device program. Pictures ofthe porous carbon sampling boat and vaporizer are shown in Fig. S3 and Fig. S4, respectiveal ly, which were made by SANYE Carbon Materials Factory. For the single gas line model in the previous study 130l, when y a sampling boat was switched into the ETV unit, the gas circuit was c sealed, however, when switching out the ETV unit, the gas circuit was opened to ambient conditions. Thus, the gas pressure appeared to fluctuate sharply, extinguishing the ICP torch, which is sensitive to gas g flow changes. To avoid extinguishing the plasma, a "double gas circuit" design, including the carrier gas and supplemental gas lines, was employed. The schematic diagram of the gas circuit design is shown in Fig. 1. The carrier gas and supplemental gas lines were controlled by separate gas mass flow controllers (GMFC). The carrier gas line connected the split-flow "T" joint, GMFC-1, solid sampling device and confluence "T" joint to carry the analyte to the detector; the supplemental gas line connected the gas supplier, GMFC-2 and confluence T joint to keep the plasma stable. The gas line, "T" joint and interface for the ICP-MS were all made from polytetrafluoroethylene (PTFE) materials.