High‐Efficiency Photo‐Induced Charge Transfer for SERS Sensing in N‐Doped 3D‐Graphene on Si Heterojunction

Nitrogen‐doped three‐dimensional graphene (N‐doped 3D‐graphene) is a graphene derivative with excellent adsorption capacity, large specific surface area, high porosity, and optoelectronic properties. Herein, N‐doped 3D‐graphene/Si heterojunctions were grown in situ directly on silicon (Si) substrates via plasma‐assisted chemical vapor deposition (PACVD), which is promising for surface‐enhanced Raman scattering (SERS) substrates candidates. Combined analyses of theoretical simulation, incorporating N atoms in 3D‐graphene are beneficial to increase the electronic state density of the system and enhance the charge transfer between the substrate and the target molecules. The enhancement of the optical and electric fields benefits from the stronger light‐matter interaction improved by the natural nano‐resonator structure of N‐doped 3D‐graphene. The as‐prepared SERS substrates based on N‐doped 3D‐graphene/Si heterojunctions achieve ultra‐low detection for various molecules: 10−8 M for methylene blue (MB) and 10−9 M for crystal violet (CRV) with rhodamine (R6G) of 10−10 M. In practical detected, 10−8 M thiram was precisely detected in apple peel extract. The results indicate that N‐doped 3D‐graphene/Si heterojunctions based‐SERS substrates have promising applications in low‐concentration molecular detection and food safety.


Introduction
Surface-enhanced Raman scattering (SERS) for chemical detection has the advantages of being fast, convenient, non-contact, reliable, and high sensitivity, which has been widely applied in food safety, environmental monitoring, medical diagnosis, etc. [1][2][3] The enhancement mechanism of SERS originates from two aspects: the electromagnetic mechanism (EM) and the chemical mechanism (CM).6][7] CM is generated by charge transfer (CT) between the substrate and probe molecule.However, the development of SERS is limited by the reliance on various noble metal nanoparticles (such as Au and Ag).The apparent lack of horizontal gap control in noble metal nanoparticles leads to signal fluctuations, resulting in poor spatial repeatability of molecular detection.In addition, the high chemical activity of noble metals inevitably affects the long-term stability and reproducibility of SERS substrates. [8]Therefore, there is still critical to construct highsensitivity and high-reliability SERS substrates.
Based on CM-dependent SERS enhancement (CT between probe molecules and substrate), 2D-graphene its derivatives have been used as SERS substrates for molecular detection.[11] N-doped 3D-graphene is assembled by growing N-doped 2D graphene perpendicular to the substrate, which has unique properties due to its novel 3D structure. [12]The large specific surface area of Nitrogen-doped three-dimensional graphene (N-doped 3D-graphene) is a graphene derivative with excellent adsorption capacity, large specific surface area, high porosity, and optoelectronic properties.Herein, N-doped 3Dgraphene/Si heterojunctions were grown in situ directly on silicon (Si) substrates via plasma-assisted chemical vapor deposition (PACVD), which is promising for surface-enhanced Raman scattering (SERS) substrates candidates.Combined analyses of theoretical simulation, incorporating N atoms in 3D-graphene are beneficial to increase the electronic state density of the system and enhance the charge transfer between the substrate and the target molecules.The enhancement of the optical and electric fields benefits from the stronger light-matter interaction improved by the natural nano-resonator structure of N-doped 3D-graphene.The as-prepared SERS substrates based on N-doped 3D-graphene/Si heterojunctions achieve ultralow detection for various molecules: 10 À8 M for methylene blue (MB) and 10 À9 M for crystal violet (CRV) with rhodamine (R6G) of 10 À10 M. In practical detected, 10 À8 M thiram was precisely detected in apple peel extract.The results indicate that N-doped 3D-graphene/Si heterojunctions based-SERS substrates have promising applications in low-concentration molecular detection and food safety.
[15][16] Moreover, incorporating N atoms is beneficial to increasing the electronic state density of the system and enhancing the CT between the substrate and the target molecules.N-atom doping technology can open the graphene bandgap, effectively promoting the generation of CTs between energy levels.Therefore, N-doped 3D-graphene interacts more strongly with incident light than 2D-graphene and can generate more photogenerated carriers.These unique properties suggest that N-doped 3D-graphene/Si will provide ideas for the structural design and multimodal molecular detection of SERS substrates. [17,18]n this work, we fabricated an N-doped 3D-graphene/Si heterojunction as a SERS substrate by a one-stage process with excellent adsorption capacity, large specific surface area, high porosity, and optoelectronic properties.Through the density functional theory (DFT) calculation and scanning Kelvin probe microscopy (SKPM) analysis, based on the unique physical/chemical features of N-doped 3D-graphene, the electron interaction between the target molecules and the substrate is enhanced to improve charge transfer effects in the heterojunction.This SERS substrate is highly stable, ultrasensitive, inexpensive, convenient, and reusable for various molecule detection.The detection results show that the substrate is highly sensitive to various molecules: the lowest detection limits measured using a 532 nm laser were found to be 10 À8 , 10 À9 , 10 À10 , and 10 À9 M for methylene blue (MB), crystal violet (CRV), rhodamine (R6G) and thiram, respectively.These results indicate that the N-doped 3D-graphene/Si substrate has broad application prospects in molecular detection and food safety.

Characterization of Structure and Optical Properties of N-Doped 3D-Graphene/Si Heterojunction
The cross-sectional scanning electron microscopy (SEM) image of N-doped 3D-graphene/Si is shown in Figure 1a, indicating a vertical heterojunction structure.The surface topography of N-doped 3Dgraphene/Si is described in Figure S1, Supporting Information.The wettability of N-doped 3D-graphene compared with 3D-graphene and pure Si, verifying the vertical features of N-doped 3D-graphene, as shown in Figure S2, Supporting Information.The 3D atomic force microscopy (AFM) image (Figure 1b) displays the morphological features of N-doped 3D-graphene, demonstrating the vertically aligned nature of N-doped graphene on the Si substrate.The height of Ndoped 3D-graphene is taken along the dashed line in Figure 1b, as shown in Figure S3, Supporting Information, indicating a uniform structure with an average height of about 580 nm.The 3D AFM image and height distribution of 3D-garphene/Si are presented in Figure S4, Supporting Information.The Raman spectra of N-doped 3D-graphene compared with intrinsic 3D-graphene are shown in Figure 1c.Intrinsic graphene has three distinct peaks: D peak (~1350 cm À1 ), G peak (~1580 cm À1 ), and 2D peak (~2700 cm À1 ).It is worth noting that the G and 2D peaks of N-doped 3D-graphene appear blue-shifted and red-shifted, respectively, originating from the introduction of N atoms. [19]The energy-dispersive X-ray spectrometry (EDS) in the inset of Figure 1d shows the presence of C, N, and Si elements.The cross-sectional EDS elemental maps confirm the uniform N doping of 3D-graphene in Figure 1d.The chemical states of N-doped 3D-graphene are shown by the high-resolution XPS spectrum of C-1s (Figure 1e) and N-1s spectrum (Figure 1f).The highresolution C-1s spectrum can be fitted to three peaks at ~284.6, ~285.5, and ~289.3 eV, corresponding to C=C, C=N and C-N, respectively, [20] as shown in Figure 1e.Meanwhile, the highresolution N-1s spectrum can be decomposed into three components at ~398.3, ~400.8, and ~403.5 eV, as shown in Figure 1f, indicating the presence of pyridinic N, pyrrolic N and graphite N. [21,22] These prove that N atoms have been successfully bonded to the graphene lattice.Approximately 80% porosity of N-doped 3D-graphene enables it to have a large specific surface area (Figure 1g), which endows it with the ability to adsorb a mass of probe molecules.In contrast, the porosity of 3D-graphene is only 70%, as shown in Figure S4, Supporting Information.The interaction of light with N-doped 3Dgraphene/Si substrate was explored using Finite-difference timedomain (FDTD) simulations.Details about the calculation method are presented in Supporting Information.Figure 1h exhibits the normalized power loss density distribution of the N-doped 3D-graphene/Si substrate.The distributions of power loss density distribution along the vertical direction are selected, as shown in Figure S5a, Supporting Information, implying that the light absorption occurs inside the graphene and the maximum absorption positions are in the top 3D graphene.Furthermore, the normalized local electric field distribution of N-doped 3D-graphene/Si substrate was simulated in 3D views, as shown in Figure 1i.The distributions of normalized local electric field distribution along the vertical direction are selected, as shown in Figure S5b, Supporting Information, suggesting that electric field enhancement is in the graphene region.It can be estimated that the enhancement factor based on electric field enhancement of N-doped 3D-graphene is 5.4 9 10 2 .The detailed analysis and calculation process are placed in Supplementary Information.The enhancement of the optical and electric fields benefits from the stronger light-matter interaction improved by the natural nano-resonator structure of Ndoped 3D-graphene.The electromagnetic mechanism of N-doped 3D-graphene mainly originated from the nanogap structure.Local surface plasmon resonance is caused by high porosity and sharp edges of N-doped 3D-graphene.Furthermore, the Fermi level of graphene is regulated by doping technology, so the system gains more active CTs.Therefore, N-doped 3D graphene-based SERS substrates exhibit high sensitivity and enhancement factors.

Explore the Charge Transfer Mechanism by DFT Calculation of Two Structures
The intense light absorption of 3D graphene provides the charge basis for photoinduced charge transfer (PICT).Typically, the efficiency of the PICT transition strongly depends on the intensity of vibronic coupling in the substrateÀmolecule system. [23,24]To investigate the vibronic coupling between the probe molecule and different graphene systems, we adopted the first-principles density functional theory (DFT) to calculate the electronic density of states (DOS) of the two graphene systems.The results are displayed in Figure S6, Supporting Information.To further explain the reason that the substrate composed of N-doped 3D-graphene/Si shows higher SERS enhancement compared with the 3D-graphene/Si, the electrostatic potentials in the Z-direction of two systems were discussed and found that the work function of N-doped 3D-graphene/Si increased than 3D-graphene/Si.The work functions of 3D-graphene/Si and N-doped 3D-graphene/Si are 4.576 eV and 4.966 eV, as shown in Figure 2a.As shown in Figure 2b, the Fermi level of 3D-graphene/Si and N-doped 3D-graphene/Si is located between the lowest unoccupied molecular orbital (LUMO) (À3.364 eV) and highest occupied molecular orbital (HOMO) (À5.143 eV) of the R6G molecules, suggesting that the CT is easily generated between 3D graphene and R6G molecules.The Fermi energy level of the N-doped 3D-graphene/Si structure was closer to the HOMO of R6G, promoting the charge transfer between R6G and SERS substrate, leading to better SERS performance. [25]To investigate the interfacial charge transfer between 3D-graphene structure and surface complex, we built the R6G & 3D-graphene/Si and R6G & N-doped 3D-graphene/Si models, respectively, as shown in Figure 2c,f.Bader analysis was used to calculate the charge transfer in atoms to directly compare numerical values for electron transfer.Figure 2d,e shows the front and side views of the difference in charge distributions for R6G & 3D-graphene/Si.Figure 2g,h shows the front and side views of the difference in charge distributions for R6G & N-doped 3D-graphene/Si.The calculation results indicated that 0.554 e À was transferred from R6G to the 3D-graphene/Si system, and 0.645 e À was transferred from R6G to the N-doped 3D-graphene/Si system.That suggested a significant increase in molecular polarizability, which promotes Raman scattering. [23]

Investigate the Charge Transfer Mechanism by SKPM and CAFM
To demonstrate that the SERS performance of the N-doped 3Dgraphene/Si substrate can also be enhanced via the CM, SKPM and conductive AFM (CAFM) were used to show that carrier transfer occurs between certain probe molecules and the substrate.This work, used a solution of the fluorescent dye molecule R6G for this purpose.Figure 3a,b depict the surface potential distribution of N-doped 3D-graphene/ Si without and with adsorption of R6G molecules using SKPM, respectively.It was observed that the surface potential increased significantly after R6G adsorption, indicating that R6G molecules and N-doped 3Dgraphene/Si occur in highly efficient CT.The average surface potential of N-doped 3D-graphene/Si increased from 240 mV to 600 mV before and after the adsorption of R6G molecules, as shown in Figure 3c.In order to further investigate the charge transfer process, different molecules (R6G, CRV, and MB) on N-doped 3D-graphene/Si are characterized by SKPM, as shown in Figure S7, Supporting Information.It was observed that the surface potential increased significantly after R6G, CRV, and MB adsorption, indicating that dye molecules and N-doped 3D-graphene/Si occur in highly efficient charge transfer.CAFM was used to study the changes in the current map of R6G molecules adsorbed on N-doped 3D-graphene/Si under dark and light conditions, as shown in Figure 3d,e.The average surface current of N-doped 3Dgraphene/Si increased from 2.8 to 8 nA under dark and light conditions, respectively, as shown in Figure 3f.The current under the light condition is much larger than in the dark condition, indicating that PICT occurs.A photoelectric response test based on the N-doped 3Dgraphene/Si structure was conducted to better understand the PICT process.Figure S8, Supporting Information, presents the photoresponse measurement of N-doped 3D-graphene/Si and R6G/N-doped 3D-graphene/Si hybrid photodetector.The probe solution was added onto the SERS substrate for each SERS detection.Take the R6G molecule as an example, R6G molecules uniformly exist on the surface of 3D-graphene/Si and N-doped 3Dgraphene/Si by physical adsorption.N-doped 3D-graphene/Si adsorption efficiency is higher than 3D-graphene/Si due to its larger porosity.[28][29][30] The enhancement factors of 3Dgraphene/Si substrate and N-doped 3D-graphene/Si substrate are 1.01 9 10 4 and 2.78 9 10 6 , respectively.The detailed calculation method is presented in Supporting Information.Compared with 3Dgraphene/Si system, the N-doped 3D-graphene/Si system has a more abundant electronic density of states near the Fermi level and more photogenerated charges produced by optical absorption enhancement, as shown in Figure S10, Supporting Information.It has the Fermi energy level closest to the HOMO of the molecule.This makes the EF of the N-doped 3D-graphene/Si substrate far greater than that of 3Dgraphene/Si substrate.The Raman spectra of various concentrations of CRV, R6G, and MB on N-doped 3D-graphene/Si substrates are shown in Figure 4e-g, indicating the characteristic peak intensities of different probe molecules are positively correlated with their concentrations.In addition, CRV, R6G, and MB with detection limits as low as 10 À9 M, 10 À10 M and 10 À8 M, respectively.In addition, the linear fitting of the characteristic peak intensity as a function of concentration is performed respectively, as shown in Figure 4h-j.The linear correlation coefficient (R 2 ) of CRV at 1621 and 1177 cm À1 is 0.865 and 0.883; The R 2 of R6G at 1651 and 1365 cm À1 is 0.976 and 0.985; The R 2 of MB at 1626 and 1397 cm À1 is 0.853 and 0.846.The high linearities indicate that N-doped 3D-graphene/Si substrate can detect low-concentration probe molecule solutions.The Raman spectra of various concentrations of CRV, R6G, MB, and thiram on 3D-graphene/Si substrates are shown in Figure S11, Supporting Information.
The uniformity, stability, and reproducibility of the SERS substrate are the key indicators for evaluating its capability.Therefore, 900 points were randomly selected on the surface of a (1 cm 9 1 cm) sample of N-doped 3D-graphene/Si substrate.The intensity of the 2D-peak was recorded, and the results are shown in Figure 5a.At the same time, we calculated that the relative standard deviation (RSD) was 5.86% (Figure 5c).The Raman spectra of N-doped 3D-graphene from 0 to 8 months are shown in Figure S12, Supporting Information.Both show that this N-doped 3D-graphene/Si is stable and changes little over time.Similarly, 900 points were randomly selected on the surface of R6G & N-doped 3D-graphene/Si substrate, recording the intensity at 1651 cm À1 and showing the results in Figure 5b.We gained a minute RSD value of 12.79% (Figure 5d), showing that the SERS signal undergoes only small fluctuations.We track the detection capability of the substrate over a long time, as shown in Figure 5e-g.The results showed that the ability of the substrate to detect these probe molecules did not change with time, which proved that the substrate has ultrahigh stability.Figure 5h shows the 3D AFM surface topography of Ndoped 3D-graphene before and after eight times of ethanol-washing processes, and no change in the vertical structure was observed.Moreover, this also demonstrates the excellent reusability and stability of the substrate.These results show that the N-doped 3D-graphene/Si substrate is a credible SERS platform.

To Explore the Practical Applications of the N-Doped 3D-Graphene/Si Substrate in Thiram Detection
Surface-enhanced Raman scattering detection of pesticide residues in fruits was studied to demonstrate further the practical application of Ndoped 3D-graphene/Si substrates in food safety.As the primary type of pesticide residue, thiram is dissolved in alcohol with its concentration adjusted.Figure 6a shows the process of SERS detection of thiram.To describe the combination of substrate and adsorbed molecule legitimately and at the same time can be used to calculate vibrational spectra properly, we established and optimized the Thiram molecular model in Figure 6b, and we calculated the vibrational frequencies for the optimized structures based on the models.Figure 6c shows the comparison of experimental values with the calculated Raman spectrum.It was found that the calculated Raman frequencies were in good agreement with experimental values, which proves that our calculated model was probably reasonable. [31]Figure 6d shows the Raman spectra of Ndoped 3D-graphene/Si substrates detecting thiram ethanol solution 10 À6 M-10 À9 M concentrations.Characteristic peaks can still be observed at concentrations as low as 10 À9 M. Furthermore, the residual thiram on the surface of apple peel was extracted with an ethanol solution and was successfully detected as 2.4 9 10 À3 mg kg À1 , which was lower than the maximum residue level of pesticides in food allowed by EU standards (0.1 mg kg À1 ) (Figure 6e).By linearly fitting the characteristic peak intensities at 1381 and 1507 cm À1 , R 2 can be found to be 0.835 and 0.812, respectively, as shown in Figure 6f.These results prove that the N-doped 3D-graphene/Si substrate has high practicability.Compared with the earlier literature, the detection limit of the N-doped 3D-graphene/Si-based SERS substrate is better than that of most graphene composite substrates and other SERS substrates, as shown in Figure S13, Supporting Information.

Conclusions
The N-doped 3D-graphene/Si substrate fabricated via PACVD exhibits excellent adsorption capacity, large specific surface area, and high porosity, which can attribute to its unique 3D structure.Combined analyses of theoretical simulation, incorporating N atoms in 3Dgraphene are beneficial to increase the electronic state density of the system and enhance the CT between the substrate and the target molecules.The as-prepared SERS substrates are highly stable, so they can maintain stability after a long time of storage.Besides, the reusability of N-doped 3D-graphene/Si substrate greatly reduces the cost and enables large-scale application.The FDTD simulation shows that light absorption occurs inside the graphene and the maximum absorption position is in the top N-doped 3D-graphene.And then, more charge carriers are produced in graphene in incident light also improve chemical/charge transfer effects in the heterojunction.Hence, SERS substrates based on N-doped 3D-graphene/Si heterojunction achieve ultra-low detection for various molecules.In practical application, apple peel extract precisely detected 10 À8 M thiram.This study provides highly stable, lowcost, reusable, ultrasensitive, and novel SERS substrates, which have broad application prospects in low-concentration molecular detection and food safety.

Experimental Section
Materials characterizations and simulation method: Materials characterizations and details of the simulation method can be found in Supporting Information.
N-doped 3D-graphene/Si is prepared via PACVD: Si substrates with the size of 1 cm 9 1 cm were placed in a PACVD quartz tube for the subsequent treatments.Equipment model BTF-1200C-II-AS-PECVD, purchased from Anhui Beyike Equipment Technology Co., Ltd.The quartz tube was first vacuumed to ~5 Pa and then heated to 750 °C in a mixed atmosphere of 10 sccm argon (Ar, 99.9999% purity) and 1 sccm hydrogen (H 2 , 99.9999% purity).After reaching the set temperature, Ar and H 2 were turned off simultaneously, and 5 sccm of methane (CH 4 ) and 1 sccm of ammonia (NH 3 ) were introduced.After that, the plasma source was turned on and set to 200 W.After 60 minutes of growth, CH 4 and NH 3 were turned off, 10 sccm of Ar was then introduced to raise the tube back to atmospheric pressure.The samples were taken out as the temperature of the quartz tube dropped to room temperature.
Probe molecular materials and raman measurement: CRV, R6G, and MB (Aladdin Reagents, Shanghai, China) were dissolved in ethanol as dye probe molecules, respectively.A certain concentration of thiram solution was used as the probe solution.For the Raman test, 5 ll of probe molecules were dropped onto the substrate using a pipette.A 532 nm laser was used for the SERS test as the Energy Environ.Mater.2024, 7, e12565 excitation source with a laser spot diameter of 12.5 lm.The laser power was set to 2.5 mW and the integration time was 10 s.
Simulation method: The FDTD simulations were used to calculate the normalized electric field and power loss density distributions of the N-doped 3Dgraphene/Si structure.The height of N-doped 3D-graphene is 580 nm.The wavelength of the excitation source is 532 nm.Details about the simulation method are presented in Supporting Information.
Computational methods: All calculations were carried out using density functional theory (DFT) with the Perdew-Burke-Ernzerhof (PBE) form of the generalized gradient approximation (GGA) functional.The Vienna ab initio simulation package (VASP) was used.The energy cutoff for plane wave expansions was set to 400 eV, and the energy (converged to 1 e À5 eV atom À1 ) and force (converged to À2 e À2 eV A À1 ) were set as the convergence criteria for geometry optimization.The 3D-graphene/Si (100) model and N-doped 3D-graphene/Si (100) model were constructed (see Supporting Information).The Brillouin zones were sampled with the gamma-centered Monkhorst-Pack (5 9 5 9 1) k-points meshes for all models.As for the slab models, a vacuum space of 15 A was added to the nonperiodic direction to avoid interaction between periodic images.In addition, the DFT-D3 method was included to improve the description of the long-range weak van der Waals (vdW) interaction for all DFT calculations.The optimizations and Raman spectra of molecules were carried out with Gaussian 16 program, using the ωB97XD hybrid function and the 6-31G(d) level, with the Raman frequency correction factor 0.9490, which makes it easy and accurate to investigate the weak interaction changes between molecules.DFT-D4 correction provided a satisfactory result and fit well with the experimental findings.The molecular orbital calculations were manipulated by the Multiwfn 3.8 package.Details about the calculation method are presented in Supporting Information.

Figure 2 .
Figure 2. a) Electrostatic potential of 3D-graphene/Si structure and N-doped 3D-graphene/Si structure, respectively.b) The ground state charge transfer between R6G and 3D-graphene/Si structure and N-doped 3D-graphene/Si structure, respectively.c) The frontal view of the R6G & 3D-graphene/Si model.d) Front view of the charge distributions for R6G & 3D-graphene.e) Side view of the charge distributions for R6G & 3D-graphene.f) The frontal view of the R6G & N-doped 3D-graphene/Si model.g) Front view of the charge distributions for R6G & N-doped 3D-graphene.h) Side view of the charge distributions for R6G & N-doped 3D-graphene.

Figure 3 .
Figure 3. Investigate the charge transfer mechanism by SKPM and CAFM.N-doped 3D-graphene/Si surface potential distribution a) before and b) after adsorption of R6G, respectively.c) The average surface potential value before and after adsorption of R6G.Current maps of the N-doped 3D-graphene/Si under different conditions: d) in the dark and e) in the presence of light.f) The average value of the currents produced under dark and light conditions.

Figure
Figure 4a,b separately show the Raman spectra of both CRV (10 À4 M) and R6G (10 À4 M) on N-doped 3D-graphene/Si and 3D-graphene/Si substrates.The experimental values and calculated Raman spectra of these molecules are compared in Figure S9, Supporting Information, and the inset shows the corresponding molecular model.Comparing the characteristic peak intensities of the two molecules (CRV molecules at 1177, 1371, 1585 and 1621 cm À1 ; R6G molecules at 1187, 1365, 1511 and 1651 cm À1) on N-doped 3D-graphene/Si and 3Dgraphene/Si, as shown in Figure4c,d, it indicates that the peaks intensity of the N-doped 3D-graphene/Si substrate is almost three times that of 3D-graphene/Si substrate.[26][27][28][29][30]The enhancement factors of 3Dgraphene/Si substrate and N-doped 3D-graphene/Si substrate are 1.01 9 10 4 and 2.78 9 10 6 , respectively.The detailed calculation method is presented in Supporting Information.Compared with 3Dgraphene/Si system, the N-doped 3D-graphene/Si system has a more abundant electronic density of states near the Fermi level and more photogenerated charges produced by optical absorption enhancement, as shown in FigureS10, Supporting Information.It has the Fermi energy level closest to the HOMO of the molecule.This makes the EF

Figure 6 .
Figure 6.a) SERS measurement process using a wavelength of 532 nm.b) The structural model of the Thiram molecule.c) The comparison of the experimental values (Measured on N-doped 3D-graphene/Si substrate) with calculated Raman Frequencies of thiram.d) Raman spectra of 10 À6 M-10 À9 M thiram measured on N-doped 3D-graphene/Si substrate.e) Raman spectrum of Thiram extracted from apple peel measured on N-doped 3D-graphene/Si substrate.f) Intensities of the 1381 and 1507 cm À1 peak were plotted against the concentration for thiram.