2017 Volume 57 Issue 4 Pages 730-736
An inorganic coating on 50CrVA spring steel was successfully fabricated by a simple and low-cost spraying technique, using bauxite, silicon carbide, sodium silicate, feldspar and phosphate binders. Experimental results showed that the weight gain of coated specimen was reduced by 67.82% and the decarburization layer of coated specimen disappeared completely at 1050°C holding for 120 min. The coating showed a reduction in decarburization of 100% at 1050°C holding for 120 min. It suggests that the coating prevented the decarburization during high-temperature treatment.
50CrVA spring steel is a kind of advanced spring steel and often used as materials of weight valves, safety valve and piston, etc. 50CrVA spring steel has high fatigue strength, high yield-stress ratio, and high harden ability. The heating temperature of 50CrVA spring steel in furnace is below 1100°C. Decarburization and oxidation of spring steels often were observed along with the heat treatment procedures, especially in O2, CO2, H2O and H2 atmosphere. Decarburization is loss of carbon elements from the surface of substrates, and resulted in changes of mechanical properties, e.g., along with the changes of fatigue resistance and hardness.1) Some traditional methods had been used to reduce the decarburization and oxidation during high temperature treatment of high carbon steel, e.g., drawing and peeling methods as physical ones, vacuum heating and controlling the atmosphere, etc.2) Although these mentioned above methods were efficient in preventing the oxidation and decarburization to some extent, they were restricted in their application because of complex process, high equipment cost, and the rigid operation conditions.
Compared with the aforementioned methods, coating method could be suitable for annealing and reheating process at high temperature. Many works devoted to the protective coatings of steel at high temperature.3,4,5,6) Low carbon steel, high carbon steel, stainless steel, multi-alloyed steel, and so on was protected by metallic coatings.7,8,9) The most used methods of applying metallic coatings to steels are thermal spray, hot-dip galvanizing, chemical vapour deposition, electroplating, cladding, and electro-less plating.10) EB-PVD, plasma electrolytic oxidation, Sol–gel, electro-spark deposition and flame or normal spray methods were used to prepare nonmetallic coatings of steel.11,12) The coatings provided excellent resistance to high temperature oxidation, abrasion and corrosion.13,14,15,16) Decarburization and oxidation of 50CrVA steel at high temperature is a serious problem. However, very little study has been performed about coating protection for high carbon steel especially to its oxidation and decarburization protection simultaneously during heating process.
In laboratory, we prepared many inorganic matrix protective coatings for alloy a slurry-spraying technique17,18,19) and investigated the effects on oxidation behavior at high temperature in air. Slurry-spraying technique is applicable for industry process because of less demanding on the surface of substrates. Decarburization and oxidation of 50CrVA spring steel haven’t been solved effectively because of unique heating process. In this paper, we prepared a high temperature corrosion resistant coating for 50CrVA spring steel by a simple and low-cost spraying technique and discussed the process of oxidation and decarburization. The resistance effects of decarburization and oxidation were analyzed. The microstructure and sintering behaviors of the coatings were investigated. The mechanism of corrosion resistant also is discussed.
The 50CrVA steel was cut into cuboids of 10 × 10 × 10 mm from cold-rolled plates. The surfaces were polished on silicon carbide paper of 800 meshes, and then washed with ethyl alcohol in an ultrasonic bath, and dried. The composition of 50CrVA spring steel is shown in Table 1.
The coating material was selected based on the linear coefficient of thermal expansion (CTE), softening point and reactivity with 50CrVA spring steel by trial and error. The raw materials mainly included bauxite, silicon carbide, sodium silicate, feldspar and phosphate binders. The compositional ranges of the inorganic coating used in this study are showed in Table 2. The raw material was mixed with pure water of double amount by weight. The powders without water were mixed with aluminum dihydrogen phosphate binder and sodium silicate dispersant with weight ratio of 10:1:5. After that, mixtures of the prepared raw material were mixed until well-distributed, to make sprayable and thick coating slurry.
The as-made slurry was sprayed on the steel surfaces at room temperature. The thickness of the coating was about 200–500 μm. And then, all the coated samples were dried naturally. To evaluate the oxidation and decarburization resistance of coating, two types of 50CrVA steel specimens, coated and bare specimens, were heated in muffle furnace. Then, the specimens were cooled in air until reaching to room temperature. The specimens were heated in muffle furnace for 120 minutes at 1050°C under atmospheric pressure. The heating rate was 10°C/min up to a final temperature of 1050°C.
The composition of steel was examined by energy dispersive spectrometer (EDS, JEOL, JSM-6700F, Japan). The composition of coating was examined by X-ray fluorescence (XRF, AXIOS-MAX, Nertherland). The particle size of powder was analyzed by a laser particle size analyzer (Beckman coulter LS 13320, USA). The weight changes of oxidation with coated and bare specimen were measured by a thermo balance (RZ, Precondar, Luoyang, China) which was equipped with a sensitivity of ±1 mg and a continuous weighing capacity of 500 g. Every specimen was cut into two parts by line cutting machine and each one was polished on SiC papers. The specimens were washed by 5 vol% admixture of ethyl alcohol and nitric acid. Finally, they were cleaned by ethyl alcohol and prepared for metallographic observation. The structures and morphologies of coating and cross section morphologies of the specimens after heating were observed by scanning electron microscopy (SEM, JEOL, JSM-6700F, Japan). Phase changes of specimen were collected by X-ray diffractometry (XRD, Philips, X’Pert Pro, Netherlands). Metallographies of the decarburization were observed by an optical microscope (Olympus, polarization microscope, BX51M, Japan). Energy changes with the temperature changing were investigated by a DTA thermal analysis system (TG-DTA, Scientific Instrument Factory, Beijing, China). Thermal softening behavior of coating was examined using hot stage microscopy (SJY, Xiangyi Instruments, Xiangtan, China).
The weight gains of oxidation with coated and bare specimen were measured by a thermo balance furnace in air atmosphere. The heating rate was 10°C/min up to the final temperature of 1050°C. And then holding time at 1050°C is 2 h. The bare sample was severely oxidized by the atmosphere and the surface was large uneven with loose irregular channels. On the contrary, the coated sample was covered by a well-proportioned ceramic coating. Weight gains versus heating times for the bare and coated 50CrVA steel after isothermal period at 1050°C for 2 h are shown in Fig. 1. At the beginning, the sample weight decreased because of dehydration. After heating 90 min, above 900°C, both coated and bare samples weight increased obviously.
Oxidation weight changes of coated and bare sample.
After heating for 103 min, the temperature reached to 1050°C. The difference of weight change was more obvious. The weight gain of the bare specimens after 103 min was 8 mg/cm2. As for the coated specimen, its weight gain was 3 mg/cm2, a decrease of 62.5% when compared to the bare one. The weight gain of the bare specimens after holding 120 min at 1050°C was 115 mg/cm2. As for the coated specimen, its weight gain was 37 mg/cm2, a decrease of 67.82% comparing to the bare one. It indicated that the 50CrVA spring steel being oxidation was markedly prevented during thermal exposure by coating. In addition, the anti-oxidation effect was near to constant with the temperature rise. During oxidation tests performed at 1050°C in a muffle furnace, it was found that the oxide scales on bare specimen were free of cracks. However, the oxide scales of coated specimen were completely loose and very easy to remove.
Figure 2 shows the surface microstructures of the coated and uncoated specimens respectively after heat treatments at 1050°C for 120 min. For the bare specimens, the depth of the ferrite decarburization layer and partial decarburization layer increased with increasing temperature. It is indicated that decarburization grew dramatically with increasing temperature. The depth of the ferrite decarburization layer and partial decarburization layer grew for 225 μm and 405 μm at 1050°C. In contrast, the coated specimens presented a slow decarburization process. The coating showed a reduction in decarburization of 100% at 1050°C. It suggests that the coating prevented the decarburization during high-temperature treatment.
Microstructure of 50CrVA spring steel, after heating for 120 min at 1050°C in a muffle furnace, (a) bare and (b) coated.
Left of Fig. 3 shows the total decarburization growth of bare and coated specimens with different temperatures during heating for 120 min. For the bare specimens, the depth of decarburization layer increased with increasing temperature. For the coated specimens, the depth of decarburization layer increased obviously from 900°C to 950°C and increased slowly from 950°C to 1000°C. From 1000°C to 1050°C, the depth of decarburization layer decreased dramatically. Until to 1050°C, the depth of decarburization layer disappeared completely. Light of Fig. 3 showed the decarburization growth of bare and coated samples with different holding time at 1050°C. For the bare specimens, the depth decarburization layer increased with increasing holding time. Conversely, the depth of the decarburization layer of coated specimens decreased dramatically with increasing holding time. The protective coating can prevent the decarburization process effectively. After holding 120 min, the decarburization layer of specimen disappeared completely.
Decarburization growth of bare and coated 50CrVA alloy specimens with different temperatures during heating for 120 min (left) and with different holding time at 1050°C (right).
Oxidation process depends on two factors after forming thin oxide layer on the surface of alloy.20,21) One factor is the reaction rate on the interface of Fe-oxide layer/O2-oxide layer. The other factor is the diffusion rate of reaction product in oxide layer. Firstly, the ferric element in the surface layer reacts with O2, CO2 and H2O. Chemical equations are listed as follows.
For 50CrVA spring steel, chemical reactions near the surface with temperature rise are listed as follows.
The two factors dominate the full reaction process. When the oxide layer is very thin, the interface reaction rate played a leading role. The diffusion rate gradually dominated the full reaction with the increase of oxide layer.
As shown in Fig. 4 the oxide layer ingredients of bare specimen consist of Fe3O4 and Fe2O3 at 1050°C. The oxide layer ingredients of coated specimen mainly consist of Fe2O3 at 1050°C. To bare specimen, the layer close to substrate is composed of Fe3O4. To coated specimen, the layer of Fe3O4 is disappeared, because the coating prevents the further oxidation. The layer of Fe2O3 is easy to remove from substrate because of loose structure.
XRD patterns of specimens without coating (a) and with coating (b) after heating at 1050°C for 0 min.
The SiO2 from bauxite is found at 1050°C for 0 min. With extending the holding time, SiO2 will be absorbed by liquid phase. Al2O3 from bauxite has a high melting point. In complex system, the melting point is low because of solid solution.22) Thus functional ingredients of raw material in complex system are easy to sintering. The Na2Al6P2O15 phase in Fig. 4 is the product during heating process according to Eq. (15). As the temperature approaches 1050°C the new phase generated and filled in the voids through wetting and surface tension. Depending upon the oxidation conditions, outward migration of Fe ions or inward diffusion of O ions, across the oxide layer, can predominate the oxidation reaction. Firstly, the Fe in porous oxide layer diffuses from inner to surface. This medium of diffusion is porous oxide layer. Secondly, the O2 in atmosphere diffuses from surface to inner. This kind of diffusion medium is alloy surface. Thirdly, the Fe in alloy and O2 in atmosphere diffuse mutually. The diffusion media is porous oxide layer. The Fe contacted with O2 easily because of the aforementioned diffusion. So, reaction rate will be increased and then oxide layer becomes thicker and thicker. For the composition of oxide layer, the Fe content is higher near to the alloy surface. The products of chemical reaction and liquid phase form complex network structure to prevent the diffusion of Fe, C and O. So, the coating prevents the decarburization and oxidation.
The oxide layers of bare and coated specimen were grounded into powders before XRD test. After heating at 1050°C for 30 min (a), 60 min (b) and 120 min (c) the phase changes are shown in Fig. 5. With the extension of heating time, the SiO2 and Na2Al6P2O15 in Fig. 4, formed the glass phase. So, the SiO2 and Na2Al6P2O15 phases were not found in Figs. 5(a) and 5(b). But, while holding for 120 min, the saturated SiO2 precipitated again as shown in Fig. 5(c) because of composition change of the glass phase. As heating at 1050°C for 30 min, phases of interface of steel/coating mainly are Al2FeO4 and Al2O3. More Al2O3 in bauxite was deposited and Na2Al6P2O15 phase was absorbed on the interface of steel/coating with increasing heating time. After heating at 1050°C for 60 min, the phases mainly are Fe3O4 and Fe2SiO4. When heating at 1050°C for 120 min, the phases mainly are Fe2SiO4 and SiO2. The enrichment layer of SiO2 on the interface of steel/coating can erode early decarburization layer and lead to the disappearing of decarburization layer eventually. The oxidation of SiC generates activated carbon atoms. The micro area is reducing atmosphere because of activated carbon atoms. The SiO2-cristobalite can prevent the diffusion between oxygen and carbon.23)
XRD patterns of coated specimens after heating at 1050°C for 30 min (a), 60 min (b) and 120 min (c).
The protective effects of coating were most obvious in gas loaded with more O2 content, H2O in sequence. The main reason was that O2 and H2O reacted with SiC. Meanwhile, the SiC was one of the functional compositions. H2O of reaction activity with SiC was lower than O2. Both the reactions can generate amorphous SiO2 as shown in Fig. 5. The amorphous SiO2 with high reaction activity differs from the SiO2 from bauxite.24) The amorphous SiO2 is easier to react with Fe2O3. So, SiO2, one of reaction products, eroded the decarburization layer, and then resulted in the removal of decarburization layer subsequently. Meantime, the reaction products cut off the contractions between oxygen and substrate.
For bare specimens, O2, and H2O reacted with Fe3C of steel surface and then resulted in decarburization as shown in Eqs. (12), (13), (14). For coated specimens, O2, and H2O could not contact with steel surface. At high temperature, O2, and H2O reacted with the functional compositions of coating. In addition, the reaction temperature declined in complex system.
Cross-section microstructures and corresponding elemental mapping of coated specimens after heating at 1050°C for 120 min are shown in Fig. 6. The upper part of the figure is the mapping image of the interface of steel/coating. The lower part of the figure is the mapping image of the interface of coating/air. The coating materials, oxide layer and decarburization layer reacted to each other and formed protective layer. From Eq. (18), the amorphous SiO2 reacted with the Fe2O3, thus eroded the earlier decarburization layer. In the upper layer, enrichment of Si and Fe are found in Si and Fe element maps of Fig. 6. Subsequently, the Fe2SiO4 layer on the interface of steel/coating was produced because of the above reaction. The corresponding elemental mapping is in accordance with XRD results of Fig. 5. Al element existed in noncrystalline phases and filled in the voids at high temperature. Near to the interface of steel/coating, the coating layer is more compact. So, there is more Al content in the upper layer. O element also mainly existed in noncrystalline phases. Thus, Al and O content have the similar distributing process. Only but, the O content has much higher level. At last, we can conclude that the coating prevents the diffusion of oxygen.
Cross-section microstructures and corresponding elemental mapping of coated specimens after heating at 1050°C for 120 min.
From the DTA curves as shown in Fig. 7, the process of the coating formation was clearly detected. The coating powder experienced endothermic process at around 100°C when dehydration process occurred. At around 780°C, an evident endothermic process of bare and coated specimens happened because of austenization. To bare specimen, a distinct exothermal process was observed at 950°C. It is shown that the oxidation reaction occurred at 950°C. In this temperature range, the raw materials of coating powder reacted to each other. By contrast, the oxidation process of coated specimen was not appeared. At 900°C and 1010°C, the reactions of Eqs. (15) and (18) occurred respectively. Extending the holding time at 1050°C, the protection of decarburization was more effective, because the main reactions of the coating, which gave the protective property, could be finished completely.12) The tapping temperature of 50CrVA spring steel is about 1050°C, so the diffraction peaks above 1050°C are useless under actual working condition.
DTA curves of specimens with and without coating.
Melting process of coating powders is very important to protection. Because, the liquid phase play a key role of protective effect at high temperature. 50CrVA spring steel was protected by coating through chemical reaction at high temperature. Melting process was conducive to chemical reaction at high temperature. Coating powders were pressed into a column with diameter of 3 mm and height of 3 mm, then were placed on an alumina disc and were heated at a rate of 5°C/min. The thermal softening behavior of coating was observed by hot stage microscope images equipment. Figure 8 shows the hot stage microscope images of coating powders at different stage. Black part of central figure is the sample of coating powder. White part of the figure is the hearth of equipment.
Hot stage microscope images of coating powder during heating with rate of 5°C/min, (a) 203°C, (b) 653°C and (c) 889°C.
Invisible changes were observed at 203°C while obvious changes were observed at 653°C, which indicates the chemical reaction from 653°C. However, when the temperature rose to 889°C, the column of sample became to melt. The height of column decreased obviously, as well as hemispherical column formed which meant the coating had melted down completely. At this moment, the coating has already melted. So, the coating can protect the steel effectively from this temperature point. So far, according to experimental data, we can conclude that the coating powders start to melt from 653°C and melt down completely at 889°C. Below the melting temperature point, the necessary oxidation could be achieved while heated to service temperatures.25,26) So, according to Fig. 7, the oxidation both bare and coated specimen was observed bellow 850°C. This is also the reason that anti-oxidation effect of coating cannot reach to the 100%. Therefore, it is believed that the as-prepared coating would melt down at high temperature (i.e. 1050°C in this study), and then transformed into a homogeneous, continuous inorganic coating layer.
The coating on the steel surface played different roles with changing temperatures. The protection of coating came true through chemical reaction and phase changes at high temperature. Fig. 9 shows the protective process and mechanism of the coating.
Protective process and mechanism of the coating.
As shown in Fig. 9, in the early stage, especially below 700°C, the coating on the surface was porous. The O2 can react with Fe and C through the voids. At lower temperatures, the diffusion of C and Fe in steel was slower than at higher temperatures. So, the thin oxide and decarburization layers were produced under the coating. The SiC, one of raw materials, reacted with the H2O and O2. The reaction products, CH4 and CO2, can provide reducing atmosphere in micro area. With increasing temperature, the raw materials of coating powders reacted to each other. Meantime, the coating ingredients reacted with steel surface. Subsequently, the coating became to compact and eroded the early decarburization layer. For the densification of the coating to occur, a higher temperature and longer holding time of heating were needed. The melting phases were formed with the sintering process at higher temperatures. The solid phases filled the voids through surface tension and wetting power between the solid and liquid phases and made the coating to become more compact. So, the formed effective film prevents the secondary diffusion of oxygen and carbon at higher temperatures. Meanwhile, it also prevented the reaction of oxygen with the substrate at 1050°C. Reaction products, Fe2SiO4 and Na2Al6P2O15, were produced on the steel surface. At last, uniform melting shielded the substrate. When the oxidation layer on the surface was removed by descaling equipment, the steel surface was smooth and without decarburization layer.
(1) An inorganic coating on 50CrVA spring steel was successfully fabricated by a simple and low-cost spraying technique and using bauxite, silicon carbide, sodium silicate, feldspar and phosphate binders, to prevent decarburization and oxidation during heating process.
(2) 50CrVA spring steel was markedly protected being oxidized and decarburized by coating during thermal exposure. Weight gain of coated specimen was reduced by 67.82% at 1050°C holding for 120 min. The decarburization layer of coated specimen disappeared completely with increasing temperature. The depth of the decarburization layer of coated specimens decreased dramatically with increasing heating temperature and holding time. After holding 120 min at 1050°C, the decarburization layer of coated specimen disappeared completely.
(3) The as-prepared coating had melted down at 1050°C, transforming into a homogeneous, and continuous inorganic coating. The coating on the steel surface played different roles with changing temperatures. At high temperature, the melting coating reacted with 50CrVA spring steel surface. The coating eroded the early decarburization layer. The small decarburization layer dissolved into the oxidation layer ahead of the coating protective effect. Finally, there was no decarburization on the surface.
The authors acknowledge the financial supports by Natural Science Foundation of China (No. 51202249), the 863 Project (2011AA06A104), and Projects in the National Science & Technology Pillar Program during the 12th Five-year Plan Period (2012BAB08B04).