Exciton-polaritons, hybrid light-matter quasi-particles, have emerged as a fascinating avenue for exploring and tuning semiconductor properties. These quasi-particles arise in the strong coupling regime, where photons and excitons hybridize, analogous to the formation of molecular orbitals in chemical bonding. This hybridization grants exciton-polaritons unique characteristics, such as the ability to delocalize across many chromophores, opening new pathways for manipulating optical and electronic behaviors in materials. By transitioning into the strong coupling regime, exciton-polaritons offer a powerful platform for advancing semiconductor technologies beyond the constraints of traditional light-matter interactions.
Much as in the orbital bonding case, polaritons inherit properties from both their parent states, allowing them to be polarizable, sensitive to electric and magnetic fields, and display ultralow effective masses on the order of 10 -4 times that of an electron. 1,2 In particular, the ultralow effective masses allow for room-temperature Bose-Einstein condensates, which can spontaneously emit coherent, monochromatic light, much like a laser. Unlike lasers, BECs do not require population inversion, allowing for ultralow thresholds. More-over, polariton-polariton interactions introduce strong optical nonlinearities that enable tuning of polarization states, energies, and spatial arrangements, while the quantum states of the polaritons are readable via standard spectroscopic techniques as the emitted light is part of the polariton wave function. The tunability, efficiency, and ease of reading quantum states make polariton systems uniquely suited for advanced computing in quantum, neuromorphic and classical formats, low energy lasing, as well as fundamental study of optical fluids. [1][2][3][4] In order to reach this strong coupling state, the constituent photon and exciton need to exchange energy faster than either decoheres. [5][6][7] In practice, this means designing systems with long photon lifetimes, large exciton binding energies to stabilize the polaritonic states, and high oscillator strengths that enhance the interaction rate. 4,6 Photon lifetimes are easily extended by embedding systems into Fabry-Perot (FP) cavities, with the added benefits of selecting for specific energy levels and generating an angle-dependent dispersion. 5,6,8 Material selection has proven challenging for excitonpolaritons. Inorganic semiconductors offer large oscillator strengths (up to ∼2.5) [9][10][11][12] but most suffer from low binding energies (5-25 meV), meaning that the excitons, and thus polaritons, are only stable at cryogenic temperatures. 4,13 Room-temperature operation has been achieved with the use of high binding energy (25-60 meV) semiconductors such as GaN, ZnO, and certain perovskites, and with quantum confined systems such as quantum wells or transition metal dichalcogenides, but these still require intensive epitaxial deposition or exfoliation techniques, for both the materials themselves and high Q-factor distributed Bragg reflector (DBR) mirrors, making practical application limited. 4 Organic semiconductors have gained recent favor for their large exciton binding energies (200-1200 meV), 14 and solution processability, allowing for room-temperature operation and easy fabrication. 4,15 While oscillator strengths are typically low (<0.25), 16,17 some molecules can reach rather high values (up to ∼3.00), 18 and the ability to pack many chromophores into a cavity has allowed systems to reach enormous collective Rabi splittings, surpassing 1 eV; [19][20][21] Large Rabi splitting signifies enhanced coupling and is desirable for many reasons such as raising the critical temperature for condensation and tunability of energy states. 4 However, many of these systems rely on excitonic features with large line widths, as great as 820 meV, 22 which can limit the quality of polaritonic states for applications. Polaritonic states display a line width that is the weighted average of the homogeneous broadening from the excitonic and photonic state they are formed from, thus homogeneous broadening in the excitonic state directly broadens the polaritonic state, [23][24][25] which corresponds to a shorter polariton lifetime, and is detrimental to factors such as the lasing threshold. 26 Since inhomogeneous broadening does not impact the line width of the polaritonic states, it has been largely disregarded, but recent studies have shown that inhomogeneous broadening severely limits coherent and/or delocalized polariton formation. 27,28 Inhomogeneous broadening also can result in spectral splitting that closely resembles strong coupling or ultrastrong coupling but whose optical origins are not from polariton formation. 29,30 Beyond these issues of coherence, organic materials exhibit much weaker Coulombic interactions, limiting the optical nonlinearities responsible for many of the most useful condensate properties; 1,4,31 device stability is a major issue, as deposition of the top mirror tends to degrade organic materials; 32,33 electrical injection is a challenge; 1,15 and organic semiconductors struggle to reach into the technologically important UV range, 34 which has been gaining interest for applications ranging from medicine to sensing and communications. [35][36][37] Some have hailed the promise of colloidal nanoplatelets for exciton-polaritonics, due to their strong exciton binding energies (195-315 meV), 38 large oscillator strengths, and precise thickness control that leads to sharp line widths. [39][40][41][42][43] However, recent studies have shown that the large oscillator strengths associated with colloidal nanoplatelets arise from a convolution of multiple localized excitons within a single platelet, with each exciton exhibiting far lower oscillator strengths (∼2.6), leaving them only slightly better than epitaxially grown quantum wells. [44][45][46] Indeed, this downside is reflected in Rabi splittings of less than 100 meV for high Qfactor optical cavities. [39][40][41][42] This level of Rabi splitting is sufficient to demonstrate polariton formation, but it is significantly lower than what has been reported with organic materials [19][20][21] and even some perovskite structures. [47][48][49][50] These low Rabi splitting values only allow a limited range to tune the strong coupling parameters. Much as with organics, nanoplatelets also struggle to reach into the UV range, limiting their versatility. 38 An alternative category has largely been overlooked in 0D nanocrystals. These nanocrystals display strong binding energies comparable to organic systems (60-1000 meV), 51,52 with the benefit of exceptionally large oscillator strengths (4-15), 53,54 outpacing any of the other categories discussed so far. As such they have been under study by the field for strong coupling for several decades, 15,55 but have chiefly been relegated to single-particle studies, and, in order to reduce the mode volume associated with those particles, more recently to plasmonic cavities. 55,56 These approaches to study single particles may be driven by the size dispersion (polydispersity) in nanocrystals, since even the best synthesis produces a spread of excitonic states that could, as in organic systems, dilute the impact of strong coupling. To address this concern, we have chosen a system of ∼1.5-nm CdS magic-size clusters (MSCs). 58 MSCs are atomically identical nanocrystals formed due to local energy minima during growth, solving the dispersity issues of colloidal quantum dots. 57 Similarly sized CdSe systems display exciton binding energies of 560-730 meV, 51,52 and oscillator strengths of 3.5-4.0. 53,54 As such, we expect similarly large values for this CdS system, situating them as an ideal system for room-temperature strong coupling.
In this work we demonstrate the incorporation of CdS MSCs into mirrored cavities, resulting in strong coupling with Rabi splitting as large as 390 meV. The ratio of this splitting to the exciton absorption line width is comparable to high performing polaritonic systems with confirmed coherence. 48,50 The processability of these systems is also a benefit, boasting an ease of solution processing and simple fabrication of the metallic FP structure. Additionally, we show our MSC cavities can be tuned to support emissive polariton states spanning a 570 meV range, from 3.07 eV (403 nm) to 3.64 eV (340 nm), while maintaining large coupling strengths. This large emissive range compensates for the primary challenge with single-sized MSCs, which is the large spectral gap between particle sizes. These clusters suggest the potential for a variety of new highperformance tunable polariton systems with the use of other MSC systems.
2.1. Overview of the Coupling System. CdS magic-size clusters (MSCs) were synthesized at high concentration following procedures previously reported, to yield monodisperse clusters ∼1.5 nm in diameter, stabilized by a fibrous organic mesophase, 58 with a sharp excitonic absorption feature at 3.83 eV (Figure 1A, in solution). The MSCs have been previously imaged using high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) 58 and structurally evaluated through total scattering and pair distribution function (PDF) analysis with reverse Monte Carlo modeling. From the PDF modeling we found a cluster form that fits the spectra with a formula of Cd 37 S 20 . 59 The
HAADF-STEM images give insight into the discrete nature of the clusters but ultimately are unable to resolve them individually, likely due to issues with excess organics. The narrow line width of these clusters positions them as excellent candidates for forming exciton-polaritons, a hybrid light-matter state arising from the strong coupling of excitons with a common radiation field.
To form exciton-polaritons, a radiation field must exchange energy with excitonic states faster than either excitation can dissipate. 5,7 To meet this condition, we incorporated thin films of MSCs into Fabry-Perot (FP) microcavities with spacings designed to extend the interaction time of photon frequencies resonant with the MSC film while rejecting other frequencies (Figure 1B,C). These enhanced photon frequencies, known as cavity modes, are determined by the cavity length, L c , as 5,60
where m is an integer and λ cav is a resonant wavelength within the cavity, which differs from the respective vacuum wavelength by the effective refractive index of the materials within the cavity, n eff (resonance for m = 2 depicted by the standing waves in Figure 1B,C).
These cavities were constructed using a 100 nm thick, thermally evaporated Al bottom mirror and a 20 nm evaporated Al top mirror. The latter thickness was chosen to afford sufficient optical confinement for strong coupling but sufficient transmission for our optical experiments. The active layer between the mirrors was deposited by spin-coating solvated CdS MSCs to yield a smooth film (Figure S1A). As needed, LiF was also deposited by thermal evaporation as a spacer between the CdS active layer and the mirrors to control cavity thickness L c , independent of the active layer thickness (Figure 1B). L c was varied (∼140-210 nm) to tune the energy of the cavity mode relative to the MSC exciton energy. Critically, the desired thickness of the MSC layer is less than 200 nm, on the order of the organic mesophase which assembles the clusters into fibers (>170 nm in diameter). 58 In order to accurately control the thickness in the desired range, the solutions are sonicated for 2 hours directly before spin coating to break up this mesophase.
The cavity modes of an FP cavity exhibit energetic dispersions with respect to the angle of incidence θ of the incident light, or more specifically, of θ′, the angle of incidence adjusted by Snell's law for the change in refractive index upon entering the cavity. The wavevector of light within the cavity may be broken into two orthogonal vectors, one perpendicular to the cavity plane with magnitude k ⊥ , and one parallel to the cavity plane with magnitude k ∥ (Figure 1C). Only the magnitude of k ⊥ is confined by L c , according to 5,60
With this constraint, for a given cavity thickness, the in-plane magnitude of the wavevector k ∥ of the resonant photon mode depends on θ′:
Increasing the angle of incidence from normal thus increases the magnitude of the cavity mode wavevector k and accordingly, for a fixed cavity thickness, increases the cavity photon energy E C :
where ℏ is the reduced plank's constant, ω ph is the frequency of the cavity mode, c eff is the speed of light within the cavity, and c is the speed of light in a vacuum (Figure 1D, blue dashed line).
The exciton energy of the active material, by contrast, does not depend on the angle of incidence and remains constant (Figure 1D, black dashed line). When the energy of the photonic mode approaches resonance with the energy of the excitonic transition and strong coupling is achieved, the two energy levels split into an upper polariton (UP) and lower polariton (LP), as shown in Figure 1D. The polaritons inherit an energetic dispersion from the cavity photon, and the splitting of the strong light-matter interaction results in characteristic anticrossing behavior between the polaritons. 6 At the point of exciton-photon resonance, the UP and LP achieve a minimum energetic separation that is defined as the Rabi splitting ℏΩ R (Figure 1D inset).
Experimentally, this dispersion relation and anticrossing behavior can be measured with angle dependent reflectivity. In MSC cavities (representative measurement shown in Figure 1E) we observe a pair of dispersive modes that appear at energies distinct from the bare MSC transition at 3.83 eV and exhibit an anticrossing at this energy. This behavior is distinct from the cavity mode of a cavity without active material. A side-by-side comparison of a cavity with and without active material is shown in Figure S2. The minimum splitting observed for this sample is 308 meV at 58°. As the defining factor for strong coupling is energy exchange between states faster than either state may decohere, and the exchange rate is the Rabi frequency, one may test for the potential of strong coupling by comparing the spectral splitting to the line widths of the photonic mode and the exciton absorption, which are related to their respective decay rates. In our system the splitting exceeds half of the combined full width at halfmaximum of the cavity mode (178 meV) and the exciton absorption of a film (141 meV), suggesting that we have reached the strong coupling regime, that the observed modes correspond to UP and LP bands, and that the energy spacing can be identified as ℏΩ R . 5,[61][62][63] To confirm that we have achieved strong coupling and formed polaritons, we investigated the collective behavior characteristic of the strong light-matter coupling regime. Polaritons are often delocalized 5,27,64,65 across the entire mode volume of the cavity as they are generated from the simultaneous interaction of a single cavity mode with many chromophores in its mode volume. As such, the Rabi splitting depends on the number of emitters, N, within the mode volume, V, and it scales with the square root of the concentration of resonant excitons, C: 61,66-68
This relationship provides an experimental tool to test the presence of strong coupling. A series of cavities with varying chromophore concentrations should display a linear relation between the magnitude of ℏΩ R and the square root of the chromophore concentration, C . To systematically explore the relationship between ℏΩ R and emitter concentration, we designed and fabricated a series of 30 cavities with varied MSC layer thickness accompanied by LiF filler layers to maintain the overall mirrored-cavity volume (Figure 2A). While the concentration of the MSC material in the spin-coated active layer remains constant in each sample, the introduction of LiF spacers reduces the effective concentration of the MSC layer within the cavity. This approach allows us to define the filling fraction, F, as a proxy for the concentration of chromophores within the cavity as
where t is the thickness of the MSC active layer. Angle-dependent reflectivity of these cavities reveals dispersion relations with clear anticrossings, characteristic of strong coupling, yielding a reduction in ℏΩ R as the filling fraction is reduced (four representative plots shown in Figure 2B). Plotting the ℏΩ R against the square root of the filling fraction = F t L / c reveals a strong linear relation (Figure 2C). This linearity is a hallmark of the strong coupling regime, 61,62,[67][68][69][70] where the coupling strength scales predictably with the concentration of the emitters, providing direct evidence of polariton formation. Our theoretical simulations using the transfer matrix (TM) method exhibit a similar trend to our experimental results (the modeling results are detailed in Supporting Information Section S3).
Achieving robust strong coupling provides the opportunity to tune the properties of MSC films nonsynthetically. One distinct drawback for the use of MSCs in applications is that, unlike larger quantum dots, there are only a limited number of sizes that have been synthesized, creating spectral gaps between the accessible states. Hybrid light-matter states offer an ability to overcome this challenge by enabling tunability of both the absorption edge and emission energies. This tunability is achieved through manipulation of the polariton band positions by varying either the concentration of chromophores within the cavity (i.e., tuning ℏΩ R ) or thickness of the cavity (i.e., controlling E C ). Tuning of ℏΩ R via excitonic absorber concentration, as described in eq 5 and demonstrated in Figure 2, disallows independent control of the coupling strength and LP energy, and is limited to the range where ℏΩ R exceeds the strong-coupling threshold. On the other hand, the resonance condition for cavity photons may be adjusted through L c , following eq 4. Shifting the cavity photon energy then shifts the polariton states without changing the concentration of the active material, and thus maintaining the light-matter coupling strength.
To explore tunability via the cavity mode we fabricated a series of cavities with L c ranging from ∼140 to 210 nm (Figure 3A) and measured their corresponding reflectivity dispersions (Figure 3B). We also measured the respective photoluminescence (PL) spectrum of each cavity (Figure 3B, overlaid plots) and the PL of a film not confined within a cavity (Figure 3B, green-filled plot at right). The LP band clearly shifts to higher energies as the cavity thickness is decreased. Across the range of structures fabricated, the peak of the LP band at normal incidence-extrapolated from our angular dispersions-shifts from 3.07 eV (403 nm) to 3.64 eV (340 nm), a change of 570 meV. The PL spectra follow the trend of the LP band, demonstrating that the LP is in all cases the primary emissive state of the coupled system. By comparison, the PL emission from a bare MSC film peaks at 3.73 eV (332 nm). This result demonstrates that the PL of the cavities never reaches the native MSC PL energies due to the energetic shift imposed by hybridization. We further analyzed whether this spectral difference can be accounted for by the cavity effectively acting as a transmission filter, preferentially enhancing the low-energy components of the broad PL tail. In Figure S7 we convolve the cavity transmission spectrum with the bare film PL, finding that in such a regime we would expect two distinct emission peaks: a broad feature near the measured LP band, and a sharp peak where the exciton emission transmits through the UP. Our results in Figure 3 clearly deviate from this scenario, which suggests that the cavity not only selects spectral components but also provides a new primary radiative decay pathway. Our measurements in Figures 3A,B and S7 verify that strong coupling provides a straightforward means to red-shift the emission energy of MSCs via the tunable polariton bandgap.
A crucial question then arises, whether the presence of the new radiative decay channel via these collective polariton states significantly alters the photophysical decay rates in MSC films. We probe this behavior with transient absorption (TA) spectroscopy, a pump-probe technique that reports on the change of optical properties in a material due to the presence of photoexcited states, with subpicosecond precision. In an MSC film excited at 3.81 eV, the TA signal is dominated by a broad photoinduced absorption (PIA) with peaks at 2.93 and 3.18 eV (Figure S8A). These features rapidly decay to yield a weak, long-lived PIA that spans the full detection range. This behavior is consistent with previous reports of other MSC compositions 71,72 and we analogously assign the signatures to an initial singlet exciton which decays into long-lived trap states. Integrating over the PIA band (3.13-3.35 eV), we obtain the biexponential decay kinetics presented in Figure 3C (green). These decay components reflect the interplay between intraparticle singlet excitons and long-lived surface traps, a common result in such colloidal systems. 71,72 Even after the radiative decay of the singlet exciton, a population of states remains throughout our detection range (Figure S8A). These states last for several nanoseconds, aligning well with the dynamics of the broad, red-shifted PL reported in CdS MSCs in solution. 58 We performed equivalent measurements on multiple cavities supporting MSC polaritons, again with the pump tuned to the MSC exciton energy of 3.81 eV. As shown in Figure S8B, we observe a striking change in spectral shape in the cavity. In the cavity spectra, distinct positive features emerge around the LP energy, which eventually decay and transform into a broad PIA, similar to the bare film. The precise origin of such spectral features is subject to considerable debate, with potential contributions including the spectral fingerprints of photoexcited polaritons, depopulation of the ground state due to photoexcitation, thermal expansion of the active layer, and hotelectron effects in the cavity mirrors. 63,[73][74][75] Disentangling these contributions is beyond the scope of this report, but crucially the dynamics of these effects-beyond the few-ps electron thermalization time scale-should follow the population dynamics of the MSC film within the cavity. Indeed, we obtain a close overlap between the reference film and cavity decay dynamics (Figure 3C). This observation suggests that the formation of exciton-polaritons does not strongly perturb the photophysics of MSC films. Instead, it simply provides a new, tunable LP state through which the MSC radiative decay is funneled. Accordingly, we also observe a strong correlation between the decay profiles obtained from TA and from timeresolved PL (Figure S9), where the MSC film and cavity exhibit similar PL decay.
Our results positively confirm polaritons in a tunable UV system and further demonstrate a new material class, which addresses many issues present in other systems. As discussed in the introduction, the line width of the excitonic feature is critically important to the quality of the strong coupling in the system under study. Homogenous broadening can lead to broader polaritons states, [23][24][25] while inhomogeneous broadening can limit the formation of coherent and/or delocalized states, 27,28 or inflate the spectral splitting from nonstrong coupling origins, leading to the impression that strong or ultrastrong coupling has been achieved when it has not. 32,33 In order to contextualize our system in the broader field, we have plotted a comparison of the Rabi splitting normalized by the line width of the excitonic feature for a series of reported systems (Figure 4). This normalization suggests that the large splittings reported for many organic systems arise in large part from their large line widths, and, when considering this factor, their coupling quality is not as generally dominant as has been assumed. Considering their other struggles with device stability, 32,33 electrical injection, 15 and weak nonlinearities, 1,4,31 organic systems no longer appear generally advantageous over other material categories. Inorganic systems solve many of the concerns with organics and, in the context of Rabi splitting normalized by excitonic line width, perform well, but their intensive processing limits their use for practical application. 4 Nanoplatelets have gained interest for combining the benefits of both sides, [39][40][41][42][43] however recent studies have debunked the presence of exceptionally strong oscillator strengths, [44][45][46] and this is reflected in their relatively low Rabi splittings, even when normalized by the narrow line widths of their excitonic features. Magic-size clusters, however, combine the same benefits of organic and inorganic systems, with generally larger exciton binding energies and oscillator strengths than nanoplatelets, and similarly narrow line widths. These advantages are reflected in Figure 4, which shows CdS MSCs have larger line width-normalized Rabi splitting than other UV systems and are even comparable with high performing systems in the visible range, suggesting that MSCs may be a uniquely valuable platform for polaritonics, combining high quality coupling with a range of available energies from deep UV 76 up to 2.19 eV (565 nm) 77 with the ease of solution processing and simple metallic FP cavities.
In conclusion, we have realized strong coupling in CdS MSCs, confirmed with a square root dependence of Rabi splitting on chromophore concentration. The polariton states are capable of spanning most of the near-UV range through cavity detuning, representing an opportunity for future applications in UV photonics. With a narrow line width of 141 meV and a large Rabi splitting of 390 meV, we expect this system to overcome inhomogeneous broadening to achieve the threshold is not available), as a function of the excitonic energy for a series of organic, [19][20][21][22][80][81][82][83][84][85][86][87] hybrid organic-inorganic perovskites (HOIP), [47][48][49] inorganic, 50,[88][89][90][91] and nanoplatelet [39][40][41][42] systems with this CdS MSC system. The systems presented have been restricted to exciton-polariton systems with collective coupling in the UV and visible range, with exciton energies down to 1.9 eV at room temperature in optical cavities to allow for direct comparison.
for polariton delocalization, which can significantly enhance long-range energy transport. 27 CdS MSCs combine several advantages from other polariton-forming systems beyond extending into the UV spectrum, such as room-temperature solution processing, stable device integration, large oscillator strengths, and exciton binding energies comparable to organic systems. To highlight the versatility of MSCs, we prepared further cavities with the particles dispersed in a PMMA matrix (Figure S6), a common structure for organic excitonpolaritons. Even in this dilute regime, MSCs maintain strong coupling due to their narrow line width and strong absorption, offering a highly adaptable approach to device engineering. In applications like OLEDs, embedding emitters in a semiconductor host matrix allows for fine-tuning of both optical and physical properties. 78 Similarly, in solar cells, efficient charge transport to the interface is essential for charge generation, and having this adaptability supports flexible integration of MSCs in future device design. 79 The robust properties of MSCs�including a large bandgap with well-defined excitonic transitions above 3.83 eV, ease of room-temperature processing, and a large oscillator strength� allows them to overcome several limitations inherent in other systems. Moreover, the postsynthetic tunability of their optical bandgap by changing the conditions for strong light-matter coupling, without altering their fundamental photophysics, offers an exceptional degree of flexibility in device design. This ability to freely tune the absorption and emission energies, combined with the stability and scalability of room-temperature processing, positions MSC polariton systems as a highly versatile and promising category for future advancements in optoelectronics and polariton-based devices.
CdS MSCs were colloidally synthesized under high concentration following procedures previously reported. 58 Once the MSCs have been washed and dried overnight, a solution may be prepared for spin coating. The MSCs and solvent are added to a vial under nitrogen with the solvent of choice, usually heptane, toluene, or chloroform, to the appropriate concentration, commonly 10 mg/mL. The vial is then allowed to stir for a minimum of 3 days in a nitrogen glovebox. Once stirred, the solution is sonicated to completion (fibrous mesophase can be broken down no further), taking approximately 2 h. After sonication the solution must immediately be used for spin coating, else the solution will gel in short order. Spin coating cycles vary from 5 to 15 s, and 800-4500 rpm depending on the solvent, concentration, and thickness desired.
FP cavities with an MSC film as the active layer were prepared through standard layer-by-layer processing techniques. First, 100 nm of Al was deposited on quartz-coated glass substrates by thermal evaporation, acting as the bottom mirror, followed by a layer of LiF of variable thickness to serve as an inert spacer. CdS MSCs were then spin-coated on to the substrate using the above procedures. A top layer of LiF matching the bottom layer and a 20 nm, semitransparent layer of Al were then deposited directly onto the active layer to complete the cavity. All measurements were performed from the semitransparent 20 nm mirror side. The total thickness between the Al mirrors was typically 200 nm to support a λ cavity mode, that is a standing wave of one wavelength, resonant with the main absorption band. The relative thicknesses of the CdS active layer and LiF were varied to explore the concentration dependence expected of the strong coupling regime. Structures with thinner or thicker mirror spacings were prepared to demonstrate the impact of detuning the cavity mode.
4.2. Characterization. The UV-vis absorption of films was measured on a Jasco-1500 Spectrophotometer, and of solutions on an Ocean Optics USB2000+ spectrophotometer with an Ocean Optics DH-2000-BAL light source. To characterize the angular dispersion in microcavities, we performed reflectance measurements using a Xe plasma white-light source (LDLS, Hamamatsu) and Avantes AvaSpec-Mini4096 detector incorporated into a home-built, fiber-coupled goniometer with two arms rotating between angles of incidence of 9°a nd 70°, where 0°is normal incidence.
Steady-state photoluminescence (PL) measurements on the bare film were taken using Edinburgh Instruments FLS1000 spectrometer with a xenon arc lamp as the excitation source and photon counting PMT detector. Steady-state PL measurements on cavities were performed using a Pharos-10 W femtosecond pulsed laser (1030 nm, 180 fs, Light Conversion) as the excitation source. The output was directed into an Orpheus optical parametric amplifier for conversion to 325 nm used in the experiments. PL from the cavity was detected using a Princeton Instruments Spectrapro HRS300 spectrometer and PICOMAX4 ICCD camera. Emission was collected using a large lens spanning an angular range of -10 to 10 degrees. Time-resolved PL measurements on both films and cavities were measured by timecorrelated single-photon counting with the Edinburgh FLS1000 lifetime spectrometer using a 280 nm pulsed LED as the excitation source and a monochromator to select the detection wavelengths.
Transient absorption (TA) measurements were performed using a commercially available Ultrafast Systems HELIOS transient absorption spectrometer, driven by the Pharos-10 W system above and operating at 8 kHz. Excitation pulses were generated using the Orpheus optical parametric amplifier. The 325 nm pump pulse had a duration of less than 200 fs. For the UV probe, a frequency-doubled 515 nm pulse was employed for white-light generation. All transient absorption measurements were done in reflection geometry. Schematics of the key instruments are shown in Figure S12.
All data will be provided by the corresponding author upon reasonable request.
The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsnano.4c17355.
Additional details on experimental setups, surface characterization of MSC films, transfer matrix calculations for optical cavities, representative reflectivity spectra of empty and filled cavities used in concentration-dependent studies, further analysis of PL and TA spectroscopy in films and cavities, and detailed methodology for estimating chromophore concentration in MSC films (PDF)
https://doi.org/10.1021/acsnano.4c17355 ACS Nano 2025, 19, 16438-16447
ACS Nano