Aluminum-carbon refractories are widely used in functional refractory products with long-term flow control and protection properties such as continuous casting nozzles, stoppers and slides due to their excellent mechanical properties, thermal shock stability and slag erosion resistance. In recent years, with the continuous improvement of the smelting requirements of special steel grades such as clean steel, low carbon steel and ultra-low carbon steel, it has become a top priority to develop new functional carbon-containing refractory materials with high performance, multi-functions and long life that are adaptable to new steel grades and new continuous casting technologies. In order to improve the service performance of refractory materials, researchers have achieved the reinforcement and toughening of refractory materials by optimizing the design of C chains and generating ceramic phases. Among them, the addition of carbon fiber is considered to be a feasible solution.
Carbon fiber not only has excellent heat resistance and corrosion resistance, but also has fiber flexibility, which allows it to maintain its strength in high-temperature inert environments above 2000°C. Therefore, carbon fiber composites are widely used in various fields, including for reinforcing and toughening ceramic-based and metal-based composites.
For example, carbon fibers were introduced into C/SiC-based ceramic materials, and their dynamic mechanical properties were studied; the introduction of carbon fibers into zirconium diboride-based ceramics significantly improved fracture toughness; the application of carbon fiber reinforced composite woven meshes to concrete beams improved their load-bearing properties to varying degrees; ZrB₂-SiC-Csf nanocomposites were prepared using a pressureless sintering method, and the results showed that the hardness and fracture toughness of the composite material were improved when the carbon fiber content was 2.5%; the mechanical properties of nylon 6 materials reinforced with carbon fibers were significantly improved; the introduction of carbon fibers into aluminum-carbon materials revealed a significant toughening and reinforcing effect; and the introduction of carbon fibers into low-carbon magnesia-carbon bricks significantly improved the material’s oxidation resistance, room temperature and high temperature strength, and thermal shock resistance.
However, research has found that the dispersion uniformity of carbon fibers affects their ability to enhance material performance. Carbon fibers are very easy to break when sheared transversely, and due to their high surface energy, they are very easy to agglomerate. This makes it very difficult to homogenize the chopped carbon fibers and the matrix when preparing composite materials. At present, the dispersion methods for carbon fibers in carbon fiber functional composites include dry mixing, wet mixing, and alcohol pre-dispersion. However, these methods have shortcomings such as easy agglomeration of carbon fibers, poor dispersion effect, complex dispersion process, and high requirements for the length of the carbon fibers.
Based on this, this paper introduces a novel carbon fiber dispersion method during material preparation. By directly introducing carbon fibers into a binder with good affinity to the carbon surface, ultrasonic dispersion is used to simply achieve uniform dispersion of the carbon fibers. This simplifies the pre-dispersion process and effectively separates the chopped fibers into individual filaments. Furthermore, ultrasonic dispersion effectively avoids carbon fiber damage. This paper investigates the effects of different amounts of chopped carbon fibers added on the properties of alumina-carbon refractories, comparing and analyzing the changes in bulk density, apparent porosity, room temperature flexural strength, compressive strength, and thermal shock resistance of the samples after high-temperature treatment.
Experimental part
1.1 Raw material and sample preparation
The experiment used fused brown corundum (30, 70, and 200 mesh particle sizes), flake graphite (C>94%, 80 mesh particle size), chopped carbon fibers, elemental silica powder (Si>98%), and silicon carbide (SiC>98%) as the main raw materials, with thermosetting phenolic resin as the binder. The properties of the chopped carbon fibers used are shown in Table 1, and the SEM (scanning electron microscope) morphology of the carbon fibers is shown in Figure 1. The samples were named CA to CE according to the chopped carbon fiber content from 0% to 0.4%, and the specific raw material composition of the samples is shown in Table 2.
| Type | Fiber diameter/μm | Standard length/mm | Carbon content/% | tensile strength/MPa | Tensile modulus/GPa | density/(g.cm-3) |
| Resin-free type | 7 | 1 | 95 | 3500 | 228 | 1.75 |
| raw material | content | ||||
| CA | CB | CC | CD | CE | |
| Fused Brown Corundum | 79 | 79 | 79 | 79 | 79 |
| Flake graphite | 15 | 15 | 15 | 15 | 15 |
| Elemental silicon powder | 3 | 3 | 3 | 3 | 3 |
| Silicon carbide | 3 | 3 | 3 | 3 | 3 |
| Chopped carbon fiber | 0 | 0.1 | 0.2 | 0.3 | 0.4 |
| Thermosetting phenolic resin | 4 | 4 | 4 | 4 | 4 |
Weigh the dry powder materials of each component according to the experimental proportions in Table 2 and premix them. Weigh the short-cut carbon fibers and thermosetting phenolic resin of each component according to the ingredient list, and mix them evenly by ultrasonic stirring. Mix the premix with the binder at high speed and dry to obtain the experimental mud. Press the experimental mud into 200mm×150mm×50mm samples by isostatic pressing, dry at 200℃ for 24h, and then fire the dried parallel samples at 1100℃ and 1500℃ in a reducing atmosphere, respectively. After cooling, remove them.
1.2 Test and characterization methods
The bulk density and apparent porosity of the samples were determined using the Archimedes method. The room-temperature compressive strength of the samples was determined using a pressure testing machine. The room-temperature and high-temperature flexural strength of the samples were determined using the three-point bending method. The phase composition of the sintered samples was determined using an X-ray diffractometer (DX-2700BH). The microstructure of the samples was observed using a scanning electron microscope (PhenomProX). Thermal shock tests were performed, and the room-temperature flexural strength of the samples after one, two, and three cycles of carbon embedding and water cooling at 1100℃ was measured.
Results and Discussion
2.1 Dispersion effect of carbon fiber after introducing binder
The microscopic morphology of the carbon fiber after being introduced into the resin binder and cured at 1000°C is shown in Figure 2. The chopped carbon fiber is evenly dispersed after being introduced by directly dispersing it in the binder, without agglomeration, and the carbon fiber and the resin are tightly bonded.
2.2 Effects of different carbon fiber contents on material properties, phase composition, and morphology after heat treatment at 1100℃
The flexural strength and compressive strength of the samples after heat treatment at 1100℃ are shown in Figure 3. The flexural strength and compressive strength at room temperature of the samples with added carbon fiber are improved compared with those without carbon fiber. However, as the carbon fiber content is ≥0.1%, the flexural strength and compressive strength at room temperature of the samples show a trend of first increasing and then decreasing.
Figure 4 shows the variation curves of high-temperature flexural strength (1500℃) of samples with different carbon fiber addition amounts. The high-temperature flexural strength of the samples with added carbon fiber is improved compared with the samples without added carbon fiber, but as the carbon fiber content is ≥0.1%, it also shows a trend of first increasing and then decreasing.
Figure 5 shows the room-temperature flexural strength of each sample after thermal shock. It can be seen that the residual flexural strength of the samples after 1 to 3 thermal shock treatments follows the same trend as the strength curve of the original sample. The residual flexural strength of the samples containing carbon fiber after 1 to 3 thermal shock treatments is higher than that of the samples without carbon fiber. The optimal residual flexural strength is achieved when the carbon fiber content is 0.1%.
The bulk density and apparent porosity of each sample after heat treatment at 1100°C are shown in Figure 6. Compared with the sample without carbon fiber addition, the bulk density of the sample with 0.1% carbon fiber addition increased and the apparent porosity decreased; while the bulk density of the sample with ≥0.2% carbon fiber addition decreased and the apparent porosity increased.
As can be seen from the above, at a heat treatment temperature of 1100°C, the flexural and compressive strengths of the material significantly improved when the carbon fiber content was 0.1%, compared to the sample with zero carbon fiber content. However, when the carbon fiber content was 0.2%, the room-temperature flexural and compressive strengths of the material were slightly higher than those of the sample with zero carbon fiber content, but significantly lower than those of the sample with 0.1% carbon fiber content. Among all parallel samples, the thermal shock resistance was also optimal when the carbon fiber content was 0.1%. This indicates that material toughening can be achieved when the carbon fiber content is appropriate. However, when the carbon fiber content is ≥0.2%, the interfacial area between the fiber and the matrix increases, resulting in a decrease in the bulk density and an increase in apparent porosity, which in turn reduces the flexural and compressive strengths. Furthermore, due to the significant internal stress accumulated during the pressing of the carbon fiber and the blank, and the complex internal stress distribution caused by the inconsistent plastic deformation directions, elastic aftereffects such as elastic rebound and hysteretic rebound caused by internal stress relaxation occur after pressure relief and demolding, reducing the bulk density and increasing the porosity. This is consistent with the changing trend of the bulk density and apparent porosity of the samples after heat treatment at 1100℃.
Figure 7 shows the XRD patterns of each sample after heat treatment at 1100℃. At the heat treatment temperature of 1100℃, the phases in each sample are basically the same, with C and Al₂O₃ as the main phases, and small amounts of Si and SiC phases.
Figure 8 shows a scanning electron micrograph of the carbon fibers in the cross section of a specimen with a 0.1% carbon fiber content (heat treated at 1100°C). On the fracture surface, the carbon fibers are observed to be dispersed in stripes. Some are encapsulated in the matrix (Figure 8(a)), others interact with the matrix through bridging mechanisms (Figures 8(a) and 8(b)), and some debond and pull out (Figure 8(b)). When the material fractures, one possible reason is that the matrix crack propagates to the carbon fibers during the stress increase, hindering the crack. Furthermore, crack deflection and torsion along the carbon fiber/matrix interface consume more energy than direct crack propagation, thereby increasing the fracture surface energy during the fracture process and improving strength and toughness. Carbon fiber bridging cracks is also an important factor in material toughening. When the stress is higher than the matrix’s fracture strength but lower than the carbon fiber’s fracture strength, the carbon fibers can prevent the matrix from debonding by bridging the cracks. Furthermore, the debonding, pullout, and frictional energy required to overcome carbon fiber debonding and pull out also consume a certain amount of energy. Under the combined effect of these mechanisms, alumina-carbon refractories with added carbon fibers can achieve improved strength and fracture toughness at a service temperature of 1100℃, thus achieving the purpose of toughening and reinforcement. Figure 8 also shows that with an addition of 0.1%, short carbon fibers introduced into the resin via direct dispersion can be uniformly dispersed in the sintered sample without agglomeration.
2.3 Effects of heat treatment at 1500℃ on the structural properties, phase composition, and morphology of carbon fiber-containing sample materials
Figure 9 shows the variation curves of bulk density and apparent porosity of samples with different carbon fiber additions at 1500℃. It can also be seen that as the carbon fiber addition increases from 0 to 0.4%, the bulk density first increases and then decreases, while the apparent porosity first decreases and then increases.
Figure 10 shows the XRD patterns of each sample after heat treatment at 1500℃. At 1500℃, the phase composition of each sample was basically the same, with C and Al₂O₃ as the main phases and a small amount of SiC. Compared to 1100℃ (Figure 6), the diffraction peaks of Si in each sample disappeared after heat treatment with embedded carbon at 1500℃, and new silicon carbide diffraction peaks appeared at 20=33.4° and 20=38°, corresponding to a-SiC (72-0018). Comparing the XRD patterns of samples with different carbon fiber contents after heat treatment with embedded carbon at 1500℃, it can be seen that the diffraction peaks of the silicon carbide phase generally increase with increasing carbon fiber content. This indicates that increasing the carbon fiber content at 1500℃ is beneficial to the formation of the new silicon carbide phase. The microstructure of the new phase was characterized using SEM.
Figure 11 shows the SEM results of the surface of a single carbon fiber in the cross-section of a sample with a carbon fiber content of 0.1% after heat treatment at 1100℃ [Figure 11(a)] and 1500℃ [Figure 11(b)]. As can be seen from the figure, after heat treatment at 1100℃, the carbon fiber surface is smooth and without alteration, showing no difference in morphology compared to the state of the short-cut carbon fiber before treatment (Figure 1). However, compared to the smooth morphology of the carbon fiber after heat treatment at 1100℃ [Figure 11(a)], the surface and interior of the carbon fiber in the cross-section of the sample treated at 1500℃ have become rough, with well-developed silicon carbide whiskers forming on the surface.
The reaction process for generating silicon carbide whiskers on the carbon fiber surface is as follows: During the sintering process, SiO and CO are generated according to reactions (1) and (2). When the SiO and CO pressures are equal, SiC whiskers are formed according to reaction (3).
2C+O₂→2CO (1)
Si+CO→SiO+C (2)
SiO+3CO→SiC+2CO₂ (3)
Therefore, after heat treatment at 1500℃, the silicon carbide whiskers generated in the carbon fiber-containing sample act as a bridge, connecting the matrix and carbon fibers to form a weak interfacial bond, inducing crack propagation, and enhancing the mechanical properties and thermal shock resistance of the material. Thanks to the dual reinforcing effect of carbon fibers and silicon carbide whiskers, the service performance of continuous casting aluminum-carbon refractories in high-temperature environments can be effectively improved.
in conclusion
(1) The short carbon fibers added directly by the binder were uniformly dispersed without agglomeration. This indicates that the direct introduction of the binder into the carbon fibers for ultrasonic dispersion has a good effect, effectively achieving the separation of short-cut fiber monofilaments while avoiding carbon fiber damage.
(2) Under heat treatment conditions of 1100℃ and 1500℃, the room temperature flexural strength, compressive strength, residual flexural strength and high temperature flexural strength of the carbon fiber added samples were effectively improved compared with those without carbon fiber addition. Among them, the performance was the best when the carbon fiber addition content was 0.1%. At the same time, as the carbon fiber addition amount increased from 0 to 0.4%, the volume density first increased and then decreased, while the apparent porosity first decreased and then increased. This shows that the appropriate amount of carbon fiber addition will optimize the volume density and apparent porosity of the material while playing a toughening role, improve the flexural strength, compressive strength and thermal shock resistance, and thus achieve toughening and reinforcement of aluminum-carbon refractory materials in medium and high temperature use environments.
(3) After heat treatment at 1500℃, the carbon fiber samples, which were originally smooth at 1100℃, underwent alteration on both the surface and interior. The silicon content decreased, and well-developed silicon carbide whiskers formed on the surface. These whiskers bridged the matrix and carbon fibers, enhancing the material’s mechanical properties and thermal shock resistance. Thanks to the dual reinforcing effect of carbon fibers and silicon carbide whiskers, the service performance of continuous casting aluminum-carbon refractories under 1500℃ high-temperature immersion conditions can be effectively improved.
