Introduction

Global emissions of persistent bioconcentratable organic chemicals have resulted in a wide range of adverse ecological effects. Consequently, industry developed less persistent, more water soluble polar or hydrophilic organic compounds (HpOCs), which generally have low bioconcentration factors. However, evidence is growing that the large fluxes of these seemingly more environmentally friendly compounds (e.g., pesticides, prescription and non-prescription drugs, personal care and common consumer products, industrial and domestic-use chemicals, and their degradation products) into aquatic systems on a world-wide basis may be responsible for incidents of acute toxicity and sublethal chronic abnormalities [1-3]. These adverse effects include altered behavior, neurotoxicity, and severely impaired reproduction [4]. Furthermore, the presence of these HpOCs likely plays a major role in the endocrine disrupting effects of complex mixtures of chemicals present in aquatic environments [5,6]. In regard to physiological effects, pharmaceuticals are of particular concern because they are designed to elicit diverse pharmacological responses. Unfortunately,

$Although the research described in this chapter has been funded in part by the United States Environmental Protection Agency through (IAG #DW14900401) to USGS-CERC, it has not been subjected to Agency review and, therefore, does not necessarily reflect the views of the Agency and no official endorsement should be inferred.

Comprehensive Analytical Chemistry 48

R. Greenwood, G. Mills and B. Vrana (Editors)

Volume 48 ISSN: 0166-526X DOI: 10.1016/S0166-526X(06)48008-9

© 2007 Elsevier B.V. All rights reserved. 171

the effects of this class of HpOCs on non-target, aquatic organisms are largely unknown.

The HpOCs enter aquatic systems through treated effluents from wastewater treatment plants (WWTPs), leaking septic tanks and sewage lagoons, direct environmental disposal of unused drugs, landfill leachates, and surface runoff. They are often present at low concentrations, posing problems with most traditional sampling and analytical procedures. Measurements of HpOCs in environmental waters generally require modifications of existing methods or development of new methods to improve detection limits. The fate of these contaminants during wastewater treatment and in the environment is largely unknown. However, some findings suggest that many of these chemicals survive treatment and some are returned to their biologically active form via deconjugation [7-11]. Of greater concern is a study showing that many of these chemicals also survive treatment in drinking water plants and are present in finished waters [12].

There has been considerable effort directed towards development of active sampling methods for HpOCs in water, but nearly all this research has centered on the use of solid-phase extraction (SPE) employing specially modified polymeric resins in either a cartridge or enmeshed in an inert membrane disk [13-16]. Although SPE is advantageous over earlier liquid-liquid extraction methods, it often requires the collection of large volumes of water to satisfy the detection limit requirements of commonly used analytical methods. In cases where bulk (or filtered) water samples are shipped to the laboratory, the preservation and transport of large volumes of water can be problematic. On the other hand, the use of on-site automated sampling systems can be costly and difficult to maintain.

Because nearly all traditional sampling methods provide data only at the moment of sampling, episodic events such as spills or stormwater runoff may not be detected. This problem is particularly relevant to HpOCs, as their residence times in riverine systems are generally lower than hydrophobic organic compounds (HOCs). However, transient but frequent occurrence of certain HpOCs in wastewater effluents may result in temporal changes in the habitat quality of receiving waters. Thus, there is a critical need for sampling and analytical methods capable of enhancing the detection and identification of HpOCs in an integrated manner, which in turn, provides highly relevant time-weighted average (TWA) concentrations. Without this type of methodological advancement, investigators face a daunting task in adequately assessing the environmental risks posed by this diverse class of chemicals.

Passive samplers offer an attractive alternative to traditional sampling methods. The success of small personal dosimeters, or passive monitors, in determining TWA exposure concentrations of organic vapors in occupational environments has contributed to the application of the same principle to dissolved organic contaminants in aquatic environments [17,18]. A wide variety of passive samplers has been developed to sample HOCs, volatile organic compounds, and labile metals [19]. Passive integrative (i.e., no significant losses of accumulated residues during the exposure period) samplers concentrate ultra-trace to trace levels of chemicals over prolonged sampling periods, generally resulting in much greater masses of sequestered chemicals than those recovered using grab sampling techniques. Consequently, the use of this approach results in increased analytical sensitivity and lower detection limits relative to those reported for most traditional methods. Furthermore, the use of passive integrative samplers enhances the probability of the detection of chemicals that rapidly dissipate or degrade.

8.2 FUNDAMENTALS OF POCIS 8.2.1 POCIS description and rationale

The classification of a compound as an HpOC is based on the presence of one or more polar functional groups (e.g., hydroxyls) or a significant molecular dipole moment. The n-octanol-water partition coefficient (Kow) provides a convenient but somewhat arbitrary means of discriminating between HpOCs and HOCs. For example, volatile organic compounds may have relatively low Kow values but they are generally non-polar. In this chapter, we use a log Kow value of 3.0 as the cutoff point between HOCs and HpOCs However, it is important to have some overlap in the compounds sequestered by samplers for HOCs and HpOCs to ensure holistic sampling of organic contaminants.

Although a few passive sampling devices have been tested for HpOCs, the first sampler reported for this chemical class was the polar organic chemical integrative sampler or POCIS [20-25]. The POCIS has been shown to sample a wide variety of HpOCs as well as some HOCs with log Kow values between 3.0 and 4.0. The POCIS consists of a disk-like configuration of a solid-phase sorbent or a mixture of sorbents sandwiched between two microporous polyethersulfone (PES) membranes. Unfortunately, PES is not amenable to heat sealing, therefore, rings are used to form a compression seal to prevent sorbent loss. Figure 8.1 depicts an array of POCIS supported on a threaded rod with an exploded view of

Upper Compression Ring Upper Membrane Disk Sorbent Layer Lower Membrane Disk Lower Compression Ring

Fig. 8.1. An array of POCIS disks mounted on a support rod ready for insertion into a deployment canister. The inset is an exploded view of a single POCIS.

the ''membrane-sorbent-membrane sandwich'', which comprises the functional component of the sampler. The compression rings can be made of either a metallic or polymeric material (free of surficial or leachable contaminants) and a combination of thumb screws, bolts, nuts, or clips are used to secure the rings to the membranes.

The microporous PES membrane acts as a semipermeable barrier between the sorbent and the surrounding environment. It allows dissolved HpOCs to pass through to the sorbent, while particulates, microorganisms, and macromolecules with cross-sectional diameters greater than 100 nm are selectively excluded. Without the protection of the membrane, biofouling of POCIS sorbents is very likely during extended exposures (>2 weeks) in surface waters. Unlike the planar surfaces of PES or low-density polyethylene (i.e., membrane used for semipermeable membrane devices (SPMDs)), no effective methods are known for the direct removal of the biofilm-particulate phases from the surfaces of spherical or granular POCIS sorbents directly exposed to surface waters. Thus, the amounts of analytes accumulated in POCIS sorbents could not be distinguished from the amounts present in the biofilm-particulate phases, greatly complicating data interpretation. Fortunately, the PES membrane appears to resist biofouling much better than other polymeric materials commonly used in passive samplers and serves as an effective barrier to particle deposition eliminating these potential problems.

Upon deployment of POCIS, water rapidly permeates the pore structure of PES membrane and makes direct contact with the sor-bents. The water-filled transport corridors through the PES membrane are more tortuous than the linear structures of many microporous membranes. Based on mass per volume measurements and density information [20,23,26], the estimated volume of the hydrated pore structure is 76.5% of the total membrane volume. The average thickness of the hydrated PES membrane is approximately 130 mm.

For a typical POCIS disk used in field studies, the effective surface area of the membranes in contact with exposure waters is 41 cm2 and the sorbent mass is E228mg. Herein we define a standard POCIS as having a surface area to sorbent mass ratio of e 180 cm2 g-1. Because the amount of chemical sampled is directly related to the surface area of the device, it is sometimes necessary or desirable to combine the extracts from the sorbents of multiple POCIS disks into a single sample to increase the mass of sequestered chemical for analysis or bioassay.

The POCIS is versatile in that the sorbents can be changed to target-specific chemicals or chemical classes. However, two types of sorbent systems are considered as standards for all POCIS field deployments to date. Because each sorbent system is better suited for specific classes or types of HpOCs, it is common to have both standard sorbents in POCIS deployed in a single protective canister. This configuration is designed to maximize the number of detectable HpOCs at sample sites. One sorbent system consists of the triphasic admixture of Isolute ENV+ polystyrene divinylbenzene resin (80% by weight) and Ambersorb 1500 carbon lightly dispersed on S-X3 Biobeads (20% by weight). Ambersorb 1500 is no longer commercially available; however, Ambersorb 572 is an equivalent substitute. Details regarding the triphasic admixture have been discussed by Alvarez [20,23]. This mixture has a higher capacity than many other sorbents evaluated and exhibits excellent concentration of waterborne HpOCs with efficient recovery of most pesticides, natural and synthetic hormones, and other wastewater-related contaminants. Because of its broad applicability, the triphasic admixture is considered to be the generic-sorbent configuration for HpOCs. The other standard POCIS configuration incorporates the Oasis HLB sor-bent for optimum sequestering of pharmaceutical HpOCs. The use of the Oasis HLB configuration is necessary because many pharmaceuticals have multiple functional groups, which have a tendency to strongly bind to the carbonaceous component of the triphasic admixture, resulting in poor recoveries of some members of this class of compounds. Because Oasis HLB is a commonly used sorbent for the SPE-based sampling of pharmaceuticals and certain other HpOCs from water, considerable data exist on the recoveries of sorbed analytes. Furthermore, solvents used to recover chemicals from the Oasis HLB are generally compatible with toxicity tests, precluding the need for rigorous solvent exchange.

8.2.2 Applicability of POCIS

Although the standard POCIS configurations will concentrate a wide range of HpOCs, they are not suitable for all environmental contaminants. Table 8.1 lists the chemical classes or selected compounds shown to concentrate in POCIS, but it is not all inclusive. For compounds with log Kow values",4,0,4?>3.0, POCIS may not perform as well as other passive samplers such as SPMDs. A graphical representation of the sampling characteristics of POCIS and SPMDs relative to their log Kow values is depicted in Fig. 8.2. The normalized sampling rates for the POCIS are visibly less than those for the SPMDs. This difference is likely due to the additional aqueous transport resistance (relative to a SPMD) of the water-filled PES membrane. Clearly shown is the significant overlap in the types of chemicals sampled by the two devices. Therefore, the use of SPMDs and POCIS in concert should provide a better understanding of the full extent of organic chemical contamination.

0 0

Post a comment