**1. Introduction**

Carbon steel always involves a slab reheating and hot rolling process to obtain thin carbon steel. During the reheating process, slabs suffer from serious oxidation over 1250 ◦C, which results in weight loss of slabs and energy waste [1–3]. Moreover, oxidation could potentially lead to serious problems for the surface quality of carbon steel, such as decarburization and micro-defects [4,5]. Different methods have been taken to protect slabs from oxidization during the reheating process. The most effective countermeasure for the reduction of the oxidation intensity is to isolate slabs from oxidants. Therefore, reheating in reducing atmosphere, such as vacuum or inert gas atmosphere, is once thought to be an ideal solution. However, the rigid operation conditions and expensive equipment limit its industrial application [6–8].

Protective coating, an effective and economical method, is considered as an alternative solution to prevent the slabs from oxidization at high temperature [9,10]. Ceramic coating (MgO, Al2O3, SiO2, CoO, ZnO, ZrO2 etc.) usually has a high melting point and excellent chemical stability at high

temperature [11,12]. By using ceramic coating, a protective layer can be formed on carbon steel surface at temperatures ranging from 1100 to 1300 ◦C so that the internal diffusion of oxygen and external diffusion of iron ions are slowed down [11,13–16].

Among the ceramic coatings, Al2O3-SiO2 coating (AS) is widely used due to its high anti-oxidation performance and low cost [17]. Silicates with low melting points are always added into Al2O3-SiO2 coating to accelerate the sintering under low temperature. However, a eutectic crystal of Fe–2FeO·SiO2 is formed over 1170 ◦C, which is harmful to the anti-oxidation performance of the protective coating at high temperature [18,19]. In order to accelerate the sintering of Al2O3-SiO2 coating under low temperature and avoid the formation of Fe–2FeO·SiO2 over 1170 ◦C, silicates with low melting point should be replaced by other additives [20,21]. Aluminite powder is an alternative additive because it can melt at 660 ◦C and react with SiO2 [22]. The exothermic reaction between Al and SiO2 leads to local high temperature and thus the sintering of ceramic materials is accelerated. Sodium silicate and colloidal silica are widely used as binder agents in ceramic coating preparation. Functional colloidal sol would form under the hydrolysis of sodium silicate or metasilicic acid and be balanced with the free OH<sup>−</sup>. If Al powders are directly added into ceramic coating slurry containing alkaline binder such as sodium silicate and colloidal silica, a exothermic reaction (2Al + 2H2O + 2OH− → 2AlO2 − + 3H2) will take place and break the balance of hydrolysis within the sodium silicate solution (SiO3 2− + 2H2O → H2SiO3 + 2OH−). It was the consumption of OH− that would accelerate the hydrolysis and enhance the viscosity of the slurry. Viscous slurry is hard to disperse by spraying due to its poor fluidity. In addition, the hydrogen produced by the reaction is flammable and it is hazardous in the reheating workshop. Therefore, Al powders should be modified to be stable with the coating slurry.

In this paper, silicon-modified Al powders (SiO2@Al) were prepared by tetraethyl orthosilicate (TEOS) hydrolysis under alkaline conditions. A new Al2O3-SiO2 ceramic coating (ASMA) was prepared with SiO2@Al as additives to prevent carbon steel from oxidization. The anti-oxidation performance of ASMA was investigated by a heating process from room temperature to 1250 ◦C with a rate of 10 ◦C/min and maintained for 120 min at 1250 ◦C. A protective mechanism of ASMA was also clarified. Some achievements in this paper were not only applied to anti-oxidation of carbon steel at high temperature but also provided new ideas for the design of low-temperature sintering ceramic.

## **2. Experiment Procedure**

#### *2.1. Preparation of Carbon Steel Sample*

Carbon steel, J55 (C 0.28 wt %, Si 0.27 wt %, Mn 1.35 wt % and Fe balance), with a specimen size of 55 × 50 × 5 mm<sup>3</sup> was used for high temperature treatment in muffle furnace. A carbon steel specimen (25 × 24 × 5 mm3) was cut from cold-rolled plates for continuous thermo-balance investigation. The specimens were cleaned with alcohol in an ultrasonic bath and dried in an oven.

#### *2.2. Preparation of Modified Al Powders (SiO2@Al)*

Al powders with the size of 45 μm, tetraethyl orthosilicate (TEOS), ethanol, ammonia water, and deionized water were used for the preparation of SiO2@Al. A mixture of 10 g Al powders and 250 mL ethanol was stirred in a beaker at 50 ◦C for 1 h. The mixture was blended with 15 mL TEOS under agitation, which was diluted with 150 mL alcohol. After that, 20 mL ammonia water diluted with 25 mL deionized water was dropped into the mixture by a peristaltic pump at 2 mL/min as the catalyst for TEOS hydrolysis. Then, the beaker was kept in a water bath at 50 ◦C for 8 h. The obtained SiO2@Al was separated from the resulting mixture by vacuum filtration, followed by washing several times with ethanol. Then SiO2@Al was dried at 60 ◦C under vacuum.

The synthesis of SiO2@Al through TEOS hydrolysis under alkaline condition is illustrated in Figure 1. The SEM-EDS results in Figure 2 indicated that Al powders were successfully encapsulated

by SiO2 and thus SiO2@Al was obtained. The size of SiO2 encapsulating on the surface of Al powders was about 100 nm.

**Figure 1.** The schematic illustration of SiO2@Al synthesis.

**Figure 2.** SEM and EDS images of SiO2@Al.

#### *2.3. Preparation of Coating Slurry*

As shown in Table 1, ceramic coating slurry with different amount of SiO2@Al (0, 5%, 10%, 15%, 20%, 25%) were prepared to determine an appropriate proportion of SiO2@Al additives. The mixture was ball-milled with water for 4 h until the particle size was <45 μm. The milling ball was ZrO2 and the diameter of ZrO2 balls was 6–10 mm. During the ball-milled process, SiO2@Al stably existed in the slurry without any hydrogen releasing. The prepared slurry was coated on carbon steel by spraying gun, which was connected to an air compressor with pressure of 8 bar. The thickness of the coating was 0.4 mm.



#### *2.4. Evaluation of Anti-oxidation Performance*

To evaluate the anti-oxidation performance of ceramic coating, the weight changes of heated carbon steel samples were investigated. The carbon steel samples were heated in a muffle furnace from room temperature to different temperatures (1050, 1100, 1150, 1200, and 1250 ◦C) and maintained for 120 min. The anti-oxidation ability *E*, which was related to the weight loss of carbon steel oxidation, was calculated by Equations (1) and (2). Two samples were used in each test and the average value of samples was adopted.

$$\text{Steel yield } \alpha \, = \frac{m\_2}{m\_1} \times 100\% \,\tag{1}$$

$$\text{Anti}-\text{oxidation ability }E = \frac{\alpha\_{\text{cated}} - \alpha\_{\text{bare}}}{1 - \alpha\_{\text{bare}}} \times 100\% \tag{2}$$

where, *m*1 and *m*2 are the weights of samples before and after high temperature treatment without scale. αcoated and αbare are the yields of the coated and bare samples after high temperature treatment.

To evaluate the non-isothermal kinetic of carbon steel oxidation, the correlation between oxidation reaction rate and heating temperature was investigated. The samples were heated at a rate of 10 ◦C/min to certain temperature (1050, 1100, 1150, 1200, and 1250 ◦C) and maintained for 5 min. The reaction rate (*v*) was calculated by Equations (3) and (4).

$$\text{Weight loss per unit area } \Delta m = M\_2 - M\_1 \tag{3}$$

$$\text{Reaction rate } v = \frac{\Delta m}{\Delta t} \tag{4}$$

where, Δ*m* is the weight loss per unit area of the sample during the holding stage at certain temperature. Δ*t* is the time that the samples were maintained at certain temperature.

To evaluate the anti-oxidation ability of ceramic coating at a certain temperature, the isothermal kinetic was also carried out. The isothermal kinetic was conducted by a continuous thermos-balance (RZ, Luoyang Precondar, Luoyang, China) at a heating rate of 10 ◦C/min to 1250 ◦C and maintained for 120 min. The weight change of the sample was calculated by Equation (5).

$$
\Delta \boldsymbol{\omega} = \boldsymbol{\omega}\_{\bar{i}} - \boldsymbol{\omega}\_{0} \tag{5}
$$

Here, ω0 is the weight change per unit area of the sample heated to certain temperature and ω*i* is that of the sample heated and maintained at certain temperature.
