The research team, led by Jann-Gerrit T. Müller of the Leibniz Institute for Baltic Sea Research in Warnemünde, set out to answer a deceptively simple question. Contaminants of emerging concern, a category that includes pharmaceuticals, personal care products, and industrial chemicals, are routinely detected in coastal waters at concentrations ranging from nanograms to micrograms per liter. Yet their fate is not governed by inputs alone. It depends on the chemistry of the receiving water, and chief among those variables is the vast, poorly characterized pool of dissolved organic matter, or DOM, that pervades every aquatic ecosystem.

DOM is far more than background noise. This complex mixture of carbon compounds can shield pollutants from light by absorbing photons before they reach their targets, or it can quench the reactive intermediates that break contaminants down. At the same time, DOM can act as a photosensitizer, generating an arsenal of photoproduced reactive intermediates, including excited triplet states of DOM itself, hydroxyl radicals, and singlet oxygen, that attack pollutants indirectly. Which of these competing roles wins out depends on the composition of the DOM, and that composition shifts dramatically with the seasons. Spring runoff delivers humic, terrestrial compounds to the Baltic coast, while summer phytoplankton blooms flood the water with protein- and carbohydrate-like material.

To disentangle these effects, the researchers designed a factorial photodegradation experiment using water from three regionally relevant sources. North Sea water from the German Bight supplied marine DOM, a forest stream near Rostock supplied terrestrial DOM, and brackish water from the Heiligendamm pier on the southwestern Baltic coast served as the unamended background matrix. After filtering all waters through 0.22-micrometer membranes and extracting DOM by solid-phase extraction, the team prepared four matrices: ultrapure water, background water alone, and background water spiked with either terrestrial or marine DOM at roughly three times the background dissolved organic carbon concentration.

Into each matrix went a cocktail of eight contaminants selected for their high detection frequency and risk potential in a recent year-long survey of the Rostock estuary: the pharmaceuticals sulfamethoxazole, carbamazepine, and diclofenac; the natural estrogens estrone and 17β-estradiol; the synthetic estrogen 17α-ethinylestradiol; the sunscreen UV filter benzophenone-4; and caffeine. Each compound was spiked to 2 micrograms per liter, a level ten to one hundred times above typical coastal concentrations but far closer to environmental reality than the concentrations used in many comparable studies. Twenty-eight bottles were incubated for 68 days under UVA-340 lamps emitting at 300 to 450 nanometers with an integrated irradiance of 17 watts per square meter, closely matching the mean diurnal UV irradiance measured at the nearby Warnemünde coast. Identical bottles wrapped in opaque boxes served as dark controls.

Optical monitoring revealed that the two DOM sources were chemically distinct from the start. Terrestrial DOM showed the highest aromaticity and average molecular weight, indicated by elevated carbon-specific UV absorbance at 254 nanometers and lower spectral slopes. Parallel factor analysis of fluorescence spectra resolved seven components, dominated by humic-like fluorophores that accounted for 90 percent of total fluorescence in the terrestrial matrix, 80 percent in the marine matrix, and 70 percent in the unamended Baltic water. Over the irradiation period, humic-like fluorescence declined by roughly a third while protein-like fluorescence rose by more than half, and the two DOM-spiked matrices converged optically, suggesting that terrestrial DOM contained a larger pool of photodegradable, optically active material.

The pollutant results were strikingly compound-specific. Diclofenac vanished fastest, degrading below detection limits within hours in every matrix, confirming direct photolysis as its dominant removal pathway. Sulfamethoxazole degraded fastest in ultrapure water, though the authors caution that pH-dependent speciation makes that result a poor guide to natural conditions. Caffeine and carbamazepine, by contrast, degraded more slowly in the unamended background water than in the DOM-spiked matrices, indicating that higher organic matter concentrations promoted their indirect photodegradation, likely through excited triplet states of DOM. The UV filter benzophenone-4 proved stubbornly persistent, degrading by no more than 82 percent even in DOM-containing matrices, a finding consistent with its design as a stable sunscreen agent and raising concerns about its accumulation in recreational coastal waters.

The estrogens told the most consequential story. All three degraded equally fast under irradiation, confirming photodegradation as an effective sink, but their breakdown was consistently and distinctly slower in the terrestrial DOM matrix than in the marine or brackish matrices, an effect that appeared independent of dissolved organic carbon concentration. The researchers attribute this inhibition to the higher aromaticity and humic-like fluorescence of terrestrial DOM, which may preferentially quench the reactive oxygen species known to drive estrogen degradation. In the dark controls, 17β-estradiol disappeared while estrone accumulated, consistent with microbial interconversion between the two compounds, whereas the synthetic estrogen 17α-ethinylestradiol showed no dark degradation at all, suggesting it could accumulate under light-limited conditions.

Estimated half-lives in unamended coastal water, doubled to approximate the natural day-night cycle, ranged from about four hours for diclofenac to fifteen days for sulfamethoxazole, with estrone and the synthetic estrogen at roughly three days each. Translated into environmental terms, diclofenac would fall below a risk quotient of 0.1 within a day, while the synthetic estrogen would need about thirty days and sulfamethoxazole about forty-three. Caffeine, with an estimated photodegradation half-life of 158 days, and carbamazepine at 28 days, may owe their high detection frequencies in regional surveys partly to this photochemical sluggishness. The authors emphasize that these estimates assume favorable irradiation conditions in clear, shallow surface waters and that the study’s small sample size limits generalization.

The climate connection is what elevates these findings from a laboratory curiosity to an environmental warning. Projected increases in terrestrial runoff, flooding, and permafrost thaw are expected to deliver ever more terrestrial DOM to coastal waters, and the Baltic Sea is already showing signs of browning. If terrestrial organic matter systematically retards estrogen photodegradation, as this study indicates, rising riverine inputs could prolong environmental exposure to compounds that disrupt endocrine systems at sub-nanogram-per-liter concentrations. The effect appears small in absolute terms, but given the extreme ecotoxicity of estrogens, even modest delays in degradation may matter. The authors call for factorial experiments with multiple independent DOM sources at varying carbon concentrations, combined with measurements of microbial community composition and metabolic activity, to separate photochemical from biological degradation and to build more reliable predictions of how coastal seas will process the pollutant mixtures of the Anthropocene.

Subject of Research: Photodegradation of emerging organic pollutants as a function of dissolved organic matter quality in coastal Baltic Sea waters

Article Title: Dissolved organic matter quality and emerging pollutants in coastal Baltic Sea waters: Insights into photodegradation

Article References: Müller, J.-G. T., Osterholz, H., & Kanwischer, M. (2026). Dissolved organic matter quality and emerging pollutants in coastal Baltic Sea waters: Insights into photodegradation. Environmental Advances, 26, Article 100755. https://doi.org/10.1016/j.envadv.2026.100755

Image Credits: AI Generated

DOI: 10.1016/j.envadv.2026.100755

Keywords: Baltic Sea, dissolved organic matter, photodegradation, contaminants of emerging concern, estrogens, pharmaceuticals, caffeine, benzophenone-4, PARAFAC fluorescence, coastal biogeochemistry, climate change, water quality

The research team, led by Bruno M. Exposto of the Polytechnic University of Bragança in Portugal, working with colleagues at the Federal University of Technology of Paraná in Brazil, took a deliberately counterintuitive approach. Rather than assuming that more processing means better performance, they compared raw, in-natura cork with pre- and post-carbonized derivatives of both cork and industrial almond shell residues. The results upended expectations: thermal activation through carbonization proved ineffective, while the humble raw cork, designated CiN in the study, delivered the best overall performance. At 25 degrees Celsius and a modest adsorbent dosage of 1.0 gram per liter, raw cork achieved a cumulative mean removal efficiency of 66.1 percent across the three estrogens simultaneously present in solution.

The numbers behind that headline figure reveal how the three hormones behaved when forced to compete for the same binding sites. The maximum experimental uptake capacities were 3.82 milligrams per gram for estrone, 3.75 milligrams per gram for 17β-estradiol, and a striking 7.23 milligrams per gram for the synthetic estrogen EE2, yielding a combined capacity of 14.79 milligrams per gram. That EE2, the artificial compound used in oral contraceptives, captured nearly twice the capacity of either natural hormone is one of the study’s most consequential observations, and the authors traced it to the molecule’s higher hydrophobicity and a lower activation energy barrier for its uptake. In a real wastewater matrix, where these compounds arrive together rather than one at a time, such competitive dynamics determine whether a treatment actually works.

Understanding the speed and shape of the uptake process was central to the work. The kinetic profiles of all three estrogens were accurately captured by the pseudo-second-order model, a mathematical framework that assumes the rate-limiting step involves chemical interactions between the adsorbate and the binding sites rather than simple diffusion-limited filling. That model predicted an equilibrium time of 440 minutes at 25 degrees Celsius, meaning the system needs roughly seven and a half hours to settle into its final partitioning between water and cork. For engineers designing contact tanks or packed-bed filters, that timescale matters: it defines how long contaminated water must linger against the material before the bulk of the hormone load has been transferred to the solid phase.

Equally important was the equilibrium modeling. Because the three estrogens compete, single-solute isotherm models, which describe one pollutant at a time, cannot capture reality. The team therefore applied the Sheindorf–Rebuhn–Sheintuch (SRS) multicomponent isotherm model, an extension of classical adsorption theory that explicitly accounts for competition between species sharing the same surface. The SRS model successfully simulated the competitive equilibrium data, providing a quantitative handle on how strongly each hormone suppresses the uptake of the others. This kind of validated multicomponent framework is rare and valuable, because most published adsorption studies report only single-contaminant results that dramatically overstate real-world performance.

The researchers also probed how acidity and alkalinity govern the process, and the answer was unambiguous. Highly alkaline media, at pH values of 12 or above, caused a significant decline in uptake capacity, an effect the authors attribute to electrostatic repulsion. At such extreme pH, the ionization states of the estrogen molecules and the cork surface change in ways that make the hormone and the adsorbent repel rather than attract each other. For practical applications, this finding brackets the operating window: the material performs best under the near-neutral conditions typical of most natural and treated waters, but strongly caustic environments would undermine it.

Thermodynamic analysis added another layer of mechanistic clarity. The sequestration process was confirmed to be spontaneous, with a negative standard Gibbs free energy change; exothermic, with a negative standard enthalpy change, meaning it releases heat and proceeds more readily at lower temperatures; and entropically favorable at the solid–liquid interface, with a positive standard entropy change. Taken together, these signatures describe a process that does not need to be driven by external energy input, a property that bodes well for low-cost, passive deployment in treatment trains, particularly in regions where sophisticated infrastructure or reagent supply chains are lacking.

Perhaps the deepest contribution of the study lies in its mechanism analysis. By combining kinetic, thermodynamic, and isotherm evidence, the authors identified a synergistic interplay between two uptake modes. The first is physisorption via pore filling, in which hormone molecules physically lodge inside the porous architecture of the cork. The second, and apparently the more decisive, is chemisorption governed by p-electron donor–acceptor interactions and hydrogen bonding networks. In these interactions, the electron-rich aromatic rings of the estrogen molecules engage with complementary electron-accepting or donating sites on the cork surface, while hydroxyl and carbonyl groups on both partners form hydrogen bonds that lock the hormones in place. Crucially, the team concluded that surface chemistry can play a more decisive role than texturized porosity in sequestering endocrine-disrupting chemicals, a conclusion that directly challenges the industry’s long-standing fixation on maximizing surface area through energy-intensive activation.

That conclusion carries real economic and environmental weight. Conventional activated carbon remains the workhorse adsorbent for micropollutant removal, but its production demands high-temperature processing and, often, chemical activation, both of which carry substantial carbon and cost footprints. Cork and almond shells, by contrast, are abundant agricultural and industrial byproducts. Portugal is among the world’s leading cork producers, and almond processing generates mountains of shell waste annually; the study’s materials were supplied by the companies Amendouro and Corticeira Amorim, underscoring the industrial relevance of the feedstocks. Converting these residues into an effective water-treatment medium without any carbonization step means the adsorbent arrives with a fraction of the embedded energy of conventional alternatives, and the waste stream becomes a resource rather than a disposal problem.

The broader context makes the work timely. Steroid estrogens are on the European Union’s watch list of substances of concern for water intended for human consumption, and monitoring studies across European rivers, including Portuguese estuaries such as the Douro and the Mira, have repeatedly detected estrogenic compounds at biologically active levels. Exposure to these chemicals has been linked to feminization of fish populations and a range of endocrine effects in wildlife, and they resist many conventional treatment steps such as coagulation and standard chlorination. Adsorption onto bio-based materials has emerged as one of the most promising remediation routes, but the field has lacked rigorous multicomponent studies that reflect the competitive reality of mixed contaminants. By delivering validated kinetics, a working multicomponent isotherm model, a full thermodynamic picture, and a coherent mechanistic narrative, all anchored in a material that requires no activation at all, the Bragança and Paraná team has provided both a practical candidate technology and a conceptual correction to the field: when it comes to capturing endocrine disruptors from water, what is on the surface of a sustainable residue may matter far more than how many pores it contains.

Subject of Research: Competitive adsorption of steroid estrogens from water using sustainable cork and almond-shell residues

Article Title: Competitive adsorption of natural and synthetic estrogens onto sustainable almond-shell and cork residues: kinetics, multi-component isotherms, and mechanism analysis

Article References: Exposto, B. M., Gomes, M. C. S., de Menezes, M. L., Brito, P., Queiroz, A., & Ribeiro, A. E. (2026). Competitive adsorption of natural and synthetic estrogens onto sustainable almond-shell and cork residues: kinetics, multi-component isotherms, and mechanism analysis. Environmental Science and Pollution Research. https://doi.org/10.1007/s11356-026-38277-8

Image Credits: AI Generated

DOI: 10.1007/s11356-026-38277-8

Keywords: estrogens, cork, almond shell, adsorption, endocrine-disrupting chemicals, water remediation, multicomponent isotherms, Sheindorf–Rebuhn–Sheintuch model, biomass valorization, p-electron donor–acceptor interactions, hydrogen bonding, wastewater treatment