Article pubs.acs.org/est
Remobilization Dynamics of Caffeine, Ciprofloxacin, and Propranolol following Evaporation-Induced Immobilization in Porous Media Hayley J. Normile,† Charalambos Papelis,‡ and Tohren C. G. Kibbey*,† †
School of Civil Engineering and Environmental Science, University of Oklahoma, Norman, Oklahoma 73019, United States Department of Civil Engineering, New Mexico State University, Las Cruces, New Mexico 88003, United States
‡
S Supporting Information *
ABSTRACT: Changing weather conditions can cause cycles of wetting and drying in the unsaturated zone. When porewater evaporates, any nonvolatile solutes present in the pores will be driven to adsorb and ultimately precipitate on solid surfaces. When media are subsequently resaturated through rainfall infiltration, the remobilization of solutes likely depends on both the hydraulics of resaturation and the dynamics of dissolution processes. The focus of this work was to study the dynamics of remobilization of three different emerging contaminants (caffeine, ciprofloxacin, and propranolol) and two model compounds (fluorescein and sulforhodamine B) from porous media following evaporation of porewater. Remobilization column experiments were conducted to study this phenomenon and were evaluated using a finite difference model developed to simulate the adsorption−desorption dynamics during resaturation and elution. Results indicate that dissolution dynamics become increasingly important with increasing adsorption affinity for solid surfaces. Trends in observed elution behavior are not well-predicted from chemical properties, such as solubility. One of the most significant observations of the work is the presence of spikes in elution concentrations well above initial porewater concentration, resulting from the hydraulics of the resaturation process. The effect is most significant in highly mobile compounds that exhibit low adsorption affinity for solid surfaces.
■
environment and in treated wastewater,12−14 and their capacities for producing physiological impacts mean that all three have the potential to create risk to human or environmental health. Two highly mobile fluorescent dyes, fluorescein and sulforhodamine B, were also studied to provide context to understand the behavior of the three emerging contaminants studied. Pore-water evaporation is a natural environmental process that has the potential to impact contaminant mobility by driving adsorption and precipitation of dissolved compounds, such as emerging contaminants. In the shallow vadose zone or in the beds of seasonal or intermittent streams, as water in the pore spaces evaporates, the concentrations of nonvolatile compounds increase. As concentrations increase and water phase-volume decreases, the changing phase equilibrium can drive adsorption to solid surfaces and ultimately formation of surface-associated solid precipitates. When the pores are subsequently resaturated through rainfall infiltration, changes in water table level, or flooding, the dynamics of contaminant mobilization will depend in part on
INTRODUCTION Emerging contaminants are contaminants that are currently not regulated in wastewater effluents or drinking water but can be widely detected in the environment. Examples include pharmaceuticals, chemical components of personal care products, and other soluble, poorly digested constituents of foods, such as artificial sweeteners. Emerging contaminants are present in waste produced by humans, animals, hospitals, and industrial effluents and can enter the environment via landfill leachate, leaks in sewage and wastewater systems, surface runoff, and illegal dumping.1,2 Even treated wastewater effluent, though compliant with regulatory levels for conventional contaminants, can contain low concentrations of a wide range of different emerging contaminants.3 Contamination by emerging contaminants has been detected in surface water and groundwater around the world.4−8 Emerging contaminants have the potential to cause harm to human and environmental health through a range of mechanisms, including increased bacterial resistance from exposure to antibiotics9,10 or cytotoxicity to fish.11 The focus of this work was on exploring how pore-water evaporation impacts the environmental mobility of three specific emerging contaminants: ciprofloxacin (a fluoroquinolone antibiotic), propranolol (a beta blocker), and caffeine (a stimulant). All three compounds have been detected in the © XXXX American Chemical Society
Received: Revised: Accepted: Published: A
December 12, 2016 May 2, 2017 May 15, 2017 May 15, 2017 DOI: 10.1021/acs.est.6b06294 Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Article
Environmental Science & Technology
acrylic and had a total internal volume of 6.72 cm3. Columns were packed with US Silica (Berkeley Springs, WV) F-65 fine sand to an average porosity of 0.37. The sand is a high purity foundry-grade sand, with very low fines and negligible organic carbon content. Note that carbon measurement with a different size fraction from the same sand series found fractional total carbon content (inorganic and organic combined) to be 3 × 10−5, at the low end of the detectable range for the instrument used.15 Before packing, the sand was rinsed with deionized water several times to remove fines and then oven-dried at 110 °C. Stainless steel screens (#40 mesh) at the top and bottom of the column supported capillary barrier membranes. Paper filters (Ahlstrom, 1.5 μm pore size) were used as water-wet capillary barriers to permit flow of water, while Teflon membranes (Tisch Scientific, 0.22 μm pore size) were used as nonwetting capillary barriers, to permit flow of air. Membranes were changed at specific points throughout the experiments as described below. Membranes were held in place by O-rings. Detection. Outlet flow from the sand column was analyzed with in-line UV spectrophotometric detection, using methods described in detail in previous work.16−18 After leaving the column, effluent entered a custom flow-through optical cell attached to a fiber optic spectrometer (Ocean Optics, Dunedin FL) with a deuterium light source. As in previous work,16 lowdead volume outlets were used in sand columns to minimize the time lag between fluid leaving the column outlet and entering the optical cell. The absorbance spectrum of the effluent was recorded at 1- or 10-s intervals throughout each experiment. An automated fitting program was then used to determine concentration over time by fitting full spectra to a linear combination of measured standard spectra. The method is capable of simultaneous detection of multiple compounds and has been found to be very good at rejecting interference from sand fines.16−19 Following each experiment, column contents were extracted with methanol or acetonitrile to quantify any compound remaining in the cell. Saturated Experiments. Saturated elution experiments were conducted with the column at full water saturation to determine the transport parameters needed for interpretation of remobilization experiments. Experiments were conducted with the column in a vertical orientation, with flow from top to bottom. The column was initially saturated with a pH-adjusted background solution for at least 5 pore volumes to stabilize the effluent pH and flush out any remaining sand fines. Inlet solution containing one compound was then flushed until the effluent concentration had reached a stable value equal to the inlet concentration. The inlet solution was then changed back to the background solution and flushed until the compound was removed from the column. Concentrations recorded during this phase of the experiment provided the elution curves that were used to determine adsorption parameters. Retardation coefficients (Rf) were determined directly from numerical integration under the saturated elution curves. Other transport parameters (dispersivity, desorption rate constant) were determined through fits to the transport model described in Finite Difference Elution Model below. All experiments (both saturated and remobilization) were conducted at a flow rate of 0.50 mL min−1 (0.47 cm min−1 porewater velocity). Remobilization Experiments. Remobilization elution experiments explored the transport of chemicals during elution from an initially fully dry state. Experiments involved initial saturation of the column, first with background solution and then with solution containing one compound, following the
the dynamics of dissolution. Because dissolution processes can be very slow, the potential exists for the dynamics of mobilization after evaporation to be substantially different from the dynamics of saturated transport. To date, work allowing prediction of the effect of evaporation-driven precipitation on mobilization has not been reported. This paper describes unsaturated/saturated column experiments specifically designed to study the dynamics of remobilization of initially dissolved chemicals following evaporation of pore water. A finite-difference model developed for this work is used to distinguish between adsorption− desorption rate effects and rate effects resulting from dissolution.
■
MATERIALS AND METHODS Chemicals. Three emerging contaminants were used for sand column experiments: caffeine (a stimulant), propranolol (a beta blocker), and ciprofloxacin (an antibiotic). Two fluorescent dyes, fluorescein and sulforhodamine B, were selected to act as model low-adsorption compounds. The five compounds were selected to cover a range of sorption behaviors and to be easily detected via UV-spectrometry for in-line analyses. Properties of the five compounds are given in Table 1. Note that both propranolol and ciprofloxacin exhibit Table 1. Properties of the Compounds Used under the Conditions Studied compound
pKa
MW (g mol−1)
fluorescein23,26 sulforhodamine B23 caffeine27,6 propranolol28
5.1
