Surface modification of CA6NM hydroturbine steel for protection against erosion

Abstract

The world is facing energy crisis due to depletion of fossil fuels at a very fast rate. In current scenario, renewable energy sources are given considerable attention world-wide to meet the ever-increasing energy demands. Survival of human race will be difficult without the effective utilization of these resources in daily life. The added advantage of using such resources is their non-polluting nature. One such energy source, widely utilized all around the globe, is hydropower. In Southeast Asia, the rivers originating from Himalayas are the main source of hydropower. However, since Himalayas are relatively young, their morphology is rather fragile in nature, which eventually leads to higher concentration of sand particles in river water. When this sand-laden water passes through turbines, the degradation of submerged components takes place. The phenomenon of loss of material from the surface of the components due to interaction with sand particles entrained in river water is known as slurry erosion (SE). Another form of erosion, known as cavitation erosion (CE) is also a predominant form of degradation, existing in hydroturbines. Both SE and CE significantly affect the performance of hydroturbines by directly lowering their efficiency and damaging the submerged turbine components. Several-a-time, a sudden rise in sand concentration due to natural causes such as heavy rains and floods in the mountains leads to the shutdown of the plants, which results in enormous economic losses. 13Cr4Ni steel (Trade name: CA6NM) introduced in early 1960s, is a primary structural material used in hydroturbines. Although several other steels such as SS316L, 16Cr5Ni, CF8M, 13Cr1Ni have also been suggested as structural materials for hydroturbines, however, the properties such as high corrosion resistance and nominal cavitation erosion resistance have made CA6NM as the most preferred choice. It is important to mention that CA6NM is not capable enough to provide the sufficient resistance against SE. Therefore, to protect the CA6NM turbine components from SE, several surface modification techniques have been proposed and investigated. Among these techniques, thermal spraying has gained wide popularity owing to its versatile nature and capability to deposit a wide spectrum of materials on a variety of substrate materials. Friction stir processing (FSP) has recently emerged as an effective tool for the surface and bulk refinement of the materials. It is known that ultra-fine grained (UFG) materials possess superior mechanical properties, which could further lead to improvement in the performance of the materials against wear. FSP is one of the severe plastic deformation (SPD) processes wherein material is subjected to high strain rates resulting into high temperatures. These extreme conditions incubate the recrystallization of the material and lead to refine its microstructure. Moreover, it can also help in breaking the existing micron-scale secondary phase, if any, which leads to the formation of composite-like structure with enhanced mechanical properties. In the present work, two different surface modification techniques namely; thermal spraying and FSP were utilized for enhancing the mechanical, as well as, tribological properties of CA6NM steel. Ni-Al2O3 based powders with proportion of Al2O3 varied from 20 wt. % to 60 wt. % were deposited on CA6NM steel by high velocity flame spraying. In addition, FSP was also employed to modify the surface of CA6NM steel. The processed surface layer was immediately cooled using methanol kept, at -20° C, to help further refinement of the material. In-depth analysis of the thermal sprayed and friction stir processed (FSPed) samples was undertaken using optical microscope (OM), scanning electron microscope (SEM) equipped with (EDS) and X-ray diffraction (XRD). In addition to this, electron back-scatter diffraction (EBSD) technique was also used for the analysis of the FSPed sample. Microand nano-indentation testing was also performed along with scratch resistance evaluation. Moreover, the tribological performance of both the sprayed, as well as, FSPed samples under slurry and cavitation erosion conditions was also evaluated. The design and development of slurry erosion test rig was done in-house for this performance evaluation. It was observed that the as-sprayed coatings consisted of typical splat-like morphology for all the investigated compositions. The splats of alumina exhibited a pancake like morphology without the presence of corona. The splats of Ni also possessed pancake like morphology, but with the presence of corona. Microstructural attributes such as splat size and shape, porosity, un-melted particles and surface roughness were observed to be highly influenced by the alumina content in the coatings. The porosity was found to increase with the alumina content with a maximum value of 2.5 % for the coating containing 60 wt. % Al2O3. With the rise in alumina content in the coating, the size of the alumina splat was reduced. XRD results indicated the presence of both α and γ phases of Al2O3 in the coatings. Some low-intensity peaks corresponding to NiO phase were also observed. Micro- and nano-indentation results showed that with the increase in alumina content, hardness and Young’s modulus for the coatings were increased. An average hardness of 1151 HV0.1kgf was observed for the coating containing 60 wt. % Al2O3. Fracture toughness showed a non-linear behavior with respect to the fraction of Al2O3 in the coatings. It was found to be highly dependent on the microstructure of the coatings. A maximum fracture toughness of 1.4 MPa-m1/2 was observed for the coating containing 40 wt. % Al2O3. XRD analysis showed that with the increase in percentage of Al2O3 in the coating, the compressive residual stresses decreased. Surface roughness of the coatings was also found to be dependent upon the amount of Al2O3 in the coatings. A model has been proposed, which can help in understanding the dependency shown by surface roughness of the coatings on the Al2O3 content. Friction stir processing resulted in significant refinement in the microstructure of CA6NM steel. The average grain size of the steel was found to be reduced by a factor of 10 after FSP. EBSD results confirmed that the average grain size of processed CA6NM steel was in the range of 2.6 - 4 µm, in contrary to 25 µm of the unprocessed steel. Maximum refinement was observed at the centre of the processed zone (stirred zone, SZ), where around 15 % grains were of size less than or equal to 1 µm. Analysis of the pole figures revealed a significant difference between the textures of the processed zones and the unprocessed zones of the steel. Proportion of high angle grain boundaries (HGBs) was also found to reduce after the processing. After FSP, the hardness of the CA6NM steel was increased to nearly 2.6 times at the surface. The relationship between the hardness and the grain size in stir, advancing and retreating zone was found to be in accordance with the Hall-Petch relation. An analysis of the data obtained from the nano-indentation studies showed that FSP also helped in enhancing the yield strength of the steel. However, the strain-hardening exponent was observed to decrease after FSP. Slurry erosion (SE) studies showed that ploughing and platelet mechanisms play a dominant role in the erosion of CA6NM steel. Maximum erosion rate for the steel was observed at 30° impingement angle. Velocity was found to have a high level of interaction with the concentration of the erodents. Models have been proposed to explain indentation and mixed cutting-ploughing mechanisms of erosion, which are not well understood in the open literature. All the coatings were effective in controlling the SE of the CA6NM steel. Among the coatings, one containing 40 wt. % Al2O3 showed the highest resistance against SE. This could be attributed to the presence of optimum combination of the hardness and fracture toughness. This coating showed around 3.5 times higher resistance against SE in comparison to CA6NM steel. Primary mechanisms responsible for the loss of coating were found to be the splat delamination, removal of chunk of material (formed by interlinking of cracks) and fracture of Al2O3 splats. Attempt was made to correlate the erosion resistance of these coatings with various microstructural and mechanical parameters. Among different parameters,( ) 1/3 2 K HIC , a function of fracture toughness (KIC) and hardness (H) showed an excellent correlation with erosion resistance of coatings at both the impingement angles (30° and 90°). In comparison to the coatings, FSP was able to reduce the SE of the CA6NM steel by only 60 %. In other words, FSP was found to be less effective in comparison with thermal spraying. No significant difference in the erosion mechanisms of the FSPed and unprocessed CA6NM steel was observed. Cavitation erosion (CE) studies showed that none of the developed coatings was effective in controlling the damage. However, similar to SE studies, the coating containing 40 wt. % Al2O3 showed the highest resistance against CE among the investigated coatings. The content of alumina was found to be having significant effect on the erosion response and the degradation mechanism of the coatings. De-bonding of the splat, initiation of the pits in the Ni matrix and the fracturing of Al2O3 splats were observed to be the primary material removal mechanisms. Fracture toughness was found to have significant effect on the resistance of the coatings against CE. The CE resistance of CA6NM steel got enhanced by around 2.4 times after FSP. This improvement in the CE resistance of CA6NM steel after FSP could be attributed to its higher hardness and yield strength in comparison to the unprocessed steel. No significant difference between the material removal mechanisms for the unprocessed and processed CA6NM steel was observed. The observed mechanisms for these cases included the formation of pits and lips. A model has been proposed in the current investigation to predict the erosion mechanisms for different materials, namely, “Erosion Mechanism Identifier, ξ” as shown below: 2 0.5 r H E K mV       = σ ξ This model showed a good correlation between the predicted and experimentally observed results. This model was validated against the results compiled from the literature for various classes of materials. It has been established that ‘ξ’ may help in predicting the erosion mechanisms for different categories of materials and coatings. Another model for predicting the slurry erosion rates of the ductile materials based on theory of plasticity has been proposed as given below: 2 2 2 9 (1 ) () 4 r y s Ve E E f xmxf nK G −   =      θ σ Further, a similar model developed using fracture and contact mechanics approach has also been proposed for the coatings. A mathematical model proposed for predicting the slurry erosion rates of ductile materials was found to follow the trend shown by the experimental results. The model proposed for predicting the erosion rates of the coatings was able to do so within a tolerance range of 20 to 30 %. The coupling of the ductile model with the computational fluid dynamics (CFD) approach further helped in improving the predictability of the model. It was concluded that since SE is a localized phenomenon, which is highly dependent on the local impingement velocity and angle, therefore, CFD could be an efficient technique for predicting the slurry erosion rates of the materials.