A team of South African researchers at the University of Venda has now unveiled a strikingly inventive answer to that sample-preparation challenge, one that begins in an unlikely place: the bin of discarded cigarette butts. In a study published in Results in Chemistry, Herbert Musarurwa and colleagues transformed ordinary cellulose acetate cigarette filters into high-performance sorbents for microextraction by packed sorbents, or MEPS, a miniaturized cousin of conventional solid-phase extraction. The work is notable not only for its chemistry but for its rigor of self-assessment: the authors scored their method with six separate metric tools covering greenness, practicality, analytical performance and innovation, an unusually comprehensive audit for a sample-preparation protocol.

The heart of the innovation lies in a hydrophobic deep eutectic solvent, or HDES, prepared from two inexpensive and comparatively benign ingredients: menthol and salicylic acid, mixed in a 4:1 molar ratio. Deep eutectic solvents form when hydrogen-bond donors and acceptors combine into a homogeneous liquid at temperatures far below those needed to synthesize conventional ionic liquids or polymeric sorbents. In this case, the mixture was simply sonicated for thirty minutes at 70 degrees Celsius. Fourier transform infrared spectroscopy confirmed the chemistry: the hydroxyl stretching bands of the two starting materials, at roughly 3229 and 3234 inverse centimetres, shifted to 3386 inverse centimetres in the solvent, the fingerprint of an extensive hydrogen-bonding network between salicylic acid and menthol.

Coating the filters proved equally straightforward. The researchers submerged cigarette filter segments in a methanol solution of the HDES and sonicated the mixture for ten minutes at 70 degrees, then air-dried the modified filters for twenty-four hours. Infrared analysis of the coated filters showed small but telling shifts in the characteristic cellulose acetate bands, including the ester carbonyl vibration near 1733 inverse centimetres and the acetyl C–O stretch near 1228, alongside the appearance of a pronounced hydroxyl peak at 3412 inverse centimetres, together confirming that the eutectic solvent had become incorporated into the filter surface. Scanning electron microscopy revealed the practical payoff: a highly fibrous, interwoven network of well-separated cellulose acetate strands forming an open, hydrophobic architecture riddled with diffusion channels that give analytes easy access to adsorption sites.

Turning the coated filters into extraction devices required little more than a syringe. Each 10-millimetre segment of modified filter, the length that performed best in preliminary tests, was packed into a cleaned 5-millilitre syringe barrel and conditioned with methanol, which the team showed does not strip the coating because the modification itself was performed in methanol. During extraction, one gram of homogenized vegetable was vortexed with six millilitres of methanol, centrifuged, and the supernatant drawn up and expelled through the packed syringe repeatedly. Ten draw-eject cycles proved optimal, and the pumping action eliminated the need for vortex-assisted extraction, reducing energy consumption while driving efficient mass transfer of pesticides from the sample onto the sorbent. Methanol then eluted the trapped analytes for analysis by high-performance liquid chromatography with photodiode-array detection.

Optimization was carried out with genuine statistical discipline rather than trial and error. A full factorial screening design first identified which of four parameters, ionic strength, elution solvent volume, elution time and sample pH, actually mattered, with Pareto charts revealing that ionic strength dominated for fenitrothion while elution volume and pH governed chlorpyrifos extraction. A central composite design of eighteen runs then mapped the response surfaces in three dimensions. The two pesticides behaved quite differently: fenitrothion recovery climbed with increasing ionic strength up to about 12 percent, a classic salting-out effect, whereas chlorpyrifos recovery fell as salt concentration rose, presumably because excessive viscosity impedes mass transfer and can clog the sorbent pores. Both compounds degraded under alkaline conditions, so slightly acidic samples were essential. A desirability function finally reconciled the competing optima, settling on a sample pH of 6.08, an ionic strength of 10.8 percent and an elution volume of 600 microlitres for the simultaneous extraction of both pesticides.

The validated method performed credibly against far more elaborate techniques. Matrix-matched calibration curves were linear over ranges spanning roughly three orders of magnitude, with correlation coefficients between 0.9929 and 0.9993. Limits of detection ranged from 0.02 to 0.10 micrograms per gram, and limits of quantification from 0.08 to 0.29 micrograms per gram, in cucumber and tomato. Spiked samples yielded extraction recoveries between 74.8 and 90.3 percent, with enrichment factors of 7.48 to 9.03 and intra-day and inter-day precisions generally below 9 percent relative standard deviation. None of the pesticides was detected in the unspiked real samples tested. In a head-to-head comparison table, the cigarette-filter method matched or approached the sensitivity of approaches based on molecularly imprinted polymers, graphene oxide composites and kapok-fiber biochar, while posting a greener environmental profile than most of them.

That green profile was quantified rather than asserted. The Complementary Green Analytical Procedure Index pictogram was dominated by green and yellow zones, and the AGREE algorithm returned an overall score of 0.72, high for a sample-preparation method. The authors were careful to note that greenness alone does not guarantee a useful technique, so they also applied the Blue Applicability Grade Index and the Click Analytical Chemistry Index, scoring 67.5 and 78 respectively for practicality, the Red Analytical Performance Index, which returned a satisfactory 70, and the Violet Innovation Grade Index, which returned 75. The method’s main environmental liability, its reliance on offline analysis with its attendant handling steps, was mitigated by the tiny 600-microlitre elution volume, and sonication was the only energy-consuming step in the entire workflow. The extraction device itself could be reused through four adsorption-desorption cycles without significant loss of performance, and even eroded devices can reportedly be re-coated and pressed back into service.

Beyond the immediate results, the study points toward a genuinely circular form of analytical chemistry. Cigarette butts are among the most littered items on Earth, and the authors suggest that used filters, not just fresh ones, could be valorized as sorbent precursors, converting a ubiquitous waste stream into laboratory infrastructure. The extraction mechanism relies on a combination of hydrophobic interactions, ion-dipole and dipole-dipole contacts and pi-pi stacking between the aromatic pesticides and the coated filter, interactions that the hydrophobic menthol-salicylic acid solvent is ideally suited to mediate. The team acknowledges that the protocol has so far been demonstrated only for two organophosphates in two vegetables, and that multiresidue applications remain to be tested. Still, the combination of a waste-derived sorbent, a green solvent, minimal energy input and rigorous multi-metric validation offers a template for how trace analysis in food might be made both cheaper and cleaner, putting sophisticated pesticide monitoring within reach of laboratories that could never afford conventional solid-phase extraction consumables.

Subject of Research: Green microextraction of organophosphate pesticide residues from vegetables using deep eutectic solvent-modified cellulose acetate cigarette filter sorbents

Article Title: Deep eutectic solvent-modified cellulose acetate cigarette filters for the microextraction by packed sorbents (MEPS) of pesticides in vegetables: Application of green metric tools

Article References: Musarurwa, H., Tavengwa, N. T., Mokgehle, T. M., Madala, N. E., & Selahle, S. K. (2026). Deep eutectic solvent-modified cellulose acetate cigarette filters for the microextraction by packed sorbents (MEPS) of pesticides in vegetables: Application of green metric tools. Results in Chemistry, 31, Article 103927. https://doi.org/10.1016/j.rechem.2026.103927

Image Credits: AI Generated

DOI: 10.1016/j.rechem.2026.103927

Keywords: deep eutectic solvents, microextraction by packed sorbents, pesticide residues, chlorpyrifos, fenitrothion, cellulose acetate, cigarette filters, green analytical chemistry, HPLC, vegetables, sample preparation, AGREE metric

The study was led by Peter M. Chilipweli and Elias C. Nyanza of the Catholic University of Health and Allied Sciences in Mwanza, together with Aiwerasia Vera Ngowi of Muhimbili University of Health and Allied Sciences and Karim Manji, also of Muhimbili. Their team recruited 337 mother–child pairs from reproductive and child health clinics across four districts in the Iringa and Njombe regions. Of these, 220 came from districts classified in advance as pesticide-exposed — Kilolo and Makambako, both centers of intensive horticulture — while 117 came from comparison districts, Njombe Town and Iringa Municipal, where agricultural chemical use is expected to be lower.

The central measurement was erythrocyte acetylcholinesterase activity, assessed during the first trimester of pregnancy. Acetylcholinesterase is the enzyme that breaks down acetylcholine, the neurotransmitter released at nerve synapses, and organophosphate insecticides work precisely by inhibiting this enzyme, causing overstimulation of nerve pathways. Because red blood cell acetylcholinesterase mirrors the enzyme found at nerve endings, its activity in maternal blood serves as a widely used proxy for the body’s internal dose of organophosphate exposure. The researchers classified each mother as having either low inhibition or high or moderate inhibition of the enzyme. Overall, 38.3 percent of the mothers showed moderate or high inhibition, a strikingly high burden of biological exposure in a general population of pregnant women rather than a group of occupationally selected workers.

Alongside the blood measurements, the team administered a structured pesticide-use inventory, classifying the products farmers reported using according to the international IRAC, FRAC and HRAC classification systems for insecticides, fungicides and herbicides. The picture that emerged was one of heavy reliance on organophosphates, particularly chlorpyrifos and profenofos, alongside pyrethroids and neonicotinoids. Notably, 24.3 percent of the reported use involved pesticide mixtures, including combinations of neonicotinoids with organophosphates — a pattern of co-formulation and co-application that toxicologists increasingly worry about, since different chemical classes can interact in ways that single-compound risk assessments do not capture.

When the infants reached approximately six months of age, the researchers assessed their development using the Malawi Developmental Assessment Tool, an instrument designed and validated for African settings that evaluates four domains: gross motor function, fine motor function, language and hearing, and social behavior. The overall picture was reassuring at the population level. Most infants — 88.4 percent — showed normal composite development, and across the individual domains, between 94.7 and 97.3 percent scored within normal ranges. Yet beneath that generally healthy surface, the statistical analysis revealed a consistent signal tied to prenatal exposure.

Using parsimonious multivariable Firth penalized logistic regression — a technique chosen because it produces more stable estimates when outcomes are imbalanced and sample sizes are modest — the researchers found that high maternal acetylcholinesterase inhibition, indicating high pesticide exposure, was associated with roughly triple the odds of mild or delayed composite neurodevelopment in the infants, with an adjusted odds ratio of 2.94 and a 95 percent confidence interval of 1.02 to 8.33, reaching statistical significance at p equal to 0.045. The confidence interval was wide, reflecting the modest number of delayed cases, but the association aligned with a substantial body of evidence from other parts of the world.

Perhaps the most intriguing finding was a significant interaction between pesticide exposure and the sex of the child, with an adjusted odds ratio of 4.30 and a 95 percent confidence interval of 1.05 to 17.56, at p equal to 0.042. This suggests that the relationship between maternal pesticide exposure and infant neurodevelopment may differ for boys and girls, echoing a recurring theme in the environmental health literature, where male fetuses have often appeared more vulnerable to neurotoxic insults, although the direction and magnitude of sex differences vary across studies. The authors themselves are careful to note that this particular estimate was imprecise and requires confirmation in larger, adequately powered, sex-stratified analyses before firm conclusions can be drawn.

The study also identified protective social factors. Mothers who had completed secondary or tertiary education had lower odds of having an infant with mild or delayed composite development, with an adjusted odds ratio of 0.47 and a confidence interval of 0.22 to 1.00, just crossing the threshold of significance at p equal to 0.050. Increasing maternal age was associated with lower odds of language and hearing delay, with an adjusted odds ratio of 0.86 per year of age and a confidence interval of 0.74 to 0.99 at p equal to 0.042. These findings underscore that developmental risk in such communities is shaped by a web of influences — schooling, maternal experience, household environment — that interact with chemical exposures rather than operating in isolation.

The broader context gives the findings weight. Agricultural intensification in Tanzania’s Southern Highlands has driven a marked increase in reliance on organophosphate pesticides, and evidence from cohorts such as the CHAMACOS study of farmworker families in California has linked prenatal organophosphate exposure to poorer cognitive, motor and behavioral outcomes in childhood. What has been missing is data from low- and middle-income countries, and especially from sub-Saharan African smallholder communities, where women often work in the fields themselves, handle concentrates without adequate personal protective equipment, and face exposure through spraying, mixing, re-entering treated plots and laundering contaminated clothing. This study provides a rare, early-window measurement in exactly that population, and it does so using an objective biomarker rather than self-reported exposure alone.

The authors emphasize that the cross-sectional design, the single exposure measurement in the first trimester, and the wide confidence intervals all counsel caution, and they call for longitudinal studies incorporating repeated, pesticide-specific exposure measurements, longer neurodevelopmental follow-up and properly powered sex-stratified analyses. Still, the practical implications are immediate. The team recommends strengthening safer pesticide-use practices in horticultural communities, providing antenatal counseling on occupational exposure, and building developmental surveillance into routine reproductive and child health services so that infants at elevated risk are identified early. The work was funded by Tanzania’s Ministry of Education, Science and Technology through the Higher Education for Economic Transformation Project, and approved by the Catholic University of Health and Allied Sciences and Bugando Medical Centre Research Ethics and Review Committee. For the farming families of Iringa and Njombe, the message is that the first weeks of pregnancy may be a critical window in which pesticide exposure leaves measurable traces in an infant’s developing nervous system — and that protecting mothers in those early months is an investment in the next generation’s minds.

Subject of Research: Prenatal organophosphate pesticide exposure and infant neurodevelopment in Tanzanian smallholder farming communities

Article Title: First-trimester maternal erythrocyte acetylcholinesterase activity and neurodevelopmental outcomes at 6 months among children of smallholder horticultural growers in the Southern Highlands of Tanzania

Article References: Chilipweli, P. M., Nyanza, E. C., Ngowi, A. V., & Manji, K. (2026). First-trimester maternal erythrocyte acetylcholinesterase activity and neurodevelopmental outcomes at 6 months among children of smallholder horticultural growers in the Southern Highlands of Tanzania. BMC Pediatrics. https://doi.org/10.1186/s12887-026-07698-2

Image Credits: AI Generated

DOI: 10.1186/s12887-026-07698-2

Keywords: acetylcholinesterase, organophosphate pesticides, chlorpyrifos, neurodevelopment, Malawi Developmental Assessment Tool, prenatal exposure, smallholder horticulture, Tanzania, infant health, occupational exposure, child sex differences, public health