TRADELABOR has more than 20 years of experience in the control and treatment of air, working with an experienced and qualified technical staff and with the most advanced technology in this area, which together guarantee the quality of the services provided.
Nowadays, pesticides are routinely applied in agriculture to protect crops and fruit from diseases, pests and weeds. However, studies reveal that the vast majority of applied pesticides enter the environment, agricultural products, and even the food chain gratuitously, which may directly pollute the agro-environment and cause health risk even at trace levels [1,2].
Thiabendazole (TBZ), a benzimidazole derivative, is a fungicide used to protect crops from different fungal diseases, such as mould, rot, and blight. TBZ is also applied to keep the freshness of fruit, which is used in postharvest prior to the waxing procedure on apples, pears and others. The USA Environmental Protection Agency (EPA) classifies TBZ as carcinogenic as it can disturb thyroid hormone balance and cause liver damage [3,4]. Likewise, thiram, a typical dithiocarbamate fungicide, is also widely applied in agriculture, and its residue can also cause serious health problems [1,5]. Therefore, scrutiny of such toxic residues in food is an essential step in regulating and monitoring the levels of pesticides.
Traditional methods, such as gas chromatography (GC), high-performance liquid chromatography (HPLC), and their combinations with ultraviolet (UV) or mass spectroscopy, are widely applied for the detection of pesticides residue in agricultural products. However, these conventional approaches have several limitations including complicated pretreatment steps, requiring expensive instruments, operational complexity, lack of instrument portability, and difficulties in real-time monitoring [4,6]. Although immunoassays, electrochemical methods, and capillary electrophoresis are common rapid-detection techniques [, , ], they also suffer from some inherent defects, such as solution instability and limited storage time. Therefore, there is a need to develop fast, simple, and cost-effective analytical methods for noninvasive rapid and sensitive detection of pesticides to prevent potential health risks.
Recently, the advancement in nanotechnology and nanomaterials has provided opportunities for developing rapid and nondestructive tools for detecting chemical contaminants in complex food matrixes. Among these tools, surface-enhanced Raman spectroscopy (SERS), which is an integration of nanotechnology and Raman spectroscopic technique [, ], has emerged as an alternative powerful detection method [, , , , , , , ]. SERS is an enhancement of the normal Raman spectroscopic method to obtain significantly strong signals and rich molecular fingerprints vibrational information with great sensitivity and specificity [, , , ]. Such a high sensitivity is due to the use of nanoscale-roughened noble metal nanostructures (typically, Ag or Au substrates), which can tremendously enhance the Raman signals [, , , ]. Furthermore, numerous methodologies have been established to fabricate high quality and sensitive SERS substrates. Among them, the liquid-liquid interfacial self-assembly is considered as an ideal technique to prepare SERS substrates with the potential to generate intense hot-spots at the junction of neighbouring nanoparticles and it has the advantages of easy operation, cost-effective fabrication, fast preparation at room temperature, good reproducibility and immediate applicability for SERS detection [, , ].
Therefore, increasing studies concerning SERS have been conducted to detect pesticide residues in fruit or fruit juice [, , , , , ]. However, in some studies, destructive pretreatment steps are needed before SERS measurements [4,28,35], while some others only focus on single-component detection of a pesticide at one time, or just perform qualification analysis without quantifying multiple pesticides [, , , , ]. In fact, the mixture of two or more pesticides is normally applied in agriculture.
During simultaneous detection of multiple pesticides in a single sample, overlapping of the characteristic peaks is the main limitation in some SERS studies. Therefore, strategies to differentiate and extract a particular component from a mixture is very important. Multivariate curve resolution (MCR), or self-modelling curve resolution (SMCR) and self-modelling mixture analysis (SMA) are often adopted to identify the involvement of each component spectrum by decomposing the data matrix, and both MCR and SMA have the potential to decompose a data matrix into the outer product of non-negative factors [, ].
Basically, MCR is able to deconvolve the two-way signals from an original data matrix including unresolved multi-component mixtures and provide a chemically meaningful additive bilinear model of pure contributions . Self-modelling mixture analysis (SMA), especially the simple-to-use interactive self-modelling mixture analysis (SIMPLISMA), is a basic iterative method of MCR. They aim at selecting the most dissimilar rows or columns in a single data matrix or a multiset structure, providing initial estimates of spectra or of concentration profiles [, ]. SMA can resolve (decompose) spectral mixture data into pure component spectra (or factors) and their contributions (or scores) without using prior information about the mixtures and is suitable for datasets that are not evolving or are naturally ordered in time . Additionally, the feasibility of quantifying mixed pesticides using the model built based on pure pesticide, and the feasibility of extracting and quantifying pure components from the spectra of mixed pesticides with overlapping characteristic peaks to eliminate mutual interference between components has been certified by the SMA method [, ].
Therefore, the objective of the current work was to develop a deployable SERS technology combined with SMA method for noninvasive detection of the residues of mixed thiram and TBZ on the surface of apple, pear, and tomato. Firstly, high-density Au NR arrays as sensitive SERS substrates were synthesized and fabricated using the seed-mediated method and organic-aqueous interfacial self-assembly method, respectively. A simple and rapid swab-extract method was then developed for the recovery of the mixed pesticides on the surface of the contaminated fruits. Finally, the analysis of spectral data for the pesticide mixture was conducted using the developed SMA method to identify and extract Raman signatures of each composition (Scheme 1). To the best of our knowledge, this is the first time that simultaneously realize qualitatively and quantitatively a single pesticide from the mixed pesticides spectra on fruit, meanwhile presented the resolved pure components of each pesticide using SERS technique with SMA method. This method could be extended to detect other contaminants in the real world. It was thus anticipated that the method should pave the way for exploiting the potential of multi-analytes measurements in practical applications.
2. Materials and methods
Gold chloride trihydrate (HAuCl4·3H2O, Mw: 393.83, CAS: 16961-25-4), hexadecyl trimethyl ammonium bromide (CTAB, Mw: 364.45, CAS: 57-09-0), rhodamine 6G (R6G, Mw: 479.01, CAS: 989-38-8), tetramethylthiuram disulphide (Thiram, Mw: 240.43, CAS: 137-26-8), thiabendazole (TBZ, Mw: 201.25, CAS: 148-79-8), and methanol (HPLC reagent, CAS: 67-56-1) were purchased from Aladdin Industrial Co., Ltd. (Shanghai, China). Silver nitrate (AgNO3, Mw: 169.87, CAS: 7761-88-8), ascorbic acid (AA, Mw: 176.12, CAS: 50-81-7), and acetone (Mw: 58.08, CAS: 67-64-1) were obtained from Sinopharm Chemical Regent Co., Ltd. (Shanghai, China). Sodium borohydride (NaBH4, Mw: 37.83, CAS: 16940-66-2) was supplied by Tianjin Fuchen Chemical Reagent Factory (Tianjin, China). Sodium oleate (NaOL, Mw: 304.44, CAS: 143-19-1), nitric acid (HNO3) and hydrochloric acid (HCl) were provided by Shanghai Macklin Biochemical Technology Co., Ltd. (Shanghai, China). Cyclohexane (Mw: 84.16, CAS: 110-82-7) and ethanol (CAS: 64-17-5) were bought from Tianjin Fuyu Fine Chemical Co., Ltd. (Tianjin, China). Silicon wafer was supplied by Topvendor Technology Co. Ltd. (Beijing, China). Red Fuji apple (Malus pumila Mill, Shandong, China), crown pear (Pyrus bretschneideri Rehd., Hebei, China), cherry tomatoes (Lycopersicon esculentum Mill., Yunnan, China) and the swab were bought from China Resources Vanguard Supermarket (Guangzhou, China). Ultrapure water (resistivity ˃ 18.2 MΩ cm−1) purified by a Milli-Q system (Millipore Co., Billerica, MA, USA) was used for all the solutions. All glassware was treated with freshly prepared aqua regia (HCl: HNO3 = 3:1, v:v) followed by rinsing with copious amounts of water.
2.2. Preparation of Au NRs
Highly consistent Au NRs were prepared using a modified seed-mediated method with a binary surfactant mixture [, ]. In order to synthesize high-quality seed particles for Au NRs growth, the amount of NaBH4 to be added for HAuCl4 reduction was very important. In details, the seed solution was synthesized as follows: 5 mL of CTAB (0.1 M) solution was mixed with 125 μL HAuCl4 (10 mM), and 0.3 mL of freshly prepared ice-cold NaBH4 (10 mM) was injected into the mixture under vigorous stirring (1200 rpm, 2 min) by a magnetic stirrer (MS7-H550-Pro, DLAB Scientific Co., Ltd., Beijing, China), and change in colour from pale to brownish was observed with the addition of NaBH4, and the resulting seed solution was then incubated in an incubator (DZF-6050, Shanghai Qixin Scientific Co., Ltd., Shanghai, China) at 30 °C for 2 h.
To prepare growth solution, 0.7 g CTAB and 0.1234 g NaOL were dissolved in 25 mL warm water (50 °C) in a 100 mL conical flask. After cooling to 30 °C, an amount of 1 mL AgNO3 (10 mM) was added in the flask with continuously stirring for 5 min at 600 rpm, then kept undisturbed in the incubator at 30 °C for 15 min. Next, 25 mL of HAuCl4 (1 mM) was mixed with the solution followed by 90 min of stirring (600 rpm), and the yellow solution became colourless. Then an amount of 2.5 mL HCl (1.0 M) was added to adjust the pH, followed by gentle stirring at 400 rpm for 15 min. After that, 80 μL of AA (0.1 M) was added into the solution under vigorous stirring at 1200 rpm for 30 s. Finally, 100 μL of seed solution was injected into the growth solution followed by stirring at 700 rpm for 1 min. The resultant mixture was kept undisturbed in the incubator (30 °C, 12 h) to complete the growth. The prepared Au NRs colloid stock solution was purified and concentrated by centrifugation (8000 rpm, 25 min) (JW-3024HR, Anhui Jia Wen Instrument and Equipment Industry Co., Ltd., Hefei, China) and washed with ultrapure water for three times to remove the residual CTAB and NaOL before further uses.
2.3. Interfacial self-assembly of Au NRs array
As shown in Fig. 1, Fig. 5, 5 mL of Au NRs colloid stock solution was concentrated to 3 mL and then put in a 5 mL beaker, which was cleaned ultrasonically in acetone for 30 min. The organic-inorganic interphase was formed by adding 1 mL of cyclohexane on the Au NRs colloid surface. By injecting 1 mL of ethanol rapidly into the system using a 1.5 mL range syringe, interfacial self-assembly reaction was induced and a gold film was formed on the cyclohexane-aqueous interphase. After cyclohexane evaporation, the Au NRs array film floating on water was transferred onto the silicon wafer (6 × 6 mm2) and allowed for drying at room temperature of 25 °C, which was used as the substrate and fixed on the Si wafer. Finally, the self-assembly Au NRs array substrate was immersed in ethanol for 30 min to remove the CTAB layer from the surface of Au NRs followed by air-dried before uses [, ].
The UV–Vis extinction spectra of Au NRs colloids were acquired from a UV spectrophotometer (UV-1800, Shimadzu Co., Kyoto, Japan). The transmission electron microscopy (TEM) images were collected from a JEM-2100 Plus high-resolution transmission electron microscope (JEOL Ltd., Tokyo, Japan) operating at an acceleration voltage of 200 kV. The field-emitting scanning electron microscope (FE-SEM) images were acquired from a scanning electron microscope (Zeiss Merlin FESEM, Carl Zeiss NTS GmbH, Oberkochen, Germany) at an acceleration voltage of 10.0 kV. The aspect ratio of Au NRs was calculated by measuring the length and width of the Au NRs from the SEM images using a particle size distribution software (Nano Measurer 1.2.5, Department of Chemistry, Fudan University, Shanghai, China, https://nano-measurer.software.informer.com/).
2.5. SERS measurements of probe molecule R6G and pesticides
In order to evaluate the SERS sensitivity, uniformity and feasibility of the proposed Au NRs array substrates developed in the current study, a series of R6G and thiram in water, and TBZ in methanol were prepared. In details, an amount of 0.5 mL solution sample was put in a 1 mL centrifugation tube and then an Au NRs array substrate was put into the tube. After incubation for 30 min, the Raman spectra was recorded using a laser confocal Raman microspectroscopy system (LabRAM HR, Horiba France SAS, Villeneuve d’ Ascq, France) equipped with a 50× long working distance (LWD) visible objective (LMPlanFL N 50×/0.50) (Olympus Co., Center Valley, PA, USA) and a 633 nm He–Ne laser (laser rated power: 17 mW) as excitation source. The laser power was attenuated to 25% for SERS measurements (4.25 mW) and the exposure time was set for 10 s for each scanning. Each spectrum was the average of 10 scans.
For Raman mapping in uniformity evaluation, Au NRs array substrates that incubated with 10−6 M R6G was laid on the x-y mapping stage. Raman spectra of the substrates with an area of 40 × 40 μm2 and a step of 0.5 μm were collected in the range of 543–1683 cm−1 under 50 × microscopic objective with the 633 nm laser.
For nondestructive and rapid detection of thiram, TBZ, and their mixture on the surfaces of fruits (apple, tomato, and pear), a modified surface swab method  was performed. In details, 50 μL of pesticide solution was spiked on the surface of the fruit (1 × 1 cm2) by a pipette followed by air-drying. Then the fruit surface was swabbed using a swab pre-soaked with methanol-water (1:1, v/v) solution, after that, the head of the swab was removed and immersed in 500 μL of the methanol-water (1:1, v/v) solution in a 1 mL centrifugation tube and vortexed for 30 s to extract the pesticides, i.e., the pesticide sample on the fruit surface was diluted in 10 times during extraction procedure. Finally, an Au NRs array substrate was put into the tube and incubated for 30 min for Raman measurements, or the 500 μL extracted pesticides solution was filtrated using a filter before HPLC analysis.
2.6. HPLC analysis of thiram and TBZ
Chromatographic analysis of thiram and TBZ was performed according to the Chinese National Food Safety Standard “Determination of thiabendazole residue in fruits and vegetables: Liquid chromatography” (GB 23200.17–2016) for TBZ, and the Chinese Entry-Exit Inspection and Quarantine Industry Standard “Determination of thiram residue in fruit and vegetable for export” (SN/T 0525–2012) for thiram, respectively. In details, the standard solution or samples of thiram and TBZ were analysed using HPLC instrument (ACQuity® Arc™, Waters Co., MA, USA) with a reverse-phase C-18 column (XBridge® BEH, 100 mm × 2.1 mm, particle size 23.5 μm, Waters Chromatography Ireland Ltd., Dublin, Ireland) equipped with a quaternary solvent manager-R (ACQuity® Arc™, Waters Pacific Pte Ltd., Singapore), a sample manager FTN-R (ACQuity® Arc™, Waters Co., MA, USA) and a dual absorbance ultraviolet detector (2998 PDA detector, Waters Pacific Pte Ltd., Singapore). The samples were eluted using a linear isocratic elution procedure of methanol and water (50/50, v/v) used as mobile phases at a flow rate of 0.417 mL/min at 25 °C. The detection wavelength was set at 270 nm for thiram and 300 nm for TBZ, respectively.
2.7. Processing of SERS spectra of mixed pesticides
All the collected Raman spectra were processed using LabSpec 6 spectroscopy software suite (Horiba France SAS, Villeneuve d’Ascq, France): de-spiking, de-noising and baseline correction.
The SERS enhancement factor (EF) was calculated using the following equation :(1)Enhancementfactor=ISERS×NRamanIRaman×NSERSwhere NSERS and NRaman are the corresponding numbers of molecules of R6G for SERS and Raman measurements, ISERS and IRaman are the corresponding SERS and Raman intensities, respectively. The conventional Raman spectrum of R6G molecule (Fig. S1) and details of the calculation are available in the supporting information.
Linear calibration curves were determined by monitoring SERS intensities of characteristic peaks as a function of R6G or pesticides concentration. The LOD was determined using the following formula:(2)LOD=3Skwhere S represents the standard deviation of Raman intensity of blank samples, k is the slope of the calibration curve .
In this study, an alternating least squares (ALS) algorithm, which decomposes mixture data into single component spectra and contributions, was used in SMA. Initial estimates were determined using a pure variable detection method based on the SIMPLISMA approach . Purity function in PLS-TOOLBOX (Eigenvector Research, Wenatchee, WA, USA) was employed to conduct a self-modelling analysis  of SERS spectra of pesticides mixture. The spectral data analysis procedures were executed using in-house software developed by MATLAB (MathWorks, Natick, MA, USA). The program of the SIMPLISMA used refers to the code available in literature . The basic formulas for SMA are as follows:(3)D=CST+Ewhere D is an (m × n) matrix of mixed spectral data, C is the contribution matrix (m × k), ST (k × n) is the transpose of S (n × k), which represents the matrix of pure component spectra, k is the number of components, E is the residual error. The contribution C can be estimated using the spectral intensities. As a result, S can be determined by the following equation:(4)S=DTC(CTC)−1where S represents the estimated pure component spectra. Next, the estimated contribution C* can be calculated by the following equation:(5)C*=DSSTS−1
More details for SMA in this study are available in supporting information.
3. Results and discussion
3.1. Characterization of Au NRs and Au NRs arrays substrates
The prepared Au NRs colloids were characterized by FE-SEM, TEM, and UV-NIR spectroscopy and the results are displayed in Fig. 1A–D. As shown in Fig. 1A, the synthesized Au NRs showed uniform morphology, and their average aspect ratio was 3.4 with an average width diameter of 23.3 ± 1.9 nm and a length diameter of 79.8 ± 6.4 nm. Fig. 1B and C shows the TEM images of fabricated Au NRs in 15 KX (Fig. 1B) and 500 KX (Fig. 1C) magnifications with the lattice fringes of 0.234 nm (high-resolution Fig. S2 is available in supporting information), and the Au NRs were single-crystalline growing along the  direction . The extinction spectrum of Au NRs colloids stock solution in Fig. 1D exhibited two distinct absorption characteristic bands: the transverse localized surface plasmon resonance (LSPR) at 515 nm and a longitudinal LSPR at 745 nm, due to the surface plasmonic resonance. The UV-NIR spectrum of the residual Au NRs colloids after the interfacial self-assembly showed the same LSPRs with the Au NRs colloids stock solution, with dramatically decreased absorbance intensity, revealing that the Au NRs particles in colloid were moved to organic-inorganic interphase to complete the self-assembly.
In aqueous solution, the surface plasmon resonance (SPR) absorption maximum (λmax) of Au NRs was linearly proportional to the aspect ratio (AR) by the following relationship :(6)λmax=95AR+420
Based on Equation (1) and the AR (3.4) of the Au NRs, λmax was calculated as 743 nm, which was in line with the longitudinal LSPR (745 nm), presented in the UV- NIR spectra (Fig. 1D). Correspondingly, the theoretical AR can also be calculated as 3.4 by using Equation (6) and the longitudinal LSPR (745 nm).
Moreover, Fig. 1E presents the preparation procedure for Au NRs array by interfacial self-assembly using as-prepared Au NRs colloids. In details, the condensed Au NRs colloid (Fig. 1E1) as the inorganic phase and the cyclohexane as the organic phase formed distinct interphase (Fig. 1E2). Then, the inducer ethanol was rapidly squirted into the system (Fig. 1E3) to decrease the surface charge of Au NRs, facilitating the Au NRs-water surface tension and driving the Au NRs rapidly to the cyclohexane-aqueous interphase, which eventually formed a compact Au NRs array (Fig. 1E4) . With the evaporation of cyclohexane spontaneously at the room temperature, the Au NRs array film that floating on aqueous phase (Fig. 1E5) was transferred onto the silicon wafer (6 × 6 mm2) by a tweezer (Fig. 1E6). After air-drying, the Au NRs array was fixed on the silicon wafer with a golden appearance (Fig. 1E7). The FE-SEM images of the Au NRs array displayed in Fig. 1F and G and Fig 2B3-D3 further demonstrated that Au NRs were closely packed on the silicon wafer with a smooth, dense and crack-free morphology, which would provide abundant hot-spots for enhancing the Raman scattering signals of analytes on the array.
3.2. SERS performance, uniformity and reproducibility of Au NRs arrays
Under a specified volume of the container (5 mL beaker) and reactants, the structure, morphology and SERS performance of the Au NRs array depended on the concentration of the Au NRs colloid. As shown in Fig. 2, the morphology of the Au NRs array was tuned by adjusting the volume of the Au NRs colloid stock solution from 3 to 5 mL, and the SERS performance and uniformity were evaluated using Raman probe molecule R6G. For the interfacial self-assembly reaction using 3 mL prefabricated Au NRs colloid stock solution, the SERS intensity of R6G at 1510 cm−1 was 7985 ± 1081 (Fig. 2A1) with the relative standard deviations (RSD) of 13.53%, collected from 100 randomly selected spots on the Au NRs array substrates (Fig. 2E1), the Au NRs were not packed closely, and there were many cavities in the array (Fig. 2B1-D1). When 4 mL Au NRs colloid stock solution (re-dispersed in 3 mL ultra-pure water after centrifugation in wash procedure) was used, the SERS intensity of R6G at 1510 cm−1 was 19272 ± 4121 (Fig. 2B1), which was increased by more than two-folds, however, the morphology of the substrates was rough (Fig. 2B2-D2) with RSD of 21.38% (Fig. 2E2), and such an RSD was too high for practical applications as quantification analysis required RSD to be less than 20% . When the Au NRs colloid stock solution was increased to 5 mL, no obvious cavities in the array were observed, showing that Au NRs were closely assembled with a compact and crack-free appearance (Fig. 2B3-D3), and the SERS intensity of R6G at 1510 cm−1 was 43185 ± 4009 (Fig. 2A3) with RSD of 9.28% (Fig. 2E3), demonstrating that this could be the ideal substrate with excellent SERS activity and uniformity. Therefore, 5 mL Au NRs colloid stock solution was selected for the interfacial self-assembly fabrication of Au NRs array in the following investigation.
For further verification of the uniformity of the fabricated Au NRs array substrates, Raman mapping was carried out on a selected area of 40 × 40 μm2 on the substrate with a resolution of 0.5 μm. Fig. S3 displays the recorded 6561 SERS spectra (81 × 81 spots) of R6G (Fig. S3A) incubated with the Au NRs array substrate and their Raman mapping images (Fig. S3C). The colour distribution displayed good uniformity and without obvious fluctuations. The RSDs by targeting Raman intensities at 612, 772, 1182, 1311, 1362, 1510 and 1650 cm−1 were calculated to be 15.02%, 13.81%, 13.40%, 14.90%, 13.45%, 13.57%, 14.77%, respectively. These values were higher than those from 100 randomly selected spots on the Au NRs array substrate (9.29% at 1510 cm−1) as the area illuminated by the laser (1.86 μm2) was larger than the step (0.5 μm) for Raman mapping. This phenomenon could be ascribed to the possible destruction of the analyte on the substrate caused by continuous laser irradiation during Raman mapping, leading to the fluctuation of Raman intensities between the centre and the edge of the mapping area. Nevertheless, these results revealed that the Au NRs array showed excellent uniformity and could serve as reliable SERS substrates.
Furthermore, for the batch-to-batch reproducibility of the Au NRs array substrates evaluation, the SERS spectra of R6G at 90 random spots (3*3*10) from three batches of Au NRs colloids and nine batches of Au NRs array were collected, and their SERS intensity distribution at 1510 cm−1 are presented in Fig. S3B with the average of 41524 ± 5928 and the RSD of 14.28%. Additionally, the SEM image of the Au NRs array substrate after incubation with the sample and SERS analysis is shown in Fig. S4. The Au NRs array had no significant difference from that before detection, indicating no occurrence of lixiviation of the Au NRs from the substrate. Therefore, the ideal SERS performance with good uniformity and reproducibility confirmed that the Au NRs array could serve as excellent SERS substrates for practical applications.
3.3. Detection of thiram and TBZ in standard solutions
To examine the performances of the prepared Au NRs array substrates in practical uses, thiram and TBZ were quantitatively detected by SERS method using R6G as the Raman probe molecule. Fig. 3A shows the SERS spectra of R6G solution with R6G concentrations from 1 × 10−9 to 1 × 10−6 M (1 × 10−3 μM–1 μM). From the spectra of the blank control sample, weak background of Au NRs array was observed, and the slight bulge at 1444 cm−1 was attributed to C–N binding mode of CTAB (Fig. S5) on the surface of Au NRs , which almost disappeared when detecting the analytes. With an increase in the concentrations of R6G, characteristic peaks at 612, 772, 1182, 1311, 1362, 1510, 1578 and 1650 cm−1 were clearly observed. The assignments of these peaks are listed in Table S1 . Considering the intensity and stability, the peak at 1510 cm−1 was selected to establish the relationship of the SERS intensity as a function of R6G concentrations. The SERS intensity is linearly dependent on the R6G concentration in a narrow range but showed a logarithmic relationship in a wide concentration range. It is attributed to the saturation of the analyte molecules on the substrate at high concentrations. A similar phenomenon can also be found for thiram (Fig. 3E and F), TBZ (Fig. 3H and I) and in the literature . Therefore, piecewise functions were adopted to express the relationship of the SERS intensity and analyte concentrations. Fig. 3 B–C demonstrate that the SERS intensity at 1510 cm−1 showed a linear correlation with the logarithmic R6G concentrations ranging of from 1 × 10−3 to 1 × 10−1 μM and the R6G concentrations from 0.1 to 1 μM. The calibration curves were Y1510 = 3369.15 Lg(X) + 10632.22 and Y1510 = 41447.5X + 3192.6 with coefficients of determination (R2) of 0.9883 and 0.9981 (Table S2), respectively. The LOD was calculated as 8.3 × 10−4 μM. Furthermore, the EF of the Au NRs array substrates was calculated as 1.64 × 106 using the Raman probe molecule R6G, and details of the calculation are available in the supporting information.
Thiram and TBZ were further detected on the Au NRs array substrates. Fig. 3D, G displays the SERS spectra of thiram (Fig. 3D) with concentrations from 1 × 10−10 to 1 × 10−6 M (1 × 10−4 to 1 μM) and TBZ (Fig. 3G) with concentrations from 1 × 10−9 to 1 × 10−5 M (1 × 10−3 to 10 μM), respectively. The SERS intensities of the corresponding characteristic peaks at 562, 932, 1144, 1378, 1512 cm−1 for thiram, and 778, 880, 1007, 1278, 1321, 1578, 1601 and 1622 cm−1 for TBZ increased with increases in the concentrations of thiram and TBZ increase. The assignments of the peaks of thiram and TBZ are also listed in Table S1 [1,, , ]. For the detection of thiram, the most intense peak at 1378 cm−1 was attributed to C–N stretching and CH3 deformation, which was selected to build the calibration curves of SERS intensities versus thiram concentrations. Fig. 3 E-F exhibit that the SERS intensity at 1378 cm−1 showed linear correlations (Table S2) with the logarithmic concentrations of thiram from 1 × 10−4 to 1 μM and with the concentrations of thiram from 1 to 10 μM. The LOD of thiram was calculated to be 3.6 × 10−5 μM. For the detection of TBZ, the peak at 778 cm−1 was assigned to in-plane CN bending and C–S stretching, and the peak at 1007 cm−1 was attributed to C–C stretching and C–N stretching, and both were selected to establish the relationships of SERS intensities with concentrations of TBZ. Fig. 3 H–I present that the SERS intensities at 778 and 1007 cm−1 displayed linear correlation (Table S2) with the logarithmic concentrations of TBZ from 1 × 10−3 to 10 μM, and the LOD of TBZ was calculated to be 1.03 × 10−3 μM and 9.5 × 10−4 μM, respectively. Additionally, it was noticed that the sensitivity for thiram detection was significantly higher than that of TBZ, indicating that more Au–S bond formed on the substrate during thiram detection. These results indicated the accuracy of SERS detection based on the Au NRs array, and its high sensitivity made the high-density Au NRs array a promising substrate for practical food safety detection.
3.4. SERS and HPLC detection of thiram and TBZ on the fruit surface
Fig. 4 shows the residues of thiram and TBZ on the surface of the apple, pear and tomato as detected using SERS technology combined with the modified surface swab method. The characteristic peaks of thiram at 1378 cm−1 and of TBZ at 1007 cm−1 were chosen to build the calibration curves (Table S2) for SERS intensities versus the concentrations of thiram (Fig. 4B1-3) or TBZ (Fig. 4D1-3), respectively. Two intense and clear peaks could still be identified when the concentrations of the thiram (Fig. 4A1-3) and TBZ (Fig. 4C1-3) were decreased to ppb level. The LODs of pesticides on the surface of apple, tomato and pear were 0.81, 0.58, 0.93 ppb for thiram, and 15.89, 15.24, 15.38 ppb for TBZ, respectively. The difference in LODs might be due to different absorptivity on pericarp structures of apple, tomato and pear. The LODs of pesticides on the fruit surface can also be expressed as mass-to-area ratio (0.041, 0.029 and 0.047 ng/cm2 for thiram, and 0.79, 0.76 and 0.80 ng/cm2 for TBZ on the surface of apple, tomato and pear, respectively), or mass of pesticide to mass of fruit (0.039, 0.073 and 0.045 ng/g for thiram, and 0.76, 1.91 and 0.78 ng/g for TBZ on the surface of apple, tomato and pear, respectively), which are listed in Table S3. Details for calculating the concentration conversion are available in the supporting information.
Nevertheless, the LODs were lower than both the maximum residue limit (MRLs) for thiram (5 ppm) and TBZ (12 ppm) in fruits prescribed by the USA Environmental Protection Agency (EPA, 2020) , and for thiram (5 mg/kg) and TBZ (3 mg/kg) in fruits prescribed by GB 2763–2016 “China National Food Safety Standard: Maximum Residue Limits for Pesticides in Food”, respectively. The comparison of the current results with those reported in the literature is listed in Table S4.
In addition, HPLC analysis was adopted to validate the SERS method. Fig. S6 shows the retention time (Fig S6 A, D) and standard curve of thiram (Fig S6 B-C) and TBZ (Fig S6 E-F) in methanol, respectively. Considering that the swab cannot release all the pesticides in the extract procedure, the extraction efficiency was calculated to calibrate the recovery on the fruit surface by using the HPLC standard curve of thiram and TBZ in methanol. To calculate the extraction factor, 50 μL of 100 ppm thiram or TBZ solution was dropped on a clean glass slide and dried. The same procedure with the swab test on fruit surface was carried out. The final pesticides concentration was calculated using the standard curve. The average extraction factor was calculated to be 72.1% for thiram and 65.8% for TBZ using the following equation:(7)Extractionfactor=CD×VDCS×VS×100%and the calibrated recovery on the fruit surface was calculated by the equation below:(8)Calibratedrecovery=CD×VDExtractionfactor×VSwhere CD is the detected concentration of pesticide on glass or fruit surface, CS is the spiked concentration of pesticide on glass or fruit surface, VD is the detected volume of pesticide solution extract from swab, VS is the spiked volume of pesticide on glass or fruit surface, respectively.
Furthermore, the recovery experiments were conducted using SERS technique as well as HPLC method, and the results are shown in Table 1. Satisfactory recoveries of 79.23–116.08% for detecting thiram and 74.38–127.7% for detecting TBZ using SERS technique, and 89.46–113.73% for thiram and 81.05–117.02% for TBZ using HPLC method were accomplished, respectively. The results provided evidence that the fabricated Au NRs array substrates combined with the surface swab method were practicable for sensing contaminants in real applications.
Table 1. Recoveries for detecting thiram and TBZ on surface of apple, tomato, and pear using SERS technique and HPLC method.
|Samples||Spiked (ppm)||Detected by SERS (ppm)||Recovery (%)||Detected by HPLC (ppm)||aCalibrated Recovery (%)|
|Thiram on apple||20||19.20||96.00||1.36||94.31|
|Thiram on tomato||20||16.58||82.88||1.34||92.93|
|Thiram on pear||20||21.89||109.45||1.29||89.46|
|TBZ on apple||20||14.86||74.30||1.54||117.02|
|TBZ on tomato||30||25.98||86.59||1.65||83.59|
|TBZ on pear||50||40.11||80.23||3.12||94.83|
Calibrated Recovery (%) for the pesticides were calculated using equation (5). In this context, the spiked volume of pesticide solution is 50 μL while the detected volume of pesticide solution is 500 μL, the pesticide sample on the fruit surface was diluted in 10 times during extraction procedure.
3.5. Spectral analysis of pesticides mixture using SMA method
For simultaneous measurements of mixed thiram and TBZ, the SMA method was adopted to analyse the obtained Raman spectra from the mixture according to the following equations:(9)IpureA∗A+IpureB∗B+…=Mixture1IpureA∗A+IpureB∗B+…=Mixture2⋯IpureA∗A+IpureB∗B+…=Mixturenwhere A, B, …, are pure spectra of each single component, Ipure A, Ipure B, …, are contributions of A, B, …, respectively.
Fig. 5 shows the self-modelling mixture analysis of the SERS spectra for the mixed pesticides. In the current study, singular value decomposition (SVD) was used to find the number of components . In order to resolve a spectral dataset with SMA, pure variables should be present . For the presence of overlapped peaks in spectra, to avoid selecting poor pure variables that would result in inferior resolution results, second-order derivatives were introduced  and to decrease/suppress any ambiguity in the SMA solution, constraints were introduced . Details of the analysis are available in the supporting information. Finally, the original SERS spectra of the mixture and its corresponding decomposed pure spectra (variables multiplied by scores) of thiram and TBZ are displayed in Fig. 5, in which the blue, pink, and purple rectangles represent the characteristic peaks of thiram, TBZ, and their overlapped peaks, respectively. The characteristic spectra and fingerprint peaks of each individual pesticide were clearly resolved and presented. Results showed that the acquired SERS spectra of each pesticide in the mixture had no obvious variation compared with the spectra of the corresponding pure pesticide, indicating the feasibility of SMA method for extracting the pure spectra from spectral mixture data in detecting complex matrixes.
By taking the advantages of the fingerprint-based SERS technique and the powerful pure spectra extracting method of SMA, mixed pesticides spiked on the surface of the apple, pear and tomato were simultaneously detected and decomposed (Fig. 6A–C), and the resolved individual pure spectra are displayed in Fig. 6A1-C1 for TBZ and in Fig. 6A2-C2 for thiram, respectively. The characteristic peaks of thiram at 1378 cm−1 and of TBZ at 1007 cm−1 were selected to calculate the recovery of the two pesticides on the surface of three fruits. Compared with the mixture spectra, satisfactory recoveries of 66.0–142.54% for detecting thiram and 83.06–136.76% for detecting TBZ at an appropriate proportion of the pesticides (Table 2) were obtained, revealing that SERS technique combined with SMA method had the potential in practical applications for multiple-analytes sensing in complex systems.
Table 2. Recoveries for detecting the mixture of thiram and TBZ on surface of apple, tomato, and pear using SERS technique combined with SMA method.
|Samples||Spiked pesticides (ppm)||Detected TBZ (ppm)||Detected thiram (ppm)||Recovery of TBZ (%)||Recovery of thiram (%)|
Note: B is for TBZ and H is for thiram.
In the current study, the mixed pesticides of thiram and TBZ on the surface of the apple, pear and tomato were simultaneously detected using SERS technique combined with organic-inorganic interfacial self-assembly high-density Au NR array substrates. A simple and rapid swab-extract method was developed for the recovery of the pesticides on the surface of contaminated fruits. An SMA method was used to analyse the Raman spectra of the pesticides mixture to extract Raman signatures of each composition. This is the first investigation that qualitative and quantitative analyses of a single component from the pesticides mixture spectra were simultaneously realized, presenting the resolved pure components of each pesticide using SERS technique with the SMA method. The novel method could be extended to detect other mixed contaminants. The method demonstrated the great potential to be served as a useful strategy for monitoring mixed pesticide residues in practical scenarios.
CRediT authorship contribution statement
Bingxue Hu: Writing – original draft, Formal analysis, Investigation. Da-Wen Sun: Supervision, Writing – review & editing. Hongbin Pu: Validation, Funding acquisition. Qingyi Wei: Resources.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
The authors are grateful to the National Key R&D Program of China (2018YFC1603400) for its support. This research was also supported by the Contemporary International Collaborative Research Centre of Guangdong Province on Food Innovative Processing and Intelligent Control (2019A050519001), the Common Technical Innovation Team of Guangdong Province on Preservation and Logistics of Agricultural Products (2019KJ145, 2019KJ101), the Innovation Centre of Guangdong Province for Modern Agricultural Science and Technology on Intelligent Sensing and Precision Control of Agricultural Product Qualities, the Key Research and Development Program of Guangdong Province (2019B020212001) and the Fundamental Research Funds for the Central Universities (2018MS056).
Appendix A. Supplementary data
The following is the Supplementary data to this article:
Multimedia component 1.
A.M. Dowgiallo, D.A. GuentherDetermination of the limit of detection of multiple pesticides utilizing gold nanoparticles and surface-enhanced Raman spectroscopyJ. Agric. Food Chem., 67 (2019), pp. 12642-12651
H. Wu, Y. Luo, C. Hou, D. Huo, Y. Zhou, S. Zou, J. Zhao, Y. LeiFlexible bipyramid-AuNPs based SERS tape sensing strategy for detecting methyl parathion on vegetable and fruit surfaceSensor. Actuator. B Chem., 285 (2019), pp. 123-128
L. He, T. Chen, T.P. LabuzaRecovery and quantitative detection of thiabendazole on apples using a surface swab capture method followed by surface-enhanced Raman spectroscopyFood Chem., 148 (2014), pp. 42-46
F.K. Alsammarraie, M. Lin, A. Mustapha, H. Lin, X. Chen, Y. Chen, H. Wang, M. HuangRapid determination of thiabendazole in juice by SERS coupled with novel gold nanosubstratesFood Chem., 259 (2018), pp. 219-225
K. Wang, D.-W. Sun, H. Pu, Q. WeiShell thickness-dependent Au@Ag nanoparticles aggregates for high-performance SERS applicationsTalanta, 195 (2019), pp. 506-515
B. Hu, D.-W. Sun, H. Pu, Q. WeiRecent advances in detecting and regulating ethylene concentrations for shelf-life extension and maturity control of fruit: a reviewTrends Food Sci. Technol., 91 (2019), pp. 66-82
E. Watanabe, S. Miyake, Y. YogoReview of enzyme-linked immunosorbent assays (ELISAs) for analyses of neonicotinoid insecticides in agro-environmentsJ. Agric. Food Chem., 61 (2013), pp. 12459-12472
M.L. Xu, J.B. Liu, J. LuDetermination and control of pesticide residues in beverages: a review of extraction techniques, chromatography, and rapid detection methodsAppl. Spectrosc. Rev., 49 (2014), pp. 97-120
H. Jayan, H. Pu, D.-W. SunRecent development in rapid detection techniques for microorganism activities in food matrices using bio-recognition: A reviewTrends Food Sci. Technol., 95 (2020), pp. 233-246
H. Pu, Z. Huang, D.-W. Sun, X. Xie, W. ZhouDevelopment of a highly sensitive colorimetric method for detecting 17β-estradiol based on combination of gold nanoparticles and shortening DNA aptamersWater air soil poll., 230 (2019), p. 124
D. Li, Z. Zhu, D.-W. SunVisualization of the in-situ distribution of contents and hydrogen bonding states of cellular level water in apple tissues by confocal Raman microscopyAnalyst, 145 (2020), pp. 897-907
M.-L. Xu, Y. Gao, X.X. Han, B. ZhaoDetection of pesticide residues in food using surface-enhanced Raman spectroscopy: a reviewJ. Agric. Food Chem., 65 (2017), pp. 6719-6726
J.F. Li, Y.J. Zhang, S.Y. Ding, R. Panneerselvam, Z.Q. TianCore-shell nanoparticle-enhanced Raman spectroscopyChem. Rev., 117 (2017), pp. 5002-5069
H. He, D.-W. Sun, H. Pu, L. Chen, L. LinApplications of Raman spectroscopic techniques for quality and safety evaluation of milk: A review of recent developmentsCrit. Rev. Food Sci. Nutr., 59 (5) (2019), pp. 770-793
H. Pu, X. Xie, D.-W. Sun, Q. Wei, Y. JiangDouble strand DNA functionalized Au@Ag NPs for ultrasensitive detection of 17β-estradiol using surface-enhanced Raman spectroscopyTalanta, 195 (2019), pp. 419-425
H. Pu, W. Xiao, D.-W. SunSERS-microfluidic systems: a potential platform for rapid analysis of food contaminantsTrends Food Sci. Technol., 70 (2017), pp. 114-126
X. Xie, H. Pu, D.-W. SunRecent advances in nanofabrication techniques for SERS substrates and their applications in food safety analysisCrit. Rev. Food Sci. Nutr., 58 (2017), pp. 2800-2813
A. Hussain, H. Pu, D.-W. SunSERS detection of sodium thiocyanate and benzoic acid preservatives in liquid milk using cysteamine functionalized core-shelled nanoparticlesSpectrochim. Acta A., 229 (2020), p. 117994
Y. Jiang, D.-W. Sun, H. Pu, Q. WeiUltrasensitive analysis of kanamycin residue in milk by SERS-based aptasensorTalanta, 197 (2019), pp. 151-158
T. Yaseen, H. Pu, D.-W. SunFunctionalization techniques for improving SERS substrates and their applications in food safety evaluation: a review of recent research trendsTrends Food Sci. Technol., 72 (2018), pp. 162-174
P.-L. Wang, L.-H. Xie, E.A. Joseph, J.-R. Li, X.-O. Su, H.-C. ZhouMetal–organic frameworks for food safetyChem. Rev., 119 (2019), pp. 10638-10690
A. Hussain, D.-W. Sun, H. PuSERS detection of urea and ammonium sulfate adulterants in milk with coffee ring effectFood Addit Contam A, 36 (6) (2019), pp. 851-862
A. Hussain, D.-W. Sun, H. PuBimetallic core shelled nanoparticles (Au@AgNPs) for rapid detection of thiram and dicyandiamide contaminants in liquid milk using SERSFood Chem, 317 (2020), p. 126429
Z. Wang, S. Zong, L. Wu, D. Zhu, Y. CuiSERS-activated platforms for immunoassay: probes, encoding methods, and applicationsChem. Rev., 117 (2017), pp. 7910-7963
R. Gillibert, J.Q. Huang, Y. Zhang, W.L. Fu, M.L. de La ChapelleFood quality control by surface enhanced Raman scatteringTrac. Trends Anal. Chem., 105 (2018), pp. 185-190
J. Langer, D. Jimenez de Aberasturi, J. Aizpurua, R.A. Alvarez-Puebla, B. Auguié, J.J. Baumberg, G.C. Bazan, S.E. Bell, A. Boisen, A.G. BroloPresent and future of surface-enhanced Raman scatteringACS Nano, 14 (2020), pp. 28-117
F. Yu, M. Su, L. Tian, H. Wang, H. LiuOrganic solvent as internal standards for quantitative and high-throughput liquid interfacial SERS analysis in complex mediaAnal. Chem., 90 (2018), pp. 5232-5238
K. Wang, D.-W. Sun, H. Pu, Q. WeiTwo-dimensional Au@Ag nanodot array for sensing dual-fungicides in fruit juices with surface-enhanced Raman spectroscopy techniqueFood Chem., 310 (2019), p. 125923
K. Wang, D.-W. Sun, H. Pu, Q. Wei, L. HuangStable, flexible, and high-performance SERS chip enabled by a ternary film-packaged plasmonic nanoparticle arrayACS Appl. Mater. Interfaces, 11 (2019), pp. 29177-29186
S. Fateixa, M. Raposo, H.I.S. Nogueira, T. TrindadeA general strategy to prepare SERS active filter membranes for extraction and detection of pesticides in waterTalanta, 182 (2018), pp. 558-566
K. Wang, D.-W. Sun, H. Pu, Q. WeiA rapid dual-channel readout approach for sensing carbendazim with 4-aminobenzenethiol functionalized core-shell Au@Ag nanoparticlesAnalyst, 145 (2020), pp. 1801-1809
T. Yaseen, H. Pu, D.-W. SunEffects of ions on core-shell bimetallic Au@Ag NPs for rapid detection of phosalone residues in peach by SERSFood Anal Method, 12 (9) (2019), pp. 2094-2105
N. Yang, T.-T. You, Y.-K. Gao, C.-M. Zhang, P.-G. YinFabrication of a flexible gold nanorod polymer metafilm via a phase transfer method as a SERS substrate for detecting food contaminantsJ. Agric. Food Chem., 66 (2018), pp. 6889-6896
A. Jiao, X. Dong, H. Zhang, L. Xu, Y. Tian, X. Liu, M. ChenConstruction of pure worm-like AuAg nanochains for ultrasensitive SERS detection of pesticide residues on apple surfacesSpectrochim. Acta A., 209 (2019), pp. 241-247
G. Fu, D.-W. Sun, H. Pu, Q. WeiFabrication of gold nanorods for SERS detection of thiabendazole in appleTalanta, 195 (2019), pp. 841-849
T. Yaseen, H. Pu, D.-W. SunRapid detection of multiple organophosphorus pesticides (triazophos and parathion-methyl) residues in peach by SERS based on core-shell bimetallic Au@Ag NPsFood Addit Contam A, 36 (5) (2019), pp. 762-778
T. Yaseen, H. Pu, D.-W. SunFabrication of silver-coated gold nanoparticles to simultaneously detect multi-class insecticide residues in peach with SERS techniqueTalanta, 196 (2019), pp. 537-545
P. Wang, L. Wu, Z. Lu, Q. Li, W. Yin, F. Ding, H. HanGecko-inspired nanotentacle surface-enhanced Raman spectroscopy substrate for sampling and reliable detection of pesticide residues in fruits and vegetablesAnal. Chem., 89 (2017), pp. 2424-2431
J. Chen, M. Huang, L. Kong, M. LinJellylike flexible nanocellulose SERS substrate for rapid in-situ non-invasive pesticide detection in fruits/vegetablesCarbohydr. Polym., 205 (2019), pp. 596-600
G. Kwon, J. Kim, D. Kim, Y. Ko, Y. Yamauchi, J. YouNanoporous cellulose paper-based SERS platform for multiplex detection of hazardous pesticidesCellulose, 26 (2019), pp. 4935-4944
Y. Ma, P. Huang, D. Hou, J. Cai, Q. Wang, G. ZhangThe spectral resolution of unknown mixture based on THz spectroscopy with self-modeling techniqueChemometr. Intell. Lab., 150 (2016), pp. 65-73
R. KhodabakhshianFeasibility of using Raman spectroscopy for detection of tannin changes in pomegranate fruits during maturitySci. Hortic., 257 (2019), p. 108670108610.101016/j.scienta.102019.108670
W. Windig, S. MarkelSimple-to-use interactive self-modeling mixture analysis of FTIR microscopy dataJ. Mol. Struct., 292 (1993), pp. 161-170
W. Windig, N.B. Gallagher, J.M. Shaver, B.M. WiseA new approach for interactive self-modeling mixture analysisChemometr. Intell. Lab., 77 (2005), pp. 85-96
C.D. Albuquerque, R.J. PoppiDetection of malathion in food peels by surface-enhanced Raman imaging spectroscopy and multivariate curve resolutionAnal. Chim. Acta, 879 (2015), pp. 24-33
C. Zhai, Y. Peng, Y. Li, K. ChaoExtraction and identification of mixed pesticides’ Raman signal and establishment of their prediction modelsJ. Raman Spectrosc., 48 (2017), pp. 494-500
X. Ye, C. Zheng, J. Chen, Y. Gao, C.B. MurrayUsing binary surfactant mixtures to simultaneously improve the dimensional tunability and monodispersity in the seeded growth of gold nanorodsNano Lett., 13 (2013), pp. 765-771
F.K. Alsammarraie, M. LinUsing standing gold nanorod arrays as surface-enhanced Raman spectroscopy (SERS) substrates for detection of carbaryl residues in fruit juice and milkJ. Agric. Food Chem., 65 (2017), pp. 666-674
K. Wang, D.-W. Sun, H. Pu, Q. WeiSurface-enhanced Raman scattering of core-shell Au@Ag nanoparticles aggregates for rapid detection of difenoconazole in grapesTalanta, 191 (2019), pp. 449-456
W. Windig, B. Antalek, J.L. Lippert, Y. Batonneau, C. BrémardCombined use of conventional and second-derivative data in the SIMPLISMA self-modeling mixture analysis approachAnal. Chem., 74 (2002), pp. 1371-1379
Barry M. Wise, N. B Gallagher, R. Bro, J.M. Shaver, W. Windig, R.S. KochChemometrics Tutorials for PLS-Toolbox and SoloEigenvector Research Inc, Wenatchee, WA (2006), pp. 213-235
W. WindigSpectral data files for self-modeling curve resolution with examples using the Simplisma approachChemometr. Intell. Lab., 1 (1997), pp. 3-16
X. Ye, L. Jin, H. Caglayan, J. Chen, G. Xing, C. Zheng, V. Doan-Nguyen, Y. Kang, N. Engheta, C.R. KaganImproved size-tunable synthesis of monodisperse gold nanorods through the use of aromatic additivesACS Nano, 6 (2012), pp. 2804-2817
X. Huang, S. Neretina, M.A. El‐SayedGold nanorods: from synthesis and properties to biological and biomedical applicationsAdv. Mater., 21 (2009), pp. 4880-4910
A.D. Strickland, C.A. BattDetection of carbendazim by surface-enhanced Raman scattering using cyclodextrin inclusion complexes on gold nanorodsAnal. Chem., 81 (2009), pp. 2895-2903
Z. Fan, W. Na, Z. Jian, Z. Jing, G.J. Weng, J.J. Li, J.W. ZhaoAu@AuAg yolk-shell triangular nanoplates with controlled interior gap for the improved surface-enhanced Raman scattering of rhodamine 6GSensor. Actuator. B Chem., 271 (2018), pp. 174-182
M.-S. Kim, M.-K. Kim, C.-J. Lee, Y.-M. Jung, M.-S. LeeSurface-enhanced Raman spectroscopy of benzimidazolic fungicides: benzimidazole and thiabendazoleBull. Kor. Chem. Soc., 30 (2009), pp. 2930-2934
S. Kumar, P. Goel, J.P. SinghFlexible and robust SERS active substrates for conformal rapid detection of pesticide residues from fruitsSensor. Actuator. B Chem., 241 (2017), pp. 577-583
V. Sharma, V. KrishnanFabrication of highly sensitive biomimetic SERS substrates for detection of herbicides in trace concentrationSensor. Actuator. B Chem., 262 (2018), pp. 710-719
EPAElectronic Code of Federal Regulations: Part 180-tolerances and exemptions for pesticide chemical residues in food(2020)https://www.govinfo.gov/app/details/CFR-2014-title40-vol24/CFR-2014-title40-vol24-part180, Accessed 28th Mar 2020
A. de Juan, J. Jaumot, R. TaulerMultivariate curve resolution (MCR). Solving the mixture analysis problemAnal. Methods., 6 (2014), pp. 4964-4976
Continue at: https://www.sciencedirect.com/science/article/pii/S0039914020302897
The text above is owned by the site above referred.
Here is only a small part of the article, for more please follow the link