See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/26777656

Numerical investigation of heating of a gold nanoparticle and the surrounding

microenvironment by nanosecond laser pulses for nanomedicine applications

Article in Physics in Medicine and Biology · October 2009

DOI: 10.1088/0031-9155/54/18/013 · Source: PubMed

CITATIONS READS

81 848

3 authors, including:

Elisabetta Sassaroli King C Li

Brigham and Women's Hospital University of Illinois, Urbana-Champaign
60 PUBLICATIONS 732 CITATIONS 281 PUBLICATIONS 8,062 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Otto, PAMELA M View project

Tomographic Reconstruction from Truncated Cone-beam Projections View project

All content following this page was uploaded by Elisabetta Sassaroli on 19 November 2014.

The user has requested enhancement of the downloaded file.

IOP PUBLISHING PHYSICS IN MEDICINE AND BIOLOGY

Phys. Med. Biol. 54 (2009) 5541–5560 doi:10.1088/0031-9155/54/18/013

Numerical investigation of heating of a gold

nanoparticle and the surrounding microenvironment
by nanosecond laser pulses for nanomedicine
applications
E Sassaroli, K C P Li and B E O’Neill1
Department of Radiology, the Methodist Hospital Research Institute, Houston, TX, 77030, USA

E-mail: beoneill@tmhs.org

Received 21 May 2009, in final form 7 August 2009

Published 28 August 2009
Online at stacks.iop.org/PMB/54/5541

Abstract
We have modeled, by finite element analysis, the process of heating of a
spherical gold nanoparticle by nanosecond laser pulses and of heat transfer
between the particle and the surrounding medium, with no mass transfer. In our
analysis, we have included thermal conductivity changes, vapor formation, and
changes of the dielectric properties as a function of temperature. We have shown
that such changes significantly affect the temperature reached by the particle
and surrounding microenvironment and therefore the thermal and dielectric
properties of the medium need to be known for a correct determination of the
temperature elevation. We have shown that for sufficiently low intensity and
long pulses, it is possible to establish a quasi-steady temperature profile in the
medium with no vapor formation. As the intensity is increased, a phase-change
with vapor formation takes place around the gold nanoparticle. As phase-
transition starts, an additional increase in the intensity does not significantly
increase the temperature of the gold nanoparticle and surrounding environment.
The temperature starts to rise again above a given intensity threshold which is
particle and environment dependent. The aim of this study is to provide useful
insights for the development of molecular targeting of gold nanoparticles for
applications such as remote drug release of therapeutics and photothermal
cancer therapy.

1. Introduction

Noble metal nanoparticles strongly absorb and scatter specific wavelengths of optical radiation
owing to their surface plasmon resonance (SPR) properties (Bohren and Huffman 1983,
Kreibig and Vollmer 1995). SPR is caused by the collective excitation of the conduction
1 Author to whom any correspondence should be addressed.

0031-9155/09/185541+20$30.00 © 2009 Institute of Physics and Engineering in Medicine Printed in the UK 5541
5542 E Sassaroli et al

electrons, called plasmon oscillation. The frequency of this oscillation is called the plasmon
resonance frequency. Because of the long-range nature of the forces involved in plasmon
oscillation, in small systems, such as nanoparticles, the SPR frequency is affected by the
size and shape of the particle, which provide a unique tunability to the nanoparticle optical
properties. Other factors that influence the SPR frequency are the dielectric properties of the
surrounding medium and particle–particle interactions. Gold nanoparticles (GNPs) are easily
produced and can be made stable by adding surface ions or molecules on them, thus preventing
their clustering. In addition, they are also relatively easy to conjugate with ligands and other
biological molecules and they are inert, which make them relatively biocompatible. These
properties make GNPs good candidates for molecular targeting.
Because of their unique optical properties, GNPs have been the subject of intense
investigation for a variety of biomedical applications, such as contrast enhancement in optical
imaging, for thermal ablation in cancer treatment, and as chemical and biological sensors (Hu
et al 2006, Jain et al 2007, Sperling et al 2008). In imaging, GNPs are very attractive as contrast
agents as they can be visualized with several different methods. The scattering of light from
GNPs is intense enough to allow sensitive imaging by conventional light microscopy (Souza
et al 2006, Tkachenko et al 2004). The large absorption of optical radiation exhibited by GNPs
is efficiently converted to heat, which leads to localized heating of the medium immediately
surrounding the GNP. This phenomenon has been exploited in two ways. Photothermal
imaging records the local variation of the refractive index by microscopy providing enough
sensitivity to allow for detection at the single-particle level (Berciaud et al 2004, Boyer et al
2002). In photoacoustic imaging, on the other hand, the acoustic waves generated by heat
absorption are detected by ultrasound transducers. This modality makes possible the detection
in vivo and in vitro of cells targeted by GNPs for application in cancer and inflammatory
processes (Eghtedari et al 2007, Kim et al 2007, Mallidi et al 2007).
In therapeutic applications, GNPs have been used for a long time to deliver molecules
into cells (Sperling et al 2008). More recently, the SPR heating has been employed for cell
killing in vitro and in vivo (Hirsch et al 2003, Pitsillides et al 2003), the opening of chemical
bonds such as double-strand DNA (Hamad-Schifferli et al 2002), and the opening of polymeric
capsules (Angelatos et al 2005).
The strong dependence of the SPR frequency on the local microenvironment has made
GNPs an attractive possibility for sensor applications in biotechnology. The binding of
molecules to the particle surface changes the plasmon resonance frequency as it can be
detected by the light scattered from the particle in dark field microscopy (Raschke et al 2003).
This change is even more dramatic when GNPs form small aggregates which give rise to a
red shift in the plasmon resonance frequency (Ghosh and Pal 2007). This effect of plasmon
coupling has been used for calorimetric detection of analytes (Mirkin et al 1996, Moller and
Fritzsche 2007) and could provide a way to differentiate cancer cells from surrounding benign
cells (Sokolov et al 2003).
In this paper, we present our numerical investigation of the heating and vapor formation
around a single GNP. More specifically, we make use of a finite element algorithm to solve
the heat equation for a spherical GNP and surrounding water environment. We assume
that the GNP is irradiated by nanosecond laser pulses and we model thermal conductivity
changes, vapor formation, and changes of the dielectric properties as a function of the
exposure conditions. Modeling laser heating of a single-gold nanoparticle surrounded by
a water medium is a first step towards understanding the thermal processes of heating in
relation to possible biomedical applications, such as, for example, the opening of chemical
bonds and to remotely control the release of drug from carriers by photo-induced heating of the
gold nanoparticles. The present model has nevertheless several limitations. Among those, we
Pulsed laser heating of gold nanoparticles 5543

have assumed that the nanoparticle suspension is sufficiently diluted so that particle–particle
interaction can be neglected. The velocity difference between the fluid and the GNP, which
is assumed to be rigid, is considered small. The possible presence of stabilizing molecules
on the GNP surface has not been considered. In addition, the oscillation and collapse of the
bubble that might form around the GNP is not modeled in the present study. The aim of this
investigation is to provide useful insights for the optimization of nanomedicine applications
of GNPs.

2. Mathematical model

2.1. Model outline

We have modeled by finite element simulations the process of heating of a single GNP which
absorbs laser radiation and of heat transfer between the particle and the surrounding medium
with no mass transfer. The temperature for gold melting is Tm = 1336 K (1063 ◦ C). In our
analysis we assume that the temperature of the GNP remains below Tm. In addition, the
particle is considered to be spherical, rigid and larger than 20 nm so that quantum mechanical
and surface effects can be neglected. Under these conditions, the process of heating of a GNP
subject to laser pulses of at least nanosecond duration with no mass transfer can be described
by (Pustovalov 2005)
∂Tp 1 ∂
ρp cp (Tp ) = 2 (r 2 kp (Tp )∇Tp ) + Slaser (1)
∂t r ∂r
where ρ p and cp are the density and the specific heat of the particle material (gold), Tp is the
temperature, kp is the thermal conductivity, and Slaser is the source term. It is important to
point out here that for nanomedicine applications to cancer treatment, it is critical to consider
particles within the range of 10–100 nm. This is because nanoparticles smaller than 10 nm are
rapidly eliminated through the kidney and particles within 10–100 nm travel throughout the
blood stream and can leak into the tumor tissue owing to the abnormal tumor blood vessels
(Heath and Davis 2008).
The fluid temperature is determined by the equation
DTf Dpf
ρf cpf − βf Tf = ρf f + ∇ · (kf ∇Tf ) (2)
Dt Dt
with ρf fluid density, cpf specific heat at constant pressure, kf thermal conductivity, β f the
thermal expansion, f is the thermal dissipation function which is proportional to the squares
of the spatial derivatives of the velocity components and D/Dt is defined by Dt D
= ∂t∂ + u · ∇.
Equations (1) and (2) are coupled through the boundary conditions at the nanoparticle surface.
Therefore, in general, the particle temperature depends on the pressure and temperature in
the fluid. However, because of the small size of the particle, a few simplifications can be
made. Small, free particles adjust themselves very quickly to fluid velocity variations so that
the velocity difference between particle and fluid is usually small. Thus, we can assume
f = 0 in equation (2). Similarly we can neglect the convective term on the left-hand side of
equation (2). The next simplification involves the pressure field in the fluid. For slow motion
the fluid may be taken as incompressible, in which case βf Tf → 0. Hence, the external fluid
temperature is described by Fourier’s heat conduction equation
 
∂Tf 1 ∂ ∂
ρf cpf = 2 2
r kf (T ) Tf (3)
∂t r ∂r ∂r
where ρf is the fluid density, cpf is the specific heat at constant pressure, and kf is the thermal
conductivity. For very short laser pulses in the picoseconds region, explosive boiling of the
5544 E Sassaroli et al

surrounding fluid with bubble and pressure wave formation may occur (Faraggi and Gerstman
2007, Hartland 2006). In our investigation, we are primarily interested in thermal effects with
little or none of the laser energy channeled into pressure waves or vaporization, and therefore
we consider sufficiently long laser pulses.
The initial condition at t = 0 is
Tp (r, 0) = Tf (r, 0) = T∞ (4)
and the boundary condition on the particle surface is
Tp (a, t) = Tf (a, t)
∂ ∂ (5)
kp (Tp ) Tp (a, t) = kf (Tf ) Tf (a, t)
∂r ∂r
where a is the particle radius. Furthermore Tp should be finite for r → 0, and Tf = T∞ for
r → ∞.
The source term Slaser can be expressed in terms of the absorption cross-section per unit
volume (Bohren and Huffman 1983) as
Cabs
Slaser = I (6)
vp
where vp is the particle volume, Cabs absorption cross-section with the dimension of an area,
and I is the laser intensity or irradiance with the dimensions of an energy per unit area and unit
time. I is the time average of the Poynting vector S. If E is the energy emitted by a laser at a
given wavelength, the mean density of energy in the laser beam is given by U  = E/π r 2 ctL
where r is the beam radius, c is the speed of light and tL is the laser pulse duration. The flux
mediated in time is then S = I = c U . The absorption cross-section Cabs is calculated by
Mie theory (Bohren and Huffman 1983), which provides an exact solution for the absorption
cross-section for a spherical particle of arbitrary refraction index and size. When a particle
is placed in a beam of electromagnetic radiation, it can scatter and absorb electromagnetic
radiation. This phenomenon is described by the scattering Csca and absorption cross-sections
Cabs, respectively. The scattering and absorption cross-sections are defined respectively as
the ratio of the power scattered and absorbed to the incident plane wave intensity. The sum
of the scattering and absorption cross-sections define the extinction cross-section Cext. Cext
is the ratio of power lost through absorption and scatter to the intensity of the incident plane
wave. The extinction cross-section is usually defined in terms of the dimensionless extinction
efficiency factor as Cext = π a 2 Qext with
Qext = Qsca + Qabs . (7)
In Mie theory, the extinction and the scattering efficiency factors can be written respectively
as (see for example Hartland 2006)
∞
2 
Qext = (2n + 1))Re(an + bn ) (8a)
x 2 n=1
∞
2 
Qsca = (2n + 1)(|an |)2 + |bn |2 ) (8b)
x 2 n=1

with
ψn (mx)ψn (x) − mψn (mx)ψn (x)
an = (9a)
ψn (mx)ςn (x) − mψn (mx)ςn (x)
Pulsed laser heating of gold nanoparticles 5545

mψn (mx)ψn (x) − ψn (mx)ψn (x)

bn = (9b)
mψn (mx)ςn (x) − ψn (mx)ςn (x)
where ψ n = (π z/2)1/2 × Jn+1/2(z), ζ n = (π z/2)1/2 × (Jn+1/2(z) − Yn+1/2(z)) with Jn+1/2 and
Yn+1/2 Bessel functions, x = 2π anf /λ and m = nnano /nf , where nnano and nf are respectively
the refractive index of the GNP and the fluid. Hence from the above equations, one can derive
an expression for the coefficient Cabs defined in equation (6). For particles of small sizes,
as considered in this study, only two terms in the series in equations (8) and (9) need to be
evaluated.
Equations (9a) and (9b) are valid for spherical nanoparticles which have strong absorption
in the visible region of the electromagnetic spectrum. A class of gold nanoparticles, called
metal nanoshells, have attracted attention for biomedical applications because they show
an intense absorption spectrum in the near-infrared region of the electromagnetic spectrum
and therefore can penetrate more deeply into biological tissues (Hirsch et al 2003). Metal
nanoshells consist of a spherical dielectric core, usually silica, surrounded by a thin shell
of noble metal such as gold (Averitt et al 1999). For a nanoshell, the coefficients given in
equations (9) are no longer valid. Expressions that are valid for a coated particle, such as a
nonoshell, are known and they can be found for example in Bohren and Huffman (1983).
In order to model numerically the phase transition from liquid to vapor in the layer
immediately surrounding the GNP, we introduce the following effective quantities for the fluid
heat capacity ρf cpf and thermal conductivity kf as (Comini et al 1974, Fachinotti et al 1999,
Voller and Swaminathan 1990):
ρf cpf = Hρv (T )(cpv (T ) + Lv δ(T − Tvap )) + (1 − H )ρw (T )(cpw (T ) + Lv δ(T − Tvap )) (10)
kf = H kv (T ) + (1 − H )kw (T ) (11)
where H is the Heaviside step function, ρ v , cpv, kv are respectively the density, the specific
heat and the thermal conductivity for the vapor, ρ w , cpw, kw are the same quantities for the
liquid, Lv is the latent heat of vaporization, Tvap is the vaporization temperature (100 ◦ C at
atmospheric pressure), and δ has the properties of a delta function, i.e. it is defined by the
condition  ∞
ρLv δ(T − Tvap ) dT = ρLv . (12)
−∞
The step function H describes the different thermal properties of vapor and liquid water with
H equal to 1 for vapor and 0 for liquid water. H amounts to the volume fraction of vapor within
the model element. The δ function is the derivative of H with respect to the temperature.
As the liquid surrounding the GNP turns into vapor, the index of refraction of the medium
is also affected. We have modeled numerically this effect, by introducing an effective index
of refraction for the fluid as
nf = H nv + (1 − H )nw (13)
where nv is the index of refraction of the vapor and nw is the index of refraction of the
liquid. The change in nf affects the coefficient m in equation (9) and therefore the absorption
cross-section in equations (7) and (8).

3. Numerical results

3.1. GNP absorption cross-section

In this section, the numerical predictions of the mathematical model described above will be
illustrated. We are primarily interested in investigating thermal effects, so we will consider
5546 E Sassaroli et al

Table 1. Input parameters and functions.

Parameter Symbol Unit

Water density ρ w (T) kg m−3 Interpolated data from Holman (1997)

Water specific heat cpw(T) J kg−1 K Interpolated data from Holman (1997)
Water thermal conductivity kw (T) W m−1 K 0.0015T + 0.1689
Vapor density ρ v (T) kg m−3 0.1 × 106 × 18 × 10−3/8.314/T
Vapor specific heat cpv (T) J kg−1 K −0.0002T2 + 0.7929T + 1673.1
W m−1 K
2−7.8203log10(abs(T))+7.1242)
Vapor thermal conductivity kv (T) 10(1.7186log10(abs(T))

Water latent heat of vaporization Lv J kg−1 2.26 ×106

Gold density ρp kg m−3 193 ×104
Gold specific cp (T) J kg−1 K 111.6515 + 0.066T-2.6554×10−5T2
Gold thermal conductivity kp(T) W m−1 K 335.4544–5.8253×10−2T-7.4773×10−6T2
Au plasma resonance frequency ωp Hz 1.3703×1016
Au bulk damping constant γ bulk Hz 3.750×1013
Au Fermi velocity vF m s−1 1.3953×106

here regions of the parameter space where such effects are predominant. The numerical
constants and functions used in the simulations are summarized in table 1.
For the purpose of illustration, in figure 1 we have plotted the absorption efficiency factor
Qabs given by equations (7)–(9) at room temperature as a function of the wavelength for a =
15, 25 nm in an aqueous solution (nw ∼ = 1.33) and in a gaseous environment (nv ∼ = 1.0002).
For the GNP sizes considered here, only the dipole (n = 1) and quadrupole (n = 2) terms need
to be evaluated. As may be seen, Qabs shows a maximum around 520–530 nm for the GNPs
in water and at around 500–510 nm in a gaseous environment. Therefore, the overall energy
absorption is not only affected by the GNP radius but also by the environment surrounding
the GNP. The absorption in a gaseous solution is much less effective and its maximum is blue
shifted. The data used in figure 1 for the dielectric function of gold were taken from Johnson
and Christy (1972), who measured the dielectric function of bulk gold in the optical range at
room temperature. Strictly speaking, the dielectric function of a bulk metal cannot be used
to describe optical properties of nanoparticles, this is because of quantum mechanical and
surface effects. However, for gold nanoparticles with diameters of 20 nm and larger, the bulk
dielectric function gives a reasonable good agreement between the experimental spectra and
the Mie theory (Sonnichsen et al 2002). The dielectric function of bulk gold is in general
a function of temperature. Recently, Rashidi-Huyeh and Palpant (2006) have calculated the
temperature dependence of the dielectric function of gold and confirmed experimentally their
results in a later study (Palpant et al 2007). We have made use of their data in figure 2, where
we have plotted the absorption efficiency factor Qabs as a function of temperature for a GNP
of radius a = 25 nm and wavelength λ = 532 nm respectively in (a) liquid water and in (b) a
gaseous suspension. In (a) it may be seen that the absorption efficiency factor decreases as a
function of temperature. In (b) Qabs is much less sensitive to temperature changes and shows
a maximum at around 260 ◦ C.
For the purpose of comparison, we have shown in figure 3 the absorption efficiency factor
for a silica–gold nanoshell. Since the gold shells of these nanoparticles are usually thinner than
the bulk electron-free path in gold (about 10 nm; Kreibig and Vollmer 1995) the contribution to
the dielectric function due to the size-dependent electron scattering becomes important. This
effect has been incorporated into the Mie theory by Averitt et al (1999) who have proposed to
Pulsed laser heating of gold nanoparticles 5547

3.5

3
Absorption Efficiency Factor

2.5

1.5

0.5

0
400 450 500 550 600 650 700 750 800
(nm)

Figure 1. Absorption efficiency factor Qabs as a function of the wavelength. The solid (-) and
dotted lines (·) show Qabs for a = 25 nm, respectively, in an aqueous and a vapor suspension. The
dashed (–) and dash-dot lines (-·) illustrate Qabs for a = 15 nm, respectively, in an aqueous and a
vapor suspension.

3.6 1
a= 25 nm a= 25 nm
=532 nm =532 nm
0.99
3.55

0.98
3.5
0.97

3.45
0.96
Qabs
Qabs

3.4 0.95

0.94
3.35

0.93
3.3
0.92

3.25
0.91

3.2 0.9
20 40 60 80 100 100 150 200 250 300 350 400
T (0C) T (0C)

Figure 2. Absorption efficiency factor as a function of the temperature for a = 25 nm and

wavelength λ = 532 nm, respectively, in (a) liquid water and in (b) a gaseous suspension.
5548 E Sassaroli et al

3.5
b=28 nm a =25 nm

Absorption Efficiency Factor 2.5

1.5

0.5

0
400 600 800 1000 1200 1400 1600
(nm)

Figure 3. Absorption efficiency factor as a function of the wavelength for a silica–gold nanoshell
with inner radius a = 25 nm and outer radius b = 30 nm in a liquid environment.

modify the bulk dielectric function of gold as

ωp2 ωp2
ε(δ, ω) = εexp (ω) + − (14a)
ω2 + iωγbulk ω2 + iω

where δ is the shell thickness, εexp is the experimental bulk dielectric function, ωp is the
bulk plasmon resonance frequency, γ bulk is the bulk damping constant, and  describes the
modification of the bulk collision frequency as
νF
 = γbulk + A (14b)
δ
where vF is the Fermi velocity and A is a constant assumed equal to 1 in the above model. In
figure 3 we have plotted the first two terms (n = 1, 2) of the efficiency absorption cross-section
in a water environment with the modified dielectric function of gold given by equation (14a)
for a silica-gold nanoshell with inner radius a = 25 nm and outer radius b = 28 nm. The values
of the coefficients an and bn valid for a nanoshell have been taken from Bohren and Huffman
(1983). The absorption cross-section peaks in the near-infrared region and the absorption
strength is comparable to the one of a spherical gold particle of similar size as may be seen
in figure 1 for a spherical GNP of radius a = 25 nm. Therefore we should expect a similar
thermal behavior for a spherical GNP and a nanoshell of comparable size.

3.2. Thermal plots

For easy implementation and reliability of the solution, the thermal equations have been solved
using a commercial finite element package Comsol Multiphysics 3.5 (Comsol Inc., Burlington,
MA), which provides state-of-the-art routines to solve partial differential equations (PDEs).
The thermal properties for water and gold have been considered as a function of temperature
(see table 1). Details about the method of solution can be found in the Appendix. The
temperature dependence of the dielectric function of gold has also been taken into account in
Pulsed laser heating of gold nanoparticles 5549

100

90

80

70
T( 0C)

60

50

40

30

20
0 5 10 15 20 25 30
r/a

Figure 4. Temperature evolution as a function of the dimensionless distance r/a during a laser
pulse of duration TL = 1000 ns and intensity I = 2.5 × 109 W m−2 with a = 25 nm. The first
ten plots represent the temperature from t = 0 to 10 ns and the remaining plots from 50 ns up to
1000 ns in step of 50 ns.

the simulations. The changes of the thermal and dielectric properties of the fluid surrounding
the GNP have been modeled by introducing the Heaviside step function as discussed in section
2. Because of the discontinuity at T = Tvap, problems with convergence may arise. To avoid
such problems, a Comsol built-in smoothed Heaviside step function has been used in the
simulations. The smoothed Heaviside step approximates the discontinuity in an appropriate
interval around T = Tvap with a 7th degree polynomial chosen so that the second derivative is
continuous. The delta function in equation (12) has been implemented by taking the derivative
of the smoothed Heaviside step using a Comsol built-in differential operator.
Figure 4 shows the time evolution of the temperature for a spherical GNP and surrounding
fluid (water) environment during a laser pulse of duration tL = 1000 ns. The temperature in
degree centigrade is given as a function of the distance r/a from the nanoparticle center. In the
plot, we have assumed a = 25 nm, and the laser intensity I = 2.5 × 109 W m−2. The first ten
plots represent the temperature from t = 0 to 10 ns and the remaining plots from 50 ns up to
1000 ns in step of 50 ns. As it may be seen, inside the nanoparticle the temperature is spatially
uniform and becomes quasi-steady, i.e. it does not significantly change in time, at about 150–
200 ns. It is worth pointing out that the characteristic time required for the formation of a
a2
quasi-steady temperature profile inside a particle is τp ≈ 4κ p
(Carslaw and Jaeger 1986), where
κp is the coefficient of thermal diffusion of the particle. For gold κp ∼ = 1.2 × 10−4 m2 s−1 ,
−10
hence for a nanoparticle with radius a = 25 nm, τp ≈ 10 s and for a laser pulse duration
tL 10−10 s, the temperature T inside the particle can be considered uniform. In figure 5,
we have considered the heating as a function of time at specific locations, respectively r/a =
1, 1.1, 1.2, 1.3, 1.4, 2.0, 2.5 for the same intensity as in figure 4 and pulse duration of 300 ns.
After 50 ns, the temperature remains above 70 ◦ C up to a distance of about 10 nm (r/a = 1.4)
from the GNP surface, above 50 ◦ C for a distance of 25 nm (r/a = 2) and above 40 ◦ C for a
distance of about 37.5 nm (r/a = 2.5).
5550 E Sassaroli et al

100
r/a=1

90 r/a=1.1

r/a=1.2
80 r/a=1.3
r/a=1.4
70
T( 0C)

60 r/a=2.0

r/a=2.5
50

40

30

20
0 50 100 150 200 250 300
Time(ns)

Figure 5. The quasi-steady temperature profile as a function of time reached by the GNP and by
several locations in the nanoenviroment of the GNP is illustrated for I = 2.5 × 109 W m−2.

The characteristic time for heat exchange between a single nanoparticle and the
surrounding medium with the formation of a quasi-steady distribution of temperature in
2
the medium is about τm ≈ 4κa m , where κm is the coefficient of thermal diffusion of the
medium. For water κm ∼ = 1.53 × 10−7 m2 s−1 , therefore for nanoparticles with radius
−9
a = 25 nm, τm ≈ 10 s and laser pulse duration tL τm , the approximation of quasi-
stationary heat exchange between the particle and its environment is justified. In such a case,
an analytical solution (Carslaw and Jaeger 1986) can be obtained for the temperature in the
liquid surrounding the particle
 
(T0 − Tin ) r −a
Tf = Tin + erf √ (15)
r 2 κm t
where Tin is the initial temperature and T0 is the particle surface temperature. This solution
is valid for temperatures below the phase transition temperature and assuming the thermal
properties of the fluid constant.
In figure 6, we have plotted the cooling of the GNP and surrounding fluid
microenvironment during the first 50 ns after the end of the laser pulse of duration tL =
300 ns at specific locations r/a = 1, 1.4, 2, 10 for the same intensity as above. The temperature
remains above 50 ◦ C for above 5 ns at distance up to 25 nm (r/a = 2) from the GNP surface.
At a distance of 225 nm from the particle surface, the temperature remains constant and a few
degrees above the initial condition up to 50 ns after the end of the laser pulse.
As the intensity is further increased, phase transition from liquid to vapor takes place
in a very narrow region surrounding the GNP where the temperature reaches 100 ◦ C. The
quasi-stationary temperature profile in the liquid as discussed above cannot take place at
such intensities. For a given pulse duration tL, there is an intensity threshold at which phase
transition occurs. In figure 7, we have plotted the temperature reached at the end of the
laser pulse by a GNP of radius a = 25 nm as a function of r/a for a pulse duration tL =
6 ns at various intensities I = 4.0, 8, 12, 14, 16, 18 × 109 W m−2. Phase and dielectric
Pulsed laser heating of gold nanoparticles 5551

100
r/a=1
r/a=1.4
90 r/a=2.0
r/a=10

80

70
T( 0C)

60

50

40

30

20
0 5 10 15 20 25 30 35 40 45 50
Time (ns)

Figure 6. The cooling of the GNP and surrounding fluid microenvironment during the first
50 ns after the end of the laser pulse of duration tL = 300 ns at specific locations for I = 2.5 ×
109 W m−2.

250

I=4.0x109 W/m2
I=8.0 x109 W/m2
I=12 x109 W/m2
200
I=14 x109 W/m2
I=16 x109 W/m2
I=18 x109 W/m2

150
T( 0C)

100

50

0
0 0.5 1 1.5 2 2.5 3 3.5 4
r/a

Figure 7. The temperature reached at the end laser pulse by a GNP of radius a = 25 nm as a
function of r/a for a pulse duration tL = 6 ns at various intensities.
5552 E Sassaroli et al

100
I=5 x109 W/m2

90

80

70
T( 0C)

60

50

40

30

20
0 1 2 3 4 5 6
Time(ns)

Figure 8. The temperature as a function of time in a narrow region of a few nanometers at the
boundary between the GNP and the fluid for I = 5.0 × 109 W m−2. Lines, from top to bottom,
give the temperature at 0–5 nm distance from the gold.

changes in the surrounding fluid start when the temperature reaches 100 ◦ C which occurs at
about I = 4.0 × 109 W m−2. As the intensity is further increased, the temperature of the
GNP and surrounding medium increase very slowly because part of the energy is used in the
phase transition. Furthermore, as the phase change takes place, the dielectric properties of
the surrounding medium also changes, and the absorption of optical radiation becomes less
effective. The temperature starts to rise again above 14 × 109 W m−2.
As vapor forms, a bubble grows around the GNP which at the end of the laser pulse can
collapse and/or undergo oscillations, which are not modeled in this study. Thermodynamical
considerations and experimental studies show that explosive boiling occurs when the
temperature reaches 80–90% of the critical value (Avedisian 1985, Debenedetti 1996, Dou
et al 2001). The critical temperature of water at 1 atm is 640–650 K (367–377 ◦ C). The
highest exposure condition in figure 7 shows a temperature at which explosive boiling and the
possible occurrence of strong cavitation may take place.
Finally, figures 8 and 9 show the temperature as a function of time in a narrow region
of about five nanometers at the boundary between the GNP and the fluid, for an intensity
of respectively I = 5.0 and 15 × 109 W m−2 and pulse duration as above. In figure 8, the
temperature reaches about 100 ◦ C at 3 ns, and in figure 9 almost immediately after the start of
the laser pulse. For longer laser pulses the start of the phase transition takes place at smaller
intensities. For example for tL = 12 ns it takes place at about 3.6 × 109 W m−2.

4. Discussion

This paper has numerically investigated the heating by laser irradiation of a diluted suspension
of spherical GNPs including thermal and dielectric changes in the medium as well as vapor
formation surrounding the GNP.
Several investigators (Ekici et al 2008, Faraggi and Gerstman 2007, Goldenberg and
Tranter 1952, Kotaidis et al 2006, Neumann and Brinkmann 2005, 2007, Richardson et al
Pulsed laser heating of gold nanoparticles 5553

110
I=15 x109 W/m2

100

90

80

70
T( 0C)

60

50

40

30

20
0 1 2 3 4 5 6
Time(ns)

Figure 9. The temperature as a function of time in a narrow region of a few nanometers at the
boundary between the GNP and the fluid for I = 15.0 × 109 W m−2. Lines, from top to bottom,
give the temperature at 0–5 nm distance from the gold.

2009, Pitsillides et al 2003, Pustovalov 2005, Sun et al 2000, Volkov et al 2007) have modeled
the problem of GNP heating and heat transfer to the surrounding environment under a variety
of laser exposure conditions and particle properties. To cite a few of them, Pustovalov (2005)
has investigated theoretically the heating of a single GNP by integrating equation (1) over the
particle volume to obtain the following ordinary differential equation (ODE):
dT0 1
ρp cp vp = I Cabs S0 − Jc S0 (16)
dt 4
where T0 is the GNP (surface) temperature, S0 = 4π a 2 , Cabs is the n = 1 absorption efficiency
∂T
factor in Mie theory, I is the laser intensity and Jc = −kf ∂rf |r=a is the heat flux. He
obtained an analytical solution for equation (16), by approximating the temperature of the
fluid surrounding the GNP as
a
Tf = Tamb + (T0 − Tamb ) (17)
r
where Tamb is the ambient temperature. The knowledge of the temperature Tf allows to
∂T 
determine the heat flux J = −kf ∂rf r=a (assuming kf constant) and therefore to solve equation
(16) analytically. The solution during the laser pulse is
T0 = Tamb + C1 [1 − e−C2 t ] (18)
with C1,2 appropriate coefficients as determined by the initial and boundary conditions applied
to equation (16). The cooling of the GNP can also be obtained as
T0 = Tamb + (Tmax − Tamb ) e−C(t−tL ) (19)
where Tmax is the GNP temperature at the end of the laser pulse of duration tL and C is an
appropriate coefficient.
Richardson et al (2009) have investigated the collective heating by continuous wave laser
irradiation of a water-droplet containing GNPs. Their theoretical analysis has some features
5554 E Sassaroli et al

in common with Pustolov’s work. In their model, the heating process is described by an
empirical equation in which the rate of energy (temperature) increase in the water droplet
equals the energy absorbed from the laser field minus the energy lost by the droplet owing to
conduction at its boundary. Only the thermal properties of water are explicitly considered in
the equation, the effect of the presence of the GNPs is modeled with two parameters: the heat
loss coefficient and the so-called efficiency of light-to-heat conversion factor. The absorbed
energy is given by an expression containing the laser intensity, an experimentally determined
absorbance coefficient and the light-to-heat conversion factor, which they determined to be
close to 1. They obtained this latter coefficient together with the rate of heat loss by conduction
by fitting their experimental data with the solutions of their heat equation for the droplet. Those
solutions are of the type given by equations (18) and (19) with T0 water droplet temperature.
They also used a simplified model to describe the heating of a single-spherical GNP by
considering an approximate quasi-steady solution for the fluid surrounding the GNP of the
type given by equation (17) with a (T0 − Tamb ) replaced by vp q0 /4π kf , where q0 is the rate
of heat generation within a single GNP. They estimated the temperature increase at the GNP
surface without solving equation (16) but they used equation (17) evaluated at r = a to obtain
an expression of the type T0 − T∞ = (vp /4π kf )(q0 /a).
In our investigation, we have solved equations (1) and (3) directly by finite element without
making any assumption on the fluid temperature. The thermal and dielectric coefficients in
equations (1) and (3) have been assumed temperature dependent and we have considered
two terms in the Mie absorption cross-section series. Furthermore, we have modeled
vapor formation around the GNP. Our simulations show that the quasi-steady temperature
profile in the fluid surrounding the GNP can be approximated by equation (15) rather than
equation (17).
The above papers are primarily concerned with the physical process of heating of GNPs
and of heat transfer to the surrounding environment without any attempt to model the in vivo
situation. Several studies however can be found in the literature that specifically investigate
GNPs as a way to enhance photothermal heating of tissue for cancer treatment (Choi et al
2007, Dickerson et al 2008, Elliott et al 2007, 2008, El-Sayed et al 2006, Gobin et al 2007,
Hirsch et al 2003, Huang et al 2006, Huff et al 2007, Pissuwan et al 2007). In those studies
the goal is to induce an overall increase of temperature in the targeted tissue enough to induce
cancer cell death. In general, long, low power exposures are used to globally heat the tissue
up to 20 ◦ C above the tissue temperature. For example, (Elliott et al 2007) have performed
gel phantom studies to characterize the thermal response of GNPs. In their work, the power
density inside phantoms containing different concentrations of nanoparticles was modeled
using the optical diffusion approximation and the temperature distribution inside the phantom
was simulated using the finite element technique. Magnetic resonance temperature imaging
was used to visualize the spatio-temporal distribution of the temperature in the phantoms. In
most cases, very good agreement was obtained between the simulations and the experiments.
In our investigation, we are primarily interested in understanding the heating produced
by a single GNP for applications such as molecularly targeted photothermal therapy. Cells
naturally ingest colloidal nanoparticles (Chithrani et al 2006) and this uptake can be made more
effective via receptor–ligand interaction which specifically targets cancer cells. Once the GNP
is internalized, it could potentially be made to selectively target, heat, and damage organelles
or molecules within the cell itself, or to heat and release drug from an attached thermally
sensitive drug carrier, all without causing significant thermal damage to adjacent cells.
Another area of interest for medical applications is the use of GNPs for externally
triggering the release of drugs from a thermal sensitive liposome, polymer capsule and gel
composites. It has been shown that by incorporating GNPs inside the carrier or at its wall,
Pulsed laser heating of gold nanoparticles 5555

remotely controlled release of cargo molecule can be achieved under a variety of exposure
conditions (Das et al 2007, Goodwin et al 2005, Paasonen et al 2007, Radt et al 2004, Sershen
et al 2000, Skirtach et al 2005, Wu et al 2008). The majority of those experiments involved
either CW laser pulses of low intensity or very short pulses in the pico and femto second range
and higher intensities. It is very likely that in the first case a quasi-steady temperature profile
is established in the medium with a temperature high enough to trigger release from the drug
carrier (Sershen et al 2000, 2002) and in the second case bubble collapse (inertial cavitation) is
responsible for the liposome opening (Wu et al 2008). More specifically, Wu et al (2008) have
employed femtosecond near-infrared laser pulses to irradiate hollow gold nanoshells which
were either encapsulated within liposomes or tethered to the liposome membrane with a linker
to remotely trigger liposome release. Femtosecond laser pulses can produce very rapid heat
transfer to the surrounding medium (Hartland 2006) which in turn can produce violent bubble
collapse. In their study, Wu et al (2008) concluded that triggered release of the liposome
content within a few seconds was not simply caused by heating but it can be attributed to
inertial cavitation. Inertial cavitation may, however, produce unwanted bioeffects in an in vivo
application.
Our model suggests a region of the parameter space where short laser pulses in the
nanosecond range and intensities below the threshold for significant cavitation may be used
for externally triggering the release of drugs from a thermal sensitive liposome, polymer
capsule and gel composites.
As a final remark, it is noteworthy to observe here that the temperature of the GNP
and surrounding medium at the end of the laser pulse is much higher if one does not model
the dielectric and thermal changes occurring in the medium as water turns into vapor. For
example, the temperature reached by the GNP at the end of the laser pulse with tL = 6 ns and I =
8 × 109 W m−2 is above 500 ◦ C if one does not include the dielectric changes from water to
vapor given by equation (13). If the thermal changes given by equations (10) and (11) are not
included for the same exposure conditions, the GNP temperature is also well above 100 ◦ C,
but is below 500 ◦ C because the thermal conductivity of water is much higher than that of
vapor. Therefore, the knowledge of the thermal and dielectric properties of the medium
surrounding the GNP is crucial to establish the correct elevation of temperature produced by
the irradiation of GNPs in a given medium.

5. Conclusions

The results of our study could be helpful in designing experiments for applications such as
molecularly targeted photothermal therapy and drug delivery where the creation of a quasi-
steady, highly focal temperature could be beneficial. We have seen that in an aqueous solution,
the temperature reached within 5 nm from the GNP surface is above 80 ◦ C, within 10 nm is
above 70 ◦ C and within 0.2 μm is just a few degrees above the ambient temperature. It is
expected that similar temperatures can be reached inside a cell as well at appropriate exposure
conditions. Soft tissue thermal conductivity is of the order of 0.6 W m−1 K−1 and therefore
comparable to the one used in our simulations for water.
In addition, the intensity that gives the maximum quasi-steady temperature profile in the
medium is about 2.5 × 109 W m−2 and the time to reach such a configuration is of the order of
a few hundreds of nanoseconds. This possibility can only be achieved within a narrow range
of laser parameters. For example, we have seen that as the laser intensity is further increased
it is no longer possible to establish a quasi steady-temperature distribution in the medium, but
a phase-change with bubble formation takes place around the GNP. For a given pulse duration,
there is an intensity threshold at which phase transition takes place. As phase-transition starts
5556 E Sassaroli et al

an additional increase in the intensity does not significantly increase the temperature in the
GNP environment, this is because part of the absorbed energy is used for phase transition. For
example, for a pulse duration of 6 ns, phase transition takes place at about 4.2 × 109 W m−2 for
the case modeled in this investigation. At such exposure conditions, the temperature reached
by the end of the laser pulse at a distance of 10 nm from the particle surface is about 69 ◦ C.
As the intensity is further increased, the temperature distribution does not change significantly
until the intensity is above 14 × 109 W m−2. At that point the temperature of the GNP and
of a very narrow region of a few nanometers surrounding the GNP starts to rise again. As
vapor forms, a bubble is nucleated around the GNP. Although our simulations cannot model
bubble dynamics and collapse, it is reasonable to assume that during the laser pulse duration
the bubble expands and when the pulse is off collapses or undergoes a few oscillations before
collapsing. Further, it is expected that as the intensity is increased, cavitation effects become
increasingly stronger.
In summary, we have numerically investigated the heating by nanosecond laser pulses of a
diluted suspension of spherical GNPs including thermal, phase and dielectric changes without
considering pressure generation, bubble dynamics and the effect of GNP concentration. The
aim is to provide useful insight for applications of GNPs to cancer drug delivery and cancer
thermal therapy.

Acknowledgments

This work was supported by the Methodist Hospital Research Institute. We thank Dr. Palpant
for providing the data for the index of refraction of gold as a function of temperature.

Appendix A

A.1. Thermal simulations

The thermal equations (1)–(13) have temperature-dependent coefficients and therefore they
can be solved only numerically.
We have chosen to use the finite element method, which is a general-purpose technique
for computing accurate solutions to PDEs. In order to avoid the singularity at r = 0, equations
(1) and (3) are multiplied by r2 and then transformed into the dimensionless form
 2 
∂Tp ∂ r̂ ∂
r̂ 2 ρp cp + − 2 kp Tp = r̂ 2 Slaser (A.1)
∂t ∂ r̂ a ∂ r̂
 2 
∂Tf ∂ r̂ ∂
r̂ 2 ρf cpf + − 2 k f Tf = 0 (A.2)
∂t ∂ r̂ a ∂ r̂

with r̂ = r/a and ∂/∂r = (1/a)(∂/∂ r̂). Equations (A.1) and (A.2) can be further simplified
by introducing the variables
mp = r̂ 2 a 2 ρp cp , mw = r̂ 2 a 2 ρf cpf , κp = r̂ 2 kp , κw = r̂ 2 kf, f = r̂ 2 a 2 Slaser (A.3)
and
 
∂Tp ∂ ∂
mp + −κp Tp = f (A.4)
∂t ∂ r̂ ∂ r̂
 
∂Tw ∂ ∂
mw + −κw Tw = 0 (A.5)
∂t ∂ r̂ ∂ r̂
Pulsed laser heating of gold nanoparticles 5557

In equation (A.5), Tw = T f − T∞ and therefore the boundary conditions for large r̂ is

naturally taken into account. Equation (A.4) has to be solved in the domain [0, 1] subject to
the inhomogeneous Neumann boundary conditions
 
∂Tp  ∂Tp 
κp (0) = 0, κ (1) r̂=1 = h (A.6)
∂ r̂  ∂ r̂ 
r̂=0 p

where the function h is obtained by imposing the continuity of the heat flux at r̂ = 1 as given
by the second of equations (5).
In the finite element method, the PDEs are transformed into the variational (weak) form,
which is obtained by multiplying equations (A.4), (A.5) by an appropriate test function and
integrating by parts in the highest-order space derivative. The variational form of equation
(A.4) can be introduced by defining the two functional spaces
  1 
L (0, 1) = v : (0, 1) → R :
2
v (x)dx ≺ ∞
2
0
 
∂v
H (0, 1) = v : [0, 1] → R : v,
1
∈ L (0, 1)
2
∂r
where L2(0,1) is the set of real functions v in (0,1) such that v 2 is Lebesgue integrable.
H1(0,1) is an example of Sobolev space which is a subspace of L2(0,1) with v and ∂v/∂r
L2(0,1)-integrable. The variational form of equation (A.4) with boundary conditions given by
equation (A.6) amounts to find a function Tp ∈ H 1 (0, 1) such that
 1  1
∂Tp ∂Tp ∂v
mp v + κp dr̂ = f v dr̂ + hv(1) t > t0 ∀v ∈ H 1 (0, 1). (A.7)
0 ∂t ∂ r̂ ∂ r̂ 0
The finite element formulation of equation (A.7) is then obtained by applying the Galerkin
method for producing an approximate solution to the weak form from a given finite-
dimensional subspace.
The choice of the subspace in not unique, but it is common to choose subspaces of
continuous, piecewise polynomials which can be linear or quadratic/cubic depending on
the accuracy required in the problem. The subspace of piecewise polynomials is defined
relative to a mesh which divides the computational domain in subdomains or elements. In
our simulations, we have considered Lagrange quadratic elements and when more accuracy
was required Hermite cubic elements. Given the mesh r̂0 = 0, r̂1 , r̂2 , . . . , r̂n = 1 and a given
piecewise subspace Sn of H 1 (0, 1), a basis φ0 , φ1 , . . . , φn for such subspace must satisfy the
condition

1 i=j
φi (r̂j ) = . (A.8)
0 i = j
The finite element solution of equation (A.7) is obtained by finding a solution Tpn ∈ Sn with
 n
Tpn = Uj (t)φj (r̂) (A.9)
j =0
such that

n
dUj 
n
Mij + Uj Kij = Fi + h, i = 0, 1, . . . , n (A.10)
j =0
dt j =0
where Mij , Kij , Fi are respectively the mass matrix, the stiffness matrix and the load vector
defined by
 1  1  1
Mij = mp φi φj dr̂, κp φi φj dr̂, Fi = f φi dr̂. (A.11)
0 0 0
5558 E Sassaroli et al

Equation (A.10) is reduced to a system of ODEs and can be solved using standard techniques
such as, for example, the backward Euler method. Equation (A.10) is coupled to equation
(A.5) at the boundary r̂ = 1, which needs to be solved in the domain [1, R̂] with R sufficiently
large subject to the mixed boundary conditions:
∂Tw 
κw (1) r̂=1 = h, Tw (R̂) = 0. (A.12)
∂ r̂
The variational form of equation (A.5) subject to the boundary conditions (A.12) is equivalent
to find a Tw ∈ H01 (1, R̂) with H01 (1, R̂) defined by H01 = {v ∈ H 1 (1, R̂) : v = 0 at r̂ = R̂}
such that
 1
∂Tw ∂v
κw dr̂ = −hv(1) t > t0 ∀v ∈ H01 (1, R̂). (A.13)
0 ∂ r̂ ∂ r̂
The finite formulation of equation (A.13) can be obtained by applying the Galerkin method to
an the appropriate subspace S̃n of H01 (1, R̂). Given the mesh r̂0 = 1, r̂1 , r̂2 , . . . , r̂n = R̂, we
denote with ϕ0 , ϕ1 , . . . , ϕn−1 a Lagrange quadratic/Hermit cubic basis for S̃n . The solution
of equation (A.13) is obtained by finding a solution Twn ∈ S̃n with

n−1
Twn = Vj (t)ϕj (r̂) (A.14)
j =0

such that

n−1
Vj K̃ij = −h, i = 0, 1, . . . , n − 1 (A.15)
j =0

with K̃ij stiffness matrix defined by

 R̂
κw ϕi ϕj dr̂. (A.16)
1
We notice here that by summing equations (A.10) and (A.15) the coefficient h cancels out and
therefore the continuity condition of the heat flux at r̂ = 1 is automatically implemented.

References

Angelatos A S, Radt B and Caruso F 2005 Light-responsive polyelectrolyte/gold nanoparticle microcapsules J. Phys.
Chem. B 109 3071–76
Avedisian C T 1985 The homogeneous nucleation limits of liquids J. Phys. Chem. Ref. Data 14 695–729
Averitt R D, Westcott S L and Halas N J 1999 Linear optical properties of gold nanoshell J. Opt. Soc. Am. B 16 1824–32
Berciaud S, Cognet L, Blab G A and Lounis B 2004 Photothermal heterodyne imaging of individual nonfluorescent
nanoclusters and nanocrystals Phys. Rev. Lett. 93 257402
Bohren C F and Huffman D R 1983 Absorption and Scattering of Light by Small Particles (New York: Wiley)
Boyer D, Tamarat P, Maali A, Lounis B and Orrit M 2002 Photothermal imaging of nanometer-sized metal particles
among scatterers Science 297 1160–3
Carslaw H S and Jaeger J C 1986 Conduction of Heat in Solids (Oxford: Clarendon)
Chithrani B D, Ghazani A A and Chan C W 2006 Determining the size and shape dependence of gold nanoparticle
uptake into mammalian cells Nano Lett. 5 331–8
Choi M R et al 2007 A cellular Trojan horse for delivery of therapeutic nanoparticles into tumors Nano Lett. 7 3759–65
Comini C, Del Giudice S, Lewis R W and Zienkiewicz O C 1974 Finite element solution of non-linear heat conduction
problems with special reference to phase change Int. J. Numer. Meth. Eng. 8 613–24
Das M, Sanson N, Fava D and Kumacheva E 2007 Microgels loaded with gold nanorods: Photothermally triggered
volume transitions under physiological conditions Langmuir 23 196–201
Debenedetti P G 1996 Metastable Liquids: Concepts and Principles (Princeton, NJ: Princeton University Press)
Pulsed laser heating of gold nanoparticles 5559

Dickerson E B, Dreaden E C, Huang X, El-Sayed I H, Chu H, Pushpanketh S, McDonald J F and El-Sayed M A

2008 Gold nanorod assisted near-infrared plasmonic photothermal therapy (PPTT) of squamous cell carcinoma
in mice Cancer Lett. 269 57–66
Dou Y, Zhigilei L V, Winograd N and Garrison B J 2001 Explosive boiling of water films adjacent to heated surfaces:
a microscopic description J. Phys. Chem. A 105 2748–55
Eghtedari M, Oraevsky A, Copland J A, Kotov N A, Conjusteau A and Motamedi A 2007 High sensitivity of in vivo
detection of gold nanorods using a laser optoacoustic imaging system Nano Lett. 7 1914–8
Ekici O, Harrison R K, Durr N J, Eversole D S, Lee M and Ben-Yakar A 2008 Thermal analysis of gold nanorods
heated with femtosecond laser pulses J. Phys. D: Appl. Phys. 41 185501–11
Elliott A, Schwartz J, Wang J, Shetty A, Hazle J and Stafford J R 2008 Analytical solution to heat equation
with magnetic resonance experimental verification for nanoshell enhanced thermal therapy Lasers Surg.
Med. 40 660–5
Elliott A M, Stafford R J, Schwartz J, Wang J, Shetty A M, Bourgoyne C, O’Neal P and Hazle J D 2007 Laser-induced
thermal response and characterization of nanoparticles for cancer treatment using magnetic resonance thermal
imaging Med. Phys. 34 3102–8
El-Sayed I H, Huang X and El-Sayed M A 2006 Selective laser photo-thermal therapy of epithelial carcinoma using
anti-EGFR antibody conjugated gold nanoparticles Cancer Lett. 239 129–35
Fachinotti V D, Cardona A and Huspe A E 1999 A fast convergent and accurate temperature model for phase-change
heat conduction Int. J. Numer. Meth. Eng. 44 1863–84
Faraggi E and Gerstman B S 2007 Acoustical resonant absorption of pulsed laser radiation by a spherical absorber J.
Appl. Phys. 102 123505
Ghosh S K and Pal T 2007 Interparticle coupling effect on the surface plasmon resonance of gold ganoparticles: from
theory to applications Chem. Rev. 107 4797–862
Gobin A M, Lee M H, Halas N J, James W D, Drezek R A and West J L 2007 Near-infrared resonant nanoshells for
combined optical imaging and photothermal cancer therapy Nano Lett. 7 1929–34
Goldenberg H and Tranter C J 1952 Heat flow in an infinite medium heated by a sphere Br. J. Appl. Phys. 3 296–8
Goodwin A P, Mynar J L, Ma Y, Fleming G R and Frechet J M J 2005 Synthetic micelle sensitive to IR light via a
two-photon process J. Am. Chem. Soc. 127 9952–3
Hamad-Schifferli K, Schwartz J J, Santos A T, Zhang S G and Jacobson J M 2002 Remote electronic control of DNA
hybridization through inductive coupling to an attached metal nanocrystal antenna Nature 415 152–5
Hartland G V 2006 Coherent excitation of vibrational modes in metallic nanoparticles Annu. Rev. Phys.
Chem. 57 403–30
Heath J M and Davis M E 2008 Nanotechnology and cancer Annu. Rev. Med. 59 251–65
Hirsch L R, Stafford R J, Bankson J A, Sershen S R, Rivera B, Price R E, Hazle J D and West J L 2003 Nanoshell-
mediated near-infrared thermal therapy of tumors under magnetic resonance guidance Proc. Natl Acad. Sci.
USA 100 13549–54
Holman J P 1997 Heat Transfer (New York: McGraw-Hill)
Hu M, Chen J, Li Z Y, Au L, Hartland G V, Li X, Marquez M and Xia Y 2006 Gold nanostructures: engineering
their plasmonic properties for biomedical applications Chem. Soc. Rev. 35 1084–94
Huang X, El-Sayed I, Qian W and El-Sayed M 2006 Cancer cell imaging and photothermal therapy in the near-infrared
region by using gold nanorods J. Am. Chem. Soc. 128 2115–20
Huff T B, Tong L, Zhao Y, Hansen M N, Cheng J X and Wei A 2007 Hyperthermic effects of gold nanorods on tumor
cells Nanomedicine 2 125–32
Jain P K, Huang X, El-Sayed I H and El-Sayed M A 2007 Review of some interesting surface plasmon resonance-
enhanced properties of noble metal nanoparticles and their applications to biosystems Plasmonics 2 107–18
Johnson P B and Christy R W 1972 Optical constants of the noble metal Phys. Rev. B 6 4370–9
Kim K, Huang S W, Ashkenazi S, O’Donnell M, Agarwal A, Kotov N A, Denny M F and Kaplan M J 2007
Photoacoustic imaging of early inflammatory response using gold nanorods Appl. Phys. Lett. 90 223901
Kotaidis V, Dahmen C, von Plessen G, Springer F and Plech A 2006 Excitation of nanoscale vapor bubbles at the
surface of gold nanoparticles in water J. Chem. Phys. 206 184702
Kreibig U and Vollmer M 1995 Optical Properties of Metal Clusters (Berlin: Springer)
Mallidi S, Larson T, Aaron J, Sokolov K and Emelianov S 2007 Molecular specific optoacoustic imaging with
plasmonic nanoparticles Opt. Express 15 6583–8
Mirkin C A, Letsinger R L, Mucic R C and Storhoff J J 1996 A DNA-based method for rationally assembling
nanoparticles into macroscopic materials Nature 382 607–9
Moller R and Fritzsche W 2007 Metal nanoparticle-based detection for DNA analysis Curr. Pharm.
Biotechnol. 8 274–85
Neumann J and Brinkmann R 2005 Boiling nucleation on melanosomes and microbeads transiently heated by
nanosecond and microsecond laser pulses J. Biomed. Opt. 10 024001
 5560 E Sassaroli et al

Neumann J and Brinkmann R 2007 Nucleation dynamics around single microabsorbers in water heated by nanosecond
laser irradiation J. Appl. Phys. 101 114701
Paasonen L, Laaksonen T, Johans C, Yliperttula M, Kontturi K and Urtti A 2007 Gold nanoparticles enable selective
light-induced contents release from liposomes J. Control. Release 122 86–93
Palpant B, Rashidi-Huyeh M, Gallas B, Chenot S and Fisson S 2007 Highly dispersive thermo-optical properties of
gold nanoparticles Appl. Phys. Lett. 90 223105
Pissuwan D, Valenzuela S, Miller C and Cortie M 2007 A golden bullet? Selective targeting of toxoplasma gondii
tachyzoites using anti body-functionalized gold nanorods Nano Lett. 7 3808–12
Pitsillides C M, Joe E K, Anderson R R and Lin C P 2003 Selective cell targeting with light-absorbing microparticles
and nanoparticles Biophys. J. 84 4023–32
Pustovalov V K 2005 Theoretical study of heating of spherical nanoparticle in media by short laser pulses Chem.
Phys. 308 103–8
Radt B, Smith T and Caruso F 2004 Optically addressable nanostructured capsules Adv. Mater. 16 2184–9
Raschke G, Kowarik S, Franzl T, Sonnichsen C, Klar T A, Feldmann J, Nichtl A and Kurzinger K 2003 Biomolecular
recognition based on single gold nanoparticle light scattering Nano Lett. 3 935–8
Rashidi-Huyeh M and Palpant B 2006 Counterintuitive thermo-optical response of metal-dielectric nanocomposite
materials as a result of local electromagnetic field enhancement Phys. Rev. B 74 075405
Richardson H H, Carlson M T, Tandler P J, Hernandez P and Govorov A O 2009 Experimental and theoretical studies
of light-to-heat conversion and collective heating effects in metal nanoparticle solutions Nano Lett. 9 1139–46
Sershen S R, Westcott S L, Halas N J and West J L 2000 Temperature-sensitive polymer-nanoshell composites for
photothermally modulated drug delivery J. Biomed. Mater. Res. 51 293–8
Sershen S R, Westcott S L, Halas N J and West J L 2002 Independent optically addressable nanoparticle-polymer
optomechanical composites Appl. Phys. Lett. 80 4609–11
Skirtach A G, Dejugnat C, Braun D, Susha A S, Rogach A L, Parak W J, Moehwald H and Sukhorukov G B 2005
The role of metal nanoparticles in remote release of encapsulated materials Nano Lett. 5 1371–7
Sokolov K, Follen M, Aaron J, Pavlova I, Malpica A, Lotan R and Richards-Kortum R 2003 Real-time vital optical
imaging of precancer using anti-epidermal growth factor receptor antibodies conjugated to gold nanoparticles
Cancer Res. 63 1999–2004
Sonnichsen C, Franzl T, Wilk T, von Plessen G and Feldmann J 2002 Plasmon resonances in large noble-metal
clusters New J. Phys. 4 93.1–8
Souza G R, Christianson D R, Staquicini F I, Ozawa M G, Snyder E Y, Sidman R L, Miller J H, Arap W and Pasqualini R
2006 Networks of gold nanoparticles and bacteriophage as biological sensors and cell-targeting agents Proc.
Natl Acad. Sci. USA 103 1215–20
Sperling R A, Gil P R, Zhang F, Zanella M and Parak W J 2008 Biological applications of gold nanoparticles Chem.
Soc. Rev. 37 1896–906
Sun J M, Gerstman B S and Li B 2000 Bubble dynamics and shock waves generated by laser absorption of a
photoacoustic sphere J. Appl. Phys. 88 2352–62
Tkachenko A G, Xie H, Liu Y L, Coleman D, Ryan J, Glomm W R, Shipton M K, Franzen S and Feldheim D L 2004
Cellular trajectories of peptide-modified gold particle complexes: comparison of nuclear localization signals
and peptide transduction domains Bioconjug. Chem. 15 482–90
Volkov A N, Sevilla C and Zhigilei L V 2007 Numerical modeling of short pulse laser interaction with Au nanoparticle
surrounded by water Appl. Surf. Sci. 253 6394–9
Voller V R and Swaminathan C R 1990 Fixed grid techniques for phase change problems: a review Int. J. Numer.
Meth. Eng. 30 875–98
Wu G, Mikhailovsky A, Khant H A, Fu C, Chiu W and Zasadzinski J A 2008 Remotely triggered liposome release
by near-infrared light absorption via hollow gold nanoshells J. Am. Chem. Soc. 130 8175–7

View publication stats