Vol.4, No.9B, 89-96 (2013) Agricultu ral Sciences
http://dx.doi.org/10.4236/as.2013.49B016
Copyright © 2013 SciRes. OPEN A CCESS
Certification and uncertainty evaluation of flavonoids
certified reference materials
Can Quan*, Huanfang Yao, Caixia Hou
Division of Chemistry, National Institute of Metrology, Beijing 100013, China;
*Corresponding Author: superfluidcan@hotmail.com
Received August 2013
ABSTRACT
Flavonoids are the most widely used in the field
of the food, medicine, and cosmetic due to their
pharmacological functions and biological activi-
ties, such as anti-tumor, anti-inflammatory and
antioxidation. This work described the assign-
ment of purity values to 3 flavonoids certified
reference materials (CRMs) including baicalein,
5,7-dihydoxy flavones and naringenine which
we re developed in this study according to the
ISO Guides 34 and 35. The qualitative analysis
was performed by liquid chromatography-mass
spectrometry (LC-MS) and infrared spectrosco-
py (IR). The CRMs’ purity values were assigned
based on the weighted average of quantitative
nuclear magnetic resonance method and mass
balance approach with high resolution liquid
chromatography-VWD. All the three CRMs with
following value mass fractions: baicalein at a
certified purity P ± U (k = 2) of 98.8% ± 0.8%;
5,7-dihydoxy flavones of 99.1% ± 0.7% and na-
ringenine of 99.5% ± 1.0% respectively. The
homogeneity of the CRMs was determined by an
in-house validated liquid chromatographic me-
thod. Potential degradation during storage was
also investigated and a shelf-life based on this
value was established.
Keywords: Flavonoids; Certified Refe rence
Material (CRM ); Quantitative Nucl ear Magnetic
Resonance (QNM R); M ass balance approa ch;
Uncertainty
1. INTRODUCTION
A certified reference material (CRM) is a material or
substance whose one or more property values are suffi-
ciently homogeneous, stable, and well established to be
used for the calibration of an apparatus, the assessment
of a measurement method, or for assigning values to ma-
terials [1].
Flavonoids are increasingly in demand in the field of
the food, medicine, and cosmetic [2] due to their phar-
macological functions and biological activities, such as
anti-tumor, anti-inflammatory and antioxidation [3 -5].
The qualities of flavonoids products will have analytical
challenges. The flavonoids CRMs have been determined
by at least two independent analytical methods. The fla-
vonoids CRMs could be used to ensure the accuracy,
traceability and comparability of the related products test
results.
This paper describes the development of three flavo-
noids CRMs including baicalein, 5,7-dihydoxy flavones
and naringenine. Qualitative analysis was carried out by
infrared spectrometry (IR) and liquid chromatography-
mass spectrometry (LC-MS). The moisture content and
inorganic e lements conten t were de termined by Kar l Fisch-
er titration, inductively coupled plasma mass spectrome-
try (ICP-MS), respectively. Homogeneity study was eva-
luated by F-test and t-test both for between vials and
within vial. The uncertainties of CRMs were evaluated
extensively.
2. EXPERIMENTAL
2.1. Materials
The chemical structure of flavonoid compounds stu-
died including baicalein, 5,7-dihydoxy flavones and na-
ringenine were shown in Figu re 1, all were purchased
from Sigma-Aldrich (USA). Benzoic acid certified ref-
erence material (99.9% ± 0.1% purity) used as QNMR
internal standard was obtained from NIST, China.
HPLC grade Acetonitrile and methanol were bought
from Merck (HPLC, Germany). Dimethyl sulfoxide-d6
solvent was purchased from Cambridge Laboratoties
(USA). Distilled deionized water (18.2 MΩcm) was ob-
tained with Milli-Q water system and filtered through
0.22 μm membrane.
2.2. Instrument
Infrared spectr a were collected w ith FT-IR spectrometer
C. Quan et al. / Agricultural Sciences 4 (2013) 89-96
Copyright © 2013 SciRes. OPEN A CCESS
90
Figure 1. Structure of the studied flavonoids CRMs.
Thermo Nicolet IS10 spectrophotometers. All the flavo-
noids reference material solutions, diluted to concentra-
tion of 10 μgmL1 in methanol for LC/MS Analysis
were quantified at flow injection analysis mode with an
Agilent 6410 triple quad system with electro-spray ioni-
zation source (ESI). The QNMR measurements were
performed with a Bruker AV600 spectrometer (Bruker,
Billerica, MA). The Agilent 7500CE ICP-MS was used
for the analysis of trace metal residues in all the samples.
A Mettler Tolerdo DL39 Karl Fischer titrator was used
for water measurement of all the samples.
2.3. Qualitative Analysis Methods
2.3.1. LC-MS Method
The flavonoids CRMs in methanol with concentration
of 10 μgmL1 were used to LC-MS analysis. The ion
spray voltage was set at 4.0 kV. Drying gas temperature
was maintained at 350˚C with flow rate of 9.0 mLmin1.
Nebulizer pressure was set at 40 psi. The fragmentor
energy was set at 135 V. The collision energy respectiv e-
ly were optimized at 30 V for baicalein and 5,7-dihydoxy
flavones, and 16 V for naringenine. The analysis the
mass spectrum peak molecular weight in an ESI positive
ion mode at specified MS scan range of 60 - 300.
2.3.2. IR Method
Infrared spectra were collected with FT-IR spectrome-
ter Thermo Nicolet IS10 spectrophotometers via KBr
pellets technique. Specimens were ground with dried
spectroscopic grade KBr powder and the mixture was
compressed to pellets for FT-IR measurements. The
sample to KBr mass ratio was optimized as 1:100. All the
spectra were collected in the 4000 - 400 cm1 range at 4
cm1 resolution.
2.4. Quantitative Analysis Methods
2.4.1. Mass Balance A pproach
The flavonoids were quantified by HPLC equipped
with an ZORBAX Eclipse XDB-C18(460 mm × 250 mm
× 5 μm, Agilent ) column with VWD detector. The
chromatography conditions are detailed in Ta b l e 1 . The
purity values of flavonoids CRMs were calculated using
the peak area normalization method. Following subtract
the moisture and inorganic elements impurities which not
measured by HPLC from 100%, as follows [6,7]:
( )
m aHPLC
Purity1 PPP=−− ∗
(1)
Where, PHPLC is the percentage area of main com-
pound acquired by HPLC-VWD, Pm is the moisture con-
tent measured by Karl Fisher method, and Pa is the inor-
ganic elements content measured by ICP-MS.
2.4.2. QNMR Method
The QNMR method is based on the directly propor-
tional relationship between the signal response (inte-
grated signal area, Ix) and the number of nuclei generat-
ing the corresponding resonance line, which is represented
by Eq.1 as [8].
std
S
std
std
x
x
std
std
x
S
x
xP
m
m
M
M
N
N
I
I
m
m
P⋅⋅⋅⋅==
(2)
Where Px is purity of analyte; Ix and Istd are the inte-
grated signal areas of analyte and internal standard, re-
spectively; Nx and Nstd are the spin numbers of the ana-
lyte and internal standard, respectively; Mx and Mstd are
molar mass of analyte and internal standard, respectively;
m and mstd are the mass of analyte and internal standard,
respectively; Pstd is the purity of the internal stand ard.
The QNMR measurements were performed with a
Bruker AV600 spe ctrometer (Bruker, Billerica, MA). Spec-
tra were run with the following optimized parameters:
probe size, 5 mm; probe temperature, was 23.0˚C; exci-
tation pulse angle, 45˚; 32 K time domain points; 16 K
spectral data points; pulse delay, 4.15 μs; relaxation delay,
32 s and number of scans, 32. A sample solution con-
taining about 10 mg benzoic acid, 70 mg of carbohydrate
samples diluted in 0.5 mL DMSO was analyzed by
QNMR. The total mass of analyte in the sa mple solution
was determined usi ng Eq.1.
2.5. Homogeneity Study and Stability Study
According to the Technical Norm of Primary Refer-
ence Material of China [9], the homogeneity study and
the stability study were carried out by HPLC-VWD me-
thod. 15 bottles, were randomly selected and assayed for
O
OOH
HO
HO 7
6510 43
2
9
81' 4'
5'
6'
2'3'
AC
B
Baicalein CRM
O
OOH
HO7
6510 43
2
9
81'
4'
5'
6'
2'
3'
AC
B
5,7-dihydoxy flavones CRM
O
OOH
HO7
651043
2
9
81'
4'
5'
6'
2'
3'
AC
B
OH
Naringenine CRM
C. Quan et al. / Agricultural Sciences 4 (2013) 89-96
Copyright © 2013 SciRes. O PEN A CCESS
Table 1. The chromatography conditions of the studied flavonoids CRMs.
CRMs Mobile phase
(Acetonitrile:0.1% phosphoric acid) Wavelength (nm) Injection volume (μL) Flow (mL/min) Concentration (mg/mL)
Baicalein 35:65 275 5 1 0.5
5,7-dihydoxy fl avones 43:57 268 5 1 0.3
Naringenine 35:65 288 2 1 1
homogeneity between bottles, and 7 portions from each
bottle were assayed for homogeneity within bottle. The
results were examined via F-test and t-test. In F-test,
standard deviations were used to examine whether the
deviation between bottles was significantly greater than
the deviation within bottle. In t-test, means were used to
examine whether the mean between bottles was signifi-
cantly different from that of within bottle [10].
A stability study monitored a significant change in
value of the flavonoids CRMs was performed over one
year.
A study of long-term stability for storage at 4˚C was
carried out by HPLC-VWD method [11 ] in which ma-
terial was determined at 1, 3, 6 and 12 months. For each
condition, two solutions were prepared and analyzed in
duplicate. A comparison was made between the purity
determined by HPLC-VWD for different storage time. A
freshly prepared control sample was analyzed at each
time. All the samples were randomized for analysis.
2.6. Uncertainty Evaluation
The uncertainty evaluation for the flavonoids CRMs
derived from the certified value, homogeneity and stabil-
ity study. The expanded uncertainty was calculated as
follows:
222
c vhs
uuuU kUk=⋅=⋅ ++
( 3)
where, k is the coverage factor, Uc is the combined un-
certainty of the sample, uv, uh, us are the uncertainty
components of the assigned value, homogeneity and sta-
bility study.
3. RESULTS AND DISCUSSION
3.1. Qualitative Analysis Methods
3.1.1. LC-MS Method
The experiment obtained the molecular weight of the
flavonoids CRMs by full scan mass spectra. The results
indicted the molecular ion peak (M + H)+ of mass charge
ratio m/z of baicalein and 5,7-dihydoxy flavones was the
molecular weight plus the atomic weight of hydrogen
(1.0), the molecular ion peak (M H) of mass charge
ratio m/z of naringenine was the molecular weight sub-
tract the atomic weight of hydrogen (1.0). Ion fragments
by product ion scan are detailed in Figu re 2. The results
showed the flavonoids mainly fractured in 1,2 C-Ochem-
ical bond and 3,4 C-C chemical bond.
3.1.2. IR Method
The Infrared spectrum of the flavonoids CRMs were
shown in Figu re 3. The results revealed the peaks at
3090, 3097, and 3114 cm1 belong to hydroxide peaks,
1660, 1653, and 1630 cm1 belong to associat ed car bon yl
peaks, 1449 ~ 1618 cm1 belong to benzene ring skeleto n
of C = C peaks of the flavonoids CRMs. The Infrared
spectrums of the flavonoids CRMs were consistent with
those spectrums from literature [13-15] and can be used
for quantification.
3.2. Quantitative Analysis Methods
3.2.1. Mass Balance A pproach
The HPLC chromatogram of the flavonoids CRMs
were shown in Figure 4. The results of the moisture
content and the inorganic elements contents were listed
in Table 2, as indicated, The water content for Baicalein,
5,7-dihydoxy flavones, and naringenine were 0.538%,
0.150%, 0.098% and 0.115%, respectively, all were less
than 0.1%, while the inorganic content all warless than
0.00002% at the studied conditions.
3.2.2. QNMR Method
QNMR was also used to determinate purity values
flavonoids CRMs. The NMR spectrums were shown in
Figu re 5. The C3-H peak areas of baicalein, C8-H peak
areas of 5,7-dihydoxy flavones, C4,-OH peak areas of
naringenine, C2,2-2H peak areas of the benzoic acid in-
ternal standard were selected for integra tion .
3.2.3. Purity Values of the Flavonoids CRMs
The purity values of flavono ids CRMs were calculated
with the average of mass balance approach and QNMR
method. The results are listed in Table 2.
3.3. Homogeneity Study and Stability Study
The F-Test and t-test results of all the studied were
summarized in Table 3. No significant differences were
found for all the studied compounds in homogeneity test
as the calculated F-values were less than the critical
F-values (Table 3). Although a slight difference was ob-
served, this was attributed to the variation of the analyti-
cal method.
C. Quan et al. / Agricultural Sciences 4 (2013) 89-96
Copyright © 2013 SciRes. OPEN A CCESS
92
Figure 2. Main fragmentation schemes of the flavonoids CRMs.
Figure 3. The infrared spectrums of the studied flavonoids CRMs.
The flavonoids CRMs stored at 4˚C were used to test
stability. Th e testing data of stability study are showed in Table 4. The results in different times were consistent
with the standard value through the t-test analysis. There
4000 3500 3000 2500 2000 1500 1000500
0
20
40
60
80
100
1470
640
827
899
1090
1160
1618
3090
3410
Wavenumbers
(
cm
-1)
Transmitt ance
%)
1300
1586
1660
1505
1390
1020
Baicalein C RM
4000 3500 3000 2500 2000 1500 1000500
0
50
100
641
807
842
1168
1500
1577
1034
1449
1612
1653
2712 2631
Transmitt ance
%)
Wavenumbers
(
cm
-1
)
3097
5,7-dihydoxy flavones CR M
4000 3500 3000 2500 2000 1500 1000500
0
50
100
1083
1065 532
731
832
1520
1499
1158
1602
1630
2833
3114
Transmitt ance
%)
Wavenumbers
(
cm
-1
)
3286
Naringenin e CR M
C. Quan et al. / Agricultural Sciences 4 (2013) 89-96
Copyright © 2013 SciRes. O PEN A CCESS
Figure 4. Purity determination HPLC chromatogram of the flavonoids CRMs.
Tab le 2. The purity values results of flavonoids CRMs.
CRMs HPLC (%) QNMR (%) Average (%)
Baicalein 98.78 98.88 98.8
5,7-dihydoxy fl avones 99.40 98.84 99.1
Naringenine 99.47 99.46 99.5
were no obvious changes in purity values monitored of
the flavonoids CRMs over one year. The results mean
that they were stable.
3.4. Uncertainty Evaluation
The uncertainty for the flavonoids CRMs is composed
of the uncertainty from the certified value, homogeneity
and stability study.
3.4.1. Uncertainty of Certified Value
1) Uncertainty from HPLC measurement [16].
a) u1 (type A uncertainty) is the uncertainty from re-
peatability of HPLC measurement.
b) u2 (type B uncertainty) is the uncertainty from dif-
ferent responses from flavonoids CRMs and impurities at
different wavelengths of HPLC measurement.
The com bined uncerta inty is as follow:
22
HPLC1 2
uuu= +
2) Uncertainty of the QNMR measurement
a) u3 (type A uncertainty) is the uncertainty from re-
peatability of QNMR measurement.
b) u4 is the uncertainty from relative molar masses of
flavonoids CRMs and benzoic acid internal standard
CRM. The value of u4 calculated is less than 0.1%. In
this case, u4 is ignored.
c) u5 is the uncertainty from benzoic acid CRM. The
uncertainty of benzoic acid CRM was 0.1%, according to
the uniform distribution, u5 = 0.1%/31/2 = 0.06%.
d) u6 is the uncertainty from balance weighing. The
balance (Max = 2.1 g, d = 0.1 μg) was used to weigh the
flavonoids CRMs and benzoic acid internal standard
CRM. Balance changing and test error estimate for the
0.5 μg, therefore u6 is estimated 0.01%.
The combined uncertainty is as follow:
22 2 2
QNMR3 45 6
uu uu u=++ +
3.4.2. The Combined Uncertainty
The combined uncertainty for the flavonoids CRMs is
composed of the uncertainty from the certified value,
homogeneity and stability study. The results of homo-
geneity and stability study completely conformed to the
requirement of homogeneity and st ability in the regula-
tion of the CRMs. The homogeneity and stability of the
flavonoids CRMs both are good. Therefore the uncer-
tainty of homoge ne i ty (uh) and stab ility (us) is neglected.
2222222
Cvh sHPLCQNMRh s
22 22 2222
123 45 6hs
22 22 22
123 45 6
Uu uuuuuu
u uuuuuuu
u uuuuu
= ++=+++
=++ ++ +++
≈++ ++ +
3.4.3. The Expanded Uncertainty
The expanded uncertainty (U) for this flavonoids
CRMs was calculated by multiplying the combined un-
certainty by a coverage factor of k = 2. The uncertainty
evaluation for flavonoids CRMs were listed in Table 5.
4. CONCLUSION
The three flavonoids CRMs were developed with the
procedures of sample preparation, certified value, homo-
geneity study, and stability study and uncertainty evalua-
tion. The three flavonoids CRMs can not only meet the
requirement of quantitative detection of food, medicine
and related materials, but also provide a more quickly
and accurately experimental methods for the research
and devel o pment of related prod uc ts.
5. ACKNOWLEDGEMENTS
The authors acknowledge Special Program for Key Basic Research
of the Ministry of Science and Technology, China (Grant No.
2011FY130100) and the National Natural Science Fou ndation of C hina
(Grant No 21275134/ B 0505) for financing this work.
05 10 15 20 25 30 3540 45
0
5
10
15
20
25
30
A / mAu
t / m in
A
Baicalein C RM
05 10 15 20 25 30 35 40 45
0
5
10
15
20
25
30
t / m in
A / mAu
B
5,7-dihydoxy flavones CR M
05 10 15 20 25 30 35
0
5
10
15
20
25
30
A / mAu
t / m in
Naringenin e CR M
C. Quan et al. / Agricultural Sciences 4 (2013) 89-96
Copyright © 2013 SciRes. OPEN A CCESS
94
Figure 5. The spectrums of the studied flavonoids CRMs.
C. Quan et al. / Agricultural Sciences 4 (2013) 89-96
Copyright © 2013 SciRes. O PEN A CCESS
Table 3. Homogeneity results for the contents, the within-bottle, and between-bottle and homogeneity test of flavonoids CRMs.
Homogeneity test Baicalein CRM 5,7-dihydoxy flavones CR M Naringenine CRM
X between-bottle (s1,%) 99.31 ± 0.018 99.58 ± 0.013 99.58 ± 0.015
within-bottle (s2,%) 99.31 ± 0.011 99.58 ± 0.010 99.58 ± 0.010
F-test
22
12
/Fss=
2.63 1.86 2.60
0.05
F (14,6)
3.96 3.96 3.96
0.05
FF<
Y Y Y
t-test
12
txx= −
0.0010% 0.0011% 0.0022%
1
22
2
12
α
12
ss
t2 nn

 

= +
 
 

 

0.0064% 0.0049% 0.0053%
tt
α
Y Y Y
Table 4. The results of stability of flavonoids CRMs.
Time interval Baicalein 5,7-dihydoxy fl avones Naringenine
Zero month 99.27 99.50 99.53
One month 99.37 99.56 99.57
Three months 99.31 99.54 99.58
Six months 99.31 99.52 99.50
Twelve months 99.35 99.46 99.55
Average (%) 99.32 99.52 99.55
RSD (%) 0.04 0.04 0.03
Table 5. The results of uncertainty evaluation of flavonoids CRMs.
Uncertainty Baicalein CRM 5,7-dihydoxy flavones CRM Naringenine CRM
u1/% 0.01 0.02 0.03
u2/% 0.04 0.04 0.01
u3/% 0.36 0.33 0.50
u4/% 0 0 0
u5/% 0.06 0.06 0.06
u6/% 0.01 0.01 0.01
uc/% 0.37 0.34 0.50
k 2 2 2
U/% 0.8 0.7 1.0
Purity value/% 98.8 99.1 99.5
REFERENCES
[1] Arunachalama, J., Bleiseb, A., Mahwar, R.S., et al. (2006)
The natural matrix reference material database of the In-
ternational Atomic Energy Agency: Reference materials
in support of dietary investigations. Journal of Food
Composition and Analysis, 19, 241-249.
http://dx.doi.org/10.1016/j.jfca.2005.07.002
[2] Sander, L.C., Sharpless, K.E. and Wise, S.A. (2006) Die-
tary supplement standard referenc e materials. Life S c ien c es,
78, 2044-2048.
http://dx.doi.org/10.1016/j.lfs.2005.12.037
[3] Chen, C.H., Huang, T.S., Wong, C.H. , et al. (2009) Syn-
ergistic anti-cancer effect of baicalein and silymarin on
human hepatoma HepG2 Cells. Food and Chemical Tox-
icology, 47, 638-644.
http://dx.doi.org/10.1016/j.fct.2008.12.024
[4] Middleton J r., E. and Kanda swami, C. (1992) Effects of
flavonoids on immune and inflammatory cell functions.
Biochemical Pharmacology, 43, 1167-1179.
http://dx.doi.org/10.1016/0006-2952(92)90489-6
[5] Heim, K.E., Tagliaferro, A.R. and Bobilya, D.J. (2002)
Flavonoid antioxidants: Chemistry, metabolism and struc-
ture-activity relationships. Journal of Nutritional Bio-
chemistry, 13, 572-584.
http://dx.doi.org/10.1016/S0955-2863(02)00208-5
[6] The Directorate for the Quality of Medicines of the
Council of Europe (2007) Establishment of reference
standards, in European Pharmacopoeia. 5th Edition, Sup-
plement 5.6, General texts 5.12.4, Council of Europe,
Strasbourg.
[7] World Health Organization (2003) Evaluation of chemi-
C. Quan et al. / Agricultural Sciences 4 (2013) 89-96
Copyright © 2013 SciRes. OPEN A CCESS
96
cal reference substance. In: International Pharmacopoeia,
3rd Editon, Vol . 5, Supplementary Information, Part A,
3.3, World Health Organization, Geneva.
[8] Liu, S.Y. and Hu, C.Q. (2007) A comparative uncertainty
study of the calibration of macrolide antibiotic reference
standards using quantitative nuclear magnetic resonance
and mass balance methods. Analytica Chimica Acta, 602,
114-121. http://dx.doi.org/10.1016/j.aca.2007.09.009
[9] JJG1006-1994 (1994) Technical norm of primary refer-
ence material of china. Beijing.
[10] Quan, H. (2002) Reference material and its application
technology. China Standard Press, Beijing. (in Chinese)
[11] Dai, X.H., Fang, X., Shao, M.W., Li, M., et al. (2011)
Certified reference materials (GBW09170 and 09171) of
creatinine in human serum. Journal of Chromatography B,
879, 429-435.
http://dx.doi.org/10.1016/j.jchromb.2010.12.031
[12] EURACHEM/CITAC Guide CG4 (2000) Quantifying
uncertainty in analytical measurement. Seconded.
[13] Li, Y.X., Suo, Q.L., He, W.Z., et al. (2008) Separation
and Purification and Structural Characterization of Bai-
calein. Spectroscopy and Spectral Analysis, 28, 1895-
1899.
[14] NIST, NIST Chemistry WebBook.
http://webbook.nist.gov/
[15] Wang, Y.G., Xie, S.W. and Su, W.W. (2005) St udy on the
chemical constituents of Polygonum chinese L. Journal of
Chinese Medicinal Materials, 28, 1000-1001.
[16] Quan, C., Su, F.H., Wang, H.F., et al. (2011) Develop-
ment of anabolic-androgenic steroids purity certified ref-
erence materials for anti-doping. Steroids, 76, 1527-1534.
http://dx.doi.org/10.1016/j.steroids.2011.09.001