saturable and reverse-saturable absorption - OSA Publishing

Third-order nonlinear optical properties of colloidal Au nanorods systems: saturable and reverse-saturable absorption E. V. García-Ramírez,1 S. Almaguer-Valenzuela,2 O. Sánchez-Dena,1 O. BaldovinoPantaleón,2,* and J. A. Reyes-Esqueda1,3 2

1 Instituto de Física, Universidad Nacional Autónoma de México, D.F.04510, Mexico UAM Reynosa – Rodhe, Universidad Autónoma de Tamaulipas, Carr. Reynosa-San Fernando S/N, Reynosa, Tamaulipas 88779, Mexico 3 [email protected] * [email protected]

Abstract: In this work, we present a study of the nonlinear absorption properties from different gold nanorod (NR) systems in aqueous suspension. The NRs were obtained with the bottom-up protocol by the seed-mediated growth method (SMG), using Ag+ ions at different concentrations, and CTAB as surfactant. By using this method, aspect ratios between 2 and 5 were obtained. The transverse surface plasmons (TSP) are located between 514 – 535 nm, while the longitudinal surface plasmons (LSP) are between 639 – 921 nm, for the different samples studied. The Zscan technique was implemented for open (OA) and closed (CA) aperture at 532 and 1064 nm, with laser pulses of 26 ps, for vertical and horizontal polarizations, with respect to the incidence plane (horizontal). At 532 nm all samples showed saturable absorption (SA), while for samples with LSP near 1064 nm, such effect was observed only at low-energy pulse experimental conditions. In the high-energy pulse regime, an apparent reverse-saturable absorption (RSA) was observed for both wavelengths. However for 532 nm, it was possible to determine that this effect results from structural changes in the samples, which are manifested through the behavior of nonlinear absorption and refraction curves. These results were used to determine the irradiances to which NRs can be modified by photodegradation. ©2015 Optical Society of America OCIS codes: (190.0190) Nonlinear optics; (160.1190) Anisotropic optical materials; (160.4236) Nanomaterials.

References and links 1. 2. 3. 4. 5. 6. 7. 8.

X. Huang, S. Neretina, and M. A. El-Sayed, “Gold Nanorods: from synthesis and properties to biological and biomedical applications,” Adv. Mater. 21(48), 4880–4910 (2009). S. A. Maier, P. G. Kik, H. A. Atwater, S. Meltzer, E. Harel, B. E. Koel, and A. A. G. Requicha, “Local detection of electromagnetic energy transport below the diffraction limit in metal nanoparticle plasmon waveguides,” Nat. Mater. 2(4), 229–232 (2003). L. François, M. Mostafavi, J. Belloni, J. F. Delouis, J. Delaire, and P. Feneyrou, “Optical limitation induced by gold clusters. 1. size effect,” J. Phys. Chem. B 104(26), 6133–6137 (2000). R. West, Y. Wang, and T. Goodson III, “Nonlinear absorption properties in novel gold nanostructured topologies,” J. Phys. Chem. B 107(15), 3419–3426 (2003). A. S. Thakor, J. Jokerst, C. Zavaleta, T. F. Massoud, and S. S. Gambhir, “Gold nanoparticles: a revival in precious metal administration to patients,” Nano Lett. 11(10), 4029–4036 (2011). S. S. Chang, C. W. Shih, C. D. Chen, W. C. Lai, and C. R. C. Wang, “The shape transition of gold nanorods,” Lagmuir 15(3), 701–709 (1999). B. M. I. van der Zande, M. R. Böhmer, L. G. J. Fokkink, and C. Schönenberger, “Aqueous gold sols of rodshaped particles,” J. Phys. Chem. B 101(6), 852–854 (1997). S. R. Nicewarner-Pena, R. G. Freeman, B. D. Reiss, L. He, D. J. Peña, I. D. Walton, R. Cromer, C. D. Keating, and M. J. Natan, “Submicrometer metallic barcodes,” Science 294(5540), 137–141 (2001).

#251135 © 2016 OSA

Received 30 Sep 2015; revised 11 Nov 2015; accepted 12 Nov 2015; published 10 Dec 2015 25 Jan 2016 | Vol. 24, No. 2 | DOI:10.1364/OE.24.00A154 | OPTICS EXPRESS A154

9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

31. 32. 33.

Y. Y. Yun, S. S. Chang, C. L. Lee, and C. R. C. Wang, “Gold nanorods: electrochemical synthesis and optical properties,” J. Phys. Chem. B 101(34), 6661–6664 (1997). K. Esumi, K. Matsuhisa, and K. Torigoe, “Preparation of rodlike gold particles by uv irradiation using cationic micelles as a template,” Lagmuir 11(9), 3285–3287 (1995). N. R. Jana, L. Gearheart, and C. J. Murphy, “Wet chemical synthesis of high aspect ratio cylindrical gold nanorods,” J. Phys. Chem. B 105(19), 4065–4067 (2001). K. S. Lee and M. A. El-Sayed, “Dependence of the enhanced optical scattering efficiency relative to that of absorption for gold metal nanorods on aspect ratio, size, end-cap shape, and medium refractive index,” J. Phys. Chem. B 109(43), 20331–20338 (2005). A. Brioude, X. C. Jiang, and M. P. Pileni, “Optical properties of gold nanorods: DDA simulations supported by experiments,” J. Phys. Chem. B 109(27), 13138–13142 (2005). S. Link and M. A. El-Sayed, “Simulation of the optical absorption spectra of gold nanorods as a function of their aspect ratio and the effect of the medium dielectric constant,” J. Phys. Chem. B 109(20), 10531–10532 (2005). S. Link and M. A. El-Sayed, “Shape and size dependence of radiative, non-radiative and photohermal properties of gold nanocrystals,” Int. Rev. Phys. Chem. 19(3), 409–453 (2000). B. N. Khlebtsov and N. G. Khlebtsov, “Multipole plasmons in metal nanorods: scaling properties and dependence on particle size, shape, orientation, and dielectric environment,” J. Phys. Chem. C 111(31), 11516– 11527 (2007). P. K. Jain, K. S. Lee, I. H. El-Sayed, and M. A. El-Sayed, “Calculated absorption and scattering properties of gold nanoparticles of different size, shape, and composition: applications in biological imaging and biomedicine,” J. Phys. Chem. B 110(14), 7238–7248 (2006). Y. Tsutsui, T. Hayakawa, G. Kawamura, and M. Nogami, “Tuned longitudinal surface plasmon resonance and third-order nonlinear optical properties of gold nanorods,” Nanotechnology 22(27), 275203 (2011). J. Li, S. Liu, Y. Liu, F. Zhou, and Z.-Y. Li, “Anisotropic and enhanced absorptive nonlinearities in a macroscopic film induced by aligned gold nanorods,” Appl. Phys. Lett. 96(26), 263103 (2010). H. I. Elim, J. Yang, J.-Y. Lee, J. Mi, and W. Ji, “Observation of saturable and reverse-saturable absorption at longitudinal surface plasmon resonance in gold nanorods,” Appl. Phys. Lett. 88(8), 083107 (2006). L. De Boni, E. L. Wood, C. Toro, and F. E. Hernandez, “Optical saturable absorption in gold nanoparticles,” Plasmonics 3(4), 171–176 (2008). D. Manzani, J. M. P. Almeida, M. Napoli, L. De Boni, M. Nalin, C. R. M. Afonso, S. J. L. Ribeiro, and C. R. Mendonça, “Nonlinear optical properties of tungsten lead–pyrophosphate glasses containing metallic copper nanoparticles,” Plasmonics 8(4), 1667–1674 (2013). J. Olesiak-Banska, M. Gordel, R. Kolkowski, K. Matczyszyn, and M. Samoc, “Third-order nonlinear optical properties of colloidal gold nanorods,” J. Phys. Chem. C 116(25), 13731–13737 (2012). N. R. Jana, “Gram-scale synthesis of soluble, near-monodisperse gold nanorods and other anisotropic nanoparticles,” Small 1(8-9), 875–882 (2005). B. Nikoobakht and M. A. El-Sayed, “Preparation and growth mechanism of gold nanorods using seed-mediated growth method,” Chem. Mater. 15(10), 1957–1962 (2003). H. L. Fang and R. L. S. Wafford, “The thermal lens in absorption spectroscopy,” in Ultrasensitive Laser Spectroscopy, D. Kliger ed. (Academic Press, 1983), pp 175–232. J. M. Khosrofian and B. A. Garetz, “Measurement of a Gaussian laser beam diameter through the direct inversion of knife-edge data,” Appl. Opt. 22(21), 3406–3410 (1983). M. Sheik-Bahae, A. A. Said, T. H. Wei, D. J. Hagan, and E. W. Van Stryland, “Sensitive measurement of optical nonlinearities using a single beam,” IEEE J. Quantum Electron. 26(4), 760–769 (1990). E. V. García Ramírez, M. L. Arroyo Carrasco, M. M. Méndez Otero, E. Reynoso Lara, S. Chávez Cerda, and M. D. Iturbe Castillo, “Z-scan and spatial self-phase modulation of a Gaussian beam in a thin nonlocal nonlinear media,” J. Opt. 13(8), 085203 (2011). J. A. Reyes-Esqueda, V. Rodríguez-Iglesias, H.-G. Silva-Pereyra, C. Torres-Torres, A.-L. Santiago-Ramírez, J. C. Cheang-Wong, A. Crespo-Sosa, L. Rodríguez-Fernández, A. López-Suárez, and A. Oliver, “Anisotropic linear and nonlinear optical properties from anisotropy-controlled metallic nanocomposites,” Opt. Express 17(15), 12849–12868 (2009). J. M. Gómez-Cervantes and J. A. Reyes-Esqueda, “Presence of Fano-like resonances into the birefringence of plasmonic materials,” in preparation. Y. B. Band, D. J. Harter, and R. Bavli, “Optical pulse compressor composed of saturable and reverse saturable absorbers,” Chem. Phys. Lett. 126(3), 280–284 (1986). G. I. Stegeman, “Material figures of merit and implications to all-optical waveguide switching,” Proc. SPIE 1852, 75–89 (1993).

1. Introduction In recent years, metal nanoparticles (NPs) have been intensively studied due to their electrical, magnetic and optical properties. In particular, wide attention has been dedicated to NPs made of alkali and noble metals, such as copper, silver, and gold. These systems have a broad absorption band in the visible region of the electromagnetic spectrum, called surface plasmon resonance (SPR), which is responsible of confining resonant photons in such a manner as to induce coherent surface plasmon oscillation of their conduction band electrons [1]. Scattering,

#251135 © 2016 OSA


strong absorption and local-field enhancement occurring at the SPR manifest in a large optical polarization associated with the collective electron oscillations. For this reason the nonlinear optical properties of metal nanorods or nanoparticles are being studied for their use in applications like plasmon waveguide, sensor protection, optical limiting device and medical therapies [2–5]. In general, optical response of metal NPs can be tuned by controlling their size, shape and environment [6]. But, shape and crystallographic facets are also key factors to determine the surface and catalytic process of the NPs. For these reasons, much research is focused on finding synthesis techniques able to control the morphology of the NPs. While the chemical composition of NPs is important, their morphology as well as their colloidal colloidal characteristics are determinant for optical studies. Nowadays, a significant number of research works is directed to one-dimensional NPs such as nanorods (NRs). Metal nanorods/wires have been synthesized by different approaches, using various methods such as templating [7,8], electrochemistry [6,9], photochemistry [10], and seed-mediated growth methods, using Ag+ ions and surfactants like hexadecyltrimethylammonium bromide (CTAB) [11]. In comparison with nanoshells and nanospheres, colloidal NRs offer better tunability of surface plasmon resonance (SPR) bands in a wide range of optical frequencies [12]. In the case of gold NRs, the excitation along the short axis induces an absorption band in the visible region, at wavelengths close to where spherical Au NPs usually absorb, while excitation along the long axis induces a stronger absorption band centered at longer wavelengths. Both modes of oscillation, namely longitudinal surface plasmon resonance (L-SPR), and transverse surface plasmon resonance (T-SPR), introduce additional variables upon which NRs optical response is sensitive to, like incident frequency and polarization state of light. The transverse band is not very sensitive to the size of NRs, while the longitudinal band is red-shifted largely from the visible to the nearinfrared region, accordingly to increments on the NRs aspect ratio. This behavior is explained by Gans theory, which describes the optical properties of ellipsoidal particles using a dipole approximation. This has been related to the fact that, in aqueous solution, the L-SPR absorption maximum λmax, is linearly proportional to the aspect ratio R by the empiric relationship λmax = 95R + 420 [13,14]. To control the optical properties and design colloidal NRs with specific applications, a systematic and quantitative description of interactions between NRs and light is necessary. Optical properties of gold NRs at low intensity have been widely investigated, including the dependence of light scattering, absorption in function of size, aspect ratio, among others [12,15–17]. Third-order nonlinear optical response of NRs, saturable absorption and reversesaturable absorption as function of intensity, has been previously reported [4,18–23]. However, only the particular experimental condition of high-energy pulses has been widely studied, while the behavior of NRs under short-energy pulses has not been well determined, yet. Therefore, a systematic study of the nonlinear optical response in colloidal NRs is transcendental so that a proper design can be fulfilled in order to exploit a specific application such as optical limiting devices [3,4]. In this work, we present a systematic study of the third-order nonlinear optical properties of four different colloidal Au NRs systems. The NRs were prepared with the bottom-up method by seed growth [11], using Ag+ ions and CTAB as surfactant, obtaining an aspect ratio between 3 and 5. The third-order nonlinear optical properties were measured by implementing the Z-scan technique, using both experimental configurations, that is, open (OA) and closed aperture (CA) [24]. Excitation of NRs was performed with pulsed light of 26 ps, with a repetition rate of 10 Hz, at wavelengths of 532 and 1064 nm. The obtained results show a saturable absorption effect for all samples at both excitation wavelengths, and a small nonlinear refraction for some of them. Small changes in the nonlinear absorption coefficients were obtained modifying the incident polarization of light. The samples were studied for several irradiances; in the case of 1064 nm, increasing the irradiance gives place to a decreasing effect regarding the nonlinear absorption Z-scan characteristic curve, which in turn, appoints for a possible presence of the reverse-saturable absorption process. For 532 nm, the increment of the irradiance causes structural changes in the samples, as shown by electron #251135 © 2016 OSA


micrographs and given that nonlinear absorption and refraction Z-scan characteristic curves are significatively modified. 2. Experimental methods 2.1 Sample preparation In this work, we use the well-established aspect control and silver-assisted seeded growth protocol that provide monodisperse gold nanorods in high yield relative to other shapes. However, the overall yields are relatively low as nearly 15%. In the preparation of Au NRs by the seed-mediated growth procedure, we use small NRs seeds (1.5 nm) stabilized with CTAB surfactant, which grow with the reduction of Au(III) salt to Au(0) induced mainly by ascorbic acid, with or without the presence of silver ions. However, a small amount of silver ions could be used to control the aspect ratio of NRs ranging from 1.5 to 5. The growth mechanism of silver-assisted Au NRs is still a matter of debate [24]. Seed solution: Colloidal Au NRs were prepared as follows: 2.5 mL of HAuCl4 solution (9 × 10−4 M), 2.5 ml of deionized water, and 5 mL of CTAB (hexadecyltrimethylammonium bromide) solution (0.2 M) were stirred together and heated to constant temperature of 45°C. 600μl of ice-cold NaBH4 solution (0.01 M) was added and stirred for 2 min at 25°C, obtaining a brown solution. Growth solution: 150 mL of CTAB solution and 150 mL of HAuCl4 solution were mixed, and 890 μL of HCl solution (0.10 M) was added. To gain a better growth control on NRs size, four different samples were produced to be analyzed in this work by addition to the growth solution of 1, 2, 3 and 4 mL of AgNO3 solution (4 × 10−3 M), respectively. Then, to each of one, 1250 μL of ascorbic acid in aqueous solution (7.8 × 10−2 M) was aggregated, obtaining colorless solutions. Finally, 200 μL of the seed solution were incorporated to all and each of them. All this mix was maintained at temperature of 40°C with gentle and constant stirring overnight. Finally, for all colloidal systems, separation of NRs from spherical nanoparticles was carried out by centrifugation (at least three times at 6 000 rpm, 1 hour). The morphologies of the Au NRs were determined by means of transmission electron microscopy (TEM) using a JEOL 2010 FE-TEM. Figure 1 shows the TEM images corresponding to the four samples analyzed after centrifugation. Typically, by using this synthesis method, approximately 1% of spherical NPs remain into the colloid system. On the other hand, samples with high content of silver ion in the growth solution not always increase the Au NRs aspect ratio due to interaction with the bromide counter ion of the surfactant monomers. Consequently, a larger fraction of nanospheres may be present, even up to 5% [25]. Samples 1 to 3 mL have approximately the same concentration of Au NRs (98-99%) with high monodispersity. On the contrary, sample 4 mL, due to the Ag ion effect, was unable to be satisfactorily purified by the same method, and the Au NRs concentration reduced to 95%, with the polydispersity being slightly increased. The micrographs were analyzed by Digital Micrograph software. Figure 2 shows the aspect ratio histograms corresponding to the statistical analysis. According to them, the length of the NRs is between 25 and 60 nm, while the diameter goes between 11 and 12 nm, approximately. The proper information obtained from these two figures is presented in Table 1.

#251135 © 2016 OSA


Fig. 1. TEM images of Au nanorods corresponding to sample a) 1 ml, b) 2 ml, c) 3 ml, d) 4 ml of AgNO3 concentration, bar scale of 20 nm. Table 1. Dimensions of Au NRs. λmax (nm)

11.47 ± 3.09

Experimental Aspect Ratio Rexp 2.25

626.78

2.17

37.46 ± 4.98

11.65 ± 2.64

3.21

722.04

3.17

3 ml

54.13 ± 6.72

12.42 ± 3.77

4.35

836.20

4.38

4 ml

60.43 ± 5.98

12.38 ± 0.99

4.88

897.00

5.02

Sample

Length (nm)

Diameter (nm)

1 ml

25.82 ± 5.35

2 ml

#251135 © 2016 OSA

Theoretical Aspect Ratio

Rth = (λmax − 420) / 95


Fig. 2. Histograms corresponding to the Au NRs TEM micrographs. They show the distribution of Au NRs’ aspect ratio as a function of the different concentrations of AgNO3.

2.2 Optical measurements Linear optical absorption measurements were performed with an Ocean Optics USB2000 + UV–visible spectrophotometer. The absorption spectra of Au NRs are characterized by a dominant L-SPR band, corresponding to longitudinal resonance, and a weaker transverse resonance T-SPR, corresponding to diametric resonance. Figure 3 shows the absorption spectra for all the samples. From this figure, it can be observed the presence of two SPR peaks in all cases. The T-SPR is located around of 514-535 nm, while the L-SPR goes from 626 to 900 nm. The redshift of longitudinal resonance with concentration makes evident its evolution with aspect ratio. The linear absorption coefficients were estimated by the expression Pt = P0 exp( −α 0l ) [26], with Pt and P0 the transmitted and incident powers respectively, α0 the linear absorption coefficient and l the thickness of the sample.

#251135 © 2016 OSA


Fig. 3. Optical absorption spectra for Au NRs at different concentrations of AgNO3.

Z-scan technique at 532 nm and 1064 nm was used to study the third-order NLO properties of these colloids, which were scanned around the waist beam ω0 along the optical axis (z direction). As light source was used a PL2143A laser system by EKSPLA, featuring 26 ps pulses with a 1-10 Hz repetition rate. The experiments reported were realized at a frequency of 10 Hz, with vertical and horizontal polarization, with respect to the incidence plane (horizontal), for several irradiances. A quartz cell of 1 mm thickness was used in the experiments. The reproducibility of results was verified, however damage was found at 532 nm for irradiances approximately larger than I0 = 2.551 GW/cm2, for 1 and 2 ml samples, and 3.811 GW/cm2 for 3 and 4 ml samples.. The laser beam was focused by a lens with a focal length of 400 mm, where the radius of the waist beam for each wavelength was determined by the knife-edge method [27]. The obtained values were ω0 = 35.20 ± 0.97 μm, and ω0 = 60.24 ± 0.70 μm for 532 and 1064 nm, respectively. The corresponding calculated Rayleigh lengths were Z0 = 0.732 cm, and Z0 = 1.071 cm approximately, so that the thin medium condition is fulfilled. The transmitted beams for CA and OA for Z-scan technique were measured with Thorlabs DET 10A fast photodiodes, where an aperture of 25 mm in diameter was used for the CA measurements. 3. Results and discussion 3.1 Low-energy pulse The characteristic OA and CA Z-scan curves are presented in Figs. 4 and 5. Figure 4 corresponds to results obtained using an incident wavelength of 532 nm, for a peak irradiance at the focus of I0 = 0.400 GW/cm2. a) and c) correspond to OA experiments for horizontal and vertical polarizations, respectively. b) and d) show CA results, again, for horizontal and vertical polarizations, respectively. It is clear from Fig. 4(a), first, that a negative, saturable absorption (increased transmittance) is present for all the samples; second, that nonlinear absorption increases with AgNO3 concentration, that is, with aspect ratio (or, more importantly, with linear optical absorption, since changes in aspect ratio are mostly reflected in longitudinal surface plasmon resonance shifts); however this behavior is not preserved for vertical polarization, see Fig. 4(c). In this case, sample with 2 ml of concentration of AgNO3 presents a higher response than the other samples. Regarding the nonlinear refraction response, it does not exhibit a clear trend with AgNO3 concentration. The curves show positive or negative responses depending on the sample, for both polarizations. This erratic response is clearly due partially to the proximity of 532 nm to the respective plasmon resonance. For this reason it was not possible to fit the CA curves by using the Sheik-Bahae theory [28]; instead, only some curves could be fitted using the phenomenological model proposed by García-Ramírez et al. [29].

#251135 © 2016 OSA


Fig. 4. NLO responses at λ = 532 nm for colloidal Au NRs. a) Nonlinear absorption and b) nonlinear refraction for horizontal polarization. c) and d) similar to a) and b) for vertical polarization.

The OA Z-scan curves were fitted using the relationships given by Sheik-Bahae et. al. [28]. The OA experimental curves were fitted by using the equation: ∞

[−q0 ( z ,0)]m

m =0

(m + 1)3/2

T ( z , S = 1) = 

which is valid if the condition q0 =

β I 0 Leff 1 + ( z / z0 ) 2

,

< 1 is fulfilled, with Leff =

(1)

1 − e −α L

α

and

L the thickness of sample. The β coefficients corresponding to 532 nm, for several irradiances, for horizontal and vertical polarizations, are reported in Table 2. It is evident how, for each AgNO3 concentration, this coefficient decreases in magnitude as irradiance increases. On the other hand, as mentioned above, this coefficients increases in magnitude with AgNO3 concentration, at least clearly for horizontal polarization and for the smallest irradiance. This is clear consequence of the increment in resonance, that is, of the increment with AgNO3 concentration of the linear absorption coefficient, as indicated by α0 values in Table 1. Experiments realized at larger irradiances are reported in section 3.2.

#251135 © 2016 OSA


Table 2. Nonlinear coefficients β for λ = 532 nm.

Sample −1

α0 (cm )

1 ml

2 ml

3 ml

4 ml

11.013 ± 0.202

11.304 ± 0.183

13.941 ± 0.251

17.680 ± 0.314

β × 10−8 (cm/W) I0(GW/cm2)

Horizontal polarization

0.400

−5.339 ± 0.010

−6.513 ± 0.011

−7.478 ± 0.009

−10.19 ± 0.014

0.547

−5.064 ± 0.014

−4.991 ± 0.013

−5.303 ± 0.011

−7.459 ± 0.017

0.673

−4.778 ± 0.012

−4.512 ± 0.015

−4.289 ± 0.008

−6.190 ± 0.012

0.733

−4.523 ± 0.016

−4.356 ± 0.012

−4.185 ± 0.013

−5.712 ± 0.010

Vertical polarization 0.400

−5.225 ± 0.013

−7.063 ± 0.017

−5.308 ± 0.009

−5.762 ± 0.016

0.547

−5.060 ± 0.015

−6.215 ± 0.016

−5.483 ± 0.013

−5.931 ± 0.014

0.673

−5.052 ± 0.011

−4.743 ± 0.015

−4.156 ± 0.011

−5.047 ± 0.012

0.733

−4.401 ± 0.012

−4.923 ± 0.017

−3.446 ± 0.009

−5.521 ± 0.010

The samples were also analyzed at 1064 nm. Figure 5 shows absorption and refraction nonlinear curves at I0 = 2.9657 GW/cm2, for vertical and horizontal polarization. In this case, samples 3 and 4 ml exhibit a better absorptive response compared to 1 and 2 ml samples, while their refractive response shows a wide peak and valley for vertical case. However, it can be observed in Fig. 5(b) that the response does not reach the linear regimen at the end of zscanning. Several experiments were realized under the same conditions, obtaining similar results. Figure 5(d) shows the refractive response for horizontal polarization. It can be seen valley suppression for samples 3 and 4 ml, and the null response for samples 1 and 2 ml. It is important to mention that the nonlinear absorption response is considerably wide.

#251135 © 2016 OSA


Fig. 5. NLO responses at λ = 1064 nm for colloidal Au NRs. a) Nonlinear absorption and b) nonlinear refraction for horizontal polarization. c) and d) similar to a) and b) for vertical polarization.

Nonlinear absorption coefficients obtained at several irradiances are reported in Table 3. Sample 1 ml did not present any response for all the irradiances when using horizontal polarization. Table 3. Nonlinear coefficients β for λ = 1064 nm.

Sample

1 ml

2 ml

α0(cm−1)

0.309 ± 0.090

0.203 ± 0.099

3 ml

4 ml

2.052 ± 0.115

4.098 ± 0.124

−9

β × 10 (cm/W) I0(GW/cm2)

Horizontal polarization

2.965

—

−0.383 ± 0.010

−0.839 ± 0.017

−2.832 ± 0.019

4.750

—

—

−0.625 ± 0.015

−1.356 ± 0.017

5.812

—

—

—

−0.792 ± 0.010

Vertical polarization 2.965

−0.544 ± 0.010

−0.962 ± 0.011

−1.628 ± 0.015

−5.092 ± 0.012

4.750

—

−0.573 ± 0.099

−0.636 ± 0.013

−1.253 ± 0.010

5.812

—

—

—

−1.204 ± 0.012

Regarding the results for different polarizations, it can be observed how they are quite different for the smallest irradiance and the largest nanorods, for both wavelengths. On one hand, this indicates that these differences are not related to a nonlinear optical phenomenon since, when increasing the irradiance, the results become similar. On the other hand, preliminary birefringence measurements performed on these colloids [30] have shown quite #251135 © 2016 OSA


small values for all the samples, but also the presence of a Fano-like resonance in the birefringence for the longitudinal plasmon resonance of the 4 ml sample [30,31]. This last result is an indication that, possibly because of the size of the NRs, there is a preferential orientation of the NRs into the colloid, although quite small, which will be supported by the different β values shown in Tables 2 and 3 at low irradiances, for that sample. 3.2 High-energy pulse These samples were also studied for high irradiances for both wavelengths. Because of the similarity in the behavior of the absorptive response, Fig. 6 shows only the results, at 532 nm, for samples 1 ml and 2 ml. The irradiances values went from I0 = 2.551 GW/cm2 to I0 = 5.215 GW/cm2. As it can be observed for both samples, the NLO response shows a peak and a valley. At first, observing Fig. 6, one could think that, when increasing I0, there could be a competition between SA and RSA such that RSA would dominate from I0>5.215 GW/cm2. However, if that was the case, there should be some symmetry in the curves, the change for one response to the other should have been gradual, and the transmittance minimum should be located at the focal position. None of these three facts happened for these curves. The response corresponding to SA did not decrease when increasing I0, actually, what happened is that, when increasing energy, the valley broadened while the peak moved further to negative values of the z position. The ensemble of these observations raised the importance of guaranteeing the reproducibility of the results as follows.

Fig. 6. Nonlinear absorptive response at several irradiances for 532 nm. Samples a) 1 ml and b) 2 ml.

To analyze the reproducibility or not of the results, the experiments were repeated several times under the same illumination conditions; this can be observed in Fig. 7, where curves corresponding to four scans made immediately one after each other for sample 2 ml are shown. For the smallest I0, I0 = 2.551 GW/cm2, the curves were reproducible after several scans. But, for larger I0s the response varied with each scan, as can be seen in Fig. 7(a). As said before, the curves show very wide valleys with a poor defined transmittance minimum, which cleary differs from a typical RSA response. Also, the fact that the response changed for every scan is a clear indication of a damage being produced on the sample, increasing the asymmetry of the measured curve. Finally, a clear proof of this damage is shown in Fig. 7(b), where it can be seen a curve obtained at a much lower irradiance, I0 = 0.896 GW/cm2, after the fourth scan at I0 = 4.991 GW/cm2. It can be clearly observed that this last curve is very different from the curves obtained at the low-energy pulses regime, Fig. 4.

#251135 © 2016 OSA


Fig. 7. Nonlinear absorption response for sample 2 ml. a) after several measurements at 532 nm with I0 = 4.991 GW/cm2. b) Response at I0 = 0.896 GW/cm2, after of 4th scan at I0 = 4.991 GW/cm2.

After performing these measurements for all the samples, a limit of reproducibility was obtained for each of them. For samples 1 and 2 ml, the limit was I0 = 2.551 GW/cm2, while for samples 3 and 4 ml, this limit was 3.811 GW/cm2. To understand what happened to the samples after irradiating them above these irradiances, sample structure was studied using electron microscopy. Sample 2 ml was studied in Central Laboratory of MicroscopyIFUNAM with JEOL JSM-7800F, Field Emission Scanning Electron Microscope. Micrographs were obtained for this sample before z-scan, and after 4th z-scan, at I0 = 4.991 GW/cm2. Figure 8 shows the structural modifications induced in the sample by large irradiances. Besides of aggregation effects, there is also a clear change in the shape of the NPs, passing from being NRs to show spherical shapes, with diameters around the original large axis of the NRs. It was not possible to obtain higher resolution images since the electron beam started to evidently affect the samples. However TEM microscopy was still feasible; Fig. 9 shows several micrographs taken before and after z-scan. In Figs. 9(d)-9(f), it can be observed the trend of the particles to form clusters after de 4th z-scan. Moreover, it is possible to observe the different morphologies present in the sample. It is important to mention that the analyzed sample’s amount and the preparation method for microscopy of the samples were the same before and after the z-scan measurements. In Fig. 10, it is shown the absorption spectra corresponding to sample 2 ml before and after z-scan, the spectrum after the 4th zscan shows a decrease relative to the spectrum of sample before experiments.

Fig. 8. Electron microscopy analysis of 2 ml sample, a) before z-scan at 532 nm, and b) after of 4th scan at I0 = 4.991 GW/cm2.

#251135 © 2016 OSA


Fig. 9. TEM microscopy of sample 2 ml, a), b), c) before z-scan measurements at scale 0.1 μm; d), e) and f) after z-scan. e) and f) with scale 50 nm.

Fig. 10. Optical absorption spectra for sample 2 ml before and after z-scan

Similar experiments were made at 1064 nm. Results shown in Fig. 11 correspond to sample 4 ml for several irradiances. In this case, it can be observed the decrease of NLA with irradiance increment. Actually, for I0 = 20.39 GW/cm2, it starts to be evident a possible RSA, which is a typical behavior for optical limiting phenomena. The origin of the RSA is attributed to the formation of strong ligth scattering centers due to the vaporization of the initial particles induced by the laser pulse [3]. The transformation from saturable to reverse absorption is an interesting effect that can be used for optical pulse compressor, optical switching and laser pulse narrowing [32]. It is worth remarking how, for 1064 nm, the change from SA to RSA is totally different from what happened at 532 nm, and was discussed above. In this case, the curves are all symmetrical and there is a clear decrease of the transmittance maximum when increasing the irradiance.

#251135 © 2016 OSA


Fig. 11. NLA for sample 4 ml for several irradiances at 1064 nm.

On the other hand, an ideal nonlinear material for an optical switching device should fulfill the following conditions: 1) the effect of linear absorption must be weak compared to the effect of nonlinearity, and 2) the effect of two-photon absorption must be weak compared to the nonlinear effect. These conditions are quantified in terms of the Stegeman figures of merit W (one photon), and T (two photon), respectively [33]:

W=

n2 I s

λα 0

,T = λ

β n2

(2)

where I s = −α 0 / β is the saturation intensity. For a certain material to be used for all-optical switching, it must satisfy both W > 1 and T < 1. In our case, due to the erratic nature of the nonlinear refractive response, it is difficult to obtain n2. However, in the samples illuminated with 1064 nm, RSA occurs at high energies of excitation. Since RSA is due to nonlinear scattering and/or further absorption of excited electrons, this observed RSA postulates these gold nanorods as good candidates for optical limiters. 4. Conclusions Third-order nonlinear optical response was studied for different colloidal Au NR systems, for wavelengths close to transverse and longitudinal SPRs. Saturable absorption was obtained for low irradiances, increasing with AgNO3 concentration. For high irradiances, at 532 nm, an apparent reverse-saturable absorption was obtained. However, further analysis of this effect showed that photodegradation of the NRs is responsible of this change of sign. The irradiances, from which photodegradation of the NRs was observed, were I0 = 2.551 GW/cm2 for samples 1 and 2 ml, and 3.811 GW/cm2 for samples 3 and 4 ml. In the case of 1064 nm, saturable absorption decreases when increasing irradiance, and for irradiances larger than 20.39 GW/cm2, reverse-saturable absorption starts to show. Acknowledgments The authors wish to acknowledge the technical assistance of Diego Quiterio Vargas and Samuel Tehuacanero Núñez. We also acknowledge financial support from CONACyT through grants CB2010-156529 and PROMEP-102491; and from DGAPA through postdoctoral fellowship for E.V.G.R. and grant PAPIIT IT102013.

#251135 © 2016 OSA
