FINAL TECHNICAL REPORT

September 1, 1996, through August 31, 1997

Project Title: USE OF FLY-ASH BASED BINDERS FOR FOUNDRY MOLDS

ICCI Project Number: 96-1/3.1A-1

Principal Investigator: S.K. Kawatra, Department of Metallurgical and Materials Engineering, Michigan Technological University

Other Investigators: T. C. Eisele, Department of Metallurgical and Materials Engineering, Michigan Technological University; N.W. Burrows, Department of Metallurgical and Materials Engineering, Michigan Technological University

Project Manager: D. Banerjee, ICCI

ABSTRACT

Fly-ash disposal from coal combustion is a serious problem for coal-fired power plants. Some means is therefore needed for disposing of or utilizing the fly-ash. Of the tremendous amounts of fly-ash produced each year, only a small fraction can be used constructively in useful applications, while the majority of fly-ash is landfilled. The beneficial use of fly-ash in an economical and environmentally acceptable manner is important for encouraging the continued use of Illinois coals.

The objective of this project was to develop a new class of markets for Illinois fly-ashes, as particulate binders for the production of sand molds in foundry operations. Each year U.S. foundries consume over 885,00 tons of binder for producing metal-casting molds. The ability to use fly-ash as a binder in the production of foundry molds would consume considerable amounts of Illinois fly-ash, but to this date no studies have been conducted using fly-ash in this application. The extensive Illinois foundry industry would benefit from being able to use locally produced fly-ashes as binders to replace the bentonite, which must be shipped long distances.

In this project, the necessary parameters for producing satisfactory fly-ash bonded foundry molds were determined. All of the fly-ashes studied could be used as satisfactory binders, provided that calcium chloride was added as an accelerator for the binding reactions. The ash from a fluid-bed combustor functioned as a binder with only calcium chloride and water added, but the Class F fly-ashes required additions of calcium-bearing compounds, preferably CaO, before they exhibited binding properties. All of the ash-based binders performed best when they were first pre-reacted with water before adding them to the sand. Molding sand specimens also had the best properties when they were rammed, and then cured for one hour before testing. Fly-ash bonded sands also had satisfactory performance when heated to elevated temperatures, showing that they would behave properly in contact with molten metals in casting operations. The necessary conditions for using fly-ash as foundry sand binders have been established, and prospects for successfully casting metal parts using these sand binders are very good.

EXECUTIVE SUMMARY

Fly-ash is produced in great quantities as a by-product during the combustion of coal, primarily by coal-fired power plants. It is entrained in the exhaust gases and would create a serious dust problem in the surrounding area if released, so clean-air guidelines require that the fly-ash be collected from the flue gases. It is obviously preferable to market the fly-ash for useful purposes instead of simply landfilling it, since the market price can help to defray the costs of shipping and handling the material.

The main existing market for fly-ash is as a cement additive. However, this market requires that the fly-ash contain less than 6% unburned carbon by weight, and a large percentage of the fly-ash produced in the U. S. contains more than 6% unburned carbon. These large quantities of high-carbon fly-ashes have little or no commercial application and must be disposed of in landfills. Therefore, applications in which a high carbon content is non-critical, or even desirable, must be developed for these fly-ashes. An application in which high levels of carbon or other impurities would be desirable is as an inorganic binder to replace or supplement bentonite clay, which is used in foundries to produce sand molds in the casting of metal components. Previous work conducted by the investigators had determined that high-carbon fly-ashes with carbon contents as high as 11.1% can be used as inorganic binders.

A useful binder for foundry sands must be able to hold sand particles together strongly enough to produce molds that can hold their shape and surface detail while molten metal is poured into them. The binder must also give the mold sufficient permeability to allow gases to escape without producing casting flaws. In addition, the molds must be able to withstand high temperatures, without melting into a mass that would be difficult to remove from the finished part. Experimentation carried out by this project to date has concentrated on determining whether binders based on fly-ash can match the baseline characteristics obtained with a conventional bentonite binder, which is known to produce acceptable-quality foundry molds.

The fly-ashes used to date were obtained from a fluidized-bed combustor operated by ADM, Inc. in Decatur, Illinois, and from the pulverized-coal combustors at the E.D. Edwards plant, operated by the Central Illinois Light Co. The ADM fly-ash is a very high-calcium, moderately high-carbon fly-ash that also has a high sulfate content, all of which tend to make it unusable in existing fly-ash applications. The plant therefore must currently dispose of it by landfilling. The Edwards plant produces fly-ashes containing 6.87% carbon and 11.11% carbon which have low calcium and sulfate contents compared to ADM fly-ash, but their carbon contents are still high enough to make them unusable in existing fly-ash applications. The compositions of the ashes studied are given in Table 1.

The sand used in these experiments was a standard foundry grade with an AFS grain fineness number of 55, which was provided by a midwestern sand producer at no charge to the project. This is a sand that is currently being used by the Caterpillar foundries, as well as many other foundries in the U. S.

Fly-ash binders were combined with the sand, and the mixture was formed into standard forms for testing. A summary of the tests and their purposes is given in Table 2.

In this project, it was determined that compressive strength and shear strength of foundry sand/binder mixtures were strongly affected not only by the dosage of fly-ash, but also by the moisture content, and by the dosage of calcium chloride which was added as a curing accelerator. It was also found that the quality of the bonded sand specimens could be greatly improved by reacting the fly-ash with calcium chloride and water for 20 minutes before mixing it with the sand, and by allowing the specimens to cure at room temperature for 1 hour before testing for compressive and shear strength.

Table 1:  Chemical analyses of the Illinois fly-ashes being used, with the analysis of a typical western bentonite included for comparison.

Compound	ADM Fluid Bed Combustor	Edwards Dry Low Carbon	Edwards Dry High Carbon	Western Bentonite
SiO2	19.62	50.07	48.00	39.62
Al2O3	6.6	24.41	23.87	23.16
Fe2O3	5.04	9.51	9.34	5.49
SiO2+Al2O3+Fe2O3	31.26	83.99	81.21	68.27
CaO	46.19	2.72	2.63	9.63
MgO	3.33	0.85	0.84	2.11
Na2O	0.69	0.61	0.55	1.06
K2O	0.81	2.68	2.61	0.39
TiO2	0.43	1.22	1.25	1.25
MnO2	0.13	0.05	0.05	0.01
P2O5	0.72	0.09	0.12	2.2
SrO	0.08	0.15	0.14	0.39
BaO	0.05	0.16	0.16	0.53
SO3	12.53	0.61	0.54	5.93
Loss on Ignition	3.08	6.87	11.11	13.24

Experiments were then conducted to determine the effects of variations in moisture content on the sand specimen properties. The ADM fly-ash was added at a constant dosage of 10% of the sand weight in all the tests where it was used. The Edwards fly-ashes were added at a constant dosage of 6% of the sand weight, along with supplemental calcium in the form of either calcium oxide (at a dosage of 4% of the sand weight) or calcium hydroxide (at a dosage of 5.28% of the sand weight). All experiments included calcium chloride as a curing accelerator, added at a dosage of 3% of the sand weight. The fly-ash results were compared to the baseline results obtained earlier using bentonite binder added at a dosage of 6% of the sand weight, with 2% moisture.

Permeability was found to be a strong function of the moisture content, with the highest permeability achieved at a moisture content of approximately 3.5% for the ADM ash, and at 4% for the Edwards ash. The ADM ash reached a maximum permeability that was comparable to that of the bentonite binder, while the permeabilities for the Edwards ash were significantly lower.

Table 2: Description of tests used to evaluate the properties of molded sand test samples. These are standard tests for foundry sands, recommended by the American Foundrymen's Society (AFS)

Test	Procedure	Use of Data
Compressive Strength (Green)	A compressive load is applied to a cylinder of compacted sand until the cylinder fails.	Determines the maximum compressive stress a mixture is capable of sustaining when rammed into a mold.  The ability of a mold to hold together is dependent on this parameter.
Permeability	Air if forced through a cylinder of compacted sand inside a metal sleeve, and the permeability determined from the flowrate and pressure drop.	Measures the flow of gasses through a porous media such as compacted or cured molding sand.  Venting qualities of a sand mold are dependent on this quality.
Moisture Content	a 50g sample of uncompacted sand is placed in a hot air dryer for seven minutes at 110 degrees C to remove the moisture from the sand.  Moisture is determined from the loss in weight of the sample.	Used in determining the amount of water needed to properly make a sand mold hold a form without interfering with any of the other mold properties.
Shear Strength (Green)	A cylinder of compacted sand is placed in a pendulum type strength machine and a diametric load is applied tot he opposite halves until the specimen shears.	Determines the maximum shear stress a sand mixture is capable of sustaining when rammed into a mold.  The ability of a mold to hold pattern details is dependent on this parameter.
Compactability	Sand is de-lumped through a 6.35 mm screen into a sample cylinder, any excess sand is removed, taking care not to compact the sand.  The specimen tube is compressed by a 140 psi load.  The percent change in height is used to determine compactability.	Determines how much a sand will compact when a load is applied to it.  If the sand has poor compaction properties, it will be difficult to ram into a mold, and may deform when metal is poured, producing a poor-quality casting.

Increasing water addition also increased the compressive strength. The compressive strength with no curing time reached its maximum at 4% moisture for the ADM ash and the Edwards ash with calcium hydroxide, and at nearly 5% moisture for the Edwards ash with calcium oxide. Before curing, the Edwards ash with calcium oxide showed a higher strength than the Edwards ash with calcium hydroxide, indicating that calcium oxide is a better choice for a calcium source than calcium hydroxide. After providing a one hour curing time, the compressive strengths increased for all of the binders tested, which all equalled or exceeded 50 psi between 4.0% and 4.5% moisture.

Shear strength of the test specimens was too low to be reliably measured for all of the fly-ash bonded specimens when no curing time was provided. However, when a one hour curing time was used, the shear strength increased. At moisture contents greater than about 3.0 - 3.5%, the cured shear strength was higher than the shear strength obtained with a bentonite binder.

From these studies, it was concluded that the fly-ash binders require moisture contents of 3.5 - 4.0% for the best permeability, and 4.0 - 5.0% for the best compressive and shear strengths. Experiments were also conducted to compare the temperature resistance of fly-ash-bonded sand specimens to that of bentonite-bonded sand specimens. Upon heating to 600C and 1000C, the fly-ash-bonded specimens maintained their strength, without premature melting problems. The results were comparable to those seen using a bentonite binder.

The Class F fly-ashes used in this project had carbon contents of 6.87% and 11.11% by weight. Using these ashes, experiments were conducted to determine the effects of carbon content on the properties of the bonded sand. The carbon content was shown to have no harmful effect on the sand specimen properties, which indicates that fly-ashes with a broad range of carbon contents can be readily used in this application.

Fly-ash binders have been demonstrated to have properties suitable for use as foundry sand binders. With proper addition rates and handling, they can produce bonded sand properties comparable to those produced with conventional bentonite binders. The prospects for successful use of fly-ash binders to produce actual cast metal parts in foundry operations are therefore excellent.