Inorganic binder system

Abstract

A composition for making cores and a process for metal casting, the composition comprising: a particulate refractory material; an inorganic binder comprising at least one alkali metal silicate; a pozzolanic additive; and a lustrous carbon former. The process includes forming a core from the composition and assembling a mould comprising the core and supplying molten metal.

Claims

1. A composition for making a core for use in a metal casting process, the composition comprising: a particulate refractory material; 0.5 wt % to 5 wt % of an inorganic binder comprising one or more of sodium silicate, potassium silicate, lithium silicate, or a combination thereof; 0.1 wt % to 2 wt % of a pozzolanic additive, comprising one or more of silica fume, fused silica, pyrogenic silica, and micro-silica; and 0.1 to 1.5 wt % of a lustrous carbon former, wherein said weight percentages are relative to the weight of the particulate refractory material.

2. The composition according to claim 1, for making a core for use in a ferrous metal casting process, and wherein the lustrous carbon former is a strong lustrous carbon former, wherein the strong lustrous carbon former has a lustrous carbon content of at least 15%.

3. The composition according to claim 2, wherein the strong lustrous carbon former comprises one or more of: asphalt; hydrocarbon resin; polystyrene; and gilsonite.

4. The composition according to claim 1, for making a core for use in a non-ferrous metal casting process, and wherein the lustrous carbon former is a weak lustrous carbon former, wherein the weak lustrous carbon former has a lustrous carbon content of less than 15%.

5. The composition according to claim 4, wherein the weak lustrous carbon former comprises one or more of: graded coal, coal dust, and seacoal.

6. The composition according to claim 1, wherein the particulate refractory material comprises sand.

7. A core formed from the composition of claim 1.

8. A process for producing a metal article by metal casting, the process comprising: (i) mixing a composition according to claim 1 to form a mixture; (ii) moulding and hardening the mixture to produce a core in the shape of an internal cavity of the article; (iii) assembling the core with a mould for metal casting, such that the mould and the core together define a casting cavity; (iv) supplying molten metal into the mould cavity until the mould cavity is filled; (v) cooling and solidifying the molten metal to form the article.

9. The process of claim 8, for producing a ferrous metal article by ferrous metal casting, wherein; step (iv) is carried out at a temperature of at least 1000 C.

10. The process of claim 8, for producing a non-ferrous metal article by non-ferrous metal casting, wherein; step (iv) is carried out at a temperature of less than 1200 C.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 is a graph of the flowability characteristics of the compositions of Table 1;

(2) FIG. 2 is a graph of the bending strengths of cores formed from the compositions of Table 1;

(3) FIG. 3 is a graph of the flowability characteristics of the compositions of Table 2;

(4) FIG. 4 is a graph of the bending strengths of cores formed from the compositions of Table 2;

(5) FIG. 5 is a graph of the bending strengths of cores formed from the compositions of Table 3; and

(6) FIG. 6 is a graph showing the Thermogravimetric Analysis of four lustrous carbon formers.

EXAMPLES

Experiment 1Initial Compositions for Ferrous Casting

(7) A series of cores were produced from quartz sand type H33 mixed with the binders and additives in Table 1 below. Mixing was performed with Hobart mixer for 1 minute and then repeated for a second minute.

(8) A Brookfield Powder Flow Tester was used to test the flowability characteristics of the compositions and measure flow function of the compositions. In turn, a sample of each of the compositions was loaded into a cell in the Brookfield PFT and a vertical force was then applied to the powder to compact it (the Major Principal Consolidating Stress). Subsequently, a rotational force is applied to the compacted powder while the same Major Principal Consolidation Stress is maintained in order to determine the force required to initiate flow of the powder (the Unconfined Failure Strength). The process is repeated at a range of consolidation stresses and the flow function constructed by plotting the unconfined failure strength against the consolidation stress as shown in FIG. 1 and thus determine the internal resistance to flow of the composition. Typically, greater flowability is desirable to reduce problems with powder handling and transport, and to avoid moulding defects. No obvious difference was observed between Ex. 1 to 4 and 6, with only Ex. 5 showing significantly lower flowability due to the inclusion of carbon black within the composition.

(9) Transverse bars were manufactured with the Laempe laboratory machine type L1 being developed for manufacturing test-cores in heated and non-heated tooling, using gas hardening processes like CO.sub.2, cold box and hot box. The sand mixture is automatically injected in the core box, which is clamped between the side presses, and can be heated at various temperatures. The release of high pressured air blows the sand from the sand storage bunker into the core box at high speed. The total elapsed shooting time was set at 1 s and with a shooting pressure of 4 bar (400 kPa). All specimens were purged with heated air for 120 s at 120 C. Core box temperature was set at 140 C.

(10) Bending strength of the samples was measured with a Tinius Olsen H5K-T strength tester, computer controlled with Q-mat software for 3-point bending tests and the results graphed in FIG. 2. The lowest strength was obtained with carbon black as the breakdown agent and was between 50 and 60 N/cm.sup.2these cores were highly friable.

(11) TABLE-US-00001 TABLE 1 Composition Ex1 Ex2 Ex3 Ex4 Ex 5 Ex6 Binder .sup.a, b 2.4% 2.4% 2.4% 2.4% 2.4% 2.2% M23NL .sup.1 94.5 94.5 94.5 94.5 94.5 77.5 Water 5.0 5.0 5.0 5.0 5.0 22.0 DSK40 .sup.2 0.5 0.5 0.5 0.5 0.5 0.5 Total 2.1% 2.1% 2.1% 2.1% 2.1% 1.8% additive .sup.a, c Microsit H10 .sup.3 85 80 80 80 80 FP70/75 .sup.4 15 AG60/70 .sup.5 20 Coke FM .sup.6 20 Petrol Coke 20 Flower .sup.7 Carbon Black .sup.8 20 Silica 22 Fume A .sup.9 EFA-Fller .sup.10 74 Collapsibility Easy Easy Easy Mod- Very Mod- erate/ easy erate/ easy easy .sup.a values in bold are in wt % relative to weight of particulate; .sup.b values for individual compounds listed below are in wt % relative to weight of total binder; .sup.c values for individual compounds listed below are in wt % relative to weight of total additive .sup.1 Sodium silicate M23NL (BASF, Monheim, Germany) .sup.2 DSK 40 (2-ethylhexyl sulphate sodium), 0.5% (Anionic surfactant, Brenntag, Enschede, the Netherlands) .sup.3 H10Microsit H10 fly ash (BauMineral GmbH, Herten, Germany) .sup.4 Graphite FP70/75 (IMCD Benelux BV, Rotterdam, Netherlands) .sup.5 Amorphous Graphite (Grafitbergbau Kaiserberg, St Stefan ob Loben, Austria) .sup.6 Coke Flour M (Mco Mcher & Enstipp GmbH, Essen, Germany) .sup.7 Petrol Coke Flour (Mco Mcher & Enstipp GmbH, Essen, Germany) .sup.8 Carbon Black (IMCD Benelux BV, Rotterdam, Netherlands) .sup.9 Silica Fume A (COFERMIN Chemicals GmbH & Co., Essen, Germany) .sup.10 EFA Fuller HP, fly ash (Krahn Chemie, Zaandam, Netherlands)

(12) The above cores were tested in a casting trial using ductile iron with a pouring temperature of 1355 C. After the casting process, core residue was removed and qualitatively measured and the results listed in Table 1 above. De-coring after the casting trials was easiest for the cores containing carbon black. Slightly less collapsibility was found with Ex 4 and Ex 6. The inner surface of the castings were investigated. It was clear that in all cases, no significant difference occurred between the various inner surfaces and in all cases the surface was relatively rough with sand grains adhering to the surface.

Experiment 2Further Ferrous Investigations

(13) Ex. 6 above was selected for further development and improvement. Sand mixtures based on quartz sand type H33 were mixed with the binders and additives in Table 2 below. The gilsonite would be considered an additive within the context of the present invention, but the gilsonite content is listed separately in Table 2 for ease of comparison and to simplify the test procedure. The total additive content can be calculated by adding the additive concentration and gilsonite concentration, since both are record in wt % relative to the weight of the particulate refractory material.

(14) Flowability measurements were measured as per Experiment 1 above and results graphed as per FIG. 3. All compositions were found to be easy flowing, with lower gilsonite contents having greater flowability. Subsequently cores were produced under the same conditions and bending strength testing carried out as Experiment 1 above, and the results graphed in FIG. 4. Increasing gilsonite concentration demonstrated an approximately linear decrease in bending strength of the cores. Sample weight for all transverse bars was between 693 g and 700 g, which shows that the gilsonite content does not have a significant effect on compaction levels.

(15) Casting trials with ductile iron were performed at 1380 C. Subsequently, the castings were cooled down to room temperature before the core residue was removed. It was found that cores with the presence of a low amount of Gilsonite (breakdown agent) showed improved breakdown properties in comparison with those without Gilsonite, and the breakdown properties increased with an increase in the Gilsonite concentration, as shown in Table 2 below. After removing core residue, it was clear that sand adhesion decreased with higher levels of the Gilsonite, and that observations of the inner casting surface revealed that surface smoothness increased with higher levels of the Gilsonite. The use of concentration levels of Gilsonite of 0.6 wt % or higher resulted in veining formation in the casting; the higher the concentration the higher the sensitivity for veining. After sand blasting, it was clear that the inner surface of the casting and related to cores without Gilsonite showed a white-shiny appearance. This was not the case when Gilsonite was present, irrespective the concentration level.

(16) TABLE-US-00002 TABLE 2 Ex Ex Ex Ex Ex Ex Composition 7 8 9 10 11 12 Binder .sup.a, b 2.2% 2.2% 2.2% 2.2% 2.2% 2.2% M23NL.sup.1 77.5 77.5 77.5 77.5 77.5 77.5 Water 22.0 22.0 22.0 22.0 22.0 22.0 DSK40 .sup.2 0.5 0.5 0.5 0.5 0.5 0.5 Additives .sup.a, c 1.8% 1.8% 1.8% 1.8% 1.8% 1.8% EFA- 74 74 74 74 74 74 Fller .sup.10 Silica 22 22 22 22 22 22 Fume A .sup.9 Graphite 4 4 4 4 4 4 FP70/75 .sup.4 Gilsonite .sup.a, 11 0.0% 0.2% 0.4% 0.6% 0.8% 1.0% Collapsibility Mod- Easy Easy Very Very Very erate easy easy easy .sup.a values in bold are in wt % relative to weight of particulate; .sup.b values for individual compounds listed below are in wt % relative to weight of total binder; .sup.c values for individual compounds listed below are in wt % relative to weight of total additive .sup.1Sodium silicate M23NL (BASF, Monheim, Germany) .sup.2 DSK 40 (2-ethylhexyl sulphate sodium), 0.5% (Anionic surfactant, Brenntag, Enschede, the Netherlands) .sup.4 Graphite FP70/75 (IMCD Benelux BV, Rotterdam, Netherlands) .sup.9 Silica Fume A (COFERMIN Chemicals GmbH & Co., Essen, Germany) .sup.10 EFA Fuller HP, fly ash (Krahn Chemie, Zaandam, Netherlands) .sup.11 Natural carbon-containing resin, American Gilsonite Company, Utah, USA

(17) The present inventors have found that strong lustrous carbon formers (LCFs) are particularly desirable for use in ferrous casting. The high temperatures are believed to require a higher lustrous carbon content in the overall composition in order to achieve the desired improvements in surface quality and in reducing casting defects. By using a strong LCF, this improvement is achieved without significantly affecting the flowability of the composition or the strength of the cores formed therefrom. Without wishing to be bound by theory, although a higher content of weak LCF may provide an equivalent lustrous carbon content and thus in theory provide a similar improvement in casting quality, the reduction in flowability and core strength lead to reductions in casting quality which negate any theoretical improvements. It is believed that the overall lustrous carbon content in the composition can be carefully selected by using a blend of LCFs, including optionally a blend of strong and weak LCFs, in order to achieve the optimal casting conditions without affecting workability or core strength.

Experiment 3Compositions for Non-Ferrous Casting

(18) Tests were carried out to investigate the suitability of lustrous carbon formers for use in non-ferrous casting processes. A series of cores was prepared using sand and the compounds in Table 3 below.

(19) TABLE-US-00003 TABLE 3 Composition Ex 13 Ex 14 Ex 15 Ex 16 Binder .sup.a, b 2.0% 2.0% 2.0% 2.0% M23NL.sup.1 77.5 77.5 77.5 77.5 Water 22.0 22.0 22.0 22.0 DSK40 .sup.2 0.5 0.5 0.5 0.5 Additives .sup.a, c 1.0% 1.0% 1.0% 1.0% EFA-Fuller .sup.10 74 74 74 74 Silica Fume A .sup.9 22 22 22 22 Graphite FP70/75 .sup.4 4 4 4 4 Tri-calcium 0.3% 0.3% 0.3% phosphate .sup.a Gilsonite .sup.a, 11 0.05% 0.10% .sup.a values in bold are in wt % relative to weight of particulate; .sup.b values for individual compounds listed below are in wt % relative to weight of total binder; .sup.c values for individual compounds listed below are in wt % relative to weight of total additive .sup.1Sodium silicate M23NL (BASF, Monheim, Germany) .sup.2 DSK 40 (2-ethylhexyl sulphate sodium), 0.5% (Anionic surfactant, Brenntag, Enschede, the Netherlands) .sup.4 Graphite FP70/75 (IMCD Benelux BV, Rotterdam, Netherlands) .sup.9 Silica Fume A (COFERMIN Chemicals GmbH & Co., Essen, Germany) .sup.10 EFA Fuller HP, fly ash (Krahn Chemie, Zaandam, Netherlands) .sup.11 Natural carbon-containing resin, American Gilsonite Company, Utah, USA

(20) The flowability was tested as in Experiment 1. All compositions had very similar profiles and showed no obvious differences in flowability, and were easy flowing. A series of transverse bars were formed using the compositions, solidified and the strengths tested as per Experiment 1. The results were graphed and are found in FIG. 5. The presence of a small amount of tri-calcium phosphate lead to a significant decrease in bending strength, and low concentrations of Gilsonite were not found to have a significant effect on strength values.

(21) Cores formed from the compositions in Table 3 were tested in a casting trial using aluminium with a pouring temperature of 745 C. Higher magnifications showed, in case of Ex. 15 and 16 which included a small amount of Gilsonite, small gas defects at the inner surface of the aluminium castings. However, severe deformation occurred of the aluminium casting material in Ex 15 and Ex 16, with the higher concentration of Gilsonite showing greater deformation. The use of Gilsonite for aluminium casting was not considered workable.

(22) TABLE-US-00004 TABLE 4 Composition Ex 17 Ex 18 Ex 19 Ex 20 Binder .sup.a, b 2.4% 2.1% 2.1% 2.2% Crystal 0230 .sup.12 90 90 90 ZSE874 .sup.13 90 Water 5 4.5 4.5 4.5 Kasil 1841 .sup.14 4.5 5 5 5 DSK40 .sup.2 0.5 0.5 0.5 0.5 Additives .sup.a, c 1.2% 1.0% 1.0% 1.2% Silica Fume A .sup.9 62 78 78 78 Aluminium metaphosphate 20 AFS200 .sup.15 6 8 8 8 AF96/97 .sup.16 7 7 14 Graphite FP70/75 .sup.4 12 GC-190 .sup.17 7 GC-145 .sup.18 7 Fused silica .sup.a 0.2% 0.2% 0.2% Aluminium metaphosphate .sup.a 0.1% .sup.a values in bold are in wt % relative to weight of particulate; .sup.b values for individual compounds listed below are in wt % relative to weight of total binder; .sup.c values for individual compounds listed below are in wt % relative to weight of total additive .sup.2 DSK 40 (2-ethylhexyl sulphate sodium), 0.5% (Anionic surfactant, Brenntag, Enschede, the Netherlands) .sup.4 Graphite FP70/75 (IMCD Benelux BV, Rotterdam, Netherlands) .sup.9 Silica Fume A (COFERMIN Chemicals GmbH & Co., Essen, Germany) .sup.12 Sodium silicate solution Crystal 0230 (PQ Corporation, Eijsden, the Netherlands) .sup.13 Sodium silicate solution ZSE874 (PQ Corporation, Eijsden, the Netherlands) .sup.14 Kasil 1841 (PQ Corporation, Eijsden, the Netherlands) .sup.15 Cerabeads AFS 200 (Ziegler & Co. Wunsiedel, Germany) .sup.17 Graded coal GC-190 (James Durrance Sons Ltd, UK) .sup.18 Graded coal GC-145 (James Durrance Sons Ltd, UK)

(23) A series of cores for use with a permanent die for aluminium casting were prepared according to Table 4 below. The cores were tested in a casting trial using aluminium with a pouring temperature of approximately 730 C. After solidification in the permanent die, the castings were stored for 30 minutes in a pre-heated furnace at 500 C. After solidification, core Ex. 17 (without a lustrous carbon former: graded coal) showed more sand adhesion. The use of graded coal-145 lead to a higher surface quality compared to graded coal-190.

(24) Without wishing to be bound by theory, it is believed that lustrous carbon formers such as Gilsonite are too strong to be effective for use in non-ferrous and/or lower temperature castings.

Experiment 4Lustrous Carbon Former Investigations

(25) Thermogravimetric analysis was carried out for four lustrous carbon formers: Gilsonite, Graded Coal-190, Superfine Graded Coal-240, and Coal Sand. The analysis was carried out from 20-1000 C. at a rate of 10 C./min, with the exception of Gilsonite, which was tested at a rate of 5 C./min. As is shown in FIG. 6, although all four lustrous carbon formers began to lose mass from approximately 400 C., the Gilsonite sample lost mass far more rapidly than the other three lustrous carbon formers.

(26) Table 5 shows a list of lustrous carbon formers and typical lustrous carbon content contained therein.

(27) TABLE-US-00005 TABLE 5 Lustrous carbon former Lustrous carbon content Polystyrene 56% Hydrocarbon resin 38-48%.sup. Gilsonite 35% Asphalt 26-32%.sup. Coal dust 8-14%.sup. Graded coal 8-12%.sup. Seacoal 10%

(28) Without wishing to be bound by theory, the inventors believe that the rate at which the lustrous carbon formers are able to volatilise under the casting conditions significantly affects the activity of the lustrous carbon former (LCF) to reduce surface defects in the casting. For lower temperatures, such as in aluminium and other non-ferrous casting processes, weak lustrous carbon formers have surprisingly been found to be more effective. For higher temperatures such as found in ferrous casting processes, the use of strong lustrous carbon formers has been found to be surprisingly effective. As used herein, the terms strong and weak reflect both the LCFs volatility and the overall content of lustrous carbon within the additive. Experiments carried out with alternative carbon sources, such as graphite, were found to be far less effective than LCFs. The most effective LCF for any particular casting process is a balance between strength of the LCF effect, pouring temperature of the casting and the requirement to minimise strength loss of the core through LCF addition rates.

Experiment 6 Foundry Moulding Compositions and Core Production

(29) Various cores were produced under the following conditions: sand, liquid binder and additives as specified in Table 6a below were mixed with a commercially available batch mixer (Hobart) with a batch size of 20 litre, whereby the additive and liquid binder were added in parallel with a mixing time of 21 min.

(30) TABLE-US-00006 TABLE 6A Composition Ex 21 Ex 22 Ex 23 Particulate refractory Silica sand Silica sand Silica sand material QQS 26 .sup.22 F32 .sup.23 HB32 .sup.24 Binder .sup.a, b 2.1% 1.6% 1.8% Crystal 0230 .sup.12 79.5 84.5 ZSE874 .sup.13 79.5 Water 20 20 5 Kasil 1841 .sup.14 5 5 5 Silres BS16 .sup.19 5 DSK 40 .sup.2 0.5 0.5 0.5 Total additives .sup.a, c 1.2% 0.6% 0.6% Fused Silica 325 .sup.20 16.7 Silica Fume A .sup.9 65 60 26 AFS 200 .sup.15 6.6 GC-190 .sup.17 11.7 Eurocell 150H .sup.21 6 6 Gilsonite .sup.11 34 68 .sup.a values in bold are in wt % relative to weight of particulate; .sup.b values for individual compounds listed below are in wt % relative to weight of total binder; .sup.c values for individual compounds listed below are in wt % relative to weight of total additive .sup.2 DSK 40 (2-ethylhexyl sulphate sodium), 0.5% (Anionic surfactant, Brenntag, Enschede, the Netherlands) .sup.11 Natural carbon-containing resin, American Gilsonite Company, Utah, USA .sup.12 Sodium silicate solution Crystal 0230 (PQ Corporation, Eijsden, the Netherlands) .sup.13 Sodium silicate solution ZSE874 (PQ Corporation, Eijsden, the Netherlands) .sup.14 Kasil 1841 (PQ Corporation, Eijsden, the Netherlands) .sup.15 Cerabeads AFS 200 (Ziegler & Co. Wunsiedel, Germany) .sup.17 Graded coal GC-190 (James Durrance Sons Ltd, UK) .sup.19 silicon organic water repellent (Wacker Chemie AG, Stuttgart, Germany) .sup.20 Fused Silica 325 (Imerys Fused Minerals Greeneville Inc., Greeneville, USA) .sup.21 Pozzolanic filler - aluminium-silicate (Stauss-Perlite GmbH, Poelten, Austria) .sup.22 QQS 26 (D50 particle size 0.26 mm) (Wolff und Muller Quarzsande GmbH, Germany); .sup.23 F32 (D50 particle size 0.24 mm) (Quarzwerke GmbH, Frechen, Germany); .sup.24 HB32 (D50 particle size 0.30 mm) (Quarzwerke GmbH, Frechen, Germany);

(31) The mixtures were introduced into core shooters as set out in Table 6b below and cores were produced under the conditions therein.

(32) Ex 21After storing the cores 24 hours at an air temperature of 20 C. at a relative humidity of 40% a strength test was performed. The cores had a cold strength of about 400 N/cm.sup.2 as measured according to Experiment 1 above. Casting trials were carried out with the alloy AlSi.sub.7Cu.sub.4Mg.sub.0.5. Pouring temperature was 760 C. and the total amount of aluminium was for each casting trial 39 kg. The inner surface of the casting specimen showed a clean surface without sand adhesion. No coating was applied to the cores.

(33) TABLE-US-00007 TABLE 6B Ex 21 Ex 22 Ex 23 Core shooter DISA CORE 20L Laempe L1 DISA CORE 20L Core box temp. 140 C. 140 C. 140 C. Shooting pressure 4 bar 4 bar 4 bar Shooting time 2 1 s 1 1 s 2 1 s Purging air temperature 165 C. 120 C. 165 C. Purging time 50 s 60 s 50 s Non-pressurised time 5 s 5 s 5 s Curing air pressure 3.0 bar (300 kPa) 4.0 bar (400 kPa) 3.0 bar (300 kPa)

(34) Ex 22After storing the cores 1 hour at an air temperature of 20 C. at a relative humidity of 50% a strength test was performed. The cores had a hot cold strength of 350 N/cm.sup.2 as measured as per Experiment 1. No coating was applied to the cores. Casting trials were carried out with the alloy CC491. Pouring temperature was 1150 C. and the total amount of the alloy was for each casting trial 20 kg. The inner surface of the casting specimen showed a clean surface without sand adhesion.

(35) Ex 23After storing the cores 24 hours at an air temperature of 20 C. at a relative humidity of 50% a strength test was performed. The cores had a strength of 500 N/cm.sup.2 measured as per Experiment 1 No coating was applied to the cores. Casting trials were carried out with nodular cast iron GJS600. Pouring temperature was 1450 C. and the total amount of the alloy was for each casting trial 160 kg. Smooth and sand-free surfaces of differential housings as test specimen were achieved with the cores produced as above described.