Journal of

Materials Chemistry C
View Article Online
PAPER
Published on 25 September 2013. Downloaded by Indian Institute of Technology Hyderabad on 8/13/2021 6:53:12 AM.

View Journal | View Issue

Microfluidics-based continuous flow formation of

Cite this: J. Mater. Chem. C, 2013, 1,
7540
triangular silver nanoprisms with tuneable surface
plasmon resonance†
Michele Carboni,ab Lorenzo Capretto,b Dario Carugo,b Eugen Stulz*a
and Xunli Zhang*b

Received 11th July 2013
Accepted 23rd September 2013
DOI: 10.1039/c3tc31335b
www.rsc.org/MaterialsC
Controlled synthesis of nanoparticles with desired geometries can provide a direct approach for the
formation of nanomaterials with tuneable size/shape-dependent characteristics, vastly expanding their
range of applications in catalysis, electronics and optics, biology and medicine. In this report, we
demonstrate the development of a microfluidics-based process for the direct synthesis (without seed-
mediated growth) of silver nanoprisms in a continuous flow format using a flow focusing microreactor,
coupled with UV-Vis-NIR spectroscopy for real time monitoring of optical properties of the produced
nanoprisms. By adjusting microfluidic conditions, the nanomaterial's absorbance spectral position and
peak broadness can be fine-tuned. TEM analysis of the produced nanoparticles quantitatively correlates
the characteristics of absorbance spectra with the shape/geometry of nanoparticles. Furthermore, faster
kinetics for the formation of nanoparticles has been observed in the microreactor compared to those in
traditional batch reactors, and the microreactor's capability of synthesising homogenous nanoparticles
over a prolonged operation time is demonstrated.

1 Introduction should meet the requirements of sustainable processes by

The synthesis and characterisation of nanomaterials has
increasingly attracted attention in a wide range of research
elds such as chemistry,1,2 physics,3,4 biology5 and medicine,6
mainly owing to the unique properties of nanomaterials which
differ substantially from both their atomic (or molecular) and
bulky counterparts.7 For instance, metal nanoparticles (NPs)
including gold and silver NPs exhibit strong absorbance in the
visible spectrum8 due to local surface plasmon resonance
(LSPR) associated with NPs' size and shape. Such size- and
shape-dependence of nanomaterial characteristics makes NPs
very versatile,9–12 opening the possibility for their applications in
different elds such as catalysis,13,14 electronics and optics,15,16
biological imaging and cancer research.17–21 However, there still
remain challenges for the synthesis of NPs with precise control
and tailoring of stable geometries (e.g. prisms,22 cubes23 and
stars24) in order to achieve geometry-dependent characteristics
in a reproducible and predictable manner. This becomes more
difficult for large-scale production where these challenges
a
School of Chemistry, University of Southampton, Higheld, Southampton SO17 1BJ,
UK. E-mail: est@soton.ac.uk; Tel: +44-(0)2380-599369
b
Engineering and the Environment, University of Southampton, Higheld,
Southampton SO17 1BJ, UK. E-mail: XL.Zhang@soton.ac.uk; Tel: +44-(0)2380-595099
† Electronic supplementary information (ESI) available: Detailed information on
the effect of pH on the synthesis of nanoprisms, batch synthesis, fouling of
particles inside the microreactor, and TEM images. See DOI: 10.1039/c3tc31335b
using low cost precursors, simple procedures and reducing the
amount of waste.25 In addition, the most commonly used batch
processes including dry and semidry methods (e.g. plasma,
ame, aerosol, spray and pyrolysis), which are currently
employed in the industry, suffer from the lack of mono-
dispersity and control over the reaction steps26 including
nucleation and/or growth.27,28
Recently, microuidic reactors have been proposed as tools
to potentially overcome the limitations commonly associated
with macroscale production of a range of nanoparticles (e.g.
polymeric,29,30 metal31–33 and oxide34). The interest in translating
conventional synthesis to microreactors stems from the unique
characteristics provided by such micro-scale devices. These
include the quick response of the system to variation in the
experimental boundary conditions,35 improved handling of
smaller volumes, more efficient heat and mass transfer, and
highly controllable and tuneable mixing compared to batch
reactors.36,37 In addition, microuidic reactors can be employed
as investigation tools,38,39 given the possibility of online process
monitoring and integration with sensors and control units.37
Previous studies have demonstrated that by precisely control-
ling the experimental parameters, such as mixing and residence
time, the two steps (i.e. nucleation and growth) that lead to the
formation of NPs can be identied, isolated and manipulated.40
Various types of metal NPs (e.g. platinum, silver and gold) have
been produced successfully in microchannels, leveraging the

7540 | J. Mater. Chem. C, 2013, 1, 7540–7546 This journal is ª The Royal Society of Chemistry 2013
 View Article Online

Paper Journal of Materials Chemistry C

faster mixing31,33 and control over heat exchange32,41,42 which
lead to the production of highly mono-disperse isotropic
particles (i.e. spheres). Particle analysis has shown that the
dimensional standard deviation of metal nanoparticles
produced with microuidics is reduced by approximately 20%
Published on 25 September 2013. Downloaded by Indian Institute of Technology Hyderabad on 8/13/2021 6:53:12 AM.
peruorooctylsilane were supplied by Sigma-Aldrich (UK). Iso-
propanol and hydrogen peroxide were purchased from Fisher
Scientic (UK). Sodium borohydride was obtained from Acros
Organic (Belgium). Polydimethylsiloxane (PDMS) monomer
Sylgar 184 and curing agents were purchased from Dow Corning

compared to batch systems.31,33 It has been suggested that more (USA), and SU-8 photoresist was obtained from Chestech (UK).
homogenous diffusion of the reagents obtained inside the
microreactor grants more uniform and efficient nucleation,
which is crucial for the subsequent growth phase, leading to a 2.2 Microreactor fabrication
narrower distribution in both particle size and shape. A hybrid PDMS-glass microreactor was fabricated for the
However, the production of anisotropic materials represents synthesis of silver nanoparticles. Firstly, an SU-8 mold with the
an additional challenge, which is commonly tackled by the use designed microchannel pattern, consisting of three inlets and
of specic ligands for selectively blocking the growth of some one reaction channel, was prepared following the standard
faces of the growing crystal.42,43 Due to the precise control procedure provided by Microchem.57 Subsequently, the mold
required over nucleation and growth steps, a seed-mediated was covered with a layer of a 10 : 1 (w/w) PDMS precursor and
approach is usually employed in order to bypass the nucleation curing agent mixture and heated for 1 h at 80  C for the polymer
step and to simplify the overall procedure.44,45 Among aniso- to cure. The PDMS sheet with the microchannel architecture
tropic particles studied, triangular prisms46 have recently fabricated on the surface was then removed from the mold and
attracted the interest of researchers in view of their wide range permanently bonded to a glass slide aer oxidizing its surface
of applications (such as surface enhanced Raman scattering or with a plasma treatment. The reaction microchannel had
SERS, and optical lters).47–49 In general, size control in batch dimensions of 150 mm  50 mm  3 cm (width  depth 
synthesis can be achieved by photo-irradiation mediated length). The angle between the lateral feeding channels and the
assembly,50 through changing the concentration of the reducing central channel was 30 , as illustrated in Fig. 1.
agent NaBH4 whilst keeping the concentrations of the other Following the bonding of the PDMS sheet to the glass layer,
reagents constant,51 or by adding halide ions (Cl
and Br
) to the channel wall was functionalized with a solution of tri-
nely tune their size and shape.52 Alternatively, a seed-mediated chloro(1H,1H,2H,2H)-peruorooctylsilane (5% v/v in dime-
approach can be used to exert control over the nal absorption thylformamide) to make the channel surface hydrophobic in
of silver nanoprisms; by changing the concentration of the seed order to reduce the interaction between silver cations and
different sizes of prisms can be achieved.53 This approach has hydroxyl groups on the glass surface. Notably, other protocols
been successfully implemented with ow reactors54,55 where usually employed to reduce fouling,33 such as increasing the
effects similar to those observed in the batch synthesis on the
optical properties of silver nanoprisms can be achieved.
Although the seed mediated approach is the most widely used
to prepare silver nanoprisms in microreactors, the combination
of microstructured reactors and efficient LED illumination has
also been successfully applied for synthesising nanoprisms of
enhanced homogeneity.56
Here we demonstrate the development of a microuidics-
based process for the direct synthesis (i.e. without seed-medi-
ated growth) of silver nanoprisms in a continuous ow format
using a ow focusing microreactor, coupled with UV-Vis-NIR
spectroscopy for real time monitoring of optical properties of
the produced nanoprisms. The spectral position of the
maximum absorbance peak (lmax) and the full width at half
maximum (FWHM) of the in-plane dipole surface plasmon
resonance can be tuned by varying major microuidic param-
eters only, namely the ow rate ratio (R) and total ow rate (Vtot).
The single device setup is greatly simplifying the synthesis of
silver nanoprisms with controlled optical properties without
changing the concentrations of the feed solutions.

2 Experimental section
2.1 Reagents
All chemicals were of analytical grade and were used as received
from their respective suppliers. Trisodium citrate dihydrate,
silver nitrate, dimethylformamide and trichloro(1H,1H,2H,2H)-
Fig. 1 (a) Schematic of the experimental setup showing the microreactor layout
and the process monitoring systems. (b) Microphotograph of the microreactor
flow focusing configuration with three co-flowing streams. Liquid–liquid inter-
faces due to different fluid optical properties can be seen.

This journal is ª The Royal Society of Chemistry 2013 J. Mater. Chem. C, 2013, 1, 7540–7546 | 7541

Journal of Materials Chemistry C
total ow rate (Vtot) or increasing the pH of the solution, proved
to be ineffective for our systems.
2.3 Synthesis of triangular silver nanoprisms
Published on 25 September 2013. Downloaded by Indian Institute of Technology Hyderabad on 8/13/2021 6:53:12 AM.
To synthesise triangular silver nanoprisms, a modied protocol
was employed following that reported by Mirkin et al.51 In a
typical batch experiment, a solution of NaBH4 (reducing agent)
was rapidly injected into a water solution of AgNO3, trisodium
citrate dihydrate (TSCD, capping agent) and H2O2 (oxidizing
agent), resulting in the production of triangular nanoprisms.
For the microuidic system, a basic isopropanolic solution of
NaBH4 was employed as a reducing agent instead of the
conventional aqueous solution in order to reduce H2 develop-
ment in solution due to spontaneous degradation of NaBH4
in H2O.
The protocol for solution preparation followed that
described by Shalom et al.58 Briey, 1 mL of a 0.01 M solution of
sodium borohydride (prepared by dissolving 0.038 g of NaBH4
in 70 mL of MilliQ H2O and 30 mL of 1 M NaOH) was added to
32 mL of a 0.1 M isopropanolic solution of NaBH4. The pH of
the solution was adjusted by varying the amount of NaOH used
for the NaBH4 solution. Notably, a higher pH resulted in the
production of solely nanospheres or a complete impairment of
the nanoprecipitation process (i.e., no particles formed in this
case, see ESI†). Solutions were freshly prepared before each
experiment.
In the microuidic reactor system, the two solutions were
injected through the lateral channels (inlet C1 for AgNO3, TSCD
and H2O2, and inlet C3 for NaBH4, Fig. 1) by means of syringe
pumps. Additionally, MilliQ H2O was injected through inlet C2
in order to (a) dilute the two solutions from C1 and C3 to the
desired concentration level, and (b) act as a spacer between the
two solution streams containing the reactants as they would
otherwise react rapidly at the focused interface resulting in
clogging or fouling at the channel junction. PTFE tubes with an
internal diameter of 750 mm were used for connecting the
feeding syringes to the reactor inlets, and for transferring the
product colloid from the reactor outlet to a ow cell for inline
absorbance measurements.
The microreactor was placed on an inverted microscope
(Olympus IX70) stage with a CCD camera (QICAM Fast, Qimage,
Canada) coupled to the microscope (Fig. 1). This allowed a
direct in situ monitoring of the uidic behavior within micro-
channels, especially if clogging and/or fouling of the channel
occurred. Both microuidic and batch experiments were per-
formed at room temperature and in the dark to avoid sponta-
neous photo reduction of AgNO3.
2.4 Nanoparticle characterisation and TEM image analysis
The nanoparticle colloid samples were characterised with a UV-
Vis-NIR inline spectrophotometer (Cary Varian Bio 300) using a
ow cell (10 mm optical path), and the spectra were recorded in
the wavelength range of 300 nm to 900 nm. Transmission
electron microscopy (TEM) images of nanoparticles were
acquired with an FEI Technai12 TEM by drop casting 10 mL of
colloid solution on sample grids (S160-4 Carbon Films on 3 mm
7542 | J. Mater. Chem. C, 2013, 1, 7540–7546
View Article Online
Paper
400 mesh grids), and imaging at an acceleration potential of
120 kV.
3 Results and discussion
3.1 Microuidic synthesis
The microuidic reactor system for the synthesis of silver
nanoprisms is illustrated in Fig. 1. In a typical experiment, a
solution of AgNO3 (36 mM), TSCD (6.5 mM) and H2O2 (76 mM)
was injected into C1 at a selected volumetric ow rate V1, while
NaBH4 (3.6 mM) was injected through C3 at a ow rate V3. Water
owing in the central channel at a ow rate V2 was anked by
the inorganic precursor (AgNO3) on one side and the reducing
agent (NaBH4) on the other side (as detailed in the Experimental
section). The total inlet ow rate (Vtot ¼ V1 + V2 + V3) was
operated in the range of 2.41–3.41 mL h
1. With this setting, it
was observed that a stable ow focusing was achieved within the
main reaction channel, and the use of water as the central
“spacer” stream effectively avoided the immediate clogging of
the channel (Fig. 1b). By varying the lateral inlet ow, the
chemical composition of the reaction system was adjusted as
desired.
Fig. 2a shows the absorbance spectra of the silver nano-
prisms obtained with the inline UV-Vis-NIR spectrophotometer
under different ow conditions, with R ranging from 0.41 to
0.58; R represents the ow rate ratio of the central ow over the
total ow rate (V2/Vtot), with a constant central ow rate (V2) at
1.41 mL h
1. These ow conditions were selected for the
concentration of reagents inside the microreactor to reect the
chemical conditions employed in the batch experiments by
Mirkin et al.51 As can be seen from Fig. 2a, all six spectra show a
single maximum absorbance peak which shis from 690 nm to
796 nm, with increasing R from 0.41 to 0.54. It should be noted
that a small peak at an identical position of 340 nm appears in
all spectra, which may be attributed to the out-of-plane
quadruple resonance effect which is signicantly weaker than
the main absorbance band due to the in-plane dipole surface
plasmon resonance.59
Previous studies have suggested that the in-plane dipole
resonance is a good indicator of general nanoprism geometries.
Its redshi, particularly, is correlated with increased edge
length.51,60 The results shown in Fig. 2 suggest that, by adjusting
the ow conditions, different sizes of nanoprisms can be
Fig. 2 (a) Normalized UV-Vis-NIR spectra of silver nanoprisms obtained under
different flow conditions where R represents the flow rate ratio of the central flow
over the total flow rate (V2/Vtot), with the central flow rate (V2) kept constant at
1.41 mL h
1. (b) TEM image of silver nanoprisms (R ¼ 0.54).
This journal is ª The Royal Society of Chemistry 2013

Paper
obtained. Fig. 2b shows an example TEM image of nano-
particles obtained through the microreactor synthesis, showing
equilateral triangular platelets with rounded tips.
It is generally understood that the process of silver nano-
prism formation involves two steps including (i) the reduction
Published on 25 September 2013. Downloaded by Indian Institute of Technology Hyderabad on 8/13/2021 6:53:12 AM.
of AgNO3 by NaBH4 to form and grow silver seeds in the pres-
ence of the stabilizing agent TSCD, and (ii) the chemical
oxidation of Ag(0) induced by hydrogen peroxide to etch the
silver nanoparticles formed. With the selective adsorption of
citrate ions on the (111) surface of Ag seeds, the most stable
shape of prisms (or triangular platelets) in solution is ach-
ieved,51,61,62 which is largely attributed to the kinetic factors of
the fast growth along the planar twinning defects.53 Previous
studies with a traditional batch synthesis method showed that
the growth phase took approximately 30 min or 5 min in the
presence or absence of the stabilizer poly(vinylpyrrolidone),
respectively.51 In our microuidic approach, the residence time
of reagents within the reaction channel was in the range of
0.23–0.34 s under the ow conditions described above, indi-
cating a signicant increase in the reaction kinetics due to
faster mixing compared to bulk reactions.29,36
3.2 TEM image analysis
To identify and quantify the type of silver nanoprism produced
under ow conditions described above, analysis of TEM images
was carried out by using an in house developed Matlab program
for quantitative geometry characterisation in order to further
correlate the NPs' geometries with the exhibited optical char-
acteristics. Specically, by discriminating particles by their
shape, the edge length of triangular prisms and their
percentage in the whole particle sample were calculated, as
detailed in the Experimental section. Representative TEM
images are shown in Fig. 2b and S5 (ESI†). Image analysis
results are illustrated in Fig. 3 for nanoprisms' average edge
length (Fig. 3a) and percentage (Fig. 3b) at different ow rate
ratios; the percentage is related to the overall particle compo-
sition including spheres and other geometries. The corre-
sponding absorbance band position and FWHM at different
ow rate ratios are also plotted for correlation.
Fig. 3a shows a growth of the prism edge length from 18 nm
to 46 nm with increasing R from 0.41 to 0.58, following an
approximately linear relationship. Notably, the absorbance
band position shis following a similar linear trend. This
correlation of nanoprisms' geometries with their optical
Fig. 3 (a) Nanoprism edge length and the corresponding absorbance band peak
position at variable R. (b) Percentage of nanoprisms and the corresponding
FWHM at variable R. The data are summarised in Table S1 (see ESI†).
This journal is ª The Royal Society of Chemistry 2013
View Article Online
Journal of Materials Chemistry C
response is in good agreement with that observed by Shuford
et al.,63 who found that the absorbance band position of
nanoprisms is proportional to the prism edge length (d) and
inversely proportional to the prism thickness (T), as represented
by eqn (1).63
 
d
lmax ¼ 33:8 þ 418 (1)
T
Using eqn (1) with the prism edge lengths obtained (Fig. 3a),
the prism thickness was calculated. It was found that the
minimum average thickness of prisms was 2.3 nm when d ¼ 18
nm, and the maximum was 3.6 nm when d ¼ 46 nm. This
suggests that the shi of the spectral position shown in Fig. 2a
is largely due to the increase in prism edge length (156% max
increase) with smaller changes in their thickness (56.5% max
increase).
Fig. 3b shows the percentage of nanoprisms in whole
samples and their absorbance band FWHM at different ow
rate ratios (R). It shows that both the percentage of nano-
prisms and FWHM are signicantly inuenced by the ow
conditions. In the operational R range, the percentage of
nanoprisms initially increases with increasing R, reaching a
maximal percentage of 92% at R ¼ 0.47, and then decreases
when R > 0.47. At the same time, the change in absorbance
band FWHM shows an opposite trend, initially decreasing with
increasing R, reaching a minimum FWHM of 360 nm at R ¼
0.47, and then increasing when R > 0.47. As the absorbance
spectral position and its broadness depend on the size and
shape distribution of the nanoparticles, the results demon-
strate the capability of the microreactor for the controllable
production of nanomaterials by adjusting microuidic opera-
tional parameters.
As described above, the results shown in Fig. 2 were obtained
at variable ow rate ratios. R was varied by changing the ow
rates of two lateral streams while keeping the central ow rate
constant. This resulted in both the ow rate ratio R and total
ow rate Vtot change simultaneously; the higher R, the lower
Vtot. Within the continuous ow reaction channel, the change of
R altered the width of ow streams and consequently the
chemical composition in the microenvironment where nano-
particles formed and grew. At the same time, Vtot determined
the overall residence time (RT) of reactants passing through the
whole reaction channel volume (V), RT ¼ V/Vtot. In the micro-
uidic reactor system, both mixing and reagents' concentration
proles are dynamic, and the synergistic effect of R and Vtot add
extra dimensions to the complexity of the system. Therefore, it
was necessary to further examine the effect of individual uidic
parameters.
3.3 Effects of varying microuidic parameters
The effect of two key parameters (i.e. R and Vtot) was examined
by keeping one of the two variables constant whilst varying the
other parameter. Based on the experiments described above,
three values of R (i.e. 0.41, 0.50 and 0.58) were selected and at
each value the total ow rate Vtot was varied in the range of
2.41–3.41 mL h
1. The results are depicted in Fig. 4.
J. Mater. Chem. C, 2013, 1, 7540–7546 | 7543

Journal of Materials Chemistry C
Published on 25 September 2013. Downloaded by Indian Institute of Technology Hyderabad on 8/13/2021 6:53:12 AM.
Fig. 4 Changes in (a) absorbance spectral position and (b) spectral band
broadness by changing R and Vtot, respectively. The data are summarised in
Tables S2 and S3 respectively (see ESI†).
As shown in Fig. 4a, for a given Vtot the spectral position of
the maximum absorbance shis to higher wavelength as R
increases, indicating an increase in particle size according to
eqn (1).63 As discussed above, within the microscale reaction
channel changes in the chemical composition of the microen-
vironment were facilitated through the diffusive mixing process
across the three laminar ow streams. At lower R levels, the
central spacer ow stream becomes relatively narrow thus the
system requires shorter time for complete mixing (i.e. faster
mixing).29,30 Subsequently, the resulting chemical composition
of the microenvironment affects the kinetics of nanoprism
formation steps, including super-saturation, nucleation and
growth. In general, faster mixing increases the super-saturation
rate and thus the kinetics of both nucleation and growth.64 It
was found in previous studies that the nucleation step was
inuenced more signicantly by an increased super-saturation
rate leading to the formation of a higher number of small
nuclei, and vice versa.64 Therefore, at higher R levels, the slower
mixing process and the associated chemical composition within
the microenvironment are favorable for the formation of bigger
size and triangular shaped nanoparticles, resulting in the
spectral redshi, as observed in the experiments and coherently
with eqn (1).
The results depicted in Fig. 4b show that at the different R
values investigated the broadness of the spectral band in terms
of FWHM decreases with decreasing Vtot in the higher Vtot range
(3.41 / 2.81 mL h
1), and reaches a minimum at Vtot ¼ 2.81 mL
h
1, which is followed by an increase with decreasing Vtot. As
Vtot determines the reagents' residence time within the reaction
channel, the lower Vtot generally provides a longer residence
and reaction time creating a microenvironment for the forma-
tion of triangular NPs with smaller FWHM, as shown in Fig. 3b.
On the other hand, Munechika et al.65 suggested that a rising
percentage of triangular nanoprisms could result in an increase
in radiation damping. Thus, an increase in FWHM is observed
in the lower Vtot range (2.81 / 2.41 mL h
1) with more trian-
gular nanoprisms produced, as conrmed by TEM analysis
(Fig. S4†). Overall, this demonstrates that the size and shape
distribution can easily be tuned to the desired specications,
simply by adjusting the operational parameters.
For comparison, experiments using a conventional batch
reactor were also performed at the same reagent concentrations
employed in the microuidic reactor, as detailed in the ESI.†
Briey, the results showed that the reagent concentration
7544 | J. Mater. Chem. C, 2013, 1, 7540–7546
View Article Online
Paper
affected the spectral band position (lmax) and broadness
(FWHM) in a non-trivial way, where no clear trend of effect was
observed. Clearly the dynamic conditions are crucial for the
reproducibility of the NPs' characteristics and cannot be ach-
ieved in the same manner under bulk conditions.
3.4 Microreactor robustness test
In order to test the robustness of the microreactor system for
long time performance, experiments were carried out by
monitoring the reaction channel optically, and the reaction
product spectroscopically. For monitoring microchannel
fouling, microscopy images were acquired every 15 min by
focusing at the bottom wall of the channel. The monitoring of
the produced nanoparticle colloid system was performed by
collecting UV-Vis-NIR spectra at intervals of 1 min. The time
evolution of fouling, and related changes in spectral position
and peak broadness are shown in Fig. 5.
Fig. 5a and b show the microscopic images of the micro-
channel at the beginning of the experiment (t ¼ 0 min) and aer
225 min run, respectively. As can be seen, no fouling was
present at t ¼ 0 min. With the experiment running, a thin grey
line appeared in the center of the channel suggesting some
minimal precipitation and aggregation at the channel surface
(Fig. 5b). There was no uidic disturbance caused by the
attached layer which is in the micron range thickness and is
conned to a very limited area. The time evolution of the fouling
process in the region of interest was quantied by measuring
the grey level intensity of each pixel in the images. The results
are illustrated in Fig. 5c as a surface plot which shows the
change in channel surfaces of both side and bottom walls over
the experimental period. The results conrm that no signicant
fouling took place within the microreactor channel during the
long-time run (up to 225 min).
In Fig. 5d the evolution of the position of the absorbance
peak maximum, lmax, and FWHM over time is shown. The trend
Fig. 5 (a) Microscopy images of the microchannels in the region of interest at (a)
t ¼ 0 min, and (b) t ¼ 225 min. (c) Surface plot showing the temporal evolution of
the grey level intensity recorded at the channel junction. (d) Profiles of spectral
peak maxima (solid line) and FWHM (dashed line) over the experimental period.
This journal is ª The Royal Society of Chemistry 2013

Paper
of the spectral maximum reached a stable stage aer an
induction period of 25 min, which was mainly due to the time
needed for ushing the spectrometer ow cell with a much
bigger volume than the microreactor priming volume. The
FWHM prole instead displays a slightly decreasing trend over
Published on 25 September 2013. Downloaded by Indian Institute of Technology Hyderabad on 8/13/2021 6:53:12 AM.
time. The standard deviation calculated for the maximum
wavelength is 40 nm over the whole period of the experiment,
which is 9.2 nm if the induction stage is excluded. The same
calculation performed on FWHM data revealed a standard
deviation of 15 nm aer the induction stage. Thus, the spec-
troscopic monitoring results together with microscopic obser-
vation demonstrate that the microreactor developed is robust,
enabling a long-term stable process for nanoparticle synthesis
in a continuous ow format.
4 Conclusions
A novel microuidic reactor design for the synthesis of trian-
gular silver nanoprisms is presented, and the effect of different
uidic conditions on the characteristics of the produced
nanoparticles has been studied. We have demonstrated that the
ow rate ratio and the total ow rate can be adjusted inde-
pendently to control the spectral position and the peak broad-
ness of the absorbance spectra of the nanoparticles. By varying
both parameters simultaneously, ne tuning of the optical
properties can be achieved. In order to demonstrate that this
effect is not due to the effect of different concentrations of
reagents inside the channel, batch experiments were carried
out, showing that controllable changes in optical properties are
due to the microuidic conditions and cannot be replicated in a
batch reactor. To the best of our knowledge, this is the rst time
that the optical properties of triangular shaped silver nano-
prisms can be tuned in a microreactor without a seed mediated
approach. The reactions carried out in the microreactor show a
faster kinetics than the traditional batch reactor, as nano-
particles are formed in seconds rather than in minutes. TEM
analyses have been carried out to compare the spectrophoto-
metric characteristics with the particle shape and size distri-
bution. Image analysis of the TEM pictures shows that the
redshi in the spectral position is due to an increase in the edge
length, while the narrowing of the peak is related to the increase
in the percentage of triangular shaped particles. Though parti-
cles obtained from seed-mediated approaches show a better
homogeneity, our setup allows for a good degree of controlla-
bility, reproducibility and shape selectivity by using a simple
microreactor design. Furthermore, the device was tested over a
long period of time to prove its robustness and its potential
industrial applications; the data collected demonstrate that the
device is capable of synthesising homogenous nanoparticles
over a long period of time with reduced clogging owing to
hydrophobic functionalization of the internal walls of the
channels.
Acknowledgements
The authors thank ERDF (IS:CE-Chem & InterReg IVa program
4061) for nancial support.
This journal is ª The Royal Society of Chemistry 2013
View Article Online
Journal of Materials Chemistry C
References
1 A. Y. Utekhina and G. B. Sergeev, Russ. Chem. Rev., 2011, 80,
219–233.
2 F. P. Zamborini, L. L. Bao and R. Dasari, Anal. Chem., 2012,
84, 541–576.
3 K. Ostrikov, U. Cvelbar and A. B. Murphy, J. Phys. D: Appl.
Phys., 2011, 44, 174001.
4 A. Patra, C. G. Chandaluri and T. P. Radhakrishnan,
Nanoscale, 2012, 4, 343–359.
5 W. J. Stark, Angew. Chem., Int. Ed., 2011, 50, 1242–1258.
6 L. M. Ensign, R. Cone and J. Hanes, Adv. Drug Delivery Rev.,
2012, 64, 557–570.
7 I. G. Gogotsi, Nanomaterials Handbook, CRC/Taylor &
Francis, Boca Raton, 2006.
8 P. Mulvaney, Langmuir, 1996, 12, 788–800.
9 W. S. Seo, H. H. Jo, K. Lee, B. Kim, S. J. Oh and J. T. Park,
Angew. Chem., Int. Ed., 2004, 43, 1115–1117.
10 Y. Pan, S. Neuss, A. Leifert, M. Fischler, F. Wen, U. Simon,
G. Schmid, W. Brandau and W. Jahnen-Dechent, Small,
2007, 3, 1941–1949.
11 C. L. Nehl and J. H. Hafner, J. Mater. Chem., 2008, 18, 2415–
2419.
12 H. Chen, X. Kou, Z. Yang, W. Ni and J. Wang, Langmuir, 2008,
24, 5233–5237.
13 D. Astruc, F. Lu and J. R. Aranzaes, Angew. Chem., Int. Ed.,
2005, 44, 7852–7872.
14 T. J. Yoon, W. Lee, Y. S. Oh and J. K. Lee, New J. Chem., 2003,
27, 227–229.
15 W. Chen, J. Z. Zhang and A. G. Joly, J. Nanosci. Nanotechnol.,
2004, 4, 919–947.
16 E. Katz, I. Willner and J. Wang, Electroanalysis, 2004, 16, 19–44.
17 N. O'Farrell, A. Houlton and B. R. Horrocks, Int. J. Nanomed.,
2006, 1, 451–472.
18 C. Gruttner and J. Teller, J. Magn. Magn. Mater., 1999, 194,
8–15.
19 M. D. Chavanpatil, A. Khdair, Y. Patil, H. Handa, G. Mao and
J. Panyam, J. Pharm. Sci., 2007, 96, 3379–3389.
20 S. Cammas, K. Suzuki, C. Sone, Y. Sakurai, K. Kataoka and
T. Okano, J. Controlled Release, 1997, 48, 157–164.
21 G. R. Reddy, M. S. Bhojani, P. McConville, J. Moody,
B. A. Moffat, D. E. Hall, G. Kim, Y. E. L. Koo,
M. J. Woolliscro, J. V. Sugai, T. D. Johnson,
M. A. Philbert, R. Kopelman, A. Rehemtulla and
B. D. Ross, Clin. Cancer Res., 2006, 12, 6677–6686.
22 D. F. Liu, Y. J. Xiang, Z. X. Zhang, J. X. Wang, Y. Gao, L. Song,
L. F. Liu, X. Y. Dou, X. W. Zhao, S. D. Luo, C. Y. Wang,
W. Y. Zhou, G. Wang and S. S. Xie, Nanotechnology, 2005,
16, 2665.
23 Z. Peng, C. Kisielowski and A. T. Bell, Chem. Commun., 2012,
48, 1854–1856.
24 C. L. Nehl, H. Liao and J. H. Hafner, Nano Lett., 2006, 6, 683–
688.
25 H. Cui, Y. Feng, W. Ren, T. Zeng, H. Lv and Y. Pan, Recent
Pat. Nanotechnol., 2009, 3, 32–41.
26 T. Ould-Ely, M. Luger, L. Kaplan-Reinig, K. Niesz, M. Doherty
and D. E. Morse, Nat. Protoc., 2011, 6, 97–104.
J. Mater. Chem. C, 2013, 1, 7540–7546 | 7545

Journal of Materials Chemistry C
27 J. Turkevich, P. C. Stevenson and J. Hillier, Discuss. Faraday
Soc., 1951, 11, 55–75.
28 V. K. LaMer and R. H. Dinegar, J. Am. Chem. Soc., 1950, 72,
4847–4854.
29 L. Capretto, D. Carugo, W. Cheng, M. Hill and X. Zhang,
Published on 25 September 2013. Downloaded by Indian Institute of Technology Hyderabad on 8/13/2021 6:53:12 AM.
J. Colloid Interface Sci., 2011, 357, 243–251.
30 R. Karnik, F. Gu, P. Basto, C. Cannizzaro, L. Dean, W. Kyei-
Manu, R. Langer and O. C. Farokhzad, Nano Lett., 2008, 8,
2906–2912.
31 S. Yujun, S. S. R. K. Challa and H. Josef, J. Nanosci.
Nanotechnol., 2004, 4, 788–793.
32 X. Z. Lin, A. D. Terepka and H. Yang, Nano Lett., 2004, 4,
2227–2232.
33 J. Wagner and J. M. Köhler, Nano Lett., 2005, 5, 685–691.
34 L. Frenz, A. El Harrak, M. Pauly, S. Bégin-Colin,
A. D. Griffiths and J.-C. Baret, Angew. Chem., Int. Ed., 2008,
47, 6817–6820.
35 H. A. Stone and S. Kim, AIChE J., 2001, 47, 1250–1254.
36 L. Capretto, W. Cheng, M. Hill and X. Zhang, in Microuidics:
Technologies and Applications, ed. B. C. Lin, 2011, pp. 27–68.
37 K. F. Jensen, Chem. Eng. Sci., 2001, 56, 293–303.
38 X. Yan, Q. Li and L.-S. Li, J. Am. Chem. Soc., 2012, 134, 16095–
16098.
39 S. Park and J. W. Hong, Thin Solid Films, 2009, 517, 3183–
3186.
40 T. L. Sounart, P. A. Saer, J. A. Voigt, J. Hoyt, D. R. Tallant,
C. M. Matzke and T. A. Michalske, Lab Chip, 2007, 7, 908–
915.
41 E. M. Chan, R. A. Mathies and A. P. Alivisatos, Nano Lett.,
2003, 3, 199–201.
42 W. Chen-Hsun, H. Chih-Chia, Y. Chen-Sheng, L. Huan-Yao
and L. Gwo-Bin, J. Micromech. Microeng., 2008, 18, 035019.
43 W. Wang, S. Banerjee, S. Jia, M. L. Steigerwald and
I. P. Herman, Chem. Mater., 2007, 19, 2573–2580.
44 N. R. Jana, L. Gearheart and C. J. Murphy, Adv. Mater., 2001,
13, 1389–1393.
45 N. R. Jana, Small, 2005, 1, 875–882.
46 J. E. Millstone, S. J. Hurst, G. S. Metraux, J. I. Cutler and
C. A. Mirkin, Small, 2009, 5, 646–664.
7546 | J. Mater. Chem. C, 2013, 1, 7540–7546
View Article Online
Paper
47 C. Ying, W. Chungang, M. Zhanfang and S. Zhongmin,
Nanotechnology, 2007, 18, 325602.
48 V. Torres, M. Popa, D. Crespo and J. M. Calderón Moreno,
Microelectron. Eng., 2007, 84, 1665–1668.
49 V. S. Tiwari, T. Oleg, G. K. Darbha, W. Hardy, J. P. Singh and
P. C. Ray, Chem. Phys. Lett., 2007, 446, 77–82.
50 R. Jin, Y. Cao, C. A. Mirkin, K. L. Kelly, G. C. Schatz and
J. G. Zheng, Science, 2001, 294, 1901–1903.
51 G. S. Métraux and C. A. Mirkin, Adv. Mater., 2005, 17, 412–
415.
52 N. Cathcart, A. J. Frank and V. Kitaev, Chem. Commun., 2009,
7170–7172.
53 D. Aherne, D. M. Ledwith, M. Gara and J. M. Kelly, Adv. Funct.
Mater., 2008, 18, 2005–2016.
54 A. Knauer and J. M. Kohler, Chem. Ing. Tech., 2013, 85, 467–
475.
55 A. Knauer, A. Csáki, F. Möller, C. Hühn, W. Fritzsche and
J. M. Köhler, J. Phys. Chem. C, 2012, 116, 9251–9258.
56 S. Silvestrini, T. Caroglio and M. Maggini, Chem. Commun.,
2013, 49, 84–86.
57 The detailed procedure (SU-8 2025-2075 datasheet) can be
downloaded from http://www.microchem.com/Prod-
SU82000.
58 D. Shalom, R. C. R. Wootton, R. F. Winkle, B. F. Cottam,
R. Vilar, A. J. deMello and C. P. Wilde, Mater. Lett., 2007,
61, 1146–1150.
59 K. L. Kelly, E. Coronado, L. L. Zhao and G. C. Schatz, J. Phys.
Chem. B, 2002, 107, 668–677.
60 R. C. Jin, Y. C. Cao, E. C. Hao, G. S. Metraux, G. C. Schatz and
C. A. Mirkin, Nature, 2003, 425, 487–490.
61 Y. Sun, B. Mayers and Y. Xia, Nano Lett., 2003, 3, 675–679.
62 A. Callegari, D. Tonti and M. Chergui, Nano Lett., 2003, 3,
1565–1568.
63 K. L. Shuford, M. A. Ratner and G. C. Schatz, J. Chem. Phys.,
2005, 123, 114713–114719.
64 H. C. Schwarzer and W. Peukert, Chem. Eng. Commun., 2004,
191, 580–606.
65 K. Munechika, J. M. Smith, Y. Chen and D. S. Ginger, J. Phys.
Chem. C, 2007, 111, 18906–18911.
This journal is ª The Royal Society of Chemistry 2013