In my work on high-performance aero-engine hot-section components, I have repeatedly observed that the manufacture of single-crystal turbine outer ring castings is limited less by the nominal solidification process than by a persistent surface casting defect known as recrystallization. The outer ring sits in the combustion chamber region, where thermal gradients are severe, mechanical loads are complex, and service life requirements are long. Under these conditions, nickel-based single-crystal superalloys are selected because they provide excellent high-temperature strength, creep resistance, fatigue resistance, and resistance to oxidation and hot corrosion. However, the same alloy chemistry and crystallographic perfection that make single-crystal components attractive also make them highly sensitive to any local loss of single-crystal integrity. When a small surface region undergoes plastic deformation before or during high-temperature heat treatment, new grains can nucleate and grow with random orientations. This produces a recrystallized layer that behaves as a weak zone inside an otherwise strong single crystal. The result is a serious casting defect that can reduce creep life, accelerate fatigue crack initiation, and ultimately cause premature failure.
I selected a second-generation nickel-based single-crystal superalloy and produced [001]-oriented turbine outer ring castings by the selector method in a directional solidification furnace. Production statistics showed that the main factors reducing single-crystal integrity were stray grains and recrystallization. Recrystallization alone accounted for more than 75% of rejected parts in some campaigns. Therefore, controlling recrystallization became the central objective of my study. I systematically analyzed the structural features of the turbine outer ring and the influence of the casting process on defect formation, clarified the formation mechanism of recrystallization, and developed a pre-heat-treatment electrolytic corrosion stress-relief process. By optimizing the electrolytic parameters, I reduced the recrystallization defect rate from approximately 70% to below 5%, significantly improving single-crystal integrity and product qualification rate. In this article I present the mechanism, the experimental method, the parameter windows, and the practical conclusions that I obtained.
The appearance of a recrystallized region is not merely a cosmetic problem. In a single-crystal superalloy, grain-boundary strengthening elements are absent or present only in very small amounts. Therefore, any high-angle boundary that forms during recrystallization has a low migration resistance and becomes a preferred path for grain-boundary sliding, oxidation, and crack propagation. I have seen that a recrystallized layer can dominate the failure process even when the bulk crystal is perfectly oriented. The local casting defect becomes a mechanical weak link. This is why recrystallization control is not optional in single-crystal turbine outer ring production. It is a mandatory condition for reliable batch manufacturing.
To describe recrystallization in a quantitative way, I considered both the thermodynamic driving force and the kinetic barriers. The total free-energy change associated with the formation of a new recrystallized grain can be written as
$$
\Delta G = \Delta G_v + \Delta G_s + \Delta G_e
$$
where \(\Delta G_v\) is the volume free-energy difference between the deformed and recrystallized states, \(\Delta G_s\) is the surface energy contribution, and \(\Delta G_e\) is the stored elastic strain energy. In a surface layer that has experienced plastic deformation, the stored energy term \(\Delta G_e\) is large. This term provides the driving force for nucleation. The critical nucleation energy for a spherical nucleus can be approximated by
$$
\Delta G^* = \frac{16\pi \gamma^3}{3(\Delta G_v – \Delta G_e)^2}
$$
where \(\gamma\) is the interfacial energy. The equation shows that as the stored strain energy increases, the nucleation barrier decreases. This is the fundamental reason why a seemingly minor surface scratch, grinding mark, or local collision can later become a recrystallization site after high-temperature exposure. I also considered the grain growth rate, which can be described by
$$
v = M \cdot P
$$
where \(M\) is the grain-boundary mobility and \(P\) is the driving pressure. For a recrystallized grain in a single-crystal matrix, \(M\) is relatively high because boundary pinning by second phases or grain-boundary segregants is limited. Thus, once a nucleus forms, growth can proceed rapidly during solution treatment. This combination of a low nucleation barrier and a high growth rate explains why recrystallization is such a persistent casting defect in single-crystal outer rings.
In my analysis of the turbine outer ring, the structural complexity was a major cause of high recrystallization tendency. The wall thickness varies significantly, the central window is deep, and there are multiple abrupt cross-section changes. These features disturb the temperature field during solidification and produce concentrated thermal stresses. In addition, the back-side bosses used for mechanical positioning are small in size and have a large height-to-diameter ratio. The transition radius at the top of each boss is a natural stress concentrator during solidification shrinkage. When the casting is later heated for solution treatment, this stored stress is released rapidly, and surface recrystallization is easily triggered. In my view, the geometry of the boss region is therefore the most critical location for recrystallization control.
I also examined the role of the mold shell and core removal process. During directional solidification, the mold shell contracts around the casting, and the ceramic core can impose local constraint. After casting, shell removal and core leaching can introduce additional mechanical impact. Grinding, shaping, and handling steps can produce shallow plastic deformation. Each of these operations can contribute to the surface deformation layer. When the component is subsequently held at high temperature during solution treatment, the deformed layer provides the necessary stored energy for recrystallization. The combination of high temperature and residual stress supplies both the thermodynamic driving force and the kinetic path. Therefore, the recrystallization process is not caused by a single factor but by the accumulation of casting, finishing, and heat-treatment effects.
My experimental design was based on a simple idea: if the surface stress layer can be removed before high-temperature heat treatment, the driving force for recrystallization will be reduced. Electrolytic corrosion is attractive because it can dissolve a controlled amount of surface material, remove the damaged layer, and simultaneously blunt sharp corners. The latter effect is important because a sharp transition radius concentrates stress and can act as a nucleation site. By rounding the edge, the stress concentration factor is reduced, and the local mechanical driving force for recrystallization is lowered. In this study, I used electrolytic corrosion as a pre-heat-treatment stress-relief and surface-conditioning step.
The electrolytic solution that I used was prepared with the following composition. The values are given as a practical batch formulation.
| Component | Amount | Function in the electrolyte |
|---|---|---|
| Hydrochloric acid | 950 mL | Provides chloride ions and acidic attack |
| Water | 800 mL | Solvent and conductivity medium |
| Iron(III) chloride | 550 g | Oxidizing agent and etching accelerator |
| Ethanol | 175 mL | Wetting and surface activation |
| Citric acid | 225 g | Complexing agent for controlled dissolution |
| Dodecylbenzenesulfonic acid | 22.5 g | Surfactant for uniform electrolyte contact |
Before electrolysis, I masked the areas that were not intended for treatment with insulating tape. This step is essential because uncontrolled dissolution can change the dimensions of functional surfaces and introduce new surface casting defects. The masked casting was then fixed with an electrode and immersed in the electrolyte. I set the power supply to a voltage of 1.5 V and a current of 4 A. The treatment was applied to the boss region, where recrystallization had been most frequently observed. To determine the effect of removal depth, I varied the electrolysis time over 3, 5, 8, 10, 13, and 15 min. After electrolysis, each casting was thoroughly rinsed and dried. Then I applied the standard solution and aging treatment. Finally, I performed macroscopic etching to evaluate single-crystal integrity.
The relationship between electrolysis time and material removal can be described by Faraday’s law. For a given current and time, the theoretical mass removal is
$$
m = \frac{I t M}{z F}
$$
where \(m\) is the removed mass, \(I\) is the current, \(t\) is the time, \(M\) is the molar mass, \(z\) is the valence number, and \(F\) is the Faraday constant. If the treated area is \(A\) and the alloy density is \(\rho\), the average removal depth is
$$
d = \frac{m}{\rho A} = \frac{I t M}{z F A \rho}
$$
In practice, the local removal depth is not perfectly uniform because the current distribution depends on the geometry, the electrolyte flow, and the presence of edges. Nevertheless, this relationship provided a useful guide for selecting the time window. The measured removal depths that I obtained are summarized in the table below.
| Electrolysis time (min) | Approximate removal depth (mm) | Surface appearance after electrolysis |
|---|---|---|
| 3 | 0.010 | Bright, minimal change |
| 5 | 0.020 | Bright, slight edge rounding |
| 8 | 0.035 | Bright, moderate edge rounding |
| 10 | 0.040 | Bright, clear edge rounding |
| 13 | 0.060 | Dark, visible over-etching |
| 15 | 0.080 | Dark, severe over-etching and dendritic relief |
The heat-treatment schedule that I used after electrolysis is given below. The solution treatment was performed at a selected temperature between 1295°C and 1305°C for 2 h, followed by air cooling to below 100°C. The first aging treatment was conducted at 1120°C for 4 h and air cooled. The second aging treatment was conducted at 1080°C for 4 h and air cooled. This schedule is typical for the second-generation nickel-based single-crystal superalloy that I studied, and it provides sufficient thermal exposure to reveal any recrystallization that may have been triggered by surface deformation.
| Process step | Temperature and time | Cooling condition |
|---|---|---|
| Solution treatment | 1295–1305°C, 2 h | Air cooling to below 100°C |
| First aging | 1120°C, 4 h | Air cooling to below 100°C |
| Second aging | 1080°C, 4 h | Air cooling to below 100°C |
After electrolysis, I inspected the castings visually. For treatment times from 3 to 10 min, the surfaces remained bright. For treatment times of 13 and 15 min, the surfaces became dark and showed clear signs of over-corrosion. In addition, the electrolytic treatment effectively removed casting burrs and sharp corners at the boss transition radius. The contour became smoother and more rounded. This geometric change is important because a rounded profile reduces the stress concentration factor. I noted that the treatment also removed a thin surface layer that could contain residual deformation from shell removal, grinding, or handling. Therefore, the process addresses both the material and the geometric contributions to recrystallization.
After heat treatment, I compared the treated castings with untreated castings. In the untreated condition, the boss radius often showed distinct recrystallized grains. In the treated condition, the transition region was visibly smoother. Macroscopic etching revealed the following results. For electrolysis times of 3 and 5 min, the castings still contained many large recrystallized grains. The removal depth was insufficient to eliminate the deformed layer completely. For electrolysis times of 8 and 10 min, only a few small recrystallized spots remained, typically fewer than five per casting. These small spots could be removed by subsequent polishing, and the castings were considered acceptable. For electrolysis times of 13 and 15 min, no recrystallization was detected. However, the over-corrosion produced deep dendritic relief on the surface, which damaged dimensional accuracy and appearance and made finishing difficult. Therefore, the optimal window was 8 to 10 min, corresponding to a removal depth of 0.035 to 0.040 mm.
| Electrolysis time (min) | Removal depth (mm) | Recrystallization observation after heat treatment | Surface quality | Acceptability |
|---|---|---|---|---|
| 3 | 0.010 | Many large grains | Bright | Not acceptable |
| 5 | 0.020 | Many large grains | Bright | Not acceptable |
| 8 | 0.035 | Few small spots, less than 5 | Bright, rounded edges | Acceptable |
| 10 | 0.040 | Few small spots, less than 5 | Bright, rounded edges | Acceptable |
| 13 | 0.060 | No recrystallization | Dark, over-etched | Not acceptable due to surface damage |
| 15 | 0.080 | No recrystallization | Dark, severe dendritic relief | Not acceptable due to surface damage |
The statistical improvement that I achieved was substantial. Before implementing the electrolytic corrosion process, the recrystallization-related rejection rate was approximately 70%. After optimizing the process, the recrystallization defect rate fell to below 5%. This reduction changed the economics of production because fewer castings were scrapped after expensive directional solidification and heat treatment. It also improved confidence in the single-crystal integrity of the delivered parts. I regard this as a clear demonstration that a surface-controlled electrochemical step can solve a difficult casting defect problem without changing the alloy composition or the basic solidification route.
| Production stage | Recrystallization-related rejection rate | Single-crystal integrity | Product qualification rate |
|---|---|---|---|
| Before electrolytic corrosion control | Approximately 70% | Frequently degraded by surface grains | Low |
| After electrolytic corrosion control | Below 5% | Significantly improved | High |
To understand why the electrolytic process works, I considered the stress state at the boss radius. During solidification, the boss cools more rapidly than the surrounding thicker sections because it has a smaller thermal mass. The contraction mismatch generates a tensile stress at the transition radius. If the local stress exceeds the yield strength at high temperature, plastic deformation occurs. The stored dislocation structure then provides the driving force for recrystallization during solution treatment. By removing a thin surface layer, the electrolytic process removes the region with the highest stored energy. By rounding the radius, it reduces the elastic stress concentration. Both effects lower the total driving force. The nucleation rate can be expressed in a simplified form as
$$
N = N_0 \exp\left(-\frac{\Delta G^*}{kT}\right)
$$
where \(N\) is the nucleation rate, \(N_0\) is a pre-exponential factor, \(k\) is the Boltzmann constant, and \(T\) is the absolute temperature. When the stored energy is reduced, \(\Delta G^*\) increases, and the nucleation rate decreases exponentially. This explains why even a modest removal depth of a few tens of micrometers can produce a large improvement. The process does not need to remove the entire deformed region; it only needs to remove enough of the highest-energy layer to push the local nucleation rate below a practically observable level.
I also considered the effect of surface curvature on stress concentration. For a rounded notch or fillet, the elastic stress concentration factor \(K_t\) can be approximated as
$$
K_t = 1 + \alpha \sqrt{\frac{h}{r}}
$$
where \(h\) is the section height, \(r\) is the fillet radius, and \(\alpha\) is a geometry-dependent coefficient. Increasing the fillet radius \(r\) reduces \(K_t\). Electrolytic corrosion naturally increases the effective radius at sharp corners. Therefore, the treatment produces a double benefit: it removes the deformed material and it reduces the mechanical driving force for further deformation or stress-assisted nucleation. In my experiments, the edge rounding was clearly visible after 8 to 10 min of treatment, and the recrystallization response was correspondingly reduced.
The role of the electrolyte composition is also worth discussing. The hydrochloric acid provides a high chloride concentration, which promotes uniform dissolution of the nickel-based superalloy. The iron(III) chloride acts as an oxidizing agent and helps to maintain a controlled dissolution rate. Citric acid complexes metal ions and prevents the formation of dense surface films that could cause uneven etching. Ethanol improves wetting, especially in narrow gaps and around the boss radius. The surfactant, dodecylbenzenesulfonic acid, reduces surface tension and helps the electrolyte reach recessed areas. Together, these components produce a controlled and relatively uniform removal process. I found that the formulation was stable over repeated runs and could be used for batch processing with appropriate filtration and replenishment.
| Parameter | Value used in this study | Effect on the process |
|---|---|---|
| Voltage | 1.5 V | Controls driving force for dissolution |
| Current | 4 A | Controls removal rate according to Faraday’s law |
| Electrolysis time | 8–10 min optimum | Determines removal depth and edge rounding |
| Removal depth | 0.035–0.040 mm | Sufficient to remove the stressed surface layer |
| Masking | Insulating tape on non-target areas | Prevents unwanted dimensional changes |
The relationship between removal depth and residual recrystallization can be summarized by a simple empirical trend. As the removal depth increases, the number of recrystallized grains decreases. However, beyond a certain depth, the surface quality deteriorates. Therefore, the process has an optimum window rather than a monotonic “more is better” behavior. I expressed this as a balance between two competing effects. The first is the beneficial removal of stored energy, which reduces the recrystallization driving force. The second is the damaging over-etching, which creates surface roughness and dimensional error. The optimum condition is the point where the recrystallization count falls to an acceptable level while the surface remains bright and dimensionally accurate. In my study, this point was reached at 8 to 10 min.
$$
\text{Benefit} \propto \int_{0}^{d} \rho_e(x) \, dx
$$
$$
\text{Damage} \propto f(d, \text{current distribution}, \text{surface roughness})
$$
In these expressions, \(\rho_e(x)\) represents the stored energy density as a function of depth, and \(d\) is the removal depth. The integral represents the amount of stored energy removed by the electrolytic treatment. The damage term increases rapidly once the removal depth exceeds the thickness of the damaged layer. This is why I recommend a controlled removal depth rather than a long treatment time.
I also evaluated the effect of the process on the single-crystal integrity of the entire casting, not only the boss region. Macroscopic etching showed that the single-crystal nature was preserved in the bulk of the casting. The electrolytic treatment affected only the surface, and the removal depth was small compared with the wall thickness. Therefore, the crystallographic orientation of the bulk material remained unchanged. The treatment did not introduce new grain boundaries in the interior. It simply removed a thin surface layer that would otherwise have acted as a recrystallization source. This is a key advantage of the process: it improves surface quality without compromising the bulk single-crystal structure.
The aging treatments after solution treatment also deserve attention. The first aging at 1120°C and the second aging at 1080°C produce the gamma-prime precipitate distribution that is required for high-temperature strength. These treatments are performed at temperatures well below the solution temperature, so they do not provide a strong additional driving force for recrystallization. However, if recrystallization has already nucleated during solution treatment, the aging treatment may allow the recrystallized grains to grow slightly or to become more visible after etching. Therefore, the critical control point is the solution treatment. The electrolytic process must be completed before solution treatment so that the surface stress layer is removed before the component experiences the highest temperature in the heat-treatment cycle.
From a manufacturing perspective, the process has several practical advantages. It is performed at room temperature, so it does not require additional high-temperature equipment. It can be integrated into the existing production line after casting cleanup and before heat treatment. The electrolyte can be maintained in a tank, and the current and time can be controlled automatically. The masking step adds labor, but it can be standardized with reusable fixtures or protective coatings. The process is also adaptable to different casting geometries by adjusting the electrode position and the treatment time. For the turbine outer ring, the boss region is the critical area, so the electrode can be shaped to focus the current on that region. This improves efficiency and reduces the risk of over-treating other surfaces.
| Manufacturing aspect | Observation in this study | Practical implication |
|---|---|---|
| Integration point | After casting cleanup, before heat treatment | Removes stress layer before solution treatment |
| Equipment | Simple DC power supply and electrolyte tank | Low capital cost and easy scale-up |
| Process control | Voltage, current, and time | Repeatable removal depth |
| Masking | Insulating tape or protective coating | Protects functional surfaces |
| Inspection | Macroscopic etching after heat treatment | Confirms single-crystal integrity |
The recrystallization mechanism can be summarized in a sequence. First, localized plastic deformation occurs at the boss radius during solidification, shell removal, or handling. Second, the deformed region stores dislocation energy. Third, during solution treatment, the high temperature activates recovery and recrystallization. Fourth, new grains nucleate at the surface and grow into the single-crystal matrix. Fifth, the recrystallized grains form high-angle boundaries that degrade mechanical performance. The electrolytic process interrupts this sequence at the second step by removing the stored-energy layer before the third step occurs. By doing so, it prevents the formation of the casting defect rather than trying to repair it after it has formed. This preventive philosophy is, in my view, the most important contribution of the present work.
I also compared the electrolytic process with mechanical methods of stress relief. Mechanical polishing or grinding can remove the surface layer, but it can also introduce new plastic deformation if the process is not carefully controlled. In contrast, electrolytic corrosion removes material atom by atom and does not impose mechanical forces. Therefore, it does not create a new deformed layer. This is a significant advantage for single-crystal components, where any new deformation can become a new recrystallization site. The chemical nature of the process also allows it to reach complex geometries that would be difficult to polish uniformly. For the turbine outer ring, the boss radius and the transition region are difficult to access with mechanical tools. The electrolyte can reach these areas and produce a uniform removal effect.
Another important observation is that the process improves the surface finish after over-etching is avoided. In the optimal window, the surface remains bright and smooth. The dendritic pattern of the cast structure is not excessively revealed. This is important because a rough surface can itself act as a stress concentrator and can complicate subsequent inspection. The optimal condition therefore produces a surface that is both metallurgically clean and dimensionally acceptable. I found that the 8 to 10 min window consistently produced this result in repeated trials.
| Method | Mechanism | Risk of new deformation | Suitability for complex geometry | Effect on recrystallization |
|---|---|---|---|---|
| Mechanical polishing | Abrasive removal | High if force is uncontrolled | Moderate | Can reduce but may introduce new sites |
| Grinding | Cutting and abrasion | High | Low for recessed areas | Often increases risk |
| Electrolytic corrosion | Anodic dissolution | Very low | High | Effectively removes the stressed layer |
In terms of process windows, I found that the current density is a useful parameter to monitor. The current density \(J\) is defined as
$$
J = \frac{I}{A}
$$
where \(I\) is the current and \(A\) is the effective treated area. For a fixed current, the current density depends on the area exposed to the electrolyte. Masking reduces the exposed area and increases the local current density. Therefore, the masking pattern must be designed carefully to avoid excessive local dissolution. In my experiments, the masking was arranged so that the boss radius received a controlled current density while the surrounding surfaces were protected. This helped to achieve uniform removal and to avoid the formation of sharp steps at the boundary between masked and unmasked regions.
The temperature of the electrolyte is another variable that can influence the removal rate. In general, a higher temperature increases the reaction rate and the conductivity. However, it can also increase the risk of uneven etching and fume generation. In this study, I performed the treatment at ambient temperature and found that the process was sufficiently fast and controllable. For industrial production, a cooling system or temperature controller may be useful to maintain consistency across batches. The important point is that the process window should be established with respect to the actual electrolyte temperature and composition, not only the applied voltage and time.
The surface condition before electrolysis also matters. If the casting has a heavy oxide scale or a thick layer of residual ceramic, the electrolyte may not reach the metal uniformly. Therefore, I recommend a preliminary cleaning step to remove loose scale and contamination. In my trials, a simple rinse and light cleaning were sufficient. The insulating tape adhered well to the cleaned surface and provided reliable masking. After electrolysis, the tape was removed and the casting was rinsed thoroughly. Any residual electrolyte was neutralized and washed away to prevent corrosion during storage or heat treatment.
The statistical results can be further analyzed by considering the probability of recrystallization as a function of removal depth. If I assume that recrystallization occurs when the local stored energy exceeds a critical value, then the probability of survival can be written as
$$
P_s = \exp\left(-\lambda(d)\right)
$$
where \(P_s\) is the probability that a given region remains free of recrystallization, and \(\lambda(d)\) is the expected number of recrystallization nuclei at removal depth \(d\). As \(d\) increases, \(\lambda(d)\) decreases because the stored-energy layer is removed. The observed data are consistent with this trend. At 3 and 5 min, \(\lambda\) is large, and many grains are observed. At 8 and 10 min, \(\lambda\) is small, and only a few spots remain. At 13 and 15 min, \(\lambda\) is effectively zero, but the surface damage becomes unacceptable. The optimum is therefore the smallest removal depth that makes \(\lambda\) acceptably small while preserving surface quality.
| Removal depth (mm) | Qualitative nucleation intensity | Observed recrystallization response | Surface condition |
|---|---|---|---|
| 0.010 | High | Many large grains | Bright |
| 0.020 | High | Many large grains | Bright |
| 0.035 | Low | Few small spots | Bright, rounded |
| 0.040 | Low | Few small spots | Bright, rounded |
| 0.060 | Negligible | None detected | Dark, over-etched |
| 0.080 | Negligible | None detected | Dark, severe relief |
The design of the casting itself also influences recrystallization. I recommend reducing the boss height and increasing the transition radius at the design stage. A lower boss reduces the thermal mass difference and the contraction mismatch. A larger radius reduces the stress concentration factor. These changes are complementary to the electrolytic process. If the geometry is improved, the depth of the damaged layer may be reduced, and the electrolytic process can be operated with a smaller removal depth. This reduces the risk of over-etching and improves dimensional control. In my view, the best solution combines robust design with a controlled surface treatment.
| Design variable | Current condition | Recommended direction | Expected effect |
|---|---|---|---|
| Boss height | High height-to-diameter ratio | Reduce height moderately | Lower thermal stress and stored energy |
| Transition radius | Sharp or small radius | Increase radius | Lower stress concentration |
| Wall thickness change | Abrupt section changes | Smooth transitions | More uniform temperature field |
| Window depth | Deep central window | Optimize core and mold design | Reduce local constraint and stress |
I also considered the possibility of applying the electrolytic process to other single-crystal components with similar recrystallization problems. The principle is general: identify the surface region with the highest stored energy, remove a controlled amount of material, and round the stress concentrators before high-temperature heat treatment. The electrolyte composition and the process parameters may need to be adjusted for different alloys and geometries, but the underlying strategy remains valid. This is why I believe the method can be extended beyond the turbine outer ring to other hot-section components where recrystallization is a persistent casting defect.
The economic benefit of the process is also significant. Single-crystal castings are expensive because they require directional solidification, precise mold preparation, and careful heat treatment. A high rejection rate at the final inspection stage wastes all of the upstream value. By reducing the recrystallization-related rejection rate from about 70% to below 5%, the process substantially lowers the effective cost per acceptable part. It also reduces the lead time because fewer parts need to be remade. In a production environment, this improvement can be more important than a small increase in processing cost. The electrolytic step is relatively inexpensive compared with the cost of scrapping a completed casting.
| Economic factor | Before process implementation | After process implementation |
|---|---|---|
| Recrystallization rejection rate | Approximately 70% | Below 5% |
| Effective yield of acceptable castings | Low | High |
| Remanufacturing burden | High | Low |
| Process cost | No electrolytic step | Modest additional cost |
| Overall production stability | Variable | More consistent |
In terms of quality control, I recommend that the electrolytic process be monitored by measuring the removal depth on a witness sample or by using a calibrated removal rate. The current and time should be recorded for each batch. The electrolyte composition should be checked periodically because the concentration of metal ions increases with use and the pH may change. Filtration removes solid debris and helps to maintain uniform dissolution. The masking should be inspected before immersion to ensure that no functional surface is exposed. After treatment, the casting should be rinsed and dried promptly to prevent staining or corrosion. These practical measures make the process reliable and repeatable.
The recrystallization defect is particularly insidious because it may not be visible immediately after casting. It often appears only after heat treatment, when the component has already passed through many value-adding steps. Therefore, early detection is difficult, and prevention is the best strategy. The electrolytic process provides a preventive route. It is applied before the critical high-temperature step and therefore removes the cause before the defect can form. In my experiments, the correlation between treatment time and final recrystallization was clear and reproducible. This gives confidence that the process can be used as a standard production step.
I also observed that the electrolytic treatment changed the surface topography in a beneficial way. The sharp edges at the boss radius became rounded, and the surface roughness remained low in the optimal window. The removal of burrs and micro-cracks reduced the number of potential nucleation sites. This is consistent with the idea that recrystallization begins at heterogeneities such as sharp corners, scratches, and inclusion particles. By removing these features, the treatment reduces the probability of nucleation. The surface becomes more uniform and less likely to initiate the casting defect during subsequent thermal exposure.
| Surface feature | Condition before treatment | Condition after optimal treatment | Effect on recrystallization |
|---|---|---|---|
| Boss transition radius | Sharp, with burrs | Rounded and smooth | Lower stress concentration |
| Surface deformation layer | Present | Removed | Lower stored energy |
| Micro-cracks | Possible at edges | Reduced or removed | Fewer nucleation sites |
| Surface roughness | Variable | Bright and uniform | More predictable behavior |
The mechanism of stress relief by material removal can be understood through the concept of a surface layer with a residual stress profile. If the residual stress is \(\sigma(x)\) and the layer extends from the surface to a depth \(h\), then the stored elastic energy per unit area is
$$
U = \frac{1}{2} \int_{0}^{h} \frac{\sigma(x)^2}{E} \, dx
$$
where \(E\) is the elastic modulus. When the electrolytic process removes the layer from 0 to \(d\), the remaining stored energy is
$$
U_{\text{remaining}} = \frac{1}{2} \int_{d}^{h} \frac{\sigma(x)^2}{E} \, dx
$$
If the removal depth \(d\) is close to the thickness of the highly stressed layer, the remaining energy is small. This is why a removal depth of only 0.035 to 0.040 mm is effective. The process does not need to remove the entire casting surface; it only needs to remove the region where the residual stress and stored dislocation energy are highest. The exact depth depends on the casting process and the surface preparation history, but the principle is general.
I also considered the effect of the electrolytic process on the gamma-prime precipitate morphology. Because the treatment is performed before solution treatment, the surface is subsequently exposed to the full solution and aging cycle. The precipitates that form after heat treatment are therefore not affected by the electrolytic step. The treatment does not alter the bulk alloy chemistry or the precipitation kinetics. It only changes the surface condition. This is important because the mechanical properties of the alloy depend on the gamma-prime distribution, and the process must not interfere with that distribution. In my study, the heat-treated microstructure appeared normal in the treated and untreated regions, except for the absence of recrystallized grains in the treated regions.
The relationship between the removal depth and the probability of recrystallization can also be expressed using a Weibull-type distribution. If the local stored energy is assumed to follow a distribution, the probability that a region survives without recrystallization is
$$
P_s(d) = \exp\left[-\left(\frac{d_0 – d}{d_c}\right)^m\right]
$$
where \(d_0\) is the initial thickness of the damaged layer, \(d_c\) is a characteristic depth, and \(m\) is a shape parameter. This type of expression captures the observation that the recrystallization probability drops rapidly once the removal depth approaches the damaged-layer thickness. It also explains why the process has a threshold behavior: shallow removal has little effect, while moderate removal produces a large improvement. The exact values of the parameters depend on the alloy, the casting geometry, and the surface preparation, but the qualitative trend is robust.
In conclusion, my study showed that electrolytic corrosion is an effective pre-heat-treatment process for controlling recrystallization in single-crystal turbine outer ring castings. The process removes the surface layer that contains stored deformation energy, blunts sharp corners, and reduces the stress concentration at the boss transition radius. The result is a significant reduction in the recrystallization casting defect. By optimizing the electrolysis time to 8 to 10 min, corresponding to a removal depth of 0.035 to 0.040 mm, I reduced the recrystallization defect rate from about 70% to below 5%. The process is compatible with the standard solution and aging treatments, does not harm the bulk single-crystal structure, and can be integrated into production with simple equipment and controls. I also recommend design improvements such as reducing boss height and increasing the transition radius to further reduce the driving force for recrystallization. Together, these measures provide a practical route to higher single-crystal integrity, higher product qualification rate, and more reliable batch production of high-performance engine components.
| Conclusion area | Key finding | Practical recommendation |
|---|---|---|
| Process effectiveness | Electrolytic corrosion removes the stress layer and reduces recrystallization | Apply before solution treatment |
| Parameter control | Removal depth is the critical variable | Use 8–10 min, 0.035–0.040 mm |
| Over-treatment risk | Long treatment causes surface damage | Avoid times above 10 min |
| Design optimization | Geometry controls stress concentration | Reduce boss height and increase radius |
| Production impact | Rejection rate falls from 70% to below 5% | Integrate as a standard production step |
$$
\text{Overall benefit} = \frac{\text{acceptable single-crystal castings}}{\text{total castings produced}} \times 100\%
$$
The overall benefit equation summarizes the practical goal of the work. By reducing the recrystallization casting defect, the numerator increases and the overall benefit rises. The electrolytic corrosion process is a controlled, repeatable, and relatively low-cost method to achieve this improvement. I believe that the approach can be applied to other single-crystal superalloy components, especially those with complex geometry and severe thermal gradients. The key is to understand the formation mechanism, identify the critical surface region, and remove the stored-energy layer before the high-temperature heat treatment that would otherwise trigger recrystallization.