Article

pubs.acs.org/ac

Target-Induced Nanocatalyst Deactivation Facilitated by Core@Shell

Nanostructures for Signal-Amplified Headspace-Colorimetric Assay
of Dissolved Hydrogen Sulfide
Zhuangqiang Gao,† Dianyong Tang,‡ Dianping Tang,*,† Reinhard Niessner,§ and Dietmar Knopp§
†
Institute of Nanomedicine and Nanobiosensing, Key Laboratory of Analysis and Detection for Food Safety (Fujian Province &
Ministry of Education), Department of Chemistry, Fuzhou University, Fuzhou 350108, P. R. China
‡
Chongqing Key Laboratory of Environmental Materials & Remediation Technologies, College of Materials and Chemical
Engineering, Chongqing University of Arts and Sciences, Chongqing 402160, P. R. China
§
Chair for Analytical Chemistry, Institute of Hydrochemistry, Technische Universität München, Marchioninistrasse 17, D-81377
München, Germany
Downloaded by UNIV OF SOUTH DAKOTA on September 5, 2015 | http://pubs.acs.org

*
Publication Date (Web): September 4, 2015 | doi: 10.1021/acs.analchem.5b03008

S Supporting Information

ABSTRACT: Colorimetric assay platforms for dissolved

hydrogen sulfide (H2S) have been developed for more than
100 years, but most still suffer from relatively low sensitivity.
One promising route out of this predicament relies on the
design of efficient signal amplification methods. Herein, we
rationally designed an unprecedented H2S-induced deactiva-
tion of (gold core)@(ultrathin platinum shell) nanocatalysts
(Au@TPt-NCs) as a highly efficient signal amplification
method for ultrasensitive headspace-colorimetric assay of dissolved H2S. Upon target introduction, Au@TPt-NCs were
deactivated to different degrees dependent on H2S levels, and the degrees could be indicated by using a Au@TPt-NCs-triggered
catalytic system as a signal amplifier, thus paving a way for H2S sensing. The combination of experimental studies and density
functional theory (DFT) studies revealed that the Au@TPt-NCs with only 2-monolayer equivalents of Pt (θPt = 2) were superior
for H2S-induced nanocatalyst deactivation owing to their enhanced peroxidase-like catalytic activity and deactivation efficiency
stemmed from the unique synergistic structural/electronic effects between Au nanocores and ultrathin Pt nanoshells.
Importantly, our analytical results showed that the designed method was indeed highly sensitive for sensing H2S with a wide
linear range of 10−100 nM, a slope of 0.013 in the regression equation, and a low detection limit of 7.5 nM. Also the selectivity,
reproducibility, and precision were excellent. Furthermore, the method was validated for the analysis of H2S-spiked real samples,
and the recovery in all cases was 91.6−106.7%. With the merits of high sensitivity and selectivity, simplification, low cost, and
visual readout with the naked eye, the colorimetric method has the potential to be utilized as an effective detection kit for point-
of-care testing.

T he detection and monitoring of dissolved H2S in aqueous

mediums (e.g., natural and waste waters, agricultural
water, blood/serum, and cell lysates) has been a research theme
for a long time, because its level is not only an important
environmental index but also a significant biomedical index.1−3
Among various detection approaches for this theme, colori-
metric methods, as a classical technique, have been widely
exploited and utilized for detecting dissolved H2S for more than
100 years due to their distinct merits over other methods, e.g.,
low cost, simplicity, practicality, and convenient readout with
the naked eye.4−8 Significantly, the performance of analytical
methods can be further enhanced by introducing highly
effective sample pretreatments because they could purify and
enrich trace target analytes and reduce the interference of
impurities toward analysis results.9−11 Recently, the Mutus
research group combined such effective techniques with
colorimetric principles to design a simple static headspace-
based colorimetric assay for the detection of dissolved H2S,
which showed high selectivity, antijamming capability, and
accuracy.6 On the basis of the static headspace strategy, the
detection event can be cleverly translated from complex
mediums into simple and clean environments, thereby greatly
reducing the bad effect of complex mediums toward analytical
systems.6,9−11 However, the detection sensitivity of current
colorimetric assay systems for H2S is still far from satisfying
because the dissolved H2S level is very low in some cases (e.g.,
nanomolar level in blood sometimes).6,12 Hence, it is highly
desirable to exploit an efficient signal amplification method for
ultrasensitive colorimetric sensing of dissolved H2S by coupling
with the static headspace strategy.
Very recently, a facile target-induced nanocatalyst deactiva-
tion has proven to be promising as a signal amplification
Received: August 5, 2015
Accepted: September 1, 2015

© XXXX American Chemical Society A DOI: 10.1021/acs.analchem.5b03008

Anal. Chem. XXXX, XXX, XXX−XXX
 Analytical Chemistry
Scheme 1. Schematic Illustration of the Colorimetric Sensing for Dissolved H2S by Integrating the H2S-Induced Deactivation of
Au@TPt-NCs with the Static Headspace Strategy
Downloaded by UNIV OF SOUTH DAKOTA on September 5, 2015 | http://pubs.acs.org
Publication Date (Web): September 4, 2015 | doi: 10.1021/acs.analchem.5b03008
method for sensors and assay systems.13−16 In the target-
induced nanocatalyst deactivation, the presence of target causes
the deactivation of nanocatalysts to different degrees depending
on target concentrations, and the deactivation degrees can be
evaluated by using nanocatalyst-triggered catalytic system as a
signal amplifier, which pave a way to indicate target
concentrations. That is to say, the target-induced nanocatalyst
deactivation establishes a relation between target concen-
trations and nanocatalyst-catalyzed signal amplification systems
through the interaction between target and nanocatalysts,
thereby serving as an innovative signal transduction and
amplification method. Although this signal amplification
method has been investigated and employed very well, its
performance is still limited to some extent due to the limited
property of the used monometallic nanocatalysts. According to
the mechanism of target-induced nanocatalyst deactivation,
exploiting novel nanostructures with enhanced catalytic
property and deactivation property would be advantageous
for improving its signal amplification performance.
Core@shell nanoparticles, representing an emerging and
versatile class of nanostructures with unique catalytic, optical,
electronic, and magnetic properties, have attracted intense
attention over the past few decades owing to their extensive
potential applications in various realms of science and
technology, e.g., catalysis, energy, biomedicine, analysis and
sensing, and spectral signal enhancement.17−22 With the
astonishingly rapid development of core@shell nanoparticles
in recent years, more and more new opportunities in these
realms have emerged. Some original and innovative examples
among them are powerful and significant, e.g., core@shell
plasmonic nanostructures for high-resolution colorimetric
sensing, shell-isolated nanoparticle-enhanced Raman spectros-
copy, and (metal core)@(semiconductor shell) nanostructures
for plasmon-enhanced photocatalysis.23−25 In these works,
core@shell nanoparticles, profiting from their unique syner-
gistic structural and electronic effects, always exhibit improved/
distinctive physical and chemical properties over their
monocomponent nanoparticles, thus endowing the constructed
methods/devices with enhanced performances and unparalleled
functionalities. Inspired by the advantages of core@shell
nanoparticles, we speculated that employing an appropriate
core@shell bimetallic nanostructure to tailor its catalytic
B DOI: 10.1021/acs.analchem.5b03008
Article
property and deactivation property would be a potential
approach to design a highly efficient target-induced nano-
catalyst deactivation for signal-amplified H2S sensing.
Herein, we for the first time design a novel signal-amplified
colorimetric method for highly sensitive and selective
monitoring of dissolved H2S based on using H2S-induced
deactivation of (gold core)@(ultrathin platinum shell) nano-
catalysts (Au@TPt-NCs) as a highly efficient signal amplifica-
tion method by coupling with the static headspace strategy
(named “Au@TPt-HCA method”, Scheme 1). Upon target
dissolved H2S introduction, the H2S gas volatilized from the
test sample causes the deactivation of Au@TPt-NCs, and the
deactivation degree can be indicated by using the catalysis of
Au@TPt-NCs toward the chromogenic reaction of H2O2-
3,3′,5,5′-tetramethylbenzidine (TMB) as an amplifier system,
thus providing a way for qualitatively/quantitatively detecting
the H2S level in the sample by monitoring the change in the
colorimetric signal (the color/absorbance of H2O2-TMB
solution). Notably, the ultrathin Pt shells are very crucial for
endowing the Au@TPt-NCs with enhanced peroxidase-like
catalytic activity and deactivation efficiency, thus our H2S-
induced deactivation of Au@TPt-NCs can show powerful
signal amplification performance.
■ RESULTS AND DISCUSSION
Working Principle of the Au@TPt-HCA Method. The
working principle of the designed Au@TPt-HCA method is
schematically shown in Scheme 1. Initially, a droplet of Au@
TPt-NCs solution is coated onto the internal side of the cap of
a tube based on surface tension, which serves as the
colorimetric probe of H2S. In the absence of target dissolved
H2S, the Au@TPt-NCs do not get attacked and remain their
peroxidase-like catalytic activity, so they can catalyze the H2O2-
TMB system to produce a high colorimetric signal. When
dissolved H2S is introduced, the H2S gas can be volatilized from
the test sample in the bottom of the tube by using static
headspace sampling and then diffuses to cause the deactivation
of Au@TPt-NCs. The deactivation degree is increased with the
increment of H2S concentration and can be detected by using
the Au@TPt-NCs + (H2O2-TMB) catalytic system as a signal
amplifier. In this way, the colorimetric signal (i.e., the color/
absorbance of H2O2-TMB solution) is gradually decreased with
Anal. Chem. XXXX, XXX, XXX−XXX
 Analytical Chemistry
the increase of H2S concentration. Therefore, the change in the
color/absorbance can be used to qualitatively/quantitatively
indicate the dissolved H2S level in the sample. On the basis of
the designed protocol, the static headspace strategy cleverly
translates the detection of dissolved H2S from complex
mediums into simple and clean environments, and the H2S-
induced deactivation of Au@TPt-NCs causes the amplification
of the detectable signal. As a result, the designed method can
realize highly sensitive and selective detection.
Preparation and Characterization of Au@TPt-NCs. The
Au@TPt-NCs used in this work were prepared by simply
reducing H2PtCl6 with ascorbic acid to generate only 2-
monolayers equivalent of Pt (θPt = 2) on the surface of ∼16 nm
gold nanoparticles (AuNPs) based on the reported seed-
mediated growth method [Note: Because Pt grows as a set of
clusters (not as a set of continuous monolayers) on Au surface
according to the Volmer−Weber growth mode, we utilize
monolayer-equivalents (denoted as θPt) to approximately
Downloaded by UNIV OF SOUTH DAKOTA on September 5, 2015 | http://pubs.acs.org
describe the Pt amount on each nanoparticle].26,27 The as-
Publication Date (Web): September 4, 2015 | doi: 10.1021/acs.analchem.5b03008
prepared nanocatalysts were first characterized by UV−vis
absorption spectrophotometer. The unshelled AuNPs exhibited
a strong localized surface plasmon resonance (LSPR) peak at
520 nm (curve “a” in Figure 1C). After the AuNPs were reacted
with H2PtCl6 and ascorbic acid, the LSPR peak is blue-shifted
to 508 nm (curve “b” in Figure 1C). The blue-shift indicated
the possible presence of Pt shells on the AuNPs according to
Figure 1. (A,B) TEM images of (A) the AuNPs and (B) the Au@TPt-
NCs; (C) UV−vis absorption spectra of (a) the AuNPs and (b) the
Au@TPt-NCs; (D) size distributions of (a) the AuNPs and (b) the
Au@TPt-NCs; (E) EDX spectrum of the Au@TPt-NCs; and (F)
cyclic voltammograms of (a) AuNPs- and (b) Au@TPt-NCs-coated
glassy carbon electrodes in 0.5 M H2SO4 solution at 100 mV/s,
respectively.
C DOI: 10.1021/acs.analchem.5b03008
Article
the previous studies.28,29 Further, we also used transmission
electron microscope (TEM; TECNAI G2F20, FEI) to
investigate the as-prepared nanocatalysts. They displayed fairly
uniform dispersity with a smooth surface morphology (Figure
1B), which was similar to that of AuNPs (Figure 1A). Also their
size was slightly larger than that of AuNPs. To further precisely
investigate the size of the as-prepared nanocatalysts, we also
used “NANO MEASURER software” to analyze their size
distribution based on counting more than 200 particles. As
displayed in Figure 1D, the size of the nanocatalysts was in
principle uniform with an average value of 17.1 nm (ranging
slightly from 14 to 20 nm), which was about ∼1.3 nm larger
than that of AuNPs (15.8 nm) evidently. These investigations
indicated that there were most likely only ∼2 monolayers of Pt
coated on the AuNPs because the diameter of Pt atom is 0.28
nm.30 Meanwhile, the energy-dispersive X-ray (EDX) spectros-
copy shows that there are platinum and gold elements in the
nanocatalysts (Figure 1E). On the basis of these results and the
preparation principle, we could infer that the structure of the
as-prepared nanocatalysts was (Au core)@(Pt shell) nanostruc-
tures with ultrathin Pt shells (∼2 Pt monolayers). To confirm
the proposed nanostructures clearly, cyclic voltammetry, as a
sensitive surface analytical technique, was used to investigate
the surface behaviors of AuNPs before and after modification
with ultrathin Pt shells. A characteristic reductive peak of Au at
0.9 V could be obtained for AuNPs (curve “a” in Figure 1F),
while the reductive peak of Au was completely disappeared and
a characteristic reductive peak of Pt at 0.45 V was observed for
Au@TPt-NCs (curve “b” in Figure 1F), revealing the formation
of a complete Pt shell on the surface of AuNPs in the Au@TPt-
NCs and further confirming the proposed (Au core)@(ultra-
thin Pt shell) nanostructures for the Au@TPt-NCs.31,32
Therefore, the Au@TPt-NCs could be successfully synthesized
via the classical seed-mediated growth method.
Feasibility of the Au@TPt-HCA Method for Detecting
Dissolved H2S. Similar with the previous Pt nanostructures,
Au@TPt-NCs were highly active for catalyzing the chromo-
genic reaction of H2O2-TMB to produce oxidized TMB
(TMBox) with a blue color and an UV−vis absorption peak
at 650 nm (see Figure S2 in Part B1, Supporting
Information).33−35 So we utilized a microplate reader to
measure the change of absorbance at 650 nm in the following
studies owing to its merits (e.g., speediness and high-
throughput). In this study, the amplification of colorimetric
signal mainly derived from the H2S-induced deactivation of
Au@TPt-NCs. To realize our design, an important precondi-
tion was whether H2S could inhibit the peroxidase-like activity
of Au@TPt-NCs. To verify this point, the peroxidase-like
activities of Au@TPt-NCs before and after reaction with H2S
(Na2S as a H2S donor) were investigated. As seen from Figure
2A, the incubation of colorless Au@TPt-NCs solution (column
“a” and photograph “a”) with colorless H2O2-TMB solution
(column “b” and photograph “b”) resulted in a deep blue
solution with a strong absorbance at 650 nm (A650 nm = 1.7092,
column “d” and photograph “d”). When Au@TPt-NCs solution
was first reacted with colorless H2S solution (column “c” and
photograph “c”), almost no color and absorbance were changed
(column “e” and photograph “e”); after the addition of H2O2-
TMB solution into the above mixture, the color turned blue
and the absorbance was increased (A650 nm = 0.8810, column “f”
and photograph “f”), but it was weaker than that in the absence
of H2S (column “d”). Further increase the concentration of H2S
resulted in the decrease of the blue color intensity and the
Anal. Chem. XXXX, XXX, XXX−XXX
 Analytical Chemistry Article

in decreasing the adsorption capacity and excitation capacity of

Figure 2. (A) Control tests for the H2S-induced deactivation of Au@
TPt-NCs: (a) Au@TPt-NCs, (b) H2O2-TMB, (c) 100 nM H2S, (d)
Au@TPt-NCs + (H2O2-TMB), (e) Au@TPt-NCs + 100 nM H2S, (f)
(Au@TPt-NCs + 100 nM H2S) + (H2O2-TMB), (g) (Au@TPt-NCs
+ 500 nM H2S) + (H2O2-TMB), and (h) (Au@TPt-NCs + 1000 nM
H2S) + (H2O2-TMB); and (B) the detection of (a) 0 nM, (b) 100 nM,
(c) 200 nM, and (d) 500 nM H2S by coupling the H2S-induced
deactivation of Au@TPt-NCs with the static headspace strategy
(Inserts: the corresponding photographs). Error bar represents the
standard deviation (n = 3).
Downloaded by UNIV OF SOUTH DAKOTA on September 5, 2015 | http://pubs.acs.org
Pt nanostructures toward substrates.36−40 Further, to testify the
signal amplification capacity of the proposed H2S-induced
deactivation of Au@TPt-NCs, the relationship between the
amount of used H2S and the corresponding decrease of final
indicator (TMBox) amount was calculated (take 100 nM H2S as
an example, see Part C3 in Supporting Information for the
calculation process). Calculation results showed that the
detection of 100 nM H2S in 10 μL of solution was finally
translated to the detection of 70.79 μM TMBox in 100 μL of
solution, thus the signal was improved about 7079 times,
demonstrating the highly efficient signal amplification perform-
ance of H2S-induced deactivation of Au@TPt-NCs. The
successful development of the powerful H2S-induced deactiva-
tion of Au@TPt-NCs provided a foundation for ultrasensitive
H2S sensing based on the proposed method.
Logically, another important precondition for the Au@TPt-
HCA method was whether the H2S-induced deactivation of

Au@TPt-NCs could be smoothly progressed on the static

Publication Date (Web): September 4, 2015 | doi: 10.1021/acs.analchem.5b03008

headspace platform. To clarify this issue, the H2S-induced

absorbance intensity (columns “g” and “h”, photographs “g”
and “h”). There results indicated that the peroxidase-like
activity of Au@TPt-NCs was decreased with the increase of
H2S concentration, confirming that H2S could strongly inhibit
the peroxidase-like activity of Au@TPt-NCs. The reason was
that H2S could be easily absorbed on the surface of Pt
nanostructures to block the active sites of Pt nanostructures
and change the electronic state of the active sites, thus resulting
deactivation of Au@TPt-NCs was used to detect H2S with a
series of concentration gradients (0, 100, 200, and 500 nM
were used as examples.) by coupling with the static headspace
strategy. As shown from Figure 2B, both the color intensity and
the absorbance intensity decreased with the increasing H2S
concentration, indicating that the H2S-induced deactivation of
Au@TPt-NCs could be successfully introduced into the static
headspace platform. In addition, the comparison study of the

Figure 3. (A−H) TEM images of Au@PtNPs with various thickness: (A) θPt = 1, (B) θPt = 3, (C) θPt = 4, (D) θPt = 5, (E) θPt = 6, (F) θPt = 8, (G)
θPt = 10, and (H) θPt = 15 (insets, the corresponding size distributions based on counting more than 200 particles); (I) comparison studies of the
developed method for the detection of 0 nM (blue column) and 100 nM (yellow column) H2S by using AuNPs, PtNPs, and Au@TPt-NCs (a) and
Au@PtNPs with various thickness (θPt = 1, 2, 3, 4, 5, 6, 8, 10, and 15) (b); (J,K) the detailed analysis of part I(b), the corresponding signal
intensity(A0 − A50, J), catalytic activity (A0, K(a)) and deactivation efficiency ((A0 − A50)/A0, K(b)) versus θPt, respectively. Error bar represents the
standard deviation (n = 3).

D DOI: 10.1021/acs.analchem.5b03008
Anal. Chem. XXXX, XXX, XXX−XXX
 Analytical Chemistry
Au@TPt-HCA method with and without the static headspace
strategy showed that the static headspace strategy could greatly
improve the selectivity and antijamming capability of the assay
system, thereby enhancing the accuracy and precision of the
analytical results, which showed its importance for improving
the analytical performance (see Figure S3 in Part B1 in
Supporting Information for the details).
Advantages of Au@TPt-NCs for H2S-Induced Nano-
catalyst Deactivation and Density Functional Theory
(DFT) Study. The signal amplification performance of target-
induced nanocatalyst deactivation is very crucial for improving
the sensitivity of assays. Theoretically, according to the
mechanism of target-induced nanocatalyst deactivation, exploit-
ing a nanocatalyst with excellent catalytic property and
deactivation property would be advantageous to ensure its
efficient signal amplification performance. In the present study,
Au@TPt-NCs were used as the nanocatalyst for H2S-induced
nanocatalyst deactivation. To highlight the potential of Au@
Downloaded by UNIV OF SOUTH DAKOTA on September 5, 2015 | http://pubs.acs.org
TPt-NCs for enhancing the signal amplification performance of
Publication Date (Web): September 4, 2015 | doi: 10.1021/acs.analchem.5b03008
H2S-induced nanocatalyst deactivation in the colorimetric
assay, some control tests using Pt/Au nanostructures with
different compositions and shell thicknesses were implemented
for detecting H2S (50 nM as an example) under the same
conditions three times per nanostructures (see Part A7 in
Supporting Information for the detailed experimental proce-
dures). The comparison was based on the change of the
absorbance of H2O2-TMB solution at 650 nm relative to zero
analyte (i.e., (A0 − A50), where Ax is the absorbance of H2O2-
TMB solution at x nM H2S condition.). In addition, the
catalytic activity and deactivation efficiency of these Au/Pt
nanostructures were also compared to better clarify this
potential of Au@TPt-NCs. The comparison of the catalytic
activity and deactivation efficiency were based on the
absorbance of H 2 O 2 -TMB solution obtained by pure
nanostructures (i.e., A0, in the absence of analyte) and the
ratio of absorbance decrement caused by 50 nM H2S (i.e., (A0
− A50)/A0) under the same conditions, respectively. Therefore,
the relation between the signal intensity (A0 − A50) and the
catalytic activity (A0) and the deactivation efficiency ((A0 −
A50)/A0) of nanostructures could be summarized as
(A 0 − A50) = A 0 × ((A 0 − A50)/A 0),
i.e., (signal intensity)
= (catalytic activity) × (deactivation efficiency)
which was in accordance with the mechanism of target-induced
nanocatalyst deactivation: the better the catalytic activity and
deactivation efficiency are, the higher the signal amplification
performance (i.e., signal intensity) is.
A characteristic of the Au@TPt-NCs was their core@shell
bimetallic nanostructures. To verify the advantages of this
structure, the signal amplification behavior of the H2S-induced
deactivation of Au@TPt-NCs was investigated and compared
with those of monometallic AuNPs (∼16 nm as an example;
TEM image, Figure 1A) and Pt nanoparticles (PtNPs, ∼3.5 nm
as an example; TEM image, Figure S4 in Part B1, Supporting
Information). As seen from Figure 3I(a), the use of AuNPs for
H2S-induced nanocatalyst deactivation could not provide any
signal toward 50 nM H2S (A0 − A50(AuNPs) ≈ 0). Since this
dosage of AuNPs could not catalyze the H2O2-TMB system to
produce colorimetric signal (A0(AuNPs) = 0.0530, which
equaled to that of the blank H2O2-TMB solution), the low
E DOI: 10.1021/acs.analchem.5b03008
Article
signal intensity was obviously ascribed to the ultralow
peroxidase-like catalytic activity of AuNPs. Also the signal
intensity by using PtNPs was very low as well (A0 −
A50(PtNPs) = 0.0183). Different from AuNPs, PtNPs showed
high peroxidase-like catalytic activity (A0(PtNPs = 0.8819)
under the same surface area. Although their high catalytic
activity made them able to provide a strong background signal,
their low deactivation efficiency ((A0 − A50)/A0(PtNPs) =
2.08%) made them impossible to provide high signal intensity
toward 50 nM H2S. In contrast, when using Au@TPt-NCs, the
signal intensity was as high as 0.6340 dramatically (A0 −
A50(Au@TPt-NCs) = 0.6340). Au@TPt-NCs could exhibit not
only relatively high peroxidase-like catalytic activity (A0(Au@
TPt-NCs) = 1.7673) but also astonishingly high deactivation
efficiency [(A0 − A50)/A0(Au@TPt-NCs) = 35.87%], thus
leading to the high signal intensity. Consequently, the signal
intensity by using Au@TPt-NCs was much higher than those
by using AuNPs and PtNPs, indicating that the H2S-induced
deactivation of Au@TPt-NCs possessed higher signal amplifi-
cation performance and further demonstrating the advantages
of core@shell bimetallic nanostructures. The reason was
obviously due to their high catalytic activity and high
deactivation efficiency stemmed from the synergistic struc-
tural/electronic effects between Au nanocores and ultrathin Pt
nanoshells in the core@shell nanostructures, which mono-
metallic AuNPs/PtNPs could not possess simultaneously.
Another characteristic of the Au@TPt-NCs was their
ultrathin Pt shells. To demonstrate that the use of ultrathin
Pt shells was essential for amplifying the signal of H2S-induced
deactivation of (Au core)@(Pt shell) nanoparticles (Au@
PtNPs), the signal intensity, catalytic activity, and deactivation
efficiency of Au@PtNPs with various thickness (θPt = 1, 2, 3, 4,
5, 6, 8, 10, and 15, TEM images: Figure 1B and Figure 3A−H)
were investigated (Figure 3I(b), J, K). As shown in Figure 3J,
the signal intensity versus θPt displayed a volcano-shaped
dependence, and the maximum signal could be obtained at θPt
= 2−4, confirming that the use of the ultrathin Pt shells (θPt =
2−4) was important for enhancing the signal amplification
performance of H2S-induced deactivation of Au@PtNPs.
Meanwhile, their catalytic activities increased rapidly with the
increase of θPt and reached the maximum when θPt = 2−4 but
decreased after θPt = 4 (curve “a” in Figure 3K); their
deactivation efficiencies stayed almost the same when θPt = 1−4
but also decreased rapidly after θPt = 4 (curve “b” in Figure 3K).
So the superior signal amplification performance of H2S-
induced deactivation of Au@PtNPs with ultrathin Pt shells (θPt
= 2−4) was mainly owing to their higher catalytic activity and
higher deactivation efficiency over those of other Au@PtNPs.
The change in catalytic activity and deactivation efficiency seen
with changing Pt shell thickness from θPt = 1 to 15 could be
attributed to the change of synergistic structural/electronic
effects between Au nanocores and Pt nanoshells. In addition,
considering the cost of the assay, Au@TPt-NCs (θPt = 2) with
their less consumption of Pt element among Au@PtNPs (θPt =
2−4) was chosen for H2S-induced deactivation of Au@PtNPs
in this work.
In order to deeply understand the synergistic electronic
effects between Au nanocores and ultrathin Pt nanoshells in
Au@TPt-NCs and clearly reveal the advantages of Au@TPt-
NCs in catalytic activity and deactivation efficiency, DFT
calculations were performed using the Cambridge Serial Total
Energy Package (CASTEP) codes.41 We utilized the (111)
crystal face as a model to study the surface electronic structures
Anal. Chem. XXXX, XXX, XXX−XXX
 Analytical Chemistry
of different Pt monolayers on the Au surface. Figure 4A displays
the partial density of state (PDOS) profiles of 5d-orbitals for
Downloaded by UNIV OF SOUTH DAKOTA on September 5, 2015 | http://pubs.acs.org
Publication Date (Web): September 4, 2015 | doi: 10.1021/acs.analchem.5b03008
Figure 4. (A,B) PDOS profiles (A) and d-band centers (B, relative to
the Fermi level (EF)) of 5d-orbitals for the outmost atoms of Au
(111), Pt (111), and Au@Pt (111) with different Pt monolayers (1, 2,
3, 4, 5, 6, 7, 11); and (C) relationship between the d-band center of
Au@Pt (111) and the number of Pt monolayers. Note: the Fermi level
is set as zero, and “Au@Pt(111)-x” is defined for representing the
Au@Pt (111) with x Pt monolayers.
the outmost atoms of pure Au (111), pure Pt (111), and Au@
Pt (111) with different Pt monolayers (1, 2, 3, 4, 5, 6, 7, 11).
According to the d-band center theory proposed by Nørskov et
al., the adsorption ability and catalytic activity of transition
metal catalysts are strongly dependent on the d-band center of
their surface atoms: (i) the higher the d-band center is, the
stronger the metal−adsorbate binding strength is and (ii) the
activity of metal catalysts versus d-band center generally
exhibits a classical volcano-shaped dependence, thus an active
metal catalyst should have a d-band center with intermediate
value (neither too high or too low).42−46 So we calculated and
compared the d-band centers of Au (111), Pt (111), and Au@
Pt (111) with different Pt monolayers based on the
corresponding PDOS profiles, respectively (Figure 4B). For
the comparison of Au (111) and Pt (111), the d-band center of
Pt (111) was much higher than that of Au (111) and compared
with the d-band of Au (111), the d-band of Pt (111) was
broader and had a much higher density near and above the
Fermi level. These results indicated the fact that Pt usually has a
higher chemical reactivity than that of Au.47,48 This was the
reason that the peroxidase-like activity of PtNPs was much
higher than that of AuNPs. For the comparison of Pt (111) and
Au@Pt (111) with different Pt monolayers, the d-band center
of Au@Pt (111) decreased and tended to approach the d-band
center of pure Pt (111) with the increasing number of Pt
monolayers (Figure 4C). These results showed that Au@Pt
(111) with 2−7 Pt monolayers could possess a higher catalytic
F DOI: 10.1021/acs.analchem.5b03008
Article
activity due to their intermediate values of d-band center, and
the adsorption capacity of surface Pt atoms on the Au@Pt
(111) decreased with the increasing number of Pt monolayers,
which were similar to the experimental results that the Au@
PtNPs with θPt = 2−4 exhibited the highest peroxidase-like
catalytic activity, and the deactivation efficiency of Au@PtNPs
by absorbing H2S decreased with the increment of Pt shell
thickness (Figure 3K). These results confirmed that the
decrease of d-band center (i.e., the change of surface electronic
structures) of Au@PtNPs with the increase of Pt shell thickness
resulted in the change of their catalytic activities and
deactivation efficiencies and further demonstrated that Au@
TPt-NCs (θPt = 2−4) possessing an optimal catalytic activity
and deactivation efficiency owing to their appropriate values of
the d-band center, outperformed AuNPs, PtNPs, and other
Au@PtNPs in enhancing the signal amplification performance
of H2S-induced nanocatalyst deactivation. In addition, since the
surface geometrical structure is also another important factor
affecting the chemical activity of metal nanocatalysts, some
differences between the experimental results and theoretical
results based on electronic effects might be attributed to the
change of surface geometrical structures in Au@PtNPs.42−44
Analytical Performance of Au@TPt-HCA Method.
Under the optimal conditions (see Figures S5, S6, and S7 in
Part B1 in Supporting Information for detailed results), the
sensitivity and dynamic measurement range of the developed
Au@TPt-HCA method were evaluated against H2S standards
with various concentrations. As seen from Figure 5A, the
detection solution on the cap of tube presented distinct color
changes from deep blue to light blue and to colorless when the
concentration of H2S solution increased from 0 to 300 nM,
which could be easily distinguished by using the naked eye,
thereby providing a convenient way for qualitatively monitoring
H2S. Inspiringly, even the low concentration of H2S (50 nM)
could generate an appreciable change of color intensity, so the
resolving power of the Au@TPt-HCA method by using the
naked eye could reach down to the 50 nM H2S level. Moreover,
the color change could also be identified by using digital camera
images for quantitative analysis of H2S (see Figure S8 in Part
B1, Supporting Information).49,50 Further, we used a micro-
plate reader to quantitatively detect the H2S standards based on
the change of absorbance intensity at 650 nm, and the
absorbance intensity decreased with the increase of the H2S
level form 0 to 300 nM (Figure 5B). The calibration plots
displayed a good linear relationship between the decrease of the
absorbance intensity and the H2S concentration in the range
from 10 nM to 100 nM with a high sensitivity of 0.0130 nM−1
and a low detection limit (LOD) of 7.5 nM H2S based on 3σb/
slope, where σb was the standard deviation of blank samples (n
= 12, Figure 5C). The LOD was obviously lower than those of
the previously reported methods (Table S3 in Part B2,
Supporting Information). The high sensitivity and the low
LOD were mainly attributed to the highly efficient signal
amplification performance of H2S-induced deactivation of Au@
TPt-NCs. More importantly, because the reported concen-
tration of biological sulfides is usually in a vast range from the
nanomolar to micromolar level,6,12 and the water quality
standards for sulfide level in various waters is in the range of
0.02−1 mg L−1 (0.63−31 μM) according to the State Standard
of the People’s Republic of China (Table S4 in Part B2,
Supporting Information), the developed colorimetric assay
could completely meet the requirement of H2S analysis in
biological samples and various water samples. Moreover, the
Anal. Chem. XXXX, XXX, XXX−XXX
 Analytical Chemistry
Downloaded by UNIV OF SOUTH DAKOTA on September 5, 2015 | http://pubs.acs.org
Publication Date (Web): September 4, 2015 | doi: 10.1021/acs.analchem.5b03008
Figure 5. (A,B) Photographs (A) and absorbance intensities at 650
nm (B) of the Au@TPt-HCA method toward H2S standards with
different concentrations, (C) calibration plots of the Au@TPt-HCA
method based on the decrease of the absorbance intensities at 650 nm
toward different-concentration H2S standards (inset, linear curve), and
(D) specificity of the Au@TPt-HCA method against H2S (300 nM)
and interfering substances (3 μM) including Cys, GSH, DTT, BSA,
IgG, F−, Cl−, Br−, HCO3−, NO3−, SO32−, SO42−, Na+, K+, Ca2+, Mg2+,
and NH4+. Error bar represents the standard deviation (n = 3).
assay was very inexpensive despite the used nanocatalysts were
made of Au and Pt. The expense for all the involved chemicals
in one sample test is estimated to be US$0.002 (see Table S5
and S6 in Part B2 in the Supporting Information for the
calculation process). These results showed that the developed
Au@TPt-HCA method was a simple, ultrasensitive, convenient,
and cost-effective approach for dissolved H2S sensing.
Next, the specificity of the Au@TPt-HCA method was
evaluated by challenging the system with other interfering
substances, including biological thiols (e.g., Cys, glutathione
(GSH), and dithiothreitol (DTT)), proteins (e.g., bovine
serum albumin (BSA) and immunoglobulin G (IgG)), and
common ions (e.g., F−, Cl−, Br−, HCO3−, NO3−, SO32−, SO42−,
Na+, K+, Ca2+, Mg2+, NH4+). As shown in Figure 5D, a
significant decrease in the absorbance was observed only with
the target H2S (300 nM), revealing that the present sensing
system exhibited excellent selectivity for the monitoring of
dissolved H2S. In addition, the reproducibility and precision of
the Au@TPt-HCA method were investigated to be acceptable
(Table S7 in Part B2, Supporting Information).
Analysis of Real Samples and Evaluation of Method
Accuracy. To investigate the application feasibility of the
developed method for detecting real samples, we applied our
method to determine H2S levels in 4 sample matrixes including
tap water, Minjiang river water, seawater, and newborn cattle
serum. The recovery experiments were carried out by using
Na2S as the donor to spike 4 blank samples (20, 40, 60, and 80
G DOI: 10.1021/acs.analchem.5b03008
Article
nM H2S used as examples). Initially, these samples were diluted
to 2-fold with pH 6.0 PBS containing 5 mM ethyl-
enediaminetetraacetic acid (EDTA). Then, Na2S donors with
different concentrations were spiked into the diluted samples,
respectively. The obtained samples were tested by using the
developed strategy. The H2S levels in the samples were
calculated referring to the regression equation (y = 0.0130x −
0.0778) in Figure 5C. The results are shown in Table S8
(Supporting Information, Part B2). The recovery for all the
samples was in the range from 91.6% to 106.7%. The results
strongly validated that the Au@TPt-HCA method could be
considered as a reliable approach for detection of dissolved H2S
in real samples.
■ CONCLUSIONS
In summary, we have successfully developed a simple and novel
signal-amplified colorimetric assay for highly sensitive and
selective detection of dissolved H2S by integrating H2S-induced
deactivation of Au@TPt-NCs with static headspace strategy.
The two components in the colorimetric assay system play
different important roles: (i) the static headspace strategy was
demonstrated to be effective as a sample pretreatment method
in this system; and (ii) the H2S-induced deactivation of Au@
TPt-NCs was systematically proved to be powerful as a signal
amplification method by experimental results and DFT results
owing to the enhanced peroxidase-like catalytic activity and
deactivation efficiency of Au@TPt-NCs stemmed from the
synergistic structural/electronic effects between Au nanocores
and ultrathin Pt nanoshells. Our results convincingly
demonstrate that the combination of these two components
endow the developed colorimetric method with several
advantages when compared with the conventional and reported
colorimetric methods. First of all, the method can show high
selectivity and antijamming capability owing to the use of static
headspace sampling, thus offering high accuracy and precision.
Second, the method can display a high sensitivity with a low
LOD of 7.5 nM due to the highly efficient signal amplification
capacity of H2S-induced deactivation of Au@TPt-NCs. Third,
the system can be simply carried out in a Eppendorf tube with
only two-steps experimental operation and without the need of
sophisticated instrumentation, thus it is very convenient,
portable, and inexpensive. At last, the developed method is
suitable for sensing dissolved H2S in various complex sample
matrixes from water samples to biological fluids. In the view of
high sensitivity and selectivity, simplification, low cost, user-
friendliness, and visual readout with the naked eye, the
colorimetric method reported here has the potential to be
served as a novel detection kit for point-of-care testing. Future
work would be focused on the design of novel target-induced
nanocatalyst deactivation as a powerful signal amplification
method for the detection of other target analytes by controlling
the catalytic activity and deactivation efficiency of the used
nanocatalysts.
■
ASSOCIATED CONTENT
*
S Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.anal-
chem.5b03008.
Experimental section, Tables S1−S8, Figures S1−S8, and
calculation processes (PDF)
Anal. Chem. XXXX, XXX, XXX−XXX
 Analytical Chemistry Article

■ AUTHOR INFORMATION
Corresponding Author
(31) Zou, S.; Weaver, M. J. Anal. Chem. 1998, 70, 2387−2395.
(32) Dai, Y.; Chen, S. ACS Appl. Mater. Interfaces 2015, 7, 823−829.
(33) Gao, Z.; Xu, M.; Hou, L.; Chen, G.; Tang, D. Anal. Chim. Acta
*Phone: +86-591-2286 6125. Fax: +86-591-2286 6135. E-mail: 2013, 776, 79−86.
dianping.tang@fzu.edu.cn. (34) Sun, X.; Guo, S.; Chung, C.; Zhu, W.; Sun, S. Adv. Mater. 2013,
Notes 25, 132−136.
The authors declare no competing financial interest. (35) Gao, Z.; Xu, M.; Lu, M.; Chen, G.; Tang, D. Biosens. Bioelectron.

■
2015, 70, 194−201.
ACKNOWLEDGMENTS (36) Bernardi, F.; Traverse, A.; Olivi, L.; Alves, M. C. M.; Morais, J. J.
Phys. Chem. C 2011, 115, 12243−12249.
We acknowledge support from the NSFC of China (Grants (37) Bartholomew, C. H.; Agrawal, P. K.; Katzer, J. R. Adv. Catal.
41176079 and 21475025), the NSF of Fujian Province (Grant 1982, 31, 135−242.
2014J07001), and the Alexander von Humboldt-Foundation of (38) Bartholomew, C. H. Appl. Catal., A 2001, 212, 17−60.
Germany. (39) Oudar, J. Catal. Rev.: Sci. Eng. 1980, 22, 171−195.

■
REFERENCES
(1) Kong, M.; Salin, E. Anal. Chem. 2012, 84, 10038−10043.
(2) Xiong, B.; Zhou, R.; Hao, J.; Jia, Y.; He, Y.; Yeung, E. Nat.
Commun. 2013, 4, 1708.
(3) Lin, V.; Chen, W.; Xian, M.; Chang, C. Chem. Soc. Rev. 2015, 44,
Downloaded by UNIV OF SOUTH DAKOTA on September 5, 2015 | http://pubs.acs.org
(40) Ouder, J.; Wise, H. Deactivation and Poisoning of Catalysts;
Marcel Dekker: New York, 1985.
(41) Clark, S. J.; Segall, M. D.; Pickard, C. J.; Hasnip, P. J.; Probert,
M. I. J.; Refson, K.; Payne, M. C. Z. Kristallogr. 2005, 220, 567−570.
(42) Nørskov, J. K.; Bligaard, T.; Rossmeisl, J.; Christensen, C. H.
Nat. Chem. 2009, 1, 37−46.

(43) Nørskov, J. K.; Bligaard, T.; Hvolbæk, B.; Abild-Pedersen, F.;

4596−4618.
Publication Date (Web): September 4, 2015 | doi: 10.1021/acs.analchem.5b03008

Chorkendorff, I.; Christensen, C. H. Chem. Soc. Rev. 2008, 37, 2163−

(4) Fischer, E. Ber. Dtsch. Chem. Ges. 1883, 16, 2234−2236. 2171.
(5) Fogo, J.; Popowsky, M. Anal. Chem. 1949, 21, 732−734. (44) Greeley, J.; Nørskov, J. K.; Mavrikakis, M. Annu. Rev. Phys.
(6) Jarosz, A.; Yep, T.; Mutus, B. Anal. Chem. 2013, 85, 3638−3643. Chem. 2002, 53, 319−348.
(7) Adarsh, N.; Krishnan, M. S.; Ramaiah, D. Anal. Chem. 2014, 86, (45) Zhang, J.; Vukmirovic, M. B.; Xu, Y.; Mavrikakis, M.; Adzic, R.
9335−9342. R. Angew. Chem., Int. Ed. 2005, 44, 2132−2135.
(8) Song, Y.; Wei, W.; Qu, X. Adv. Mater. 2011, 23, 4215−4236. (46) Shao, M.; Liu, P.; Zhang, J.; Adzic, R. R. J. Phys. Chem. B 2007,
(9) Nerin, C. Anal. Bioanal. Chem. 2007, 388, 1001−1002. 111, 6772−6775.
(10) Nerin, C.; Salafranca, J.; Aznar, M.; Batlle, R. Anal. Bioanal. (47) Sholl, D. S.; Steckel, J. A. Density Functional Theory: A Practical
Chem. 2009, 393, 809−833. Introduction; John Wiley & Sons, Inc.: Hoboken, NJ, 2009.
(11) Deng, Z.; Chen, X.; Wang, Y.; Fang, E.; Zhang, Z.; Chen, X. (48) Mirkin, M. V.; Amemiya, S. Nanoelectrochemistry; CRC Press:
Anal. Chem. 2015, 87, 633−640. Boca Raton, FL, 2014.
(12) Peng, H.; Cheng, Y.; Dai, C.; King, A. L.; Predmore, B. L.; Lefer, (49) Steiner, M. S.; Meier, R. J.; Duerkop, A.; Wolfbeis, O. S. Anal.
D. L.; Wang, B. Angew. Chem., Int. Ed. 2011, 50, 9672−9675. Chem. 2010, 82, 8402−8405.
(13) Lien, C.; Tseng, Y.; Huang, C.; Chang, H. Anal. Chem. 2014, 86, (50) Wang, X.; Meier, R. J.; Link, M.; Wolfbeis, O. S. Angew. Chem.,
2065−2072. Int. Ed. 2010, 49, 4907−4909.
(14) Wu, G.; He, S.; Peng, H.; Deng, H.; Liu, A.; Lin, X.; Xia, X.;
Chen, W. Anal. Chem. 2014, 86, 10955−10960.
(15) Deng, H.; Weng, S.; Huang, S.; Zhang, L.; Liu, A.; Lin, X.; Chen,
W. Anal. Chim. Acta 2014, 852, 218−222.
(16) Li, W.; Chen, B.; Zhang, H.; Sun, Y.; Wang, J.; Zhang, J.; Fu, Y.
Biosens. Bioelectron. 2015, 66, 251−258.
(17) Chaudhuri, R. G.; Paria, S. Chem. Rev. 2012, 112, 2373−2433.
(18) Zhang, Q.; Lee, I.; Joo, J. B.; Zaera, F.; Yin, Y. Acc. Chem. Res.
2013, 46, 1816−1824.
(19) Chen, G.; Ågren, H.; Ohulchanskyy, T. Y.; Prasad, P. N. Chem.
Soc. Rev. 2015, 44, 1680−1713.
(20) Lucky, S. S.; Soo, K. C.; Zhang, Y. Chem. Rev. 2015, 115, 1990−
2042.
(21) Hao, J.; Xiong, B.; Cheng, X.; He, Y.; Yeung, E. S. Anal. Chem.
2014, 86, 4663−4667.
(22) Wang, Y.; Yan, B.; Chen, L. Chem. Rev. 2013, 113, 1391−1428.
(23) Gao, Z.; Deng, K.; Wang, X.; Miro, M.; Tang, D. ACS Appl.
Mater. Interfaces 2014, 6, 18243−18250.
(24) Li, J. F.; Huang, Y. F.; Ding, Y.; Yang, Z. L.; Li, S. B.; Zhou, X.
S.; Fan, F. R.; Zhang, W.; Zhou, Z. Y.; Wu, D. Y.; Ren, B.; Wang, Z. L.;
Tian, Z. Q. Nature 2010, 464, 392−395.
(25) Li, B.; Gu, T.; Ming, T.; Wang, J.; Wang, P.; Wang, J.; Yu, J. C.
ACS Nano 2014, 8, 8152−8162.
(26) Li, J.; Yang, Z.; Ren, B.; Liu, G.; Fang, P.; Jiang, Y.; Wu, D.;
Tian, Z. Langmuir 2006, 22, 10372−10379.
(27) Fang, P.; Duan, S.; Lin, X.; Anema, J. R.; Li, J.; Buriez, O.; Ding,
Y.; Fan, F.; Wu, D.; Ren, B.; Wang, Z. L.; Amatore, C.; Tian, Z. Chem.
Sci. 2011, 2, 531−539.
(28) Chen, H.; Wang, F.; Li, K.; Woo, K.; Wang, J.; Li, Q.; Su, L.;
Zhang, X.; Lin, H.; Yan, C. ACS Nano 2012, 6, 7162−7171.
(29) Wang, F.; Shen, Y. Phys. Rev. Lett. 2006, 97, 206806.
(30) Wakisaka, M.; Asizawa, S.; Uchida, H.; Watanabe, M. Phys.
Chem. Chem. Phys. 2010, 12, 4184−4190.

H DOI: 10.1021/acs.analchem.5b03008
Anal. Chem. XXXX, XXX, XXX−XXX