearling
Pearled barley
Washing
(Koji making)
1st mash
2nd mash
Yeast
Water
Steeping
(Fermentation)
Steaming
Distillation
Cooling
Barley shochu
(Material treatment)
Fig. 1. Diagram of the process to produce barley shochu.
4L]WMGEP TVSTIVXMIW SJ FEVPI]
The Single Kernel Characterization System (SKCS
*IVQIRXEXMSR GLEVEGXIVMWXMGW
Digestion examinations and citric acid producing
power of barley koji was performed to assess fermentation
properties. Koji was produced by using barley samples
pearled at a pearling rate of 65% using a Thermo-Hygro-
stat LH-30 (Nagano Science Equipment MFG. Co., Ltd.,
Japan) The digestion examination was conducted using
barley koji and was based on the method of Horie et al.2
Citric acid producing power was assayed by the measure-
ment of the acidity of the koji.
4100) (Perten Instruments, Springfield IL) was used for
measuring the physical properties of the whole grains. Af-
ter randomly choosing 20 g of whole grains, the hardness,
weight (mg), diameter (mm) and water content (%) values
of 300 whole grains were measured with the SKCS 4100,
to obtain the average and standard deviation of these 300
grains.
4IEVPMRK GLEVEGXIVMWXMGW
Pearling examination was performed using the Test
Mill TM-05 (Satake Corporation, Hiroshima, Japan). Us-
ing a revolving speed of 1,450 rpm, a roller count of 40,
and whole grain samples of 150 g per examination as
pearling conditions, the pearling was performed for two
minutes with the given method. Then, the pearled sample
was prepared by bran removal with a 2.0 mm mesh screen.
The pearling operation is shown in Fig. 2. The pearling
yield and the relevant broken kernel ratio, highly regarded
in the pearling properties, were calculated as follows:
6)79087
4L]WMGEP TVSTIVX] HMWXVMFYXMSR SJ [LSPI KVEMRW
ERH TL]WMGEP GLEVEGXIVMWXMG JSV IEGL ZEVMIX]
For the whole grain samples provided, the distribution
of physical properties and physical characteristics for each
variety were examined. For the moisture value, no data is
given since moisture control was performed.
Whole grains (150 g)
Sound kernel weight
Pearling yield (%) =
™100
Whole grains weight (150 g)
Test Conditions
Rpm: 1,450
Stone: #40
Broken kernel weight
™100
Broken kernel ratio(%) =
Time: 2 min
Pearled barley weight
Bran
*After pearling, the bran was
removed using a 2.0 mm mesh
Table I. Varieties of barley in this study.
Pearled barley
sieve
Variety
Harvest area
No. of samples
Stirling
Schooner
Alapiles
Western Australia
South Australia, Victoria
Victoria
33
88
18
Broken kernel
Sound kernel
Total
139
Fig. 2. The pearling process.

A)
B)
C)
100
80
60
40
20
0
100
80
60
40
20
0
100
80
60
40
20
0
<20 20-30 30-40 40-50 50-60 60-70 70-80
<38 38-40 40-42 42-44 44-46 46-48 48-50
<2.3
2.3-2.4 2.4-2.5 2.5-2.6 2.6-2.7
Hardness
Weight (mg)
Diameter (mm)
□
Fig. 3. Distribution of physical characteristics of whole grains in each variety. Stirling (ꢀ); Schooner (ꢀ); Alapiles ( ). A) Hardness,
B) weight, C) diameter.
1) Hardness distribution of whole grains. The hard-
ness distribution of whole grains for each variety is shown
in Fig 3A. Seven hardness classifications from below 20 to
below 80 were adopted. Approximately 83% of total sam-
ples lay within the hardness range of 50 to 70. The ave-
rage hardness and standard deviation of the total sample
was 56.98 and 8.76, respectively (Table II). Looking at
each variety, e.g. with Stirling, the hardness level of 60–70
was the most frequent, being 94% and that of 70–80 was
the next best at 6%. The average hardness and standard
deviation of Stirling was 65.41 and 2.83, respectively. For
Schooner, the hardness level of 50–60 was the most fre-
quent at 44% and that of 60–70 was the next best at 33%.
The remaining samples lay within the hardness range of
20 to 50. The average hardness and standard deviation of
Schooner was 54.44 and 8.70, respectively. For Alapiles,
83% of the samples were below the hardness level of 50–
60 but showed a range of hardness levels of 30–40, 40–50
and 60–70 each at 6%. The average hardness and standard
deviation of Alapiles was 53.93 and 5.73, respectively.
Comparison among varieties using the average and
standard deviation values showed the following result:
higher hardness and narrower hardness distribution of
Stirling; lower hardness and wider hardness distribution of
Schooner. For Alapiles, the hardness was found to be al-
most the same as that of Schooner and the hardness distri-
bution lay between that of Stirling and Schooner.
Looking at each variety, it was found that the most fre-
quent levels of Stirling, Schooner and Alapiles were those
of 42–44 mg for (55%), 44–46 mg (34%) and 40–42 mg
(39%), respectively. The average weights of Stirling,
Schooner and Alapiles were 42.09 mg, 43.80 mg and 42.36
mg, respectively (Table II). Comparison among varieties
using the average and standard deviation values indicated
a relatively low weight per grain for Stirling, higher
weight per grain for Schooner and an in-between weight
for Alapiles. For the weight distribution, Schooner,
Alapiles, and Stirling showed the widest, the second wid-
est and the least, respectively.
3) Diameter distribution of whole grains. The diame-
ter distribution of whole grains is shown in Fig. 3C. Five
diameter classifications from below 2.3 mm and below 2.7
mm were adopted. Of total samples, 89% lay within the
diameter range of 2.4 mm to 2.6 mm. Looking at each
variety, it was found that the most frequent levels were
Stirling (61%) 2.4–2.5 mm, Schooner (44%) 2.5–2.6 mm
and Alapiles (67%) 2.4–2.5 mm. The average diameters of
Stirling, Schooner and Alapiles were 2.49 mm, 2.49 mm
and 2.46 mm respectively (Table II).
Table II. Physical characterisics of whole grains in each variety.
Weight*
(mg)
Diameter*
(mm)
Variety
Hardness*
2) Weight distribution of whole grains. The weight
distribution of whole grains is shown in Fig. 3B. Seven
classifications, from below 38 mg to below 50 mg, were
adopted. The level of 42–44 mg was the most frequent at
approximately 33% and the levels of 40–42 mg and 44–46
mg were the next best at approximately 26%, respectively.
Stirling
Schooner
Alapiles
65.41 2.83
54.44 8.70
53.93 5.73
42.09 1.31
43.80 2.04
42.36 1.80
2.49 0.04
2.49 0.07
2.46 0.07
Total
56.98 8.76
43.21 2.03
2.49 0.06
*Mean SD.

A)
B)
100
80
60
40
20
0
100
80
60
40
20
0
<55
55-60
60-65
65-70
70-75
75-80
<5
5-10
10-15
15-20
20-25
25-30
Pearling yield (%)
Broken kernel (%)
Fig. 4. Distribution of pearling characteristics of whole grains in each variety. Stirling (ꢀ); Schooner (ꢀ); Alapiles
□
( ). A) Pearling yield, B) broken kernel.
most frequent level ranged from 1050–1100. Looking at
each variety, it was found that the most frequent levels of
Stirling, Schooner and Alapiles were those of 1050–1100
(33%), 1100–1150 (30%) and 1050–1100 (44%), respec-
tively. The average conversion power readings of Stirling,
Schooner and Alapiles were 1029, 1116 and 1097, respec-
tively (Table III).
2) Acidity distribution of whole grains. The acidity
distribution of whole grains is shown in Fig. 5B. Five acid-
ity classifications from below 4 to below 8 were adopted.
Approximately 92% of the total samples lay within the
pearling yield range of 4 mL to 7 mL. Looking at each
variety, the average acidity values of Stirling, Schooner
and Alapiles were 5.6 mL, 5.4 mL and 5.0 mL, respec-
tively (Table III).
4IEVPMRK GLEVEGXIVMWXMGWꢂ HMWXVMFYXMSR SJ [LSPI
KVEMRW ERH TIEVPMRK GLEVEGXIVMWXMG JSV IEGL ZEVMIX]
For the whole grain samples, the distribution of pearl-
ing characteristics for each variety was examined.
1) Pearling yield distribution of whole grains. The
pearling yield distribution of whole grains is shown in Fig.
4A. Six pearling yield classifications from below 55% to
below 80% were adopted. About 86% of total samples lay
within the pearling yield range of 60% to 70%. Looking at
each variety, it was found that the most frequent levels
of Stirling, Schooner and Alapiles were those of 65%–
70% for (79%), 70%–75% for (45%) and 60%–65% for
(44%), respectively. The average pearling yields of Stir-
ling, Schooner andAlapiles were 66.9%, 67.1% and 65.7%,
respectively (Table III).
2) Broken kernel ratio distribution of whole grains.
The broken kernel ratio distribution of whole grains is
shown in Fig. 4B. Six pearling yield classifications from
below 5% to below 30% were adopted. Over 90% of total
samples lay within the broken kernel ratio below 10%. For
each variety, the most frequent levels were below 5%,
each frequency was 55% (Stirling), 64% (Schooner) and
94% (Alapiles), respectively. The average broken kernel
ratios of Stirling, Schooner and Alapiles were 5.30%,
5.54% and 1.97%, respectively (Table III).
Table III. Pearling and fermentation characteristics of each variety.
Pearling
yield*
(%)
Broken
kernel*
(%)
Conversion
power*
(–)
Acidity*
(mL)
Variety
Stirling
Schooner
Alapiles
66.9 3.21 5.30 3.06
67.1 7.37 5.54 6.38
65.7 7.03 1.97 1.36
1029 56
1116 82
1097 49
5.58 0.91
5.42 0.71
4.95 0.72
Total
66.9 6.62 5.02 5.47
1093 82
5.39 0.79
*Mean SD.
*IVQIRXEXMSR GLEVEGXIVMWXMGW HMWXVMFYXMSR SJ
[LSPI KVEMRW ERH JIVQIRXEXMSR GLEVEGXIVMWXMGW
JSV IEGL ZEVMIX]
Table IV. Correlation coefficient for relationships between the whole
grain physical characteristics and shochu characteristics.
For the whole grain samples, the distribution of fer-
mentation characteristics for each variety was examined.
1) Conversion power distribution of whole grains.
The conversion power distribution of whole grains is
shown in Fig. 5A. Eight comprehensive power classifica-
tions from below 950 to below 1300 were adopted. The
Hardness
Weight
Diameter
Pearling yield
Broken kernel ratio
Acidity
0.63**
–0.45**
0.08
0.09
0.03
–0.10
0.00
0.00
0.03
0.03
Conversion power
–0.56**
–0.16
** Significant at 1% level.

A)
B)
100
80
60
40
20
0
100
80
60
40
20
0
<4
4-5
5-6
6-7
7-8
<950
950- 1000- 1050- 1100- 1150- 1200- 1250-
1000
1050 1100
1150 1200 1250
1300
Acidity (mL)
Acidity (mL)
Conversion power (-)
Conversion power (–)
Fig. 5. Distribution of fermentation characteristics of whole grains in each variety. Stirling (ꢀ); Schooner (ꢀ);
□
Alapiles ( ). A) Conversion power, B) acidity.
B)
C)
A)
100
80
60
40
20
0
40
30
20
10
0
1400
1200
1000
800
600
0
20
40
60
80
0
20
40
60
80
0
20
40
60
80
Hardness
Hardness
Hardness
ꢂ
Fig. 6. Correlation between hardness and pearling/fermentation characteristics in each variety. Stirling (ꢁ); Schooner (ꢀ); Alapiles ( ).
A) Pearling yield, B) broken kernel, C) conversion power.
among the physical property values, pearling yield and
broken kernel ratio indicated a correlation with the hard-
ness of the whole grain. A relatively high positive correla-
tion was observed between the hardness of the whole grain
and the pearling yield (Fig. 6A). Between the hardness of
the whole grain and the broken kernel ratio, a negative
correlation was observed (Fig. 6B). However, no correla-
'SVVIPEXMSR FIX[IIR XLI TL]WMGEP TVSTIVXMIW SJ
[LSPI KVEMR ERH XLI TIEVPMRK ERH JIVQIRXEXMSR
GLEVEGXIVMWXMGW
The coefficient of correlation between the physical
properties of the whole grain and the pearling and fermen-
tation characteristics are shown in Table IV. Comparisons

tion was found between the weight of whole grain, diame-
ter of whole grain and pearling characteristics (Table IV).
Comparisons between the physical property values, the
acidity from koji and conversion strength indicated a nega-
tive correlation between the hardness of the whole grain
and the conversion strength (Fig. 6C). No correlation was
observed with any other physical property. For acidity
from koji, no correlation was observed with the physical
properties.
The hardness of the whole grain showed a correlation
with pearling and fermentation characteristics. Therefore,
pearling yield, broken kernel ratio and conversion power
trends of barley material may be used as one indicator to
evaluate the characteristics of barley for shochu. In the
selection of material that is suitable for shochu, the mea-
surement of hardness using SKCS can be performed in a
few minutes, allowing easy and rapid selection of many
samples at one locality for the production of either shochu
or pearled barley.
(-7'977-32
Results presented revealed that the hardness of whole
grain influences pearling and fermentation characteristics.
In the samples studied the hardness characteristics were
observed for each variety. Results suggest that although
varietal characteristics are important in determining hard-
ness, environmental factors are also important. Further
work is required to assess the relationship between varie-
tal and environmental factors on hardness and the relation-
ship between hardness and fermentation properties of the
pearled grain.
In the pearling industry, barley with a higher pearling
yield and the relevant lower crushing rate is generally re-
garded as good material. The greater the hardness of nor-
mal barley, the greater the high-pearling yield and the re-
sistance to crushing. Therefore, the results suggest that
barley with higher hardness appears to be the more advan-
tageous in regard to pearling performance.
On the other hand, in the manufacture of barley sho-
chu, barley with a higher conversion strength of barley
koji is considered to be the better material4,5. Therefore,
the results suggested that barley with lower hardness
seems to be the more advantageous in regard to fermenta-
tion performance. Namely, fermentation suitability appears
to depend on the hardness of the unpolished barley. This
difference in the conversion strength of koji due to hard-
ness may influence moromi control, and the shochu qual-
ity itself.
1. Gaines, C.S., Finney, P.F., Fleege, L.M. and Andrews, L.C., Pre-
diction a hardness measurement using the Single-Kernel Charac-
terization System. Cereal Chem., 1996, 73(2), 278–283.
2. Horie, S., Tosa, N. and Hosoya, T., Trial of quality evaluation of
ginjo koji (Koji for specially made sake). J. Brew. Soc. Japan,
1992, 87(1), 57–61 (in Japanese).
3. Martin, C.R., Rousser, R. and Brabec, D.L., Development of a
single-kernel wheat characterization system. Transactions ASAE,
1993, 36(5), 1399–1404.
4. Ogawa, K., Omori, T., Shimoda, M., Wada, H., Shiroishi, M.,
Saito, K., Kwatu, K., Hida, N., Furue, K. and Nagamori, Y., Esti-
mates two-rowed barley variety for Shochu-making. J. Brew.
Soc. Japan, 1996, 91(9), 652 (in Japanese).
5. Ogawa, K., Omori, T., Shimoda, M., Shiroishi, M., Saito, K.,
Kwatu, K., Hida, N., Furue, K. and Nagamori, Y., Estimates two-
rowed barley variety for Shochu-making (Vol. 2). J. Brew. Soc.
Japan, 1998, 93(9), 750 (in Japanese).
6. Osbone, B.G., Kotwal, Z., Blakeney, A.B., O’Brien, L., Shah. S.
and Fearn, T., Application of single-kernel characterization sys-
tem to wheat receiving testing and quality prediction. Cereal
Chem., 1997, 74(4), 467–470.
7. Osbone, B.G., Turnbull, K.M., Anderssen, R.S., Rahman, S.,
Sharp, P.J. and Appels, R., The hardness locus in Australian
wheat lines. Australian J. Agric. Res., 2002, 52, 1275–1286.
8. Palmer, G.H., Microscopic and other studies of rice and rice
kojis. Bulletin of Research, Institute for Food Science: Kyoto Uni-
versity, 1994, 57, 29–68.
(Manuscript accepted for publication, May 2003)