1 Introduction

Metal nanoparticles (MNPs) with at least one dimension (length, width, or thickness) within a range of 1–100 nm, present exclusive physiochemical properties, such as high stability, exceptional optical properties and catalytic activities, enabling potential applications in catalysis [1], sensing [2, 3], energy storage [4], surface enhanced Raman scattering [5]. Among various MNPs, silver nanoparticles (AgNPs) were paid great attention to a variety of fields due to their unique physical and chemical properties, such as surface plasmon resonance (SPR) [6] and fluorescent properties [7], owing to their excellent photo-stability, good process ability, and extremely low toxicity. Generally, AgNPs with sizes of less than 2 nm, where the particle size is on the same order of the electron Fermi wavelength (0.5 nm), exhibit a fascinating fluorescent property, and become non-luminescent with the particle size approaching to the electron mean free path length (~ 50 nm), where collective excitations of electrons become dominant, leading to plasmon resonance [8, 9]. It has been reported that the SPR and fluorescent properties of silver nanostructures depend strongly on its geometry and environment, such as the nanoparticle size [10] and shape [11]. Shang et al. [12] synthesized fluorescent Ag nanoclusters using the common and commercially available polyelectrolyte poly(methacrylic acid) as the template. The obtained Ag nanoclusters exhibit good photo-stability, pH- and excitation-dependent emission behavior, and a stimuli-responsive property, showing potential applications in a wide range such as bio-sensing. Luo et al. [13] obtained Ag nanorods with thin diameters and tunable aspect ratio using a seed-mediated approach. It is found that the transverse mode kept below 400 nm, while the longitudinal mode could be readily tuned from the visible to the near-infrared region by varying the aspect ratio. Tsuji et al. [14] studied the rapid sphere-to-prism (STP) transformation of AgNPs by monitoring the changing of SPR bands in the UV–Vis region. They reported that a key process for the STP transformation is shape-selective oxidative etching of spherical particles in a mixture of spheres and prisms with that the etching rate of spherical particles by H2O2 is faster than that of prisms. Scardaci et al. [15] found that the shape change of AgNPs from spherical to flat nanoparticles can tune the position of the main plasmon resonance across the visible and near-infrared part of the spectrum. Therefore, it is crucial for the controllable synthesis of AgNPs with desirable physicochemical properties.

Recently, the third-order nonlinear optical (NLO) properties of AgNPs and its nanocomposites have attracted a lot of attention due to its potential applications in optical switching [16] and ultrafast laser generation [17]. Gieseking et al. [18] found in theory that in linear silver nanowires, the longitudinal excited states show positive NLO polarizability which increases with nanowire length, while the transverse modes show a negative one. Wada et al. [19] fabricated a layered hybridized thin films composed of oriented polymers and AgNPs. They found that the existence of AgNPs in the hybrid thin films can significantly enhance the third-order NLO properties due to “excitation enhancement” induced by the localized SPR. Ferreira et al. [20] synthesized Au–Ag bimetallic nanoparticles and reported that the third-order NLO susceptibility can be tailored by modifying the thickness of shell. The NLO absorption of AgNPs contains a variety of processes like saturable absorption (SA), two-photon absorption (TPA), multiphoton absorption, excited state absorption (ESA), free-carrier absorption (FCA), and reverse saturable absorption (RSA), which have different applications in different areas. Generally, SA leads to decrease in absorption in the sample along with increase in the input intensity which can be used to generate ultrashort pulse lasers, whereas for TPA and other processes, the absorption in the sample increases along with increase in the input intensity resulting in optical limiting effect, which can be used for optical limiter to protect the electro-optical sensors and the eyes [21]. For example, Bhavitha et al. [22] reported the optical limiting performance of Ag cluster and investigated the solvent effect at 532 nm using 5 ns laser pulse. Zvyagin et al. [23] observed optical limiting effect in AgNPs with spherical and triangular shapes using 10-ns laser pulses at 532 nm, respectively. It is found that the optical limiting effect was induced by the nonlinear light scattering. Maurya et al. [24] investigated the polymer effect on the nonlinear absorption of AgNPs using femtosecond (fs) laser pulses at 400 nm and found that the SA process can be suppressed at high concentration of the polymer. However, how to differentiate which process dominates the NLO performance remains challenging for AgNPs, especially for the optical limiting effect.

On the other hand, two-dimensional (2D) transition metal dichalcogenides (TMDs) show excellent NLO performances, such as giant TPA absorption in monolayer MoS2 [25] and SA in few-layer MoS2 [26, 27], with potential applications in ultrafast laser generation and optical limiting. Formation of hetero-structured nanocomposites of AgNPs and 2D TMDs is expected to further enhance the NLO absorption properties compared with their counterparts. Shao et al. [28] reported that introduction of AgNPs in WS2-Ag nanocomposites can further enhance the nonlinear absorption of 2D WS2 at nanosecond 532 nm. Poudel et al. [29] found a strong coupling of excitonic states in Ag-MoS2 hybrids.

In this work, we used a simple chemical synthesis method to synthesize AgNP colloids using sodium citrate as stabilizing agent and sodium borohydride as reducing agent in water–ethanol solution by a simple mixing process. It is found that the addition of the amount of ethanol significantly affects the formation and growth of AgNPs in the colloids. The AgNP colloids synthesized with optical parameters can keep stable for several weeks without heating or stirring. Then, we investigated its linear and nonlinear optical properties by ultraviolet visible (UV–Vis) spectrophotometer and open-aperture Z-scan (Z is Z-axis, which is the optical axis) technique irradiated at 515 nm with 340 fs at a repetition rate of 1 kHz, respectively. A tunable nonlinear optical performance is observed by adjusting the SPR frequencies. In addition, the selectively synthesized AgNPs were found to be able to greatly enhance the NLO properties of 2D MoS2 by forming hetero-structured nanocomposites.

2 Experimental

2.1 Synthesis of AgNPs colloids

All chemicals were analytical-grade reagents and used as received without further purification. Water used for this study was double-distilled water. Silver nitrate was used as precursor of silver to prepare AgNPs in colloidal solution with a concentration of 0.8 × 10–4 mol·L−1 in the existence of sodium citrate as surfactant.

To investigate the effect of the ethanol on the morphology of the AgNPs and the stability of the colloids, a series of samples with different amounts of ethanol were prepared. Typically, 0.5 mL 5 × 10–3 mol·L−1 AgNO3 was mixed with 30 mL of deionized water and 0.1 mL 1 wt% sodium citrate, followed by addition of 0.1 mL 1 mg·mL−1 sodium borohydride gradually without stirring at room temperature. The color of the mixture turned slowly from colorless to yellow. The obtained sample was denoted as AgNPs(0), where 0 means 0 mL ethanol. By increasing the amount of ethanol from 0 to 15, 20, and 25 mL, a series of AgNPs aqueous colloids were obtained, denoted as AgNPs(15), AgNPs(20), AgNPs(25), respectively.

2.2 Synthesis of AgNPs/MoS2 nanocomposites

The AgNPs/MoS2 nanocomposites were synthesized via a simple two-step process. First, it is the synthesis of 2D MoS2 by liquid-exfoliated method [26]. Typically, MoS2 powders (5 mg·mL−1) were dispersed in 90% ethanol aqueous solution and ultra-sonicated in a water bath for 24 h without addition of any other surfactants. The dispersions after standing for 24 h were centrifuged at 3000 r·min−1 for 90 min to remove the un-exfoliated powders. The top two-thirds of the dispersions were collected for use. Then, similar to the synthesis of AgNPs(25) colloids, Ag/MoS2 hetero-structured nanocomposites were in situ synthesized in 2D MoS2 solutions by adding sequentially Ag precursors, sodium citrate and NaBH4 with the volume of ethanol as 25 mL.

2.3 Characterizations

UV–Vis absorption spectra were measured for the samples with a PerkinElmer Lambda 750 instrument in the range of 250–800 nm. Transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) were carried out with a JEOL 3010 instrument. The samples were drop-cast on carbon-coated copper grids (400 meshes) and allowed to dry under ambient conditions.

2.4 NLO measurements

An open-aperture Z-scan system was used to investigate the third-order NLO properties of AgNPs and AgNPs/MoS2 aqueous solution in a quartz cuvette with a thickness of 1 mm using a mode-locked fiber laser operating at 515 nm with 340 fs pulses at a 1 kHz repetition rate [30]. Total transmittance through the samples as a function of incident intensity was measured while the samples were gradually moved through the focus of a lens along the Z-axis. The beam waist radii of the focus at 515 nm were estimated to be ~ 30 μm. In our experiments, the excitation power is limited in a range within which Z-scan curves are readily reproducible without any change in linear transmittance to avoid laser damage.

3 Results and discussion

3.1 Characterization of the fabricated AgNPs colloids

A series of AgNP colloids was synthesized with different amounts of ethanol. Figure 1 shows the UV–Vis absorption spectra of the series of AgNP colloids with the detailed synthesis parameters shown in Table 1. Obviously, samples AgNPs(0), AgNPs(15) and AgNPs(20) exhibit only one SPR peak at ~ 400 nm [31], while AgNPs(25) revealed two SPR bands with the color of the solution displaying light purple (Fig. 1a, c). According to previous reports, the one at ~ 400 nm is ascribed to the transverse plasmon mode, and the other one at ~ 520 nm is ascribed to the longitudinal mode, demonstrating a non-spherical morphology of AgNPs in AgNPs(25) [13, 32]. Gao et al. [33] also reported the SPR bands evolution along with the shape of AgNPs changing from isotropic sphere to anisotropic triangle from only one mode at ~ 450 nm to two modes with another at ~ 600–800 nm. On the other hand, the symmetry of the SPR peak at ~ 400 nm becomes poorer along with increasing the amount of ethanol, suggesting the shape of the AgNPs grows gradually non-spherical [33, 34]. The results demonstrate the addition of ethanol plays a vital role on controlling the shape of AgNPs. By increasing the volume ratio of ethanol in the system to a suitable value, two SPR modes will be observed for the obtained AgNPs colloids.

Fig. 1
figure 1

Effect of ethanol on the absorption for the prepared AgNPs colloidal solution. a Linear absorption spectra for the Ag colloidal solution prepared with the amount of ethanol as 0, 15, 20 and 25 mL, respectively. b A407/l versus sedimentation time for the obtained AgNPs colloid solution. c Photographs of the colloids with different storage times. Note: d means days

Full size image
Table 1 Synthesis parameters for the AgNP colloids
Full size table

By comparing the positions of the SPR peak at ~ 400 nm for the samples (Fig. 1a), a little red-shift from ~ 401 to ~ 403 nm then to ~ 416 nm occurs along with increasing the volume of ethanol (VE) from 0 to 15 mL, and then to 20 mL. The red-shift is probably ascribed to the increased anisotropy with higher aspect ratio of the AgNPs induced by the increase of the ethanol amount [11, 35], which is further proved by the following TEM results. The refraction index of the host media may also contribute to the red-shift with the refraction index increasing from ~ 1.33 for AgNPs(0) to about 1.36 for AgNPs(25) [36].

To monitor the colloidal stability of the samples in a quantitative manner, we measured the absorption intensity at 407 nm per unit cell length (l). Figure 1b shows Absorption407 (A407)/l as a function of time for dispersions for AgNPs(0), AgNPs(15), AgNPs(20) and AgNPs(25), respectively. Obviously, sample AgNPs(15) reveals the highest dispersion ability. This is due to that the increasing amount of ethanol can accelerate the reduction rate of AgNO3, which results in the highest stability of AgNPs(15) with moderate reaction rate among the series of samples. As demonstrated by Morales-Bonilla et al. [37], the NPs with sizes smaller than 17 nm will stay suspended in the solution with a good stability, which is in consistent with our results. Combined with the following TEM results, the poor dispersion ability of sample AgNPs(25) can be ascribed to the larger sizes of un-spherical AgNPs with obvious larger aspect ratio compared with the other three samples.

The morphology of the prepared AgNPs in the colloidal solution was characterized by TEM and HRTEM. Figure 2a–c reveals the typical TEM images of the samples synthesized with different amounts of ethanol. The AgNPs show an average diameter of ~ 15 nm and well crystal structure with the average size slightly increasing with increasing the amount of ethanol. Clearly, the AgNPs in AgNPs(0) and AgNPs(15) present spherical morphology. While for AgNPs(25), the AgNP displays an obvious elliptical morphology (Fig. 2c), leading to the appearance of the longitudinal mode at ~ 520 nm. The results are consistent with the results of the UV–Vis absorption spectra. The lattice spacing of 0.23 nm was observed from the HRTEM image in Fig. 2d, which is corresponding to the {111} planes of Ag [13].

Fig. 2
figure 2

Typical TEM images of a–c AgNPs(0), AgNPs(15) and AgNPs(25). d Typical HRTEM image of a AgNP for sample AgNPs(15)

Full size image

3.2 NLO performances of the AgNPs colloids

Based on the above results, we chose samples AgNPs(15) and AgNPs(25) as the typical samples to investigate their NLO properties and the dominant mechanism using Z-scan setup. The open-aperture Z-scan system was illustrated in Fig. 3. This experimental set-up has been successfully used in measuring the NLO performances of graphene and 2D TMDs [38, 39].

Fig. 3
figure 3

Schematic of the adopted open-aperture Z-scan set-up

Full size image

Figure 4 shows the typical Z-scan curves of AgNPs(15) and AgNPs(25) excited by 515 nm (340 fs). As shown in Fig. 4a, the AgNPs(15) colloids exhibit obvious optical limiting response, while sample AgNPs(25) shows obvious SA response (Fig. 4b). In our case, the thermal effects can be neglected because the samples were excited by femtosecond laser pulses with low irradiance intensity which have weak peak intensity [40]. So, the observed optical limiting response can be attributed to the absorption, such as TPA, multi-photon absorption, ESA, FCA, or RSA. For FCA effect, the critical fluence is:

$$ F_{{\text{C}}} = 4\sqrt {2} {\text{h}} \omega \left[ {\sigma \left( {1 - R} \right)} \right], $$
(1)

where σ is the FCA absorption cross section and R is the surface reflectivity which is much higher when compared with the onset fluence for TPA. The estimated FC in previous report is ~ 0.1 J·cm−2, which is approximately two orders of magnitude larger than the fluence in our experiment [41]. This means the FCA effect can also be neglected. At the same time, AgNPs(15) shows only one SPR peak at ~ 400 nm (~ 3.1 eV) with the photon energy larger than that of the excitation laser (515 nm, 2.4 eV) which needs to absorb two photons to excite the surface plasmons. For AgNPs(25), two SPR modes were observed with one at ~ 400 nm and the other at ~ 520 nm with the photon energy smaller than that of the excitation laser, making it possible to absorb one photon to excite the surface plasmons. This means that only TPA exists in AgNPs(15), while TPA and SA co-exist in AgNPs(25). Considering that the irradiation wavelength of 515 nm is near to the longitudinal SPR mode at ~ 520 nm, it is reasonable to assume that SA is dominant in AgNPs(25) [42]. This explains why AgNPs(15) exhibits TPA response while AgNPs(25) shows SA response.

Fig. 4
figure 4

Typical Z-scan results for the samples a AgNPs(15) and b AgNPs(25) under excitation of femtosecond laser at 515 nm with a pulse width of 340 fs. The solid symbols denote the experimental data and the solid lines represent the theoretical fits, and Z' represents the distance from sample to focus

Full size image

To quantitatively determine the NLO characteristics of AgNP colloids, the Z-scan data were fitted using the propagation equation in the form of Eq. (2) [43, 44],

$$ \frac{{dI\left( {z^{\prime}} \right)}}{dz^{\prime}} = - \alpha_{0} I\left( {z^{\prime}} \right) - \beta_{{{\text{eff}}}} I^{2} \left( {z^{\prime}} \right), $$
(2)

where I(\({z}^{^{\prime}}\)) is the laser beam intensity through the sample, \({z}^{^{\prime}}\) is the propagation depth in the sample, \({\alpha }_{0}\) is the linear absorption coefficient of the sample, and βeff is the nonlinear absorption coefficient. In the case of SA, βeff is negative, while in the case of TPA, βeff is positive. This Eq. (2) can be solved exactly, and the normalized power transmission T(z) is seen in Ref. [38]

$$ T\left( z \right) = {{\left[ {\ln \left( {1 + \beta_{{{\text{eff}}}} I_{0} L_{{{\text{eff}}}} \left( {{{1 + z^{2} } \mathord{\left/ {\vphantom {{1 + z^{2} } {z_{0}^{2} }}} \right. \kern-\nulldelimiterspace} {z_{0}^{2} }}} \right)} \right)} \right]} \mathord{\left/ {\vphantom {{\left[ {\ln \left( {1 + \beta_{{{\text{eff}}}} I_{0} L_{{{\text{eff}}}} \left( {{{1 + z^{2} } \mathord{\left/ {\vphantom {{1 + z^{2} } {z_{0}^{2} }}} \right. \kern-\nulldelimiterspace} {z_{0}^{2} }}} \right)} \right)} \right]} {\left[ {\beta_{{{\text{eff}}}} I_{0} L_{{{\text{eff}}}} \left( {{{1 + z^{2} } \mathord{\left/ {\vphantom {{1 + z^{2} } {z_{0}^{2} }}} \right. \kern-\nulldelimiterspace} {z_{0}^{2} }}} \right)} \right]}}} \right. \kern-\nulldelimiterspace} {\left[ {\beta_{{{\text{eff}}}} I_{0} L_{{{\text{eff}}}} \left( {{{1 + z^{2} } \mathord{\left/ {\vphantom {{1 + z^{2} } {z_{0}^{2} }}} \right. \kern-\nulldelimiterspace} {z_{0}^{2} }}} \right)} \right]}}, $$
(3)

where \({L}_{\mathrm{eff}}=(1-{e}^{-{\alpha }_{0}L})/{\alpha }_{0}\) is the sample’s effective thickness, which is the optical length of the adopted cuvettes (1 mm), I0 is the on-axis irradiance at the focus, L is the sample’s real thickness, and z0 is the beam’s diffraction length.

The solid lines in Fig. 4a and b are the fitting curves based on the nonlinear absorption model and agree well with the experimental data. The fitting parameters are shown in Table 2. The nonlinear absorption coefficient βeff was fitted to be (1.96 ± 0.4) × 10–2 cm·GW−1 and − (7.90 ± 0.4) × 10–2 cm·GW−1 at the irradiance intensity of ~ (2.2 ± 1.2) GW·cm−2 for AgNPs(15) and AgNPs(25), respectively. The TPA coefficient βeff for AgNPs(15) is increased to (5.81 ± 0.4) × 10–2 cm·GW−1, while the SA coefficient βeff for AgNPs(25) kindly decreased to − (3.45 ± 0.6) × 10–2 cm·GW−1 with increased irradiance intensity to (6.8 ± 1.2) GW·cm−2, respectively. The irradiance-dependent TPA coefficient was reported previously by Zhang et al. [38]. The deviation of the nonlinearity at different irradiances for AgNPs(25) can be mainly due to the experimental errors (Table 2). The strong SA absorption response of AgNPs(25) can be ascribed to the relatively strong longitudinal resonant absorption (~ 520 nm) near the excitation wavelength at 515 nm, resulting in prominent SA response induced by one-photon absorption [45]. López-Suárez et al. [46] observed the variation of TPA to SA in AgNPs with only one SPR band at ~ 400 nm under off-resonance excitation, which is caused by the small difference in the magnitude of the absorbance values and SPR peak positions. In our work, the two AgNPs samples exhibit significantly different linear absorption in the UV–Vis region as shown in Fig. 1. Thus, in our experimental conditions, only TPA occurs for the sample with only one SPR band, while both TPA and SA co-exist in the sample with two SPR bands with SA dominant due to the on-resonance irradiation. This phenomenon reveals that the on/off resonance of SA/TPA near the SPR frequency plays an important role in adjusting the NLO response [44]. The imaginary third-order susceptibility Imχ(3) can be approximately expressed in Eq. (4):

$$ {\text{Im}} \chi^{(3)} = \left[ {{{\left( {10^{ - 7} c\lambda n^{2} } \right)} \mathord{\left/ {\vphantom {{\left( {10^{ - 7} c\lambda n^{2} } \right)} {\left( {96\pi^{2} } \right)}}} \right. \kern-\nulldelimiterspace} {\left( {96\pi^{2} } \right)}}} \right]\beta_{{{\text{eff}}}} , $$
(4)

where c, λ, and n are the light speed in vacuum, wavelength of the laser pulses, and the refractive index, respectively. Here, n is taken as 1.36. The figure of merit (FOM) for the third-order optical nonlinearity is defined as:

$$ {\text{FOM}} = \left| {{{{\text{Im}} \chi^{\left( 3 \right)} } \mathord{\left/ {\vphantom {{{\text{Im}} \chi^{\left( 3 \right)} } {\alpha_{0} }}} \right. \kern-\nulldelimiterspace} {\alpha_{0} }}} \right|. $$
(5)
Table 2 Linear and NLO parameters of the AgNP colloid excited at 515 nm in the fs region
Full size table

The calculated Imχ(3) and FOM for the two typical samples are shown in Table 2.

To further study the SA characteristic of the AgNPs colloids, we fitted the Z-scan data using the following equation:

$$ T\left( z \right) = {{\left[ {1 - \frac{{I_{{{\text{sat}}}} \times \Delta R}}{{{{I_{{{\text{sat}}}} + I_{0} } \mathord{\left/ {\vphantom {{I_{{{\text{sat}}}} + I_{0} } {\left( {{{1 + Z^{2} } \mathord{\left/ {\vphantom {{1 + Z^{2} } {Z_{0}^{2} }}} \right. \kern-\nulldelimiterspace} {Z_{0}^{2} }}} \right)}}} \right. \kern-\nulldelimiterspace} {\left( {{{1 + Z^{2} } \mathord{\left/ {\vphantom {{1 + Z^{2} } {Z_{0}^{2} }}} \right. \kern-\nulldelimiterspace} {Z_{0}^{2} }}} \right)}}}}} \right]} \mathord{\left/ {\vphantom {{\left[ {1 - \frac{{I_{{{\text{sat}}}} \times \Delta R}}{{{{I_{{{\text{sat}}}} + I_{0} } \mathord{\left/ {\vphantom {{I_{{{\text{sat}}}} + I_{0} } {\left( {{{1 + Z^{2} } \mathord{\left/ {\vphantom {{1 + Z^{2} } {Z_{0}^{2} }}} \right. \kern-\nulldelimiterspace} {Z_{0}^{2} }}} \right)}}} \right. \kern-\nulldelimiterspace} {\left( {{{1 + Z^{2} } \mathord{\left/ {\vphantom {{1 + Z^{2} } {Z_{0}^{2} }}} \right. \kern-\nulldelimiterspace} {Z_{0}^{2} }}} \right)}}}}} \right]} {\left( {1 - \Delta R} \right)}}} \right. \kern-\nulldelimiterspace} {\left( {1 - \Delta R} \right)}}, $$
(6)

where ΔR and Isat are the modulation depth and saturation intensity, respectively. As shown in Fig. 4b, the modulation depth is 2.1%. By fitting the curve, the saturation intensity of the AgNPs colloids was calculated to be 39.42 GW·cm−2 at irradiance of (6.8 ± 1.2) GW·cm−2. As shown in Table 2, the saturation intensity Isat becomes larger at higher irradiance, which can be explained by a weak saturation effect [44]. Additionally, both the relative absorption intensity and the position of the longitudinal SPR band are dependent on the aspect ratio of the anisotropic AgNPs [13]. The NLO absorption in the anisotropic nanoparticle can be greatly influenced by its aspect ratio. Our work demonstrates that the varied NLO performances from SA to TPA can be realized for AgNPs by adjusting its SPR frequencies, which is meaningful in practical applications. Gorham et al. [47] reported that if AgNPs is oxidized, the SPR bands will disappear with obvious color loss. In our work, no color loss is observed for the AgNPs colloids during measuring process with the Z-scan results repeatable, suggesting that no oxidation of AgNPs occurs under the experimental conditions.

3.3 NLO properties of AgNPs/MoS2 nanocomposites

It has been reported that 2D MoS2 shows SA response under the excitation of femtosecond laser pulses in a wide range from visible light region to near-infrared region [25, 48]. Thus, we choose AgNPs(25) with SA behavior to form composites with 2D MoS2. Figure 5a shows the linear absorption spectra of MoS2 and Ag/MoS2 nanocomposites. The characteristic bands at 623 and 670 nm corresponding to few-layered MoS2 are observed for both two samples, which are ascribed to the A and B excitonic transitions arising from the K point of the Brillouin zone [49]. Compared with pure MoS2 dispersions, the nanocomposites exhibit higher absorption in the UV–Vis range, which is due to the surface plasmon band of AgNPs as shown by the arrows.

Fig. 5
figure 5

a Linear absorption spectra and b typical Z-scan results for the samples MoS2 and Ag/MoS2 suspensions under excitation of femtosecond laser at 515 nm with a pulse width of 340 fs. The solid symbols denote the experimental data and the solid lines represent the theoretical fits

Full size image

Their NLO properties were investigated using Z-scan technique under the same conditions with those of AgNPs. Figure 5b shows the typical Z-scan curves of MoS2 and Ag/MoS2 suspensions. As expected, the nanocomposites show enhanced SA response compared with that of pure MoS2 under the same measuring conditions. By fitting the experimental data, the nonlinear absorption coefficient βeff was obtained to be − (3.71 ± 0.6) × 10–2 cm·GW−1 and − (10.93 ± 0.6) × 10–2 cm·GW−1 at the irradiance intensity of (26.4 ± 1.5) GW·cm−2 for MoS2 and Ag/MoS2 suspensions (Table 2), respectively. The βeff for Ag/MoS2 nanocomposites was almost three times that of MoS2 suspensions, and two times that of AgNPs(25) with an average SA coefficient of − (3.71 ± 0.6) × 10–2 cm·GW−1. Combined with previous reports, the NLO enhancement compared with single components can be ascribed to the SPR of AgNPs and the charge transfer between AgNPs and MoS2 [50].

4 Conclusion

In summary, we investigated the NLO performances of the fabricated AgNP colloids and realized the tunable NLO performances from TPA to SA. The linear optical spectra results demonstrate that only a typical SPR band is observed at ~ 400 nm for AgNPs(0) and AgNPs(15) colloids with low ethanol amount. When increasing the ethanol amount, another longitudinal SPR mode occurs for sample AgNPs(25) at ~ 520 nm. For the ultrafast NLO investigation, the optical limiting effect for sample AgNPs(15) was switched into SA effect for sample AgNPs(25). The controllability can be attributed to their different SPR frequencies which induce dominant TPA occurrence for AgNPs(15) and SA for AgNPs(25). The prominent SA behavior is caused by the strong longitudinal resonant absorption near the excitation wavelength at 515 nm. Additionally, we controllably synthesized nanocomposites of Ag/MoS2 utilizing the resonant absorption near the excitation wavelength at 515 nm, which exhibited enhanced NLO property, almost three times that of MoS2 suspensions. The present results are beneficial to in-depth understanding of the ultrafast NLO absorption properties of AgNPs and to promote the practical applications of AgNPs and its nanocomposites in different fields.