Fluorimetric determination of carbohydrates
with 2,3-diaminonaphthalene
Yang
Jinghe, Cao Xihui, Huang Fang, Wu Xia, Zhou Guangjun#
(Department of Chemistry, Lab. Colloid & Interface Chemistry for
Education Ministry, Shandong University, Jinan, 250100, #Shandong Supervision
&Inspection Institute for Product Quality, China)
Received Oct. 21, 1999; Supported by the National Natural Science
Foundation of China and Shandong Province, the Research Laboratory of SEDC
of Analytical Science for Material and Life Chemistry
Abstract A new
simple and sensitive fluorimetric method for the determination of carbohydrates is
described. The method is based on the reaction of carbohydrates with
2,3-diaminonaphthalene (DAN) in the presence of sulfuric acid. Under optimum conditions,
an excellent linear relationship was obtained between the fluorescence intensity and the
concentration of carbohydrates. The detection limit is 1.1×10-8mol/L for
fructose and 9.7×10-8mol/L for galactose and that of other carbohydrates lies
between 1.1×10-8mol/L and 9.7×10-8mol/L.
Keywords carbohydrates, 2,3-diaminonaphthalene, Fluorimetric determination
Carbohydrates are very important in nature
and they take part in almost all of the life processes. They are not only central to
energy generation and storage but also offer us lots of information about other species
and ourselves. So, they are also called as information molecules. The study of
carbohydrates is very important and valuable in chemistry and bioscience.
It is difficult to detect carbohydrates with usual means because they
have no groups that could be detected sensitively. So, it is important to derive the
carbohydrates and let them have some properties that could be detected sensitively in the
determination. Recently, there are many reports about the derivatization of carbohydrates.
The usual derivative reagents are 1,2-di(4-methoxyphenyl)ethylenediamine (DME)[1],
1,2-phenulenediamine (PDM)[2], malonamide[3], guanidine[4],
arginine[5], benzamidine[6], 8-aminonaphthalene-1,3,6-trisulfonic
acid (ANTS)[7], 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate[8]
(AQC) and 2-metthyl-3-oxo-4-phenyl-2,3-dihydrofunan-2-yl acetate[9] (PDFAc),
etc, of which AQC and PDFAc are applied in the determination of the amino sugar. However,
when they are applied to the determination of carbohydrates, they all have some drawbacks
such as long reaction time, strong reaction conditions or low sensitivity and precision.
Our research of finding new derivative reagents emphasizes on following points: a) rapid
derivatization b) mild reaction conditions and c) high sensitivity and precision. The new
derivative reagent 2,3-diaminonaphthalene answers above requirements well. Several
carbohydrates were detected by the method of DAN. The limit of detection reaches 10-8mol/L
level and in the linear range the correlation coefficients are more than 0.99. The
sensitivity of the proposed method is similar to that of the DME method, but 3-10 times
higher than other analogous methods.
1. EXPERIMENTAL
1.1 Apparatus
All fluorescence intensities were measured on a Hitachi Model 850spectrofluorimeter and
all absorption spectra were taken on a Shimadzu UV-240 spectrophotometer.
1.2 Reagents
The 2,3-diaminonaphthalene (DAN) solution was prepared by dissolving 0.995g DAN in 250ml
0.1 mol/L HCl, then treated with 50ml cyclohexane, shaken for about 1min, the was organic
phase discarded, and the remaining aqueous phase treated twice more with cyclohexane
according to the above procedure. The solution was stored in a brown vessel at room
temperature.
Glucose, fructose, galactose, glucosamine hydrochloride and
N-acetylglucosamine were all obtained from Detecting Department of Medicines and
Byproducts of China.Fucose and xylose were purchased from Medicine Company of China. a-Rhamnose was obtained from Chemical
Reagents Company of Beijing.L-sorbose was obtained from Beijin Fangcao medicine and
chemitechnology Research Company.D-arobinose was purchased from Medicine Store of Military
Medical College of China.The carbohydrate solution was prepared with distilled water and
the concentration is 1.00×10-2 mol/L.Sulfuric acid solution is prepared by
solving 75ml sulfuric acid to 25ml distilled water.
All the other reagents were of analytical reagent grade and distilled
deionized water was used throughout.
1.3 Procedure
To a 25ml test-tube were added 0.5ml of glucose standard solution, 2ml of Sulfuric acid
solution and 2ml of DAN. This mixture was shaken and heated in boiling water bath for
150min, diluted to 25ml after cooled to room temperature. The fluorescence intensity of
the system was measured in a 1cm-quart cell with excitation and emission wavelengths of
402 and 480nm, respectively.
2. RESULTS AND DISCUSSION
2.1 Fluorescence Spectra
Glucose H2SO4 have no fluorescence and
DAN, when pH=1,has an emission peak at 400nm and two excitation peak at 292nm and 340nm
respectively. When H2SO4 is added to DAN, the fluorescence was
weakened evidently, and the maximum excitation peak is at 292nm.
The fluorescence spectra of glucose-H2SO4-DAN are
shown in Fig.1. It can be seen that the glucose- H2SO4-DAN system
has two excitation peaks, both at 400 and 415nm respectively, and two emission peaks, at
480 and 498nm, respectively. From Fig. 1, one can also see that because the reagent blank
is very low and stable, the redundant DAN do not affect the determination of carbohydrates
when 400nm is selected as the excitation peak of glucose- H2SO4-DAN
system.
Many carbohydrates can react with DAN and give fluorescent compounds.
The results are shown in table1. From table 1 we can see that except the a-Rhamnose, N-acetylglucosamin and
HCl-glucosamine all the other carbohydrates react with DAN to give fluorescent compounds,
the fluorescence spectra of which are similar to that of glucose- H2SO4-DAN
although the intensity is different.
If |
 |
If |

|
Fig.1 Fluorescence Spectra
(a) excitation spectra (l
=480nm), (b) emission spectra (l=400nm)
1. the system of glucose-H2SO4 , 2. the system of DAN- H2SO4
,3. the system of glucose-DAN- H2SO4 |
Table 1. Fluorescence Spectra of
different carbohydrates
carbohydrates |
maximum
excitation wavelength (nm) |
maximum
emission wavelength (nm) |
relative
intensity (or=10) |
Glucose |
400 |
480 |
35 |
Fructose |
402 |
486 |
49 |
Galactose |
398 |
460 |
45 |
Xylose |
397.5 |
484 |
41 |
Fucose |
404 |
464 |
56 |
a -Rhamnose |
× |
× |
× |
L-Sorbose |
399 |
483 |
24 |
D-Arabinose |
401 |
479 |
34 |
N-acetylglucosamine |
× |
× |
× |
HCl-glucosamine |
× |
× |
× |

|
Fig.2 effect of the concentration of H2SO4 Conditions: glucose: 5.00×10-4mol/L,
DAN: 2.0×10-3mol/L, Heating time: 150min |

|
Fig.3
Effect of the concentration of DANConditions: glucose: 5.00×10-4mol/L, H2SO4: (1:3,v: v) 2ml, Heating time:
150min |
2.2 Factors Affecting the Fluorescence
Intensity
2.2.1 Effect of the concentration of H2SO4
DAN is basic, and does not solve in neutral or basic solution. When pH<2, it reacts
with some carbohydrates and forms fluorescent substance. The concentration of H2SO4
has an important effect on the fluorescence intensity of the carbohydrate- H2SO4-DAN
system and the results are shown in Fig 2. From which, we can see that the fluorescence
intensity increases with the increasing of the concentration of the H2SO4.
The fluorescence intensity of the system reaches the maximum and retains constant when the
volume of H2SO4 solution is 1.5-4ml. In this paper, 2ml of H2SO4
solution were selected.
2.2.2 Effect of the concentration of DAN
The effect of the concentration of DAN was studied and the results were shown in Fig3.
From which one can clearly see that a final concentration of 0.002mol/L gave the maximum
fluorescence under the conditions defined.
2.2.3Effects of reaction temperature and heating time
The effects of different reaction temperature were studied and the results were given
in Fig 4. It shows that under 80°C the fluorescence is very weak and the balance time is
very long, but under 100°C the fluorescence is quite strong and the balance time is very
short. 100°C was selected in this paper. The results also show that the fluorescence
increases with the increasing of the heating time. The fluorescence intensity reaches the
maximum after the system is heated in boiling water bath for 100-150min, in which the
concentration of glucose is 5.0×10-4 mol/L (Fig 4). Other carbohydrates are
similar to glucose. In this paper the heating time is 150min.

|
Fig 4 Effects
of temperature and heating time
Conditions: glucose: 5.00×10-4mol/L, DAN:
2.0×10-3mol/L, H2SO4: (1:3,v: v)
2ml |
Compared with the
analogous derivative reagents, DAN has the properties of mild reaction conditions and
rapid derivatization. It is reported that the reaction temperature is usually between
120°C (1,2-phenylenediamine method) to 160°C (guanidine method 150°C, DME method 150°C
and arginine method 160°C). At the same time this proposed method is more rapid than
other methods, such as the 1,2-phenylenediamine method, which requires180 min.
2.3 The stability
It is discovered that the fluorescence intensity of this system retained stable for
two hours after it reaches the maximum.
3. ANALYTICAL APPLICATION
3.1 Calibration curve and detection limit
Under the optimum conditions defined, a linear relationship was obtained between the
fluorescence intensity and the concentration of carbohydrates such as glucose, fructose,
galactose, xylose, fucose, L-sorbose, D-arabinose. The results are shown in table 2, from
which we understand that in the linear ranges all kinds of carbohydrates above have an
excellent correlation coefficient, and their limits of detection could reach 10-8
mol/L (S/N=2).
Table 2. the linear range and detection limit of different carbohydrates
carbohydrates |
linear
range(mol/l) |
correlation
coefficient |
detection
limit (mol/l) |
Glucose |
7×10-7-6×10-4 |
0.9991 |
9.5×10-8 |
Fructose |
1×10-7-1×10-4 |
0.9993 |
1.1×10-8 |
Galactose |
6×10-7-4×10-4 |
0.9998 |
9.7×10-8 |
Xylose |
4×10-7-1×10-4 |
0.9981 |
9.0×10-8 |
L-Sorbose |
4×10-7-7×10-5 |
0.9975 |
5.8×10-8 |
Fucose |
5×10-7-1×10-4 |
0.9977 |
9.2×10-8 |
D-arabinose |
5×10-7-3×10-4 |
0.9912 |
7.8×10-8 |
3.2 Recovery tests and sample
determination
Tests of the recovery efficiency for known amounts of glucose, fructose and xylose
solutions were made. The results are listed in table 3 which indicate that about 94.8% can
be recovered.
Table 3. Recovery tests
Samples |
Added
(10-5mol/l) |
Found
(10-5mol/l) |
X±S.D. |
Recovery
(%) |
Glucose |
1.00 |
0.95, 0.96,
0.94, 0.94, 0.97 |
0.948±0.013 |
94.8% |
Fructose |
1.00 |
0.94, 0.93,
0.98, 0.95, 0.96 |
0.952±0.019 |
95.2% |
Xylose |
1.00 |
0.93, 0.95,
0.96, 0.95, 0.96 |
0.95±0.012 |
95% |
The proposed method was also used to
determine glucose in glucose injection solution (Jinan Mingshui LiMing Pharmaceutical,
China). The satisfied results are shown in table 4.
Table 4. Sample determination
|
Standard
(%) |
Found
(%) |
X±S.D. |
Glucose
injection solution |
50 |
47.0, 47.3,
47.1
47.7, 48.2 |
47.5±0.49 |
S.D.: standard deviation
REFERENCES
[1] Umegae Y. Anal. Sci., 1989, 5: 371.
[2] Towne J C, Spikner J E. Anal. Chem., 1963, 35 (2): 211.
[3] Honda S, Matsuda Y, Terao M et al. Anal. Chim. Acta, 1979, 108: 421.
[4] Yamauchi S, Nimure N, Kinoshita T et al. Analyst, 1993, 118 (7): 773.
[5] Mikami H, Ishida Y. Bunseki Kagaku, 1983, 32: E207.
[6] Coquet A, Venthey J L, Haerdi W. J. Chromatogr., 1991, 553 (1-2): 255.
[7] Stefansson M, Novotny M. Anal. Chem., 1994, 66: 1134.
[8] Diaz J, Lliberia J L, Comellas L et al. J.Chromatogr., 1996, 719 (1): 171.
[9] Chen P, Novotny M V. Anal. Chem., 1997, 69 (14): 2806.
|