*2.2. Antibacterial Performance*

The plate-counting bacteria colonies of *E. coil* were used to evaluate the bacteriostatic ability of UCNPs@R-TiO2 composites under 980 nm light irradiation (1 W). Under the different preparation conditions (Table S1), the highest antibacterial efficiency (98.1%) was achieved for the UCNPs@R-TiO2 (180 ◦C, 20 h) composites. The low reaction temperature (180 ◦C) was unfavorable to the crystal growth of R-TiO2, and the high reaction temperature (300 ◦C) caused the agglomeration of R-TiO2, which was not good for photocatalytic reactions [11,31]. Simultaneously, the high crystallinity of R-TiO2 was achieved at the optimal reaction time (20 h) [11]. The highest antibacterial efficiency (98.7%) was also achieved on the UCNPs@R-TiO2 (40%) composites among the different mass ratios of R-TiO2 and UC-NPs (Figure S5). Consequently, the optimal UCNPs@R-TiO2 (30%, 180 ◦C, 20 h) composites

were used for further research. As shown in Figure 4A, the bacterial photoinactivation effect was suitably correlated with the dosage of nanomaterials and the UCNPs@R-TiO2 nanocomposite, resulting in the highest bactericidal effect (97.3%) at the concentration of 50 μg/mL. Interestingly, we further found that both the UCNPs (Figure 4(Ba)) and R-TiO2 (Figure 4(Bb)) were capable of killing *E. coil* colonies on the agar plate under 980 nm laser irradiation for 12 min when compared to the saline control (Figure 4B). Strikingly, the UCNPs@R-TiO2 nanocomposite (Figure 4(Bc) and Figure S6) treatment eliminated about 98.1% of *E. coil* colonies on the plate, reflecting its enhanced bactericidal activity. Simultaneously, the bactericidal performance of the optimized UCNPs@R-TiO2 nanocomposite was compared to the previously reported works (Table S2). We also found that the UCNPs@R-TiO2 nanocomposite had the best bactericidal effect among these antibacterial agents.

**Figure 3.** (**A**) X-ray powder diffraction (XRD) patterns of UCNPs@R-TiO2 and the standard hexagonal phase (JCPDS 00-016-0334) and anatase phase (JCPDS 01-021-1272). (**B**) The upconversion luminescence (UCL) spectrum of the UCNPs (a) and UCNPs@R-TiO2 (b). (**C**) The UV-Vis absorption spectrum of TiO2 (a) and R-TiO2 (b), with insert showing the corresponding bandgap determined by Tauc plot. (**D**) The zeta potential of UCNPs (a), R-TiO2 (b), and UCNPs@R-TiO2 (c).

The sterilization of these three materials under the same condition was also investigated via matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) analysis. As shown in Figure 5a, there were distinct characteristic peaks (*m*/*z* = 4331, 5060, 6220, 7250, 9190, 9519) of *E. coli* K12 [32–34] present in our analysis, indicating the massive survival of *E. coli* K12. However, the number of peaks was dramatically decreased upon adding either the UCNP nanorods (Figure 5b) or R-TiO2 nanoparticles (Figure 5c), indicating that some of the bacteria were killed. Importantly, there were no characteristic peaks detected between the 4000 to 14,000 Da region after the treatment of the UCNPs@R-TiO2 composites, indicating that the *E. coli* K12 was nearly entirely killed.

All these data demonstrate that the UCNPs@R-TiO2 nanocomposites possessed a highly effective bactericidal ability under the 980 nm NIR illumination.

**Figure 4.** (**A**) *E. coli* viability under all different sample concentrations and (**B**) photographs of agar plates of *E. coli* incubated with 40 μg/mL of UCNPs (**a**), R-TiO2 (**b**), and UCNPs@R-TiO2 nanocomposite (**c**) using a 980 nm laser (1 W, 12 min).

**Figure 5.** MALDI-TOF MS analysis of *E. coli* without (**a**) and with R-TiO2 (**b**), UCNPs (**c**), and UCNPs@R-TiO2 composites (**d**) under 980 nm NIR light irradiation for 20 min (40 μg/mL). The red band is the characteristic peak of *E. coli* at *m*/*z* = 7250.
