Patent Application
20110108232
The present invention relates to a binder system for binding particles together, and in particular, to binding system containing a collagen or lignin and an inorganic material.
The price increase of energy sources and raw materials in recent years is known. For example, iron foundries with cupola furnaces use foundry coke as a conventional fuel and the price of the foundry coke increased by about 450% from the year 2002 to 2008.
In typical iron cupola furnaces, for every 1000 pounds of scrap iron charged into the furnace, approximately 100 pounds of coke, 1 pound of silicon, 20 pounds of limestone, along with other additives, are added or charged into the furnace[1]. The overall dwell time for charged materials is typically 40-80 minutes which avails about 20-40 minutes in a preheat zone where the material experiences temperatures in the range of 1200-1400° C. in a reducing/starved-air (pyrolysis) environment.
Before coke is provided to the cupola furnace, it is produced by pyrolysis of special coking bituminous coals [2-4]. Once produced, the coke has desirable chemical and physical properties suitable for burning in the cupola furnace. For example, the coke has a fused porous carbon structure that affords for a fast burning fuel bed to be developed above the molten iron within the cupola furnace[4].
One draw back from using foundry coke is the fact that only a small fraction of mineable coal resources offer desirable coking properties. In addition, the manufacture of coke consumes large amounts of energy. As such, research on a coking replacement has been conducted, but as of yet achieved limited success, and coke is still the most used fuel in cupola furnaces.
It is of interest to note that before the development of coke, anthracite was widely used as a fuel source in cupola furnaces. Anthracite offers similar chemical properties as coke, however, the combustion rate of chunks or pellets of anthracite is relatively slow due to the materials condensed structure. The combustion rate of anthracite chunks can be increased by reducing chunk size[1], however if anthracite pieces become too small, they can be blown out of the furnace via the strong tuyere air flow before adequate or complete combustion can occur. Indeed, there are minimum size specifications for solid fuels used in iron cupola furnaces.
As such, anthracite fines represent a heretofor unusable residue from anthracite mining processes and are often left as waste in collection ponds, valleys, etc., of coal mining regions. Although organic binders have been used to bind anthracite fines into bricks, heretofor developed or employed organic binders have not been able to survive the high temperature conditions in the cupola furnace. Therefore, there is a need for a binder to bind particles such as anthracite fines, to make pellets, bricks, and the like such that the pellets, bricks, etc., can withstand high temperatures and mechanical forces experienced in cupola furnaces.
In addition to the need for a foundry coke alternative, the metal casting industry extensively uses sand molds and sand cores made from silica sand and organic binders, with conventional binders used for the manufacture of sand cores emitting up to 70% of a foundry's Volatile Organic Compound (VOC) emissions[5].
A sand core can containing between 1 to 3% of a binder with the remaining portion being sand. The binder must provide a core with adequate strength at room temperature and structural integrity when exposed to molten metal. Paradoxically, after withstanding molten metal thermal exposure and subsequent cooling, the binder must then breakdown during shake-out so that the sand can be easily removed from cavities of a cast product.
Several petroleum based binder systems in use today, such as phenolic urethanes and furans, provide the desired structural integrity, shakeout propensity and casting quality. However, the phenolic urethane and furan binders breakdown after metal solidification and can emit between 30 to 70% of a foundry's VOC's and Hazardous Air Pollutants (HAP's) [5,6]. Furthermore, VOC and HAP emissions increase proportionately with an increase in core-loading[5,7,8].
Researchers have shown that VOC emissions for a collagen core binder are 70-95% less than for a phenolic urethane binder[6,9]. Collagen, which is available in large quantities as a by-product of the meat-packing industry, is a fibrous triple helical protein that strengthens skin, tendon and bones, and has a tensile strength greater than steel. However, despite significant emission reductions enabled by collagen, the material has been marginally employed in foundries due to its inability to withstand high temperatures imparted by molten iron (1290° C.-1510° C.) at the molten metal/sand interface. Specifically, breakdown of collagen at high temperatures results in erosion of the core and costly casting defects. As such, there is also a need for a binder for sand cores that can withstand the harsh environment of metal casting and yet has lower VOC emission in comparison to phenolic urethanes and furans.
The present invention discloses a binder material for binding a plurality of particles together to form a conglomerate such as a carbon-containing briquette, a sand casting core and the like. The binder material can include a first component that can contain a collagen and/or lignin and a second component that can contain a metal, a metalloid and/or an inorganic compound. In some instances, the binder material can be used to make a composite material. The composite material can include the plurality of particles and the binder material. The binder material affords for the plurality of particles to be bound together into a desired shape, the desired shape having desirable properties for a particular use of the composite material.
The plurality of particles can be a plurality of sand particles, i.e. grains of sand, or in the alternative, a plurality of carbon-containing particles such as coal particles, coke particles, anthracite fines, anthracite fines that have been ground into a powder, and the like. The plurality of sand particles can be bound together to form a casting component, for example a sand casting core, a sand casting mold, etc. The plurality of carbon-containing particles can be bound together to form a fuel pellet which may or may not have a plurality of silicon carbide nanowires as part of the fuel pellet structure/morphology. In some instances, silicon carbide nanowires nucleate and grow when the fuel pellet is heated above 1200° C. in a reducing and/or pyrolysis gaseous atmosphere.
The first component can contain denatured collagen and the second component can contain inorganic particles such as particles of an amorphous silica-rich ash, silicate, ferrosilicon, silicon, combinations thereof and the like. In the event that the particles include a silicate, the silicate can be an alkaline metal silicate, alkaline earth metal silicate, combinations thereof, etc. In some instances, the second component provides a source of silicon which can react with the carbon-containing particles to form silicon carbide, which may or may not be in the form of nanowires. In addition, the nanowires can assist in binding of the plurality of component particles together to form the component. The binder material can also include an organic component, e.g. a biomatieral and/or a carbohydrate. The carbohydrate can be a saccharide which may or may not include a monosaccharide such as glucose, fructose, galactose, xylose, ribose, etc.
A process for making a conglomerate using the binder system can include mixing the plurality of particles with the binder material and forming or molding the mixture into a desired shape. In some instances, the binder material is pre-heated before mixing with the plurality of particles. Thereafter, the desired shape can be dried and may or may not be exposed to ultraviolet radiation that can assist in curing of the shape.
The present invention discloses a binder system, also referred to as a binder material or simply as a binder, for binding together particles such as anthracite fines, silica sand grains, and the like. As such, the present invention has utility for making a conglomerate such as a carbon-base briquette or a sand casting core. For the purposes of the present invention, the term conglomerate is defined as shape, part, tool and the like made from at least two different materials, wherein the materials adhere to one another. The carbon-containing pellet can be used as a fuel source in a cupola foundry furnace, a steel making blast furnace and the like, or in the alternative, the sand casting core can be used in a molten metal casting process. For the purposes of the present invention, conglomerate and composite material are used interchangeably. In addition, conglomerates include cores, sand casting cores, carbon-containing briquettes, briquettes, pellets, and bricks. The terms cores and sand casting cores are used interchangeably, as are the terms carbon-containing briquettes, briquettes, pellets, and bricks.
The binder material can include a first component that can contain collagen and/or lignin and a second component that can contain a metal, a metalloid and/or an inorganic compound. The collagen may or may not be denatured and can be mixed with an aqueous solution that can further be mixed with the second component and/or a plurality of component particles. Other additives can be included in the binder material, for example and for illustrative purposes only, organic compounds that include biomass, biomass derived material, carbohydrates such as saccharides, monosaccharides and the like. For the purposes of the present invention, the term biomass is defined as living or dead vegetation or vegetation derived matter, animal, and/or animal derived material, excluding petroleum and petroleum products.
The binder material can be mixed with a plurality of particles to make a composite material that can be formed and/or molded together into a conglomerate having a desired shape, the conglomerate having desirable properties. For example and for illustrative purposes only, the conglomerate can be a fuel briquette or brick, also referred to as a fuel pellet herein, that can be used in a cupola furnace. The fuel pellet can be made from a mixture of the binder material and waste anthracite fines, and the pellet can replace at least part of the foundry coke used by the cupola furnace. The fuel pellet can also exhibit mechanical properties that afford for the pellet to be stored, shipped and/or fed into an industrial cupola furnace without excessive breakage and/or burning.
The plurality of particles used to make the conglomerate can have different shapes, sizes, chemical compositions, mechanical properties and the like. In some instances, 50% by weight or mass of the plurality of particles bindered by the binding material in order to make a conglomerate have an average mean diameter of less than 0.25 inches. In other instances, the plurality of particles are sand particles that have an American Foundry Society (AFS) grain fineness number between 25 and 150, inclusive.
The conglomerate can also be a sand casting core made from the binder material and a plurality of sand grains/particles. The sand core can exhibit adequate and/or desired high temperature strength, liquid metal erosion resistance and shake-out qualities. In addition, the sand core can exhibit or provide a significant reduction in volatile organic compounds (VOC) emissions and hazardous air pollutants (HAP) when compared to traditional binders such as a phenolic urethane, furan, etc.
The second component can contain a silicate, ferrosilicon, silicon, combinations thereof and the like. In the event that the second component contains a silicate, the silicate can be an alkaline metal silicate, alkaline earth metal silicate, an amorphous silica-rich ash and combinations thereof. For example and illustrative purposes only, the silicate can be sodium silicate, potassium silicate, lithium silicate, magnesium silicate, calcium silicate, kaolinite, and combinations thereof. In addition, the amorphous silica-rich ash as can be an ash derived from rice hulls and rice stalks. In some instances, the second component provides a source of silicon which can react with carbon-containing particles and form silicon carbide, which may or may not be in the form of nanowires. In addition, the nanowires can assist in binding of the plurality of particles together to form the conglomerate and provide desirable mechanical properties thereto.
A process for making a conglomerate using the binder system can include mixing the plurality of particles with the binder material and forming or molding the mixture into a conglomerate having a desired shape. In some instances, the binder material is pre-heated before mixing with the plurality of particles. Thereafter, the conglomerate can be dried and may or may not be exposed to ultraviolet radiation that may assist in curing of the conglomerate.
In order to better teach and/or disclose the present invention, and yet in no way limit the scope thereof, examples of at least two embodiments are provided below.
Number 5 (#5) anthracite fines (hereafter referred to simply as “anthracite fines”) were obtained from Jeddo Coal Company located in Wilkes-Barre, Pa. A sieve analysis of the anthracite fines illustrated a particle size between US mesh # 10-80 (2000 to 177 μm). In some instances, the anthracite fines were crushed into powders that could pass through a U.S. mesh #100 (150 μm) sieve (hereafter referred to as “anthracite powder”).
Chemical analysis and heat content of the anthracite fines were compared with a typical foundry coke used by Ward Manufacturing in Blossburg, Pa., with results shown in Table 1 below. As shown in the table, the dry heat content of the anthracite fines was approximately the same as for foundry coke.
A dry granular collagen-based binder was provided by Entelechy, representing Hormel Foods Company in Austin, Minn. Fructose (98%-102%) was obtained from Mallinckrodt Baker, Inc., in Phillipsburg, N.J.
Three different lignins were acquired and tested. The first lignin was low sulfonate alkali lignin purchased from Sigma-Aldrich, the second was a hardwood lignin extracted from Kraft black liquor via sulfuric acid precipitation, and the third was a hardwood lignin extracted via the Lignoboost™ process which extracts lignin from black liquor in part by lowering the pH of the liquor with CO2.
Three forms of silicon having three different redox levels were obtained and tested. Metallic silicon (Si) having zero valence and in the form of 10 cm or less lumps was purchased from Alfa Aesar in Ward Hill, Mass. The lumps were crushed into powders of less than #100 mesh before being mixed with the anthracite fines and/or anthracite powder. Amorphous Si having +4 valence and in the form of a sodium silicate (Na2SiO3) solution containing 42.5% by weight of dry solids was provided by J.B. DeVenne Inc., in Berea, Ohio. And crystallized Si having +4 valence in the form of kaolinite powder (KGa-1, Al2Si2O5(OH)4) was obtained from Washington County, Ga.
The anthracite fines were dried at 105° C. overnight to remove moisture and 1 g of collagen binder was dissolved in 12 g of water at 70° C. to form a gelatin solution. In some instances, fructose was added and dissolved in the water plus collagen binder solution. A Si source, either the metallic Si or the kaolinite powder, was mixed with the anthracite fines, or in the alternative, the Si-containing liquid solution (sodium silicate) was added to the gelatin solution. Thereafter, the anthracite fines, with or without the Si source depending on which form of Si was being used, were mixed with the gelatin solution without or with the Si-containing liquid solution, respectively. The final mixture of anthracite fines and gelatin solution was then packed into a cylindrical mold (2.86 cm diameter×4.76 cm long) with 275 kPa (40 psi) pressure applied on both ends. Finally, a conglomerate in the form of a pellet was extruded from the mold and cured under ambient conditions with at least three anthracite pellets produced from a single batch or final mixture. During curing, evaporation released approximately 10 g of the initial 12 g of water.
Mechanical strength of the pellets at room temperature was determined using unconfined compressive strength testing and drop shatter testing[10]. The unconfined compressive strength of anthracite pellets was determined using a Simpson-Gerosa electronic universal sand strength machine. A horizontally moving arm applied pressure on a pellet until failure. Final compressive force was calculated based on the diameter of the original pellet sample, assuming negligible shape change during the test. The sample preparation and unconfined compressive strength procedure disclosed herein are defined as an unconfined compressive strength test protocol.
The drop shatter test was a standard breakage test used for foundry coke with results often referred to as an anti-breakage strength. The anthracite pellets were dropped onto a steel plate through a 1.83 meter (m) long polyvinylchloride (PVC) tube that had an inner diameter of 3.8 cm. A total of 10 drops for each pellet was performed with a first end of the pellet pointing in the down direction for a given drop, followed by a second end of the pellet pointing in the down direction for the next drop. After the tenth and final drop, the weight of the largest remaining piece of the pellet was obtained. In addition, if no single piece having a mass larger than 50% of the original weight of the pellet remained after a drop, the test was stopped and 0% was recorded for the test. The sample preparation and drop testing procedure disclosed herein are defined as a drop test protocol.
Pyrolysis of the anthracite pellets was conducted in a horizontal tube furnace or a vertical tube furnace. When using the horizontal tube furnace, a pellet was placed in an alumina tube extending through the furnace and a slow nitrogen gas flow of approximately 2 standard cubic centimeters per minute was used to prevent the anthracite from burning. In addition, a three-step pyrolysis procedure was employed with the furnace ramped at 3° C./min up to 1400° C., holding at 1400° C. for 2 hours, and then furnace cooling to room temperature at 3° C./min. The pyrolyzed anthracite pellets were then removed from the tube-furnace for further tests. Unless otherwise identified, the prescribed maximum temperature was 1400° C., which is generally close to preheating zone temperatures of a cupola furnace. This pyrolysis procedure is defined as a horizontal furnace pyrolysis protocol.
For pyrolysis in the vertical tube furnace, the furnace was heated to a predefined temperature with a laminar flow of nitrogen gas provided up through a vertical alumina tube extending through the furnace. Thereafter, a 5 g sample was placed in a basket and descended into the heated tube. A thermocouple was used to record the temperature of the region where the sample was located and it was determined that the sample region increased from 30-40° C. to 800° C. within about 2-3 minutes, where the temperature remained for the prescribed time. This pyrolysis procedure is defined as a vertical furnace pyrolysis protocol.
Burning rates for the anthracite fines and foundry coke were determined by measuring the time required for complete combustion of approximate 5 g pieces of anthracite pellets and foundry coke. Specifically, a piece of an anthracite pellet or foundry coke was placed in a 5 cm diameter quartz tube furnace and heated to 1050° C. under an inert and/or reducing atmosphere. Thereafter, the piece was held at 1050° C. within the tube for between 2-3 minutes and then air at about 2 L/s was passed up through the tube and into contact with the material. The air thus provided oxygen for burning of the piece and a time was recorded for when combustion was no longer observed.
Scanning electron microscopy (SEM) analysis was performed using a FEI Quanta 200 Environmental SEM with a Gaseous SE detector, a voltage of 20 kV and an electron beam spot size of 4 nm. A transmission electron microscope (TEM, Model 2010, JEOL, Tokyo, Japan) was also used to determine material morphology, particle size and electron diffraction.
Ambient temperature XRD patterns were obtained using a PANalytical X'Pert Pro MPD diffractometer and powders of the anthracite pellets were obtained by crushing a pellet with a ball mill. It is appreciated that the XRD patterns provided a bulk measurement of the material.
Anthracite pellets were prepared with an array of compositions and protocols. For example, anthracite pellet compositions included 100 g of anthracite fines, 1 g of collagen, 0 to 1 g of fructose, 0 to 10 g lignin and 0 to 10 g of silicon, kaolinite or sodium silicate. Results of unconfined compressive strength tests as described above for anthracite pellets having different compositions, excluding lignin, are shown in Table 2 below. As shown by the data in the table, the use of collagen and/or fructose as a binder material provides strength to the pellets before pyrolysis and as such can provide strength at lower temperatures that would be experienced by anthracite pellets during shipping, storage and/or initially entering a cupola furnace. The data also show that pellets made with 10 g silicon metal powder achieved the most favorable unconfined compressive strengths: 1300 kPa before pyrolysis and 690 kPa after pyrolysis. In addition, 1% fructose increased the strength of 10 g silicon powder pellets even more. In contrast, the 10 g sodium silicate addition offered considerable unconfined compressive strength (2344 kPa) before pyrolysis, but only 28 kPa after pyrolysis. Pellets made with kaolinite exhibited negligible strength after pyrolysis. It is of interest to note that for all three Si sources, less Si corresponded to lower strengths.
Regarding the addition of lignin to the binder material,
Looking now at
When 10 g of kaolinite powder was bindered to 100 g anthracite fines with 1 g of collagen, the kaolinite XRD peaks that appeared before pyrolysis were converted mostly to mullite. This kaolinite-anthracite combination also did not yield silicon carbide.
In contrast, when collagen bindered anthracite fines included sodium silicate as the Si source, pyrolysis of the material yielded beta silicon carbide (3C—SiC) (
The formation of SiC increased when 10 g of silicon metal powder was bindered to 100 g anthracite fines with 1 g of collagen and pyrolyzed (
Turning now to
When sodium silicate was the additive, the pyrolyzed products exhibited sponge-like ceramic structures (
In contrast, the 10 g silicon powder additive resulted in a labyrinth of SCNWs. The nanowires were generally 20 μm in length with lengths up to 100 μm observed. Per visual inspection, the nanowires altered the color of the pellets from the black of anthracite to the light green of the nanowires. In addition, the SCNWs were observed to have formed after heating of the anthracite pellet above 1200° C. for times as low as 6 minutes and as such can provide improved strength at higher temperatures that would experienced by anthracite pellets fed into a cupola furnace. As such, the SCNWs were formed and/or present when the anthracite pellets were pyrolized above 500° C., e.g. above 900° C., 1000° C., 1100° C. and/or 1200° C.
The nanowires were typically 20-50 nm in diameter and were coated by an amorphous thin layer (about 2 nm thick) of silicon oxide as shown in the high resolution TEM (HRTEM) image in
The effect of anthracite particle size on anthracite pellet binding strength was also examined. With a smaller particle size, e.g. using anthracite powder to make a pellet, gaps between anthracite particles were reduced and SiC nanowires were able span between adjacent anthracite particles.
SEM images in
Turning now to Table 3, a comparison of burning rates for different compostion anthracite pellets and foundry coke at 1050° C. is shown. It is appreciated from the data that pieces of anthracite pellets had burn rates generally equivalent to pieces of coke. In contrast, unbound anthracite fines burned considerably faster.
Full scale testing of anthracite fine bricks made from anthracite fines, collagen binder, fructose additive, and silicon powders was performed by feeding the bricks into an industrial foundry cupola. In total, 500 pounds of anthracite bricks generally measuring 5.75 inches (14.6 cm) in diameter, 2.25 inches (5.7 cm) in height and weighing 1.8 pounds (0.82 kg) were prepared.
The testing included feeding the bricks into the cupola for about one hour at a rate that replaced 10% of the coke that would have otherwise been used (
When the anthracite bricks were dropped into the bucket, no breakage was observed as per visual inspection. Breakage was also not observed, as per remote video camera, when the bricks were conveyed into the top of the cupola. As the anthracite bricks descended through the cupola melt zone and past the tuyere windows, 10 to 12 shapes that exhibited a tell-tale smooth curvature of a 14 cm circle were observed and it appeared that a number of these shapes were anthracite bricks. None of the bricks had broken into pieces by the time they passed the tuyere windows and the bricks appeared to be burning throughout their structure while in the melt zone, as were chunks of coke.
The product or resulting metal quality and chemical composition produced by the cupola furnace, along with cupola temperatures and pressures, were analyzed during the full-scale demonstration. The metal product quality and chemistry parameters were within experimental error, although back pressure and temperature (represented by the upper-stack temperature) of the cupola increased slightly during the addition of the anthracite bricks. It is appreciated that the increase in back pressure and temperature could have resulted from the anthracite-silicon bricks burning faster than the coke. However, it is also appreciated that cupola operation is complicated and affected by many factors, and as such, no conclusions could be drawn relating the changes in temperature and back pressure to the addition of the anthracite bricks.
Samples of binder material and sand casting cores made with the binder material were prepared for thermogravimetric analysis (TGA), dynamic mechanical analysis (DMA), differential scanning calorimetry (DSC) and hot distortion testing (HDT). In general, TGA and DSC samples contained no sand and were prepared with only the binder components listed in Table 4. DMA and HDT samples were prepared with sand and binder. DMA samples were molded into gum-stick sized core samples generally measuring 12 mm×5 mm×35 mm and HDT samples were molded into candy bar sized samples generally measuring 15.24 cm×2.54 cm×1.27 cm.
A modified sodium silicate solution containing 42.5% by weight of dry solids was provided by J.B. DeVenne and Associates in Berea, Ohio. Collagen in the form of dry powder was provided by Entelechy in Canton, Mich. A phenolic urethane binder that was heat was provided by Ashland Chemical Company in Columbus, Ohio. The phenolic urethane binder came as a two-part binder (Part A and Part B) that cured when mixed and heated. Potassium silicate was provided by PQ Corporation in Malvern, Pa. and came as a liquid with 29% by weight solids content. Silica sand having a specified American Foundry Society (AFS) AFS grain fineness number ranging from 67-73 was provided by Wedron Silica Company in Wedron, Ill.
In one embodiment of a binder material/system, 10 g of collagen was added into 20 mL of modified sodium silicate with 5 grams of triglycerides, and the entire mixture was heated to 70° C. and vigorously stirred for 5 minutes. The hybrid mixture was then poured into an aluminum pan and dried at 110° C. for 12 hours. Phenolic urethane samples were prepared with 55 parts A to 45 parts B, as is common in industry practice, and then dried at 110° C. for 24 hours. Another hybrid binder material was prepared by mixing 10 g collagen with 2.9 grams of K/Li silicate solids, the mixture also dried at 110° C. for 12 hours.
For TGA tests, a small portion of dried binder (10 to 20 mg) was removed with a scalpel and placed in a TA instruments SDT 2960—Simultaneous DSC-TGA instrument. The test commenced at room temperature under an argon atmosphere and was heated at 10° C./min to 600° C. Trials were tested in duplicate with an average between the duplicates taken/used as final TGA results. The sample preparation and TGA procedure disclosed herein are defined as a TGA test protocol.
For DSC tests, binder samples were shaved and finely powderized with a razor blade. DSC analyses used a 2.0 mg sample and employed a TA Instruments TA-Q100 DSC instrument. A hermetic pan was used to hold samples with the pan lid punctured with an 8 penny finish nail to prevent the pan from ‘jumping’ from the heating element during analysis. Samples were purged with nitrogen gas at 50 mL/min. Two tests were performed using a temperature ramp of 10° C./min to 510° C. A third test was performed using a temperature ramp of 10° C./min to 510° C., followed by cooling at 20° C./min to 50° C. and then ramping at 10° C./min to 510° C. a second time. The sample preparation and DSC procedure disclosed herein are defined as a DSC test protocol.
For DMA and HDT experiments, bindered cores were made by mixing 1 to 3% binder material with 99 to 97% silica sand (Table 4). Bindered core samples were prepared by mixing 1.0 kg of washed foundry silica sand with a given binder material for 90 seconds at Level ‘2’ Solid State Speed Control, in a Commercial Kitchen Aid Mixer, Model Number KSMC50S.
For DMA analyses, the mixed material was hand packed into an 8-part wooden core mold that produced samples sizes described above. Samples were then cured at 110° C. for one hour, with the exception of heat cured phenolic urethane bindered samples which cured for 24 hours. The samples were then removed from the core mold and placed in a dessicator for overnight storage.
TEA-cured phenolic urethane samples were provided by the Penn State University HMAC foundry and were produced using 1.1% phenolic urethane cured with TEA gas in a full-scale/industrial operation. The cores of the samples had a 3 inch×1 inch diameter rod protruding from a main core, which was removed and manually filed down to a size that could be used for DMA testing.
For DMA analyses, samples were analyzed using a TA Instruments DMA q800 unit with a dual cantilever clamp. The samples were heated at 10° C./min to 500° C., while a middle support oscillated at a frequency of 20 Hz and a 5 μm amplitude. During testing, the samples were maintained under a nitrogen gas atmosphere flowing at 2 L/min. The sample preparation and DMA procedure disclosed herein are defined as a DMA test protocol.
For HDT analyses, samples were formed into hand packed cores in a 16 sample wooden core box. The bindered cores were cured in the core box at 110° C. for 1 hour, with the exception of phenolic urethane bindered cores which were cured for 24 hours.
The HDT was performed using a custom HDT device that clamped 2.54 cm of a first end portion of the sample onto an iron ledge. The samples were then heated in the center of the exposed 12.7 cm of the sample using a natural gas flame from a 2.54 cm perforated Bunsen burner having an adiabatic flame temperature of 1960° C. and a micrometer detected deformation of the sample at a location 1.27 cm from a second end of the sample. In addition, the micrometer transmitted deformation readings every second to a computer which logged the data. For each binder composition, 10 to 15 samples were analyzed. It is appreciated that the results exhibited inherent variability in both time to given deflection and deflection at failure. The sample preparation and HDT procedure disclosed herein are defined as an HDT protocol.
Molten metal erosion testing was also performed on sand core samples. The sand cores were produced with collagen, sodium silicate, sodium silicate plus collagen, lithium silicate plus collagen (two variations), heat cured phenolic urethane and TEA cured phenolic urethane. The cores were hand packed and had a general final size of 5 in. long×5 in. wide×2 in. thick—with the exception of TEA cured phenolic urethane cores which were 3 in. wide×2.75 in. thick. The cores were placed on a stand leaning at a 45° angle and molten iron at 2620-2720° F. was poured onto the surface of the core from a height of about 8 inches using a constant head core. After pouring the molten metal over a core sample, the metal was captured in a ladle and weighed. The sand cores were then removed and evaluated for erosion and surface characterization.
Core evaluation was performed using core scratch hardness and a micrometer to measure erosion depth. In addition, sand was used to fill any eroded away region in order to quantify erosion volume. The cores were evaluated for core hardness before and after the pour with hardness measured one inch from where molten metal flowed off the core. The erosion depth was measured using Mitutoyo calipers and sand was poured into an erosion cavity, if present, and scraped off using a metal bar. The quantity of sand required to fill the cavity was then measured in order to quantify sand volume eroded during metal pouring. The sample preparation and molten metal erosion testing procedure disclosed herein is defined as an molten metal erosion testing protocol.
Turning now to
In addition to mass retained,
The TGA-MS testing utilized a TA Instruments TGA 2050 coupled to a mass spectrometer (ThermoStar™ 301T, Pfeiffer Vacuum) which measured the gaseous effluent. In addition, an Ashland Isocure Focus(R) WTC Binder having 0.55 g of part 1 and 0.45 g part II per gram of binder was used as the phenolic urethane binder. Conglomerate samples weighing 400 mg were used for the TGA-MS testing and the samples contained approximately 1% organic binder, dry mass basis. The samples were cured by conventional processing technology and were not exposed to temperatures higher than curing conditions. Once cured, a sample was placed in an alumina pan which was subsequently placed in the TGA-MS furnace and heated from room temperature to 1000° C. at a rate of 10° C./min. The furnace, and thus the sample, were purged with 100% ultra-high purity argon gas for a minimum of one hour prior to analysis and for the duration of analysis using a gas flow rate of 90 mL/min.
The MS provided a response measured in nanoamps (nA) as a function of temperature for each atomic mass unit (AMU) analyzed. From the measured nA as a function of temperature data, a plot as shown in
The AMU responses for benzene were taken for an AMU of 78, for phenol an AMU of 94 and for xylene an AMU of 106. As shown in
As such, an inventive binder disclosed herein can provide a reduction in benzene of at least 60%, a reduction in xylene of at least 60% and/or a reduction in phenol of at least 50% compared to the phenolic urethane tested herein. In the alternative, an inventive binder can provide a reduction in benzene of at least 70%, a reduction in xylene of at least 70% and/or a reduction in phenol of at least 60% compared to the phenolic urethane tested herein. In still another alternative, an inventive binder can provide a reduction in benzene of at least 80%, a reduction in xylene of at least 80% and/or a reduction in phenol of at least 65% compared to the phenolic urethane tested herein. In still yet another alternative, an inventive binder can provide a reduction in benzene of at least 90%, a reduction in xylene of at least 90% and/or a reduction in phenol of at least 67.5% compared to the phenolic urethane tested herein.
When conducting these analyses, the TGA-MS responses were normalized by analyzing a standard sample of 27.9 mg calcium oxalate. For this standard calcium oxalate sample, the maximum response for AMU 44, attributed to CO2, was about 5 nA total when compared to a baseline of 0 nA at about 800° C. When the nA response for calcium oxalate is different than this, the test samples could be normalized accordingly. In addition, the sample preparation and TGA-MS procedure disclosed herein are defined as a TGA-MS protocol.
Each sample was heated in 100% argon at 10° C./min to 1000° C. with air emissions monitored via mass spectrometry. As shown in the figures, collagen emits significantly less VOCs than phenolic urethane. For example,
Regarding DMA testing,
Not being bound by theory, it is postulated that bindered core DMA performance in the 300 to 500° C. range correlates to a bindered core's resistance to cracking-type failure which can lead to erosion. As such, a higher storage modulus in this temperature range corresponds to less failure. In that perspective, collagen alone was the lowest performing material since it offered the lowest storage modulus of 10-100 MPa in this range. Sodium silicate was the highest performer at 400-900 MPa and phenolic urethane (heat cured) was 50-100 MPa. The collagen+silicate binders performed well with values between 50-300 MPa in the stated temperature range. It is appreciated that the low storage modulus for collagen in the temperature range of 275° C. to 400° C. may contribute to core erosion when the collagen is exposed to molten metal.
Referring now to
As shown in
HDT core samples distorted while being heated by the natural gas flame with results shown by the graph in
Looking now at Table 7, when molten iron was poured onto the cores, the sodium silicate and both collagen+silicate cores did not exhibit measurable erosion as shown by the zero readings in the table. In contrast, the phenolic urethane (P-U) cores exhibited slight erosion and the collagen cores exhibited significant erosion. The phenolic urethane cores lost only 0.02 cm of core depth and 0.42 to 0.72 grams of sand per kilogram of iron poured. The collagen core exhibited the greatest erosion, while experiencing the least molten iron. For example, the collagen core erosion depth and sand erosion were approximately 20 times greater and over 10 times greater, respectively, per kilogram of molten iron when compared with the phenolic urethane core samples.
The sodium silicate core decreased slightly in hardness whereas the collagen and collagen+silicate cores exhibited more significant decreases in hardness when experiencing a similar amount of molten iron. Also, the hardness of the phenolic urethane core decreased significantly when experiencing molten iron, to less than half the sodium silicate core hardness.
1Phenolic urethane
2Heat cured phenolic urethane
3TEA cured phenolic urethane
4silicate
5collagen
In view of the teaching presented herein, it is to be understood that numerous modifications and variations of the present invention will be readily apparent to those of skill in the art. As such, the foregoing is illustrative of specific embodiments of the invention but is not meant to be a limitation upon the practice thereof. It is the following claims, including all equivalents, which define the scope of the invention.
| Number | Date | Country | |
|---|---|---|---|
| 61249037 | Oct 2009 | US | |
| 61308115 | Feb 2010 | US |
