Article
pubs.acs.org/Langmuir
Controlling the Pulsed-Laser-Induced Size Reduction of Au and Ag
Nanoparticles via Changes in the External Pressure, Laser Intensity,
and Excitation Wavelength
Daniel Werner and Shuichi Hashimoto*
Department of Optical Science and Technology, The University of Tokushima, Tokushima 770-8506, Japan
S Supporting Information
*
ABSTRACT: The laser-induced size reduction of aqueous
noble metal nanoparticles has been the subject of intensive
research, because of the mechanistic interest in the light−
nanoparticle interactions and its potential application to size
control. The photothermal evaporation hypothesis has gained
solid support. However, the polydispersity of the final products
is considered as an inherent drawback of the method. It is
likely that the polydispersity arises from the uncontrolled heat
dissipation caused by vapor bubble formation in the ambient
atmosphere. To overcome this problem, we applied high
pressures of 30−100 MPa. The particle size was regulated by
adjusting three parameters: the pressure, laser intensity, and
excitation wavelength. For example, starting from a colloidal solution of 100 nm diameter gold nanoparticles, highly
monodisperse (±3−5%) spheres with various diameters ranging from 90 to 30 nm were fabricated by tuning the laser intensity at
100 MPa, using an excitation wavelength of 532 nm. Further size reduction of the diameter to 20 nm was achieved by reducing
the pressure and switching the excitation wavelength to 355 nm. It was found that the application of high pressures led to the
heat loss-controlled size-reduction of the gold nanoparticles. More complicated results were obtained for 100 nm silver
nanoparticles, possibly because of the different size-dependent light-absorbing nature of these particles. Based on our extensive
experimental studies, a detailed picture was developed for the nanosecond laser-induced fabrication of gold and silver
nanoparticles, leading to unprecedented size control.
■
INTRODUCTION
The interactions of noble metal nanoparticles (NPs) with
pulsed lasers have been the subject of intensive research for
more than a decade, with the aim of improving the basic
understanding of light−nanomaterial interactions. Ultrafast
excitation with a femtosecond laser revealed the generation
and instantaneous relaxation of hot electrons (electron−
electron relaxation), lattice heating by thermalized electrons
(electron−lattice relaxation), and heat transfer to the
surrounding medium (lattice−lattice relaxation).1−4 Picosecond
and nanosecond laser excitation of the localized surface
plasmon resonance (LSPR) band resulted in the melting and
size-reduction of NPs in solution. This laser-induced size
reduction of noble metal NPs has attracted a great deal of
attention from the mechanistic point of view.5 A model was
proposed in which the reshaping from a faceted to a spherical
geometry takes place because of the melting of the particle
(mp: 1337 K); the size-reduction was assumed to take place via
evaporation from particles heated above their boiling point (bp:
≈3100 K) when the particles are subjected to high-intensity
nanosecond pulses.6 This photothermal model for the
morphological changes in NPs was supported by the results
of rigorous calculations for the time profile of a particle
temperature as a function of the laser intensity.7 Despite these
© 2012 American Chemical Society
promising theoretical results, however, typical photothermal
size reduction experiments show the formation of polydisperse
particles and interconnected structures.6,8−11 The polydispersity may arise as a consequence of vapor bubble formation
around the particles as they reduce in size. It has been shown
that photothermal vapor bubble formation takes place during
the pulsed-laser excitation of Au NPs dispersed in water, when
laser fluences that are likely to cause size reduction are
used.12−15 It has also been demonstrated that NPs are formed
inside the cavitation bubbles that form during the laser ablation
of Ti, Au, and Ag targets.16,17
The size reduction depends on both the exposure period and
the laser fluence, as shown in Scheme 1. It has been shown that
the average particle size reduces with increasing irradiation time
when growth processes such as fusion are absent (Scheme 1a).6
In the photothermal process, insufficient cooling of the laserheated NPs inside the bubbles (which act as a thermal shield)
could be one reason for the efficient vaporization from the
particles. This would suggest that poor management of the
laser-generated local heating could cause uncontrollable size
Received: November 19, 2012
Revised: December 19, 2012
Published: December 21, 2012
1295
dx.doi.org/10.1021/la3046143 | Langmuir 2013, 29, 1295−1302
Langmuir
Article
Information, Figure S1; 5.6 × 109 mL−1), and Ag NPs were purchased
from Sigma-Aldrich (#730777, nominal diameter of 100 ± 7 nm, 3.6 ×
109 mL−1). The laser irradiation of 2 mL aliquots of aqueous colloidal
solutions contained in a quartz cuvette (optical path length 1.0 cm,
width 1.0 cm, height 2.0 cm) was conducted using the second and
third harmonics of an Nd:YAG laser (Continuum, Surelite I-10). The
cuvette was placed in a hydrostatically pressurized container (≤100
MPa) similar to the one described previously, but this container was
modified to incorporate cuvettes with >2 mL volumes instead of 0.4
mL, and to allow the agitation of the solution.25 This change in the
design permitted a forced-convective heat transfer. The NP solutions
were agitated using a magnetic stirrer during the irradiation. The laser
parameters were as follows: wavelength, 532 and 355 nm; repetition
rate, 10 Hz; pulse width, 5−6 ns full width at half-maximum; beam
area, (0.12 ± 0.01) cm2. TEM photographs of the particles were
recorded on a JEM-2100F microscope (JEOL) operated at 200 kV.
The specimens for the TEM measurements were prepared by placing
one drop of a sample solution on a support consisting of Cu mesh
coated with an evaporated 13 nm thick carbon film. The support was
dried under vacuum. The preparation of the TEM specimens was
carried out on the same day as the laser irradiation. The software
package Image J (http://rsbweb.nih.gov/ij/) was used to process the
particle images.
Scheme 1. Graphical Representation of the Size Reduction
Behavior in an Aqueous Colloidal Solution, Represented by
the Average Particle Diameter versus the Irradiation Perioda
a
Observations under ambient atmospheric pressure (a) and the new
strategy for discriminating particle sizes using the laser intensity with
the application of high pressures (b). Abbreviation for laser intensities:
high (H), medium (M), and low (L).
■
reduction, resulting in polydispersity. However, the size
reduction should stop when the particle size at which the
particle cooling dominates the laser heating is reached. The
minimum particle size should depend on the laser fluence
regardless of the number of pulsesif the particle cooling is
size-dependent.18 If this is the case, it will be possible to control
the particle size by changing the laser fluence, as shown in
Scheme 1b.
It is possible to achieve effective cooling by applying high
pressures, which suppresses the formation of bubbles. High
pressures (below 30 MPa) have been applied by the Sasaki
group to suppress the formation of laser-induced plasma during
the nanosecond laser ablation of a Ti target in water.19,20 The
Saitow group implemented a pressure chamber (≤30 MPa) for
the laser-ablation generation of NPs in supercritical media.21,22
Their method is different from ours in that they preheated the
chamber to generate supercritical states in media such as CO2
prior to the laser ablation. Interestingly, particle aggregation
and fusion were observed in these laser ablation studies in
supercritical media.22 To achieve control over the sizereduction rate of NPs without creating bubbles, it is necessary
to apply pressures that exceed the critical pressure of water,
22.1 MPa.23,24 Our preliminary examination showed that the
size reduction of 58 nm diameter Au NPs was remarkably
retardedand even stopped at approximately 30 nmat water
pressures of 100−200 MPa.25 We ascribed this effect to the
increased boiling points at high pressures, and the suppression
of bubble formation. However, a quantitative description of the
laser fluence-dependent size-reduction under high pressures is
still lacking.
In this paper, starting with relatively large Au and Ag NPs of
≈100 nm in diameter, we demonstrate laser fluence-dependent,
controlled size-reduction that is critically dependent on both
pressure and excitation wavelength. Specifically, Au sphere
diameters ranging from 90 to 20 nm and narrow size
distributions were achieved, by adjusting the laser fluence and
pressure, and by choosing the excitation wavelength. Moreover,
by incorporating a heat loss-controlled reaction protocol, we
made progress in the interpretation of the data.
■
RESULTS AND DISCUSSION
Size Reduction of 100 nm Au NPs. Aqueous colloidal Au
NPs with an average diameter of 100 nm (100 ± 8 nm) were
subjected to excitation with a 532 nm nanosecond pulsed laser,
under a hydrostatic pressure of 100 MPa. Irradiation was
carried out with different laser energy densities (fluences: mJ
cm−2) for various numbers of laser pulses, and TEM images
were acquired after irradiation. Figure 1 summarizes the results
for 72 000−108 000 laser pulses, showing TEM images and the
corresponding particle size distribution histograms (see
Supporting Information, Figure S2 for extinction spectra).
The inspection of Figure 1 reveals that remarkably spherical Au
NPs were formed. The spheres were highly monodisperse
(±3−5%) after irradiation with a sufficient number of laser
pulses (as described above). Further irradiation did not reduce
the particle size or improve the size distributions. We found
that the final size distribution was determined by the pulse-topulse laser fluctuations. Most importantly, the diameters of the
Au NPs thus prepared were laser fluence-dependent: the
diameter progressively decreased with increases in the laser
fluence. It should be noted that besides these very spherical
particles, small Au particles with diameters below 10 nm were
formed. These small particles can be seen in the background of
the TEM images shown in Figure 1.
In a previous study, we investigated the size reduction of 58
nm Au NPs by applying high pressures ranging from 30 to 200
MPa.25 We employed a small cuvette volume of 0.4 mL (4.0
mm × 4.0 mm × 2.5 mm), and the solutions were not agitated.
The small volume was used to maximize the volume of the
solution that could be irradiated with a laser beam 4.2 mm in
diameter; however, effective heat convection, the importance of
which will be mentioned below, was not considered. Because of
these technical deficiencies, monodisperse size distributions
were not achieved in these experiments.
In the nanosecond laser excitation regime, size reduction is
considered to occur via photothermal evaporation from Au NPs
heated above their boiling temperature by laser heating.6−8 This
model predicts the evaporation of atoms and clusters from the
particle surface, leaving core particles.8 The model might
explain the observations of Figure 1, but it is also possible that
small evaporated clusters could coalesce to form particles of
EXPERIMENTAL METHODS
Au NPs were obtained from British Biocell International (EMGC100,
diameter 100 ± 8 nm determined using TEM, see Supporting
1296
dx.doi.org/10.1021/la3046143 | Langmuir 2013, 29, 1295−1302
Langmuir
Article
Figure 2. Mean diameter of Au NPs as a function of laser fluence at 30
MPa (red curve with closed green diamonds), 60 MPa (green curve
with closed blue triangles), and 100 MPa (black curve with open
circles), based on the measurement of the size distributions. The
vertical error bars show the standard deviations of Gaussian
distributions. The horizontal error bars represent the uncertainties in
the laser beam diameter and the pulse energy fluctuations. The particle
images and size distributions obtained at 60 MPa are given in
Supporting Information Figure S4.
Figure 1. Particle images (TEM photographs) and corresponding size
distributions at 100 MPa after irradiation with 532 nm nanosecond
laser pulses (10 Hz) at various laser fluences, for the original 100 nm
Au NPs. Laser fluences and pulse numbers: (a) 43.1 mJ cm−2, 72 000
pulses, (b) 60.2 mJ cm−2, 72 000 pulses, (c) 78.0 mJ cm−2, 90 000
pulses, (d) 90.2 mJ cm−2, 108 000 pulses, (e) 106.5 mJ cm−2, 108 000
pulses, (e) 136.6 mJ cm−2, 108 000 pulses. The number of particles
measured ranged from 700 to 1300. NPs with diameters bigger than
10 nm were counted. The 100 nm scale bar is applicable to all of the
TEM images. Corresponding extinction spectra are given in
Supporting Information Figure S2.
have been partly due to the differences in the boiling points
(bp) of the particles, because the boiling point of bulk gold is
pressure-dependent. By applying the Clausius−Clapeyron
equation, the bp was calculated as 5500 K at 30 MPa, 6200
K at 60 MPa, and 6800 K at 100 MPa.26 Additionally, the
convective/conductive heat transfer in water is pressuredependent, as will be described below. These results suggested
that the pressure values, as well as the laser fluences, were
decisive in controlling the final core particle diameter. At 100
MPa, we found it difficult to prepare monodisperse core
particles with diameters ≤30 nm, because at laser fluences of
≥140 mJ cm−2 the size distribution suddenly became wide. In
contrast, at 30 MPa, particles with a minimum core diameter of
20 nm were prepared with a size distribution of ±15%. These
results suggested that the core particle sizes could be fine-tuned
by adjusting either the laser fluence or the external pressure.
Although the size distribution was very narrow at 100 MPa, a
much wider distribution was observed at 30 MPa.
To investigate the dependence of the size reduction on the
excitation wavelength, laser irradiation was performed at 355
and 532 nm, at a pressure of 30 MPa. As shown in Figure 3,
smaller core particles were formed at 355 nm than at 532 nm,
with similar laser fluences. These results showed that the size
reduction efficiency was higher under excitation at 355 nm. The
effects of the excitation wavelength will be detailed and
explained below.
Size Reduction of 100 nm Ag NPs. The generation of
size-controlled, monodisperse core particles from initial 100 nm
Au NPs via the adjustment of the laser fluence at high pressures
prompted us to investigate the applicability of the method to
other materials. We measured the laser fluence-dependent size
reduction of ≈100 nm Ag NPs at 60 MPa, as shown in Figure 4
(for the changes in the extinction spectra, see Supporting
Information Figure S5). The laser fluence-dependent core
diameter curves for the Ag NPs were also excitation
wavelength-dependent, but were noticeably different from
those observed for the Au NPs. Under excitation at 355 nm,
a sudden decrease in the core diameter was observed when the
laser fluence was increased from 35 to 45 mJ cm−2; the average
diameter dropped from 90 to 20 nm. In contrast, at an
<10 nm; particles such as these were observed in the
background of our TEM images. The laser fluence-dependent,
controlled size reduction of core particles was only observed in
water that was pressurized above the critical pressure (22.1
MPa). At an ambient pressure of 0.1 MPa, the laser-induced
size reduction was uncontrollably rapid, and no mechanisms (as
shown in Scheme 1a) operated to stop the size reduction. For
instance, a complete size reduction from 100 nm to less than 10
nm occurred within 30 min under irradiation of 59.9 mJ cm−2
(Supporting Information, Figure S3), while only a 30%
reduction in diameter was observed at 100 MPa. The
observation that the size reduction stopped at a fixed size,
even after repeated irradiations, is unusual; at ambient
pressures, repeated irradiations of already-size-reduced Au
NPs typically cause further size reductions, due to light
absorption, until the absorption cross-section of the particles is
negligible (<2 nm for excitation at 532 nm). To achieve a better
understanding of the photothermal mechanism at high
pressures, we performed laser irradiation at various pressures.
Figure 2 summarizes the mean diameters of the generated
spheres as a function of the laser fluence at 30, 60, and 100
MPa. The mean diameter decreased nonlinearly with increases
in the laser fluence. This observation showed that increasingly
greater laser fluences were required to form smaller core
particles. The laser fluence-dependent size reduction was also
found to be pressure-dependent (for example, see Supporting
Information Figure S4 for the size distribution at 60 MPa). For
increasing pressures (30 < 60 < 100 MPa), higher laser fluences
were required to achieve similar particle diameters. This might
1297
dx.doi.org/10.1021/la3046143 | Langmuir 2013, 29, 1295−1302
Langmuir
Article
instantaneous heating of the particle, and this is followed by
heat transfer to the surrounding medium.4,5 Although it has
been revealed that bubbles of water vapor are formed around
particles heated to 400−674 K at ambient pressure (0.1
MPa),11,12,23 a supercritical water layer can form around the NP
at temperatures above 647 K when the external pressure
exceeds 22.1 MPa.24 In this case, it is likely that the slow
evaporation observed at high pressures was mainly due to the
efficient particle cooling caused by the greater heat dissipation
to the medium in the supercritical water compared with that in
the bubble. As the pressure increased, we observed that it was
increasingly difficult for the Au NPs to undergo further size
reduction as the particle size was reduced (see Figure 2). This
can be explained by the increased convective/conductive heat
transfer to the medium, which resulted from the fact that the
core particles undergoing evaporation were smaller. Thus,
Scheme 2 functions as follows:
The particle temperature TP and the medium temperature
Tm under nanosecond pulsed-laser-heating are defined by a set
of differential equations:5
Figure 3. Mean diameters of Au NPs as a function of laser fluence at
30 MPa, after excitation at two wavelengths: 355 nm (red curve with
closed rectangles) and 532 nm (black curve with open circles), starting
from 100 nm diameter particles. The accumulated laser pulse numbers
are 108 000.
CP
dTP
=S−F
dt
(1a)
Cm
∂Tm
= ∇⃗(k m∇⃗Tm) + F
∂t
(1b)
where CP (J m−3 K−1) and Cm (J m−3 K−1) represent the heat
capacities of the particle and medium and km (W m−1 K−1)
represents the thermal conductivity of the medium. The
absorbed laser energy term S that contributes to the
temperature increase of the Au NPs and the heat transfer
term F that contributes to the temperature decrease of the Au
NPs are given by5
Figure 4. Mean diameters of Ag NPs as a function of laser fluence at
60 MPa, after excitation at two wavelengths: 355 nm (blue solid curve)
and 532 nm (red dotted curve), starting from 100 nm diameter
particles. The accumulated laser pulse numbers are 108 000.
S=
excitation wavelength of 532 nm, the average diameter of the
Ag NPs remained nearly constant at ≈90 nm, with a slight
decrease for fluences of up to 80 mJ cm−2, and a gradual size
reduction occurring at much higher laser fluences (100−180 mJ
cm−2) (Figure 5). At these high laser fluences, an approximately
bimodal size distributioninstead of a monodisperse distributionwas observed after the size reduction. Under atmospheric pressure, the threshold laser fluence for the size
reduction of the 100 nm Ag NPs was ≈15 mJ cm−2 at an
excitation wavelength of 355 nm and ≈50 mJ cm−2 at 532 nm.
The retardation resulting from the pressurization of the water
was remarkable for the Ag NPs, but the size reduction could
not be controlled by changing the laser fluence, in contrast with
the Au NPs. For the Ag NPs at 60 MPa, the slope of the
particle diameter versus laser fluence curve in the range from 35
to 45 mJ cm−2 for excitation at 355 nm was too steep to allow
control over the particle size; the curve measured at 532 nm
was not steep enough, and this resulted in polydispersity, rather
than the monodispersity observed for the Au NPs. In spite of
the unsatisfactory size control results, the observations made for
the Ag NPs were useful in giving general insights into the size
reduction phenomena.
Mechanistic Aspects. Scheme 2 shows a schematic
representation of the laser-induced size-reduction of an Au
NP in water at pressures p > 22.1 MPa. The absorption of the
laser energy from a nanosecond pulse by the Au NP allows the
CabsP
VP
F=h
(2a)
AP
(TP − Tm)
VP
(2b)
where P (W m−2) represents the laser power density, Cabs (m2)
is the temperature-dependent absorption cross section of a Au
NP, VP (m3) is the particle volume, AP (m2) is the NP surface
area, and h (W m−2 K−1) is the heat transfer coefficient.
Equation 2b predicts that the heat loss through the NP−water
interface is proportional to the surface-to-volume ratio AP/VP
and the heat transfer coefficient h. The coefficient h is defined
as (kmNu)/DP, where Nu denotes the Nusselt number27 and DP
is the particle diameter. When a constant temperature gradient
(Tp − Tm) is assumed (at constant external pressure), the heat
transfer term F of a sphere with diameter DP is given by
F∝
6k mNu
DP 2
≡K
(3)
Equation 3 predicts that the heat transfer is inversely
proportional to the square of the diameter of the sphere (F
∝ 1/DP2). Note that the thermal conductivity of the fluid km(T,
p) is a function of both temperature and pressure.28 Here, an
increase in the pressure results in an increased thermal
conductivity km, giving rise to greater heat dissipation.
Figure 6 gives an approximate picture of K (W m−3 K−1, scale
on the right), a measure of heat loss, calculated as a function of
particle diameter. The rate of heat loss (or cooling) for a
1298
dx.doi.org/10.1021/la3046143 | Langmuir 2013, 29, 1295−1302
Langmuir
Article
Figure 5. Histograms representing the particle size distributions of Ag NPs at higher laser intensities than those given in Figure 4 for 532 nm laser
excitation at 60 MPa: 113 mJ cm−2 (a), 131 mJ cm−2 (b), 145 mJ cm−2 (c), and 179 mJ cm−2 (d). Extinction spectral changes are given in
Supporting Information Figure S5. The laser pulse numbers are 108 000.
Scheme 2. Schematic Illustration Representing the
Nanosecond Laser-Induced Size Reduction of an Aqueous
Au NP at p > 22.1 MPaa
a
TP is the particle temperature, and Tevap is the evaporation
temperature for the AuNPs. The laser-induced heating of the particle
to TP ≈ Tevap sets the particle evaporation in motion, and if the
condition of TP ≥ Tevap is met, enhanced particle evaporation takes
place. This enhanced evaporation causes considerable size reduction,
which also brings about enhanced heat loss. As a result, the size
reduction stops when the condition TP < Tevap is met.
Figure 6. Heat loss rate as a function of particle diameter, calculated
using eq 3 (blue curve). The Nusselt number Nu is set to 18.1, and the
thermal conductivity km of the water medium is set as 0.58 W m−1 K−1.
The cooling rate constant as a function of DP (dashed red curve) is
also included, reproduced from ref 18. Both curves show a dependence
on DP−2.
also included in Figure 6 (scale shown on the left). The
experimental DP-dependent cooling rate constants agreed
qualitatively with those of our model. According to eq 1a, the
minimum core diameter (which is reached via surface
evaporation caused by increases in the temperature) is
determined by the laser heating term S, and the cooling term
F. At a given laser fluence P, the laser-induced size reduction of
an Au NP stops when the size at which the heat loss term
exceeds the heating term is reached, leading to a particle
temperature TP that is below the evaporation threshold. A
particle is greater for smaller particles. For example, the rate of
heat loss from a 70 nm Au NP is twice that of a 100 nm sphere,
and the rate of heat loss for a 50 nm particle is four times as
large as that of a 100 nm particle. This picture is consistent with
previous experimental results from Hartland’s group.18 They
determined a quadratic dependence of the cooling time
constants τcooling on the diameter of Au NPs (i.e., τcooling ∝
DP2) by applying femtosecond pump−probe spectroscopy. The
cooling rate constant τcooling−1 reproduced from their work is
1299
dx.doi.org/10.1021/la3046143 | Langmuir 2013, 29, 1295−1302
Langmuir
Article
Figure 7. Calculated Cabs/Vp versus particle diameter curves for various refractive indices of water, from 1.33 (liquid water at room temperature) to
1.0 (assuming supercritical water), for Au NPs after excitation at 532 nm (a), Au NPs after excitation at 355 nm (b), Ag NPs after excitation at 532
nm (c), and Ag NPs after excitation at 355 nm (d). Only the refractive index reduction was considered for the calculation of Cabs, and the increase in
particle temperature was not considered. The Mie theory calculation was carried out using a computational code given in ref 30.
Information Figure S6. During the size reduction, the
temperature rise causes alterations in the Cabs values as a result
of the progressive reduction of the refractive index of water.
The refractive index in supercritical water can be ≈1 at
pressures > 60 MPa and Tm > 1000 K. The water temperature
at the NP interface is able to reach such high values because of
the heat transfer.
Figure 7 shows the Cabs/Vp values as a function of particle
diameter, calculated for the Au and Ag NPs at excitation
wavelengths of 532 and 355 nm. For the Au NPs at 532 nm
(Figure 7a), the curve for n ≈ 1.0 was nearly flat, with a slight
decreasing trend with decreasing particle diameter from 80 to
30 nm. This could have been a foundation for the observed
controlled size reduction phenomenon, because in this case the
heating term S gradually increases with increasing P, and the
size reduction proceeds until the size-dependent cooling term F
becomes dominant. The Cabs/Vp versus DP curve was also
nearly flat for excitation at 355 nm (Figure 7b), but the values
at n ≈ 1.0 were 1.5 times greater than those observed with
excitation at 532 nm. This could have been a reason for the
slightly higher efficiency observed in Figure 3. The Cabs/Vp
versus DP curves for the Ag NPs were markedly different from
those observed for the Au NPs, and the effects of the two
excitation wavelengths on the Cabs/Vp versus DP curve were
remarkably different. Under excitation at 355 nm (Figure 7d),
the Cabs/Vp values increased significantly with decreasing DP.
This suggested that the contribution of the heating term
increased monotonically with decreasing DP values, resulting in
almost uncontrollable size reduction. This deduction is
larger laser power is therefore required to achieve further
reductions in size. The cooling rate vs DP relation qualitatively
explains the relationship between the core particle size and the
laser fluence, as shown in Scheme 2. At ambient pressure, the
heating term S far exceeds the cooling term F, because F ≈ 0
for Au NPs inside bubbles; this results in very rapid
evaporation. The high-pressure results shown here represent
a heat loss-controlled reaction; its nonequilibrium nature is
quite unique. Note that the highly efficient cooling mechanism
observed here at high pressures may not work under
femtosecond laser excitations. Under such excitation conditions, the time scale for heating the particle to its maximum
temperature is shorter than the cooling time constant of Au
NPs (20 ps to 1 ns).5,7 Thus, particle evaporation can occur
before the cooling of the particle sets in. In addition, the
Coulomb explosion mechanism can operate instantaneously
after femtosecond laser excitation, before the particle
evaporation takes place.5,7,29
We observed that the gradual, laser fluence-dependent coresize reduction did not occur for the Ag NPs. Here, the role of
the laser heating term S in eq 2a must be examined. This term
is dependent on Cabs/Vp at a given laser power density P, as
shown in eq 2a. The absorption of light by the NPs initially
takes place at ambient temperatures. However, the absorption
of light causes both Tp and Tm to increase during the excitation
pulse (FWHM: 5−6 ns), until the particle boiling temperature
is reached. The temperature increase affects the Cabs values,
because the values are dependent on both Tp and Tm. The
effect of Tm is much greater, as shown in Supporting
1300
dx.doi.org/10.1021/la3046143 | Langmuir 2013, 29, 1295−1302
Langmuir
Article
Figure 8. Particle images showing the effect of centrifugation on the sample shown in Figure 1d: (a) before, and (b) after the separation of the core
particles from the expelled fragments. The solution was centrifuged at 2000 rpm for 60 min. The supernatant that contained the fragments was
removed, and the core particles were dispersed again in distilled water. This process was repeated three times. The TEM picture in Figure 8b shows
the Au NPs 2 weeks after fabrication. The inset gives the corresponding size distribution. The particles showed no size changes after the
centrifugation treatment. To construct the histogram, 907 particles were measured. Additional TEM photographs of the Au NPs and extinction
spectra, together with Mie fits for the Au NP solutions after the cleaning process, are given in Supporting Information Figure S6.
■
consistent with the observations of the sudden size reduction
that took place at 35−45 mJ cm−2 for excitation at 355 nm
(Figure 4). Under excitation at 532 nm, the Cabs/Vp values were
an order of magnitude smaller (note the difference between the
vertical scale in Figure 7c and that in Figure 7d). This showed
that the excitation efficiency at this wavelength was very low,
which likely led to the inefficient size reduction that was
observed (Figures 4 and 5). Accordingly, the analysis based on
eq 1 qualitatively explains the entire size reduction process for
both the Au and Ag NPs. However, quantitative calculations
might be necessary to obtain a full understanding.
From a practical point of view, the current work suggests that
nanosecond pulsed lasers can be used efficiently for the sizecontrolled preparation of Au NPs. With eventual practical
applications in mind, the additional processing step of
separating the fragments expelled from the core particles was
therefore carried out. We demonstrated the separation of
monodisperse particles using centrifugation, as shown in Figure
8 and Supporting Information Figure S7. Such particles can be
used as building blocks for plasmonic crystals, for which NPs
with a polydispersity of less than 10% are required.31,32
■
ASSOCIATED CONTENT
S Supporting Information
*
Particle images (TEM micrographs) and corresponding size
distribution of 100 nm Au NPs, extinction spectra of 100 nm
Au NPs after laser irradiation, particle images (TEM micrographs) and corresponding size distributions after laser
irradiation at 0.1 MPa, particle images and corresponding
histograms for Au NPs prepared at 60 MPa, changes in the
extinction spectra for ≈100 nm diameter Ag NPs after laser
irradiation at 60 MPa, particle images (TEM micrographs),
absorption cross section spectra (Cabs) for Au NPs (d = 100, 80,
50 nm) in water, depending on the refractive index (n) of the
medium and the particle temperature (n = 1.33), and particle
images (TEM micrographs) and changes in the extinction
spectra for Au NPs after centrifugation. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected].
Notes
The authors declare no competing financial interest.
■
CONCLUSION
ACKNOWLEDGMENTS
This work was supported by KAKENHI (No. 23310065) from
the Japan Society for the Promotion of Science (JSPS) and the
Amada Foundation (AF-2011201). A JSPS fellowship to D.W.
(No. 201107976) is also gratefully acknowledged. Mr.
Tomoyoki Ueki, Mr. Satoshi Sugano, and Mr. Taka Fujimoto
are acknowledged for TEM photograph acquisition. We thank
Dr. Vasudevan P. Biju of AIST Shikoku Center for critical
reading of the manuscript.
We have achieved a technological advance in the pulsed laserinduced size-reduction of Au NPs in water by applying high
pressures of 30−100 MPa. Below <400 MPa, the high pressure
technique can be more easily implemented than ultrahigh
vacuum and ultralow temperature techniques. We demonstrated that the combination of three parametershigh
external pressures, the excitation laser intensity, and the
excitation wavelengthis instrumental to the performance of
photothermal layer-by-layer size reduction in a controlled
manner. As a result, we successfully prepared highly
monodisperse Au spheres with diameters in the 20−100 nm
range, starting from ca. 100 nm diameter Au particles.
Mechanistically, the management of the laser-generated heat
through convective/conductive transfer is important, as
demonstrated for the Au NPs. The success critically depended
on the size-dependent light absorption behavior of the NPs at
high temperatures. In this regard, the Ag NPs were not
particularly suitable for achieving controlled size reduction.
Overall, the introduction of an additional parameterthe high
pressurespromoted the better understanding of the nature of
liquid-phase laser processing of NPs.
■
REFERENCES
(1) Link, S.; El-Sayed, M. A. Shape and size dependence of radiative,
non-radiative and photothermal properties of gold nanocrystals. Int.
Rev. Phys. Chem. 2000, 19, 409−453.
(2) Burda, C.; Chen, X.; Narayanan, R.; El-Sayed, M. A. Chemistry
and properties of nanocrystals of different shapes. Chem. Rev. 2005,
105, 1025−1102.
(3) Bauer, C.; Abid, J.-P.; Ferman, D.; Giault, H. H. Ultrafast
chemical interface scattering as an additional decay channel for nascent
nonthermal electrons in small metal nanoparticles. J. Chem. Phys. 2004,
120, 9302−9315.
(4) Hartland, G. V. Optical studies of dynamics in noble metal
nanostructures. Chem. Rev. 2011, 111, 3858−3887.
1301
dx.doi.org/10.1021/la3046143 | Langmuir 2013, 29, 1295−1302
Langmuir
Article
(28) Haar, L.; Gallagher, J. S.; Kell, G. S. NBS/NRC Steam Tables;
Hemisphere Publishing: Washington, DC, 1976.
(29) Werner, D.; Furube, A.; Okamoto, T.; Hashimoto, S.
Femtosecond laser-induced size reduction of aqueous gold nanoparticles: In situ and pump-probe spectroscopy investigations revealing
Coulomb explosion. J. Phys. Chem. C 2011, 115, 8503−8512.
(30) Bohren, C. F.; Huffman, D. R. Absorption and Scattering of Light
by Small Particles; Wiley: New York, 1983.
(31) Park, S. Y.; Lytton-Jean, K. R.; Lee, B.; Weigand, S.; Schatz, G.
C.; Mirkin, C. A. DNA-programmable nanoparticle crystallization.
Nature 2008, 451, 553−556.
(32) Macfarlane, R. J.; Lee, B.; Jones, M. R.; Harris, N.; Schatz, G. C.;
Mirkin, C. A. Nanoparticle superlattice engineering with DNA. Science
2011, 334, 204−208.
(5) Hashimoto, S.; Werner, D.; Uwada, T. Studies on the interaction
of pulsed lasers with plasmonic gold nanoparticles toward light
manipulation, heat management, and nanofabrication. J. Photochem.
Photobiol., C 2012, 13, 28−54.
(6) Takami, A.; Kurita, H.; Koda, S. Laser-induced size reduction of
noble metal particles. J. Phys. Chem. B 1999, 103, 1226−1232.
(7) Werner, D.; Hashimoto, S. Improved working model for
interpreting the excitation wavelength- and fluence-dependent
response in pulsed laser-induced size reduction of aqueous gold
nanoparticles. J. Phys. Chem. C 2011, 115, 5063−5072.
(8) Inasawa, S.; Sugiyama, M.; Yamaguchi, Y. Bimodal size
distribution of gold nanoparticles under picosecond laser pulses. J.
Phys. Chem. B 2005, 109, 9404−9410.
(9) Giammanco, F.; Giorgetti, E.; Marsili, P.; Giusti, A. Experimental
and theoretical analysis of photofragmentation of Au nanoparticles by
picosecond laser radiation. J. Phys. Chem. C 2010, 114, 3354−3363.
(10) Matsuo, N.; Muto, H.; Miyajima, K.; Mafune, F. Single laser
pulse induced aggregation of gold nanoparticles. Phys. Chem. Chem.
Phys. 2007, 9, 6027−6031.
(11) Werner, D.; Hashimoto, S.; Uwada, T. Remarkable photothermal effect of interband excitation on nanosecond laser-induced
reshaping and size reduction of pseudospherical gold nanoparticles in
aqueous solution. Langmuir 2010, 26, 9956−9963.
(12) Kotaidis, V.; Dahmen, C.; Von Plessen, G.; Springer, F.; Plech,
A. Excitation of nanoscale vapor bubbles at the surface of gold
nanoparticles in water. J. Chem. Phys. 2006, 124, 184702.
(13) Siems, A.; Weber, S. A. L.; Boneberg, J.; Plech, A.
Thermodynamics of nanosecond nanobubble formation at laserexcited metal nanoparticles. New J. Phys. 2011, 13, 043018.
(14) Lukianova-Hleb, E. Y.; Lapotko, D. O. Influence of transient
environmental photothermal effects on optical scattering by gold
nanoparticles. Nano Lett. 2009, 9, 2160−2166.
(15) Lapotko, D. O. Optical excitation and detection of vapor
bubbles around plasmonic nanoparticles. Opt. Express 2009, 17, 2538−
2556.
(16) Sasaki, K.; Takada, N. Liquid-phase laser ablation. Pure Appl.
Chem. 2010, 82, 1317−1327.
(17) Ibrahimkutty, S.; Wagener, P.; Menzel, A.; Plech, A.;
Barcikowski, S. Nanoparticle formation in a cavitation bubble after
pulsed laser ablation in liquid studied with high time resolution small
angle x-ray scattering. Appl. Phys. Lett. 2012, 101, 103104.
(18) Hu, M.; Hartland, G. V. Heat dissipation for Au particles in
aqueous solution: Relaxation time versus size. J. Phys. Chem. B 2002,
106, 7029−7033.
(19) Sasaki, K.; Nakano, T.; Soliman, W.; Takada, N. Effect of
pressurization on the dynamics of a cavitation bubble induced by
liquid-phase laser ablation. Appl. Phys. Express 2009, 2, 046501.
(20) Soliman, W.; Takada, N.; Sasaki, K. Growth processes of
nanoparticles in liquid-phase laser ablation studied by laser-light
scattering. Appl. Phys. Express 2010, 3, 035201.
(21) Wei, S.; Saitow, K. In situ multipurpose time-resolved
spectrometer for monitoring nanoparticle generation in a highpressure fluid. Rev. Sci. Instrum. 2012, 83, 073110.
(22) Saitow, K.; Yamamura, T.; Minami, T. Gold nanospheres and
nanonecklaces generated by laser ablation in supercritical fluid. J. Phys.
Chem. C 2008, 112, 18340−18349.
(23) Caupin, F.; Herbert, E. Cavitation in water: a review. C. R. Phys.
2006, 7, 1000−1017.
(24) Weingärtner, H.; Franck, E. U. Supercritical water as a solvent.
Angew. Chem., Int. Ed. 2005, 44, 2672−2692.
(25) Werner, D.; Ueki, T.; Hashimoto, S. Methodological improvement in pulsed laser-induced size reduction of aqueous colloidal gold
nanoparticles by applying high pressure. J. Phys. Chem. C 2012, 116,
5482−5491.
(26) Atkins, P. W. Physical Chemistry, 4th ed.; Oxford University
Press: Oxford, 1990.
(27) Incropera, F. P.; DeWitt, D. P.; Bergman, T. L.; Lavine, A. S.
Fundamentals of Heat and Mass Transfer, 6th ed.; Wiley: New York,
2007.
1302
dx.doi.org/10.1021/la3046143 | Langmuir 2013, 29, 1295−1302