CHEMISTRY O

1. Home Search Collections Journals About Contact us My IOPscience Collective optical Kerr effect exhibited by an integrated configuration of silicon quantum dots and gold nanoparticles embedded in ion-implanted silica This content has been downloaded from IOPscience. Please scroll down to see the full text. 2015 Nanotechnology 26 295701 (http://iopscience.iop.org/0957-4484/26/29/295701) View the table of contents for this issue, or go to the journal homepage for more Download details: IP Address: 130.179.16.201 This content was downloaded on 05/09/2015 at 02:36 Please note that terms and conditions apply.

2. Nanotechnology Nanotechnology 26 (2015) 295701 (8pp) doi:10.1088/0957-4484/26/29/295701 Collective optical Kerr effect exhibited by an integrated configuration of silicon quantum dots and gold nanoparticles embedded in ion-implanted silica C Torres-Torres1, A López-Suárez2, B Can-Uc3, R Rangel-Rojo3, L Tamayo-Rivera4and A Oliver2 1Sección de Estudios de Posgrado e Investigación, ESIME ZAC, Instituto Politécnico Nacional, México, D.F. 07738, Mexico 2Instituto de Física, Universidad Nacional Autónoma de México, D. F. 04510, Mexico 3Depto. de Óptica, Centro de Investigación Científica y de Educación Superior de Ensenada A.P. 360, Ensenada, B.C., 22860, Mexico 4Escuela Superior de Apan, Universidad Autónoma del Estado de Hidalgo, Mineral de la Reforma, Hidalgo, 42090, Mexico E-mail: rrangel@cicese.mx Received 9 February 2015, revised 15 May 2015 Accepted for publication 1 June 2015 Published 2 July 2015 Abstract The study of the third-order optical nonlinear response exhibited by a composite containing gold nanoparticles and silicon quantum dots nucleated by ion implantation in a high-purity silica matrix is presented. The nanocomposites were explored as an integrated configuration containing two different ion-implanted distributions. The time-resolved optical Kerr gate and z-scan techniques were conducted using 80 fs pulses at a 825nm wavelength; while the nanosecond response was investigated by a vectorial two-wave mixing method at 532nm with 1ns pulses. An ultrafast purely electronic nonlinearity was associated to the optical Kerr effect for the femtosecond experiments, while a thermal effect was identified as the main mechanism responsible for the nonlinear optical refraction induced by nanosecond pulses. Comparative experimental tests for examining the contribution of the Au and Si distributions to the total third- order optical response were carried out. We consider that the additional defects generated by consecutive ion irradiations in the preparation of ion-implanted samples do not notably modify the off-resonance electronic optical nonlinearities; but they do result in an important change for near-resonant nanosecond third-order optical phenomena exhibited by the closely spaced nanoparticle distributions. Keywords: nonlinear optics, optical Kerr effect, ultrafast processes (Some figures may appear in colour only in the online journal) 1. Introduction In this direction, the challenge for designing highly improved nanoparticles (NPs) is probably among the most motivating considerations that require to be taken into account to implement many all-optical devices. The assemblage of dis- tinct elements in advanced NCs gives us the capability of tailoring their characteristics, which strongly depend on the density of NPs, their size and shape [1]. There are several Due to their interesting third-order optical nonlinear proper- ties, ion-implanted nanocomposites (NCs), have given rise to outstanding potential applications ranging from ultra-small instruments, sensors and nonlinear optical systems that have originated a great interest from a technological point of view. 0957-4484/15/295701+08$33.00 1 © 2015 IOP Publishing Ltd Printed in the UK

3. Nanotechnology 26 (2015) 295701 C Torres-Torres et al NPs and a sample containing both is comparatively investigated. preparation methods devised over the years to produce NCs with different features; they currently constitute a very active field of research for fabricating nanomaterials [2]. In particular, for NCs containing metallic NPs, the sur- face plasmon resonance (SPR) represents a promising tool for a group of applications, based on the different plasmonic processes associated with this resonance [3–7]. Since the medium surrounding the NPs plays an important role in determining their properties, diverse processing routes for encapsulating Au-NPs have been followed [8]. This includes studies of the influence of the supporting medium on the physical properties of the NCs [9], and even the fabrication of ordered NP arrays [10]. On the other hand, the development of silicon quantum dots (Si-QDs) is also a clear case that has allowed the opti- mization of numerous optical processes related to resonant effects, like optical absorption [11], photoluminescence [12], or third-order optical nonlinearities [13]. Furthermore, multiple procedures for enhancing the scenario of luminescent phenomena have emerged from grouping particular Si-QDs and Au-NPs in particular [14, 15]. Besides, it is well known that strong luminescence can be achieved, and tuned, in Si-QDs by quantum confinement effects in the quantum dots. Although it has been pointed out that quenching of Si nanocrystal photoluminescence is accomplished by doping with Au [16], by contrast, the ion- implantation technique has demonstrated that luminescent properties of Si nanocrystals are preserved under irradiation with Au ions [17]. Hence, it could be expected that the nonlinear optical response of binary (compounded) compo- sites can be tailored by the combination of Si-QDs and Au- NPs using an ion-implantation route for their preparation. Comparatively, it has been previously indicated that ion- implanted NCs containing both Ag-NPs and Si-QDs result in picosecond third-order optical enhanced compared to that of samples containing only either Ag-NPS or Si-QDs [18]. Moreover, for a particular wave- length of excitation, the ultrafast nonlinear optical absorption in NCs conformed by Ag-NPs and Si-QDs can be tailored by a double-ion implantation route [19]. It is worth noting that Ag-NPs and Au-NPs can show opposite signs of the nonlinear refractive index, and therefore possible cancellation of both effects is in principle con- ceivable in a bilayer system [20]. However, Au-NPs and Si- QDs display nonlinear refractive indexes of the same sign for wavelengths near to the SPR. With this motivation, it can be inferred that an enhancement in the optical Kerr effect could be generated by the combination of Au-NPs and Si-QDs. Then, this work is devoted to giving insight into the third-order nonlinear optical response exhibited by Au-NPs and Si-QDs prepared by ion-implantation and embedded in a high-purity silica matrix. The scope of this research is related to femto- and nano-second experiments that confirm the possibility of enhancing their refractive contribution to the nonlinearity by the collective physical mechanisms associated to the thermal and optical Kerr effects generated in a multi- component configuration of Si-QDs and Au-NPs. Within this work the nonlinear optical response exhibited by Si-QDs, Au- 2. Experiment 2.1. Sample preparation and characterization During implantation the ions interact with the electrons and nucleus of the target producing physical, electrical and che- mical changes in the material by transferring their energy and momentum to the electrons and atomic nuclei of the target material. The ion interaction causes defects in the matrix, such as Si- or O-broken bonds, but these defects are passivated by the presence of hydrogen during annealing. In this work, three high-purity silica glass plates (25×25×1mm) with OH content less than 1ppm and total impurity content less than 20 ppm, were implanted at room temperature. First, two identical samples were prepared, each one exposed to 1.5 MeV Si+2ions at a fluence of 2.5×1017ions/cm2. After implantation, the samples were thermally annealed in a reducing atmosphere (50% N2+50% H2) at 1100 °C for 90 min. Thermal annealing promotes the nucleation of the Si- QDs from the supersaturated solution, and at the same time assures the reduction of the defects produced during the implantation, as well as the passivation of the sample. Afterwards, one of the samples containing Si-QDs was implanted at room temperature with Au+2ions with energy of 1.5 MeV at a fluence of 8.5×1016ions/cm2. A third sample was produced by 2MeV Au+2ions implanted onto a silica substrate at a 5×1016ions/cm2fluence. Since nanoclusters may not be nucleated during implantation process, a sub- sequent thermal treatment must be conducted for synthesizing the Au-NPs. Then, in order to form the Au-NPs, immediately after each implantation was concluded, the samples were thermally annealed in an oxidizing atmosphere (air) at 1000 °C for 60 min. The end results were three samples, one containing Si-QDs nucleated by 1.5 MeV Si+2ions at a 2.5×1017ions/cm2fluence, a second one containing Au-NPs nucleated by 2MeV Au+2ions at 5×1016ions/cm2, and a third sample containing Si-QDs nucleated by 1.5 MeV Si+2 ions at 2.5×1017ions/cm2 1.5 MeV Au+2ions at a 8.5×1016ions/cm2fluence. The main reason of having synthesized samples with Au distributions using different energies (1.5 and 2MeV) is to assure that Si distribution would not overlap Au distribution inside the silica matrix. If Au ions would have been implanted at 2MeV in the same sample that had been previously implanted with Si at 1.5 MeV, the tails of both distributions would overlap. Once the Si-QDs and the Au-NPs were embedded in the same silica matrix, the samples were studied and character- ized using different techniques. Monte Carlo simulations were performed using the SRIM code in order to determine the Gaussian depth distribution of the implanted ions. Rutherford backscattering spectrometry (RBS) was performed at 3 MeV using a4He2+beam, in order to obtain the sample composi- tion and to corroborate the distribution of the Au in the substrate. The SIMNRA code [21] was used to calculate the nonlinearities that are and Au-NPs nucleated by 2

4. Nanotechnology 26 (2015) 295701 C Torres-Torres et al depth profile and the concentration of the implanted Au. These measurements, as well as the ion implantations, were performed at the 3MV Tandem accelerator (NEC 9SDH-2 Pelletron) facility at the Instituto de Física of the Universidad Nacional Autónoma de México. The linear optical absorption spectra of the resulting samples were acquired by using an UV–visible spectrometer (Cary 5000-Varian). z, and the on-axis optical phase shift of the beam ΔΦo, can be written as [23]: ΔΦ 4 x ( ) o ΔΦ = − , 1 , (2) T z ( )( ) o + + 2 2 9 1 x x = = π λ = 2 with x wavelength, all in free space. In addition, the peak nonlinear phase change ΔΦ0is given by: ΔΦ here Δn0= n2I0represents the on-axis maximum refractive index change, where n2is the nonlinear refractive index, and I0is the peak on-axis irradiance at focus. and λ the laser z z , o 2 , 2 k z kw o o 2.2. Time-resolved optical Kerr gate = k n L Δ In order to evaluate the electronic behavior associated to the Kerr nonlinearities together with the recovery times exhibited by the Au-NPs and the Si-QDs, the time-resolved Optical Kerr gate (OKG) technique was employed, using what has become the standard configuration [22], with femtosecond pulses. In this configuration, a half wave plate, λ/2, with a polarizer A1, was employed to control the plane of polar- ization of the probe beam. The pump and probe beams, with an irradiance relation of 15:1, their propagation directions making a small angle, and their linear polarizations making an angle of 45°, were focused on the sample under study with a spot size of 80 μm. An analyzer with its transmission axis crossed with respect to the initial polarization of the probe beam was placed in front of a photodetector providing the Kerr signal. The probe beam energy was measured using a lock-in amplifier tuned at a frequency determined by a mechanical chopper. By delaying the probe beam with respect to the pump beam, it was possible to observe the probe transmittance as a function of pump to probe pulse delay time, and hence the recovery of the response. The source employed is a Ti:sapphire laser with λ=825nm, 80 fs pulses, 3nJ maximum pulse energy and a repetition rate of 94 MHz. , (3) eff o o 2.4. Nanosecond nonlinear optical response In order to further investigate the third-order nonlinear optical behavior of the samples, excitation in near-resonance to the Au-NP surface plasmon peak was employed. The nanosecond third-order nonlinear optical response was explored by using a vectorial two-wave mixing (TWM) experiment [24], with a Nd-YAG laser Continuum Model SL II-10 at a 532 nm wavelength and 1ns pulse duration. Linearly polarized pump and probe beams were made to coincide in the sample with their propagation vectors making a 20° angle. A single pulse with maximum energy of 65 mJ was employed as a pump beam, while the probe beam consisted of a 10 nJ pulse. The beam waists for the pump and probe beams were measured, giving 6 and 1mm for the pump and the probe, respectively. The polarized irradiance of the probe-beam in the two-wave interaction was measured in different cases of polarization of the incident waves making an angle ϕ between their planes of polarization. The axis of transmission of the analyzer placed in front of the photodetector was aligned in order to capture the orthogonal component of the transmitted polarization. The expressions for describing the amplitudes of the transmitted fields can be written as [25]: ⎡⎣ ⎤⎦ 2.3. Femtosecond z-scan experiments In order to identify the different third order nonlinear optical processes involved in the response to fs pulses, the refractive and absorptive contributions to the nonlinearity were mea- sured through the z-scan technique [23]. The same Ti:sap- phire laser system as in the time-resolved OKG experiments was employed, using here focusing conditions producing a 47μm beam waist, resulting in 0.87GW cm−2, at a 825 nm wavelength. The open- and closed-aperture configurations in the z-scan experiments were separately obtained under the same energy and irradiance conditions. The approximation of the transmittance for the open z-scan measurements was evaluated by means of the following equation [23], ( ) ( ) ( J ) = Ψ + − Ψ α 0 ± (1) ± 0 0 (1) ± ( ) i i E z E J E E ± ± ± 1 0 1 1 2 3 ⎛ ⎞ I z ( ) 2 ( ) ⎜ ⎝ ⎟ ⎠ − Ψ − Ψ − 0 (1) ± (0) ± exp i , (4) E J ± 2 4 ⎡⎣ ( ) ( ) ) ( J ) = Ψ + − Ψ α 0 (1) ± 0 0 ± (1) ± ( ) i i E z E J E E ± ± ± 2 0 1 2 4 1 ⎛ ⎞ ⎤⎦ I z ( ) 2 ( ⎜ ⎝ ⎟ ⎠ − Ψ − Ψ − 0 (1) ± (0) ± exp i , (5) E J ± 2 β 3 I L o ( ) eff + ΔΦ = − , 1 , (1) T z ( ) o 2 2 2 1 x where E1±(z) and E2±(z) are the complex amplitudes of the circular components of the transmitted waves beams; E3±(z) and E4±(z) are the amplitudes of the self-diffracted waves, while E , 1 2 and self-diffracted waves at the surface of the sample; α(I) is the irradiance dependent absorption coefficient; stands for the Bessel function of order m, z is the thickness of the nonlinear media, and Ψ± and changes. where β represents the two-photon absorption coefficient, and Leffis the effective length given by L with L the sample length, and αo the linear absorption coefficient. For the closed-aperture z-scan technique, it has been shown [23] that for a Gaussian beam with waist radius wo, travelling in the +z direction, the geometry-independent nor- malized transmittance T, as a function of the sample position − α = − α L (1 )/ , e eff 0 0 0 ±E , 0 0 0 ±and E4 ±are the amplitudes of the incident ±E3 ( ) Ψ± (1) Jm Ψ± are the nonlinear phase (0) (1) 3

5. Nanotechnology 26 (2015) 295701 C Torres-Torres et al Figure 1. RBS experimental spectrum of the sample implanted with Si and Au ions at 1.5 MeV. Figure 2. Numerical Monte Carlo simulation of the Si and Au ions implanted at 1.5 MeV. 3. Results The concentration and the Au depth profile in the sample containing Si-QDs and Au-NPs were experimentally eval- uated by Rutherford backscattering spectroscopy (RBS) measurements; figure 1 illustrates the results, where the pre- sence of oxygen and silicon comes from the conforming elements of the silica (SiO2) matrix. The Si amount that comes from the Si-QDs is veiled by the Si signal of the silica matrix. The evaluation of the Au atoms showed an asym- metric distribution centered at a depth of about 0.36μm. Monte Carlo simulations were performed with the SRIM code in order to determine the Gaussian depth distribution of the implanted Au and Si ions. The SRIM code is based on a Monte Carlo simulation method, named binary collision approximation, with a random selection of the impact para- meter of the next colliding ion. The same experimental parameters used during implantation experiment, were used during SRIM simulation to obtain the ion distributions. These parameters correspond to type and ion energy (1.5 MeV for Si and 1.5 and 2MeV for Au), angle of incidence of the pro- jectile (0° in all cases) and host material (SiO2). According to SRIM results, the projectile range of the implanted Si is 1.65μm; meanwhile the ion ranges of the Au are 0.36 and 0.47μm for the Au atoms implanted at 1.5 and 2MeV, respectively. Monte Carlo simulation also allows us to esti- mate the thickness of the separated Au and Si distributions, which is 1.29 μm, as it can be seen from figure 2. During the ion-implantation process, particle distribu- tions are formed. The Si-QDs/Au-NP distributions were performed separately. Firstly, Si ions were implanted at 1.5MeV and afterwards 1.5 MeV Au ions were implanted. After energy loss calculations, we determined that Si ions implanted at 1.5 MeV would reach 2 μm in silica. With this data in mind, now we calculate the energy loss of Au ions in silica. Since Au atoms are heavier than Si atoms, and lose more energy during implantation, the first ones would stop closer to the silica surface than the last ones. After Figure 3. Linear absorption spectra of the samples studied. calculations, we obtained that 1.5 MeV Au ions would stop at 0.66 μm in silica. In this way, we assure that both distribu- tions (Si and Au) never overlap, so there is no way to get a wider distribution, since we can control the position of the distributions. Figure 3 depicts the linear optical absorption spectra of the samples studied. Close to 520 nm, one can clearly observe the peak of the absorbing bands associated with the SPR of the Au-NPs. To carry out the nonlinear optical experiments, the femto- and nano-second measurement systems were pre- viously calibrated using carbon disulfide (CS2) with a thick- ness of D=1mm, as a reference of nonlinear medium with a well-known third-order nonlinear susceptibility of =1.9×10−12esu [26]. Figure 4 shows the time-resolved OKG results for the samples with NP distributions corresponding to the Au-NPs and Si-QDs by themselves. In the figure it can be seen that the signals associated with the nonlinear optical response raise χ ∣ ∣ (3) 4

6. Nanotechnology 26 (2015) 295701 C Torres-Torres et al Figure 4. Time-resolved optical Kerr gate signal for the samples studied. and decay within the duration of the pulse for both cases, pointing to an ultrafast purely electronic optical response for both samples. Due to the fact that the Kerr gate signal only gives information about |χ(3)|, we conducted further z-scan studies in order to resolve the refractive and absorptive contributions to the response. The open-aperture z-scan results are presented in figure 5. An important saturated absorption effect can be notably seen for this pulse duration and wavelength in all the samples studied. The open-aperture z-scan results show the presence of saturable absorption, i.e. decreasing absorption for high irradiances for the three samples studied. The saturation effect is clearly stronger for the Si-QD sample than for the other two. This seems somehow strange, since the third sample also contains Si-QDs, and hence it should show a stronger nonlinear optical absorptive effect. The plots also show fits made to the data using expression (1), from which the respective β are extracted. For the case of saturable absorption, the irradiance-dependent absorption coefficient has the form α(I)=α0/(1+ I/Is), which does not strictly cor- respond to a third-order nonlinear process. However, for small irradiance values, i.e. I0/Is≪ 1, this expression can be approximated to α(I)≈α0−(α0/Is)I, and therefore we can identify β = −α0/Is as the corresponding negative TPA coefficient, related to Imχ(3). Figure 6 shows the experimental data for the closed- aperture z-scan results obtained under the same irradiance values. The signature of a positive nonlinear refractive index n2for the fs pulses, i.e. a pre-focal minimum followed by a post-focal maximum, can be clearly observed in the three cases studied. The plots also show fits made to the data using equation (2), from which the respective n2 values are extracted. For the nanosecond two-wave mixing experiment, figure 7 plots the experimental probe beam transmittance as function of the angle between polarizations ϕ, together with the fits obtained using equations (4) and (5). An error bar of Figure 5. Femtosecond open-aperture z-scan results at 825 nm (a) Si- QDs (b) Au-NPs (c) Si-QDs/Au-NPs. ±20% was estimated for the nanosecond experimental irra- diance data. Table 1 illustrates a summary of the fitted parameters obtained for the femto- and nano-second nonlinear optical measurements. 5

7. Nanotechnology 26 (2015) 295701 C Torres-Torres et al Figure 7. Nanosecond nonlinear optical probe transmittance versus angle of polarization of the interacting beams in the samples studied. nonlinear optical absorption coefficients are presented prob- ably as a consequence of their differences in linear optical absorption. In the cases studied, the formation of distributions with different sizes of Si-QDs can be expected after following a different number of thermal annealing processes, and also due to the dissimilar ion-irradiation energies employed for the implantation in each sample. In addition, we consider that subsequent ion-implantation processes produces additional diffusion and nucleation processes, which can result in changes in the structure, and modification in size of the initially implanted NPs, as it has been previously suggested for dual implantation processes [27]. Moreover, since for a given ion different implantation energies result in different penetration depths and distribution widths (as observed from both RBS and SRIM results [28]), the likely result is asymmetric distributions with different particle sizes and densities. On the other hand, it has been reported that thermal annealing of silicon nanocrystals seems to produce an increment in the size of the particles, with a corresponding modification of their resulting nonlinear optical absorption [29, 30]. Based on these considerations, it would be feasible to predict that samples with single- or dual-ion implantation would present different nonlinear absorption (and refraction) properties, even though the linear absorption differences observed could not in principle account for the large differences resulting as nonlinear optical effects. Alternatively, the change in the linear and nonlinear optical properties should be also accompanied by a variation in the resulting dielectric constants, and correspondingly deriving in an alteration of the mechanical response too. Thus, the change in mass density originated by the optical Kerr effect could be useful for manipulating dielectrical or mechanical functions, with different operations based on these effects actually being proposed [31–34]. Moreover, for all-optical applications based on optical Kerr effect designed by multi-components conforming NCs, the selected elements that ought to be contemplated to take part of the same NCs should avoid a decrease in the collective Figure 6. Femtosecond closed-aperture z-scan results at 825 nm. (a) Si-QDs (b) Au-NPs (c) Si-QDs/Au-NPs. 4. Discussion From the femtosecond data presented in table 1, it can be seen that the nonlinear refractive index exhibited by the three samples possess a value that is within the same order of magnitude. By contrast, remarkable differences for the 6

8. Nanotechnology 26 (2015) 295701 C Torres-Torres et al Table 1. Nonlinear optical parameters evaluated in the samples studied. Nonlinear optical response at λ=825 nm, t=80 fs Nonlinear optical response at λ=532 nm, t=1 ns χ χ n2[m2W−1] β [m W−1] (3) (3) Sample [esu] [esu] 3.37×10−15 5.55×10−15 8.94×10−15 −1.26×10−8 −1.51×10−7 −6.43×10−9 2.6536×10−9 8.682×10−9 1.21×10−10 3.5×10−9 5.3×10−10 2.2×10−9 Si-QDs/Au-NPs Si-QDs Au-NPs nonlinear optical response. This implies that they should present nonlinear refractive coefficients of the same sign, i.e. either both focusing or both self-defocusing effects. As we see from the results obtained, this is actually the case of the combination of Au-NPs and Si-QDs. However for mechano- optical functions, opposite signs in the optical Kerr effect exhibited by multi-components could provide the possibility to perform functions for mechanical expansion and contrac- tion. According to comparative previous results in dual- implanted NCs [18, 19], the combination of Si-QDs with Au- NPs results in an optical Kerr effect that is approximately one order of magnitude higher than Si-QDs and Ag-NPs. Within this work we demonstrate that concerning their nonlinear optical response, the coupling of SPR in Au-NPs in contribution to the optical Kerr effect is similarly developed by Au-NPs even under interaction with Si-QDs in an inte- grated configuration. Furthermore, the optical Kerr effect of ion-implanted Au-NPs seems to present very similar proper- ties whether they were implanted in a first stage, or whether they were nucleated in a second-stage process resulting from a consecutive irradiation in a previously Si-implanted matrix. However, the Au-implanted samples developed by con- secutive irradiations could provide a possible way to tune the nonlinear optical absorption properties of NCs in different controlled distributions. Furthermore, it is worth noting that the addition of diverse elements for preparing NCs also allows the accumulation of the resonances that separately correspond to each element. This would broaden the wave- length range for exciting optical nonlinearities. Potential applications to perform multifunctional integrated devices could be contemplated. remarkable changes in the nonlinear optical properties of the system, potential applications for designing all-optical nano- devices can be considered. Acknowledgments The authors kindly acknowledge the financial support from CICESE, Universidad Nacional Autónoma de México, Insti- tuto Politécnico Nacional and CONACyT. Also, the authors wish to acknowledge the technical assistance of J.G. Morales, K. López, F.J. Jaimes. This work has been partially supported under projects DGAPA-UNAM IN-100213 and CONACyT- México 99224, 102937 and 222485. References [1] Vasquez Y, Kolle M, Mishchenko L, Hatton B D and Aizenberg J 2014 Three-phase co-assembly: in situ incorporation of nanoparticles into tunable, highly ordered, porous silica films ACS Photonics 1 53–60 [2] Cesca T, Pellegrini G, Bello V, Scian C, Mazzoldi P, Calvelli P, Battaglin G and Mattei G 2010 Nonlinear optical properties of Au–Ag nanoplanets made by ion beam processing of bimetallic nanoclusters in silica Nucl. Instrum. Methods B 268 3227–30 [3] Kauranen M and Zayats A V 2012 Nonlinear plasmonics Nat Photonics 6 737–48 [4] Marín M J, Rashid A, Rejzek M, Fairhurst S A, Wharton S A, Martin S R, McCauley J W, Wileman T, Field R A and Russell D A 2013 Glyconanoparticles for the plasmonic detection and discrimination between human and avian influenza virus Org. Biomol. Chem. 11 7101–7 [5] Mendes M J, Hernández E, López E, García-Linares P, Ramiro I, Artacho I, Antolín E, Tobías I, Martí A and Luque A 2013 Self-organized colloidal quantum dots and metal nanoparticles for plasmon-enhanced intermediate- band solar cells Nanotechnology 24 345402 [6] Gurevich E L, Temchura V V, Überla K and Zybin A 2011 Analytical features of particle counting sensor based on plasmon assisted microscopy of nano objects Sensors Actuators B 160 1210–5 [7] Hashimoto S, Werner D and Uwada T 2012 Studies on the interaction of pulsed lasers with plasmonic gold nanoparticles toward light manipulation, heat management, and nanofabrication J. Photochem. Photobiol. C 13 28–54 [8] Crouch J J, Carley A F, Davies P R, Morgan D J, Lalev G, Dimov S and Pham D-T 2013 Encapsulation of Au nanoparticles on a silicon wafer during thermal oxidation J. Phys. Chem. C 117 21577–82 5. Conclusions Important differences in third order optical response exhibited by different ion-implanted distributions containing Au-NPs and Si-QDs are presented. Nano- and femto-second experi- ments were performed in order to evaluate the nonlinear optical absorption and nonlinear optical refraction. An enhancement resulting from the additive response of the studied distributions in the nanosecond regime was found; while on the contrast a diminishment in resulting addition for the same configuration for femtosecond pulses was detected. Apparently these nonlinear refractive phenomena can be explained considering the different physical mechanisms corresponding to the electronic polarization and thermal effects derived in each temporal regime. In regards to the 7

9. Nanotechnology 26 (2015) 295701 C Torres-Torres et al [9] Ji Q, Hill J P and Ariga K 2013 Shell-adjustable hollow ‘soft’ silica spheres as a support for gold nanoparticles J. Mater. Chem. A 1 3600–6 [10] Mangold M A, Calame M, Mayor M and Holleitner A W 2011 Resonant photoconductance of molecular junctions formed in gold nanoparticle arrays J. Am. Chem. Soc. 133 12185–91 [11] Goffard J, Gérard D, Miska P, Baudrion A L, Deturche R and Plain J 2013 Plasmonic engineering of spontaneous emission from silicon nanocrystals Sci. Rep. 3 2672 [12] Mertens H, Biteen J S, Atwater H A and Polman A 2006 Polarization-selective plasmon-enhanced silicon quantum- dot luminescence Nano Lett. 6 2622–5 [13] Zhang P, Zhang X, Lu P, Xu J, Xu X, Li W and Chen K 2014 Interface state-related linear and nonlinear optical properties of nanocrystalline Si/SiO2 multilayers Appl. Surf. Sci. 292 262–6 [14] Harun N A, Benning M J, Horrocks B R and Fulton D A 2013 Gold nanoparticle-enhanced luminescence of silicon quantum dots co-encapsulated in polymer nanoparticles Nanoscale 5 3817–27 [15] Shi Y, Zhang H, Yue Z, Zhang Z, Teng K S, Li M J, Yi C and Yang M 2013 Coupling gold nanoparticles to silica nanoparticles through disulfide bonds for glutathione detection Nanotechnology 24 375501 [16] Tchebotarevaa A L, de Dooda M J A, Biteenb J S, Atwaterb H A and Polman A 2005 Quenching of Si nanocrystal photoluminescence by doping with gold or phosphorous J. Lumin. 114 137–44 [17] Mikhaylov A N, Kostyuk A B, Korolev D S, Zhavoronkov I Y, Chugrov I A, Belov A I, Burdov V A, Ershov A V and Tetelbaum D I 2012 Formation of gold nanoparticles in single layer and multi layer ensembles of light emitting silicon nanocrystals using ion implantation Bull. Russ. Acad. Sci. Phys. 76 214–7 [18] Tamayo-Rivera L, Fernández-Hernández R C, Rodríguez-Fernández L, Rangel-Rojo R, Oliver A and Reyes-Esqueda J A 2011 Wavelength-varying third-order nonlinear optical response of Ag nanoparticles-Si quantum dots integrated plasmonic system Opt. Mater. Express 1 980–9 [19] López-Suárez A, Torres-Torres C, Can-Uc B, Rangel-Rojo R, Valencia C E and Oliver A 2015 Third order nonlinear optical properties exhibited by a bilayer configuration of silver nanoparticles integrated to silicon nanocrystals embedded in ion-implanted silica J. Opt. Soc. Am. B 32 805–11 [20] Torres-Torres C, Tamayo-Rivera L, Rangel-Rojo R, Torres-Martínez R, Silva-Pereyra H G, Reyes-Esqueda J A, Rodríguez-Fernández L, Crespo-Sosa A, Cheang-Wong J C and Oliver A 2011 Ultrafast optical phase modulation with metallic nanoparticles in ion-implanted bilayer silica Nanotechnology 22 355710 [21] Mayer M 2003 SIMNRA user’s guide, version 6.0, Max Planck-Institute für Plasmaphysik [22] Yu B L, Bykov A B, Qiu T, Ho P P, Alfano R R and Borrelli N 2003 Femtosecond optical Kerr shutter using lead–bismuth– gallium oxide glass Opt. Commun. 215 407–11 [23] Sheik-Bahae M, Said A A, Wei T, Hagan D J and Stryland E W V 1990 Sensitive measurement of optical nonlinearities using a single beam IEEE J. Quantum Electron. 26 760–9 [24] Muñoz-César J C, Torres-Torres C, Moreno-Valenzuela J, Torres-Torres D, Urriolagoitia-Sosa G and Trejo-Valdez M 2013 Identification of inhomogenous optical absorptive response by chaotic photonic signals in Au nanoparticles Meas. Sci. Technol. 24 035603 [25] Trejo-Valdez M, Torres-Martínez R, Peréa-López N, Santiago-Jacinto P and Torres-Torres C Contribution of the two-photon absorption to the third order nonlinearity of Au nanoparticles embedded in TiO2films and in ethanol suspension J. Phys. Chem. C 114 10108–13 [26] Boyd R W 1992 Nonlinear Optics (San Diego, CA: Academic Press) [27] Magruder R H III, Robinson S, Smith C, Meldrum A and Haglund R F Jr 2012 Fabricating a dichroic plasmonic mirror in fused silica by dual-ion implantation Appl. Phys. A 107 935–42 [28] Trejo-Luna R, de la Vega L R, Rickards J, Falcony C and Jergel M 2001 9 MeV ion implantation into Ti and Ti-6AI- 4V J. Mater. Sci. 36 503–10 [29] Ito M, Imakita K, Fujii M and Hayashi S 2010 Nonlinear optical properties of silicon nanoclusters/nanocrystals doped SiO2films: annealing temperature dependence J. Appl. Phys. 108 063512 [30] López-Suárez A, Torres-Torres C, Rangel-Rojo R, Reyes-Esqueda J A, Santana G, Ortíz A, Alonso J C and Oliver A 2009 Modification of the nonlinear optical absorption and optical Kerr response exhibited by nc-Si embedded in a silicon-nitride film Opt. Express 17 10056–68 [31] Mochizuki Y and Agren H 2001 Polarization of silicon clusters Chem. Phys. Lett. 336 451–6 [32] Sylvester-Hvid K O, Åstrand P O, Ratner M A and Mikkelsen K V 1999 Frequency-dependent molecular polarizability and refractive index: Are substituent contributions additive? J. Phys. Chem. A 103 1818–21 [33] Liu Y and Daum P H 2008 Relationship of refractive index to mass density and self-consistency of mixing rules for multicomponent mixtures like ambient aerosols Aerosol Sci. 39 974–86 [34] Maroulis G, Begué D and Pouchan C 2003 Accurate dipole polarizabilities of small silicon clusters from ab initio and density functional theory calculations J. Chem. Phys. 119 794 8