Patent Application
20110049057
1. Field of the Invention
The present invention relates to metals immobilization media and methods for immobilizing heavy or toxic metals (for example As(III), As(V), Cd, Cu, P, Pb, Ni, W, Se(IV), Se(VI), and Zn), in particular with the use of slag fines.
2. Background Art
Environmental remediation is an ongoing concern as our waterways and soils become contaminated with metals and metal alloys. Leaching of metals and metal alloys from contaminated soils or soil-like media into ground water is one particular concern. Options exist for preventing metals present in contaminated soils or soil-like media from leaching into groundwater and for removing or reducing the concentrations of metals present in contaminated water, however there is a continuing need for innovative ways to accomplish environmental remediation.
Metals immobilization using slag fines, such as basic oxygen furnace (BOF) slag fines, also known as steel slag fines (SSF), or blast furnace slag fines (BFF) is an innovative use for slag fines. Slag fines may be a granular media that may correspond to a ⅜ inch minus material, with typically less than 20% by weight passing through a U.S. No. 200 sieve (0.075 mm), though shifts in the gradation to somewhat larger or finer gradations should perform similarly. Slag fines can contain Fe(0) and are geochemically active. Slag fines may immobilize both cationic and oxyanionic heavy metals from aqueous, soils, and soil-like media. The slag fines may immobilize the metals by sorption, precipitation, complexation, oxidation and reduction processes, pH controlled processes, or combinations thereof, without the need for additives. The use of slag fines in environmental remediation is particularly advantageous because it can be potentially accomplished with low cost industrial byproducts and recycled materials.
In some embodiments, slag fines may be used as a firing range soil and/or firing range backstop berm, either alone or in combination with other soils and soil-like media, into which bullets or projectiles are directly fired. The leaching of metals from such slag fine berms is significantly lowered over conventional granular media (i.e., sands, gravels, natural soils) as a result of the metals immobilizing characteristics of the slag fines.
In some embodiments, slag fines may be used as a filter medium (e.g., layer, mat, canister, trench, chimney, reactive subbase, trickling filter) through which heavy metals laden water (mine, surface, ground, leachate or process) flows. The slag fines may partially reduce or wholly eliminate the aqueous phase concentrations of the metals.
In some embodiments, slag fines may be directly blended with heavy metals laden soils and soil-like media (residues, dusts, powders, filter cakes, dredged material, fly ash or other ash materials, sludges, quarry fines, mine spoil, etc.) to simultaneously provide geotechnical enhancements to the soils and soil-like media (e.g., strength, drainage, and/or reduced settlement) while also immobilizing the heavy metals in the soils and soil-like media, and/or water present in pores of the soils and soil-like media, or waters that may eventually pass through the combined media.
In some embodiments, the present invention is directed to a method for immobilizing heavy metals in soil, comprising blending slag fines with soil to form a media, wherein the slag fines contain Fe(0), are geochemically active, immobilize metal present in the media and reduce leaching of the metal into water.
In some embodiments, the slag fines are steel slag fines, blast furnace slag fines, or combinations thereof.
In some embodiments, the slag fines are bordered by, enclosed by, layered with, or prefabricated with a geosynthetic.
In some embodiments, the metal immobilized by the slag fines is a cationic metal, an oxyanionic metal, or alloys thereof. In some embodiments, the cationic metal is selected from the group consisting of cadmium, copper, lead, nickel, zinc, and alloys thereof. In some embodiments, the oxyanionic metal is selected from the group consisting of arsenic, phosphorus, selenium, tungsten, uranium, and alloys thereof. In some embodiments, the metal immobilized by the slag fines is selected from the group consisting of lead, copper, cadmium, zinc, arsenic, selenium, tungsten, uranium, nickel, phosphorus, and alloys thereof.
In some embodiments, the slag fines immobilize the metal through a process selected from the group consisting of sorption, precipitation, complexation, pH controlled processes, oxidation and reduction processes, and combinations thereof.
In some embodiments, the soil comprises soil-like media. In some embodiments, the soil-like media is selected from the group consisting of residue, dust, powder, filter cake, dredged material, ash materials, sludge, quarry fines, mine spoil, and combinations thereof.
In some embodiments, the method further comprises making a berm or berm system for a firing range comprising the media.
In some embodiments, the method further comprises applying a ground covering at a firing range with the media.
In some embodiments, the method further comprises making a geotechnical fill comprising the media. In some embodiments, the geotechnical fill is usable for applications selected from the group consisting of earthwork construction, land reclamation, mine reclamation, brownfields redevelopment, and portfields redevelopment.
In some embodiments, the present invention is directed to a metal immobilization media comprising a blend of slag fines and soil, wherein the slag fines contain Fe(0), are geochemically active, immobilize a metal present in the media, and reduce leaching of the metal into water.
In some embodiments, the slag fines are steel slag fines, blast furnace slag fines, or combinations thereof.
In some embodiments, the slag fines are bordered by, enclosed by, layered with, or prefabricated with a geosynthetic.
In some embodiments, the slag fines are capable of immobilizing a cationic metal, an oxyanionic metal, or alloys thereof. In some embodiments, the cationic metal is selected from the group consisting of cadmium, copper, lead, nickel, zinc, and alloys thereof. In some embodiments, the oxyanionic metal is selected from the group consisting of arsenic, phosphorus, selenium, tungsten, uranium, and alloys thereof. In some embodiments, the slag fines are capable of immobilizing metal selected from the group consisting of lead, copper, cadmium, zinc, arsenic, selenium, tungsten, uranium, nickel, phosphorus and alloys thereof.
In some embodiments, the soil comprises soil-like media. In some embodiments, the soil-like media is selected from the group consisting of residue, dust, powder, filter cake, dredged material, ash material, sludge, quarry fines, mine spoil, and combinations thereof.
In some embodiments, the media is a berm for a firing range.
In some embodiments, the media is a ground covering at a firing range.
In some embodiments, the media is a geotechnical fill. In some embodiments, the geotechnical fill is capable for use in applications selected from the group consisting of earthwork construction, land reclamation, mine reclamation, brownfields redevelopment, and portfields redevelopment.
In some embodiments, the present invention is directed to a method of filtering water, comprising placing a filter comprising a blend of slag fines in a flow path of water, wherein the slag fines immobilize a metal present in the water to reduce an aqueous phase concentration of the metal in the water and the slag fines contain Fe(0) and are geochemically active.
In some embodiments, the slag fines are steel slag fines, blast furnace slag fines, or combinations thereof.
In some embodiments, the slag fines are bordered by, enclosed by, layered with, or prefabricated with a geosynthetic.
In some embodiments, the metal immobilized by the slag fines is a cationic metals, an oxyanionic metal, or alloys thereof. In some embodiments, the cationic metal is selected from the group consisting of cadmium, copper, lead, nickel, zinc, and alloys thereof. In some embodiments, the oxyanionic metal is selected from the group consisting of arsenic, phosphorus, selenium, tungsten, uranium, and alloys thereof. In some embodiments, the metal immobilized by the slag fines is selected from the group consisting of lead, copper, cadmium, zinc, arsenic, selenium, tungsten, uranium, nickel, phosphorus, and alloys thereof.
In some embodiments, the present invention is directed to a metal immobilization filter, comprising a blend of slag fines, wherein the slag fines are capable of immobilizing a metal present in a water passing through the filter to reduce an aqueous phase concentration of the metal in the water and the slag fines are geochemically active and contain Fe(0).
In some embodiments, the slag fines are steel slag fines, blast furnace slag fines, or combinations thereof.
In some embodiments, the filter further comprises a geotextile bordering, enclosing, layered with, or prefabricated with a layer of the slag fines.
In some embodiments, the slag fines are capable of immobilizing a cationic metal, an oxyanionic metal, or alloys thereof. In some embodiments, the cationic metal is selected from the group consisting of cadmium, copper, lead, nickel, zinc, and alloys thereof. In some embodiments, the oxyanionic metal is selected from the group consisting of arsenic, phosphorus, selenium, tungsten, uranium, and alloys thereof. In some embodiments, the slag fines are capable of immobilizing metal selected from the group consisting of lead, copper, cadmium, zinc, arsenic, selenium, tungsten, uranium, nickel, phosphorus, and alloys thereof.
The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate the present invention by way of example, and not by way of limitation. The drawings together with the description, further serve to explain the principles of the invention and to enable a person skilled in the pertinent art to make and use the invention.
A process of immobilizing metals in a variety of sources, such as aqueous bodies, soils, and soil-like media may include providing a metals immobilization media that immobilizes and limits the mass transfer of the metals present in these sources. A metal immobilization media may include slag fines such as basic oxygen furnace slag fines, also known as steel slag fines, or blast furnace slag fines.
Slag is a collection of molten byproducts exiting a furnace during a steel making or metals purification process. As their name suggests, basic oxygen furnace (BOF) slag fines (or steel slag fines) originate from a basic oxygen furnace during the steel-making process. Similarly, blast furnace slag fines may originate from a blast furnace during a steel-making process. Other slags, such as argon oxygen decarburization (AOD) slag, can possess a reactive geochemistry which allows them to be similarly used for metals immobilization. Slag may be recycled, for example, by air cooling followed by crushing and screening processes, as a synthetic construction aggregate for roadway and civil construction. Slag fines are the residual material from such crushing and screening processes and are a granular material. In some embodiments, the slag fines may have a grain size on the order of ⅜ inch minus with less than 20% by weight passing through a U.S. No. 200 sieve (0.075 mm), although somewhat coarser grain sizes are likely to provide similar functions. Also, slag fines may be milled to smaller sizes which may to increase their sorption and buffering capacity. The bulk chemistry in terms of weight percentage of one source of steel slag fines (SSF) and blast furnace slag fines (BFF) is shown for illustrative purposes in
On their own, slag fines may immobilize both cationic and oxyanionic heavy metals and metal alloys present in aqueous, soils, and soil-like media. Exemplary cationic heavy metals that may be immobilized by slag fines may include, but are not limited to, lead, copper, cadmium, nickel and zinc. Exemplary oxyanionic heavy metals that may be immobilized by slag fines may include, but are not limited to, arsenic, phosphorus (as phosphate), selenium, tungsten, and uranium. The slag fines may immobilize the metals by sorption, precipitation, complexation, oxidation and reduction processes, pH controlled processes, or combinations thereof, without the need for additives. Slag fines have sufficient immobilization properties on their own, however, other reagents may also be added to the slag fines such as lime, cementitous materials or salts to facilitate the immobilization. Inert media may also be added as a filler to influence the blend properties. In some embodiments, a metals immobilization media may include slag fines when cationic metals immobilization is a primary concern. In some embodiments, a metals immobilization media may include slag fines when oxyanionic metals immobilization is a primary concern.
The bulk chemistry of iron (Fe) shown in
In order to perform the XANES analyses, duplicate samples of the SSF and BFF media were first pulverized using a Fritsch Planetary Ball Mill. Approximately 10 g of the air-dried sample was placed into milling containers and was pulverized for 20 minutes at a rotation speed of about 250 RPM. The entire pulverized sample was passed through 100 sieve (0.15 mm) to avoid any fractionation. Three iron standards were used as reference for the quantification of Fe. The reference standards for Fe(0), Fe(II) and Fe(III) were nano zerovalent iron (synthesized in laboratory), ferrous sulphate heptahydrate (FeSO4.7H2O, 99% purity, Fisher Scientific, GA, USA) and ferric chloride hexahydrate (FeCl3.6H2O, 98% purity, Fisher Scientific, GA, USA), respectively.
XANES analyses were performed at the Pohang Accelerator Laboratory (PAL) in the Republic of Korea using a beam line 7C (Electrochemistry) with a Si(111) monochromator and ring current of about 120-170 mA at about 2.5 GeV. The Fe K-edge XANES spectra were collected in both transmission and fluorescence modes over the range from about 100 eV below the Fe K-edge at about 7,112 eV to as much as about 900 eV above the edge. In order to remove the energy shift problem, energy calibration was carried out by measuring the XANES spectra of Fe foil (99.99% purity, Exafs Materials, Inc., CA, USA) and the samples simultaneously. Quantitative XANES analyses were conducted using the ATHENA program in the IFEFFIT computer package.
Table 1 summarizes the quantitative XANES results of Fe(0), Fe(II) and Fe(III) in the slag fines, where the total concentrations of iron in the SSF and BFF media were taken to be 131,200 mg/kg and 15,322 mg/kg, respectively, based on the averaged data shown in
Zero-valent iron [Fe(0)] has been used to reduce oxidized forms of heavy metals, including oxyanions such as chromate, selenate, uranyl, etc., and halogenated organic compounds such as TCE, etc., with variable levels of efficiency. In some embodiments, the present invention utilizes SSF, BFF, or other slag fines that contain Fe(0) and that are geochemically active. In some embodiments, the present invention utilizes slag fines wherein at least 1%, at least 5%, at least 10% or at least 15% by weight of the iron present in the slag fines is Fe(0). In some embodiments, the present invention utilizes slag fines wherein 1% to 15%, 1% to 10%, 1% to 5%, 5% to 15%, 5% to 10%, or 10% to 15% by weight of the iron present in the slag fines is Fe(0).
SSF and BFF media are an under-utilized resource for environmental remediation. The unexpected metals immobilizing properties and high buffering capacity of these slag fines (and others) make slag fines ideal component in a metals immobilization media. For example, slag fines may be used alone or in combination with other materials as a metals immobilization media for using in firing ranges as a backstop berm or other range media, as a filtering medium, and as a geotechnical fill material, as will be described in more detail below. In any of these applications, the slag media can be bordered by, enclosed by, layered with and/or prefabricated with geosynthetics (e.g., geotextiles) to enhance the strength, drainage, flow, particle separation and metals immobilization of the slag containing system.
In one embodiment, slag fines of a quality similar to that shown in
Most firing ranges utilize soil impact berms behind the target line to stop bullets from leaving the firing range. These impact berms are usually constructed of mixtures of sand, silt, and clay soils. These soils have very little sorption and immobilization potential for heavy metals, and they must be routinely sieved to remove bullet fragments and/or be changed out. The use of a metals immobilization media such as slag fines in firing range applications as a substitute for the gravelly/sandy/clayey soils currently used as backstop media may prevent or minimize undesirable leaching of metals into the subsurface, surface waters and/or groundwater.
Firing range soils are often contaminated primarily with lead, but also with metal alloys used for jacketing materials of bullets, such as copper, nickel, iron, and cobalt, and other specialty munitions and penetrators, such as tungsten and depleted uranium. These materials may leach into groundwater at unacceptable levels that do not satisfy the Federal or regional water quality criteria.
SSF or BFF media may be utilized as an impact berm (e.g., the trapezoidal embankment material situated behind targets into which bullets ultimately penetrate and come to rest). For example, an impact berm may measure about 100 ft long by 20 ft deep by 20 ft high, and therefore may utilize 40,000 cubic feet of slag fines assuming a unit weight of 120 lb/ft3, or about 2,400 tons. If slag fines are utilized as the material for impact berms in the approximately 3,000 firing ranges in the U.S., approximately 7.2 million tons of slag fines would be used. In addition, if the entire firing range area was made out of slag fines, in order to minimize any potential metals contamination, the amount of slag fines utilized may increase by at least one order of magnitude, i.e. 72 million tons of slag fines. In some embodiments, SSF and BFF media can also be used in berm systems including its attendent runoff, drainage, and water conveyance systems. In some embodiments, SSF and BFF media can be used as a ground covering at a firing range. In some embodiments SSF and BFF media can be mixed or blended with contaminated soils at firing ranges or other locations with contaminated soils.
Lead is a main contaminant of concern at firing ranges. As shown in
The experimentally-determined lead solubility curve of
Table 2 illustrates that SSF and BFF media are capable of immobilizing lead at unexpected levels. To obtain this data, the SSF and BFF media (SP soil by Unified Soil Classification System) were spiked with a lead salt (PbNO3) at target concentrations of 500, 1,000, 5,000 and 10,000 mg Pb/kg. The lead salt and a SPLP solution (20:1 liquid to solid ratio) were simultaneously applied to the SSF and BFF media and then they were rotated in a tumbler for 18 hours prior to SPLP extraction using EPA protocols. Tests were run in triplicate and average results are reported in Table 2.
The BFF media achieved somewhat lower lead solubilities due primarily to pH control in and around the pH associated with the minimum solubility of lead in aqueous systems, which is a pH of approximately 9.0. Lead leaching from the BFF media are unexpectedly low given the fact that there was no mellowing or curing time after lead spiking (
SSF and BFF media illustrate unexpected levels of immobilization for lead and would be of particular use as a lead immobilization media to prevent lead from leaching from contaminated soils or soil-like media into water.
The metals “thresholding” potential of the SSF media was further evaluated to immobilize significant concentrations of several target metals that are relevant to firing range, highway pavement runoff and industrial processes [As(III), As(V), Cd, Cu, P, Pb, Ni, W, Se(IV), Se(VI), Zn]. Aqueous metal solutions were individually prepared by dissolving the corresponding highly-soluble salt in de-ionized (DI) water to achieve target doses equivalent to 100 mg/kg to 100,000 mg/kg to the SSF media, depending on the metal (see Table 3). The SSF media was first wetted (sprayed) with the metals-spiked DI water solution, and then was mixed using a stainless steel spoon to achieve a moisture content of 16%. The metals-spiked SSF media was then stored in sealable plastic bags and allowed to mellow for 30 days. After mellowing, all the samples were air-dried and used for analytical testing.
To batch sufficient SSF media at the 100,000 mg/kg Pb level, the aqueous equivalent of 5,000 mg/L Pb was prepared based on a L:S ratio of 20:1. Due to the volume required, the solution was contacted with the SSF media in a (500 ml polypropylene) standard TCLP bottle that was rotated for 18 hours. The mixture was then transferred to an open stainless steel bowl in a vacuum hood for approximately 24 hours until the free liquid evaporated. Using a spatula, the moist SSF media was then returned to its TCLP bottle, and was mellowed for the balance of the 30 days. A totals analysis of the process revealed negligible loss of Pb.
All reagents supplied by Fisher Scientific (GA).
The sample size for the TCLP and SPLP analyses was reduced to 25 g. Three replicates each were prepared for each target dose for totals, TCLP and SPLP testing followed by ICP-OES analysis for all metals except for the totals digestion of W. The Cd concentration in the TCLP and SPLP leachates was also determined using a graphite furnace atomic absorption spectrometer (GFAAS, Varian Zeeman Spectra AA 220Z). Prior to analysis, the solutions were acidified using concentrated HNO3 in an amount of 1%.
The total digestion procedure for the W-containing samples used a modified US Occupational Safety and Health Administration (OSHA) method, namely, Method ID-213 (OSHA, 1994). Acid digestion was performed using 0.5 gram soil aliquots in 125 mL beakers (Phillips) in a three step process. The first extraction used 2 mL (certified ACS grade) for a minimum of 12 hours (overnight). To this soil-extract solution was added 2 mL each of DI water, 85% H3PO4, and concentrated HCl, HNO3 and H2O2, and the resulting solution was left overnight. Afterwards, 5 mL of concentrated HNO3 was added, and the soil-extractant mixture was then refluxed over a hotplate (190° C.) using a ribbed watch glass cover until the mixture was reduced to approximately 2 mL (essentially H3PO4). After cooling, the digestate was filtered through No. 42 Whatman filter and remaining solution was diluted to 100 mL using DI water prior to ICP/OES analysis. For quality control purposes the analysis also included the digestion of tungsten powder (99.99% purity, Sigma Aldrich, MO) and matrix-spiked samples.
The single metals thresholding results (except for P) are shown in Tables 4 to 6. The As dosing maximums corresponded to measured totals concentrations exceeding 14,000 and 19,000 mg/kg for the As(III) and As(V) series, respectively. Totals concentrations measured above the target dose can occur as an artifact of the small totals sample size (0.5 g), because the finest fraction of the SSF media tends to wick the spraying water by capillarity and accumulate metals due to specific surface effects. The sample size for the TCLP and SPLP allows a more representative grain size distribution and therefore better reflects composite leaching effects. Under TCLP conditions and interpolating between the dosing targets, the SSF media immobilized approximately 8,250 mg/kg As (III) and 9,000 mg/kg As(V) before exceeding the TCLP-As criteria of 5.0 mg/L. Under SPLP conditions, very likely to due pH considerations, the Federal drinking water criteria were satisfied up to As(III) and As(V) totals concentrations on the order of approximately 1,000 mg/kg.
Industrial concentrations of copper, cadmium and zinc were immobilized by the SSF media under both TCLP and SPLP conditions to below several Federal drinking water criteria without any attempt at optimization as shown in Tables 5 and 6. For example, cadmium and nickel were removed below their detection limits (DLs) of 0.05 mg/L. Lead leaching complied with the TCLP standard (5.0 mg/L) for totals concentrations up to 100,000 mg/kg. The SPLP-Pb results were lower, due to the pH issues described above with respect to the differences between the SSF and BFF media. Tungsten removal was also extremely high without any attempt at optimization. Selenium, which is extremely difficult to immobilize, was removed at modest levels. But since selenium was tested at high concentrations herein, the SSF media would likely perform well in natural systems with trace Se contamination or as a filtering medium, as further detailed below as another embodiment of the art.
A series of multi-element thresholding experiments were also conducted with combinations of P, Pb and W (including a P-only control). The main reason for this was that P may have been applied to treat Pb-contaminated firing range soils which may have also contained W, especially on military firing ranges. Concern obviously arises since P has the potential and is known to solubilize W (e.g., aforementioned OSHA method), enhancing the migration potential of W. The goal was to illustrate the ability of SSF to immobilize all three elements simultaneously as a potential remedial approach to P-treatment induced migration of W in firing range soils.
Table 7 shows the P—Pb—W thresholding suite where the pre-existing Pb and/or W contamination was established in the SSF media and mellowed for 30 days as outlined above. The equivalent aqueous doses of Pb and W were 5,000 and 500 mg/L, respectively. The exception to this was the P—Pb—W suite in which the W-spiked SSF media was allowed to mellow and air-dry for approximately 24 hours prior to Pb application by the aforementioned procedure. The Pb—W spiked SSF media was then allowed to mellow for a total of 30 days. A totals analysis of the experimental apparatus revealed no significant mass loss of Pb and/or W by this spiking process.
The pH values of the P spiking solutions (as PO4) to the P—Pb—W spiked SSF media are shown in Table 8. After P spiking, the P—Pb—W spiked SSF media was allowed to mellow for an additional 30 days (60 days total) prior to being subject to TCLP and SPLP extractions. The mellowed pH of the SSF media and the post-TCLP extraction pH values were also monitored. Total P concentrations were not measured because the SSF media used in the experiments itself contained on the order of 10,334±712 mg/kg PO4. Tables 8 to 11 contain the experimental results (average of triplicate measurements) for the P—Pb—W thresholding suite.
The ability of SSF to immobilize PO4 (alone) was very high, as shown in Table 8, which is consistent with the literature on PO4 removal by slags. The PO4 immobilization under TCLP and SPLP extraction conditions was <2 and <0.5 mg/L for equivalent aqueous doses up to 500 mg/L, respectively. The final pH of the system appears to have a strong influence on the PO4 removal efficiency of the SSF media.
Table 9 shows the joint immobilization potential of Pb and PO4 for SSF media initially containing 100,000 mg/kg Pb. In every case, the PO4 concentration was less than 0.6 mg/L (>90% removal) and the corresponding Pb concentrations were <0.25 mg/L (100.00% removal) for TCLP conditions and were <290 mg/L (94.33% removal min.) for SPLP conditions. An important difference between the TCLP and SPLP conditions was the pH of the extracted leachate. Pb concentrations were lower at moderate pH, PO4 concentrations remained below the DL (0.5 mg/L) for nine of ten experiments. The presence of PO4 improved the immobilization of Pb over the Pb-only spiked SSF media (Table 4) by an order of magnitude under TCLP conditions, but the performance was mixed under SPLP conditions.
Table 10 presents the joint immobilization potential of W and PO4 for SSF media initially containing 10,000 mg/kg W. In every case, the PO4 concentration was <0.5 mg/L, except for the 10,000 mg/kg PO4 spiked system under TCLP conditions (1.2 mg/L). The corresponding W concentrations varied between 2 and 16 mg/L with no apparent trend under TCLP conditions, but the W concentrations increased with increasing PO4 concentrations under SPLP conditions. The presence of PO4 improved the immobilization of W over a W-only system (Table 5) by approximately 25% to 65% under TCLP conditions, but the concentration of W appeared to increase with increasing PO4 dose under SPLP conditions from approximately 0.65 times less to 4.5 times more than the Pb-only system.
Table 11 shows the joint immobilization potential of Pb, W and PO4 for SSF media initially containing 100,000 mg/kg Pb and 10,000 mg/kg W. The mellowed pH values for the Pb—W control values (no P) and the P—Pb—W spiked SSF media mirrored the pH values for the P—Pb spiked SSF system, and were approximately 1 pH unit less than the P—W system. The TCLP extracted pHs were the most uniform and lowest as a group and for the P—Pb—W spiked SSF media followed by the P—Pb systems. The highest pHs were obtained for the P—W system at the lowest P spiking levels. For SPLP extraction conditions, the highest pH values were obtained for the P—W system at low P doses, but thereafter the system results were generally similar but were ordered by the lowest average pH as P—Pb—W, P—Pb and P—W.
It is important to note that P is not required to achieve high Pb and W immobilization levels in SSF media as evidenced by the data contained in Table 5 and the Pb—W control data shown in Table 11. Rather, the SSF media is also effective in immobilizing P, without significant reduction in (and even enhancement to) Pb and W immobilization. While the PO4 concentrations fluctuated around the DL (0.5 mg/L) under TCLP conditions for all P dosing levels, PO4 was BDL under SPLP conditions for all P dosing levels.
The TCLP-Pb concentrations were lowered by an order of magnitude with P addition in the Pb—W controls and Pb-only spiked SSF media (Table 5), the results being similar but slightly higher than the P—Pb spiked SSF media for corresponding P dosing levels. However, the SPLP-Pb concentrations generally increased with increasing P dosing level (versus the Pb—W control), and overall, the SPLP-Pb concentrations were less than the corresponding P dosing levels in the P—Pb spiked system.
W immobilization was highest in the P—Pb—W spiked media, followed in order by the Pb—W control data (Table 11), the P—W spiked SSF media (Table 10) and W-only spiked media (Table 5) for corresponding dosing levels for TCLP and all SPLP data sets except the P—W spiked SSF media, which has the lowest removal rates, potentially due to P-induced increases in W solubility, or oxyanionic competition for precipitating cations. Also, the presence of Pb appears to have a larger influence on W immobilization than P, perhaps due to the formation of stolzite (PbWO4). The increased immobilization of W from the P—W system to the P—Pb—W system due to the presence of Pb is generally on the order of a magnitude or more. The W concentrations in the P—Pb—W spiked SSF are the lowest observed for the tested experimental conditions and seemingly independent of P dose.
The ability of the SSF media to immobilize heavy metals is quite rapid, as evidenced by batch equilibration experiments. Batch kinetics experiments were conducted using aqueous solutions at a dose equivalent to 10,000 mg/kg per metal for As(III), As(V), Cd, Cu, Ni, W and Zn, 100,000 mg/kg for Pb, see Table 12. The selected sampling intervals were 1, 2, 5, 10, 60 minutes, and 3, 6, 12 and 18 hours. Sacrificial samples were prepared in triplicate for each metal and time interval. For each replicate, 6.5 g of dry SSF media was placed in 130 mL bottle. For all samples, aqueous metal solution was added to the sample (total liquid volume 130 mL) using a L:S of 20:1, identical to the TCLP/SPLP procedure. The resultant slurries (in triplicate; A to C) were mixed in an end-over-end rotator at 30 rpm. At each interval, triplicate samples were removed and the supernatants were filtrated through a 0.45 μm nylon membrane filter. The pH of the leachate was recorded using an Accumet AR20 pH-meter. All samples were stored in the refrigerator at a temperature of 4° C. before they were analyzed by ICP-OES. Table 12 also summarizes the pH of the equilibrated salt solutions and the final leachate pH (18 hours).
The initial pH of the aqueous salt solution (pHo) of the cationic metals, Cd, Cu, Ni and Pb was less than 5 and for Zn, the pHo was less than 7. The pHo for the oxyanions, As (III), As(V) and W was greater than 7. Metals removal was imparted due to the pH change in some systems. For the cationic metals, Cu, Ni, and Zn, the leachate pH was increased to 11 within 5 to 10 min, due to the strong buffering capacity of the SSF media. For Cd, the leachate pH was increased to 11 within 1 min, probably due to the hydration behavior of the Cd cation. For Pb, whose dose was 10 times higher than the other metals, the leachate pH increased to 11 within 3 hrs.
The percent removal of all the metals by SSF was about 99% after 18 hrs mixing time, except for Pb, which was about 93% (
The removal of Cd, Cu, Ni and Zn appear to be largely pH controlled (
In another embodiment, slag fines may be used as a filter medium (e.g., layer, mat, canister, trench, chimney, trickling filter, reactive subbase) through which heavy metals laden water (mine, surface, ground, leachate or process) flows either in a loose or compacted state. The slag fines may partially reduce or wholly eliminate the aqueous phase concentrations of the metals. The slag fines may be used alone or mixed or blended with a sand or other inert or sorptive granular media to form a metals immobilization media that acts as a filtering medium.
In one example, a layer of slag fines could be placed as a drainage layer directly underneath a contaminated soil or soil-like media structure or pile, such as a traditional impact berm in a firing range, fly ash monofill, or heap leach pad to treat infiltrating water that contains metal contaminants prior to it passing into the subsoil. The layer can be placed loose or compacted to act as a semi-permeable metals immobilizing layer. In another example, a trench may be filled with slag fines at the bottom of a sloped structure or pile of contaminated soil or soil-like media, such as the bottom of an embankment, to intercept, treat, and divert water runoff. In another example, slag fines may be placed between two layers, such as a non-woven fabric or other geosynthetic, to form a reactive mat or pad that may be used as a drainage layer directly underneath a contaminated soil or soil-like media structure fill or pile, such as fly ash monofill. In such an instance, the slag fines would act as a reactive layer that intercepts and treats contaminated water prior to release. In general, the slag fines may be placed in the flow path of contaminated water to act as a filter medium to partially reduce or wholly eliminate the aqueous phase concentrations of metals in the contaminated water.
In one embodiment, SSF and BFF media and their blends, for example those having the bulk chemistry shown in
Acid neutralization capacity (ANC) testing was performed on the SFF and BFF media to measure the buffer capacity of each system in response to an acid attack, which could be taken to simulate time in the case of an acid rain exposure condition, e.g., repeated additions (equivalents) of an acid of known strength, as shown in
The raw SSF, BFF, and DM media having the environmental quality shown in
The ANC test procedure was based on the Generalized Acid Neutralizing Capacity test of Isenberg and Moore (1992). The procedure consisted of equilibrating the soil samples to increasing equivalents of reagent per kilogram of dry soil. Specifically, 6.5 grams of each dry sample were placed in a series of 130 mL bottles. For strongly alkaline media, incremental amounts of 15.8N nitric acid (HNO3) were added to the sample (total liquid volume 130 mL), using a liquid:solid ratio of 20:1, identical to the TCLP and SPLP procedures. For the DM, incremental amounts of 10N sodium hydroxide (NaOH) were added to the sample to illustrate the impacts of alkalinity on the sample.
The resultant slurries were tumbled in a standard TCLP rotating extractor for 48 hours. The supernatants were then filtrated through a 0.45 μm nylon membrane filter, and the pH of the leachate was recorded using an Accumet AR20 pH-meter. All samples were stored in the refrigerator at a temperature of 4° C. before they were analyzed for testing of pH total and dissolved As, Cr, and Fe.
The SSF media consumed approximately 7 acid eq/kg to reach a neutral pH (7) and approximately 11 acid eq/kg to achieve a pH of about 3. The response of the SSFM sample was much stronger, requiring approximately 10 and 18 acid eq/kg to attain a pH of about 3. This suggests that destroying the morphology of the SSF media liberates alkalinity due to grain size effects, whether it be in the form of lime or other pH buffering minerals.
From the ANC data, the SSF media will likely persist at an elevated pH for long periods. This has the double effect of preventing As leaching from the DM and keeping the pH above the range to support microbial transformations of the DM which lead to its ultimate acidification.
To illustrate how the SSF media buffered the DM, the DM and SSF media were proportioned and blended to prepare the 80/20, 50/50 and 20/80 DM-SSF blends (dry weight basis) where the DM content is reported first. These blends were homogenized and cured for 7 days prior to ANC testing (Example 3). The DM and SSF media were evaluated as is without a curing time. None of these samples were As spiked.
The As in the DM is liberated very easily and quickly exceeded the drinking water criteria, whereas the As concentration from the SSF and SSFM samples never exceed the detection limit (which fluctuates due to matrix issues). The leaching of Cr from DM increases dramatically with increasing pH, as expected.
The USEPA Cr drinking water standard is shown at 0.1 mg/L. The Cr is likewise liberated from the DM relatively easily with increasing pH. On the other hand, it takes 12 to 14.5 eq/kg to release Cr from the SSF media (6× to 7× TCLP condition) to values above the Federal drinking water criteria.
To further illustrate the strong buffer capacity of SSF and its corresponding release of metals under acid attack, the doses of metals shown in Table 12 were used to individually spike the SSF media followed by the mellowing procedure outline above. After mellowing, they were subject to ANC testing as previously described using triplicate samples for each acid equivalent added. The release of Ca, Fe and Mg, which were indigenous to the SSF media itself, were also measured.
To determine if the release of metals was solubility controlled (aqueous system), the concentration-pH behavior of each heavy metal was simulated using Visual Minteq, Version 2.61 (David and Allison, 1999). The target dose of each heavy metal was used as the control concentration (Co), see Table 12. The control concentrations of Ca, Fe and Mg were selected based on a comparison of the converted measured totals (XRF) and total digestion data, as shown in Table 13. The totals data were taken to be more representative based on the accuracy (and aggressiveness) of the extraction process. Iron was further speciated as Fe(II) and Fe(III) directly in the sold phase using (XANES), see Table 1. Also shown in Table 13 are the respective Ca, Fe and Mg concentrations in the ANC controls.
abased on XANES speciation
The aqueous database of Minteq was augmented with the aqueous data of the relevant species cited from literature. The solubility curves for different elements obtained from the Minteq model were plotted with the experimental data. For Pb only, an experimentally-determined Pb solubility curve obtained from synthetic Pb solutions (Dermatas and Meng, 2003) is also shown for comparison purposes because Minteq modeling results did not match other Pb solubility data in the literature (USEPA, 1986 and others). The Minteq predicted Pb curve shows it to be less soluble than other elements it is known not to be less soluble than (e.g., Cd). It is worth noting here that all Minteq simulations were run for pure systems and did not consider complex phenomena due to multi element systems such as the ones at hand. Moreover, adsorption modules were not considered in the simulation runs.
In terms of buffering capacity, the control sample (raw SSF) consumed approximately 7 eq/kg to achieve a neutral pH, and approximately 12 eq/kg to achieve pH˜3 as shown in
In one embodiment, the metal immobilization media may include a mixture of slag fines directly blended with heavy metals laden soils and soil-like media (residues, dusts, powders, filter cakes, dredged material, fly ash or other ash materials, sludges, quarry fines, mine spoils etc.) to simultaneously provide geotechnical enhancements to the soft, weak, or wet soils and soil-like media (e.g., strength, reduced settlement) while immobilizing the heavy metals contained in the soils and soil-like media, and/or water present in pores of the soils and soil-like media, and/or in waters that may eventually pass through the combined media.
The mixture of slag fines and soil or soil-like media may be used in geotechnical fill applications, such as for example, general or bulk, embankment, and structural fill applications. The fill materials may be used in transportation, airport, building, and maritime construction; land reclamation; and brownfields and portfields redevelopment in urban areas. In order to qualify as an engineered fill, the compacted material must generally have a minimum compacted density greater than or equal to 90% relative compaction (RC) of the maximum dry density (MDD) attained by standard Proctor compaction (ASTM D698) up to a typical maximum of 95% RC of MDD attained by modified Proctor compaction (ASTM D1557).
Compaction data for 100% DM, 100% SF, and 20/80, 50/50 and 80/20 DM-SF blends, separately using both SSF and BFF was also compiled to determine the unit weights of each mixture as fill material.
In one example, slag fines may be mixed with soft soils or soil-like media (e.g., clays, loess, collapsible soils, quarry fines, mine spoil and other wet, soft media) to make engineered fills. Major benefits may be expected when blending slag fines with fresh DM, which is material removed from the bottom of a body of water and may include sediment, soil, silt, and clay. The slag fines may be SSF or BFF media. Partially de-watered DM, such as that having the bulk chemistry shown in
A summary of the total concentrations (mg/kg) based on five replicates of the randomly sampled, SSF and BFF media and DM are shown in
The SSF and BFF media pass MD residential criteria for everything except arsenic and iron. Arsenic is non-leachable, see below. Iron is a secondary water quality parameter, and its total concentration is not even regulated in neighboring states due to its non-hazardous nature. Chrome (1,100 mg/kg) was successfully spectated using the alkaline digestion process (USEPA 3060/7196 series) and were confirmed using X-ray absorption near edge spectroscopy (XANES), a new non destructive testing technique that has the ability to directly speciate and measure the concentrations of Cr(VI) and Cr(III) in the solid phase.
The TCLP and SPLP extractions were performed on the DM, the SSF and the BFF media. These results are summarized in
Excluding the individual hits on iron in both the DM and SSF which are significantly above the detection limits and appear to be outliers, only four metals (copper, mercury, nickel and zinc) were detected above their respective TCLP limits, all of them in the DM. Three metals (copper, mercury and zinc) were detected via SPLP in the DM, all below the MD Type HI aquifer criteria. Only zinc was detected in the SSF SPLP extract, which was likewise below the MD Type NI aquifer media. Additionally, despite the iron content of the SSF media of approximately 13.12 wt % (131,000 mg/kg), the SPLP leaching of iron was below the MD Type I/II aquifer criteria.
Based on the totals, TCLP and SPLP leaching behavior of the metals SFF and BFF media excess the environmental quality of the DM, making blending an attractive approach to recycle DM in large fill applications. The main purpose of this testing was to illustrate the ability of the SSF and BFF media to immobilize significant concentrations of arsenic occurring in the DM without additional additives and to set/propose an upper limit (totals) that could be tolerated based on understanding the arsenic leaching from the compacted dredged material-slag fines (DM-SF) blends. The main environmental motivation for blending slag fines with dredged material is that the geochemical characteristics of the slag fines (especially pH) are such that they have the ability to immobilize significant concentrations of arsenic (either As(III) or As(V)) in the DM.
In another example, DM naturally containing approximately 20 to 30 mg/kg As, was intentionally and individually spiked with 100 mg/kg As(III) or As(V) to evaluate how successful the SSF and BFF media would be in immobilizing As. In this way, the DM acted as the source of As contamination. Immobilization of arsenic was tested using 100% DM, 100% SF, and 20/80, 50/50 and 80/20 DM-SF blends, using both SSF and BFF media. Sodium arsenite (NaAsO2) and sodium arsenate heptahydrate (Na2HAsO4.7H2O) were used as the As(III) and As(V) sources, respectively (ACS grade, Fisher Scientific, GA). The DM was spiked separately with 100 mg/kg (dry weight basis) of As(III) and As(V). Aqueous solutions of each arsenic salt were prepared using deionized water and were targeted to achieve final bucket moisture content on the order of 130 to 135%. This allowed the raw DM to have a liquid consistency sufficient to enable its homogenization with the much drier, granular SSF and BFF media.
Raw DM was placed in a Globe SP-30 mixer (Dayton, Ohio) and was homogenized as-is for approximately 10 minutes. Thereafter, during mixing, the arsenic solution was gradually introduced to the DM using a series of polyethylene squirt bottles. Mixing was paused every 10 minutes to manually scrape the excess DM off of the bucket sidewalls using a spatula. This material was blended by hand for some moments, and then the automated mixing bucket was restarted until a total of 40 minutes elapsed. The DM was then removed and the device was cleaned prior to mixing the next batch. Individual batches of As-spiked DM were stored in sealable 5-gallon bucket for a mellowing period of 30 days to enable equilibration.
In the DM-SF blends (where SF=SSF or BFF), the As source was always the DM. However, the SSF and BFF media were spiked with As to explicitly evaluate its ability to immobilize As as a raw material. The spiking procedure for the SSF and BFF media involved As(III) and As(V) concentrations of 100 mg/kg and is described above. The As spiked SSF and BFF media were then stored in 1 gallon sealable plastic bags and allowed to mellow for 30 days.
The thresholding suite consisted of the DM, SF and the 20/80, 50/50 and 80/20 DM-SF blends (dry weight percent of DM reported first for the blends). These samples were blended immediately following the mellowing period of DM. Upon homogenization, they were compacted to greater than 90% of the maximum dry density by Standard Proctor compaction (ASTM D698), removed from the molds then cured in sealable plastic bags for an additional 30 days.
In another example, the mixtures of Example 4 (100% DM, 100% SF, and 20/80, 50/50 and 80/20 DM-SF blends, using both SSF and BFF media) were tested with the addition of 0% NewCem (NC) or slag cement, 0.5% NC, 1.0% NC, 1.5% NC, and 2.0% NC. The results are shown in
Swell data was obtained for 100% DM, 100% SF, and 20/80, 50/50 and 80/20 DM-SF blends, using both SSF and BFF media using the same sample preparation process described in Example 7. This data was compared to the swell behavior of reference of 3:1 kaolinite-bentonite clay blend.
As can be shown From the above data and examples, the SSF and BFF media illustrate unexpected results in stabilizing DM to immobilize oxyanionic metals, particularly arsenic. Blending the SSF and/or BFF media with soils or soil-like media contaminated with arsenic may be a successful way of immobilizing the arsenic therein. In addition, the combination of DM and slag fines may harden over time enhancing its usefulness as a fill material. It is believed that the lime present in the slag fines may be a contributing factor in the immobilization properties of the slag fines. In addition, the slag fines, particularly the SSF media, exhibit strong pH control and buffering capacity. This may enable the structures and fills formed from slag fines and DM-SF blends to also act as a filtering medium for contaminated water that passes through the structures or fills.
As discussed above, slag fines have many beneficial uses in metals immobilization. For example, slag fines may be used alone or in combination with other materials as a metals immobilization media for using in firing ranges as a backstop berm, as a filtering medium, and as a geotechnical fill material.
The foregoing description of the specific embodiments will so fully reveal the general nature of the invention that others can, by applying knowledge within the skill of the art, readily modify and/or adapt for various applications such specific embodiments, without undue experimentation, without departing from the general concept of the present invention. Therefore, such adaptations and modifications are intended to be within the meaning and range of equivalents of the disclosed embodiments, based on the teaching and guidance presented herein. It is to be understood that the phraseology or terminology herein is for the purpose of description and not of limitation, such that the terminology or phraseology of the present specification is to be interpreted by the skilled artisan in light of the teachings and guidance.
This application claims the benefit of U.S. Appl. No. 61/239,285, filed on Sep. 2, 2009 and of U.S. Appl. No. 61/307,231, filed on Feb. 23, 2010, both of which are hereby incorporated by reference in their entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61307231 | Feb 2010 | US | |
| 61239285 | Sep 2009 | US |
