Food Technology and Nutrition Dept. 62-1
October 25, 1961
A Rapid Spectrophotometric Method of High Sensitivity for the
Determination of Sugar Constituents.
A. M. Furuholmen and R. A. Dennison
The procedure described in this report is based on the oxidation
of reducing sugars by highly-buffered alkaline ferricyanide. The
reduced form, ferrocyanide, reacts with arsenomolybdate forming a
blue-green complex which is measured spectrophotometrically at the
wavelength 745 mu and related through standard curves, to the
concentration of sugar present.
Glucose and fructose may be determined simultaneously through
differences in their oxidation rates at 100C and 55C under the
conditions used. Ten minutes at the higher temperature is sufficient
to oxidize both sugars completely. A 30-minute heating period at
the lower temperature is sufficient for complete oxidation of
fructose. During this time, however, only a small but reproducible
fraction of the glucose is oxidized. Thus the actual concentration
of each sugar can be calculated from the resulting absorptivity
coefficients and absorbance measured at the two temperatures. Total
sugars can be determined as the difference in reducing sugar content
before and after inversion.
The basic reactions were developed by S. V. Ting (1) into a
method for analysis of citrus juices. Ting's method involved the use
of a Lumetron colorimeter with a 515 mu filter. In our investigations
of the possible applications of this procedure, particularly to
potato extract, further modification was made and sensitivity -gatly
increased by measurement with a spectrophotometer at mu which.
the center of the absorption maximum of the complex. ;bLai; 'I
determined, through comparative analyses, that the extract can be used
directly without clarification or removal of alcohol.
Reagents are prepared as recommended by S. V. Ting (1):
Alkaline Ferricyanide Solution. Dissolve 160 grams of anhydrous
sodium carbonate and 150 grams of disodium phosphate heptahydrate in
850 ml. of distilled water, add 4 grams of potassium ferricyanide, and
dilute to 1 liter.
Arsenomolybdate Solution. Dissolve 25 grams of ammonium molybdate
tetrahydrate in 450 ml. of distilled water. Add 21 ml. of concentrated
sulfuric acid, followed by 3 grams of disodium arsenate in 25 ml. of
distilled water. Heat at 55C for 30 minutes in a water bath with
constant stirring, or in an incubator maintained at 37C for 24 to 48
Sulfuric Acid Solution, 2N. Dilute 56 ml. of concentrated
sulfuric acid (specific gravity 1.84) to 1 liter.
Sodium hydroxide solutions, O1N and 1N.
Hydrochloric acid, 1 to 1 by volume.
The sample for analysis must contain from 0.01 to 0,05% reducing
sugar. Thus, in some cases, dilution will be necessary. A blank
determination is run with the samples, one at each temperature, and
used for setting 0.0 absorbance.
Total Reducing Sugars
Pipet 1 ml aliquot into 100 ml volumetric flask. Add, by pipet,
5 ml ferricyanide solution. Mix well and place in boiling water bath,
with solution below water level during entire heating period. Heat
for exactly 10 minutes, remove and cool quickly and completely by
immersing in cold water. When solution has cooled, add 10 ml 2N
sulfuric acid and mix carefully until no more gas is evolved. Pipet
4 ml arsenomolybdate into flask, mix, and dilute to volume. Measure
absorbance, using Beckman DU (or equivalent spectrophotometer) set at
745 mu. Measurement should be made after one-half hour and before two
The procedure is the same as for total reducing sugars except
that water bath is maintained at 55C and sample heated for 30 minutes.
Sugars by inversion
Pipet 50 ml aliquot into beaker. Add 10 ml 1:1 HCI and allow to
stand for 18 hours (may stand for 24 hours). Neutralize by adding 5
ml of 1ON NaOH and additional lN NaOH to bring sample to pH range of
5-7 (a pH meter is recommended). Quantitatively transfer to 100 ml
volumetric flask and dilute to volume. Procedure is then the same as
for total reducing sugar, again using a 1 ml aliquot. Other methods
of inversion should give equivalent results.
Standard curves are run for glucose and fructose at 100C and
55C by regular procedure. The concentrations used are 0.01, 0.02,
0.03, 0.04 and 0.05 gm/100 ml.
From valuesobtained by analysis of standard solutions, the
average K values are calculated, where K is the reciprocal of the
absorptivity coefficient, as, from Beer's Law: As = asbC
As = measured absorbance
b = cell width, cm
C = concentration
C = A- =K As where 1 cm cell used,
asb b C = KAs
1. total reducing sugar
Rt = KrAsD Rt = concentration total
reducing sugar (gm/100 ml)
Kr = K value at 1000C
D = dilution factor
2. fructose and glucose (550C)
apparent fructose = Fa = KfAsD Kf = K value of
fructose at 55C
glucose = G = (Rt-Fa) Q Q = KG/KF as obtained at 550C
KG = K value of glucose at 55C
fructose = F = (Rt-G)
3. sugar after inversion
Calculated as the difference in total reducing sugar
before and after hydrolysis x 0.95.
A study was made of the complex spectra, using a Beckman DK-2
recording spectrophotometer in the visible region (340-900 mu). The
spectra showed an absorption maximum in the region of 740-750 mu.
With the instrument set for the scale of 0-1.0 absorbance units, the
peak was on scale for sugar concentration of approximately 0.05% and
lower. High sensitivity is obtained by measurement at 745 mu as shown
by an absorbance value of 0.37 for 0.02% solution and 0.77 for 0.04%.
Spectra were run of the complex formed by standard glucose and fructose
solutions and by potato extract, at 100o and 550C. The extract
paralled the standard solutions in the region above 500 mu; however,
divergence occurred in the region toward the uv with increased
absorption by the extract. In considering the application of this
procedure to other systems where there are natural components present
with strong absorption near 745 mu, a check of the complex spectra
should be made from 500-800 mu.
Several types of spectrophotometers were used to determine their
accuracy in measurement of absorbancy. To obtain reproducible results
in the range of 0.01 to 0.05%, it is recommended that an instrument
such as Beckman Model B or its equivalent be considered minimum
adequate specifications for selection and that a maximum cell size of
1 cm be used. A Beckman DU was selected for the current work with
A study of the variation of absorbance with time was made. There
was no shift in the peak during the 24 hours of study, but absorbance
decreased after two hours. Based on data obtained, then, the extract
should be measured between one-half hour and two hours after reaction
with arsenomolybdate. It is to be expected that the stability of the
complex will vary according to the system; therefore a check should be
made to be assured of reproducible results in the two hour period.
Reproducibility by this analytical procedure is extremely good.
Six replicates had a standard deviation of 0.0015, which is the same
range of accuracy as the reading of the instrument.
Comparison with the Shaffer-Somogyi method showed good agreement
between procedures, and recovery determined by use of model systems
showed an average recovery of 98.3%, the major loss (2-3%) being at
the upper limit of concentration.
Several substances were investigated by Ting (1) to determine
their effect on the values obtained. He found that ascorbic acid will
interfere but only if present in a concentration of 100 mg/100 ml or
higher. Other substances checked but which did not develop color
include naringin, hesperidin, citric acid and several amino acids.
Polyphenolic compounds were included in our investigation. It
was found, as expected, that they are oxidized by ferricyanide under
the conditions of the reaction and thus would constitute an interference
if present in sufficient quantity (approximately 50 mg/100 ml).
However, analysis of the extract by chromatography and by the
colorimetric method of Arnow (2) gave no indication of the presence
of these substances in the alcohol extract.
In summation, the procedure described will provide an accurate,
rapid method for determining quantitatively sugar constituents as
glucose, fructose, total reducing sugars, and hydrolyzable sugar.
No clarification was necessary and alcohol extracts may be used
directly. The concentration range is 10-50 mg/100 ml. The method
is applicable where the above mentioned sugars are the major reducing
sugar components and where dilution may be made to yield samples in
the desired range. Very small differences may be determined because
of the high sensitivity obtained.
1. Ting, S. V.; Rapid Colorimetric Methods for Simultaneous
Determination of Total Reducing Sugars and Fructose in
Citrus Juices; J. Agr. Food Chem. 4, 263 (1956).
2. Arnow, L. E.; Colorimetric Determination of the Components
of 3,4-Dihydroxyphenylalanine-Tyrosine Mixtures; J. Biol.
Chem. 118, 531 (1937).