Guided mode resonance-driven photoluminescence enhancement and angular emission control in a 2D dielectric photonic lattice†

Jagriti Ahuja^a, Saurabh Pandey^a, Shital Devinder^b, Jyoti Sardana^a, Anupriya Tiwari^a, Nagarajan Subramaniyam^c and Joby Joseph*^ade
aPhotonics Research Lab, Physics Department, Indian Institute of Technology, Delhi, New Delhi, Delhi 110016, India. E-mail: joby@iitd.ac.in
^bCentre for Sensors, Instrumentation, and Cyber Physical System Engineering, Indian Institute of Technology Delhi, New Delhi, Delhi, India
^cX-Fold Imaging oy, FL-00076 Espoo, Finland
^dOptics and Photonics Centre, Indian Institute of Technology Delhi, New Delhi, Delhi 110016, India
^eIndian Institute of Technology Delhi – Abu, Dhabi, Zayed City, Abu Dhabi, United Arab Emirates

First published on 30th July 2025

---

---

1. Introduction

Photoluminescence (PL) engineering has become an essential component of nanophononics, facilitating improved control over light emission in optoelectronic devices,^1,2 optical sensors,^3–5 and quantum optics.^2,6,7 Generally, spontaneous emission in bulk and nanostructured materials occurs in all directions,⁸ resulting in less-than-ideal use of emitted photons. A major hurdle in this field is the effective extraction and control of the directionality of emitted photons, which is essential for enhancing the performance of light-emitting devices.

Integrating photonic nanostructures with emitter layers has significantly improved light emission control, enhancing both PL intensity and directionality.^9,10 Various periodic nanostructures, including plasmonic^11–13 and dielectric configurations,^14–17 have been explored to tailor emission properties. Plasmonic nanostructures leverage plasmon resonance to enhance light–matter interactions; however, they suffer from intrinsic ohmic losses¹⁸ and PL quenching due to nonradiative energy transfer to metals,¹⁹ limiting their efficiency. In contrast, dielectric periodic nanostructures, such as photonic crystals, topologically engineered platforms^20,21 and metasurfaces, enable efficient light confinement, high-quality factor, and coupling without these losses.^22,23 The periodic arrangement of nanostructures leads to resonant nanophotonic effects mediated by diffraction effects such as Rayleigh anomalies and surface lattice resonance. The close proximity of the waveguide layer to such a lattice arrangement leads to a diffraction-coupled waveguide mode, known as guided-mode resonance (GMR).²⁴ Guided-mode resonance (GMR) in dielectric nanostructures integrated with a waveguide layer provides robust PL directional amplification²⁵ by modifying the local density of states (LDOS) of emitters^26,27 and by redistributing the emitted light into specific angles with minimal loss, outperforming other periodic structures.

Recent studies utilizing periodic structures have shown varying degrees of PL enhancement (N-fold) with controlled light emission via resonant light–matter interactions. Rigid dielectric metasurfaces and photonic crystal cavities have shown high PL enhancement factors by leveraging high-Q Fano resonances (1000 fold),²⁸ cavity mode resonance (2400 fold),²⁹ surface lattice resonance,³⁰ symmetry breaking and spin–momentum locking (45 fold)^20,21 or through collective Mie resonances (10 fold),³¹ but these require complex fabrication techniques such as e-beam lithography and semiconductor epitaxy, lack mechanical flexibility with limited scalability and exhibit limited Purcell effects with no angular control. However, for practical applications, it is crucial to fabricate these photonic nanostructures over large areas such as light-emitting diodes (LEDs),³² single-photon sources,^33,34 lasers^6,25,35,36 and bio-imaging systems,^37,38 ensuring uniform enhancement across the entire emission region, while maintaining the mechanical flexibility to ensure scalability and integration into real-world devices.

In this work, we present the design and study of a dielectric photonic array composed of patterned flexible polyethylene terephthalate (PET) in a square lattice with a TiO₂ thin layer coated over it. The fabrication is carried out through the nanoimprint lithography process, resulting in a large area periodic structure. This configuration integrates the flexibility and cost-effectiveness of PET with the high refractive index of TiO₂, functioning as a foundational element for improved light–matter interaction. When irradiated with transverse electric (TE)/transverse magnetic (TM) polarized light with respect to the sample, the nanostructure supports the corresponding GMR modes in the visible spectrum. A comprehensive insight into the origin of these modes has been elaborated to highlight their impact on the LDOS, which prominently plays a vital role in PL enhancement via the Purcell effect. The optical dispersion curves obtained experimentally and the associated mode analyses are corroborated by simulation findings utilizing the Lumerical finite-difference time-domain (FDTD) solver³⁹ and with analytical calculations.

Furthermore, we leveraged the strong local field confinement in the dielectric array to demonstrate the phenomenon of enhanced light emission by coating a thin layer of the fluorescent dye Coumarin 481. The enhancement mechanism is attributed to the strong interaction between the PL emission from the dye layer and the diffraction-related photonic modes supported by the structure. The presence of these resonance modes in the structure not only shows an enhancement of 110 times in the emitted light but also spatially redistributes the earlier broad PL emission from the dye into well-defined angles in specific angular directions.⁴⁰ Additionally, by collecting the emitted light at different angles, we establish a direct correlation between the optical extinction spectra and the PL dispersion curves for TE and TM polarization of incident light, further providing validation for the interaction between the GMR modes and the emitted fluorescence. Furthermore, back focal plane (BFP) imaging provides direct visualization of these spatially controlled emission patterns from the dielectric array. The insights gained from this work would significantly enhance tunable light emission applications for next-generation optical devices, including large area light-emitting diodes, polarization-sensitive biosensors, advanced spectroscopic tools, directional emitters, and fluorescence-based bioimaging.

2. Experimental methods

2.1 Sample design and fabrication

Fig. 1(a and b) illustrates a schematic of the designed PET–TiO₂ nanopillar photonic structure on a PET substrate, over which a homogeneous medium of 7-(diethylamino)-4-(trifluoromethyl)coumarin (Coumarin 481) was introduced by spin coating for the photoluminescence studies. The inset shows the photograph of the fabricated PET–TiO₂ array. The fabrication of the PET array involves two steps: (i) nano-imprint lithography and (ii) pulsed laser deposition to form a 30 nm thick layer of low-loss TiO₂ material on the PET pillars. The comprehensive fabrication details are outlined in section S1 of the ESI.†

The array's configuration shape is meticulously designed to ensure that the GMR under normal incidence aligns with the primary emission peak of the Coumarin dye at around 500 nm. The square periodic PET nanopillar, as shown in the scanning electron microscopy (SEM) image in Fig. 2(a), possesses a height (H) of 30 nm and a radius (R) of 85 nm, situated within a square lattice characterized by a periodicity (P) of 300 nm along the x and y axes (fill factor ff = 67.11%), validated by atomic force microscopy (AFM) (Fig. 2b). For the analysis, the plane of incidence (POI) is set to be in the x–z plane, such that the incident wave vector (k) resides in this plane. When the polarization of the electric field vector (E) is oriented orthogonal to the POI, i.e., along the y-axis, this configuration corresponds to TE polarization. In contrast, when E is located within the POI, it corresponds to TM polarization. A solution of C481 dye with a concentration of 1 mM in ethanol was mixed with an aqueous PVA solution (0.25 wt%) in equal parts. This mixture was spin-coated onto a nanostructured sample substrate at a speed of 1500 rpm for 30 seconds, resulting in a film thickness of 300 nm (section S2). The coated sample was then heated at 95 °C for 2 minutes to evaporate the solvents. The PVA–dye solution shows an absorption peak at 400 nm, while the thin film exhibits a fluorescence peak at 500 nm, with a full width at half maximum (FWHM) of 80 nm, as illustrated in Fig. S2 of the ESI.†

2.2 Optical extinction measurements

The extinction spectrum was obtained by varying the incidence angle θ, determined in the x–z plane, using a home-built setup. An x–y–z translational stage was used to place the sample, which was further placed on a rotational stage to vary θ_in in the x–z plane of the sample. The sample was exposed to polarized broadband light (300–2000 nm) from a halogen source (Ocean Optics Flame). A fiber with a 200 μm diameter positioned along the forward axis was employed to collect the zeroth-order transmission and direct it to the spectrometer (Flame-S, Ocean Insight). The extinction (E) of the sample is defined in percentage with respect to the PET bare substrate as a reference and calculated as 100 − T, where T is the normalized transmission in percentage. A schematic diagram of the transmission measurement setup is provided in Fig. S3.†

2.3 Emission measurement

A custom photoluminescence setup was used to measure the emission from the designed photonic array covered by the emissive layer. The sample was mounted on a specially designed goniometer to adjust the incidence and the collection angle. A diode-pumped continuous wave (CW) solid-state laser with a wavelength of 405 nm illuminated the sample with an intensity of 1.48 mW cm⁻² at an incident angle of θ_in. The emitted light was passed through a 425 nm long-pass filter to block the excitation light and then directed into a fiber connected to a spectrometer (Flame-S, Ocean Insight), which was calibrated initially with a fluorescence standard (rhodamine 6G) to ensure a consistent and linear spectral response. A rotation of the collection end of the fiber around the sample was performed to obtain the emission spectrum through the photonic array. This spectrum was obtained as a function of the emission angle, denoted as θ_em, which was determined in the x–z plane. The measurement setup is explained in detail in Fig. S3 of the ESI.†

To ensure the reliability of the emission data, all the measurements were performed on the ensembles of dye molecules, which effectively average out transient blinking events. Each PL measurement was acquired rapidly, with minimal exposure time for each sample to minimize the impact of photobleaching. Additional excitation power-dependent and photostability measurements on the dye-coated PET–TiO2 array were also conducted and are provided in section S3 of the ESI.†

2.4 FDTD simulations

FDTD simulations were carried out for numerical analysis of optical modes and their characteristics supported in the considered photonic structure, as shown in Fig. 1b. The structure consisted of PET cylindrical pillars coated with a TiO₂ layer. In the simulations, the nanocylinders of PET were modelled with a refractive index of n = 1.56, a diameter (D) of 85 nm, and a height (H) of 30 nm, covered with a 30 nm TiO₂ layer with a refractive index of 2.4. The nanocylinders were arranged in a square lattice with a period (P) of 300 nm. The model used a square unit cell with periodic Bloch boundary conditions in the x and y directions and perfectly matched layer (PML) boundary conditions in the z-direction. A broadband plane wave source illuminated the structure at normal incidence, with polarization angles of 90° and 0° for TE and TM polarized light, respectively. Frequency-domain field monitors were employed to capture the system's optical responses. The refractive indices and extinction coefficients of PET, TiO₂, and the superstrate layer were measured using the ellipsometry technique.

3. Results and discussion

The specified thickness and refractive indices of the TiO₂ and PET layers enable the composite structure to support two significant waveguide resonance modes at 491 nm and 515 nm, which appear as peaks in the extinction spectra (Fig. 3(a)) for both TE and TM polarization of incident light under normal illumination. The peaks observed in the extinction spectrum are due to the destructive interference between the leaky waveguide mode and the zeroth transmitted order and referred to as guided-mode resonance.^41,42 The analytical solution of diffraction from a 2D periodic lattice was determined and waveguide mode analysis was carried out to understand the spectral position of excited guided-mode resonance. For a 2D array, the diffraction equation is defined by the relationship (1):⁴³

	(1)

where is the wave vector of diffracted orders (m₁, m₂) with the diffraction angle ‘θ_d’ in transmission mode, which passes through the waveguide of refractive index n_w. The in-plane component of the incident wave vector falling on the grating at the angle of incidence ‘θ_in’ is defined as , while is the reciprocal lattice vector of the array, with the primitive reciprocal lattice vectors and , provided as

	(2)

The relationship (1) for the normal angle of incidence (θ_i = 0) is further reduced to eqn (3) for the periodicity ‘a’, providing the effective refractive index satisfied by the diffracted orders in the waveguide media.¹³

	(3)

The effective index of the diffracted orders is calculated and plotted along with the effective index of the TM and TE waveguide modes supported by the designed photonic array, as shown in Fig. 3(b). A waveguide mode solver was used to evaluate the effective index of the waveguide modes⁴⁴ supported by the photonic array. Whenever the effective index of the diffracted order equals the effective index of the waveguide mode, it excites a GMR mode. It can further be inferred that the mode at 515 nm corresponds to the TE waveguide mode, while the mode at 491 nm is attributed to the TM waveguide mode⁴⁵ for TM polarization of input light. To further confirm the nature of the modes, the electric field distributions of each GMR mode corresponding to their respective resonant wavelengths along the XZ, XY, and YZ cross-sections are plotted in Fig. 4. The peak at 491 nm corresponds to propagation along the X-axis ( is parallel to (grating vector), TM mode), whereas for 515 nm, propagation along the Y-axis results in the TE mode ( is perpendicular to ). Furthermore, the polarization degeneracy of GMR under normal illumination is due to the cylindrical symmetry and square periodicity of the photonic array.⁴⁶ Detailed analysis of these modes, calculated using electromagnetic mode treatment of a slab waveguide, is presented in sections S5–S7 of the ESI.† Fig. 3(c) and (d) present the experimentally obtained optical extinction spectra of the photonic array as a function of the incident angle (θ_in) for both TE and TM polarized incident light, respectively. The simulated extinction spectra exhibit strong concordance with the observed extinction spectra (Fig. 3(e) and (f)). A translucent dashed square box in the dispersion curve indicates the dye emission band. As denoted in Fig. 3(c and e) for TM polarization, the (±1, 0) order shows a non-degenerate nature, i.e., they split into two branches at oblique incidence, and is associated with the TM-GMR (491 nm, propagation along the X-axis), while the (0, ±1) order is degenerate in nature, showing a nearly flat dispersion curve with no splitting at oblique incidence and is associated with the TE-GMR (515 nm, propagation along the Y-axis). In contrast, the reverse occurs for the TE polarization excitation as anticipated. In TM waveguide modes, the electric field vector under normal TM excitation (P-polarization) lies parallel to the plane of incidence (x–z plane)⁴⁷ and propagates along the x–z plane, producing a resonance at 491 nm. To gain more insight into the TM-GMR mode, we have plotted the simulated |E|² for a resonance wavelength of 491 nm in the x–y, y–z, and x–z planes (Fig. 4a). The high refractive index contrast between the TiO₂ layer and the surrounding medium results in a substantial near-electric field enhancement (|E|² > 40) at the top edges of the TiO₂–PET nanostructures, which in turn facilitates mode propagation along the x direction. TE waveguide modes, on the other hand, show distinct behaviour. The electric field vector is directed orthogonally to the direction of propagation, in the y–z plane.⁴⁵ Fig. 4b shows the electric field confinement for TE-GMR mode in the x–y, y–z, and x–z planes, revealing the mode confinement along the y direction. It ensures that the electric field is localized within the structure, which is crucial for fluorescence enhancement in the presence of an emitting dye.

The improvement of photoluminescence in the nanostructured array is determined collectively by extended light–emitter interaction, modifications in radiative decay rates, and efficient light extraction.¹⁰ Excitation of GMR aligned with the emission wavelength of the emitter results in near-field coupling of confined electromagnetic field density. This boost in the interaction supports the increased local density of states (LDOS) with an improved radiative recombination rate. This is based on Fermi's “golden rule”, where the spontaneous emission rate of an emitter in the weak coupling regime can be modified by tailoring the LDOS.⁴⁸ Consequently, the presence of nanostructures modifies the LDOS of the emitters by providing new decaying channels in their near field, which directly affects the decay rate of the emitters.⁴⁹ The spontaneous emission decay rate is measured as⁵⁰

	(4)

where is the LDOS in the photonic environment and is Dyadic Green's function, which represents the electric field at due to a dipole source placed at ^50,51 and n is the dipole orientation unit vector.

The Purcell factor enhancement is one of the crucial parameters in quantifying the modification in the emission decay rate,⁵² which is the ratio of the LDOS in the presence of nanostructures to the free space LDOS. The Purcell factor is expressed as:^52,53

	(5)

where Γ and Γ₀ are the spontaneous emission decay rates in the presence of nanostructures and in free space, respectively, and and ρ₀(ω) are the corresponding LDOS values. The Green's function-based method takes into consideration the distance between the dye molecules and the photonic lattice, which is an important factor in measuring emission enhancement.

The Purcell factor is often approximated using the formula

	(6)

where signifies the effective mode volume,^54,55 Q represents the quality factor of the GMR mode, λ denotes the GMR wavelength, and n indicates the refractive index of the surrounding medium. As discussed earlier, the designed photonic array supports TE- and TM-GMR modes with intense electric field confinement, thereby substantially modifying the LDOS by concentrating electromagnetic energy in the vicinity of the emitters. The relationship between the Purcell factor and GMR modes is expressed as⁵⁶

	(7)

where is the amplitude of the GMR mode at the position where the emitters are placed. To understand the emission mechanism behind the PL enhancement in the designed photonic array, we have calculated the spontaneous decay rate (Γ) numerically using Green's function approach implemented in Ansys Lumerical FDTD, by modelling the dye emitters as a point dipole source placed over the periodic structure.⁵⁷ The dyadic Green's function G was extracted from the time-domain field response using the Fourier transform method, whose imaginary part gives the LDOS (ρ_d).

As shown in Fig. 5a, an enhancement in the LDOS of approximately 100 at a wavelength of 515 nm was observed for emitters placed at the top nanocylinders, indicating a sharp increase in the density of states in comparison with the free space. This peak aligns with the TE-GMR condition at λ = 515 nm of the photonic array. This increase of the LDOS, as indicated by eqn (5), is precisely proportional to the ratio of the radiative decay rate and gives a Purcell Factor enhancement of 100. The analytically derived decay rate curve, obtained using eqn (4), shows a notable increase the TE-GMR wavelength for the photonic array relative to the background decay in air, demonstrating effective emitter-mode coupling and increased spontaneous emission, as shown in Fig. 5b. The spatial and orientational dependence of the relative LODS for the emitter is further discussed in section S8 of the ESI.† Both the relative LDOS and the spontaneous decay rate curve show a relatively smaller peak, λ = 495 nm, corresponding to the TM-GMR condition. This reduced LDOS is attributed to its lower Q factor and broader spectral width compared to the TE-GMR mode. This indicates that the observed increase in emission from the emitter layer is attained through the utilization of the waveguide modes facilitated by the photonic array. For completeness, the Purcell factor is also calculated using eqn (6) through FDTD simulation, and the results are discussed in section S9.

To further support the calculated LDOS enhancement, time-resolved photoluminescence (TRPL) measurements were performed, which reveal a significant drop in the average dye PL lifetime from 3.37 ns on the reference substrate to 0.15 ns when coupled to the photonic array at the GMR wavelength (515 nm). This substantial reduction in lifetime gives direct experimental proof of the Purcell factor enhancement, confirming that the enhanced LDOS indeed leads to a higher spontaneous emission rate. The detailed decay fitting and analysis are provided in section S10 of the ESI.†

4. Directional amplification and shaping of emission

The photonic array fulfills two functions: it not only increases the light emission but also acts as an outcoupling mechanism mediated by the emission wavelength, satisfying the GMR phenomena and following the same angular dispersion in the far field, so enabling directional amplification with angular emission control of emitter light. Fig. 6(a) illustrates the relationship between the extinction peaks at normal incidence (θ_in = 0°) and the photoluminescence enhancement at normal emission (θ_em = 0°), which confirms the occurrence of the outcoupling mechanism in the designed photonic array. The photoluminescence spectrum of the dye-coated glass substrate works as a reference sample, as shown by the grey curve in Fig. 6(a). The photonic array facilitates resonant coupling between the dye's incoherent emission and GMR modes, enabling directed amplification of photoluminescence intensity. The distinct resonance characteristic of the designed photonic array, characterized by a full width at half maximum (FWHM) of 9 nm for TE-GMR and 10 nm for TM-GMR (Q ∼50), leads to two spectrally narrow photoluminescence peaks with electromagnetic confinement near the emitter in the normal emission. Furthermore, it establishes an interaction between PL emission and leaky waveguide modes, leading to a 110-fold increase in the directional amplification of PL peak intensity compared to the reference sample. The integral PL intensity ratio is also calculated over the emission band (400–560 nm), showing a similar increase of almost 116 times, which suggests that the observed increase represents a significant increase in the total light emitted because of resonant coupling.

The interaction of PL emission with GMR is further confirmed by examining angle-resolved photoluminescence spectra for TE- and TM-polarized excitation (λ = 405 nm), as shown in Fig. 6(b and c), where the sample was excited normally and emission was recorded as a function of θ_em. The momentum-space behavior of the emission spectra is consistent with the angular GMR extinction dispersion observed, as shown in Fig. 3(c and d). This directly supports the excitation of waveguide leaky mode by the emission wavelengths of the emitter, which corresponds to various k vectors of the emitted light, ultimately leaking and coupling with zeroth-order diffracted light. The PL amplification takes place precisely at wavelengths that align with the extinction peaks observed in the dispersion curves, with no anti-crossing or Rabi splitting behavior, showing the absence of strong coupling.^58,59 Furthermore, the dye emission exhibits a relatively broad spectral bandwidth (∼80 nm), which exceeds the resonance splitting required to resolve vacuum Rabi modes.^60,61 This suggests that the emission is notably influenced by the resonant excitation of moderate Q-factor of GMR modes, thereby affirming that the robust near-field coupling between the emitters and the waveguide modes is via the Purcell effect (weak coupling regime) and plays a pivotal role in engineering the directionality of light emission.

Furthermore, we recorded the back focal plane images of the PL emission pattern to validate its directional amplification abilities, where the emission pattern was recorded in the momentum space from the photonic array consisting of the emitter layer. These images contain information about the intensity of emitted light in different directions in the Fourier plane. BFP images were recorded using a home-built setup (Fig. 7a) with a collimated 405 nm continuous wave laser as the excitation source. The PL emission from the sample is collected using a Nikon Plan Fluor 40× objective with 0.75 numerical aperture (NA) and directed toward a CCD camera after propagating through a dedicated 4f lens system and a 425 nm long pass filter. The PL intensity distribution in momentum space, PL(k_x,k_y), is represented by the back-focal plane of the collection objective, which is imaged using the CCD camera. This distribution is then transformed into the angular distribution, PL(θ,ϕ), using the following coordinates:^62,63

PL(ky,ky) = PL(θ,ϕ)(cosθ)−1	(8)

where,

ky = k0sinθsinϕ	(9)

kx = k0sinθcosϕ	(10)

and .

where k₀ is the wave vector in free space and θ and ϕ represent the zenith and azimuth angles (which signify the k vector angle with the normal z direction and its projection angle in the xy plane) in spherical coordinates, respectively. The maximum angle θ_em for collecting the photoluminescence is constrained by the NA of the collection objective employed in the experiment, i.e., 0.75. A polarizer was employed in the detection path to differentiate various components of the emission. The far-field emission from the reference sample is utilized to compare with the emission pattern derived from the photonic array. The BFP image of the unpolarized PL from the dye-coated photonic array is shown in Fig. 7c. The white dotted circles indicate the greatest angle (θ_em = 48.5 degrees) that can be captured using an objective with a numerical aperture (NA) of 0.75 from the sample normal. The BFP image exhibits a 4-fold symmetry because of the diffraction from a square periodic lattice and is normalized to its maximal value to facilitate comparative analysis. It is observed that the bands of high emission intensity in the BFP image are superimposed on the background emission originating from dye molecules that remain uncoupled to any resonant mode supported by the structure. The difference in the emission pattern compared to the reference sample indicates strong near-field coupling between the waveguide GMR modes and the light emitted by the emitters. The periodicity of the photonic array imposes momentum-matching conditions that selectively enhance certain propagation directions, effectively redistributing the emitted photons along those directions. The interaction of localized emitter dipoles with the structured optical modes of the array generates this directional control by means of constructive and destructive interference, leading to modified radiation patterns of the dye emitters. Furthermore, we also analyzed BFP images of the emission for TE and TM polarization states to examine the polarization filtering response in the emission pattern. Fig. 7d and e illustrate the emission patterns obtained with a polarizer positioned just before the camera with its pass axis along the x and y directions (with respect to the sample), satisfying TM and TE conditions, respectively. The dependence of the out-coupled emission on polarization is clearly seen in these two images. The central lobe in the BFP image (Fig. 7(a)) shows a measured FWHM of 20° for the dye-coated photonic array, compared to 90° for the dye on glass, indicating stronger angular confinement with a side lobe suppression ratio (SLSR) of 2.83 dB. Further details on the quantification are provided in section S12 of the ESI.† This suggests the structure's photonic modes and its angular anisotropic features dictate emission behaviour in angular shaping, which can be tailored through polarization filtering of emitted light to a great extent. The detailed description of all the high-intensity bands observed in the BFP images is discussed in section S11 of the ESI.† Thus, the BFP imaging shown above not only offers clear proof of mode-specific coupling between the emitters and the photonic array but also sets the stage for a variety of advanced applications, including optical diagnostics, fluorescence polarization microscopy,^64,65 biosensing, and the design of very directional light-emitting diodes (LEDs).¹⁰

5. Conclusions

In conclusion, the study demonstrated a directional amplification and spectral shaping of photoluminescence from fluorescent dye molecules, in conjunction with the GMR modes supported by the large area fabricated dielectric photonic array. A direct corroboration of the guided mode satisfied by the emitters has been validated through a comprehensive angle-resolved emission spectrum analysis of the photonic array, which further demonstrates the polarization-sensitive emission spectra as the structure exhibits TE and TM polarization of incident light. The findings are substantiated by the study of the electromagnetic field and its influence on the local density of states that determines the dynamics of the PL emission rate, as demonstrated through numerical simulations. The LDOS of the emitters has been enhanced by a factor of 100 at the GMR wavelength due to the presence of the array, resulting in a significant increase in the spontaneous emission rate. Experimentally, we have achieved a 110-fold enhancement in PL counts from the dye emitters placed on the array, contrasting with the isotropically dispersed PL light from emitters on a glass substrate, which can be attributed to GMR effects. Furthermore, it spatially disperses the isotropic emission into clearly defined angular directions, which aligns with its angular dispersion characteristics. The enhancement in directional emission facilitated by GMR highlights the capability of the engineered array to precisely customize the far-field light emission profile.

While the PL enhancement observed in our structure closely matches the calculated LDOS enhancement, it is important to note that, in general, PL enhancement can differ from LDOS enhancement due to additional factors such as light extraction efficiency, nonradiative losses, and the specifics of experimental detection. Notably, this work provides a comprehensive quantitative demonstration of spatially uniform optical amplification and directional emission in large-area and industrially scalable photonic arrays, thereby linking large-scale nanofabrication with practical photonic applications.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting the findings of this study are available within the article and its Supplementary Information file ESI.† Additional datasets or simulation files generated and analyzed during the current study are available from the corresponding author upon reasonable request.

Acknowledgements

J. Ahuja, S. Pandey, S. Devinder, and J. Sardana acknowledge the Ministry of Education (MoE), Government of India, and IIT Delhi for support through a fellowship. A. T. acknowledges the Department of Science and Technology, Government of India, for financial aid through the INSPIRE Fellowship (IF190039). The authors gratefully acknowledge Dr Shereena Joseph for her valuable contributions to the theoretical and experimental discussions that greatly enriched this work. The authors acknowledge Dr Mohamed Abdelkhalik for having various insightful discussions to strengthen the findings of this work. The authors acknowledge the Central Research Facility (CRF), IIT Delhi, for access to FESEM and EDAX facilities. The authors also acknowledge funding from the Institute of Eminence (IOE).

References

1. C. H. Kang, Y. Wang, O. Alkhazragi, H. Lu, T. K. Ng and B. S. Ooi, APL Photonics, 2023, 8, 020903 CrossRef CAS .
2. T. T. H. Do, M. Nonahal, C. Li, V. Valuckas, H. H. Tan, A. I. Kuznetsov, H. S. Nguyen, I. Aharonovich and S. T. Ha, Nat. Commun., 2024, 15, 2281 CrossRef CAS PubMed .
3. D. Huang, Y. Bing, H. Yi, W. Hong, C. Lai, Q. Guo and C. Niu, Anal. Methods, 2015, 7, 4621–4628 RSC .
4. E. Benito-Peña, M. G. Valdés, B. Glahn-Martínez and M. C. Moreno-Bondi, Anal. Chim. Acta, 2016, 943, 17–40 CrossRef PubMed .
5. P. Kassal, E. Horak, M. Sigurnjak, M. D. Steinberg and I. M. Steinberg, Rev. Anal. Chem., 2018, 37, 20170024 CAS .
6. S. Wang, Q. Le-Van, T. Peyronel, M. Ramezani, N. Van Hoof, T. G. Tiecke and J. Gómez Rivas, ACS Photonics, 2018, 5, 2478–2485 CrossRef CAS .
7. T. Peyronel, K. J. Quirk, S. C. Wang and T. G. Tiecke, Optica, 2016, 3, 787 CrossRef .
8. S. Noda, M. Fujita and T. Asano, Nat. Photonics, 2007, 1, 449–458 CrossRef CAS .
9. Y. Zhai, C. Xu, Z. Zhang, P. Li, S. Murai, J. Gómez Rivas, X. Li and S. Wang, Nano Lett., 2024, 36, 11311–11318 CrossRef PubMed .
10. A. Vaskin, R. Kolkowaski and I. Staude, Nanophtonics, 2019, 8, 151–1198 Search PubMed .
11. W. Zhou, M. Dridi, J. Y. Suh, C. H. Kim, D. T. Co, M. R. Wasielewski, G. C. Schatz and T. W. Odom, Nat. Nanotechnol., 2013, 8, 506–511 CrossRef CAS PubMed .
12. G. Lozano, D. J. Louwers, S. R. K. Rodríguez, S. Murai, O. T. A. Jansen, M. A. Verschuuren and J. Gómez Rivas, Light: Sci. Appl., 2013, 2, e66 CrossRef .
13. K. Imaeda, R. Miyazaki, S. Ryuzaki and K. Ueno, J. Phys. Chem. C, 2025, 129, 5095–5104 CrossRef CAS .
14. S. Murai, G. W. Castellanos, T. V. Raziman, A. G. Curto and J. Gómez Rivas, Adv. Opt. Mater., 2020, 8, 1902024 CrossRef CAS .
15. P. Ding, M. Li, J. He, J. Wang, C. Fan and F. Zeng, Opt. Express, 2015, 23, 21477 CrossRef CAS PubMed .
16. S. I. Azzam, K. Chaudhuri, A. Lagutchev, Z. Jacob, Y. L. Kim, V. M. Shalaev, A. Boltasseva and A. V. Kildishev, Laser Photonics Rev., 2021, 15, 2000411 CrossRef CAS .
17. M. Jeong, B. Ko, C. Jung, J. Kim, J. Jang, J. Mun, J. Lee, S. Yun, S. Kim and J. Rho, Nano Lett., 2024, 24, 5783–5790 CrossRef CAS PubMed .
18. A. I. Kuznetsov, A. E. Miroshnichenko, M. L. Brongersma, Y. S. Kivshar and B. Luk'yanchuk, Science, 2016, 354, 116611 CrossRef PubMed .
19. P. Anger, P. Bharadwaj and L. Novotny, Phys. Rev. Lett., 2006, 96, 113002 CrossRef PubMed .
20. H. Siampour, C. O’Rourke, A. J. Brash, M. N. Makhonin, R. Dost, D. J. Hallett, E. Clarke, P. K. Patil, M. S. Skolnick and A. M. Fox, npj Quantum Inf, 2023, 9, 15 CrossRef .
21. H. Siampour, O. Wang, V. A. Zenin, S. Boroviks, P. Siyushev, Y. Yang, V. A. Davydov, L. F. Kulikova, V. N. Agafonov, A. Kubanek, N. A. Mortensen, F. Jelezko and S. I. Bozhevolnyi, Nanophotonics, 2020, 9, 953–962 CrossRef CAS .
22. A. I. Kuznetsov, A. E. Miroshnichenko, M. L. Brongersma, Y. S. Kivshar and B. Luk'yanchuk, Science, 2016, 354, 116611 CrossRef PubMed .
23. I. Staude, T. Pertsch and Y. S. Kivshar, ACS Photonics, 2019, 6, 802–814 CrossRef CAS .
24. P. K. Sahoo, S. Sarkar and J. Joseph, Sci. Rep., 2017, 7, 2045–2322 CrossRef PubMed .
25. J. Guan, L. K. Sagar, R. Li, D. Wang, G. Bappi, W. Wang, N. Watkins, M. R. Bourgeois, L. Levina, F. Fan, S. Hoogland, O. Voznyy, J. M. De Pina, R. D. Schaller, G. C. Schatz, E. H. Sargent and T. W. Odom, ACS Nano, 2020, 14, 3426–3433 CrossRef CAS PubMed .
26. M. Khokhar, F. A. Inam and R. V. Nair, Adv. Opt. Mater., 2022, 10, 2200978 CrossRef CAS .
27. W. L. Barnes, J. Mod. Opt., 1998, 45, 661–699 CrossRef CAS .
28. S. Yuan, X. Qiu, C. Cui, L. Zhu, Y. Wang, Y. Li, J. Song, Q. Huang and J. Xia, ACS Nano, 2017, 11, 10704–10711 CrossRef CAS PubMed .
29. X. Liu, T. Shimada, R. Miura, S. Iwamoto, Y. Arakawa and Y. K. Kato, Phys. Rev. Appl., 2015, 3, 014006 CrossRef CAS .
30. F. Laux, N. Bonod and D. Gérard, J. Phys. Chem. C, 2017, 121, 13280–13289 CrossRef CAS .
31. Y. T. Lin, A. Hassanfiroozi, W. R. Jiang, M. Y. Liao, W. J. Lee and P. C. Wu, Nanophotonics, 2022, 11, 2701–2709 CrossRef CAS PubMed .
32. M. S. Abdelkhalik, A. Vaskin, T. López, A. M. Berghuis, A. Abass and J. Gómez Rivas, Nanophotonics, 2023, 12, 3553–3562 CrossRef CAS PubMed .
33. Y. Kan and S. I. Bozhevolnyi, Adv. Opt. Mater., 2023, 11, 2202759 CrossRef CAS .
34. L. Novotny and N. Van Hulst, Nat. Photonics, 2011, 5, 83–90 CrossRef CAS .
35. R. Heilmann, K. Arjas, T. K. Hakala and N. Päivi Törmä, ACS Photonics, 2023, 10, 3955–3962 CrossRef CAS .
36. M. Wu, S. T. Ha, S. Shendre, E. G. Durmusoglu, W. K. Koh, D. R. Abujetas, J. A. Sánchez-Gil, R. Paniagua-Domínguez, H. V. Demir and A. I. Kuznetsov, Nano Lett., 2020, 20, 6005–6011 CrossRef CAS PubMed .
37. P. Reineck and B. C. Gibson, Adv. Opt. Mater., 2017, 5, 1600446 CrossRef .
38. W. Li and C. F. Kaminski, Chem. Rev., 2022, 122, 2495–12543 Search PubMed .
39. Ansys Lumerical FDTD | Simulation for Photonic Components, https://www.ansys.com/en-in/products/optics/fdtd.
40. G. Lozano, S. R. K. Rodriguez, M. A. Verschuuren and J. Gómez Rivas, Light: Sci. Appl., 2016, 5, e16080 CrossRef CAS PubMed .
41. D. Wang, Q. Wang and Z. Zhan, IEEE Photonics J., 2018, 10, 1–9 Search PubMed .
42. S. Sarkar, S. Poulose, P. K. Sahoo and J. Joseph, OSA Continuum, 2018, 1, 1277 CrossRef CAS .
43. R. G. Hunsperger, Integrated optics: Theory and technology, Springer US, 6th edn, 2009 Search PubMed .
44. M. Hammer, 1-D mode solver for dielectric multilayer slab waveguides, https://www.computational-photonics.eu/oms.html.
45. S. Sarkar, A. K. Ghosh, M. Adnan, O. Aftenieva, V. Gupta, A. Fery, J. Joseph and T. A. F. König, Adv. Opt. Mater., 2022, 10, 2200954 CrossRef CAS .
46. C. Matricardi, J. L. Garcia-Pomar, P. Molet, L. A. Pérez, M. I. Alonso, M. Campoy-Quiles and A. Mihi, Adv. Opt. Mater., 2020, 8, 2000786 CrossRef CAS .
47. G. Lérondel, S. Kostcheev and J. Plain, Springer Ser. Opt. Sci., 2012, 167, 269–316 Search PubMed .
48. W. L. Barnes, S. A. R. Horsley and W. L. Vos, J. Opt., 2020, 22, 073501 CrossRef CAS .
49. M. Khokhar, Priya and R. V. Nair, Phys. Rev. A, 2020, 102, 013502 CrossRef CAS .
50. P. Lodahl, A. F. Van Driel, I. S. Nikolaev, A. Irman, K. Overgaag, D. Vanmaekelbergh and W. L. Vos, Nature, 2004, 430, 654–657 CrossRef CAS PubMed .
51. W. L. Vos, A. F. Koenderink and I. S. Nikolaev, Phys. Rev. A, 2009, 80, 053802 CrossRef .
52. D. Englund, D. Fattal, E. Waks, G. Solomon, B. Zhang, T. Nakaoka, Y. Arakawa, Y. Yamamoto and J. Vučkovićc, Phys. Rev. Lett., 2005, 95, 013904 CrossRef PubMed .
53. V. Rutckaia, F. Heyroth, A. Novikov, M. Shaleev, M. Petrov and J. Schilling, Nano Lett., 2017, 17, 6886–6892 CrossRef CAS PubMed .
54. J. Ren, S. Franke, B. Vandrunen and S. Hughes, Phys. Rev. A, 2024, 109, 013513 CrossRef CAS .
55. P. T. Kristensen, C. Van Vlack and S. Hughes, Opt. Lett., 2012, 37, 1649 CrossRef CAS PubMed .
56. J. T. Y. Tse, S. Murai and K. Tanaka, Phys. Rev. A, 2025, 111, 013511 CrossRef CAS .
57. Y. Xu, J. S. Vučković, R. K. Lee, O. J. Painter, A. Scherer and A. Yariv, J. Opt. Soc. Am. B, 1999, 16, 465 CrossRef CAS .
58. P. A. Thomas, W. J. Tan, H. A. Fernandez and W. L. Barnes, Nano Lett., 2020, 20, 6412–6419 CrossRef CAS PubMed .
59. D. Felbacq, Superlattices Microstruct., 2015, 78, 79–87 CrossRef CAS .
60. K. Gietka, Phys. Rev. A, 2024, 110, 063703 CrossRef CAS .
61. A. V. Kuhlmann, J. H. Prechtel, J. Houel, A. Ludwig, D. Reuter, A. D. Wieck and R. J. Warburton, Nat. Commun., 2015, 6, 8204 CrossRef PubMed .
62. H. Nicolai, Dissertation, LMU München, 2013 .
63. M. A. Lieb, J. M. Zavislan and L. Novotny, J. Opt. Soc. Am. B, 2004, 21, 1210 CrossRef CAS .
64. D. Jameson and J. Croney, Comb. Chem. High Throughput Screening, 2012, 6, 167–176 CrossRef PubMed .
65. K. Zhanghao, L. Chen, X. S. Yang, M. Y. Wang, Z. L. Jing, H. Bin Han, M. Q. Zhang, D. Jin, J. T. Gao and P. Xi, Light: Sci. Appl., 2016, 5, e16166 CrossRef CAS PubMed .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5nr02160j

This journal is © The Royal Society of Chemistry 2025