However, there were significant correlations between the [4] Wootton, M., and D.

Wootton, M., and D. Mahdar: Properties of starch from Australian

enthalpy involved and each of VAG maximum viscosity, wheats. III. In vitro digestibility and hydroxypropyl derivatives.
breakdown, viscosity at 50 °C and set back (r = 0.53, Starch/Stärke 45 (1993), 337–341.
[5] Dengate, H. N.: Swelling, pasting and gelling of starch, in:
P < 0.05; r = 0.49, P < 0.05; r = 0.62, P < 0.05; r = 0.65, Advances in Cereal Science and Technology Vol VI. American
P < 0.01, respectively). For RVA data, correlations were Association of Cereal Chemists, St. Paul, MN 1984.
found only between this enthalpy and maximum viscosity [6] Hong, S. H.: The effects of wheat genotype and growth environ-
and breakdown (r = 0.53, P < 0.05 for both). This suggests ment on factors governing the quality of white dry salted noodle.
that higher levels of amylose/lipid complex in the starch are PhD Thesis, The University of New South Wales 1992.
associated with higher paste viscosities. [7] AACC Approved Methods 8th edn. American Association of
Cereal Chemists, St. Paul, MN, 1983.
[8] Williams, P. C., F. D. Kuzina, and I. Hlynka: A rapid calorimetric
4 Conclusions procedure for estimating the amylose content of starches and
flours. Cereal Chem. 47 (1970), 411–420.
[9] RACI Official Testing Methods. Cereal Chemistry Division,
Comparison of the parameters measured for each starch in Royal Australian Chemical Institute, Parkville Vic 1988.
this study, and investigation of correlations between them [10] Deffenbaugh, L. B., and C. E. Walker: Comparison of starch
using linear regression analysis, indicated that starch pro- pasting properties in the Brabender Viscoamylograph and the
perties varied between different wheat cultivars grown at the Rapid Visco-Analyser. Cereal Chem. 66 (1989), 493–499.
same location. Apparent amylose content appeared to be the [11] SAS User’s Guide: Statistics 5th edn. Carey: SAS Ins.
most important characteristic since it correlated significantly [12] Soulaka, A. B., and W. R. Morrison: The amylose and lipid
with several VAG and RVA parameters, and proportion of contents, dimensions and gelatinisation characteristics of some
wheat starches and their A- and B- granule fractions. J. Sci. Food
small granules. For the DSC gelatinisation endotherm, all Agric. 36 (1985), 709–718.
three temperatures correlated significantly with proportion [13] Evers, A. D.: The size distribution among starch granules in wheat
of small starch granules but not with apparent amylose. This endosperm. Starch/Stärke 25 (1973), 303–304.
confirms that higher gelatinisation temperatures are required [14] Evers, A. D., and J. Lindley: The particle size distribution in wheat
for smaller granules. No significant correlations were found endosperm starch. J. Sci. Food Agric. 28 (1977), 98–102.
between gelatinisation enthalpy and any other starch property. [15] Stevens, D. J., and G. A. H. Elton: Thermal properties of the
Significant correlations between VAG measurements and the starch/water system. I. Measurement of heat of gelatinisation by
enthalpy of amylose/lipid melting confirm the importance of differential scanning calorimetry. Starch/Stärke 23 (1971), 8–11.
[16] Ghiasi, K., R. C. Hoseney, and E. Variano-Marston: Gelatinisation
this complex in affecting starch pasting properties. of wheat starch. III. Comparison by differential scanning calori-
metry and light microscopy. Cereal Chem. 59 (1982), 258–262.

Bibliography Addresses of authors: Assoc. Prof. Michael Wootton*. Department

of Food Science and Technology. The University of New South Wales.
[1] Moss, H. J., and D. M. Miskelly: Variation in starch quality in Sydney, NSW 2052, Australia. Joseph F. Panozzo, Victorian Institute
Australian flour. Food Technol. Aust. 36 (1984), 90–91. for Dryland Agriculture, Horsham Viv 3400, Australia. Sung-Hie Hong,
[2] Wootton, M., and D. Mahdar: Properties of starches from Aus- Korea Food Research Institute, San 64-1, Beakhyun-dong, Budang-ku,
tralian wheats. I. Separation of starch and gluten. Starch/Stärke 45 Sungnam-si, Kyonggi-do 463-420, Republik of Korea.
(1993), 255–258.
[3] Wootton, M., and D. Mahdar: Properties of starches from Aus- *Correspondence author.
tralian wheats. II. Some physico-chemical properties. Starch/
Stärke 45 (1993), 295–299. (Received: September 8, 1997).

A Simple and Rapid Colorimetric Method

for the Determination of Amylose in Starch Products
Scott J. McGrance, Hugh J. Cornell,
and Colin J. Rix, Melbourne (Australia)

A method for the determination of the amylose content in starch ces. A linear relationship was obtained between absorbance and amy-
has been developed which is based on the colorimetric measurement of lose concentration for mixtures of amylose and amylopectin standards,
the iodine complexes formed with amylose and amylopectin. The and this forms the basis of the determination. The method is rapid,
method requires measurement at only one wavelength and avoids the simple, accurate and does not require the use of multi-component
use of harsh dispersants for the starch. Dimethyl sulphoxide is used as analysis of spectra, since a wavelength is chosen that suits the particular
the dispersant and a wavelength of 600 nm can be used for measure- starch being analysed. It can be adapted to a micro-scale method if
ment of the amylose content of starches from different botanical sour- necessary.

1 Introduction proposed that the iodine component of the complex is

present in a unidimensional array within an amylose helix
The reaction between starch and iodine has been known with six glucose residues per turn [1]. Much later, I 3– was
for over a century. Some fifty years ago, Rundle and Baldwin proposed as the iodine species responsible [2]. However, the

158 Starch/Stärke 50 (1998) Nr. 4, S. 158–163 © WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998 0038-9056/98/0404-0158$17.50+.50/0
reaction has become better understood in more recent times 11.0–11.5 % moisture as determined by both Karl-Fischer
when Teitelbaum et al. [3, 4] studied the structure of the titration and oven drying at 105 °C for 18 h. The results were
iodine complex using resonance Raman and iodine-129 expressed on a dry basis.
Mössbauer spectroscopy and showed that the major chromo- Starch, or a fraction thereof (0.1 g), was accurately
phore was the pentaiodide anion, I5–. weighed and dissolved by heating in DMSO for 15 min in a
Many techniques employing the starch-iodine reaction water bath at 85 °C. When dissolved, this solution was then
have been used to determine the amount of amylose in diluted to 25 mL in a volumetric flask with deionised water.
starch, where amylose is present in a natural mixture with An aliquot (1.00 mL) of this solution was then diluted
amylopectin. Most of these techniques rely on potentio- with 50.00 mL of water, 5.00 mL of a solution of iodine
metric titration of bound iodine or colorimetric estimation of (0.0025 mol/L) in potassium iodide (0.0065 mol/L) added
the complex. The potentiometric method suffers from being with mixing and the absorbance of a sample of this solution
a slow, non-ionic reaction and the broad inflection point in a 1 cm path length quartz cell read at 600 nm using a Cary
leads to some inaccuracies [5]. 1C UV/Visible spectrophotometer. If a slight turbidity was
Problems of dispersibility of the starch are relevant to observed in the starch sample, the mixture was heated gently
both the potentiometric and colorimetric methods. The with about 5 mL of water and cooled quickly before adding
above potentiometric method uses 0.5 mol/L potassium the remaining water and the iodine reagent. Some samples
hydroxide for dispersion of the starch, but such a strong were left for 15 min after the addition of the iodine before
alkali is able to cause degradation of the starch components taking a reading on the spectrophotometer. We found only a
and mixtures require neutralisation before the iodine assay is slight increase in absorbance of the samples after standing
performed. Acids such as perchloric [6] and hydrochloric [7] for 15 min and this made no significant difference to the
are also considered harsh dispersants, and whilst they may results.
be suitable for certain plant materials, they are not good A standard curve was plotted for mixtures of amylose and
dispersants for high amylose starches or preparations rich in amylopectin from potato containing 0, 10, 25, 50, 75, and
amylose. 100 % amylose. Similarly, a standard curve was plotted for
Considerable progress has been made with the use of mixtures of corn amylose (70 % amylose content) and corn
dimethyl sulphoxide (DMSO) as a dispersant for starches. amylopectin. Spectra from 270–900 nm were recorded for
DMSO has been shown to be superior to strong alkalies most samples, including blanks for the amylose and amylo-
[8, 9] in this regard. Mixtures of 90 % DMSO and 10 % water pectin solutions and the iodine reagent.
have been used [10] but at room temperature dissolution is
slow, even with good stirring. Morrison and Laignelet [11] 2.2 Other experiments
have developed a similar method based on the use of a mix- (a) Preliminary experiments were carried out using
ture of 90 % DMSO and 10 % aqueous 6 mol/L urea. Howe- UV-visible spectrophotometric scans of all samples over
ver, this method is somewhat time-consuming and requires the wavelength range 270–900 nm. These spectra were
careful attention to detail if accurate results are to be obtai- necessary for determining the amount of iodine required for
ned. At the temperature used for dispersion (100 °C) some a given amount of amylose or amylopectin and for checking
degradation of the starch may occur, and after addition of the wavelength maxima of the mixtures for the different
iodine there is some drift in absorbance values which require starches. Potato amylose was chosen for iodine-binding
a 15 min waiting period. This latter phenomenon may be due capacity as it will show less interference from lipid than corn
to the presence of the urea. Lipids also form inclusion com- starch.
plexes with starch and this can influence the result. Hence, (b) In order to exploit the observed differences between
the Morrison and Laignelet method includes the use of 85 % the binding of iodine to amylose and amylopectin, ab-
methanol-water (v/v) or water-saturated butanol to extract sorbance values corresponding to one of the iodine peaks
these materials from the starch prior to analysis. (291 nm) were determined on mixtures of corn amylose and
The estimation of amylose by iodine, based on the for- amylopectin as described in Section 3.1., but using 0.2 mL of
mation of helical inclusion complexes, also includes a con- 0.0025 mol/L iodine with 1.00 mL of a solution of 20 mg of
tribution from the amylopectin. Jarvis and Walker [12] have carbohydrate mixture diluted with 5.00 mL of deionised
placed special emphasis on this aspect of amylose deter- water.
mination and have developed an assay based on multi-com- (c) Some experiments were also carried out using potas-
ponent analysis using 6 wavelengths, which enables both sium hydroxide (1 mol/L) as dispersant, according to the
amylose and amylopectin to be determined directly. method of Jarvis and Walker [12] in order to make compari-
In the present paper, we have followed the approach of sons of the spectra.
Knutson and Grove [13] by using DMSO as dispersant, (d) Samples of commercial wheat starch and corn starch
but without the use of sonication, and have used a simple were defatted with water-saturated butanol for 6 h at 100 °C
standard calibration graph which takes into account the [11] and amylose determinations carried out for comparison
contribution made by the amylopectin. with those of the commercial starches.
(e) Sigma corn amylose (70 % amylose) was purified
from contaminating amylopectin by formation of its in-
2 Experimental clusion complex with 1-butanol according to the following
procedure. The crude amylose (15 % (w/v)) was dispersed in
2.1 Main method DMSO (100 mL) and heated in a water bath at 85 °C for
Pure potato amylose and amylopectin, practical grade 15 min. Water (400 mL) at 80 °C was added and then heated
corn amylose (70 % amylose) and corn amylopectin (100 % at 100 °C for 3 min. 1-butanol (50 mL) was then added and
amylopectin) were all obtained from Sigma Chemical Com- the mixture allowed to stand for 2–3 h to allow precipitation
pany, St. Louis, MO. Potato starch, wheat starch, and two of the amylosebutanol complex from solution. The mixture
samples of corn starch were obtained from commercial was then centrifuged at 6,000 rpm for 30 min and the amy-
sources. All standards and samples tested contained lose complex washed with ethanol to remove the DMSO.

Starch/Stärke 50 (1998) Nr. 4, S. 158–163 159

The purified amylose sample was obtained by heating in an
oven at 35–40 °C until dry (about 6 h). The product was then
analysed for its amylose content. A second precipitation was
also carried out on the final product.

3 Results and Discussion

3.1 Amount of iodine required
The amount of iodine required can readily be gauged by
the gradual addition of iodine to a known amount of amylose
until maximal absorbance in the visible range of the spec-
trum is obtained. In the present work, this corresponded to
4.00 mL of 0.0025 mol/L iodine reagent to a 1/25 portion of
the amylose solution, i.e. 0.004 g amylose.
Hence total mass of iodine = (4/1000) ´ 0.0025 ´ 254 g
= 0.00254 g
and moles I5– = 0.00254/(5 ´ 127)
= 4.0 ´ 10–6 moles I5–
–
and molecules I5 = 4.0 ´ 10–6 ´ NA,
where NA =
Avogadro’s number.
If M = molar mass of amylose, the number of anhydro-
glucose units (molar mass = 162) in 0.004 g amylose
= (0.004/M) ´ (M/162) ´ NA
= 2.47 ´ 10–5 ´ NA anhydroglucose units
And on the basis of the suggestion that there are 6 anhy-
droglucose units per turn of the helix [1, 14], the number of
turns on 0.004 g of amylose will be
= (2.47 ´ 10–5 NA)/6
= 4.1 ´ 10–6 ´ NA turns
This result suggests one turn of the helix accommodates Fig. 1. An optimised space-filled molecular model of a left-handed
one I5– anion. However, the pitch of the collapsed amylose helix of amylose containing six anhydroglucose units per turn. A
–
helix is about 800 pm [14] and the length of an I5– anion is pentaiodide ion (I5 ) is shown inside the helix, parallel to its long axis.
about 1460 pm. Allowing for an extension of the helix up to End-on (top) and side-on (bottom) views are shown, with the side-on
view containing seven anhydroglucose units to illustrate the continua-
a pitch of 2100 pm (the so-called extended form) [14] and tion of the helical structure. A Hyperchem® molecular modelling
different affinities of binding, the results are in accord with program was used.
a model in which an I5– is within almost every turn of the
amylose helix, parallel to its long axis. The conformation
shown (refer Fig. 1) is one of four possible structures all
having the same pitch (1500 pm). This conformation pro-
vides a very hydrophobic interior and quite a hydrophilic
outer surface. The differences are the result of various
orientations of the –CH2OH groups (shown in green).
The way in which the cation (K+) is associated with the
I5–-amylose complex remains uncertain.
3.2 Spectral patterns of starch-iodine complexes
The spectral patterns for iodine-potato amylose mixtures
(refer Fig. 2) showed three main peaks:
(1) 288–292 nm
(2) 349–352 nm
(3) 628–636 nm (broad peak)
Some of the variations in wavelength maxima seemed to
depend upon the relative amounts of iodine and amylose
used.
The spectra of iodine-potato amylopectin mixtures are
different in that the broad peak occurs at lower wavelength
(552 nm) and is less intense for the same amount of carbo- Fig. 2. UV-visible spectra of iodine-potato amylose and amylopectin
hydrate (refer Fig. 2). At low amounts of amylose, this diffe- complexes in comparison with the iodine reagent blank. Where A =
rence is detectable with the naked eye, with the iodine-amy- amylose, AP = amylopectin, and iodine = iodine reagent blank. Pre-
lose complex being observed as a bright blue colour and the paration as in Section 2.1.

160 Starch/Stärke 50 (1998) Nr. 4, S. 158–163

Table 1. Absorbance maxima (mn)* of complexes from potato and corn
starch components together with potato amylose/amylopectin mixtures
and starches. The iodine reagent blank is also shown for comparison.

Sample Peak 1 Peak 2 Peak 3

Potato amylose 288 351 636
75 % potato amylose & 288 351 628
25 % potato amylopectin
50 % potato amylose &
50 % potato amylopectin 289 351 613
25 % potato amylose &
75 % potato amylopectin 288 352 592
10 % potato amylose &
90 % potato amylopectin 288 351 566
Potato amylopectin 288 351 552
Corn amylose (70 %) 289 351 604
Corn amylopectin 288 351 521
Corn starch 1 288 351 606
Wheat Starch 289 351 617
Iodine reagent 288 352 –
(* obtained with 4 mg carbohydrate in 1.00 mL, diluted with 50.00 mL
of water, after addition of 5.00 mL of 0.0025 mol/L iodine reagent. All
percentages are w/w, dry basis.).

Fig. 4. Plot of absorbance at 291 nm against percentage amylose (w/w)

iodine-amylopectin complex as a purple colour. The spectra for mixtures of corn amylose and amylopectin with iodine (r = 0.9833).
were very similar to those obtained by dispersion of potato Preparation as in Section 3.2. (b).
amylose and potato amylopectin in 1 mol/L KOH [12]. Dif-
ferences in the absorbance maxima at all wavelengths for
complexes of potato amylose, potato amylopectin, and mix- of amylose/amylopectin mixture) an approximately linear
tures thereof are shown in Table 1. Absorbance maxima for relationship was observed between absorbance at 291 nm
complexes from corn amylose, corn amylopectin, corn and percentage amylose in the mixture, low absorbance
starch, and wheat starch are also shown in Table 1. Spectra values being obtained with high percentages of amylose
for complexes from corn amylose and corn amylopectin are (refer Fig. 4). For this graph, the correlation coefficient was
shown in Fig. 3 and are similar to those produced by the 0.9833.
potato components. The iodine reagent shows absorption maxima at approxi-
mately 288 nm and 352 nm (refer Table 1). The peak at
3.2.1 U.V. absorbance patterns 288–292 nm is dependent upon the percentage of amylose
When amylose and amylopectin are present in the sample, present. When iodine concentrations are low, this peak is
the absorbance at 291 nm is much less than the iodine blank. high with low amounts of amylose but absorbance dimin-
The peaks were greater with the amylopectins from potato ishes as the percentage of amylose increases, together with a
and corn compared with the amylose fractions. In fact, with loss of peak sharpness. At higher concentrations of iodine, as
small amounts of iodine (0.2 mL iodine reagent per 0,004 g would be required for saturation of the amylose molecules,
the peak at 288–292 nm did not seem to be of great value in
estimations of amylose or amylopectin.
A spectrophotometric method has been reported based on
measuring the free iodine by its absorbance at 286 nm [15].
A plot of free iodine versus bound iodine (total iodine – free
iodine) showed an inflection which gave bound iodine on
extrapolation. Differences in iodine binding capacity bet-
ween the essentially linear amylose and the extensively
branched amylopectin may thus be utilised to determine the
composition of a starch. In our hands, the results with this
type of method were not precise and differences in absorb-
ance at 291 nm were not sensitive enough to be of value
when excess iodine was used.
The peak at 349–352 nm is fairly constant over a wide
range of amylose concentrations, suggesting that it is a
characteristic of both the amylose and amylopectin com-
plexes. Hence, this peak does not seem to be of value in the
determination of amylose content.
3.2.2. Visible region absorbance patterns
Although the spectra for potato amylose and potato
Fig. 3. UV-visible spectra of iodine-corn amylose (70 % amylose) and
amylopectin complexes in comparison with the iodine reagent blank
amylopectin were somewhat similar (Fig. 2), the absorbance
(shown from 450–900 nm for clarity). Where A = amylose, AP = maxima in the visible region were quite different, with the
amylopectin, and iodine = iodine reagent blank. Preparation of samples amylose maximum occurring at higher wavelength (refer
as in Section 2.1. Table 1).

Starch/Stärke 50 (1998) Nr. 4, S. 158–163 161

 Table 2. Amylose contents for commercial starches determined by
measurement of the iodine complex at 600 nm*. A corn amylose pre-
paration is also included.

Sample Amylose content (% (w/w), dry basis)+

Potato starch 29.7 ± 10.3
Corn starch 1 24.7 ± 0.4
Corn starch 2 23.1 ± 0.5
De-fatted corn starch 1 29.5 ± 0.5
Wheat starch 22.5
(estimated from corn calibration graph)
De-fatted wheat starch 28.7
(estimated from corn calibration graph)
Refined corn
amylose preparation 80.4 ± 1.6
(* reagents used as in table 2.).
(+ mean ± standard error of mean of three determinations.).

results on the commercial starches are undoubtedly due to

lipid inclusion complexes with amylose.
Fig. 5. Calibration graph of absorbance at 600 nm against percentage The results on the commercial starches are a little higher
amylose (w/w) for potato amylose/amylopectin mixtures showing (3–4 %) than those reported for a similar method using
duplicate results for 0 % and 100 % amylose (r = 0.9994). 0.5 mol/L potassium hydroxide as dispersant [16], which
may be indicative of some degradation in the harsh alkali.
The peak in the visible region at about 600 nm varies from
However, when mixtures of amylose and amylopectin one starch to another and between amylose and amylopectin.
were prepared, the difference in these maxima became less In the case of potato amylose and corn amylose, the diffe-
and made it possible for a single wavelength to be used for rence between maxima for amylose and amylopectin appears
the estimation of amylose. This wavelength could be taken as to be around 80 nm with amylose having the higher absorb-
being 600 nm for potato amylose/amylopectin mixtures. ance maximum. Potato amylose has an absorbance maxi-
Even though the absorbance maximum in the visible portion mum of 636 nm, while corn amylose has a maximum of
of the spectrum is lower than this for potato amylopectin, 604 nm.
this does not appear to alter the linearity of the graph from The amylose and amylopectin complexes have peaks
0–100 % amylose (refer Fig. 5). The correlation coefficient which are extremely broad and thus there is no necessity to
for this graph was 0.9994. use the maximum absorbance for quantification of amylose.
Amylose and amylopectin from corn starch gave spectra Furthermore, the samples containing 25 % potato amylose
as shown in Fig. 3. They were similar to the spectra of corres- and 21 % corn amylose had absorbance maxima in their res-
ponding fractions from potato starch, but with each absor-
bance maxima in the visible region being about 30 nm lower
than those for the corresponding potato starch components
(refer Table 1). Again, the peaks at the shorter wavelengths
were present as described previously. The amylose and amy-
lopectin blanks (without iodine) were very low in absor-
bance over the range 270–900 nm and did not significantly
influence the results.
Plots of absorbance at 600 nm against percentage of
amylose were also linear for corn amylose/amylopectin
mixtures (refer Fig. 6). The graph had a correlation coeffi-
cient of 0.9991.
3.2.3 Tests on starches
Using the calibration graph in Fig. 5, the result for the
amylose content of commercial potato starch, which is
essentially lipid free, was 29.7 %. The results for the two
corn starch samples were 24.7 % and 23.1 %. When the
former corn starch was de-fatted, the result increased to
29.5 % (refer Table 2).
A similar difference was found between commercial
wheat starch and de-fatted wheat starch. No amylose or amy-
lopectin standards were available for wheat starch analysis,
but if the corn starch calibration graph can be taken as a
rough guide, the amylose contents of the wheat starch so Fig. 6. Calibration graph of absorbance at 600 nm against percentage
obtained were 22.5 % for the commercial starch and 28.7 % amylose (w/w) for corn amylose/amylopectin mixtures showing dupli-
for the de-fatted commercial starch (refer Table 2). Lower cate values for all results (r = 0.9991).

162 Starch/Stärke 50 (1998) Nr. 4, S. 158–163

pective mixtures at 592 nm and 594 nm (Table 1), very close In the micro-method, weighing errors would be increased
to those obtained with the natural potato and corn starches. to about 1 % and therefore we can expect a further possible
Thus, a wavelength of 600 nm is suitable for estimation of increase of about 3 % error in the absorbance values and
the amylose content of both starches and takes into account standard errors of about ± 2 %. The overall accuracy of the
the contribution from the amylopectin complex, which method depends considerably on having high precision in
however, is much smaller than that from the amylose com- the preparation of the standard graph, which is indicated by
plex. correlation coefficients of at least 0.999.
3.2.4 Tests on corn amylose preparations
Acknowledgements
The analysis of a purified sample of corn amylose prepar- The authors would like to thank the Australian Food Industry
ed from Sigma corn amylose (70 %) by precipitation with Science Centre (Afisc) for providing the postgraduate research scholar-
1-butanol gave a result of 80.4 % amylose (Table 2). This ship to S. J. McGrane which fundded this work. We would also like to
result was not improved by a second precipitation. When acknowledge Dr. Graham Wills-Johnson of our Department for his
analysing preparations of high amylose contents such as this expertise in producing the molecular model of the pentaiodide-amylose
material, it is suggested that about half the normal amount of inclusions complex illustrated here.
sample be used.
3.3 Advantages of the spectrophotometric method Bibliography
Two important aspects of the method are its versatility
[1] Rundle, R. E., and R. J. Baldwin: J. Am. Chem. Soc. 65 (1943),
and simplicity. It can be used for starches from a wide variety 544–557.
of botanical sources, and requires no special equipment other [2] Reddy, J. M, K. Knox, and M. B. Robin: J. Chem. Phys. 40 (1964),
than a simple spectrophotometer capable of 1082–1089.
measuring absorbance in the vicinity of 600 nm. Samples of [3] Teitelbaum, R. C., S. L. Ruby, and T. J. Marks: J. Am. Chem. Soc.
high and low amylose content may be analysed and require 100 (1978), 3215–3217.
only a change in the volume of the aliquot chosen to give [4] Teitelbaum, R. C., S. L. Ruby, and T. J. Marks: J. Am. Chem. Soc.
optimal results. 102 (1980), 3322–3328.
The sensitivity of the iodine-starch reaction is quite high [5] Bates, F. L., D. French, and R. E. Rundle: J. Am. Chem. Soc. 65
(1943), 142– 148.
and is applicable to amounts of starch containing down to [6] Hovenkamp-Hermelink, J. H. M., J. N. de Vries, P. Adamse, E.
100 mg amylose. As presented in the main method (Section Jacobsen, B. Witholt, and W. J. Feenstra: Potato Research 31
2.1), the amount of amylose taken to prepare the final (1988), 241–246.
mixture is 4 mg, but this amount has been chosen to [7] Magel, E.: Starch/Stärke 43 (1991), 384–387.
minimise weighing and volumetric errors. [8] Adkins, G. K., C. T. Greenwood, and D. J. Hourston: Cereal Chem.
47 (1970), 13– 18.
3.4 Micro method [9] Kurtzman, R. H., F. T. Jones, and G. F. Bailey: Cereal Chem. 50
The following adaptation of the method could be used if (1973), 312– 322.
[10] Wolf, M. J., E. H. Melvin, W. J. Garcia, R. J. Dimler, and W. F.
only small amounts of amylose or other starch products are Kwolek: Cereal Chem. 47 (1970), 437–446.
available: [11] Morrison, W. R., and B. Laignelet: J. Cereal Science 1 (1983),
Weigh 20 mg of amylose or starch-derived product, add 9–20.
DMSO (0.4 mL) and dissolve by heating at 85 °C for 15 min. [12] Jarvis, C. E., and J. R. L. Walker: J. Sci. Food Agric. 63 (1993),
Dilute the mixture to 100.0 mL with de-ionised water. Take a 53–57.
1.00 mL aliquot (º 200 mg) and add 3.00 mL of water follow- [13] Knutson, C. A., and M. J. Grove: Cereal Chem. 71 (1994),
ed by 1.00 mL of 6.5 ´ 10–4 mol/L iodine/1.3 ´ 10–2 mol/L KI 469–471.
reagent. Allow to stand for 10 min and then measure ab- [14] Zobel, H. F.: Starch/Stärke 40 (1988), 44–50.
[15] Karve, M. S., and N. R. Kale: Starch/Stärke 44 (1992), 19–21.
sorbance at the specified wavelength in a 1 cm path length [16] Williams, P. C., F. D. Kuzina, and I. Hlynka: Cereal Chem. 47
cell. (1970), 411–420.
Under these conditions, 200 mg of amylose (11 % mois-
ture) gives an absorbance reading of about 0.5.
3.5 Estimation of errors Addresses of authors: Scott J. McGrane, Hugh J. Cornell, and Colin
J. Rix. Department of Applied Chemistry, RMIT University, GPO Box
In the main method, the weighing of 0.1 g of sample on a 2476V, Melbourne 3001, Australia.
standard analytical balance incurs an error of ± 0.2 %. Use of Mr. Scott J. McGrane (BAppSci(Hons)), corresponding author, Docto-
A-grade volumetric glassware incurs further errors: a 25 mL ral candidate, Department of Applied Chemistry, RMIT University.
volumetric flask (0.1 %), a 1 mL pipette (0.1 %), a 50 mL Private address: 5 Outside Court, Teesdale, Victoria, 3328, Australia.
pipette (0.1 %); giving an uncertainty of about 0.3 % in the Assoc. Prof. Hugh J. Cornell (MSc, Ph.D.), Professorial Fellow,
concentration of the final solution. Department of Applied Chemistry, RMIT University.
Uncertainties such as these contribute to the uncertainty in Private address: 17 Gardenview Court, Templestowe, Victoria, 3106,
the recorded absorbance, which itself has uncertainty. A total Australia.
Dr. Colin J. Rix (BSc(Hons), Ph.D.), Senior Lecturer Inorganic Chemi-
of about 1 % uncertainty in the absorbance would be expec- stry, Department of Applied Chemistry, RMIT University.
ted to be contributed by these factors. Overall, however, tri- Private address: 38 Francis Crescent, Glen Iris, Victoria, 3146, Austra-
plicate results on the starches indicate a range of about ± 2 % lia.
in absorbance values leading to the standard errors of the
mean amylose contents as shown in Table 2. (Received: December 19, 1997).

Starch/Stärke 50 (1998) Nr. 4, S. 158–163 163