*3.5. Detection of Mixed Pesticides on Orange Surface Using AgNPs-PDMS Substrate*

After determining the calibration curve of each pesticide using the AgNPs-PDMS substrate, our next aim is to accomplish semi-quantitative detection of three pesticides simultaneously. A mixed solution of malathion (500 μg L−1), thiram (100 μg L−1), and phoxim (200 μg L−1) was prepared and measured by SERS using the AgNPs-PDMS substrate. In Figure 5a, the characteristic peaks in the SERS spectrum of the mixed sample can be used for distinguishing each pesticide. As labeled in different colors, for each component there are at least two or three exclusive peaks without overlapping. In other words, benefiting from the molecular fingerprint property of SERS, the characteristic information of each pesticide can be extracted from the SERS signal of the mixed sample. According to integrated peak areas at 860, 1378, and 1506 cm−1, the recovery rates were calculated as 104.6 ± 1.9%, 96.5 ± 3.3%, and 118.9 ± 2.4% for malathion, thiram, and phoxim, respectively. The recovery rate for phoxim is higher than expected, presumably due to slightly overlapping with the weak band of thiram at 1513 cm<sup>−</sup>1. Nonetheless, this study confirms that the semi-quantitative analysis of three pesticides from a complex environment is feasible using this method.

**Figure 5.** SERS spectra of mixed (black curve) and single pesticides (thiram, malathion, and phoxim) using an AgNPs-PDMS substrate (**a**). SERS spectrum of orange surface spiked with mixed pesticides after swab sampling using an AgNPs−PDMS substrate; characteristic SERS bands are labeled for each pesticide: thiram (green square), malathion (red circle), and phoxim (blue triangle) (**b**). Inset of (**b**): photo of performing swab sampling using the flexible SERS substrate.

Benefitting from the flexible property of the AgNPs-PDMS substrate, it is possible to transfer pesticide residues from the orange surface onto the substrate for SERS measurement. Comparing with other techniques, this method does not require the additional pre-treatment of oranges samples. In this work, a mixed solution of three pesticides was spiked onto the surface of oranges as the tested sample. After solvent evaporation, the orange surface was sprayed with the minimum amount of ethanol and covered by the AgNPs-PDMS substrate. As shown in the inset of Figure 5b, through gentle pressing and swabbing, adequate contact between the orange surface and the substrate was made, thus sampling of the spiked pesticides was completed. From the SERS spectrum presented in Figure 5b, it is obvious that the characteristic peaks of each pesticide can be identified. In detail, SERS signals at 552 and 1376 cm−<sup>1</sup> are attributed to thiram; peaks at 811, 860, 1093, and 1718 cm−<sup>1</sup> are assigned to malathion; and peaks at 640, 995, 1506, and 1591 cm−<sup>1</sup> indicate the presence of phoxim. It is worth noting that no obvious matrix interferences can be observed in the SERS spectrum. Presumably, the surface of the tested orange did not break down owing to the swab sampling approach, resulting in an almost neglectable matrix effect for SERS measurements. This experiment demonstrates that it is feasible to conduct the rapid screening of mixed pesticides on orange surfaces using this flexible and low-cost SERS substrate.

### **4. Conclusions**

In summary, a simple and robust method was proposed for the rapid screening of three pesticide residues on orange surfaces using SERS. Flexible SERS substrates were prepared by modifying AgNPs on amino-functionalized PDMS films. The optimized substrates showed high sensitivity, good uniformity, and stability. For pesticide detection, thiram, malathion, and phoxim were measured using this substrate. Linear relationships were established for each analyte, enabling semi-quantitative detection of these pesticides by SERS. Moreover, the simultaneous determination of three pesticides on orange surfaces was achieved by performing a swab sampling process. Compared with other methods, this approach allows rapid and nondestructive detection of mixed pesticides, with low-cost material and an easy to operate procedure. No extra treatment of samples is required before testing, and the substrate can be directly discarded after use. It is feasible to further extend this method for the rapid and on-site detection of other hazardous substances on various fruits.

**Supplementary Materials:** The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/foods11223597/s1, Figure S1: Optical images of AgNPs-PDMS substrates prepared with different concentrations (0.002~0.5%) of APTES. Figure S2: SEM images of AgNPs-PDMS substrates prepared with 0.01% (a) and 2% (b) APTES modified PDMS. Figure S3: SERS spectrum of 100 μg L−<sup>1</sup> R6G on AgNPs-PDMS substrate (red curve); and Raman spectrum of R6G powder acquired under the same condition (black curve). Figure S4: Normalized SERS (red curve) and Raman (black curve) spectra of three tested pesticides: thiram (a), malathion (b), and phoxim (c) [49,50].

**Author Contributions:** Conceptualization, W.Z. and M.W.; methodology, M.C.; validation, W.Z.; investigation, W.Z. and M.C.; writing—original draft preparation, W.Z.; writing—review and editing, Z.X., D.L. and M.W.; supervision, M.W.; project administration, W.Z.; funding acquisition, Z.X. and M.W. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was financially supported by the Beijing Natural Science Foundation (6214036), Beijing Science and Technology Project (Z211100007021005), the Finance Additional Project and the Innovation and Capacity-building Project by Beijing Academy of Agriculture and Forestry Science (CZZJ202102, KJCX20210407), National Natural Science Foundation of China (21976123), and the Science and Technology Commission of Shanghai Municipality (19391901800).

**Data Availability Statement:** Data is contained within the article or Supplementary Materials.

**Acknowledgments:** Zhiyong Yan and Chaojie Yang from the Beijing Center for Physical and Chemical Analysis are acknowledged for performing SEM and TEM characterization of the synthesized nanomaterials.

**Conflicts of Interest:** The authors declare no conflict of interest.
