Open Access
Add to BibSonomy
Abstract
Surface-enhanced Raman scattering (SERS) signals are fundamental for spectroscopy applications. However, existing substrates cannot perform a dynamically enhanced modulation of SERS signals. Herein, we developed a magnetically photonic chain-loading system (MPCLS) substrate by loading magnetically photonic nanochains of Fe3O4@SiO2 magnetic nanoparticles (MNPs) with Au nanoparticles (NPs). We achieved a dynamically enhanced modulation by applying an external stepwise magnetic field to the randomly dispersed magnetic photonic nanochains that gradually align in the analyte solution. The closely aligned nanochains create a higher number of hot spots by new neighboring Au NPs. Each chain represents a single SERS enhancement unit with both a surface plasmon resonance (SPR) effect and photonic property. The magnetic responsivity of MPCLS enables a rapid signal enhancement and tuning of the SERS enhancement factor.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
SERS is a powerful tool with a prominent role in analytical techniques [1–3]. Recent researches have focused on improving SERS signals for spectroscopic detection, which can promote developments in a wide range of practical applications [4,5]. SERS improvement is mainly dependent on the number of hot spots formed by aggregates of metal NPs or tailored metal nanostructures [6,7]. Although metal substrates prepared using the above approaches can increase the sensitivity of SERS signals by several orders of magnitude, the poor reproducibility of these substrates represents a major challenge [8]. And existing metal substrates are unable to perform adjustment during SERS detection. Hence, the benefit of SERS suffers from weak signals that may be collected. In particular, dynamically enhanced modulation of the measured signals is limited by the inability to control substrate adjustments immediately, compromising SERS performance and applications.
To address this concern, controllable magnetic materials made of iron oxide colloids with plasmonic metals loading (e.g., Au, Ag) were introduced. The resulting SERS substrates showed a high controllability when an external magnetic field was applied [9–12]. In fact, magnetism-induced aggregation of metal-loaded iron oxide colloids results in a large area of closely packed particles, presenting increased interparticle hot spots density due to their unique magnetic responsivity [13]. The enhanced hot spots can compensate for undesired SERS signals, observed during detection, by increasing the strength of an external magnetic field. Recent researches have reported that Fe3O4 MNPs represent a suitable supporting platform for plasmonic metals loading to form composite nanostructures, demonstrating magnetic responsivity and high SERS sensitivity under strong magnetic fields [14–21]. However, despite the introduction of these nanostructures to endow metal substrates with magnetically controllable properties, available researches do not consider that these nanocomposites can also be used to form magnetically induced self-assembled photonic structures.
A particular class of photonic stop band (PSB) materials can modulate the propagation of photons and improve light-matter interactions [22–24]. Researches have demonstrated that the SERS effect can be further enhanced when the laser wavelength aligns with the PSB position of the substrate [25,26]. Hu et al. [27] considered this aspect and proposed a smart liquid SERS substrate consisting of suspensions of Fe3O4/Au MNPs with a photonic structure that enabled tunable SERS activity, displaying structural colors when a varying magnetic field was applied. The electromagnetic fields were generated from the SPR effect of Au NPs, which were increased with decreasing interparticle distance of the magnetically photonic crystal. Nevertheless, the reproducibility of this type of substrates is limited by two aspects: the irregular agglomeration of clusters induced by Fe3O4 MNPs and poor mechanical stability of Fe3O4 MNPs against variations in the environment pH or temperature. In general, both issues can be solved by coating Fe3O4 MNPs with an inert material shell (e.g., SiO2, carbon, etc.) to form a stable core-shell structure [28–30].
In this study, we used a strategy combining the aforementioned aspects to develop MPCLS. The MPCLS acts as SERS substrate to enable the magnetically photonic structure and a high reproducibility simultaneously, which is composed of a large number of magnetically responsive photonic nanochains. Each one-dimensional nanochain was prepared by loading fixed linear nanochain of Fe3O4@SiO2 MNPs with Au NPs. The resulting nanochain displays both magnetically responsive photonic property and SPR effect. Even under strong magnetic fields, the fixed structure of the chain effectively prevents the disruption of the photonic nanostructure, which might result from irregular agglomerations of individual particles. Further, dynamically enhanced modulation of the SERS signal can be achieved using MPCLS substrate. In fact, magnetically photonic nanochains, randomly dispersed in the analyte solution, align themselves gradually on stepwise increase of the strength of an external magnetic field, as illustrated in Fig. 1. Adjacent Au NPs progressively create an increasing number of hot spots and nanochains provide the orderly photonic chain structures, so that each chain, functioning as a single SERS enhancement unit, contributes to the general enhancement modulation effect.
Fig. 1. Schematic illustration of controllable SERS detection using a MPCLS substrate for dynamically enhanced modulation of the SERS signal. Enlarged and cross-sectional pictures of a single particle from the nanochain are illustrated at the bottom left of the inset.
2. Experimental setup
2.1 Materials
Ethylene glycol (EG, spectroscopically pure, ≥ 99%), tetraethyl orthosilicate (TEOS, reagent grade, 98%), hexadecyl trimethyl ammonium bromide (CTAB, reagent grade, 99%), ammonia solution (NH3·H2O, 28% in H2O, ≥ 99.999%), sodium borohydride (NaBH4, reagent grade, 98%), sodium acetate anhydrous (anhydrous grade, ≥ 99%), and rhodamine 6 G (R6G, reagent grade, 99%) were purchased from Shanghai Aladdin Biochemical Co. Ethanol (anhydrous grade, ≥ 99.5%), 3-aminopropyl triethoxysilane (APTES, reagent grade, 98%), and ferric chloride hexahydrate (FeCl3·6H2O, analytically pure, 99%) were purchased from Sinopharm Chemical Reagent Co. Gold chloride trihydrate (HAuCl4·3H2O, optima grade, ≥ 99.9%) and trisodium citrate dehydrate (analytically pure, 99%) were purchased from Sigma-Aldrich Chemicals Co. Deionized water (DI water, 18.2 MΩ) was prepared by using the Milli-Q system.
2.2 Fabrication of MPCLS substrate
The first step was synthetizing the Fe3O4 MNPs using the following solvothermal method. We added 0.819 g of FeCl3·6H2O and 0.318 g of trisodium citrate to 30 mL of EG under magnetic stirring for 30 min, until the particulates dissolved completely in solution. Then, 1.5 g of sodium acetate and 1 mL of DI water were added in sequence while stirring to form a homogeneous red-brown solution. The mixture was sealed in a Teflon-lined autoclave and heated at 200°C for 10 h. After cooling to room temperature, the products were collected using a magnet, washed five times with a mixture of DI water and ethanol, and then dried under vacuum at 55°C until use. Subsequently, 50 mg of Fe3O4 MNPs were added to 40 mL of ethanol under mechanical stirring, until particles were dispersed uniformly. After injecting 6 and 4 mL of DI water and NH3·H2O, respectively, 240 µL of TEOS were slowly dropped into the mixture, and the whole was gently stirred for 2 h.
Notably, the fabrication of individual photonic nanochains requires that the chaining of the magnetic particles be induced by brief exposure to an external magnetic field during the silica coating process, so that the particles remain connected [31–35]. Importantly, during the initial stage of the silica coating, lasting from 10 to 35 min, a neodymium disk magnet was applied, with an approximate field strength of 550 G and a large-diameter of 10 cm. To ensure that the magnetic field was constant throughout the entire reaction mixture and the resulting nanochains exhibited uniform lengths, the central portion of the disk magnet was placed underneath the center of the flask. The distance from the disk magnet to the flask was 2 cm. When the magnet was well-placed, the particles were gradually aligned along the field direction, and thus one-dimensional linear chain structures were gradually assembled exploiting the magnetic dipole-dipole interaction. With the TEOS hydrolysis, Fe3O4 MNPs were gradually coated by silica, forming a connection between neighboring particles and thus further fixing the nanochain structures. The solution color gradually changed from yellow to dark brown during this process. The linear chain structures remained unaltered even after the magnet was removed due to the presence of silica connections. The final products were one-dimensional linear nanochains consisting of orderly arranged Fe3O4@SiO2 MNPs with a magnetically photonic structure.
We used a typical one-step reduction reaction to synthesize Au NPs [27], 15 mL of 1 mM CTAB solution were added into 80 mL of 0.25 mM HAuCl4 solution under mechanical stirring for 30 min. Subsequently, 5 mL of 6 mM ice-cold NaBH4 solution were rapidly poured into the mixture and stirred for 30 min. The final mixture was stored until use.
Finally, the MPCLS substrate was prepared, when needed, by adding 1 g of the fixed Fe3O4@SiO2 nanochains to 80 mL of the Au solution in a beaker that was kept protected from light under ultrasound for 5 h, at room temperature. The solution color gradually changed from dark red to purple during this process. The products were collected using a magnet and washed four times with a mixture of DI water and ethanol to remove the excess of unloaded Au NPs. The Au-loaded nanochains obtained from different sets of preparation were mixed, and then 2 g of Au-loaded nanochains were uniformly dispersed in 50 mL of DI water. The resulting MPCLS solution was stored until use.
2.3 SERS measurement
In this study, we performed experiments of the typical SERS liquid sample detection [36]. R6G was employed as the analyte probe to detect SERS, so that 2 mL of the as-prepared MPCLS solution and 500 µL of R6G solution were added into a custom one-face-opened quartz cuvette. The mixture was diluted with 3 ml of DI water and left to settle for 0.5 h, until the analyte molecules were sufficiently bound to the MPCLS substrate, specifically, to the surface of Au NPs. When the magnetic field was applied during detection, a small-diameter neodymium magnet (with approximate magnetic field strength of 117 G) was placed underneath the middle of the cuvette. The small diameter enabled the close alignment of dispersed Au-loaded nanochains within the laser excitation range. The laser was focused on the substrate located at the center of the cuvette. The magnetic field strength could be increased or decreased by adding or removing the magnet.
SERS spectra were recorded using a Renishaw confocal Raman spectrometer with a 633 nm He-Ne laser, laser power of 250 mW, integration time of 5 s, and a 50× microscope objective. The wavelength of Raman laser used in the SERS measurements should be higher than the wavelength of plasmon resonance band of SP materials. That is, a higher wavelength of the Raman laser is acceptable. But if the wavelength of Raman laser is lower than the plasmon resonance, the measured Raman signal will be compromised. It can be interpreted by following two aspects. First, the Raman scattering is an inelastic process. The scattered light has a red shift (Stokes shift) to the incident light, which is getting enhanced with a factor of the squared permittivity of the metal by the plasmonic near-field. Second, the optical properties of the plasmonic far-field and the plasmonic near-field are different. The near-field of plasmonic metal nanoparticles is red-shifted compared to their far-field. The red shift is determined by the total damping as inferred from the width of the plasmon resonance in the far-field spectra, such as absorption and extinction.
3. Results and discussion
One-dimensional linear nanochains before and after loading the Au NPs were characterized using transmission electron microscopy (TEM). Figures 2(a)-(b) indicate the morphology of bare nanochain is compactly assembled by Fe3O4@SiO2 MNPs. For visual comparison, the nanochains with different amounts (low, moderate, and high) of Au NP depositions in MPCLS, which were prepared by adding different amounts of Au solution (20, 50, and 80 mL), as shown in Figs. 2(c)-(e). We can see the fixed nanochains exhibit a highly uniform morphology of one-dimensional linear chain structure. It can be found that Au NP depositions increase with the addition amount of Au solution. The Au NPs do not adversely affect the orderly photonic chain structure, showing a disperse distribution on the surface of nanochain and no obvious Au aggregation. These results demonstrate that the coupling of SPR effect from plasmonic Au NPs with photonic structure for further SERS enhancement was successful. Note that there are connections formed by silica between neighboring particles can be observed in these nanochains. Also, previous studies have demonstrated that the connection between neighboring particles might be enhanced through the condensation reaction of surface silanol groups on the growing silica surface [31,37]. The silica connection plays a key role in fixing linear chain structure, ensuring that the fixed nanochains were not influenced by the external forces (e.g., mechanical stirring, sonication, etc.), showing a high stability. According to Fig. 2, the nanochain lengths were measured to be 1.040 and 1.041 (a), 1.069 (b), 1.053 and 1.050 (c), 1.067 (d), and 1.049 µm (e), respectively. Further, the average length of nanochains was calculated to be 1.053 µm.
Fig. 2. (a)-(b) TEM images of nanochains before Au NP depositions. (c)-(e) TEM images of nanochains with different amounts (low, moderate, and high) of Au NP depositions, respectively.
We examined the morphological uniformity of nanochains by scanning electron microscopy (SEM). The test sample was mixed by five groups of the as-prepared nanochains. The SEM images in Figs. 3(a)-(b) show that, after solvent evaporation, a large number of uniformly shaped one-dimensional nanochains were successfully prepared using the external magnetic field during the initial stage of silica coating (described in Section 2.2 above). The results verify the preparation of nanochain structures was not random. Note that slight local Au agglomerations are evident on the surface of individual nanochains in Fig. 3(b), which form upon solvent drying. This finding can be neglected for liquids. Based on Fig. 3(a), chain length distribution histogram is shown in the top-right inset of Fig. 2(b). The average length of nanochains is 1.065 µm, which is very close to 1.053 µm, considering a large number of samplings. Taken together, our nanochain preparation shows a high reproducibility. The same experiment was performed without the external magnetic field during preparation. Figures 3(c)-(d) show that the resulting aggregates consisted of a large number of Fe3O4@SiO2 and Au-loaded Fe3O4@SiO2 particles, respectively, where no chain formation was observed. Hence, the external magnetic field is the key to develop the one-dimensional linear chain structure.
Fig. 3. Top row: SEM images of nanochains prepared applying a magnetic field, before (a) and after (b) loading Au NPs, showing morphological uniformity. Bottom row: SEM images of aggregates formed by Fe3O4@SiO2 (c) and Au-loaded Fe3O4@SiO2 (d) particles, in samples where the external magnetic field was not applied during the fabrication process. The inset in the upper right of Fig. 2(b) shows the chain length distribution histogram of nanochains that was evaluated from Fig. 2(a).
The core-shell structure of a single particle from individual nanochains of MPCLS was better observed by TEM. As presented in Fig. 4(a), the magnified area of the nanochain with a high amount of Au NPs can be seen clearly. The silica shell provides the Fe3O4 core with good dispersion property, stability, and a large surface area, available for Au NP depositions. The corresponding elemental distribution was confirmed by energy-dispersive X-ray (EDX) mapping, as shown in Fig. 4(b). It can be seen that Fe (pink) is located at the center and entirely covered by a large area of Si (red); Au (green) dispersedly decorates the surface. The measured lattice spacing of the Au NPs was 0.235 nm, which can be attributed to the (111) plane of the face-centered cubic Au crystallite [38]. The average particle size of Au NPs was 18.5 nm (see Figs. 4(c)-(d)).
Fig. 4. Top row: High resolution TEM (a) and EDX (b) images of single particle from individual nanochains with high amount of Au NPs. Bottom row: High resolution TEM image of single Au NP (c) and particle size distribution histogram of Au NPs (d), evaluated from Fig. 2(e).
We tested whether our MPCLS substrate, consisting of magnetically responsive nanochains, could be well-controlled by an external magnetic field. As shown in Fig. 5(a), magnetic hysteresis measurements indicated that the magnetic saturation values of Fe3O4, Fe3O4@SiO2, and Au-loaded Fe3O4@SiO2 MNPs were approximately 77, 56, and 41 emu g-1, respectively. All the samples exhibited both low coercivity and remanence (less than 20 Oe and 3 emu g-1, respectively), as shown in the inset of Fig. 5(a). These results revealed a rapid magnetic response, which enables fast control of nanochains in response to an external magnetic field. Thanks to the magnetic responsivity, the reflection spectra of the magnetically photonic nanochains of Fe3O4@SiO2 MNPs in response to a varying magnetic field strength were measured in Fig. 5(b). We can see that the peak resulting from the diffraction of photonic nanochains blue shifts as the strength of the magnetic field increases from 117 to 936 G. These results demonstrated that the dispersed photonic nanochains align themselves gradually with increasing field strength. Furthermore, each nanochain functions as a single diffraction unit, contributing to the collective diffraction of individual nanochains. Lastly, an Au plasmon absorption peak was observed at a wavelength of 530 nm following Au NPs loading on Fe3O4@SiO2 MNPs, as shown by the UV-visible absorption spectra in Fig. 5(c). This result can be mainly attributed to the SPR effect of Au NPs [39].
Fig. 5. (a) Magnetic hysteresis loops of Fe3O4, Fe3O4@SiO2, and Au-loaded Fe3O4@SiO2 MNPs. The inset shows magnified fields at the origin. (b) Reflection spectra of the magnetically photonic nanochains of Fe3O4@SiO2 MNPs in response to varying magnetic field strength. (c) UV-visible spectra of Fe3O4, Fe3O4@SiO2, and Au-loaded Fe3O4@SiO2 MNPs.
Our results show that MPCLS substrates are suitable for SERS enhancement, combining photonic chain structures and SPR effects. To assess the effect of our MPCLS substrate on SERS performance (e.g., activity, spot-to-spot reproducibility, and sensitivity), the following experiments were conducted. For comparison, we measured the SERS spectra of 10−6 M R6G molecules adsorbed on MPCLS and Au-loaded Fe3O4@SiO2 MNPs without chain structure (i.e., without external magnetic field applied during detection) under the same conditions, separately. As depicted in Fig. 6(a), the MPCLS substrate displayed a stronger SERS signal intensity, indicating higher activity compared to the other sample. This can be interpreted as the synergistic effect of the multiple light scattering, resulting from the photonic chain structure and the SPRs of Au NPs, both contributing to the enhancement of SERS. Furthermore, we calculated the enhancement factor (EF) for a quantitative estimation of SERS activity. According to [37]:
where ISERS and IRS represent the peak intensities of the collected SERS and Raman signals, respectively, and CSERS and CRS are the concentrations of the corresponding R6G analytes. Here, we used a 10−2 M R6G as a reference. The EFs of the MPCLS substrate at different peaks of 1362 and 1510 cm-1 were estimated around 105. SERS spectra were also recorded by conducting measurements at ten random spots to evaluate the spot-to-spot reproducibility of the MPCLS substrate. The results, presented in Fig. 6(b), show the Raman intensities of the characteristic peaks of R6G, which revealed high reproducibility with corresponding values of relative standard deviation (RSD) below 16%. In addition, we examined the SERS spectra of R6G at reduced concentrations to investigate the SERS detection sensitivity of the MPCLS substrate. As shown in Fig. 6(c), the detectable concentration limit with the MPCLS substrate for R6G can reach as low as 10−9 M. As shown in Fig. 6(d), the common logarithm of Raman intensities at the peaks of 1362 and 1510 cm-1 were plotted towards the common logarithm of concentration, respectively. The error bar represents the standard deviation of the average Raman intensity obtained from three sets of SERS measurements. Firstly, for each R6G concentration, the three groups of the measured Raman intensities at the peaks of 1362 and 1510 cm-1 were transformed to a log-scale by taking common logarithm, respectively. Secondly, the Excel STDEV function was used to calculated the error bars for each group. Finally, the values of goodness of fit (R2) were calculated to be 0.98 and 0.96 respectively, indicating a high linearity between Raman intensity and R6G concentration.Fig. 6. (a) SERS spectra of 10−6 M R6G using the MPCLS substrate and Au-loaded Fe3O4@SiO2 MNPs, under the same conditions. (b) SERS spot-to-spot reproducibility measurements with the MPCLS substrate at ten random spots using 10−6 M R6G. (c) SERS spectra of R6G at different concentrations, adsorbed on the MPCLS substrate. (d) The linear relationship between the Raman intensities at the peaks of 1362 and 1510 cm-1 and the R6G concentration ranging from 10−9 to 10−5 M.
The different MPCLS substrates, consisting of the magnetically photonic nanochains with different amounts (low, moderate, and high) of Au NP depositions, were performed to fairly conduct SERS measurements under varying magnetic field, respectively. The test samples correspond to Figs. 2(c)-(e), respectively. Strikingly, the dynamically enhanced modulation of SERS signals can be achieved by stepwise increasing the strength of the external magnetic field using our MPCLS substrates above, as shown in Figs. 7(a)-(c). The signal intensity is gradually enhanced as the field strength increases from 117 to 819 G. The dynamic enhancement effect can be explained as follows. In the absence of an external magnetic field, the nanochains are randomly dispersed in the solution. With a stepwise increase in the strength of the external magnetic field, the dispersed nanochains closely align themselves gradually along the field direction, so that more hot spots are progressively formed by the newly neighboring Au NPs, when particles from different chains are close to each other.
Fig. 7. (a)-(c) SERS spectra of 10−6 M R6G recorded from MPCLS substrates with different amounts (low, moderate, and high) of Au NP depositions in response to increasing magnetic field strength. (d)-(f) Raman intensities at peak of 1510 cm-1 obtained from Figs. 7(a)-(c), respectively.
According to Figs. 7(a)-(c), the Raman intensities at peak of 1510 cm-1 were chosen to plot against the magnetic field strengths. The results shown in Figs. 7(d)-(f) correspond to Figs. 7(a)-(c), respectively. Take Fig. 7(d) as an example, we can see that when the field strength is increased from 351 to 468 G, Raman intensity shows the most drastic enhancement in comparison to other field strength changes. This is mainly attributed to the spectral alignment between the PSB position and the laser wavelength used in the SERS measurements can enable further SERS enhancement [25,26]. Based on Fig. 5(b), when the field strengths are 468 and 585 G, the corresponding PSB positions of the nanochains almost align with the Raman laser wavelength of 633 nm. Hence, a significant SERS enhancement occurs under synergistic effect of the SPR effect and the PSB property. Note that the PSB property does not play a role when the field strength is increased from 585 to 702 G, and thus the enhanced signal intensity mainly depends on the increasing number of hot spots formed by newly neighboring Au NPs. The same conclusions can be drawn from the results displayed in Figs. 7(e)-(f). By comparing these results, it can be found that the signal intensity increases with the amount of Au NP depositions. And the Raman intensity recorded from the MPCLS with high amount of Au NPs exhibits apparent advantages under the field strength of 819 G. Thereby, this MPCLS was confirmed to be the most appropriate substrate in our study. According to Fig. 7(c), a tunable SERS EF can be obtained by varying the magnetic field strength. The EF values corresponding to the peaks of 1362 and 1510 cm-1 improved from 5.2 × 105 to 2.6 × 106 and from 6.4 × 105 to 3.2 × 106, respectively, when the field strength was increased from 117 to 819 G.
Furthermore, five groups of MPCLS samples with high amount of Au NPs were prepared separately. All samples were evenly mixed, and divided into twenty substrates with the same nanochain concentration. To evaluate the reproducibility, we examined SERS signals on twenty substrates above under the field strength of 819 G, as shown in Figs. 8(a)-(b). According to the peak of 1510 cm-1, the SD and the average of twenty Raman intensities were calculated to be 9771.074 and 69607.87, respectively. Therefore, the RSD was derived from SD divided by average, which was calculated to be 14%, indicating a good reproducibility.
Fig. 8. (a) SERS spectra recorded from twenty MPCLS substrates under the field strength of 819 G. (b) Raman intensities obtained from peak of 1510 cm-1.
4. Conclusions
We developed a robust approach to prepare MPCLS substrate containing a large number of magnetically responsive photonic nanochains for controllable SERS detection. Each one-dimensional linear nanochain results from the combination of the fixed photonic chain structure of Fe3O4@SiO2 MNPs with Au NPs loading. The fabrication of the fixed nanochains were achieved by applying an external magnetic field during the silica coating stage. This field aligns dispersed magnetic particles, and the silica shell connection ensures chain structure remains unaltered after Au NPs loading. Individual nanochains combine magnetically photonic properties with the SPR of plasmonic Au, acting as a single SERS enhancement unit. Furthermore, the MPCLS successfully achieves dynamically enhanced modulation of the SERS signal, using a stepwise increase in the strength of the external magnetic field. The convenient rapid control over the nanochains behavioral state, in response to an external magnetic field, dynamically enhances signal intensity and EF during SERS detection. Excellent mechanical and optical stability complement the above features, making our MPCLS promising for many practical SERS applications.
Funding
National Natural Science Foundation of China (12174461, 12234012, 61835013); National Key Research and Development Program of China (2021YFA0718300, 2021YFA1400900, 2021YFA1402100).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. S. Nie and S. R. Emory, “Probing Single Molecules and Single Nanoparticles by Surface-Enhanced Raman Scattering,” Science 275(5303), 1102–1106 (1997). [CrossRef]
2. K. Kneipp, W. Yang, H. Kneipp, L. T. Perelman, I. Itzkan, R. R. Dasari, and M. S. Feld, “Single molecule detection using surface-enhanced Raman scattering (SERS),” Phys. Rev. Lett. 78(9), 1667–1670 (1997). [CrossRef]
3. J. Kneipp, H. Kneipp, and K. Kneipp, “SERS – a single-molecule and nanoscale tool for bioanalytics,” Chem. Soc. Rev. 37(5), 1052–1060 (2008). [CrossRef]
4. Q. Zhang, X. Li, Q. Ma, Q. Zhang, H. Bai, W. Yi, J. Liu, J. Han, and G. Xi, “A metallic molybdenum dioxide with high stability for surface enhanced Raman spectroscopy,” Nat. Commun. 8(1), 14903 (2017). [CrossRef]
5. D. Liu, X. Chen, Y. Hu, T. Sun, Z. Song, Y. Zheng, Y. Cao, Z. Cai, M. Cao, L. Peng, Y. Huang, L. Du, W. Yang, G. Chen, D. Wei, A. T. S. Wee, and D. Wei, “Raman enhancement on ultra-clean graphene quantum dots produced by quasi-equilibrium plasma-enhanced chemical vapor deposition,” Nat. Commun. 9(1), 193 (2018). [CrossRef]
6. Y. Sharma and A. Dhawan, “Plasmonic “nano-fingers on nanowires” as SERS substrates,” Opt. Lett. 41(9), 2085–2088 (2016). [CrossRef]
7. Y. Huang, Y. Chen, X. Xue, Y. Zhai, L. Wang, and Z. Zhang, “Unexpected large nanoparticle size of single dimer hotspot systems for broadband SERS enhancement,” Opt. Lett. 43(10), 2332–2335 (2018). [CrossRef]
8. N. Chen, T. H. Xiao, Z. Luo, Y. Kitahama, K. Hiramatsu, N. Kishimoto, T. Itoh, Z. Cheng, and K. Goda, “Porous carbon nanowire array for surface-enhanced Raman spectroscopy,” Nat. Commun. 11(1), 4772 (2020). [CrossRef]
9. Y. Li, J. Zhao, W. You, D. Cheng, and W. Ni, “Gold nanorod@iron oxide core-shell heterostructures: synthesis, characterization, and photocatalytic performance,” Nanoscale 9(11), 3925–3933 (2017). [CrossRef]
10. R. D. Rodriguez, E. Sheremet, T. Deckert-Gaudig, C. Chaneac, M. Hietschold, V. Deckert, and D. R. Zahn, “Surface- and tip-enhanced Raman spectroscopy reveals spin-waves in iron oxide nanoparticles,” Nanoscale 7(21), 9545–9551 (2015). [CrossRef]
11. J. Reguera, D. J. de Aberasturi, N. Winckelmans, J. Langer, S. Bals, and L. M. Liz-Marzan, “Synthesis of Janus plasmonic-magnetic, star-sphere nanoparticles, and their application in SERS detection,” Faraday Discuss. 191, 47–59 (2016). [CrossRef]
12. W. Jiao, C. Chen, W. You, J. Zhang, J. Liu, and R. Che, “Yolk-shell Fe/Fe4N@Pd/C magnetic nanocomposite as an efficient recyclable ORR electrocatalyst and SERS substrate,” Small 15(7), 1805032 (2019). [CrossRef]
13. C. Wang, J. Wang, P. Li, Z. Rong, X. Jia, Q. Ma, R. Xiao, and S. Wang, “Sonochemical synthesis of highly branched flower-like Fe3O4@SiO2@Ag microcomposites and their application as versatile SERS substrates,” Nanoscale 8(47), 19816–19828 (2016). [CrossRef]
14. Y. Song, J. Chen, X. Yang, D. Zhang, Y. Zou, D. Ni, J. Ye, Z. Yu, Q. Chen, S. Jin, and P. Liang, “Fabrication of Fe3O4@Ag magnetic nanoparticles for highly active SERS enhancement and paraquat detection,” Microchem. J. 173, 107019 (2022). [CrossRef]
15. L. Sun, J. He, S. An, J. Zhang, and D. Ren, “Facile one-step synthesis of Ag@Fe3O4 core-shell nanospheres for reproducible SERS substrates,” J. Mol. Struct. 1046, 74–81 (2013). [CrossRef]
16. Z. H. Li, J. H. Bai, X. Zhang, J. M. Lv, C. S. Fan, Y. M. Zhao, Z. L. Wu, and H. J. Xu, “Facile synthesis of Au nanoparticle-coated Fe3O4 magnetic composite nanospheres and their application in SERS detection of malachite green,” Spectrochim. Acta, Part A 241, 118532 (2020). [CrossRef]
17. C. Fan, S. Zhu, H. Xin, Y. Tian, and E. Liang, “Tunable and enhanced SERS activity of magneto-plasmonic Ag-Fe3O4 nanocomposites with one pot synthesize method,” J. Opt. 19, 1 (2017). [CrossRef]
18. D. Han, B. Li, Y. Chen, T. Wu, Y. Kou, X. Xue, L. Chen, Y. Liu, and Q. Duan, “Facile synthesis of Fe3O4@Au core-shell nanocomposite as a recyclable magnetic surface enhanced Raman scattering substrate for thiram detection,” Nanotechnology 30(46), 465703 (2019). [CrossRef]
19. H. Guo, A. Zhao, Q. Gao, D. Li, M. Zhang, Z. Gan, D. Wang, W. Tao, and X. Chen, “One-step synthesis of Ag-Fe3O4 nanocomposites and their SERS properties,” J. Nanopart. Res. 16(8), 2538 (2014). [CrossRef]
20. H. Guo, A. Zhao, R. Wang, D. Wang, L. Wang, Q. Gao, H. Sun, L. Li, and Q. He, “Generalized green synthesis of Fe3O4/Ag composites with excellent SERS activity and their application in fungicide detection,” J. Nanopart. Res. 17(12), 494 (2015). [CrossRef]
21. L. Lou, K. Yu, Z. Zhang, R. Huang, J. Zhu, Y. Wang, and Z. Zhu, “Dual-mode protein detection based on Fe3O4-Au hybrid nanoparticles,” Nano Res. 5(4), 272–282 (2012). [CrossRef]
22. J. Ge and Y. Yin, “Magnetically responsive colloidal photonic crystals,” J. Mater. Chem. 18(42), 1 (2008). [CrossRef]
23. J. Ge, Y. Hu, and Y. Yin, “Highly tunable superparamagnetic colloidal photonic crystals,” Angew. Chem., Int. Ed. 46(39), 7428–7431 (2007). [CrossRef]
24. J. P. Ge, Y. X. Hu, T. R. Zhang, T. Huynh, and Y. D. Yin, “Self-assembly and field-responsive optical diffractions of superparamagnetic colloids,” Langmuir 24(7), 3671–3680 (2008). [CrossRef]
25. J. Wang, H. Le-The, T. Karamanos, R. N. S. Suryadharma, A. van den Berg, P. W. H. Pinkse, C. Rockstuhl, L. Shui, J. C. T. Eijkel, and L. I. Segerink, “Plasmonic Nanocrystal Arrays on Photonic Crystals with Tailored Optical Resonances,” ACS Appl. Mater. Interfaces 12(33), 37657–37669 (2020). [CrossRef]
26. D. Tuyen le, A. C. Liu, C. C. Huang, P. C. Tsai, J. H. Lin, C. W. Wu, L. K. Chau, T. S. Yang, Q. Minh le, H. C. Kan, and C. C. Hsu, “Doubly resonant surface-enhanced Raman scattering on gold nanorod decorated inverse opal photonic crystals,” Opt. Express 20(28), 29266–29275 (2012). [CrossRef]
27. F. Hu, H. Lin, Z. Zhang, F. Liao, M. Shao, Y. Lifshitz, and S. T. Lee, “Smart liquid SERS substrates based on Fe3O4/Au nanoparticles with reversibly tunable enhancement factor for practical quantitative detection,” Sci. Rep. 4(1), 7204 (2014). [CrossRef]
28. Z. Sun, J. Du, L. Yan, S. Chen, Z. Yang, and C. Jing, “Multifunctional Fe3O4@SiO2-Au satellite structured SERS probe for charge selective detection of food dyes,” ACS Appl. Mater. Interfaces 8(5), 3056–3062 (2016). [CrossRef]
29. Y. Ye, J. Chen, Q. Ding, D. Lin, R. Dong, L. Yang, and J. Liu, “Sea-urchin-like Fe3O4@C@Ag particles: an efficient SERS substrate for detection of organic pollutants,” Nanoscale 5(13), 5887–5895 (2013). [CrossRef]
30. C. Niu, B. Zou, Y. Wang, L. Cheng, H. Zheng, and S. Zhou, “Highly sensitive and reproducible SERS performance from uniform film assembled by magnetic noble metal composite microspheres,” Langmuir 32(3), 858–863 (2016). [CrossRef]
31. Y. Hu, L. He, and Y. Yin, “Magnetically responsive photonic nanochains,” Angew. Chem., Int. Ed. 50(16), 3747–3750 (2011). [CrossRef]
32. D. Liu, S. Li, T. Zhang, H. Jiang, and Y. Lu, “3D Magnetic Field-Controlled Synthesis, Collective Motion, and Bioreaction Enhancement of Multifunctional Peasecod-like Nanochains,” ACS Appl. Mater. Interfaces 13(30), 36157–36170 (2021). [CrossRef]
33. J. Li, X. Lu, Y. Zhang, F. Cheng, Y. Li, X. Wen, and S. Yang, “Transmittance Tunable Smart Window Based on Magnetically Responsive 1D Nanochains,” ACS Appl. Mater. Interfaces 12(28), 31637–31644 (2020). [CrossRef]
34. H. Ma, K. Tang, W. Luo, L. Ma, Q. Cui, W. Li, and J. Guan, “Photonic nanorods with magnetic responsiveness regulated by lattice defects,” Nanoscale 9(9), 3105–3113 (2017). [CrossRef]
35. T. Zhang, Q. Yue, P. Pan, Y. Ren, X. Yang, X. Cheng, F. A. Alharthi, A. A. Alghamdi, and Y. Deng, “One-dimensional nanochains consisting of magnetic core and mesoporous aluminosilicate for use as efficient nanocatalysts,” Nano Res. 14(11), 4197–4203 (2021). [CrossRef]
36. K. Kim, H. S. Han, I. Choi, C. Lee, S. Hong, S. H. Suh, L. P. Lee, and T. Kang, “Interfacial liquid-state surface-enhanced Raman spectroscopy,” Nat. Commun. 4(1), 2182 (2013). [CrossRef]
37. J. Wang, A. Sugawara, A. Shimojima, and T. Okubo, “Preparation of anisotropic silica nanoparticles via controlled assembly of presynthesized spherical seeds,” Langmuir 26(23), 18491–18498 (2010). [CrossRef]
38. M. Ye, Z. Wei, F. Hu, J. Wang, G. Ge, Z. Hu, M. Shao, S. T. Lee, and J. Liu, “Fast assembling microarrays of superparamagnetic Fe3O4@Au nanoparticle clusters as reproducible substrates for surface-enhanced Raman scattering,” Nanoscale 7(32), 13427–13437 (2015). [CrossRef]
39. Z. Sun, J. Du, B. Lv, and C. Jing, “Satellite Fe3O4@SiO2-Au SERS probe for trace Hg2+ detection,” RSC Adv. 6(77), 73040–73044 (2016). [CrossRef]
