# Fileset

[Advanced Optical Materials 12(2024)2400868 - Hu - Iwanaga - Tang - Metasurface Platform Incorporating AIE.pdf](https://mdr.nims.go.jp/filesets/e579dfa3-c891-4d72-8de5-2a907d7be920/download)

## Creator

Qi Hu, [Masanobu Iwanaga](https://orcid.org/0000-0002-8930-6940), Youhong Tang

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Metasurface Platform Incorporating Aggregation Induced Emission Based Biosensor for Enhanced Human Serum Albumin Detection](https://mdr.nims.go.jp/datasets/bc1d2a52-50fe-499f-9c87-f52dfe613188)

## Fulltext

Metasurface Platform Incorporating Aggregation Induced Emission Based Biosensor for Enhanced Human Serum Albumin DetectionRESEARCH ARTICLEwww.advopticalmat.deMetasurface Platform Incorporating Aggregation InducedEmission Based Biosensor for Enhanced Human SerumAlbumin DetectionQi Hu, Masanobu Iwanaga,* and Youhong Tang*Metasurfaces exhibit excellent optical performance to enhance thelight–matter interaction of target molecules in biosensing based on itswell-optimized nanostructured unit cells. In the meantime, fluorescence (FL)biosensors with aggregation induced emissions (AIE) features alsodemonstrate outstanding performance in biomarker detection due to theirfast response, high selectivity, and low background noise. Nevertheless,extremely low-level analytes are difficult to detect in practical applicationssince complex urine samples include a number of uncontrolled variables suchas impurities, autofluorescence, other urine components, etc. At present,improving optical signal sensitivity of human serum albumin (HSA) detectionis always a big challenge to overcome such interference in human urinescenarios. In this work, first an AIE-based FL biosensor TPE-4TA is combinedwith an all-dielectric metasurface platform to achieve quantitative detection oftrace HSA in urine by utilizing biofunctionalization protocols on the silicon(Si) nanostructures. The results indicate significant FL enhancement in themetasurface platform that offers a promising pathway for improvingbiomarker detection in the future.1. IntroductionMetasurface is a 2D version of metamaterials that con-sists of periodic/non-periodic subwavelength metallic/dielectricQ. Hu, M. IwanagaResearch Centre for Electronic and Optical MaterialsNational Institute for Materials Science (NIMS)1-1 Namiki, Tsukuba 305-0044, JapanE-mail: iwanaga.masanobu@nims.go.jpQ. Hu, Y. TangInstitute for NanoScale Science and TechnologyMedical Device Research InstituteCollege of Science and EngineeringFlinders UniversityBedford Park, South Australia 5042, AustraliaE-mail: youhong.tang@flinders.edu.auThe ORCID identification number(s) for the author(s) of this articlecan be found under https://doi.org/10.1002/adom.202400868© 2024 The Authors. Advanced Optical Materials published byWiley-VCH GmbH. This is an open access article under the terms of theCreative Commons Attribution License, which permits use, distributionand reproduction in any medium, provided the original work is properlycited.DOI: 10.1002/adom.202400868structures.[1,2] It exhibits the capabilitiesof concentrating, inhibiting, absorbing,scattering or guiding waves by design-ing nanoscale cells to achieve dynamicmodulation in optics, mechanics, electron-ics, etc.[3,4] The strong wavefront manip-ulation is the remarkable characteristicwhere the unit cells of shape and size, di-versified patterns and geometry are tun-able and controllable.[5–7] It is preciselybased on such characteristics that the at-tributes of incident light can be customizedthrough light confined mode and local-ized surface plasmon (LSP) electromag-netic (EM) near-field, thus enhancing light–matter interactions of target molecules.[8,9]To date, metasurface has been extensivelydeveloped in recent decades on the aspectsof polarization conversion,[10] wavefrontshaping,[11] and controllable radiation.[12]The advancement of metasurface plat-form has attracted marvelous attention tothe applications of biosensing and ledto a significant stride toward biomarker detection.[13,14] Ogun-toye and his team designed a resonant dielectric photonic Si-nanoantenna metasurface to detect a tuberculosis biomarkerCFP-10 peptide and this platform’s cost was 87–96% lower thancurrent assays with equivalent sensitivity. The results illustratedthat the maximum sensitivity and limit of detection (LOD) were0.1 μm and 10 pm, respectively.[15] Wang et al. reported an optoflu-idic silicon-on-insulator (SOI) metasurface that provided mul-tiple nanoscale lateral flow channels to deliver ErbB2 breastcancer biomarkers to the sensor surface. This platform illus-trated a resonance mode ≈1550 nm wavelength and LOD forErbB2 is 0.7 ng mL−1.[16] Negm et al. demonstrated a double-resonating metasurface with the characteristic of permittivityasymmetry and geometric asymmetry simultaneously to achievea thin protein layer sensing. Germanium Antimony Telluride(GST326) ellipse-shaped nanopillars were sensitive to the analyteand showed great robustness of phase transition losses from theamorphous to the crystalline state in the mid-infrared range.[17]In our previous publications, Iwanaga et al. fabricated an effec-tive all-dielectric metasurface fluorescence (FL) biosensor withperiodic silicon (Si) nanorods array, which had also successfullydetected several biomarkers such as immunoglobulin G (IgG),[18]cancer biomarker (PSA and CEA),[19] cell-free DNA (cfDNA),[20]etc. Specifically, Si or SOI-fabricated metasurfaces possess theAdv. Optical Mater. 2024, 12, 2400868 2400868 (1 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbHhttp://www.advopticalmat.demailto:iwanaga.masanobu@nims.go.jpmailto:youhong.tang@flinders.edu.auhttps://doi.org/10.1002/adom.202400868http://creativecommons.org/licenses/by/4.0/http://crossmark.crossref.org/dialog/?doi=10.1002%2Fadom.202400868&domain=pdf&date_stamp=2024-06-04www.advancedsciencenews.com www.advopticalmat.deFigure 1. A) Schematic diagram of metasurface platform. Inset figure magnifies the nanostructure of metasurface area; B) The schematic diagram offluorescence (FL) setup for imaging; C) Actual photograph of experimental configuration for human serum albumin (HSA) detection. Sample liquidsflow on metasurface platform through inlet (right) and outlet tubes (left) controlled by rotary pump and FL is collected by setup in (B).tuning adaptability of their optical properties and provide high-quality resonances.[21] When the target analyte enables immobi-lized or is coated on Si-nanoscale matrices, high refractive indexof Si facilitates light modulation,[22,23] resulting in deep light–matter interactions, thereby reflecting on the analyzed peaks. Col-lectively, these above-mentioned satisfactory results indicate theoutstanding capability of metasurface on biomarker nanopho-tonic detection.Aggregation-induced emission (AIE) FL biosensors also ex-hibit excellent detection capabilities in biosensing where theintermolecular interactions restrict the movement of AIEmolecules rotors, thus producing an extremely fast and brightFL response through the radiation channel.[24,25] For instance,we have synthesized and reported an AIE-based FL biosensorthat achieves the highly efficient detection of human serum al-bumin (HSA) with a wide linear dynamic range 0–1000 mg L−1and LOD 0.253 mg L−1.[26] However, HSA measurement usingAIE FL biosensors has always faced considerable challenges inhuman urine scenarios owing to unpredictable reasons regard-ing the complexity of urinary excrement. For starters, it has beendiscovered that albumin excretion varies greatly across individ-uals, with a typical within-person coefficient of variation (CV)of 40–60%,[27] which leads to a broad concentration extensionof urinary HSA from extremely low to high depending on indi-vidual differences. In addition, the urine matrix is made up of avariety of organic and inorganic compounds, ranging from low-molar mass molecules to polymers.[28] It could also contain cellsand bacteria that have the ability to rapidly change the composi-tion of urine.[28] On the basis of this background, undesirable FLenhancement or quenching will be further triggered because ofthe interference of numerous impurities,[29] autofluorescence,[30]other urine components,[28,31] etc. Consequently, it is inevitableto overestimate or underestimate HSA content, leading to in-accurate readings, especially at an extremely low-level.[32] Notonly that, but the optical signal is also hardly detectable due tosmall amplitude of light absorption when HSA concentration islow, which requires high detection accuracy by boosting sensitiv-ity or reducing background noise. To address these issues withcomplex bio-samples in practical applications and evaluate thebiomarker detection performance of metasurface in biosensing,herein, we first report nanophotonic all-dielectric metasurfaceplatforms coupled with AIE featured FL biosensors for achiev-ing the enhanced detection of HSA. This platform integrates amicrofluidic system and a metasurface substrate to enable an-alyte delivery and the monitoring/detection of FL enhancementin real-time with the outstanding characteristics of high through-put, good reusability and low reagent consumption, as shown inFigure 1A. Under the FL setup (Figure 1B,C), the results fromhuman urine samples show that our proposed metasurface hassignificant potential as a biomedical chip-based platform for pro-viding a promising pathway of enhancing sensing and biomarkerdetection quantitatively.2. Results and Discussion2.1. Optical Properties of Metasurface PlatformThe optical optimization of the metasurface has a history, whichwas reported in previous publications.[1–3,33–35] In principle, FL-intensity enhancement is a consequence of multiple control ofphotoexcited states;[33–35] that is, FL enhancement factor is equalto the product of excitation efficiency, inner quantum yield, andFL-emission efficiency (namely, Purcell factor). One of the keys isto the FL-emission efficiency, which was optimized by adjustinga particular resonant mode of the metasurface (i.e., a reflectancepeak in Figure 2B) to the wavelength of FL. This was conductedAdv. Optical Mater. 2024, 12, 2400868 2400868 (2 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbH 21951071, 2024, 23, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adom.202400868 by National Institute For, Wiley Online Library on [19/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advopticalmat.dewww.advancedsciencenews.com www.advopticalmat.deFigure 2. A) 3D illustration of metasurface of Si nanocolumn (diameter d = 220 nm, height h = 200 nm) array forming a square lattice of periodicity300 nm. B) Simulated reflectance spectrum at the normal incidence in a wide wavelength range from the UV to the near infrared. C,D) Electric andmagnetic field intensities (|E|2 and |H|2), respectively, excited at 360 nm, which is indicated by an arrow in Figure 2B. This xz-section was set to cutthrough the center of the Si nanocolumns. E,F) Electric and magnetic field intensities, respectively, shown in the xz-section view similar to (C,D) andexcited at 530 nm, indicated by an arrow in Figure 2B. Incident field was set to be unity, that is, |Ein|2 = 1 and |Hin|2 = 1. Color bars indicate the fieldintensities. White lines show the interfaces of the Si nanocolumns with air and the underlying SiO2 layer.by tuning the diameter and height of the Si nanocolumns.[35] Be-sides, the periodicity of the metasurface was set to a reasonablevalue, in accordance with the diameter of the Si nanocolumns.Thus, the present metasurface has a set of optimized struc-tural parameters, based on the previous studies.[18,35] In thiscase, the optical structural design of metasurface is suitable forAIE FL biosensor TPE-4TA to achieve FL signal enhancement atgreen wavelengths ≈530 nm. Figure 2A shows the dimensionparameters of SOI-nanorod metasurface with the thickness ofa base Si wafer 675 μm and SiO2 layer 375 nm, respectively. Sinanocolumns consisting of diameter of 220 nm and height of200 nm forms a square lattice of 300 nm periodicity. Normal in-cident light propagates from the air to the metasurface, ensuringthat the vector of the electric field is perpendicular to the inci-dence plane. Furthermore, the reflectance spectrum, as shownin Figure 2B, was numerically computed in the configuration ofFigure 2A, exhibiting reflection peaks or dips according to theresonances of metasurface. A reflectance band of 30% is gen-erated at 360 nm which is the excitation wavelength for TPE-4TA, whereas the reflection response at 530 nm reaches up to70% with a single sharp peak significantly. A high reflectanceexceeding 90% also observed between 700–800 nm, which origi-nates from the magnetic dipole resonance (Figure S1, SupportingInformation). Figure S2 (Supporting Information) shows simu-lated and measured reflectance spectra, respectively. Overall, thetwo spectra agree with each other in the spectral shapes. Sincethe spectrometer allowed us to measure reflectance at 5 degreesand more, we set the incident angle to be 5°, which is close tothe normal incidence, and, indeed, verified that the simulatedspectrum in (A) is approximately the same as the spectrum inFigure 2B. Importantly, the main feature regarding reflectancepeaks at 360 and 530 nm, which was related to the FL enhancingeffect in this study, was reproduced in the measured reflectancespectrum (B). The incident polarization was set to s polariza-tion, which means that, when plane of incidence is the xz plane,the incident electric-field vector is parallel to the y-axis. Besides,we mention that some interference signatures at 840–1050 nmcame not from the metasurface itself but from the optical con-figuration in the spectrometer. Afterward, to visualize the un-derlying mechanism of the FL enhancement induced by the Sinanocolumns, resonant EM field distributions are investigatedin Figure 2C–F. The intensities of electric fields (|E|2, Figure 2C)Adv. Optical Mater. 2024, 12, 2400868 2400868 (3 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbH 21951071, 2024, 23, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adom.202400868 by National Institute For, Wiley Online Library on [19/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advopticalmat.dewww.advancedsciencenews.com www.advopticalmat.deand magnetic fields (|H|2, Figure 2D) excited at 360 nm are re-inforced to 54.9 and 6.1, respectively, in comparison with theincident field intensity of 1.0. These EM-field distributions aredark inside the Si nanocolumns and indicate that the incidentlight was absorbed in the Si nanocolumns. However, the elec-tric field at the outermost surface is highly enhanced at the side-wall of the Si nanocolumns and is not significantly reduced atthe top of the Si nanocolumns. At the wavelength of 530 nm, theelectric field intensity decreases at 41.9 slightly but this strongelectric field distributions shift toward the upper part of the Sinanocolumns on which the majority of FL-labeled HSA analytestend to be immobilized (Figure 2E). Notably, the magnetic fieldsare significantly enhanced up to 115.3 inside Si nanocolumns asFigure 2F shows, which indicates that this resonance is a mag-netic mode (specifically, a higher magnetic mode than magneticdipole mode), corresponding to the high-reflectance resonance at530 nm in Figure 2B. The reinforced resonant EM fields are ad-vantageous to transit electric dipole in FL molecules,[35] thus con-tributing to enhanced FL emission at the outermost surface of theSi nanocolumns. The numerical calculation was based on rigor-ous coupled-wave analysis and scattering matrix algorithm.[36]2.2. Detection Strategy and Binding Affinities on MetasurfacePlatformThe detection strategy is to utilize biofunctionalization proto-cols of binding molecules, labeled antibody (Ab), and targetanalyte for achieving multi-level immobilizations on metasur-face (Figure 3A). After initial phosphate-buffered saline (PBS)flow, the binding molecules of Cys-streptavidin (Cys-SA) flowthrough the microfluidic paths onto the metasurface regionswhere a uniform and stable physiosorbed layer is formed di-rectly by the interaction between streptavidin and the silica shellof nanocolumns.[37] Next, HSA antibodies labeled with biotin(Biotin-HSA Ab) are bound to the Cys-SA through the fast andstrong non-covalent protein-ligand interactions, and this formedcomplex biolayer provides a good high-density binding plat-form for HSA. Furthermore, when tetrazolate nitrogens from FLbiosensors TPE-4TA bind with polar dominant-contacting lysineresidues (Lys) in HSA binding conformation through hydrogenbonding and electrostatic interactions, triggered restriction of in-tramolecular movement effect lights up HSA.[38] Afterward, pre-bound products of HSA and TPE-4TA can be captured persis-tently through antibody-antigen reaction. Notably, PBS rinsingis compulsory to remove unbound parts after immobilization ateach level. Besides, the results in Figure S3 (Supporting Informa-tion) demonstrate that the binding sequence of TPE-4TA affectsthe final FL response. Two pre-bound products ([HSA+TPE-4TA]and [HSA Ab+HSA+TPE-4TA]) possess the strongest FL inten-sities in this protocol compared to the case where TPE-4TA isindividually immobilized on HSA in the last step. The potentialreason for the FL-intensity differences is the inward orientationof HSA as big biological molecules when immobilized. The bind-ing sites of endogenous and exogenous substances are mainly lo-cated in subdomain IIA and IIIA on HSA tertiary structure[39–41]where IgG binding positions of ligands on HSA are localized pre-cisely in Domain III.[42,43] Nevertheless, the favorable binding po-sitions of ligands between HSA and TPE-4TA are found in theintersection of Domain I and Domain III.[38] Once HSA immo-bilizes onto its Ab successfully, Domain I and III of HSA mainlyface the surface of Si nanocolumns. The presence of steric hin-drance may limit TPE-4TA movement by crossing directly intothe interface of Domain I and III, thereby the binding opportu-nities between HSA and TPE-4TA will be further reduced dueto immobility of adaptive spatial conformation of HSA-HSA Ab(Figure S4, Supporting Information). However, the prepared mix-ing solution of HSA and TPE-4TA beforehand can be fully re-acted to completely avoid this binding loss. In short, the wholedesigned biofunctionalization protocol indicates that the detec-tion strategy is feasible and optimal for binding identification,stability, and intensity, which has been validated from subsequentexperimental results.In order to ensure that the preset immobilizations can pro-ceed as the desired protocols, the surface plasmon resonance(SPR) measurement was conducted to evaluate the effectivenessof immobilization quantitatively as shown in Figure 3B. The mea-surement was conducted on a flat gold surface where SPR isinduced. The initial flow of PBS at 10 μL min−1 for 10 min isthe highest priority step to clean the gold film surface and main-tain unobstructed microfluidic paths. Next, Cys-SA, Biotin-HSAAb, and the mixing solutions of HSA and TPE-4TA are flowed at10 μL min−1 for 10 min, respectively. Furthermore, each bindingstage contains a standard rinse with PBS for 10 min to removeresidual reagents from the previous step. The results at the steadystate indicate that binding affinity reflects on a quantity, ΔRU, de-fined as absorbent mass per unit area in the scale of pg mm−2. Asthe sequential immobilizations accomplish, ΔRU continues toincrease with three binding stages, which are 1303.8 of Cys-SAimmobilization (Molecular weight, MCys-SA = 17.9 kDa), 1694.2of Biotin-HSA Ab immobilization (MBiotin-HSA Ab = 69 kDa) and281.5 conjugates of HSA and TPE-4TA (MHSA+TPE-4TA = 67 kDa),respectively. The inset figure in Figure 3B represents the differ-ence of HSA with/without TPE-4TA in binding response. Com-pared with HSA solution only, the binding ratio of HSA+TPE-4TA increases by 34%, which is attributed to the ligands bindingof TPE-4TA through hydrogen bonding and electrostatic interac-tions. Based on the analysis of BIACORE, the theoretical bindingcapacity of TPE-4TA as small molecules on gold surface can becalculated using the following Equation 1:RTPE−4TA =MTPE−4TA × RHSA × SrMHSA(RU) (1)where MTPE − 4TA and MHSA represent the molecular weight ofTPE-4TA and HSA, respectively; RHSA represents the binding ca-pacity of immobilized HSA and Sr refers to the stoichiometricratio between TPE-4TA and HSA.The result indicates that 95.7 of the measured binding re-sponse of TPE-4TA are 24 times more than the theoretical esti-mate. This increase could be explained by the AIE phenomenonsince numerous FL molecules of TPE-4TA aggregate into thestructural cavity of HSA.Moreover, the immobilized molecular density is evaluated bythe Equation 2:DA =MARA × NA(1 molecule for nm2)(2)Adv. Optical Mater. 2024, 12, 2400868 2400868 (4 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbH 21951071, 2024, 23, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adom.202400868 by National Institute For, Wiley Online Library on [19/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advopticalmat.dewww.advancedsciencenews.com www.advopticalmat.deFigure 3. A) Schematic diagram of multiple immobilizations on Si nanocolumns; B) Binding response of immobilizations on an Au-film sensor chip usingsurface-plasmon-resonance instrument. Inset: the difference of binding ratio between HSA and the mixture of HSA + TPE-4TA in the final immobilizationstep; C) Non-specific absorption between 2 μg mL−1 Biotin-HSA Ab and 5 μg mL−1 HL555-HSA Ab with HSA concentration varying. Inset: the changeof FL efficiency with different concentrations of HL555-HSA Ab.where DA represents the immobilized molecular density of theanalyte; MA represents the molecular weight of the analyte; RArepresents the binding capacity of the analyte and NA is the con-stant of Avogadro.The result shows that one molecule of Cys-SA, Biotin-HSA Aband conjugates of HSA and TPE-4TA are situated on an averagesquare surface area of 4.61 × 4.61, 8.20 × 8.20, and 19.9 ×19.9 nm2, respectively, indicating this multi-level immobilizationstrategy is feasible and can be also applied in our platform effec-tively.The indirect verification of immobilization is also carried outthrough non-specific Ab-Ab absorption as shown in Figure 3C.Adv. Optical Mater. 2024, 12, 2400868 2400868 (5 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbH 21951071, 2024, 23, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adom.202400868 by National Institute For, Wiley Online Library on [19/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advopticalmat.dewww.advancedsciencenews.com www.advopticalmat.deIn this case, the same species of HSA Abs have been labeled in-dividually with two different moieties: biotin and FL-label HyLiteFlour 555 (HL555). The flowing order of samples is as follows:Cys-SA, Biotin-HSA Ab, HSA, and HL555-HSA Ab then rinseis performed after each sample flow. Concretely, HL555 labeledHSA Abs emit FL at the peak of 570 nm and they will be absorbedonto Biotin-HSA Abs in metasurface area that has not reactedwith HSA yet, hence being forced to dock on the metasurface like-wise. When different concentrations of HSA are already fixed on2 μg mL−1 Biotin-HSA Abs, FL quenching effect occurs as HSAconcentration increases upon extra introduction of 5 μg mL−1HL555 labeled HSA Ab into the microfluidic paths, which provesthat antibody-antigen reactions with higher priority interrupt Ab–Ab nonspecific absorption (Inset graph of Figure 3C). The re-sults illustrate that FL intensities reach the strongest value inthe absence of HSA meanwhile it descends by 68.1% when HSAconcentration reaches 1000 ng mL−1. Figure S5 (Supporting In-formation) reveals that the concentration of HL555-labeled HSAAb is positively correlated with its FL efficiency under this non-specific absorption. The inset figure in Figure S5 (Supporting In-formation) indicates that the introduction of HSA is the primaryfactor for this FL quenching effect. Consequently, multi-level im-mobilization can also be confirmed indirectly.2.3. FL Kinetics on Metasurface PlatformThe FL kinetics behavior of AIEgen TPE-4TA in HSA detectionunder the modulation of metasurface platform is evaluated inthis section. TPE-4TA has excellent FL response toward HSAand its concentration effect has been optimized (Figures S6 andS7, Supporting Information). Its good selectivity toward HSA ex-cludes potential interference from biomolecules over various pro-teins with isoelectric points ranging 1–10 in buffer solution.[38]Figure 4A shows an excellent photostability of HSA+TPE-4TAconjugate within 2 h, which allows multi-level immobilizers togenerate FL persistently on the metasurface. Moreover, FL in-tensity is positively related to HSA concentration as displayedin Figure 4B. FL enhancement in the microalbumin range is di-vided into two phases: a sharp phase (0–20 μg mL−1) and a slowphase (20–160 μg mL−1). The slowing trend indicates bindingcapabilities of Biotin-HSA Ab are approaching their maximumlimitation. The steady FL intensities near 160 μg mL−1 describethat the fixed amount of HSA and TPE-4TA conjugates has beenregionally saturated on metasurface. Significantly, the metasur-face platform remains ultra-sensitive toward the trace of HSA(< 20 μg mL−1) where FL signal climbs rapidly at 82% of maxi-mum response when HSA content jumps toward 20 μg mL−1. Tofurther explore the extent to which the metasurface platform con-tributes to FL amplification, two other platforms (i.e., microplateplatform and microfluidic platform) are also established for hori-zontal comparison and the corresponding linear fittings of threeplatforms are plotted in Figure 4C and Figures S8 and S9 (Sup-porting Information). The microplate platform is composed ofthe conventional transparent polystyrene materials and microflu-idic platform possesses the same construction as metasurfaceplatform without Si-nanocolumns. In Figure 4D, 39% and 136%FL enhancement of HSA detection are observed in microplateplatform and microfluidic platform respectively, however, emit-ted FL boosts up to 1001% under the light-confined mode in themetasurface platform. Through comparison of three platforms(Figure 4E), FL signal of the metasurface platform and the mi-crofluidic platform are 7.93-fold and 1.70-fold stronger than thatof the microplate platform respectively. Obviously, FL regulationof metasurface is superior to the other two platforms. We alsonote that the numbers of the FL molecules involved in these mea-surements are the least in the metasurface platform because thePBS rinse removed the unbounded FL molecules. Thus, net FL-detection efficiency is far superior in the metasurface platform tothe other microplate and microfluidic platforms.On top of that, the practical evaluation in the application ofhuman urine is investigated among three platforms to verifyfeasibility of sensing augmentation. In urinary FL analysis, aut-ofluorescence is induced inevitably under UV light and leads toFL overlap within the detection scope of 529–571 nm (FigureS10, Supporting Information). Hence, treated urine samples di-luted 80 times with PBS buffer are employed as the standard toeliminate the impact of auto-FL as efficiently as possible (FigureS11, Supporting Information). PBS buffer also maintains dilutedurine samples into a neutral environment avoiding the poten-tial impact of changes in urinary pH on FL measurements. Ad-ditionally, the robustness of TPE-4TA has already been proved inslightly acidic environment, and main urinary components, suchas urea, uric acid, creatinine, etc., do not interfere assay sensitiv-ity of HSA.[38] Validation from the commercial kit shows HSAlevel of 45.8 μg mL−1 in the urine samples and this concentra-tion can be calculated independently by referring to the standardcurves established in three platforms to obtain 40.7, 41.9, and43.8 μg mL−1 with recovery rates of 88.9%, 91.5%, and 95.6%respectively (Table S1, Supporting Information). Through sys-tematic measurements presented in Figure 4F, the FL retentionrate refers to the percentage variations in detectable FL signalsfrom intense to dimmed at extremely low HSA concentrationsas the dilution ratio increases. First, the FL retention rates ofthree platforms are 100% in the initial phase (80x diluted urinesamples) and drop toward 80% with doubling of the dilution ra-tio (160x). When the dilution ratio is adjusted to 320x, the mi-crofluidic and metasurface platforms retain their 70%, while themicroplate platform experiences a dramatic decline to 29%. Atthis time, the FL retention rate of the microplate is less thanhalf of that of the other two platforms. Next, 7%, 18% and 38%FL retention rates are recorded under the dilution ratio of 640xwhere metasurface platform is 2.1 folds and 5.4 folds more in-tense than the microfluidic and microplate, respectively. Afterthat, the metasurface maintains 27% of FL signal with furtherdilution (1280x) whereas the microplates and microfluidics al-most lose it completely (2%). Similar circumstances still persistuntil the final phase of 2560x with near 10% retention rate inmetasurface. Moreover, microplate, microfluidic and metasur-face platforms have 300 ng mL−1 (LOD1), 150 ng mL−1 (LOD2),and 18.75 ng mL−1 (LOD3) as their respective detection limits,corresponding to the FL retention rate of 80%, 70% and 9% re-spectively. The above results of FL retention rate change are con-sistent with the performance of LODs in three platforms. Finally,the sensitivity level is expressed as: metasurface > microfluidic> microplate. The metasurface platform can effectively amplifythe FL signal at extremely low concentrations of HSA in humanurine environment. FL retention rate is still detectable even ifAdv. Optical Mater. 2024, 12, 2400868 2400868 (6 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbH 21951071, 2024, 23, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adom.202400868 by National Institute For, Wiley Online Library on [19/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advopticalmat.dewww.advancedsciencenews.com www.advopticalmat.deAdv. Optical Mater. 2024, 12, 2400868 2400868 (7 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbH 21951071, 2024, 23, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adom.202400868 by National Institute For, Wiley Online Library on [19/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advopticalmat.dewww.advancedsciencenews.com www.advopticalmat.deurine specimens are diluted by 2560 times, which fully illustratesthat Si-nanocolumn matrices which enable HSA to immobilizeare the primary reason for producing FL enhancement.2.4. Underlying Mechanisms on Metasurface PlatformAfter understanding the optical properties, binding affinities, andFL kinetics of metasurface platform throughout the dynamicsensing process, the potential mechanism can be explained inFigure 5 displayed with the comparison of three models. Initially,large-scale wells located on microplate platforms enable numer-ous conjugates of HSA and TPE-4TA to freely move in multi-dimensional directions. Uniformly dispersed FL molecules arehardly restrained in camera zone, resulting in the limited FLsignal in large liquid dimensions. Remarkably, microfluidic andmetasurface platforms further compress the scope of FL ac-tivities along the height direction, thus enabling us to acquirethe focused FL images, which aligns with the results presentedin Figure 4E. More significantly, the Si nanocolumns allow FLmolecules to be stacked in a non-uniform manner within an ex-tremely small area and a local enrichment state occurs, thus en-hancing FL output. Particularly, the reinforced effect of resonantEM fields further amplifies FL emission of HSA+TPE-4TA fromthe outermost surface of Si nanocolumns through multi-level im-mobilizations. Moreover, the results from FL kinetics (Figure 4B)reveal that FL enhancement is subject to the total fixed num-ber of conjugates of HSA and TPE-4TA in metasurface area. FLmolecules immobilization reaches its saturation when the con-centration of HSA is excessive (>160 μg mL−1). The amplifyingeffect can only occur on the bound sections and does not causepromotion with HSA concentration continuing to increase anylonger. Conversely, the microalbumin or trace range does nothave the saturation effect, which is the reason why the metasur-face is more significant for it.2.5. Future Prospects on Metasurface PlatformThis research offers a feasible strategy to achieve enhanced HSAdetection using AIE FL biosensor incorporating with metasur-face platform. Through the reinforced resonant EM fields of Sinanocolumns, optical signal sensitivity of analytes can be am-plified, which is particularly significant in the presence of manyinterferences in the urine scenarios. Improved LOD meets thehigher standard requirements of HSA detection at a low level.Besides that, the removal of HSA from human urine for en-hancing sample loading capabilities in analytical methods be-comes a promising pathway to identify low-abundant proteinsand improve their detection sensitivity. To be more specific,urine contains thousands of different types of proteins, whileHSA is one of the main proteins as the high-abundant pro-teins. At present, more than 3400 individual proteins have beenfound in urine that could become potential biomarkers for dif-ferent diseases and the majority of them belong to low-abundantproteins.[44,45] The masking effect of HSA poses an impeding fac-tor in urine for discovery/screening of less-abundant proteins.Si nanocolumns enable to deplete existing HSA content effi-ciently to act as the function of filtration. The superior opticalproperties provided by the metasurface platform are apparentlybeneficial for monitoring downstream process analysis of HSAremoval.Our metasurface platform also provides a universal mi-crofluidic design to be biofunctionalized for targeted biomarkerdetection via the combination of standard streptavidin-biotinmodel, immunobinding and FL labeling, which is fully ca-pable of applying to other biomarkers. First, the array of Si-nanocolumns constructed on SOI substrate has a larger sur-face area on which biomolecules are allowed to immobilize.Second, this platform can achieve high reflectance in visi-ble wavelength range to be compatible with optical propertiesof target objects by turning radius and periodicity. Third, FLmolecules specific to target biomarkers are selected carefullyto ensure that steady output of optical signal will be turnedon only after multi-level immobilization is completed success-fully. More significantly, the running cost can be reduced dra-matically because our metasurface substrates are reusable bywashing.All-dielectric metasurface platform has been illustrated onbiomarker detection in biosensing with extraordinary opticalcharacteristics that may lead to better sensitivity, higher through-put, and more significant specificity for extremely low-level ana-lytes in the future. The performance of several instances of all-dielectric metasurface platforms based on microfluidic FL sens-ing is summarized in this section as shown in Table 1. As meta-surface platform becomes better explored, it will have a potentialpromotion on biomarker detection in the medical diagnosis cri-terion to be a higher standard.3. ConclusionThis designed all-dielectric metasurface platform shows a reso-nance of 70% reflection response at 530 nm and can enhance FLemission at the outmost surface of Si nanostructures owing to thesignificant amplification of its resonant EM field. Subsequently,the multi-level immobilization as a detection strategy success-fully implements strong high-density bindings of FL moleculeson metasurface area. FL kinetics results prove its good photosta-bility and a good dynamic range of 0–160 μg mL−1 especially ultra-sensitive to the trace HSA. In the comparison of three platforms,metasurface platform is superior to microplate and microfluidicplatforms, which exhibits 1001% FL enhancement, and its FLFigure 4. A) Stability of FL in HSA detection within 2 h using the metasurface platform; B) The correlation of HSA and TPE-4TA from 0 to 160 μg mL−1using the metasurface platform. Inset graphs are change in FL images at 0 and 160 μg mL−1; C) The corresponding standard curve of TPE-4TA forHSA detection in the range of 0–160 μg mL−1 using the metasurface platform based on (B); D) The difference of FL response with/without HSA ineach of platform; E) The comparison of FL enhancement among the microplate platform, the microfluidic platform and the metasurface platform; F) FLretention rate between three platforms in urine scenarios with different dilution ratios where dilution ratios refer to total volume/treated urine volumewith PBS buffer as diluent. Three broken horizontal lines represent corresponding LODs in different platforms (LOD1: the microplate platform (orange),LOD2: the microfluidic platform (green), LOD3: the metasurface platform (purple)).Adv. Optical Mater. 2024, 12, 2400868 2400868 (8 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbH 21951071, 2024, 23, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adom.202400868 by National Institute For, Wiley Online Library on [19/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advopticalmat.dewww.advancedsciencenews.com www.advopticalmat.deFigure 5. Potential mechanism of the metasurface platform compared to the microplate and microfluidic platforms under their nanostructures.signal is 7.93-fold and 4.69-fold stronger than the others. Further-more, the results in urine testing demonstrate the highest recov-ery rate (96%), the lowest LOD (18.75 ng mL−1) and the biggest FLretention rate (9%) in urine up to 2560 times diluted among threeplatforms. The underlying mechanism concludes that a local en-richment state on the metasurface area and space constraintsfrom microfluidic paths are the main reasons for enhancing FLemission. In summary, the AIE-based FL biosensor incorporat-ing a metasurface platform can effectively achieve FL enhance-ment in HSA detection. The combination of AIE fluorogens andmetasurface platform opens up a new route for biosensing in realscenarios.4. Experimental SectionMaterials and Instruments: The FL biosensor TPE-4TA had been syn-thesized according to previous publications.[38] Human serum albumin(HSA) was purchased from Sigma–Aldrich (A1653-500MG, Germany). Bi-otin labeled HSA antibody (Biotin-HSA Ab) was purchased from Abcam(ab27632, purified, UK). Cys-streptavidin (Cys-SA) was purchased fromClick Biosystems (PRO1005, Richardson, USA). Raw human urine sam-ples were purchased from LEE BioSolutions (991-03-S, USA). Phosphate-buffered saline (PBS) and PBS-Tween 20 (PBS-T) were purchased fromFujifilm Wako Pure Chemical, Japan (164-25511 and 163–24361, respec-tively). Nanosep Centrifugal devices were purchased from Pall Life Sci-ences (OD300C34, USA). HiLyte Fluor 555 (HL555) Labeling Kit waspurchased from Dojindo Molecular Technologies (LK14, Japan) and theTable 1. Summary of FL detection of biomarkers on the metasurface platform. AF555 denotes Alexa Flour 555, HL555 HyLite Flour 555, and HEX a FLprobe on DNA.No. FLcomponentBiomarker Performance Refs.Dynamic range LOD1 TPE-4TA HSA 0–160 μg mL−1 18.75 ng mL−1 This work2 AF555 IgG 5–2000 pg mL−1 5 pg mL−1 [18]3 HL555 PSA and CEA 0.16–1000 ng mL−1;0.002–25 ng mL−11 ng mL−1;0.002 ng mL−1[19]4 HL555 COVID-19 glycoproteinpeptide and correspondingAb0.16–100 ng mL−1;6.25–100 ng mL−10.64 pg mL−1;1.56 ng mL−1[46]5 HEX cfDNA 0–2 fm 0.488 am [20]6 HEX SARS-CoV2 5–4000 am 5.86 am [47]Adv. Optical Mater. 2024, 12, 2400868 2400868 (9 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbH 21951071, 2024, 23, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adom.202400868 by National Institute For, Wiley Online Library on [19/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advopticalmat.dewww.advancedsciencenews.com www.advopticalmat.delabeling operation was done according to standard protocols. Human Al-bumin SimpleStep ELISA Kit was purchased from Abcam (ab227933, UK)and the operation procedure followed the standard instructions in themanual.An all-dielectric metasurface manufacturing procedure was carried outthrough the use of electron-beam lithography and dry etching on SOIwafers according to our previous publication.[18,48] The six metasurface re-gions in this investigation were distributed in the middle of each substratewith the dimension of 45 × 45 mm2, corresponding to six microfluidicflow in-out channels, as seen in Figure 1A. Notably, the metasurface sub-strates could be reused after being washed with a complex acid solution(H2SO4 + H2O2).The FL setup is illustrated in Figure 1B. The Light-emitting device wasacquired from Thorlabs (M365FP1, USA). 10× objective lens was acquiredfrom Mitsutoyo (M Plan Apo, Japan). 16-bit FL images were recorded byan uncooled CCD camera (Infinity-3S, Teledyne-Lumenera, Canada). FLspectra were measured by Spectrofluorometer (FP-8500, Jasco Interna-tional, Japan). The binding affinity was obtained from the surface plasmonresonance instrument (SPR, BIACORE-X100, General Electric Healthcare,USA).Preparation of Samples: AIE FL biosensor TPE-4TA solution(1000 μg mL−1) as the stocking solution was dissolved in PBS bufferand stored at 4 °C in the dark environment. Stocking solution wouldbe diluted freshly with PBS buffer to 20 μg mL−1 in every experimentunless otherwise noted. The freshly made HSA solution was formulatedto 1000 μg mL−1 in PBS buffer and diluted to the required concentrationof 0–1000 μg mL−1 with a specific concentration gradient. 200 μg mL−1 ofCys-SA stocking solution and 1 mg mL−1 of Biotin-HSA Ab were dilutedto 20 and 5 μg mL−1 with PBS buffer for daily use, respectively. PBS-Tbuffer was diluted with PBS buffer (pH 7.4) three times for rinsing use.Mixing solutions of HSA and TPE-4TA were incubated for 10 min beforeuse.In the Results Section 2.2, in order to ensure that all componentswere fixed tightly on the gold thin film, Cys-SA, Biotin-HSA Ab, HSA, andTPE-4TA were all formulated to relatively high working concentration of20 μg mL−1 during the experiments of binding response. For non-specificabsorption, HL555- HSA antibodies were set to 0–5 μg mL−1 in Figure 2C.In human urine testing, raw urine samples (991-03-S, Lot. 18-06-615,Lee Biosolutions, MO, USA) would be thawed from −20 °C to room tem-perature. Afterward, centrifugal filters with 300K Omega film were used tocentrifuge for 10 min with the rotation speed of 120 × 100 g and this op-eration must be repeated three times for each experiment until a sufficientamount of treated urine was collected for further use. Finally, the treatedurine could be prepared into different urine samples through specific dilu-tion factors with PBS. The mixing solutions of urine samples and TPE-4TAwere incubated for 10 min before measurement.Preparation of Metasurface Platform: The metasurface platform was adynamic measurement that composed of a self-absorbed pair of a meta-surface substrate and a polydimethylsiloxane (PDMS) microfluidic chip(thickness = 2 mm) that had six flow-in and flow-out independent mi-crofluidic channels. These microfluidic paths (height = 30 μm) were con-nected with tubes and the flowing speed could be controlled by a ro-tary pump (RP-6R01S-5A-DC3V, Takasago Fluidic Systems, Japan). In themeanwhile, this metasurface substrate was a three-layer planar structure(from top to bottom) consisting of the Si-nanocolumns array (with a di-ameter of 220 nm, a height of 200 nm, and a period of 300 nm), a 375 nmSiO2 layer and a 675 μm base Si wafer (Figure 2A).Preparation of Microplate Platform and Microfluidic Platform: Mi-croplate platform was based on the conventional transparent 96-wellsplate (P96F03N, Sansho, Japan) for static measurement. Microfluidic plat-form was assembled by a PDMS microfluidic chip and a normal substrate(a SiO2 layer and a base Si wafer without the Si-nanocolumns array) fordynamic detection. Microfluidic configuration remained consistent acrossmicrofluidic platform and metasurface platform.Microfluidic Flowing Program: The microfluidic flowing program wasdivided into the following steps: control of reagent flow time and rate,multiple reagents switching, FL imaging and system cleaning and drying.In order to achieve the multiple immobilizations on Si-nanorods, differ-ent sample liquids needed to be injected step by step through stainless-steel pins connected to the external inlet and outlet tubes for passing oversix metasurface regions where flowing variation among different channelswas ≈5%. First, PBS buffer was prepared for 3 min preflow with the flowingrate of 102.8 μL min−1 to fill the microfluidic paths. Second, Cys-SA wasflowed at 10.9 μL min−1 on metasurface area for 11 min and PBS rinsewas performed at the flowing rate of 18 μL min−1 for ≈7 min. Afterward,FL background measurements were recorded in PBS environment. Next,Biotin-HSA Ab was also performed to flow at 10.9 μL min−1 for 11 minthen PBS rinsing at 18 μL min−1 for 7 min. The same step was repeatedwhen the prepared mixture solution of FL biosensor TPE-4TA and HSAflowed on the Si-nanocolumn array (using PBS-T buffer for rinsing insteadof PBS buffer). Ultimately, actual FL images were captured then analyzedby deducting the preceding background effects. Besides, system cleaningwould be carried out with neutral washing solutions and pure water fortwo sections (rinse cleaning of tubes and ultrasonic cleaning of microflu-idic chips & metasurface substrates) after FL measurement. Overall, thismicrofluidic flowing program required ≈70 min flowing process, 30 minFL measurement, and 50 min cleaning and drying, respectively.FL Measurement: In FL setup, UV LED emitted 360 nm excitation light,which focused on the metasurface through an objective lens of numericalaperture (NA) 0.28. After that, beam splitter filtered out the reflected lightwith the wavelength less than 409 nm, meanwhile emissive FL was col-lected by objective lens, and transmitted toward a 529–571 nm bandpassfilter, then was detected eventually by an uncooled CCD camera. The expo-sure time was appropriately adjusted according to different platforms andthe gain was fixed at 10 during every experiment. The excitation wavelengthwas set to 360 nm and the emission wavelength was detected within therange of 529–571 nm.Supporting InformationSupporting Information is available from the Wiley Online Library or fromthe author.AcknowledgementsQ.H. gratefully acknowledges the financial support from the Flinders Uni-versity Research scholarship (FURS), and International Cooperative Grad-uate Program (ICGP) Fellowship under the Flinders University – NIMSCooperative Graduate Program. M.I. thanks for financial support fromthe NIMS Priority Research Project “Biomaterials” and JSPS KAKAENHIGrant Number JP24K01389. This study was supported in part by AdvancedResearch Infrastructure for Materials and Nanotechnology (ARIM) of theMEXT, Japan (Proposal Number JPMXP1223NM5163) and by the super-computing resources in Cyberscience Centre, Tohoku University, Japan.The authors thank AIEgen Biotech Co., Ltd (Hong Kong) for supplyingTEP-4TA for this study.Open access publishing facilitated by Flinders University, as part of theWiley - Flinders University agreement via the Council of Australian Univer-sity Librarians.Conflict of InterestThe authors declare no conflict of interest.Data Availability StatementThe data that support the findings of this study are available on requestfrom the corresponding author. The data are not publicly available due toprivacy or ethical restrictions.KeywordsAIE FL biosensors, HSA detection, metasurface platform, multi-level im-mobilizations, Si-nanocolumnsAdv. Optical Mater. 2024, 12, 2400868 2400868 (10 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbH 21951071, 2024, 23, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adom.202400868 by National Institute For, Wiley Online Library on [19/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advopticalmat.dewww.advancedsciencenews.com www.advopticalmat.deReceived: March 30, 2024Revised: May 14, 2024Published online: June 4, 2024[1] J. Hu, S. Bandyopadhyay, Y.-H. Liu, L.-Y. Shao, Front Phys 2021, 8,586087.[2] J. Canet-Ferrer, Metamaterials and metasurfaces, (Ed: J. Canet-Ferrer),BoD–Books on Demand, Norderstedt, Germany 2019.[3] A. Ali, A. Mitra, B. Aïssa, Nanomaterials 2022, 12, 1027.[4] H. H. Hsiao, C. H. Chu, D. P. Tsai, Small Methods 2017, 1,1600064.[5] P. Gutruf, C. Zou, W. Withayachumnankul, M. Bhaskaran, S. Sriram,C. Fumeaux, ACS Nano 2016, 10, 133.[6] L. Li, T. Jun Cui, W. Ji, S. Liu, J. Ding, X. Wan, Y. Bo Li, M. Jiang, C.-W.Qiu, S. Zhang, Nat Commun 2017, 8, 197.[7] H.-S. Ee, R. Agarwal, Nano Lett. 2016, 16, 2818.[8] A. V. Kabashin, P. Evans, S. Pastkovsky, W. Hendren, G. A. Wurtz, R.Atkinson, R. Pollard, V. A. Podolskiy, A. V. Zayats, Nat. Mater. 2009,8, 867.[9] M. L. Tseng, Y. Jahani, A. Leitis, H. Altug, ACS Photonics 2020, 8,47.[10] H. Zhu, S. Cheung, K. L. Chung, T. I. Yuk, IEEE Trans Antennas Propag2013, 61, 4615.[11] M. Farmahini-Farahani, H. Mosallaei, Opt. Lett. 2013, 38, 462.[12] Y. Bao, X. Ren, S. Liao, Q. Xue, ACS Appl. Mater. Interfaces 2023, 15,40022.[13] S. Zhang, C. L. Wong, S. Zeng, R. Bi, K. Tai, K. Dholakia, M. Olivo,Nanophotonics 2020, 10, 259.[14] R. Liu, X. Ye, T. Cui, Research 2020, 2020, 7949037.[15] I. O. Oguntoye, B. K. Simone, S. Padmanabha, G. Z. Hartfield, P.Amrollahi, T. Y. Hu, A. J. Ollanik, M. D. Escarra, ACS Appl. Nano Mater.2022, 5, 3983.[16] Y. Wang, M. A. Ali, E. K. Chow, L. Dong, M. Lu, Biosens. Bioelectron.2018, 107, 224.[17] A. Negm, M. M. Howlader, M. Bakr, S. Ali, Mater. Des. 2023, 227,111747.[18] M. Iwanaga, ACS Nano 2020, 14, 17458.[19] M. Iwanaga, Biosensors 2023, 13, 377.[20] M. Iwanaga, T. Hironaka, N. Ikeda, T. Sugasawa, K. Takekoshi, NanoLett. 2023, 23, 5755.[21] S. Cheung, B. Guan, S. Djordjevic, K. Okamoto, S. Yoo, 2012 Con-ference on Lasers and Electro-Optics (CLEO), IEEE, San Jose, CA, USA2012.[22] F. Van Laere, T. Claes, J. Schrauwen, S. Scheerlinck, W. Bogaerts, D.Taillaert, L. O’Faolain, D. Van Thourhout, R. Baets, IEEE PhotonicsTechnol. Lett. 2007, 19, 1919.[23] J. Ferrara, W. Yang, L. Zhu, P. Qiao, C. J. Chang-Hasnain, Opt. Express2015, 23, 2512.[24] Y. Hong, J. W. Lam, B. Z. Tang, Chem. Soc. Rev. 2011, 40, 5361.[25] X. Zhang, B. Yao, Q. Hu, Y. Hong, A. Wallace, K. Reynolds, C. Ramsey,A. Maeder, R. Reed, Y. Tang, Mater. Chem. Front. 2020, 4, 2548.[26] Q. Hu, B. Yao, T. C. Owyong, S. Prashanth, C. Wang, X. Zhang, W. W.Wong, Y. Tang, Y. Hong, Chem Asian J 2021, 16, 1245.[27] A. R. Dyer, P. Greenland, P. Elliott, M. L. Daviglus, G. Claeys, H.Kesteloot, H. Ueshima, J. Stamler, Am J Epidemiol 2004, 160, 1122.[28] J. A. Simerville, W. C. Maxted, J. J. Pahira, Am. Fam. Physician 2005,71, 1153.[29] S. Bouatra, F. Aziat, R. Mandal, A. C. Guo, M. R. Wilson, C. Knox, T.C. Bjorndahl, R. Krishnamurthy, F. Saleem, P. Liu, PLoS One 2013, 8,e73076.[30] A. Birková, J. Oboril, R. Kréta, B. Čižmárová, B. Hubková, Z.Šteffeková, J. Genči, J. Paralič, M. Mareková, Biomed Signal ProcessControl 2020, 56, 101693.[31] H. P. Patel, Pediatr Clin North Am 2006, 53, 325.[32] S. Aitekenov, A. Gaipov, R. Bukasov, Talanta 2021, 223, 121718.[33] B. Choi, M. Iwanaga, H. T. Miyazaki, Y. Sugimoto, A. Ohtake, K.Sakoda, Chem. Commun. 2015, 51, 11470.[34] M. Iwanaga, B. Choi, H. T. Miyazaki, Y. Sugimoto, Nanoscale 2016, 8,11099.[35] M. Iwanaga, Appl. Sci. 2018, 8, 1328.[36] M. Iwanaga, Plasmonic Resonators: Fundamentals, Advances, and Ap-plications, Jenny Stanford Publishing, New York 2016.[37] A. Irrera, A. A. Leonardi, C. Di Franco, M. J. Lo Faro, G. Palazzo, C.D’Andrea, K. Manoli, G. Franzo, P. Musumeci, B. Fazio, ACS Photonics2018, 5, 471.[38] Y. Tu, Y. Yu, Z. Zhou, S. Xie, B. Yao, S. Guan, B. Situ, Y. Liu, R. T. Kwok,J. W. Lam, ACS Appl. Mater. Interfaces 2019, 11, 29619.[39] Z. Luo, B. Liu, K. Zhu, Y. Huang, C. Pan, B. Wang, L. Wang, Dyes Pigm.2018, 152, 60.[40] Y. Huang, T. Lv, T. Qin, Z. Xu, L. Wang, B. Liu, Chem. Commun. 2020,56, 11094.[41] K. Zhu, T. Lv, T. Qin, Y. Huang, L. Wang, B. Liu, Chem. Commun. 2019,55, 13983.[42] A. Zorzi, S. Linciano, A. Angelini, MedChemComm 2019, 10, 1068.[43] M. Fasano, S. Curry, E. Terreno, M. Galliano, G. Fanali, P. Narciso, S.Notari, P. Ascenzi, IUBMB Life 2005, 57, 787.[44] L. Santucci, G. Candiano, A. Petretto, M. Bruschi, C. Lavarello, E.Inglese, P. G. Righetti, G. M. Ghiggeri, J. Proteomics 2015, 112, 53.[45] A. Marimuthu, R. N. O’Meally, R. Chaerkady, Y. Subbannayya, V.Nanjappa, P. Kumar, D. S. Kelkar, S. M. Pinto, R. Sharma, S. Renuse,J. Proteome Res. 2011, 10, 2734.[46] M. Iwanaga, W. Tangkawsakul, Biosensors 2022, 12, 981.[47] M. Iwanaga, Biosensors 2022, 12, 987.[48] M. Iwanaga, Biosens. Bioelectron. 2021, 190, 113423.Adv. Optical Mater. 2024, 12, 2400868 2400868 (11 of 11) © 2024 The Authors. Advanced Optical Materials published by Wiley-VCH GmbH 21951071, 2024, 23, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/adom.202400868 by National Institute For, Wiley Online Library on [19/08/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advopticalmat.de Metasurface Platform Incorporating Aggregation Induced Emission Based Biosensor for Enhanced Human Serum Albumin Detection 1. Introduction 2. Results and Discussion 2.1. Optical Properties of Metasurface Platform 2.2. Detection Strategy and Binding Affinities on Metasurface Platform 2.3. FL Kinetics on Metasurface Platform 2.4. Underlying Mechanisms on Metasurface Platform 2.5. Future Prospects on Metasurface Platform 3. Conclusion 4. Experimental Section Supporting Information Acknowledgements Conflict of Interest Data Availability Statement Keywords