*3.2. Surface Characterization*

To understand how topological factors affect cell adhesion and spreading, the surface morphology of PEEK and CFR-PEEK composite was determined using SEM. Figure 3 presents the SEM images of untreated, polished, and sandblasted PEEK and CFR-PEEK samples. Printing borders, as shown in Figure 3a,d were formed on the surface due to the deposition between two printing lines. The clear peaks and valleys, which completely disappeared after polishing and sandblasting, could be identified on both untreated PEEK and CFR-PEEK sample surfaces. The polished surfaces displayed the smoothest morphology, although a few defects remained on the polished CFR-PEEK surfaces (Figure 3b,e). The surfaces of specimens, however, after sandblasting treatment possessed surface topography features in the micrometer scale with a homogeneous distribution of protuberances and cavities (Figure 3c,f).

**Figure 3.** SEM images of PEEK and CFR-PEEK composite: (**a**) untreated PEEK; (**b**) polished PEEK; (**c**) sandblasted PEEK; (**d**) untreated CFR-PEEK; (**e**) polished CFR-PEEK; (**f**) sandblasted CFR-PEEK. Bars represent 200 μm and 20 μm (inserts), respectively.

Figure 4 illustrates the roughness of specimens of different groups. It is obvious that the untreated specimens displayed the roughest surfaces, both for PEEK and CFR-PEEK materials with the Sa value of 17.67 ± 5.7 μm and 32.36 ± 17.02 μm, which were significantly higher than the values of the polished and sandblasted groups (*p* < 0.05). The Sa values of sandblasted PEEK (0.85 ± 0.14 μm) and CFR-PEEK (0.97 ± 0.26 μm) samples were slightly higher than those of the polished surfaces (0.42 ± 0.26 μm and 0.67 ± 0.42 μm). In contrast, the polished Ti samples showed a very smooth surface (0.2 ± 0.04 μm), which was more homogenous compared with that of the polished PEEK and CFR-PEEK samples. The same trend could also be seen in the Sq data.

**Figure 4.** Reconstructed three-dimensional (3D) surface topographies of analyzed samples, and Sa and Sq values: (**a**) untreated PEEK; (**b**) polished PEEK; (**c**) sandblasted PEEK; (**d**) untreated CFR-PEEK; (**e**) polished CFR-PEEK; (**f**) sandblasted CFR-PEEK; (**g**) polished Ti; (**h**,**i**) Sa and Sq values of as-printed, polished, and sandblasted PEEK and CFR-PEEK samples, the polished Ti was used as an additional reference. The data are presented as means ± standard deviation, \* *p* < 0.05.

The result of contact angle analysis is shown in Figure 5. Data revealed that the untreated surfaces of pure PEEK reflected an obvious hydrophobic response to water with a mean contact angle of 105 ± 26◦. The polished PEEK specimens exhibited a hydrophilic behavior (78 ± 3◦). After sandblasting, the contact angle rose slightly (88 ± 7◦), but the difference was not significant (*p* > 0.05). As for the CFR-PEEK samples, the untreated group also indicated the most hydrophobic sample surface (92 ± 12◦) compared with polished (82 ± 5◦) and sandblasted (75 ± 3◦) specimens. Both PEEK and CFR-PEEK samples, whether with or without surface modifications, revealed a more hydrophobic response to water compared to Ti (51 ± 5◦).

**Figure 5.** Water contact angle measured on untreated, polished, and sandblasted PEEK and CFR-PEEK samples: (**a**) untreated PEEK; (**b**) polished PEEK; (**c**) sandblasted PEEK; (**d**) untreated CFR-PEEK; (**e**) polished CFR-PEEK; (**f**) sandblasted CFR-PEEK. (**g**) Ti (additional reference); (**h**) quantitative contact angle values (means ± standard deviation). The dotted line indicates the contact angle of 90◦, which is the division of hydrophilicity and hydrophobicity, \* *p* < 0.05.
