Dual-Function Fe₃O₄-Cu₂O-Ag/GO Nanocomposites: Efficient Photocatalytic Degradation and Ultrasensitive SERS Detection of Methylene Blue and Malachite Green Dyes

Abstract

The wastewater discharged from the aquaculture and textile industries often contains toxic organic dyes, such as methylene blue (MB) and malachite green (MG), which pose significant risk to public health and ecosystem stability due to their high chemical stability, bioaccumulation potential and resistance to degradation. To address these challenges, the development of an integrated system capable of both efficient degradation and highly sensitive detection of organic dyes is essential for ecological restoration and early pollution monitoring. Herein, bifunctional Fe₃O₄-Cu₂O-Ag-GO (FCA 2-GO) nanocomposites (NCs) were developed by depositing Cu₂O, Ag nanocrystals and graphene oxide (GO) onto the surfaces of Fe₃O₄ nanocrystals. This multifunctional material acted as both a photocatalyst and a surface-enhanced Raman scattering (SERS) platform, enabling simultaneous degradation and ultrasensitive detection of organic dyes. Under simulated sunlight irradiation, FCA 2-GO NCs achieved over 98% degradation of both MB and MG within 60 min, driven by the synergistic action of reactive oxygen species (·O₂⁻ and ·OH). The degradation kinetics followed pseudo-first-order behavior, with rate constants of 0.0381 min⁻¹ (MB) and 0.0310 min⁻¹ (MG). Additionally, the FCA 2-GO NCs exhibited exceptional SERS performance, achieving detection limits as low as 10⁻¹² M for both dyes, attributed to electromagnetic–chemical dual-enhancement mechanisms. Practical applicability was demonstrated in soil matrices, showcasing robust linear correlations (R² > 0.95) between SERS signal intensity and dye concentration. This work provides a dual-functional platform that combines efficient environmental remediation with trace-level pollutant monitoring, offering a promising strategy for sustainable wastewater treatment and environmental safety.

1. Introduction

Dyes are utilized extensively in the textile dyeing process due to the explosive growth of the worldwide textile dye industry [1]. However, a significant amount of dye effluents are produced throughout the dyeing process, which can cause toxicity to aquatic organisms, disrupt the balance of aquatic ecosystems, and even endanger human health if they are released directly into water bodies without being properly treated [2]. Among the most widely used dyes, cationic methylene blue (MB) and malachite green (MG) are particularly prevalent in textile applications, including the dyeing of paper, silk, wool, jute, and leather [3]. Beyond textiles, MB serves as a disinfectant in aquaculture, while MG functions as a parasiticide, fungicide, and bactericide in both aquaculture and medical fields [4]. However, MB and MG exhibit high chemical stability, environmental persistence, and bioaccumulative potential [5]. Thus, there is an urgent need to develop integrated strategies capable of both efficient degradation and sensitive detection of these hazardous dyes.

Current approaches for organic dye wastewater treatment can be broadly classified into biological, physical, and chemical methods [6]. Biological methods utilize microorganisms and enzymes to decolorize and biodegrade dye molecules through metabolic processes [7]. Although they offer advantages such as low cost, environmental friendliness, and a wide application range, they also have disadvantages including demanding reaction conditions for wastewater treatment and long processing times. Physical approaches, including adsorption, ultrafiltration, and ion exchange, are valued for their operational simplicity and efficiency. Nevertheless, challenges such as high adsorbent costs, incomplete mineralization, and secondary pollution limit their widespread application [8]. In contrast, advanced chemical oxidation techniques—particularly photocatalytic degradation—have emerged as a promising solution due to their high efficiency, capacity for complete mineralization, and adaptability to diverse dye pollutants [9].

While removing organic dyes efficiently, achieving the highly sensitive detection of dyes is also crucial. Typical techniques in the field of dye detection mainly include high-performance liquid chromatography (HPLC), ultraviolet-visible (UV-Vis) spectroscopy, gas chromatography (GC), and mass spectrometry (MS) [10,11]. Although each of these methods has unique advantages, they also each have their own technical difficulties. For instance, HPLC often requires complex sample pretreatment and lengthy analysis times, even with its exceptional separation efficiency [12]. Despite its ease of use, UV-Vis spectroscopy requires quantitative sample calibration and cannot detect dyes without fluorescent characteristics [13]. Although GC is good at detecting volatile organic dyes, it is not appropriate for non-volatile or thermally labile dyes. Similarly, MS provides precise molecular mass data but involves high operational costs and substantial sample requirements [14]. Surface-enhanced Raman spectroscopy (SERS) can, however, get around these restrictions by providing quick results and avoiding the need for complicated sample pretreatment. Additionally, SERS offers unique advantages, including label-free detection and the ability to provide molecular vibrational fingerprint information. The crux of successfully applying SERS technology in dye detection lies in the fabrication of efficient active substrates. Currently, plasmonic gold (Au) and silver (Ag) nanocrystals, particularly due to their localized surface plasmon resonance (LSPR) effect, are universally acknowledged as the optimal SERS materials. They can enhance the Raman scattering signals of adsorbed dye molecules, enabling ultrasensitive detection [15,16]. The SERS enhancement mechanism comprises two primary contributions: electromagnetic enhancement (EM) and chemical enhancement (CE) [17]. EM stems from the localized electric field generated by noble metals under illumination at specific wavelengths, whereas CE encompasses charge transfer (CT) or chemical interactions between the metal surface and adsorbed molecules [18]. The synergistic effect of these two enhancement mechanisms underpins the exceptional sensitivity of SERS in dye detection.

To achieve the dual objectives of efficient removal and highly sensitive detection of organic dyes, the rational design and construction of multifunctional nanocomposites (NCs)—capable of functioning as both photocatalysts and SERS substrates—are vital. In recent years, ternary and quaternary nanocomposite systems have attracted widespread attention due to their ability to integrate multiple functional components. For instance, Fe₃O₄@TiO₂@Ag ternary composite leverages the synergistic effects among magnetic, semiconductor, and plasmonic constituents, demonstrating enhanced photocatalytic activity alongside moderate SERS performance [19]. Similarly, graphene-based ternary systems like Ag/GO/TiO₂ have been reported to exhibit improved adsorption capacity and recyclability, facilitating both pollutant removal and detection [20]. However, these systems often face limitations such as insufficient interfacial coupling, limited charge separation efficiency and inadequate stability under operational conditions. To address these challenges, research has gradually shifted toward quaternary nanocomposites, which incorporate additional functional units to optimize overall performance [21]. For example, Fe₃O₄@Cu₂O@Ag-GO system can achieve remarkable enhancements in photocatalytic degradation and SERS sensitivity due to more efficient electron transfer, increased active sites, and improved electromagnetic field confinement. Fe₃O₄, a magnetic nanomaterial with superparamagnetic characteristics, can be added to NCs to give them significant magnetism, which makes separation and recovery easier when exposed to an external magnetic field, thereby enhancing the stability and recyclability of photocatalytic materials [22]. Cuprous oxide (Cu₂O), a p-type semiconductor with a narrow bandgap (~2.1 eV), is a promising visible-light photocatalyst due to its cost-effectiveness, low toxicity, and environmental compatibility [23,24]. However, its practical application is hindered by rapid charge recombination, which drastically reduces photocatalytic efficiency [25]. The introduction of noble metal (e.g., Ag) has been proven to be an efficacious solution to this problem, because the Schottky barriers at the Ag-Cu₂O interface can impede electron transfer from Ag to Cu₂O [25]. Ag nanocrystals can enhance photocatalytic performance by acting as electron capture agents. Additionally, the combination of Ag and Cu₂O facilitates enhanced light absorption and utilization via surface plasmon resonance (SPR) occurring at the interface, thus generating more reactive oxygen species (ROS) and significantly enhancing the photocatalytic properties of the composite [26,27]. Graphene oxide (GO) is an efficient adsorbent for eliminating organic dye molecules from wastewater owing to its large specific surface area and numerous oxygen-containing functional groups. These properties enhance the dye adsorption capability of NCs, thereby improving the overall efficiency of wastewater treatment [28]. Remarkably, after the addition of SERS-active substrates (e.g., Ag), the NCs can likewise serve as SERS substrates. It is also noteworthy that Fe₃O₄ and Cu₂O can synergize with plasmonic nanostructures to amplify electromagnetic field enhancement, further increasing SERS detection sensitivity [29].

Herein, we engineered bifunctional Fe₃O₄-Cu₂O-Ag-GO NCs that synergistically combined photocatalytic degradation and SERS detection capabilities. The photocatalytic activity dependence on Ag nanocrystal content was systematically investigated, and a plausible mechanism for MB and MG degradation by Fe₃O₄-Cu₂O-Ag-GO NCs was proposed. To evaluate SERS performance, we employed Fe₃O₄-Cu₂O-Ag structures with varying Ag contents and Fe₃O₄-Cu₂O-Ag-GO NCs as substrates, using 4-mercaptobenzoic acid (4-MBA) as a Raman probe to assess the effect of Ag contents and GO on SERS activity. Finite-difference time-domain (FDTD) simulations were conducted to quantify electromagnetic field enhancement in these nanostructures. Furthermore, the detection capability was demonstrated through ultrasensitive SERS analysis of MB and MG target molecules. This study offers two significant advances: (1) fundamental insights into the synergistic mechanisms governing both photocatalytic degradation and SERS enhancement, and (2) a versatile platform for environmental applications combining sensitive dye detection with efficient photocatalytic removal.

2. Results and Discussion

2.1. Microstructural and Morphological Characterization of Synthesized Materials

The crystal structure of the synthesized samples was investigated using X-ray diffractometer. As shown in Figure 1a, the diffraction pattern for GO only has a diffraction peak at 11.3°, corresponding to (001) plane of GO [30]. The crystal lattice planes of Fe₃O₄ nanocrystals (220), (311), (400), (422), (511), (440), (620), and (533) may be accurately correlated with the diffraction peaks at 2θ values of 30.08°, 35.43°, 43.05°, 53.41°, 56.94°, 62.5°, 70.93° and 73.96°, respectively [31]. Following the deposition of Cu₂O nanocrystals on the surfaces of Fe₃O₄, four additional peaks emerge at 36.4°, 42.3°, 61.4°, and 77.4°, which are indexed to the (111), (200), (220), and (222) planes of cubic Cu₂O nanocrystals (JCPDS 78-2076), respectively [32]. Once Ag nanocrystals are further immobilized onto the surface of Fe₃O₄-Cu₂O NCs, four new diffraction peaks at 38.1°, 44.3°, 64.5° and 77.4° are observed. These peaks are attributed to the (111), (200), (220), and (311) planes of Ag (JCPDS 87-0718), respectively [31]. The successful incorporation of GO in the FCA-GO NCs is confirmed by the presence of the GO (001) peak. Figure 1b presents the X-ray diffraction (XRD) patterns of Fe₃O₄-Cu₂O NCs with varying Ag loadings. All samples maintain the characteristic peaks of Fe₃O₄, Cu₂O, and Ag phases. Especially, the intensity of the Ag diffraction peak rises with increasing Ag loading, suggesting that more Ag nanocrystals are adhered to the Fe₃O₄-Cu₂O NCs surfaces.

Scanning electron microscopy (SEM) was utilized to characterize the morphology of as-prepared samples. Figure 2a–d show the SEM images of Fe₃O₄ nanocrystals, Fe₃O₄-Cu₂O, FCA 2 and FCA 2-GO NCs. As shown in Figure 2a, the Fe₃O₄ nanocrystals exhibit a spherical flower-like shape with a diameter of about 1.5 μm. These microspheres are composed of interconnected two-dimensional nanopetals (20–50 nm thickness) with pore sizes of 5–15 nm that self-assemble into three-dimensional nanomaterials. Cu₂O nanocrystals with a diameter of roughly 220 ± 15 nm nm are successfully modified on the surfaces of Fe₃O₄ nanocrystals, as shown in Figure 2b. Figure 2c and Figure S1a reveal that Ag nanocrystals with 16 ± 3 nm in size are uniformly distributed across the composite, as statistically confirmed by size distribution analysis of over 50 measurements (Figure S2). Figure 2d clearly illustrates that GO nanosheets (as evidenced in Figure S1b) intimately encapsulate the FCA 2 NCs to form FCA 2-GO NCs.

N₂ adsorption–desorption analysis was employed to compare the porosity properties and the specific surface areas of FCA 2 and FCA 2-GO NCs by the Brunauer–Emmet–Teller (BET) method. As shown in Figure S3a,b, both FCA 2 and FCA 2-GO NCs exhibit type IV isotherms with H3 hysteresis loops, which implies that these two samples have mesoporous structures and are advantageous for the adsorption of dyes. Notably, FCA 2-GO NCs exhibits a substantially larger specific surface area of 74.21 m²·g⁻¹, compared to the mere 53.26 m²·g⁻¹ of FCA 2 NCs. The reason for the increase in the specific surface area is that the introduction of GO mitigates nanoparticle aggregation and exposes more active sites. The Barrett–Joyner–Halenda (BJH) pore size distributions in the inset reveal that both FCA 2 and FCA 2-GO NCs contain broadly distributed pores, the majority of which are concentrated near 100 nm.

The microstructure and crystallographic characteristics of FCA 2 and FCA 2-GO NCs were further analyzed using transmission electron microscopy (TEM) and high-resolution TEM (HRTEM). As shown in Figure 3a, the FCA 2 NCs consist of Fe₃O₄, Cu₂O and Ag nanocrystals, and the Cu₂O and Ag nanocrystals are attached onto the surfaces of Fe₃O₄ nanocrystals. Lattice fringes with 0.25, 0.30, and 0.24 nm spacing are ascribed to (311) crystal plane of Fe₃O₄ nanocrystals [33], (110) crystal plane of Cu₂O nanocrystals [34], and (111) crystal plane of Ag nanocrystals [35], respectively, according to the HRTEM image (the enlarged image of the region shown as the solid orange box) of FCA 2 NCs depicted in Figure 3b. Upon incorporation of GO, a transparent and ultrathin folded layer is observed in FCA 2-GO NCs (Figure 3c), and the FCA 2 NCs are wrapped with GO. The HRTEM image of FCA 2-GO NCs in Figure 3d further demonstrates intimate interfacial contact between the GO sheet and FCA 2 NCs. Notably, the lattice continuity at the interface suggests strong electronic interactions among Fe₃O₄, Cu₂O, Ag, and GO. These structural observations align with earlier XRD and SEM analyses, corroborating the successful synthesis of FCA 2 and FCA 2-GO NCs.

The surface elemental states of the FCA 2-GO NCs were examined using X-ray photoelectron spectroscopy (XPS). The existence of C, O, Fe, Cu, and Ag in FCA 2-GO NCs is confirmed by the survey spectrum shown in Figure 4a. Fe₃O₄ is characterized by two distinct peaks at 711.1 eV (Fe 2p_3/2) and 725.0 eV (Fe 2p_1/2) in the high-resolution Fe 2p spectrum (Figure 4b), with a spin–orbit splitting of~13.9 eV between Fe 2p_3/2 and Fe 2p_1/2 [36]. The Cu 2p spectrum of Figure 4c shows peaks at 933.2 and 953.3 eV, which are believed to belong to Cu 2p_3/2 and Cu 2p_1/2 of Cu₂O, respectively [37]. The presence of a weak satellite peak at 944.2 eV suggests possible surface oxidation of the Cu₂O nanocrystals. Furthermore, the Ag 3d spectrum (Figure 4d) displays two distinct peaks at 367.5 eV (Ag 3d_5/2) and 374.6 eV (Ag 3d_3/2), confirming the presence of metallic Ag [38]. The high-resolution C 1s XPS spectrum (Figure S4) exhibits two fitted peaks at binding energies of 284.52 and 285.36 eV, corresponding to C–C and C–O bonds, respectively. Therefore, XPS results further demonstrate the coexistence of Fe₃O₄, Cu₂O, and Ag nanocrystals in FCA 2-GO, which are consistent with XRD, SEM, and TEM analyses.

2.2. Photocatalytic Activity and Photocatalytic Degradation Mechanism of FCA 2-GO NCs

The degradation of MB and MG under simulated sunshine irradiation was utilized to methodically assess the photocatalytic performance of FCA-GO NCs. UV-Vis absorption spectroscopy was employed to track the degradation kinetics, with characteristic absorption peaks at 654.5 nm (MB) and 615 nm (MG). As shown in Figure 5a, the FCA 2-GO NCs achieve complete MB degradation within 60 min, as evidenced by the progressive attenuation of the 654.5 nm absorption peak. Figure 5b compares the photocatalytic activity of Fe₃O₄ nanocrystals, Fe₃O₄-Cu₂O, FCA 1, FCA 2, FCA 3 and FCA 2-GO NCs. Control experiments (MB solution without catalyst) exhibit negligible degradation (<2%), indirectly confirming the catalytic role of FCA 2-GO NCs in degrading MB. Notably, FCA 2-GO NCs exhibit optimal photocatalytic degradation ability towards MB. The linear correlation between ln(C_t/C₀) and time indicates that all the degradation of MB catalyzed by different photocatalysts follows pseudo-first-order kinetics in Figure 5c. The kinetic rate constant (k) of FCA 2-GO NCs reaches up to 0.0381 min⁻¹, which is higher than that of Fe₃O₄-Cu₂O (0.0293 min⁻¹), FCA 1 (0.0321 min⁻¹), FCA 2 (0.0375 min⁻¹) and FCA 3 (0.0358 min⁻¹). The superior photocatalytic performance of FCA 2-GO NCs is largely attributed to the following two synergistic effects. On one hand, Fe₃O₄ and Cu₂O act as electron donors, and Ag nanocrystals promote charge carrier separation by trapping photogenerated electrons [39]. On the other hand, GO serves as both an adsorbent and an electron mediator to further enhance interfacial electron transfer due to its high surface area and oxygen-functionalized groups [40]. Analogous trends are observed for MG degradation (Figure 5d–f). FCA 2-GO NCs can also achieve complete MG removal in 60 min (k = 0.0310 min⁻¹), outperforming other photocatalysts. It follows that the FCA 2-GO NCs demonstrate exceptional photocatalytic efficiency for MB and MG degradation, achieving > 98% removal during 60 min of irradiation.

The degradation mechanism of dyes catalyzed by FCA 2-GO NCs was revealed by a series of optical measurements. Figure 6a shows the UV-vis DRS of Cu₂O nanocrystals, Fe₃O₄-Cu₂O, FCA 2 and FCA 2-GO NCs. It is evident that there are distinct absorption edges for Cu₂O nanocrystals in visible region. The incorporation of Fe₃O₄ results in a red shift in absorption edges of Fe₃O₄-Cu₂O NCs. Furthermore, dedicated analysis of FCA 2-GO NCs (see Supporting Information, Figure S5) reveals a distinct absorption feature around 690 nm, indicating that the incorporation of GO extends the visible light response region and enhances the light absorption capacity. The optical gaps (E_g) of Fe₃O₄-Cu₂O, FCA 2 and FCA 2-GO NCs were calculated via Tauc plots (Figure 6b) using the following equation [41].

$${(\alpha \mathrm{h}\nu )}^2=\mathrm{A}(\mathrm{h}\nu -{\mathrm{E}}_{\mathrm{g}})$$

(1)

α, hν and A denote absorption coefficient, photon energy and a proportionality constant, respectively. The calculation results indicate that the derived E_g values decrease sequentially from 2.38 eV (Cu₂O) to 2.34 eV (Fe₃O₄-Cu₂O), 2.30 eV (FCA 2) and 2.26 eV (FCA 2-GO), confirming enhanced visible-light utilization. Valence-band XPS was used to determine the valence band (VB) edge of pristine Cu₂O, as shown in Figure 6c, and VB value of 0.85 eV can be determined. The conduction band (CB) position can be measured as E_CB = E_VB − E_g, and CB value of Cu₂O is calculated to be −1.53 eV [42]. Figure S6a indicates that rapid and stable photocurrent responses are detected in all the samples under visible light illumination. Especially, FCA 2-GO NCs show the highest photocurrent density due to Ag-mediated electron-hole separation and superior charge transport capabilities of GO [43,44]. Furthermore, the PL spectra of Cu₂O illustrate a prominent emission peak at 575 nm, which is suggestive of rapid electron-hole recombination, as seen in Figure S6b. In contrast, FCA 2-GO NCs exhibit the weakest emission intensity, indirectly indicating that FCA 2-GO NCs have lowest carrier recombination efficiency because GO can act as an electron sink to boost the separation efficacy of photocarriers [45].

Based on the test results mentioned above, the mechanism of photocatalytic degradation of the dyes catalyzed by FCA 2-GO NCs was proposed. A Schottky barrier emerges at the Cu₂O-Ag interface when two types of nanomaterials with different work functions come into contact with one another, as illustrated in Figure 6d [29]. Until a new Fermi equilibrium is reached, the electrons (e⁻) will transfer from Ag, which has a lower work function, to Cu₂O, which has a higher work function. Energy band bending will occur downward at their interface as a result of the shift-up of Fermi level of Cu₂O and the shift-down of Fermi level of Ag [24]. When FCA 2-GO NCs are exposed to visible light, the photo-induced e⁻ move from VB of Cu₂O to its CB while generating a corresponding number of holes (h⁺) in the VB and producing photoexcited e⁻−h⁺ pairs. The e⁻ are further pushed to migrate in the direction of Ag nanocrystals since the CB of Cu₂O is located at an energy level higher than the equilibrated Fermi potential. Crucially, the charge recombination is successfully suppressed by the Schottky barrier that forms at the Cu₂O-Ag interface, which allows electron flow from Cu₂O to Ag while preventing reverse migration [46]. Superoxide anion radicals (·O₂⁻) are produced when the surrounding O₂ captures the photoexcited e⁻ that accumulate on the surfaces of Ag nanocrystals. Meanwhile, a hydroxyl group captures the h⁺ that remain in the VB of Cu₂O, yielding hydroxyl radicals (·OH). Lastly, MG and MB can be degraded under the combined attack of ·O₂⁻ and ·OH. Additionally, GO can also serve as an electron transfer channel to capture and transfer the e⁻ to further enhance photocatalytic performance [47].

In addition, the magnetic properties of Fe₃O₄ nanocrystals, Fe₃O₄-Cu₂O, FCA 2, and FCA 2-GO NCs were evaluated using VSM (vibrating sample magnetometer). As shown in Figure S3c, all of the samples are superparamagnetic and the saturation magnetization (Ms) of Fe₃O₄ nanocrystals is high as 63.1 emu g⁻¹. Upon modifying Cu₂O, Ag and GO, Ms values of Fe₃O₄-Cu₂O, FCA 2 and FCA 2-GO NCs decrease to 48.5, 41.3 and 34 emu g⁻¹, respectively. Despite reduction in Ms values, all NCs still have strong magnetic response. The inset of Figure S3c shows that FCA 2-GO NCs can be separated in just 15 s with a magnet. Benefiting by the rapid magnetic response and the recyclability of FCA 2-GO NCs, they are expected to achieve dyes degradation multiple times. Therefore, the photocatalytic activity of FCA 2-GO NCs was assessed over six consecutive degradation cycles for MB under identical conditions (Figure S3d). Remarkably, the degradation efficiency for MB remains above 84% across all cycles, demonstrating that FCA 2-GO NCs have exceptional recycling and reuse.

2.3. SERS Activity and SERS Enhancement Mechanism of FCA 2-GO NCs

In order to evaluate the effect of Ag contents and GO on SERS activity of the substrates and thus reveal the potential SERS enhancement mechanism, FCA NCs with different Ag loadings and FCA 2-GO NCs were used as the substrates to detect SERS spectra of 4-MBA at a concentration of 10⁻⁵ M, respectively. As shown in Figure 7a, all of the SERS substrates exhibit the distinctive vibrational fingerprint of 4-MBA. The most noticeable peaks at 1072 and 1593 cm⁻¹ are attributed to excitation of ring breathing and axial ring deformation modes, respectively [48]. The findings demonstrate that SERS intensity of 4-MBA first increases and then decreases with the increases in Ag contents. The increase in SERS intensity originates from the significant increase in Ag nanocrystal density, which induces the formation of high-density plasmonic hotspots at the interfacial junctions of adjacent nanocrystals. These localized electromagnetic field enhancements synergistically amplify the LSPR effect, thereby achieving a substantial augmentation in SERS signal intensity [49]. Additionally, the abnormal decline in the SERS intensity can be attributed to the over-addition of Ag nanocrystals, because excess Ag nanocrystals can lead to the agglomeration of Ag nanocrystals on surfaces of Fe₃O₄-Cu₂O NCs, which will reduce the number of hotspots. The synergistic CT interactions between 4-MBA molecules and the SERS-active substrate are responsible for the further increase in the SERS intensity of the 4-MBA after GO inclusion. In addition to having high SERS sensitivity, good homogeneity is crucial for high-quality SERS substrates [50]. To evaluate the spatial uniformity of the FCA 2-GO SERS substrate, SERS spectra of 4-MBA at a concentration of 10⁻⁵ M were acquired from 15 randomly sampled locations. Figure 7b,c show that the SERS signal intensity at 1593 cm⁻¹ is almost uniform and the relative standard deviation (RSD) value is less than 10%. The aforementioned results demonstrate that FCA 2-GO SERS substrate has high sensitivity and superior homogeneity. Subsequently, SERS detection was carried out every 5 days while the substrate was exposed to air for 30 days (Figure 7d). The SERS intensity of FCA 2-GO NCs demonstrates stability even over the 30 day period.

In order to validate the proposed SERS enhancement mechanism, the FDTD method was employed to conduct the electromagnetic field simulation analyses for FCA 1, FCA 2, FCA 3 and FCA 2-GO NCs. As shown in Figure 8a, only dispersed and sparse hot spots are formed on the surface of Fe₃O₄-Cu₂O in FCA 1 NCs with a relatively low Ag loading. The sparsity of this hot spot distribution leads to a limited excitation and coupling degree of the LSPR effect, which in turn keeps the electromagnetic field enhancement factor at a relatively low level and ultimately limits the SERS activity of FCA 1 NCs [51]. Moderate Ag loading (Figure 8b) significantly increases Ag nanocrystal density, generating a large number of interfacial hot spots between Ag nanocrystals that enhance LSPR and electromagnetic field intensity. However, excessive Ag deposition (Figure 8c) induces Ag nanocrystals to aggregate and reduces the quantity of the hot spots, thereby adversely affecting the SERS performance [29]. Remarkably, after introducing GO into FCA 2 NCs, its unique electronic structure and excellent charge transport performance play a key role. The addition of GO effectively enhances the interfacial CT efficiency between Ag nanocrystals and GO [52]. Specifically, as shown in Figure S7, in the FCA 2-GO NCs system, the CT pathway is dominated by the electron dynamics process induced by plasma resonance at the Cu₂O/Ag interface. Under the excitation of LSPR, photogenerated electrons exhibit dual-channel transport characteristics: (i) Through the quantum tunneling effect regulated by the Cu₂O/Ag Schottky barrier, they are directly injected into the CB of Cu₂O, and then relaxed to the surface-state energy level (E_SS) of Cu₂O through the non-radiative energy dissipation pathway; (ii) Cross-dimensional carrier transport is achieved through interfacial CT mediated by GO substrates. The above two electron transfer paths undergo quantum coherent coupling at the lowest unoccupied molecular orbital (LUMO) energy level of the 4-MBA molecule, forming a hybrid interface state. In this system, the electromagnetic coupling mechanism of the plasma near-field local enhancement effect and CT resonance shows a nonlinear synergistic effect, causing the SERS enhancement factor to exhibit significant quantum cascade enhancement characteristics. This enhanced charge transfer process further amplifies the electromagnetic field intensity. Thus, under the combined effect of electromagnetic field enhancement and charge transfer synergy, this enables the FCA 2-GO NCs to exhibit the optimal SERS performance.

To validate the applicability of FCA 2-GO NCs in detecting MB and MG in environmental samples, the soil samples were obtained from Jilin Normal University (Siping, China), which were calcined at 600 °C for 2 h to eliminate spectral interference of organic and inorganic impurities and ensure the minimization of the base background signal. Subsequently, 10 mg of the pre-treated soil was mixed and incubated with MB/MG solutions (10⁻⁴−10^–12 M) in centrifuge tubes for 1 h, followed by addition of 2 mg of FCA 2-GO NCs. The mixed system was continuously stirred in the dark for 24 h to achieve adsorption equilibrium. Subsequently, analyte-adsorbed NCs were magnetically separated and subjected to SERS analysis using 514 nm laser excitation. As shown in Figure 9a,b, characteristic SERS peak of MB and MG were observed [47]. The peak intensities show a decrease trend with the decrease in the analyte concentration, but can still be clearly identified at the ultra-trace concentration of 10⁻¹² M, demonstrating ultratrace detection capability in complex soil matrices. As presented in Figure 9c,d, further quantitative analysis shows that the peak intensity of MB at 1619 cm⁻¹ is linearly correlated with the concentration (linear correlation coefficient, R² = 0.96), while SERS response of MG at 1628 cm⁻¹ also conforms to the linear model (R² = 0.95). Both calibration curves exhibit robust logarithmic-linear relationships between characteristic peak intensities and analyte concentrations, confirming the quantitative detection capability of the FCA 2-GO NCs and thus providing reliable methodological support for actual environmental monitoring.

3. Experimental Section

3.1. Chemicals and Characterization

Detailed information regarding the chemicals and characterization methods utilized in this study can be found in Supplementary Materials.

3.2. Synthesis of Fe₃O₄-Cu₂O-Ag NCs

Here, Fe₃O₄ and GO were synthesized according to the methods outlined in our previous studies [30,51]. The preparation process is listed in Supplementary Materials in detail. Fe₃O₄-Cu₂O-Ag NCs were synthesized via wet-chemical reduction as illustrated in Scheme 1a. Specifically, 0.05 g of Fe₃O₄ nanocrystals were dispersed in 80 mL of deionized water under stirring. Following addition of 10 mL of CuCl₂ solution (0.1 M), the mixture was agitated for 10 min. Subsequently, 1.8 mL of NaOH (1 M) and 12 mL of NH₂OH·HCl (0.1 M) were introduced sequentially. NaOH and NH₂OH·HCl acted as precipitating and reducing agents, respectively. The mixture was stirred for additional 1 h to achieve Fe₃O₄-Cu₂O NCs. To incorporate silver, 20, 40, and 60 mg of AgNO₃ were added to the suspension to synthesize FCA 1, FCA 2, and FCA 3 NCs, respectively. Following 40 min of continuous agitation, the product was triple-rinsed with deionized water and vacuum-dried.

3.3. Synthesis of Fe₃O₄-Cu₂O-Ag-GO NCs

Fe₃O₄-Cu₂O-Ag-GO NCs were synthesized by impregnation method. As illustrated in Scheme 1b, the synthesis process of Fe₃O₄-Cu₂O-Ag-GO NCs was described as follows. 5 mg of GO was magnetically stirred for 1 h to ensure complete dissolution in 80 mL of deionized water. After adding 50 mg of Fe₃O₄-Cu₂O-Ag (FCA 2) NCs, the mixture was stirred for 10 min. Subsequently, the mixture was dried at 60 °C for 8 h after being rinsed three times with alcohol. Finally, the Fe₃O₄-Cu₂O-Ag-GO NCs were obtained and named FCA 2-GO NCs.

3.4. Photocatalytic Experiments

Photocatalytic activity was evaluated through MB and MG degradation (10 mg/L each) under ambient conditions using a 300 W xenon arc lamp equipped with an Air Mass (AM) 1.5G filter (irradiance: 100 mW/cm²) to simulate light illumination. For each test, 30 mg of catalyst was dispersed in 300 mL of dye solution. At 10 min intervals, 3 mL of aliquots were extracted and filtered through a 0.22 μm membrane filter. Residual dye concentrations were quantified by UV-Vis spectroscopy.

3.5. Photoelectrochemical Measurement

Photoelectrochemical characterization employed a CHI660E (CH Instruments, Inc., Austin, TX, USA) workstation with a standard three-electrode system: platinum plate counter electrode, Ag/AgCl reference electrode, and sample-coated ITO working electrode. For electrode preparation, 30 mg of sample was ultrasonicated in 3 mL of isopropanol (10 min) and drop-cast onto ITO glass preheated to 100 °C. The experiments were performed in a 0.5 M Na₂SO₄ aqueous electrolyte solution.

3.6. SERS Measurements

The functionalization process involved immersing FCA 1, FCA 2, FCA 3, and FCA 2-GO NCs in 4-MBA solution (10⁻⁵ M) for 24 h under dark conditions, followed by 1 h of ultrasonic treatment to facilitate molecular adsorption. After centrifugation, the samples were rinsed with ethanol to eliminate physically adsorbed 4-MBA molecules and subsequently dried for SERS analysis. SERS spectra were acquired using a 514 nm laser. To assess the spatial uniformity of the SERS signal, 15 random positions on each sample were selected for spectral collection. Additionally, the temporal stability of the SERS signal was evaluated by monitoring 4-MBA spectra at 10 day intervals over a 30 day period. To determine the detection limit of MB and MG, we diluted the original solution concentration from 10⁻⁴ to 10⁻¹² M. SERS detection method of MB and MG was identical to that of 4-MBA.

3.7. FDTD Simulation

Electromagnetic field enhancement of FCA 1, FCA 2, FCA 3 and FCA 2-GO NCs was simulated by using FDTD software (Lumerical solutions, Version 2018; Ansys; Canonsburg, PA, USA). The simulation region was a 2000 × 2000 × 2000 nm³ cuboid space surrounded by absorber layers to avoid numerical reflections. And the structural parameters in the FDTD simulation were similar to the observed values of the FCA 2-GO NCs in the experiment. A continuous-wave source (λ = 514 nm) with size of 2000 × 2000 nm² was placed in the positive z-axis. The polarization direction of the source was along the y-axis.

4. Conclusions

In summary, we successfully synthesized dual-functional Fe₃O₄-Cu₂O-Ag-GO NCs capable of efficient photocatalytic degradation and ultrasensitive SERS detection of organic dyes in aqueous systems. XRD results demonstrated the coexistence of Fe₃O₄, Cu₂O and Ag nanocrystals and GO. SEM and TEM results confirmed that ~220 nm Cu₂O, ~16 nm Ag nanocrystals and sheet-like GO were successfully attached to the surfaces of ~1.5 μm Fe₃O₄ nanocrystals. Among Fe₃O₄-Cu₂O, FCA 1, FCA2, FCA3, and FCA 2-GO NCs, FCA 2-GO NCs exhibited optimal photocatalytic performance. Under simulated sunlight, they could achieve almost complete MB and MG degradation in 60 min. From the perspective of the photocatalytic mechanism, the enhanced photocatalytic performance originated from the formation of Schottky barrier at Cu₂O-Ag interfaces and the introduction of GO. Notably, the FCA 2-GO NCs maintained >84% MB degradation efficiency after six cycles, demonstrating excellent recyclability. In addition, our investigation revealed a direct correlation between the density of plasmonic hot spots on the Fe₃O₄-Cu₂O NCs’ surfaces and the resulting SERS signal intensity. The FCA 2-GO NCs exhibited exceptional SERS performance for both MB and MG, achieving remarkable detection sensitivities as low as 10⁻¹² M. This enhanced activity stemmed from the synergistic combination of electromagnetic field enhancement and charge transfer mechanisms. This research not only deepens our understanding of photocatalytic degradation mechanisms and SERS enhancement mechanisms but also provides a potent platform for the effective elimination and the accurate identification of organic dyes.