EP2425910A1

EP2425910A1 - Method for producing salt core for casting

Info

Publication number: EP2425910A1
Application number: EP10769833A
Authority: EP
Inventors: Koichi Anzai; Katsunari Oikawa; Youji Yamada
Original assignee: Tohoku University NUC; Yamaha Motor Co Ltd
Current assignee: Buehler AG
Priority date: 2009-05-01
Filing date: 2010-04-30
Publication date: 2012-03-07
Also published as: WO2010126135A1; JP5419549B2; JP2010279951A; US20120048502A1

Abstract

First, a melt is formed by heating a salt mixture (S101). Next, a mold for core molding is heated to a temperature higher than 0.52 × Tm and lower than 0.7 × Tm (S102). Note that Tm represents the liquidus temperature of the salt mixture as an absolute temperature (K). Then, the melt is poured under pressure into the mold heated as described above (S103). A salt core for casting is molded by solidifying the melt inside the mold (S104).

Description

Technical Field

The present invention relates to a method for producing a water-soluble salt core for casting.

Background Art

Aluminum alloy die casting has features such as a light weight, high strength, high productivity, and high dimensional accuracy, and hence is widely used in applications such as automobile industries, and the production amount of aluminum alloy die casting is increasing. Recently, demands have arisen for applications to parts having more complicated shapes. For example, demands have arisen for applications to, e.g., water-cooling water jackets such as a cylinder block of an internal-combustion engine. However, it is difficult for aluminum alloy die casting to mold a hollow structure like this, and a cast product having an undercut shape.
Molding of a hollow structure or undercut shape requires a expendable core removable after casting. However, a core for use in die casting must be a high-strength core that can withstand the impact force of a melt injected at high speed and . More specifically, it is important to withstand a melt impact of a gate velocity of 38.7 m/s as a recommended value of NADCA (The North American Die Casting Association), or a melt impact of a gate velocity of 25.4 to 40.6 m/s recommended by the PQ2 manual of NADCA (see non-patent literatures 1 and 2). Also, a condition required of a core for die casting is that it does not deform at an injection pressure of 75 MPa or more, in order to smash cavities readily formed on the surface of a heat insulating core due to blowholes.
As is well known, however, the strength and expendability of a general sand core are conflicting characteristics. Therefore, it is regarded as important to optimize the strength and expendability by adjusting the type and amount of binder (see non-patent literatures 3 and 4).
On the other hand, the application of a salt is also being examined. A salt is water-soluble and readily removable by high-speed, running water. Also, a core using a salt (a salt core) does not significantly lose the removability in most cases even when strengthened. From the foregoing, a salt core is regarded as suitable for the die casting process. A salt is a brittle material like ceramics, and structure control such as increasing the density or downsizing the crystal grains is effective in increasing the strength (see non-patent literatures 5 and 6). Recently, a salt core dispersion-strengthened by Mullite has been reported (see non-patent literature 7).
The present inventors have systematically studied, e.g., the mechanical properties and microstructure of a melt-molded salt, in order to apply a salt core to aluminum die casting (see non-patent literatures 8 to 12). For example, aluminum borate whisker is effective to dispersion-strengthen a salt core using an alkali chloride (see non-patent literature 8). Also, the present inventors have revealed that a salt mixture of an alkali chloride and alkali carbonate shows a very high strength of 20 to 30 MPa without any reinforcing material such as whisker (see non-patent literature 10). Furthermore, the present inventors have revealed conditions under which a KC1-NaCl-K₂CO₃-Na₂CO₃ multi-element salt mixture becomes a high-strength material, by comparing a computational phase diagram with the microstructure (see non-patent literature 11). These high-strength salt mixtures have been produced by gravity casting. When these high-strength salt mixtures are cast and molded by die casting, however, the improvements of the dimensional accuracy and productivity of cores can be expected.

Prior Art Literatures

Non-Patent Literatures

Non-Patent Literature 1: E.A. Herman NADCA Gating Die Casting Dies Pub. #E-514 (1996), 29.
Non-Patent Literature 2: J. Wronowicz, M. Cox and R. Fish NADCA PQ2 EC700 (July 2006), 8.
Non-Patent Literature 3: T. Manabe, M. Nitta and M. Yaguchi: SOKEIZAI, 44(2003) 12, 26.
Non-Patent Literature 4: T. Komasaki, T. Miyamoto and M. Nitta: J. JFS, 78(2006), 533.
Non-Patent Literature 5: H.G. Muller: Z. Physik, 96(1935), 321.
Non-Patent Literature 6: T. Sata: J. Ceram. Soc. Jpn, 107(1999), 166.
Non-Patent Literature 7: C. Hayashi, T. Yamazaki, T. Ishikuro and A. Urakami: ALUTOPIA, 35(2006)6, 22.
Non-Patent Literature 8: J. Yaokawa, K. Anzai, Y. Yamada, H. Yoshii and H. Fukui: J. JFS, 76(2004), 823.
Non-Patent Literature 9: J. Yaokawa, T. Sawada, K. Anzai, Y. Yamada, H. Yoshii and H. Fukui: J. JFS, 78(2006), 59.
Non-Patent Literature 10: J. Yaokawa, D. Miura, K. Anzai, Y. Yamada and H. Yoshii: J. JFS, 78(2006), 516.
Non-Patent Literature 11: J. Yaokawa, D. Miura, K. Oikawa, K. Anzai, Y. Yamada and H. Yoshii: J. JFS, 79(2007), 184.
Non-Patent Literature 12: Y. Yamada, J. Yaokawa, H. Yoshii, K. Anzai, Y. Noda, A. Fujiwara, T. Suzuki and H. Fukui: 20076584(JSAE).
Non-Patent Literature 13: J. Yaokawa, K. Oikawa and K. Anzai: CALPHAD, 31(2007), 155.
Non-Patent Literature 14: R. Danzer, T. Lube, P. Supancic and R. Damani: Adv. Eng. Mater., 10(2008), 275.
Non-Patent Literature 15: A.G. Evans and T.G. Langdon: Progress in Mater. Sci., 21(1976), 171.

Summary of the Invention

Problem to be Solved by the Invention

Unfortunately, salt cores formed by die casting have variations in strength and hence have not completely put to practical use yet.
The present invention has been made to solve the problem as described above, and has its object to make it possible to more stably obtain a practical strength of a salt core for casting molded by die casting by melting a salt such as sodium.

Means of Solution to the Problem

A method for producing a salt core for casting according to the present invention is a method for producing a salt core for casting formed by a molten salt made of a salt mixture containing sodium salt, comprising at least a first step of forming a melt by heating the salt mixture, a second step of heating a mold for core molding to a temperature higher than 0.52 × Tm and lower than 0.7 × Tm where Tm is a liquidus temperature of the salt mixture as an absolute temperature (K), a third step of pouring the melt into the heated mold under pressure, and a fourth step of molding a salt core for casting by solidifying the melt inside the mold.
In the above-mentioned method for producing a salt core for casting, the mold need only be heated to a temperature of 225°C to 250°C. Also, the salt mixture preferably has a composition in which Na⁺ : K⁺ = 70 mol% : 30 mol%, and Cl^- : CO₃ ^2- = 46.2 mol% : 53.8 mol%.

Effects of the Invention

In the present invention as explained above, letting Tm be the liquidus temperature of a salt mixture as an absolute temperature (K), a mold for core molding is heated to a temperature higher than 0.52 × Tm and lower than 0.7 × Tm, and the melt of the salt mixture is poured into the heated mold under pressure. Accordingly, it is possible to more stably obtain a practical strength of a water-soluble salt core for casting made of a cast product obtained by salt die casting by which a salt such as sodium is melted and molded.

Brief Description of Drawings

Fig. 1 is a flowchart for explaining a method for producing a salt core for casting according to an embodiment of the present invention;
Fig. 2 is a photograph showing the state of a molded product for obtaining a specimen produced by an experiment according to the embodiment of the present invention;
FIG. 3 is a graph showing examples of typical pressure changes in a runner portion and specimen portion during a die casting process;
FIG. 4 is a view showing the shape of the specimen produced by the experiment according to the embodiment of the present invention;
FIG. 5 is a correlation diagram showing the relationship between the average bending strength and mold temperature;
Fig. 6 is a graph plotting the average bending strengths of samples molded at mold temperatures of 175°C and 250°C as a function of the injection pressure;
Fig. 7 is a graph showing the relationship between the hardness of a specimen and the mold temperature;
Fig. 8 shows the results of the visual observation of surface defects of specimens performed by dye penetrant inspection;
Fig. 9 shows photographs each showing an SEM image of the bottom surface of a specimen;
Fig. 10 shows photographs each showing an SEM image of the sample surface of a section perpendicular to the longitudinal direction, or an SEM image of the center of the interior;
Fig. 11 shows Weibull plots indicating the relationship between the cumulative fracture probability and bending stress in a bending test of a sample of this embodiment;
Fig. 12 is a perspective view showing the arrangement of a closed deck type cylinder block produced by aluminum alloy die casting;
Fig. 13 is a correlation diagram showing the relationship between the average bending strength and Weibull coefficient of each sample;
Fig. 14 shows photographs showing the results of the visual observation of surface defects of specimens having composition A performed by dye penetrant inspection;
Fig. 15 shows photographs showing the results of the visual observation of surface defects of specimens having composition B performed by dye penetrant inspection;
Fig. 16 shows photographs showing the results of the visual observation of surface defects of specimens having composition C performed by dye penetrant inspection; and
Fig. 17 shows photographs showing the results of the visual observation of surface defects of specimens having composition D performed by dye penetrant inspection.

Best Mode for Carrying Out the Invention

Embodiments of the present invention will be explained below with reference to the accompanying drawings. Fig. 1 is a flowchart for explaining a method for producing a salt core for casting according to an embodiment of the present invention. First, in step S101, a melt is produced by heating a salt mixture. Then, in step S102, a mold for core molding is heated to a temperature higher than 0.52 × Tm and lower than 0.7 × Tm. Note that Tm represents the liquidus temperature of the salt mixture as an absolute temperature (K). Subsequently, in step S103, the melt is poured under pressure into the mold heated as described above. In step S104, a salt core for casting is molded by solidifying the melt inside the mold.
The salt mixture is obtained by mixing, e.g., 50 mol% of Na₂CO₃, 20 mol% of NaCl, and 30 mol% of KCl. When the composition of this salt mixture is analyzed, the ion composition ratios are Na⁺ : K⁺ = 80 mol% : 20 mol%, and Cl^- : CO₃ ^2- = 50 mol% : 50 mol%. In this case, the mold need only be heated to a temperature of 225°C to 250°C. In this embodiment as described above, the variations in strength of obtained salt cores are reduced, and a higher strength can be obtained.
The present invention will be explained in more detail below by way of its experimental examples.
That is, a KCl-NaCl-Na₂CO₃-based salt mixture was molded by die casting, and the influences of the mold temperature and the injection pressure of the melt on the bending strength were inspected. In addition, a salt core for a single-cylinder, closed-deck type cylinder block was manufactured on an experimental basis as an application example. The results will be explained below.

[Experimental Method]

[Production of Samples]

Nacl, KCl, and Na₂CO₃ having a purity of 99.5% were melted in an alumina crucible by using a resistance furnace. The melting ambient was the atmosphere, and the melting temperature was 688°C. The melted sample composition was 50mol%Na₂CO₃-20mol%NaCl-30mol%KCl reportedly having high strength. According to a computational phase diagram (see non-patent literature 13), the primary crystal was a carbonate having a high sodium ion concentration, and a eutectic reaction occurred between the carbonate and a chloride at a temperature equal to or lower than the eutectic temperature. The liquidus temperature, eutectic start temperature, and eutectic end temperature were respectively found to be 638°C, 574°C, and 573°C by calculations. Also, the degree of superheat during melting was about 50°C (688°C).
A cold chamber type machine having a clamping force of 110 tons was used as a die casting machine, and casting was performed using a mold capable of molding two rectangular specimens as shown in Fig. 2. The casting conditions were that the mold temperatures were 175°C, 200°C, 225°C, and 250°C, and the injection pressures were 39.2, 58.8, and 78.4 MPa, and these conditions were systematically changed. Molding was performed in order from the lowest mold temperature and lowest injection velocity. Other molding conditions were a sleeve diameter of 50 mm, an injection velocity of 34 mm/sec, and a filling rate of about 60%. Since the filling rate was low, the amount of air involvement was presumably small, so no two-step injection as performed by general die casting was performed.
Also, to remove the chilled layer (cold flakes) generated on the sleeve surface, sliding cores(partition) was inserted behind (in the sprue of) the sleeve. If the cold flakes generated in the sleeve entered the mold, the bending strength of molded products largely varies. To prevent the cold flakes from entering the mold, a weir projecting toward the circumferential edge of the sprue of the sleeve was inserted into the sprue. The insertion of the weir makes it possible to suppress the variations in bending strength of molded products.
Note that the mold temperature was measured by inserting a thermocouple in each of the upper and lower molds between the two specimens, and controlled based on the temperature measurement results. Note also that the sleeve temperature was controlled to be equal to the mold temperature.
Fig. 3 shows examples of typical pressure changes in a runner portion and specimen portion during the die casting process. After a molten salt was completely filled, a pressure applied by a plunger became maximum during holding the pressure, and gradually decreased. In this state, the maximum values in the runner portion and specimen portion were almost the same, so the plunger pressure was probably sufficiently transmitted to the specimen portion at the time of solidification.

[Testing Method]

Next, a testing method will be explained. The diecast material cast as described above was cut from the gate portion, thereby obtaining a specimen having a shape as shown in Fig. 4. The mechanical strength of the specimen obtained as described above was evaluated by the four point bending test. The cross head speed of a bending test machine was 1 mm/min, and the support points diameter was 4 mm. Bending strength σ was obtained by equation (1) below from a maximum load F [N] when the specimen broke while a force was applied. $\begin{array}{l} [Equation 1] \\ σ = \frac{3 (L_{1} - L_{2}) F}{2 {WT}^{2}} \end{array}$
In equation (1), L₁ is a lower support points spacing, and 50 mm. L₂ is an upper support points spacing, and 10 mm. W is the width of the specimen, and 20 mm. T is the height of the specimen, and 18 mm. About 10 to 20 specimen were produced under each of the above-described casting conditions, and the average bending strength was calculated for each condition.
Also, a portion of the bending specimen was cut, and surface observation and microstructure observation were performed on the cut surface. When performing microstructure observation, the section was dry-polished with #4,000 water-resistant emery paper, and cleaned with ultrasonic waves in acetone. After that, vacuum carbon deposition was performed on the section, and the section was observed with a scanning electron microscope (SEM). In addition, a micro Vickers hardness test was conducted. The testing conditions were a load of 4.9 N, and a load holding time of 30 sec. The surface and center were measured 10 times each, and the average value of the measurement results of each of the surface and center was calculated. Furthermore, surface defects were visually observed by dye penetrant inspection.

[Results]

[Bending Test]

The results of the bending test will now be explained. Fig. 5 is a correlation diagram showing the relationship between the average bending strength and mold temperature. Referring to Fig. 5, an error bar indicates a standard deviation. At each injection pressure, the average bending strength tended to increase as the mold temperature increased when the mold temperature was 175°C to 225°C. The bending strengths of materials cast at mold temperatures of 225°C and 250°C had a slight difference, but this difference was insignificant, so the values were almost equal. When the mold temperatures were 225°C and 250°C, the average bending strength exceeded 25 MPa and exhibited a high value almost equal to that of a gravity cast material at each injection pressure (see non-patent literature 9).
Fig. 6 is a graph plotting the average bending strengths of samples molded at mold temperatures of 175°C and 250°C as a function of the injection pressure. Although there was a slight variation, this variation can be regarded as an error range, so there was no large influence on the bending strength within this injection pressure range. The results were the same when the mold temperatures were 200°C and 225°C.

[Hardness Test]

Next, the results of the hardness test will be explained. As described above, the injection pressure had almost no influence on the bending strength. Therefore, the Vickers hardness was mainly checked for samples molded at an injection pressure of 78.4 MPa. Fig. 7 is a graph showing the relationship between the hardness of a specimen and the mold temperature. On the surface of a sample (solid circles), there was a slight variation, but the hardness did not largely depend on the mold temperature and almost remained the same. On the other hand, in the center of the interior of the sample (solid triangles), the hardness was highest when the mold temperature was lowest, i.e., 175°C. When the mold temperature was 175°C, there was no large difference in hardness between the surface and interior. By contrast, the surface of a sample was obviously harder than the center of its interior at other mold temperatures.

[Surface and Microstructure Observation]

The results of surface and microstructure observation will be explained below. Dye penetrant testing and structure observation by an SEM were also mainly performed on samples molded at an injection pressure of 78.4 MPa. Fig. 8 shows the results of the visual observation of surface defects of specimens performed by dye penetrant inspection. Referring to Fig. 8, a colored portion is a cracked portion. When the mold temperature was lowest, i.e., 175°C, many large cracks were found on the surface as indicated by (a) in Fig. 8. Large cracks were also observed in a sample molded at a mold temperature of 200°C as indicated by (b) in Fig. 8, although these cracks were not so large as those formed at 175°C. On the other hand, when the mold temperatures were 225°C and 250°C, almost no clear cracks were observed on the surfaces as indicated by (c) and (d) in Fig. 8. When the molten salt solidifies, a solidification shell generated on the mold surface in the initial stages acts to shrink during solidification, but cannot shrink because a pressure is applied, thereby generating a tensile stress on the surface. When the mold temperature is low, the tensile stress perhaps forms surface cracks because the solidification shell cannot deform.
Fig. 9 shows SEM images of the bottom surfaces of specimens. When the mold temperature was lowest, i.e., 175°C, patterns like unevenness were observed as indicated by (a) in Fig. 9, and microcracks were formed near the patterns. As the mold temperature rose to 200°C, 225°C, and 250°C, the patterns like unevenness significantly reduced when compared to the 175°C sample. The number of microcracks also decreased as the mold temperature rose. Furthermore, there was a region on the surface of which micro pits of a few micron size were found. These micro pits were presumably formed by deliquescence. These surface defects probably have large influence on the bending strength. The patterns like unevenness were perhaps caused by, e.g., the surface quenching effect during solidification.
Fig. 10 shows SEM images each showing the close to the surface portion of a sample in a section perpendicular to the longitudinal direction, or the close to the center portion of the interior of a sample. When the mold temperature was 175°C, there was almost no difference in microstructure between the surface and center. When the mold temperatures were 200°C, 225°C, and 250°C, however, coarse primary dendrite was clearly observed in the interior, when compared to the surface. Also, the surface structure did not largely change due to the mold temperature. The structure near the surface was presumably a chilled layer made of a solidification shell rapidly formed on the mold surface in the initial stages of solidification. The coarseness of the dendrite increased probably because the solidification speed of the interior decreased as the mold temperature rose. Furthermore, the tendency of the hardness matches that of the coarseness of the structure.

[Consideration]

The above-described experimental results reveal that the average bending strength increases as the mold temperature rises. The experimental results also indicate that the change in bending strength is associated with, e.g., cracks on the surface. To examine the change in bending strength in more detail, the bending strength was analyzed by Weibull statistics used in the field of brittle materials (see non-patent literature 12).
To increase the accuracy of analysis, the number of times of the bending test was increased to 20. Also, the analysis was performed under two conditions, i.e., (condition 1) in which the injection pressure was 74.8 MPa and the mold temperature was 250°C, and (condition 2) in which the injection pressure was 74.8 MPa and the mold temperature was 200°C. The former is an example in which the mold temperature and average bending strength are high. The latter is an example in which the mold temperature and average bending strength are low.
A two-parameter Weibull distribution function used for a brittle material is given by equation (2) below (see non-patent literature 12). $\begin{array}{l} [Equation 2] \\ F (σ) = 1 - \exp (- V_{E} {(\frac{σ_{f}}{σ_{0}})}^{m}) \end{array}$
In equation (2), F(σ) is the cumulative fracture probability, V_E is the effective volume, σ₀ is a scale parameter, σ_f is the bending stress, and m is a shape parameter. The larger the shape parameter m, the smaller the variation in strength, and the higher the reliability. The value of m is about 5 to 20 for general ceramics. F(σ) was obtained from experimental data σ_i by using the mean rank method. The cumulative fracture probability F(σ_i) for each individual data is given by equation (3) below (see non-patent literature 23) . $\begin{matrix} [Equation 3] \\ F (σ_{i}) = \frac{i}{N + 1} \end{matrix}$
In equation (3), N is the number of data, and i is the order of data.
Fig. 11 shows the Weibull plots of the cumulative fracture probability and bending stress in the above-described sample bending test. Under condition 1 (solid squares) indicating high strength, the Weibull plot is a straight line, demonstrating that the sample perhaps fractured from one type of a fracture source. Also, when the Weibull plot was analyzed by the least square method, m = 8.49, i.e., a high value was obtained. As an example of the m value of a salt, m = 5.34 (see non-patent literature 4) obtained by a bending test of a sintered material of NaCl has been reported. Even when compared with this value, the reliability is regarded as high.
On the other hand, under condition 2 (hollow circles) indicating low strength, the Weibull plot bends near 16 MPa, indicating that the sample obviously fractured from a plurality of fracture sources. Assuming that the fracture sources were two types, least square analysis was performed in regions I and II as shown in Fig. 11. The m value was 1.04 in region I, and 6.75 in region II. In region I, the m value was significantly small, indicating that the reliability was low. In region II, the m value was relatively large, so the reliability was presumably high.
Generally, the strength of a brittle material such as ceramics is dominated by a defect and the fracture toughness value, and the relationship between them is given by equation (4) below. $\begin{matrix} [Equation 4] \\ σ_{c} = K_{IC} / \sqrt{πc} \end{matrix}$
In equation (4), K_IC is the fracture toughness value, σ_c is the stress caused by fracture, and c is the depth of a crack. Since the tensile stress is applied to the surface in the bending test, the bending strength is presumably dominated by the fracture toughness value and a defect near the surface. The fracture toughness value K_IC is a value determined by the elastic constant at the distal end of the crack and the crack propagation energy. Since the mold temperature has no large influence on the microstructure and hardness of the surface, the K_IC probably changes little due to the mold temperature. Accordingly, a surface defect is perhaps a dominant cause of the change in bending strength resulting from the mold temperature.
Under condition 2, as shown in Fig. 8, irregular cracks having a depth of a few mm or more were observed by dye penetrant testing. These cracks were presumably fracture sources in region I where the reliability and strength were low. In region II in which the shape parameter value m was high, the bending strength was also high, and the depth of a crack as a fracture source was smaller. A possible fracture source like this is a microcrack observed by an SEM. The depth of this microcrack is at most a few ten µm. In region II, therefore, when no large crack clearly observable by dye penetrant testing existed between the support points, fracture perhaps occurred from a microcrack as a fracture source.
Even when the bending strength was high (condition 1), microcracks were observed although the number was small, and the observed microcracks were probably fracture sources in the bending test. As the size of these microcracks decreases, the bending strength perhaps increases. Accordingly, it is important to examine the relationship between the microcrack and mold temperature in more detail in the future.

[Application to Core for Cylinder Block]

The object of the present invention is, e.g., an application to a closed deck type cylinder block manufactured by aluminum alloy die casting. For example, a single-cylinder cylinder block as indicated by (a) in Fig. 12 is a target. A water jacket portion has an undercut shape. As a core for forming this water jacket, a core for a water jacket portion of a single-cylinder cylinder block was successfully produced by the die casting method of the present invention as indicated by (b) in Fig. 12. The surface was smooth, and there was no visible defect. Accordingly, the present invention is expected to be suitably applicable to aluminum alloy die cast products.

[Conclusions]

A salt core (casting salt core) for aluminum alloy die casting was produced by die casting, the influences of the injection pressure and mold temperature on the bending strength and Vickers hardness were checked, and surface defects and microstructures were observed. Consequently, the following conclusions were obtained.

(1) As the mold temperature rose, the bending strength increased. On the other hand, the bending strength did not largely change within the injection pressure range adopted in this study.
(2) When the mold temperature was low, large cracks were observed on the surface in dye penetrant testing, and these cracks probably significantly decreased the bending strength. On the other hand, surface unevenness and microcracks were found by surface observation using an SEM when the mold temperature was low, and they reduced when the mold temperature rose. These surface unevenness and microcracks were presumably start points of fracture in the bending test. When a molten salt solidifies, a solidification shell formed on the mold surface in the initial stages acts to shrink during the solidification, but cannot shrink because a pressure is applied, and a tensile stress is generated on the surface. When the mold temperature is low, the tensile stress generates surface unevenness perhaps because the solidification shell cannot deform.
(3) The analysis of the results of the bending test by Weibull statistics revealed that when the mold temperature rose, not only the bending strength but also the reliability of the bending strength increased.
(4) Based on the above-mentioned findings, a core for a water jacket of a single-cylinder closed deck type cylinder block was successfully produced.

The above-described rise in salt core strength caused by the mold temperature is presumably obtained by heating at a temperature higher than 0.52 × Tm and lower than 0.7 × Tm where Tm is the liquidus temperature (as an absolute temperature K) of a salt mixture. For example, as indicated by the above-described example, a practical bending strength can be obtained within the range of 225°C to 250°C for a salt mixture formed by mixing 50 mol% of Na₂CO₃, 20 mol% of NaCl, and 30 mol% of KCl. The liquidus temperature Tm of this salt mixture is 911 K (638°C), a temperature of 0.52 × Tm is about 474 K (201°C), and a temperature of 0.7 × Tm is about 638 K (365°C). "The range of 225°C to 250°C" falls within this range.
Next, the results of experiments conducted on casting cores produced by changing the composition ratio of a salt mixture will be explained.

[Samples]

First, samples will be explained. Samples having compositions having the following ion composition ratios were produced.
Composition A: Na⁺ : K⁺ = 70 mol% : 30 mol%, and Cl^- : CO₃ ^2- = 46.2 mol% : 53.8 mol%
Composition B: Na⁺ : K⁺ = 80 mol% : 20 mol%, and Cl^- : CO₃ ^2- = 45 mol% : 55 mol%
Composition C: Na⁺ : K⁺ = 80 mol% : 20 mol%, and Cl^- : CO₃ ^2- = 50 mol% : 50 mol%
Composition D: Na⁺ : K⁺ = 100 mol% : 0 mol%, and Cl^- : CO₃ ^2- = 50 mol% : 50 mol%

[Production of Samples]

Following the same procedures as described above, samples were produced by die casting from the salt mixtures having compositions A, B, C, and D. In this production, the injection pressure and mold temperature were changed.

[Testing Methods]

Following the same procedures as described above, the samples having the different compositions and produced under the different conditions were tested.

[Results]

[Bending Test]

A bending test was conducted in the same manner as described above. Table 1 below shows the results of the bending test conducted on samples having the different compositions and produced by changing the injection pressure. For these samples, the mold temperature was 250°C. Also, Table 2 below shows the results of the bending test performed on samples having the different compositions and produced by changing the mold temperature. For these samples, the injection pressure was 78.4 MPa. Note that the results of the samples shown in Tables 1 and 2 indicate the average bending strength (MPa).

[Table 1]

Table 1 Injection Pressure & Average Bending Strength (MPa)

Composition	Injection Pressure (MPa)
Composition	39.2	58.8	78.4
A	27.95	27.97	30.67
B	24.37	25.58	29.74
C	25.76	27.27	31.42
D	20.98	23.52	23.68

[Table 2]

Table 2 Mold Temperature & Average Bending Strength (MPa)

Composition	Mold Temperature (°C)
Composition	175	200	225	250	265
A	----	32.75	26.8	30.67	30.68
B	13.16	29.74	28.34	----	----
C	6.98	14.56	27.34	31.42	----
D	11.81	14.80	18.41	23.68	----

As shown in Table 1, the higher the injection pressure, the higher the strength. Also, as shown in Table 2, the strength increased as the mold temperature rose regardless of the composition. For composition A, the strength was in the order of the latter halh 20MPa or above, i.e., satisfied a minimum requirement (15 MPa) for the strength of a core for aluminum die casting, over the entire tested mold temperature range of 200°C to 265°C. For composition B, the strength was in the order of the latter halh 20MPa or above, i.e., satisfied the above-described minimum requirement (15 MPa), at 200°C and 225°C or more except for 175°C. For composition C, the strength was 20 MPa or more, i.e., satisfied the above-described minimum requirement at 225°C and 250°C or more except for 175°C and 200°C. For composition D, the strength satisfied the above-described minimum requirement at 225°C and 250°C or more except for 175°C and 200°C. For compositions C and D, the strength was slightly lower than the minimum requirement at 200°C. These results reveal that the strength probably meets the minimum requirement (15 MPa) regardless of the composition when the mold temperature is 201°C or more.
Table 3 shows the results of bending strength analysis performed for each composition by Weibull statistics. Note that the numerical values in Table 3 are Weibull coefficients.

[Table 3]

Table 3 Mold Temperature & Weibull Coefficient

Composition	Mold Temperature (°C)
Composition	175	200	225	250	265
A	----	4.50	20.02	16.188	5.69
		(1.33)
B	15.45	5.85	----	----	----
	(0.55)
C	----	6.00	----	8.59	----
		(1.05)
D	1.81	3.81	4.56	6.22	----
		(2.91)		(4.13)

As shown in Table 3, the higher the mold temperature, the larger the Weibull coefficient. For composition A, the Weibull coefficient was low at a mold temperature of 265°C. This is so perhaps because a high mold temperature caused burning the mold, and defects occurred on the surface of the formed sample.
For composition A at 225°C and 250°C, the Weibull coefficient was large, the reliability was high, and no surface defect occurred. For composition A at 200°C, two Weibull coefficients existed, and they were small, so the reliability was low.
For composition B at 200°C, the Weibull coefficient was m = 5.85, i.e., smaller than those of composition A at 225°C and 250°C. For composition B at 175°C, the reliability was low because two Weibull coefficients existed.
For composition C at 250°C, the Weibull coefficient was m = 8.59, i.e., smaller than that of composition A, but the reliability was very high. For composition C at 200°C, the reliability was low because two Weibull coefficients existed.
For composition D at 250°C and 200°C, two Weibull coefficients existed, so the reliability was low. For composition D at 225°C, the Weibull coefficient was m = 4.56, i.e., smaller than that of composition A, and the reliability was low. For composition D at 175°C, the Weibull coefficient was small, i.e., m = 1.81.
Fig. 13 shows the relationship between the average bending strength (Table 2) and the Weibull coefficient (Table 3) of each sample. As the material of a salt core formed by die casting, a composition in the upper right corner (the region enclosed with an ellipse) of Fig. 13 is presumably optimum. The composition in the upper right corner is composition A. From the viewpoint of the average strength and Weibull coefficient, the optimum composition is composition A, and the optimum conditions are a casting pressure of 78.4 MPa, and mold temperatures of 250°C and 225°C.
As explained above, the strength and reliability increased as the mold temperature rose for all the compositions. When the mold temperature was 265°C, however, no high strength was obtained in some cases due to seizing with the mold. Also, the strength and reliability were particularly high for composition A within the above-described composition range.

[Penetrant Testing]

The results of penetrant testing conducted on the samples having the different compositions will now be explained. Figs. 14, 15, 16, and 17 show the results of visual observation of surface defects performed on specimens having compositions A, B, C, and D by penetrant testing. In these drawings, colored portions indicate cracked portions. Also, each numeral (°C) in the drawings indicates the mold temperature. Table 4 below shows rough degrees (evaluation results) of defect amount of compositions A, B, C, and D based on the observation results shown in Figs. 14 to 17.

[Table 4]

Table 4 Degrees of Defect Amount

Mold Temperature (°C)	175	200	225	250	265
A(Solidification Zone 46°C )	-	×	Δ	⊚	⊚
B(Solidification Zone 78°C)	×	⊚	⊚	-	-
C(Solidification Zone 65°C)	×	Δ	⊚	⊚	-
D(Solidification Zone 12°C)	×	×	Δ	Δ	-

×...many large cracks △...relatively large cracks
⊚...almost no cracks

As shown in Figs. 14 to 17, crack defects on the surface reduced as the mold temperature rose for all the compositions. For composition A, many surface cracks were found when the mold temperature was 200°C, and almost no surface cracks were found when the mold temperature was 250°C or more. Similarly, almost no surface cracks were found for composition B when the mold temperature was 200°C or more, and for composition C when the mold temperature was 225°C or more. By contrast, defects remained for composition D even when the mold temperature was 250°C. To eliminate all surface defects, therefore, a higher mold temperature is probably necessary. These results reveal that surface defects reduce when the mold temperature is raised. This is so perhaps because the tensile stress generated when the mold temperature is low reduces, as described previously.
Also, some compositions may have mold temperature thresholds at which many surface cracks are formed. As shown in Table 4, the amount of defects (surface cracks) and the solidification zone may have a correlation. Table 4 shows that a mold temperature at which surface cracks almost disappeared rose in the order of B, C, A, and D as the composition conditions, and this order is regarded as the order of the difficulty of crack formation. Furthermore, this order probably matches the order of the width of the solidification zones.

Claims

A method for producing a salt core for casting formed by a molten salt made of a salt mixture containing sodium salt, comprising at least:
a first step of forming a melt by heating the salt mixture;

a second step of heating a mold for core molding to a temperature higher than 0.52 × Tm and lower than 0.7 × Tm where Tm is a liquidus temperature of the salt mixture as an absolute temperature (K);

a third step of pouring the melt into the heated mold under pressure; and

a fourth step of molding a salt core for casting by solidifying the melt inside the mold.
A method for producing a salt core for casting according to claim 1, wherein the mold is heated to a temperature of 225°C to 250°C.
A method for producing a salt core for casting according to claim 1 or 2, wherein the salt mixture has a composition in which Na⁺ : K⁺ = 70 mol% : 30 mol%, and Cl^- : CO₃ ^2- = 46.2 mol% : 53.8 mol%.