*3.3. Adhesion and Thermal Shock Properties*

Cross-cut tape test (ASTM D3359) [21] was first chosen for qualitative estimation, and pull-off test (ASTM D4541) [22] was then carried out to quantify the adhesion of the coatings. In cross-cut tape test, a grid of square scratches was made on the coating surface. Optical microscopy was used to evaluate about the adhesion of the coating. As shown in Figure 6a, the edges of the scratches are completely smooth and none of them is detached. According to standard ASTM D3359, the highest adhesion scale (5B) is referred to the coating. In pull-off test, the adhesive glue used in this work is Araldite 2021 with flexural strength more than 26 MPa. The glued specimens were cured at room temperature for 30 min. The specimens were mounted on testing machine and were pulled in a direction normal to the coating surface at a constant speed 0.013 mm/s until failure occurs. When failure occurs, the maximum load was recorded. As a result, the interface failure occurred in the 27–29 MPa range. Figure 6b shows cross-sectional SEM of coating after the pull-off test. A delamination appeared near the coating–steel interface, i.e., failure surface occurred within the coating. This is referred as a cohesive failure of the coating. Good adhesion of the coating is ascribed to an interdiffusion at the glass-steel interface due to the mobility of some metal ions in the glass and the metal atoms of the steel at high temperature. Metal ions from the glass will diffuse into the metal while metal atoms will diffuse into the glass.

**Figure 6.** (**a**) Optical microscope image after cross-cut tape test; (**b**) Cross-sectional SEM image after pull-off test.

The thermal shock resistant test was examined in water quenching. The coated specimen was heated to 800 ◦C for 5 min and then cooled rapidly in water (20 ◦C) for 5 min. After ten uninterrupted thermal cycles no macroscopic cracks or failure are visible. The good thermal shock resistance of the coating is an evidence to suggest that the CTE composite coating is close to that of the steel [18]. Some small spherical pores distributed in the coating are the reason for relaxing thermal stresses and improve the thermal shock resistance of the coated samples.

#### *3.4. Anti-Fouling*

Figure 7a,b present the cross-sections of the non-coated steel plate without and with ash particulate deposited, respectively. As depicted in Figure 7a, bare C-steel experienced breakaway oxidation behavior at 800 ◦C with the top surface of oxide layer was very intact. On the contrary, in case of exposure with fly ashes (Figure 7b), the top surface of oxide layer was seriously damaged with many hollow blisters. The fly ash adhered strongly to the steel, which rapidly increased the fouling rate. A top view of the surfaces (Figure 8) showed that much ash was attached all over the surface of the non-coated steel (Figure 8a). This suggests that the following chemical reaction takes place between the fouling particulates and steel at high temperatures [28]:

$$\text{<3Na}\_2\text{O} + \text{Fe} \rightarrow \text{Na}\_4\text{FeO}\_3 + 2\text{Na}\uparrow\tag{2}$$

In contrast, almost no fouling ash adhered on the whole surface of the coated steel (Figure 8b); only some fine fly ash particles stuck in tiny holes were observed. Figure 9 shows that the outer area of the coating was empty of fly ash, while inner area still adhered well to the steel. This demonstrates that the developed ceramic coating provides excellent resistance against the fouling precipitation of fly ash. It could also withstand the high-pressure air jet, which means that a strong bond formed between the coating and steel.

**Figure 7.** EDS maps for the cross-sectional surface of (**a**) the non-coated steel without fly ash deposition and (**b**) with fly ash deposition at 800 ◦C.

**Figure 8.** Top-view SEM images of (**a**) non-coated steel and (**b**) coated steel after the anti-fouling testing.

**Figure 9.** EDS maps of the cross-sectional surface of the coated steel after the anti-fouling testing.

#### *3.5. Thermal Conductivity*

Figure 10a describes the thermal conductivity of the coated and non-coated samples as a function of the temperature up to 800 ◦C. The thermal conductivity of the non-coated steel was ~5 W/mK less than that of the coated steel at 800 ◦C. Figure 10b indicates the thermal conductivity of the ceramic coating itself and the ash deposited specimen. At 800 ◦C, the thermal conductivity of the composite coating was 1.2 W/mK which is about twice as high as that of the corrosion specimen (0.6 W/mK). Those values can be explained as follows. The physical structure and microstructure of a surface is believed to affect its thermal conductivity [29,30]. As shown in Figure 7b, the surface of the non-coated steel with highly porous deposits of loose particulate matter had low thermal conductivity. The deposits that formed on the steel surface limited the absorption of the incident radiation and the transfer of this energy to tubes containing the working fluid. Meanwhile, Figure 9 shows that the surface of the coated steel had a dense interconnected structure without fouling deposits, so the thermal conductivity was high. Therefore, thanks to the ceramic coating, the energy efficiency of the boiler tubes was significantly improved.

**Figure 10.** Thermal conductivities of all prepared samples as a function of the temperature up to 800 ◦C: (**a**) Specimens with and without ceramic coating; (**b**) Ceramic coating materials and corrosion specimen.
