Evaluating an infrared reclaimer for clay-bonded sands.

Of the foundrys and reclamation processes currently available, thermal systems are particularly suitable for the treatment of sands bonded with organic materials. An additional important use of thermal sand reclamation involves heating the clay in clay-bonded sands (green sand) to a high temperature to render it inactive or dead.' This process, called calcining, is particularly important for sand reclamation systems that would be required to process the various types of sands used in a given foundry.

After calcining, the clay is removed by a mechanical scrubbing process, but because the clay is inactive, the scrubbing can occur at lower energy input levels and is much more efficient and complete.

While thermal systems can use either gas or electricity for the calcining stage of the operation, thermal reclaimers using electric power offer some unique advantages over gas-fired units. The electric units do not require large quantities of air to support combustion, and because fuel is not burned in the reclaimer, contamination of the sand with products of combustion is not possible. Complete reclamation of organically bonded sand can be achieved at lower temperatures (500C or 932F), minimizing thermal shock cracking and reducing heat loss. Further, lower-temperature operation avoids the unwanted silica phase change from quartz to cristobalite.

Additionally, the reclamation process can be programmed to integrate with plant power demands to improve load factor and avoid increasing peak demand.

A disadvantage of operating at the lower temperature, however, is that some of the volatile organic compounds driven off the sand may not be fully burned in the reclamation unit proper. In these cases, an afterburner may be required to meet local air quality standards. Some electric technologies such as resistance units do not easily reach temperatures sufficiently high (871C or 1600F) to treat green sand.

Existing electric thermal reclamation units use resistance heating elements and fluidized-bed technology to process the sand. This results in a low throughput system designed to operate at relatively low temperatures on a continuous basis. These systems do not reach a temperature high enough to calcine the clay used with green sand and, therefore, are only suitable for organically bonded sands.

The EPRI Center for Materials Production has recognized the desirability of developing a thermal sand reclamation system that would operate at high temperatures and at high throughput levels suitable for processing both organically bonded and green sands. CMP has funded a project to evaluate the potential for thermal processing of foundry sand using the following energy sources:

* resistance

* plasma

* microwave

* infrared

* radio frequency

The results of these evaluations and the development of an infrared thermal reclamation system are described in this article.

Phase One

The first phase of the project consisted of thermally treating actual foundry sand samples using the electrically powered heat sources described above to determine if any of the methods proved to be particularly suitable for this process.

In all cases, the goal was to be able to treat the sand at the higher temperatures 1300-1600C or 2372-2912F) required to calcine or kill the clay in the clay-bonded sand. If this was accomplished, the system would automatically be suitable for the treatment of the organically bonded sand.

Three types of foundry sands, each using different types of clay, were processed to determine the effectiveness of the heating processes. A steel foundry sand bonded with sodium (Western) bentonite, an aluminum sand bonded with calcium (Southern) bentonite, and a gray iron sand bonded with a combination of calcium and sodium bentonites were treated with all of the listed heating processes. Each of the different clay bonds has a different deactivation (calcining) temperature, and any differences in behavior were noted.

Sand Testing

The treated sands were all evaluated using standard American Foundrymen's Society (AFS) sand tests. The tests were performed at the AFS Laboratories in Des Plaines, Illinois, by M.J. Granlund, coauthor of this report.

The tests used to evaluate the sands were:

Methylene Blue Clay, which determines the amount of clay remaining, before and after treatment, that still has the ability to develop an effective clay-water bond. A properly reclaimed sand should exhibit a value of 0%.

AFS Clay, which determines the total amount of fines and water- absorbing materials in the clay. While the methylene blue clay measures active bonding capacity, the AFS clay measures the total of calcined (dead) clay and other fines and indicates any problems with rebonding that may be anticipated. A value of 0.4% or less indicates adequate reclamation.

Loss on Ignition (LOI) measures the amount of gas-forming material in the sand. High levels of LOI indicate that considerable amounts of organic binder additives remain in the sand; therefore, reclamation has not been effective. A value of 0.10% or less indicates adequate reclamation.

Acid Demand Value (ADV) measures the amount of alkaline compounds remaining on the sand. These compounds may adversely affect the curing of organically bonded sands, and ADV is a measure of sand cleanliness. The ADV is a property of the base sand used and should not be changed by the reclamation process.

In addition, sand screen analyses were performed on a selected basis to ensure that the sand screen distribution was not significantly changed by the thermal treatment.

System Development

Based on the results of the screening phase of the project, it was decided that further work should concentrate on the development of the infrared/ fluidized bed system, which is based on a proprietary infrared heat treating process. In concept, the system consists of the following segments:

* lump breaker and

metal removal

screens/magnets

* thermal processing

unit

cooling section

* scrubber/classifier

The system is illustrated schematically in Fig. 1.

Design items of special interest are the patented high-intensity infrared elements that are immersed in the sand bed which is fluidized by air. This intimate contact provides excellent thermal efficiency. The fluidizing air can be preheated in the cooling section to improve thermal efficiency. The thermal unit size is very compact and does not require a refractory lining. Quantities of fluidizing air are minimal, thus requiring a minimum of gas handling and cleaning equipment. Casting Trials

To validate the previous test results with actual foundry tests, several barrels of a typical gray iron foundry sand were obtained from Northern Iron, Inc., St. Paul, Minnesota. The sand was bonded with a combination of sodium/ calcium bentonite and had a relatively high LOI value. It was felt that this sand was representative of the majority of iron foundry operations and was selected to give the reclamation system a severe test of its capabilities. Samples of the new sand used by Northern Iron were also obtained to serve as a control.

To verify the preliminary laboratory tests, samples of this sand were treated at various temperatures from 482-871C (900-1600F) in a batch mode. Typical results are represented in Table 1. The results indicate that processing at temperatures of 760C (1400F) and above will lower the active clay, total clay and LOI values to those comparable to new sand. Microscopic examination of the thermally processed sand revealed that a thin shell or coating of calcined clay remained on the grains that could be detrimental to the use of the sand with organic binders. This shell was removed with a simple mechanical scrubber following conventional mechanical reclamation technology.

While the results of sand laboratory tests are of interest to the sand technician, the real proof to a foundrymen comes from making actual castings and examining their surface quality. To this end, several 55-gallon drums of the Northern iron waste sand were processed in the unit at 871C 1600F) followed by mechanical scrubbing. The processed sand, along with the new sand control, were then shipped to the casting laboratory of the AFS Cast Metals institute (CMI) in Des Plaines for casting trials.

The CMI staff, under the supervision of Granlund, made and poured two different types of test castings using both new and reclaimed sand. Both molds and cores were made using a phenolic urethane organic binder system at a 1.7% binder plus catalyst level.

The first type of test casting was the standard "Gertsman" test, which used 2-in. diameter by 2-in. long test cores to evaluate the thermal stability and reactivity of the sand. The second test casting used cylindrical castings with a 2-in. inside diameter and 3.25 in. outside diameter. The castings are 3.25 in. long. This is a standard test for the performance of organically bonded sands. Both test castings have been extensively used in the industry and allow direct comparison to sand test results reported in the literature. The test castings also provide an excellent evaluation of sand cleanliness. All castings were made blind with code numbers to ensure the elimination of bias in evaluating of the casting surfaces.

All castings were rated byv several AFS/CMI personnel and the coauthors of this report. No significant differences in castings surface quality were observed. The conclusion from these tests is that the infrared reclaimer processes a sand that will perform as well as new sand when compared on the basis of sand properties after treatment and surface quality of castings produced using the processed sand.

Nobake Trials

In another trial, several barrels of wastes and from a nobake steel foundry, Minncast, Minneapolis, Minnesota, were processed in the infrared unit and returned for casting production. This foundry, which uses a furan organic nobake system for both molds and cores, operates a mechanical reclamation system. The report from Minncast is that the reclaimed sand is visually cleaner than their own mechianically reclaimed sand and produces better castiiig surface quality. Testing with this found continues to determine the optimum time/temperature treatment cycle.

Further Development

Based on the successful preliminary trials reported above, the project continues with two parallel efforts. A pilot unit is being constructed in a laboratory that will continuously process about 500-1220 lb of sand per hour, depending on operating temperature. The unit is properly instrumented to evaluate and develop optimum operating parameters for different types of sands. This unit will be available on a contract basis to foundries to have their own particular sand processed and returned for plant casting trials.

Parallel with the construction of the above pilot unit, a full-scale, 1.5 tph industrial unit will be installed at Northern Iron, Inc. An economic and energy evaluation of the unit will be conducted by CMP.

This article has been excerpted from the report "Electrotechnologies for Sand Reclamation An Overview and Evaluaof Infrared (No. 91-7) published by the EPRI Center for Materials Production and is available through CMP, 4400 Fifth Ave, Pittsburgh, PA 152132683: telephone: [phone].