Millifluidics for time-resolved mapping of the growth of gold nanostructures

Article abstract of DOI:10.1021/ja400434c

Innovative in situ characterization tools are essential for understanding the reaction mechanisms leading to the growth of nanoscale materials. Though techniques, such as in situ transmission X-ray microscopy, fast single-particle spectroscopy, small-angle X-ray scattering, etc., are currently being developed, these tools are complex, not easily accessible, and do not necessarily provide the temporal resolution required to follow the formation of nanomaterials in real time. Here, we demonstrate for the first time the utility of a simple millifluidic chip for an in situ real time analysis of morphology and dimension-controlled growth of gold nano- and microstructures with a time resolution of 5 ms. The structures formed were characterized using synchrotron radiation-based in situ X-ray absorption spectroscopy, 3-D X-ray tomography, and high-resolution electron microscopy. These gold nanostructures were found to be catalytically active for conversion of 4-nitrophenol into 4-aminophenol, providing an example of the potential opportunities for time-resolved analysis of catalytic reactions. While the investigations reported here are focused on gold nanostructures, the technique can be applied to analyze the time-resolved growth of other types of nanostructured metals and metal oxides. With the ability to probe at least a 10-fold higher concentrations, in comparison with traditional microfluidics, the tool has potential to revolutionize a broad range of fields from catalysis, molecular analysis, biodefense, and molecular biology.

Full text of DOI:10.1021/ja400434c

Article
pubs.acs.org/JACS
Millifluidics for Time-resolved Mapping of the Growth of Gold
Nanostructures
Katla Sai Krishna,†,∥ Chelliah V. Navin,†,‡ Sanchita Biswas,† Varshni Singh,† Kyungmin Ham,†
G. Lisa Bovenkamp,† Chandra S. Theegala,‡ Jeffery T. Miller,§ James J. Spivey,∥
and Challa S. S. R. Kumar*,†,∥
†Center for Advanced Microstructures and Devices (CAMD), Louisiana State University, Baton Rouge, Louisiana 70806, United
States
‡Department of Biological and Agricultural Engineering, and LSU AgCenter, Louisiana State University, Baton Rouge, Louisiana
70803, United States
§Argonne National Laboratory, 9700 South Cass Avenue, CSE, Argonne, Illinois 60439-4837, United States
∥Center for Atomic-Level Catalyst Design, #324, Cain Department of Chemical Engineering, Louisiana State University, Baton
Rouge, Louisiana 70803, United States
S
* Supporting Information
ABSTRACT: Innovative in situ characterization tools are
essential for understanding the reaction mechanisms leading to
the growth of nanoscale materials. Though techniques, such as
in situ transmission X-ray microscopy, fast single-particle
spectroscopy, small-angle X-ray scattering, etc., are currently
being developed, these tools are complex, not easily accessible,
and do not necessarily provide the temporal resolution
required to follow the formation of nanomaterials in real
time. Here, we demonstrate for the first time the utility of a
simple millifluidic chip for an in situ real time analysis of
morphology and dimension-controlled growth of gold nano- and microstructures with a time resolution of 5 ms. The structures
formed were characterized using synchrotron radiation-based in situ X-ray absorption spectroscopy, 3-D X-ray tomography, and
high-resolution electron microscopy. These gold nanostructures were found to be catalytically active for conversion of 4-
nitrophenol into 4-aminophenol, providing an example of the potential opportunities for time-resolved analysis of catalytic
reactions. While the investigations reported here are focused on gold nanostructures, the technique can be applied to analyze the
time-resolved growth of other types of nanostructured metals and metal oxides. With the ability to probe at least a 10-fold higher
concentrations, in comparison with traditional microfluidics, the tool has potential to revolutionize a broad range of fields from
catalysis, molecular analysis, biodefense, and molecular biology.
micrometer dimensions in which the reactive liquid precursors
needed to produce solid clusters of metals are mixed and
allowed to react. Recently, they have become an attractive
technology due to their ability to rapidly mix reagents, provide
homogeneous reaction environments, continuously vary
reaction conditions, add reagents at precise time intervals
during reaction, and the ability to control the residence time by
varying the reactant flow-rates and/or the length of the flow
channel.5−7 These features have been cleverly utilized in the
wet-chemical synthesis of nanomaterials not only to control
their size, size-distribution, and shape8 but also to control their
crystal structure9 and for faster clinical translation.10
INTRODUCTION
■
A number of synchrotron radiation-based X-ray character-
ization tools1,2 are used for both chemical and physical
characterization of materials, especially in a liquid environment.
Among these, X-ray absorption spectroscopy (XAS) is a
versatile tool to probe oxidation states, coordination, and local
order of materials, even in the absence of a crystalline lattice.
XAS has been used to investigate the electronic and geometric
structures of nanomaterials; more specifically, to map local
coordination number, the extent of alloying and composition of
multimetallic systems in solution.3 Further, in situ XAS offers an
opportunity to observe time-resolved bond formation, changes
in oxidation states, coordination, and local order.4 The ability to
follow the fundamental processes in the synthesis of nano-
structured materials with atomic precision is an unfulfilled need
in the field of nanoscience.
The use of XAS in conjunction with microfluidic systems
provides a powerful new method for in situ time-resolved
experiments probing the structure and reaction dynamics at
Microfluidic systems have been used to prepare a wide range
of materials. These systems consist of small channels of
Received: January 14, 2013
Published: March 15, 2013
© 2013 American Chemical Society
5450
dx.doi.org/10.1021/ja400434c | J. Am. Chem. Soc. 2013, 135, 5450−5456

Journal of the American Chemical Society
Article
Figure 1. (a) Millifluidic chip marked with different zones where in situ XAS was performed. (b) In-situ XAS analysis at different zones within the
millifluidic channel. (c) Reaction scheme of precursors. (d) XANES spectra showing the Au L3-edge at zones 3 (red), 5 (blue) and at zone 5 after 12
h (black). (e) EXAFS spectra at the same zones. (f) EXAFS of Au foil (black) and sample after NaBH4 reduction (red); () Fourier transform
magnitude and (- - -) imaginary component of the Fourier transform.
atomic level.11 However, requirement for a sample concen-
tration in excess of 0.1 M to obtain high-quality data has been a
major drawback in utilizing microfluidic chips for in situ XAS
analysis.7 We have recently introduced the concept of chip-
based millifluidics, which are very different from tubular
millifluidics,12 and demonstrated their advantages over tradi-
tional microfluidics for higher throughput controlled synthesis
of ultrasmall nanoclusters.13,14 Continuing from these inves-
tigations, herein, we demonstrate a step-up advance in utilizing
this technique, which is simple and easily accessible unlike
traditional microfluidics, as a probe for mapping time-resolved
growth of nanomaterials. We show for the first time an in situ
XAS analysis, in real time, of morphology-controlled growth of
gold nanostructures within a millifluidic chip reactor. The
reaction sequence of gold nanostructure formation with time
resolution has been mapped. The central theme of these novel
investigations is the utility of millifluidics for mapping the time-
resolved chemistry of the growth of metallic structures in
general and catalysts in particular. We also show continuous
flow catalytic activity of the as-formed gold nanostructures for
reduction of 4-nitrophenol and ferricyanide. While micro-
fluidics-based continuous flow catalysis of gold nanoparticles
impregnated on alumina has been previously utilized for
synthesis of polypyridine derivatives,15 the ability to control the
dimension, and morphology of the embedded gold nano-
structured catalysts within continuous flow channels was never
explored before. The results reported are anticipated to lead to
superior continuous flow catalysis applications. With the
possibility to embed atomically precise catalysts, these tools
can revolutionize catalysis from the point of view of practical as
well as fundamental investigations.16−18 Such systems are also
likely to lead to advances in the field of biosensing,19
electrophoresis,20 and enhanced optical detection.21 Addition-
ally, the flower-like gold morphology obtained through this
formation process has applications in surface-enhanced Raman
spectroscopy,22 catalysis,23 bioimaging,24 and superhydrophobic
coatings.25
RESULTS AND DISCUSSION
■
The millifluidic chip used in this study consists of serpentine
channels with dimensions of 2 mm (W) × 0.15 mm (H) × 220
mm (L). Each chip had two inputs and a single output which
connects them to a manifold (Figures 1 and S1). Fully
automated syringe pumps were programmed to dispense the
reagent solutions through the chip at required flow-rates
through a computer controlled software. A hand-held
millifluidic device purchased from Millifluidica, LLC (www.
millifluidica.com) was also found to be useful for investigating
these reactions.
After optimizing the flow conditions, the precursor reagents
(chloroauric acid (HAuCl4) and meso-2, 3-dimercapto succinic
acid (DMSA)) were dispensed within the chip at a flow-rate of
10 mL/h and allowed to react at zone 1 (Figure 1a,c). At the
flow-rates utilized, the Reynolds numbers are low (for example,
5451
dx.doi.org/10.1021/ja400434c | J. Am. Chem. Soc. 2013, 135, 5450−5456

Journal of the American Chemical Society
Article
Table 1. Results of the in Situ Au EXAFS Analysis
sample
Au+3 (HAuCl4)
Au+1 (AuxSy−
)
Au0
edge energy, keV
scattering pair
N
R, Å
Δσ2 (× 103)
Eo, eV
HAuCl4
zone 3
1.0
1.0
1.0
0.4
−
−
−
−
0.6
1.0
−
−
−
−
−
−
11.9182
11.9182
11.9185
11.9189
11.9195
11.9191
Au−Cl
Au−Cl
Au−Cl
Au−Cl(S)
Au−S
Au−Au
Au−S
Au−Au
4.0
4.2
4.0
3.0
2.0
8.7
0.4
12.0
2.28
2.27
2.29
2.29
2.30
2.86
2.31
2.88
0.0
0.0
0.0
0.0
0.0
1.0
0.0
0.0
−0.5
−0.2
1.5
zone 4
zone 5
−0.2
−2.9
−0.8
−4.1
0.0
after 12 h
NaBH4
−
1.0
Au foil
−
−
1.0
11.9190
2.57 × 10−4 at 1 mL/h). Under these conditions, the flow is
completely laminar, and the two precursor solutions do not mix
except at the interface. Rather, they remain as separate solution
streams on either side of the rectangular cross-section, and the
reaction between them occurs only at the interface of the two
streams. It is at this interface that the gold nanostructures are
formed, producing first a controlled deposit of solid gold sulfide
at the bottom center of the channel that is later reduced to
metallic gold by flowing NaBH4 as a reducing agent through the
entire channel. To understand the time-resolved chemical
nature of the gold nanostructures formed, in situ XAS was
carried out at different positions on the chip during the
reaction. For this purpose, the chip was mounted onto a metal
stage that could be adjusted in three dimensions. The mount
was then placed in the path of the monochromatized
synchrotron beam and adjusted such that the beam passed
through the desired zone on the chip (Figure S2). Prior to the
start of the reaction, the channel mapping at Au L3-edge was
carried out by dispensing only HAuCl4 solution through the
channels. The channel was marked for probing at five different
zones (Figure 1a) and the XAS spectra were collected at each of
the five zones, probed by an X-ray beam size of 0.05 × 0.05
mm, while flowing the precursor solutions into the channels
(Figure 1b). The spectra from zones 1−4 were found to be
exactly similar to that of HAuCl4 solution. The first changes in
the spectra were noticed at zone 5. After this initial study, zones
3 and 5 were chosen for more detailed analysis. Figure 1d
shows the Au L3-edge XANES spectra collected at zones 3 and
5 on the chip immediately after passing the precursors and at
zone 5 after 12 h. The spectrum obtained at zone 3 matched
well with that of the precursor, HAuCl4, having Au(III)
oxidation state (Table 1). The spectrum obtained at zone 5
showed a mixture of HAuCl4 and also the formation of AuxSy−
nanoclusters26 having an Au/S ratio close to 2 with Au(I)
oxidation state. The transmission electron microscopy (TEM)
image of the sample collected from the same zone (by
dissecting the chip at this location) also confirmed the
formation of AuxSy− nanoclusters of 1−2 nm size (Figure
S3). The EXAFS spectrum of the sample probed at zone 5
showed the presence of 40% of HAuCl4 and 60% of the AuxSy−
nanoclusters (Figure 1e, and Table 1) based on the linear
combination fitting with gold foil and gold sulfide reference
compounds. The formation of AuxSy− nanoclusters was
observed 17 s after injecting the precursor solutions and the
reaction rate (calculated based on the consumption of
precursor) at this point was 0.235 mmol−1. However, complete
formation of the AuxSy− nanoclusters was observed only after
12 h duration (Figures 1d,e, and Table 1). Interestingly, the
AuxSy− nanoclusters were stable even after 24 h, and there was
no evidence of Au−Au bonding until the addition of NaBH4.
Tsukuda and co-workers27 reported the formation of metallic
Au13(DMSA)8 clusters with Au−Au bonding upon mixing of
the same precursors in a traditional flask synthesis. Our in situ
XAS results showed that the Au(III) (from the HAuCl4) was
converted to only Au(I) and not to Au(0) until the addition of
NaBH4. The matrix-assisted laser desorption ionization
(MALDI) mass spectra of the nanoclusters prepared within a
conventional flask under similar conditions showed formation
of a mixture of ultrasmall Aun(SR)m clusters with 10−13 gold
atoms (Figure S4). After passing NaBH4 through the chip, the
bond length of the nanoclusters increased from 2.30 (Au−S) to
2.86 Å (Au−Au) indicating the reduction of Au(I) to Au(0)
(Figure 1f). The generality of the technique, therefore, is that it
provides an opportunity to observe the reaction intermediates
at different zones with a spatial resolution corresponding
directly to time resolution. This approach can be used to probe
reactions in real-time to understand reaction kinetics and
mechanism. For example, the products formed within the
channel at a flow-rate of 10 mL/h can be analyzed with a time
resolution of 5.4 ms (Figure S5). By increasing the flow-rate
further, there is a potential opportunity to increase this time
resolution from milliseconds to nanoseconds.28
The same millifluidics tool was also applied to study the
growth of gold nanostructures, at a larger time interval, as they
further grew into macroscopic structures. This was possible by
monitoring the growth after 1, 5, and 9 h on line while keeping
the flow-rate constant at 1 mL/h. During these three time
periods, the morphology of the gold nanostructures changed
dramatically. The morphology of the resultant gold structures
formed at different flow-times were probed using scanning
electron microscope (SEM). Figure 2 illustrates the SEM
images of the gold structures formed at three different flow-
times (1, 5, and 9 h respectively). Figure 2a shows the low
magnification SEM image of the gold structures formed at 1 h
flow-time. Randomly shaped polygonal nanoparticles of ∼300
nm (inset of Figure 2a) were observed at this stage. There were
no structures observed at flow-times <1 h. When the flow-time
was increased to 5 h, the gold structures formed appeared to
have flower-like morphology as seen in Figure 2b. The
structures were ∼3 μm in diameter and had corrugated petal-
like nanostructured features. The figure inset of Figure 2b
shows a magnified view of the same microflowers where one
could clearly see a porous structure containing multiple grooves
of sizes 100−200 nm. Figure 2c shows a low-magnification
SEM image of the gold coating obtained at 9 h flow-time, where
one could see further transformation of flower-like morphology
into microspherical structures. A closer view of these structures
(Inset of Figure 2c) showed the microstructures to be
hemispherical in shape. The gold hemispheres were ∼3 μm
in diameter and had a rough nanostructured porous surface,
which was evident from the higher magnified images. The
structures seem to be formed from the fusion of numerous
5452
dx.doi.org/10.1021/ja400434c | J. Am. Chem. Soc. 2013, 135, 5450−5456

Journal of the American Chemical Society
Article
increasing flow-rate. When the flow-rate was increased to 12
mL/h, formation of an increased number of even smaller
spherical structures was noticed more quickly; just after 45 min
(Figure S8). It is clear that the gold structures appear to evolve
with time depending on the flow-rate as well as flow-times.
A schematic summarizing the time-resolved mapping of the
growth of gold nanostructures within the millifluidic chip is
given in Figure 3. As seen from the Figure 3a, zone 3 is a
Figure 3. Schematic showing the different stages of spatially (and
time) resolved growth process of gold nanostructures within the
millifluidic chip.
representative of zones 1−4 (in Figure 1a), where the gold salt
solution HAuCl4 remained as it is, as shown by the XAS spectra
taken at these zones (Figure 1d). TEM analysis also did not
show formation of any particles at these zones 1−4. At ∼17 s,
the first changes in the reduction process were noticed at zone
5 through XAS analysis (Figure 1d) as well as TEM imaging
(Figure 3f), indicating the formation of AuxSy− nanoclusters.
Therefore, the time-resolved growth was monitored at this
zone. After an hour, the initial AuxSy− nanoclusters further
increased in size to ∼300 nm (Figures 3e, 2a, and 1f). After a
time interval of 5h, the formation of ∼3 μm in diameter gold
structures that had corrugated petal-like nanostructured
features can be noticed (Figures 3d, 2b). When the time of
flow was increased to 9 h, one could notice that these structures
further transformed into micro-hemispherical structures
(Figures 3c, 2c). The Figure 3b indicated that the structural
features remained same before and after the treatment with
NaBH4, and only the chemical nature changed from
predominantly gold sulfide structures to metallic gold structures
(Figure 1f). It is to be noted that all these different structures
were formed in the center of the channel as the reaction was
taking place at the interface of the flows from two inlets (based
on X-ray tomography experiments as explained below and
shown in Figure 4).
Figure 2. SEM images of the gold structures formed within the
millifluidic channels with flow-times: (a) 1, (b) 5, and (c) 9 h. Insets
show the magnified images of a single structure.
smaller-sized particles generating occasional pores on its curved
surface. Such irregular rough surfaces are usually preferred for
their high surface area and are commonly found in electro-
catalysts.29 Hemispherical shape was probably formed due to
the lower surface area per given volume (in order to reduce
surface energy) attributed to that shape. Formation of such
gold hemispheres is rarely reported in the literature,30 and this
is the first report of their formation within a continuous flow
chip.
The structures formed at intermediate flow-times, such as 3
and 7 h, were also analyzed. However, they showed
incompletely formed flowers and hemispheres (Figure S6).
The SEM image of AuxSy− structures formed at 9 h flow-time
without the NaBH4 reduction also showed formation of similar
hemispherical structures indicating the retention of similar
morphologies before and after reduction (Figure S7). While
these morphology changes with flow-time were observed at
constant flow-rate, we have also investigated the influence of
In order to understand the electronic and geometric
properties of the gold structures formed at 9 h flow-time, the
chip was dissected to obtain samples for XAS analysis. The
XANES spectrum of the dissected samples showed formation
5453
dx.doi.org/10.1021/ja400434c | J. Am. Chem. Soc. 2013, 135, 5450−5456

Journal of the American Chemical Society
Article
thickness of the film with deposited gold structures stopped
at 27.5 μm even though its width gradually increased up to 100
μm with increasing flow-time.
The catalytic activity of nanostructured gold continues to be
explored ever since the first report by Haruta et al.32 While a
majority of the catalysis applications of gold were focused on
isolated gold nanoparticles supported through impregnation
process,33,34 there are very few reports regarding the catalytic
activity of gold nanoparticles embedded within continuous flow
microreactors.15,35−37 We therefore, investigated the catalytic
activity of the gold nanostructures formed within the millifluidic
chip using the reduction of 4-nitrophenol (4-NP) to 4-
aminophenol (4-AP)38 as an example (Scheme 1).
Scheme 1. Reduction Reaction of 4-NP to 4-AP
Figure 4. (a,d,g) 3-D X-ray tomographic images of the gold formed
(marked by arrows) within the millifluidic channels after 1, 5, and 9 h
flow-time; (b,e,h) width profile; and (c,f,i) thickness profile.
of metallic gold with Au−Au bonding without the need for
NaBH4 as a reducing agent. Calculations based on linear
combination fitting using gold foil and gold sulfide reference
compounds revealed 50% Au−S and 50% Au−Au bonding for
this sample. There was a gradual increase in thickness of the
AuxSy− growth observed within the millifluidic channels with
increased flow-time of the precursors. Upon reduction with
NaBH4 for 15 min, the XANES linear combination fitting
showed 17% Au−S and 83% Au−Au bonding, indicating
reduction from Au(I) to Au(0). It is interesting to note that
nearly 100% Au−Au bonding (corresponding to metallic gold)
was observed only after performing catalysis (where the
reducing agent NaBH4 was utilized) over the resultant gold/
gold sulfide nanostructures (Figure S9).
In addition to the morphological changes at different flow-
times, we also observed dimensional changes of the nano-
structured gold coating within the channel. After 9 h of flow-
time, a thin yellow-colored deposit of ∼100 μm in width
appeared at the center of the bottom surface of the millifluidic
channel, i.e., at the bottom of the interface of the two precursor
solutions (Figure 4). This deposition primarily consisted of
AuxSy− nanoclusters formed by the reaction of the two
precursor solutions. After NaBH4 flow was initiated, in order
to reduce these structures to metallic gold, the color changed to
black, corresponding to the reduction of the sulfide to metallic
gold (Figure S10). Such a precise positioning of the gold
coating within the channel was attributed to reaction occurring
at the interface of the solutions due to the laminar flow within
the channel. The black line, formed due to the reduction of Au
(I) to Au (0) (Figure 4), contained gold nanostructures as
described above. The EDAX elemental analysis confirmed that
the black structures contained predominantly gold and traces of
sulfur (Figure S11). Further, the width and thickness of the
gold coating formed varied with the flow-time. In order to
understand the time-resolved dimensional change of the
coating, synchrotron radiation-based X-ray tomography was
utilized.31 Figure 4 shows the 3-D X-ray tomography images of
the central gold coating obtained at 1, 5, and 9 h flow-time,
respectively. The measured width and depth profiles (Figure 4)
increased gradually with the flow-ime. Interestingly, the
Initially, 4-NP was mixed with NaBH4 to form its more
reactive 4-nitrophenolate ion (4-NPI) form, which was later
introduced into the chip at a flow-rate of 5 mL/h. It was
converted into 4-AP, catalyzed by the deposited gold
nanostructures. Figure 5a shows the UV−vis spectra of the
standard solutions of 4-NP, 4-NPI, and 4-AP. The absorption
spectrum of 4-NP with a λmax of 316 nm was shifted to 399 nm
indicating the formation of 4-NPI on mixing with NaBH4. The
4-NPI was further converted to 4-AP (λmax of 301 nm) by
flowing it through the millifluidic channel containing the
nanostructured gold deposited at the center. In the absence of
the gold catalyst, conversion of 4-NPI was 20%; whereas in the
presence of gold catalyst (formed at 9 h deposition time), there
was 90.5% conversion of the 4-NPI (Figure 5b). The gold
catalyst was also found to be catalytically active even after 80 h
of continuous flow reaction. In order to understand the
influence of flow-rate on catalysis, we carried out the catalysis at
four different flow-rates (5, 20, 40, and 60 mL/h) and
monitored the conversion of phenolate ion at these flow-rates.
As shown in Figure S12, the highest conversion was found to
take place at the lowest flow-rate of 5 mL/h. This is not
surprising as one can expect better contact with the catalyst at
lower flow-rate where the residence time is higher. The
deposited nanostructured gold catalysts were also found to be
active for conversion of ferricyanide to ferrocyanide (Figure
S13). Here again, the gold catalyst (formed at 9 h flow-rate)
had shown 85.5% conversion of the ferricyanide compared to
12% with no catalyst (Table 2).
CONCLUSIONS
■
In summary, we have demonstrated the use of a millifluidic chip
for time-resolved mapping of the growth of gold nanostructures
from solution using in situ XAS. The approach is novel and
simple. It can be used to map the growth of a number of
metallic nanostructures. Even though we demonstrated a time
resolution of ∼5 ms, the tool developed has the potential to
5454
dx.doi.org/10.1021/ja400434c | J. Am. Chem. Soc. 2013, 135, 5450−5456

Journal of the American Chemical Society
Article
different time intervals.41 While the traditional field of
microfluidics has not yet become widely used even after two
decades of investigations, simple and easily accessible
millifluidic-based tools as the one described here can
revolutionize the ubiquity of lab-on-a-chip devices for high-
value applications ranging from pharmaceuticals to micro-
analytical tools.
ASSOCIATED CONTENT
* Supporting Information
Experimental details and additional figures. This material is
■
S
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
ckumar1@lsu.edu
■
Notes
The authors declare the following competing financial
interest(s): One of the authors, Challa S. S. R. Kumar, is the
founder of the company, Millifluidica, LLC (www.millifluidica.
com), that is commercializing millifluidics-based technologies.
ACKNOWLEDGMENTS
■
This research work is supported as part of the Center for
Atomic Level Catalyst Design, an Energy Frontier Research
Center funded by the U.S. Department of Energy, Office of
Science, Office of Basic Energy Sciences under award number
DE-SC0001058 and also supported by Board of Regents under
grants award number LEQSF (2009-14)-EFRC-MATCH and
LEDSF-EPS(2012)-OPT-IN-15. MRCAT operations are sup-
ported by the Department of Energy and the MRCAT member
institutions. The use of the Advanced Photon Source at ANL is
supported by the U.S. Department of Energy, Office of Science,
Office of Basic Energy Sciences, under contract no. DE-AC02-
06CH11357. Financial support for JTM was provided as part of
the Institute for Atom-efficient Chemical Transformations
(IACT), an Energy Frontier Research Center funded by the
U.S. Department of Energy, Office of Science, Office of Basic
Energy Sciences.
Figure 5. UV−vis spectra of (a) 4-NP, 4-NPI, and 4-AP. (b)
Conversion of 4-NPI to 4-AP in a millifluidic chip reactor with and
without gold.
provide nanosecond time resolution for in situ analysis of
reactions in general and nanoparticle formation in particular by
manipulating the flow-rates and size of the X-ray beam utilized
for probing. In addition to synchrotron radiation-based in situ
XAS probe, the tool ‘in principle’ can be combined with other
spectroscopy probes such as UV−vis spectroscopy, Raman
spectroscopy, SAXS, and so on.39,40 Finally, this tool can also
be utilized to obtain dimension and morphology controlled
catalytic surfaces within the channels, and therefore it can be an
excellent tool for continuous flow catalysis. In terms of
versatility, it can be used to carryout and investigate the
mechanistic aspects of a vast number of catalytic reactions. We
envisage the utility of the tool not only to investigate the time-
resolved formation of catalytically active structures but also for
probing the catalysis in an operando mode with similar time
resolutions. With the option to dissect the chips, one can
envisage carrying out fundamental, atomically precise electronic
and geometric investigations of metal clusters grown at
REFERENCES
■
(1) Ade, H.; Zhang, X.; Cameron, S.; Costello, C.; Kirz, J.; Williams,
S. Science 1992, 258, 972−975.
(2) Thornton, G. Science 2003, 300, 1378−1379.
(3) Sarma, L. S.; Chou, H. L.; Cheng, M. Y.; Taufany, F.; Lai, F. J.;
Tsai, M. C.; Chang, S. H.; Hwang, B. J. In Nanomaterials for the Life
Sciences, Wiley-VCH: Weinheim, Germany, 2008; Vol. 3, pp 377−409.
(4) Sun, Y. Mater. Today 2012, 15, 140−147.
(5) Whitesides, G. M. Nature 2006, 442, 368−373.
(6) deMello, A. J. Nature 2006, 442, 394−402.
(7) Zinoveva, S.; De Silva, R.; Louis, R. D.; Datta, P.; Kumar, C. S. S.
R.; Goettert, J.; Hormes, J. Nucl. Instrum. Methods Phys. Res., Sect. A
2007, 582, 239−241.
Table 2. Summary of Catalysis Results
conversion percentage of reactant (%)
concentration of
reactant (mol)
concentration of
reductant (mol)
temperature
(K)
flow-rate
(ml/h)
with gold catalyst (9 h
deposited chip)
without gold
catalyst
reaction
reduction of 4-nitrophenol to 4-
aminophenol
9 × 10−5
0.65
298
5
90.5
20
reduction of ferricyanide to
ferrocyanide
8.33 × 10−4
0.01
298
5
85.5
12.2
5455
dx.doi.org/10.1021/ja400434c | J. Am. Chem. Soc. 2013, 135, 5450−5456

Journal of the American Chemical Society
Article
(8) Wang, J. T.; Wang, J.; Han, J. J. Small 2011, 7, 1728−1754.
(9) Song, Y.; Modrow, H.; Henry, L. L.; Saw, C. K.; Doomes, E. E.;
Palshin, V.; Hormes, J.; Kumar, C. S. S. R. Chem. Mater. 2006, 18,
2817−2827.
(10) Valencia, P. M.; Farokhzad, O. C.; Karnik, R.; Langer, R. Nature
Nanotechnol. 2012, 7, 623−629.
(40) Polte, J.; Erler, R.; Thunemann, A. F.; Sokolov, S.; Ahner, T. T.;
Rademann, K.; Emmerling, F.; Kraehnert, R. ACS Nano 2010, 4,
1076−1082.
(41) Losovyj, Y. B.; Li, S. C.; Lozova, N.; Katsiev, K.; Stellwagen, D.;
Diebold, U.; Kong, L.; Kumar, C. S. S. R. J. Phys. Chem. C 2012, 116,
5857−5861.
̈
(11) Oyanagi, H.; Sun, Z. H.; Jiang, Y.; Uehara, M.; Nakamura, H.;
Yamashita, K.; Zhang, L.; Lee, C.; Fukano, A.; Maeda, H. J. Synchrotron
Rad. 2011, 18, 272−279.
(12) Lorber, N.; Sarrazin, F.; Guillot, P.; Panizza, P.; Colin, A.;
Pavageau, B.; Hany, C.; Maestro, P.; Marre, S.; Delclos, T.; Aymonier,
C.; Subra, P.; Prat, L.; Gourdon, C.; Mignard, E. Lab Chip 2011, 11,
779−787.
(13) Biswas, S.; Miller, J. T.; Li, Y.; Nandakumar, K.; Kumar, C. S. S.
R. Small 2012, 8, 688−698.
(14) Li, Y.; Sanampudi, A.; Reddy, V. R.; Biswas, S.; Nandakumar, K.;
Yemane, D.; Kumar, C. S. S. R. ChemPhysChem 2012, 13, 177−182.
(15) Abahmane, L.; Knauer, A.; Kohler, J. M.; Groß, G. A. Chem. Eng.
̈
J. 2011, 167, 519−526.
(16) Gaur, S.; Miller, J. T.; Stellwagen, D.; Sanampudi, A.; Kumar, C.
S. S. R.; Spivey, J. J. Phys. Chem. Chem. Phys. 2012, 14, 1627−1634.
(17) Nie, X.; Qian, H.; Ge, Q.; Xu, H.; Jin, R. ACS Nano 2012, 6,
6014−6022.
(18) Jin, R. Nanotechnol. Rev. 2012, 1, 31−56.
(19) Chen, S. P.; Yu, X. D.; Xu, J. J.; Chen, H. Y. Biosens. Bioelectron.
2011, 26, 4779−4784.
(20) Liang, R. P.; Meng, X. Y.; Liu, C. M.; Qiu, J. D. Electrophoresis
2011, 32, 3331−3340.
(21) Luo, C.; Fu, Q.; Li, H.; Xu, L.; Sun, M.; Ouyang, Q.; Chen, Y.;
Ji, H. Lab Chip. 2005, 5, 726−729.
(22) Duan, G.; Cai, W.; Luo, Y.; Li, Z.; Li, Y. Appl. Phys. Lett. 2006,
89, 211905.
(23) Xia, Y.; Xiao, H. J. Mol. Catal. A: Chem. 2010, 331, 35−39.
(24) Wang, Z.; Zhang, J.; Ekman, J. M.; Kenis, P. J. A.; Lu, Y. Nano
Lett. 2010, 10, 1886−1891.
(25) Wang, L.; Guo, S.; Hu, X.; Dong, S. Electrochem. Commun. 2008,
10, 95−99.
(26) Jiang, D.; Walter, M.; Dai, S. Chem.Eur. J. 2010, 16, 4999−
5003.
(27) Negishi, Y.; Tsukuda, T. J. Am. Chem. Soc. 2003, 125, 4046−
4047.
(28) Chan, E. M.; Marcus, M. A.; Fakra, S.; ElNaggar, M.; Mathies, R.
A.; Alivisatos, A. P. J. Phys. Chem. A 2007, 111, 12210−12215.
(29) Shan, C.; Han, D.; Song, J.; Ivaska, A.; Niu, L. J. Mater. Res.
2010, 25, 1755−1760.
(30) Li, H.; Wu, N. Nanotechnology 2008, 19, 275301.
(31) Prieto, G.; Zecevic, J.; Friedrich, H.; de Jong, K. P.; de Jongh, P.
E. Nat. Mater. 2013, 12, 34−39.
(32) Haruta, M.; Kobayashi, T.; Sano, H.; Yamada, N. Chem. Lett.
1987, 16, 405−408.
(33) Valden, M.; Lai, X.; Goodman, D. W. Science 1998, 281, 1647−
1650.
(34) Hughes, M. D.; Xu, Y. J.; Jenkins, P.; McMorn, P.; Landon, P.;
Enache, D. I.; Carley, A. F.; Attard, G. A.; Hutchings, G. J.; King, F.;
Stitt, E. H.; Johnston, P.; Griffin, K.; Kiely, C. J. Nature 2005, 437,
1132−1135.
(35) Adleman, J. R.; Boyd, D. A.; Goodwin, D. G.; Psaltis, D. Nano
Lett. 2009, 9, 4417−4423.
(36) Wang, N.; Matsumoto, T.; Ueno, M.; Miyamura, H.; Kobayashi,
S. Angew. Chem. 2009, 121, 4838−4840.
(37) Jamal, F.; Sebastien, G. J.; Mael, P.; Edmond, P.; Christian, R.
Microsyst. Technol. 2012, 18, 151−158.
(38) Kuroda, K.; Ishida, T.; Haruta, M. J. Mol. Catal. A: Chem. 2009,
298, 7−11.
(39) Merlin, A.; Angly, J.; Daubersies, L.; Madeira, C.; Schoder, S.;
̈
Leng, J.; Salmon, J. B. Eur. Phys. J. E: Soft Matter Biol. Phys. 2011, 34,
58.
5456
dx.doi.org/10.1021/ja400434c | J. Am. Chem. Soc. 2013, 135, 5450−5456