Environ. Sci. Technol. 2010, 44, 2119–2125
Minimising Alkalinity and pH Spikes from Portland Cement-Bound Bauxsol (Seawater-Neutralized Red Mud) Pellets for pH Circum-Neutral Waters L A U R E M . D E S P L A N D , * ,†,‡,§ MALCOLM W. CLARK,† MICHEL ARAGNO,‡ AND TONY VANCOV§ School of Environmental Science and Management, Southern Cross University, P.O. Box 157, Lismore, NSW 2480, Australia, Microbiology Laboratory, University of Neuchaˆtel, Rue Emile-Argand 11, Case postale 158, CH-2009 Neuchatel, Switzerland, and Industry and Investment, 1243 Bruxner Highway, Wollongbar NSW 2477, Australia
Received October 22, 2009. Revised manuscript received January 26, 2010. Accepted January 27, 2010.
Bauxsol reagents (powder, slurry, or pellet forms) are powerful tools in environmental remediation and water and sewage treatment. However, when used in circum-neutral water treatments, cement-bound Bauxsol pellets produce a sustained pH and alkalinity spike due to the presence of unreacted CaO in the cement binder. This study developed a pellet treatment system to minimize the alkalinity/pH spike. The recipe for pelletization consisted of Bauxsol powder, ordinary Portland cement (OPC), hydrophilic fumed silica, aluminum powder, a viscosity modifier, and water. Several batches (including different ratios and sizes) were run using modified makeup waters (H2O + CO2 or NaHCO3) or curing brines (CO2, NaHCO3, or Mg/ CaCl2). Alkalinity, pH stability, and slake durability tests were performed on pellets before and/or after curing. The best result for reducing the alkalinity/pH spike was obtained from a MgCl2, CaCl2 bath treatment using a Bauxsol:cement ratio of 2.8:1 (pH 8.28; alkalinity 75.1 mg/L) for a 100 g batch or 2.45:1 (pH 8.05; alkalinity 35.4 mg/L) for a 1 kg batch. Although brine curing does provide a control on pH/alkalinity release, the pellets may still contain unreacted CaO. Therefore, a freshwater rinse of pellets before treating circum-neutral waters is recommended as is the continued investigation of alternative pellet binders.
Introduction Bauxsol is a chemico-physically modified product derived from bauxite refinery residues, a waste of alumina refining using the Bayer process, and is different to the red mud it is derived from. Modification results in a pH decrease from ∼13 to ∼9 and conversion of soluble alkalinity to solid alkalinity. Bauxsol is a complex mix of minerals with major constituents, including hematite, boehmite, gibbsite, sodalite, quartz, cancrinite, and whewellite, while minor components * Corresponding author phone: +612 6620 3650; fax: +612 6621 2669; e-mail:
[email protected]. † Southern Cross University. ‡ Microbiology Laboratory, University of Neuchaˆtel, Switzerland. § Industry and Investment NSW, Australia. 10.1021/es9032138
2010 American Chemical Society
Published on Web 02/18/2010
include aragonite, brucite, calcite, diaspore, ferrihydrite, anhydrite/gypsum, hydrocalumite, hydrotalcite, p-aluminohydrocalcite, and a few low solubility trace minerals (1, 2). However, the exact composition and geochemical character of a Bauxsol reagent depends on the origin of the bauxite, operational procedures used in the alumina refinery, and the concentration and balance of calcium and magnesium brine used for the red mud conversion (3). Despite compositional variability, raw Bauxsol reagents are extremely fine-grained, typically 90% are less than 10 µm in diameter and have very high surface/volume ratios (up to and greater than 100 m2/g). Bauxsol reagents also have a high metal binding capacity (>1500 meq/kg), moderate acid neutralizing capacity (from 4-7 mols H+/kg at pH 7, to approximately 14 mols H+/kg at pH 5; dependent on time and pH), are largely insoluble, and are highly nondispersive (1, 3-5). Bauxsol may undergo further physical and geochemical modification such as acid treatments, blending with additional mineral components, and pelletization (1, 6, 7). Bauxsol-based technologies (powder, slurry, or pellets) have been successfully applied to environmental remediation and water and sewage treatment. McConchie et al. (1) proved the efficiency of Bauxsol powder to remove metal and neutralize pH from acidic mine effluent and from acid sulfate soils. Fuhrman (3) demonstrated the removal of arsenate from water using Bauxsol powder mixed with sand. Akhurst et al. (2) and McConchie et al. (1) showed that Bauxsol powder or slurry strongly adsorbs phosphate from aqueous solutions by a ligand exchange mechanism. Clark et al. (8, 9) developed porous cement-bound Bauxsol pellets (using hydrogen peroxide as a foaming agent) to overcome low hydraulic conductivity encountered when using Bauxsol powder as a filter device for mild acid mine drainage treatment. Lapointe et al. (4) confirmed the efficiency of these pellets in permeable reactive barriers to treat acid rock drainage. However, ordinary Portland cement (OPC) used as a binder in Bauxsol pellets often produces a sharp increase in water pH and alkalinity, especially when used in circumneutral water treatments (Virotec International, personal communication, 2006). The pH and alkalinity loss from red mud has been modeled by Khaitan et al. (10) and in general terms may be applied to Bauxsol reagents. However, the specificities of the model do not apply because of changes in mineralogy during Bauxsol formation and because no tricalcium aluminate has been observed in the red mud source material. Moreover, the pelletization process adds excess CaO from the cement binder, which directly leads to the spike in pH and alkalinity (Dr. Malcom Clark, personal communication, 2008). Therefore, the successful suppression of the pH and alkalinity spike from OPC-bound materials has a broader application in concrete manufacture, especially where concretes are to be placed in pH or alkaline sensitive environments. The aim of this paper is to investigate techniques to control pH/alkalinity spikes from Portland cement-bound Bauxsol pellets without affecting the performance of the pellet. Specifically, the objectives are to develop a general recipe to make Bauxsol pellets using OPC and to determine the efficiency of alkalinity conversion using modified makeup waters (H2O + CO2 or NaHCO3) during the production of the pellets or using curing baths (CO2, NaHCO3, or Mg/CaCl2) postpellet production. The intended purpose of these pellets is deployment in the polishing phase of wastewater treatment. The porous Bauxsol pellets should strongly bond phosphates as well as improve hydraulic conductivity and provide a support matrix for existing natural microbial communities. VOL. 44, NO. 6, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2119
2120
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 6, 2010
a Water saturated with carbon dioxide (CO2). b Water saturated with 1 M NaHCO3. c Thirty minutes at 130 °C and at 210 kPa. d Milli-Q water. e Total of 141 g/L of magnesium chloride hexahydrate (MgCl2 · 6H2O) and 42 g/L of calcium chloride dihydrate (CaCl2 · 2H2O). f Total of 106 g/L of sodium bicarbonate (NaHCO3). g Water bath saturated with carbon dioxide (CO2).
brine Ie brine Ie brine Ie brine Ie H2Od or Brine Ie or Brine IIfor CO2 sat.g H2Od or Brine Ie or Brine IIf or CO2 sat.g H2Od or Brine Ie or Brine IIf or CO2 sat.g H2Od or Brine Ie or Brine IIf or CO2 sat.g H2Od or Brine Ie or Brine IIf or CO2 sat.g
H2Od or Brine Ie or Brine IIf or CO2 sat.g
4.35 4.35 4.35 4.35 4.325 4.325 4.325 4.325 4.325 4.325 -
13.7 CO2a or NaHCO3b 13.7 CO2a or NaHCO3b steamc or no steam 13.7 CO2a or NaHCO3b steamc or no steam
13.7 CO2a or NaHCO3b steamc or no steam
5 5 5 5 0.5 0.5 0.5 0.5 0.5 0.9 0.9
0.9
-
0.5
696 249 50 689.5 255.5 50 682.5 262.5 50 671 274 50 69.6 24.9 5 68.95 27.5 5 68.25 26.25 5 67.1 27.4 5 65.9 28.6 5 65 29.5 5 70 23.5 5 67.5 26 5 65 28.5 5
70 23.5 5
2.7:1 2.6:1
2.978:1
2.978:1
2.2:1
2.3:1
2.45:1
2.6:1
2.7:1
2.8:1
2.45:1
2.6:1
Trial 3 Trial 2 Trial 1
2.28:1
curing process
Bauxsol Pellets Made with Modified Makeup Waters: Trial 1. The use of modified makeup waters had little effect on reducing soluble pH of the pellet material. Saturated CO2
ratio (Bauxsol: cement) Bauxsol (g) cement (g) hydrophilic fumed silica (g) aluminum powder (g) carbopol (g/L) make-up water
Results
TABLE 1. Bauxsol Pellets Composition and Description of the Three Different Trials
The general recipe was an adaptation from Clark et al. (8) and consisted of a mix of six ingredients: Bauxsol powder (pH 10) substantial aluminum may be mobilized from the pellet. To effectively influence alkalinity precipitation, we must add Ca and Mg chlorides. Alkalinity conversions (eqs 4, 5, and 6) and salt addition theory (17, 18) suggest a reduction in reaction pH and soluble alkalinity and an increase of EC (Table 2; Figures 1, 3, and 4). The data from Trials 2 and 3 confirmed this. However, the increase in reaction pH with increase in the Bauxsol:cement ratio in Trial 3 possibly came from the dilution of brine solution, which prevented the full neutralization reactions. This result may also reflect the increase in the volumes of the samples to be neutralized or a combination of the two. Johnston et al. (19) showed that for the same ionic strength, the CaCl2 and MgCl2 are 9 and 40 times, respectively, more EC responsive than NaOH. The substantial increase in EC for Trial 2 (Table 2; Figure 4) was most likely caused by divalent Ca and Mg as residual interstitial brine. The washing data showed that by rinse four 93% of the soluble salts were removed. Therefore, rinsing brine-cured pellets with a small quantity of fresh water prevents water EC increases. During the pH stability test, each mix ratio of Brine I and Brine II (Table 4) showed a rapid equilibration, which suggests that when all hydroxides have precipitated the minerals deposited are almost insoluble. Moreover, the pH remained stable even after a few days unlike the CO2 neutralized materials of Khaitan et al. (10). However, Trial 3 batches (Table 4) showed a steadier pH stability time, presumably due to the larger surface to volume ratios, thereby creating longer diffusion paths into the pellet material. Consequently, the brine took more time to precipitate alkalinity. Explanations for the high pH (above 9 after 3 h) of Bauxsol:cement ratios 2.6:1 and 2.8:1 include a fresh exposure of unreacted hydroxides ions from internal fracturing and pore activation to previously unexposed areas. Because aluminum powder has been used as a foaming agent (11) and in relatively large quantity (2% of cement mass or 0.5% of total dry weight), the slake durability index was poor. Yam (21) found comparable results with 0.6% aluminum powder of cement mass. While small improvements in slake durability were observed in Trial 3, no discernible differences between the treatment brine or the Bauxsol:cement ratio were noted. The data suggests that the pellets produced in this work may have handling issues (i.e., breakage and size reductions, leading to clogging and lower hydraulic conductivities). This is at odds with work reported by Clark et al. (9), where
