Industrial solid waste refers to the solid waste that is produced in industrial production activities. Without correct treatment and let-off, industrial solid waste may cause environmental pollution due to a variety of pollutants and toxic substances that are contained in it. Conventional detection methods for identifying harmful substances are high performance liquid chromatography (HPLC) and gas chromatography-mass spectrometry (GC-MS), which are complicated, time-consuming, and highly demanding for the testing environment. Here, we propose a method for the quantitative analysis of harmful components in industrial solid waste by using terahertz (THz) spectroscopy combined with chemometrics. Pyrazinamide, benazepril, cefprozil, and bisphenol A are four usual hazardous components in industrial solid waste. By comparing with the Raman method, the THz method shows a much higher accuracy for their concentration analysis (90.3–99.8% vs. 11.7–86.9%). In addition, the quantitative analysis of mixtures was conducted, and the resulting prediction accuracy rate was above 95%. This work has high application value for the rapid, accurate, and low-cost detection of industrial solid waste.

Adding metal into nonpolar materials is an effective way to improve its properties. The concentration of metal particles inside materials cannot be effectively measured through conventional methods, such as scanning electron microscope (SEM) or optical microscope (OM). In this paper, based on terahertz time domain spectroscopy (THz-TDS) system, we propose a “Terahertz Linear Scatter“ method to realize the quantitative analysis of metal particles concentration in nonpolar materials. Taking the composite graphene-nanosilver film as an example, terahertz (THz) wave can penetrate graphene with high transmittance, while it will be effectively scattered by nanosilver particles. Importantly, the small volume of the metal particles make the attenuation of received signals change in a linear fashion. Based on this principle, the concentration of metal particles in nonpolar materials can be identified quickly and accurately. The corresponding Root-Mean-Square Error can be as low as 0.05%. These results are of great significance for the analysis of various materials containing metal particles.