The surface-enhanced Raman scattering (SERS) technique is a promising method for the detection of explosives such as 2,4,6trinitrotoluene (TNT) and 3-nitro-1,2,4-triazol-5-one (NTO) because of its high sensitivity to trace substances. However, most SERS detection processes are often nonautomated as well as exhibit low efficiency and toxic exposure, which often poses potential danger to operators. Herein, we propose the integration of SERS with digital microfluidics (SERS-DMF) for automated, high-throughput, and high-sensitivity detection of explosives. First, we carefully designed a DMF chip comprising 40 drive electrodes and 8 storage electrodes to achieve a high-throughput process. And different concentrations of target molecules, silver nanoparticles (Ag NPs), and salts were loaded into the DMF chip. Then, the droplet aggregation, incubation, and detection processes were automatically controlled using the SERS-DMF platform. In addition, Ag NPs were efficiently aggregated by screening different types and concentrations of salts, resulting in "hotspots" and the SERS effect. With the help of the SERS-DMF platform, two explosive samples were automatically detected with high throughput and high sensitivity. The detection limits of TNT and NTO were 10 −7 and 10 −8 M, respectively. In addition, compared with nonautomatic operations, the SERS-DMF platform exhibited better reproducibility and higher efficiency for the detection of explosives. The proposed SERS-DMF thus has considerable potential as an analytical technique for detecting hazardous substances.

In this work, fumed silica was compounded with silicone rubber and processed under two different extraction conditions to quantitatively analyze bound rubber for studying filler−rubber interaction. When nanocomposites were treated with toluene at 90 °C under ultrasonic (US) irradiation, physisorbed polymer chains were more substantially removed when compared to the traditional room temperature (RT) extraction method. The bound rubber fraction of silicone rubber filled with 40 phr of silica was 32.85% by RT treatment, while the US-treated one was reduced to 6.20%. A thin layer of tightly bound rubber adsorbed on the silica surface could be obtained by US treatment as observed by scanning electron microscopy and transmission electron microscopy. Low-field 1 H NMR results confirmed that the obtained tightly bound rubber chains were strongly constrained. The attenuated total reflectance-infrared spectra showed that the formation of tightly bound rubber caused redshift of the Si−O−Si characteristic peaks of silica, and the obtained tightly bound rubber has lower total surface energy than the rubber matrix, indicating the formation of a more stable structure. The reagglomeration potential energy of silica during the curing process was significantly reduced due to the presence of the bound rubber layer. In addition, the amount of tightly bound rubber is mainly related to the surface physical and chemical properties of silica and is less affected by the concentration. Fumed silica tends to form nanofiller networks in rubber that are woven of multiple fillers entangled with bound rubber chains to reinforce the rubber. Therefore, fillers with higher bound rubber content and a better-dispersed aggregate structure in the rubber could more significantly enhance the physical properties of nanocomposites.