Laser-Synthesized Plasmono-Fluorescent Si-Au and SiC-Au Nanocomposites for Colorimetric Sensing

Abstract

Sensing represents one of the most rapidly developing areas of modern life sciences, spreading from the detection of pathogenic microorganisms in living systems, food, and beverages to hazardous substances in liquid and gaseous environments. However, the development of efficient and low-cost multimodal sensors with easy-to-read functionality is still very challenging. In this paper, stable aqueous colloidal suspensions (ζ-potential was between −30 and −40 mV) of ultrasmall (~7 nm) plasmonic Si-Au and SiC-Au nanocomposites were formed. Two variants of pulsed laser ablation in liquids (PLAL)—direct ablation and laser co-fragmentation—were used for this purpose. The co-fragmentation approach led to a considerable decrease in hydrodynamic diameter (~78 nm) and bandgap widening to approximately 1.6 eV. All plasmonic nanocomposites exhibited efficient multi-band blue emission peaking at ~430 nm upon Xe lamp excitation. Co-fragmentation route considerably (~1 order of magnitude) increased the PL efficiency of the nanocomposites in comparison with the laser-ablated ones, accompanied by a negligible amount of dangling bonds. These silicon-based nanostructures significantly affected the optical response of rhodamine 6G, depending on the synthesis route. In particular, directly ablated nanoparticles revealed a stronger influence on the optical response of dye molecules. The observed findings suggest using such types of semiconductor-plasmonic nanocomposites for multimodal plasmonic and colorimetric sensing integrated with luminescent detection capability.

1. Introduction

Different inorganic and organic semiconductor (nano-)structures have unique structural and optoelectronic properties, making them suitable for nanomedical, organic, and/or flexible electronic, light-emitting, photovoltaic and other applications [1,2,3,4,5,6,7,8]. Moreover, semiconductor nanostructures can be widely used for colorimetric sensing based on changes in their absorbance and/or fluorescence properties. In particular, metal-oxide semiconductor-based colorimetric sensors can detect a small leakage of industrially important gases (ammonia, methylamine, trimethylamine, and hydrazine) coupled with immediate smartphone notification [9]. Organic sensors can effectively reveal hazardous substances, such as chemical warfare agents, volatile organic compounds, or hydrogen sulphide [10]. Semiconductor optical wavelength sensors can also be employed for monitoring the quality of flowing water [11]. Colorimetric sensors are also promising for clinical applications by detecting biomarkers, which have several advantages, such as rapid detection and naked-eye determination [12]. Nevertheless, these sensors based on pristine semiconductor nanostructures often have some drawbacks, such as limited sensitivity, spectral instability, and insufficient signal reproducibility in complex biological environments. Creating semiconductor-metallic nanocomposites (NCs) with a controllable chemical composition might develop more reliable platforms for biomedical sensing and diagnostics by overcoming drawbacks via the plasmon-enhanced optical response, improved biocompatibility, and contamination-free synthesis.

Pulsed laser ablation in liquids (PLAL) is one of the simplest and versatile methods for synthesizing different single- and multi-element nanostructures using high-purity chemical conditions. It is mainly based on direct irradiation of a solid target immersed in a liquid—so-called direct laser ablation (LA)—by a continuous wave (cw) or pulsed laser, resulting in the formation of nanoparticles (NPs). Their functional properties are easily controlled by a wide set of different experimental parameters such as irradiation wavelength, laser fluence, laser pulse width, irradiation duration, targets, solvents and solutes, etc. [13,14,15,16,17,18,19,20,21,22]. In addition, another PLAL-based synthesis method is devoted to the laser irradiation of colloidal suspensions consisting of homogeneously dispersed micro-/nano-particles—laser fragmentation (LF) [23,24]. This approach is less sensitive to the precision of the focus position, leading to a better reproducibility of the NP synthesis, excluding attention to the target thickness changes. Moreover, it can also result in a greater NP yield due to the formation of supercontinuum irradiation because of the interaction of ultrashort laser pulses with water [25]. Furthermore, PLAL allows easy formation not only of various bimetallic NPs [26,27,28,29,30] but also metallic-semiconductor nanostructures [31,32,33,34]. Therefore, the performance of semiconductor nanostructures can be considerably improved by using various PLAL-based synthesis aimed at multimodal sensing applications.

In this paper, stable colloidal suspensions of silicon-based nanocomposites were synthesized using various approaches of the PLAL technique, such as direct laser ablation and laser co-fragmentation. All the multi-element nanostructures revealed strong plasmonic properties suitable for surface-enhanced improvement of optical signals compared to the pristine semiconductor ones. Gold-containing nanocomposites exhibited strong blue emission upon UV lamp excitation, whose efficiency depended on both (i) the used PLAL-based synthesis and (ii) the used semiconductor material. The bandgap of all the nanocomposites formed by the LcF approach increased as compared to (i) that prepared by the direct LA approach and (ii) pristine semiconductor NPs. The interaction between the plasmono-semiconductor nanostructures with organic dye molecules suggested their use for absorbance/fluorescence colorimetric sensing applications, which can be accompanied by plasmonic sensing and fluorescent detection.

2. Materials and Methods

Semiconductor-metallic nanocomposites were formed by means of the two following approaches of the PLAL technique:

(1)

Direct laser ablation (LA) of a gold target immersed in colloidal suspensions of semiconductor NPs;

(2)

Laser co-fragmentation (LcF) of the mixture of the semiconductor and metallic NPs colloidal suspensions.

In both cases, radiation of a picosecond laser source (1030 nm wavelength, 6 ps pulse duration, 50 µJ/pulse laser energy, 10 kHz repetition rate) was used. The laser beam was focused on the surface of the targets located under 4 mm water level or 4 mm below the air/water interface in the case of the LcF approach. Laser radiation was scanned over a 10 × 10 mm² surface for 5 min. Firstly, the colloidal suspensions of Si NPs, SiC NPs, and Au NPs were prepared by ablation of silicon, silicon carbide, and gold targets, respectively. Then, the gold target was ablated in the Si NPs or SiC NPs colloidal suspensions to form Si-Au and SiC-Au NPs, respectively. In the case of the LcF, the colloidal suspensions of Si or SiC NPs were mixed with Au NPs solutions in an equal volume ratio and irradiated under similar conditions.

The nanoparticles deposited on a copper-based TEM grid were visualized by a transmission electron microscope (TEM) working at a 300 kV accelerating voltage. The amount of NPs was calculated using the Image J software (version 1.54p) for over 500 particles. To assess the chemical stability of the colloidal suspensions of the NPs, their zeta (ζ) potential values were measured using a Malvern Zetasizer Ultra (Malvern Panalytical, Malvern, Worcestershire, United Kingdom) (laser Doppler electrophoresis, 633 nm, Smoluchowski approximation). The reported zeta values were directly calculated from the electrophoretic mobility. Their hydrodynamic diameters were investigated using multi-angle dynamic light scattering (MA-DLS) at the following angles: 13°, 90°, 173°. The dispersant was water (viscosity 0.8872 mPa·s, 25 °C). Both cumulant analysis (for PDI) and CONTIN (for distribution) were employed.

Optical properties of the NCs were investigated by UV-Vis and photoluminescence (PL) spectroscopies. The absorbance spectra were recorded by a Shimadzu-260 device (Shimadzu Corporation, Kyoto, Japan) in the 350–800 nm range using a 2 nm slit and a 2 s acquisition time at a medium scanning speed. The PL spectra were collected in the spectral range 350–750 nm by means of a customized Edinburgh FLS1000 spectrometer (Edinburgh Instruments Ltd., Livingston, Scotland, UK) equipped with a Xenon arc lamp (450 W). The spectra were detected using a 1 s acquisition time, 5 scans, and a 2 nm step. To study the interaction between the NCs and dye molecules, different volumes of the as-prepared colloidal suspensions (10–250 µL) were added to 1.5 mL rhodamine 6G solution (concentration was 1 µM). The aforementioned optical measurements were performed three times using aqueous colloidal suspensions (pH = 5) of the NCs with 0.15 g/L initial concentration. For studying the absorbance properties and the hydrodynamic diameters of the NCs, the colloidal suspensions were diluted 10-fold to avoid any concentration-related distortions.

To estimate bandgaps of the semiconductor-based NCs, the experimentally obtained absorbance spectra were replotted according to the Tauc plot equation for semiconductor nanostructures with indirect allowed transitions [35]:

αhυ~(hυ − E_g)²

where α is an absorption coefficient, υ is the light frequency, and E_g is the optical bandgap value. To extract the values of the bandgap of the NCs, linear regions of (αhν)^1/2 vs. hν dependencies were intersected with the y = 0 axis.

3. Results and Discussion

To visualize the shape and size of the laser-synthesized plasmonic silicon-based NCs, they were investigated by means of TEM (Figure 1). One can see quite homogeneous size distributions across all types of NPs prepared by both direct LA and LcF approaches, which have a perfectly spherical shape (Figure 1). To reveal the structure of the laser-synthesized NCs, their diffraction patterns were obtained by means of TEM (Figure 2). Quite similar images related to the nanostructures prepared by the LA and LcF approaches were observed (Figure 2). Indeed, the NCs formed by the direct LA approach showed several distinct rings related to different crystalline planes. Both Si-Au and SiC-Au NCs synthesized by the direct LA had four diffraction rings associated with the following crystalline planes of gold: (111), (200), (220), and (311). It also correlated well with (111), (200), (220), (311), and (222) gold crystalline planes for the Si-Au NCs prepared at different laser ablation times, as it was recently justified by X-ray diffraction (XRD) technique [36]. At the same time, the LcF-formed nanostructures demonstrated less pronounced crystalline structure in comparison with the LA ones. Thus, it indicated that different structures of the NCs were formed by these two different methods, which might be related to different mechanisms of NC synthesis. Here, one can hypothesize that direct ablation can result in a larger number of gold nanoclusters released from the solid target. Their further agglomeration can result in the formation of different nanocrystalline gold structures merged with some silicon atoms. Moreover, the silicon nanoclusters can also interact with the dissolved oxygen molecules. As a consequence, according to the X-ray photoelectron spectroscopy (XPS), different silicon (sub-)oxides and even gold oxide can be formed upon high-energy ultrashort pulses [36]. Contrary to this, the mixture of the semiconductor and metallic NPs irradiated by the same laser pulses can consist of homogeneously dispersed nanoclusters of different elements. Thus, it can lead to the restriction of the long-range order, considerably limiting the size of the formed gold nanocrystals.

To determine the size distributions of all NCs prepared by PLAL-based synthesis, more than 500 NPs from the TEM images were analyzed for each sample. As a result, the NCs formed by the various PLAL approaches revealed diverse behaviour for the different types of NCs (Figure 3). In the case of the direct LA, Si-Au NCs had a lower mean size (~8 nm) compared to SiC-Au NCs (~11 nm) (Figure 3a), while similar mean sizes (~6.7 nm) with identical size distributions were found for both Si-Au and SiC-Au NCs prepared by the LcF approach (Figure 3b). Confirmation of the dual semiconductor-metallic composition of these nanostructures was carried out previously by means of the energy-dispersive X-ray spectroscopy (EDX) [37]. It was shown that all types of such NCs revealed a response from silicon and gold in their EDX spectra, depending on the NP size [37]. Considering the aforementioned mean size values (Figure 3), one can obtain the following amounts (in atomic %) of the plasmonic metal in the NCs:

LA, Si-Au:    43%

LA, SiC-Au:  68%

LcF, Si-Au:    14%

LcF, SiC-Au: 33%

The gold content in the NCs should also considerably change the optical and structural properties of the initial pristine semiconductor nanostructures (Figure 4). To study the absorption efficiency of the composite NPs, they were investigated by means of UV-Vis spectroscopy in comparison with the pristine Si and SiC NPs. Here, one can see no plasmonic features for both pristine Si and SiC NPs and their appearance for all NCs due to merging with the plasmonic element (Figure 4a,b). Their behaviour had a similar tendency to the aforementioned size distributions (Figure 3). The experimentally observed stronger plasmonic properties of SiC-Au NCs compared to Si-Au ones (Figure 4a,b) can directly point to a larger amount of gold in the NCs [36,38].

In order to explain the aforementioned modifications of the size distributions, chemical composition, and optical properties, one can suggest the following hypothesis. During the first step, the colloidal suspensions of the pristine semiconductor Si and SiC NPs are formed by the direct LA. Their concentration and size may differ due to various (i) absorption coefficients of the corresponding bulk materials [39] as well as (ii) laser ablation threshold values at similar laser ablation conditions (~2.7 J/cm² and ~1.0 J/cm² for SiC and Si, respectively) [40,41]. Thus, the SiC NPs can have a larger size and concentration that might affect the further synthesis of the composite nanostructures.

During the second step, using the LA approach, gold nanoclusters are released from the solid target being irradiated by the high-energy ultrashort laser pulses. Additionally, the latter also affects the semiconductor NPs located close to the focal plane, leading to their decomposition into the corresponding nanoclusters. A stronger absorbance of the Si NP colloidal suspensions in comparison with the SiC ones (Figure 4a,b) results in the more prominent light attenuation and, hence, a smaller laser fluence reaching the target surface. Consequently, it leads to the experimentally observed different quantities of the gold nanoclusters released from the solid target in different cases: smaller for Si NPs and greater for SiC NPs. A larger amount of the nanostructured gold is responsible for the stronger plasmonic properties of the SiC-Au NCs synthesized by the direct LA approach (Figure 4a). Their smaller mean sizes as compared to the Si-Au nanostructures can be associated with a greater concentration of both semiconductor and metallic nanoclusters in the colloidal suspensions. As a result, a larger amount of the nanoclusters in the colloidal suspensions can lead to a strong restriction of the further regrowth of the NPs, similar to a previously published case, where the interaction between nanoclusters and different bioproteins was investigated [42]. It is worth noting the pronounced silicon oxidation due to the laser-induced interaction between silicon nanoclusters and dissolved oxygen molecules. Moreover, the experimental conditions strongly affected the oxidation states of the formed NCs [36].

In the case of the LcF approach (second step), the mechanisms of the laser-assisted physicochemical processes conditioning the growth of the NC might be as it was proposed previously [37]. Here, the colloidal suspensions (also affected by magnetic stirring) have a homogeneous distribution of both semiconductor and metallic NPs. Being affected by laser irradiation, they create a larger amount of smaller plasma plumes (containing nanoclusters of the corresponding elements) in front of the NPs, similar for both Si- and SiC-based colloids. Thus, the conditions of the laser-induced decomposition of the semiconductor and metallic NPs and further merging of the created nanoclusters can be identical. As a result, both Si-Au and SiC-Au NCs formed by the LcF have similar structural and optical properties. In particular, the same gold concentration in the initial colloidal suspensions used for the synthesis of different NCs by the LcF approach also results in similar plasmonic properties of both Si-Au and SiC-Au nanostructures (Figure 4b). Moreover, smaller dimensions of the plumes can reduce the time of the interaction between the involved chemical elements (including dissolved oxygen), contrary to the LA case. Hence, the size of the formed NCs is smaller than that in the case of the direct LA. It also might decrease the laser-induced oxidation of the nanostructures, which will be studied separately by the X-ray photoelectron spectroscopy (XPS).

Despite using the same amount of Au NPs for the LcF synthesis of the NCs, a different relative gold content was observed in the Si-Au and SiC-Au NCs. It might be due to the laser-assisted changing of the NP structure affecting the concentration of the SiC NPs. Although their initial concentration exceeded that of the Si NPs, the high-energy laser pulses could burn out carbon atoms, thus reducing the total amount of the semiconductor atoms. Hence, the concentration of the silicon atoms in the SiC-Au NCs can be less in the Si-Au ones, giving a greater gold contribution. The achieved findings show a possible way of the unification of the size distributions of the semiconductor-metallic NCs formed by PLAL-based synthesis, highlighting different mechanisms of the NC growth.

To estimate the bandgap of the semiconductor-based NPs, their experimentally obtained absorbance spectra were replotted using the Tauc plot method [35]. The intersection of the linear part of the spectra with the y = 0 axis gave corresponding bandgap values (Figure 4c,d). Quite low, similar bandgap values were observed in the case of pristine Si and SiC NPs (~1.1 eV) (

Table 1. Results of TEM, MA-DLS, and ζ-potential measurements, as well as of bandgap and plasmonic maxima measurements.

	Laser Ablation		Laser Co-Fragmentation
	Si-Au NCs	SiC-Au NCs	Si-Au NCs	SiC-Au NCs
Mean size (nm)	10.5 ± 1.1	8.1 ± 0.9	6.8 ± 0.7	6.6 ± 0.7
H. diameter (nm) (intensity-weighted)	104 ± 9	104 ± 9	86 ± 8	78 ± 7
H. diameter (nm) (number-weighted)	73 ± 8	73 ± 8	56 ± 6	51 ± 5
Count rate (kpcs)	240	220	230	250
PDI	0.22	0.24	0.19	0.188
Ζ-potential (mV)	−33 ± 3	−35 ± 3	−39 ± 4	−40 ± 4
Bandgap (eV)	1.06 ± 0.1	1.06 ± 0.1	1.58 ± 0.1	1.60 ± 0.1
Plasmonic maximum (nm)	525 ± 2	525 ± 2	523 ± 2	523 ± 2

), which were similar to the bandgap of bulk silicon. However, it was considerably lower than the bandgap values of different SiC polytypes (2.36 eV for 3C, 3.02 eV for 6 H, and 3.26 eV for 4 H) [43] marked in Figure 4c,d. It might also be explained by the possible aforementioned burning out of the carbon atoms, resulting in the formation of Si-like nanostructures.

The remarkable difference in the bandgap values of the NCs prepared by the LA and LcF approaches can be associated with their different sizes. Indeed, both Si-Au and SiC-Au NCs formed by the LcF approach showed quite low mean sizes (~5–6 nm) (Figure 3b). The increase in the bandgap for the LcF prepared NCs (~1.6 eV) (Table 1) might be due to the quantum confinement effect, which has been very well studied for porous silicon nanostructures of different sizes [44,45]. Thus, PLAL-based synthesis allows easy engineering of the size and bandgap of the semiconductor-based NCs.

To check chemical stability and hydrodynamic diameter of the single- and double-element NPs, they were analyzed using the DLS technique. Good stability of the single-element NPs having ζ-potential values from −30.2 mV (semiconductor NPs) to −37.8 mV (Au NPs) (Table 1) was observed (Figure 5a,b). Further formation of the NCs from the semiconductor NPs due to the direct LA slightly increased the ζ-potential values (−35.1 mV for SiC-Au NCs), while the LcF approach increased it more significantly (−39.5 mV for Si-Au NCs) (Figure 5a,b). The larger ζ-potential values can be due to a greater amount of charge carriers provided by the plasmonic metal, ensuring better stability of the colloidal suspensions. The laser-synthesized colloidal suspensions of the plasmonic NCs demonstrated good temporal stability: they were stable for several (at least 6) months without any additional surface treatment.

The hydrodynamic diameter of the composite NPs also showed different behaviour depending on the used PLAL approach (Figure 5c,d). All the samples exhibited narrow particle size distributions with polydispersity index (PDI) values in the range 0.18–0.24 and count rates of 220–260 kcps (Table 1). Slightly smaller diameter was observed for the NCs formed by direct laser ablation (~7% for SiC-Au NCs), while the co-fragmentation route resulted in a greater difference, reaching ~32% for SiC-Au NCs (Figure 5c,d). This discrepancy can be related to the different mechanisms of NP synthesis, as was proposed previously [37]. The laser co-fragmentation approach can lead to the destruction of all semiconductor and metallic NPs affected by the high-energy laser spot. Their further interaction can be significantly time-restricted due to the changed dimension and lifetime of the cavitation bubbles mentioned above, which served as “reactors” for nanoparticle growth, influenced by the presence of other elements. As a result, this approach can considerably restrict the growth of the plasmonic-semiconductor NCs. Contrary to this, already dispersed semiconductor NPs can be just slightly affected by direct laser irradiation, leading to their fragmentation. Further interaction with the gold nanoclusters ablated from the solid target will again increase the size of the formed NCs. Thus, one can control both chemical stability and hydrodynamic diameter of the laser-synthesized NCs by choosing an appropriate laser processing approach.

To assess the emission efficiency of the laser-synthesized plasmonic NCs, they were studied by means of the PL spectroscopy. Their PL was excited using a 450 W Xe lamp through an emission monochromator (310 nm). Firstly, significantly stronger (~1 order of magnitude) luminescence from the nanostructures prepared by the LcF approach, as compared to the LA one, was observed (Figure 6a,b). Secondly, stronger PL of Si-Au NCs formed by the LA approach also correlated well with their smaller size compared to SiC-Au NCs, which gave a larger amount of radiative transitions. All the spectra can be deconvoluted using several Gaussian curves, evidencing several emission bands at around 375, 400, 430, 480, and 590 nm (Figure S1). Additionally, a small PL contribution at ~590 nm was resolved for the LA-synthesized NCs. Thus, several radiative recombination channels might be responsible for their emission, which was also affected by the plasmonic metal.

The studied NCs are a complex material consisting of semiconductor (Si; C) and metallic (Au) elements coupled with oxygen molecules. Concentration of each element as well as their relative positions might strongly influence the structure of the final NPs, reflected in the number of the PL bands, their intensities, and spectral positions. Hence, the recombination and radiation processes in such laser-synthesized NCs might be very complex procedures [46]. Their PL can range from a red/near red band (S-band with µs lifetime) to a blue/green one (F-band with ns lifetime) [47]. Thus, the main PL mechanisms of the studied nanostructures can occur via the quantum confinement effect (QCE) [46,48] or surface and interface states [46,49]. Moreover, the synthesis of the NCs using the high-power laser pulses might also result in the formation of different point defects like oxygen deficiency centres (ODCs) [46,50] or a nonbridging oxygen hole centre (NBOHC) accompanied by the formation of paramagnetic dangling bonds [46,51].

In the case of QCE, PL is mainly observed for nanocrystals smaller than 5 nm in the red spectral range, whose spectral position and efficiency are considerably dependent on their size [46]. However, in our case, the emission is mainly detected in the blue-violet spectral range with some orange PL (Figure S2), thus, excluding this mechanism. Nevertheless, the observed PL contribution at 590 nm might be associated with the so-called Y-band, whose efficiency is low due to the effective depletion of the number of excited electrons through electron transfer and intraband relaxation [47]. Another mechanism of this PL band can also be related to the E_c → Si⁰ transition [52,53]. Taking into account the NC synthesis in the oxygen-saturated aqueous medium, they can represent a complex system consisting of nanocrystalline silicon and silicon oxide of different stoichiometry [36]. Hence, the most probable mechanisms of the silicon-based NC photoluminescence in the 350–480 nm range might be mainly related to the direct recombination of the photoexcited charge carriers at the interface of the crystalline silicon nanoclusters and oxide shell via the different surface and interface states in the Si/SiO_x system [54]. In particular, some of the aforementioned bands can be due to the following transitions: Si⁰ → E_v (~430 nm) and Si⁰ → Si–O–Si (~480 nm) [55]. The transition at 400 nm might be attributed to two-fold coordinated silicon lone-pair centres [54].

Since all the NCs also contain a plasmonic metal, the latter can also considerably affect the observed PL spectra. Their spectral position and intensity can be changed for the following reasons:

• Surface-enhanced plasmonic enhancement leading to a higher PL intensity [52,56];
• Metal-induced photoluminescence quenching [57,58];
• Bandgap narrowing due to formation of gold acceptor (E_c −0.55 eV) and donor (E_v +0.35 eV) states [59].

Based on our previous study on the influence of plasmonic metals on the properties of fluorescent carbon dots (CDs) [60], one can mainly assume the metal-induced photoluminescence quenching. Indeed, increasing the metal concentration in the Ag-CDs and Au-CDs nanostructures due to longer laser ablation resulted in a weaker luminescence response [60]. A main reason for that can be related to a considerable concentration (~150 µg/mL) of gold in the studied NCs [36]. Moreover, relative localization of the elements in the NCs might strongly affect the distribution of local electrical fields and the number of radiative and nonradiative recombination channels. Thus, it requires a detailed study of the NC structure and computer simulation (e.g., FDTD) to establish the impact of different competitive mechanisms.

The main reasons for the observed ~1 order of magnitude difference in the PL intensity of the NCs prepared by the different approaches (LcF vs. LA ones) might be associated with the different mechanisms of the NP synthesis, which were proposed above. It is assumed that the LcF-prepared NCs consist of different elements (Si; Au; O) homogeneously distributed in the NCs. It might be favourable for the (i) creation of hot-spots due to stronger local electrical fields provided by the Au nanoclusters and (ii) better bonding between the involved elements, leading to fewer amounts of dangling bonds. The latter assumption is based on the preliminary results of our ongoing study of the NCs using X-band electron paramagnetic resonance (EPR) spectroscopy (Figure S2). The corresponding EPR results will be reported separately. A significant amount of the dangling bonds in the LA-synthesized NCs was found, which could serve as nonradiative recombination channels. At the same time, the LcF-prepared nanostructures have an insignificant concentration of the paramagnetic defects, confirming the aforementioned hypothesis. The obtained findings can provide new insights into strategies for the development of novel semiconductor-metallic NCs by means of PLAL with the required optoelectronic properties for further biomedical applications.

To assess the capability of the laser-synthesized plasmonic silicon-based NCs for colorimetric sensing, the optical properties of the Rh6G organic dye molecules (1 µM) were investigated depending on the amount of added plasmonic NCs (10–250 µL). Here, similar tendencies were found for absorbance, PLE, and PL spectra of Rh6G for all NCs regardless of their composition (Figure 7, Figure 8 and Figure 9). Firstly, a small drop (10 µL) of any plasmonic NC reduced (~5–10%) the absorbance intensity of Rh6G (Figure 7). Then, the increase in the colloids volume (up to 250 µL) resulted in a stronger absorbance of Rh6G (~15–30% compared to pure Rh6G) (Figure 7). At low concentrations of the colloidal suspensions of the Si–Au and SiC–Au NCs (10 µL), the absorption of Rh6G decreased by ~5–10%, which can be attributed to the partial adsorption of dye molecules onto the NP surface and local changes in the dielectric environment. Upon increasing the concentration (250 µL), the absorption intensity increased by ~15–30% compared to the pristine dye. This enhancement can originate from the spectral overlapping between the absorption maxima of Rh6G (525 nm) and the plasmonic band of the NCs (523−525 nm), which can lead to localized surface plasmon resonance (LSPR)-induced field amplification and increased transition probability for the dye.

To better clarify the interaction mechanisms between Rh6G molecules and Si–Au/SiC–Au NCs, four main contributions were distinguished:

(i)

A minor hypsochromic shift (<2–3 nm) of the Rh6G absorption and emission bands indicates only a slight modification of the local dielectric environment around the dye;

(ii)

Competitive absorption between Rh6G (~525 nm) and the Au plasmon band (523–525 nm) leads to an initial decrease in intensity at low nanoparticle concentrations;

(iii)

Plasmon–dye energetic coupling, including possible resonance energy transfer or exciton–plasmon interactions, might be expected due to a noticeable decrease in the excitation spectra (350–550 nm) without significant peak shifts;

(iv)

Physisorption of Rh6G on negatively charged nanocomposite surfaces results in intensity changes without strong spectral deformation, suggesting enhanced local dye concentration rather than chemical degradation.

The observed optical response resulted from the balance between these processes, which strongly depended on both the synthesis method (LA vs. LcF) and the material composition (Si vs. SiC).

Secondly, the Rh6G PLE efficiency decreased (almost twice) with the increase of the NCs volume (Figure 8). This behaviour is consistent with the competitive photon absorption and scattering by the plasmonic NCs, which reduces the number of photons available for the direct excitation of the dye molecules. In addition, nonradiative energy transfer from the excited dye states to the plasmonic modes of the NCs may further quench the excitation efficiency, thereby lowering the overall PL excitation response.

All plasmonic NCs also led to a weaker PL intensity of the Rh6G dye molecules, which correlated well with the concentration-dependent PLE spectra (Figure 9). The Rh6G PL decreased in the presence of both Si–Au and SiC–Au NCs colloids, irrespective of the fabrication approach. The observed quenching can be explained by nonradiative energy transfer processes, where excited dye molecules couple to nanoparticle surface states or plasmonic modes, thus providing efficient nonradiative relaxation channels. While localized plasmon fields are expected to enhance emission under certain conditions, in this case, the dominating effect is energy dissipation via plasmon–dye interactions, resulting in a general decrease in luminescence intensity.

The observed findings highlight the critical role of the NP concentration, composition, and spectral overlapping in tuning dye–nanoparticle interactions. These results can be exploited to optimize hybrid plasmonic–semiconductor nanosystems for practical applications, including optical sensors with adjustable sensitivity, bioimaging probes with controlled emission properties, and plasmon-assisted photonic platforms for biomedical diagnostics.

4. Conclusions

In conclusion, the stable aqueous-based colloidal suspensions (−39.5 mV for Si-Au NCs) of the Si-based NCs were synthesized by different PLAL approaches. The LcF one demonstrated smaller NP mean sizes (~6.7 nm) and larger bandgap (1.6 eV) values, which were similar for both Si-Au and SiC-Au NCs. It also resulted in their stronger PL efficiency (~1 order of magnitude for SiC-Au NCs) as compared to the NCs synthesized by direct LA, which can be related to the negligible amount of dangling bonds. All the Si-based NCs had strong plasmonic properties, which can be used for various surface-enhanced plasmonic sensing applications. Additionally, they also demonstrated good capability for colorimetric sensing and photocatalytic sensing applications using Rh6G organic dyes, respectively. These findings open prospects for designing cost-effective hybrid nanosensors with controllable chemical composition for multimodal sensing applications.