Multiwavelength SERS of Magneto-Plasmonic Nanoparticles Obtained by Combined Laser Ablation and Solvothermal Methods

The present study introduces a novel method for the synthesis of magneto-plasmonic nanoparticles (MPNPs) with enhanced functionality for surface-enhanced Raman scattering (SERS) applications. By employing pulsed laser ablation in liquid (PLAL) to synthesize plasmonic nanoparticles and wet chemistry to synthesize magnetic nanoparticles, we successfully fabricated chemically pure hybrid Fe3O4@Au and Fe3O4@Ag nanoparticles. We demonstrated a straightforward approach of an electrostatic attachment of the plasmonic and magnetic parts using positively charged polyethylenimine. The MPNPs displayed high SERS sensitivity and reproducibility, and the magnetic part allowed for the controlled separation of the nanoparticles from the reaction mixture, their subsequent concentration, and their precise deposition onto a specified surface area. Additionally, we fabricated alloy based MPNPs from AgxAu100–x (x = 50 and 80 wt %) targets with distinct localized surface plasmon resonance (LSPR) wavelengths. The compositions, morphologies, and optical properties of the nanoparticles were characterized by using transmission electron microscopy (TEM), scanning electron microscopy (SEM), X-ray diffraction (XRD), UV–vis spectroscopy, and multiwavelength Raman spectroscopy. A standard SERS marker, 4-mercaptobenzoic acid (4-MBA), validated the enhancement properties of the MPNPs and found an enhancement factor of 2 × 108 for the Fe3O4@Ag nanoparticles at 633 nm excitation. Lastly, we applied MPNP-enhanced Raman spectroscopy for the analysis of the biologically relevant molecule adenine and found a limit of detection of 10–7 M at 785 nm excitation. The integration of PLAL and wet chemical methods enabled the relatively fast and cost-effective production of MPNPs characterized by high SERS sensitivity and signal reproducibility that are required in various fields, including biomedicine, food safety, materials science, security, and defense.


INTRODUCTION
−5 These nanoparticles (NPs) are exploited in various fields of biomedicine, sensing, catalysis, light harvesting, magnetic resonance imaging, and surface-enhanced Raman scattering (SERS) applications. 6,7SERS is a label-free technique that delivers high specificity and sensitivity for molecules adjacent to plasmonic surfaces.It provides in-depth information on the chemical bonds and enables a comprehensive understanding of the molecular structure.SERS is based on the localized surface plasmon resonance (LSPR) effect, which is the collective oscillation of free-electron plasma excited by an incident electromagnetic radiation.−10 Magnetism also allows for the spatial manipulation of nanoparticles, facilitating their targeted placement and concentration on a specific surface area. 11here are various methods, broadly categorized as bottomup (synthesis from atoms) and top-down (synthesis from the bulk material), for the generation of nanoparticles.−16 PLAL is a rapid, versatile, and simple method that uses no additional chemicals and results in in situ, ultrapure nano-particle dispersion. 17In PLAL, the target material is submerged in a liquid medium and subjected to high-intensity pulsed laser irradiation, which leads to the rapid melting and vaporization of the material, followed by the formation of a cavitation bubble.The collapse of the cavitation bubble results in the ejection of molten or vaporized material into the surrounding liquid, where it condenses into nanoparticles. 18,19PLAL offers partial control over the nanoparticle size, size distribution, shape, and aggregation through the modulation of the parameters, including the laser wavelength, pulse width and rate, and fluence. 12Additional factors such as the liquid medium, target, and external fields also strongly affect the final result. 13,20,21he ablation medium plays a crucial role in regulating the product outcome.A chemical analysis of Au nanoparticles produced by PLAL in saline solution revealed the presence of halide ions on the nanoparticle surfaces, supporting a hypothesis that halogen anions contribute to colloidal stability by carrying a negative charge. 22−25 Moreover, even certain cation−anion pairs, specifically LiCl, have been recognized as particularly effective in increasing the SERS signal. 26Hot-spot formation alone cannot be attributed to the increased signal, as the chlorides have also promoted the SERS signal for the roughened Ag surface. 26Halides form surface complexes characterized by low-lying excited states that readily undergo photoinduced charge transfer from the target molecule and contribute to SERS enhancement via a chemical enhancement mechanism. 27They also disturb the electric double layer and change the electrostatic interactions between the target molecule and the surface.
There are several approaches for generating hybrid bimetallic nanoparticles by PLAL, e.g., the sequential ablation of two metal targets in the same liquid medium, which leads to the mixing and coalescence of two different material nanoparticles, although the quality of the process may sometimes be limited. 28−32 However, the alloying and dual structure formation are challenging to manage, and PLAL more often results in a mixture of different nanoparticle types, which translates into poor SERS enhancement.For example, MPNPs prepared by Fe and Au multilayer laser ablation exhibited an enhancement factor of 5.8 × 10 4 for the 4-mercaptobenzoic acid (4-MBA) reporter molecule. 31Moreover, in a bimetallic alloy having Au/Ag and Fe components, the LSPR was found to sharply decline with increasingly higher Fe concentrations (above approximately 15%).Therefore, only a narrow concentration window of plasmonic material is practical for SERS. 33,34Another issue with MPNP production using PLAL is that the ablation of an iron target is limited to organic solutions because ablation in water results in the formation of weak magnetization iron oxides, such as maghemite. 32Organic solutions, however, contaminate the plasmonic nanoparticles with carbon species originating from the laser-induced thermal decomposition of the solvent. 31n this study, we propose an MPNP synthesis method in which the plasmonic part is obtained by PLAL in saline water and the magnetic part is obtained by a wet chemistry synthesis method.Polyethylenimine mediates the nanoparticle attachment and formation of the hybrid MPNP structure.Such an approach allows one to obviate the use of additional reducing and stabilizing agents that are typically necessary for plasmonic NPs and achieve a chemically pure colloid material.The magnetic nanoparticle synthesis, in contrast, relies on inorganic reagents and polyethylenimine (PEI), whose function is to attach the magnetic and plasmonic parts, resulting in hybrid Fe 3 O 4 @Au or Fe 3 O 4 @Ag nanoparticles.The enhanced SERS sensitivity and signal reproducibility of MPNPs compared to those of conventional plasmonic nanoparticles have already been proven. 9,10,35In this study, the MPNPs were probed using multiwavelength SERS and the spectroscopic marker molecule 4-MBA.We also show the possibility of forming MPNPs from Ag x Au 100−x alloy targets (x = 50 and 80 wt %), which adds another dimension of fine-tuning the LSPR frequency according to one's needs.Finally, we apply MPNP-enhanced Raman spectroscopy to analyze the biologically relevant adenine molecule.

Synthesis of Plasmonic and Magneto-Plasmonic
Nanoparticles.Plasmonic nanoparticles were obtained by laser ablation in liquid of Ag, Au (Micro to Nano, Haarlem, Netherlands), Ag 80 Au 20 , and Ag 50 Au 50 (Testbourne Ltd., U.K.) targets (subscripts indicate wt %).A nanosecond laser (Ekspla, Lithuania) operating at a wavelength of 1064 nm with an average power of 7 W, a pulse duration of 10 ns, and and a 10 kHz repetition rate was used in the generation process.The diameter of the Gaussian beam at the surface was ∼100 μm (at the level of 1/e 2 ).The laser fluence was ∼17.8 J/cm 2 .The hatch (distance between the laser scan lines) was chosen to be 50 μm, and the scanning speed of the beam was 500 mm/s.Laser ablation was performed in various concentrations of aqueous KCl solution using a 40 mL capacity of Petri dishes.The ablated nanoparticles were immediately used for modification with magnetic nanoparticles.
The magnetic nanoparticle synthesis is described elsewhere. 9Briefly, 0.175 g of FeSO 4 •7H 2 O (Carl Roth GmbH, Karlsruhe, Germany) was dissolved in 20 mL of Ar-purged water.Then, 2.5 mL of 2 M KNO 3 (Reachem, Bratislava, Slovakia), 2.5 mL of 1 M NaOH (Carl Roth GmbH, Karlsruhe, Germany), and 5 mL of 8 mg/mL PEI (M w = 2500; Sigma-Aldrich, Steinheim, Germany) were added and magnetically stirred under a N 2 gas flow at 90 °C for 2 h.After the solution cooled, the magnetic nanoparticles were triple-washed with deionized water and decorated with either silver, gold, or silver−gold alloy nanoparticles according to the procedure described elsewhere. 9o obtain hybrid nanoparticles (MPNPs), 10 mL of the plasmonic nanoparticles was mixed with 0.8 mL of the Fe 3 O 4 colloidal solution.After a 45 min ultrasound treatment, the decorated nanoparticles (termed Fe 3 O 4 @Ag, Fe 3 O 4 @Au, Fe 3 O 4 @Ag 80 Au 20 , and Fe 3 O 4 @Ag 50 Au 50 ) were collected by using a strong magnet and washed with distilled water three times.The final colloid solution was resuspended to 5 mL.The magneto-plasmonic nanoparticle preparation method is illustrated in Scheme 1.All preparations were conducted using deionized water (18.2MΩ cm) from the Direct-Q 3 UV purification system (St.Louis, Missouri, U.S.).

Nanoparticle Characterization.
Thin Au films (∼150 nm) were magnetron-sputtered on microscope glass slides by using a PVD 75 system (Lesker, U.K.).Afterward, the slides were incubated in 4-mercaptobenzoic acid (4-MBA) (99%; Sigma-Aldrich, St. Louis, Missouri, U.S.) in an ethanol solution (0.2 mM) overnight.The MPNPs were deposited in a magnetic field on a 4-MBA self-assembled monolayer and allowed to dry.The magnetic field was removed prior to the SERS measurements.SERS was carried out using a multiwavelength inVia Raman spectrometer (Renishaw, Wottonunder-Edge, U.K.) equipped with a confocal Leica microscope and a thermoelectrically cooled (−70 °C) CCD (chargecoupled device) camera.Laser excitation sources were used in combination with diffraction gratings as follows: 442 nm (2400 lines/mm), 532 and 633 nm (1800 lines/mm), 785 nm (1200 lines/mm), and 830 nm (830 lines/mm).The laser radiation was focused on the sample using a 50×/0.75NA (numerical aperture) objective lens (Leica).SERS spectra were obtained by averaging 100 spectra from different sample locations with 25 s of integration each.SERS enhancement factors (EFs) for different laser excitations of the dominant 4-MBA spectral modes near 1073 and 1581 cm −1 were calculated according to a procedure detailed elsewhere 36,37 and by taking the geometric parameters of 4-MBA from the literature. 38In short, the EF calculation procedure was as follows.EF = (I SERS N R )/(I R N SERS ), where I SERS and I R are the intensities of the SERS and Raman spectral bands at 1585/1595 cm −1 , respectively, and N R and N SERS are the number of molecules excited in the Raman and SERS measurements, respectively.
The SERS spectra of adenine (≥99%; Sigma-Aldrich, St. Louis, Missouri, U.S.) were collected using a HyperFlux PRO Plus Raman spectrometer (Tornado Spectral Systems, Mississauga, Ontario, Canada) equipped with a thermoelectrically cooled CCD camera, a fiber-optic cable, and a 785 nm wavelength laser source that produced a 105 μm diameter spot on the sample.First, 3 μL of the as-prepared nanoparticle solution was dried on a steel substrate under an external magnetic field.Then, 3 μL of the desired concentration of an aqueous adenine solution was cast onto the top and dried.After that, the magnet was removed, and the samples were probed using a laser power of 70 mW at the sample (0.7 kW/ cm 2 ) and an accumulation time of 10−300 s.Several consecutive measurements were carried out and scrutinized on possible spectral changes to ensure there was no photothermal damage to the sample.

RESULTS AND DISCUSSION
3.1.Characterization of Plasmonic Nanoparticles.The plasmonic nanoparticles were prepared by laser ablation of Au, Ag, and two Ag/Au alloy targets in saline water.The rationale for using PLAL in saline water is as follows: halide ions (i) increase colloidal stability, preventing aggregation, (ii) contribute to increased SERS sensitivity, and (iii) help to displace weakly adsorbed species and prevent nanoparticle  (NP) surface fouling, leading to a more reproducible SERS signal.Figures 1 and 2 show the UV−vis and electron microscopy characterizations of the plasmonic NPs, respectively.The LSPR peak for the Ag NPs was observed at 394 nm and was independent of the Cl − concentration in the ablation medium.While a narrow size distribution, evidenced by the width of the LSPR peak, was visible for each saline waterablated NP batch, the ones ablated in pure water exhibited an asymmetric tail in the higher wavelength region, indicating a population of NPs with increased size or aggregation.
The Au NPs were more sensitive to the Cl − concentration than the Ag NPs.The detected peak blueshift was about 6 nm for the Au NPs when the chloride concentration in the solution was altered from 1 to 20 mM.We also observed a broadening of the absorption peak and an emergence of a second peak near 670 nm, related to the NP aggregates, in the case of 20 mM chloride.Increasing the concentration to 30 mM led to rapid NP aggregation and sedimentation (Figure S1), rendering a weak absorption spectrum.Hence, these NPs were excluded from further investigation.
The UV−vis spectra of the NPs obtained from Ag and Au alloy targets in pure water are presented in Figure 1B.Both compositions show only one LSPR peak, indicating there was no segregation into monometallic or core−shell NPs.The detected absorption maximum shift is consistent with the target composition.Interestingly, the alloy target NPs exhibited greater absorption peak full width at half-maximum (fwhm) values: specifically, the fwhm values were 73 and 136 nm for the Ag 80 Au 20 and Ag 50 Au 50 NPs, respectively, while they were 59 and 62 nm for the Au and Ag NPs, respectively.Such a broadening for bimetallic alloy NPs was observed previously. 39,40he nanoparticles were also characterized by using SEM and TEM techniques (Figure 2).As can be seen from the SEM images, the nanoparticles vary in size from several nanometers to 50−70 nm.However, the spherical form is maintained in most of the nanoparticles.From the TEM images, the size of the nanoparticles can be determined more precisely.Sizes range from 10 to approximately 100 nm, and it can be seen that the nanoparticles tend to aggregate.In addition to this, the nanoparticles seem to be homogeneous without separate core−shell or Janus-type compositions.This was also confirmed by energy dispersive X-ray (EDX) analysis.EDX analysis indicated that the Ag/Au alloy target composition was retained in the NPs, and no anisotropic separation of the two metals was detected (Figure S2).XRD measurements also confirmed the composition of the NPs.Although the crystal lattice parameters for these samples are similar, 4.0790 and 4.0786 Å, respectively, the peaks of the Ag 80 Au 20 and Ag 50 Au 50 NPs are slightly shifted (Figure S3).

Characterization of Hybrid Magneto-Plasmonic Nanoparticles Consisting of Magnetite and Gold or
Silver.The magnetic nanoparticles were synthesized using polyethylenimine (PEI), resulting in magnetite NPs with a positive surface charge, whereas plasmonic NPs are typically negatively charged.Thus, the formation of hybrid magnetoplasmonic NPs is governed by the electrostatic pull, and it was accomplished by simply mixing the magnetic and plasmonic NPs solutions.To ensure complete association, we used an ultrasonic bath for 45 min.Subsequently, the hybrid NPs were isolated by using a magnet and were washed three times with distilled water to remove any impurities.Figure 3A presents a bottle with the nanoparticles dispersed in water.When a magnetic field was applied, all of the nanoparticles were attracted to the side of the bottle, and only a transparent solute remained.Such an observation is evidence that the plasmonic nanoparticles were strongly attached to the magnetic nanoparticles, and they move together in the solution.This was also confirmed by the TEM image of the hybrid NPs (Figure 3B).The magnetic nanoparticles were uniform cubic structures that were 50 nm in size, while the plasmonic nanoparticles were spheres of various sizes.In the TEM image, dark structures represent the silver nanoparticles, while light gray cubic forms and surrounding plasmonic parts are the magnetic nanoparticles.In TEM images, the darkness of the color depends on the weight of the atom.Because the Ag atom is heavier than the Fe atom, a color difference appears.
The UV−vis spectra of the hybrid NPs (Figure 4) show less pronounced absorptions compared to the plasmonic NP spectra (Figure 1), mostly due to the increased light scattering by magnetite at shorter wavelengths.In general, due to interactions with magnetite, the plasmonic absorption peak tended to shift to longer wavelengths and broaden.For example, for the Ag and Au MPNPs, the maxima were found in the 400−500 nm range and near 550 nm, respectively, compared to 394 and ca.520 nm for the Ag and Au NPs, respectively.This shifting can be ascribed partially to the high dielectric constant of magnetite affecting the plasmonic properties of the Ag and Au NPs. 3 Unfortunately, the LSPR absorption peaks for the hybrid alloy NPs were very weakly pronounced.The wet chemical synthesis approach, outlined in our earlier work, generates magnetite cubes averaging around 50 nm in diameter. 9The TEM image in Figure 3B demonstrates a uniform magnetite distribution.Conversely, pulsed laser ablation in liquid results in plasmonic nanoparticles of a higher average size with a broader distribution.Hence, the resultant MPNPs comprise a magnetite nanoparticle network with multiple larger plasmonic nanoparticles trapped inside due to electrostatic interactions.This arrangement facilitates the formation of hot-spot junctions, contributing to a significant enhancement in the Raman signal intensity; this may also be the reason for the redshifted LSPR peaks in the absorption spectra.No evidence for possible core−shell structure formation was found.

SERS of Magneto-Plasmonic Nanoparticles Consisting of Magnetite and Gold or Silver.
The SERS sensitivity of the hybrid MPNPs deposited under an external magnetic field was probed using a 4-MBA self-assembled monolayer adsorbed on an atomically flat Au surface.This molecule does not possess a resonant Raman enhancement effect in the visible spectral region and strongly adsorbs onto metal surfaces, providing opportunity for the reliable probe of SERS substrates. 41Figure 5A shows how the SERS spectra were amplified by the Fe 3 O 4 @Ag NPs obtained by PLAL in a 20 mM KCl aqueous solution and excited at different wavelengths.Most spectral modes were assigned to 4-MBA molecules, whereas the broad spectral feature near 669 cm −1 is evidence of the presence of magnetite. 42No vibrational bands from maghemite (γ-Fe 2 O 3 ) or hematite (α-Fe 2 O 3 ) were observed in the SERS spectra, indicating that the magnetic part of the hybrid nanoparticles was composed from a pure magnetite structure.Figure 5B presents the SERS enhancement factors (EFs) calculated at different laser excitation wavelengths for the two most prominent spectral modes of 4-MBA at 1073 and 1581 cm −1 .These modes are assigned to ν 12 ring breathing vibration coupled with C−S stretching and ν 8a ring vibration, respectively. 5,43,44Fe 3 O 4 @Ag exhibits optimal performance within the red spectral region.The highest SERS EFs were found for 633 nm excitation, having the values of 1.9 × 10 8 and 2.3 × 10 8 for the 1073 and 1581 cm −1 modes, respectively.The excitation in the blue region (442 nm) still showed a moderate enhancement of approximately 3 × 10 5 .
Figure 6A addresses the influence of potassium chloride on the PLAL outcome and the SERS sensitivity.Compared to the pure-water-ablated Fe 3 O 4 @Ag NPs, the addition of 0.1, 1, and 20 mM KCl in the ablation medium increased the SERS sensitivity at 633 nm excitation by 4.3, 11.7, and 7.4 times, respectively.The Fe 3 O 4 @Au NPs, in contrast, showed a much poorer performance in general.Nonetheless, chloride-related signal intensification was also evident, and the calculated factors were 4.4, 3.1, and 7 times higher than for the nanoparticles prepared without chloride.The SERS sensitivity increase induced by halide adsorption on the Au/Ag NPs surface was hypothesized to be related to the SERS chemical mechanism and aggregation of plasmonic NPs (electromagnetic mechanism). 27otably, the prepared MPNPs were stable during prolonged storage at room temperature and were functional after at least seven months after production.No segregation occurred; only rapid sedimentation was observed, which could be readily dispersed by shaking or an ultrasonic bath.
Using saline-water-prepared MPNPs also affected the relative intensities of the 4-MBA spectral pattern (Figure 6B).For example, the modes at 1181 and 1487 cm −1 strengthened, and the mode at 1378 cm −1 decreased in   intensity compared to the spectra recorded for the nanoparticles prepared in the presence of 20 mM KCl with purewater-prepared NPs.The 1181 cm −1 mode is related to ring inplane C−H deformations β(CH), 45 whereas the mode at 1487 cm −1 is assigned to ν(C−C) + γ(CH), and the mode at 1378 cm −1 is related to the surface-bound carboxylate group stretching vibration ν s (CCO − ). 44,46The observed differences in relative intensities are related to structural changes within the self-assembled monolayer.These changes might be attributed to the co-adsorption of chloride ions, potentially originating from the nanoparticles, onto the gold substrate alongside the 4-MBA monolayer.
3.4.Characterization of Hybrid Magneto-Plasmonic Nanoparticles Consisting of Gold and Silver Alloys.One of the advantages of using bimetallic Ag/Au alloy NPs is the ability to fine-tune the plasmonic properties by changing the metal ratio, as the two metals are entirely miscible (Figure 1B).The LSPR maximum can be easily varied from ca. 400 to 530 nm for spherical plasmonic NPs, 14 while in conjugation with magnetite, it shifts to longer wavelengths.Panels A and B of Figure 7 show the SERS spectra of 4-MBA excited by using hybrid alloy NPs, specifically Fe 3 O 4 @Ag 50 Au 50 and Fe 3 O 4 @ Ag 80 Au 20 , respectively.The corresponding intensity distributions of the 1585 cm −1 mode acquired from 100 spots on the surface are shown in panels C and D of Figure 7.In Figure 8A, the intensity of the dominant 1585 cm −1 mode is plotted against the KCl concentration in the ablation medium.An increase in KCl concentration from 0 to 0.1 mM led to a 4-fold spectral intensity increase in the case of the Fe 3 O 4 @Ag 50 Au 50 NPs, followed by a reduction of the SERS signal at higher concentrations.In contrast, the SERS sensitivity remained relatively stable within the 0−1 mM KCl range for the Fe 3 O 4 @ Ag 80 Au 20 nanoparticles, yet it was markedly higher compared to the Ag 50 Au 50 alloy.The optimal KCl concentration in the ablation medium lies between 0.1 and 1 mM for both alloy compositions; concentrations exceeding 10 mM had a detrimental effect on SERS sensitivity.We further employed the alloy MPNPs ablated in 0.1 mM KCl solution to achieve multiwavelength SERS enhancement factors (Figure 8B).Both compositions displayed almost the exact same dependency on the excitation wavelength, with EF maxima of 9.6 × 10 6 and 9.9 × 10 6 for the 50/50 and 80/20 compositions of the Ag/Au alloy, respectively, at 633 nm.The EFs were about 1 order of magnitude lower than the EF found for Fe 3 O 4 @Ag, and they noticeably shifted toward shorter wavelengths.In the case of the alloy MPNPs, shifting from 633 to 532 nm excitation did not show a considerable decrease in the SERS EF compared to Fe 3 O 4 @Ag, which showed a clear EF reduction by approximately 45×.Finally, only very weak SERS spectra were obtained at 422 nm excitation.
3.5.SERS Analysis of Adenine Adsorbed at MPNPs.In evaluating the SERS efficacy of the Fe 3 O 4 @Ag nanoparticles synthesized in a chloride-containing solution, adenine of varying concentrations was used as a biologically relevant probe molecule. 41Figure 9A shows the vibrational modes   associated with the molecular structure of adenine.The most intense band at 731−733 cm −1 is associated with ring breathing motion, and its position is characteristic of Ag surface-adsorbed molecules rather than bulk adenine, whose Raman band appears around 723 cm −1 . 31,47No such band was observed in the given concentration window of our experiment.−49 The mode at 668 cm −1 is associated with the stretching motion of Fe−O of magnetite.No additional iron oxide phases, such as maghemite (broad spectral features at 350, 500, and 700 cm −1 ) or hematite (290 and 412 cm −1 ), were recognized from the spectra. 50The low-frequency region (Figure 9B) contained a silver−chloride stretching mode at 240 cm −1 .
The intensity of the 733 cm −1 mode decreased with the adenine concentration, and the determined detection limit was 10 −7 M. Table 1 presents a literature survey of the SERS detection limits for adenine utilizing a range of nanoparticles.Depending on the nanoparticles, the SERS detection limit of adenine varies within the range from 10 −6 to 10 −14 M. 51−66 In this study, the detection limit appears at the higher adenine concentration range compared with the values reported in the literature.One of the factors influencing the detection limit is the incorporation of the non-SERS-active magnetite compo-nent in the MPNPs.An increasing adenine concentration resulted in a diminishing ν(Ag−Cl) intensity, which hints at the competitive replacement of chlorides from the nanoparticle surface.Interestingly, we observed a noticeable shift from 731 to 733 cm −1 when the adenine concentration changed from 10 −6 to 10 −5 M, accompanied by a −4 cm −1 shift in the Ag−Cl stretching mode frequency in a much broader concentration range; such a concentration dependency points to interactions between co-adsorbed adenine and chloride on the surface.The intensity dependencies of these two modes, when normalized to magnetite's ν(Fe−O) band and plotted against the logarithm of the adenine concentration, exhibit a linear trend in the range from 10 −3 and 10 −4 to 10 −7 M (Figure 9C).
The spectrum of bare Fe 3 O 4 @Ag is presented at the bottom of Figure 9A,B (0 M). Notably, it contains stretching modes of Fe−O and Ag−Cl, along with some extent of impurity modes.The disclosed adenine experiment shows the functionality of the nanoparticles and their ability to detect trace amounts of biologically relevant molecules outside of their electronic resonance.Additionally, the nanoparticles demonstrated longterm stability and effectiveness seven months after fabrication, assuming that they were adsorbed with chloride ions from the ablation solution.

CONCLUSIONS
Hybrid magneto-plasmonic nanoparticles (MPNPs) were synthesized by merging wet-chemically prepared Fe 3 O 4 colloids with laser-ablated plasmonic nanoparticles (Au, Ag, or alloys).The magnetic and plasmonic properties acted in synergy to facilitate nanoparticle extraction, concentration, and targeted deposition while also enhancing molecular sensitivity via surface-enhanced Raman scattering (SERS).We quantified a SERS enhancement factor of 2 × 10 8 at 633 nm laser excitation for Fe 3 O 4 @Ag using a 4-MBA molecular probe.Switching to Ag x Au 100−x alloys (x = 50 and 80 wt %) in the MPNP composition permitted the tuning of the plasmon resonance frequency but also led to a 10-fold SERS sensitivity drop compared to Fe 3 O 4 @Ag.
Furthermore, we correlated the chloride ion concentration in the ablation medium to the SERS sensitivity.Lower chloride concentrations (0.1−1 mM) amplified the SERS intensity across all nanoparticle compositions, while higher concentrations (20 mM) diminished the intensity for most of the nanoparticles.The MPNPs demonstrated long-term stability and functionality after a storage period of seven months at room temperature.Finally, the biologically relevant molecule adenine was probed to evaluate the Fe 3 O 4 @Ag nanoparticle SERS sensitivity.We determined the detection limit was 10 −7 M and that there was a linear relationship between the spectral intensity and the logarithm of the adenine concentration in the 10 −3 to 10 −7 range.These findings have implications for developing MPNP-based SERS sensors for detecting trace amounts of biologically relevant molecules.
UV−vis spectra of Fe 3 O 4 @Ag and Fe 3 O 4 @Au NPs obtained by PLAL in pure and saline (30 mM)

Scheme 1 .
Scheme 1. Flowchart of Magneto-Plasmonic Nanoparticle Preparation and Application for SERS a

Figure 1 .
Figure 1.UV−vis spectra of nanoparticles obtained by PLAL of (A) Au and Ag targets in 0−20 mM KCl solutions and (B) Ag, Au, Ag 80 Au 20 , and Ag 50 Au 50 targets in pure water.The spectra are normalized to 1 for a better comparison.

Figure 3 .
Figure 3. (A) Photographs of the MPNP solution (Fe 3 O 4 @Ag) before and after concentration using a neodymium magnet.(B) TEM image of MPNPs.Magnetite (Fe 3 O 4 ) nanoparticles are cubic in shape and light gray; Ag nanoparticles are round and dark gray.

Figure 4 .
Figure 4. UV−vis absorption spectra of hybrid magneto-plasmonic nanoparticles.The plasmonic part was obtained from (A) pure Au and Ag and (B) alloyed Ag 80 Au 20 and Ag 50 Au 50 targets by PLAL in saline water.

Figure 6 .
Figure 6.(A) Relative SERS intensity of the 1587 cm −1 spectral mode of the 4-MBA monolayer enhanced by Fe 3 O 4 @Ag and Fe 3 O 4 @Au NPs.The plasmonic nanoparticles were obtained from solutions containing varied KCl concentrations.Notice the scaling of the Fe 3 O 4 @Au dependency.(B) 4-MBA monolayer spectra excited with MPNPs obtained from (a, c) 0 and (b, d) 20 mM KCl solutions using PLAL.The results are averages of 100 measurements from different spots on a sample.Laser excitation wavelength: 633 nm, power: 45 μW, and integration time: 25 s.

Figure 7 .
Figure 7. Spectroscopic characterization of Ag x Au 100−x alloy MPNPs.Average SERS spectra of 4-MBA monolayer excited using (A) Fe 3 O 4 @Ag 50 Au 50 and (B) Fe 3 O 4 @Ag 80 Au 20 NPs that were prepared by PLAL in 0−20 mM KCl aqueous solutions (salt concentrations are indicated above the spectra).(C, D) The corresponding SERS intensity distributions of the 4-MBA 1585 cm −1 mode at 100 individual spots on the surface.The red line indicates the average value.The MPNPs were prepared in a 0.1 mM KCl solution.Laser excitation was 633 nm (45 μW), and the integration time was 25 s.

Figure 8 .
Figure 8. (A) The dependence of the SERS intensity of the 1585 cm −1 mode on the Cl − concentration in the ablation medium.Laser excitation was 633 nm (45 μW), and the integration time was 25 s.(B) SERS enhancement factor for Fe 3 O 4 @Ag 80 Au 20 and Fe 3 O 4 @ Ag 50 Au 50 NPs calculated at different excitation wavelengths and based on the 1585 cm −1 intensity (KCl concentration was 0.1 mM).

Table 1 .
water, TEM EDX profiles of Fe 3 O 4 @Ag 80 Au 20 and Fe 3 O 4 @ Literature Survey of SERS Detection Limits of Adenine a