Au Nanoparticles (NPs) Decorated Co Doped ZnO Semiconductor (Co400-ZnO/Au) Nanocomposites for Novel SERS Substrates

Au nanoparticles were decorated on the surface of Co-doped ZnO with a certain ratio of Co2+/Co3+ to obtain a novel semiconductor-metal composite. The optimal substrate, designated as Co400-ZnO/Au, is beneficial to the promotion of separation efficiency of electron and hole in a semiconductor excited under visible laser exposure, which the enhances localized surface plasmon resonance (LSPR) of the Au nanoparticles. As an interesting finding, during Co doping, quantum dots of ZnO are generated, which strengthen the strong semiconductor metal interaction (SSSMI) effect. Eventually, the synergistic effect effectively advances the surface enhancement Raman scattering (SERS) performance of Co400-ZnO/Au composite. The enhancement mechanism is addressed in-depth by morphologic characterization, UV-visible, X-ray diffraction, photoluminescence, X-ray photoelectron spectroscopy, density functional theory, and finite difference time domain (FDTD) simulations. By using Co400-ZnO/Au, SERS detection of Rhodamine 6G presents a limit of detection (LOD) of 1 × 10−9 M. As a real application, the Co400-ZnO/Au-based SERS method is utilized to inspect tyramine in beer and the detectable concentration of 1 × 10−8 M is achieved. In this work, the doping strategy is expected to realize a quantum effect, triggering a SSSMI effect for developing promising SERS substrates in future.


Introduction
The surface-enhanced Raman scattering (SERS) technique has superior sensitivity and affords the molecular fingerprint information of a target sample adsorbed or approaching on the surfaces of noble nanostructures (Ag, Au, and Cu), which has been widely explored in the fields of biological, pharmaceutical, contaminant, and toxin detections [1,2]. Two acceptable dominant enhancement mechanisms are the charge transfer (CT) process [3,4] and the localized surface plasmon resonance (LSPR) field, which is connection with incident laser lines [5,6]. In the literature, the greatest enhancement factor that has been reported is 10 14 , due to specific molecules located within the gaps of neighbor Ag nanoparticles, namely LSPR hot spots [7]. Nevertheless, metallic nanoparticles expose some shortcomings such as instability, expensive cost, and limited excitation wavelength [8].
As an alternative, more attention has been focused on the possibility of semiconductor materials as SERS substrates, owing to their chemical and mechanical stabilities, such as being less-poisonousness, having high photo-efficiency, and better resistance to the environment [9]. However, most semiconductors with nanostructures only contribute an enhancement factor for Raman scattering below 10 5 [10,11]. For further improving the SERS feature of semiconductors, morphology optimization, element doping, and the composites with noble metals were investigated [12][13][14][15][16]. Amongst these, the metal and semiconductor composites exhibit the best merits because of the strong semiconductor metal interaction

Synthesis of ZnO/Au
First, 0.2 g zinc acetate was dispersed in 70 mL ultrapure water by ultrasonic wave for 30 min. A total of 10 mL of NaOH solution (2 mol/L) was added to the zinc acetate solution under constant agitation. The above solution was transferred to a reaction kettle and then put into an oven for the reaction (160 • C, 20 h). After natural cooling to room temperature, the sample was washed several times with ultrapure water to remove residual ions and molecules, and dried under a 70 • vacuum. About 0.015 g of ZnO was dissolved in 25 mL of ultrapure water and heated to boiling under stirring. Finally, 1 mL of 10 −3 M HAuCl 4 solution was injected for 30 min under agitation until the solution turned purplish-red to obtain ZnO/Au.

Synthesis of Co-ZnO
Co-doped ZnO was synthesized as follows: following standard procedure, ZnNO 3 ·6H 2 O (0.40 g) and C 4 H 6 CoO 4 with different amounts including 0, 40, 120, 200, 280, 400, 480, and 600 mg were dissolved in 10 mL ultrapure water at room temperature. After 8 mL NaOH (0.5 mol L −1 ) was added, the suspension was stirred for 40 min, after which 2.4 g NH 4 HCO 3 was added and stirred until it completely dissolved. The suspension was then dried at 60 • C for 10 h. The product was calcined in a corundum crucible with a cover at 500 • C for 2 h, followed by rapid cooling to room temperature to yield the Co-ZnO product. The obtained products were, respectively, marked as Co 40 -ZnO, Co 120 -ZnO, Co 200 -ZnO, Co 280 -ZnO, Co 400 -ZnO, Co 480 -ZnO, and Co 600 -ZnO.

Synthesis of Co 400 -ZnO/Au
A total of 0. 02 g of Co 400 -ZnO was dispersed in 25 mL ultrapure water, and heated to a boiling while constantly stirring. Then, 5 mL of 5% HAuCl 4 solution was injected under stirring for 30 min until the solution turned brown-red to obtain Co 400 -ZnO/Au successfully. After cooling to room temperature naturally, Co 400 -ZnO/Au was washed with ultrapure water several times to remove residual ions and molecules, and dried at 70 • C under vacuum.

SERS Measurement
For SERS detection, the analyte solution was mixed with Co 400 -ZnO/Au nanocomposite suspension by a volume ratio of 1:2. Raman test was conducted by using 633 nm laser with power at 5 mW and a collection time of 3 s with 2 accumulations.

Instrumentation
UV-vis spectra were collected by a UV-vis spectrophotometer (SHIMADZU, UV-1800, Kyoto, Japan). The morphologies of SERS substrates were taken by a JEM-2100EXII transmission electron microscope (JEOL Co., Ltd., Tokyo, Japan), operating at 200 kV. The high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images and elemental mapping of SERS substrates were acquired on a Tecnai G2 S-Twin F20 field-emission transmission electron microscope (FEI, Hillsboro, OR, USA). X-ray photoelectron spectroscopy (XPS) (model PHI 5000, Versa Probe, NEC Corporation, Tokyo, Japan) was performed to identify the chemical composition of Co-ZnO/Au. X-ray diffraction (XRD) analysis was conducted on D/Max-2000 VPC (RIGAKU, Tokyo, Japan). Raman experiment was performed by using a confocal laser Raman system (Super LabRamII, Jobin Yvon, Longjumeau, France). HPLC-MS results were collected by a Q EXACTIVE PLUS HPLC-MS spectrometer (Thermo Scientific, Waltham, MA, USA).

Calculation Methods
The density of states (DOS) of ZnO and Co-doped ZnO were calculated by firstprinciple calculation based on density functional theory (DFT). The pseudopotentials and the starting DFT calculation were performed based on the Perdew-Burke-Ernzerhof (PBE) exchange-correlation functional. The plane-wave cutoff energy was set to 340 eV, and the Monkhorst-Pack method with a k-points mesh of 4 × 4 × 2 was used to sample the Brillouin-zone.
The electric field strengths of ZnO, Co 400 -ZnO, and Co-ZnO/Au were calculated by using a finite difference time domain (FDTD) method. The grid precision for FDTD simulation was 2 nm in the X, Y, and Z directions, and the time step was set at 200 fs. Periodic boundary conditions were applied in both the X and Y directions, while perfect matching layer boundary conditions in the Z direction were conducted. The plane-wave source propagated along the z-axis at incident wavelengths including 532, 633, and 785 nm on the nanoparticles.

Characterization of Co-ZnO
The ionic radius of Co 2+ (0.72 Å) is similar to that of Zn 2+ (0.74 Å). Therefore, Co element can be easily doped into a ZnO lattice to substitute the position of Zn 2+ ions, which avoids lattice mismatch to an extent [19,20]. In addition, the rich electronic states of Co element benefit the optimization of the magnetic, electrical, and optical properties of ZnO [21]. Consequently, the elevated impurity level caused by Co dopant shortens the energy gap of ZnO and simultaneously improves the charge-carrier separation due to creating many electron traps [22]. Herein, first, we tuned the amount of Co element in ZnO to improve SERS performance of the resultant composite. As depicted in Figure S1 of Supplementary Material, with an increasing amount of Co dopant, the color of composite ZnO materials changes from white to dark greenish. This is due to the high spin state Co 2+ 3d 7 (4F) involving d-d transition for oxygen coordination in tetrahedral symmetry [23,24]. In Figure 1, the XRD patterns of different Co-ZnO substrates display their wurtzite structures in good agreement with the JCPDS 36-1451. There being no obvious change in diffraction peaks of Co-ZnO substrates indicates that the amorphous Co oxides have a slight effect on the crystal structure of ZnO [25]. Clearly, in Figure S2 and Table S1, the crystallite size (D), micro strain (ε), and dislocation density (ρ) of Co doping inhibiting crystallite growth of ZnO results in a size decrease in Co-ZnO composite, which shows a connection between their differences in ionic radii and valence states [26,27]. The small size of Co-ZnO increases the surface area and boundaries, which accelerates the carrier mobility [28,29]. Additionally, elevating the amount of Co doped in ZnO initially increases the strain, resulting in alteration of the lattice constant of the composite, which is proven by the visualization of the broadened XRD peaks and slight position shifts [30,31].
ZnO materials changes from white to dark greenish. This is due to the high spin state Co 2+ 3d7 (4F) involving d-d transition for oxygen coordination in tetrahedral symmetry [23,24]. In Figure 1, the XRD patterns of different Co-ZnO substrates display their wurtzite structures in good agreement with the JCPDS 36-1451. There being no obvious change in diffraction peaks of Co-ZnO substrates indicates that the amorphous Co oxides have a slight effect on the crystal structure of ZnO [25]. Clearly, in Figure S2 and Table S1, the crystallite size (D), micro strain (ε), and dislocation density (ρ) of Co doping inhibiting crystallite growth of ZnO results in a size decrease in Co-ZnO composite, which shows a connection between their differences in ionic radii and valence states [26,27]. The small size of Co-ZnO increases the surface area and boundaries, which accelerates the carrier mobility [28,29]. Additionally, elevating the amount of Co doped in ZnO initially increases the strain, resulting in alteration of the lattice constant of the composite, which is proven by the visualization of the broadened XRD peaks and slight position shifts [30,31]. XPS analysis was performed to investigate the elemental composition and chemical state. In the survey spectrum of ZnO ( Figure S3), two significant peaks, centered at 1021.18 and 1044.08 eV, are attributed to the binding energies of core-level Zn 2p3/2 and Zn 2p1/2, respectively. The fitted O 1s spectrum in the ZnO matrix resolves into both XPS analysis was performed to investigate the elemental composition and chemical state. In the survey spectrum of ZnO ( Figure S3), two significant peaks, centered at 1021.18 and 1044.08 eV, are attributed to the binding energies of core-level Zn 2p3/2 and Zn 2p1/2, respectively. The fitted O 1s spectrum in the ZnO matrix resolves into both peaks at 530.28 and 531.28 eV, which are, respectively, ascribed to O 2ions associated with Zn 2+ ions and O 2− ions in oxygen-deficient regions [32]. Obviously, in Figure S4 and Tables S2 and S3, for Co 400 -ZnO, after Co doping, the binding energy position and intensity changes in Zn 2+ and O arise from the alternation of electron density around Zn 2+ [33,34].
UV-vis diffuse absorption spectra provided the evidence for the substitution of Co in the ZnO lattice. In Figure S5, ZnO, when alone, showed an adsorption band at 392 nm. In the case of Co-ZnO, the red shift of the band edge (marked with the arrow in Figure S6) indicates the decrease in band gap energy [35]. Detailed information involving the band gap (Eg) was estimated by Tauc formula, [36] and the optical absorption edge (nm) of pure ZnO and Co-ZnO samples is tabulated in Table S4. Co 400 -ZnO presents the highest absorption edge at 479 nm and the broadest visible absorption region, which peaked at  [18] showing that visible light excitation in the solar spectrum could generate more electron-hole pairs within Co 400 -ZnO [37,38].
In Figure S7 and Table S4, compared with ZnO, the CT process between d electrons of the Co element and the conduction band (CB) or valence band (VB) of ZnO decreases the band gap for Co-ZnO composites, and Co 400 -ZnO has the lowest band gap. The diagram of VB-XPS spectra for the band structure evolution of Co-doped ZnO samples are given in Figures S8 and S9. The ease degree of electrons jumping from the VB to the CB is closely dependent on the band gap width [39,40]. In Figure S10, the corresponding calculated density of states (DOS) is consistent with the experiment results. The VB width of Co-ZnO is slightly increased compared to ZnO alone, implying mobility enhancement of the hole. Identically, the broadened CB also suggests the accelerating electron mobility [41,42].
Photoluminescence (PL), as a direct method for estimating the recombination rate of photo generated charge pairs in the crystal structure, is related to lattice defects and surface states [43]. High intensity in the PL signal indicates a rapid recombination rate of charge carriers, resulting in poor SERS performance [44]. The PL emission spectra of the samples were recorded by using an excitation wavelength of 233 nm. In Figure S11, comparably, the lowest PL signal from Co 400 -ZnO samples can be attributed to the coexistence of Co 3+ and Co 2+ , with the ratio of 0.9578 greatly inhibiting the recombination between electron-hole pairs.
The chemical structure of the Co-ZnO composites was also studied by the Fourier transform infrared (FTIR) method. In Figure S12, for pristine ZnO, the FTIR bands at 1438, 1649, and 3450 cm −1 belong to -OH deforming, O-H stretching, and -OH stretching, respectively [45]. After Co-doping, the FTIR bands regarding ZnO vibrations shift to a low wavenumber because of a partial electron transfer between ZnO and Co [46]. One of the possible principles is that defects produced in ZnO by introducing Co could act as electron traps and become an intermediate state of electron transfer bridge [47,48], which would improve photon-induced charge transfer (CT) and the photo-generated charge carrier separation efficiency.
The Raman spectra of R6G (10 −6 M) on Co-ZnO/Au substrates in Figure S13 indicate that the resultant optimal Co 400 -ZnO could greatly improve the separation efficiency of electron and hole under visible light excitation. Furthermore, the Co 400 -ZnO/Au composite, as the SERS substrate, exhibits a long-term stability and remarkable detection sensitivity.
The photon-induced charge-transfer mechanism of Co-ZnO is shown in Figure S14A. Obviously, for ZnO alone, the visible light hardly excites the electrons from the VB to CB because of the large band gap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) level of the target molecules. In the case of Co-ZnO, the narrowed band gaps would benefit electronic transitions from the VB of ZnO to the surface state energy level (Ess) [49,50], and the electrons would then be injected into the LUMO of the adsorbed molecules.
A conceivable energy level diagram with the carrier transfer mechanism is displayed in Figure S14B. The Co 2+ ion is unstable owing to easy loss of d7 electronic configuration to Co 3+ (d7). In detail, Co 2+ tends to transform electrons to the surface absorbed oxygen (Equation (2)) [51] and, simultaneously, to the formation of superoxide (·O 2 − ). The Co 3+ tends to convert to Co 2+ (Equation (3)) by capturing the photo-induced electrons. In the case of a low amount of Co dopant, the occurrence of Co 2+ ions as electron traps enhances the separation of electron and hole. However, at a higher concentration of Co dopant, with the ratio of Co 2+ /Co 3+ decreasing, the availability of electron traps descends due to excessive Co 3+ ions with vacancies as novel centers, facilitating the recombination of electrons and holes.
ZnO + hv → e − CB + h+ VB (1) In all, the due ratio of Co 2+ /Co 3+ in Co 400 -ZnO composite correspondingly resulted in the smallest grain size, the narrowest band gap, the lowest PL intensities, and superior light absorption capability. As mentioned above, the CT mechanism of Co 400 -ZnO composite is the dominant contribution to the following superior Raman enhancement of target molecules. Therefore, Co 400 -ZnO was chosen to prepare Co 400 -ZnO/Au as the next SERS substrate.

Characterization of Co-ZnO/Au
UV-vis diffuse spectra of Co 400 -ZnO and Co 400 -ZnO/Au ( Figure S15) show successful preparation of Co 400 -ZnO/Au substrate due to the occurrence of a SPR band at 523 nm from Au nanoparticles. The hydrothermal preparation protocol was employed to synthesize a three-dimensional Co 400 -ZnO/Au composite. SEM and TEM images ( Figure S16) reveal that the morphology of Co 400 -ZnO is cylindrical and the Co 400 -ZnO/Au is a Coral-shaped porous structure. In Figure S17, compared with Co 400 -ZnO, broadened XRD patterns for Co 400 -ZnO/Au at 31.66 • and 34.22 • with a slight shift indicate the partial incorporation of Au element into the crystal lattice of Co 400 -ZnO [52]. Owing to the fact that the Fermi energy of ZnO is lower than Co and Au, the modification of gold species changes the charge distribution and, then, the electron transfer on the surface, to achieve balance state [53]. As a result, a remark of numerous free electrons on the boundaries between metal and semiconductors is conducive to enlarging the localized SPR (LSPR) effect [41,54]. The detailed band structure distributions of the Co 400 -ZnO and Co 400 -ZnO/Au are illustrated in Figure 2.
Co 3+ + e−CB Co 2+ In all, the due ratio of Co 2+ /Co 3+ in Co400-ZnO composite correspondingly resulted in the smallest grain size, the narrowest band gap, the lowest PL intensities, and superior light absorption capability. As mentioned above, the CT mechanism of Co400-ZnO composite is the dominant contribution to the following superior Raman enhancement of target molecules. Therefore, Co400-ZnO was chosen to prepare Co400-ZnO/Au as the next SERS substrate.

Characterization of Co-ZnO/Au
UV-vis diffuse spectra of Co400-ZnO and Co400-ZnO/Au ( Figure S15) show successful preparation of Co400-ZnO/Au substrate due to the occurrence of a SPR band at 523 nm from Au nanoparticles. The hydrothermal preparation protocol was employed to synthesize a three-dimensional Co400-ZnO/Au composite. SEM and TEM images ( Figure S16) reveal that the morphology of Co400-ZnO is cylindrical and the Co400-ZnO/Au is a Coral-shaped porous structure. In Figure S17, compared with Co400-ZnO, broadened XRD patterns for Co400-ZnO/Au at 31.66° and 34.22° with a slight shift indicate the partial incorporation of Au element into the crystal lattice of Co400-ZnO [52]. Owing to the fact that the Fermi energy of ZnO is lower than Co and Au, the modification of gold species changes the charge distribution and, then, the electron transfer on the surface, to achieve balance state [53]. As a result, a remark of numerous free electrons on the boundaries between metal and semiconductors is conducive to enlarging the localized SPR (LSPR) effect [41,54]. The detailed band structure distributions of the Co400-ZnO and Co400-ZnO/Au are illustrated in Figure 2.  In Figure 3D, the lattice spacing merits of ZnO indicate the presence of stacking faults and defects. Clearly, in Figure S18, there are a large amount of quantum dots (QDs) of ZnO, ranging from 2.3 to 3.3 nm, generated in the Co 400 -ZnO/Au composite, which should contribute to the quantum confinement effect [55]. According to the Hamiltonian of semiconductors, in the presence of ZnO QDs, very high mobility of charge carriers leads to the fusing of exciton and plasmon resonances [56].
The corresponding energy-dispersive X-ray (EDX) elemental mapping images ( Figure 4) and TEM-EDS results ( Figure S19) of Co 400 -ZnO/Au were recorded to confirm the uniform distribution of the Co, Zn, and Au elements.
In Figure 3D, the lattice spacing merits of ZnO indicate the presence of s faults and defects. Clearly, in Figure S18, there are a large amount of quantum dots of ZnO, ranging from 2.3 to 3.3 nm, generated in the Co400-ZnO/Au composite, should contribute to the quantum confinement effect [55]. According to the Hami of semiconductors, in the presence of ZnO QDs, very high mobility of charge c leads to the fusing of exciton and plasmon resonances [56]. The corresponding energy-dispersive X-ray (EDX) elemental mapping image ure 4) and TEM-EDS results ( Figure S19) of Co400-ZnO/Au were recorded to conf uniform distribution of the Co, Zn, and Au elements. On the other hand, QDs with many defects and a lack of long-range atomi [57,58] further strengthen the strong semiconductor metal interaction (SSSMI within Co400-ZnO/Au composite. The HRTEM images demonstrate that the p tightly contacted to form an interfacial hetero junction, efficiently retard the reco leads to the fusing of exciton and plasmon resonances [56]. The corresponding energy-dispersive X-ray (EDX) elemental mapping imag ure 4) and TEM-EDS results ( Figure S19) of Co400-ZnO/Au were recorded to con uniform distribution of the Co, Zn, and Au elements. On the other hand, QDs with many defects and a lack of long-range atom [57,58] further strengthen the strong semiconductor metal interaction (SSSM within Co400-ZnO/Au composite. The HRTEM images demonstrate that the tightly contacted to form an interfacial hetero junction, efficiently retard the rec On the other hand, QDs with many defects and a lack of long-range atomic order [57,58] further strengthen the strong semiconductor metal interaction (SSSMI) effect within Co 400 -ZnO/Au composite. The HRTEM images demonstrate that the particles tightly contacted to form an interfacial hetero junction, efficiently retard the recombination of photo-generated electron/hole pairs, reduce the photo-generated charge diffusion length [59,60], and augment the exposure area of active sites. Therefore, quantum confinement inducing the SSSMI effect enabled Co 400 -ZnO/Au to provide a greater SERS effect.
In the XPS results, shown in Figure 5A and in Table S5, compared with ZnO and Co 400 -ZnO, the binding energies of Zn, O, and Co in Co 400 -ZnO/Au shift, demonstrating the intra-atomic CT process [61]. In Figure 5B, for the XPS spectrum of Zn2p in Co 400 -ZnO/Au, the binding energies of Zn 2p 3/2 and Zn 2p 1/2 present at 1021.3 and 1044.3 eV, respectively [62]. Notably, the binding energy of Zn 2p in Co 400 -ZnO/Au showed a positive shift of 0.31 eV in comparison to 1044.08 eV of Zn 2p in Co 400 -ZnO ( Figure S3), further proving the strengthened strong semiconductor metal interaction (SSSMI) effect between ZnO and Au NPs [63]. In Figure 5C, a faint Co2p central peak appears in the span from 775 to 800 eV. In detail, two binding energies of Co2p 3/2 and Co2p 1/2 orbitals were located at 781.2 and 796.7 eV, respectively. A jolting companion peak at 786 eV is indicated as Co 2+ [64,65]. In Figure 5D, for Co 400 -ZnO/Au, XPS bands at 87.38 and 88.38 eV, corresponding to electronic states of Au 2+ (minor amount) and Au 3+ (high amount), hint at the abundant free electrons in the composite. Additionally, the binding energy of the Au4f 5/2 in the composite centered at 88.38 eV, shifts (the standard XPS peak of Au4f 5/2 positioned at 87.4 eV), which is also due to the SSSMI effect [66]. The electron exchange between Au 3+ and Co 2+ ions is given as follows: Au 3+ + Co 2+ = Au 2+ + Co 3+ (4) length [59,60], and augment the exposure area of active sites. Therefore, quantum confinement inducing the SSSMI effect enabled Co400-ZnO/Au to provide a greater SERS effect. In the XPS results, shown in Figure 5A and in Table S5, compared with ZnO and Co400-ZnO, the binding energies of Zn, O, and Co in Co400-ZnO/Au shift, demonstrating the intra-atomic CT process [61]. In Figure 5B, for the XPS spectrum of Zn2p in Co400-ZnO/Au, the binding energies of Zn 2p3/2 and Zn 2p1/2 present at 1021.3 and 1044.3 eV, respectively [62]. Notably, the binding energy of Zn 2p in Co400-ZnO/Au showed a positive shift of 0.31 eV in comparison to 1044.08 eV of Zn 2p in Co400-ZnO ( Figure S3), further proving the strengthened strong semiconductor metal interaction (SSSMI) effect between ZnO and Au NPs [63]. In Figure 5C, a faint Co2p central peak appears in the span from 775 to 800 eV. In detail, two binding energies of Co2p3/2 and Co2p1/2 orbitals were located at 781.2 and 796.7 eV, respectively. A jolting companion peak at 786 eV is indicated as Co 2+ [64,65]. In Figure 5D, for Co400-ZnO/Au, XPS bands at 87.38 and 88.38 eV, corresponding to electronic states of Au 2+ (minor amount) and Au 3+ (high amount), hint at the abundant free electrons in the composite. Additionally, the binding energy of the Au4f5/2 in the composite centered at 88.38 eV, shifts (the standard XPS peak of Au4f5/2 positioned at 87.4 eV), which is also due to the SSSMI effect [66]. The electron exchange between Au 3+ and Co 2+ ions is given as follows: Au 3+ + Co 2+ = Au 2+ + Co 3+ (4)

Simulation of Electromagnetic Field Enhancement
The FDTD simulation was used to simulate the surface electric field distribution of ZnO, Co-ZnO, and Co-ZnO/Au under exposure to lasers at 532, 633, and 785 nm. As shown in Figure 6, under irradiation with a 633 nm laser, the electric field enhancement factor of Co-ZnO can reach about six at the gap of neighboring nanoparticles, which is

Simulation of Electromagnetic Field Enhancement
The FDTD simulation was used to simulate the surface electric field distribution of ZnO, Co-ZnO, and Co-ZnO/Au under exposure to lasers at 532, 633, and 785 nm. As shown in Figure 6, under irradiation with a 633 nm laser, the electric field enhancement factor of Co-ZnO can reach about six at the gap of neighboring nanoparticles, which is due to the Co doping effectively changing the photoelectric properties in comparison with the case of ZnO alone.  When it comes to Co-ZnO/Au, the SSSMI effect between ZnO QDs and AuNPs c tributes to a great enhancement of the electric field, and an enhancement factor app imately equal to 40 could be reached, which is seven-fold greater than Co400-ZnO, sho Figure S20. Figure S20 shows the concentration-dependent SERS spectra of R6G soluti recorded on Co400-ZnO. Clearly, Co400-ZnO, due to the CT mechanism, could also c tribute to the Raman signal enhancement of target molecules, to an extent.
The FDTD simulation is validated by the SERS results of 10 −7 mol/L R6G acquired Co400-ZnO/Au under different irradiations with 532, 633, and 785 nm lasers. Cle shown in Figure S21, the matching of the 633 nm laser to the electromagnetic resona absorption of the Co400-ZnO/Au substrate contributes the greatest SERS signal [67]. shown in Figure S22A, by using Co400-ZnO/Au, the limit of detection (LOD, determi on the ratio of the signal to noise (S/N) equaling to 3) for R6G is 1 × 10 −9 mol/L.

Co400-ZnO/Au-Based SERS Detection of Tyr
Tyramine (Tyr), as one of bioamines, is commonly produced in food and bever as a consequence of microorganism fermentation and decomposition processes [ Overdose of Tyr from food stuffs taken by a person results in various adverse phy logical effects such as hypertension, rash, cardiac palpitation, intracerebral hemorrh and even death in some severe cases [69]. The European Union poses a maximum l tation of Tyr content of 100-800 mg/kg in foods. Routinely, liquid chromatogra When it comes to Co-ZnO/Au, the SSSMI effect between ZnO QDs and AuNPs contributes to a great enhancement of the electric field, and an enhancement factor approximately equal to 40 could be reached, which is seven-fold greater than Co 400 -ZnO, shown Figure S20. Figure S20 shows the concentration-dependent SERS spectra of R6G solutions recorded on Co 400 -ZnO. Clearly, Co 400 -ZnO, due to the CT mechanism, could also contribute to the Raman signal enhancement of target molecules, to an extent.
The FDTD simulation is validated by the SERS results of 10 −7 mol/L R6G acquired on Co 400 -ZnO/Au under different irradiations with 532, 633, and 785 nm lasers. Clearly shown in Figure S21, the matching of the 633 nm laser to the electromagnetic resonance absorption of the Co 400 -ZnO/Au substrate contributes the greatest SERS signal [67]. As shown in Figure S22A, by using Co 400 -ZnO/Au, the limit of detection (LOD, determined on the ratio of the signal to noise (S/N) equaling to 3) for R6G is 1 × 10 −9 mol/L.

Co 400 -ZnO/Au-Based SERS Detection of Tyr
Tyramine (Tyr), as one of bioamines, is commonly produced in food and beverage as a consequence of microorganism fermentation and decomposition processes [68]. Overdose of Tyr from food stuffs taken by a person results in various adverse physiological effects such as hypertension, rash, cardiac palpitation, intracerebral hemorrhage, and even death in some severe cases [69]. The European Union poses a maximum limitation of Tyr content of 100-800 mg/kg in foods. Routinely, liquid chromatographic-fluorescence detectors (LC-FLD) [70] and liquid chromatographic-mass spectrometry (LC-MS) [71] are employed to analyze Try residue in foods. However, LC-based methods suffer from tedious sample pre-concentration, reagent-consumption, and the need for well-training persons.
As shown in Figure S22B, Co 400 -ZnO/Au has the strongest Raman enhancement effect for Try. Concentration-dependent SERS spectra of Tyr, using Co 400 -ZnO/Au, are shown in Figure 7A and the normal Raman spectrum of powder Tyr is also given in Figure S23. Figure 7B shows a linearity concentration relationship ranging from 1.0 × 10 −8 to 1 × 10 −5 mol/L, with the correlation coefficient of 0.9838 based on the characteristic band intensity at 1208 cm −1 . Tyr, with a concentration at 1 × 10 −8 M, could be detectable, which meets the detection sensitivity requirement of the EU for total tyrosine content in foods. In Figure S24, the relative standard derivation (RSD) of the SERS intensities at 613 cm −1 recorded from 20 randomly selected points on Co 400 -ZnO/Au substrate is 8.05%, which indicates a reasonable signal uniformity. After storage in ambient conditions for 70 days, the Raman signal recorded on Co 400 -ZnO/Au substrate kept 90% of its level of signal intensity obtained on freshly prepared substrate, exhibiting excellent shelf-time ( Figure S25). We can obtain reproducible SERS spectra of R6G (10 −6 M) on the three batches prepared Co 400 -ZnO/Au substrates in Figure S26.
suffer from tedious sample pre-concentration, reagent-consumption, and the need well-training persons.
As shown in Figure S22B, Co400-ZnO/Au has the strongest Raman enhancemen fect for Try. Concentration-dependent SERS spectra of Tyr, using Co400-ZnO/Au shown in Figure 7A and the normal Raman spectrum of powder Tyr is also give Figure S23. Figure 7B shows a linearity concentration relationship ranging from 1.0 × to 1 × 10 −5 mol/L, with the correlation coefficient of 0.9838 based on the characteristic b intensity at 1208 cm −1 . Tyr, with a concentration at 1 × 10 −8 M, could be detectable, w meets the detection sensitivity requirement of the EU for total tyrosine content in fo In Figure S24, the relative standard derivation (RSD) of the SERS intensities at 613 recorded from 20 randomly selected points on Co400-ZnO/Au substrate is 8.05%, w indicates a reasonable signal uniformity. After storage in ambient conditions for 70 d the Raman signal recorded on Co400-ZnO/Au substrate kept 90% of its level of signa tensity obtained on freshly prepared substrate, exhibiting excellent shelf-time (Fi S25). We can obtain reproducible SERS spectra of R6G (10 −6 M) on the three batches pared Co400-ZnO/Au substrates in Figure S26. As shown in Figure 8, in beer, tyramine at concentration as low as 1 × 10 −8 M ca detected. As shown in Table 1, the relative standard deviation is 0.29~5.05%, and reasonable recovery is 91.20~107.15%. In Table 2, compared with the other assays for in the literature, the Co400-ZnO/Au-based SERS method shows a good sensitivity.  As shown in Figure 8, in beer, tyramine at concentration as low as 1 × 10 −8 M can be detected. As shown in Table 1, the relative standard deviation is 0.29~5.05%, and the reasonable recovery is 91.20~107.15%. In Table 2, compared with the other assays for Tyr in the literature, the Co 400 -ZnO/Au-based SERS method shows a good sensitivity. well-training persons.
As shown in Figure S22B, Co400-ZnO/Au has the strongest Raman enhancement e fect for Try. Concentration-dependent SERS spectra of Tyr, using Co400-ZnO/Au, a shown in Figure 7A and the normal Raman spectrum of powder Tyr is also given Figure S23. Figure 7B shows a linearity concentration relationship ranging from 1.0 × 10 to 1 × 10 −5 mol/L, with the correlation coefficient of 0.9838 based on the characteristic ban intensity at 1208 cm −1 . Tyr, with a concentration at 1 × 10 −8 M, could be detectable, whi meets the detection sensitivity requirement of the EU for total tyrosine content in food In Figure S24, the relative standard derivation (RSD) of the SERS intensities at 613 cm recorded from 20 randomly selected points on Co400-ZnO/Au substrate is 8.05%, whi indicates a reasonable signal uniformity. After storage in ambient conditions for 70 day the Raman signal recorded on Co400-ZnO/Au substrate kept 90% of its level of signal i tensity obtained on freshly prepared substrate, exhibiting excellent shelf-time (Figu S25). We can obtain reproducible SERS spectra of R6G (10 −6 M) on the three batches pr pared Co400-ZnO/Au substrates in Figure S26. As shown in Figure 8, in beer, tyramine at concentration as low as 1 × 10 −8 M can detected. As shown in Table 1, the relative standard deviation is 0.29~5.05%, and t reasonable recovery is 91.20~107.15%. In Table 2, compared with the other assays for T in the literature, the Co400-ZnO/Au-based SERS method shows a good sensitivity.

Conclusions
In summary, the resultant optimal Co 400 -ZnO could reasonably improve the separation efficiency of electron and hole under visible light excitation. Furthermore, the Co 400 -ZnO/Au composite was prepared as an SERS substrate, which exhibited a long-term stability and a remarkable detection sensitivity for R6G with the LOD being as low as 1 × 10 −9 M. Based on XPS characterization, DFT simulation, and FDTD theoretical exploration, this promising SERS effect can be attributed to the doping of Co to generate ZnO semiconductor with many defects accompanying the formation of certain QDs, triggering SSSMI between ZnO QDs and AuNPs. The synergistic effect boosted the huge localized electromagnetic field. As a real application case, by using Co 400 -ZnO/Au-based SERS assay, the lowest detectable concentration was 1 × 10 −8 M. In this work, an effort was made to explore whether the composite of noble metal and semiconductor quantum dots could be developed as the excellent SERS substrate for trace detection.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.