Anal. Chem.
2009, 81, 8916–8922
Development of Methodology Based on the
Formation Process of Gold Nanoshells for
Detecting Hydrogen Peroxide Scavenging Activity
Hui Li, Xiaoyuan Ma, Jian Dong, and Weiping Qian*
State Key Laboratory of Bioelectronics, School of Biological Science and Medical Engineering, Southeast University,
Nanjing 210096, P. R. China
In the present work, we have developed a novel nano-
composite-based method for estimating antioxidant activ-
ity. The assay implements a new enzyme-free optical
nanoprobe for assessing hydrogen peroxide (H2O2) scav-
enging activity based on the formation process of gold
nanoshells (GNSs). H2O2 could enlarge the gold nano-
particles (GNPs) on the surface of GNSs precursor
nanocomposites (SiO2/GNPs), and the preadsorbed
GNPs served as nucleation sites for Au deposition. As
the concentration of H2O2 increases, more GNPs on
the SiO2 cores are enlarged until continuous GNSs are
formed. During the growth procedure, the spectra
changes correlate well with H2O2 concentrations which
indicate that this nanocomposite is a good nanoprobe
for detecting H2O2. H2O2 scavenging activities of
several antioxidants were determined by restraining
the H2O2-mediated formation of GNSs from SiO2/
GNPs, and the changes of the corresponding plasmon
absorption bands correlated well with H2O2 scavenging
activity of antioxidants. The spectra were monitored by
a UV-vis-near-infrared (NIR) spectrophotometer, and
the wavelength changes were adopted as detection
signal. The results obtained expressed the difference
of H2O2 scavenging activity between various tested
compounds, and the relationship between function and
structure of antioxidants was also discussed in this
article. The new method based on the formation
process of GNSs is simple, rapid, and sensitive and,
additionally, can be used in visual analysis to a certian
extent for antioxidant functional evaluation.
Oxidative stress has been implicated in the pathogenesis of
several human diseases and conditions including aging, diabetes,
chronic inflammation, and cancer. It occurs when excessive
quantities of reactive oxygen species (ROS), such as superoxide
radical anion (O2•-), hydrogen peroxide (H2O2), and hydroxyl
radical (•OH), overwhelm host antioxidant defenses, resulting
in cellular damage, protein damage, lipid peroxidation, DNA
alteration, and enzyme inactivation.1-5 ROS can be scavenged
by antioxidants, which are defined as any substances that when
* To whom correspondence should be addressed. Phone: +8625-83795719.
Fax: +862583795719. E-mail: wqian@seu.edu.cn.
(1) Magalhaes, L. M.; Segundo, M. A.; Reis, S.; Lima, J. L. F. C. Anal. Chim.
Acta 2008, 613, 1–19.
8916 Analytical Chemistry, Vol. 81, No. 21, November 1, 2009
present at low concentration compared to those of an oxidizable
substrate significantly delay or prevent oxidation of that substrate.
Antioxidants are not only used in biomedicine and clinical
medicine but also have extensive applications in food industry and
food nutriology. ROS scavenging activity is an important aspect
of antioxidant activity. H2O2 is a key part of ROS and generated
in vivo under physiological conditions by peroxisomes, several
oxidative enzymes including glucose oxidase and D-amino acid
oxidase, and by dismutation of superoxide radical catalyzed
by superoxide dismutase; it together with O2•- can damage
many cellular components and furthermore convert into more
ROS such as •OH. Thus, the determination of H2O2 scavenging
activity is a very important part for evaluation of reactive oxygen
scavenging activity.1,2
Because of the importance of the measurement of antioxidant
activity, much research has gone into the study of antioxidant
activity assessment and a considerable amount of relevant articles
have been published. In view of many different antioxidants having
various scavenging activities against O2•-, H2O2, or •OH, many
methods for measuring these properties have been developed
with different mechanisms, respectively, such as chromatog-
raphy,6 chemiluminescence (CL),7 electrochemiluminescence
(ECL),8 spectrophotometry,1-5 cyclic voltammetry (CV),1 auto-
matic flow injection based methodologies,9 electron spin resonance
(ESR),1 etc. Recently, several nanomaterials have been utilized
in the measurement of antioxidant activity, such as metal
nanoparticles,10,11 quantum dots,8 polymer spheres,12 and so on.
(2) Pazdzioch- Czochra, M.; Widenska, A. Anal. Chim. Acta 2002, 452, 177–
184.
(3) Mansouri, A.; Makris, D. P.; Kefalas, P. J. Pharm. Biomed. Anal. 2005,
39, 22–26.
(4) Wood, L. G.; Gibson, P. G.; Garg, M. L. J. Sci. Food Agric. 2006, 86, 2057–
2066.
(5) Arnous, A.; Petrakis, C.; Makris, D. P.; Kefalas, P. J. Pharmacol. Toxicol.
Methods 2002, 48, 171–177.
(6) Winston, G. W.; Regoli, F.; Dugas, A. J.; Fong, J. H.; Blanchard, K. A. Free
Radical Biol. Med. 1998, 24, 480–493.
(7) Magalhaes, L. M.; Segundo, M. A.; Reis, S.; Lima, J. L. F. C.; Estela, J. M.;
Cerda, V. Anal. Chem. 2007, 79, 3933–3939.
(8) Jiang, H.; Ju, H. X. Anal. Chem. 2007, 79, 6690–6696.
(9) Magalhaes, L. M.; Lucio, M.; Segundo, M. A.; Reis, S.; Lima, J. L. F. C.
Talanta 2009, 78, 1219–1226.
(10) Lee, H. K.; Lee, K.; Kim, I.-K.; Park, T. G. Adv. Funct. Mater. 2009, 19,
1–7.
(11) Wang, J.; Zhou, N. D.; Zhu, Z. Q.; Huang, J. Y.; Li, G. X. Anal. Bioanal.
Chem. 2007, 388, 1199–1205.
(12) Kim, S. H.; Kim, B.; Yadavalli, V. K.; Pishko, M. V. Anal. Chem. 2005, 77,
6828–6833.
10.1021/ac901534b CCC: $40.75  2009 American Chemical Society
Published on Web 10/13/2009
Compared with conventional methods, these approaches with high instrumentation, which are quite time-consuming. Therefore, the
sensitivity, rapid determination, relative ease of measurement, development of new and effective methods for the detection of
inexpensive, and simple apparatus have attracted a widespread H2O2 scavenging activity still remains a great challenge. Herein,
attention. As a new metal and semiconductor composite nanoma- the purpose of this investigation is to apply SiO2/GNPs as a
terial, gold nanoshells (GNSs) consisting of a silica core with a nanoprobe for estimating antioxidant activity and implement a
gold shell have attracted considerable attentions owing to their new reliable, enzyme-free spectrophotometric methodology for
unique chemical, physical properties, and good biocompatibility, assessing H2O2 scavenging activity in vitro based on the
especially highly tunable plasmon resonance. The localized surface formation process of GNSs.
plasmon resonance (LSPR) of GNSs has strongly depended on
the core radius-shell thickness ratio and can be placed from
EXPERIMENTAL SECTION
visible to near-infrared (NIR) region. Compared with gold nano-
Chemicals and Reagents. Silica colloidal spheres (∼110 nm)
particles (GNPs), the utility of GNSs is more extensive because
and 3-(aminopropyl)-triethoxysilane (APTES, 98%) were obtained
their plasmon resonance can be changed in a controlled manner
from Nissian Chemical Ind., Ltd., Japan, and Sigma, respectively.
easily during a relatively broad wavelength range. For example,
Sodium borohydride (NaBH4), potassium carbonate (K2CO3),
the LSPR of GNPs with a diameter range from 8 to 99 nm in the
chloroauric acid terahydrate (HAuCl4 · 4H2O), hydrogen per-
spectra region is only from 518 to 575 nm. While, the LSPR of
oxide (H2O2, 30%), and anhydrous ethanol were all purchased
GNSs spans a range of ∼300 nm in the wavelength with the core
from Nanjing Sunshine Biotechnology Ltd., China. Tannic acid,
radius-shell thickness ratio varied between 3 and 12.13 The
ferulic acid, L-apple acid, tartaric acid, salicylic acid, and citric
unique optical properties of GNSs that lead to this new nanoma-
acid were received from Shanghai Aladdin Chemical Ltd.,
terial are both useful and sensitive, accordingly GNSs have
China, and used without further purification. All of the chemical
prompted considerable attention for a variety of applications in
reagents were analytical grade. K2CO3/HAuCl4 solution was
biomedicine, material science, diagnoses and therapy of disease,
prepared as follows: 100 mL of aqueous K2CO3 solution (0.25
and chemical and biological sensors.13-17 Among various GNSs
g/L) was mixed with 1.5 mL of HAuCl4 (1%) stock solution
synthesis methods, the seed-mediated growth technique has been
under continuous stirring for 20 min and aging in the dark at
widely exploited because it is readily adaptable to produce uniform
4 °C for 24 h. The final concentration of H2O2 used in the
gold layers on the surfaces of silica spheres compared with the
system is 200 µM. Aqueous solutions used in the experiments
direct synthesis procedure. This technique has also been used to
were prepared using Milli-Q water from Milli-Q system (resis-
prepare other core/shell nanostructures such as nanorice and
tivity >18 MΩ).
nanorod, etc.18,19 As the intermediate in the synthesis of GNSs
Synthesis of GNPs. GNPs (∼5 nm) were prepared by the
adopting the seed-mediated growth method, GNSs precursor
reduction of HAuCl4 with NaBH4 according to our previous
nanocomposites (SiO2/GNPs) consist of a SiO2 core and at-
method:13,16 3 mL of HAuCl4 (1%) was mixed with 200 mL of
tached GNPs; the preadsorbed GNPs act as nucleation sites
water under vigorous stirring, followed by the addition of 1
for further reduction of gold onto the silica core by reduction
mL of K2CO3 (0.2 M). After that, 9 mL of freshly prepared
of AuCl4- solution in the presence of reducing agent.20 Our
NaBH4 (0.5 mg/mL) was quickly added to the mixture. The
previous work has shown that13,16 as more gold is reduced, the
mixture tuned to wine red rapidly, which indicated the genera-
surface coverage increases until coalescence into complete GNSs.
tion of GNPs. The obtained solution was stored at 4 °C until
During the process, the LSPR spans at least 200 nm, and this
use.
highly optical tunability offers this nanocomposite extensive
Synthesis of GNSs Precursor Nanocomposites (SiO2/
application perspectives.
GNPs). The SiO2/GNPs were prepared as reported in our
In general, the detection methods of H2O2 scavenging activity
previous work13 with some modifications. Silica colloidal
are dependent on peroxidase originally.1,2 The peroxidase-based
spheres were purified by centrifuging and redispersing five
approaches with a common disadvantage are that the peroxidase
times in ethanol. Then 39 mL of monodisperse silica colloidal
may interfere with the determination. Thus, a variety of enzyme-
spheres (0.013 g/mL in ethanol) were amino-functionalized by
free luminescent methods coupled with numerous techniques
have been presented for assessing the H2O2 scavenging activity mixing with 60 µL of APTES under vigorous magnetic stirring
in recent years.1,3,5,9 However, these techniques require compli- at 70 °C for 3 h to present positively charged amino groups.
cated sample preparation processes or expensive and sophisticated After centrifugation and redispersion in ethanol, aiming to
remove excess reactants, the pure amino-functionalized silica
(13) Wang, Y.; Qian, W. P.; Tan, Y.; Ding, S. H. Biosens. Bioelectron. 2008, 23,
spheres (593 µL) were redispersed in ethanol (20 mL) and
1166–1170. added dropwise to 200 mL of GNPs with vigorous stirring,
(14) Stewart, M. E.; Anderton, C. R.; Thompson, L. B.; Maria, J.; Gray, S. K.; resulting in GNPs adsorption on the surface of silica nanopar-
Rogers, J. A.; Nuzzo, R. G. Chem. Rev. 2008, 108, 494–521.
(15) Scampicchio, M.; Wang, J.; Blasco, A. J.; Arribas, A. S.; Mannino, S.; Escarpa, ticles and the formation of SiO2/GNPs. After centrifugation and
A. Anal. Chim. Acta 2006, 78, 2060–2063. removal of the supernatant three times, pure SiO2/GNPs were
(16) Ding, S. H.; Qian, W. P.; Tan, Y.; Wang, Y. Langmuir 2006, 22, 7105– redispersed in 150 mL of water. The absorbance spectrum of
7108.
(17) Jain, P. K.; El-Sayed, M. A. Nano Lett. 2007, 7, 2854–2858. obtained SiO2/GNPs stock solution was 0.5 at ∼533 nm. The
(18) Skrabalak, S. E.; Xia, Y. N. ACS Nano 2009, 3, 10–15. SiO2/GNPs nanocomposites can be dispersed homogeneously
(19) Wang, H.; Brandl, D. W.; Le, F.; Nordlander, P.; Halas, N. J. Nano Lett. in solvents such as water and ethanol to form colloidal
2006, 6, 827–832.
(20) Hirsch, L. R.; Gobin, A. M.; Lowery, A. R.; Tam, F.; Drezek, R. A.; Halas, suspensions. In this case, they can be kept at least 1 month at
N. J.; West, J. L. Ann. Biomed. Eng. 2006, 34, 15–22. 4 °C.
Analytical Chemistry, Vol. 81, No. 21, November 1, 2009 8917
Scheme 1. Schematic Illustration of the Formation For H2O2-Mediated Gold Nanoshells
H2O2 Scavenging Activity Detection Assay. The testing
procedures were as follows: 3 mL of water was mixed with 200
µL of sample and 100 µL of H2O2 (13.6 mM), after incubation
for 2 min, 0.5 mL of SiO2/GNPs solution and 3 mL of K2CO3/
HAuCl4 were immediately added and mixed for 5 min under
vigorous stirring at room temperature, finally the spectra were
performed on a Shimadzu UV3150 UV-vis-NIR spectropho-
tometer in transmission mode. A JEM2100EX transmission
electron microscope (TEM) was used to image the morphology
of SiO2/GNPs at different conditions. Control was prepared by
mixing 0.5 mL of SiO2/GNPs, 3 mL of K2CO3/HAuCl4, and 3.3
mL of water. Reference samples were tested by adding 200 µL
of water instead of an antioxidant solution. All measurements
were performed at least in triplicate and values were averaged.
Results were given as means ± the standard deviation (SD).
RESULTS AND DISCUSSION
H2O2-Mediated Growth of SiO2/GNPs. As we know that
many chemicals could induce the formation of GNSs, such as
formaldehyde, hydroxylamine-hydrochloride, sodium borohydride,
and carbon monoxide gas.21 Our work observes that H2O2 could
reduce AuCl4- to Au which deposit on the surface of SiO2/
GNPs and enlarge GNPs. As more gold is reduced, the GNPs
grow larger and ultimately merge, resulting in the formation
of a continuous shell layer. During this procedure the pread-
sorbed GNPs serve as nucleation sites for Au deposition. As
the concentration of H2O2 increases, more AuCl4- can be
reduced and more GNPs on cores are enlarged; if the
concentration of H2O2 reaches to some level, continuous
nanoshells are formed as Scheme 1 illustrates.
The concentration of H2O2 could affect the growth of SiO2/
GNPs and the corresponding spectra. Figure 1A shows the
evolution of the spectra in the presence of variable concentrations
of H2O2. Also, the corresponding derived calibration curve is
shown in Figure 1B. The TEM images indicate the morphology
variation of SiO2/GNPs upon reaction with the increasing
concentration of H2O2 (Figure 2). As the concentration of H2O2
increases, the wavelength is constantly red-shifted, and the
(21) Brinson, B. E.; Lassiter, J. B.; Levin, C. S.; Bardhan, R.; Mirin, N.; Halas,
N. J. Langmuir 2008, 24, 14166–14171.
8918 Analytical Chemistry, Vol. 81, No. 21, November 1, 2009
absorbance is intensified. When the final concentration of H2O2
reaches to 200 µM, the wavelength maximum occurs which
reveals the formation of continuous nanoshells (core diameter
is ∼110 nm, shell thickness is ∼20 nm). The reaction process
completes in 5 min with color changes from faint pink to blue,
and the spectra transfers obviously from ∼530 to ∼750 nm.
Thus, the 200 µM H2O2 is the optimal concentration for the
formation of GNSs in the reaction system.
The K2CO3/HAuCl4 aged time is a key parameter for the
GNSs formation. When the K2CO3/HAuCl4 aged time reaches
to 24 h, the wavelength maximum occurs which reveals that
continuous nanoshells are formed. Less than or exceeding 24 h,
the obtained wavelengths are all lower than this value. While
the K2CO3/HAuCl4 aged time exceeds 72 h, the SiO2/GNPs
stop growing and the spectra do not change. Therefore, 24 h
is the optimal aged time of the K2CO3/HAuCl4 solution.
Detection of H2O2 Scavenging Activity. During the past
years, there have been some optical studies reporting on
methods for assessing the H2O2 scavenging activity, involving
UV spectrophotometry and fluorescence spectrophotometry.
The most common UV spectrophotometric determination
method is a direct way based on the intrinsic absorption of
H2O2 in the UV region at 230 nm, yet this UV enzyme-free
assay has a problem that samples may also absorb at this
wavelength, which makes it difficult to distinguish a small
change when the background absorption is large.1 Some
fluorimetric methods are based on the oxidation of scopoletin
or homovanillic acid to their nonfluorescent product or
fluorescent biphenyl dimer in the presence of H2O2 and
peroxidase. However, these peroxidase-based approaches
have a disadvantage that antioxidants may be a substrate
for the peroxidase enzyme, resulting in error determination.1,2
There are also some enzyme-free luminescent methods that
rely on peroxyoxalate chemiluminescence (POCL) using 9,10-
diphenylanthracene, luminol, or lucigenin as a fluorophore
(probe). POCL involves H2O2 imidazole-catalyzed oxidation
of a substance to a high-energy intermediate and transfers
its energy to the fluorophore. The transition of the excited
state of the fluorophore to its ground state causes the
emission of light. Accordingly, any compound with H2O2
Figure 1. (A) Absorbance spectra for the growth of SiO2/GNPs in the presence of variable concentrations of H2O2 from 0 to 300 µM. (B) The
calibration curve corresponding to the peak absorbance and wavelength of reaction system with different concentrations of H2O2. All systems
include 3.2 mL of water, 3 mL of K2CO3/HAuCl4, 0.5 mL of SiO2/GNPs solution, and 100 µL of H2O2 or water. Spectra were recorded after the
mixture reacted for 5 min under vigorous stirring.
Figure 2. TEM images of (a) SiO2/GNPs upon reaction with variable
concentrations of H2O2 of (b) 25, (c) 100, and (d) 200 µM.
scavenging capacity promotes CL inhibition. The above
POCL methods are proposed to evaluate mainly lipophilic
antioxidants because of the nonpolar environment employed.1,3,5
This article reports on the H2O2-mediated growth of SiO2/
GNPs to form GNSs and apply this process to develop a
method for estimating H2O2 scavenging activity of antioxi-
dants. The presence of substances with H2O2 scavenging
activity prevents SiO2/GNPs from forming complete GNSs
and causes a variation in spectra. Thus, both the shifts of
wavelength and absorbance can be used as an optical
signature for the change of SiO2/GNPs morphology. Some
conventional spectrophotometry for measuring antioxidant
capacity mainly use the absorbance change at a fixed
wavelength as an optical signature, such as the ABTS•+
assay, DPPH• assay, ferric reducing antioxidant power
(FRAP assay), folion-ciocalten reducing capacity (FC assay),
etc.1 Compared with the variation of the absorption value,
the variation of wavelength is more direct for visual
discrimination, which has more potential application in visual
colorimetric detection. In addition, to estimate the antioxi-
dant activity of a pure compound, it is also essential to assess
the antioxidant ability of relevant food and beverage.
However, some tested samples such as tea, fruit juice, and
red wine also absorb in the measured wavelength range and
the spectra are not horizontal. The instinct color of these
natural antioxidant substances may interfere with determi-
nation and lead to error estimation. Therefore, the wave-
length shift is chosen as the detection signal to avoid this
type of determination error. Besides, if there is an interfering
agent that may influence our detection, the SiO2/GNPs
nanocomposite can be separated through centrifugation
(∼3000 rpm) and ultrasonic redispersion from the superna-
tant conveniently. According to our experiments, the wave-
length of the nanocomposite is unchanged through centri-
fuging and redispersing, which does not affect the detection
results.
As the concentration of antioxidant increases, the residual
H2O2 decreases and the growth of SiO2/GNPs is more
inhibited, so the H2O2 scavenging activity of samples was
measured in different concentration ranges. For comparison
of the H2O2 scavenging activity of antioxidants, the obtained
results are presented as the percentage wavelength inhibi-
tion (Iw) using the following formula: Iw ) ∆w /∆w0 (%),
where ∆w0 is the wavelength difference of the reaction
mixture with and without H2O2 in the absence of sample,
∆w is the reaction mixture wavelength difference without
and with sample in the presence of H2O2. The amount of
each antioxidant to provoke a 50% wavelength inhibition ratio
is called IC50, while hydrogen peroxide scavenging activity
(SAHP) is defined as SAHP ) 1/IC50 (Figure 3). A relationship
between the concentration of antioxidant and the percentage
Analytical Chemistry, Vol. 81, No. 21, November 1, 2009 8919
Figure 3. Schematic diagram of the method for expressing H2O2 scavenging activity: (A) The percentage wavelength inhibition (Iw) and (B) the
hydrogen peroxide scavenging activity (SAHP).

Figure 4. The percentage wavelength inhibition of antioxidants including tannic acid, ferulic acid, citric acid, L-apple acid, tartaric acid, and
salicylic acid.

wavelength inhibition are obtained (Figure 4). According to
those results, IC50 values can be read. The concentration
range of tested antioxidants with their IC50 values and the
corresponding SAHP values are shown in Table 1.
On the grounds that the spectral change resulting from the
addition of antioxidants with their H2O2 scavenging activity
is in the visible range from ∼530 to ∼750 nm, the change in
color corresponding to the spectra can be observed directly
without a spectrophotometer, which makes the visual colo-
8920 Analytical Chemistry, Vol. 81, No. 21, November 1, 2009
rimetric detection of antioxidant activity feasible in qualitative
or semiquantitative analysis. So compared with traditional
methods for the determination of antioxidant capacity, such
as CL and chromatography, this means has more potential
practical application in rapid screening analysis of antioxi-
dants. Figure 5 shows the spectra changes and the relevant
color changes of the solution in the presence of variable
concentrations of tannic acid. The color change from faint pink
to blue could be directly observed by the naked eye with
Table 1. Concentration Ranges, IC50, and SAHP Values
of Samples

sample range (µM) IC50 (µM) SAHP (× 10-2 µM-1)

salicylic acid 100-800 470 0.21
tartaric acid 50-1200 273 0.36
L-apple acid 5-400 162 0.62
citric acid 5-400 106 0.94
ferulic acid 3-100 8 12.5
tannic acid 5-200 7 14.28

decreasing concentrations of tannic acid, and this color

alteration is consistent with the H2O2-mediated growth of the
SiO2/GNPs assay as the H2O2 concentration increases. The
TEM images indicate that increasing the concentration of
tannic acid indeed decreases the amount of H2O2 and inhibits
the growth of SiO2/GNPs (Figure 6).
Relationship between Activity and Structure of Antioxi-
dants. The results indicate that tannic acid and ferulic acid
express superior H2O2 scavenging activity than other tested
antioxidants, salicylic acid shows the weakest activity, three
organic acids (L-apple acid, tartaric acid, citric acid) represent
the median activity. These results with phenolic compounds
are in agreement with many other researchers that tannic acid Figure 5. Color images and the corresponding UV-vis spectra for
was more potent in radical scavenging activity than ferulic H2O2-mediated growth of (a) SiO2/GNPs in the presence of a fixed
concentration of 200 µM H2O2 and different concentrations of tannic
acid.22-24 The H2O2 scavenging activity may have a relationship
acid from (b) 200 to (h) 0 µM.
with the chemical structures and in particular the number of
hydroxyl groups.5,15 The structure of phenolic compounds, which
comprise an aromatic ring with one or more hydroxyl substituents
that range from simple phenolic molecules to highly polymerized
compounds, is an important determinant of their radical scaveng-
ing activity, and the antioxidant activity of phenolic acids increases
with an increasing degree of hydroxylation and extent of
conjugation.23,25 As is well-known, monophenols are less efficient
as antioxidants than polyphenols and tannic acid as large-
molecular weight polyphenols are strong radical scavengers,
which may be related to the formation of stable radicals; ferulic
acid and salicylic acid are monophenols and show lower antioxi-
dant activity than tannic acid. As to the monophenols, the inductive
effect of the hydroxyl group is an important factor that enhances
antioxidant ability and methoxyl substitution is another factor. For
this reason, salicylic acid with one hydroxyl is less efficient than
ferulic acid with one hydroxyl and methoxyl in the aromatic ring.22
Besides, three organic acids also express H2O2 scavenging
activity due to their hydroxyl structures although are not
phenolic acids. In conclusion, the degree of hydroxylation and Figure 6. TEM images of (a) SiO2/GNPs upon reaction with 200
extent of conjugation seem to be the key factor for the radical µM H2O2 and variable concentrations of tannic acid for (b) 200, (c)
25, and (d) 0 µM.
scavenging activity.
The method reported here is comparing the H2O2 scavenging pounds from dietary sources may reinforce antioxidants in the
activity of plant-based antioxidants, and this is a portion function body, keep the balance between antioxidants and oxidants in
of these multifunctional compounds. According to many living organisms, and help prevent cardiovascular disease and
researches, fruits and vegetables are excellent sources of cancer.26 The tested antioxidants were chosen because they
exogenously obtained antioxidants. Ingestion of these com- are all naturally occurring antioxidants that come from plants
and have superior antioxidant ability, meanwhile, many of them
(22) Sanchez-Moreno, C.; Larrauri, J. A.; Saura-Calixto, F. Food Res. Int. 1999, exist in daily foodstuff universally. For example, tannic acid is
32, 407–412.
the most important component of various tea beverages, and
(23) Pulido, R.; Bravo, L.; Saura-Calixto, F. J. Agric. Food Chem. 2000, 48, 3396–
3402. tannic acid, tartaric acid, and apple acid are major ingredients
(24) Kumar, G. S.; Nayaka, H.; Dharmesh, S. M.; Salimath, P. V. J. Food Compos.
Anal. 2006, 19, 446–452. (26) Prior, R. L.; Wu, X. L.; Schaich, K. J. Agric. Food Chem. 2005, 53, 4290–
(25) Scherer, R.; Godoy, H. T. Food Chem. 2009, 112, 654–658. 4302.

Analytical Chemistry, Vol. 81, No. 21, November 1, 2009 8921

of red wine, while many fruits and vegetables contain citric
acid, ferulic acid, and apple acid. Therefore, the relevant plant-
based foods and drinks should have the same function as pure
compounds. It is widely accepted that different antioxidants
represent variable behavior in tests performed under different
conditions, and no single assay will accurately reflect all of the
radical sources and estimate all the multifunctional antioxidants
in a complex system.27 Therefore, more than one method is
required to provide adequate information for in vitro antioxidant
characteristics evaluation. In this article, the method presented
here may serve as an additional means for estimating the
activities of multifunctional antioxidants.
CONCLUSIONS
Here, we have developed a new nanocomposite-based method
for estimating the H2O2 scavenging activity that relies on the
H2O2-mediated growth of SiO2/GNPs. The corresponding
plasmon absorption bands change correlate well with H2O2
(27) Wolfe, K. L.; Liu, R. H. J. Agric. Food Chem. 2007, 55, 8896–8907.
8922 Analytical Chemistry, Vol. 81, No. 21, November 1, 2009
scavenging activity of antioxidants. This methodology is simple
to design, easy to operate, and furthermore, could be used as
visual colorimetric detection without a spectrophotometer in
qualitative or semiquantitative analysis. It is a convenient tool
for the screening analysis of H2O2 scavenging activity. Studies
are under way to optimize the new method and evaluate more
antioxidants, not only pure compounds but also relevant
beverage and food with their functional estimation.
ACKNOWLEDGMENT
H.L. and X.M. contributed equally to this work. We gratefully
acknowledge support from the National Natural Science Founda-
tion of China (Grants 90923010 and 20475009) and the Ministry
of Science & Technology of China (Grant 2007AA022007).
Received for review July 10, 2009. Accepted September
29, 2009.
AC901534B