Evaluation of waste materials for acid mine drainage remediation

Fuel 188 (2017) 294–309

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Evaluation of waste materials for acid mine drainage remediation Stephanie N. Jones a, Bora Cetin b,⇑ a b

Geosyntec Consultants, Columbia, MD, USA Department of Civil, Construction, and Environmental Engineering, Iowa State University, Ames, IA 50011, USA

a r t i c l e

i n f o

Article history: Received 6 July 2016 Received in revised form 19 September 2016 Accepted 3 October 2016

Keywords: High carbon fly ash Recycled concrete aggregates Metals Leaching pH Alkalinity

a b s t r a c t Laboratory scale tests were conducted to assess the efficiency of two different types of waste materials to remediate acid mine drainage (AMD). The waste materials used in the current study were recycled concrete aggregates (RCAs), and fly ashes. Four different RCA materials and three different fly ash materials were evaluated. Column leach tests (CLTs) were conducted to determine the effects of the remediation materials on pH, electrical conductivity, alkalinity, oxidation reduction potential (Eh), and concentrations of sulfate (SO24 ), chromium (Cr), iron (Fe), manganese (Mn), copper (Cu), and zinc (Zn) in AMD. Results of the CLTs suggest RCAs and one of the highly alkaline fly ash can effectively raise pH of the AMD and reduce concentrations of Cr, Cu, Fe, Mn, and Zn in AMD. In addition, sulfate concentrations of AMD decreased significantly after being treated by RCAs while sulfate concentrations of the AMD increased when it was remediated by fly ashes. It is speculated that leaching of sulfate from fly ash samples during treatment may decrease the metal sorption capacity of fly ashes. X-ray fluorescence spectroscopy quantified the impact of CaO and loos on ignition (LOI) in the remediation materials on sorption capacity of metals from the AMD. Sorption capacity for Cr, Cu, Fe, and Zn was found to be greater in materials with high CaO and LOI content, and low unburned carbon. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Acid mine drainage is a caustic by-product of hard rock mining that can terminally impair waterways and alter landscapes. It occurs as a result of mining operations when sulfide-bearing rock oxidizes to form sulfuric acid in the presence of atmospheric oxygen and water (precipitation, groundwater, or surface water) [1–3]. The sulfuric acid dissolves surrounding rock, releasing metal ions into solution. The resulting leachate is referred to as acid mine drainage (AMD). It is highly acidic (pH typically < 3) and can contain high concentrations of trace metal ions including iron (Fe), zinc (Zn), manganese (Mn), nickel (Ni), cadmium (Cd), cobalt (Co), copper (Cu), and aluminum (Al) [4]. When AMD enters surface water, groundwater, and soil it can significantly lower pH and raise metal concentrations [5]. When feasible it is of best practice to prevent the formation of AMD using source control measures such as sealing or flooding of underground mines, solidification of mine tailings, and disposal of mine wastes in sealed waste heaps [1]. Unfortunately, these treatment methods can be extremely costly, requiring continuous chemical input and large volumes of virgin material, such as lime ⇑ Corresponding author. E-mail address: [email protected] (B. Cetin). http://dx.doi.org/10.1016/j.fuel.2016.10.018 0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.

and limestone. In an effort to reduce the high costs of AMD treatment, interest has been developed in various applications of lowcost waste products, such as fly ash and recycled concrete aggregate (RCA), for AMD remediation. Both RCA and fly ash are highly alkaline, exhibiting unique binding properties that could make them effective alternatives to costly, lime and limestone treatment. The application of fly ash for AMD remediation is a sustainable approach that could be advantageous to coal mines and power plants generating fly ash as a means to repurpose waste. Coal combustion by-products, including fly ash, are one of the nation’s most widely produced industrial waste products. In 2012, the United States coal industry produced over 53.4 million tons of fly ash [6]. Unfortunately, less than 50% of fly ash is reused, and the remaining is disposed of in landfills or retained in ponding facilities [6]. Recent studies have experimented with mixing fly ash with mine tailings to improve pH and reduce metal concentrations in AMD. Shang et al. [7] found that co-disposing of fly ash and mine tailings neutralized the pH of pore fluid, and reduced effluent concentrations of heavy metals to meet local regulatory requirements for leachate quality. Similarly, Mohamed et al. [8] found that mixing mine tailings with a mixture of aluminum, lime, and fly ash (ALFA) effectively reduced leaching of heavy metal from the mine tailings. Siriwardane et al. [9] and Bulusu et al. [10] investigated

295

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

used in this study. (2) The Methods section discusses the experimental test procedures used in the current study, including (a) preparation of samples for CLTs, (b) collection of effluent solutions from CLTs, (c) measurements of Ca, Cu, Cr, Fe, Mg, Mn, and Zn concentrations by atomic absorption spectrometry, (d) measurements of pH, electrical conductivity (EC), and oxidation-reduction potential (Eh), (e) measurement of alkalinity values and sulfate (SO24 ) anion concentrations, and (f) XRF and SEM analytical methods to investigate the surface characteristics of materials tested in the current study. (3) The Results section discusses the results, including: (a) the change in pH and Eh in AMD after remediation, (b) the change in metal concentrations in AMD and its correlation with EC, sulfate concentrations and explanation of changes on the particle surface characteristics via XRF and SEM analyses, (c) the change in alkalinity in AMD and its correlation with leached Ca concentrations, (d) impact of CaO content on AMD remediation, and (e) impact of LOI content on AMD remediation. (4) The Conclusions section provides a summary of the results section and recommendations of utilization of RCA and fly ash for such applications for future field studies.

the use of fly ash grouts to reduce mine flow rates and serve as permeable reactive barriers (PRBs) in abandoned mines. Siriwardane et al. [9] found when injected into an abandoned mine fly ash grout could successfully reduce mine flow rates by 95% while exceeding laboratory strength requirements for subsidence control. Furthermore, Bulusu et al. [10] used unconfined compressive strength, slump, and modified flow tests to demonstrate that fly ash grout acted as a permeable reactive barrier by effectively reducing the production of AMD in field applications. In addition to fly ash applications, the application of recycled concrete aggregate (RCA) for remediation of AMD could conserve landfill space while also providing effective, low-cost treatment for AMD. Currently, only about 50% of the over 200 million metric tons of RCA produced annually in the U.S. is reused [11]. The primary reuses for AMD are as road base, components of new concrete or asphalt mixes, high-value riprap, and as low-value fill products [12]. Studies have shown that RCA is strong, porous, and composed of an assortment of alkaline materials (including limestone, calcium-bearing minerals, and Portland cement) with high neutralization and sorption capacities that could make it an effective material for AMD treatment. Indraratna et al. [13] revealed that when used as a reactive material in PRBs, recycled concrete removed Fe and Al from groundwater and improved its pH from acidic to mildly alkaline. In addition, Bestgen et al. [14] and Chen et al. [15] found that fine particles of RCA (less than 0.075 mm) had greater acid neutralization capacities than coarse particles, and Cu and Zn experienced the lowest leaching at pH less than 5 and highest leaching at pH of 2. The objective of this study, through laboratory testing and analysis, was to evaluate the low-cost construction waste products, recycled concrete aggregate and fly ash, as remediation materials for AMD treatment. Column Leach Tests (CLTs) were conducted to assess the impact of fly ash and RCA on pH, electrical conductivity, alkalinity, oxidation reduction potential (Eh); and concentrations of Ca, Cu, Cr, Fe, Mg, Mn, and Zn in AMD. Following the CLTs, additional analytical methods were conducted to better understand the treatment process on a molecular basis. X-ray fluorescence spectroscopy (XRF) was used to evaluate the impact of oxide, alkalinity, and unburned carbon content of the remediation materials on their capacity to sorb metals from AMD. In addition, scanning electron microscopy (SEM) was utilized to observe the effects of fly ash and RCA treatment on the physical properties of the remediation materials. This article includes the following sections. (1) The Materials section covers a description of the physicochemical properties of the mine waste, four different RCAs, and three different fly ashes

2. Materials Mine waste recovered from the Homestake Mine in Lead, SD was used to generate acid mine drainage. Before closing in 2002, the Homestake Mine served as the largest and deepest gold mine in North America [16]. Table 1 provides the percent composition of oxides and metals in the mine waste from Homestake Mine. XRF and ICP-MS analysis of the mine waste revealed it is primarily composed of SiO2 (54.9%) and Al2O3 (16.3%). It also contains high Cu, Fe, Zn, and Mn contents which could pose a significant risk to the environment and human health if leached from the material. Four RCAs collected from various highway surfaces and recycled concrete plants in South Dakota were used in the study. The RCAs are referred to as RCA1, RCA2, RCA3, and RCA4. RCA1 and RCA2 consist of pulverized concrete from former highway pavement surfaces in Philip and Pierre, South Dakota. RCA3 and RCA4 were obtained from stockpiles in Sioux Falls and Rapid City, South Dakota. Fig. 1 provides a particle size distribution of the RCAs. The particle size distribution shows that RCA2 and RCA3 are predominately (<40% by weight) composed of gravel particles (>4.75 mm), while RCA1 and RCA4 are primarily (>40% by weight) composed of smaller sand particles (75 lm to 4.75 mm). The physical properties of the RCAs are summarized in Table 2. According to the Unified Soil Classification System (USCS), RCA1, RCA3, and RCA4 are

Table 1 Chemical composition and total metal content of mine waste, fly ash, and RCA’s. Material

MW

FA1

FA2

FA3

RCA1

RCA2

RCA3

RCA4

pH LOI (%)

2.02 4.9

6.1 6.2

6.6 6.8

9.5 8.1

10.85 25.2

12.52 12.0

12.36 7.3

11.69 32.7

Major oxide content (% by weight) Al2O3 16.3 CaO 0.11 Cr2O3 <0.01 Fe2O3 7.8 MgO 0.37 MnO 0.01 SiO2 54.9

23.1 1.07 <0.01 3.2 0.60 0.06 45.1

26.9 0.70 <0.01 5.5 0.20 0.06 50.8

25.5 12.5 <0.01 13.7 1.9 0306 50.4

0.44 37.9 <0.01 4.6 2.6 0.05 32.3

3.2 45.1 <0.01 6.1 2.0 0.08 37.9

2.0 52.5 <0.01 4.5 1.3 0.04 48.3

1.4 55.8 <0.01 4.2 2.0 0.03 35.1

Metal content (mg/kg) Cr 43 Cu 36.1 Mn 35 Zn 38

15.5 59.6 33.9 53.94

68 33 215.6 28.78

42 36 207.7 83.96

22 16 355 28

16 9.3 579 36

14.7 8.7 330 38

11 12.4 244 53

Note: LOI: loss on ignition. MW: mine waste, FA: fly ash, RCA: recycled concrete aggregate.

296

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

sess the greatest capacity to sorb aqueous components of AMD [17]. Table 1 also shows that the RCAs contain over 37% CaO by weight. The high CaO content in the RCAs is typical of cementitious components of most concretes derived from portlandite (Ca(OH)2) [18,19]. CaO is a known neutralizing agent that is likely responsible for the alkaline pH (>10) of the RCAs. Unlike the RCAs, Table 1 suggests that the fly ashes contain CaO content by weight < 5% except the FA3.

100 RCA1 RCA2 RCA3 RCA4

Percent Finer (%)

80

60

40 3. Methods 3.1. Column leach test procedures

20

0 100

10

1

0.1

0.01

Particle Size (mm) Fig. 1. Particle size distribution of recycled concrete aggregates (RCAs).

well-graded gravel (GW), and RCA2 is a poorly-graded gravel (GP). The coefficients of uniformity (Cu) of RCA materials ranged from 7.8 to 31.3, and coefficients of curvature (Cc) of RCA materials ranged from 1.6 to 1.9. Three fly ashes collected from the two different power plants (FA1 and FA2) in Maryland, and one from West Virginia (FA3) were used in the study. The fly ashes studied are primarily composed (>80% by weight) of fine particles. Table 2 summarizes the physical properties of the fly ashes used in the study. All three fly ashes are classified as off-specification pursuant to ASTM C618, and as lowplasticity silt (ML) in consonance with the USCS Classification System. The specific gravity of the fly ashes and RCAs used in this study range from 2.18 to 2.42 kg/m3 (ASTM D854). The optimum water content (wopt) of the materials is between 10 and 26%, while the maximum dry unit weights (cdry-max) vary from 9.96 to 20.4 kN/ m3 (ASTM D698). A summary of the chemical composition and total metal content of the RCAs and fly ashes is provided in Table 1. XRF and ICP-MS analysis concede that the RCAs and fly ashes have high loss on ignition values (LOI > 6%) which are indicative of unburned carbon content in ash. LOI values of the RCAs and fly ashes ranged from 7 to 32% by weight and 6 to 8% by weight, respectively. Considering that carbon has a high specific surface area that increases sorption capacity, it can be inferred that the studied materials with the highest unburned carbon content pos-

Column leach tests (CLTs) were performed to determine the leached metal concentrations from compacted columns of mine waste, RCA (RCA1, RCA2, RCA3, and RCA4), and fly ash (FA1, FA2, and FA3). CLTs were also used to determine the effectiveness of RCAs and fly ashes in reducing metal concentrations of AMD produced from mine waste under simulated in situ leaching conditions. CLTs were conducted in two series. In the first series, deionized water (ASTM II Type) was pumped through compacted columns of RCA and fly ash to measure the leachability of metals from the RCA and fly ash materials. In the second experimental series of CLTs, AMD was generated from compacted columns of mine waste and pumped through compacted columns of fly ash and RCA. Periodically, samples of untreated AMD (before it was pumped through a RCA or fly ash column), and treated AMD (after it passed through a fly ash or RCA column) were collected for analysis. The pH, electrical conductivity, oxidation reduction potential; and alkalinity, sulfate, and leached metal concentrations of the AMD samples were compared to determine the effectiveness of the RCA and fly ashes in treating AMD. Pictures of the column leach testing apparatuses are shown in Fig. 2. 3.2. Column preparation The fly ash columns used in both series of CLTs were compacted at their optimum moisture contents (22–26%) into PVC pipes (4in.  12-in.). The fly ashes were not sieved prior to compaction because they are primarily composed of fine particles. The RCA columns used in both series of CLTs were created by compacting RCA particles (sieved to less than 3/8 in.) at their optimum moisture contents (10–15.8%). The fly ash and RCA columns were compacted at optimum moisture content in 8 layers with 29 blows per layer (ASTM D698). The mine waste columns required to generate AMD in the second series of CLTs were constructed by compacting

Table 2 Physical properties of RCAs and fly ashes. Physical property

FA1

FA2

FA3

RCA1

RCA2

RCA3

RCA4

Cu Cc Gs (kg/m3) Wopt (%) Ydmax (kN/m3) Gravel (>4.75 mm) (%) Sand (4.75 mm–75 lm) (%) Fines (675 lm) (%) AASHTO classification USCS group symbol ASTM C 610 classification

N/A N/A 2.18 26 11.87 N/A N/A 80 A-2-4 (0) ML Off-Spec.

N/A N/A 2.42 22 9.96 N/A N/A 95 A-2-4 (0) ML Off-Spec.

N/A N/A 2.37 25 13.18 N/A N/A 80 A-2-4 (0) ML Off-Spec.

29 1.6 2.35 10 19.8 39 57.6 3.4 A-1-a (0) GW N/A

7.8 1.9 2.38 14.6 17.8 72.6 26.3 1.1 A-1-a (0) GP N/A

31.3 1.7 2.38 15.8 18.2 60.2 38.1 1.7 A-1-a (0) GW N/A

19.6 1.7 2.41 11.5 20.4 45.8 51.3 2.9 A-1-a (0) GW N/A

Note: Cu: coefficient of uniformity, N/A: not available, Cc: coefficient of curvature, Gs: specific gravity, Wopt: optimum water content, cdmax: maximum dry unit weight, AASHTO: American Association of State Highway and Transportation Officials, A-2-4 (0): silty sand material, A-1-a (0): high quality highway base layer material, USCS: Unified Soil Classification System, ML: low plasticity silt, GW: well-graded gravel, GP: poorly-graded gravel, Off-Spec.: off-specification, ASTM: American Society of Testing and Materials. MW: Mine waste, FA: fly ash, RCA: recycled concrete aggregate.

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

297

Fig. 2. Diagram of column leach testing apparatus for (a) the first series of tests (control) with deionized water used as the influent solution and (b) the second series of tests (experimental) with AMD generated from mine waste used as the influent solution.

mine waste rock (less than 3/8 in.) at its optimum moisture content (14%) into PVC pipes (4-in.  4-in.) in 3 layers with 25 blows per layer. As shown in Fig. 2(a), the testing apparatus for the first series of CLTs consisted of one compacted column of RCA or fly ash, two porous stones, two filters (Whatman #4 Qualitative Filter), three o-rings sealed with high vacuum grease, and one reservoir of deionized water. A peristaltic pump (Cole-Parmer Instrument Company model number 7553-80 fitted with a Masterflex Speed Controller and Masterflex Easy-Load II model 77202-50) was used to pump deionized water from the reservoir through tubing (Masterflex Precision Pump Tygon Tubing) and into the column.

The pumping rate was set at 60 ml/h to simulate a typical field velocity of 3  10 3–2  10 2 m/d [20,21]. The testing apparatus for the second series of CLTs is shown in Fig. 2(b). This apparatus consisted of one mine waste column and one column of remediation material (one of the four RCAs or three fly ashes), four porous stones, four filters (Whatman #4 Qualitative Filter), six o-rings sealed with high vacuum grease, one reservoir of deionized water, and one sampling valve. A peristaltic pump (ColeParmer Instrument Company model number 7553-80 fitted with a Masterflex Speed Controller and Masterflex Easy-Load II model 77202-50) was used to pump deionized water from the reservoir through tubing (Masterflex Precision Pump Tygon Tubing) and into

298

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

the mine waste column. The pumping rate was also set to 60 ml/h. The effluent AMD generated from the mine waste column was pumped through the RCA or fly ash column for treatment. 3.3. Sampling column leach tests Systematically during the CLTs, aqueous samples were collected from the testing columns using 150 ml acid-washed beakers. In the first series of CLTs, effluent leachate samples from the fly ash or mine waste columns were collected from the sampling port on the top of the columns. In the second series of CLTs, samples of untreated AMD (before it was pumped through a RCA or fly ash column), and treated AMD (after it passed through a fly ash or RCA column) were simultaneously collected for analysis to determine the real-time impact of the treatment columns (columns containing RCA or fly ash) on AMD chemistry. The untreated AMD samples were collected from the sampling valve fixed to the tube stretching from the top of the mine waste column to the bottom of the treatment column. The treated AMD samples were collected from the sampling port on the top of the treatment column. During the first three days of sampling, samples were collected six to eight times per day to account for fluctuations in metal concentrations and chemical properties of the effluent leachate. Over time, less frequent sampling was required as metal concentrations and pH equilibrated. From day three to five, samples were collected four times per day, and after the fifth day sample collection was reduced to twice daily for the duration of the tests. Once collected, samples were prepared for analysis following the procedure outlined in Cetin et al. [22]. The samples were pressure filtered through 0.2 lm membranes fixed in 25 mm Easy Pressure syringe filter holders using 60 ml plastic syringes. The filtered samples were poured into acid-washed polypropylene centrifuge tubes for analysis. One, 50 ml centrifuge tube was filled for metal analysis and two, 15 ml centrifuge tubes were filled for alkalinity analysis. The pH, oxidation reduction potential, and electrical conductivity of the 50 ml samples were immediately measured. The 50 ml samples intended for metal analysis were preserved by acidifying to a pH less than 2 using a 10% solution of trace metal grade HNO3. The samples for alkalinity and sulfate analysis were not acidified. All samples were refrigerated at less than 4 °C. 3.4. Analytical methods for aqueous samples A Perkin-Elmer AA Analyst 100 atomic absorption spectrometer (AA) was used to measure the total metal ion concentrations of Ca, Cr, Cu, Fe, Mg, Mn, and Zn in aqueous samples from the CLTs. Minimum detection limits (MDLs) for AA were determined for each element. The MDLs for Ca, Cr, Cu, Fe, Mg, Mn, and Zn were 2.5 mg/L, 2 lg/L, 3 lg/L, 4 lg/L, 8 lg/L, 11 lg/L and 10 lg/L, respectively. Ca and Mg were analyzed because dissolution of Ca and Mg bearing minerals in concrete are known to increase alkalinity and buffer pH [19]. Cr, Cu, Mn, Fe, and Zn were selected for analysis because they are in high volumes by weight in the testing materials and the U.S. Environmental Protection Agency (EPA) has identified them as contaminants that pose significant health risks [23]. Maximum contaminant levels established by the EPA limit the legal concentrations of these elements in primary and secondary drinking water. The pH and oxidation reduction potential (ORP) of samples collected from the CLTs were measured with an Orion Star LogR Meter, and electrical conductivity (Eh) was determined with an Oakton Bench 700 m. The alkalinity of samples was measured using a Hach Alkalinity Test Kit (Model AL-AP). The kit was used to conduct high range alkalinity tests (20–400 mg/L) and low range alkalinity tests (5–100 mg/L) per the manufacturer’s instructions. Initially the high range test was conducted on samples. However,

if alkalinity was found to be below the testing range, the low range test was conducted. Both tests use phenolphthalein indicator powder pillows and a sulfuric acid standard solution to determine the mg/L of phenolphthalein alkalinity as calcium carbonate (CaCO3) in the sample, and bromcresol green-methyl red indicator powder pillows and the same sulfuric acid standard solution to determine the mg/L of methyl orange alkalinity as CaCO3 in the sample. SO24 anion concentrations in the effluent solutions were measured in accordance with U.S. EPA-600/4-79-020, 375.3 method. SO24 was precipitated as BaSO4 in hydrochloric acid medium and was measured gravimetrically. 3.5. Analytical methods for material samples After the completion of the column leach tests, samples of column material were collected for scanning electron microscope (SEM) imaging, and X-ray fluorescence (XRF) and inductively coupled plasma-mass spectrometry (ICP-MS). Three samples were collected total from each column to observe and quantify changes in metal content due to leaching. The three samples were collected from 2 in. below the top of the column, 2 in. above the bottom of the column, and from the middle of the column. All samples were air-dried and placed on carbon tape for analysis with a Zeiss field emission SEM oriented with tungsten (W) crystal coated zirconium oxide (ZrO). SEM imaging was conducted under variable pressure mode and high beam current with a 60 lm aperture. The working distance was optimized to 8.5 mm. 4. Results and discussion 4.1. pH, oxidation-reduction potential (Eh) Analysis of untreated AMD during the CLTs aligns with the findings of past studies [13,18,24] and suggests that pH and oxidationreduction potential (Eh) of AMD are directly related to dissolution of sulfide bearing minerals in mine waste. Fig. 3 shows that the pH of aqueous samples of untreated AMD increased (from pH 2 to 4) with increasing pore volumes (PVs) of flow during the CLTs. In addition, as shown in Fig. 4, Eh declined (from 298 mV to 198 mV) at a rate inversely proportional to the observed increase in pH. Exhaustion of sulfide minerals in the mine waste was demonstrated by an asymptotical decline in sulfate concentrations. During the first 15 PVs of flow, sulfate concentrations in untreated AMD asymptotically declined from an initial peak of 5500 mg/L to below the detection limit (Fig. 5). Therefore, it can be inferred that the increase in pH and subsequent decline in Eh are likely attributed to exhaustion of sulfide bearing minerals in the mine waste as suggested by past studies and demonstrated in Figs. 3–5. RCA treatment proved to be effective in neutralizing the acidity of AMD. Fig. 3 shows that the pH of AMD significantly increases after passing though compacted columns of the RCA materials [23]. The pH of AMD prior to treatment ranged from 2 to 4, and after treatment with the RCAs the pH of the AMD increased to above 11. Past studies have attributed similar increases in pH of AMD following RCA treatment to dissolution of cementitious components of RCAs, such as available lime (CaO) and calcium carbonate (CaCO3) [9,18]. XRF analysis of specimen samples of the RCAs used in this study is summarized in Table 1. The XRF results coincide with the findings of past studies [9,18] and identify CaO as a primary alkaline component (37.9–55.8% of total composition by weight) of the RCAs. MgO was also identified as additional alkali contributing oxides in the studied RCAs (Table 1). The studied RCAs can be sorted into the following order based on their CaO content RCA4 > RCA3 > RCA2 > RCA1. The CaO content of the RCA materials is indicative of their pH buffering capacities. Fig. 3 shows that AMD

299

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

300

14

(a) AMD Treated with RCAs

(a) AMD Treated with RCAs 200

12

8

Eh (mV)

pH

100

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

10

6

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

0 -100 -200

4

-300 -400

2 0

5

10

15

0

20

5

10

15

20

Number of Pore Volumes of Flow

Number of Pore Volumes of Flow 300

12

(b) AMD Treated with fly ashes

(b) AMD Treated with fly ashes 200

10

pH

8

100 Eh (mV)

AMD AMD-FA1 AMD-FA2 AMD-FA3

AMD AMD-FA1 AMD-FA2 AMD-FA3

0

6

-100

4

-200

2 0

5

10

15

20

Number of Pore Volumes of Flow Fig. 3. Effluent pH of acid mine drainage (AMD), (a) AMD treated with RCAs and (b) AMD treated with fly ashes.

treated with RCA1 yielded slightly less alkaline pH (pH < 12) than the ones treated with RCA2, RCA3, and RCA4 (pH > 12). These results suggest that the pH of AMD is directly impacted by CaO content of RCA materials during treatment. The Eh of AMD was transformed from oxidizing state (positive) to reducing state (negative) after passing through the RCA columns, as shown in Fig. 4. This indicates that RCA induces a net reduction of oxidized metal species in AMD [25]. Fig. 5 shows that sulfate concentrations in AMD treated with the RCAs were higher than the untreated AMD following the first few PVs. Sulfate concentrations in AMD treated with the RCAs stabilized around 450 mg/L following the first few PVs, while sulfate concentrations in untreated AMD asymptotically declined to below the detection limit. This indicates that RCA starts contributing on constant leaching of sulfate due to the presence of cement in the RCA matrix.

0

5

10

15

20

Number of Pore Volumes of Flow Fig. 4. Effluent oxidation reduction potential (Eh) of acid mine drainage (AMD), (a) AMD treated with RCAs (a) and (b) AMD treated with fly ashes.

Fly ash treatment did not seem to be effective as RCA materials except the FA3 fly ash in neutralizing the acidity of AMD by decreasing sulfate concentrations, and Eh. Fig. 3 shows that the pH of AMD was effectively increased by the FA3 fly ash. The treatment the pH of AMD with FA1 and FA2 fly ashes remained acidic. The pH of AMD was most significantly buffered by FA3 fly ash. Table 1 shows that FA3 has higher (CaO) content (12.5% by weight) than the FA1 and FA2 fly ashes (1.07 and 0.7% by weight, respectively), which past studies suggests may make it a more effective buffering material due to increased dissolution of alkaline components [26,27]. Other studies [7,28] observed similar increases in pH of AMD following fly ash treatment. They attributed these increases to the formation of hydroxyl ions by dissolution of lime and calcium hydroxide in the fly ash material which were not observed with FA1 and FA2 fly ashes due to lack of lime content. In addition to increasing pH, Fig. 4 shows that only FA3 fly ash treatment significantly lowered Eh of AMD to the reducing state.

300

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

6000

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

5000

4000 Sulfate (mg/L)

concentrations were below the detection limit in the treated and untreated AMD. This indicates that both mine waste and fly ashes contribute on sulfate leaching but it dissipates over time.

(a) AMD Treated with RCAs

4.2. Metal concentrations Atomic absorption (AA) and electrical conductivity analysis identified a strong correlation between metal concentrations in AMD and electrical conductivity. Fig. 6 shows that the electrical conductivity of AMD more than doubled after passing through the RCA columns. Considering that electrical conductivity is approximately related to the quantity of total dissolved solids in solution, it is most evident that metal ions were released from the four RCAs and three fly ashes during AMD treatment [9,25]. It is speculated that the increase in electrical conductivity of AMD following RCA treatment was not likely a result of dissolution of cationic metals (such as Cr, Se, Co, Pb, As, Ni, and Cu). This is

3000

2000

1000

0 0

7000

5 10 15 Number of Pore Volumes of Flow

(b) AMD Treated with fly ashes

6000

20

AMD AMD-FA1 AMD-FA2 AMD-FA3

4000

(a) AMD Treated with RCAs AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

15

EC (mS)

5000 Sulfate (mg/L)

20

10

3000 5 2000 1000

0 0

5

10

15

20

Number of Pore Volumes of Flow

0 0

5

10

15

20 20

Number of Pore Volumes of Flow

(b) AMD Treated with fly ashes

Fig. 5. Leached sulfate concentrations in acid mine drainage (AMD), (a) AMD treated with RCAs and (b) AMD treated with fly ashes.

AMD AMD-FA1 AMD-FA2 AMD-FA3

EC (mS)

15 Following treatment with FA3, the average Eh of AMD was transformed from oxidizing (214.6 mV) to reducing ( 155.3 mV). FA2 and FA3 also lowered Eh of AMD; however, not significantly enough to establish reducing conditions. The average Eh of the AMD solutions after permeating through the FA1 and FA2 columns were 165.10 and 118.4 mV, respectively. The control group of CLTs (conducted with deionized water and the three fly ashes) followed suit, and FA3 yielded column effluent that was over one order of magnitude more oxidizing than FA1 and FA2. According to the findings of Lee et al. [29], Eh is dependent upon the reactive oxides in treatment materials. Therefore, it can be inferred that FA1 and FA2 were less effective than FA3 in inducing reducing conditions during treatment because they contain less than half the oxidizing agent Fe2O3 (3.2 and 5.5% by weight, respectively) of FA3 (13.7% by weight), as shown in Table 1. Fig. 5 shows that during the first 5 PVs sulfate concentrations in AMD treated with the three fly ashes asymptotically declined (from over 3000 mg/L to below 1000 mg/L). By the 15th PV, sulfate

10

5

0 0

5 10 15 Number of Pore Volumes of Flow

20

Fig. 6. Electrical conductivity of acid mine drainage (AMD), (a) AMD treated with RCAs and (b) AMD treated with fly ashes.

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

350

(a) Calcium

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

300

Ca (mg/L)

250

200

150

100

50

0 0

5

10

15

20

Number of Pore Volumes of Flow 70

(b) Magnesium 60

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

Mg (mg/L)

50

40

30

20

10

0 0

5 10 15 Number of Pore Volumes of Flow

20

Fig. 7. Concentrations of (a) Ca and (b) Mg in acid mine drainage (AMD) and AMD treated with RCAs.

because cationic metals tend to be more soluble and mobile as oxyanions in their oxidized state, and the pH values of AMD treated by the RCAs were very alkaline (pH > 11) and exhibited reduced states (Eh < 200 mV) [30]. Fig. 7 shows that Ca and Mg concentrations in the AMD treated with the RCAs was significantly higher than those of the untreated AMD. This suggests that the dissolution of alkaline components of the RCAs, such as Ca and Mg bearing minerals, is primarily responsible for the observed increase in electrical conductivity of the AMD. RCA treatment is proved to be effective in decreasing Cr, Fe, Cu, Mn, and Zn concentrations in the AMD, as shown in Fig. 8. Following RCA treatment, Cr concentrations in the AMD decreased by 22– 62%, and Fe concentrations declined by over 98%. Overall, RCA treatment was proved to be most effective in reducing Mn, Zn, and Cu concentrations in AMD, and their aqueous concentrations dropped to below their MCLs (0.05 mg/L Mn, 5 mg/L Zn, and

301

1.3 mg/L Cu). RCA treatment reduced Mn concentrations in AMD by 80–100% to below its MCL of 0.05 mg/L. In addition, RCA treatment decreased Cu concentrations in AMD by 33–80%, and Zn concentrations by over 95%. However, the effectiveness of the RCAs in reducing aqueous Cu and Zn concentrations declined with increasing PVs of flow. After the 10th PV, Cu concentrations in AMD treated with the RCAs increased to above the MCL. Zn concentrations in the AMD treated with the RCAs increased to above the MCL after 17 PVs of flow, and the reduction in Zn concentrations afforded by the RCAs declined to 73–88%. Other studies have experienced similar declines in sorption capacity of RCAs for Zn and Cu, which they related to limited available sorption sites on solid surfaces and components of the treatment materials preferentially sorbing other metals [26]. Fig. 9 shows the SEM images of samples of RCA3 and RCA2 column material after the termination of the CLTs with AMD. The images show an abundance of ettringite (Ca6Al2(SO4)3(OH)12
26H2O) in the RCA materials that literature suggests preferentially sorbs Cr ions from solution [25,31]. By comparing XRF analysis of specimen samples from the CLTs with deionized water with those from the CLTs with AMD (Table 3) it is apparent that Cr, Cu, Mn, and Zn content in the materials increased following treatment. This confirms that the reduction in aqueous metal ion concentrations in the AMD was the result of sorption of AMD constituents to RCA particles. Further, the observed decline in sorption capacity of the RCAs Cu is likely the result of more competitive metals (e.g. Cr) preferentially being sorbed by components of the RCAs (e.g. ettringite). Of the four studied RCAs, RCA4 proved to be most effective in treating AMD. RCA3 treatment caused a 148% increase in pH of AMD from below 2 to above 12. It also transformed Eh from oxidizing (>150 mV) to reducing (< 200 mV), reduced sulfate concentrations to below the detection limit, and demonstrated the greatest release of Ca ions (216.45 mg/L) (Fig. 7a). RCA4 contains the highest CaO (55.8% by weight), and the highest unburned carbon content (32.7% by weight) of the studied RCAs that could be responsible for its effectiveness in treating the AMD. RCA1 was the least effective RCA in absorbing Cr and Cu from the AMD. This is likely a result of the relatively higher contents of Cr, Cu, and Mn in RCA1. Table 1 shows that RCA1 contains 22 mg/kg Cr, 16 mg/kg Cu, and 355 mg/kg Mn. Of the three fly ashes studied, it is apparent that FA3 is the most effective fly ash for AMD treatment. FA3 treatment increased the average electrical conductivity of AMD from 2.16 mS to 4.25 mS, while FA1 and FA2 decreased the average electrical conductivity of AMD to 1.97 mS and 1.80 mS, respectively (Fig. 6). The observed decrease in electrical conductivity of the AMD by FA1 and FA2 fly ashes can be explained by decreased metal ion mobility at nearneutral pH [32]. The pH of AMD following treatment with FA1 and FA2 was 4.32 and 5.04, respectively. Fig. 10 shows the effects of fly ash treatment on metal ion concentrations in the AMD. Following the first flush, fly ash treatment reduced average Cr and Fe concentrations in AMD by over 90%; however, they remained exceeding of the EPA’s MCLs (0.1 mg/L Cr, and 0.3 mg/L Fe). Zn followed a cationic leaching pattern and Zn concentrations decreased with increasing pH. During treatment, AMD treated with FA3 exhibited highly alkaline pH (10), and Zn concentrations in AMD were reduced by 99% to below the EPA’s MCL of 5 mg/L. Conversely, pH of AMD treated with FA1 and FA2 was acidic (pH < 6.5) and Zn concentrations in AMD increased by over 25% following treatment with FA1 and FA2. All three fly ashes released significant Mg and Mn ions into the AMD, and concentrations increased by more than 25% following treatment. Cu leaching decreased after AMD was treated with fly ashes in the AMD increasing at alkaline pH. Fly ash treatment reduced average Cu concentrations by 99%. Following treatment with fly ashes, Cu con-

302

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

16

6

(a) Chromium

14 12

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

5 4

10

Cu (mg/L)

Cr (mg/L)

(b) Copper

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

8 6

3 2

4 1

2 0

0

0

300

5 10 15 Number of Pore Volumes of Flow

20

0

10

15

20

0.4

(c) Iron

(d) Manganese

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

250

0.35 0.3

Mn (mg/L)

200

Fe (mg/L)

5

Number of Pore Volumes of Flow

150 100

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

0.25 0.2 0.15 0.1

50

0.05

0

0 0

5 10 15 Number of Pore Volumes of Flow

20

0

5

10

15

20

Number of Pore Volumes of Flow

100

(e) Zinc

Zn (mg/L)

80

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

60

40

20

0

0

5 10 15 Number of Pore Volumes of Flow

20

Fig. 8. Concentrations of (a) Cr, (b) Cu, (c) Fe, (d) Mn, and (e) Zn in acid mine drainage (AMD), and AMD treated with RCAs.

centrations in AMD fell below the MCL (1.3 mg/L). Adsorption of cationic Cu is limited by available sorption sites on solid surfaces [33]. Considering that FA1 and FA2 demonstrated significant releases of Zn ions into solution, it is evident that these materials contain adequate sorption sites for Cu ions.

Fig. 11 compares SEM images of column specimen samples of FA3 material after the CLTs with deionized water and AMD. The sample of FA3 from the CLT with AMD appears to have significantly larger and more abundant clusters of precipitates on the surface of fly ash particles than the sample of column material from the CLT

303

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

4.3. Influence of alkalinity and CaO content of materials on treatment of AMD

Fig. 9. Scanning electron microscopy images of RCA2 (a), and RCA3 (b) column material after completion of the CLTs with acid mine drainage (AMD).

with deionized water. This suggests that the precipitation and sorption of AMD constituents onto fly ash particles is most likely the cause of the decrease in electrical conductivity of the AMD solutions following permeation through the fly ash columns.

Indraratna et al. [13], Golab et al. [18], and Regmi et al. [19] identified a close relationship between Ca-containing minerals and alkaline components of concretes. They assert that dissolution of lime (CaO) and portlandite (Ca(OH)2) in the cementitious components of concretes is primarily responsible for the release of Ca and calcium carbonate (CaCO3) from concretes’ matrices [18,19]. The increase in alkalinity as CaCO3 in the AMD after passing through the RCA columns, shown in Fig. 12, coincides with the literature. Thus, the dissolution of lime and other Ca-bearing minerals in the concrete components of the RCAs are likely the cause of the increase in pH of the AMD. The results of XRF analysis of specimen samples from the CLTs with deionized water and AMD (Table 3) show a reduction in Ca and CaO content in all four RCAs as a result of AMD treatment. Following AMD treatment Ca and CaO content in the RCAs decreased by 5–9%, and by 2–13%, respectively. Fig. 13 demonstrates that Ca concentrations in the samples from the fly ash columns passed with deionized water and AMD had initially high Ca concentrations that correspond to high initial alkalinity. It also shows that in successive pore volumes of flow alkalinity and Ca concentrations declined. This illustrates that the increase in pH in later pore volumes, was a result of the consumption of Ca and other alkaline components of the fly ashes by neutralization reactions [34]. Figs. 12 and 13 show that alkalinity and Ca concentrations generally increased in the AMD solutions after permeating through the fly ash columns. FA3 showed the most significant release of the alkaline minerals Ca and Mg into AMD, and after passing through the FA3 column concentrations increased by one to two orders of magnitude. This is likely a result of the high CaO and MgO content of FA3, shown in Table 1. Other studies have observed elevated levels of Ca, Mg, K, and Na in alkaline waters as a result of dissolution of alkaline components of materials [25,35]. Furthermore, the high aqueous concentrations of Ca and Mg in the AMD treated with FA3 are likely why it was more effective in raising the pH of the AMD than the other fly ashes. The average pH of the AMD treated with FA3 was 9.8, while the average pH of the AMD treated with FA1 and FA2 were 4.3 and 5.0, respectively. In addition to impacting pH, the dissolution of alkaline components of the RCAs and fly ashes greatly influences metal ion sorption. Fig. 14 demonstrates that sorption capacity for Cr, Cu, Fe,

Table 3 Percent composition of oxides and ions, and metal content in the mine waste, RCAs, and fly ashes determined using XRF after columns leached with deionized water and acid mine drainage. Influent solution

Deionized water

Column material

MW

Acid mine drainage

FA3

RCA1

RCA2

RCA3

RCA4

FA3

RCA1

RCA2

RCA3

RCA4

Major oxides/ions (% by weight) Ca 0.08 Fe 5.4 Mg 0.19 Al2O3 16.25 CaO 0.11 Cr2O3 <0.01 Fe2O3 7.75 MgO 0.37 MnO 0.01 SiO2 54.9

4.3 5.24 0.97 25.5 1.09 0.02 7.81 1.88 0.06 56.3

8.5 1.37 0.72 0.44 12.7 <0.01 2.1 1.28 0.09 66.9

11 1.27 0.78 3.17 31 <0.01 1.82 1.26 0.05 35.1

20 1.41 0.71 2.04 32.2 <0.01 2.05 1.12 0.05 37.9

22 2.01 1.12 1.42 36.25 <0.01 3.01 2.02 0.06 50.9

3.65 5.82 0.84 20.4 9.8 0.03 8.06 1.56 0.06 56

7.95 1.5 0.52 5.1 11.2 <0.01 2.14 0.9 0.08 69

10 1.5 0.75 3.5 28 <0.01 2.05 1.1 0.04 35.1

17 1.7 0.7 3.7 30 <0.01 2.1 1 0.05 37

20 2.9 1 9.2 33 <0.01 4 2 0.07 50.6

Metal content (mg/kg) Cr Cu Mn Zn

42 36 207 84

20 9 320 35

13 5.6 344 40

15 6.3 373 27

14 20 402 59

125 93 450 391

24 12 651 57

19 6.8 366 101

17 7 439 35

19 22 511 93

43 36.1 35 38

Note: Samples from acid mine drainage leached columns were collected from the middle for XRF analyses. MW: mine waste, FA: fly ash, RCA: recycled concrete aggregate.

304

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

6

200

(a) Chromium

AMD AMD-FA1 AMD-FA2 AMD-FA3

(b) Copper

AMD AMD-FA1 AMD-FA2 AMD-FA3

5

150

Cu (mg/L)

Cr (mg/L)

4 100

3 2

50 1 0

0 0

5

10

15

0

20

300

10

15

20

60

(c) Iron

(d) Magnesium

AMD AMD-FA1 AMD-FA2 AMD-FA3

250

AMD AMD-FA1 AMD-FA2 AMD-FA3

50 40

Mg (mg/L)

200

Fe (mg/L)

5

Number of Pore Volumes of Flow

Number of Pore Volumes of Flow

150

30

100

20

50

10

0

0 0

5

10

15

20

0

Number of Pore Volumes of Flow

5

10

15

20

Number of Pore Volumes of Flow 140

6

(e) Manganese

(f) Zinc

AMD AMD-FA1 AMD-FA2 AMD-FA3

5

120

AMD AMD-FA1 AMD-FA2 AMD-FA3

100

Zn (mg/L)

Mn (mg/L)

4 3

80 60

2 40 1

20

0

0 0

5

10

15

20

Number of Pore Volumes of Flow

0

5

10

15

20

Number of Pore Volumes of Flow

Fig. 10. Concentrations of (a) Cr, (b) Cu, (c) Fe, (d) Mg, (e) Mn, and (f) Zn in acid mine drainage (AMD), and AMD treated with fly ashes.

Mn, and Zn was greatest in treatment materials with high CaO content. Figs. 9 and 11 show that during the CLTs with high CaO materials (RCA1 and FA3) precipitates coated the surface of particles which likely decreased the sorption capacities of them for other

metal ions [18]. The surface of the RCA and fly ash particles are being deprotonated with an increase in pH. The cationic species, such as Cu2+ and Mn2+, attach to negatively charged surfaces which yields a reduction in the leached Mn concentrations [36,37]. In

305

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

500

Alkalinity as CaCO3 (mg/L)

400

300

AMD AMD-RCA1 AMD-RCA2 AMD-RCA3 AMD-RCA4

200

100

(a) AMD Treated with RCAs 0 0

5 10 15 Number of Pore Volumes of Flow

20

600

(b) AMD Treated with fly ashes

AMD AMD-FA1 AMD-FA2 AMD-FA3

Alkalinity as CaCO3 (mg/L)

500

400

300

200

100 Fig. 11. Scanning electron microscopy images of FA3 column material (a) after the first series of CLTs with deionized water and (b) the second series of CLTs with acid mine drainage (AMD).

0 0

5

10

15

20

Number of Pore Volumes of Flow

4.4. Loss on ignition (unburned carbon content) Table 1 shows that the fly ashes and RCAs used in this study have high loss on ignition (LOI) values, which indicates that they contain significant volumes of unburned carbon. The fly ashes contain between 6.2 and 8.1% unburned carbon, with the greatest percent in FA3. High unburned carbon content is typical of fly ashes generated in the United States during the last decade due to the

Fig. 12. Alkalinity as CaCO3 in acid mine drainage (AMD), (a) AMD treated with RCAs and (b) AMD treated with fly ashes.

350 AMD AMD-FA1 AMD-FA2 AMD-FA3

300

250 Ca (mg/L)

addition, these metals may precipitate by precipitating with Ca [33] in the aqueous phase. Therefore, release of Ca from CaO in the treatment materials pose great importance for remediation of AMD. Overall it can be claimed that, CaO content of treatment materials is a key mechanism driving neutralization reactions in AMD. The four RCAs (RCA1, RCA2, RCA3, and RCA4) which contain the most CaO of the studied treatment materials (27–58% by weight) were the most effective in increasing the pH of AMD. After passing through the RCA and FA3 columns, the pH of AMD increased to pH > 10. Other studies observed similar pH buffering of AMD and acid sulfate soil when treated with materials composed of 25– 70% by weight CaO [9,18]. This suggests that in addition to RCA and fly ash, other materials containing more than 25% CaO content may be effective in neutralizing the pH of AMD.

200

150

100

50

0 0

5

10

15

20

Number of Pore Volumes of Flow Fig. 13. Concentrations of Ca in acid mine drainage (AMD) and AMD treated with fly ashes.

306

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

100

100

(a) Chromium

(b) Copper 80 % of Cu Sorbed

% of Cr Sorbed

80

60

40

60

40

20

20

0

0

0 10 20 30 40 50 60 CaO Content of Treatment Material (% by weight)

?

0 10 20 30 40 50 60 CaO Content of Treatment Material (% by weight)

100

100 MT fly ash 80

% of Mn Sorbed

% of Fe Sorbed

80

60

40

20

60

40

20

(d) Manganese

(c) Iron 0

0 10 20 30 40 50 60 CaO Content of Treatment Material (% by weight)

0

0 10 20 30 40 50 60 CaO Content of Treatment Material (% by weight)

100

% of Zn Sorbed

80

(e) Zinc

60

40

20

0

0 10 20 30 40 50 60 CaO Content of Treatment Material (% by weight)

Fig. 14. Influence of calcium oxide content of treatment materials (RCAs and fly ashes) on adsorption of Cr (a), Cu (b), Fe (c), Mn (d), and Zn (e) from acid mine drainage (AMD).

increasingly common use of low nitrogen oxide and sulfur oxide burners. Unfortunately, high carbon fly ashes cannot be used in concrete production because the carbon absorbs metal ions in admixtures used to prevent crack formation [22]. The RCAs used

in this study contain between 7.34 and 32.7% unburned carbon, with the greatest percent in RCA4. Fig. 15 shows the effects of unburned carbon content of the RCAs and fly ashes on sorption of Cr, Cu, Fe, Mn, and Zn from

307

100

100

80

80 % of Cu Sorbed

% of Cr Sorbed

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

60

40

20

60

40

20

(a) Chromium 5

10

15 20 25 30 LOI Treatment Material (%)

0

35

100

100

80

80

% of Mn Sorbed

% of Fe Sorbed

0

(b) Copper

60

40

5

10

15 20 25 30 LOI Treatment Material (%)

60

40

20

20

(c) Iron 0

35

5

10

(d) Manganese

15 20 25 30 LOI Treatment Material (%)

35

0

5

10

15 20 25 30 LOI Treatment Material (%)

35

100

(e) Zinc

% of Zn Sorbed

80

60

40

20

0

5

10

15 20 25 30 LOI Treatment Material (%)

35

Fig. 15. Influence of unburned carbon content of treatment materials (RCAs and fly ashes) on adsorption of Cr (a), Cu (b), Fe (c), Mn (d), and Zn (e) from acid mine drainage (AMD).

AMD. With increasing unburned carbon contents, the sorption capacity of the treatment materials for all metals were increased. The increased sorption capacity for Cr, Cu, Fe, and

Zn could be the result of the higher specific surface area of unburned carbon than the mineral fraction of the treatment materials. According to Wang et al. [17], carbon has a high

308

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309

specific surface area, which could increase the metal sorption capacity of materials. 5. Conclusions This study provides laboratory analysis of the low-cost construction waste products, fly ash and recycled concrete aggregate (RCA), as remediation materials for AMD treatment. Column leach tests (CLTs) were conducted to determine the effects of the remediation materials on pH, electrical conductivity, alkalinity, oxidation reduction potential; and concentrations of sulfate, and selected metals in AMD. The impact of oxide, alkalinity, and unburned carbon content on retention of metal ions in the remediation materials was quantified using X-ray fluorescence spectroscopy. Furthermore, the effects of treatment on the physical properties of the materials were observed using scanning electron microscope imaging. The following conclusions were made: (1) RCA and fly ash treatment proved effective in neutralizing the acidity of AMD by reducing aqueous sulfate concentrations, and oxidation reduction potential; and by raising pH, alkalinity, and calcium concentrations. FA3 fly ash and the four RCAs (RCA1, RCA2, RCA3, and RCA4) were most effective in increasing the pH of AMD because of their high alkalinity and Ca content. After passing through the RCA and FA3 columns, the pH of AMD increased to pH > 9. (2) As a result of the dissolution of alkaline components of the RCAs and FA3 fly ash for neutralization reactions, the electrical conductivity of AMD increased after passing through the RCA and FA3 fly ash columns by nearly one order of magnitude. Permeation through the FA1 and FA2 fly ash columns decreased the electrical conductivity of AMD as metal ion mobility was decreased at near-neutral pH. (3) Eh and sulfate concentrations of the AMD decreased significantly after passing through RCA columns. However, such trends were not observed for the fly ash treated AMD tests. Results showed that fly ashes contribute to leached sulfate concentrations from AMD due to high sulfate content of the materials themselves. (4) Results of column leach tests showed that metal concentrations leached from AMD decreased significantly increased due to sorption of these constituents to the surface of the RCA and fly ash particles. Overall, it was observed that during the RCA treatment, the concentrations of Cr, Cu, Fe Mg, Mn and Zn in the AMD were greatly reduced. On the other hand, fly ash treatment was not as effective but it also reduced concentrations of Cr and Fe in the AMD; however, their sorption capacities were limited for Cu, Mn, and Zn sorption since these fly ashes also release considerable amount of these metals to the aqueous phase. (5) It was observed that sorption capacities of materials were directly correlated the CaO content and LOI content of the materials. Results of this study indicated that RCA materials were more efficient in terms of remediating the AMD compared to the fly ashes which could be related to the relatively higher CaO and LOI content of RCAs compared to the fly ashes tested in the current study. (6) Although RCA may be more effective as a treatment material in neutralizing pH and reducing metal concentrations in AMD to the trace metal (parts per billion) level, there is significant potential in the use of fly ash for less stringent applications where cleanup levels are in the parts per million range. While future studies may consider RCA treatment as a standalone technology for AMD remediation, fly ash treatment may also be incorporated into a multi-step treatment process as a

low-cost, initial treatment step. All three fly ashes increased the pH of AMD by at least one standard unit of pH and promoted sorption of AMD constituents to their matrices. (7) Overall, RCAs and high-alkalinity fly ash, such as FA3 fly ash, were found to be most effective in raising pH and reducing concentrations of Cr, Cu, Fe, Mn, and Zn in AMD. The results of this study are intended to provide comparisons between the effectiveness of recycled concrete aggregate and fly ash as remediation materials for treatment of acid mine drainage. The conclusions of this study can be used as foundations for long-term, field scale studies that investigate the use of RCA and fly ash in treating AMD. The long-term effectiveness and feasibility of using RCAs and fly ashes in remediating AMD in in situ conditions could be assessed by conducting field scale applications. Acknowledgements This study was financially supported by the South Dakota Board of Regents (SDBOR) and RESPEC Consulting and Services Inc (RESPEC). Endorsement by SDBOR and RESPEC, and or the RCA and fly ash suppliers is not implied and should not be assumed. References [1] Johnson DB, Hallberg KB. Acid mine drainage remediation options: a review. Sci Total Environ 2005;338:3–14. [2] Luìs AT, Teixeira P, Almeida FP, Ector L, Matos JX, Ferreira da Silva EA. Impact of acid mine drainage (AMD) on water quality, stream sediments, and periphytic diatom communities in the surrounding streams of Aljustrel mining area (Portugal). Water Air Soil Pollut 2009;200:147–67. [3] Romero A, Gonzàlez I, Galàn E. Stream water geochemistry from mine wastes in Peña de Hierro, Riotinto area, SW Spain: a case of extreme acid mine drainage. Environ Earth Sci 2011;62:645–56. [4] Komnitsas K, Bartzas G, Paspaliaris I. Clean up of acidic leachates using fly ash barriers: laboratory column studies. Global Nest J 2004;6(1):81–9. [5] Yeheyis M, Shang J, Yanful E. Feasibility of using coal fly ash for mine waste contamination. J Environ Eng 2010;136(7):8. doi: http://dx.doi.org/10.1061/ (ASCE)EE.1943-7870.0000211. [6] ACAA American coal ash association-production and usage brochure. In: Association (Ed.); 2013. [7] Shang J, Wang H, Kovac V, Fyfe J. Site-specific study on stabilization of acidgenerating mine tailings using coal fly ash. J Mater Civ Eng 2006;18(2):11. doi: http://dx.doi.org/10.1061/(ASCE)0899-1561(2006)18:2(140). [8] Mohamed A, Hossein M, Hassani F. Evaluation of newly developed aluminum, lime and fly ash technology for solidification/stabilization of mine tailings. J Mater Civ Eng 2007;19(1):6. doi: http://dx.doi.org/10.1061/(ASCE)0899-1561 (2007)19:1(105). [9] Siriwardane H, Kannan K, Ziemkiewicz P. Use of waste materials for control of acid mine drainage and subsidence. J Environ Eng 2003;129(10):5. doi: http:// dx.doi.org/10.1061/(ASCE)0733-9372(2003)129:10(910). [10] Bulusu S, Aydilek A, Petzrick P, Guynn R. Remediation of abandoned mines using coal combustion by-products. J Geotech Geoenviron Eng 2005;131 (8):11. doi: http://dx.doi.org/10.1061/(ASCE)1090-0241(2005)131:8(958). [11] USGS. Recycled aggregates, profitable resource conservation. In: Survey (Ed.), (Vol. Fact Sheet FS-181-99). Denver; 2000. [12] Deal TA. What it costs to recycle concrete: C&D Debris Recycling; 1997. p. 10– 13. [13] Indraratna B, Regmi G, Nghiem L, Golab A. Performance of a PRB for the remediation of acidic groundwater in acid sulfate soil terrain. J Geotech Geoenviron Eng 2010;135(7):9. [14] Bestgen J, Cetin B, Tanyu BF. Effects of extraction methods and factors leaching of metals from recycled concrete aggregates. Environ Sci Pollut Res 2016. doi: http://dx.doi.org/10.1007/s11356-016-6456-0 [published online]. [15] Chen J, Bradshaw S, Benson C, Tinjum J, Edil T. PH-dependent leaching of trace elements from recycled concrete aggregate. GeoCongress 2012;9. [16] Homestake. Homestake Mining Company. Retrieved September 30, 2014, from http://www.homestakevisitorcenter.com/; 2014. [17] Wang J, Teng X, Wang H, Ban H. Characterizing the metal adsorption capability of a Class F coal fly ash. Environ Sci Technol 2004;38(24):5. [18] Golab A, Peterson MA, Indraratna B. Selection of potential reactive materials for a permeable reactive barrier for remediating acidic groundwater in acid sulphate soil terrains. Quart J Eng Geol Hydrogeol 2006;39:209–23. [19] Regmi G, Indraratna B, Nghiem LD, Prasad BG. Treatment of acid groundwater in acid sulphate soil terrain using recycled concrete: column experiments. J Environ Eng 2011;137(6):433–43. [20] Gelhar LW, Welty C, Rehfeldt KR. Critical review of data on field-scale

S.N. Jones, B. Cetin / Fuel 188 (2017) 294–309 dispersion in aquifers. Water Resour Res 1992;28(7):1955–74. [21] Morar DL, Aydilek AH, Seagren EA, Demirkan MM. Leaching of metals from fly ash-amended permeable reactive barriers. J Environ Eng 2012;138:815–25. [22] Cetin B, Aydilek AH, Guney Y. Leaching of trace metals from high carbon fly ash stabilized highway base layers. Resour Conserv Recycl 2012;58:8–17. [23] EPA. Water drinking limits, ; 2009 [Last accessed on July 5, 2016]. [24] Madzivire G, Gitari W, Vadapalli V, Ojumu T, Petrik L. Fate of sulfate removed during the treatment of circumneutral mine water and acid mine drainage with coal fly ash: modeling and experimental approach. Miner Eng 2011;24:10. doi: http://dx.doi.org/10.1016/j.mimeng.2011.07.009. [25] Lottermoser B. Mine wastes: characterization, treatment and environmental impacts. 3 ed. Berlin Heidelberg: Springer-Verlag; 2010. [26] Hèquet VPR, Leceyer I, Le Cloirec P. Removal of Cu2+ and Zn2+ in aqueous solutions by sorption onto fly ash and fly ash mixtures. Paper presented at the International Ash Utilization Symposiutm, Center for Applied Energy Research, University of Kentucky; 1999. [27] Theis TL, Wirth JL. Sorptive behavior of trace metals on fly ash in aqueous systems. Environ Sci Technol 1977;11(12):1096–100. [28] Vadapalli V, Klink M, Etchebers O, Petrik L, Gitari W, White R, Key D, Iwuoha E. Neutralization of acid mine drainage using fly ash, and strength development of the resulting solid residues. S Afr J Sci 2008;104:5. [29] Lee J, Youm S, Choi S, Oh B. Removal of mixed contaminants by Fe-based biobarrier in flow-through columns using recycled waste materials. J Mater

[30]

[31] [32]

[33]

[34]

[35]

[36] [37]

309

Cycles Waste Manage 2009;11:7. doi: http://dx.doi.org/10.1007/s10163-0090253-5. Violante A, Cozzolino V, Perelomov L, Caporale AG, Pigna M. Mobility and bioavailability of heavy metals and metalloids in soil environments. J Soil Sci Plant Nutr 2010;10(3):268–92. Prasad B, Mortimer RJG. Treatment of acid mine drainage using fly ash zeolite. Water Air Soil Pollut 2011;218:667–79. Engelsen C, van der Sloot H, Wibetoe G, Justnes H. Leaching characteristics and geochemical modeling of minor and trace elements released from recycled concrete aggregates. Cem Concr Res 2010;40:10. Komonweeraket K, Cetin B, Aydilek AH, Benson CH, Edil TB. Effects of pH on the leaching mechanisms of elements from fly ash mixed soils. Fuel 2015;140:788–802. Jurjovec J, Ptacek CJ, Blowes DW. Acid neutralization mechanisms and metal release in mine tailings: a laboratory column experiment. Geochim Cosmochim Acta 2002;66(9):1511–23. Wang J, Wang T, Burken J, Chusuei C, Ban H, Ladwig K, Huang C. Adsorption of arsenic (V) onto fly ash: a speciation-based approach. Chemosphere 2008;72:7. Su T, Shi H, Wang J. Impact of trona-based SO2 control on the elemental leaching behavior of fly ash. Energy Fuels 2011;25:3514–21. Cetin B, Aydilek AH, Guney Y. Stabilization of recycled base material using high carbon fly ash. Resour Conserv Recycl 2010;54:879–92.