Mobile Colloid Generation Induced by a Cementitious Plume: Mineral

Purchase temporary access to this content. ACS Members purchase additional access options · Ask your library to provide you and your colleagues site-w...
0 downloads 4 Views 3MB Size
Article pubs.acs.org/est

Mobile Colloid Generation Induced by a Cementitious Plume: Mineral Surface-Charge Controls on Mobilization Dien Li,*,† Daniel I. Kaplan,† Kimberly A. Roberts,† and John C. Seaman‡ †

Savannah River National Laboratory, Aiken, South Carolina 29808, United States Savannah River Ecology Laboratory, University of Georgia, Aiken, South Carolina 29802, United States



S Supporting Information *

ABSTRACT: Cementitious materials are increasingly used as engineered barriers and waste forms for radiological waste disposal. Yet their potential effect on mobile colloid generation is not well-known, especially as it may influence colloid-facilitated contaminant transport. Whereas previous papers have studied the introduction of cement colloids into sediments, this study examined the influence of cement leachate chemistry on the mobilization of colloids from a subsurface sediment collected from the Savannah River Site, USA. A sharp mobile colloid plume formed with the introduction of a cement leachate simulant. Colloid concentrations decreased to background concentrations even though the aqueous chemical conditions (pH and ionic strength) remained unchanged. Mobile colloids were mainly goethite and to a lesser extent kaolinite. The released colloids had negative surface charges and the mean particle sizes ranged primarily from 200 to 470 nm. Inherent mineralogical electrostatic forces appeared to be the controlling colloid removal mechanism in this system. In the background pH of ∼6.0, goethite had a positive surface charge, whereas quartz (the dominant mineral in the immobile sediment) and kaolinite had negative surface charges. Goethite acted as a cementing agent, holding kaolinite and itself onto the quartz surfaces due to the electrostatic attraction. Once the pH of the system was elevated, as in the cementitious high pH plume front, the goethite reversed to a negative charge, along with quartz and kaolinite, then goethite and kaolinite colloids were mobilized and a sharp spike in turbidity was observed. Simulating conditions away from the cementitious source, essentially no colloids were mobilized at 1:1000 dilution of the cement leachate or when the leachate pH was ≤8. Extreme alkaline pH environments of cementitious leachate may change mineral surface charges, temporarily promoting the formation of mobile colloids.



have been reported.24−27 However, the mobilization of colloids from heterogeneous sediments by cement leachate has not been adequately studied. Therefore, evaluating the long-term stability and impact of concrete degradation on the mobilization of sediment-derived colloids and their role in colloid-facilitated radionuclide transport becomes increasingly important. The objective of this study was to determine if cementitious leachate would promote the in situ mobilization of natural colloidal particles from a Savannah River Site (SRS) sandy sediment, and to test whether cementitious surface or subsurface structures would create plumes that favor sediment dispersion and mobile colloid release. Laboratory column studies with reconstructed soil cores were conducted. The cation chemistries of influents and effluents were analyzed by inductively coupled plasma-optical emission spectroscopy (ICP-OES), while the mobile colloids were characterized

INTRODUCTION Mobile mineral colloids (mean size at 1−1000 nm generally) such as quartz, phyllosilicate clays, Al-/Fe-oxides, and hydroxides, are ubiquitous in natural systems. Natural colloid suspensions are formed due to (1) the mobilization of fine particles from the porous medium by chemical (e.g., decrease in ion strength, ionic composition change from divalent to monovalent cations, increase in pH), physical (e.g., increase in flow rate), and biological (e.g., biofilm coating) disturbance;1−13 and (2) the reaction between fluids and rocks that causes dissolution of primary minerals and precipitation of new mineral colloids.14,15 It has become evident that natural mineral colloids facilitate the transport of subsurface contaminants.16−21 Because of the potential importance of particle mobilization, the mechanisms of release and redeposition of colloidal particles in porous media have been under investigation in recent decades. However, much of the knowledge of deposition and release cannot be used to predict colloid mobilization in heterogeneous natural porous media. Meanwhile, concrete has been or is proposed to be used widely as an engineered barrier or a low-level nuclear waste form.22,23 Observations of cementitious colloids derived from leached cement hydrates © 2012 American Chemical Society

Received: Revised: Accepted: Published: 2755

November 17, 2011 January 27, 2012 February 9, 2012 February 9, 2012 dx.doi.org/10.1021/es2040834 | Environ. Sci. Technol. 2012, 46, 2755−2763

Environmental Science & Technology

Article

was observed during initial column saturation or leaching with the artificial groundwater surrogate. The treatment solutions were injected upward from the bottom of vertically oriented columns by a peristaltic pump. The column experiments were performed in two steps at room temperature. First, artificial groundwater was flushed through the column at the flow rate of ∼1 mL min−1 for 25 PV, which corresponds to the Darcy velocity of ∼2.85 m day−1 for a column residence time of ∼20 min. Second, cement leachates were injected into the column at the flow rate of 0.35 ± 0.01 mL min−1 for up to 25 PV, which corresponds to the Darcy velocity of ∼1 m day−1 for a column residence time of 59.0 ± 1.6 min. Effluents were collected from the top of a column into test tubes through a fraction collector for one sample every 0.25 PV or every 14.75 ± 1 min. Turbidity, pH, and Electrical Conductivity. Within 24 h of their collection, the effluent samples were agitated for 30 s, and their turbidity was measured using a Hach model 2100AN turbidimeter. The effluent pH and electrical conductivity were measured using Radiometer Copenhagen PHM 95 pH meter and CDM 210 conductivity meter (CDM 241-9 probe), respectively. Particle Size Distribution and ζ-Potential. The particle size distribution and ζ-potential of the effluent suspensions were measured using ZetaPlus (Brookhaven Instruments Corp., Holtsville, NY). The particle size distribution measurement was set for 5 runs, requiring a total of 5 min, and the reported data were averages of 5 measurements. The ζ-potential was set to have 10 measurements and the reported data were the averages of the 10 measurements. In ZetaPlus, the ζ-potential is calculated from the measured electrokinetic mobility according to Smoluchowski equation

using X-ray diffraction (XRD), scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDX), dynamic light scattering (DLS), and laser Doppler velocimetry. The mobilization mechanisms of colloids in a SRS sandy sediment by cement leachates were investigated.



MATERIALS AND METHODS Artificial Groundwater (AGW). The AGW solution was made based on the monitoring survey results for uncontaminated groundwater samples at the SRS.28 The chemical composition (in mg L−1) of AGW was Na 1.25, K 0.25, Ca 0.93, Mg 0.66, Cl 5.51, and SO4 0.73. It had a pH ∼6.3, an electrical conductivity of 0.03 mS cm−1, and a turbidity of 4.1 NTU. Cement Leachate Simulant (CLS). The CLS was made following Serne et al.29 The CLS was filtered using 0.45-μm pore-size membrane filter. The chemical composition of CLS (in mg L−1) was Na 1830, K 760, and Ca 81.2. It had an initial pH of 12.8, an electrical conductivity of 21.8 mS cm−1, and a turbidity of 2.6 NTU. The CLS was diluted with DI water at the ratios of 1:10, 1:100, and 1:1000, hereafter referred to as treatments CLS1/10, CLS1/100, and CLS1/1000, respectively. For the CLS1/10 solution, pH values were adjusted using HNO3 over a range of 11.0 (referred as to CLS1/10-pH11.0) to 5.4. The dilute, pH-adjusted CLS solutions were used to evaluate colloid dispersion scenarios downstream from the source where some dilution and mixing has occurred. SRS Sandy Sediment. Subsurface sandy sediment was collected from the Tobacco Road Formation, which is a major formation in the lower vadose zone and water table aquifer at the SRS. This uncontaminated course-textured sediment sample was air-dried and passed through a #10 sieve, 2 mm. Its sand-size fraction comprised 97%, silt 2%, and clay only 1%. The mineralogy of the clay-size fraction (