QCM-D experiments were carried out to assess the effective- ness of the membrane surface modification procedure. A clean gold sensor was spin-coated with PSf, dried in a desiccator, and then tested in pure DI water to adjust the initial reading to 0 ng/cm2. After the reading had stabilized for 15 min, the DI water was replaced by TiO2 solution to deposit the first TiO2 layer on the PSf surface of the sensor. As evidenced in Fig. 3, the mass on the sensor surface increased very fast, indicating that TiO2 nanoparti- cles were quickly adsorbed on the PSf surface. After about 2 h, the mass change slowed down significantly and stabilized at about 34 mg/cm2, indicating that the adsorption of TiO2 on the PSf reached its capacity. Since the QCM-D measurements (i.e., sensor frequency and dissipation) are affected by solution properties, the absolute mass change can only be calculated by comparing the measurements obtained in the same type of solution. Therefore, after the deposition of TiO2 on PSf, the TiO2 solution was replaced by DI water so that the absolute mass change due to TiO2 deposition could be quantified. Note that the use of DI water also helped remove any loosely packed nanoparticles from the sensor surface. Fig. 3 shows that the first round of switching from TiO2 solution to DI water caused a drastic increase in mass but the mass quickly stabilized again, indicating such a mass change was a direct result of solution difference. Comparison of the measure- ments in DI water before and after TiO2 deposition revealed a total TiO2 deposition of "' 42 mg/cm2 on PSf, thereby confirming the successful assembly of the first TiO2 layer.