Cooperative Regulation of Defect‐Mediated Molecular Recognition With Single‐Walled Carbon Nanotubes

ABSTRACT

Defects are increasingly recognized as powerful modulators of the chemistry and photophysics of fluorescent single‐walled carbon nanotubes (SWCNTs), enabling tunable optical responses toward exogenously introduced analytes. However, defect‐mediated interactions arise not solely from defects, but from the interplay among multiple interconnected system variables, including defect type and density, analyte identity and size, corona composition and conformation, SWCNT chirality, and the surrounding environment. In this forward‐looking perspective, we discuss how these variables collectively govern the fluorescence modulation and molecular recognition behavior of defect‐engineered SWCNTs. We first summarize major strategies for defect introduction and recent advances in defect quantification. Subsequently, through case‐by‐case analyses, we examine how systematic variation of individual parameters reshapes analyte–SWCNT interactions and sensing outcomes. Specifically, we discuss systems exhibiting analyte‐dependent fluorescence modulation, defect‐enabled ratiometric sensing, analyte‐size‐dependent interactions, corona‐dependent defect reactivity, and the cooperative behavior of multiple defect types within the same nanotube framework. Finally, we highlight key challenges associated with defect heterogeneity, predictability, and controllable molecular recognition, while outlining emerging opportunities involving programmable nanosensing, artificial intelligence‐assisted design, and multifunctional imaging platforms. Overall, this perspective emphasizes that defect‐mediated sensing in SWCNTs is fundamentally governed by the cooperative interplay of multiple system variables.

Defect‐mediated sensing in SWCNTs arises from the cooperative interplay of defect characteristics, density, heterogeneity, analyte properties, corona composition, and environment. Understanding these interconnected variables enables tunable fluorescence modulation, selective molecular recognition, ratiometric sensing, and sensitive detection in complex biofluids.

1. Introduction

Defects are widely recognized as powerful determinants of functionality in nanomaterials. At the nanoscale, where surface atoms, reduced coordination, and quantum confinement dominate, even subtle deviations from an ideal lattice can profoundly reshape the chemical landscape of a material [1, 2, 3]. These defects—ranging from vacancies and interstitials to dislocations, grain boundaries, and heteroatom substitutions—introduce localized electronic perturbations, redistribute charge density, and create chemically active sites that are absent in pristine structures. Consequently, they serve as tunable centers that influence reactivity, optical response, catalytic behavior, and charge transport. Rather than being passive irregularities, defects actively reconfigure bonding environments and energy states, enabling new reaction pathways, altering adsorption energetics, and stabilizing otherwise inaccessible intermediates [4]. In this sense, defects transform structural “imperfection” into chemical opportunity, positioning themselves as central design elements in the rational engineering of nanomaterials with emergent and enhanced functionalities [5, 6].

Within this broader context, single‐walled carbon nanotubes (SWCNTs) offer a uniquely sensitive and structurally well‐defined platform to interrogate defect‐mediated chemistry [7, 8, 9, 10, 11, 12, 13, 14, 15, 16]. Their quasi‐one‐dimensional architecture, coupled with discrete and chirality‐dependent electronic structures, renders their optical and electronic properties exceptionally responsive to local perturbations [17]. The introduction of defects—particularly through controlled covalent functionalization—creates localized states within the nanotube framework, effectively embedding chemically addressable sites into an otherwise inert sp² lattice [18, 19, 20]. These sites not only alter excitonic behavior but also act as focal points for molecular interactions, enabling the transduction of chemical events into measurable optical signals [18, 21, 22]. As such, defect‐engineered SWCNTs have emerged as a versatile platform for probing molecular recognition at the nanoscale.

This growing interest in defect‐engineered SWCNTs is further amplified by the broader surge in their application as near‐infrared (NIR) fluorescent probes. SWCNTs exhibit intrinsic photoluminescence in the NIR region, coinciding with the biological transparency window, where light penetration through tissues is enhanced because absorption, scattering, and autofluorescence are generally reduced relative to the visible spectral region [23, 24, 25, 26, 27, 28]. This unique optical advantage has positioned SWCNTs at the forefront of efforts toward real‐time, non‐invasive imaging and sensing in complex biological environments [29, 30, 31, 32, 33]. Consequently, SWCNT‐based platforms have been extensively explored for the detection of a diverse range of analytes and disease markers [34, 35, 36], including pathogens [37, 38, 39], hormones [40, 41, 42], lipids [43, 44], sugars [45, 46, 47, 48, 49], and many others. Despite this rapid expansion, much of the current progress remains rooted in empirical optimization, often without a unified understanding of how structural parameters, particularly defect chemistry, govern analyte recognition and signal transduction. In this context, developing a mechanistic and predictive framework for defect‐mediated responses is not only timely but also essential for translating the growing body of SWCNT‐based sensing strategies into rationally designed, application‐specific platforms [50].

Although the precise mechanisms by which defects modulate sensing responses remain an active area of investigation, several plausible and potentially complementary hypotheses have emerged. First, defects locally perturb the electronic structure of SWCNTs, generating exciton trapping sites that alter exciton diffusion, localization, and radiative relaxation. Consequently, molecular interactions occurring in the vicinity of these localized states may be coupled more efficiently to fluorescence modulation than interactions occurring on pristine nanotube surfaces [51]. Second, defects introduce chemically distinct reaction centers that can directly participate in molecular recognition through redox chemistry, coordination chemistry, reversible covalent interactions, or supramolecular binding [5, 49, 52]. Such interactions provide new recognition pathways that are absent in pristine SWCNTs and enable selective responses toward specific analytes. In addition to modifying the SWCNTs themselves, defects are also expected to influence the surrounding molecular corona, which constitutes the primary recognition interface in many SWCNT biosensors [53]. Localized changes in surface chemistry may alter the organization, conformational flexibility, hydration environment, or electrostatic properties of the corona, thereby affecting analyte accessibility and the extent of analyte‐induced corona rearrangement. Because fluorescence modulation in SWCNT biosensors arises from the intimate coupling between molecular recognition at the corona and exciton dynamics within the nanotube, perturbations to either component can substantially influence the sensing response.

Thus, despite the central role of defects, the chemistry they mediate is not governed solely by defect type. Instead, it arises from an interdependent interplay between multiple parameters: the nature of the defect, the intrinsic properties of the nanotube (including chirality), the organization of the surrounding corona phase, and the physicochemical characteristics of the interacting analyte. These factors collectively define how molecules approach, interact with, and ultimately influence defect‐localized states. The resulting response is therefore not universal, but contingent, shaped by both the local chemical environment and the properties of the analyte itself. Disentangling these contributions remains a key challenge in advancing SWCNT‐based sensing from empirical observation toward predictive design.

In this perspective, we explore this multidimensional landscape through a series of studies, each designed to isolate and interrogate specific components of this interplay. It is worth mentioning here that unlike previous reviews that have focused either on the chemical reactivity imparted by individual defect chemistries [2] or on the broad applications of SWCNTs in biosensing [54], this Perspective introduces the emerging concept of cooperative defect‐mediated molecular recognition. Rather than considering quantum defects solely as optical emitters or chemically reactive sites, we discuss how they function cooperatively with the surrounding corona phase and exciton dynamics to regulate analyte recognition and signal transduction. By systematically varying defects, corona phase, nanotube structure, and analyte characteristics, these works collectively reveal an emerging framework for understanding molecular recognition in defect‐engineered SWCNTs. Together, they point toward a hierarchical picture in which defects enable analyte differentiation, analyte properties govern response magnitude, and the surrounding interfacial environment modulates accessibility and interactions. Such insights not only refine our understanding of defect‐driven chemistry but also lay the groundwork for the rational design of next‐generation nanoscale sensing platforms dictated by defect‐controlled chemistry in SWCNTs.

2. Strategies for Defect Introduction and Their Quantification

Several chemical and physical strategies have been developed to introduce controlled defects into SWCNTs. Oxygen defects have been generated through ozone or sodium hypochlorite/UV treatments (Figure 1A) [55, 56], while solid‐state UV irradiation on photoreactive oxide substrates and mechanochemical approaches provide alternative routes for defect incorporation [57]. Covalent functionalization has also been achieved through ssDNA‐assisted photooxidation of guanine (Figure 1B) [58], hydrogen peroxide‐mediated formation of sp³ aryl defects (Figure 1C) [59], arylazo sulfonate chemistry (Figure 1D,E) [60], and, more recently, nanotube‐catalyzed Fenton reactions enabling direct conjugation of small molecules (Figure 1F–H) [61].

FIGURE 1.

(A) Evolution of the fluorescence spectra of SC‐stabilized (6,5) SWCNTs following ozone treatment and subsequent irradiation at 775 nm, showing the gradual emergence of defect‐related emission features. Reproduced with permission [55]. Copyright 2010, American Association for the Advancement of Science. (B) Schematic illustration of the single‐pot conversion of DNA‐wrapped SWCNTs into guanine‐functionalized SWCNTs through rose Bengal‐mediated photochemical defect formation. Reproduced with permission [58]. Copyright 2022, American Association for the Advancement of Science. (C) Schematic representation of hydrogen peroxide‐assisted aryl diazonium functionalization of DNA‐coated SWCNTs for the introduction of covalent sp³ defects. Reproduced with permission [59]. Copyright 2025, American Chemical Society. (D) Non‐normalized fluorescence spectra of SDBS‐dispersed (6,5)‐SWCNTs recorded during defect formation induced by 375 nm photoreaction in the presence of arylazosulfonates. (E) Raman spectra of SDBS‐dispersed (6,5)‐SWCNTs functionalized using arylazo sulfonates (CoumNNS) or aryldiazonium (CoumNN), normalized to the G‐band intensity. Reproduced with permission [60]. Copyright 2025, John Wiley & Sons. (F) Schematic depicting handle‐free covalent functionalization of SWCNTs through hydrogen abstraction‐mediated radical reactions. (G) Fluorescence spectra of SWCNTs following reaction with methanol, ethanol, and propanol, compared with untreated SWCNTs and SWCNTs exposed only to hydrogen peroxide and iron in the absence of alcohol additives. (H) Corresponding D/G Raman band intensity ratios for pristine SWCNTs and modified SWCNTs after conjugation with different alcohols and control treatments, illustrating changes in defect density following functionalization. Reproduced with permission [61]. Copyright 2026, Springer Nature.

Besides the chemical identity of the defect, both defect density and defect heterogeneity are emerging as important parameters for tailoring the sensing performance of SWCNT‐based biosensors [62]. Recent studies have shown that increasing the density of guanine defects results in a nearly linear increase in the Stokes shift, indicating that the cumulative exciton energy loss is directly related to the number of defect sites. In contrast, introducing only a few sp³ quantum defects has little influence on the overall DNA wrapping or the extent of guanine functionalization, suggesting that sparse defect incorporation preserves the structural integrity of the corona while locally modifying the excitonic landscape [62]. This balance may be advantageous for biosensing, as it enables localized modulation of exciton dynamics without significantly disrupting the molecular recognition interface.

Thus, alongside advances in synthetic methods, significant progress has been made in the quantitative characterization of defect densities. Raman spectroscopy has emerged as a central tool, where correlations between the D/G⁺ intensity ratio and defect density provide a unified and increasingly absolute metric across nanotube systems, while also revealing diameter‐dependent defect activation arising from curvature effects [63]. Complementary approaches that combine Raman analysis with photoluminescence quantum yield measurements enable direct estimation of sp³ defect densities with high precision, offering defect counts on the order of a few defects per micrometer and allowing comparison across different functionalization strategies [64]. At the single‐particle level, spectroscopic studies of monochiral SWCNTs further reveal that defect incorporation follows a stochastic (Poisson) distribution, with individual nanotubes typically hosting one to a few defects depending on reaction conditions [65]. Such findings highlight the disparity between ensemble‐averaged measurements and single‐nanotube behavior, underscoring the need for quantitative and multiscale characterization frameworks.

Collectively, these strategies demonstrate not only the growing control over defect introduction but also the increasing ability to quantify and correlate defect density with emergent properties, establishing defect engineering as a central tool for tailoring the chemical and functional behavior of SWCNTs.

3. Influence of System Parameters on Defect‐Mediated Molecular Interactions

As discussed above, the mere presence of defects in SWCNTs does not solely determine the nature of the optical response toward a given analyte. Rather, the interaction is highly dependent on the broader physicochemical environment surrounding the defect sites. In this section, we systematically examine how variations in individual system parameters, while holding other factors constant, modulate defect‐mediated interactions and consequently alter the optical response of SWCNTs to different analytes. Through these case‐by‐case studies, we highlight the dynamic and context‐dependent nature of defect chemistry in governing molecular recognition and sensing behavior (Scheme 1).

SCHEME 1.

Schematic illustration showing how interconnected system variables govern defect‐mediated molecular interactions and optical responses in SWCNTs.

3.1. Analyte‐Dependent Defect Interactions

In the first case, we consider a system in which the SWCNTs have identical corona phase and the same defect configuration, while only the analyte identity is varied. This framework enables a direct evaluation of how differences in analyte structure, functionality, and molecular configuration influence defect‐mediated interactions and the resulting optical response of the nanotubes. By isolating the analyte as the sole changing parameter, this scenario provides valuable insight into the role of defects in enabling selective molecular recognition and analyte differentiation.

In this regard, recent results from our laboratory demonstrated that oxygen‐defect‐engineered (6,5)‐enriched SWCNTs suspended by sodium cholate (SC‐SWNTs) exhibit markedly divergent fluorescence responses to structurally analogous neurotransmitters such as dopamine (DA) and serotonin [51]. While DA induced a pronounced enhancement in the defect‐related fluorescence of defect‐induced SWCNTs (D‐SWCNTs), serotonin resulted in significant fluorescence quenching, despite both molecules sharing closely related aromatic and redox‐active frameworks (Figure 2A–D). Importantly, these contrasting responses were absent in pristine SWCNTs, underscoring the indispensable role of oxygen defects in mediating analyte‐specific interactions.

FIGURE 2.

(A) Fluorescence spectra of oxygen defect‐induced SWCNTs recorded upon exposure to increasing concentrations of dopamine, illustrating progressive modulation of defect‐related emission. (B) Concentration‐dependent normalized fluorescence responses of the intrinsic E₁₁ (brown) and defect‐state E₁₁* (orange) emission bands of defect‐induced SWCNTs toward dopamine. (C) Fluorescence spectra of oxygen defect‐induced SWCNTs following treatment with increasing concentrations of serotonin. (D) Corresponding normalized fluorescence responses of the E₁₁ (brown) and E₁₁* (orange) emission peaks upon serotonin addition. (E) Raman spectra of defect‐engineered SWCNTs before and after dopamine treatment, indicating changes associated with defect passivation and chemical interaction at defect sites. Reproduced with permission [51]. Copyright 2024, American Chemical Society.

Mechanistic investigations revealed that the fluorescence enhancement induced by DA could not be fully explained by conventional reactive oxygen species scavenging pathways, which are commonly invoked for pristine SWCNTs [66]. Although radical scavengers such as mannitol produced only negligible changes in fluorescence, multiple spectroscopic analyses indicated a direct defect‐mediated chemical transformation of DA. Electrospray ionization mass spectrometry (ESI‐MS) confirmed the formation of polydopamine (PDA) in the presence of D‐SWCNTs, while no such signatures were observed for pristine SWCNTs, highlighting the critical role of oxygen defects in catalyzing DA oxidation and polymerization. The formation of PDA was further supported by UV–vis absorption spectroscopy through the emergence of a characteristic absorption shoulder around 450 nm. Moreover, Raman spectroscopic analysis revealed a decrease in the defect‐associated D band intensity following DA addition, accompanied by the appearance of a new peak near 1089 cm^{− 1}, suggestive of local lattice distortions and possible pentagonal ring formation arising from PDA adsorption onto the nanotube surface (Figure 2E). Collectively, these observations suggest that oxygen defects facilitate the conversion of DA into PDA, which subsequently adsorbs and partially passivates nonradiative pathways, thereby enhancing the fluorescence of D‐SWCNTs.

In contrast, serotonin induced fluorescence quenching through an entirely different mechanism. Raman spectra showed no appreciable changes in the defect‐associated D band upon serotonin addition, providing no spectroscopic evidence for direct chemical interactions between serotonin and the oxygen defects. Instead, the enhanced quenching observed in D‐SWCNTs relative to pristine SWCNTs was attributed to increased water accessibility around the defect sites. The oxygen functionalities likely promote hydrogen bonding with surrounding water molecules, thereby facilitating solvent‐induced nonradiative relaxation pathways, including possible electronic‐to‐vibrational energy transfer (EVET) to water. Furthermore, ESI‐MS measurements showed no evidence of polyserotonin formation, ruling out polymerization‐mediated surface passivation analogous to the DA case.

A related study further highlighted the analyte‐dependent nature of defect‐mediated sensing by examining guanine‐defect‐engineered SWCNTs, stabilized by DNA (G_d‐SWCNTs), that were exposed to dopamine and riboflavin [67]. While dopamine induced fluorescence enhancement in both pristine and defect‐containing SWCNTs, the magnitude of enhancement decreased with increasing defect density. In contrast, riboflavin‐induced fluorescence quenching became more pronounced in defect‐engineered nanotubes compared to pristine SWCNTs (Figure 3A). These opposing trends suggested that the underlying sensing mechanisms differ substantially between the two analytes. In the case of dopamine, fluorescence modulation was proposed to involve conformational rearrangements of the DNA corona phase through interactions between dopamine hydroxyl groups and the phosphate backbone. Since covalent defects restrict corona flexibility, increasing defect density likely suppresses these structural rearrangements, thereby diminishing the dopamine response. Conversely, riboflavin recognition appeared to be less dependent on corona conformational dynamics, resulting in stronger fluorescence quenching even in defect‐rich nanotubes.

FIGURE 3.

(A) Relative fluorescence modulation of guanine defect‐induced SWCNTs upon exposure to analytes that either enhance (dopamine) or quench (riboflavin) fluorescence. (B) NIR fluorescence spectra of (GT)₁₀‐coated pristine SWCNTs before and after dopamine addition (100 µM). (C) Corresponding fluorescence spectra of (GT)₁₀‐coated SWCNTs containing low densities of sp³ defects following dopamine addition. (D) Concentration‐dependent E₁₁ and E₁₁ ^* fluorescence responses of pristine and defect‐engineered SWCNTs toward dopamine. (E) Simplified exciton decay pathways in defect‐engineered SWCNTs highlighting trapping, detrapping, and radiative/nonradiative relaxation channels. (F) Schematic representation of the Monte Carlo simulation model used to describe exciton diffusion and defect‐mediated decay processes in SWCNTs. Reproduced with permission [68]. Copyright 2021, American Chemical Society.

Additionally, defect incorporation has also provided important mechanistic insight into the photophysical processes underlying analyte‐induced fluorescence modulation in SWCNT biosensors [68]. In this regard, DNA‐coated pristine and NO₂‐aryl‐defect‐induced (6,5) SWCNTs were employed to investigate dopamine sensing through comparative exciton dynamics. While pristine SWCNTs exhibited fluorescence enhancement of up to +250% at the E₁₁ transition upon dopamine interaction, defect‐engineered SWCNTs displayed the opposite behavior, with decreases of approximately 35% in E₁₁ emission and 50% in defect‐related E₁₁* emission (Figure 3B–D). These contrasting responses demonstrated that analyte‐induced fluorescence modulation cannot be explained solely through changes in radiative decay pathways. Instead, the study revealed that dopamine interaction simultaneously altered multiple nonradiative exciton decay processes together with exciton diffusion dynamics. Interestingly, the fluorescence response of pristine SWCNTs increased with nanotube length, further supporting the role of exciton diffusion in the sensing mechanism. By combining experimental measurements with Monte Carlo simulations, the authors proposed a three‐rate‐constant model that involves modulation of two nonradiative decay pathways alongside increased exciton diffusion (Figure 3E–G). Overall, this work highlighted how controlled defect incorporation can serve as a mechanistic probe for unraveling exciton transport and signal transduction pathways in SWCNT‐based nanosensors.

Thus, despite the diversity of analytes investigated, a common mechanistic theme emerges across these studies: defects do not simply create new emissive states but fundamentally alter the pathways through which analyte recognition is transduced into an optical response. Depending on the chemical nature of the analyte, defects may directly participate in redox reactions, facilitate specific chemical transformations, modulate corona conformational dynamics, or perturb exciton diffusion and relaxation. Consequently, analyte‐dependent sensing arises not from a single mechanism, but from the cooperative interplay between defect chemistry, analyte properties, corona organization, and exciton dynamics. The above examples thus illustrate how varying only the analyte identity selectively activates these different mechanistic pathways while the remaining system variables are kept constant.

3.2. Defect‐Enabled Ratiometric Sensing

In the next case, the corona phase, defect configuration, and analyte identity remain constant, allowing the influence of defect‐mediated optical modulation to be examined within a single sensing framework. Under such conditions, the simultaneous presence of intrinsic and defect‐state emission pathways can generate distinct ratiometric fluorescence responses upon analyte interaction. Unlike intensity‐based sensing, which is often susceptible to environmental fluctuations, concentration variations, and instrumental inconsistencies, ratiometric sensing provides an internal self‐referencing mechanism by monitoring the relative changes between E₁₁ and E₁₁* emissions.

In this regard, oxygen‐defect‐engineered (6,5)‐enriched SC‐stabilized SWCNTs have recently been employed as dual‐emissive ratiometric probes for cholesterol (Chol) sensing in aqueous and serum environments [69]. Upon interaction with Chol, substantial and opposing variations in intensity were observed in the E₁₁ and E₁₁* emission bands, yielding a pronounced ratiometric fluorescence response (Figure 4A). The resulting sensors exhibited limits of detection of 0.28 ± 0.01 µM in water and 0.72 ± 0.05 µM in serum, values comparable to those reported for established clinical methodologies for cholesterol detection. Importantly, the use of a ratiometric readout significantly improved selectivity against a broad range of competing biomolecules and interferents, including amino acids, sugars, ions, proteins, steroid hormones, surfactants, and phospholipids. Mechanistic studies revealed that cholesterol recognition was facilitated by its incorporation into sodium cholate micellar assemblies surrounding the nanotubes, while the oxygen defects themselves directly participated in analyte interactions and signal transduction.

FIGURE 4.

(A) Fluorescence spectra of oxygen defect‐induced SWCNTs before and after exposure to cholesterol. Reproduced with permission [69]. Copyright 2024, American Chemical Society. (B) Fluorescence spectra of defect‐engineered SWCNTs before and after Cu^{2 +} addition followed by the addition of varying concentrations of pyrophosphate. Reproduced with permission [52]. Copyright 2024, Springer Nature. (C) Normalized fluorescence response of PEG‐PL‐PBA‐functionalized SWCNTs toward 100 µM dopamine in PBS. (D) Fluorescence trajectories of individual nanosensors embedded within brain tissue following dopamine addition at t = 240 s, along with representative images of a single SWCNT before and after dopamine exposure. Reproduced with permission [70]. Copyright 2024, American Chemical Society.

Similarly, sp³‐defects induced (6,5) SWCNTs have also been utilized as ratiometric near‐infrared probes for the detection of inorganic pyrophosphate, an important biomolecule involved in DNA synthesis and cellular metabolism [52]. In this system, covalently attached aryl‐alkyne groups and triazole moieties enabled the immobilization of Cu^{2 +} ions onto the SWCNT surface. The presence of Cu^{2 +} induced fluorescence quenching through photoinduced electron transfer (Figure 4B), whereas pyrophosphate binding sequestered Cu^{2 +} ions and restored the nanotube fluorescence (Figure 4B). Notably, the differing responses of pristine and defect‐state emissions enabled reproducible ratiometric sensing over a broad concentration range. Furthermore, phospholipid‐polyethylene glycol (PEG‐PL)‐coated‐defects induced SWCNTs demonstrated biocompatibility and enabled pyrophosphate detection in complex biological media such as cell lysates, as well as monitoring DNA synthesis during polymerase chain reaction.

Beyond metabolite sensing, defect‐engineered SWCNTs have also been explored for neurotransmitter detection [70]. In one such study, phenylboronic acid (PBA)‐based quantum defects were incorporated into semiconducting SWCNTs to selectively recognize catecholamine neurotransmitters through interactions with catechol moieties. Subsequent functionalization with PEG‐PL imparted biocompatibility and enabled extracellular operation in biological systems. Interestingly, catecholamines such as dopamine did not significantly perturb the intrinsic E₁₁ emission of the nanotubes, whereas the defect‐related E₁₁* emission decreased by up to 35%, thereby generating a robust ratiometric fluorescence response (Figure 4C). Moreover, the selectivity of the sensor could be further tuned by modifying the charge of the PEG polymer, highlighting the cooperative interplay between defect chemistry and corona‐phase engineering in governing analyte recognition. Importantly, these sensors remained extracellular and retained functionality within brain tissue slices, demonstrating their potential for ratiometric imaging of biologically relevant neurotransmitters (Figure 4D).

Defect engineering has also been integrated with the concept of corona phase molecular recognition to generate ratiometric nanosensors [71]. In this approach, SWCNTs wrapped with C16‐PEG₂K‐ceramide corona phases, previously known for selective insulin recognition, were functionalized with sp³ quantum defects through UV irradiation in the presence of 5‐iodouracil. The incorporation of defects introduced an additional E₁₁* emission feature alongside the intrinsic E₁₁ transition, thereby enabling dual‐emission sensing. Furthermore, the use of (6,5) chirality‐sorted SWCNTs significantly simplified the fluorescence background and improved the modulation of the E₁₁*/E₁₁ ratio. Insulin‐induced fluorescence quenching was notably stronger in chirality‐purified SWCNTs than in unsorted nanotube samples, demonstrating the importance of spectral uniformity for defect‐enabled ratiometric sensing. To quantitatively evaluate the sensing response, multiple parameters, including spectral shifts, E₁₁ and E₁₁* intensities, and their intensity ratio, were systematically analyzed. Interestingly, the study proposed that an E₁₁*/E₁₁ ratio greater than 3 is desirable for effectively converting conventional SWCNT sensors into robust ratiometric platforms through the incorporation of quantum defects. These findings illustrate how defect engineering can synergistically enhance established corona‐phase molecular recognition strategies by introducing internally calibrated optical outputs.

Collectively, these studies demonstrate that ratiometric sensing is not a consequence of any single defect chemistry, but rather of the cooperative regulation of molecular interactions and exciton dynamics. While analyte recognition may arise through diverse mechanisms—including molecular insertion, coordination chemistry, reversible covalent interactions, or corona‐phase molecular recognition—the resulting perturbations are transduced through the differential modulation of E₁₁ and E₁₁ ^*. This convergence of chemically distinct recognition pathways onto a common photophysical output establishes defect engineering as a versatile strategy for developing internally calibrated and highly robust SWCNT biosensors.

3.3. Influence of Analyte Size on Defect‐Mediated Interactions

In this case, the corona phase, defect configuration, and analyte class remain unchanged, while only the size of the analyte is varied. This framework enables a direct assessment of how molecular dimensions alone influence defect‐mediated interactions and the resulting optical response of the nanotubes. By minimizing variations due to differences in surface chemistry or analyte functionality, such systems provide insight into the roles of steric effects and analyte accessibility in governing nanoscale recognition processes.

In this regard, chirality‐pure (6,5) and (7,5) SWCNTs suspended using the fluorene‐based polymers PFO‐BPy6,6′ and PFO‐FH were recently subjected to a post‐functionalization modification (PFM) strategy involving UV irradiation in the presence of sodium hypochlorite to introduce oxygen defects [72]. While pristine polymer‐wrapped SWCNTs generally remain inert toward analytes (Figure 5A–D) due to dense corona coverage, defect introduction partially displaced the polymer corona and exposed regions of the nanotube surface, thereby enabling the recognition of those analytes (Figure 5E–H). Evidence for this partial corona displacement was obtained through low‐temperature fluorescence spectroscopy, Raman analysis, dynamic light scattering, and transmission electron microscopy. Furthermore, riboflavin was employed as a fluorescent reporter to probe the degree of surface accessibility [73] after defect introduction. Increasing NaClO concentration used during the defect‐introduction step resulted in stronger riboflavin quenching, indicating progressively reduced polymer coverage and greater exposure of the SWCNT surface. Importantly, modulation of corona coverage directly influenced the size‐selective interaction of polyaromatic hydrocarbons (PAHs) with the defect‐engineered SWCNTs (Figure 5I–K). At intermediate NaClO treatment, the PFM‐SWCNTs responded selectively to naphthalene, a relatively small two‐ring PAH. In contrast, higher NaClO treatment enabled interactions not only with naphthalene but also with larger PAHs such as fluorene and pyrene, containing three and four aromatic rings, respectively. Notably, similar response trends were observed for both (6,5) and (7,5) SWCNTs, indicating that the interaction behavior was predominantly governed by analyte dimensions rather than chirality‐dependent electronic differences.

FIGURE 5.

Fluorescence spectra of pristine (6,5) SWCNTs before and after exposure to 400 µM of (A) naphthalene, (B) phenanthrene, (C) pyrene, and (D) fluorene. Fluorescence spectra of defect‐induced (6,5) SWCNTs upon addition of increasing concentrations of (E) naphthalene, (F) phenanthrene, (G) pyrene, and (H) fluorene. (I–K) Corresponding bar plots summarizing fluorescence modulation under different NaClO treatment conditions, demonstrating the influence of corona coverage and analyte size on fluorescence response. Reproduced with permission [72]. Copyright 2025, Royal Society of Chemistry.

These findings collectively highlight that, even in chemically identical defect‐engineered SWCNT systems, variations in analyte dimensions can profoundly influence access to defect‐associated interaction sites, thereby governing the extent and selectivity of the resulting optical response.

An important area worth discussing here is Corona Phase Molecular Recognition, a widely used sensing mechanism in SWCNT‐based biosensors [74]. In this approach, SWCNTs are non‐covalently wrapped with polymers, surfactants, peptides, or DNA to form a molecular corona that serves as the primary recognition interface for target analytes. The adsorption of an analyte induces subtle changes in the structure, dynamics, or local dielectric environment of the corona, which are subsequently transduced into changes in the near‐infrared fluorescence of the underlying SWCNTs. Understanding how defects influence the corona‐SWCNT coupling is therefore central to the rational design of defect‐engineered SWCNT sensors.

3.4. Influence of Corona Phase on Defect‐Mediated Molecular Interactions

In this case, the defect configuration and analyte identity remain unchanged, while the corona phase surrounding the SWCNTs is systematically varied. Such a framework enables direct evaluation of how the local interfacial environment created by different surface coatings influences analyte access to defect‐associated regions and modulates the resulting optical response. Since the corona phase governs parameters such as steric accessibility, local polarity, hydration, and molecular affinity, even chemically identical defect sites can exhibit markedly different sensing behavior depending on the surrounding corona architecture.

In this regard, oxygen‐defect‐engineered (6,5) SWCNTs have recently been employed to investigate how identical defects behave within distinct corona environments toward organophosphates (OPs), a representative class of reactive small‐molecule analytes [53]. In sodium cholate dispersions, pristine nanotubes exhibited a rapid fluorescence decrease upon OP exposure, whereas defect‐induced SWCNTs showed an amplified response (Figure 6A), indicating that oxygen defects enhanced analyte interaction within this relatively accessible surfactant environment. In contrast, lipid–polyethylene glycol‐wrapped SWCNTs displayed entirely different behavior. While pristine SWCNTs suspended with PEG showed negligible fluorescence modulation in response to OPs, defect‐induced SWCNTs exhibited a gradual fluorescence enhancement (Figure 6B,C) that persisted even under complex conditions such as seawater (Figure 6D). Mechanistic investigations revealed that the corona phase fundamentally altered the nature of analyte–defect interactions. Fourier transform infrared (FTIR) spectroscopy and high‐resolution X‐ray photoelectron spectroscopy (XPS) indicated direct interactions between OP molecules and oxygen defect sites in SC suspended SWCNTs. Conversely, in the PEGylated systems, fluorescence enhancement was attributed to surface passivation effects rather than direct chemical interactions at the defect sites. These findings demonstrate that identical defects can participate in entirely different interfacial processes depending on the surrounding corona architecture.

FIGURE 6.

(A) Real‐time fluorescence response of sodium cholate‐stabilized oxygen defect‐induced SWCNTs following the addition of water (control, blue) and organophosphate (OP, red). (B) Fluorescence intensity of pristine PEG‐phospholipid stabilized SWCNTs measured 90 min after addition of water (control, green) or OP (red). (C) Fluorescence intensity of oxygen defect‐induced PEG‐phospholipid stabilized SWCNTs measured 90 min after addition of water (control, purple) or OP (red). (D) Fluorescence response of equilibrated oxygen defect‐induced PEG‐phospholipid stabilized SWCNTs following addition of OP in seawater (red) or seawater alone as control (purple). Reproduced with permission [53]. Copyright 2026, Royal Society of Chemistry.

These observations establish that the corona phase acts not merely as a passive stabilizing layer, but as an active regulator of defect accessibility, analyte diffusion, and local chemical environment. Consequently, the optical response of defect‐engineered SWCNTs emerges from a cooperative interplay between intrinsic defect chemistry and the extrinsic nanoscale environment imposed by the corona phase.

3.5. Interplay between Multiple Defect Types

While most studies on defect‐engineered SWCNTs focus on a single type of defect, an emerging direction involves introducing multiple, chemically distinct defects within the same SWCNT and understanding their cooperative interactions. Rather than acting independently, coexisting defects may influence one another's local electronic environment, exciton dynamics, and local chemical reactivity. Such coupled defect–defect effects can provide valuable insight into the relationship between nanotube chemistry and photophysics, while expanding the design space for multifunctional and programmable nanosensors.

To this end, defect heterogeneity further adds another level of complexity [62]. The coexistence of chemically distinct defects, such as shallow guanine defects and deeper sp³ quantum defects, creates multiple excitonic trapping pathways that may interact through exciton transfer and recapture processes. It has been hypothesized that excitons thermally detrapped from deep sp³ defect states can be recaptured by neighboring guanine defects, thereby modifying the overall relaxation pathway while preserving the characteristic photophysical properties of the guanine defects [62]. Moreover, different defect configurations generated through radical‐mediated functionalization are expected to produce distinct local electronic and chemical environments, potentially leading to variations in analyte recognition and signal transduction. These observations suggest that defect density, spatial distribution, and defect configuration represent complementary design parameters that can be exploited to optimize sensor sensitivity, selectivity, dynamic range, and multiplexed sensing capabilities. Continued advances in defect‐specific synthetic strategies and structural characterization will therefore be essential for the rational design of next‐generation defect‐engineered SWCNT sensors.

To elaborate on this, SWCNTs simultaneously functionalized with sp³ aryl defects and DNA‐based guanine defects have recently been explored as a model system for studying defect–defect interplay [62]. Four distinct synthetic pathways were developed to introduce both defects into the same nanotube while preserving optical properties and chemical stability. Interestingly, the coexistence of the two defects resulted in measurable electronic coupling, where the sp³ defect‐associated E₁₁* emission exhibited a small red‐shift (∼3 nm) in the presence of guanine defects. Although substantially smaller than the ∼40 nm shift in intrinsic E₁₁ emission induced by guanine defects alone, this observation demonstrated that one defect environment can subtly perturb another within the same SWCNT lattice.

The coexistence of multiple defect types also strongly influenced analyte recognition and sensing behavior. To investigate this aspect, the fluorescence responses of guanine‐defect/sp³‐defect SWCNTs toward dopamine, epinephrine, riboflavin, and serotonin were systematically examined. Dopamine and epinephrine induced fluorescence enhancement, whereas riboflavin caused fluorescence quenching. However, the magnitude of fluorescence modulation progressively decreased with increasing guanine defect density, particularly for catecholamines, suggesting that covalent guanine incorporation restricts the conformational flexibility of the DNA corona phase and thereby limits analyte‐induced corona rearrangement. In contrast, riboflavin responses showed weaker dependence on defect density, indicating distinct sensing pathways.

It was observed that dopamine did not induce net fluorescence quenching even at higher defect densities, highlighting the strong influence of synthesis pathway, defect heterogeneity, and corona environment on sensing behavior. Furthermore, serotonin consistently induced fluorescence quenching, consistent with earlier observations in oxygen‐defect SWCNT systems, suggesting a distinct sensing mechanism less dependent on corona conformational dynamics.

Collectively, these findings demonstrate that combining multiple defect types not only introduces defect–defect coupling but also generates highly tunable and analyte‐specific sensing behavior governed by the cooperative interplay between exciton dynamics, corona flexibility, and local defect chemistry. A summary of the different sensing strategies and the corresponding system variables governing cooperative defect‐mediated molecular recognition is provided in Table 1.

TABLE 1.

Summary of the strategies and key system variables governing cooperative defect‐mediated molecular recognition.

Section	Variable varied	Parameters held constant	Mechanistic origin of molecular recognition	Cooperative interactions governing sensing	Representative optical response	Key design implication	Analytes tested	Concentration (range)
Analyte‐dependent defect interactions [51, 67, 68]	Analyte identity	Defect chemistry and corona phase	Different analytes engage defects through distinct chemical pathways (e.g., oxidation, polymerization, hydrogen bonding, charge transfer, corona rearrangement)	Cooperation between defect chemistry, analyte chemistry, corona dynamics, and exciton relaxation	Fluorescence enhancement or quenching depending on analyte	Defects impart analyte‐specific recognition by altering chemical reactivity and exciton dynamics.	Dopamine and serotonin [51] Dopamine and riboflavin [67] Dopamine and riboflavin [68]	50 µM–10 mM for dopamine, 5 µM–10 mM for serotonin [51]. 100 µM for dopamine and riboflavin [67]. 1 nM–100 µM for dopamine [68].
Defect‐enabled ratiometric sensing [52, 69, 70, 71]	Optical readout (E11 vs E11*)	Defect chemistry, corona, analyte	Defects introduce a second emissive state that responds differently to analyte perturbation	Cooperation between intrinsic and defect‐localized excitons, together with defect chemistry and corona‐mediated analyte recognition	Opposing modulation of E11 and E11 * enables ratiometric sensing	Defect states provide internally referenced optical outputs that improve robustness and quantitative sensing.	Pyrophosphate [52] Cholesterol [69] Dopamine, ascorbic acid, and epinephrine [70] Insulin [71]	17 µM–320 µM [52] 8 µM–1.6 mM [69]. 100 µM [70] 0.34 nM–3.44 µM [71].
Influence of analyte size [72]	Analyte dimensions	Defect chemistry, corona chemistry, analyte class	Defect formation partially exposes the nanotube surface, altering steric accessibility	Cooperation between defect‐induced corona perturbation, surface accessibility, and steric interactions	Size‐dependent fluorescence modulation	Defects can regulate molecular accessibility, enabling size‐selective sensing without changing analyte chemistry.	Naphthalene, phenanthrene, pyrene, and fluorene [72]	80 µM–400 µM [72]
Influence of corona phase [53]	Corona composition	Defect chemistry and analyte	The surrounding corona controls analyte access, hydration, and local polarity around identical defect sites	Cooperation between defect chemistry and interfacial corona environment	Same defects produce distinct fluorescence responses in different coronas	Molecular recognition is determined by the coupled defect–corona interface rather than defect chemistry alone.	bis[2‐ (methacryloyloxy) ethyl] phosphate [53]	0.015625 mM–16 mM [53]
Interplay between multiple defect types [62]	Defect identity and heterogeneity	Analyte class	Multiple defects generate coupled electronic environments and competing exciton trapping pathways	Cooperation between different defect types, exciton transfer, corona flexibility, and local defect chemistry	Tunable analyte‐dependent fluorescence and spectral shifts	Defect density, configuration, and spatial distribution emerge as additional design parameters for engineering sensing performance.	Dopamine, epinephrine, riboflavin, and serotonin [62]	100 µM [62]

4. Challenges and Future Outlook

Despite remarkable progress in defect‐engineered SWCNTs, several challenges remain before these systems can be fully harnessed as predictable, programmable nanoscale sensing platforms. A central theme emerging throughout this work is that the optical and chemical behavior of defect‐engineered SWCNTs is governed not by defects alone, but by the collective interplay of multiple system variables, including defect identity, corona composition, analyte properties, and environmental conditions. Consequently, a foremost challenge is achieving precise control over these interconnected variables, particularly defect identity, density, and spatial distribution. Although numerous synthetic strategies have been developed, defect formation often remains stochastic, leading to heterogeneity both within and across nanotube populations. Such variability complicates the establishment of universal structure–property relationships and limits reproducibility in sensing performance. Addressing this challenge will require synthetic and analytical approaches capable of controlling and quantifying defects while simultaneously accounting for their coupling with surrounding system variables. A promising strategy to address this challenge may involve developing post‐synthetic separation and purification methodologies for defect‐engineered SWCNTs, analogous to the chirality‐sorting approaches widely used for isolating monochiral nanotube populations. Just as aqueous two‐phase extraction [75, 76, 77, 78], chromatography [79, 80] and selective polymer wrapping enabled the isolation of SWCNTs with uniform chiralities [81, 82, 83, 84], future approaches could potentially separate nanotubes based on defect density, defect type, or local defect environment. Such “defect‐sorted” SWCNT populations would significantly reduce heterogeneity and allow more precise correlation between defect structure and sensing behavior. Coupling these separation strategies with advanced spectroscopic mapping and single‐particle analysis could further enable standardized defect libraries, ultimately improving reproducibility and accelerating the development of predictive structure–property relationships in defect‐engineered nanosensors.

The strong dependence of defect‐mediated responses on surrounding interfacial conditions also opens opportunities for designing adaptive nanoscale interfaces whose behavior can be dynamically tuned by external variables. In this regard, dynamically reconfigurable coronas, programmable defect chemistries, and hybrid architectures integrating defects with plasmonic, catalytic, or biomolecular components could enable next‐generation nanosensors capable of multimodal recognition and real‐time environmental adaptation. Importantly, such systems would exploit system variables not as sources of complexity to be minimized, but as tunable inputs to regulate nanoscale interactions and optical outputs.

Given the multidimensional nature of these interactions, an important future direction will involve integrating artificial intelligence (AI) and machine learning (ML) frameworks to predict analyte–defect interactions as a function of multiple interconnected system variables [85, 86, 87, 88]. Recent advances have already shown the utility of AI and ML in accelerating the discovery and optimization of SWCNT‐based biosensors, including identifying corona phases and sensing motifs that would be difficult to obtain through empirical screening alone [31, 85, 87, 88, 89, 90]. As demonstrated throughout this work, the fluorescence response of defect‐engineered SWCNTs emerges from a highly coupled interplay between chirality, defect chemistry, corona structure, analyte characteristics, and the local environment. However, these studies have largely focused on pristine SWCNT systems and have not explicitly incorporated defect chemistry, defect density, defect heterogeneity, or their interplay with corona composition as predictive variables. Capturing these multidimensional relationships remains a major challenge. These relationships are difficult to optimize empirically, making them particularly well‐suited for data‐driven approaches. Machine learning models trained on experimental and computational datasets could uncover hidden correlations between system variables and sensing behavior, while AI‐assisted inverse design strategies may identify optimal combinations of chirality, defect architecture, and corona composition required for targeted applications. Such developments would move defect engineering beyond empirical optimization toward a predictive design framework.

The integration of defect engineering with corona phase molecular recognition further highlights the importance of systematically understanding how system variables collectively govern molecular recognition. Recent studies have demonstrated that combining defect chemistry with corona phase molecular recognition can generate new sensing modalities and improve molecular recognition [71]. Conventionally, this approach relies heavily on complex corona conformations that are often difficult to predict. However, introducing chemically defined defect sites provides additional control over the local chemical environment, exciton localization, and analyte accessibility. Nevertheless, the mechanisms governing defect–corona coupling remain poorly understood. In particular, it remains unclear how defect chemistry, defect density, spatial distribution, and corona composition collectively regulate analyte binding, corona rearrangement, exciton dynamics, and fluorescence transduction. Addressing these questions will be essential for establishing predictive structure–response relationships and enabling the rational engineering of defect‐mediated molecular recognition.

Beyond their influence on sensing performance, defect engineering may also affect the interactions of SWCNTs with biological systems, making biocompatibility and nanotoxicology important directions for future investigation [91]. The introduction of chemically distinct defect sites modifies the surface chemistry, charge distribution, and interfacial interactions of SWCNTs, which may influence protein corona formation, cellular uptake, immune recognition, biodistribution, and clearance. While chemically engineered defects can improve sensing functionality and molecular recognition, they may also alter biological responses in ways that remain poorly understood. Consequently, systematic studies that correlate defect chemistry, defect density, and corona composition with biological outcomes will be essential for establishing design principles that simultaneously optimize sensing performance and biocompatibility. Such efforts will be particularly important for the clinical translation of defect‐engineered SWCNTs in biosensing, bioimaging, and theranostic applications.

Finally, the coexistence of numerous tunable system variables raises the exciting possibility of constructing logic‐based nanoscale architectures. Parameters such as defect type and density, chirality, corona composition, analyte identity, analyte size, pH, ionic strength, and solvent environment could collectively serve as programmable inputs that generate distinct optical outputs. In such systems, fluorescence modulation may resemble nanoscale logic‐gate behavior, in which specific combinations of variables produce well‐defined sensing states through selective modulation of E₁₁ and E₁₁* emissions. The strong dependence of defect‐engineered SWCNT responses on multiple interconnected system variables [92] presents a powerful opportunity for the development of intelligent nanosensors with adaptive recognition behavior, multiplexed detection capability, and environment‐responsive imaging functionalities. By harnessing these multidimensional interactions, future studies may enable increasingly programmable and predictive nanoscale sensing platforms.

5. Conclusions

In this perspective, we discussed how defects in SWCNTs act as chemically active perturbations that reshape the local nanoscale environment, thereby modulating molecular recognition, analyte accessibility, and optical response. Rather than functioning as isolated photophysical modifiers, defects were shown to operate within a broader network of interconnected system variables, including chirality, corona composition, analyte identity, analyte size, and surrounding environment. Through selected examples, we highlighted how varying individual parameters while keeping others constant enables deeper insight into the mechanisms governing defect‐mediated interactions. The studies summarized here collectively demonstrate that identical defects can produce markedly different responses depending on the surrounding corona phase or analyte environment, while controlled variation of analyte size, corona structure, or defect density can systematically tune sensing behavior. We further discussed how defect engineering enables advanced functionalities such as analyte differentiation, ratiometric sensing, enhanced selectivity, and adaptive optical responses. Finally, future opportunities involving programmable defect architectures, AI‐assisted predictive design, tailored corona‐phase integration, and logic‐gated sensing platforms underscore the growing potential of defect‐engineered SWCNTs as versatile and intelligent nanoscale interfaces for sensing and imaging applications.

Author Contributions

Srestha Basu: writing – original draft, validation, visualization, writing – review and editing. Gili Bisker: conceptualization, funding acquisition, visualization, writing – review and editing, project administration, supervision, resources.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.