MIP-on-the-flow: Molecularly imprinted polymers in microfluidic sensing systems

MIP-on-the-flow: Molecularly imprinted polymers in microfluidic sensing systems Meltem Okan a,* , Vildan Sanko a , Ender Yıldırım a,b,e , H. Cumhur Tekin a,c , Haluk Külah a,d,e,** a METU MEMS Center, Ankara, 06530, Türkiye b Department of Mechanical Engineering, Middle East Technical University, Ankara, 06800, Türkiye c Department of Bioengineering, ˙ Izmir Institute of Technology, ˙ Izmir, 35430, Türkiye d Department of Electrical and Electronics Engineering, Middle East Technical University, Ankara, 06800, Türkiye e Department of Micro and Nanotechnology, Middle East Technical University, Ankara, 06800, Türkiye ARTICLE INFO Keywords: Molecular imprinting in polymers Microfluidic MIP-based sensor Lab-on-a-Chip Point-of-Care Continuous-flow analysis ABSTRACT The integration of molecularly imprinted polymers (MIPs) with microfluidic systems has emerged as a powerful strategy for developing selective and sensitive analytical platforms. As “artificial receptors,” MIPs offer robustness, reusability, and cost-effectiveness, while microfluidics enable precise fluid handling and miniaturized analysis. Together, they yield hybrid sensors capable of real-time detection. Recent advances in polymerization, nanoimprinting, and surface functionalization have tailored MIPs for seamless microfluidic integration. In parallel, innovations in soft lithography and 3D printing have expanded design possibilities for lab-on-chip architectures. Cutting-edge detection modalities, including electrochemical, optical, and mass-based transduction, have unlocked applications in biomedical diagnostics, environmental monitoring, and food safety. Examples include continuous biomarker monitoring, trace pollutant detection, and rapid food contaminant identification. Despite progress, challenges in reproducibility, large-scale fabrication, and commercialization remain. Addressing these through material innovations and scalable engineering will accelerate translation into point-ofcare testing, environmental protection, and global food security. 1. Introduction The merging of molecularly imprinted polymers (MIPs) with microfluidic technologies represents a pivotal stride toward nextgeneration analytical platforms. MIPs are synthetic materials capable of selectively recognizing analyte molecules through tailor-made cavities formed during polymerization [1]. Often referred to as “synthetic antibodies,” these materials provide a stable, low-cost, and versatile alternative to natural receptors, resisting harsh environments and offering long-term usability [2]. Microfluidics, on the other hand, allows the controlled manipulation of fluids at the microscale, enabling precise, low-volume, and high-throughput analyses. When combined, MIPs and microfluidic systems enable powerful analytical devices that offer high selectivity, operational stability, and real-time responsiveness [3]. Microfluidic devices have redefined analytical chemistry by reducing assay time, minimizing reagent use, and supporting miniaturized, automated workflows [4]. Their inherent design allows for integration of multiple functions, such as sample processing, separation, and detection, into compact chips [5]. MIPs can be easily adapted to these platforms through a variety of integration methods, including electropolymerization on electrodes, adsorptive immobilization on modified channel surfaces, or incorporation into nanoparticles or porous matrices. This compatibility has driven the development of hybrid systems that deliver enhanced recognition capabilities, making them especially attractive for point-of-care (POC) diagnostics, environmental testing, and food quality assurance. Numerous applications now demonstrate the strength of this hybrid approach. In clinical diagnostics, MIP-based microfluidic sensors have enabled non-invasive, continuous monitoring of biomarkers such as cortisol and glucose [6,7]. Environmental applications include ultra-trace detection of pollutants like perfluorinated compounds and pesticides [8,9]. In food safety, MIPs have been used to detect contaminants such as antibiotics, toxins, and pathogens within complex matrices [10,11]. In some cases, these systems outperform traditional immunosensors, especially where stability, shelf-life, or reusability is * Corresponding author. ** Corresponding author. METU MEMS Center, Ankara, 06530, Türkiye. E-mail addresses: [email protected] (M. Okan), [email protected] (H. Külah). Contents lists available at ScienceDirect Trends in Analytical Chemistry journal homepage: www.elsevier.com/locate/trac https://doi.org/10.1016/j.trac.2025.118511 Received 31 July 2025; Received in revised form 26 September 2025; Accepted 23 October 2025 Trends in Analytical Chemistry 194 (2026) 118511 Available online 24 October 2025 0165-9936/© 2025 Elsevier B.V. All rights are reserved, including those for text and data mining, AI training, and similar technologies. Abbreviations ACEK alternating current electrokinetics ACET alternating current electrothermal APTES (3-aminopropyl)triethoxysilane ATD 2-amino-1,3,4-thiadizole BET Brunauer-Emmett-Teller BLI biolayer interferometry BPA bisphenol A CBD cannabidiol CDs carbon dots CE counter electrode CIP cell-imprinted polymer CFU colony-forming unit CMIP conductive molecularly imprinted polymer CNFM carbon nanofiber membrane CNNs convolutional neural networks CPF chlorpyrifos CRP C-reactive protein DAS N1,N12-diacetylspermine DHEBA dihydroxy ethylene-bisacrylamide DI deionized water DL deep learning DLP digital light processing DMSO dimethyl sulfoxide DOSY diffusion-ordered spectroscopy dPAD distance-based paper analytical device DPV differential pulse voltammetry DSC differential scanning calorimetry DVB divinylbenzene E. Coli Escherichia coli EC-SPME electrochemically assisted solid-phase microextraction EDC 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide EGDMA ethylene glycol dimethacrylate EIS electrochemical impedance spectroscopy ELISA enzyme-linked immunosorbent assay FCE flexible carbon electrode FLD fluorescence detection FRP free radical polymerization FTIR Fourier transform infrared spectroscopy GNP gold nanoparticle HEMA 2-hydroxyethyl methacrylate HIV human immunodeficiency virus HPLC high-performance liquid chromatography HR-MAS 1H high-resolution magic angle spinning IL-6 interleukin-6 ITO indium tin oxide LC-MS liquid chromatography-mass spectrometry LIG laser-induced graphene LoC lab-on-a-chip LOD limit of detection LRP living/controlled RP MAA methacrylic acid MIPs molecularly imprinted polymers ML machine learning MMA methyl methacrylate MM-SGM-chip microfluidic biomimetic chip MOFs metal-organic frameworks MR methyl red NHS N-hydroxysuccinimide NMR nuclear magnetic resonance OECT organic electrochemical transistor PBNPs prussian blue nanoparticles PC polycarbonate PDMS polydimethylsiloxane PEDOT poly(3,4-ethylenedioxythiophene) PET polyethylene terephthalate PiET photo-induced electron transfer PFAS perand polyfluoroalkyl substance PFOS perfluorooctanesulfonic acid PI polyimide PM piezoelectric microgravimetry PMMA polymethyl methacrylate PoPD poly(o-phenylenediamine) PS polystyrene PSS poly(styrene sulfonate) PVC polyvinyl chloride Py pyrole QCM quartz crystal microbalance QD quantum dot RAFT reversible addition fragmentation chain transfer RE reference electrode RP radical polymerization RSD relative standard deviation SAM self-assembled monolayer SARS-CoV-2 severe acute respiratory syndrome coronavirus 2 SEM scanning electron microscopy SDS sodium dodecyl benzene sulfonate SMI surface molecular imprinting SPP solvent-programmable polymers SPR surface plasmon resonance SVMs support vector machines TEA triethyleneamine TEGDMA triethylene glycol dimethacrylate TEM transmission electron microscopy TEOS tetraethyl orthosilicate TGA thermogravimetric analysis THC tetrahydrocannabinol TMOS tetramethyl orthosilicate TNFα tumor necrosis factor-alpha TRIM trimethylolpropane trimethacrylate UV–Vis ultraviolet–visible VIPs virus-imprinted polymers VSV-G vesicular stomatitis virus glycoprotein WE working electrode ZIF-8 zeolitic imidazolate framework 8 Symbols C concentration ΔG Gibbs free energy change E potential (V) ip peak current K binding constant Qe adsorption capacity V voltage ν scan rate λwavelength t time. M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 2 critical [12]. MIP-on-chip technologies have also extended to agriculture, where nutrient uptake in plants is tracked in real time, reflecting the growing versatility of these platforms [13]. However, successfully integrating MIPs into microfluidic systems remains a demanding task. Maintaining high fidelity of recognition sites during synthesis, ensuring robust immobilization within microchannels, and removing template molecules without damaging the polymer matrix are all technical challenges. The choice of polymerization methods, monomer-template pairs, and porogens critically influences recognition performance. Furthermore, the structural compatibility of MIPs with the microfluidic environment must be balanced with their mechanical durability and permeability. These considerations are compounded by scalability concerns, as many fabrication methods are still tailored for laboratory use rather than mass production. The field of MIP-based microfluidic sensors is yet new and rare but rapidly evolving (Fig. 1), driven by advancements in materials science, microfabrication, and analytical chemistry [2]. This review systematically discusses the design and integration strategies of MIP-based microfluidic sensors, with particular emphasis on non-paper-based platforms due to their advantages in material versatility, precise fluid handling, compatibility with advanced sensing technologies, scalability, durability, reusability, and support for complex and customizable designs. We examine established and emerging methods for MIP synthesis, including bulk, precipitation, miniemulsion, surface, and electropolymerization routes, and their suitability for microfluidic adaptation. We also detail fabrication techniques for microfluidic chips, from soft lithography to 3D printing, and summarize bonding and surface modification techniques essential for stable MIP incorporation. By highlighting recent advances in real-world applications, including wearable biosensors, environmental field assays, and food monitoring platforms, this review captures the rapid evolution and growing practical relevance of MIP-integrated microfluidic systems. We further explore innovations in signal transduction, including electrochemical non-Faradaic (capacitive and potentiometric) and Faradaic (amperometric and voltammetric), as well as optical readouts, and how these enhance detection accuracy and sensitivity. Finally, we outline current limitations and future prospects, polymer architecture, and scalable device manufacturing. Compared to existing literature, this review offers a focused perspective by examining the integration of MIPs into microfluidic platforms under analytical continuous-flow conditions. While previous reviews, such as those by Saylan et al. (2023) [14] and Karasu et al. (2023) [15], have explored MIP-based sensing in broader contexts from optical systems to general lab-on-a-chip (LoC) applications, they provide limited analysis of the design strategies, fabrication methods, and operational challenges unique to flow-based MIP sensors. Recent reviews like Cetinkaya et al. (2025) [16] and Wang et al. (2023) [17] have provided comprehensive overviews of electrochemical MIP sensors and their biomedical uses, but mainly focus on conventional sensor designs and transduction techniques. Although these studies highlight the analytical advantages of electrochemical platforms, such as high sensitivity, affordability, and portability, they offer limited discussion on integrating MIP sensing layers within dynamic microfluidic environments [18,19]. Our review advances this discussion by focusing on incorporating MIP-based sensors, including both electrochemical and optical platforms, within continuous-flow microfluidic systems. Beyond biomedical diagnostics, we expand the scope to include recent developments in MIP-based food safety and environmental monitoring methods. While many reviews emphasize clinical applications, our work emphasizes the importance of highly selective, miniaturized, and cost-effective sensors for detecting food contaminants, like antibiotics, toxins, pathogens, and environmental pollutants such as pesticides, heavy metals, and endocrine disruptors. By including examples from these fields, we demonstrate flow-integrated MIP sensors’ versatility and interdisciplinary applicability. This expanded perspective highlights the growing importance of these platforms for public health and environmental sustainability, showcasing their potential as universal analytical tools across various real-world scenarios. 2. The intersection of molecular recognition and microfluidics In the quest for highly sensitive, selective, and miniaturized analytical tools, the combination of MIPs and microfluidic systems has emerged as a transformative approach in sensor technology [16]. The convergence of these two technologies addresses critical demands in diverse fields, including biomedicine, environmental monitoring, food safety, and chemical analysis. MIPs, often regarded as “artificial receptors,” are synthetic materials engineered to exhibit a high affinity for chosen target analytes. Their robustness, reusability, and tailored selectivity offer distinct advantages over traditional biological recognition elements, such as antibodies and enzymes [20]. When integrated with microfluidic platforms, miniaturized systems capable of handling and analyzing small volumes of fluids with remarkable precision, MIPs enhance the sensitivity and selectivity of sensor devices while maintaining cost efficiency and scalability [21]. Microfluidic devices provide a unique platform for chemical and biological analysis by enabling rapid, automated, and multiplexed measurements [4]. The inherent properties of microfluidics, such as low sample and reagent consumption, high throughput, and portability [22], align seamlessly with the characteristics of MIPs. This synergy has catalyzed significant advancements in the development of highly efficient, compact, and precise sensors, paving the way for real-time and Fig. 1. Histogram for MIP-based microfluidic sensors published in the last decade and increasing citation graph for MIP-based microfluidic sensors (plotted using ISI Web of Knowledge platform) (Refined using keywords ‘Molecularly Imprinted Polymer’ & ‘Microfluidic’ under Topic search). M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 3 on-site applications. Despite their complementary strengths, integrating MIPs with microfluidic systems presents several challenges. The fabrication and incorporation of MIPs into microfluidic devices demand meticulous control over their functionalization to ensure compatibility with microfluidic architectures. Additionally, achieving a balance between the mechanical stability of MIPs and their molecular recognition capabilities is crucial for the successful implementation of these systems. Addressing the challenges requires a comprehensive understanding of the basic principles underlying molecular imprinting, surface functionalization and modification. One of the key challenges in integrating MIPs into microfluidic devices is ensuring their compatibility with microfluidic architecture. This involves optimizing the mechanical and chemical properties of MIPs to withstand the operational conditions of microfluidic systems while maintaining their molecular recognition capabilities. Strategies such as surface functionalization, nanoimprinting, and the use of flexible polymeric substrates have been developed to address these challenges. Additionally, innovative fabrication techniques, such as 3D printing and soft lithography, have expanded the possibilities for creating customized microfluidic devices with integrated MIPs. The applications of MIP-based microfluidic sensors span across diverse domains, reflecting their versatility and potential to address critical analytical challenges (Fig. 2). In biomedicine, these sensors have been used for drug monitoring [23] and detection of hormones [24] and biomarkers associated with diseases [25]. Their ability to perform rapid, sensitive, and selective analyses makes them invaluable for POC diagnostics and personalized medicine. In environmental monitoring, MIP-based microfluidic sensors offer a powerful tool for detecting toxins [26,27] and other hazardous substances [28]. Their high selectivity and stability enable the detection of trace levels of contaminants in complex environmental matrices, contributing to the protection of public health and the environment. In the food industry, these sensors have been employed for the detection of foodborne pathogens [29], allergens [30], and chemical residues [9,10], ensuring food safety and quality control. Their rapid and accurate detection capabilities make them ideal for on-site monitoring and compliance with regulatory standards. 3. Fundamentals of molecularly imprinted polymers While previous reviews have focused on general synthesis strategies and recognition mechanisms for MIPs, this review offers a more structured and mechanistic view of the fundamental principles behind molecular imprinting. The reviews by Hassan et al. [31], Akg¨ onüllü et al. [32], Ozcelikay et al. [33] and Wang et al. [34] make valuable contributions by outlining basic polymerization strategies and interaction types, however, this review expands on these foundations by providing a more detailed discussion of the designs and molecular-level factors involved in MIP synthesis. Additionally, this review includes a systematic assessment of monomer-template interactions, including hydrogen bonding, ionic forces, and π – π interactions, along with in-depth explanations of how these forces influence binding sites, selectivity, and stability. It also highlights the physicochemical rationale for selecting monomers and cross-linkers, offering insights into how the imprinting strategy impacts template removal efficiency and site accessibility. By presenting these principles within a more application-driven and design-oriented framework, our review aims to deepen understanding of how to tailor MIP structures for improved performance in practical sensing applications. 3.1. Principles of molecular imprinting MIPs shift a paradigm in the design of recognition elements for sensor systems. Unlike natural receptors such as enzymes, antibodies and nucleic acids etc. Which are often unstable and expensive to produce, MIPs are synthetic materials that can be tailored to recognize target molecules with high precision [35,36]. The molecular imprinting process involves the formation of a polymeric matrix around a template molecule, followed by the removal of the template to leave behind cavities that are complementary in size, shape, and recognizing sites to the target molecule size, shape and its binding sites. This process imparts MIPs with a unique combination of selectivity, stability, and cost-effectiveness, making them ideal for various analytical applications. The versatility of MIPs extends to their synthesis, which can be adapted to produce materials with specific properties, such as porosity, surface area, and distribution of imprinted cavities [37]. Techniques such as bulk polymerization, precipitation polymerization, and surface Fig. 2. Schematic illustration for the MIP-based microfluidic electrochemical and optical sensors and application fields under both bioand chemical sensors. M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 4 imprinting offer a wide range of options for tailoring MIPs to meet the requirements of specific applications. Additionally, advanced characterization methods, including spectroscopic, chromatographic, and microscopic techniques, provide valuable insights into the structural and functional attributes of MIPs, enabling their optimization for enhanced performance. These synthesis and characterization techniques are briefly mentioned here to highlight their relevance, detailed discussions of their principles, advantages, and application-specific considerations are provided in the subsequent sections. Fig. 3 illustrates the general principle of molecular imprinting, which involves the formation of a pre-polymerization complex between the target template and functional monomers, followed by polymerization in the presence of a cross-linking monomer. After template removal, vacant imprinted cavities, complementary in shape and functionality to the binding sites of the target molecule, are generated, enabling selective analyte binding. Template molecule sites are formed by polymerizing functional monomers in the presence of a template molecule, which serves as a model for the target analyte. Upon removal of the template, the polymer retains cavities that are complementary in size, shape, and chemical functionality to the template molecule, enabling selective binding to the target analyte. At the core of molecular imprinting lies the chemistry of intermolecular interactions that govern the pre-polymerization complex. Functional monomers interact with the template molecule through non-covalent forces such as hydrogen bonding, the van der Waals forces, and ionic interactions, or through reversible covalent bonds, depending on the chosen imprinting strategy [38]. Non-covalent imprinting is the most widely employed method due to its simplicity and the ease of template removal. The choice of monomers and their interaction strength directly influence the stability and selectivity of the pre-polymerization complex, which in turn affects the performance of the final MIP [39]. In 1995, Whitcombe and co-workers proposed a hybrid imprinting strategy that combined the advantages of covalent and non-covalent methods. In this approach, covalent bonding is employed during the polymerization step. At the same time, the recognition of the target molecule upon rebinding relies on non-covalent interactions, a process referred to as semi-covalent imprinting. Nevertheless, the non-covalent strategy remains the predominant method for MIP synthesis [40,41]. The quantity and quality of MIP recognition sites depend on how well the monomers and templates interact in the pre-polymerization mixture. Nicholls and Andersson et al. studied the thermodynamics of MIP recognition [42–44] and explained that the degree of template binding at equilibrium is determined by the Gibbs free energy change for each type of interaction between the template and the functional monomer. During the pre-polymerization phase, these interactions are thermodynamically controlled, so the monomer-template complex avoids strain or unfavourable interactions. Additionally, because MIPs are highly cross-linked, they maintain their structure and undergo minimal changes during template recognition. Since the polymerization and analyte binding typically occur in non-polar solvents, hydrophobic interactions are not the dominant cause for template recognition, with hydrogen bonding and π - π interactions playing a more significant role [45,46]. Hydrogen bonding plays a pivotal role in non-covalent imprinting strategies. Functional monomers such as methacrylic acid (MAA) can form strong, directional hydrogen bonds with templates, enhancing the stability of the pre-polymerization complex. For instance, Arshady and Mosbach demonstrated the efficacy of hydrogen bonding in creating high-affinity binding sites for theophylline, a bronchodilator drug, in MIPs [47]. The study highlighted how the selection of a monomer with appropriate functional groups enhances the recognition ability of the polymer. van der Waals forces, though weaker than hydrogen bonds, contribute significantly to the stability of the pre-polymerization complex. These interactions arise from transient dipole-induced dipole attractions and are particularly relevant for templates with hydrophobic regions. The presence of van der Waals forces ensures that the imprinting process captures the subtle structural features of the template molecule, thereby improving the selectivity of the MIP [48]. Ionic interactions, which involve electrostatic attraction between oppositely charged functional groups on the template and monomers, are another critical component of molecular imprinting. This type of interaction is especially useful for imprinting charged molecules, such as pharmaceuticals or environmental contaminants. For instance, Dai et al. reported the use of ionic interactions in the synthesis of MIPs for the selective binding of diclofenac, a common pharmaceutical pollutant, showcasing how electrostatic forces can enhance template-monomer binding [49]. In addition to non-covalent methods, reversible covalent bonding is employed in certain imprinting strategies to create highly stable pre-polymerization complexes. This approach involves forming covalent bonds between the template and monomers, which are later broken during template removal. The reversible nature of these bonds ensures the retention of imprinted cavity selectivity, pioneered by Wulff and Sarhan [50]. While covalent imprinting often results in superior selectivity, it is less commonly used due to the complexity of synthesizing and removing the template. Functional monomers interact with the template molecule through not only hydrogen bonding but also via ionic or π - π interactions. Table 1 shows frequently used monomers, where they are categorized based on their template interaction capabilities. Based on this table, templates interact with the aldehyde group of acrolein, the amide group of acrylamides, the carboxylic acid group of acrylic acid and MAA, the amine group of allylamine, the nitrile group of acrylonitrile and the dicarboxylic acid group of itaconic acid. Pyridine ring of 2-vinylpyridine and 4-vinylpyridine and imidazole ring of 1-vinylimidazole take part in basic interactions, lactam group in N-vinylcaprolactam is used in hydrogen bonding and tertiary amine group of N–N-diethylamino [ethyl methacrylate] allows for electrostatic interactions. Crosslinking monomers form rigid polymer matrix and provide structural integrity while maintaining the recognition sites. The types of cross-linking monomers vary based on molecules: N,N ′ -methylene bisacrylamide contains amidebased, ethylene glycol dimethacrylate (EGDMA) contains di-functional and m-divinylbenzene and p-divinylbenzene contain aromatic crosslinking monomers. Often used in biological applications, monomers with hydrophilic properties improve compatibility with an aqueous environment. 2-Hydroxyethyl methacrylate (HEMA), uraconic acid and uraconic acid ethyl ester contain hydrophilic hydroxyl, dicarboxylic acid Fig. 3. Schematic representation of a molecular imprinting process. M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 5 and ester groups, respectively. π - π interactions are non-covalent forces that result from the attraction between delocalized π -electron clouds in aromatic rings. These interactions typically involve the overlap of π -electron systems, which can occur in configurations like parallel stacking (face-to-face) or edge-to-face arrangements (T-shaped). The strength of these interactions is influenced by substituents on the aromatic rings, as they can modify the electron density of the π -system, either enhancing or reducing the attraction. In molecular imprinting, π - π interactions are especially important when aromatic monomers are used to bind aromatic template molecules. These interactions stabilize the monomer-template prepolymerization complex during polymerization, ensuring the formation of highly selective binding sites in the final polymer structure. Examples in Table 1 are vinylbenzene, m-divinylbenzene and p-divinylbenzene, which interact with aromatic templates or participate in π - π stacking. After polymerization, template removal is a critical step that determines the quality of the binding sites. Solvent extraction, often involving polar solvents like methanol or acetic acid, is commonly used to remove templates bound through non-covalent Table 1 Commonly used functional monomers, cross-linking monomers and example target molecules. For all cases, the interaction type is either covalent linkage or covalent network via free radical polymerization (FRP). Monomers Functional group Cross-linking monomers Target molecules Ref Functional monomer N Acrylamide Vinyl (CH – – CH 2 ), Amide (CONH 2 ) EGDMA, N,N ′ - Bisacrylamide BPA, progesterone, mnitrophenol [51,52] N-Vinyl caprolactam Vinyl (CH – – CH 2 ), Lactam ring (cyclic amide) N, N ′ -methylene bisacrylamide Gemcitabine, rhodamine [53,54] A Acrylic acid Vinyl (CH – – CH 2 ), Carboxylic acid (COOH) DVB, EGDMA, TRIM Folic acid, creatinine, diuron, S-ketamine [55–58] Itaconic acid Vinyl (CH 2 – – C–), Two carboxylic acid (COOH) EGDMA, pdivinylbenzene Sulpiride, amphetamine [59,60] Methacrylic acid Methacrylate (C(CH 3 ) – – CH 2 ), Carboxylic acid (COOH) EGDMA, TRIM Erythromycin, ciprofloxacin, 5-fluorouracil [61–63] B Allylamine Allyl (CH 2 –CH – – CH 2 ), Primary amine (NH 2 ) Epichlorohydrin Xylose [64] 2-(N,N-Diethylamino) ethyl methacrylate Methacrylate (C(CH 3 ) – – CH 2 ), Tertiary amine (N (CH 2 CH 3 ) 2 ), Ester linkage (COO) TEGDMA Aflatoxin B1 [65] 1-Vinyl imidazole Vinyl group (CH – – CH 2 ), Imidazole ring EGDMA, pdivinylbenzene BPA, amphetamine [60,66] 2-Vinylpyridine Vinyl (CH – – CH 2 ), Pyridine ring EGDMA Ibuprofen, naproxen, diclofenac, Cu 2+ , ametryn [67–69] 4-Vinylpyridine Vinyl (CH – – CH 2 ), Pyridine ring EGDMA Mandelic acid, TBBPA [70,71] H/philic N 2-Hydroxyethyl methacrylate Methacrylate (C(CH 3 ) – – CH 2 ), Hydroxyl (OH), Ester linkage (COO) EGDMA, TRIM Timolol maleate, uric acid [72,73] Co. N Acrylonitrile Vinyl (CH – – CH 2 ), Nitrile (C – – – N) Sclareol [74] π - π A Uraconic acid ethyl ester Carboxylic acid (COOH), ethyl ester (COOCH 2 CH 3 ) TEGDMA Tyrosinase [75] (A, Acidic; B, Basic; N, Neutral; H/philic, Hydrophilic; Co, Comonomer). Table 2 Commonly used MIP synthesis techniques, resulting MIP morphology, advantages and disadvantages listed based on their level of control. Method Level of control Morphology Advantages Disadvantages Ref FRP Bulk Polymerization Low–moderate. Solid block, irregular particles post-crush Pure polymer, no solvent needed, easy scale-up Gel effect, uneven heat/mass transfer, non-uniform particles [90] Template-Assisted Solvent Evaporation Low-moderate. Depends on evaporation rate, solvent properties, and template interaction Capsule-like structures, porous films Simple and low-cost, templates in core-shell structures, solvent-friendly Time-consuming process, low reproducibility and control, surface can be irregular [91] Precipitation Polymerization Medium. Relies on solubility threshold of polymers Monodisperse microbeads No grinding needed, uniform morphology, scalable Template solubility limits, not suitable for all systems [92] Suspension Polymerization Medium. Controlled dropletbased microreactors Spherical beads (tens to hundreds of microns) Good size control, no grinding, robust beads Surfactants may disturb complexation, phase separation issues [93] Miniemulsion Polymerization Medium–high. Droplet size controls morphology Nanospheres (~50–200 nm), narrow size High surface area, uniform, monodispersed colloidal particles, no grinding Requires surfactants and homogenization, interface effects may affect binding [94–96] LRP RAFT Polymerization High. Narrow molecular weight, homogeneity, tuneable architecture Uniform beads or polymers with controlled porosity Controlled structure, better binding capacity, grafting possible Requires RAFT agent optimization [97] Other Sol–gel Transition Moderate-high. Enables finetuning of porosity, surface area, and chemical composition Amorphous or mesoporous networks with high surface area, monolithic, particulate, or thin-film formats Excellent control over porosity and surface chemistry, high chemical/ thermal stability. Slow processing, possible cracking upon drying, template removal can be difficult due to pore entrapment [86] Surface Imprinting High. Templated sites on particle surfaces Thin bindings layers Fast binding/rebinding, accessible sites, good for sensors & separations More complex synthesis, possible template leakage from support [98–100] Electropolymerization High. Electrochemical control of polymer growth Thin, uniform films on conductive surfaces No initiators or crosslinkers, excellent for sensors, direct integration with electrodes Limited to electropolymerizable monomers, thin films with limited binding capacity [101, 102] M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 6 interactions. For covalently imprinted polymers, mild hydrolysis or other chemical treatments are employed. The efficiency of template removal directly affects the accessibility and functionality of the binding sites, which are essential for the polymer’s performance in subsequent applications. The resulting cavities in the polymer matrix exhibit remarkable molecular memory, allowing them to selectively bind the target molecule from complex mixtures. This selectivity is governed by the spatial arrangement and distribution of recognizing sites, which mimic the original template-monomer interactions (see Table 2). 3.2. Synthesis methods of MIPs The polymerization process itself significantly influences the morphology and performance of the final MIPs. Radical polymerization (RP), initiated by thermal or photochemical methods, is the most common approach. During polymerization, the functional and cross-linking monomers form a highly cross-linked polymer network around the template molecule. Cross-linkers, such as EGDMA, play a dual role by stabilizing the pre-polymerization complex and providing structural rigidity to the polymer matrix. The degree of cross-linking must be optimized to balance the mechanical stability of the MIP with the accessibility of the imprinted cavities [76]. FRP is one of the most convenient methods for polymer synthesis due to its simple setup, minimal requirements for reagent purity, wide variety of polymerizable monomers, and flexible reaction conditions. It is responsible for nearly half of the commercially produced polymers. On the other hand, FRP offers very limited control over the polymerization process. This lack of control is primarily due to side reactions, such as bimolecular termination and chain transfer, that occur during the polymerization. These reactions compete with the propagation step, leading to the formation of inactive polymer chains with a broad range of molecular weights [77]. The heterogeneous structures formed during FRP arise from the significant mismatch between the slow initiation rate of radicals and the rapid propagation and termination steps. This imbalance leads to uneven polymer chain growth, which depends on the local concentration of monomers around the active radicals. Consequently, polymers produced via FRP show a broad molecular weight distribution. Additionally, the high reaction rate and the reaction environment variability cause non-stereospecific addition of vinyl monomers to the growing radical chain. This process produces mostly atactic polymers, whose lack of stereoregularity significantly impacts the mechanical, thermal, and electrical properties of the final material [78]. Molecular imprinting in weakly crosslinked polymers is more effective in heteropolymer systems. A single type of functional monomer often cannot establish optimal interactions with the template molecule, as each functional monomer tends to form energetically favorable interactions with specific chemical groups on the template, depending on its chemistry. During imprinting, the monomers are organized within the growing polymer chains into a low-energy conformation that facilitates multiple-point interactions with the template. Allowing the polymer chains with template-binding complexes to reach a global energy minimum during network formation improves the retention of chain conformations and enhances template binding properties in both highly and weakly crosslinked polymers. Therefore, multiple monomer-template interactions are essential for creating well-structured imprinted polymers [79]. Meanwhile, using multiple monomers to improve template binding introduces additional heterogeneity to the system, complicating polymer formation. In FRP, the carbon radical is highly reactive toward vinyl monomers, leading to rapid propagation and polymer chain growth. However, this same reactivity causes rapid termination through combination or disproportionation, particularly at high monomer conversions. Chain transfer reactions, where the radical abstracts hydrogen from polymers or solvents, further result in undesirable branching. Unstabilized monomers create even more reactive radicals, increasing the likelihood of head-to-head propagation. The presence of monomers with varying reactivities adds complexity, making polymer control more challenging. While network formation can limit the impact of these factors as mobility decreases near the gelation point, heterogeneity remains a significant challenge in achieving precise polymer structures [78]. Conventional FRP (Fig. 4a) is far from being ideal, with discrepancies between theoretical predictions and experimental data pointing to heterogeneity within network structures [80,81]. In contrast, living/- controlled RP (LRP) offers greater control over polymer chain growth through a fast, reversible activation and deactivation process of the reactive species. Amongst various LRP techniques, Reversible Addition Fragmentation Chain Transfer (RAFT) polymerization (Fig. 4b) is particularly popular due to its use of reversible chain transfer agents, which stabilize intermediates and establish a dynamic equilibrium between active and dormant species. The photosensitivity of chain transfer agents in RAFT further enhances control over the reaction [82]. RAFT polymerization is highly versatile, accommodating a wide range of monomers and reaction conditions. This improved control allows living polymerization techniques like RAFT to be combined with other optimized reaction conditions to produce more homogeneous polymer networks with enhanced binding and transport properties [78]. Noteworthy, conventional FRP encompasses several polymerization formats, including bulk polymerization (monomer and initiator without solvent), solution polymerization (involving a solvent phase), suspension polymerization (monomer droplets containing initiator dispersed in water), and emulsion polymerization (where an aqueous initiator polymerizes hydrophobic monomer droplets stabilized by surfactants). Bulk polymerization (Fig. 4c) is advantageous for producing large quantities of MIPs with high mechanical strength, but it often requires extensive grinding and sieving to achieve uniform particle size. Precipitation polymerization, on the other hand, provides MIPs with welldefined spherical morphology and eliminates the need for postsynthesis processing, though it may result in lower yields. Miniemulsion polymerization (Fig. 4d), a technique employed rather commonly in the world of MIPs, involves the creation of submicron oilin-water dispersions, referred to as miniemulsions, that can remain stable for months [83]. These monomer-in-water dispersions are stabilized against droplet coagulation and diffusional degradation by incorporating a low-molecular-weight, water-insoluble co-stabilizer and an effective surfactant. The formulation and preparation method are critical in creating stable miniemulsions. A typical formulation includes water, monomers, a co-stabilizer, surfactants, and an initiator. Surfactants used in miniemulsion polymerization must meet specific criteria: they should have a polar and non-polar structure, be soluble in water to ensure adsorption on droplet surfaces, adsorb strongly to resist displacement during droplet collisions, provide sufficient electrokinetic stability, work effectively at low concentrations. While non-ionic surfactants are less effective in conventional emulsion polymerization due to their dissolution in monomer droplets, miniemulsion polymerization enhances their efficiency. This is because the initial surface area of the miniemulsion droplets is significantly larger, increasing the proportion of surfactant available at the droplet-water interface. Additionally, droplet nucleation reduces the need for surfactant mass transfer. The preparation process involves dissolving the surfactant in water and the co-stabilizer in the monomer, followed by mixing and homogenization. The resulting miniemulsions typically undergo polymerization reaction for a certain time interval. However, droplet size may increase during this step due to two primary mechanisms: (i) droplet coalescence caused by collisions from Brownian motion and van der Waals forces, which depend on the surfactant system, and (ii) droplet degradation due to monomer diffusion, often the dominant factor [83]. Precipitation polymerization is a widely used technique for the preparation of MIPs, particularly valued for producing uniform, monodisperse microspheres without the need for mechanical grinding or sieving. In this method, the functional monomer, cross-linking monomer, template, and initiator are initially dissolved in a porogenic solvent such as acetonitrile. As polymerization proceeds, growing polymer M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 7 chains become insoluble and precipitate out of solution, forming spherical particles. This approach enables the generation of MIPs with high surface area and accessible binding sites, improving mass transfer and template recognition. The technique is simple, reproducible, and suitable for a wide range of monomers and templates, although it requires that the template be soluble in the reaction medium and is typically limited to relatively low template concentrations to avoid interfering phase separation. Overall, precipitation polymerization offers a robust and scalable route for producing MIP microspheres with well-defined morphology and enhanced binding performance [84]. Suspension polymerization is a heterogeneous polymerization technique employed for the synthesis of spherical MIP beads, particularly suitable for large-scale applications. In this method, the organic phase, comprising functional monomer, cross-linking monomer, initiator, and template, is dispersed as droplets in an aqueous phase containing a stabilizer, typically under mechanical stirring. Polymerization occurs within each droplet, resulting in well-defined microspheres with relatively uniform size distribution and improved flow properties. This method eliminates the need for post-polymerization grinding and allows for facile recovery of the product by filtration. Compared to bulk polymerization, suspension polymerization offers advantages such as better control over particle morphology, higher mechanical stability of the beads, and enhanced suitability for column applications. However, stabilizers or surfactants may interfere with the formation of template–monomer complexes if not carefully chosen, and phase separation during polymerization can impact particle uniformity. Despite these considerations, suspension polymerization remains a valuable approach for preparing MIPs with controlled size, shape, and surface characteristics [85]. The sol–gel transition is a widely used method in the synthesis of MIPs, particularly when aiming for materials with stable and porous structures. In this method, organosilane compounds (typically metal alkoxides like tetraethyl orthosilicate (TEOS) or tetramethyl orthosilicate (TMOS) are mixed in a low molecular weight solvent along with a catalyst, commonly an acid, base, or ions like fluoride. This mixture undergoes hydrolysis and subsequent condensation reactions, leading to the formation of a rigid, porous silica network. Sol–gel chemistry offers a simple, cost-effective, and versatile route to design MIPs with high structural integrity. Compared to traditional polymerization routes, sol–gel-based MIPs demonstrate better mechanical and thermal stability, which is particularly advantageous for complex sample matrices like biological fluids. Common challenges in conventional MIP use, such as polymer swelling, structural collapse, and pore blockage, can be significantly minimized through the sol–gel transition method. Another major benefit of this method is its ability to facilitate template removal. Since silica-based materials tolerate high temperatures, elevated conditions can be used to efficiently eliminate the template molecule without degrading the polymer framework. Furthermore, silica-derived MIPs typically exhibit enhanced porosity and surface area, overcoming two of the major limitations seen in conventional MIPs [86,87]. Surface molecular imprinting (SMI) (Fig. 4e) involves conducting polymerization reactions on the surface of solid-phase matrices, resulting in recognition sites distributed on the outer layer or surface of the substrate. The technique is ideal for creating thin MIP layers on substrates, enhancing accessibility to binding sites and improving kinetics, but it can be challenging to maintain the structural integrity of the imprint under operational conditions. Unlike traditional molecular imprinting methods, SMI offers higher separation efficiency, faster binding kinetics, and minimizes issues like template embedding. The preparation of SMI polymers typically involves three key steps: (i) template-monomer complex formation where the template molecules interact with functional monomers through covalent, non-covalent, or semi-covalent bonding to form stable complexes under specific conditions, (ii) surface polymerization which occurs on the surface of the solid substrate in the presence of initiators and cross-linkers, creating an imprinting layer that incorporates the templates, and (iii) template removal where the templates are removed using physical or chemical methods, leaving three-dimensional cavities on the substrate’s surface that are complementary in shape, size and functionality to the template molecules. These imprinted cavities on the polymer surface enable selective recognition via binding of the analyte molecules from complex samples, facilitating efficient separation and detection. The SMI process is straightforward, rapid, and practical, making it a valuable tool for applications requiring precise molecular recognition [88]. Electropolymerization is an electrochemically driven technique for synthesizing MIPs, offering exceptional control over film thickness, morphology, and deposition site. In this method, a mixture containing the functional monomer and template is contacted with an electrode surface, where polymerization is initiated by an applied potential, typically under potentiodynamic or potentiostatic conditions. This localized polymer growth enables the formation of uniform thin films directly adhering to conductive substrates, eliminating the need for Fig. 4. Schematic representation of a) FRP mechanism, which serves as the underlying chemistry for most traditional methods, b) RAFT polymerization as a controlled radical polymerization, c) Bulk polymerization as a conventional FRP-based method, d) Miniemulsion polymerization format based on radical-initiated droplets, and e) Surface imprinting strategy, where binding sites are localized on or near the surface of a support material. M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 8 initiators or cross-linking monomers. The method is particularly wellsuited for sensor applications, as it facilitates the creation of highly accessible recognition sites and enhances the direct transduction of binding events into electrical signals. Electropolymerized MIPs exhibit excellent reproducibility, fast fabrication, and compatibility with miniaturized devices. However, the approach is generally limited to electropolymerizable monomers e.g., pyrrole (Py), aniline, phenol derivatives and to relatively thin films, if those are non-conductive which may constrain the binding capacity. Despite these limitations, electropolymerization remains a powerful and versatile tool for preparing MIP-based electrochemical sensors with high sensitivity and selectivity [89]. 3.3. Characterization of MIPs Characterization of MIPs is a critical step in understanding their structural, morphological, and functional properties. It provides insights into the efficiency of the imprinting process and the performance of the resulting MIPs in selective recognition and binding of target analytes. Various analytical and spectroscopic techniques are employed to assess key parameters such as binding capacity, surface morphology, chemical composition, and thermal stability. One of the fundamental characterization methods is Fourier transform infrared (FTIR) spectroscopy, which is used to identify functional groups and confirm the presence of template-monomer interactions [103]. FTIR spectroscopy analysis before and after template removal reveals changes in the polymer structure, indicating successful imprinting. In a study by Shafqat et al. MIPs were synthesized for the selective removal of methyl red (MR) from aqueous media [104]. FTIR spectroscopy analysis was employed to evaluate the chemical composition of MR-MIPs and confirm imprinting and then template removal. Spectra were recorded for MR-MIPs with varying EGDMA crosslinker ratios, both before and after leaching of the MR dye, as well as for the non-imprinted polymers (NIPs). The spectra showed that the key bands, notably those at 1730 cm −1 (C – – O stretching) and 1160 cm −1 (C–O stretching), were consistently observed across all samples, confirming the presence of EGDMA in the polymer backbone. Bands at 1452 cm −1 (CH 2 scissoring) and other bands in the fingerprint region further supported EGDMA’s incorporation. A distinct difference was observed in –OH stretching bands associated with acrylic acid (AA) and hydrogen bonding with MR. In the unleached MR-MIPs, the –OH stretching bands at 3564 cm −1 and 1157 cm −1 appeared broader and more intense, indicating the formation of intermolecular hydrogen bonds between AA and MR. After leaching the template, these –OH bands shifted (e.g., from 3564 cm −1 to approximately 3454 cm −1 ), and the intensity of several bands decreased, reflecting the removal of the dye. In contrast, the NIP spectra showed all the same bands but with overall lower intensity, supporting the absence of template interactions. Scanning electron microscopy (SEM) and Transmission electron microscopy (TEM) are widely used to study the surface morphology and structural features of MIPs [105]. SEM provides detailed images of the polymer surface, enabling the analysis of particle size, porosity, and uniformity. TEM, on the other hand, offers high-resolution insights into the internal structure of the polymer matrix. These techniques are particularly useful in surface-imprinted MIPs, where the accessibility of binding sites plays a crucial role in performance. Wang et al. devised MIP microspheres with an eccentric hollow structure for selective recognition of bisphenol A (BPA) [95]. TEM analysis revealed that both MIP and NIP microspheres exhibit a bowl-like morphology with an average diameter of around 210 nm. A distinctive light spot with a thinner shell thickness suggested a hollow interior, displaced from the center, indicating an eccentric hollow structure. Additionally, a dimple-like concave on the microsphere surfaces further confirmed this asymmetry. SEM images, taken after gold sputtering, showed that both MIP and NIP microspheres had smooth surfaces and similarly displayed the dimpled bowl-like morphology, supporting the TEM observations. Importantly, the morphological features were consistent across MIP and NIP, indicating that the hollow structure was not a result of template involvement but inherent to the polymerization design. Nitrogen adsorption-desorption analysis is employed to determine the surface area, pore size distribution, and porosity of MIPs using the Brunauer-Emmett-Teller (BET) method [106]. BET theory is utilized to determine the surface area of solid or porous materials. This measurement provides valuable insights into the material’s physical structure, as the surface area plays a crucial role in its interactions with the surrounding environment. A high surface area and appropriate pore size distribution are critical for maximizing the binding capacity and enhancing the accessibility of target molecules. Che Lah et al. developed a MIP-based sensor for the determination of atrazine in water, using MAA and EGDMA in various porogenic solvent systems [107]. Their goal was to evaluate how different porogen compositions, particularly varying the ratio of dimethyl sulfoxide (DMSO) and toluene, influence the polymer’s selectivity, structural stability, and sensor performance. Focusing on the BET surface analysis, the authors investigated how changes in DMSO concentration affect the MIP’s surface area, pore volume, and pore diameter. In their study, no consistent trend was observed with increasing DMSO content. For example, the specific surface area initially decreased from 237.5 m 2 /g at 10 % DMSO to 126.6 m 2 /g at 50 % DMSO but then increased again to 218.2 m 2 /g at 75 % DMSO. Similarly, pore volume and pore diameter did not follow a clear progression, indicating that the solubility parameter differences between MAA and the porogen mixtures (DMSO:toluene) were not the dominant factor influencing pore structure. These findings suggest that toluene played a more significant role in forming porous structures favorable for molecular recognition, supporting previous reports that minimal DMSO with predominantly toluene yields better imprinting and selectivity due to improved template–monomer interactions and colloidal stability during polymerization. Binding capacity and selectivity studies are critical for evaluating the performance of MIPs. These studies typically involve equilibrium binding assays, where the MIP is exposed to a solution containing the target analyte at varying concentrations. The amount of analyte bound by the MIP is determined using techniques such as high-performance liquid chromatography (HPLC) or spectrophotometry. Selectivity is assessed by comparing the binding affinity of the MIP for the target analyte against structurally similar compounds. For example, in a study by Thongchai et al. MIPs were developed for the selective extraction and detection of andrographolides [108]. To evaluate binding performance, the researchers applied a solid-phase extraction protocol using the synthesized MIPs, followed by quantification with HPLC. The MIPs showed a significant increase in retention and peak intensity for andrographolides compared to NIPs, especially in complex biological matrices such as human urine. After spiking urine samples with andrographolide standards, the MIP-treated extracts produced well-defined chromatographic peaks with high sensitivity. Recovery rates for the MIP ranged between 91 % and 94 %, confirming both high binding capacity and excellent selectivity under real sample conditions. These HPLC results underscored the effectiveness of the MIP in selectively concentrating the target analyte, even in challenging biological media. Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) are used to evaluate the thermal stability of MIPs. TGA measures the weight loss of the polymer as a function of temperature, providing insights into its decomposition behaviour. DSC, on the other hand, identifies thermal transitions such as glass transition and melting points. These techniques ensure the stability of MIPs under operating conditions, particularly in applications involving high temperatures or harsh environments. For instance, Mohebali et al. investigated the thermal behaviour of amitriptyline-imprinted polymers using both TGA and DSC analyses [109]. TGA data revealed that MIPs exhibited two distinct weight loss events: an initial degradation between 125 and 155 ◦C corresponding to the release of template drug molecules, and a major M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 9 photoresist was spin-coated to a thickness of 50 μ m and patterned via UV lithography. PDMS was prepared by mixing the base and curing agent in a 10:1 w/w ratio, degassed with N 2 for 40 min, poured onto the mold, and cured at 70 ◦C for 90 min. Once cured and demoulded, the PDMS layer was aligned and bonded to the sensing substrate using oxygen plasma treatment. 5. Applications for MIP-based microfluidic sensors This section contains examples of MIP based microfluidic sensor applications in biomedical, environmental monitoring and food safety and chemical analysis. 5.1. Clinical applications Li et al. created a MOF/MIPs paper-based microfluidic biomimetic chip (MM-SGM-chip) that utilizes metal-organic frameworks (MOFs) paired with MIPs to enable rapid and sensitive glucose determination in saliva [7]. A visual and sensitive glucose determination method was developed by integrating an enzyme-assisted colorimetric sensing platform. The sensor mimicked biological recognition by integrating ZIF-8 MOF particles and a glucose-imprinted polymer layer, stabilized by boron affinity interactions, onto patterned paper channels. Upon exposure to saliva, glucose was selectively captured by the MOF/MIP composite, triggering a colorimetric signal measured by a portable reader. The MM-SGM chips enabled fast and sensitive visual quantification of glucose concentrations between 0.1 and 3.2 mM, achieving an LOD down to 0.02 mM. In addition, the MM-SGM-chip executes quantitative analysis in under 30 min, with notable selectivity against common interferences (metal ions, fructose, and sucrose) and minimal effects (3.8 %–12.8 %) from saliva components. Regeneration tests showed that the sensor retained an adsorption capacity of 167.89 mg g −1 after five reuse cycles, with only a 10.2 % reduction. The slight decline was attributed to material loss or residual glucose in binding sites, indicating the sensor’s stable performance over multiple uses. The chip was optimized in the lab for trace glucose detection and validated through spiked and artificial Fig. 7. Schematic illustrations of a) Microfluidic saliva sensing platform utilizing MIPs embedded in dental floss for non-invasive salivary cortisol determination, redrawn based on an illustration from Sharma et al. [6] and b) wearable MIP-based electrochemical sensor integrated with a nanofiber-based microfluidic chip for the real-time determination of cortisol in human sweat, redrawn based on an illustration from Mei et al. [24]. (FCE, flexible carbon electrode; PDMS, polydimethylsiloxane; PLEG, poriferous laser-engraved graphene electrode; PET, polyethylene terephthalate; PBNP, prussian blue nanoparticle; GNP, gold nanoparticle; CNFM, carbon nanofiber membrane; RE, reference electrode; WE, working electrode; and CE, counter electrode). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 16 saliva tests. It was then applied to real diabetic saliva samples to demonstrate feasibility. Sharma et al. introduced an innovative saliva sensing platform utilizing MIPs embedded in dental floss for non-invasive salivary cortisol determination, a crucial biomarker for stress assessment (Fig. 7a) [6]. In this study, cortisol-imprinted MIPs are electropolymerized directly onto a porous laser-engraved graphene electrode, creating selective binding sites that mimic the size and shape of cortisol molecules. Users can floss normally, capillary action pulls saliva through microfluidic threads into the device, allowing cortisol to bind to the MIP sites, which modifies the electrochemical signal. Results are wirelessly transmitted to a mobile device in about 11–12 min. From an analytical perspective, the system exhibits outstanding performance: it offers a determination range of 0.10–10.000 pg mL −1 , a LOD of ~0.048 pg mL −1 , and remarkable linearity (R 2 =0.9916). The accuracy is impressive, with recovery rates between 98.6 and 102.4 % and a relative standard deviation (RSD) of approximately 5 %. Additionally, tests on human saliva show a strong correlation with conventional ELISA (R =0.991). This study exemplifies the growing trend of incorporating MIPs into wearable and point-of-need analytical systems, particularly in formats that enable autonomous sample collection. Significantly, the platform embodies the “MIP-on-the-flow” concept by merging passive sampling, selective molecular recognition, and signal generation within a compact flow setup. The dental floss design is disposable, presenting a new physical design format for MIP-based biosensors that could be applied to other salivary or oral biomarkers. Mei et al. developed a wearable MIP-based electrochemical sensor integrated with a nanofiber-based microfluidic chip for the real-time, non-invasive determination of cortisol in human sweat [24] (Fig. 7b). The sensor fabrication began with laser-cutting a flexible PET substrate, onto which a three-electrode system was screen-printed, this configuration was termed flexible carbon electrodes (FCE). 3D carbon nanofiber membrane decorated with gold nanoparticles (GnPs@CnFM) was then applied to the FCE surface and dried, forming the GnPs@CnFM/FCE structure. Subsequently, a MIP incorporating Prussian Blue nanoparticles (PBnPs@MIP) was formed on this structure via electropolymerization. The potentiodynamic process was carried out between −0.2 V and +0.9 V at 50 mV/s for 10 cycles. Cortisol was then removed by further potentiodynamic scanning (−0.2 to +0.8 V, 50 mV/s, 30 cycles). Polyimide/sodium dodecyl benzene sulfonate (PI/SDS) nanofiber membranes were electrospun directly onto the PBnPs@MIP-modified FCE. A PDMS ink (10:1 ratio of base to curing agent) was stirred for 30 min and refrigerated at 4 ◦C for 36 h before being directly written onto the nanofiber layer. After curing, a complete microfluidic electrochemical sensor integrated with nanofiber structures was obtained. The developed electrochemical sensor was used to detect cortisol under optimal conditions. Chronoamperometry at 0.15 V showed a clear drop in current as cortisol concentration increased, with a linear response between 1.0 and 1.0 ×10 3 nM (R 2 =0.993), and a LOD 0.35 nM. The sensor exhibited high repeatability (RSD =3.56 %), reproducibility (RSD =4.87 % across five sensors), and long-term stability (97.8 % signal retention over 20 days at 4 ◦C). Recovery tests in artificial sweat yielded 95.3–104.6 % with RSD <5 %, confirming accuracy. Khachornsakkul et al. introduced an innovative distance-based paper analytical device (dPAD) that merges MIPs with fluorescent carbon dots Fig. 8. Distances-based paper analytical device (dPAD) merging MIPs with fluorescent CDs for the instrument-free measurement of cytokine biomarkers, a) dPAD fabrication, b) The principle and workflow. Redrawn based on an illustration from Khachornsakkul et al. [12]. (CDs, carbon dots; CRP, C-reactive protein; TNFα , tumor necrosis factor-alpha; IL-6, interleukin-6). M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 17 (CDs) for the concurrent, instrument-free measurement of cytokine biomarkers, C-reactive protein (CRP), tumor necrosis factor-alpha (TNFα ), and interleukin-6 (IL-6), present in human samples (Fig. 8) [12]. Quantitative evaluation can be performed simply by determining the quenched fluorescence distance using a standard ruler and direct visual inspection, allowing for quick identification of cytokine syndrome and its associated infection. In this microfluidic design, dopamine-derived MIPs were created by surface imprinting directly onto the detection zones of paper channels, while CDs functioned as optical transducers with fluorescence that diminishes upon target attachment via photo-induced electron transfer (PiET). This fluorescence reduction travels along the flow path, allowing for a quantitative distance measurement using just a ruler and UV light, thus removing the necessity for intricate instrumentation. Remarkably, the sensor achieved detection limits as low as 0.25–2.5 pg mL −1 , demonstrated high selectivity compared to non-imprinted controls, and showed excellent reproducibility (RSD <5.14 %). The design incorporates passive capillary flow, synthetic receptor chemistry, and visual quantification, showcasing the main principles of MIP-on-the-flow sensing. Furthermore, the sensor proved effective across various biological matrices such as serum, urine, saliva, and sweat, with recovery percentages ranging from 99.22 to 103.58 %, and it was reusable up to five times without diminishing performance. In contrast to antibody-based immunosensors, this MIP-CD hybrid system provides enhanced stability, simplicity, and scalability for POC diagnostics. From a critical standpoint, although the optical distance readout is innovative and user-friendly, its resolution and multiplexing capabilities may be limited. Haghgouei and Alizadeh describe the development of a paper-based LoC device that combines electrochemically assisted solid-phase microextraction (EC-SPME) with a potentiometric ion-selective electrode (ISE) for quick on-site naproxen determination in biological fluids [178]. The platform utilizes conductive molecularly imprinted polymer (CMIP) films, formed by electropolymerizing polypyrrole in the presence of naproxen, which create selective imprinted cavities featuring recognition sites. This device consists of two integrated chambers: the first allows for selective EC-SPME cleanup, while the second contains a naproxen-ISE for potentiometric determination. The analytical evaluation demonstrated high selectivity, exhibiting a linear dynamic concentration range of 4 ×10 −7 to 1 ×10 −2 M. The LOD was established at 2.0 ×10 −7 M, with a short response time of approximately 7 s. When tested on actual saliva and serum samples, the LOC device showcased impressive recovery rates (90–110 %) and maintained high precision (RSDs of ≤7 %). In this study, EC-SPME improves selectivity by removing interferents, while the integrated naproxen-ISE offers rapid, portable quantitation. The fully portable LOC-PAD system represents a promising tool for on-site drug monitoring in limited sample volumes, without relying on traditional laboratory equipment. Garg et al. discuss a wearable and label-free electrochemical sensor tailored for real-time cortisol monitoring in sweat [179]. This device features a cortisol-MIP applied to a laser-induced graphene (LIG) electrode, facilitating electrochemical impedance spectroscopy (EIS) as the method for transduction. It demonstrates high detectability, detecting cortisol levels as low as 1 pM. A distinctive aspect of the system is its combination of iontophoresis-based sweat induction with paper microfluidic channels. It enables the simultaneous measurement of sweat volume, secretion rate, sodium ion concentration, and cortisol levels, all achieved without labels or redox probes. The paper-based fluidics facilitate passive, continuous collection and efficient analyte delivery to the sensing zone. This work integrates synthetic receptor selectivity, microfluidic sampling, and portable electrochemical determination into a wearable format. As a proof of concept, the system accurately monitors sweat biomarkers with minimal user intervention and shows potential for expansion to other analytes, including proteins or therapeutics. 5.2. Environmental monitoring Amin et al. developed an innovative capacitive MIP sensor designed for the ultra-sensitive and rapid determination of perfluorooctanesulfonic acid (PFOS), a toxic and persistent perand polyfluoroalkyl substance (PFAS). This sensor integrates label-free capacitive transduction with alternating current electrokinetic (ACEK) acceleration, significantly improving point-of-need environmental analysis [8]. This sensor’s integration with alternating AC electrothermal (ACET) enrichment sets this sensor apart, which facilitates rapid preconcentration of PFOS near the sensor surface by inducing micro-scale fluid flow through localized heating. The main sensing platform features interdigitated gold microelectrodes coated with a MIP layer, which is prepared by electropolymerization using o-phenylenediamine as the functional monomer and PFOS as the template molecule. After removing the template, imprinted cavities featuring high-affinity recognition sites selective for PFOS. The main sensing platform features interdigitated gold microelectrodes covered with a MIP layer. This layer is electropolymerized using o-phenylenediamine as the functional monomer and PFOS as the template molecule. After removing the template, the MIP offers high-affinity binding sites selectively for PFOS. The sensor measures interfacial capacitance changes upon PFOS binding, achieving a remarkably low LOD of 0.5 fg L −1 and a wide linear dynamic concentration range of 0.5–500 fg L −1 . The total detection time is less than 10 s, and the sensor demonstrated strong selectivity over structurally similar PFAS analogs. The sensor was tested in tap water in real sample applications. Spike-and-recovery experiments yielded high recovery rates, and these results underscore the platform’s potential for rapid and ultra-trace PFOS determination in real-world environmental monitoring scenarios. Doostmohammadi et al. developed a portable cell-imprinted polymer (CIP)-based microfluidic sensor for the rapid determination of E. coli in real water samples [29]. A custom enclosure was fabricated using 3D printing with poly (methyl methacrylate) (PMMA) to create a robust, portable platform for real-world use. This structure was designed to accommodate the microfluidic device, fluid handling system, and a fluorescence microscopy unit, enabling efficient integration and operation. The core microfluidic system was built using a monolayer PDMS-on-glass structure, featuring a straight 5 cm-long, 1 mm-wide, and 400 μ m-deep channel. Within the central section of the channel, seven oval iron-PDMS soft ferromagnetic microstructures were embedded to trap cell imprinted polymer-magnetic nanoparticle complex (CIP-MNPs) and facilitate fluorescence-based bacterial determination. The device included three inlets for loading CIP-MNPs, introducing the bacterial solution, and injecting deionized (DI) water or washing buffer, and one outlet. The microchannel mold was produced via photolithography. PDMS and iron-PDMS composite were cast and cured on the mold, then bonded to glass. External magnets were affixed to ensure functionality and alignment. CIP-MNPs were synthesized by mixing acrylamide, MAA, methyl methacrylate (MMA), N-vinylpyrrolidone, crosslinker DHEBA, and DMSO with fluorescent magnetic polystyrene microparticles and E. coli. After polymerization, the particles were magnetically separated and washed with a methanol-acetic acid solution (9:1 v/v) for 30 s to remove the bacteria templates. The E. coli cells were confirmed dead but structurally intact, allowing them to serve as effective templates. To evaluate the sensor’s real-world applicability, they tested a pond water sample for bacterial content and compared the results. The total bacterial load in the fresh pond water was determined to be 2.33 ×10 6 CFU/mL, while the total viable count was 1.91 ×10 5 CFU/mL. They successfully identified 15 bacterial species in the non-enriched pond water. In another study, Doostmohammadi et al. developed a portable microfluidic fluorescence sensor that incorporates a CIP-coated platform for the selective determination of E. coli in real water samples [29]. The CIPs were synthesized by polymerizing a mixture of methyl MMA, MAA, acrylamide, N-vinylpyrrolidone, and dihydroxy ethylene-bisacrylamide M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 18 in the presence of E. coli cells as templates. After removing the template bacteria, highly selective recognition cavities were formed on the surface. CIP-MNPs, imprinted with E. coli, capture the target cells as the sample flows at an optimized rate of 0.01 mL min −1 . Fluorescent imaging conducted before and after incubation allows for quantitative analysis based on changes in fluorescence intensity. The sensor showed a LOD of 8 ×10 3 CFU mL −1 and a limit of quantification (LOQ) of 6 ×10 5 CFU mL −1 , with a dynamic range from 10 3 to 10 9 CFU mL −1 . Selectivity tests demonstrated that E. coli binds selectively compared to other bacteria like Salmonella and Sarcina. During practical validation, the sensor detected E. coli in a natural pond water sample, providing a measurement of 2 ×10 6 CFU mL −1 , which closely aligned with the laboratory testing result of 2.33 ×10 6 CFU mL −1 . These results emphasize the sensor’s capability for on-site, point of environmental monitoring, achieving a blend of selectivity, speed, and portability while maintaining analytical accuracy. Kamat et al. (Fig. 7b) developed a sensor-integrated microfluidic chip for real-time monitoring of nitrate and phosphate uptake in plants, with the goal of advancing precision agriculture and soil nutrient analysis [13]. The fabrication process of the microfluidic chip as well as the synthesis route for the MIPs were detailed under section 3.2. The system enabled continuous, non-invasive nutrient tracking while maintaining healthy root and shoot growth over seven days. The sensors demonstrated strong linearity (R 2 =0.99 for nitrate, 0.95 for phosphate) and a broad linear dynamic concentration range (1–1000 mM), with high selectivity against interfering ions. Nutrient uptake analysis showed preferential nitrate absorption (~700 μ M vs. 30 μ M phosphate in 3.5 days). Phenotypic comparisons confirmed that microfluidic-grown plants had 10 % higher germination rates and similar nitrogen, chlorophyll, and sugar levels as conventionally grown plants, validating the chip’s suitability for plant nutrition studies. 5.3. Food safety Nagabooshanam et al. developed a microfluidic affinity sensor based on a MIP thin film for the ultrasensitive determination of chlorpyrifos (CPF), an organophosphate pesticide (Fig. 9a) [9]. Pyrrole was electropolymerized with CPF on gold microelectrodes (Auμ E) via CV (−0.4 V to +1.8 V, 50 mV/s, 15 cycles) to form the MIP layer, followed by template removal with 50:50 ethanol-water solution for 8 min to create selective binding sites. A PDMS microfluidic chip, fabricated by soft lithography and bonded to glass, allowed low-volume (2 μ L) sample handling and integration with the sensor. The sensor’s performance was assessed using differential pulse voltammetry (DPV). As CPF concentration increased from 1 fM to 1 μ M, the peak current decreased from 3.9 to 0.27 μ A, successful binding of the CPF analyte into the PPy cavities, which hindered redox probe access. The interaction is primarily due to hydrogen bonding between the pyridine nitrogen in CPF and the N–H groups in PPy. The sensor exhibited LOD of 0.93 fM, LOQ of 2.82 fM and sensitivity of 3.98 μ A/( μ M mm 2 ). To assess selectivity, the sensor was exposed to 1 nM of CPF, parathion, malathion, and monocrotophos. Only CPF caused a significant decrease in DPV peak current, confirming the selective binding of CPF at the imprinted cavities. Feng et al. developed an inertial microfluidic-assisted molecularly imprinted quantum dot (QDs@MIPs) sensor for the rapid determination of enrofloxacin, an antibiotic contaminant in food matrices (Fig. 9b) [10]. Fluorescent CdSe 1-x S x /ZnS QDs were embedded into a MIP shell formed on silica microspheres using APTES and TEOS, with enrofloxacin as the template. After polymerization and template removal, the QDs@MIPs exhibited selective fluorescence quenching upon target Fig. 9. Schematic illustrations of a) Microfluidic affinity sensor based on a MIP thin film for the ultrasensitive determination of chlorpyrifos, redrawn based on an illustration from Nagabooshanam et al. [9], b) Microfluidic-assisted QDs@MIPs sensor for the rapid detection of enrofloxacin, redrawn based on an illustration from Feng et al. [10]. (DPV, differential pulse voltammetry; CPF, chlorpyrifos; Auμ E, gold microelectrode; Py, pyrrole; F L , lift force and F P , dean drag force). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.) M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 19 binding. A spiral-channel PDMS-based microfluidic chip enabled size-based separation of QDs@MIPs from complex matrices using inertial and Dean flow forces. The device consisted of three connected spiral microchannels, two outer inlets (for injecting the sample and QDs@MIPs suspension), an inner inlet (for buffer solution), a five-loop spiral channel, and two outlets. The microchannel was fabricated via soft lithography, using a PDMS mold bonded onto a glass substrate. The spiral microchannel had a width of 500 μ m, a spacing of 1 mm between loops, and a height of 130 μ m, ensuring optimal fluid flow dynamics. The platform achieved a detection range of 0.1–50 μ g mL −1 and an LOD of 0.059 μ g/mL, below the 0.1 μ g/mL regulatory limit, along with a high separation efficiency (97.7 % ±1.6 %). Validation with swine muscle samples showed strong agreement with HPLC results, demonstrating its suitability for food safety monitoring. Azhdary et al. developed a highly selective microfluidic gas sensor for tetrahydrocannabinol (THC) determination using MIP nanoparticles (NPs) (Fig. 10) [196]. The MIP NPs were synthesized via precipitation polymerization using MAA as the functional monomer and EGDMA as the crosslinker in the presence of THC as the template. After polymerization at 60–65 ◦C for 15 h, the template was removed using solvent extraction and sonication to create selective recognition cavities for THC. The microfluidic platform was fabricated using a 3D-printed microchannel coated with Parylene C, and the MIP NPs were immobilized on the microchannel surface via drop-casting. A second channel was coated with NIP NPs as a control. The sensor’s performance was evaluated by exposing it to THC, cannabidiol (CBD), methanol, and ethanol vapors at 300–700 ppm at 300 ◦C. The MIP-coated channel showed strong selectivity toward THC, with an 8.48 % relative response at 500 ppm and a recovery time of 555 s. The classification model achieved 96.3 % accuracy in distinguishing THC from other analytes, demonstrating the platform’s potential for selective and reliable THC determination. The sensor’s response to 600 ppm of THC, CBD, methanol, and ethanol revealed that the MIP channel displayed strong selectivity toward THC, as indicated by a modest resistance drop and longer recovery time, reflecting stronger analyte-cavity interactions. In contrast, CBD showed moderate interaction, while methanol and ethanol produced minimal responses due to their small size and weak binding. Repeatability tests across four cycles of THC exposure confirmed full recovery and consistent performance. In dynamic response experiments, the MIP channel showed stable responses to THC of increasing concentrations, while the NIP channel’s response increased linearly, highlighting the selectivity of the imprinted sites. When tested with 300–700 ppm analyte concentrations, only THC triggered a distinctly different response curve in the MIP channel, with a smaller change in resistance but longer recovery times, demonstrating strong and selective binding. In contrast, methanol and ethanol induced the least change, supporting their weak interaction with the MIP. At 500 ppm, MIP responses showed: THC: 8.48 % response (78 % lower than NIP), 555 s recovery time (46 % longer), CBD: 19.43 % response, 453 s recovery and methanol/ethanol: ~31 % response, ~290–353 s recovery. Hua et al. created a reusable microfluidic LOC sensor designed for the quick, on-site detection of zearalenone, a mycotoxin frequently present in food items (Fig. 11) [11]. This system integrates the molecular selectivity offered by MIPs with the fluorescence quenching properties of quantum dots, enabling precise determination without requiring sophisticated instrumentation. A dummy template strategy was implemented to synthesize MIPs. PDMS chips were fabricated using standard soft lithography with modifications tailored to available equipment. PDMS and glass layers were plasma-bonded, and fluidic access points were created using biopsy punches, connected via PVC tubing. To evaluate real-sample performance, zearalenone was spiked into corn powder and whole-wheat flour at 1, 5, and 10 mg/kg, in line with regulatory limits. The microfluidic chip method demonstrated high recovery rates (91–105 %) and RSDs of 6–11 %, closely matching the performance of standard HPLC-FLD. Notably, the chip method avoided analyte loss typical of HPLC’s SPE clean-up step. The chip could detect zearalenone at levels as low as 0.25 mg/kg, suitable for regulatory thresholds. While it is less sensitive than LC–MS, it matched prior studies using MIP-based determination for toxicants and biomarkers. The full assay took 20 min, including only 3 min of manual work, making it significantly faster and more user-friendly than LC–MS or HPLC. The platform operated without specialized training, pipetting, or complex lab setups. It also relied on portable, battery-powered instruments. Fig. 10. Microfluidic gas sensor for THC determination using MIP nanoparticles, redrawn based on an illustration from Azhdary et al. [196]. M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 20 6. Towards a future of precision and miniaturization 6.1. Challenges and mitigation strategies Microfluidic systems provide unique benefits for miniaturized, highthroughput, and POC applications, however, their development faces several technical challenges. One main difficulty is integrating sensors within confined microchannel structures, where ensuring leak-proof bonding and maintaining sensor performance without disrupting fluid flow are essential concerns. Material compatibility also presents significant obstacles: adhesives or bonding methods must be carefully chosen to prevent chemical incompatibility or optical interference with detection components. Additionally, the absence of standardized fabrication procedures and the variety of substrates, including PDMS, glass, and plastics, hinder reproducibility and large-scale production. Another challenge involves miniaturizing sensors to fit within microchannels while keeping high sensitivity and selectivity. Fluid control at the microscale adds more complexity, as pressure stability and bubble prevention are vital for consistent operation. Finally, although microfluidics aim for portability, many systems still depend on large external equipment for signal processing and readout, which limits true POC use [197,198]. Additionally, there are other challenges observed in MIP-integrated microfluidic systems. One of the primary concerns in MIP development is imprinted cavity (ies) heterogeneity, which arises due to the formation of multiple imprinted cavity (ies) with varying affinities during polymerization. This issue is primarily attributed to the dynamic equilibrium between template and functional monomers in non-covalent imprinting, leading to both strong and weak interactions within the polymer matrix [199–201]. Furthermore, polymerization kinetics affect the spatial arrangement of monomers, resulting in Fig. 11. Reusable microfluidic LOC sensor designed for on-site detection of zearalenone, redrawn based on an illustration from Hua et al. [11]; a) QD@MIPs synthesis route and b) architecture of the microfluidic platform proposed. M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 21 inconsistent imprinted cavity (ies) formation. Physical factors, such as grinding, sieving, and solvent-induced swelling or collapse, further contribute to the heterogeneity, impacting MIP reproducibility. Strategies such as stoichiometric non-covalent imprinting and computational modeling have been explored to reduce this variability, with the goal of improving site uniformity and enhancing recognition performance [202–205]. Another major challenge is the selection of functional monomers and cross-linkers, which plays a crucial role in determining MIP selectivity and stability. The binding affinity of monomers with the template influences imprinting efficiency, with hydrogen bonding (e.g., MAA), electrostatic interactions (e.g., 2-vinylpyridine), and π - π stacking (e.g., aromatic monomers) being commonly used approaches [205–207]. Cross-linking density also significantly impacts polymer rigidity and binding site accessibility. Highly cross-linked polymers provide better structural integrity but may hinder analyte diffusion, whereas lower cross-linking results in reduced stability. Porogen effect solvents also play a key role in template-monomer interactions, with non-polar solvents (e.g., toluene) stabilizing complex formation, while polar solvents (e.g., DMSO) disrupt weak non-covalent interactions, thereby lowering imprinting efficiency [208,209]. Porogen effect solvents also play a key role in template-monomer interactions, with non-polar solvents (e.g., toluene) stabilizing complex formation, while polar solvents (e.g., DMSO) disrupt weak non-covalent interactions, thereby lowering imprinting efficiency [208,209]. Efforts to optimize these parameters through systematic studies and AI-based modeling are ongoing. Also, recent studies also highlight the potential of combining systematic optimization with AI-driven modeling to predict solvent effects and improve imprinting results. In particular, AI-powered microfluidic platforms have shown the ability to quickly analyze complex interaction patterns and optimize formulation parameters in real time, offering a strong method to enhance monomer–template recognition and overall polymer performance [210,211]. A persistent issue in MIP-based sensors and separation technologies is template removal efficiency. Incomplete extraction of template molecules can result in background contamination, reducing selectivity and causing non-selective interactions. Conventional removal methods include Soxhlet extraction, ultrasonication, proteolytic digestion, and electrochemical elution, each with its own advantages and limitations [212–215]. Soxhlet extraction is solvent-intensive and time-consuming, while ultrasonication can damage the polymer structure. Proteolytic digestion is effective for protein templates but requires additional washing steps. Electrochemical removal is promising for conductive MIPs but has limited applicability to non-conductive polymers [112, 216–218]. To address this, dummy template imprinting, where structurally similar analogs replace the original template, has been proposed to minimize leaching and improve recognition performance [219,220]. The physical properties of MIPs, including their porosity, rigidity, and mass transport characteristics, also limit their practical applications. Highly cross-linked polymers exhibit reduced diffusion rates, making analyte imprinted cavity (ies) and release slower. If binding sites are buried too deep within the polymer matrix, target molecules may not access them efficiently, reducing sensitivity [221]. Solvent-induced swelling further complicates binding site stability, as polymer networks expand or contract based on solvent polarity, affecting recognition performance [222,223]. To overcome these issues, researchers have explored hierarchical porous MIPs, core-shell structures, and stimuli-responsive MIPs that change their conformation based on environmental conditions such as pH or temperature [224,225]. Computationally guided design has also been utilized to optimize polymer morphology for improved analyte accessibility and mass transport efficiency [226–228]. Several advanced synthesis techniques have emerged to address the limitations of traditional MIP fabrication methods. Click chemistry offers a precise and highly efficient strategy for functionalizing imprinted cavities, improving both stability and selectivity [229]. Monomolecular imprinting using dendrimers enhances site uniformity by utilizing well-defined dendritic structures that create highly selective recognizing cavities [199,202,230–235]. Solvent-programmable polymers (SPP) introduce restricted molecular rotation, allowing control over binding site conformation and improving selectivity [236]. Computational approaches, including AI-driven modeling, have been increasingly employed to predict optimal monomer-template interactions and reduce non-selective binding [237–241]. Several advanced synthesis techniques have emerged to address the limitations of traditional MIP fabrication methods. Scaling up microfluidic device fabrication presents significant material-related challenges. Traditional materials like PDMS are widely used due to their optical transparency and flexibility, yet they exhibit limitations such as permeability to gases, solvent-induced swelling, and difficulties in long-term surface modifications [162,242,243]. These drawbacks necessitate the exploration of alternative materials that are cost-effective, chemically stable, and capable of forming robust seals for mass production. In this context, thermoplastics such as polystyrene [242], polycarbonate [244], and cyclic olefin copolymer [245] have gained prominence, as they are more compatible with scalable manufacturing techniques [246]. While these materials support mass production, their surface properties must be carefully tailored to enable efficient MIP integration under continuous flow conditions an essential requirement for realizing reproducible and high-performance sensing platforms. Despite the growing interest in microfluidic systems with MIPs, studies focusing predominantly on thermoplastic-based microfluidic devices remain limited. Most existing research still relies heavily on PDMS-based platforms, which, while advantageous for prototyping, fall short in terms of scalability and long-term operational stability. The scarcity of comprehensive studies exploring MIP integration into thermoplastic microfluidics, especially under continuous flow conditions, suggests that significant engineering and surface chemistry challenges remain unresolved, including the incompatibility of conventional polymerization chemistries with thermoplastics, bonding difficulties following MIP surface modification, inconsistent and poorly scalable surface activation techniques, swelling or deformation of MIPs under flow, and limited mass transfer within microchannel-embedded recognition sites. This gap partly explains the absence of commercially available MIP-based microfluidic sensors, underlining the need for further research in this direction. Future studies should focus on developing reliable surface modification strategies tailored for thermoplastics, enabling robust MIP attachment without compromising flow dynamics or sensor performance. Addressing these challenges will be critical for transitioning from proof-of-concept devices to scalable, realworld diagnostic platforms. 6.2. Commercialization prospects Although several companies offer accessible, high-quality microfluidic technologies for diagnostics, material synthesis, and biomedical research, commercially available MIP-based sensor platforms, especially personalized systems ready for integration, remain scarce. One notable company working in this field is Tozaro (formerly known as MIP Diagnostics), based in Sharnbrook, UK. Founded in 2015, Tozaro specializes in the development of high-affinity nanoparticle-based MIPs using a solid-phase imprinting technique to detect melamine, vancomycin and a peptide, an approach pioneered by the Piletsky group at the University of Leicester [247]. This method offers significant advantages, such as the ability to recycle the template, reducing costs and enabling semi-automated production through the elution of polymers rather than template extraction [248]. The company initially focused on contract development services for industries seeking alternatives to antibodies, particularly in applications where antibodies face limitations such as batch-to-batch variability or sensitivity to extreme temperature and pH conditions. Their nanoMIPs have been successfully employed in direct M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 22 antibody replacements in diagnostic assays and sensor development. Their recent scientific posters illustrate the application of smart polymers, MIPs, for the determination and purification of viral targets such as VSV-G, SARS-CoV-2 spike protein, and HIV gp120. These MIPs have been integrated into a variety of analytical platforms including SPR, biolayer interferometry (BLI), lateral flow assays, membrane-based capture systems, and LC-MS workflows. While these implementations are currently limited to batch and sensor-based formats rather than continuous-flow microfluidic systems, they highlight the versatility and industrial promise of MIPs in diagnostics and bioprocessing. The rapid prototyping, high selectivity, and chemical robustness of these systems position smart polymers as a compelling alternative to conventional antibodies in regulated environments. With recent investments, Tozaro aims to scale up its production and expand its in-house portfolio of MIP-based products, starting with commonly used biomarkers, which could significantly accelerate the development of MIP-based sensing technologies in the near future [241]. Alongside Tozaro, several other companies are working to commercialize molecularly imprinted polymers. MIP Technologies, a Swedish company and one of the first spin-offs in this field, has led the way in developing MIPs for uses like separation, purification, and chemical sensing. The company has also played a key role in showing that the technology is viable for industrial use [249]. Meanwhile, MIP Diagnostics has broadened the potential of MIPs to include biomedical and diagnostic applications [250]. These companies, taken together, demonstrate how MIP research is being turned into products and services that matter to the commercial world, helping MIP technology move from the lab to the market. 7. Conclusion The integration of MIPs with microfluidic technologies presents a compelling strategy for creating highly selective, miniaturized, and robust sensing platforms. This synergistic union harnesses the molecular recognition power of MIPs with the precision and automation capabilities of microfluidics, resulting in devices that are well-suited for POC diagnostics, environmental monitoring, and food safety assessments. Throughout this review, we have highlighted the broad spectrum of MIP synthesis strategies and microfluidic integration methods, along with their associated advantages and limitations. Recent advancements demonstrate that electropolymerization, nanoimprinting, and surface imprinting approaches are particularly well-matched to microfluidic platforms due to their spatial control, reproducibility, and compatibility with enclosed channel geometries. In parallel, soft lithography, 3D printing, and PDMS/glass-based fabrication have enabled rapid prototyping of microfluidic chips tailored for MIP-based sensing. Integration strategies continue to diversify, with surface functionalization chemistries, flow-based immobilization, and in situ synthesis providing multiple paths toward robust and sensitive platforms. Despite the promise, several challenges remain: ensuring uniform template removal, minimizing binding site heterogeneity, and developing scalable, reproducible fabrication protocols. The sensitivity and selectivity of MIP-based microfluidic sensors can be significantly enhanced through rational monomer-template design, AI-assisted modeling, and incorporation of advanced materials such as nanocomposites or hybrid functional layers. Integrating AI with microfluidic sensor platforms and MIP-based sensors has become a transformative method in environmental and bioanalytical sensing. AI supports realtime analysis, classification, and prediction using complex data from microfluidic systems, improving their ability to detect contaminants with high precision and sensitivity. Machine learning (ML) and deep learning (DL) techniques, such as convolutional neural networks (CNNs) and support vector machines (SVMs), enable automated decisionmaking and pattern recognition from diverse, dynamic samples. This collaboration offers benefits like shorter analysis times, increased portability, and on-site monitoring with minimal user input. The compact design of microfluidic devices complements AI’s need for large, structured data, helping create cost-effective, high-throughput, autonomous systems. However, challenges remain, including the need for extensive high-quality training data for reliable AI performance. Additionally, issues around model interpretability and the technical complexity of integrating AI with microfluidic hardware can slow adoption. Despite these restrictions, AI-enhanced microfluidics shows great potential for future intelligent sensor systems in fields from environmental monitoring to healthcare diagnostics [210,251,252]. Future efforts should focus on turning MIP-based sensors into realworld applications by developing standardized manufacturing processes and using scalable, chemically durable substrates like thermoplastics that can be produced industrially. Their flexibility allows use across many fields, from medical diagnostics and environmental monitoring to food safety and industrial process control. Although challenges remain, such as achieving consistent imprinting in complex samples and maintaining long-term stability, MIPs provide distinct benefits like molecular-level selectivity, affordable production, and compatibility with various transduction platforms. These qualities, along with ongoing progress in modular microfluidic integration and the growth of wearable, label-free, and smartphone-compatible systems, highlight why MIP-on-the-flow technologies have strong potential to transform molecular detection, determination and lead the next wave of highprecision sensing solutions. CRediT authorship contribution statement Meltem Okan: Writing – review & editing, Writing – original draft, Conceptualization. Vildan Sanko: Writing – review & editing, Writing – original draft, Conceptualization. Ender Yıldırım: Writing – review & editing, Conceptualization. H. Cumhur Tekin: Writing – review & editing. Haluk Külah: Supervision, Project administration. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgement This work is supported by the OrChESTRA (Organ-on-a-Chip Focused Strategic Partnership) project, which has received funding from the European Union’s Horizon Europe research and innovation program under Grant Agreement No. 101079473. Data availability No data was used for the research described in the article. References [1] A. Herrera-Chac´ on, X. Cet´ o, M. del Valle, Anal. Bioanal. Chem. 413 (2021) 6117–6140. [2] C. Virumbrales, R. Hern´ andez-Ruiz, M. Trigo-L´ opez, S. Vallejos, J.M. García, Sensors 24 (2024) 3852. [3] Y. Saylan, A. Denizli, Micromachines 10 (2019) 766. [4] C.M. Raju, D.P. Elpa, P.L. Urban, ACS Sens. 9 (2024) 1033–1048. [5] M. Sonker, V. Sahore, A.T. Woolley, Anal. Chim. Acta 986 (2017) 1–11. [6] A. Sharma, N.I. Hossain, A. Thomas, S. Sonkusale, ACS Appl. Mater. Interfaces 17 (2025) 25083–25096. [7] N. Li, Y. Zhou, H. Sun, N. Wang, N. Yang, P. Ren, L. Fu, Y. Zhang, W. Liu, Y. Li, J. Jin, Chem. Eng. J. 504 (2025) 159023. [8] N. Amin, J. Chen, Q. He, J.S. Schwartz, J.J. Wu, Sensor. Actuator. B Chem. 420 (2024) 136464. [9] S. Nagabooshanam, S. Roy, S. Deshmukh, S. Wadhwa, I. Sulania, A. Mathur, S. Krishnamurthy, L.M. Bharadwaj, S.S. Roy, ACS Omega 5 (2020) 31765–31773. [10] S. Feng, Y. Lin, J. Xu, X. Lu, Sensor. Actuator. B Chem. 394 (2023) 134342. [11] M.Z. Hua, S. Li, M.S. Roopesh, X. Lu, Lab Chip 24 (2024) 2700–2711. M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 23 [12] K. Khachornsakkul, R. Del-Rio-Ruiz, L. Chheang, W. Zeng, S. Sonkusale, Lab Chip 24 (2024) 2262–2271. [13] V. Kamat, L. Burton, V. Venkadesh, K. Jayachandran, S. Bhansali, ECS Sens. Plus 2 (2023) 043201. [14] Y. Saylan, ¨ O. Altıntas ¸, A. Denizli, Results Opt. 13 (2023) 100541. [15] T. Karasu, E. ¨ Ozgür, L. Uzun, J. Pharm. Biomed. Anal. 226 (2023) 115257. [16] A. Cetinkaya, S.I. Kaya, S.A. Ozkan, Anal. Chim. Acta 1357 (2025) 344080. [17] L. Wang, M. Pagett, W. Zhang, Sens. Acutators Rep. 5 (2023) 100153. [18] A. S ¸enocak, S.O. Tümay, E. Sarı, V. Sanko, M. Durmus¸, E. Demirbas, J. Electrochem. Soc. 168 (2021) 087513. [19] V. Sanko, C. Erkmen, F. Kuralay, Electroanalysis 36 (2024) e202400004. [20] D. Refaat, M.G. Aggour, A.A. Farghali, R. Mahajan, J.G. Wiklander, I.A. Nicholls, S.A. Piletsky, Int. J. Mol. Sci. 20 (2019) 6304. [21] A. Zare, B. Babamiri, M. Hassani, M. Khalghollah, M. Mohammadi, S. Haghjooy Javanmard, A. Sanati Nezhad, Biosens. Bioelectron. 286 (2025) 117599. [22] Z. Li, Y. Bai, M. You, J. Hu, C. Yao, L. Cao, F. Xu, Biosens. Bioelectron. 177 (2021) 112952. [23] J. Liu, Y. Zhang, M. Jiang, L. Tian, S. Sun, N. Zhao, F. Zhao, Y. Li, Biosens. Bioelectron. 91 (2017) 714–720. [24] X. Mei, J. Yang, X. Yu, Z. Peng, G. Zhang, Y. Li, Sensor. Actuator. B Chem. 381 (2023) 133451. [25] C.-C. Hong, C.-P. Chen, J.-C. Horng, S.-Y. Chen, Biosens. Bioelectron. 50 (2013) 425–430. [26] S. Wagner, J. Bell, M. Biyikal, K. Gawlitza, K. Rurack, Biosens. Bioelectron. 99 (2018) 244–250. [27] S. Li, J. Li, J. Luo, Z. Xu, X. Ma, Microchim. Acta 185 (2018) 295. [28] Y. Shiraki, K. Tsuruta, J. Morimoto, C. Ohba, A. Kawamura, R. Yoshida, R. Kawano, T. Uragami, T. Miyata, Macromol. Rapid Commun. 36 (2015) 515. [29] A. Doostmohammadi, H. Huang, S. Naushad, P. Rezai, Microchem. J. 206 (2024) 111611. [30] J. Ashley, Y. Shukor, R. D’Aurelio, L. Trinh, T.L. Rodgers, J. Temblay, M. Pleasants, I.E. Tothill, ACS Sens. 3 (2018) 418–424. [31] R.A. Hassan, S. Abu Hanifah, L.Y. Heng, Talanta 287 (2025) 127592. [32] S. Akg¨ onüllü, S. Kılıç, C. Esen, A. Denizli, Polymers 15 (2023) 629. [33] G. Ozcelikay, S.I. Kaya, E. Ozkan, A. Cetinkaya, E. Nemutlu, S. Kır, S.A. Ozkan, TrAC, Trends Anal. Chem. 146 (2022) 116487. [34] J. Wang, R. Liang, W. Qin, TrAC, Trends Anal. Chem. 130 (2020) 115980. [35] E.K. Reville, E.H. Sylvester, S.J. Benware, S.S. Negi, E.B. Berda, Polym. Chem. 13 (2022) 3387–3411. [36] E. Salmanli, T. Tezcan, T. Karaoglu, Anal. Methods 16 (2024) 551–557. [37] A. Lusina, T. Nazim, M. Cegłowski, Synthesis and characterization of MIPs, in: S. Patra, S.K. Shukla, M. Sillanp¨ a¨ a (Eds.), Molecularly Imprinted Polymers: Path to Artificial Antibodies, Springer Nature, Switzerland, 2024, pp. 29–67. [38] S.F.F.S. Yaacob, M. Suwaibatu, R.Z.R. Jamil, N.N.M. Zain, M. Raoov, F.B.M. Suah, J. Chem. Technol. Biotechnol. 98 (2023) 312–330. [39] G. Wullf, A. Biffis, Molecular imprinting with covalent or stoichiometric noncovalent interactions, in: B. Sellergren (Ed.), Molecularly Imprinted Polymers: Man-Made Mimics of Antibodies and their Application in Analytical Chemistry, Elsevier, Amsterdam, 2000, pp. 71–111. [40] D. Hong, C. Wang, L. Gao, C. Nie, Molecules 29 (2024) 3555. [41] Z. Wang, X. Sun, Y. Xu, L. Yang, M. Wang, Y. Xia, Y. Wang, Y. Tang, C. Qiao, Y. Lin, Colloids Surf. A Physicochem. Eng. Asp. 696 (2024) 134292. [42] I.A. Nicholls, Chem. Lett. 24 (1995) 1035–1036. [43] I.A. Nicholls, H.S. Andersson, Thermodynamic principles underlying molecularly imprinted polymer formulation and ligand recognition, in: B. Sellergren (Ed.), Molecularly Imprinted Polymers: Man-Made Mimics of Antibodies and Their Application in Analytical Chemistry, Elsevier, Amsterdam, 2000, pp. 59–70. [44] I.A. Nicholls, K. Adbo, H.S. Andersson, P.O. Andersson, J. Ankarloo, J. HedinDahlstr¨ om, P. Jokela, J.G. Karlsson, L. Olofsson, J. Rosengren, S. Shoravi, J. Svenson, S. Wikman, Anal. Chim. Acta 435 (2001) 9–18. [45] G. van Wissen, J.W. Lowdon, T.J. Cleij, K. Eersels, B. van Grinsven, Polymers 17 (2025) 1057. [46] Z.M. Karazan, M. Roushani, Synthesis methods and strategies for MIPs, in: S. Patra, M. Sillanpaa (Eds.), Molecularly Imprinted Polymers as Artificial Antibodies for the Environmental Health, Springer Nature, Switzerland, 2024, pp. 31–52. [47] R. Arshady, K. Mosbach, Makromol. Chem. Phys. 182 (1981) 687–692. [48] J. Matsui, Y. Miyoshi, O. Doblhoff-Dier, T. Takeuchi, Anal. Chem. 67 (1995) 4404–4408. [49] C.-m. Dai, X.-f. Zhou, Y.-l. Zhang, S.-g. Liu, J. Zhang, J. Hazard. Mater. 198 (2011) 175–181. [50] G. Wulff, Polymer assisted molecular recognition: the current understanding of the molecular imprinting procedure, in: U.K. Pandit, F.C. Alderweireldt (Eds.), Bioorganic Chemistry in Healthcare and Technology, Springer, Boston, 1991, pp. 55–66. [51] C. C´ aceres, C. Bravo, B. Rivas, E. Moczko, P. S´ aez, Y. García, E. Pereira, Polymers 10 (2018) 679. [52] T. Kanai, C. Sanskriti, P. Vislawath, A.B. Samui, A. Baran, J. Nanosci. Nanotechnol. 13 (2013) 3054–3061. [53] M. Gomar, H.A. Panahi, E. Pournamdari, ChemistrySelect 3 (2018) 2571–2577. [54] A. Carnicero, M. Martinelli, ACS Appl. Polym. Mater. 7 (2025) 654–665. [55] P. Panjan, R.P. Monasterio, A. Carrasco-Pancorbo, A. Fernandez-Gutierrez, A. M. Sesay, J.F. Fernandez-Sanchez, J. Chromatogr. A 1576 (2018) 26–33. [56] M.G. Cho, S. Hyeong, K.K. Park, S.H. Chough, Polymer 237 (2021) 124348. [57] A. Wong, F.M. de Oliveira, C.R.T. Tarley, M. Del Pilar Taboada Sotomayor, React. Funct. Polym. 100 (2016) 26–36. [58] H.Y. Alharbi, M.S. Aljohani, M. Monier, React. Funct. Polym. 191 (2023) 105686. [59] W. Zhang, X. She, L. Wang, H. Fan, Q. Zhou, X. Huang, J.Z. Tang, Materials 10 (2017) 475. [60] E. De Rycke, A. Trynda, M. Jaworowicz, P. Dubruel, S. De Saeger, N. Beloglazova, Biosens. Bioelectron. 172 (2021) 112773. [61] J. Kuhn, G. Aylaz, E. Sari, M. Marco, H.H.P. Yiu, M. Duman, J. Hazard. Mater. 387 (2020) 121709. [62] M. Okan, E. Sari, M. Duman, Biosens. Bioelectron. 88 (2017) 258–264. [63] M. Cegłowski, J. Kurczewska, A. Lusina, T. Nazim, P. Ruszkowski, Polymers 14 (2022) 1027. [64] J. Liu, Q. Song, Y. Li, J.J. Bao, Int. J. Food Sci. Technol. 58 (2023) 1423–1433. [65] T. Sergeyeva, D. Yarynka, E. Piletska, R. Linnik, O. Zaporozhets, O. Brovko, S. Piletsky, A. El’skaya, Talanta 201 (2019) 204–210. [66] R. Üzek, E. Sari, S. S ¸enel, A. Denizli, A. Merkoçi, Microchim. Acta 186 (2019) 218. [67] L.M. Madikizela, P.S. Mdluli, L. Chimuka, React. Funct. Polym. 103 (2016) 33–43. [68] M. Yolcu, N. Dere, Can. J. Chem. 96 (2018) 1027–1036. [69] S. Khan, S. Hussain, A. Wong, M.V. Foguel, L. Moreira Gonçalves, M.I. Pividori Gurgo, M.d.P. Taboada Sotomayor, React. Funct. Polym. 122 (2018) 175–182. [70] M.M. Qronfla, B. Jamoussi, R. Chakroun, Polymers 15 (2023) 2398. [71] Y. Shao, L. Zhou, Q. Wu, C. Bao, M. Liu, J. Hazard. Mater. 339 (2017) 418–426. [72] R. Aeinehvand, P. Zahedi, S. Kashani-Rahimi, M. Fallah-Darrehchi, M. Shamsi, Polym. Adv. Technol. 28 (2017) 828–841. [73] A.P. Leshchinskaya, N.M. Ezhova, O.A. Pisarev, React. Funct. Polym. 102 (2016) 101–109. [74] S ¸.-O. Dima, Polym. Eng. Sci. 55 (2015) 1152–1162. [75] T.A. Sergeyeva, O.A. Slinchenko, L.A. Gorbach, V.F. Matyushov, O.O. Brovko, S. A. Piletsky, L.M. Sergeeva, G.V. Elska, Anal. Chim. Acta 659 (2010) 274–279. [76] A.G. Mayes, M.J. Whitcombe, Adv. Drug Deliv. Rev. 57 (2005) 1742–1778. [77] S. Beyazit, B. Tse Sum Bui, K. Haupt, C. Gonzato, Prog. Polym. Sci. 62 (2016) 1–21. [78] V.D. Salian, M.E. Byrne, Macromol. Mater. Eng. 298 (2013) 379–390. [79] S. Venkatesh, S.P. Sizemore, M.E. Byrne, Biomaterials 28 (2007) 717–724. [80] K.S. Anseth, C.M. Wang, C.N. Bowman, Polymer 35 (1994) 3243–3250. [81] J.B. Hutchison, A.S. Lindquist, K.S. Anseth, Macromolecules 37 (2004) 3823–3831. [82] M.A. S¨ oylemez, M. Okan, O. Güven, M. Barsbay, Curr. Res. Green Sustain. Chem. 4 (2021) 100196. [83] J.M. Asua, Prog. Polym. Sci. 27 (2002) 1283–1346. [84] H. Zhang, Eur. Polym. J. 49 (2013) 579–600. [85] X. Liu, F. Wu, C. Au, Q. Tao, M. Pi, W. Zhang, J. Appl. Polym. Sci. 136 (2019) 46984. [86] M.M. Moein, A. Abdel-Rehim, M. Abdel-Rehim, Molecules 24 (2019) 2889. [87] C. Erkmen, V. Sanko, B.O. Ozturk, J. Quinchía, J. Orozco, F. Kuralay, TrAC, Trends Anal. Chem. 179 (2024) 117876. [88] C. Dong, H. Shi, Y. Han, Y. Yang, R. Wang, J. Men, Eur. Polym. J. 145 (2021) 110231. [89] A.A. Faysal, A. Cetinkaya, T. Erdo˘ gan, S.A. Ozkan, A. G¨ olcü, Electrochim. Acta 512 (2025) 145516. [90] B. Fresco-Cala, A.D. Batista, S. C´ ardenas, Molecules 25 (2020) 4740. [91] A. Chiappini, L. Pasquardini, A.M. Bossi, Sensors 20 (2020) 5069. [92] Y. Wu, J. Xiong, S. Wei, L. Tian, X. Shen, C. Huang, J. Chromatogr. A 1714 (2024) 464550. [93] M.D. Ariani, A. Zuhrotun, P. Manesiotis, A.N. Hasanah, Polym. Adv. Technol. 35 (2024) e6201. [94] H. Lu, H. Tian, C. Wang, S. Xu, Mater. Adv. 1 (2020) 2182–2201. [95] Z. Wang, X. Zhang, Z. Li, G. Wang, Y. Wang, P. Li, X. Yue, J. Environ. Chem. Eng. 12 (2024) 113405. [96] N. Ezati, M. Abdouss, M. Rouhani, P.G. Kerr, E. Kowsari, React. Funct. Polym. 181 (2022) 105437. [97] M. Barsbay, O. Güven, Radiat. Phys. Chem. 169 (2020) 107816. [98] H.F. El-Sharif, N.W. Turner, S.M. Reddy, M.V. Sullivan, Talanta 240 (2022) 123158. [99] C. Unger, P.A. Lieberzeit, React. Funct. Polym. 161 (2021) 104855. [100] F. Cui, Z. Zhou, H.S. Zhou, Sensors 20 (2020) 996. [101] M. Roushani, N. Zalpour, React. Funct. Polym. 169 (2021) 105069. [102] C. Mwanza, W.-Z. Zhang, K. Mulenga, S.-N. Ding, Green Chem. 26 (2024) 11490–11517. [103] A.N. Hasanah, N. Safitri, A. Zulfa, N. Neli, D. Rahayu, Molecules 26 (2021) 5612. [104] S.R. Shafqat, S.A. Bhawani, S. Bakhtiar, M.N.M. Ibrahim, S.S. Shafqat, BMC Chem. 17 (2023) 46. [105] H.-L. Peng, J.-P. Fan, Morphological characterization of molecularly imprinted polymer composites, in: M.P. Sooraj, A.S. Nair, B. Mathew, S. Thomas (Eds.), Molecularly Imprinted Polymer Composites Synthesis, Characterisation and Applications, Woodhead Publishing, Cambridge, 2021, pp. 123–142. [106] R. Moradirad, H. Asilian Mahabadi, S.J. Shahtaheri, A. Rashidi, S. Fakhraie, M. Khadem, J. Sajedifar, Int. J. Environ. Sci. Technol. 21 (2024) 8943–8958. [107] N.F. Che Lah, A.L. Ahmad, S.C. Low, N.D. Zaulkiflee, Membranes 11 (2021) 657. [108] W. Thongchai, P. Poolprasert, S. Thongchai, J. Chromatogr. Sci. 59 (2021) 877–886. [109] A. Mohebali, M. Abdouss, Y. Kazemi, S. Daneshnia, Polym. Adv. Technol. 32 (2021) 4386–4396. [110] J. Svenson, J.G. Karlsson, I.A. Nicholls, J. Chromatogr. A 1024 (2004) 39–44. [111] J. Niu, M. Du, W. Wu, J. Yang, Q. Chen, J. Separ. Sci. 47 (2024) 2400353. M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 24 [112] A.S. L´ opez, M.P. Ramos, R. Herrero, J.M.L. Vilari˜ no, Sep. Purif. Technol. 257 (2021) 117860. [113] S. Espinoza-Torres, R. L´ opez, M.D.P.T. Sotomayor, J.C. Tuesta, G. Picasso, S. Khan, Polymers 15 (2023) 3332. [114] M. Ferreira, V. Carvalho, J. Ribeiro, R.A. Lima, S. Teixeira, D. Pinho, Micromachines 15 (2024) 873. [115] B. S¸en-Do˘ gan, M. Okan, N. Afs¸ar-Erkal, E. ¨ Ozgür, ¨ O. Zorlu, H. Külah, Micromachines 11 (2020) 703. [116] E. Sari, R. Üzek, M. Duman, A. Denizli, Talanta 150 (2016) 607–614. [117] J. McClements, P.M. Seumo Tchekwagep, A.L. Vilela Strapazon, F. Canfarotta, A. Thomson, J. Czulak, R.E. Johnson, K. Novakovic, P. Losada-P´ erez, A. Zaman, I. Spyridopoulos, R.D. Crapnell, C.E. Banks, M. Peeters, ACS Appl. Mater. Interfaces 13 (2021) 27868–27879. [118] T. Wasilewski, S. Orbay, N.F. Brito, K. Sikora, A.C.A. Melo, M.E. Melendez, B. Szulczy´ nski, A. Sanyal, W. Kamysz, J. Gębicki, TrAC, Trends Anal. Chem. 177 (2024) 117783. [119] M. Caldara, G. van Wissen, T.J. Cleij, H. Dili¨ en, B. van Grinsven, K. Eersels, J. W. Lowdon, Adv. Sens. Res. 2 (2023) 2200059. [120] N. Leibl, K. Haupt, C. Gonzato, L. Duma, 9 (2021) 123. [121] S. Kia, M. Fazilati, H. Salavati, S. Bohlooli, RSC Adv. 6 (2016) 31906–31914. [122] N.P. Kalogiouri, A. Tsalbouris, A. Kabir, K.G. Furton, V.F. Samanidou, Microchem. J. 157 (2020) 104965. [123] M. Okan, M. Duman, Development of molecularly imprinted polymer-based microcantilever sensor system, in: A. Tiwari (Ed.), Advanced Molecularly Imprinting Materials, Scrivener Publishing, Beverly, 2016, pp. 637–679. [124] A. Ebner, P. Hinterdorfer, H.J. Gruber, Ultramicroscopy 107 (2007) 922–927. [125] S. Choudhary, Z. Altintas, Biosensors 13 (2023) 229. [126] M. Okan, M. Duman, Hacettepe Journal of Biology and Chemistry 52 (2024) 405–414. [127] W. Li, J. Xiang, J. Han, M. Man, L. Chen, B. Li, Analyst 148 (2023) 5896–5904. [128] I. Perçin, N. Idil, M. Bakhshpour, E. Yılmaz, B. Mattiasson, A. Denizli, Sensors 17 (2017) 1375. [129] T. Wang, C. Shannon, Anal. Chim. Acta 708 (2011) 37–43. [130] C. Xu, L. Ye, Chem. Commun. 47 (2011) 6096–6098. [131] M.A. Abomuti, Polym. Int. 74 (2025) 696–709. [132] A. Poma, A. Guerreiro, M.J. Whitcombe, E.V. Piletska, A.P. Turner, S.A. Piletsky, Adv. Funct. Mater. 23 (2013) 2821–2827. [133] ¨ O. Erdem, I. Es¸, Y. Saylan, M. Atabay, M.A. Gungen, K. ¨ Olmez, A. Denizli, F. Inci, Nat. Commun. 14 (2023) 4840. [134] M. Farrokhnia, B. Babamiri, M. Mohammadi, A. Sanati Nezhad, ACS Sens. 10 (2025) 3112–3122. [135] Q. Wang, X. Bai, F. Liu, P. Li, Q. Tang, Microchem. J. 205 (2024) 111260. [136] C.-C. Hong, C.-C. Lin, C.-L. Hong, Z.-X. Lin, M.-H. Chung, P.-W. Hsieh, Biosens. Bioelectron. 86 (2016) 623–629. [137] R.R. Khan, H. Ibrahim, G. Rawal, J. Zhang, M. Lu, L. Dong, Sensor. Actuator. B Chem. 389 (2023) 133920. [138] F. Arcadio, L. Zeni, C. Perri, G. D’Agostino, G. Chiaretti, G. Porto, A. Minardo, N. Cennamo, Chemosensors 9 (2021) 218. [139] M. Pesavento, L. Zeni, L. De Maria, G. Alberti, N. Cennamo, Biosensors 11 (2021) 72. [140] O. Çelik, Y. Saylan, I. G¨ oktürk, F. Yılmaz, A. Denizli, Talanta 253 (2023) 123939. [141] M. Bakhshpour-Yücel, M. Küçük, E. Tümay ¨ Ozer, B. Osman, Talanta Open 11 (2025) 100417. [142] H.S. Birinci, S. Akg¨ onüllü, H. Yavuz, O. Arslan, A. Denizli, Microchem. J. 213 (2025) 113668. [143] F. Bonyadi, M. Kavruk, S. Ucak, B. Cetin, G. Bayramo˘ glu, A.D. Dursun, Y. Arica, V. C. Ozalp, Crit. Rev. Anal. Chem. 54 (2024) 2888–2899. [144] S. Akg¨ onüllü, E. ¨ Ozgür, A. Denizli, Chemosensors 10 (2022) 106. [145] S. Haghdoust, U. Arshad, A. Mujahid, L. Schranzhofer, P.A. Lieberzeit, Chemosensors 9 (2021) 362. [146] ¨ O. Acet, M. Odabas ¸ı, Polym. Bull. 80 (2023) 6657–6674. [147] A. Ebner, L. Wildling, R. Zhu, C. Rankl, T. Haselgrübler, P. Hinterdorfer, H. J. Gruber, Functionalization of probe tips and supports for single-molecule recognition force microscopy, in: P. Samorì (Ed.), STM and AFM Studies on (Bio) Molecular Systems: Unravelling the Nanoworld, Springer, Germany, 2008, pp. 29–76. [148] D. Cetin, M. Okan, E. Bat, H. Kulah, Colloids Surf. B Biointerfaces 188 (2020) 110808. [149] D. Bas¸, ˙ I.H. Boyacı, Anal. Bioanal. Chem. 400 (2011) 703–707. [150] B.S. Simpkins, S. Hong, R. Stine, A.J. M¨ akinen, N.D. Theodore, M.A. Mastro, C. R. Eddy, P.E. Pehrsson, J. Phys. D Appl. Phys. 43 (2010) 015303. [151] P. Kar, A. Pandey, J.J. Greer, K. Shankar, Lab Chip 12 (2012) 821–828. [152] S. Ramanavicius, A. Jagminas, A. Ramanavicius, Polymers 13 (2021) 974. [153] X. Yuan, N. Wolf, D. Mayer, A. Offenhausser, R. W¨ ordenweber, Langmuir 35 (2019) 8183–8190. [154] J.V. Staros, R.W. Wright, D.M. Swingle, Anal. Biochem. 156 (1986) 220–222. [155] T. Kamra, S. Chaudhary, C. Xu, L. Montelius, J. Schnadt, L. Ye, J. Colloid Interface Sci. 461 (2016) 1–8. [156] M. Okan, M. Duman, Sensor. Actuator. B Chem. 256 (2018) 325–333. [157] S. Hosseini, F. Ibrahim, I. Djordjevic, H.A. Rothan, R. Yusof, C. van der Marel, L. H. Koole, Appl. Surf. Sci. 317 (2014) 630–638. [158] P. Y´ a˜ nez-Sede˜ no, A. Gonz´ alez-Cort´ es, S. Campuzano, J.M. Pingarr´ on, Sensors 19 (2019) 2379. [159] H.C. Kolb, M.G. Finn, K.B. Sharpless, Angew. Chem. Int. Ed. 40 (2001) 2004–2021. [160] K.y. Tomizaki, K. Usui, H. Mihara, Chembiochem 6 (2005) 782–799. [161] R.B. Alnoman, M.S. Aljohani, H.Y. Alharbi, M. Monier, J. Chromatogr. A 1743 (2025) 465657. [162] G.M. Whitesides, Nature 442 (2006) 368–373. [163] E.K. Sackmann, A.L. Fulton, D.J. Beebe, Nature 507 (2014) 181–189. [164] A. Bor´ ok, K. Laboda, A. Bony´ ar, Biosensors 11 (2021) 292. [165] X. Han, Y. Zhang, J. Tian, T. Wu, Z. Li, F. Xing, S. Fu, Polym. Eng. Sci. 62 (2022) 3–24. [166] T. Trantidou, M.S. Friddin, K.B. Gan, L. Han, G. Bolognesi, N.J. Brooks, O. Ces, Anal. Chem. 90 (2018) 13915–13921. [167] K. Gao, J. Liu, Y. Fan, Y. Zhang, Biomed. Microdevices 21 (2019) 83. [168] Y. Fan, S. Liu, K. Gao, Y. Zhang, Microsyst. Technol. 24 (2018) 1783–1787. [169] A. Gokaltun, M.L. Yarmush, A. Asatekin, O.B. Usta, Technology 5 (2017) 1–12. [170] H. Chen, F. Bian, L. Sun, D. Zhang, L. Shang, Y. Zhao, Adv. Mater. 32 (2020) 2005394. [171] J. Liu, Q. Deng, D. Tao, K. Yang, L. Zhang, Z. Liang, Y. Zhang, Sci. Rep. 4 (2014) 5487. [172] J.A. Thompson, H.H. Bau, J. Chromatogr. B 878 (2010) 228–326. [173] T. Kang, C. Park, N. Meghani, T.T.D. Tran, P.H.L. Tran, B.-J. Lee, Pharmaceutics 12 (2020) 555. [174] Y. Liu, X. Su, P. Fan, X. Liu, Y. Pan, J. Ping, Sci. Bull. 70 (2025) 2004–2013. [175] J. Sanchez-Almirola, A. Gage, R. Lopez, D. Yapell, M. Mujawar, V. Kamat, A. Kaushik, Mater. Sci. Eng. B 296 (2023) 116670. [176] A.V. Okhokhonin, M.I. Stepanova, T.S. Svalova, A.N. Kozitsina, J. Electroanal. Chem. 924 (2022) 116853. [177] R. Park, S. Jeon, J.W. Lee, J. Jeong, Y.W. Kwon, S.H. Kim, J. Jang, D.-W. Han, S. W. Hong, Biosensors 13 (2023) 1013. [178] H. Haghgouei, N. Alizadeh, Anal. Chim. Acta 1330 (2024) 343275. [179] M. Garg, H. Guo, E. Maclam, E. Zhanov, S. Samudrala, A. Pavlov, M.S. Rahman, M. Namkoong, J.P. Moreno, L. Tian, ACS Appl. Mater. Interfaces 16 (2024) 46113–46122. [180] M. Adampourezare, J. Ezzati Nazhad Dolatabadi, K. Asadpour-Zeynali, Microchem. J. 212 (2025) 113168. [181] M. Karimian, K. Dashtian, R. Zare-Dorabei, S. Norouzi, Anal. Chim. Acta 1285 (2024) 342022. [182] A. Riaz, I. Zareef, A. Munawar, A. Rakha, M.F. Khan, S. Akhtar, A. Anwar, S. Nazir, S.u. Din, A.S. Abbasi, Appl. Nanosci. 13 (2023) 6331–6339. [183] T. Rypar, V. Adam, M. Vaculovicova, M. Macka, Sensor. Actuator. B Chem. 341 (2021) 129999. [184] Q. Zhou, X. Wang, K. Tang, Y. Chen, R. Wang, H. Lei, Z. Yang, Z. Zhang, Talanta 278 (2024) 126402. [185] T. Moya-Cavas, L.R. Arias-Aranda, E. Benito-Pe˜ na, L. Bouffier, N. Sojic, G. Salinas, G. Orellana, Sensor. Actuator. B Chem. 433 (2025) 137566. [186] K. Khachornsakkul, W. Zeng, S. Sonkusale, Microchim. Acta 191 (2024) 253. [187] Y. He, M.Z. Hua, S. Feng, X. Lu, Food Chem. 451 (2024) 139446. [188] Y. Zhou, D. Wang, D. Wang, Y. Wang, Y. Li, J. Li, Y. Zhang, Chem. Eng. J. 496 (2024) 153918. [189] S. Akhtarian, S. Kaur Brar, P. Rezai, Biosensors 14 (2024) 445. [190] S. Akhtarian, A. Doostmohammadi, D.-E. Archonta, G. Kraft, S.K. Brar, P. Rezai, Biosensors 13 (2023) 943. [191] B. Uka, J. Kieninger, G.A. Urban, A. Weltin, ACS Sens. 6 (2021) 2738–2746. [192] A. Lamaoui, A. Karrat, A. Amine, Sensor. Actuator. B Chem. 368 (2022) 132122. [193] Y. Cai, L. Cao, H. Cai, W. Yang, H. Lu, A. Adila, B. Zhang, Y. Cao, W. Huang, W. Xu, W. Yang, J. Food Compos. Anal. 139 (2025) 107108. [194] A. Doostmohammadi, K. Youssef, S. Akhtarian, G. Kraft, P. Rezai, Talanta 268 (2024) 125290. [195] L. Wang, B. Li, J. Li, J. Qi, Z. Zhang, L. Chen, Analyst 147 (2022) 3756–3763. [196] P. Azhdary, S. Janfaza, S. Fardindoost, N. Tasnim, M. Hoorfar, Anal. Chim. Acta 1278 (2023) 341749. [197] M.A. Buttkewitz, C. Heuer, J. Bahnemann, Curr. Opin. Biotechnol. 83 (2023) 102978. [198] J.M. Ayuso, M. Virumbrales-Mu˜ noz, J.M. Lang, D.J. Beebe, Nat. Commun. 13 (2022) 3086. [199] S. Xu, L. Wang, Z. Liu, Angew. Chem. Int. Ed. 60 (2021) 3858–3869. [200] C.J. Stephenson, K.D. Shimizu, Polym. Int. 56 (2007) 482–488. [201] B.C.G. Karlsson, J. O’Mahony, J.G. Karlsson, H. Bengtsson, L.A. Eriksson, I. A. Nicholls, J. Am. Chem. Soc. 131 (2009) 13297–13304. [202] L. Chen, X. Wang, W. Lu, X. Wu, J. Li, Chem. Soc. Rev. 45 (2016) 2137–2211. [203] M. Arabi, L. Chen, Langmuir 38 (2022) 5963–5967. [204] M. Arabi, A. Ostovan, A.R. Bagheri, X. Guo, L. Wang, J. Li, X. Wang, B. Li, L. Chen, TrAC, Trends Anal. Chem. 128 (2020) 115923. [205] L. Carballido, T. Karbowiak, P. Cayot, M. Gerometta, N. Sok, E. Bou-Maroun, Chem 8 (2022) 2330–2341. [206] L. Pasquardini, A.M. Bossi, Anal. Bioanal. Chem. 413 (2021) 6101–6115. [207] M. Yoshikawa, K. Tharpa, S.-O. Dima, Chem. Rev. 116 (2016) 11500–11528. [208] L. Wang, K. Zhi, Y. Zhang, Y. Liu, L. Zhang, A. Yasin, Q. Lin, Polymers 11 (2019) 602. [209] C. Zheng, X.-L. Zhang, W. Liu, B. Liu, H.-H. Yang, Z.-A. Lin, G.-N. Chen, Adv. Mater. 25 (2013) 5922–5927. [210] N. Pouyanfar, S.Z. Harofte, M. Soltani, S. Siavashy, E. Asadian, F. GhorbaniBidkorbeh, R. Keçili, C.M. Hussain, Trends Environ. Anal. Chem. 34 (2022) e00160. [211] Z. Wang, Y. Dong, X. Sui, X. Shao, K. Li, H. Zhang, Z. Xu, D. Zhang, npj Flex. Electron. 8 (2024) 35. [212] F.W. Scheller, X. Zhang, A. Yarman, U. Wollenberger, R.E. Gyurcs´ anyi, Curr. Opin. Electrochem. 14 (2019) 53–59. M. Okan et al. Trends in Analytical Chemistry 194 (2026) 118511 25