Enzyme-Based Single Solid-State Nanochannel Biosensors

Abstract

Sensing technologies play a critical role in healthcare, not only for diagnosis and treatment but especially for prevention and early intervention. Recent advances in biology, medicine, and materials science have expanded the landscape of measurable biological markers and enabled the development of nanotechnology-based biosensing platforms. Among the most prominent strategies in biosensing are those that take inspiration from nature, particularly through the integration of biological components such as enzymes. This review focuses on the intersection between enzymatic catalysis and single solid-state nanochannel (SSN) technologies as a promising approach for the development of advanced biosensing devices. We provide an overview of the historical background, current state of the art, and major achievements in enzyme-based biosensors and artificial nanochannel platforms, highlighting their synergistic potential. Particular attention is given to the challenges associated with enzyme integration into artificial environments, including stability and functionality retention, and the strategies employed to overcome them. Finally, we discuss the prospects and limitations of combining enzymes with SSNs, aiming to inspire future research in this emerging and multidisciplinary field.

1. Introduction

The solution is meaningless if the problem is invisible; treatments intended to improve health may become a serious danger without proper diagnosis. These considerations are particularly relevant when it is already too late to prevent a health problem. However, the real importance of sensing technologies in the medical and health fields lies in their ability to assess potential hazards and enable preventive measures to be taken. Great advances in human genome sequencing [1], cellular metabolism, immunology [2], proteomics [3], toxicology [4], pharmacology, neuroscience [5], and human microbiology [6] have enabled the identification of numerous new measurable targets that can be objectively monitored and evaluated as indicators of biological system functionality, pathogenic conditions, or pharmacological responses to therapeutic treatments. In this context, the necessity for high-performance measurement devices is imperative. Fortunately, existing sensing techniques are continuously improving, and new ones are in constant development to meet the growing demand. These advancements enable continuous monitoring, prevention, early diagnosis, and appropriate treatments. All of this is supported by intensive research into new materials and platforms with highly adaptable and multifunctional physicochemical properties, powerful design tools involving AI assistance, and high-precision versatile fabrication techniques [7,8].

Despite all the breakthroughs and successful developments in sensing methodologies, with platforms and devices achieving high sensitivity and sub micromolar detection limits, there will always be limitations or challenges to face in this field. The intrinsic complexity of biological systems and samples represents a primary concern. Specificity, complex matrix operation, miniaturization, in vivo detection capability, wearability, accessibility, and ease of operation are essential features of sensors currently being developed [9]. These attributes facilitate the efficient integration of sensors into diverse applications and ensure their usability by a wide range of users without requiring specialized expertise. This progress has enabled the use of common smart technology devices to digitize, process, store, and transmit healthcare data in real time [10]. Historically, mankind has found in nature the greatest support to its unstoppable evolution, and living systems appear to be an inexhaustible source of inspiration. The field of sensing for healthcare and clinical purposes is no exception. The ability to study biochemical mechanisms in living organisms has not only enhanced our understanding of complex metabolic pathways but also enabled the isolation, modification, design, and production of functional biochemical structures that can be integrated into abiotic systems to develop powerful biosensing tools for specific analytical purposes [11,12,13,14,15].

In this review, the focus will be on two functional biochemical structures that are of paramount importance: firstly, enzymes, which have a long and fascinating history and are foundational to modern biochemistry [16,17,18]. Their relevance in biomarker sensing stems not only from their direct implications in life but also for their ability to function without the need for a living organism, thus forming the basis for the extensive field of biosensing. After the publication of the iconic work “Electrode systems for continuous monitoring in cardiovascular surgery” in 1962 by the man now known as the “father of biosensors” Prof. Dr. L.C. Clark Jr., and his colleague C. Lyons [19], and the subsequent development of the well-known “enzyme electrode” in 1967 by S.J. Updike and G.P. Hicks [20], the scientific community realized the enormous potential of enzymes as analytical tools, highlighting their great versatility and specificity. Today, more than half a century later, enzymes continue to be a key part of advanced biosensing technologies. Their high performance in both sensing and industrial applications, along with the development of new methodologies for enzyme production, preservation, and manipulation, ensures that these efficient biological catalysts remain at the forefront of biosensing technology [21,22].

On the other hand, the study of structural properties of protein channels in biological membranes has led to an understanding of their complex transport mechanisms and facilitated the construction of versatile platforms with numerous applications in the analytical field. The last decade of the past century could be described as “The boom period of protein channel research”. The remarkable work by Prof. Dr. Roderick MacKinnon and Prof. Dr. Peter Agre (Nobel laureates in Chemistry in 2003) on protein channels in biological membranes revealed the implications of shape, size, and surface physicochemical properties on the regulation of matter transport through them [23,24]. Consequently, this understanding opened the possibility of designing and constructing abiotic structures with similar characteristics, aimed at providing robustness while preserving functionality. These new abiotic systems brought the research field into a new stage, where mimicking cellular processes became a central focus, leading to the development of amphiphilic systems as the first approach, because most biochemical reactions take place inside cells or at lipid interfaces [25]. However, sooner rather than later, researchers realized that these synthetic structures are potentially useful for developing other platforms not necessarily related to functions found in living systems. Then, the materials to build channels quickly changed from being amphiphilic molecules, crown ethers, oligomers, or peptides [26,27] to being structures more often completely composed of thermoplastic polymers such as polyethylene terephthalate (PET) and polycarbonate (PC) [28]. Those last polymers have the advantage of disregarding the complexity of protein channels while retaining the desirable functional features [27]. Furthermore, in addition to their increased physical resistance and tolerance to different chemical environments, polymeric membrane-supported channels have the advantage of being able to be designed and constructed in a wider range of different sizes and shapes to modulate transport as required [29,30]. Figure 1 shows the timeline of the nanochannel technology.

Enzymes are highly specialized proteins that perform distinct functions within living systems, exhibiting optimal performance in healthy individuals with adequate biochemical conditions. However, complications arise when they are extracted from their natural environment and placed in different systems with the expectation of achieving similar functionality. Factors such as denaturation, pH, temperature, ionic concentration, lack of cofactors and coenzymes, presence of inhibitors, active site blocking, and accumulation of reaction end products significantly hinder enzyme activity in artificial environments [40]. Consequently, enzyme supports and carriers have been extensively studied to develop immobilization platforms that preserve their maximum functional performance [22,41,42]. In this context, the integration of enzymes into single solid-state nanochannels (SSNs) has emerged as a powerful strategy that combines the strengths of both components to create responsive biosensing platforms. Solid-state nanochannels offer signal amplification and mechanical robustness, providing versatile stable support for enzyme immobilization, while enzymes contribute high selectivity and specific analyte recognition. Notably, the enzyme structure itself often acts as part of the transduction mechanism. This transduction role has been further enhanced in recent years through the incorporation of new elements, such as polyelectrolytes with multiple amphoteric groups and nanoconfinement-induced phenomena like nanoprecipitation, which, together, contribute to improved sensing performance [43,44]. Although this is still a relatively young research area with many open questions, recent progress has demonstrated promising outcomes, including cascade enzymatic reactions and multianalyte detection within a single nanochannel. The aim of this review is to highlight both the recent advancements and the remaining challenges in the development of these platforms. It aims to explore the exciting prospects for the future of the field while reviewing the remarkable work that laid the foundations for this area of research. Multichannel biosensor systems in membranes or porous materials and single nanochannel platforms involving other biological recognition elements than enzymes are not in the scope of this review, and the reader is referred to complete recompilations of nanochannel works [43,44,45,46,47,48].

2. Physicochemical Aspects of Single Nanochannel System

2.1. Single Nanochannel Fabrication: Materials and Methods

Constructing single nanochannels presents a significant challenge. Precise control over nanochannel density requires advanced instrumentation and highly controlled Experimental conditions. The methodologies currently employed for this purpose enable the fabrication of single nanopores with varying sizes, shapes, physicochemical properties, and support materials by controlling processing conditions and precursor materials accordingly.

Table 1. Single nanochannel fabrication methods.

Technology	Materials	Diameter (nm)	Shape	Reference
EBL + RIE	Si3N4	Around 20	Rectangular	[55]
RIE/FIB + IBS	Si3N4	1.8–100	Cylindrical-bowl-shaped cavity	[56]
FIB + IBS	Si3N4/Si	>1	Cylindrical-bowl-shaped cavity	[57]
FIB	Au/Si3N4/Au; PFA/Si3N4/PFA	20–140	Conically narrowing	[58]
EE-PEO	PDMS	>100	Cylindrical	[59]
Bench-top	Glass based	>86	Conical	[60]
Bench-top/laser pulling	Glass based	>40	Hourglass	[61]
Thermal pulling	Glass based borosilicate	Around 30	Surficial (rectangular and conical)	[62]
EB + dry etching + FIB	Silica glass substrate	58–655	Conical-trumpet-shaped	[63]
Ion-track + dry plasma etching	PC	10–1000 variable	Etching-dependent	[64]
Ion-track + UV exposure + chemical etching	PET, PC, PP, PVDF, PI	5–300 variable		[65]
	PET	50–600 variable		[66]
Ion-track + UV exposure + chemical/electrochemical etching	PC	>2 nm variable		[67]
Ion-track + chemical/electrochemical etching	PET, PC, PI, PVDF			[30]
Ion-track + chemical etching	PET, PC, PI, PVDF, PP			[68]
	PET, PC, PI, PVDF, SiO2, Mica			[69]
	PET, PC, PI, PVDF, PP, CR39			[70]
	PET			[71]
	PET, PC	>100 variable		[72]
	PI	10–1000 variable		[73]
	SiN	>100		[74]

Electron beam lithography (EBL) and focused ion beam (FIB) are high-resolution patterning methods used for precise nanochannel definition. Reactive ion etching (RIE), dry etching, and ion beam sculpting (IBS) are used to shape or refine nanochannels post-patterning. Electron beam (EB) irradiation and embedding electrospun polyethylene oxide nanofibers (EE-PEO) enable channel formation through polymer modification or template removal. Bench-top methods, laser pulling, and thermal pulling involve mechanical or thermal drawing of glass or polymer capillaries to achieve tapered nanochannel geometries. Polyethylene terephthalate (PET), polycarbonate (PC), polypropylene (PP), polyvinylidene fluoride (PVDF), polyimide (PI), silicon nitride (Si₃N₄), perfluoroalkyl passivation (PFA), and polydimethylsiloxane (PDMS).

summarizes some of the most frequently used techniques to obtain 1D single nanochannels, along with their relevant properties. Additionally, methodologies for nanopipette fabrication (e.g., bench-top method, laser pulling, and thermal pulling) are included. While nanopipettes share several functional properties with SSNs, such as robustness, a wide range of functionalization possibilities, asymmetric ion transport, and current rectification, technically, they are classified as a distinct group of nanopores [49]. Unlike SSNs, nanopipettes lack geometrical versatility, flexibility, and ease of integration into devices due to their extended length. Although enzymes can also be immobilized at the tip of nanopipettes for sensing applications, these systems will not be covered in this review. Instead, readers are referred to comprehensive and up-to-date reviews on the topic [49,50,51,52,53,54]. While various fabrication approaches will be mentioned below, special attention will be given to the description of methodologies used to fabricate individual SSNs reported as support for enzyme integration in sensing applications.

2.1.1. Ion-Track Chemical Etching

The ion-track etching method is one of the most versatile technologies for producing 1D pores or channels in substrates made of various materials, commonly employed in sensing applications [30,44,45,75]. It is based on the highly accelerated ion irradiation of thin membranes. However, early research in this technology had a different focus. During the 1950s, the optimism surrounding the implementation of nuclear energy and the burgeoning arms race fueled extensive research into nuclear science [76,77]. This surge in interest also influenced many other scientific fields. Naturally, the need for particle detection systems with enhanced performance compared to cloud chambers and nuclear emulsions became imperative—not only to detect particle emissions but also to identify them, enabling a deeper understanding of particle behavior. Consequently, materials susceptible to irradiation damage were employed as detector screens to record the passage of charged particles generated by nuclear fission processes. When a charged particle impacts most insulating solids, it undergoes a deceleration or braking process, transferring its kinetic energy to the target material [78,79]. Depending on the initial energy of the particle and the nature of the target, the linear momentum transfer during the collision can either deflect the particle, altering its trajectory, or allow it to pass through the material in a straight line [69]. In both cases, the insulating nature of the bulk material leads to the ejection of excited electrons from their original positions, creating an “electronic collision cascade” that dissipates energy and generates “atomic collision cascades” through ion repulsion of the newly formed cations [79]. As a result, a typically permanent zone of ionization defects forms along the path of the projectile particle, known as “the track” (Figure 2a) [80].

Almost every latent track in a solid can be transformed into a pore or channel, with the key being the selection of optimal conditions to achieve the preferential dissolution of the “damaged” material [81]. This procedure was first demonstrated by Dr. Young in his 1958 work, “Etching of Radiation Damage in LiF.” Using a solution of FeF₃ in CH₃COOH and HF, he made ion tracks generated by particles from U₃O₈ in a LiF crystal visible through electron micrographs; however, there is some controversy about his work [82]. Four years later, Price and Walker presented electron microscope images of etched charged-particle tracks in various materials, and their findings are also considered the first description of chemical etching [83]. These ground-breaking studies not only expanded the possibilities in particle detection but also laid the foundation for the ion-track etching method, which remains widely used today to fabricate pores and nanochannels [44,45,48,84]. Enlarging ion tracks through chemical etching offers several advantages. This method is effective in revealing tracks created by ions across a wide energy range, including low-energy ions such as those from solar wind. Under optimal conditions, ion tracks can be widened to the point of being visible under an optical microscope while still preserving critical information about the irradiating ion [78]. Beyond its simplicity, this technology is highly versatile. By employing asymmetric chemical etching, incorporating additives, or precisely controlling time and temperature, it is possible to create nanopores with various sizes, shapes, and profiles [30,85,86,87].

Surfactant-assisted etching is the most commonly used method for shaping nanochannels in the development of enzyme-based biosensors. As shown in Table 2, alkyldiphenyloxide disulfonate (Dowfax 2A1) is the most frequently employed surfactant in this context. It is a large anionic surfactant (MW ≈ 543 g/mol) featuring a highly charged polar head composed of two aromatic rings, each bearing a sulfonate group, and a hydrophobic 12-carbon alkyl chain attached to one of the rings (Figure 2a). Dowfax 2A1 is particularly useful in etching processes due to its exceptional chemical stability in both acidic and basic media. In these systems, polyethylene terephthalate (PET) foils are typically used as the nanochannel substrate, and concentrated NaOH solutions are applied to preferentially dissolve the ion-damaged regions of the polymer via basic hydrolysis [88,89,90,91,92]. Under these conditions, the surfactant acts as an etch-rate modulator, selectively slowing the chemical degradation of the bulk material and enabling the formation of asymmetric channel geometries. Most of the understanding of the protection effect generated by surfactant molecules during the etching process is due to the extensive work by Apel and coworkers [66,72,93,94]. As was mentioned before, the difference between the etch rate of the track core and the bulk material mainly determines the channel shape. The mechanism behind it considers that surfactant molecules adsorb on the polymer surface, altering its susceptibility to chemical attack; at the same time, the protective layer formation depends on the available amount of surfactant species in a determinate zone, which is significantly regulated by the restricted diffusion in nanometric spaces due to their size (surfactant layer thickness 2–4 nm) [72]. Figure 2 shows a schematic representation of the process used to obtain a bullet-shaped nanochannel in a PET foil. When the etching solution is in contact with the foil surface, a quasi-solid layer of surfactant molecules is formed on the tip side (the smallest opening) and the etchant molecules have to diffuse through it to begin the material degradation, whereas, at the base, the degradation rate depends only on the free diffusion of OH⁻ and the degraded material. At the moment when the tip diameter is big enough for surfactant penetration, molecules get inside and cover the newly formed wall, hindering the further diffusion of surfactant due to the comparable sizes of the pore and the protective layer; however, OH⁻ diffusion is not hindered. For that reason, the pore growing rate inside is higher and leads to the formation of a parabolic growing profile (Figure 2b). If the process continues, the entrance hole will eventually become large enough to allow the surfactant to pass through and coat the inner walls, reducing the inner radial etching rate and ensuring that the channel profile remains consistent as radial growth progresses (Figure 2c). On the other hand, the axial enlargement rate remains higher along the track core axis. This occurs because, in this region of the “narrow growth front”, the diameter is consistently smaller than the size of the surfactant molecules, preventing the formation of a protective layer. In the final stage, once the channel is fully developed along the axis, only radial enlargement occurs (Figure 2d).

Note that similar principles can be employed to obtain many other shapes. In the case of conical channels, a one-side etching procedure is used: the etching solution is placed on one side of the membrane while a stopping solution is applied to the opposite side (the future tip side), where it serves to neutralize the etching agent. Consequently, the radial growth rate becomes higher along the pore axes from the tip to the base side, leading to the desired profile (Figure 2e) [73,95]. On the other hand, double conical ones can be obtained by two methodologies: a symmetrical etching involving the same etch solution at each side of the membrane [96,97] or a three-step procedure that entails a first fast pre-symmetrical etching from both sides without stopping solution and subsequent one-side etching of each side in the second and third steps, as was described for the conical ones [98] (Figure 2e). These methods often involve on-line transmembrane electrochemical controls as a channel diameter indirect on-line measurement and as an electro-stopping perforation tool as described elsewhere [99,100]. The uses of etching solution additives as organic solvents and UV irradiation have also been reported as mechanisms for controlling geometrical properties in polymeric membrane nanochannels by affecting wettability, surfactant action, and polymer susceptibility to chemical degradation [70,101,102].

Table 2. Fabricated single nanochannels for enzyme integration in sensing applications.

Support	Fabrication Specifications (Tip Side|Base Side) 1	Shape	Etching Time (Minutes)	Etching T (°C)	Size (nm) 2 (Tip/Base/Length)	Ref
PET	Ion-track asymmetrical chemical etching (6 M NaOH + 0.05% w/w Dowfax 2A1|6 M NaOH) 6 min, 60 °C	Bullet	6	60	(90/600/12,000)	[103]
					(76/560/12,000)	[104]
					(55/500/12,000)	[92]
					(60/600/12,000)	[91]
					(No specified)	[90]
					(85/600/12,000)	[89]
					(55/900/12,000)	[88]
Si3N4	Photolithography-RIE-FIB Ga+ 30 kV 1pA	Cylindrical	0.025	Ion beam	(50/50/100,000)	[105]
PI	Ion-track chemical etching (NaClO4|NaClO4), etching stop with addition of 1 M KI	Double conical	Electrochemically controlled etching stop time	25	(25/1300/12,000)	[96]
PET	Ion-track UV-sensitized asymmetrical etching (1 M KCl + 1 M HCOOH|9 M NaOH), etching stop with addition of 1 M KCl + 1 M HCOOH	Conical		30	(30/860/12,000)	[99]
PET		Conical		25	(8/210/12,000)	[106]
PET	Ion-track UV-sensitized etching 35 h (4 M NaOH + 0.02% v/v Dowfax 2A1|6 M NaOH), etching stop with addition of 1 M HCl	Conical		60	(20/2500/12,000)	[107]

Polyethylene terephthalate (PET), polyimide (PI), focused ion beam (FIB), reactive ion etching (RIE). ¹ (Not applicable for double conical and cylindrical channels.) ² (Average diameter of different nanochannels fabricated with the same methodology.)

Table 2. Fabricated single nanochannels for enzyme integration in sensing applications.

Support	Fabrication Specifications (Tip Side|Base Side) 1	Shape	Etching Time (Minutes)	Etching T (°C)	Size (nm) 2 (Tip/Base/Length)	Ref
PET	Ion-track asymmetrical chemical etching (6 M NaOH + 0.05% w/w Dowfax 2A1|6 M NaOH) 6 min, 60 °C	Bullet	6	60	(90/600/12,000)	[103]
					(76/560/12,000)	[104]
					(55/500/12,000)	[92]
					(60/600/12,000)	[91]
					(No specified)	[90]
					(85/600/12,000)	[89]
					(55/900/12,000)	[88]
Si3N4	Photolithography-RIE-FIB Ga+ 30 kV 1pA	Cylindrical	0.025	Ion beam	(50/50/100,000)	[105]
PI	Ion-track chemical etching (NaClO4|NaClO4), etching stop with addition of 1 M KI	Double conical	Electrochemically controlled etching stop time	25	(25/1300/12,000)	[96]
PET	Ion-track UV-sensitized asymmetrical etching (1 M KCl + 1 M HCOOH|9 M NaOH), etching stop with addition of 1 M KCl + 1 M HCOOH	Conical		30	(30/860/12,000)	[99]
PET		Conical		25	(8/210/12,000)	[106]
PET	Ion-track UV-sensitized etching 35 h (4 M NaOH + 0.02% v/v Dowfax 2A1|6 M NaOH), etching stop with addition of 1 M HCl	Conical		60	(20/2500/12,000)	[107]

Polyethylene terephthalate (PET), polyimide (PI), focused ion beam (FIB), reactive ion etching (RIE). ¹ (Not applicable for double conical and cylindrical channels.) ² (Average diameter of different nanochannels fabricated with the same methodology.)

2.1.2. Photolithography-Reactive Ion Etching FIB

Direct writing using focused ion beams (FIBs) has become a widely used nanofabrication tool [108,109]. This technology was introduced during the 1970s and early 1980s following the invention of the liquid metal ion source (LMIS), driven primarily by the growing demand for advanced integrated circuit fabrication techniques [110]. FIBs can be utilized to precisely fabricate nano- and microchannels in solid materials with diverse compositions, including metals, ceramics, semiconductors, and glass-based substrates. Compared to other high-energy particle beams (such as electron or photon beams) used for similar purposes, FIBs offer clear advantages. Their heavier ion mass allows for a higher strike energy with relatively short wavelengths, minimizing forward and back scattering and thereby enhancing precision. Furthermore, the possibility to follow pore growing in real time by SEM or TEM techniques allows for additional control over pore morphology [111,112]. A key parameter in FIB applications is the ability to precisely regulate beam size, current, and energy to effectively remove or add material in localized zones as required [113,114]. FIB methodologies for single nanochannel production are often integrated with other nano- and microfabrication techniques, as these nanochannels are typically supported in thin substrates that are part of microdevices with stringent precision requirements [114]. For example, photolithography, reactive ion etching (RIE), and FIBs were combined to produce cylindrical nanochannels [105]. The process involves sequential steps, beginning with substrate preparation. Low-pressure chemical vapor deposition (LP-CVD) is first used to prepare a three-layer material consisting of a silicon core (300 nm) sandwiched between two silicon nitride (Si₃N₄) membranes (100 nm each) (Figure 3a). Subsequently, optical lithography is used to precisely define the channel’s base aperture by removing the Si₃N₄ membrane on one side by reactive ion etching to expose the underlying silicon substrate (Figure 3b,c), which is then chemically etched with potassium hydroxide (KOH) to reach the Si₃N₄ membrane on the opposite side (Figure 3d). Finally, FIBs are used to create the cylindrical nanochannel through the Si₃N₄ membrane (Figure 3e).

2.2. Single Nanochannel Physical and Chemical Properties

2.2.1. Morphology and Surface Chemistry

Biological channels are renowned for their structural complexity and diversity in shape and size. Their predominantly protein-based architecture provides flexibility to pore walls, enabling remarkable conformational changes in response to specific stimuli and adaptation to environmental variations. Features, such as geometry, diameter, chirality, and surface chemistry have significant influence on selective chemical compound transport, water regulation, and ionic rectification [115]. These transport mechanisms, strongly influenced by channel morphology, became evident early in biological pore research and are now useful in the design and fabrication of synthetic nanopores [115]. However, single solid-state nanochannels used in enzyme-based sensing applications lack the structural flexibility of their biological counterparts, but this comes with the advantage of greater chemical stability and mechanical strength. As evidenced during fabrication, storage, and use, these biomimetic systems do not undergo denaturation (as protein channels do) or structural failure when exposed to harsh acidic or basic conditions or elevated temperatures (see Section 2.1 and Section 2.2). For that reason, the desired size and shape must be controlled during the fabrication procedure [44,45]. As shown in Table 2, the cylindrical and double conical nanochannels are the only symmetric designs employed. In contrast to the asymmetric bullet-shaped and conical-shaped nanochannels, the cylindrical design features an almost uniform diameter along its entire length, whereas the double conical design is characterized by two opposing, gradually tapering conical sections that converge at the central region, forming the narrowest point. These structural features, combined with the greater length of the cylindrical channel, influence the transduction mechanisms, as described below [34]. The inner surface chemistry of nanochannels is another critical parameter in channel sensor development. The functional chemical groups exposed on the surface after pore formation depend on the composition of the support material and significantly influence ion transport mechanisms in unmodified channels. Moreover, subsequent surface modifications depend on the availability of chemical anchor points [84,86].

PET and PI have been extensively utilized as base materials for single and multi-nanochannel arrays in membrane technology due to their advantageous properties [30,70,116]. Both materials share excellent chemical stability, relatively low dielectric constants, insulating nature, and high mechanical strength. However, PI is recognized for its superior performance compared to PET. This performance advantage comes at the cost of a more intricate chemical synthesis process, which translates into higher production expenses. Furthermore, the enhanced resistance of PI presents additional challenges during the etching process. Hot concentrated sodium hypochlorite solutions are often used because certain etchants may become unstable at the elevated temperatures required for processing this material [70]. Silicon nitride is also widely recognized for its excellent mechanical, thermal, and chemical resistance, making it highly suitable for applications subjected to high temperatures and stress [117]. This durability enables it to withstand FIB and EB fabrication techniques without catastrophic structural failure (Table 1). Chemical etching has also been employed for channel fabrication in Si₃N₄ membranes. However, due to its robust properties, more stringent conditions are required to generate nanopores via chemical etching. For example, a study reports the treatment of a thin silicon nitride membrane with concentrated HF or H₃PO₄ solutions at 150 °C for over 20 min to achieve channels with base diameters under 100 nm [74].

Figure 4 illustrates the chemical structures and precursors of PET (a) and PI (b). The synthesis of these polymers is achieved through condensation reactions: terephthalic acid reacts with ethylene glycol to form PET while pyromellitic dianhydride reacts with 4,4′-diaminodiphenyl ether to produce PI [70,118]. During the etching process, basic hydrolysis of these materials generates abundant carboxylate terminal surface groups (-COO⁻). In the case of PI, amine terminal groups (-NH_x) can also be formed on the etched surface [73,118,119,120]. These chemical terminations serve as anchor points for subsequent surface modifications, enabling techniques such as direct chemical attachment or electrostatic self-assembly, as described in Section 3.1. On the other hand, Si₃N₄-based channels, due to their less versatile surface chemistry dominated by SiO₂ groups after FIB fabrication, require additional preconditioning for effective integration of functional structures. For instance, as reported in [105], a hydroxylation treatment with H₂SO₄/H₂O₂ followed by silanization with APTES was employed to introduce amine terminal groups, enabling further surface modification [121] (Figure 4c).

2.2.2. Wettability

Fluids exhibit complex and often unexpected behaviors in nanoscale environments. In particular, the wetting behavior of water in nanoconfined spaces plays a crucial role in matter transport and signal transmission in biological systems [122,123]. Many protein channels utilize gating mechanisms based on wetting and dewetting processes to effectively regulate flux of ions and small neutral molecules [124,125]. Since water serves as the primary solvent in biological systems, solutes cannot pass through dry (dewetted) pores. A key factor in regulating wettability within water channels is the hydrophobicity of the pore walls. It has been demonstrated that uncharged surfaces within nanochannels promote water evaporation, resulting in the formation of local vapor pockets. This leads to a non-conductive (closed) state at which no transport occurs [126]. Additionally, studies have shown that, even in channels with low surface charge, a closed state is observed when the channel diameter falls below 20 nm [127]. These insights highlight the importance of wettability as a critical parameter in the development of single nanochannel enzymatic biosensors. Since biosensing mechanisms depend on continuous mass transport through the channel to facilitate enzyme–substrate interactions, physicochemical transduction processes, and ionic current measurements, maintaining the conductive (open) state is essential. Consequently, hydrophilic channels with larger diameters and highly charged surfaces are preferred to ensure optimal functionality.

2.3. Transport in Single Nanochannels

‘There is plenty of room at the bottom,’ were the iconic words of Professor Richard Feynman, highlighting the vast potential to decipher and mimic the complex processes taking place in biological systems at the micro- and nanoscale [128]. Although he was not referring to physical space, it is interesting that a deeper understanding of these processes has revealed that, in the case of pores and nanochannels, reduced space is the key factor underlying the unique physicochemical mechanisms governing matter transport. For example, in biological channels with an effective diameter below 10 nm, the confined size has significant implications on the regulation of water, ion, and neutral molecules transporting through them [129]. At the nanoscale, interactions that are negligible in bulk volumes become dominant, while conventional bulk laws lose influence as dimensions decrease [130,131].

2.3.1. The Nanoconfinement Effects

Beyond the influence of surface physicochemical properties on various processes occurring in nanochannels, nanometric confinement itself acts as a powerful regulator of multiple physical and chemical phenomena. For instance, it has been reported in theoretical and experimental studies that acid–base equilibria can shift significantly when dissociable groups are confined at the nanoscale, primarily due to repulsion effects arising from restricted space [132,133,134]. Additionally, it was observed that the electrostatic self-assembly of polyelectrolytes behaves differently inside nanochannels compared to bulk surfaces, with these deviations attributed to the distinct structural organization that polymers adopt under nanoconfinement [135,136]. Similarly, crystallization, precipitation, and adsorption processes are influenced by nanoscale confinement, leading to phenomena such as the formation of metastable crystalline phases not observed in bulk systems, localized precipitation/dissolution gating mechanisms, and preferential adsorption [137,138,139]. Enzyme activity is also affected at this scale, though responses vary depending on the specific system. Some studies report a reduction in activity due to steric hindrance, whereas other studies observe an enhancement attributed to shorter substrate diffusion distances [140,141]. In terms of sensing performance, most single-nanochannel-based systems exhibit linear detection zones in the nano- to micromolar range. This sensitivity is largely attributed to the concentration effect caused by species accumulating in the femtoliter-sized volumes within the narrowest sections of the channel. Additionally, at these shorter distances, both substrate–enzyme interactions and the coordinated functions among functional elements are enhanced, further improving sensing performance.

2.3.2. Forces Regulating Ion Behavior

Due to the continuous increase in the surface-to-volume ratio, interfacial effects such as steric interactions and hydration (~1–2 nm range), Van Der Waals forces (~1–50 nm range), and electrostatic interactions (~1–100 nm range) gain crucial importance in nanoconfined spaces [142]. Phenomena such as ultrafast liquid flow, ion rectification, nanoprecipitation, nanoevaporation, and polarization concentration deviate considerably from bulk behavior, necessitating approaches beyond classical theories to accurately explain them [143,144]. In this section, we will discuss some of the most widely accepted models used for this purpose, focusing on those that help to elucidate the elegant transduction mechanism converting chemical processes into a measurable electrical response. Models based on evaporation and pressure-driven flow will not be considered here as they are not relevant for enzyme-based sensors; for a detailed description of general transport phenomena in nanometric and sub-nanometric structures, the reader is referred to [142,145,146,147].

As mentioned above, water remains essential as a medium for matter transport at these scales. Consequently, a nanochannel placed between two bulk reservoirs that is completely filled with an aqueous electrolyte solution will be considered, and the non-conductive state resulting from drying is disregarded (Figure 5a). Given the high sensitivity of electrostatic charges to changes in solution pH, the inner channel surface acquires an electrostatic charge as described by Equation (1) (given in C m⁻²) due to the dissociation of groups on the inner wall and/or the specific non-electrostatic adsorption of ions from the solution [148,149,150,151] (except at the point of zero charge, when electroneutrality is reached at a specific pH). It will be shown that the presence of surface charges has significant implications on transport.

$${\sigma }_s=\sum _{i}F{\Gamma }_i{q}_i$$

(1)

where $q_i=z_{i}e$ is the net charge of i, z is the valence of ion i, e is the electron charge (1.602 × 10⁻¹⁹ C), ${\Gamma }_i$ is the surface density of i (mol m⁻²), and $F$ is the Faraday constant.

When fixed charges are present on the channel wall, counterions migrate from the bulk to restore electroneutrality, resulting in the formation of a screening cloud composed of regions with varying ionic concentration and mobility, collectively known as the electrical double layer (EDL). The EDL is typically divided into two main zones: a compact layer (Stern layer) of counterions with limited mobility immediately adjacent to the channel wall, next to a diffuse layer (Gouy–Chapman layer) where the concentrations of ions and counterions decay exponentially toward the bulk [152,153] (Figure 5b). The thickness of the EDL depends on both the surface charge density and the electrolyte concentration. For instance, at high electrolyte concentration or low surface charge density, the screening cloud is thin because the potential generated by the fixed charges only has a short-range effect. Conversely, at high surface charge density or low electrolyte concentration, the EDL extends further into the solution [146,154]. The electrostatic potential profile in space, arising from the charge distribution at the solid–liquid interface, is described by the Poisson–Boltzmann equation (Figure 5c).

$${\nabla }^2\psi ={\displaystyle \frac{e}{{\epsilon }_0{\epsilon }_r}}{\sum }_i{n}_i^{\infty }{z}_{i}exp\left[{\displaystyle \frac{{-z}_{i}e\psi \left(x\right)}{k_{B}T}}\right]$$

(2)

where $\psi$ is the electric potential (V) due to the surface charge, ${\epsilon }_0$ is the vacuum permittivity (8.854 × 10⁻¹² F m⁻¹), ${\epsilon }_r$ is the relative permittivity, $e$ is the electron charge, $z_i$ is the valence of $i$, $n_i^{\infty }$ is the bulk volume density, $x$ is the spatial coordinate perpendicular to the charged plane surface (m), $k_B$ is the Boltzmann constant (1.381 × 10⁻²³ J K⁻¹), and $T$ is the absolute temperature (K).

This equation can be solved analytically by assuming a small surface potential in the EDL ($z_i\psi$ < 0.0257 V at 25 °C) and expanding the exponential function using the first-order linearization $e^{-\alpha }=1-\alpha$ for small $\alpha$. This leads to the Debye–Hückel approximation (Equation (3)), where the $\kappa$ parameter, defined as the Debye–Hückel parameter, is related to the Debye length $\left({\lambda }_D\right)$, which corresponds to a measure of the thickness of the EDL ${\lambda }_D={\kappa }^{-1}$ [155,156].

$${\nabla }^2\psi ={\kappa }^2\psi \left(z\right)$$

(3)

being $\kappa ={({\displaystyle \frac{e^2{\sum }_i{n}_1^{\mathrm{\infty }}{z}_i^2}{{\epsilon }_0{\epsilon }_r{k}_B}})}^{\frac{1}{2}}$.

For a symmetric $z_i:z_i$ electrolyte in water at 25 °C, ${\lambda }_D$ (nm) can be calculated using Equation (4). Under these conditions, the electric potential extends approximately ${3\lambda }_D$ before decreasing to about 2% of its initial value. Considering the typical EDL thicknesses in KCl solutions (

Table 3. Debye length ${\lambda }_D$ in usual KCl solutions calculated from Equation (3) based on the approximation ($z_i\psi$ < 0.0257 V at 25 °C).

KCl (M)	$\mathbf{Debye}\mathbf{Length}{\mathit{\lambda }}_{\mathit{D}}$ (nm)
10	0.3
10−1	1.0
10−2	3.1
10−3	9.6
10−4	30.5
10−5	96.3

), it is evident that, for concentrations below 1 mM, the surface-charge-induced potential variation can extend up to 30 nm from the surface. Consequently, in a nanochannel with a diameter of ~60 nm, the entire radial aperture will be influenced by the EDL potential (Figure 5a) [145].

$${\lambda }_D={\displaystyle \frac{0.215}{\sqrt{I_s}}}$$

(4)

where $I_s$ is the ionic strength, defined as $I_s={\displaystyle \frac{1}{2}}\sum c_i{z}_i^2$.

Although the Debye screening model is useful for describing symmetric electrolyte distributions at low surface potential and enables the calculation of key EDL parameters, it has limitations when accurately representing high surface potential distributions or distinguishing the individual contributions of ions and counterions to charge balance in asymmetric electrolyte systems. To overcome these limitations, the Gouy–Chapman model, derived from a numerical solution of the Poisson–Boltzmann equation, is used [154,157]. This model demonstrates that charge balance is predominantly achieved through counterion accumulation at the interface (Figure 5d). As a result, the ionic concentration in the EDL cloud is significantly higher than in the bulk [158]. This effect plays a crucial role in the high surface conductance observed at charged interfaces in an electrostatic aqueous medium when an electric field is applied. The electric field overcomes the attractive electrostatic force between counterions and the charged surface, inducing the movement of the accumulated ions. This motion, in turn, drags water and neutral solutes along the direction of the ionic flux, thereby facilitating and enhancing matter transport [159,160].

Note that liquid, solute, and ion transport are strongly coupled at the nanoscale. Although concentration gradients can drive osmotic flow, during single nanochannel sensing, the bulk reservoir composition is generally symmetric, and changes in the concentration of uncharged substrates during measurement by enzyme reactions are not enough to generate a measurable flux. Therefore, electro-osmotic flow by voltage gradient is the governed regime of transport in these systems [143,145,161].

2.3.3. Signal Sources in Nanochannels: Iontronic Current, ICR, and TEs

A basic understanding of the forces responsible for ionic movement at interfaces with dimensions comparable to the Debye length has been briefly summarized from the perspective of the channel wall because this viewpoint highlights the direct influence of surface charges on the local ionic environment. Now, let us review the models used to describe ion concentration distribution and ionic current throughout the entire nanochannel. For instance, let us consider the same cylindrical channel with a diameter of 60 nm in a 1 mM KCl solution (Figure 5a). In this scenario, the surface charge directly influences the overall potential distribution; consequently, it also governs the ion concentration inside the channel, which in turn affects the ionic current [162,163]. Under these conditions—characterized by high surface charge or low salt concentration—the ion concentration inside the channel is higher than in the bulk electrolyte [164]. This occurs because counterions accumulate inside the channel to neutralize the fixed surface charge, while the co-ion amount also increases in the diffuse layer due to satisfied electroneutrality. This concentration imbalance, known as the excess mobile counterion concentration, $c_e={\displaystyle \frac{2{\sigma }_s}{r{N}_{A}e}}$, is a concentration threshold that must be considered for obtaining the model for nanochannel conductance $\left(G\right)$ (Equation (5)) [143]. This approximation accounts for the fact that, at ionic strengths higher than $c_e$, conductance primarily depends on the channel geometry (first term). However, at ionic strengths lower than $c_e$, surface charge dominates the conductance, making it less dependent on the geometrical factor (second term). It is important to highlight that, although this equation refers to a symmetric cylindrical nanochannel behaving similarly to an ohmic resistor, the same principle applies to asymmetric ones as well [145].

$$G=\left({\mu }_{K^{+}}{+\mu }_{{Cl}^{-}}\right)n_{KCl}e{\displaystyle \frac{\pi r^2}{l}}+2{\mu }_{K^{+}}{\sigma }_s{\displaystyle \frac{2\pi r}{l}}$$

(5)

where ${\sigma }_s$ is the surface charge, $\mu$ is the mobility (m² V⁻¹ s⁻¹), $n$ is the volume density (m⁻³), $r$ is the radius (m), and $l$ is the length of the nanochannel (m).

The conductance of a single SSN nanochannel is defined as $G={\displaystyle \frac{I}{∆V}}$. Current/voltage (I/V) curves are obtained by applying a DC voltage and recording the steady-state current using the setup shown in (Figure 6a). The silver–silver chloride (Ag|AgCl) electrode is commonly used in nanofluidic measurements due to its stability, non-polarizability, and minimal capacitive effect [143]. These properties are essential as the electrodes are responsible for both current measurement and voltage application. A typical I/V curve recorded with a symmetrical cylindrical nanochannel with homogeneous surface charge distribution exhibits a straight line with slope G, characterized by identical absolute current values for voltages of equal magnitude but opposite polarity. As the geometry and surface charge are symmetrical across the channel, the electrostatic field and ionic cloud are also homogeneously distributed. Consequently, the channel exhibits ohmic behavior, at which the current varies linearly with the applied voltage [144,165] (Figure 6b). It is important to note that no faradaic processes occur in this system; therefore, the recorded signal arises solely from ion migration. As a result, this output is referred to as an iontronic current, which, at these conditions, depends strongly on the surface charge. This characteristic makes it inherently useful as a response variable in sensing applications, as any physicochemical process that modifies the surface charge can be detected through changes in the iontronic output [43,44,45,166].

Figure 6c presents a different system where the I/V curve is recorded using an asymmetrical bullet-shaped nanochannel. Due to the broken geometrical symmetry, the potential distribution is no longer homogeneous across the channel volume, leading to special transport regimes [167]. Note that our previous discussion on surface charge control over ion transport applies primarily to the narrow tip region (~60 nm). As the channel widens toward the base (~700 nm), the electric double layer loses influence over ion transport due to its limited extension relative to the channel diameter. When a +1 V potential is applied to the electrode near the tip of a negatively charged channel, an ionic current (I) is recorded. However, when the polarity is reversed (−1 V), the net ionic current decreases significantly, often by several orders of magnitude (Figure 6c, red curve). This phenomenon, known as ion current rectification (ICR), has been extensively studied in previous works [146,168,169]. Briefly, EDL overlaps at the channel tip, enhancing counterion permeability while hindering the flow of co-ions [131]. This results in an ionic accumulation/depletion mechanism governed by the applied potential polarity, producing a diode-like behavior characterized by a high- and low-conductance branch in the I/V curve. ICR is a fundamental mechanism in enzyme biosensing with single nanochannels, as most nanochannels used in these applications feature asymmetric potential distributions (see Table 4 below).

Similar to the iontronic output in symmetrical systems, ICR is highly sensitive to surface charge changes [162,170]. Thus, the rectification efficiency (f_rec), defined as the ratio of the current in the high-conductance state to that in the low-conductance state at a fixed voltage, serves as a key response variable. Moreover, ion current rectification enables the qualitative characterization of the nanochannel’s electrostatic surface charge properties, as both cation-driven and anion-driven rectification can be achieved and distinguished within the same channel, depending on surface charge polarity. Notably, the broken symmetry required for ICR can result not only from asymmetric geometry but also from asymmetric surface charge modifications or from the presence of different electrolyte solutions in the reservoirs [171].

The physical analysis of transport phenomena in nanochannels has been approached from different perspectives, each successfully explaining experimental results and predicting ionic current responses under specific system conditions [172]. In the case of single nanochannels embedded in polymeric membranes, the Poisson–Nernst–Planck (PNP) model is widely employed to describe I/V curves and ion current rectification properties. Furthermore, it allows for the prediction of ion concentration profiles and electric potential distribution within the channel. Since obtaining analytical solutions for the 3D PNP model will not be possible, simplified 1D equations are typically used. The ionic flux for each species is described by the Nernst–Planck equation (Equation (6)), while the relationship between electric potential and ion concentration is governed by the Poisson equation (Equation (7)). Finally, the Navier–Stokes equation accounts for the flow distribution, incorporating both pressure-driven and electrokinetic contributions (Equation (8)). To fully define the system, numerical solutions, e.g., finite-difference or finite-element methods with appropriate boundary conditions considering steady-state conditions $\nabla j_i=0$ and incompressible fluid flow $\nabla u=0$, are used [173,174].

$$J_i={-D}_i\nabla c_i-{\displaystyle \frac{z_{i}F}{RT}}{D}_i{c}_i\nabla \Phi +c_{i}u$$

(6)

$${\nabla }^2\Phi =-{\displaystyle \frac{F}{\epsilon }}\sum z_i{c}_i$$

(7)

$$u\nabla u={\displaystyle \frac{1}{\rho }}(-\nabla P+\eta {\nabla }^{2}u-F({\sum }_i{z}_i{c}_i)\nabla \Phi )$$

(8)

where $J_i$, $D_i$, $c_i$, and $z_i$ are the flux (mol s⁻¹ m⁻²), diffusion coefficient (m² s⁻¹), concentration, and charge of species $i$, respectively. $\Phi$ and $u$ are the total electric potential (V) and fluid velocity (m s⁻¹), and F, R, and T are the Faraday constant, gas constant, and absolute temperature, respectively. $\epsilon$ is the dielectric constant, $\rho$ is the density (kg m⁻³), $\eta$ is the viscosity of the fluid (kg m⁻¹ s⁻¹), and $P$ is the pressure (N m⁻²).

Another important mechanism of response modulation in both symmetrical and asymmetrical systems is the variation in the effective channel size. Since the iontronic signal depends on ion transport, any variation in the cross-section area inside the channel will be reflected in changes in the net recorded current. Partial or total channel blockage—or, conversely, unblocking—can result from analyte attachment or release, reorganization of supramolecular structures on the surface, or the passage of large molecules or particles through the channel. This sensing strategy has been less explored in single nanochannel enzyme biosensors. However, translocation events (TEs) associated with a partial reduction in the effective channel size have been also used as a signal correlated with analyte concentration [175].

3. Sensing Platform Nanoarchitecture

A bare nanochannel alone can serve as a valuable tool for ion sensing; however, uncharged biomarker molecules typically pass through undetected at clinically relevant concentrations, unless they are exceptionally large (e.g., macromolecules such as proteins, RNA, or DNA). As observed in nature, achieving specificity and sensitivity in detection or regulation of a particular substrate requires a well-organized assembly of functional elements operating in concert [176,177]. Therefore, the design and construction of biomimetic sensing platforms must incorporate a similar level of structural organization. In this section, we will review the architecture of biosensing single nanochannel systems, focusing on the integration of recognition, signal transduction, and amplification elements onto the channel wall and how their coordinated functions allow for the transformation of the substrate amount into measurable current.

3.1. Functional Blocks Integration

3.1.1. One-Block Systems

As explained in Section 2.2 and illustrated in (Figure 4), the availability of reactive chemical groups after nanochannel fabrication serves as anchoring sites for the integration of subsequent building blocks. Although most reported systems incorporate transduction–amplification elements in addition to the nanochannel wall itself, some studies have demonstrated that a recognition element alone (one block) can suffice for substrate sensing. In this simple architecture, one or more enzymes are directly immobilized on the surface via covalent coupling through the well-established 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC)-N-Hydroxysuccinimide (NHS) zero-length crosslinking methodology [96,99,105,107]. Briefly, EDC reacts with carboxyl groups to form a highly reactive O-acylisourea intermediate, which readily undergoes nucleophilic addition. To prevent hydrolysis, NHS is introduced, forming a semi-stable N-hydroxysuccinimidyl (NHS) ester. Finally, primary amine groups from enzymes react with the NHS-ester, generating a stable amide bond that covalently links the enzyme to the surface [178,179] (Figure 7a,b). Simplicity and effective immobilization are the main advantages of this methodology; however, it has been reported that enzyme activity can be affected by active site blockage or structural alterations resulting from covalent bond formation.

3.1.2. Two-Block Systems

Platforms incorporating additional transduction–amplification elements (two-block systems) introduce an extra constructional block aimed at enhancing overall performance and expanding the possibilities for recognition element attachment. While these systems are more complex than single-block configurations, their complexity varies depending on the number of layers of a specific block incorporated. For instance, different designs have been developed by alternating one, two, three, or four layers of each block. Among the additional blocks reported to date, cationic polyelectrolytes have been the most widely employed (Table 4). These large polymer chains provide a high density of cationic charges with diverse acid–base equilibrium constants and enable enzyme immobilization through simple layer-by-layer electrostatic self-assembly [88,89,90,91,92]. Although the exact procedure depends on the specific polyelectrolyte or enzyme used, the general methodology involves the initial self-assembly of a polyelectrolyte layer at a specific pH, ensuring a positively charged polymer and a negatively charged bare pore wall. The modified channel is then exposed to an enzyme-containing solution at an optimized pH to promote electrostatic interactions. This process can be repeated as needed to achieve the desired number of layers for each block (Figure 7d,e). Despite not requiring complex chemistry, these procedures demand precise pH control and, in some cases, extended fabrication times. Additionally, since electrostatic interactions depend on surface charges, these architectures exhibit high pH sensitivity. Regarding catalytic performance, enzyme immobilization via electrostatic self-assembly onto polyelectrolyte surfaces has been reported as one of the methods with minimal impact on enzymatic activity [180,181]. This is attributed to the intrinsic ability of polymers to reorganize their structure and the absence of covalent bond formation, which helps to preserve enzyme functionality [182,183].

A recent approach reported by Huamani et al. introduces a novel procedure that incorporates a one-pot metal–organic framework (MOF) as a structural component. This porous framework serves as a source of positive charge groups due to the presence of Zr⁴⁺ ions and amino functional groups. Additionally, the amino groups act as anchor points for subsequent enzyme attachment. The immobilization strategy utilizes divinyl sulfone (DVS) as a cross-linking agent to covalently bind urease to the amino-functionalized UiO-66-NH₂ framework (Figure 7c). DVS is a bifunctional electrophilic agent that reacts with amines, thiols, and hydroxyls under mild aqueous conditions, forming stable sulfonate linkages [184,185,186]. This enables the formation of robust MOF-DVS–enzyme covalent bonds through a two-step anchoring process. First, the nanochannel wall is activated with DVS and, subsequently, the pre-activated surface is exposed to an aqueous enzyme solution. This method offers advantages such as mild reaction conditions and enhanced structural stability, thereby expanding the operational range of the sensing platform. However, one-pot MOF synthesis is a labor-intensive procedure requiring specific nanochannel preconditioning as well as precise control over temperature and pressure conditions [103].

Table 4. Single nanochannel enzyme-based biosensor nanoarchitectonics.

Building Block Layers	Blocks Integration	Transduction Mechanism	Ref.
UiO-66-NH2-Urease	One-pot MOF synthesis, DVS-mediated crosslinking by drop-coating	pH change effect over frec, single reaction	[103]
PAH/ADA/PAH	Layer-by-layer self-assembly by dip-coating and drop-coating	pH change effect over frec, single reaction	[104]
PAH/Urease/PAH/Urease: Arginase	Layer-by-layer self-assembly by dip-coating	pH change effect over frec, cascade concerted functions	[92]
PEI/CD/PEI	Layer-by-layer self-assembly by dip-coating and drop-coating	pH change effect over frec, single reaction	[91]
PAH/Urease	Layer-by-layer self-assembly by dip-coating	pH change effect over frec, steric obstruction	[90]
PEI/AchE	Layer-by-layer self-assembly by dip-coating and drop-coating	pH change effect over frec, single reaction	[89]
PAH/Urease	Layer-by-layer self-assembly by dip-coating	pH change effect over frec, single reaction	[88]
HRP	EDC-NHS covalent coupling	Translocation events (ABTS●+ aggregates)	[105]
GOx/HRP		pH change influence over transmembrane current cascade concerted functions	[96]
GOx		pH-sensitive, single reaction	[99]
HRP		Steric obstruction, specific interaction	[106]
HRP		ABTS●+-sensitive, electrostatic interaction, and steric obstruction	[107]

Amino-terephthalate-based zirconium metal–organic framework (UiO-66-NH₂), polyallylamine hydrochloride (PAH), polyethyleneimine (PEI), adenosine deaminase (ADA), creatinine deaminase (CD), acetylcholinesterase (AchE), horseradish peroxidase (HRP), glucose oxidase (GOx), rectification factor (f_rec).

Table 4. Single nanochannel enzyme-based biosensor nanoarchitectonics.

Building Block Layers	Blocks Integration	Transduction Mechanism	Ref.
UiO-66-NH2-Urease	One-pot MOF synthesis, DVS-mediated crosslinking by drop-coating	pH change effect over frec, single reaction	[103]
PAH/ADA/PAH	Layer-by-layer self-assembly by dip-coating and drop-coating	pH change effect over frec, single reaction	[104]
PAH/Urease/PAH/Urease: Arginase	Layer-by-layer self-assembly by dip-coating	pH change effect over frec, cascade concerted functions	[92]
PEI/CD/PEI	Layer-by-layer self-assembly by dip-coating and drop-coating	pH change effect over frec, single reaction	[91]
PAH/Urease	Layer-by-layer self-assembly by dip-coating	pH change effect over frec, steric obstruction	[90]
PEI/AchE	Layer-by-layer self-assembly by dip-coating and drop-coating	pH change effect over frec, single reaction	[89]
PAH/Urease	Layer-by-layer self-assembly by dip-coating	pH change effect over frec, single reaction	[88]
HRP	EDC-NHS covalent coupling	Translocation events (ABTS●+ aggregates)	[105]
GOx/HRP		pH change influence over transmembrane current cascade concerted functions	[96]
GOx		pH-sensitive, single reaction	[99]
HRP		Steric obstruction, specific interaction	[106]
HRP		ABTS●+-sensitive, electrostatic interaction, and steric obstruction	[107]

Amino-terephthalate-based zirconium metal–organic framework (UiO-66-NH₂), polyallylamine hydrochloride (PAH), polyethyleneimine (PEI), adenosine deaminase (ADA), creatinine deaminase (CD), acetylcholinesterase (AchE), horseradish peroxidase (HRP), glucose oxidase (GOx), rectification factor (f_rec).

3.2. Recognition and Transduction Mechanism

The transduction mechanism in reported single-block platforms follows two distinct pathways depending on the specific system. The first mechanism arises from the incorporation of enzyme electrostatic charge groups into the overall surface charge distribution upon immobilization. Since the proteins’ structure contains amino acids with amphoteric functional groups, the net surface charge results from the balance of the entire charge cloud under specific environmental conditions, effectively functioning as a supramolecular amphoteric arrangement. In this context, the presence of charged groups or charged substrates can modify the net surface charge, consequently altering ion concentration within the EDL, which leads to variations in the recorded current. Thus, the generation of new charged species due to pH fluctuations or chemical reactions is sensitively transduced into changes in the I/V response (Figure 8a) [47,187]. The second transduction pathway involves a reduction in the effective channel size. In this case, partial and transient obstruction at the channel tip occurs due to the passage of small agglomerates formed within the channel as a result of nanoconfinement-driven accumulation of reaction products. When the channel aperture is temporarily narrowed, a translocation event (TE) is recorded as a spontaneous decrease in current (Figure 8b). The frequency and amplitude of these TEs depend on the applied potential; however, by setting a defined bias, a correlation with substrate concentration has been reported, enabling indirect analyte determination. It is important to clarify that the simplicity of single-block platforms, as previously mentioned, refers to their construction rather than their sensing mechanism. In fact, both transduction mechanisms can coexist within the same system, and complex concerted cascade enzymatic reactions may be required for the detection of specific analytes [105,107,188,189].

Beyond the assembly advantages provided by the integration of additional blocks in two-block systems, certain functional properties are also enhanced. For instance, in platforms exhibiting ICR, the presence of additional charges with opposite polarity enables both cationic and anionic driving sensing signals due to a broader distribution of pH-sensitive charges. In contrast, single-block systems typically allow for only cation- or anion-driven signal recording. This enhancement in pH response not only broadens the substrate detection range but also improves sensitivity and detection limits in some cases. The transduction mechanisms in two-block systems are generally based on the same principles as those in single-block platforms. However, in this case, the net surface charge arises from the collective contributions of charged groups from the pore wall, enzymes, and the additional constructional block (e.g., polyelectrolytes or MOFs) (Figure 8c). This composite amphoteric arrangement becomes more susceptible to charge state variations due to protonation and deprotonation processes across a wider range of acid–base equilibria, thereby increasing sensitivity to pH changes induced by the generation of acidic or basic reaction products [47,187]. Due to the pH-driven sensing mechanism, special care must be taken when electrostatic self-assembly-based architectures are built, as excess substrate concentrations can generate a sufficient amount of reaction products to induce significant pH shifts within the nanochannel, potentially leading to the detachment and loss of functional elements. This effect is further exacerbated by the continuous flux during measurements [91,104]. As mentioned earlier, the number of building block layers can vary among different systems, directly influencing recognition and transduction phenomena. The incorporation of additional recognition element layers has been shown to significantly improve detection limits, sensitivity, and the concentration response range. Additionally, in platforms utilizing enzymatic cascade reactions, the specific enzyme ratio and layer organization play a critical role in optimizing these performance parameters [92]. However, the incorporation of additional layers also impacts the effective channel size and fabrication complexity. Also, increasing the number of layers can lead to lower recorded currents due to a reduction in channel diameter, and, in some cases, complete channel blockage may occur, posing a challenge to system performance and reproducibility [91,104].

From a kinetic perspective, iontronic signals arising from enzyme recognition and transduction mechanisms in single nanochannel platforms have recently been addressed theoretically by Pérez Sirkin and coworkers [190]. In their study, the influence of kinetic parameters related to enzymatic reactions and the diffusion/migration of substrates and products was analyzed to accurately describe and model experimental results obtained in the one-block system developed by Lin et al. [96]. Interestingly, their findings showed that substrate concentration at the channel wall is not a limiting factor. Instead, radial diffusion rapidly equilibrates substrate levels at the surface. In fact, calculations revealed that substrate concentration remains above 60% of the bulk concentration even at the narrowest region of the channel, ensuring its availability for enzyme interaction and continuous reaction. Moreover, when the bulk concentration is much lower than K_M, substrate concentration does not appear to be a limiting factor for system saturation.

It could be expected that reaction products will diffuse rapidly to the center of the channel and exit to the bulk, potentially compromising the detection signal. However, reaction products charged with a charge opposite to that of the channel wall have limited diffusion due to electrostatic interactions with the surface, and this accumulation is the main reason for the surface charge changes, leading to variations in the iontronic output. Nevertheless, excessive accumulation also induces system saturation. For example, the retention of H⁺ at the surface results in a localized pH drop, which influences enzyme activity and further affects sensitivity. As a result, larger concentration changes are required to produce a significant iontronic response. Moreover, the neutralization of negatively charged groups occurs more rapidly due to accelerated pH reduction, rendering the channel unresponsive to further substrate additions. This effect is more pronounced in longer nanochannels and in enzymes with higher rate constants (k), as extended channel lengths hinder proton diffusion and kinetically favored reactions lead to faster pH changes. Although these findings are based on a one-block system, the development of a two-block system incorporating multilayers revealed that the presence of additional amphoteric groups with varied acid–base equilibria significantly enhances the sensing range before saturation occurs (See Section 4, Table 5).

4. Enzyme-Based SSN Platforms

Enzyme-modified single nanochannel sensing systems are highly versatile technologies. As enzymes play fundamental roles in metabolic pathways, enzymatic reaction substrates and products represent a valuable source of information useful in a clinical context. For that reason, enzyme integration into a nanochannel-based system, where biochemical processes can be transduced into measurable electrical signals, is of great interest. To develop an effective sensing platform by incorporating enzymes into single nanochannels, several key considerations must be addressed to optimize the stimulus–response mechanism. As discussed earlier, enzymatic reaction products must induce detectable changes in the nanochannel’s physical or chemical properties to alter conductance and generate a measurable electrical signal. This effect can occur directly, as in the case of pH variations resulting from the production of acidic or basic species (Table 4), or indirectly, mediated by an intermediary. For instance, some studies have demonstrated how ABTS acts primarily as a charge carrier [105,107]. By interacting with the enzyme and undergoing electron transfer, ABTS becomes ionized and subsequently aggregates. This process alters the electric double layer potential distribution and affects the effective nanochannel size, ultimately influencing the ionic transport and sensor response.

Table 5 summarizes small molecule biomarkers that have been analyzed with enzyme-modified nanochannels and some important functional parameters. In general, these systems achieve sub-micromolar detection limits and wide dynamic ranges, which represent a significant advantage—not only due to their ability to detect low-concentration analytes but also because they allow for sample dilution to minimize interferences [91]. A major benefit of these platforms is their simplified measurement setup. In most cases, only voltage control and low-noise current amplification are required, which are functionalities that are widely available in commercial potentiostats. Furthermore, measurements can be performed under ambient conditions, and the membrane materials used are highly stable and readily accessible. Regarding measurement time, iontronic response changes occur within seconds of substrate exposure. However, full signal stabilization typically takes between 1 and 10 min, depending on the specific system and with some exceptions.

Table 5. Single nanochannel enzyme-based biosensors.

Year	Target	LOD (µM)	Dynamic Ranges (µM)	Response Time (min)	Ref.
2025	Urea	10	10–10,000	≈1	[103]
2025	Adenosine	0.01	0.01–500	4	[104]
2025	Arginine	3	30–3000	5	[92]
2025	Creatinine	0.005	0.005–100	2.5	[91]
2024	Urea	0.0001	0.0001–0.01 and 0.001–1	≈3	[90]
2022	Acetylcholine	0.016	0.001–22.5 and 25–100	≈3	[89]
2018	Urea	0.001	0.001–1	1.5	[88]
2017	H2O2	500	500 Qualitative analysis	0.01–1.4	[105]
2014	Glucose	15	15–100	≈10	[96]
2014	Glucose	0.001	0.001–1000	N/F	[99]
2011	Concanavalin A 1	10	10 Qualitative analysis	180	[106]
2011	H2O2	0.01	0.01–1	0.8–1.6	[107]

Approximated response times are estimated from percentage of response change over time plots if the response time is not specifically defined. ¹ Although concanavalin A is not the enzyme’s substrate, its immobilization inside the nanopore via glycoprotein–lectin interactions reduces the available area for ionic transport. This blocking effect can be exploited to modulate the nanopore’s conductance and selectivity.

Table 5. Single nanochannel enzyme-based biosensors.

Year	Target	LOD (µM)	Dynamic Ranges (µM)	Response Time (min)	Ref.
2025	Urea	10	10–10,000	≈1	[103]
2025	Adenosine	0.01	0.01–500	4	[104]
2025	Arginine	3	30–3000	5	[92]
2025	Creatinine	0.005	0.005–100	2.5	[91]
2024	Urea	0.0001	0.0001–0.01 and 0.001–1	≈3	[90]
2022	Acetylcholine	0.016	0.001–22.5 and 25–100	≈3	[89]
2018	Urea	0.001	0.001–1	1.5	[88]
2017	H2O2	500	500 Qualitative analysis	0.01–1.4	[105]
2014	Glucose	15	15–100	≈10	[96]
2014	Glucose	0.001	0.001–1000	N/F	[99]
2011	Concanavalin A 1	10	10 Qualitative analysis	180	[106]
2011	H2O2	0.01	0.01–1	0.8–1.6	[107]

Approximated response times are estimated from percentage of response change over time plots if the response time is not specifically defined. ¹ Although concanavalin A is not the enzyme’s substrate, its immobilization inside the nanopore via glycoprotein–lectin interactions reduces the available area for ionic transport. This blocking effect can be exploited to modulate the nanopore’s conductance and selectivity.

To the best of our knowledge, the work published by Mubarak Ali et al. was the first report of a sensing platform incorporating enzymes into single nanochannel membranes [107]. This system was developed as a sensor for H₂O₂, utilizing the enzymatic reaction that occurs in the presence of immobilized HRP with ABTS as a secondary substrate. In this platform, the sensing signal is generated by the accumulation of ABTS^●+ at the tip region, leading to a decrease in the recorded current at the high-conductance branch (Figure 9a). This effect arises due to charge screening over the negatively charged surface and steric obstruction of counterion flux. The reaction mechanism involved in the process has been proposed as follows:

HRP(Fe³⁺)Porp + H₂O₂ → HRP(Fe⁴⁺=O) Porp^●+ + H₂O

(9)

HRP(Fe⁴⁺=O)Porp^●+ + AH → HRP(Fe⁴⁺=O) Porp + A^●

(10)

HRP(Fe⁴⁺=O)Porp + AH → HRP(Fe³⁺) Porp + A^● + H₂O

(11)

where HRP(Feⁿ⁺)Porp species represent different oxidation states of peroxidase while AH and A^● correspond to ABTS and ABTS^●+, respectively. This system exhibited a low detection limit of approximately 10 nM, a reversible response, and the potential for reuse without significant loss of sensing performance. However, the signal-to-noise ratio remained an area for improvement. Following a similar reaction approach, L. Zhu and coworkers reported another H₂O₂ detection platform a few years later [105]. Instead of monitoring the current via CV scans, they focused on translocation events generated by ABTS^●+ agglomerates inside a nanochannel that was ten times larger (Figure 9c). Due to the increased length and reduced diameter, product accumulation was enhanced, leading to agglomerate formation within the channel and a consequent partial reduction in the effective channel diameter. This work provided a comprehensive analysis of translocation events in such platforms and their dependence on the applied bias; however, further quantitative studies were still required.

HRP has been used in two additional sensing platforms. A few weeks after the first report by the M. Ali group, they published another work employing this enzyme as a sensing tool for concanavalin A (Con A) [106]. In this case, HRP was covalently immobilized to serve as an anchor point for Con A attachment via glycoprotein–lectin interactions. The study reported that Con A binding led to a reduction in the effective channel diameter and a corresponding decrease in the recorded current in both symmetrical and asymmetrical nanochannels. The findings were further supported by theoretical calculations, which showed good agreement with the experimental results. The authors also highlighted the potential of the platform for quantitative analysis, though additional experiments were required. However, the system is susceptible to interference from monosaccharides such as glucose.

In another study, HRP and GOx were incorporated into a biosensor for glucose detection based on an enzymatic cascade reaction involving the following steps:

Glucose + O₂ + H₂O → Gluconic acid + H₂O₂

(12)

2H₂O₂ → O₂ + 2H₂O

(13)

In this case, the sensing signal arose from surface charge reduction due to the pH drop caused by the continuous production of gluconic acid. As expected, the O₂ requirement for GOx presents a limitation in aqueous solutions due to the low solubility of O₂ in water. Additionally, H₂O₂ accumulation can lead to GOx degradation. To mitigate both effects, HRP was incorporated to catalyze the subsequent decomposition of H₂O₂ while simultaneously supplying O₂. This approach enabled the first functional device involving concerted enzyme cascade functions within a nanochannel. The system exhibited good switching properties and a short renewal time under cyclical stimulation, demonstrating its potential as a sensing device. Furthermore, it showed good tolerance to galactose, fructose, and mannose, while challenges remained regarding long-term stability and improving the dynamic range [96]. GOx was also employed in another glucose sensing system that similarly relied on pH changes caused by gluconic acid production (Figure 9e) [99], as shown in Reaction (12). However, since this platform lacked HRP, the measurement electrolyte solution was pre-saturated with O₂ to ensure an adequate supply for the reaction. The system demonstrated a low detection limit (~1 nM) and a wider dynamic range (Table 5). Additionally, the influence of channel size on sensing performance was analyzed across a broad range of tip diameters. The modified nanochannel was also exposed to L-glucose and D-glucose, showing high specificity for dextrose and good reversibility in experiments with alternating substrate exposure. However, further studies on response time and storage stability are required to advance toward full device integration.

Urease (Figure 10) and acetylcholinesterase (Figure 11a) have also been integrated into single nanochannel systems for urea and acetylcholine sensing. Unlike the previously mentioned one-block platforms, these systems incorporate an additional functional block to enhance performance. The first approach to urea biosensing was reported by Pérez-Mitta and coworkers [88]. In this work, poly(allylamine) was electrostatically assembled onto the anionic pore walls, followed by the electrostatic immobilization of urease. This design enabled rapid and highly sensitive urea detection, with the weak polyelectrolyte acting as a ‘reactive signal amplifier’ by enhancing the pH changes induced by the enzymatic reaction (Reaction (14)). A similar platform was later developed for acetylcholine sensing, employing polyethyleneimine as the secondary functional block [89]. While sharing similar functional principles, this system differed in that the enzymatic reaction led to a decrease in pH due to the production of acetic acid, as shown in Reaction (15).

Urea + 3H₂O → 2NH₄⁺ + HCO₃⁻ + OH⁻

(14)

Acetylcholine → Choline + CH₃COO⁻ + H⁺

(15)

The presence of additional charged groups with different acid–base equilibria resulted in a broad dynamic pH operational range, allowing for signal detection in both anion- and cation-driven ion current rectification states. These platforms not only introduced a novel approach for urea and acetylcholine determination but also laid the foundation for the development of single nanochannel biosensors incorporating additional functional elements to enhance sensing performance. In subsequent years, additional urease-based platforms employing a two-block architecture have been reported, utilizing the same pH-dependent sensing mechanism described in Reaction (14) but with improved performance. For instance, building on previous studies of nanoprecipitation phenomena in nanochannels with broken symmetry in the presence of polyvalent or asymmetric electrolytes [137,191], the PAH/urease system was applied to urea detection in KCl electrolyte solutions containing divalent salts (MgCl₂, CaCl₂) [90]. A significant enhancement in sensitivity was observed, achieving sub-nanomolar detection limits. In summary, the enzymatic reaction increases the pH inside the nanochannel, leading to an increase in negative surface charge. This, in turn, promotes nanoprecipitation, which functions as a reactive signal amplification mechanism, enabling lower detection limits compared to previous systems. Additionally, in a more recent study [103], Huamani et al. developed a stability-enhanced urea sensing platform by employing a covalently attached UiO-66-NH₂–urease arrangement instead of the electrostatically immobilized PAH/urease system. This new configuration not only retained its functional sensing capabilities but also exhibited greater stability under various measurement conditions, alternating substrate exposures, and continuous flow experiments, which are challenges that remain significant in electrostatically immobilized enzyme platforms.

Following the concept of concerted functions arising from cascade enzymatic reactions, a new strategy for arginine biosensing based on a multilayer architecture incorporating urease and arginase via electrostatic self-assembly, with PAH serving as the positively charged block, has recently been reported [92]. Arginine detection is achieved through a sequential catalytic process, where arginine is firstly converted into urea by arginase (Reaction (16)), and urea is subsequently hydrolyzed by urease to produce ammonia, leading to a local pH increase (Reaction (14)).

L-Arginine + H₂O → L-Ornithine + Urea

(16)

Beyond demonstrating effective arginine detection with high selectivity against non-target amino acids, this platform also enabled a detailed investigation of key design parameters. Specifically, the number of layers, their organization, and the composition ratio significantly influenced the detection limits and dynamic range. Moreover, along with the creatinine and adenosine biosensors [91,104], this system is among the few enzyme-based single SSN biosensors that have been tested in complex real samples, assessing their performance in complex biological matrices. The creatinine biosensing platform (Figure 11b) was developed by integrating creatinine deaminase and polyethyleneimine in a multilayer “sandwich” configuration, as illustrated in Figure 7d. Notably, the incorporation of an additional polyelectrolyte layer over the enzyme significantly enhanced structural stability and expanded the functional dynamic range compared to the simpler bilayer arrangement (Figure 7c). The sensing mechanism is based on charge variations induced by local pH changes, as creatinine undergoes enzymatic deamination, producing ammonia through the following reaction:

Creatinine + H₂O → N-methylhydantoin + NH₄⁺ + OH⁻

(17)

Special consideration must be given to electrostatically assembled multilayer architectures, as strong pH variations can weaken interactions between charged layers, potentially leading to block detachment and enzyme loss, ultimately affecting device performance. This phenomenon was also observed in a recent work, where adenosine deaminase was immobilized in a similar sandwich-type architecture using polyallylamine (Figure 11c) [104]. The functional principle remained the same: adenosine deaminase catalyzes the deamination of adenosine, generating ammonia and inosine (Reaction (18)).

Adenosine + H₂O → Inosine + NH₄⁺ + OH⁻

(18)

For both sandwich-structured biosensors, interference, response time, and storage stability studies were carried out. In general, the response time was short (<5 min) and good selectivity was achieved when testing common metabolites found in real samples. However, sensor performance was significantly affected by phosphate-containing species and high concentrations of proteins such as BSA, likely due to their irreversible adsorption onto the positively charged polyelectrolyte surface. Regarding storage stability, dry conditions and low temperatures (~4 °C) helped to preserve sensor functionality in the medium term, though a gradual loss of response capability was observed over time.

5. Challenges: Toward Future Developments

Despite demonstrating excellent functionality features, these sensing platforms still face major challenges for future clinical applications. One of the most pressing concerns is system stability over time. Storage stability studies have only recently been explored and remain superficially addressed [91,104]. Initial results indicate a decline in performance after short storage periods. Additionally, as nanofluidic devices, these sensors require further research into continuous substrate measurement to facilitate real-time biomarker monitoring. In this context, continuous and unidirectional flow measurements are essential for assessing long-term response stability and device durability. Another critical factor is tolerance to biological sample interferences. Although enzymatic selectivity ensures high specificity for the target substrate, systems based on polyelectrolytes are susceptible to divalent anion interactions and protein adsorption. Specifically, amino–phosphate interactions [192,193,194] and negatively charged proteins can significantly affect performance in real sample applications, necessitating sample pretreatment in current designs.

A promising approach to enhanced interference tolerance and expanding stimulus–response mechanisms is the incorporation of additional functional structural components. For example, recent applications of metal–organic frameworks have demonstrated improved system stability. Moreover, various enzyme immobilization strategies remain unexplored in nanochannel-based sensors and could further enhance performance. The potential of enzyme-based sensing systems for continuous health monitoring is evident. Platforms currently used in continuous glucose monitoring devices have demonstrated that enzyme-based sensors can function as daily monitoring tools. Although it took decades for enzyme electrodes to evolve into widely used medical devices, they now provide millions of people with greater independence and improved healthcare management. While nanofluidic sensing technologies are still in their early stages, they are continuously advancing, and their potential continues to grow. With further development, these systems could become reliable analytical tools for clinical diagnostics and real-time biomarker analysis.