Influence of the furan moulding sand on the ductile iron casting surface quality

Transcription

1 Influence of the furan moulding on the ductile iron casting surface quality R. Dańko AGH University of Science and Technology, Faculty of Foundry Engineering, Krakow, Poland, M. Górny AGH University of Science and Technology, Faculty of Foundry Engineering, Krakow, Poland, M. Holtzer AGH University of Science and Technology, Faculty of Foundry Engineering, Krakow, Poland, S. Żymankowska-Kumon AGH University of Science and Technology, Faculty of Foundry Engineering, Krakow, Poland, Copyright 214 World Foundry Organization ABSTRACT The results of investigations of the influence of molding with furan resin, prepared both with fresh and reclaimed matrix, on the formation of microstructure and surface quality of ductile iron castings are presented in the paper. Different furan moulding s were used within the studies: moulding prepared with fresh and moulding s prepared with reclaimed s of different purification degree, determined by loss of ignition value. A series of experimental heats of ductile iron in molds made of molding characterized by different levels of surface-active elements (sulfur) were performed. Various concentrations of sulfur, which is considered as a unfavourable element from the spheroidisation point of view, in the moulding s as a function of the loss on ignition has been shown in the paper. Investigations carried out by means of scanning electron microscopy (EDS), showed a concentration gradient profiles of surface-active elements in the surface layer of castings which are responsible for their quality. Moreover an analysis is shown of the effect of the wall thickness, and the initial temperature of the metal in the mould cavity on the formation of the graphite degeneration nodules in the surface layer of the casting. Keywords: ductile iron, casting skin, graphite degeneration, moulding. INTRODUCTION Production of castings of ductile iron, in loose selfhardening moulding s with furan resin hardened by paratoluenosulphonic acid, brings a danger of forming defected casting microstructure, which the most often occurs in its surface layer 1-1). This unfavourable effect is caused by surface-active elements like sulphur contained in a hardener of resin binders, which enters the surface layer causing perlite of flake graphite formation. In such case the surface layer with flake graphite can cause stress raising in the casting, similar to a notch, so all useful properties are reduced, especially the fatigue limit and impact resistance. The flake graphite in the surface layer is the most critical for thin wall castings, where it could become more than 1% of the total section. It affects also castings of thicker walls, due to the long solidification time providing an extended metal/mould interaction time. In the present paper the effort was undertook of making test castings of ductile iron in moulds made of a fresh moulding and of a moulding with a reclaim of various purification degrees. Then the castings were subjected to investigations leading to determination the microstructure of their skin layer directly adhering to the mould. Simultaneously, the gas evolution and sulphur and nitrogen contents were determined for moulding s, of which moulds were made. Moreover an analysis is shown of the effect of the wall thickness, and the initial temperature of the metal in the mould cavity on the formation of the flake graphite in the surface layer of ductile iron castings.

2 EXPERIMENTAL The experimental melts were performed in an induction furnace. The furnace charge consisted of the following materials: Sorelmetal, silicon of technical purity, Fe-Mn and steel scrap. After metal heating to a temperature of 149 o C the bath was hold for 2 minutes and then spheroidising and innoculation by the bell method was carried out. For the spheroidisation treatment the foundry alloy Fe-Si-Mg (6 % by mass Mg) was used, while for the inoculation the Foundrysil modifier was used in amount of.5% by mass. The pouring temperature was app. 14 C. Four Y block ingots were prepared from selfhardening moulding s (MT1-1 through MT1-4) of a characteristic given in Table 1. Moreover in order to determine the effect of the pouring temperature (represented by the initial temperature of the metal in mould cavity) plate shaped castings with 5x5x1 mm and 8x5x1 mm were cast. Temperature of metal in mould cavity was determined by using thermoelements (Pt-PtRh1 - with.2 mm in thickness connected to the digital data acquisition system AGILENT 3497 A) wires in regular 1 cm distances. This experimental cast was performed in mould prepared with moulding MT1-3. Table 1. Grain size analysis and loss on ignition of moulding s and matrices applied for preparing test moulding s: MT1-1 MT1-4 notation MT1-1 MT1-2 MT1-3 Characteristic of moulding s Kind of matrix Highsilica Reclaim after 1 cycle Reclaim after 2 cycles Reclaim after Loss on ignition SP (mz) Characteristic of matrix subjected to different number of cycles Grain size analysis (Analisette 22 NanoTec) d a d g S t Loss on ignition SP (P) SP (R) % mm mm cm 2 /g % MT1-4 3 cycles d a average arithmetic grain diameter, mm d g average geometric grain diameter, mm S t specific surface, cm 2 /g Ductile iron of the following chemical composition was obtained: C 3.56%; Si 2.99%; Mn.29%; P.46%; S.11%; Cr.3%; Mg.46%; Cu.2%. The performed tensile strength tests allowed to determine the grade of the melted ductile iron as ASTM A (UTS=594 MPa, YS = 38 MPa, A5 =15.7 % ). The view of the Y ignots test casting along with plate shaped casting are presented in Figure 1. b) Fig 1. Four Y ignots casted in moulds prepared with MT1-1 through MT1-4 of different matrix quality (b) plate-shaped casting Samples cut out from lower parts of test bars were subjected to metallographic and strength investigations. Metallographic investigations were carried out by means of the optical microscope Leica MEF-4M connected to computer equipped with image analysis software Leica QWin (v. 3.5.). In addition, the near surface microstructures were examined by a JEOL JSM-71F scanning electron microscope (SEM) operated at 2 kv.the cast steel classification, on the basis of the static tensile test was performed by means of the Zwick/Roell Z5 device equipped with the extensometer Macro. The testing rate was.8 s-1. Moreover samples for metallographic examinations were cut from plate shaped castings. Characterization of the graphite and matrix microstructure was performed on cross-sections at different distances from the entrance to the spiral. RESULTS AND DISCUSSION Determination of sulphur and nitrogen content in moulding s The literature data 4,8-9) emphasize a disadvantageous influence of too high sulphur and nitrogen content in moulding s on the microstructure of the obtained castings surfaces.

3 Within quality investigations of the reclaimed matrix the sulphur and nitrogen content in 4 moulding s (listed in Table 1) were performed. The obtained results presented graphically in Figure 2 confirm the accumulation effect of sulphur and nitrogen in moulding s after several reclamation treatments 11). The increase of sulphur and nitrogen content in moulding s as a function of loss on ignition exhibits nearly linear character. Thus, it seems possible to determine indirectly these elements content in moulding s on the basis of knowledge of loss on ignition and its characteristic similar to the one presented in Figure 2. b) c).3.25 Sulphur content in moulding, % Nitrogen content in moulding, % Sulphur and nitrogen content, % y =.421x R² =.9964 y =.667x R² = Loss of ignition, % Fig 2. Sulphur (S) and nitrogen (N) content in moulding s with Kaltharz U44 resin (1% by mass) and 1T3 hardener (.5% by mass). Loss on ignition as in Table 1. Gas volumetric emission from moulding s During a casting production process an intensive thermal destruction of organic components of moulding s occurs, causing a large emission of gases. These gases constitute a threat for the casting quality, since they can migrate into a casting and worsen its surface. Own investigations of gases emission were carried out according to the original method developed in the Faculty of Foundry Engineering AGH 12). Investigations of the gases emission were performed according to the original method developed in the Faculty of Foundry Engineering, AGH UST. The schematic presentation of the experimental stand is given in Figure 3. A sample of the investigated moulding of a roll shape of dimensions Φ 5 x 5 mm, compacted by a moulder s rammer stroke, is poured with liquid cast iron of a temperature of 1 35 o C. Gases emitting from the sample - after pouring it with liquid metal - are led by means of a steel pipe via the drying system and the capsule with active carbon into a tightly sealed container with liquid, from which they push out the liquid. The weight of displaced liquid was measured as a function of time. Fig 3. Scheme of stand of measurement of gas volume (, mould with the sample before casing (b), mould with the sample after casting (c) Investigations of the amounts and kinetics of gases generated in the process of pouring test bars were performed for the given above variants of moulding s (MT1-1 through MT1-3), which characteristic is presented in Table 1. The obtained gases emissivity pathways are presented in Figure 4 as a time function. It can be noticed, that the amount of gases generated from moulding with furan resin depends on the loss on ignition of this. With an increase loss on ignition an amount of gases generated in the mould pouring process also increases. The highest intensity of gases emission takes place directly after the mould pouring with liquid metal. The kinetics of gases emission, presented in Figure 5, allows to state that under tests conditions the highest emission occurs in the first 9 seconds after pouring. The analysis of pathways indicates that both the volume and kinetics of emitted gases depends essentially on the loss on ignition of moulding s.

4 Volume of gas, cm 3 /g MT Time, s MT1-1 MT1-2 Fig 4. Volume of gases emitted by the investigated moulding s in the process of their pouring with liquid metal. Pouring temperature: 135 C increases. It can be noticed, that the thickness of these layers is increasing when the loss on ignition of moulding out of which the mould was produced is increasing. In case of the casting produced of fresh moulding s (for which sulphur and nitrogen content was determined as being.9% and.7, respectively, and loss on ignition = 1.47%) the thickness of these layers is the smallest. However, also in this case there is a zone in which the nodular graphite, much-desired from the casting point of view, was degenerated. In case of the casting 4? microstructure (Fig. 5), for which the mould was produced of the moulding containing.24% of sulphur and.14% of nitrogen (loss on ignition= 4.26%) the maximum thickness of the degenerated graphite layer was nearly 1.5 mm. b).3.25 moulding MT1-1 moulding MT1-2 moulding MT1-3 Gas evolution rate, cm 3 /gs c) d) Time, s Fig 5. Kinetics of gases emitted by the investigated moulding s in the process of their pouring with liquid metal. Pouring temperature: 135 C CASTING MICROSTRUCTURES Y ignots test casting Microstructures of ductile iron obtained for the investigated moulding s is presented in Figures 6 in such a way that the dark zone on the bottom means the metal and moulding contact. The presented results of investigations of microstructures of castings indicate that with the increase loss on ignition of moulding s, of which the mould was made, (moulding MT1-1- moulding MT1-4) the thicknes of the flake graphite layer located at the metalmould contact increases. The mould production with using reclaimed materials of higher loss on ignition (it means of larger amounts of spent binder left on matrix grains) causes that the thickness of flake graphite layer Fig 6. Microstructure of the casting in mold-metal surface made in the mould of moulding MT1-1 MT1-4 (a-d), respectively Plate shaped castings Figure 7 shows the microstructure of ductile iron in the surface layer in plate shaped castings. Figure 8 shows the relationship between the thickness of the flake graphite layer located at the metal-mould contact and the distance from the beginning of plate inlet. Figure 8 shows that the thickness of the flake graphite layer in ductile iron decreases with the beginning of plate inlet. At a distance 7 cm from the beginning of the plate shaped casting (x = 7 cm) flake graphite is not present in the surface layer of a casting with wall thickness of 5 mm. It is worth noting that, in the casting wall thickness of 8 mm thickness of flake graphite layer is much higher, and disappears after reaching 9 cm from the casting inlet. This demonstrates a significant influence the wall thickness on the occurrence and the thickness of the layer of flake graphite in the surface layer. The smaller wall

5 thickness of the casting the higher temperature drop of the flowing metal stream filling the mold cavity. This is shown in Figure 9. A) B) Temperature of the melt Ti, oc wall thickness 5 mm wall thickness 8 mm b) c) Fig. 7. Microstructure of ductile iron in plate shaped castings with different wall thickness (A, B- plates with 8 and 5 mm wall thickness, respectively) and distances from the beginning of plate inlet: 1 cm, b) 3 cm, c) 7 cm Distance from the beginning of plate inlet x, cm Fig. 9. Initial temperature of metal in mould cavity (T i) during metal filling for castings with different wall thickness The temperature of the melt decreases approximately linearly with the distance from the entrance to the mould cavity. From the figure 9 it follows that there is a high temperature drop during mould filling that strongly depends on wall thickness. For casting with higher wall thickness decrease of the temperature during mould filling is smaller. The consequence of this is a higher temperature and the longer the contact time of the liquid metal with the mould as compared to a casting with smaller wall thickness. Flowing metal stream through the mould cavity heats it up. As a consequence, the conditions of heat exchange along the flowing path change 13). With increasing distance from the entrance to the mould cavity there is a shorter contact time of liquid metal with the mould, which results in decreasing temperature and velocity profiles. Figure 1 shows the effect of wall thickness and the distance from the entrance to the mould cavity to the contact time of the liquid metal with the mould. wall thickness 5 mm wall thickness 8 mm 3 Thickness of the flake graphite layer g, µm Contact time of the liquid metal with the mould tc, s Distance from the beginning of plate inlet x, cm Fig. 8. Thickness of the flake graphite layer (g) as a function of the distance from the beginning of plate inlet (x) wall thickness 5 mm wall thickness 8 mm Distance from the beginning of plate inlet x, cm Fig 1. Contact time of the liquid metal with the mould (tc) as a function of distance from the entrance to the mould cavity (x) for castings with different wall thickness

6 From figure 1 results that the contact time of the liquid metal with the mould is much longer for a casting with wall thickness of 8 mm and decreases approximately linearly with the distance from the entrance to the mould cavity. Thickness of the flake graphite zone depends mainly on despheroidizing elements (especially sulphur) content in the moulding and on the magnesium to sulphur ratio in the cast iron. Research has shown that for a given wall thickness of the casting thickness of the flake graphite layer in the surface of the casting depends on the initial temperature of the metal in the mould cavity and on its contact time with the casting mould. The flake graphite in the near surface layer of ductile iron occurs due to the magnesium (spheroidizing element) concentration decreases. From EDS analysis (Fig 11) results, that this occurs by interaction of sulfur with magnesium, and also with among others with manganese and iron through the creation of complex sulphides in the surface layer. b) Fig. 11. SEM microstructure ( and EDS spectrum (b) of complex sulfides EDS studies indicate that the number such sulphides decreases as distance from the casting inlet. This decrease depends on diffusing of sulphur from the moulding into metal. Summing up it can be state that the higher concentration of sulphur in the moulding, the higher initial temperature of metal in mold cavity and the longer contact time of liquid metal with moulding the higher thickness of the flake graphite layer near the surface of casting. CONCLUSIONS 1. The loss on ignition of the moulding after three cycles of the reclamation treatment is nearly three times larger than that of the moulding prepared on the fresh high-silica. Also matrix indicates significant increases of loss on ignition when the number of cycles increases, which indicates the accumulation of a spent binder - on matrix grains - not removed during the reclamation treatment preceding the moulding preparation. 2. Multiple reclamation treatment causes accumulating of sulphur and nitrogen content in moulding s. Sulphur and nitrogen content in moulding s increase analysed as a function of their loss on ignition indicates nearly linear character. 3. The most intensive gas evolution occurs directly after the mould pouring with liquid metal. Under the investigated conditions, the highest emission of gases occurs in the first 9 seconds after pouring with metal. An application of a moulding of higher loss on ignition causes increasing of gas emissions. 4. Castings performed in the mould of the fresh moulding indicate the best surface quality. When the number of the reclamation cycles was increased the worsening of casting surfaces occured. 5. Along with an increase of ignition losses of moulding s, of which the mould was produced, the thickness of the flake graphite layer increases. Producing mould with using reclaims of higher ignition losses (it means of larger amounts of left over spent binders on matrix garins) causes that the thickness of the flake graphite layer increases. 6. Thickness of flake graphite layer depends on the initial temperature of the metal in mould cavity and on contact time of the mould with the liquid metal. In the case of casting a wall thickness of 5 mm flake graphite in the surface layer does not arise from a distance of 7 cm. This corresponds to an initial temperature of the metal in the form of 126 C and the contact time of about 8 seconds of the liquid

7 metal with mould. In the case of casting with a wall thickness of 8 mm flake graphite in the surface layer does not arise from a distance of 9 cm. This corresponds to an initial temperature of 123 C and the contact time of about 9 seconds of the liquid metal with mould. It is worth noting that the above temperatures and contact times for which there is disappearance of flake graphite in a surface layer for casting with different wall thicknesses are similar. 7. The flake graphite in the near surface layer of ductile iron occurs due to the magnesium (spheroidizing element) concentration decreases. It result from the interaction of sulfur with magnesium, and also with among others manganese and iron through the creation of complex sulphides in the surface layer. 1. Ivan N., Chisamera M, Riposan, I., International Journal of Cast Metals Research. 213, 26, Dańko R.: Innovative developments in reclamation technologies. Metallurgy. 211, 5, Holtzer M., Grabowska B., Żymankowska-Kumon S., Kwaśniewska-Królikowska D., Dańko R., Solarski W., Bobrowski A., Metallurgy. 212 vol. 51 no. 4 p Górny M., Journal of Iron and Steel Research International. 212, vol. 19 nr 8, s ACKNOWLEDGMENTS This work was supported by Polish NCN project UMO-211/3/B/ST8/5869. REFERENCES 1. Stefanescu D.M., Wills S., Massone J., International Journal of Metalcasting, Fall 28, Stefanescu D.M., Juretzko F.R., Study of the Effect of some process variables on the surface roughness and the tensile properties of thin wall ductile iron castings. AFS Transactions, 27, Torrance J.W., Stefanescu D.M., Investigation on the effect of surface roughness on the static mechanical properties of thin-wall ductile iron castings. AFS Transactions, 24, 112, Riposan I., Chisamera M., Stan S., Skaland T., Factors influencing the surface graphite degeneration in ductile iron castings in resin mold technology. Proceedings of the Eighth International Symposium on Science and Processing of Cast Iron, Pekin, China, Boonmee S., Stefanescu D., International Journal of Metalcasting/Spring 213, Górny M., Structure formation of ultra-thin wall ductile iron castings. Kraków, Akapit Ed., Holtzer M., Zych J., Retel K., Przegląd Odlewnictwa, 1996, 49, Grabowska B., Holtzer M., Dańko R., Górny M., Bobrowski A., Olejnik E., Metalurgija Riposan I., Chisamera M., Stan S.: China Foundry, 21, 7, 2,