Armela Tafa successfully defended her PhD-Thesis – Congratulations!

Optimization of Polar Organic Chemical Integrative Samplers for Compound-Specific Stable Isotope Analysis: Investigating the Effects of Polyethersulfone Membrane Porosity and Sorbent Selection

Abstract

The widespread presence of organic micropollutants (MPs), such as pesticides and pharmaceuticals, in natural waters raises critical questions regarding their sources, behavior, and long-term environmental fate. Compound-specific isotope analysis (CSIA) provides a powerful tool to address these questions by measuring variations in stable isotope ratios (e.g., 13C/12C, 2H/1H, 15N/14N,  18O/16O) of individual compounds at natural abundance levels. These isotopic fingerprints can reveal subtle degradation processes and transformation pathways that may not be detected through concentration-based approaches alone. However, the practical application of CSIA to trace contaminants in the field remains limited due to the relatively high mass required for accurate analysis by gas or liquid chromatography coupled to isotope ratio mass spectrometry (GC-IRMS or LC-IRMS). As a result, working with environmentally relevant concentrations often demands actively sampling
tens to hundreds of liters of water, followed by exhaustive solid-phase extraction (SPE) procedures. This approach is labor-intensive, time-consuming, and ultimately restricts the feasibility of CSIA in large-scale or remote monitoring campaigns. 
Therefore, passive sampling offers a promising alternative for in situ analyte enrichment, and the Polar Organic Chemical Integrative Sampler (POCIS) is widely used for capturing time-weighted average concentrations of polar MPs. Yet, standard POCIS configurations require long deployment  durations (typically 30–60 days) to accumulate sufficient analyte mass, increasing the risk of matrix interferences such as natural organic matter (NOM) and biofouling—factors that can compromise isotopic accuracy. These limitations highlight the need for optimized POCIS designs tailored for CSIA, with enhanced analyte uptake, reduced deployment time, and improved selectivity. This dissertation addresses these challenges by systematically refining membrane and sorbent properties to establish a robust and field-compatible POCIS-CSIA methodology. Three interlinked experimental chapters were developed to systematically evaluate the influence of membrane porosity, sorbent selectivity, and sampler geometry on the analyte uptake process and matrix exclusion performance. 
In Chapter 2, POCIS devices equipped with high-porosity membranes (8 μm PES) were shown to yield up to 3.5-fold higher analyte accumulation compared to the conventional 0.1 μm PES membranes. This enhancement was attributed to reduced membrane diffusion resistance, with minimal co-accumulation of natural organic matter (NOM), thereby increasing analyte selectivity by a factor of two. No isotopic fractionation was observed for δ13C and δ15N values under these conditions, confirming the suitability of the optimized setup for isotope analysis.
Chapter 3 introduced cyclodextrin-based polymers (CDPs) as alternative POCIS sorbents to the widely used Oasis HLB. Under flow conditions up to 2 ms−1, CDPs exhibited faster equilibrium attainment and reduced co-uptake of humic acid, while maintaining a stable isotopic signature across both constant and variable concentration regimes. Mechanistic interpretation was supported by a three-compartment mass transfer model, which partitioned overall uptake resistance into contributions from the water boundary layer (1/kW), the membrane (1/kM · KMW), and the sorbent (1/kS · KSW). Parameters for each phase were derived from fitting compound-specific accumulation data under controlled flow conditions. Modeling confirmed that membrane diffusion was the dominant rate-limiting step for Oasis HLB, while uptake into CDPs was primarily limited by sorbent interactions. Importantly, CDPs showed superior performance only for analytes with relatively high partitioning coefficients to the sorbent phase (e.g., metolachlor and boscalid). For other compounds with lower affinity, Oasis HLB remained the more efficient sorbent, underscoring the need for compound-specific sorbent selection in POCIS-CSIA applications. Experimental and modeled sampling rates (Rs) aligned within a ±5% uncertainty, supporting the robustness of the mechanistic approach. 
Chapter 4 further explored the influence of membrane thickness on sampling rates by comparing PES membranes with similar pore sizes but differing thicknesses (130 and 190 μm). This investigation aimed to decouple the effects of pore size and thickness on analyte uptake. Both experimental data and mechanistic modeling using the three-compartment framework confirmed that membrane thickness is a critical parameter controlling diffusion resistance when pore size remains constant. Thinner membranes facilitated more efficient mass transfer across the membrane phase, thereby increasing analyte accumulation in the sorbent. Simulations predicted that reducing membrane thickness from 200 to 30 μm could increase sampling rates by up to a factor of three. These findings underscore the importance of optimizing not only sorbent selection and membrane porosity but also membrane geometry when designing POCIS configurations for improved performance. The insights from Chapter 4 reinforce that reducing membrane diffusion resistance is essential to enhance uptake kinetics, particularly for analytes with limited sorbent affinity or in environments with low external turbulence.  
Overall, this dissertation establishes a field-ready POCIS-CSIA configuration that enables isotopic analysis of polar organic contaminants at environmentally relevant concentrations with reduced sampling effort. The demonstrated gains in mass accumulation, selectivity, and analytical reliability offer new opportunities to extend CSIA applications into routine environmental monitoring.