Open Access
Add to BibSonomyAbstract
Silica glasses were pre-implanted with 60 keV Zn ions at different fluences of 1 × 1016 and 1 × 1017 cm−2, respectively, and were then subjected to implantation of 45 keV Cu ions at a fluence of 5 × 1016 cm−2. The formation of metallic nanoparticles (NPs) as well as their optical properties has been studied in details. Our results clearly show that Zn ion preimplantation at a low fluence of 1.0 × 1016 cm–2 can give rise to the formation of large Cu NPs with a double-layer arrangement, which contribute a greatly enhanced surface plasmon resonance (SPR) absorption at about 572 nm. As the Zn ion fluence increases to 1.0 × 1017 cm–2, Cu/Cu–Zn core/shell NPs with a high particle density and a narrow size distribution can be obtained, resulting in a strong and broad SPR absorption band around 528 nm. Besides, the dually implanted samples also exhibit excellent third-order nonlinear optical properties comparing with the Cu solely implanted sample. The possible mechanisms for the nucleation and growth of NPs as well as for the enhancements of linear and nonlinear optical properties have been discussed.
© 2015 Optical Society of America
1. Introduction
Nanometer-sized particles have received increasing attention in recent years because of their peculiar physicochemical properties in comparison with their bulk counterparts. Especially, metallic nanoparticles (NPs) embedded in dielectric matrices are specific of interest due to localized surface plasmon resonance (SPR) which has been proved to be responsible for the dramatically enhanced linear and nonlinear optical properties with promising applications in catalysis, ultrafast optical switching, and solar cells, etc [1–4]. The localized SPR, which can significantly enhance the local electric field, is a collective oscillation of conduction electrons stimulated by incident light at the interface between a metal (e.g., Au, Ag, Cu) and a dielectric. Its wavelength position in the optical spectrum is a sort of fingerprint for each metallic nanocomposite. Nowadays, a variety of methods including sol–gel synthesis [5], ion exchange [6], and ion implantation [7] have been well developed to synthesize such SPR-based nanocomposites. Among these synthesis methods, ion implantation presents its own advantages: it can introduce desired amount of the guest phase into a host matrix discarding the equilibrium limit of solubility, and provides controllable concentration and depth distributions of NPs in the substrate by precisely selecting ion fluence and energy [7–9]. In addition, dual implantation with two different metallic ions can be utilized to synthesize different types of nanostructures, e.g., binary alloy and core-shell NPs, which can produce quite distinctive optical responses [10,11].
Cu–Zn is a typical representative of the Hume-Rothery alloy system, displaying a sequence of phases along an alloy composition [12]. By changing the ratio of these two metals, the SPR absorption of Cu–Zn bimetallic NPs is expected to be modulated in a wide wavelength range. Besides, because of the lower electronegative value of Zn (1.65) than that of Cu (1.90) [13], preferential oxidation of Zn component in alloys could take place. According to this principle, it is possible to fabricate Cu–Zn/ZnO core/shell NPs [14,15]. Owing to the diversities of phase compositions and structures, Cu–Zn bimetallic NPs have exhibited many intriguing physicochemical characteristics, such as antimicrobial, anticorrosion, and catalytic properties, etc [16–18]. In view of these, we recently explored the synthesis of Cu–Zn NPs in amorphous SiO2 (a-SiO2) glasses by dual implantation with first Cu and then Zn ions and evidenced that the SPR absorption of implantation-synthesized Cu NPs can be remarkably enhanced by post Zn ion implantation to a low fluence of 1 × 1016 cm−2, whereas a strong SPR absorption band induced by Cu–Zn alloy NPs would appear when the post-implanted fluence of Zn ions increased to five folds (5 × 1016 cm−2) or higher [19–21]. Especially, a core-shell nanostructure consisting of Cu core and Zn-related shell can be prepared after 500 °C annealing in a nitrogen atmosphere [21]. However, the nucleation of new phases in the host matrices by dual implantation is sensitive to the implantation sequence. For instance, O. Peña et al have shown that Au/Ag core/shell NPs could be created in SiO2 by keeping the implantation sequence of Ag first and then Au ions, whereas an Au–Ag sequence could only generate Au–Ag alloy NPs [11]. Because our previous works only performed the dual implantation with Cu first and then Zn ions, it is a completely open question what kinds of NPs would be created by a Zn–Cu sequence. Besides, the formation of core-shell NPs in [21]. should be attributed to the thermal-driven diffusion of Zn atoms. Up to now, it is still unknown whether the desired core-shell NPs can be prepared in the host matrix without any post treatment.
In this paper, we mainly focus on the fabrication of metallic NPs in a-SiO2 glasses by 60 keV Zn ion preimplantation to fluences of 1.0 × 1016 and 1.0 × 1017 cm–2, respectively, and then following with 45 keV Cu ion implantation to a fluence of 5.0 × 1016 cm–2. The conditions that allow the formation of core-shell nanostructures by dual implantation with Zn and Cu ions are clearly revealed. Moreover, the structures, linear and nonlinear optical properties of the fabricated NPs are also investigated in detail.
2. Experimental
By using a metal vapor vacuum arc (MEVVA) implanter, optical-grade a-SiO2 slices of 1.0 mm in thickness were sequentially implanted with Zn and Cu ions. For comparison, single implantation of Cu ions and sequential implantation of Cu and Zn ions were also performed. The implantation energies of Zn and Cu ions are 60 and 45 keV, respectively. As a result, four sets of samples were prepared and their names and implantation parameters are listed in Table 1. During implantation, the incident ion beam was injected into the substrate perpendicularly with the flux density kept at ~4 μA/cm2. After implantation, annealing was carried out at different temperatures for 1 hour in a nitrogen atmosphere.
Table 1. Sample names and implantation parameters
To character the shape, size and spatial distributions of the formed NPs as well as their structures, cross-sectional transmission electron microscope (XTEM), selected area electron diffraction (SAED) and energy dispersive X-ray spectroscopy (EDXS) analyses were performed on a Tecnai G2 F20 S-Twin microscope operated at an acceleration voltage of 200 kV and equipped with an EDXS system. Optical absorption responses related to the formation of metallic NPs were recorded in a wavelength range from 200 to 700 nm on a double-beam spectrophotometer (UV–3600). Moreover, the third-order optical nonlinearities of the implanted samples were also measured by using the standard Z-scan method to detect the nonlinear susceptibilities of nanocomposites. The pump source of the optical parametric amplification is a passive-active mode-locked Nd:YAG laser with a repetition rate of 10 Hz (EKSPLA, PL2143B) delivering pulses at 532 nm with a beam-waist radius of about 20 μm. The pulse width and laser intensity were 18 ps and ~13 GW/cm2, respectively.
3. Results and discussion
In the case of dual ion implantation, the concentration profiles of the pre-implanted species as well as the distribution of preimplantation-induced defects in the substrate could strongly affect the deposition of post implants. Therefore, it would be very useful to theoretically evaluate the distributions of implants and preimplantation-induced defects in advance. The concentration profiles of implants can be calculated by the fluence-dependent depth distribution function G(Z) where Z represents the instantaneous depth coordinate. And more detailed calculation method can be found in [21]. As for the depth distribution of defects induced by pre-implanted Zn ions, it can be approximately estimated by the vacancy counts in unit depth produced by per ion, which is simulated by using SRIM 2013 code [22]. Figure 1 gives the simulated results. It can be seen that the vacancy reaches a maximum at the depth of about 30 nm before Cu ion implantation. After Cu ion implantation, the maximum vacancy concentration shifts to the sample surface due to the heavy sputtering loss. And the maximum concentration of Cu atoms appears at about 24 nm depth. Besides, when the fluence of pre-implanted Zn ions increases to 1.0 × 1017 cm−2, post Cu ion implantation would remove almost half of Zn atoms along with shifting the profile of Zn toward the sample surface.
Fig. 1 Simulated depth profiles of Zn-induced vacancies, Zn and Cu implants. They have the same vertical coordinate as shown on the right, but the units of vacancy and implants are (nm•ion)−1 and at.%, respectively.
Figure 2(a) shows the XTEM micrograph of the Cu sample. One can see that NPs with large sizes are mainly aligned in a layer at the depth of about 28 nm, whereas those with small sizes are distributed at both sides of this layer. It is clear that the position of large Cu NPs is slightly deeper than that (~24 nm) revealed in Fig. 1 due to the forward recoils induced by the subsequently incident ions. The SAED pattern inserted in Fig. 2(a) confirms that the formed NPs are face-centered cubic (fcc) copper. And the particle size statistic gives the mean diameter of Cu NPs to be 6.46 ± 3.01 nm (see Fig. 3(a)). As for the Zn1 + Cu sample, the XTEM result (see Fig. 2(b)) reveals that NPs are mainly formed in two layers with the depth ranges around ~20 and ~40 nm, respectively. The SAED pattern inserted in Fig. 2(b) exhibits the same fcc structure and lattice parameter as the copper, demonstrating that the formed NPs are still mainly Cu ones. By comparing the particle size distributions of the Cu and Zn1 + Cu samples (Figs. 3(a) and 3(b)), one can see that the percentage of large NPs in the Zn1 + Cu sample increases while that of small ones decreases sharply. When the fluence of pre-implanted Zn ions increases to 1.0 × 1017 cm−2 (Figs. 2(c) and 3(c)), it is found that spherical NPs with a size range from 4 to 12 nm in diameter are mainly distributed in a region of about 40 nm thickness. In addition, NPs outside of this region are quite small and infrequent. Compared with those in the Cu sample (Figs. 2(a) and 3(a)), NPs formed in the Zn10 + Cu sample show a higher volume fraction and a narrower size distribution. The SAED diagram inserted in Fig. 2(c) mainly exhibits the diffraction rings of CuZn alloy and fcc Cu, indicating a mixture of Cu–Zn alloy and fcc Cu phases in the Zn10 + Cu sample.
Fig. 2 XTEM results of the (a) Cu, (b) Zn1 + Cu and (c) Zn10 + Cu samples. The corresponding SAED pattern is inserted in each image.
Fig. 3 Particle size distributions of the (a) Cu, (b) Zn1 + Cu and (c) Zn10 + Cu samples. The corresponding average diameters D and standard deviations σ are also given in each figure.
Figure 4 gives the high-angle angular dark-field (HAADF) image (inset) and the line-scanned average atomic-number profiles of Cu and Zn crossing the whole implanted layer in the Zn10 + Cu sample. It can be found that the profiles of Cu and Zn atoms are nearly overlapping and the maximum concentrations of Cu and Zn atoms appear around 20 nm depth. Besides, the detected Cu signal is stronger than Zn signal although the implanted fluence of Cu ions is lower than that of Zn ions. In the case of Cu ions followed by Zn ions (see [21]. and its supporting information), we have ever evidenced that the line-scanned average atomic-number profiles of implants nearly coincide with the simulated results by G(Z) function. For the present case, however, the depth profiles of Zn and Cu in Fig. 4 quietly differ from that revealed by Fig. 1. The reasons will be discussed in detail later. Figure 5 shows the average atomic-number profiles of Cu and Zn crossing two selected NPs. We can see that Cu atoms are predominated in the particle centers, whereas Cu and Zn atoms are contributing to the peripheral regions of NPs together. It is also known that there is a high solid solubility (~38.4 atom %) between Cu and Zn metals [23], combining with the SAED result which only gives the diffraction signals of CuZn alloy and fcc Cu (see the inset in Fig. 2(c)), NPs formed in the Zn10 + Cu sample probably consist of Cu–Zn alloy shells around Cu cores.
Fig. 4 Average atomic-number distributions of Cu and Zn across the whole implanted layer of the Zn10 + Cu sample. The symbols are the experimental data and the lines are drawn only for a guide of eyes. Inset shows the HAADF image and corresponding line-scanned position.
Fig. 5 (a) HAADF image of the Zn10 + Cu sample; (b) and (c) are the average atomic-number profiles of Cu and Zn crossing the selected particles 1 and 2, respectively.
It is clear that Zn ion preimplantation can dramatically influence the structures, spatial and size distributions of post-implantation-synthesized Cu NPs. Firstly, large Cu NPs with a double-layer arrangement are observed in the Zn1 + Cu sample. Secondly, core-shell NPs probably consisting of Cu cores and Cu–Zn alloy shells are formed in the Zn10 + Cu sample. Finally, NPs fabricated in the dually implanted samples show a higher volume fraction and a relatively uniform spatial distribution comparing with those formed in the Cu sample. These changes could be closely related with pre-implanted Zn atoms and Zn-induced defects in the SiO2 substrate. As mentioned above, Zn atom has a large solid solubility in Cu material [23]. Moreover, Zn atom also has a much higher diffusivity than Cu atom in SiO2 [20,21]. During post Cu ion implantation, the forward recoil progress would induce the Zn atoms near the sample surface to migrate toward the deep region, which can “carry” the Cu atoms depositednear the sample surface to move into the deep region. Therefore, the retained amount of Cu implants is enhanced and the sputtering loss of Cu atoms is effectively suppressed. This enhanced deposition process of post implants caused by the prior ion implantation has been well supported by our previous studies where Xe ion irradiation or Zn ion preimplantation followed by Ag ion implantation into SiO2 glasses [24,25]. In addition, the ion beam heating induced by the posterior Cu implantation not only can greatly increase the mobility of Cu atoms but also can cause the aggregation of Zn atoms due to the thermal diffusion, affecting in turn the simultaneous Cu precipitation. Therefore, an increased Cu atomic concentration as well as a promoted Cu nucleation process could be expected, which is conductive to the aggregation and growth of Cu particles in the deep region of the Zn1 + Cu sample. Additionally, many defects, e.g., nonbridging-oxygen hole centers and oxygen-deficient centers [26,27], can be created in the matrix after Zn ion implantation. Post Cu ion implantation would induce a large sputtering loss, leading the maximum concentration of Zn-induced defects to migrate toward the sample surface as shown in Fig. 1. Then a depth-directional driving force associated with the potential gradient could be created in the substrate [28], inducing the shift of maximum concentration of Cu atoms toward the substrate surface in the Zn1 + Cu sample. Along with the Cu particle growth in the near-surface and deep regions, a concentration gradient dependent on the particle size becomes pronounced (smaller particles have a relatively higher solute concentration according to the Gibbs-Thomson equation) [24,28]. So the mass will transport from the small to the large particles via solute diffusion, giving rise to the appearance of a depletion zone in the middle region of the implanted layer. As a result, Cu NPs with a double-layer arrangement and large sizes are formed in the Zn1 + Cu sample as shown in Fig. 2(b). When the fluence of Zn ions increases to 1 × 1017 cm−2 (i.e., the Zn10 + Cu sample), it could be hard to form a depletion zone due to the high atomic concentration of metals. Nevertheless, more defects, which can act as nucleation centers, would be created in SiO2, providing a probability to form NPs with a narrower size distribution as shown in Fig. 3(c).
Based on the above analyses, the differences between the depth profiles of implants revealed in Figs. 1 and 4 can be understood. Firstly, the simulated results in Fig. 1 ignore the dynamic process of ion implantation. Owing to the ion-beam heating, forward recoil progress, and the high diffusivity of Zn atoms in SiO2, the factual maximum concentration of Zn atoms appears at a deeper position (Fig. 4) rather than the sample surface as revealed in Fig. 1. In addition, because of the heavy sputtering loss of Zn atoms and the enhanced deposition of Cu atoms, the detected signal of Cu is stronger than that of Zn as shown in Fig. 4.
As for the formation of Cu/Cu–Zn core/shell NPs in the Zn10 + Cu sample, it is probably associated with the difference between copper’s and zinc’s diffusivities. Zn NPs can be formed in the substrate after Zn ion implantation to a fluence of 1 × 1017 cm−2 [29]. However, most of them could be broke down by the collision cascade process during post Cu ion implantation. Then small Zn particles even numerous Zn atoms could be created in the matrix. Meanwhile, some Cu NPs may be formed in the early state of post Cu ion implantation. Because of the lower diffusivity of Cu than that of Zn, the possibly formed Cu NPs in turn act as nucleation centers to trap surrounding Zn particles/atoms around them to form outer shells of Cu–Zn alloy. In our opinion, this configuration of core-shell NPs may be also formed in the Zn1 + Cu sample. But this Cu–Zn shell could be very thin or even unclosed due to the quite low Zn concentration in the Zn1 + Cu sample so that no alloy or Zn-related diffraction signal is observed in the SAED pattern (see the inset in Fig. 2(b)). In contrast to our previous studies in the case of Cu ions followed by Zn ions [21], as Cu has a relatively low diffusivity in SiO2 than Zn, Cu atoms could not diffuse considerably during post Zn ion implantation. Therefore, post-implanted Zn atoms are mainly dissolved in the preformed Cu particles to form Cu–Zn alloy NPs. As a result, it is only possible that core-shell nanostructures can be fabricated by segregating Zn atoms from Cu–Zn alloy NPs through post thermal annealing.
Figure 6(a) shows the optical absorption spectra of the Cu, Zn1 + Cu and Zn10 + Cu samples. For comparing the influence of implantation sequence, the optical absorption spectrum from the Cu + Zn1 sample is also given. It can be seen that Cu NPs formed in the Cu sample contribute a weak SPR absorption peak at 568 nm. In contrast, a narrow absorption peak appears at 572 nm in the absorption spectrum of the Zn1 + Cu sample. And its relative intensity is about three times as strong as the SPR peak from the Cu sample. The XTEM and SAED results (Fig. 2(b)) indicate that this 572 nm absorption peak originates from the SPR of double-layer Cu NPs. Clearly, Zn ion preimplantation to a fluence of 1 × 1016 cm−2 can greatly enhance the SPR absorption of Cu NPs and shift the related absorption peak to a long wavelength. Such features could result from the rearrangement of Cu NPs in spatial and size distributions. More importantly, there could be very thin or unclosed Zn-related shells to be formed around Cu NPs, which can increase the surrounding dielectric constants of Cu NPs, resulting in the enhancement and redshift of SPR absorption peak. Nevertheless, when the fluence of pre-implanted Zn ions increases to 1.0 × 1017 cm–2, post Cu ion implantation gives rise to an intense and broad absorption band around 528 nm. This absorption band most probably comes from the SPR of Cu/Cu–Zn core/shell NPs according to the XTEM and EDXS results (Figs. 2(c) and 5).
Fig. 6 Optical absorption spectra of (a) the as-implanted samples and (b) the Zn10 + Cu sample before and after annealing at different temperatures.
The SPR absorption peak of one Cu/Cu–Zn core/shell NP can be further calculated by using the discrete dipole approximation (DDA) method [30]. For the Cu–Zn alloy shell, we used the bulk dielectric constant of Cu0.5Zn0.5 β-brass experimentally determined by Sasovskaya and Korabel [31]. The diameter D of shell is set as 7.5 nm according to the most probable diameter of NPs formed in the Zn10 + Cu sample (Fig. 3(c)). The calculated results reveal that when the diameter of Cu core takes 5.4 nm the SPR absorption band appears near 528 nm. For a comparison, the optical absorption spectra from a Cu NP with D = 5.4 nm and a β-brass NP with D = 7.5 nm were also calculated, and the corresponding results are shown in Fig. 7(a). From the figure, one can see that the Cu NP causes a very weak SPR absorption peak at about 570 nm while the β-brass NP presents a sharp absorption peak at 538 nm. However, the core-shell nanostructure gives a broad absorption band at the shorter wavelength, which is in line with the experimental result. Figures 7(b-d) further present the near-field intensity distributions of the Cu/β-brass core/shell, β-brass, and Cu NPs at 532 nm. We can see that the maximum field enhancements outside all NPs occur along the incident polarization and only locate within a few nanometers of the particle surfaces. These phenomena can be interpreted by the dipole plasmon resonance [32]. Besides, it is also clear that the waves are strongly suppressed in the interiors of the β-brass and Cu NPs. Nevertheless, the electrical field is increased in the core and only suppressed in the nanoshell. Owing to the discontinuous field distribution between the core and the nanoshell, special plasmon-polariton modes can therefore be excited in the core as well as in the shell and they couple strongly via their interfaces [33], resulting in a broad and complex SPR absorption band, as shown in Fig. 7(a). Meanwhile, the enhancements of local electric fields both inside and outside of the nanoshell could also enhance the third-order optical nonlinearities [34,35], which were indeed observed by the Zn10 + Cu sample. The corresponding results will be shown later on.
Fig. 7 (a) Optical absorption coefficients Qabs calculated for Cu/β-brass core/shell, β-brass, and Cu NPs embedded in SiO2 by using DDA method. (b), (c) and (d) respectively present the plots of the normalized electric field intensity |E| of the core-shell, β-brass, and Cu NPs at 532 nm. The electric field polarization E and propagation k vectors are indicated in (d).
The discrepancies between different implantation sequences can be further found from Fig. 6(a). It is seen that the Cu + Zn1 sample gives a very wide absorption band around 532 nm, which may be an overlap of the absorption peaks induced by the formed Cu and Cu–Zn alloy NPs in silica [19,21]. As discussed above, Cu atoms could not diffuse considerably during post Zn ion implantation. Unlike the Zn1 + Cu sample, post Zn ion implantation could not induce dramatical changes in the spatial and size distributions of Cu NPs formed in the Cu + Zn1 sample [19]. Therefore, prior Zn ion implantation to a fluence of 1.0 × 1016 cm–2 is more favorable to greatly enhancing the SPR of Cu NPs.
To characterize the thermostability of the formed Cu/Cu–Zn core/shell NPs, the Zn10 + Cu sample was further annealed at elevated temperatures for 1 hour in a nitrogen atmosphere. The corresponding optical absorption spectra are shown in Fig. 6(b). It can be seen that 300 °C annealing shifts the absorption band to about 510 nm accompanied with the appearance of an absorption shoulder around 578 nm. According to the analyses on Fig. 5, small Zn particles even numerous Zn atoms could be created in the substrate by the collision cascade process induced via subsequently incident Cu ions. During the low-temperature annealing, these Zn particles/atoms could be absorbed by Cu–Zn alloy shells around Cu cores (i.e., the Ostwald ripening process) [36]. Then the content of Zn atoms in alloy shells would increase. As a consequence, the absorption band related to Cu–Zn alloy shells shifts to the short wavelength while an absorption shoulder which could be attributed to the Cu cores appears around 578 nm. When the temperature increases to 400 °C, the alloy shell-related absorption band disappears while the Cu SPR peak becomes clear and appears at 583 nm. This change could associate with the thermal diffusion of Zn atoms in NPs. It is known that the melting temperature Tm of NPs is lower than the bulk melting point Tb and strongly depends on the particle diameter D [37]. Determining the melting point of alloy NPs could be difficult, but it is possible to tentatively estimate the melting temperatures of Cu and Zn NPs according to the equation Tm = Tb(1-6r/D), where r is the atomic radius [37]. The maximum diameter of NPs formed in the Zn10 + Cu sample is about 12 nm (see Fig. 3(c)). According to the values of Tb and r for metal Cu (Tb = 1083.4 °C, r = 1.8648 Å) and Zn (Tb = 419.5 °C, r = 1.8004 Å) [38–40], the melting temperatures of Cu and Zn NPs are calculated to be 982.4 °C and 381.7 °C, respectively. It is seen that the estimated melting point of Zn NPs is far smaller than that of Cu ones and is also lower than the annealing temperature 400 °C. Again noting the fast diffusivity of Zn atoms, the segregation of Zn toward the particle surfaces could occur according to the Kirkendall effect [41,42], promoting the growth of Cu cores as well as the aggregation of Zn atoms around Cu cores. Consequently, an absorption peak coming from Cu SPR is detected in the optical absorption spectrum. As the annealing temperature increases to 500 °C, more Zn atoms aggregate around Cu cores, inducing the increase of partial dielectric constants around Cu cores. Thus a much stronger Cu SPR peak is observed at 589 nm. After 600 °C annealing, the content of Zn atoms in the host matrix could decrease and the dielectric constants around Cu NPs could be partially recovered due to the diffusion of Zn atoms toward the substrate surface [20,21]. Therefore, the overall absorption is decreased and Cu SPR peak blue shifts to 581 nm from 589 nm as compared with that in the 500 °C annealed sample. It should be noted that Cu–Zn alloy NPs fabricated by Cu first and then Zn ion implantation are decomposed only when the annealing temperature increases to 500 °C or higher [20,21], whereas Cu–Zn alloy shells in the present work are completely decomposed at 400 °C.
To our knowledge, so far, there have been no reports on the third-order nonlinearities of implantation-synthesized Cu–Zn bimetallic NPs. Nevertheless, materials with extremely large third-order nonlinear susceptibility χ(3) are essential for future applications in optoelectronics, e.g., in all-optical memories, nonlinear waveguide devices, and optical switches, etc [1]. Therefore, the third-order nonlinear optical properties of the as-implanted samples were investigated by Z-scan technique. The normalized nonlinear transmittances T(z) measured with an open-aperture and a closed-aperture are displayed in Figs. 8(a) and 8(b), respectively. Meanwhile, the corresponding transmittance curves are theoretically fitted by [43]
Here, x = z/z0, z is the distance along the lens axis in the far field and z0 is the Rayleigh length of the focused laser beam. The nonlinear absorption coefficient α can be obtained by q0 = αI0Leff, where I0 is the intensity of the laser beam at the focus (z = 0), Leff is the effective thickness of the sample. Leff can be calculated from the real thickness L and the linear absorption coefficient α0 through Leff = (1-exp(-α0L))/α0. The nonlinear refractive index γ is determined by Δψ = kγI0Leff, where k = 2π/λ is the wave vector of the incident laser. It can be seen from Fig. 8(a) that all the open-aperture measurements show clearly enhanced transmittances near the focus, revealing negative nonlinear absorption coefficients for all samples. In addition, the peak-valley configuration of the closed-aperture transmittances (Fig. 8(b)) indicates that all of the nonlinear refractive indices are also negative. These conditions correspond to the self-defocusing of laser radiation [43]. Based on these numerical fittings, the values of α and γ can be deduced, and then the third-order nonlinear susceptibility χ(3) = χR(3) + iχI(3) can be calculated using the following equations [44] where n and c are the linear refractive index of the substrate and the velocity of light in a vacuum, respectively.Fig. 8 Normalized (a) open-aperture and (b) closed-aperture Z-scan transmittances of the Cu, Cu + Zn1, Zn1 + Cu and Zn10 + Cu samples. Symbols are the experimental results and solid lines are the theoretical fittings.
The actual values of γ, α, and χ(3) strongly depend on the particle size, the volume fraction, and the microstructure of NPs, etc [1,35,45]. Especially, when the pumping wavelength is near the SPR absorption of NPs, the nonlinear refraction and absorption can be remarkably enhanced [35]. Table 2 lists the nonlinear optical parameters for the prepared samples. One can see that the absolute values of γ, α, and χ(3) from the dually implanted samples are larger than those from the Cu sample. For the Zn1 + Cu and Zn10 + Cu samples, the enhanced third-order nonlinearities should benefit from the high particle densities and the relatively uniform spatial and size distributions as shown in Figs. 2(b) and 2(c). Particularly, the Zn10 + Cu sample in which the Cu/Cu–Zn core/shell NPs are formed presents the maximum absolute values of the γ, α, and χ(3). This should result from the closer SPR peak (528 nm) with pumping wavelength (532 nm). Finally, post Zn ion implantation to a fluence of 1 × 1016 cm−2 could not effectively improve the spatial and size distributions of NPs. However, the SPR peak of the Cu + Zn1 sample just locates at the pumping wavelength. Therefore, the Cu + Zn1 and Zn1 + Cu samples show the similar nonlinear transmittances (Fig. 8), even the |χ(3)| value of the former is slightly larger than that of the latter.
Table 2. Nonlinear optical parameters of the samples measured at the wavelength of 532 nm
4. Conclusions
In summary, the structures and optical properties of Cu–Zn bimetallic NPs fabricated in a-SiO2 glasses by sequential implantation of Zn and Cu ions have been investigated in detail. The results reveal that pre-implanted Zn atoms as well as their produced defects in the substrate can dramatically affect the nucleation and growth of post Cu implants. At the Zn ion fluence of 1.0 × 1016 cm–2, double-layer Cu NPs are fabricated and give rise to a greatly enhanced Cu SPR absorption at 572 nm. When the fluence of Zn ions increases to 1.0 × 1017 cm–2, Cu/Cu–Zn core/shell NPs with a high particle density and a narrow size distribution can be formed, causing a strong and broad SPR absorption band around 528 nm. Because of the unique configurations, both the dual-implanted samples present enhanced third-order nonlinear refraction and absorption as compared with the Cu-solely implanted sample. Especially, the specimen with the Zn ion fluence of 1.0 × 1017 cm–2 gives the most extremely enhanced third-order optical nonlinearity due to the formation of Cu/Cu–Zn core/shell NPs. This work may provide significant insight into the binary Cu–Zn nanosystem synthesized by ion implantation and open up more application prospects of the Cu–Zn nanocomposites in optoelectronics.
Acknowledgments
Authors acknowledge the financial supports from Natural Science Foundation of China (No. 11175129) and Natural Science Foundation of Tianjin (No. 12JCZDJC 26900).
References and Links
1. A. Meldrum, R. F. Haglund Jr, L. A. Boatner, and C. W. White, “Nanocomposite materials formed by ion implantation,” Adv. Mater. 13(19), 1431–1444 (2001). [CrossRef]
2. M. A. Garcia, “Surface plasmons in metallic nanoparticles: fundamentals and applications,” J. Phys. D Appl. Phys. 44(28), 283001 (2011). [CrossRef]
3. M. Kauranen and A. V. Zayats, “Nonlinear plasmonics,” Nat. Photonics 6(11), 737–748 (2012). [CrossRef]
4. S. Dhara, C.-Y. Lu, P. Magudapathy, Y.-F. Huang, W.-S. Tu, and K.-H. Chen, “Surface plasmon polariton assisted optical switching in noble bimetallic nanoparticle system,” Appl. Phys. Lett. 106(2), 023101 (2015). [CrossRef]
5. O. A. Yeshchenko, I. S. Bondarchuk, V. S. Gurin, I. M. Dmitruk, and A. V. Kotko, “Temperature dependence of the surface plasmon resonance in gold nanoparticles,” Surf. Sci. 608, 275–281 (2013). [CrossRef]
6. L. A. H. Fleming, G. Tang, S. A. Zolotovskaya, and A. Abdolvand, “Controlled modification of optical and structural properties of glass with embedded silver nanoparticles by nanosecond pulsed laser irradiation,” Opt. Mater. Express 4(5), 969–975 (2014). [CrossRef]
7. B. Can-Uc, R. Rangel-Rojo, H. Márquez, L. Rodríguez-Fernández, and A. Oliver, “Nanoparticle containing channel waveguides produced by a multi-energy masked ion-implantation process,” Opt. Express 23(3), 3176–3185 (2015). [CrossRef]
8. S. Zuccon, E. Napolitani, E. Tessarolo, P. Zuppella, A. J. Corso, F. Gerlin, M. Nardello, and M. G. Pelizzo, “Effects of helium ion bombardment on metallic gold and iridium thin films,” Opt. Mater. Express 5(1), 176–187 (2015). [CrossRef]
9. A. L. Stepanov, M. F. Galyautdinov, A. B. Evlyukhin, V. I. Nuzhdin, V. F. Valeev, Y. N. Osin, E. A. Evlyukhin, R. Kiyan, T. S. Kavetskyy, and B. N. Chichkov, “Synthesis of periodic plasmonic microstructures with copper nanoparticles in silica glass by low-energy ion implantation,” Appl. Phys., A Mater. Sci. Process. 111(1), 261–264 (2013). [CrossRef]
10. G. Mattei, “Alloy nanoclusters in dielectric matrix,” Nucl. Instrum. Methods Phys. Res Sect. B 191, 323–332 (2002).
11. O. Peña, U. Pal, L. Rodríguez-Fernández, H. G. Silva-Pereyra, V. Rodríguez-Iglesias, J. C. Cheang-Wong, J. Arenas-Alatorre, and A. Oliver, “Formation of Au-Ag core-shell nanostructures in silica matrix by sequential ion implantation,” J. Phys. Chem. C 113(6), 2296–2300 (2009). [CrossRef]
12. V. F. Degtyareva, O. Degtyareva, M. K. Sakharov, N. I. Novokhatskaya, P. Dera, H. K. Mao, and R. J. Hemley, “Stability of Hume-Rothery phases in Cu–Zn alloys at pressures up to 50 GPa,” J. Phys. Condens. Matter 17(50), 7955–7962 (2005). [CrossRef]
13. D. S. Wang, Q. Peng, and Y. D. Li, “Nanocrystalline intermetallics and alloys,” Nano Res. 3(8), 574–580 (2010). [CrossRef]
14. M. Cokoja, H. Parala, M. K. Schröter, A. Birkner, M. W. E. van den Berg, K. V. Klementiev, W. Grünert, and R. A. Fischer, “Nano-brass colloids: synthesis by co-hydrogenolysis of [CpCu(PMe3)] with [ZnCp*2] and investigation of the oxidation behaviour of α/β-CuZn nanoparticles,” J. Mater. Chem. 16(25), 2420–2428 (2006). [CrossRef]
15. N. Yang, H. B. Yang, Y. Q. Qu, Y. Z. Fan, L. X. Chang, H. Y. Zhu, M. H. Li, and G. T. Zou, “Preparation of Cu–Zn/ZnO core–shell nanocomposite by surface modification and precipitation process in aqueous solution and its photoluminescence properties,” Mater. Res. Bull. 41(11), 2154–2160 (2006). [CrossRef]
16. Z. Wang, L. Zhu, W. Li, and H. Liu, “Rapid reversible superhydrophobicity-to-superhydrophilicity transition on alternating current etched brass,” ACS Appl. Mater. Interfaces 5(11), 4808–4814 (2013). [CrossRef] [PubMed]
17. İ. H. Karahan and R. Özdemir, “Effect of Cu concentration on the formation of Cu1−xZnx shape memory alloy thin films,” Appl. Surf. Sci. 318, 100–104 (2014). [CrossRef]
18. B. Derkus, E. Emregul, and K. C. Emregul, “Copper-zinc alloy nanoparticle based enzyme-free superoxide radical sensing on a screen-printed electrode,” Talanta 134, 206–214 (2015). [CrossRef] [PubMed]
19. G. Y. Jia, J. Wang, L. H. Zhang, H. X. Liu, R. Xu, and C. L. Liu, “Remarkably enhanced surface plasmon resonance absorption of Cu nanoparticles in SiO2 by post Zn ion implantation,” EPL 101(5), 57005 (2013). [CrossRef]
20. J. Wang, L. H. Zhang, X. D. Zhang, Y. Y. Shen, and C. L. Liu, “Synthesis, thermal evolution and optical properties of CuZn alloy nanoparticles in SiO2 sequentially implanted with dual ions,” J. Alloys Compd. 549, 231–237 (2013). [CrossRef]
21. G. Jia, R. Xu, X. Mu, and C. Liu, “Zn ion post-implantation-driven synthesis of CuZn alloy nanoparticles in Cu-preimplanted silica and their thermal evolution,” ACS Appl. Mater. Interfaces 5(24), 13055–13062 (2013). [CrossRef] [PubMed]
22. J. F. Ziegler, computer code, SRIM 2013, available online at: http://www.srim.org/.
23. W. Hume-Rothery, G. W. Mabbott, and K. M. Channel Evans, “The freezing points, melting points, and solid solubility limits of the alloys of silver, and copper with the elements of the B sub-groups,” Philos. Trans. R. Soc. A 233(721-730), 1–97 (1934). [CrossRef]
24. J. Wang, G. Y. Jia, X. Y. Mu, and C. L. Liu, “Quasi-two-dimensional Ag nanoparticle formation in silica by Xe ion irradiation and subsequent Ag ion implantation,” Appl. Phys. Lett. 102(13), 133102 (2013). [CrossRef]
25. J. Wang, G. Y. Jia, B. Zhang, H. X. Liu, and C. L. Liu, “Formation and optical absorption property of nanometer metallic colloids in Zn and Ag dually implanted silica: Synthesis of the modified Ag nanoparticles,” J. Appl. Phys. 113(3), 034304 (2013). [CrossRef]
26. M. Fujinami and N. B. Chilton, “Ion implantation induced defects in Si02: The applicability of the positron probe,” Appl. Phys. Lett. 62(10), 1131–1133 (1993). [CrossRef]
27. P. Mazzoldi and G. Mattei, “Potentialities of ion implantation for the synthesis and modification of metal nanoclusters,” Riv. Nuovo Cim. 28(7), 1–69 (2005).
28. N. Kishimoto, N. Umeda, Y. Takeda, V. T. Gritsyna, T. J. Renk, and M. O. Thompson, “In-beam growth and rearrangement of nanoparticles in insulators induced by high-current negative copper ions,” Vacuum 58(1), 60–78 (2000). [CrossRef]
29. H. Amekura, M. Ohnuma, N. Kishimoto, Ch. Buchal, and S. Mantl, “Fluence-dependent formation of Zn and ZnO nanoparticles by ion implantation and thermal oxidation: An attempt to control nanoparticle size,” J. Appl. Phys. 104(11), 114309 (2008). [CrossRef]
30. B. T. Draine and P. J. Flatau, “User Guide for the Discrete Dipole Approximation Code DDSCAT 7.3,” http://arxiv.org/abs/1305.6497 (2013).
31. I. I. Sasovskaya and V. P. Korabel, “Optical properties of α- and β-CuZn brasses in the region of quantum absorption,” Phys. Status Solidi, B Basic Res. 134(2), 621–630 (1986). [CrossRef]
32. K. L. Kelly, E. Coronado, L. L. Zhao, and G. C. Schatz, “The optical properties of metal nanoparticles: The influence of size, shape, and dielectric environment,” J. Phys. Chem. B 107(3), 668–677 (2003). [CrossRef]
33. U. Kreibig and M. Vollmer, Optical Properties of Metal Clusters (Springer, 1985).
34. Y. Liu, D. Li, R. Y. Zhu, G. J. You, S. X. Qian, Y. Yang, and J. L. Shi, “Third-order nonlinear optical response of Au-core CdS-shell composite nanoparticles embedded in BaTiO3 thin films,” Appl. Phys. B 76(4), 435–439 (2003). [CrossRef]
35. T. V. Murzina, I. A. Kolmychek, J. Wouters, T. Verbiest, and O. A. Aktsipetrov, “Plasmon-assisted enhancement of third-order nonlinear optical effects in core (shell) nanoparticles,” J. Opt. Soc. Am. B 29(1), 138–143 (2012). [CrossRef]
36. P. W. Voorhees, “The theory of Ostwald ripening,” J. Stat. Phys. 38(1–2), 231–252 (1985). [CrossRef]
37. W. H. Qi and M. P. Wang, “Size and shape dependent melting temperature of metallic nanoparticles,” Mater. Chem. Phys. 88(2–3), 280–284 (2004). [CrossRef]
38. I. Khader, A. Renz, A. Kailer, and D. Haas, “Thermal and corrosion properties of silicon nitride for copper die casting components,” J. Eur. Ceram. Soc. 33(3), 593–602 (2013). [CrossRef]
39. Z. Gu, M. P. Paranthaman, J. Xu, and Z. W. Pan, “Aligned ZnO nanorod arrays grown directly on zinc foils and zinc spheres by a low-temperature oxidization method,” ACS Nano 3(2), 273–278 (2009). [CrossRef] [PubMed]
40. D. C. Ghosh and R. Biswas, “Theoretical calculation of absolute radii of atoms and ions. Part 1. The atomic radii,” Int. J. Mol. Sci. 3(2), 87–113 (2002). [CrossRef]
41. Y. H. Xu and J. P. Wang, “Direct gas-phase synthesis of heterostructured nanoparticles through phase separation and surface segregation,” Adv. Mater. 20(5), 994–999 (2008). [CrossRef]
42. M. Heggen, M. Oezaslan, L. Houben, and P. Strasser, “Formation and analysis of core−shell fine structures in Pt bimetallic nanoparticle fuel cell electrocatalysts,” J. Phys. Chem. C 116(36), 19073–19083 (2012). [CrossRef]
43. 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). [CrossRef]
44. E. Cattaruzza, G. Battaglin, F. Gonella, G. Mattei, P. Mazzoldi, R. Polloni, and B. F. Scremin, “Fast third-order optical nonlinearities in metal alloy nanocluster composite glass: Negative sign of the nonlinear refractive index,” Appl. Surf. Sci. 247(1–4), 390–395 (2005). [CrossRef]
45. G. Ma, W. Sun, S. H. Tang, H. Zhang, Z. Shen, and S. Qian, “Size and dielectric dependence of the third-order nonlinear optical response of Au nanocrystals embedded in matrices,” Opt. Lett. 27(12), 1043–1045 (2002). [CrossRef] [PubMed]
