The present invention relates to a process for preparing useful shaped cementitious .articles from the waste products of the combustion of coaL More specifically., this invention relates to the preparation of shaped -articles such as pellets, bricks, tiles, blocks -and the like by a closely controlled compaction process directly from the fly ash-containing powder produced in a lime-based dry scrubbing process for removal of fly ash and SO- from flue gases.

Among the products of coal combustion, fly ash and the acid gases, prim.ar.ily sulfur dioxide, in the flue gas are the major causes of -air pollution. Flue gas cleaning systems presently employed produce large quantities of solid waste which include recovered fly -ash and sulfur-containing reaction products from scrubbing operations. The worldwide shortage of oil and gas for heating .and power generation is causing a shift to coal burning, with a resultant increase in the production of these solid wastes. It has been established that some 73 million metric tons of fly ash .and some 22 million metric tons of flue gas desulfurization sludge will be produced in the U.S. by the year 1986. Disposing of this m.ass of materi-al is costly, requires 1-and, .and causes secondary environmental problems that must be dealt with to avoid pollution of groundwaters and loss of land values.

Solid wastes from coal combustion systems include two basic types, namely, those removed by dry collection upstream or downstream of the SOg scrubber, or collected in the scrubber itself, and scrubber waste, usually as sludge. Wet scrubbers can be categorized as either throw- away or recovery. The most common process, the single closed-loop

throw-away, disposes of its w-aste after reclaiming a certain portion of the water which is recycled to the scrubber to maintain water balance. The double-loop throw-away process, called dual-alkali, uses one loop to recirculate scrubber reagent and another loop to regenerate this reagent and remove waste sludge.

The one thing -all wet scrubbing flue gas cleaning systems have in common is the generation of a calcium siilfϊte/ calcium sulfate sludge, with a low solids content .and a limited structural load-bearing capability. This .sludge is fine-grained and highly water retentive. After settling, it will solidify somewhat, but it reverts to a fluid consistency when disturbed. Sludge solids contain leachable and potentially toxic salts and the disolved salts in the waste water occluded in the sludge contain trace quantities of soluble heavy metal elements from the coal used as fueL The prior art describes attempts to produce relatively low density shaped products from wet flue gas scrubber w-aste, see, e.g., Minnick U.S. 3,785,840. These prior art processes, however, have been generally unsuccessful

The advent of the dry scrubbing system for SO„ -and fly ash removal from flue gas-es represents a major advance in the art of poEution -controL It has been estimated that dry scrubbing systems will occupy a major role in SO ₂ and fly ash removal from the some 200,000 megawatts of coal burning -capacity which is expected to be added to the U.S. electric utility systems by 1990. Dry scrubbing systems are unique in that a dry waste is produced which can comprise fly ash in intimate association with the materials produced by the reaction of SO with calcium oxide and other metal oxid-es p.resent in the system. A powder is produced in the dry scrubbing system, as contrasted to a high water content, low solids content waste product produced in wet scrubbing systems. The dry scrubbing w-aste powder is still produced in large quantities and therefore it would be extremely desirable to provide a process for ecomonically converting this waste into useful material.

Accordingly, it is an object of this invention to economically utilize the solid fly ash-containing waste materials from the dry SO„ scrubbing of flue gases from coal combustion.

More specifically, it is .an object of the present invention to provide from these waste materials dense shaped structural articles bonded by cementitious reaction products and a process for producing these articles. These and other objects of the present invention can be achieved by providing a process for preparing a dense shaped article, which article comprises fly ash particles bonded together by cementitious reaction products, the process comprising the steps of providing a substantially dry fly ash-containing powder obtained from the dry scrubbing of fly ash-containing flue gas with lime, the fly ash-containing powder comprising fly .ash in intimate admixture with microp.articulate crystallites of scrubber reaction materials selected from the group consisting of calcium sulfite, calcium sulfate, hydrates of calcium sulfite and calcium sulfate, calcium oxide, c-alcium hydroxide and mixtures thereof; uniformly contacting the fly ash-eontaining powder with a predetermined amount of water; forming the powder/ water mixture into a shaped article by confined compaction at a predetermined compactive effort sufficient to provide a handleable green body; the predetermined .amount of water being selected as at least that amount sufficient to satisfy the short term hydration demands of the powder and further selected in combination with a selected predetermined compactive effort to produce a green body having an uncured dry density in the range of from about 95 to about 140 pounds per cubic foot and wherein the interspatial voids between the fly ash particles in .the green body are sufficient to accommodate the subsequently formed cementitious reaction products without deleterious expansion of the article and curing the green body product in a moist atmosphere to form a shaped article bonded by the cementitious reaction products.

The present invention also provides a dense hardened cementitiously bonded shaped article made from a composite material which comprises a fly ash-containing powder, the powder comprising fly .ash in intimate admixture with mieroparticulate crystallites of scrubber reaction materi.als selected from the group consisting of calcium sulfite, calcium sulfate, hydrates of calcium sulfite and calcium sulfate, calcium oxide, calcium hydroxide and mixtures thereof, the composite material exhibiting a

OMPI

splitting tensile strength of at least about 300 psi and a cured dry density of from about 95 to about 140 pounds per cubic foot.

The product of cementitious processes involving interaction of metal oxides in aqueous systems is highly time and spati-ally sensitive relative to any solid phase such as fly .ash contained in the artide produced. As cementitious reactions proceed, the reaction products formed undergo volumetric changes which must be accommodated in the final object produced when curing is completed. It is essential to achieving a product having the desired characteristics to strictly observe the critical relationships between the amount of water added to the powder and the compaction operation which determine the spatial •arrangement of fly ash particles in the green body. Added water in addition to providing an essential reactant in early hydration reactions, serves as a space control .agent which in part determines the void space i .n the compacted article. The density of the compacted article must be -controlled (by amount of added water and compaction ratio) to give an article with sufficient void space to accommodate the reaction products formed.. If too much or too little water is employed or .if too much or too little compaction is effected, the product will not exhibit the unexpected superior properties achieved by the present invention. Critical control, insofar as time of water addition in relation to the compacting operation is also required to utilize latent bonding properties of cementitious reactions avail-able. In the process of this invention, the cementitious activity of the waste powder may be increased by converting sulfites to sulfates and calcium sulfate dihydrate to hemihydrate. In toto, these factors are controlled to produce hard dense shaped articles in which the fly ash is bound in solid fixed relationship, one partide to another, by the cementitious -agents comprising the final reaction products of- calcium oxide and other oxides, sulfates, silicates, aluminates and other constituents of the system.

In the drawings which illustrate the preferred embodiments of the present invention:

Figure 1 is a flow diagram of one form of the process of the present invention. Figure 2 is a series of photomicrographs showing the physical

characteristics of a dry scrubber waste powder useful in the process of the present invention.

Fiagures 3-9 are graphs which illustrate the unique properties of the products produced according to the present invention. As used throughout the instant specification and daims, the term fly ash is intended to refer primarily to flue gas-entrained ash from the combustion of coal. The term lime is intended to refer both to calcium oxide (CaO) and its hydrated form, -calcium hydroxide (Ca(OH) ₂ ). One of the primary components of the product of the present invention is fly ash. Fly ash from coal combustion occurs as spherical partides, usually ranging in di.ameter up to 100 microns. The chemical makeup of fly ash can vary widely depending on the geologic and geographic factors .affecting the coal deposit .and on the combustion conditions. Coal can be classified under the ASTM ranking system as anthracite, bituminous, subbituminous -and lignite. Fly ashes from each of these ranks usually contain as major constituents SiO„, Al ₂ O ₃ and Fe„O,. Some ashes, particularly those from subbituminous or lignite ∞als can contain significant quantities of CaO. Various fly ash materials may also contam minor constituents such as magnesium, tit-anium, sodium, potassium, sulfur and phosphorous -and may further contain trace concentrations of from 20 to 50 additional elements.

The process of the present invention can utilize fly ash from the combustion of any of the coal types. It is preferred to employ fly ash which contains at least about 10% CaO. Typically, subbituminous and lignite coal ashes meet this requirement and therefore these fly ashes are the most preferred. In order to improve the reactivity with water, it may be necessary to add lime or other cementitious adjuvants to fly ashes derived from bituminous coals. In most cases, however, the reactivity of the fly ash-containing dry scrubber product can be enhanced to a suitable level during the dry scrubbing operation.

One preferred form of the process of the present invention induding preferred steps for producing the fly ash-containing dry scrubber powder is shown in Figure 1. In the illustrated embodiment, calcium hydroxide 1 is mixed with water in a slurry tank 2 to provide an aqueous feed suspension 3. This feed suspension is then atomized in a stream

of hot, fly ash-containing flue gas 4 in a drying chamber 5 to effect substantial drying of the resulting atomized droplets and partial absorption of the sulfur dioxide in the flue gas. A portion of the resulting dry powder eomprisi.ng fly ash, reaction products and unreaeted materials 6 is passed along with the flue gas to particulate removal device 7 in which further reaction may take place. A first portion 8 of the free flowing dry powder produced in the drying chamber is collected direetly from the bottom of this chamber, and a second portion 9 of the powder is coEeeted from the bottom of the particulate removal device. Finally, a portion of said fly ash-cont.aining powder 10 is recycled for preparation of the aqueous feed suspension. The drying and sulfur dioxide absorption are effected while mamtaining the temperature of the flue gas leaving the drying chamber at from about 8 to 40° C above the adiabatic saturation temperature of the gas by controlling the amount of feed suspension forwarded to the drying chamber and the total soHds content of the feed suspension in response to the amount, temperature, and moisture content of the flue gas feed to the drying chamber. More details of this preferred dry stubbing process can be obtained from U.S. 4,279,873,. whidi is hereby incorporated by reference. Other dry scrubbing processes may be used as long as they produce a fly ash- contaMng waste powder as hereinafter described.

In genera^ the dry waste powder obtained from the above described dry stubbing process is predominately fly ash partides coated with the reaction products and unreaeted reagent. A s . mall percentage of the powder (up to about 10-45%) may comprise fly ash partides which appear subSa tantially unchanged by the scrubbing reaction; a small percentage of reaction materials unassociated with fly ash may also be present. The fly ash partides indude those captured directly from incoming flue gas and those recycled through the reagent preparation procedure at least once. Reference to the series of photomicrographs in Figure 2 (Fig. 2 is at 30Q.X; Fig. 2B is at 1000X) shows the coated fly ash partides 21 and the apparently unchanged fly ash partides 22. Figure 2 clearly shows the microparticulate nature of the reaction material crystallites on the surface of each individual fly ash partide. Since these crystallites are formed in a very short time in the spray drying/ absorption step,

there is little time for well ordered large crystal growth. The very high reactivity of the powders used according to the process of the present invention is believed to be due, at least in part, to this unique physical structure. The gross composition of this dry s-crubber powder will vary with the nature of the coal burned and its resultant ash, with the amount of scrubbing reagent, and with the conditions of the scrubbing operation. Broadly speaking, the powder comprises fly ash, scrubbing reaction products, primarily calcium sulfite and calcium sulfate, and unreaeted reagent lime as either CaO or Ca(OH)„. Some CaO may react to form calcium-aluminosilicate hydrates or similar compounds on the surface of fly ash partides during scrubbing or reagent preparation.

In general, the dry s-crubber waste powder useful according to the present invention comprises by weight from about 10 to 85% fly ash; from about 10 to 60% CaSO (various forms, including calcium sulfite hemihydrate, and the dihydrate, hemihydrate and anhydrite of calcium sulfate-depending primarily on the temperatures in the absorber .and collection system); from .about 1-30% unreaeted lime (primarily as Ca(OH) ₂ ); and up to about 7% free water. A preferred analysis of such a powder based on a low-sulfur western coal ash is as follows:

Fly ash 70-80%

CaSO _χ 12-24%

Lime 3-10%

Free water 3-5% The amount of water which- is chemicaEy bound as hydrates in this powder prior to product formation can be estimated from the chemical compounds present. Assuming that 90% of the lime is hydrated, that CaSO _χ is 65:35 sulfite to sulfate and that the sulfate is 50:50 dihydrate to hemihydrate, the chemicaEy bound water can vary from about 1.5 to -about 13% by weight of the product. The preferred analysis for low sulfur western coal ash products is about 2 to 5% bound water.

The first collected portion of powder 8 from the bottom of the drying chamber may differ somewhat in composition from the second portion of powder 9 eoEected in the particulate removal device. In a typical operation, material eoEected from the bottom of the drying

OMPI _

chamber contains about 70% by weight fly ash, 13% by weight calcium sulfite, 7% by weight calcium sulfate, 5% by weight unreaeted lime, and 5% by weight free water. Material eoEected from the particulate eoEeetion device, (e.g., a bag house) contains about 72% by weight fly ash, 15% by weight caldum sulfite, 9% by weight calcium sulfate, 1% lime and 3% free water.

The powder from a dry scrubbing op-eration can be stored for short periods, e.g., on the order of weeks or months, and still be useful in the process of the present invention. It is preferred, however, to utilize this powder immediately after it is formed in the scrubbing operation to take fuE advantage of the nascent cementitious properties of this material.

Both the first and second eoEected portions of powder from the flue gas deaning operation, as detafled in the foregoing, contain latent cementitious materials as they normaEy exit the process. It may be desirable, in some cases, to further develop the latent bonding potentials before using the cfcy scrubber powder according to the present invention. It is possible by adjusting the scrubber parameters . to achieve this objective. Referring again to Fig. 1, a separate con-ditioning step E optionaEy can be used to enhance the cementitious properties of the fly ash-containing dry scrubber waste powder 12. In such a separate conditioning step the primary objectives are to convert sulfites to sulfates and to at least partiaEy dehydrate sulfates or sulfites. The preferred process for accomplishing these objectives is subjecting the powder from the scrubber to a mEd heating process. This heating process can be suitably carried out, e.g., in a fluidized bed, at a temperature from about 120° C to about 350° C. While either oxidizing or reducing conditions can be employed during this heating, a mEd oxidizing atmosphere is preferred. The heating can be carried out in a continuous manner for a period suffident to complete the dehydration reaction as is known. TypicaEy, heating at about 140° C for about 10 minutes is effective to achieve the objectives of this treatment. The exact parameters necessary to effect this conditioning step win, of course, vary with the specific powder produced in the scrubber. In the next step of the process of the instant invention, shown

generaEy at 13, the optionaEy conditioned waste material 14 is intimately and rapidly contacted with a predetermined amount of water 15. The amount of added water, and the mode of adding water are critical in accomplishing the process of the present invention. The addition of water serves several very important functions in the present invention. First of aE, the added water serves as a reactant in the rapidly proceeding hydration reactions with the various cementitious materials present. These rapid rate hydration reactions which can be termed "short term" reactions to distinguish them from the slower rate hydration reactions which continue during curing in a moist atmosphere, are, in combination with the densities achieved, responsible for the high initial strength of products formed according to the present invention. Secondly, the added water serves an important function as a space control agent. As indicated above, as the cementitious reactions proceed the reaction products formed tend to sweE. In order to avoid disruption of the hardened matrix, and attendant decrease in strength properti-es it is essential to provide sufficient void space in the formed green body to accommodate the volumetric changes which result upon extended curing. To this end, added water preserves void spaces during the compaction step. Added water also serves as a lubricant and a temperature/ rate of reaction moderator during the compaction step.

Water addition is preferably carried out immediatdy before the compaction step described below. Added water begins to react with the powder of the present invention immediately and if the fly ash partides have been compacted into the requisite spatial relationships as described below whEe these cementitious reactions are taking plaee, then optimum bonding of the partides occurs. It is possible to delay the compaction step for a short period of time after the addition of water without any significant adverse affect on the strength properties of the cured product. Dday times of up to about one hour or more can be utilized depending on the reactivity of the fly ash-eontaining powder and the amount of added water. Short delay times may actuaEy enhance the short term properties of the formed product, probably due to more uniform distribution of the water by capElary action during the aging step.

Water should be added to the powder in a manner that ensures rapid and uniform distribution. Large slugs of water preferably should be avoided. In addition to aging described .above, the requisite uniform water distribution can be effected by mixing. For batch type operations this mixing can take place in any suitable equipment such as a paddle mixer or the like. Another suitable manner of effecting the addition and distribution step is by the use of a finely divided spray directed into a mass of powder that is being effidently mixed -and tumbled. Apparatus such as eommerci.aEy avaEable pug mflls can be employed for this purpose. Water addition and mijdng -are preferably carried out at ambient temperatures although somewhat higher or lower temperatures can be employed. Mixing time can vary with the equipment used but typicaEy this step can be accomplished in a very short period, i.e., on the order of a few seeonds up to several minutes. Preferably, mijdng times of up to about 5 minutes are employed. The use of longer mixing times, i.e., up to .about 15 minutes, can result in compacted products exhibiting higher early strengths than simEar products mixed for shorter periods.

Addition and distribution of water can also be effectively accomplished by passing steam through the powder under conditions favoring condensation of water throughout the mixture.

The amount of added water used is critical to achieving a strong dense cured product. Powders from different sources have somewhat different water requirements due to the variabiUties of the reactivity and physical characteristics of the powder (which v-ary with the nature of the fly ash and scrubber operating conditions).

The prim.ary criteria for selecting the proper amount of added water are G) the short term hydration demands of the powder and (2) the req site inteirspatial relationship between the fly ash partides. The second factor is critieaEy interrelated to the type and extent of the compaction operation. In practice the amount of added water should be at least that amount necessary to satisfy the short term hydration demands of the particular powder and further defined as the amount which provides, for a given compaction operation, a shaped green body having adequate interspatial voids to accommodate substantiaEy aE the

- E - ^'

subsequently formed cementitious reaction products. If, for a given compaction operation, too little water is employed, the green body may be overdensified in the compaction step. Overdensification can result in a product which "comes apart" as it cures due to expansion of interspatial reaction products to a volume larger than the initial. inteirspatial voids. The strength of the cured product is also adversely affected by adding too much water — which may lead to under- densifi cation. The amount of interspatial void area necessary to accommodate the subsequently formed reaction products will depend in part on the reactivity of the powder and the conditions of curing. In general, an initial dry void space of from about 5 to about 40% by volume can be satisfactorily employed. Preferably, the void space can range from about 10 to about 35%. Percent volumetric void space can be calculated by the foE owing formula:

- dr article density absolute density

An alternative measure of the initial shaped product void space is the dry density of this product. Typically products according to the present invention should have a dry density in the r-ange of from about 95 to about 140 pounds per cubic foot .and preferably from about 100 to about 130 pounds per cubic foot.

As a practical lower limit, the amount of added water must be sufficient to form a mixture which can be compacted into a coherent handleable mass on a given type of compaction equipment. This practical lower limit varies widely with the reactivity and physical characteristics of the powder and nature of the compaction equipment. As a practical upper limit the added water also should not exceed that amo.unt which can be compacted on the particular compaction equipment chosen. For example, a very high water content mixture (mud-like) cannot normaEy be compacted into a coherent handleable shape with the preferred range of product densities. The effect of particular compaction equipment on water requirements is further demonstrated by the case of a high speed, extrusion- type peEetizing- apparatus described hereinafter. Because of

the high temperatures generated by friction^ forces, some of the free water in the powder mixture is lost by evaporation. Accordingly, the optimum water additive levels are generaEy higher for this type of process than for low water loss compaction systems. SimEarly, water requirements may .also be affected by the die size in an extrusion process. GeneraEy, larger die sizes can be operated with less added water.

In genera^ for the powders and compaction equipment described in the examples, water addition amounts of from about 4% to about 20% by weight b-ased on the powder have been suceessfuEy employed. Optimum properties for these powders compacted on commercial scale peEetizing apparatus generaEy are achieved with added water in the range of about 10% to about 20%. As described below, lubricants and other additives may be employed to reduce the water requirements somewhat.

The next step in the process, the water/powder mi.xture 16 is compacted at 17 to form a shaped handleable green body 18. The compaction preferably is performed ϊmmediatdy after the water addition step. Keepaing the time between water addition and compaction r datively short is instrumental in obtaining the most effective bonding of the fly ash partides using the cementitious products formed by the reaction of the caldum oxide with sEϊea and other materials present, and the relatively rapid calcium sulfate hemihydrate to dihydrate transformation.

It is essential that the compaction operation be controEed by selecting a compactive effort (i.e., compressive force) which, in conjiBiction with the amount of added water, will provide a handleable green body having the necessary void space between the fly ash partides to accommodate the subsequently formed cementitious reaction products. Under proper compaction conditions the fly ash spheres are positioned, relative to an adjacent sphere, such that firm bonds are formed by the cementitious agents formed during curing. If the spheres are too dose together after compaction, the formed artide wEl sweE and disrupt the cementitious matrix upon curing. If the spheres are too far apart, insufficient weak bonds will form. As shown in the examples, overdensifi-eation leacfe to strength and density regression upon extended

curing. The exact amount of compressive force needed to achieve these objectives wiE, of course, vary with the powders and type of product being foήned. In general, however, the proper degree of compaction, for a given amount of added water, can be determined with reference to the artide density or percent void space. As indicated above, the compacted shaped green product should have a dry density in the range of from about 95 to about 140 pounds per cubic foot or when expressed as percent void space, from about 5 to about 40% by volume. The optimum compactive effort for any given powder/water mixture can best be determined e.xperimentaEy in accordance with the teachings of this invention. Compaction ratios can be calculated by taking the ratio of the dry shaped artide density to the dry loose bulk density of the powder. In general, compaction ratios in the range of from about 2 to about 3, depending on the particular powder employed are suitable for the practice of the present invention. For the powders described in the examples, compaction ratios of from about 2.4 to about 2.8 achieve the desired optimum spatial relationship.

The nature of the compaction operation will, of course, vary with the type of product being formed. WhEe any of the weE known compacting and shaping systems may be employed, it is preferred to use a compaction system which forms the product very rapidy under high energy conditions. In the case of one preferred product according to this invention — gravel peEets — this rapid high energy compaction system can be a high speed extrusion type peEetizing mEl exemplified by the roEer ring mEl sold by the California PeEet MEl Company. In this type of equipment the product is formed very rapidly and in the presence of lubricating steam generated by the frictional heating of process water. TypicaEy, bulk material temperatures of up to about 140 to 160° F or more may develop in this type of process. Under these conditions the formed product exhibits very high initial strength properties due in part to temperature induced acceleration of the rate of early short term hydration and other curing reactions.

The compaction step should be carried out as rapidly as practical on the equipment chosen. For a typical commercial size peEet forming operation this compaction step can be accomplished in less than about

one second.

In the final step of the process, the compacted green body 18 is cured at 19. Ciring is the process whereby cementitious reaction products are formed which bind the powder mass together into a hardened artide. Formed products of this invention acquire ultimate strength only after weeks, months, and yeairs foEowing manufacture, as is the case with conventional pozzolan cements. However, due to the relatively rapid curing rate possible with the unique finely divided reactive components of this invention, the formed products -are serviceable after a few days, curing in humid air.

TypicaEy, the products of the present invention may reach minimum aeeepta&e strength in as little as ten days at 100° F. Curing of the products of the present invention can be acederated by the usό of heat and moisture, e.g., minimum acceptable strength may be achieved in as Ettle as two days at 120 to 180° F. In the manufacture of peEets for use as gravel substitutes the initial curing can be effected in storage sEcs for about one to ten days. By placing the peEets in sealed storage shortly after formation, the trapped moisture-saturated air, the undissipated heat generated in the peEetizing process and the heat of the cementitious reaction can be utEized to facEitate the curing process. The use of periodic water sprays or soaking in a preferred method of ensuring the requisite moisture for long-term hy-dration demands and optimum curing.

As is generaEy known in the compaction and shaping art, minor amounts of additives may be employed to facilitate processing and modify product properties. These known additives can be employed in the compositions of the present invention in conventional amounts and for conventional purposes. Examples of such additives which may be utEized indude lubricants such as stearic acid, oEs, waxes, oEy grains (e.g., oats), carboxymethyl ceEulose .and other ceEulosic materials such as starches and sugars; bmdeirs such as calcium sulfate hemihydrate, water glass, synthetic resins and latices and Portland cement; solid adjuvants such as lime kEn dust or cement kEn dust; water reducing agents such as ealdum lignosulfonat-e; water proofing agents such as calcium stearate; corrosion control agents sueh as ealdum nitrate; wetting agents, and

the like. Additive amounts should be kept at a minimum, i.e., no more than necessary to achieve the intended function. TypicaEy, lubricants are conventionaEy employed at levels of about 3% or less by weight; binders are conventionaEy employed in amounts of up to about 10% by weight.

The product of the present invention is a dense shaped artide containing fly ash partides cementitiously bonded together. As described above, one preferred form of this product is a shaped peEet or granule which can be used as a gravel substitute. This peEet finds utility in all normal uses of gravel such as in road bed compositions, as structural fiE, as an aggregate in concrete and concrete structural products, -and the like. Other shpaed artides which may be produced according to the present invention include bricks, tiles, blocks .and other structural shapes. For purposes of the foEowing discussion the product of this invention will be described primarily as a gravel substitute peEet.

The gravel substitute peEet made according to the present invention has strength and density properties which compare favorably with natural gravel products. In material respects the products of this invention meet or exceed the specifications for limestone or dolomitic gravel. Depending on the nature of the waste powder starting materials and the length and conditions of cure, the material of the artides of the present invention can achieve splitting tensEe strengths of at least about 200 psi. The preferred high strength materials of the present invention have splitting tensEe strengths of at least about 300 psi and preferably at least about 400 psi. TypicaEy, the high strength materials of this invention may achieve splitting tensile strengths of from about 500 to 1000 psi or more. In general, to maximize economic considerations the process parameters can be varied to produce a product with strength values that are no higher than required for a specific application. The attrition resistance of the product of the present invention is also very high as demonstrated in the examples. The products of the invention also have good hardness properties — ranging from about 3 to 4 on the Mohs hardness scale. Diry densities of these products can range from about 95 to 140 pounds per cubic foot or more and for the preferred high strength materials at least 100 pounds per cubic foot.

O

The foEowing examples are intended to Elustrate more fuEy the nature of the present invention without acting as a limitation on its scope.

EXAMPLE 1 This Eχ.ample demonstrates the eriticality of proper compaction in the process of the present invention. A fifty gram sample of material representing the waste product from a dry scrubber flue gas desulftffization operation (identified as Antelope Sample #2230) was placed in a iggi shear Waring Blender wherein, under rapid mixing conditions, two grams of water was added via a volumetric pipette. Mixing was continued for thirty seconds and a portion of the material immediatdy formed into peEets approximatdy 7.4 miEimeters in diameter by 7.7 milEmeters length, using a volumetric compaction ratio of approximatdy three volumes of loose powder to form each peEet. Three groups of peEets were made over a time span of ten minutes, using punch pressures of 20, 12, and 5 tons per square inch. Respectively, the test peEets average 7.7, 7.9, and 9.5 millimeters in length, thus obtaining approximatdy 23% change in overaE peEet volume at constant mass, which translates directly into avaEalΛe space between individual fly ash partides to accommodate cementitious products formed by reactions described supra. The test peEets were cured in a water saturated atmosphere over a period of several months. Curing progress was observed by destructive testing of portions of each batch of peEets at 10 days, 30 days, and 60 days. Measurements were made of attrition resistance and density as they reflect important required use properties. Attrition resistance was determined using a Spex Model 8000 grinding apparatus with a large (2 1/4" di.ameter x 3" high) grinding chamber. An individual peEet of the samples to be tested is dried to constant weight, weighed and placed into the grinding chamber with no grinding baE charge. The machine is then activated for 15 seconds after which the contents of the chamber are removed and brushed onto a U.S. Standard No. 60 sieve screen. The minus 60 material is brushed through the screen and the remaining plus 60 material is weighed. The percent of original peEet passing through the No. 60 sieve is then calculated.

If the plus 60 material is more than 60% of the original peEet, this fraction is returned to the chamber and the above procedures are repeated using cumulative grinding times of 30 sec, 1 min. 2 min., 5 min., 10 min., untE 75% or more of the original peEet has passed through the No. 60 sieve. The average time required for 60% of the peEet to pass the No. 60 sieve is determined by calculation and the values reported are the average of three such tests. Density was calculated as foEows. The volume in mis. of saturated peEets was measured by displacement in water. The S.S.D. weight (saturated surface-dry weight) of the peEets in grams was obtained. Based on these figures .an S.S.D. density was calculated. This value was converted to dry densities by using a percent absorption factor which is the difference in S.S.D. weight and 130° F dry weight divided by the dry weight.

Figure 3 shows the effect of compaction on attrition resistance as curing progresses. Rdatively poorly compacted peEets, pressed at 5 tons per square inch, were initiaEy friable and after some brief early gain did not appreciably gain in attrition resistance as curing progressed beyond 30 days. PeEets highly compacted at 20 tons per square inch showed the highest initial resistance to attrition, but lost in attrition resist-ance as curing progressed beyond 30 days and internal pressures developed as a result of insufficient space avaEable to accommodate forming-cementitious materials. PeEets pressed at 12 tons per square inch had intermediate initial attrition resistance, but gained much more rapidly in this property, as curing progressed to 30 days and thereafter feE off to a higher value than the powder compacted at 20 tons per square inch.

Figure 4 shows the effect of compaction on peEet density as a function of curing progress. Note that peEets pressed at 5 tons per square inch have relatively low initial density, and density changes only slightly during curing. In highly and intermediately compacted peEets the density decre-as-es as curing progresses, due to partide sweEing. PeEets formed at 12 tons per square inch faE off at a slower rate as curing progresses beyond 30 days due to the cohesive and adhesive forces of reaction between the formed cementitious materials and the fly ash partides.

The Antelope sample utilized in this example graphically demonstrate the regression of strength and density values that takes place when the formed product is overdensified (i.e., at 20 tons/in ). This obervable effect is beEeved to be due to the disruptive forces which develop as the reaction products in the interspatial voids sweE with curing. The absolute value of compactive force necessary to reach the point at which values begin to diminish varies with each specific powder. Other powders tested to date were not compacted with the purpose of approaching or passing this point of dimϊmsliing results.

E.XAMPLE 2

This Example demonstrates the positive effects on strength of using mEd heat conditioning to improve the reactivity of certain low- activity fly ash-containing powders. The procedures of Example 1 were repeated using 50 grams of a dry slubber powder identified as Tucson S.ample #2157. One portion of this sample was employed as received and a second portion was used after a mEd heat treatment (at 120° C), to convert the dihydrate forms of calcium sulfate to the hemihydrate form. Figure 5 is a plot of attrition values, as measured in Example 1, against curi.ng time for given compaction levels using the as received powder. The same data for the heat treated powder is presented in Figure 6. A comparison of these two figures demonstrates an improvement in long-term attrition properties for the heat treated sample.

EXAMPLE 3 Tϊύs Example demonstrates the critical effect of added water amounts on the product properties. Wdghed ^* portions of three separate powder samples of material representing the total waste product from separate planned dry scrubber flue gas desulfurization operations (identified as Antelope, Tucson and Sunflower) were placed in a Hobart Model N50 paddle type mixer, wherein, measured percentages of water were added. Mixing was continued for about six minutes and a portion of the material immediately formed into peEets using a roE mEl peEetizing machine identified as a California PeEet MEl CPM Model CL

equipped with 3/16" diameter, 3/4" long or 1/4" diameter, 1 1/2" long dies. The peEet lengths varied from about 0.15 to about 0.55 inches. The volumetric compression ratio for the different powder samples and water contents ranged from about 1.8 to 2.9. The test peEets were cured in a water saturated atmosphere over a period of several months. Curing progress was observed by destructive testing of portions of each batch of peEets. Measurements were made on splitting tensEe strength, attrition resistance and density. Splitting tensile strength of cylindrical peEets is determined using a Rinck-Mellwaine spring tester with modified adaptors to provide un.iform loading paraEd to the axis of the cyEnder, simflar to ASTM Method C496. Pressure is applied to each side of the cyHndrical peEet through the adaptor at a rate which wiE cause specimen failure in about 5 to 15 seconds. The splitting tensile strength of the peEets is given by the foEowing: T = 2P/7T Id where:

T ~ SpUtting tensEe strength, in pounds per square inch

P = Maamum load applied indicated by the testing machine, in pounds -— force

1 = Specimen length d - Specimen diameter, in inches

Attrition resistance and density were determined as described in Example 1. Figures 7-9 demonstrate the critical effect of added water on the physical properties of the products of this invention at 60 days of cure. The left side of each curve has been extrapolated back towards 0% water. Because of equipment related limitations coherent peEets cannot be formed below a certain minimum water content which wiE vary depending on the powder used. As the extrapolated curve approaches this minimum water value, the strength and attrition resistance values w l faE off to essentiaEy zero whEe the density values will faE back

towards the loose bulk density of the powder. In each of these figures it can be seen that increasing the added water levd beyond an identifiable optimum value results in a decrease of the product properties.

EXAMPLE 4 A second sample of material representing the total waste product from a planned dry scrubber flue gas desulfurization operation (identified as Sunflower) was peEeted in a manner simEar to Example 3, but under conditions designed to .simulate a production run of graveL Individual 5000 gm batches of the sample were placed in a Hobart paddle-type mixer, and 15% water (by dry weight of powder) was added by pouring it over a 5 second interval into the mixing bowl whEe the paddle was operating. Mixing continued at slow speed (140 + 5 rpm) for a total mix time of 2 minutes. Timing of the batch-mixing w-as adjusted so that freshly mixed material was continuously avaEable to the California PeEet Mill Model CL peEeting machine, requiring on the average, one fresh batch every 7 minutes. The peEeting die ^* used was a rifle-bore 1/4" diameter by 1 1 4" length die, and the mEl was operated continuously without any change in the above procedure for 4 hrs., 45 minutes. The miE came up to working temperature within 15 minutes, producing steaming peEets with a bulk material temperature in the 140 to 145 ° F range, and the peEet temperature in composite samples of approximatdy 50-60 pounds each (representing the product from 4 to 5 batches) reached temperatures of 157 to 164° F within minutes of production. A total of 503 pounds of gravd was produced. Bulk densities of the peEets as produced, measured at 10 minute intervals, varied narrowly from 70 to 74 Ib/cu ft., and the product appeared ur form. The peEet lengths varied .randomly, mainly from about 0.15 to 0.60 inches, and the compression factor, based on a loose powder bulk density of 43.8 lb/cu ft., was about 2.8. The composite samples were aEowed to cure in plastic bags which were left open at the top overnight (except as noted below) .and then dos-ed tightly to ensure that the peEets would be cured in a water saturated atmosphere. The peEets were cured in the sealed bags at ambient temperatures (70 to 95° F) for 4 days, except for Composite #1 and for two individual batches which were set aside for

special tests. The 4-day measurements on splitting tensile strength, attrition resistance, density, hardness, moisture content and absorption, using the previously outlined test procedures -are shown in Table 1.

TABLE 1

Composite No.

2 3 4 5 6 7 8 Ave.

4-DAY CURE, AMBIENT TEMPERATURES

Dry peEet 125 125 125 125 123 123 122 124 density, lb/cu ft

Splitting 756 762 777 689 760 726 684 736 tensEe str. lb/sq in

Attrition E5 171 109 l 129 E8 101 122 resistance, time to reach 60%, sec.

Mohs 4 4 4 4 4 4 4 4 hardness

Moisture, — 7.0 5.9 6.5 7.8 6.5 6.4 6.7 as cured, %

Absorption, 8.9 9.9 10.0 10.4 10.7 E.0 10.5 10.2 %

After the 4-day tests, the remaining peEets were saturated by immersion for 4 minutes (to essentiaEy const-ant weight), and portions of the surface-wet peEets from each composite were tightly containerized and placed in curing chambers at 100° F and 140° F.

At day 10, the peEets from the 100° F and 140° F curing chambers were tested for the same properties as above, and the results are shown in Table 2.

TABLE 2

Composite No.

4 Ave.

10-DAY CURE, 100° F Dry peEet; 126 128 127 126 129 126 127 127 density, Ib/cu ft

Splitting 1000 991 1000 951 946 1010 1070 995 tensEe str.

Attrition 244 250 236 227 232 245 238 239 resistance, time to reach 60%, sec.

Mohs 4+ 4+ 4+ 4+ 4+ 4+ 4+ 4+ hardness

Moisture, 9.8 7.4 9.0 9.5 9.6 9.3 8.8 9.1 as cured, . %

Absorption, 10.2 7.4 9.1 9.6 9.8 9.7 9.2 9.3

%

10-DAY CURE, 140° F

Dry peEet 126 128 127 125 127 126 127 127 density, Ib/cu ft

Splitting E30 1010 E60 1220 E90 1240 1230 E70 tensile str.

Mohs 4+ 4+ 4+ 4+ 4+ 4+ 4+ 4+ hardness

Moisture, 6.6 6.2 6.9 8.2 6.6 8.1 7.8 7.2 as cured, % Absorption, 7.9 7.5 8.0 8.2 6.9 8.2 8.0 7.8

%

The results of tests on the composite samples show that a gravel with consistent properties was made during the production run; the properties obtained indicate that the gravel is satisfactory for commercial use. As noted earEer in this Example, two individual batches of peEets, each weighing approximately 15-20 lbs., were set aside for special tests. Batch No. 1, taken immediatdy prior to Composite No. 1, was sealed in a plastic bag immediately after production to prevent any subsequent water less and to simulate curing effects that could occur in a dosed sEo. Batch No. 2, taken immediately after Composite No. 1, was air cooled and dried at ambient temperature for 20 minutes and then placed in a sealed bag. Composite No. 1 was handled in the same manner as the other composites, discussed above, except that the peEets were saturated for additional curing at 2 days rather than 4 days. Batches No. 1 and No. 2 were also saturated at 2 days. The results of tests, performed at 2 days and 10 days, are shown in Table 3.

TABLE 3

Batch No. 1 Composite Batch No. 2 Sealed Immed. No. 1 Cooled, then Sealed

2-DAY RESULTS

Dry peEet density, 122 125 126 lb/eu ft

Splitting tensEe 686 541* 467 str., lb/sq in

Attrition resistance, 167 E4 191 time to reach 60%, sec.

Mohs hardness 3 3 4

Moisture, as 7.8 7.0* 6.2 cured, %

Absorption, % 9.9 9.8* 9.6

Average of two tests.

TABLE 3 (continued)

Batch No. 1 Composite Batch No. 2 Sealed Imme No. 1 Cooled, then Sealed

10-DAY RESULTS, 100° F

Dry peEet density, 126 124 127 lb/cu ft

SpEtting tensEe 833 876 946 str., Ib/sq in

Attrition resistance, 226 220 259 time to reach 60%, sec.

Mohs hardness 4+ 4+ 4+

Moisture, as 10.3 9.1 9.2 cured, %

Absorption, % 10.5 9.6 9.2

Batch No. 1 Composite Batch No. 2 Sealed Immed, No. 1 Cooled, then Sealed

10-DAY RESULTS, 140 ° F

Dry peEet density, 127 128 124 lb/eu ft

SpEtting tensEe E30 E20 1200 str., Ib/sq in

Attrition resistance, 277 296 268 time to reach 60%, sec.

Mohs hardness 4+ 4+ 4+

Moisture, as 8.9 7.4 7.9 cured, %

Absorption, % 9.4 7.5 8.6

These results show that the amount of moisture present during the early curing period can influence the 2-day strength of the peEets, but the difference in strength is equaEzed with additional moist curing.

Additional curing effects, not readEy explained at the present time, have also been noted in the composite samples. The handling, testing, and saturation of Composites No. 1, No. 2, and No. 4 at 2 days or 4 days of age has been described ^' above. In addition to the tests discussed, tensEe strength tests were also performed after two additional days of curing of Composites No. 1 and No. 2, and peEets of Composite No. 4 were also tested at 6 days of age, without earlier saturation. In aE cases except the nonsaturated Composite No. 4, the test results Ested below in Table 4 are on peEets cured at 100° F after the initial saturation at 2 or 4 days of age. AE peEets were tested in a saturated surface dry condition.

TABLE 4

Splitting Tensile Strength, Ib/sq in

Composite Saturation Age Test Age

2 days 4 davs 6 days 10 days

No. 1 2 days 545 1226 — 876

No. 2 4 days — 756 1214 1000

No. 4 4 days — 777 — 1000

No. 4 None — — 809 —

The results indicate that resaturation of the peEets at an early age can provide a boost in tensEe strength, but that a portion of the increase is lost with additional curing.

The effect of the amount of delay between mixing and peEeting was also explored during the production run. Two extra batches of production-type material were prepared in the Hobart mixer during the run and were temporarily set to one side for peEeting later in the run.

One such batch that had been mixed 45 minutes earEer was peEeted at the end of the run; the second batch was discarded after visual inspection at 70 minutes revealed relatively large, firm lumps that might have blocked up the peEet mEL Results of tests on the 45 minute delay material are shown in Table 5.

TABLE 5 CURING CONDITIONS

45 minute 4 days 10 days 10 days

Dday Composition Ambient 100° F 140°?

Dry peEet density, 125 127 126 Ib/cu ft

SpEtting tensEe 932 1080 E70 str., Ib/sq in

Attrition resistance, E8 238 242 time to reach 60%, sec.

Mohs hardness 4 4+ ^{' '} 4+ ^'

Moisture, as 6.1 9.5 8.5 cured, %

Absorption, % 9.8 9.6 8.7

The results indicate that the 45 minute delay material eould be suecessfuEy peEeted. A 27% advantage over immediatdy peEeted material is observed in early strength; however, the mixes are nearly equal in strength after 10 days of curing. The effect of mixing time in the Hobart Type N50 mixer was also explored. Starting with a cold peEet mEl, 5000 gm batches of the powder sample plus 15% water were mixed in the Hobart mixer for various amounts of time. Other operating variables were the same as described -above, and the batches were handled in the same manner as composite from the main run, except that initial testing was at 2 days, and saturation of the remaining peEets was accompEshed at that time. Results are shown in Table 6.

TABLE 6

15 min. 5 min. 2 min. 1 min. 15 min,

Mix Mix Mix Mix Mix

2-DAY RESULTS , AMBIENT TEMPS

Dry peEet density, 125 120 122 124 125 lb/cu ft

SpEtting tensEe 806 779 801 732 890 ^* str., Ib/sq in

Attrition resistance, 129 103 105 101 137 time to reach 60%, sec.

Mohs hardness 4 3+ 4 4 4+

Moisture, as 6.5 — 7.3 7.2 6.8 cured, %

Absorption, % 8.8 10.7 10.4 8.6 9.5

10-DAY RESULTS !, 100° F

Dry peEet density, 127 122 124 128 123 lb/cu ft

Splitting tensile 950 896 1020 989 939 str., Ib/sq in

Attrition resist-ance, 214 197 239 259 240 time to reach 60%, sec.

Mohs hardness 4+ 4+ 4+ 4+ 4+

Moisture, as 9.5 10.7 10.2 10.5 9.5 cured, %

Absorption, % 9.6 10.9 10.7 E.0 9.6

TABLE 6 (continued)

10-DAY RESULTS, 140° F

Dry peEet density, 126 126 125 125 127 lb/cu ft Splitting tensEe E20 E50 E60 1110 E20 str., Ib/sq in

Mohs hardness 4+ 4+ _r 4+ 4+

Moisture, as 8.5 9.1 9.4 8.8 8.8 cured, % Absorption, % 8.7 9.4 9.8 9.2 9.0

Differences in the appearance of the mixed material prior to peEeting were obvious, with the 1 minute material containing many soft lumps at noticeably variable water contents. The 15 minute mix also contained smaE lumps but was very uniform by comparison. The results shown a 22% increase in early strength with increased mixing and a 10% increase in strength at the high, degree of miang in a hot peEet mEl as contrasted with the cold (start-up) conditions. These differences in performance are equ-aEzed with additional moist curing.

A number of ancfllary mixε- s were also made to demonstrate the abEity to make different sized peEets in the Model CL mEl. 3/16" peEets were made using a 3/16" diameter x 1" length rifle bore die; 1 4" peEets have been described in the earEer part of this Example 3/8" peEets were made using a 3/8" diameter x 1 1/2" length tapered die; and .1/2" peEets were made using a 1/2" diameter x 2" rifle bore die. In aE cases, the same two minute mixing procedure as outiined above was used with the Hobart type N50 mixer, and the mixed material was peEetized immediately. Two or three water content levels were utilized with each of the larger peEet sizes to demonstrate the effect of density on performance. The foEowing summarizes the compositions that were evaluated:

PeEet Diameter Water Added

20% 17 1/2% 15%

3/16" No Yes No

1/4" Yes Yes Yes

3/8" No Yes Yes

1/2" No Yes Yes

The properties and performance of the peEets produced are provided in Tables 7-9; aE mixes were tested and saturated at 2 days of age, and sealed containers were used to cure portions of the saturated samples at 100° F and 140° F.

TABLE 7

2-DAY CURE, AMBffiNT TEMPS.

PeEet Di.ameter Water Added 20% 17 1/2% 15%

Diry PeEet Density, lb/cu ft

3/16" — E6 1/4" - E0 E5 124 3/8* — E5 121 1/2" — E8 121

SpEtting ^■ TensEe Strength, lb/sq in

3/16" — 462 1/4" 334 409 684 3/8" — 288 405 1/2" — 435 532

Mohs Hardness

3/16" — 4 1/4" 4+ . 3+ 3+ 3/8" — 4 4 1/2" — 4 4-

Moisture, as cured, %

3/16" — 10.1 1/4" 10.8 10.2 7.2 3/8" — 10.1 7.5 1/2" — 9.5 7.3

Absorption, %

3/16" — 13.1 V4" 14.9 12.8 9.7 3/8" — 12.2 10.5 1/2" — E.7 10.6

'

TABLE 8

10-DAY CURE, 100° F

PeEet Diameter Water Added 20% 17 1/2% 15%

Dry PeEet Density, lb/cu ft

3/16" — 122 1/4" - E6 121 3/8" - E7 123 1/2" — 120 122

Splitting TensEe Strength, lb/sq in

3/16" — 1000 1/4" 796 ^* 794 970 3/8" — 544 676 1/2" — 537 708

Mohs Hardness

3/16" — 4+ 1/4" 4+ 4+ 4+ 3/8" — 4+ 4+ 1/2" — 4+ 4+

Moisture, as cured, %

3/16" — 12.4 1/4" 15.8 13.4 9.9 3/8" — E.7 10.3 1/2" — 10.2 9.3

Absorption, %

3/16" — 13.0 1/4" 15.8 13.8 10.3 3/8" — 12.6 10.5 1/2" — E.2 10.3

TAB.LE 9

10-DAY CURE, 140° F

Pdlet Diameter Water Added 20% 17 1/2% 15%

Dry PeEet Density, Ib/cu ft

3/16" — 122 1/4" E6 121 123 3/8" — E8 125 1/2" — 120 125

SpEtting TensEe Strength, lb/sq in

3/16" — 1360 1/4" 876 E00 1200 3/8" — 603 700 1/2" — 602 754

Mohs Hardness

3/16" — 4+ 1/4" 4+ 4+ 4+ 3/8" — 4+ 4+ 1/2" — 4+ 4+

Moisture, as cured, %

3/16" — 12.5 1/4" 14.6 12.4 10.3 3/8" — 10.7 9.0 1/2" — 10.0 8.0

Absorption, %

3/16" — 12.9 1/4" 15.3 12.8 10.7 3/8" — E.2 9.4 1/2" — 10.4 8.4

The compression factor, based on the two-day peEet densities and a starting bulk powder density of 43.8 lb/cu ft, ranged from 2.5 to 2.8 for the entire series of tests. The results indicate that, although the larger peEets exhibit lower tensile strengths, peUets with good perform ance properties can be made over a wide range of .sizes. Water content is shown to have a significant effect on the peEet densities that are reaEzed in the peEet mEl.

AE of the peEets described in this Example had good handling properties immediately upon formation of the peEets. Very limited fines were produced; these would normaEy be recirculated to the peEet mEl in a commercial operation. None of the peEets described in this Example exhibited any surface wetting after formation, and no dinkering or sticking together of peEets developed during the curing stages.

In order to demonstrate the immediate handling qualities of the peEets, 500 gms of each of four of the different batches used to evaluate mixing time effects were tested for PeEet DurabEity Index (PDI) at approximately 1 1/2 to 2 hours after formation. The test device is a standard unit manufactured by Carthage Foundry Co., Carthage, Mo. In this test, the peEets are screened over a No. 5 sieve, then rotated for 5 minutes in a rectangular chamber with internal lifts and rescreened. The PDI is the % retained on the sieve at the completion of the test. Values obtained on the test peEets ranged from 88 to 93% PDI, which is considered acceptable for immediate handEng of the peEets. WhEe certain specific embodiments of the invention have been described with particularity herein, it will be recognized that various modifications thereof wEl occur to those sk led in the art. Therefore, the scope of the invention is to be Umited soldy by the scope of the appended daims.