HPLC of Mono- and Disaccharides Using UNIT E1.
2
Refractive Index Detection
The choice of method for sugar analysis is dependent upon the composition of the sample
and the mixture of sugars present. The main technique presented here is used for the
determination of fructose, glucose, sucrose, maltose, and lactose in most foods, feeds,
beverages, and nutritional products (see Basic Protocol). It is an efficient screening
method for quality control and monitoring purposes and is used for fulfilling the nutrition
labeling data requirements of some products. In addition, a protocol for isolating sugars
(i.e., neutral compounds) from acids (e.g., citric acid) using anion-exchange mini-col-
umns is described (see Alternate Protocol 1), along with a procedure for HPLC using
calcium-loaded cation-exchange columns to overcome the difficulties in isolating sorbitol
and galactose from glucose (see Alternate Protocol 2). Careful consideration of the matrix
is required in order to determine which protocol is most appropriate. Because separation
is accomplished using different mechanisms, the complete set of protocols allows analysis
of a broad range of samples. The Basic Protocol exhibits a wide range of applicability,
and the Alternate Protocols expand this range further. Data regarding the sugar composi-
tion of various foods can be found on the U.S. Department of Agriculture Web site,
http://www.nal.usda.gov/fnic/foodcomp.
HPLC OF SUGARS USING AN AMINOPROPYLSILYL COLUMN BASIC
PROTOCOL
Sugars in the sample are extracted, dried, reconstituted in an aqueous acetonitrile solvent,
and injected on a high-performance liquid chromatograph (HPLC) equipped with an
aminopropylsilyl column and refractive index (RI) detector. Otherwise, standard labora-
tory equipment is used in conduct of this method. Development of this protocol was based,
in part, on previous research conducted by Mason and Slover (1971) as well as methods
developed by Brobst (1972) and Donnelly (1973). Work on carbohydrate analysis con-
ducted by Brower (1966) was also considered. The aminopropylsilyl column is used due
to its excellent sensitivity and ability to detect individual sugars at low concentration.
Disadvantages include an incomplete separation of sorbitol and mannitol from glucose
along with incomplete separation of galactose from glucose in some products (see
Alternate Protocol 2). Maltitol interferes with maltose and lactose while inositol interferes
with sucrose.
Due to interference, samples that are high in salt (especially sodium chloride) or samples
with a high free amino acid content may be difficult to analyze using this method. In those
cases, validated gas chromatography (GC) methods (Mason and Slover, 1971) are more
applicable. In addition, samples containing sorbitol should be analyzed by GC or using
a calcium-loaded anion-exchange column (see Alternate Protocol 2).
Materials
Sample or samples
0.28 M copper sulfate solution: store in a low actinic bottle up to 1 month at room
temperature
1 N hydrochloric acid
1 N or 50% (w/v) sodium hydroxide
Acid-washed celite (J.T. Baker)
Nitrogen gas
Injection solvent: 50% (v/v) HPLC-grade acetonitrile in water
Reference standard solutions (see recipe)
Mobile phase (see recipe): 7:3 (v/v) acetonitrile/water Mono- and
Oligosaccharides
Contributed by Wayne Ellefson E1.2.1
Current Protocols in Food Analytical Chemistry (2002) E1.2.1-E1.2.9
Copyright © 2002 by John Wiley & Sons, Inc. Supplement 6
200-ml volumetric flask
Whatman 2V filter paper
5-oz plastic cup and cap
4-dram vials with Teflon-lined screw caps
50°C heating block
2-ml injection vials
Syringe with 0.20-µm PTFE or nylon filter
High-performance liquid chromatograph (HPLC; Perkin-Elmer) with:
Isocratic LC pump 250
200 series refractive index (RI) detector
Advanced LC sample processor 1SS200
Analytical column: 250 × 4.6–mm aminopropylsilyl column (Supelco or
equivalent)
Guard column: 10-cm aminopropylsilyl column (Supelco or equivalent;
optional)
Prepare sample
1. Place 2 to 7 g of each sample in a separate 200-ml beaker. Add 40 ml deionized water
and a magnetic stir bar to each. Place on a stirring plate and stir ∼1 hr.
A larger sample size is required for samples with low sugar content. A smaller sample size
is required for samples with high sugar content in order to avoid solubility problems.
2. Add 10 ml of 0.28 M copper sulfate while stirring.
3. Adjust sample pH to 6.4 using 1 N hydrochloric acid or 1 N or 50% sodium hydroxide
and a pH meter.
Excess chloride can cause interference. Therefore, when titrating with hydrochloric acid,
it is important to not overshoot the 6.4 pH.
4. Carefully transfer the sample to a 200-ml volumetric flask. Bring to volume with
deionized water and mix well.
5. Filter the sample through Whatman 2V filter paper overlaid with ∼0.5 g acid-washed
celite (to aid filtration) into a 5-oz plastic cup with cap.
6. Pipet appropriate aliquots (8.0 and 1.0 ml) from the filtrate into separate 4-dram vials.
Dry aliquots down under nitrogen gas for ∼8 hr in a 50°C heating block.
The 8-ml aliquot corresponds to the assay’s lowest confidence level (<0.1%). The 1-ml
aliquot is used to quantitate any of the sugars that are present in amounts greater than 1%.
Adjustments in initial weights, volumes, or dilutions can be made in order to bring the
sugar response on scale.
7. Add 5.0 ml injection solvent (50% acetonitrile) to each 4-dram vial and tighten the
Teflon-lined screw cap. Vortex the sample vials every 10 to 15 min until no residue
is found on the wall of the vials (∼1 hr).
8. Filter into a 2-ml injection vial using a syringe and 0.2-µm PTFE or nylon filter.
The clear solution is ready to be analyzed against reference standards using HPLC.
Prepare calibration standards
9. For each reference sugar, prepare a set of calibration standards in duplicate using the
HPLC of stock and working reference standard solutions.
Mono- and
Disaccharides This gives five sets of calibration standards (i.e., fructose, glucose, sucrose, lactose, and
Using Refractive maltose) each at concentrations of 0.05, 0.1, 0.3, 0.5, and 1.0 mg/ml.
Index Detection
E1.2.2
Supplement 6 Current Protocols in Food Analytical Chemistry
130.00
125.00
120.00
115.00
110.00
105.00
100.00
95.00
90.00
85.00
80.00
75.00
70.00
65.00
60.00
mV
fructose
55.00
glucose
sucrose
50.00
45.00
40.00
maltose
lactose
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
−5.00
−10.00
−15.00
5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00
Minutes
Figure E1.2.1 Chromatogram of a sugar standard analyzed as described (see Basic Protocol).
Perform chromatography
10. Analyze calibration standards and samples by HPLC with refractive index detection
using the following conditions:
Mobile phase: 7:3 (v/v) acetonitrile/water
Flow rate: 1.0 ml/min
Column temperature: ambient
Elution mode: isocratic
Run time: 25 min
Injection volume: 75 µl.
A run is typically composed to 55 to 60 injections, including replicate samples (aliquots in
step 7), standards, and a minimum of 10% quality assurance samples, including duplicate
analyses, validated control samples, or recoveries (see Critical Parameters).
Newly prepared standards should be checked against previously qualified standards every
six months. New standards are acceptable if the difference between standards is ±3.0% of
the established values. Sample chromatograms are given for a sugar standard (Fig. E1.2.1),
presweetened cereal (Fig. E1.2.2), ice cream (Fig. E1.2.3), and a meal replacement bar
(Fig. E1.2.4).
The aminopropylsilyl guard column is required for complex matrices with low sugar
content. Simple matrices with high sugar content do not require the guard column. Other
parameters may need adjustment in order to optimize the chromatography. Mono- and
Oligosaccharides
E1.2.3
Current Protocols in Food Analytical Chemistry Supplement 6
56.00
54.00
52.00
50.00
48.00
sucrose
46.00
44.00
42.00
40.00
38.00
36.00
34.00
32.00
30.00
28.00
26.00
mV
24.00
22.00
20.00
18.00
16.00
14.00
12.00
10.00
8.00
fructose
6.00
glucose
4.00
2.00
0.00
−2.00
−4.00
−6.00
−8.00
5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00
Minutes
Figure E1.2.2 Chromatogram of a presweetened cereal analyzed as described (see Basic
Protocol).
Analyze data
11. Generate five calibration curves (i.e., each sugar standard) by plotting a first-order
curve of integrated peak area versus concentration (mg/ml) and performing linear
regression analysis.
The correlation coefficient (R value) should not be <0.999.
12. Determine the concentration (c in mg/ml) of each sugar in the sample by extrapolating
from the appropriate calibration curve as a function of area response.
13. Determine the weight percent of each sugar in the sample using the equation:
c × V × D × 100
% sugar =
w × 1000 mg/g
where V is the initial volume, D is the dilution factor (can vary according to the
amount of each sugar that is present), and w is weight of the sample in grams (step
1).
14. Compare sample results against expectations and compare control sample results
against the acceptable range (see Critical Parameters). Evaluate replicate results and
HPLC of
Mono- and recoveries for acceptability. Expectations are based upon historical data with a
Disaccharides specific matrix, standard references, or expected results (e.g., claims). Acceptable
Using Refractive
Index Detection ranges are determined during method validation.
E1.2.4
Supplement 6 Current Protocols in Food Analytical Chemistry
360.00
sucrose
340.00
320.00
300.00
280.00
260.00
240.00
220.00
200.00
180.00
mV
160.00
140.00
lactose
120.00
100.00
80.00
60.00
glucose
40.00
maltose
20.00
0.00
−20.00
5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00
Minutes
Figure E1.2.3 Chromatogram of an ice cream analyzed as described (see Basic Protocol).
If any of these do not match established acceptability criteria, the entire analytical run
must be investigated. Acceptability criteria will change depending upon the sample matrix
and the level of sugar anticipated. Based upon the investigation, another round of analysis
may be necessary. If analysis is undertaken again, multiple replicates should be analyzed,
taking into account the facts uncovered in the investigation.
SAMPLE PREPARATION USING ANION-EXCHANGE MINI-COLUMNS ALTERNATE
PROTOCOL 1
Sample matrices with high levels of citric or other organic acids (e.g., cranberry juice)
can result in poor resolution. This method of sample preparation isolates sugars (i.e.,
neutral compounds) from acids using anion-exchange mini-columns. Development of this
protocol is based on research conducted by Hong and Wrolstad (1986). Although the
procedure does require additional time (i.e., twelve samples per hour), it does result in
improved resolution and a more stable baseline.
Additional Materials (also see Basic Protocol)
Juice sample
7 mg/ml mannitol internal standard (Mallinckrodt)
Activated Sep-Pak C18 cartridge (Waters)
10-ml polyprep column containing 1.5 ml hydrated BioRex 5 anion-exchange
resin (Bio-Rad)
Mono- and
0.45-µm Millipore filter (type HA) Oligosaccharides
E1.2.5
Current Protocols in Food Analytical Chemistry Supplement 6
480.00
460.00
440.00
420.00
400.00
380.00
360.00
340.00
320.00
glucose
300.00
280.00
fructose
260.00
240.00
mV
220.00
200.00
sucrose
180.00
160.00
140.00
120.00
100.00
maltose
80.00
60.00
lactose
40.00
20.00
0.00
−20.00
5.00 6.00 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00
Minutes
Figure E1.2.4 Chromatogram of a meal replacement bar analyzed as described (see Basic
Protocol).
1. Mix 5 ml juice sample with 5 ml of 7 mg/ml mannitol internal standard solution and
pass through an activated Sep-Pak C18 cartridge.
2. Discard the first 2 to 3 ml and apply the remaining eluate to a 10-ml polyprep column
containing 1.5 ml hydrated BioRex 5 anion-exchange resin.
3. Discard the first 2 to 3 ml, collect remaining eluate, and filter through a 0.45-µm
Millipore filter (type HA).
4. Inject 50 µl into the HPLC system. Include standards, prepare calibration curves, and
analyze data (see Basic Protocol).
ALTERNATE HPLC OF MONO-AND OLIGOSACCHARIDES IN FRUIT JUICES USING A
PROTOCOL 2 CALCIUM-LOADED CATION-EXCHANGE COLUMN
A disadvantage of the amino-bonded column (see Basic Protocol) is its limitation with
samples containing small amounts of sorbitol in the presence of large quantities of glucose
and the incomplete separation of both sorbitol and galactose from glucose. As an
alternative, the use of an HPLC system with a calcium-loaded cation-exchange column
HPLC of can alleviate this constraint. Development of this protocol is based on research conducted
Mono- and
Disaccharides by Durst et al. (1995). This application has been found to be useful for the determination
Using Refractive of glucose, fructose, sorbitol, and sucrose in processed fruit products, and is commonly
Index Detection
E1.2.6
Supplement 6 Current Protocols in Food Analytical Chemistry
used for analysis of fruit juices. The mechanism of separation is based on the strength of
the bonding between cis-glycols of the sugar molecules with the Ca2+ loaded on the
column. The elution order is related to the number and strength of cis-glycol complexes
formed. The procedure is performed using a 7.8 × 300–mm Aminex HPX 87C column
fitted with a 4.6 × 40–mm carbo-c Micro-Guard precolumn (both from Bio-Rad) and is
identical to the Basic Protocol, except that the mobile phase is 0.2 mg/ml Ca(NO3)2, the
flow rate is 0.7 ml/min, and the column temperature is 85°C.
REAGENTS AND SOLUTIONS
Use Milli-Q-purified water in all recipes and protocol steps. For common stock solutions, see APPENDIX
2A; for suppliers, see SUPPLIERS APPENDIX.
Mobile phase
Using a graduated cylinder, measure 300 ml deionized water into a vacuum flask.
Add 700 ml HPLC-grade acetonitrile to the flask. Mix solution with stir bar and
magnetic stir plate. Degas the mobile phase daily in an ultrasonic bath under vacuum
or using an in-line degasser. Prepare fresh for each analytical run.
Reference standard solutions
1.0 mg/ml stock solution:
Dry ACS-grade sugars (fructose, glucose, sucrose, lactose, and maltose; Sigma) 6
hr in a 60° ± 3°C vacuum oven. Weigh 0.2500 g of each sugar, adjusting the weight
according to standard’s purity (as supplied by the manufacturer) and the water
present in the crystals. Transfer quantitatively into separate 250-ml volumetric
flasks. Add 125 ml ultrapure water (e.g., Milli-Q-purified) to the flasks and com-
pletely dissolve the sugars. Bring the standard solutions up to volume with acetoni-
trile and mix thoroughly (final 50% acetonitrile). Store up to 6 months at room
temperature.
0.05, 0.1, 0.3, and 0.5 mg/ml working solutions:
Pipet 5.0, 10.0, 30.0, and 50.0 ml of each 1.0 mg/ml stock solution into four different
100-ml volumetric flasks. Dilute to volume with injection solvent (50% acetonitrile)
and mix well. Store up to 6 months at room temperature.
COMMENTARY
Background Information tion of water and sugar will refract more than
The analytical methodology for the determi- a light beam passing through water alone.
nation of sugars has improved dramatically and When an HPLC detector is set up so that a light
is continually evolving. The traditional wet- beam passes through a cell with pure mobile
chemistry and enzymatic assays that detected phase and a flow cell with chromatographically
classes of sugars (e.g., reducing sugars) have separated analytes moving through it, the dif-
largely been replaced by chromatographic ference between the respective refractions can
methods utilizing both gas-liquid (GLC) and be measured and expressed as an electronic
high-performance liquid (HPLC) chromatog- response. This detection method provides ex-
raphy. These methods offer high specificity and cellent sensitivity and the ability to achieve
the ability to detect several sugars at the same better precision at low levels of sugar content.
time, and have been actively developed over the UV detection is not commonly used for sugar
past two decades. The HPLC technique is the analysis due to the greater potential for inter-
official AOAC International method for routine ference in the wavelength range required.
sugar analysis (AOAC, 1995). The most current advance in sugar analysis
Refractive index (RI) detection is based on is the use of high-performance anion-exchange
the principle that a light beam is refracted chromatography (HPAEC). This method, more
differently depending on the substance through commonly known as ion chromatography, is
Mono- and
which passes. A beam passing through a solu- used for routine sugar profiles as well as a Oligosaccharides
E1.2.7
Current Protocols in Food Analytical Chemistry Supplement 6
variety of unique sugar or sugar alcohol ana- a requisite uniformity. The various HPLC col-
lytical situations. Ion chromatography tends to umns typically last for at least 500 injections,
use different mobile phases than HPLC with a sometimes as many as 1000 injections. Schiff’s
calcium-loaded column. In addition, a different base formation with sugars does slightly limit
detection system (pulse amperometric detec- the life of the aminopropylsilyl column. Many
tion) is commonly used with ion chromatogra- column manufacturers list column regeneration
phy. This provides another tool for use with procedures. These procedures may extend the
difficult samples. column life somewhat.
Critical Parameters Troubleshooting
Control samples are used as a quality control The HPLC sugar assay was developed and
tool. These samples have a certified or validated validated as an efficient screening technique.
level of the analyte of interest. Inserting a con- As a result, some limitations may be encoun-
trol sample into each analytical run provides a tered. Samples high in salt content may create
tool by which to judge the overall acceptability difficulties, particularly interfering with glu-
of the run. Appropriate control samples must cose or sucrose. In some cases, some of the salt
be developed prior to conducting any sample can be removed with an HPLC clean-up column
analysis. National Institute for Standards and or the use of a column different from the one
Technology (NIST) Standard Reference Mate- specified (e.g., anion-exchange or calcium-
rials (SRMs) may be used in the study or to loaded column). The preferred solution is to
assist in the validation of in-house control sam- analyze such samples using GC instead of
ples. In-house control samples may include HPLC.
items such as fortified breakfast cereal, bever- Fructose and glucose elute close to one an-
age products, milk powder, candy, or other other, which, depending upon the amounts pre-
shelf-stable foods that contain sugar. Prepara- sent, may cause chromatographic overlap in
tion of the control sample includes grinding and some matrices. HPLC chromatographic peaks
subsequent analysis of the sugars present. It is for the five common sugars are not always as
prudent to develop both high- and low-level sharp as is desirable. This can lead to different
control samples encompassing all the sugars results amongst scientists, depending upon how
included in the profile. The control range the baseline is drawn. In some cases, an adjust-
should be established with a minimum of six- ment of the chromatographic parameters (e.g.,
teen replicate analyses conducted over multiple mobile phase) is necessary.
days. Spike recoveries should be conducted at High levels of citric acid may cause prob-
0.5×, 1×, and either 1.5× or 2× of the mean lems with fructose or glucose. In these cases,
(depending upon the level of sugar). The data the use of anion-exchange minicolumns (Alter-
are often verified by the use of another method nate Protocol 1) or GC can be used to overcome
(e.g., Alternate Protocols). Statistics are applied this problem. For samples that are high in lipid
to determine the acceptable range for results. content (e.g., dairy products) and are thus dif-
The solubility and homogeneity of the sam- ficult to dissolve, use warm water for disper-
ple are key to accurate analysis. This solubility sion.
can be affected by other ingredients in the
product to be analyzed. Certain situations may Anticipated Results
require use of a much smaller sample size to Typical retention times for the common sug-
account for solubility issues. Proper steps must ars are shown in Table E1.2.1. These times may
be taken to grind or homogenize the sample to vary in laboratory applications. Reproducibil-
Table E1.2.1 Typical Retention Times of Common Sugars on Supelco Columns
Retention time (min)
Sugar
Amino column Calcium column
Fructose 8.3 14.9
Glucose 9.8 12.0
HPLC of Sucrose 14.0 9.8
Mono- and
Disaccharides Maltose 17.4 9.8
Using Refractive Lactose 19.5 10.2
Index Detection
E1.2.8
Supplement 6 Current Protocols in Food Analytical Chemistry
ity among duplicates should be high on prod- sucrose, and maltose in presweetened cereal,
ucts with significant sugar levels. In matrices chapter 32. In Official Methods of Analysis, 16th
Ed., p. 30. AOAC International, Arlington, Va.
containing very low levels of sugar, the relative
reproducibility (%RSD) will not be as good. Brobst, K. 1972. Gas liquid chromatography of
trimethylsilyl derivations. In Methods in Carbo-
This method is intended for routine sugar
hydrate Chemistry, Vol. 6, pp. 3-8. Academic
analysis. It is not intended for products contain- Press, New York.
ing less common sugars, or for sugar alcohol Brower, H.E. 1966. Gas chromatographic sugar
analysis. These products are better analyzed analysis in hydrolyzates of wood constituents.
using gas chromatography. Anal. Chem. 38:362.
Donnelly, B.J. 1973. The carbohydrate composition
Time Considerations of corn cob hemicellulose. J. Am. Assoc. Cereal
A typical run of ∼20 samples requires 2 days Chemists 50:548.
to complete the analysis process. Sample Durst, R.W., Wrolstad, R.E., and Krueger, D.A.
preparation (i.e., weighing, mixing, adjusting 1995. Sugar, nonvolatile acid, 13C/12C ratio and
mineral analysis for determination of authentic-
pH, adjusting volume, and aliquotting) takes ity and quality of red raspberry juice composi-
∼4 hr. Two or three aliquots are taken from each tion. J. AOAC Int. 78:1195-1204.
sample and each aliquot is injected once. The Hong, V. and Wrolstad, R.E. 1986. Cranberry juice
number of aliquots needed is dependent upon composition. J. AOAC. Int. 69:199-207.
the percentage of individual sugars (i.e., 0.1% Mason, B.S. and Slover, H.T. 1971. A gas chroma-
to 1.8%, 1.0% to 20%, and >20%) contained in tographic method for the determination of sugars
the sample. For example, the higher the differ- in foods. J. Agric. Food Chem. 19:551-554.
ential between the individual sugars, the more
aliquots are required. Aliquots should be al- Key References
lowed to dry for ∼8 hr. After drying, at least 1 AOAC, 1995. See above.
hr is needed to reconstitute them in injection The AOAC methods are considered the official as-
says to be used for nutrition labeling in the United
solvent. The run is composed of 55 to 60 injec- States. The methods are adopted only after a rigor-
tions including the standards, and each injec- ous collaborative study.
tion takes 17 to 25 min. The total run time is
∼20 hr.
Contributed by Wayne Ellefson
Literature Cited Covance Laboratories
AOAC (Association of Official Analytical Chem- Madison, Wisconsin
ists). 1995. Method 982.14: Glucose, fructose,
The author gratefully thanks Dave Levin, Rick Crowley, and Mike Donart for
their insights and assistance.
Mono- and
Oligosaccharides
E1.2.9
Current Protocols in Food Analytical Chemistry Supplement 6
