Materials Science and Engineering C 42 (2014) 185–191
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Simulation and experimental results of optical and thermal modeling of gold nanoshells Lida Ghazanfari, Mohammad E. Khosroshahi ⁎ Laser and Nanobiophotonics Laboratory, Biomaterial Group, Faculty of Biomedical Engineering, Amirkabir University of Technology, Tehran, Iran
a r t i c l e
i n f o
Article history: Received 22 January 2014 Received in revised form 31 March 2014 Accepted 4 May 2014 Available online 22 May 2014 Keywords: Superparamagnetic iron oxide nanoparticles (SPION) Magneto-optical nanoshells (MNSs) Optical and thermal modeling
a b s t r a c t This paper proposes a generalized method for optical and thermal modeling of synthesized magneto-optical nanoshells (MNSs) for biomedical applications. Superparamagnetic magnetite nanoparticles with diameter of 9.5 ± 1.4 nm are fabricated using co-precipitation method and subsequently covered by a thin layer of gold to obtain 15.8 ± 3.5 nm MNSs. In this paper, simulations and detailed analysis are carried out for different nanoshell geometry to achieve a maximum heat power. Structural, magnetic and optical properties of MNSs are assessed using vibrating sample magnetometer (VSM), X-ray diffraction (XRD), UV–VIS spectrophotometer, dynamic light scattering (DLS), and transmission electron microscope (TEM). Magnetic saturation of synthesized magnetite nanoparticles are reduced from 46.94 to 11.98 emu/g after coating with gold. The performance of the proposed optical–thermal modeling technique is verified by simulation and experimental results. © 2014 Published by Elsevier B.V.
1. Introduction The increasing availability of nanostructures with highly controlled magnetic and optical properties has created widespread interest in the use of nanoshells for diagnostic and therapeutic applications. This is due to the unique properties exhibited by the nanoscale particles, which is not seen at bulk scale. The absorption property of nanoshells can be provided at specific wavelength and their movement can be controlled by an external magnetic field. Therefore, magneto-optical nanoshells (MNSs) can be guided to a specific tissue target [1]. Superparamagnetic magnetite has attracted increasing interest because of its outstanding properties especially in biotechnology and biomedicine applications [2–4]. In recent years, much effort has been devoted to the synthesis and characterization of MNSs such as gold coated iron-oxide nanoparticles (NPs) [1,5–7]. Generally, a nanoshell is a type of spherical nanoparticle which consists of dielectric core covered by a thin metallic shell, e.g. gold or silver. These nanoshells involve a plasmon which is a collective excitation or quantum plasma oscillation when excited by an electromagnetic field. This resonant interaction between the light field and the oscillating surface charges gives rise to a state known as surface plasmon. By different shapes and sizes of nanoshell, i.e. in core-shell ratio one can expect to achieve a tunable nanosystem ranging from visible (VIS) to the near infrared (NIR), based on the Mie theory [8]. When nanoshells are exposed to the appropriate wavelengths of a laser beam, they absorb energy and then heat up. These nanoshells
⁎ Corresponding author. Tel.: +98 2164542398; fax: +98 2166468186. E-mail address: [email protected] (M.E. Khosroshahi).
http://dx.doi.org/10.1016/j.msec.2014.05.002 0928-4931/© 2014 Published by Elsevier B.V.
mediate strong plasmon-induced surface heat flux upon absorption of light [2]. Therefore, the main application of nanoshells is for cancer treatment, known as nanoshell-based hyperthermia [9]. During hyperthermia, temperatures above 42 °C induce cell death in tumor tissues in the range of 41 °C to 47 °C [10–12]. So, nanoshell-based hyperthermia is noninvasive, safe, and efficient. It is readily shown that gold nanoshells can be an attractive alternative for both diagnostic and therapeutic applications such as medical imaging and skin wound soldering [13–15]. Recent studies in modeling attempt to analyze the optical and/or thermal effects of nanostructures [16–18]. Nevertheless, comprehensive evaluation of both optical and thermal characteristics of nanoshells (multiple-layer NPs) have been scarce. This paper presents a generalized and comprehensive technique which includes the optimal and thermal modeling to consider effects of nanoshell geometry and absorption cross section, irradiation source electric field, and concentration of nanoshells. The proposed optical–thermal modeling technique of synthesized MNSs is implemented in the MATLAB environment and verified by experimental results. 2. Materials and methods All of analytic reagents are purchased from the indicated suppliers and used without further purification: ferric chloride hexahydrate (FeCl3·6H2O) (99%), ferrous chloride tetrahydrate (FeCl2·4H2O) (99%), sodium hydroxide (NaOH, 99%), hydrochloric acid (HCl, 37%), and absolute ethanol are purchased from Merck. 3-Aminopropyltriethoxysilane (APTS), tetrakis (hydroxymethyl) phosphonium chloride (THPC), chloroauric acid (H AuCl4), potassium carbonate (K2CO3), and formaldehyde (37%) are obtained from Sigma-Aldrich (St. Louis, MO). Milli-Q
186
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
ethanol into 40 ml THPC gold solution and 4 ml 1 M NaCl and left at 4 °C. After 12 h the nanoparticles are washed and redispersed in 10 ml water. A plating solution is prepared by mixing 3 ml H AuCl4 (1%) with 200 ml aqueous solution of K2CO3 (1.8 mM). A gold shell is grown around the SPIONs by adding 1 ml of precursor suspension to 9 ml of plating solution. 50 μl of H2CO is quickly added into a 10 ml prepared suspension of precursor nanoparticles in plating solution in a glass vial and gently vortexed by hand for 10 s for reduction of Au3 +, and left for 15 min. UV–Vis absorption spectrum of the solution is recorded 20 min after reaction to verify the formation of gold nanoshell. The excess formaldehyde is removed by washing twice. The fabricated magnetite/gold nanoshells are collected by centrifuge at 6000 g. 2.4. Characterization methods
water (specific conductance 0.1 μS/cm) is deoxygenated by bubbling N2 gas for 1 h prior to the use. An Argon laser (514 nm) with power of 300 mW is used to illuminate the sample area.
X-ray diffraction (XRD) measurements are performed at ambient temperature using a PHILIPS PW1800 X-ray diffractometer with CuKα radiation. Transmission electron microscopy (TEM) is performed using a CM 200 FEG STEM Philips-M.E.R.C. operating at a voltage of 200 kV. Magnetization measurements are carried at 300 K in a magnetic field (H) of up to 20 kOe with a vibrating sample magnetometer (Meghnatis Daghigh Kavir Co. VSM/AGFM) that can measure magnetic moments as low as 10− 3 emu. For the magnetization measurements, uncoated Fe 3 O4 NPs are in dry powder form obtained by evaporating the water from the solution. The samples are dried by freeze dryer (Pishtaz Engineering Co. Model: FD-4 or FD-10). Calculation of the absorption efficiency is performed using MieLab software. Each number presented in this paper is the average of at least three measurements and reported using SPSS 15.0.
2.1. Synthesis of superparamagnetic iron oxide nanoparticles (SPIONs)
3. Optical and thermal modeling
SPIONs are fabricated following a previously reported procedure with some modification [19]. Briefly, FeCl2 ·4H2 O and FeCl3 ·6H2 O are added to 25 ml aqueous solution of 0.4 M HCl under vigorous stirring. Then the iron solution is added dropwise to 250 ml 1.5 M NaOH under magnetic stirring (1500 rpm) for 30 min at room temperature. The precipitated nanoparticles are separated by a magnet. The isolated precipitate is washed with water for 5 times and followed by washing twice using ethanol. The final precipitate is dried under vacuum at room temperature. All the synthesis steps are carried out under passing N2 gas through the solution medium.
In order to investigate optical and thermal properties of nanoshells, the dynamic model of nanoshells in the embedding medium must be first derived. This model should consider both accuracy and simulation time to manifest the precise dynamic properties.
Fig. 1. Nanoshell geometry including the core, shell, and embedding regions.
2.2. Functionalization of SPIONs Amino groups are functionalized on the nanoparticle surface by silanization reaction. Magnetite nanoparticles (0.074 g) are dispersed in 25 ml ethanol by probe sonicator for 30 min. This suspension is diluted to 150 ml by ethanol and 1 ml H 2 O. 35 μl APTES is added to the magnetic suspension under vigorous stirring [20]. One milliliter H2O is introduced into the reaction medium to initiate the hydrolysis. The synthesis operation proceeds for 7 h at room temperature. After that, modified magnetite nanoparticles are isolated by magnet and washed for 5 times and dried under vacuum at room temperature. 2.3. Synthesis of SPION/gold nanoshells SPION/gold nanoshells are fabricated by a multistep procedure through electroless plating of Au onto precursor nanoparticles (i.e. gold-seeded SPIONs) as reported previously by some modifications [15]. Briefly, a THPC gold solution composed of 2–3 nm gold colloids is produced according to Duff and Baker method [21] and aged for 2 weeks. Nanoshell precursor particles are synthesized by adding 1 ml of 0.0128 M amine-terminated magnetite nanoparticles in
3.1. Nanoshell configuration There are two distinct size regimes to be considered in the determination of the optical and thermal properties of metallic NPs and nanoshells [22]. In the extrinsic size regime, the optical properties can be fully described by specification of the nanoparticle radius and by the use of the bulk frequency dependent dielectric constant (ε(ω)). Conversely, in the intrinsic regime (2r b 50 nm), dipole plasmon absorption is by far the dominant contributor to the extinction cross-section. Thus, size dependent effects must be accounted for by employment of a size dependent dielectric function as [23,24]: 2
εða; ωÞ ¼ ε exp ðωÞ þ
ωp
ω þ iωγbulk 2
2
−
ωp
ω þ iωγ 2
ð1Þ
where εexp (ω) is the experimental dielectric function and a is the shell thickness. ωp and ω are the bulk plasma frequency and electromagnetic wave frequency, respectively. γ bulk is the bulk collisional frequency which is expressed by: γbulk ¼
VF l0
ð2Þ
where V F is the Fermi speed and l 0 is the gold electron mean free path at room temperature [25]. Dielectric function of the gold nanoshell becomes size-dependent when the nanoshell size is smaller than the gold electron mean-free
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
187
where ηwater and ηair are the intrinsic impedances of the water and air, respectively. η in the medium is defined as: η¼
rffiffiffi μ ε
ð6Þ
where μ is the medium permeability and ε is the medium dielectric function. The mentioned solution should be described for the electric field of uniform plane waves which travel in lossy media. The electromagnetic waves that travel in lossy media experience attenuation [28,30,31]. Therefore, the attenuation is considered as an ambient absorption coefficient which leads to Ewater ¼
2ηwater −αd E e ηair þ ηwater Laser
ð7Þ
where d is the distance from the surface of embedding medium and α is the real part of the propagation constant called the attenuation [28]: Fig. 2. Variation of total absorbed power versus inner and outer radii of nanoshell.
path (~42 nm) [26]. The width of the absorption peak can be described as a modification of the bulk collisional frequency which is given by [27].
ð3Þ
where r1 and r2 are the inner and outer radii of shell, respectively. reff is the effective mean free path of collisions. The nanoshell geometry, which is simulated as a suspension in water, is shown in Fig. 1. The core is magnetite and indicated by a radius r1 with a dielectric function ε1. The shell is gold and coated on the core from r1 to r2 with a dielectric function ε2. The embedding medium is water with a dielectric function ε3 and this medium is exposed to air with a dielectric function ε0.
Although the laser incident field is time dependent, it does not vary spatially over the nanoshell diameter where it is much smaller than the incident field wavelength. Therefore, the electric field of the laser is assumed to have a constant component where the wave is traveling in the z direction. Its expression in the air takes the form ELaser ¼
qffiffiffiffiffiffiffiffiffiffiffiffi ^z 2Iμ 0 c a
ð4Þ
where I is the incident laser power (W/cm2), μ0 is the air permeability, and c is the speed of light in a vacuum. A 300 mW continuous wave laser with a 1.5 mm diameter spot size is applied to simulate irradiation of the nanoshells. An Argon laser is chosen as irradiation source because of two main reasons. First, it is a continuous wave laser. Second, its wavelength is close to surface plasmon resonance wavelength of the synthesized nanoshells. At the surface between air and water, a part of the incident field will be reflected from the interface, and a part will be transmitted into water medium. The boundary conditions must be satisfied thus the incident field in the embedding medium can be determined. The boundary conditions make it necessary that the tangential components of the total vectors electric and magnetic fields on two sides of the interface be the same. By solving that, the incident electric field in the water medium is written [28,29]: Ewater ¼
2ηwater E ηair þ ηwater Laser
According to the permeability of water, one can write α as ωλ pffiffiffiffiffi εr α ¼ Re j 2π
ð9Þ
pffiffiffiffiffi where λ is the electromagnetic wavelength. ε r is the numerical value for the refractive index of water as a function of wavelength in the visible part of the spectrum. The refractive index over a broad wavelength range of 0.2 to 200 μm is written by [32]: 2
9
4
14
6
n ¼ 1:3199 þ 6878=λ −1:132 10 =λ þ 1:11 10 =λ :
ð10Þ
The imaginary part of the water refractive index is interpolated from [33]. Therefore, water refractive index at 514 nm is 1.34 + i4.55 × 10−10. The wave impedance is defined as: Zw ¼
3.2. Optical properties of nanoshells
ð8Þ
120π n
ð11Þ
where n is the real part of the water refractive index. The average power density associated with the transmitted field is given by [28]: 2
S¼
2jEwater j : Zw
ð12Þ
The absorption and the scattering cross sections from the polarizability are achieved by using scattering theory. In intrinsic size regime, the nanoshell diameter is much smaller than the incident wavelength. Therefore, it is subjected to a constant field, i.e. independent of position. 1.1 1
0.75
θ
r eff
V γ ¼ γ bulk þ F reff 2 2 1=3 ¼ ðr 2 −r1 Þ r 2 −r 1
pffiffiffiffiffi α ¼ Refjω μεg:
0.5
0.25
0
100
200
300
time (s) ð5Þ Fig. 3. The dimensionless temperature versus time.
400
188
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
that the variation of total absorbed power geometrically depends on shell diameter rather than core diameter. The total power equals the heat convection away from the system surface plus the heat radiation from the system surface [16].
Lin (counts)
(311)
(440)
(220)
Q ¼ Q con þ Q rad
(511)
(400)
ð18Þ
where
(422)
Q con ¼ hAðT−T amb Þ 4 4 Q rad ¼ εσA T −T amb
2-Theta (degrees)
A¼
ð19Þ
2 4πr 2
Fig. 4. XRD pattern for synthesized magnetite NPs.
Quasi-static approach (Rayleigh limit) should be employed in the calculation [34,18]. It is worth mentioning that quasi-static approach is helpful in nanoshell nonlinear optical effects. Therefore, the absorption cross section is given as follows [35]: σ abs ¼
pffiffiffiffiffi 8π2 ε 3 ε2 εa −ε 3 εb Im λ ε2 εa þ 2ε3 εb
ð13Þ
where
∑mc
ε a ¼ ε1 ð3−2P Þ þ 2ε 2 P ε b ¼ ε1 P þ ε2 ð3−P Þ 3 P ¼ 1−ðr 1 =r 2 Þ :
ð14Þ
As light being absorbed, heat is generated by nanoshells. All the energy absorbed from the incident field is converted to thermal energy and simultaneously released into the medium. Therefore, a single particle absorbed power under steady-state conditions is: q ¼ S σ abs :
ð15Þ
The total absorbed power given by: ð16Þ
where N is the total number of nanoshells which in our case equals 9 × 1012. When Eqs. (12), (13), and (15) are substituted in Eq. (16), they lead to:
where QF is obtained when the external heat flux is subtracted from the dissipated laser-induced heat as well as heat dissipated from absorbed light, which is given by ð21Þ
where Tmax is the maximum system temperature. A dimensionless driving force temperature (the terminal temperature difference between hot and cold spots is being considered as the driving force temperature) is given [17]: θ¼
T−T amb : T max −T amb
dθ hA ð1−θÞ ¼ dt Σmc
ð22Þ
ð17Þ
The simulation studies are implemented in the MATLAB environment. Fig. 2 shows total absorbed power as a function of core and shell diameters at different heat coefficient parameters. It demonstrates (111)
ð23Þ
× Eq. (23) can be solved using the initial condition θ = 0 at t = 0 which results in θ ¼ 1− expð−hAt=ðΣmcÞÞ:
ε2 εa −ε 3 εb : Im ε 2 εa þ 2ε3 ε b
Lin (counts)
ð20Þ
By substituting Eqs. (21) and (22) in Eq. (20):
Q ¼Nq
2 pffiffiffiffiffi ε3
dT ¼ QF dt
Q F ¼ hAðT max −T amb Þ−hAðT−T amb Þ
3.3. Thermal dynamics of nanoshells
2πNnjEwater j Q¼ 15λ
where ε for gold nanoparticle is 0.9, σ is the Boltzmann's constant, T is the nanoshell temperature, Tamb is the ambient temperature, and h is the heat transfer coefficient. It is assumed that the nanoshells do not interact with one another either plasmonically or thermally. Furthermore, for temperature that increases less than 12 K in our case, the ratio Qrad/ Tamb factored out from Eq. (19) remains constant. This fact allows a convenient linearization of the energy balance using a heat-transfer model as below:
ð24Þ
Fig. 3 presents the temperature versus time. The concentration of nanoshells has a significant effect on the amount of heat generated at a given power density. Furthermore, the heat transfer by convection will be insignificant if nanoshells and their surrounding medium are immobile [17]. This effectively indicates that the MNSs play not only an important role in generating thermal effect, but also an enhanced absorption cross section for spherical metal nanoshells can be achieved at optimized value of concentration for a better laser-hyperthermia. The nanoshell efficiency parameter has the form [36]: " !# 3k∞ t p ΔT 0 K abs r 2
¼ 1− exp − 2 I0 4k∞ r 2 c1 ρ1 r 31 þ c2 ρ2 1−r31 =r32
ð25Þ
(200) (220)
20
30
40
50
2-Theta (degrees) Fig. 5. XRD pattern for Fe3O4/Au nanoshells.
60
70
where I0 is the constant intensity of laser radiation during pulse duration tp, k∞ is the coefficient of thermal conduction of the ambient medium, c1, ρ1 and c2, ρ2 are the heat capacity and density of the material of the core (magnetite) and shell (gold) accordingly, r1 and r2 are the radii of the core and shell. Kabs is the absorption efficiency factor. Its value is calculated using MieLab software based on classical Mie scattering
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
189
Fig. 6. TEM image of Fe3O4 NPs.
theory [37]. The required parameters are the dimensions of nanoshell as well as the complex dielectric function of gold and magnetite. The values of complex dielectric functions for gold and magnetite at different wavelengths are obtained from literature [23,38]. In our case, the absorption efficiency corresponds to about 0.656 at surface plasmon resonance wavelength (λSPR) of 531 nm. Maximum value of the parameter ΔT0/I0 indicates the efficiency of transformation of absorbed optical energy by nanoshells into the thermal energy. In this case, the nanoshell efficiency parameter equals 2 × 10−5 °C cm2/W.
4. Results and discussion Structure and crystallinity of magnetic NPs are determined by XRD. As shown in Fig. 4, the peaks are indexed with the facecentered structure corresponding to magnetite phase (JCPDS card
no. 89-3854). The average size of the crystals is measured by Scherrer's formula: D ¼ kλ=β cosθ
ð26Þ
where D is the mean size of the crystalline domains, which may be smaller or equal to the grain size; k is a dimensionless shape factor with a typical value of about 0.9 which varies with the actual shape of the crystallite; λ is the X-ray wavelength; and β is the line broadening at half the maximum intensity (FWHM), after subtracting the instrumental line broadening, in radians. θ is the Bragg angle. Crystallite size varies from 65 to 70 Å over the range of 20 to 70° 2θ; and the average crystallite size of 67.91 Å is obtained using high index plane (311). The calculated unit cell parameter as 8.4 Å is about that of magnetite [39] which confirms the purity of synthesized magnetite NPs. Fig. 5 shows XRD pattern of Fe 3O 4 /Au
Fig. 7. TEM image of Fe3O4/Au nanoshells.
190
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
1
Table 1 Fe3O4 NPs and Fe3O4/Au nanoshell size measurements. Material
dTEM ± SD (nm)
DLS volume-weighted size (PDI) nm
SPIONs SPION/gold nanoshells
9.5 ± 1.4 15.8 ± 3.5
201.4 (0.224) 57.7 (0.25)
nanoshells. The absence of any peaks for magnetite is most likely due to the heavy atom effect from gold [40] as a result of the formation of Au-coated Fe3O 4 nanoparticles. The fact that the diffraction peaks from Fe3 O 4 are not observed is consistent with observations by others, [41] thus providing strong evidence for complete coverage of the Fe3O 4 core by Au supporting our TEM data, which supports the formation of Fe3O4/Au core-shell nanoparticles. To observe the agglomeration state, particle size distribution, and morphology of both magnetite nanoparticles (Fig. 6) and gold nanoshells (Fig. 7), bright field TEM and DLS are used (Table 1). The stability of magnetic particles is increased by sheltering magnetic dipole interaction through the growing gold shells. 2–3 nm gold colloids can be seen in Fig. 7. In addition, based on the DLS measurement, these gold colloids are 2.34 ± 0.62 nm in size. Fig. 8 shows the magnetization curve for magnetite core and gold nanoshells at ambient temperature. It indicates the superparamagnetic behavior of nanostructures. It can be deduced that the magnetite NPs have magnetization saturation (Ms) of about 46.94 emu/g. The Ms value of the resultant gold nanoshells decreases to 11.98 emu/g, when the gold is added. Fast magnetic response of the synthesized NPs can be an advantage in many applications. With the same magnetite core size and 3.5 nm gold shell thickness, the Ms value for the synthesized nanoshells in this paper is larger than that reported by Xu and Hou [42]. In Fig. 9, the variation intensity of the absorption peaks with wavelength is shown. It is obvious that the absorption peak decreases nonlinearly. Magnetite NP has a strong absorption peak about 200 nm. However, its visible absorption is quite different from magnetite/gold nanoshell which confirms the effect of SPR played by gold shell. The optical absorption spectrum range of Fe3O4/Au nanoshell is relatively broad compared with pure gold colloid. In addition, the plasmon linewidth is dominated by electron surface scattering. The broadening of resonance absorption is related to the size, shape, and aggregation of the nanometer scale metallic particles, according to Mie's theory [43]. The nanoshells produced by Lim et al. [7] had an 18 nm magnetic core and a gold shell thickness of 5 nm with a λSPR at 605 nm. A SPR wavelength of 528 nm was obtained for Au/Fe oxide nanoshells synthesized by Pham et al. [1] with a size range of 15–40 nm. In our case, a SPR wavelength of 531 nm is obtained for nanoshells with a total
Absorbance (a. u.)
0.9 0.8 0.7 0.6 0.5
magnetite/gold
0.4 0.3 150
magnetite
250
350
450
550
650
750
Wavelength (nm) Fig. 9. UV–VIS spectra of Fe3O4 NPs (solid line) and Fe3O4/Au nanoshells (dashed line).
size of 15.8 ± 3.5 nm. The higher λSPR reported for some synthesized nanoshells with the same size may be due to the formation of an incomplete shell layer during the reduction of Au3+ [44]. The temperature variation versus different nanoshell concentrations is studied at 17 W/cm2 for a period of 400 s, as shown in Fig. 10. The experimental system is set up for measuring the temperature in the water dispersed with gold nanoshells. The laser is incident on the cuvette containing a gold nanoshell solution. A thermistor probe is placed in the solution to monitor the temperature change. The thermistor with an accuracy of 0.15% (Redfish Sensors Inc., Model QTGB-14 D3) is connected to a thermometer (Model No. HP-34420A) to record the temperature. The thermometer is connected at the other end to a Windows-based laptop where the results are displayed on the screen. The thermometer gives the temperature readings as a function of time as the laser-induced resonance heating proceeds. Experiments in aqueous solution medium represent that the temperature exponentially saturates within about 400 s of laser beam exposure. This is true for all nanoshell concentrations which is in good agreement with simulation result shown in Fig. 3. A maximum temperature rise of 12 °C is achieved at 300 μg/ml. The MNS embedding medium, i.e. pure water, has negligible effect on temperature variation during laser light exposure. The temperature reaches to 5 °C after 400 s at both concentrations of 25 and 50 μg/ml. Increase of the MNS concentration, up to 50 μg/ml, has a negligible effect on temperature. However, increase of the MNS concentration, above 50 μg/ml, rises the maximum temperature. Such a phenomenon can be due to the arrangement, density, and potential for aggregation of Au nanoshells. Therefore, it results in a stronger surface plasmon absorption of light and more efficient photothermal effects. All the experiments demonstrate that the temperature increases with increasing exposure time. The results show that the fabricated MNSs are excellent candidates for hyperthermia tumor therapy. 5. Conclusion
Fig. 8. Variation of magnetization with applied magnetic field for bare magnetite NPs and magnetite-gold nanoshells.
This paper presents the generalized method for optical and thermal modeling of synthesized MNSs. The proposed modeling method considers effects of nanoshell geometry and absorption cross section, irradiation source electric field, and concentration of nanoshells. Furthermore, it can be extended to any n-layer nanoshells at any laser wavelength. In this paper, gold shells are formed on the surface of 9.5 ± 1.4 nm Fe3O4 NPs to obtain 15.8 ± 3.5 nm MNSs. Structural, magnetic and optical properties of superparamagnetic MNSs are characterized. Moreover, the experimental results, based on SPR-induced thermal effect and Mie theory, exhibit higher concentration of gold nanoparticles that produced a higher temperature rise. Simulation of the proposed optical–thermal modeling technique of synthesized MNSs is carried out in the MATLAB environment and validated by experimental results.
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
191
Fig. 10. Temperature variation (ΔT) versus different nanoshell concentrations over a period of 400 s at a maximum laser power density of 17 W/cm2.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
T. Pham, C. Cao, S. Sim, J. Magn. Magn. Mater. 320 (2008) 2049–2055. S. Wang, Y. Zhou, W. Sun, Mater. Sci. Eng. C 29 (2009) 1196–1200. L. Sun, C. Huang, T. Gong, S. Zhou, Mater. Sci. Eng. C 30 (2010) 583–589. J. Viota, A. Carazo, J. Munoz-Gamez, K. Rudzka, R. Gómez-Sotomayorb, A. RuizExtremera, et al., Mater. Sci. Eng. C 33 (2013) 1183–1192. Z. Hassannejad, M. Khosroshahi, Opt. Mater. 35 (2013) 644–651. H. Liu, P. Hou, W. Zhang, J. Wu, Colloids Surf. A Physicochem. Eng. Asp. 356 (2010) 21–27. J. Lim, A. Eggeman, F. Lanni, R. Tilton, S. Majetich, J. Adv. Mater. 20 (2008) 1721–1726. Y. Ye, Y. Huang, W. Lu, B. Casse, D. Xiao, S. Bennett, D. Heiman, L. Menon, S. Sridhar, Opt. Mater. 33 (2011) 1667–1670. A. Lin, N. Halas, R. Drezek, J. Biomed. Opt. 10 (2005) 064035-1–064035-10. R. Milleron, S. Bratton, Cell. Mol. Life Sci. 64 (2007) 2329–2333. B. Hildebrandt, P. Wust, O. Ahlers, A. Dieing, G. Sreenivasa, T. Kerner, et al., Crit. Rev. Oncol. Hematol. 43 (2002) 33–56. H. Lee, S. Kim, B. Choi, M. Park, J. Lee, S. Jeong, et al., Int. J. Hyperthermia 27 (2011) 698–707. A.-M. Gabudean, F. Lerouge, T. Gallavardin, M. Iosin, S. Zaiba, O. Maury, P.L. Baldeck, C. Andraud, S. Parol, Opt. Mater. 33 (2011) 1377–1381. M. Nourbakhsh, M. Khosroshahi, Lasers Med. Sci. 26 (2011) 49–55. M. Khosroshahi, L. Ghazanfari, Mater. Chem. Phys. 133 (2012) 55–62. D.K. Roper, W. Ahn, M. Hoepfner, J. Phys. Chem. C 111 (2007) 3636–3641. N. Harris, M.J. Ford, M.B. Cortie, J. Phys. Chem. B 110 (2006) 10701–10707. R.D. Averitt, S.L. Westcott, N.J. Halas, J. Opt. Soc. Am. B 16 (1999) 1824–1832. M. Khosroshahi, L. Ghazanfari, Phys. E. 42 (2010) 1824–1829. M. Khosroshahi, L. Ghazanfari, Mater. Sci. Eng. C 32 (2012) 1043–1049. D. Duff, A. Baiker, Langmuir 9 (1993) 2301–2309. W. Huang, J.T. Lue, Phys. Rev. B 49 (1994) 17279–17285. P. Johnson, R. Christy, Phys. Rev. B 6 (1972) 4370–4379.
[24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44]
Z. Xi, G. Ying, G.Q. Huang, Chin. Phys. B 17 (2008) 2567-07. P. Mulvaney, Phys. Rev. Lett. 12 (1996) 788–800. A. Moroz, J. Phys. Chem. C 112 (2008) 10641–10652. R. Averitt, D. Sarkar, N. Halas, Phys. Rev. Lett. 78 (1997) 4217–4220. C.A. Balanis, Advanced Engineering Electromagnetics, John Wiley & Sons, New York, 2012. S. Jiang, S. Georgakopoulos, J. Electromagn. Anal. Appl. 3 (2011) 261–266. R. Adler, L. Chu, R. Fano, Electromagnetic Energy Transmission and Radiation, John Wiley & Sons, 1960. J. Nagel, S. Blair, M. Scarpulla, Opt. Express 19 (2011) 20159–20171. T. Vo-Dinh, Biomedical Photonics, CRC PRESS, 2003. R. Pope, E. Fry, Appl. Opt. 36 (1997) 8710–8723. C. Kumar, Mixed Metal Nanomaterials, Wiley-VCH, 2009. H. van de Hulst, Light Scattering by Small Particles, Dover, New York, 1981. V. Pustovalov, L. Astafyeva, B. Jean, Nanotechnology 20 (2009) 225105–225116. C. Bohren, D. Huffman, Absorption and Scattering of Light by Small Particles, John Wiley & Sons, New York, 1983. V. Goossens, J. Wielant, S. Gils, R. Finsy, H. Terryn, Surf. Interface Anal. 38 (2006) 489–493. G. Gnanaprakash, S. Mahadevan, T. Jayakumar, P. Kalyanasundaram, J. Philip, B. Raj, Mater. Chem. Phys. 103 (2007) 168–175. X. Teng, D. Black, N.J. Watkins, Y. Gao, H. Yang, Nano Lett. 3 (2003) 261–264. L. Wang, J. Luo, Q. Fan, M. Suzuki, I.S. Suzuki, M.H. Engelhard, Y. Lin, N. Kim, J.Q. Wang, C. Zhong, J. Phys. Chem. B 109 (2005) 21593–21601. S.S.Z. Xu, Y. Hou, J. Am. Chem. Soc. 129 (2007) 8698–8699. T.J. Norman Jr., C.D. Grant, A.M. Schwartzberg, J.Z. Zhang, Opt. Mater. 27 (2005) 1197–1203. B. Brinson, J. Lassiter, C. Levin, R. Bardhan, N. Mirin, N. Halas, Langmuir 24 (2008) 14166–14171.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Simulation and experimental results of optical and thermal modeling of gold nanoshells Lida Ghazanfari, Mohammad E. Khosroshahi ⁎ Laser and Nanobiophotonics Laboratory, Biomaterial Group, Faculty of Biomedical Engineering, Amirkabir University of Technology, Tehran, Iran
a r t i c l e
i n f o
Article history: Received 22 January 2014 Received in revised form 31 March 2014 Accepted 4 May 2014 Available online 22 May 2014 Keywords: Superparamagnetic iron oxide nanoparticles (SPION) Magneto-optical nanoshells (MNSs) Optical and thermal modeling
a b s t r a c t This paper proposes a generalized method for optical and thermal modeling of synthesized magneto-optical nanoshells (MNSs) for biomedical applications. Superparamagnetic magnetite nanoparticles with diameter of 9.5 ± 1.4 nm are fabricated using co-precipitation method and subsequently covered by a thin layer of gold to obtain 15.8 ± 3.5 nm MNSs. In this paper, simulations and detailed analysis are carried out for different nanoshell geometry to achieve a maximum heat power. Structural, magnetic and optical properties of MNSs are assessed using vibrating sample magnetometer (VSM), X-ray diffraction (XRD), UV–VIS spectrophotometer, dynamic light scattering (DLS), and transmission electron microscope (TEM). Magnetic saturation of synthesized magnetite nanoparticles are reduced from 46.94 to 11.98 emu/g after coating with gold. The performance of the proposed optical–thermal modeling technique is verified by simulation and experimental results. © 2014 Published by Elsevier B.V.
1. Introduction The increasing availability of nanostructures with highly controlled magnetic and optical properties has created widespread interest in the use of nanoshells for diagnostic and therapeutic applications. This is due to the unique properties exhibited by the nanoscale particles, which is not seen at bulk scale. The absorption property of nanoshells can be provided at specific wavelength and their movement can be controlled by an external magnetic field. Therefore, magneto-optical nanoshells (MNSs) can be guided to a specific tissue target [1]. Superparamagnetic magnetite has attracted increasing interest because of its outstanding properties especially in biotechnology and biomedicine applications [2–4]. In recent years, much effort has been devoted to the synthesis and characterization of MNSs such as gold coated iron-oxide nanoparticles (NPs) [1,5–7]. Generally, a nanoshell is a type of spherical nanoparticle which consists of dielectric core covered by a thin metallic shell, e.g. gold or silver. These nanoshells involve a plasmon which is a collective excitation or quantum plasma oscillation when excited by an electromagnetic field. This resonant interaction between the light field and the oscillating surface charges gives rise to a state known as surface plasmon. By different shapes and sizes of nanoshell, i.e. in core-shell ratio one can expect to achieve a tunable nanosystem ranging from visible (VIS) to the near infrared (NIR), based on the Mie theory [8]. When nanoshells are exposed to the appropriate wavelengths of a laser beam, they absorb energy and then heat up. These nanoshells
⁎ Corresponding author. Tel.: +98 2164542398; fax: +98 2166468186. E-mail address: [email protected] (M.E. Khosroshahi).
http://dx.doi.org/10.1016/j.msec.2014.05.002 0928-4931/© 2014 Published by Elsevier B.V.
mediate strong plasmon-induced surface heat flux upon absorption of light [2]. Therefore, the main application of nanoshells is for cancer treatment, known as nanoshell-based hyperthermia [9]. During hyperthermia, temperatures above 42 °C induce cell death in tumor tissues in the range of 41 °C to 47 °C [10–12]. So, nanoshell-based hyperthermia is noninvasive, safe, and efficient. It is readily shown that gold nanoshells can be an attractive alternative for both diagnostic and therapeutic applications such as medical imaging and skin wound soldering [13–15]. Recent studies in modeling attempt to analyze the optical and/or thermal effects of nanostructures [16–18]. Nevertheless, comprehensive evaluation of both optical and thermal characteristics of nanoshells (multiple-layer NPs) have been scarce. This paper presents a generalized and comprehensive technique which includes the optimal and thermal modeling to consider effects of nanoshell geometry and absorption cross section, irradiation source electric field, and concentration of nanoshells. The proposed optical–thermal modeling technique of synthesized MNSs is implemented in the MATLAB environment and verified by experimental results. 2. Materials and methods All of analytic reagents are purchased from the indicated suppliers and used without further purification: ferric chloride hexahydrate (FeCl3·6H2O) (99%), ferrous chloride tetrahydrate (FeCl2·4H2O) (99%), sodium hydroxide (NaOH, 99%), hydrochloric acid (HCl, 37%), and absolute ethanol are purchased from Merck. 3-Aminopropyltriethoxysilane (APTS), tetrakis (hydroxymethyl) phosphonium chloride (THPC), chloroauric acid (H AuCl4), potassium carbonate (K2CO3), and formaldehyde (37%) are obtained from Sigma-Aldrich (St. Louis, MO). Milli-Q
186
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
ethanol into 40 ml THPC gold solution and 4 ml 1 M NaCl and left at 4 °C. After 12 h the nanoparticles are washed and redispersed in 10 ml water. A plating solution is prepared by mixing 3 ml H AuCl4 (1%) with 200 ml aqueous solution of K2CO3 (1.8 mM). A gold shell is grown around the SPIONs by adding 1 ml of precursor suspension to 9 ml of plating solution. 50 μl of H2CO is quickly added into a 10 ml prepared suspension of precursor nanoparticles in plating solution in a glass vial and gently vortexed by hand for 10 s for reduction of Au3 +, and left for 15 min. UV–Vis absorption spectrum of the solution is recorded 20 min after reaction to verify the formation of gold nanoshell. The excess formaldehyde is removed by washing twice. The fabricated magnetite/gold nanoshells are collected by centrifuge at 6000 g. 2.4. Characterization methods
water (specific conductance 0.1 μS/cm) is deoxygenated by bubbling N2 gas for 1 h prior to the use. An Argon laser (514 nm) with power of 300 mW is used to illuminate the sample area.
X-ray diffraction (XRD) measurements are performed at ambient temperature using a PHILIPS PW1800 X-ray diffractometer with CuKα radiation. Transmission electron microscopy (TEM) is performed using a CM 200 FEG STEM Philips-M.E.R.C. operating at a voltage of 200 kV. Magnetization measurements are carried at 300 K in a magnetic field (H) of up to 20 kOe with a vibrating sample magnetometer (Meghnatis Daghigh Kavir Co. VSM/AGFM) that can measure magnetic moments as low as 10− 3 emu. For the magnetization measurements, uncoated Fe 3 O4 NPs are in dry powder form obtained by evaporating the water from the solution. The samples are dried by freeze dryer (Pishtaz Engineering Co. Model: FD-4 or FD-10). Calculation of the absorption efficiency is performed using MieLab software. Each number presented in this paper is the average of at least three measurements and reported using SPSS 15.0.
2.1. Synthesis of superparamagnetic iron oxide nanoparticles (SPIONs)
3. Optical and thermal modeling
SPIONs are fabricated following a previously reported procedure with some modification [19]. Briefly, FeCl2 ·4H2 O and FeCl3 ·6H2 O are added to 25 ml aqueous solution of 0.4 M HCl under vigorous stirring. Then the iron solution is added dropwise to 250 ml 1.5 M NaOH under magnetic stirring (1500 rpm) for 30 min at room temperature. The precipitated nanoparticles are separated by a magnet. The isolated precipitate is washed with water for 5 times and followed by washing twice using ethanol. The final precipitate is dried under vacuum at room temperature. All the synthesis steps are carried out under passing N2 gas through the solution medium.
In order to investigate optical and thermal properties of nanoshells, the dynamic model of nanoshells in the embedding medium must be first derived. This model should consider both accuracy and simulation time to manifest the precise dynamic properties.
Fig. 1. Nanoshell geometry including the core, shell, and embedding regions.
2.2. Functionalization of SPIONs Amino groups are functionalized on the nanoparticle surface by silanization reaction. Magnetite nanoparticles (0.074 g) are dispersed in 25 ml ethanol by probe sonicator for 30 min. This suspension is diluted to 150 ml by ethanol and 1 ml H 2 O. 35 μl APTES is added to the magnetic suspension under vigorous stirring [20]. One milliliter H2O is introduced into the reaction medium to initiate the hydrolysis. The synthesis operation proceeds for 7 h at room temperature. After that, modified magnetite nanoparticles are isolated by magnet and washed for 5 times and dried under vacuum at room temperature. 2.3. Synthesis of SPION/gold nanoshells SPION/gold nanoshells are fabricated by a multistep procedure through electroless plating of Au onto precursor nanoparticles (i.e. gold-seeded SPIONs) as reported previously by some modifications [15]. Briefly, a THPC gold solution composed of 2–3 nm gold colloids is produced according to Duff and Baker method [21] and aged for 2 weeks. Nanoshell precursor particles are synthesized by adding 1 ml of 0.0128 M amine-terminated magnetite nanoparticles in
3.1. Nanoshell configuration There are two distinct size regimes to be considered in the determination of the optical and thermal properties of metallic NPs and nanoshells [22]. In the extrinsic size regime, the optical properties can be fully described by specification of the nanoparticle radius and by the use of the bulk frequency dependent dielectric constant (ε(ω)). Conversely, in the intrinsic regime (2r b 50 nm), dipole plasmon absorption is by far the dominant contributor to the extinction cross-section. Thus, size dependent effects must be accounted for by employment of a size dependent dielectric function as [23,24]: 2
εða; ωÞ ¼ ε exp ðωÞ þ
ωp
ω þ iωγbulk 2
2
−
ωp
ω þ iωγ 2
ð1Þ
where εexp (ω) is the experimental dielectric function and a is the shell thickness. ωp and ω are the bulk plasma frequency and electromagnetic wave frequency, respectively. γ bulk is the bulk collisional frequency which is expressed by: γbulk ¼
VF l0
ð2Þ
where V F is the Fermi speed and l 0 is the gold electron mean free path at room temperature [25]. Dielectric function of the gold nanoshell becomes size-dependent when the nanoshell size is smaller than the gold electron mean-free
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
187
where ηwater and ηair are the intrinsic impedances of the water and air, respectively. η in the medium is defined as: η¼
rffiffiffi μ ε
ð6Þ
where μ is the medium permeability and ε is the medium dielectric function. The mentioned solution should be described for the electric field of uniform plane waves which travel in lossy media. The electromagnetic waves that travel in lossy media experience attenuation [28,30,31]. Therefore, the attenuation is considered as an ambient absorption coefficient which leads to Ewater ¼
2ηwater −αd E e ηair þ ηwater Laser
ð7Þ
where d is the distance from the surface of embedding medium and α is the real part of the propagation constant called the attenuation [28]: Fig. 2. Variation of total absorbed power versus inner and outer radii of nanoshell.
path (~42 nm) [26]. The width of the absorption peak can be described as a modification of the bulk collisional frequency which is given by [27].
ð3Þ
where r1 and r2 are the inner and outer radii of shell, respectively. reff is the effective mean free path of collisions. The nanoshell geometry, which is simulated as a suspension in water, is shown in Fig. 1. The core is magnetite and indicated by a radius r1 with a dielectric function ε1. The shell is gold and coated on the core from r1 to r2 with a dielectric function ε2. The embedding medium is water with a dielectric function ε3 and this medium is exposed to air with a dielectric function ε0.
Although the laser incident field is time dependent, it does not vary spatially over the nanoshell diameter where it is much smaller than the incident field wavelength. Therefore, the electric field of the laser is assumed to have a constant component where the wave is traveling in the z direction. Its expression in the air takes the form ELaser ¼
qffiffiffiffiffiffiffiffiffiffiffiffi ^z 2Iμ 0 c a
ð4Þ
where I is the incident laser power (W/cm2), μ0 is the air permeability, and c is the speed of light in a vacuum. A 300 mW continuous wave laser with a 1.5 mm diameter spot size is applied to simulate irradiation of the nanoshells. An Argon laser is chosen as irradiation source because of two main reasons. First, it is a continuous wave laser. Second, its wavelength is close to surface plasmon resonance wavelength of the synthesized nanoshells. At the surface between air and water, a part of the incident field will be reflected from the interface, and a part will be transmitted into water medium. The boundary conditions must be satisfied thus the incident field in the embedding medium can be determined. The boundary conditions make it necessary that the tangential components of the total vectors electric and magnetic fields on two sides of the interface be the same. By solving that, the incident electric field in the water medium is written [28,29]: Ewater ¼
2ηwater E ηair þ ηwater Laser
According to the permeability of water, one can write α as ωλ pffiffiffiffiffi εr α ¼ Re j 2π
ð9Þ
pffiffiffiffiffi where λ is the electromagnetic wavelength. ε r is the numerical value for the refractive index of water as a function of wavelength in the visible part of the spectrum. The refractive index over a broad wavelength range of 0.2 to 200 μm is written by [32]: 2
9
4
14
6
n ¼ 1:3199 þ 6878=λ −1:132 10 =λ þ 1:11 10 =λ :
ð10Þ
The imaginary part of the water refractive index is interpolated from [33]. Therefore, water refractive index at 514 nm is 1.34 + i4.55 × 10−10. The wave impedance is defined as: Zw ¼
3.2. Optical properties of nanoshells
ð8Þ
120π n
ð11Þ
where n is the real part of the water refractive index. The average power density associated with the transmitted field is given by [28]: 2
S¼
2jEwater j : Zw
ð12Þ
The absorption and the scattering cross sections from the polarizability are achieved by using scattering theory. In intrinsic size regime, the nanoshell diameter is much smaller than the incident wavelength. Therefore, it is subjected to a constant field, i.e. independent of position. 1.1 1
0.75
θ
r eff
V γ ¼ γ bulk þ F reff 2 2 1=3 ¼ ðr 2 −r1 Þ r 2 −r 1
pffiffiffiffiffi α ¼ Refjω μεg:
0.5
0.25
0
100
200
300
time (s) ð5Þ Fig. 3. The dimensionless temperature versus time.
400
188
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
that the variation of total absorbed power geometrically depends on shell diameter rather than core diameter. The total power equals the heat convection away from the system surface plus the heat radiation from the system surface [16].
Lin (counts)
(311)
(440)
(220)
Q ¼ Q con þ Q rad
(511)
(400)
ð18Þ
where
(422)
Q con ¼ hAðT−T amb Þ 4 4 Q rad ¼ εσA T −T amb
2-Theta (degrees)
A¼
ð19Þ
2 4πr 2
Fig. 4. XRD pattern for synthesized magnetite NPs.
Quasi-static approach (Rayleigh limit) should be employed in the calculation [34,18]. It is worth mentioning that quasi-static approach is helpful in nanoshell nonlinear optical effects. Therefore, the absorption cross section is given as follows [35]: σ abs ¼
pffiffiffiffiffi 8π2 ε 3 ε2 εa −ε 3 εb Im λ ε2 εa þ 2ε3 εb
ð13Þ
where
∑mc
ε a ¼ ε1 ð3−2P Þ þ 2ε 2 P ε b ¼ ε1 P þ ε2 ð3−P Þ 3 P ¼ 1−ðr 1 =r 2 Þ :
ð14Þ
As light being absorbed, heat is generated by nanoshells. All the energy absorbed from the incident field is converted to thermal energy and simultaneously released into the medium. Therefore, a single particle absorbed power under steady-state conditions is: q ¼ S σ abs :
ð15Þ
The total absorbed power given by: ð16Þ
where N is the total number of nanoshells which in our case equals 9 × 1012. When Eqs. (12), (13), and (15) are substituted in Eq. (16), they lead to:
where QF is obtained when the external heat flux is subtracted from the dissipated laser-induced heat as well as heat dissipated from absorbed light, which is given by ð21Þ
where Tmax is the maximum system temperature. A dimensionless driving force temperature (the terminal temperature difference between hot and cold spots is being considered as the driving force temperature) is given [17]: θ¼
T−T amb : T max −T amb
dθ hA ð1−θÞ ¼ dt Σmc
ð22Þ
ð17Þ
The simulation studies are implemented in the MATLAB environment. Fig. 2 shows total absorbed power as a function of core and shell diameters at different heat coefficient parameters. It demonstrates (111)
ð23Þ
× Eq. (23) can be solved using the initial condition θ = 0 at t = 0 which results in θ ¼ 1− expð−hAt=ðΣmcÞÞ:
ε2 εa −ε 3 εb : Im ε 2 εa þ 2ε3 ε b
Lin (counts)
ð20Þ
By substituting Eqs. (21) and (22) in Eq. (20):
Q ¼Nq
2 pffiffiffiffiffi ε3
dT ¼ QF dt
Q F ¼ hAðT max −T amb Þ−hAðT−T amb Þ
3.3. Thermal dynamics of nanoshells
2πNnjEwater j Q¼ 15λ
where ε for gold nanoparticle is 0.9, σ is the Boltzmann's constant, T is the nanoshell temperature, Tamb is the ambient temperature, and h is the heat transfer coefficient. It is assumed that the nanoshells do not interact with one another either plasmonically or thermally. Furthermore, for temperature that increases less than 12 K in our case, the ratio Qrad/ Tamb factored out from Eq. (19) remains constant. This fact allows a convenient linearization of the energy balance using a heat-transfer model as below:
ð24Þ
Fig. 3 presents the temperature versus time. The concentration of nanoshells has a significant effect on the amount of heat generated at a given power density. Furthermore, the heat transfer by convection will be insignificant if nanoshells and their surrounding medium are immobile [17]. This effectively indicates that the MNSs play not only an important role in generating thermal effect, but also an enhanced absorption cross section for spherical metal nanoshells can be achieved at optimized value of concentration for a better laser-hyperthermia. The nanoshell efficiency parameter has the form [36]: " !# 3k∞ t p ΔT 0 K abs r 2
¼ 1− exp − 2 I0 4k∞ r 2 c1 ρ1 r 31 þ c2 ρ2 1−r31 =r32
ð25Þ
(200) (220)
20
30
40
50
2-Theta (degrees) Fig. 5. XRD pattern for Fe3O4/Au nanoshells.
60
70
where I0 is the constant intensity of laser radiation during pulse duration tp, k∞ is the coefficient of thermal conduction of the ambient medium, c1, ρ1 and c2, ρ2 are the heat capacity and density of the material of the core (magnetite) and shell (gold) accordingly, r1 and r2 are the radii of the core and shell. Kabs is the absorption efficiency factor. Its value is calculated using MieLab software based on classical Mie scattering
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
189
Fig. 6. TEM image of Fe3O4 NPs.
theory [37]. The required parameters are the dimensions of nanoshell as well as the complex dielectric function of gold and magnetite. The values of complex dielectric functions for gold and magnetite at different wavelengths are obtained from literature [23,38]. In our case, the absorption efficiency corresponds to about 0.656 at surface plasmon resonance wavelength (λSPR) of 531 nm. Maximum value of the parameter ΔT0/I0 indicates the efficiency of transformation of absorbed optical energy by nanoshells into the thermal energy. In this case, the nanoshell efficiency parameter equals 2 × 10−5 °C cm2/W.
4. Results and discussion Structure and crystallinity of magnetic NPs are determined by XRD. As shown in Fig. 4, the peaks are indexed with the facecentered structure corresponding to magnetite phase (JCPDS card
no. 89-3854). The average size of the crystals is measured by Scherrer's formula: D ¼ kλ=β cosθ
ð26Þ
where D is the mean size of the crystalline domains, which may be smaller or equal to the grain size; k is a dimensionless shape factor with a typical value of about 0.9 which varies with the actual shape of the crystallite; λ is the X-ray wavelength; and β is the line broadening at half the maximum intensity (FWHM), after subtracting the instrumental line broadening, in radians. θ is the Bragg angle. Crystallite size varies from 65 to 70 Å over the range of 20 to 70° 2θ; and the average crystallite size of 67.91 Å is obtained using high index plane (311). The calculated unit cell parameter as 8.4 Å is about that of magnetite [39] which confirms the purity of synthesized magnetite NPs. Fig. 5 shows XRD pattern of Fe 3O 4 /Au
Fig. 7. TEM image of Fe3O4/Au nanoshells.
190
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
1
Table 1 Fe3O4 NPs and Fe3O4/Au nanoshell size measurements. Material
dTEM ± SD (nm)
DLS volume-weighted size (PDI) nm
SPIONs SPION/gold nanoshells
9.5 ± 1.4 15.8 ± 3.5
201.4 (0.224) 57.7 (0.25)
nanoshells. The absence of any peaks for magnetite is most likely due to the heavy atom effect from gold [40] as a result of the formation of Au-coated Fe3O 4 nanoparticles. The fact that the diffraction peaks from Fe3 O 4 are not observed is consistent with observations by others, [41] thus providing strong evidence for complete coverage of the Fe3O 4 core by Au supporting our TEM data, which supports the formation of Fe3O4/Au core-shell nanoparticles. To observe the agglomeration state, particle size distribution, and morphology of both magnetite nanoparticles (Fig. 6) and gold nanoshells (Fig. 7), bright field TEM and DLS are used (Table 1). The stability of magnetic particles is increased by sheltering magnetic dipole interaction through the growing gold shells. 2–3 nm gold colloids can be seen in Fig. 7. In addition, based on the DLS measurement, these gold colloids are 2.34 ± 0.62 nm in size. Fig. 8 shows the magnetization curve for magnetite core and gold nanoshells at ambient temperature. It indicates the superparamagnetic behavior of nanostructures. It can be deduced that the magnetite NPs have magnetization saturation (Ms) of about 46.94 emu/g. The Ms value of the resultant gold nanoshells decreases to 11.98 emu/g, when the gold is added. Fast magnetic response of the synthesized NPs can be an advantage in many applications. With the same magnetite core size and 3.5 nm gold shell thickness, the Ms value for the synthesized nanoshells in this paper is larger than that reported by Xu and Hou [42]. In Fig. 9, the variation intensity of the absorption peaks with wavelength is shown. It is obvious that the absorption peak decreases nonlinearly. Magnetite NP has a strong absorption peak about 200 nm. However, its visible absorption is quite different from magnetite/gold nanoshell which confirms the effect of SPR played by gold shell. The optical absorption spectrum range of Fe3O4/Au nanoshell is relatively broad compared with pure gold colloid. In addition, the plasmon linewidth is dominated by electron surface scattering. The broadening of resonance absorption is related to the size, shape, and aggregation of the nanometer scale metallic particles, according to Mie's theory [43]. The nanoshells produced by Lim et al. [7] had an 18 nm magnetic core and a gold shell thickness of 5 nm with a λSPR at 605 nm. A SPR wavelength of 528 nm was obtained for Au/Fe oxide nanoshells synthesized by Pham et al. [1] with a size range of 15–40 nm. In our case, a SPR wavelength of 531 nm is obtained for nanoshells with a total
Absorbance (a. u.)
0.9 0.8 0.7 0.6 0.5
magnetite/gold
0.4 0.3 150
magnetite
250
350
450
550
650
750
Wavelength (nm) Fig. 9. UV–VIS spectra of Fe3O4 NPs (solid line) and Fe3O4/Au nanoshells (dashed line).
size of 15.8 ± 3.5 nm. The higher λSPR reported for some synthesized nanoshells with the same size may be due to the formation of an incomplete shell layer during the reduction of Au3+ [44]. The temperature variation versus different nanoshell concentrations is studied at 17 W/cm2 for a period of 400 s, as shown in Fig. 10. The experimental system is set up for measuring the temperature in the water dispersed with gold nanoshells. The laser is incident on the cuvette containing a gold nanoshell solution. A thermistor probe is placed in the solution to monitor the temperature change. The thermistor with an accuracy of 0.15% (Redfish Sensors Inc., Model QTGB-14 D3) is connected to a thermometer (Model No. HP-34420A) to record the temperature. The thermometer is connected at the other end to a Windows-based laptop where the results are displayed on the screen. The thermometer gives the temperature readings as a function of time as the laser-induced resonance heating proceeds. Experiments in aqueous solution medium represent that the temperature exponentially saturates within about 400 s of laser beam exposure. This is true for all nanoshell concentrations which is in good agreement with simulation result shown in Fig. 3. A maximum temperature rise of 12 °C is achieved at 300 μg/ml. The MNS embedding medium, i.e. pure water, has negligible effect on temperature variation during laser light exposure. The temperature reaches to 5 °C after 400 s at both concentrations of 25 and 50 μg/ml. Increase of the MNS concentration, up to 50 μg/ml, has a negligible effect on temperature. However, increase of the MNS concentration, above 50 μg/ml, rises the maximum temperature. Such a phenomenon can be due to the arrangement, density, and potential for aggregation of Au nanoshells. Therefore, it results in a stronger surface plasmon absorption of light and more efficient photothermal effects. All the experiments demonstrate that the temperature increases with increasing exposure time. The results show that the fabricated MNSs are excellent candidates for hyperthermia tumor therapy. 5. Conclusion
Fig. 8. Variation of magnetization with applied magnetic field for bare magnetite NPs and magnetite-gold nanoshells.
This paper presents the generalized method for optical and thermal modeling of synthesized MNSs. The proposed modeling method considers effects of nanoshell geometry and absorption cross section, irradiation source electric field, and concentration of nanoshells. Furthermore, it can be extended to any n-layer nanoshells at any laser wavelength. In this paper, gold shells are formed on the surface of 9.5 ± 1.4 nm Fe3O4 NPs to obtain 15.8 ± 3.5 nm MNSs. Structural, magnetic and optical properties of superparamagnetic MNSs are characterized. Moreover, the experimental results, based on SPR-induced thermal effect and Mie theory, exhibit higher concentration of gold nanoparticles that produced a higher temperature rise. Simulation of the proposed optical–thermal modeling technique of synthesized MNSs is carried out in the MATLAB environment and validated by experimental results.
L. Ghazanfari, M.E. Khosroshahi / Materials Science and Engineering C 42 (2014) 185–191
191
Fig. 10. Temperature variation (ΔT) versus different nanoshell concentrations over a period of 400 s at a maximum laser power density of 17 W/cm2.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]
T. Pham, C. Cao, S. Sim, J. Magn. Magn. Mater. 320 (2008) 2049–2055. S. Wang, Y. Zhou, W. Sun, Mater. Sci. Eng. C 29 (2009) 1196–1200. L. Sun, C. Huang, T. Gong, S. Zhou, Mater. Sci. Eng. C 30 (2010) 583–589. J. Viota, A. Carazo, J. Munoz-Gamez, K. Rudzka, R. Gómez-Sotomayorb, A. RuizExtremera, et al., Mater. Sci. Eng. C 33 (2013) 1183–1192. Z. Hassannejad, M. Khosroshahi, Opt. Mater. 35 (2013) 644–651. H. Liu, P. Hou, W. Zhang, J. Wu, Colloids Surf. A Physicochem. Eng. Asp. 356 (2010) 21–27. J. Lim, A. Eggeman, F. Lanni, R. Tilton, S. Majetich, J. Adv. Mater. 20 (2008) 1721–1726. Y. Ye, Y. Huang, W. Lu, B. Casse, D. Xiao, S. Bennett, D. Heiman, L. Menon, S. Sridhar, Opt. Mater. 33 (2011) 1667–1670. A. Lin, N. Halas, R. Drezek, J. Biomed. Opt. 10 (2005) 064035-1–064035-10. R. Milleron, S. Bratton, Cell. Mol. Life Sci. 64 (2007) 2329–2333. B. Hildebrandt, P. Wust, O. Ahlers, A. Dieing, G. Sreenivasa, T. Kerner, et al., Crit. Rev. Oncol. Hematol. 43 (2002) 33–56. H. Lee, S. Kim, B. Choi, M. Park, J. Lee, S. Jeong, et al., Int. J. Hyperthermia 27 (2011) 698–707. A.-M. Gabudean, F. Lerouge, T. Gallavardin, M. Iosin, S. Zaiba, O. Maury, P.L. Baldeck, C. Andraud, S. Parol, Opt. Mater. 33 (2011) 1377–1381. M. Nourbakhsh, M. Khosroshahi, Lasers Med. Sci. 26 (2011) 49–55. M. Khosroshahi, L. Ghazanfari, Mater. Chem. Phys. 133 (2012) 55–62. D.K. Roper, W. Ahn, M. Hoepfner, J. Phys. Chem. C 111 (2007) 3636–3641. N. Harris, M.J. Ford, M.B. Cortie, J. Phys. Chem. B 110 (2006) 10701–10707. R.D. Averitt, S.L. Westcott, N.J. Halas, J. Opt. Soc. Am. B 16 (1999) 1824–1832. M. Khosroshahi, L. Ghazanfari, Phys. E. 42 (2010) 1824–1829. M. Khosroshahi, L. Ghazanfari, Mater. Sci. Eng. C 32 (2012) 1043–1049. D. Duff, A. Baiker, Langmuir 9 (1993) 2301–2309. W. Huang, J.T. Lue, Phys. Rev. B 49 (1994) 17279–17285. P. Johnson, R. Christy, Phys. Rev. B 6 (1972) 4370–4379.
[24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44]
Z. Xi, G. Ying, G.Q. Huang, Chin. Phys. B 17 (2008) 2567-07. P. Mulvaney, Phys. Rev. Lett. 12 (1996) 788–800. A. Moroz, J. Phys. Chem. C 112 (2008) 10641–10652. R. Averitt, D. Sarkar, N. Halas, Phys. Rev. Lett. 78 (1997) 4217–4220. C.A. Balanis, Advanced Engineering Electromagnetics, John Wiley & Sons, New York, 2012. S. Jiang, S. Georgakopoulos, J. Electromagn. Anal. Appl. 3 (2011) 261–266. R. Adler, L. Chu, R. Fano, Electromagnetic Energy Transmission and Radiation, John Wiley & Sons, 1960. J. Nagel, S. Blair, M. Scarpulla, Opt. Express 19 (2011) 20159–20171. T. Vo-Dinh, Biomedical Photonics, CRC PRESS, 2003. R. Pope, E. Fry, Appl. Opt. 36 (1997) 8710–8723. C. Kumar, Mixed Metal Nanomaterials, Wiley-VCH, 2009. H. van de Hulst, Light Scattering by Small Particles, Dover, New York, 1981. V. Pustovalov, L. Astafyeva, B. Jean, Nanotechnology 20 (2009) 225105–225116. C. Bohren, D. Huffman, Absorption and Scattering of Light by Small Particles, John Wiley & Sons, New York, 1983. V. Goossens, J. Wielant, S. Gils, R. Finsy, H. Terryn, Surf. Interface Anal. 38 (2006) 489–493. G. Gnanaprakash, S. Mahadevan, T. Jayakumar, P. Kalyanasundaram, J. Philip, B. Raj, Mater. Chem. Phys. 103 (2007) 168–175. X. Teng, D. Black, N.J. Watkins, Y. Gao, H. Yang, Nano Lett. 3 (2003) 261–264. L. Wang, J. Luo, Q. Fan, M. Suzuki, I.S. Suzuki, M.H. Engelhard, Y. Lin, N. Kim, J.Q. Wang, C. Zhong, J. Phys. Chem. B 109 (2005) 21593–21601. S.S.Z. Xu, Y. Hou, J. Am. Chem. Soc. 129 (2007) 8698–8699. T.J. Norman Jr., C.D. Grant, A.M. Schwartzberg, J.Z. Zhang, Opt. Mater. 27 (2005) 1197–1203. B. Brinson, J. Lassiter, C. Levin, R. Bardhan, N. Mirin, N. Halas, Langmuir 24 (2008) 14166–14171.
