Nanophosphor-based Photonics

Nanophosphor-based Photonics Doctoral thesis presented by Elena Cabello Olmo Supervised by Prof. Dr. Hern´an R. M´ıguez Garc´ıa Dr. Gabriel S. Lozano Barbero Advised by Dr. Felipe Guti´errez Mora Multifunctional Optical Materials Group Institute of Materials Science of Seville Consejo Superior de Investigaciones Cient´ıficas - Universidad de Sevilla The work described in this thesis was financially supported by the grant FPU19/00346 funded by MICIU/AEI/10.13039/501100011033 and ESF Investing in your Future. This work has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Programme (NANOPHOM, grant agreement no.715832). This work has also received funding from the Spanish Ministry of Science and Innovation under grant PID2020-116593RB-I00, funded by MCIN/AEI/10.13039/ 501100011033. The cover image is a composition of a digital camera image of a self-standing photonic QR code made of nanophosphors with a scanning electron microscope image of a cross-section of a nanoimprinted surface of a QR code that gives rise to the iridescent color. Sus plumas de cobre, brillo en la tarde, se funden al sol que lento las arde, viajera eterna, libre al vuelo, la abubilla es viento y anhelo. iv | Contents Abbreviations vii Abstract ix Resumen xi 1 Introduction 1 1.1 Phosphors beyond lighting: Nanophosphors . . . . . . . . . . . . . 1 1.2 The role of photonics in light emitting materials . . . . . . . . . . . 7 1.3 Transparent nanophosphor films . . . . . . . . . . . . . . . . . . . 17 1.4 Motivation and objectives . . . . . . . . . . . . . . . . . . . . . . . 20 1.5 ThisThesis .............................. 21 2 Processing of nanophosphor thin films compatible with photonic integration 23 2.1 Downshifting vanadate nanophosphors . . . . . . . . . . . . . . . . 24 2.2 Rapid thermal processing . . . . . . . . . . . . . . . . . . . . . . . 26 2.3 Soft-lithographed nanophosphors . . . . . . . . . . . . . . . . . . . 31 2.4 Inkjet-printed nanophosphors . . . . . . . . . . . . . . . . . . . . . 35 2.5 Upconverting oxyfluoride nanophosphors . . . . . . . . . . . . . . . 38 2.6 Conclusions .............................. 45 3 Nanophosphor-based photonic architectures for tuned light emission 47 3.1 Periodic Surface Reliefs . . . . . . . . . . . . . . . . . . . . . . . . 48 3.1.1 Monolithic squared patterned nanophosphors . . . . . . . . 48 3.1.2 Inkjet-printed photonic nanophosphors . . . . . . . . . . . . 53 3.2 Metallic periodic arrays . . . . . . . . . . . . . . . . . . . . . . . . 60 3.3 Conclusions .............................. 65 v 4 Methods 67 4.1 Nanophosphor synthesis and processing . . . . . . . . . . . . . . . 67 4.1.1 GdVO4:Eu3+ nanophosphor synthesis . . . . . . . . . . . . 67 4.1.2 Oxyfluoride nanophosphor synthesis . . . . . . . . . . . . . 68 4.1.3 Nanophosphor based paste preparation . . . . . . . . . . . 68 4.1.4 Flexible upconverting nanophosphor coating . . . . . . . . . 69 4.2 Optical Characterization . . . . . . . . . . . . . . . . . . . . . . . 70 4.2.1 UV-Vis Spectroscopy . . . . . . . . . . . . . . . . . . . . . 70 4.2.2 Photoluminescence Spectrometry . . . . . . . . . . . . . . 71 4.2.3 Angular photoluminescence with excitation control . . . . . 75 4.2.4 Back Focal Plane Imaging . . . . . . . . . . . . . . . . . . 76 4.3 Calculations.............................. 77 4.3.1 Transfer Matrix Method . . . . . . . . . . . . . . . . . . . 77 4.3.2 FDTD............................. 81 General conclusions 83 List of publications 85 Bibliography 87 Agradecimientos 107 vi | Abbreviations AAbsorptance CRI Color rendering index ETL Electron transport layer FDTD Finite-difference time-domain IIE Integrated intensity enhancement LED Light emitting diode LDOS Local density of states LSPR Localized surface plasmon resonance MCS Multi-channel scaling NIR Near-infrared NP Nanoparticle OLED Organic light emitting diode PDF Powder diffraction file PL Photoluminescence PLQY Photoluminescence quantum yield PMMA Poly(methyl methacrylate) PS Polystyrene RA Rayleigh anomaly RE Rare earth RReflectance RTA Rapid thermal annealing SEM Scanning electron microscopy SLR Surface lattice resonance SPP Surface plasmon polariton SSL Solid-state lighting TTransmittance T0Zeroth-order transmittance TEM Transmission electron microscopy UC Upconversion UCNP Upconverting nanoparticle UCPL Upconversion photoluminescence XRD X-ray diffraction vii Chapter 1 Figure 1.1:1901 Antique advertisement of the lighting system with the mantles developed by Carl Auer von Welsbach, published by Illustrated London News. Photograph of the original reproduced with permission. The current fields of industrial application of RE cover electronics (batteries, magnets, etc), catalysis, energy (wind turbines, lighting, etc), polishing and glass industry, lasers and optical fibers or medical applications among others.[4,5,6,7,8,9,10] Looking back to the first industrial utilization of these materials, it was related to the generation of artificial light.[11]Figure 1.1 shows a 1901 antique advertisement of the Welsbach light that consisted of an incandescent mantle that covered a gas lighter. The intense white light came from the incandescent mantle made by an oxide network. The combination of thorium and cerium oxides resulted in bright white emission.[12,13]The physics underlying this phenomenon was different and quite inefficient compared to the light emitting diodes (LEDs) bulbs that are used today with cerium-doped coatings atop electroluminescent displays.[14]However, the advertisement is revealing as it shows two current topics. On the one hand, the fallacy of ’the more brilliant, the better’, and on the other, the impact of light on our surroundings, illustrated by the comical figure of the owl that does not know if it is day or night.[15] The composition of the phosphors, that is, the RE elements they contain, determine their main luminescent properties. Depending on the type of transition we distinguish two groups of phosphors. In most cases (Eu3+, Dy3+, Er3+, Tm3+, etc) the origin of the emission is due to inner f→felectronic transitions with well-defined energies which leads to atomic-like sharp peaks, while some RE cations presents broad 2| Introduction bands with origin in the d→ftransitions, due to the less shielded dlevel that leads to a broadening because of the crystal field splitting effect, for instance, Ce3+ or Eu2+.[16,1]Figure 1.2 illustrates some basics concepts and possibilities. Figure 1.2a shows a sketch of a lighting system based on a blue LED and a color-converting material, i.e. phosphors, that converts part of the incoming light in yellow light; most of the current LED bulbs are made up of multiple chips arranged in this configuration.[14]Figures 1.2b and 1.2c show simplified schemes of emission from exited states for the above-mentioned types of transitions while the normalized photoluminescence (PL) spectra of Eu3+ and Ce3+ are depicted in Figure 1.2d. The shape of these spectra directly affects the application of the emitters. For instance, in devices such as displays that use red, green, and blue (RGB) pixels, a narrow emission with higher color purity is preferred to achieve a wider color gamut. Meanwhile, for general lighting in homes or workplaces, a continuous spectrum similar to sunlight is desirable because it allows for more natural color perception (high color-rending index —CRI—). Focusing on the methods of photoexcitation of these materials, there are several possibilities that arise from a limitation that is the low absorption cross section that RE cations exhibit. The concentration of the RE cation in the crystalline matrix is limited to avoid quenching —a reduction in luminescence efficiency due to interactions between neighboring ions, resulting in non-radiative energy loss—. Then, the increasement of the percentage of cations is not a solution to ensure a Figure 1.2:(a) Sketch of a lighting system based on a blue LED and a color-converting material (phosphors,) that converts part of the incoming light in yellow light. (b) Energy levels of Eu3+ in GdVO4matrix illustrates transition between f-f that give raise to stretch emission peaks. In particular, the transitions take place between 5D0→7FJ(J=1,2,3,4)(c) Energy levels of Ce3+ in YAG matrixillustrates transition between d-f that give raise to broad emission peaks corresponding to the tranistions 5d2(A1g)→2FJ(J=5/2,7/2). (d) Photoluminescence spectra of Ce3+ (yellow) and Eu3+ (pink) doped emitters. (e) Schemes of different path to photoexcite RE cations in a crystalline matrix: (i) direct excitation (ii) energy transfer (ET) from matrix and (iii) use of sensitizers. |3 Chapter 1 higher absorption of the incoming excitation. A straightforward approach may be to increase the amount of material while maintaining a fixed concentration of substituents. For example, in commercial LED-based lamps, the most commonly used color-converting coatings consist of micron-thick cerium-doped yttrium aluminum garnet (YAG:Ce3+), which provides partial absorption of blue excitation, resulting in white light emission. In this case we perform a direct excitation, i.e. the Ce3+ is excited by the 4f→5d transition centered around 455 nm in the blue part of the visible range. However, there are other approaches as shown in Figure 1.2e. One of them is the use of the inorganic host as an absorber that efficiently transfers to the RE cation leads to high absorption of the incident light (sketch ii). Similar processes have been studied with organic complexes, which also transfer energy to the emitting cations, it is sometimes referred to as the antenna effect. Other RE cations (sketch iii), namely sensitizers, are also used when the levels of sensitizer and activator are similar in energy, e.g. Yb3+ is used as a sensitizer together with Er3+, Tm3+ and Ho3+ in upconverting particles in which two or more low-energy photons results in the emission of one higher-energy photon.[12] The use of phosphors is not limited to the field of house-lighting or displays. Their compatibility with LEDs as excitation source, the physicochemical properties, i.e. the conversion efficiency, and thermal and chemical stability, open up new possibilities in fields where light has been typically used and others where light is explored as new tool.[17,18]Some of them are described below while illustrative diagrams of them are shown in Figure 1.3. Horticulture The development of light sources with specific properties for horticulture is of paramount importance. Light sources can be fine-tuned to match the different types of photoreceptors, including harvesting and photosensory pigments, found in plants. Then, the spectral control of light leads to the modulation of the rates and determines the development of germination, flower induction, and other critical biological functions. Moreover, it enables us to overcome the limitations of relying on the climatology or generic indoor lighting approaches, thereby enhancing our ability to achieve optimal conditions for plant growth and yield.[19,20] Light pollution control The sensitivity to specific colors (wavelengths) is not always controlled or deliberated. A clear example is light pollution and its impact that is regarded as particularly problematic for broad spectrum lighting, because it can stimulate responses across a wide range of wavelengths, specially the blue light, which is strongly scattered 4| Introduction in the sky. This phenomenon has several adverse effects on the environment and society. It disrupts the natural behavior of nocturnal wildlife, impacting creatures like bats and birds, affecting their reproduction and migration. Light pollution also obstructs our view of the night sky, hindering astronomical observations. In addition its represents a significant energy waste, costing us not only financially but also in terms of its environmental footprint.[21,15]Thus, addressing light pollution through more efficient lighting and responsible design of the chromatic content, intensity and direction is crucial. Human Centric Lighting Human Centric Lighting refers to the design of the illumination to enhance the wellbeing and productivity of individuals.[22]It is scientifically supported by the so-called non-visuals effects of light, and it is closely related to the field of the chronobiology, which consist of the study of the adaptation of the biological rhythms to solar and lunar-related cycles and how they influence various biological processes. In humans, the effects are related to the photosensitive retinal ganglion cells ipRGCs which are a specialized type of neuron in the eye’s retina. The metrology of these effects to obtain accurate, unbiased, and ethical results remains a challenge today.[23,24] Art: visualization and protection The perception of the color of objects exposed to artificial light depends largely on the chromatic content of the incident light since the color of pigments is produced by selective absorption. An RGB illumination looks different than a continuous lighting generated by a blue LED with a coating that covers a large part of the spectrum, such as YAG:Ce. They have different CRI, which means that colors are perceived differently when viewed under artificial lighting with different chromatic composition. In the case of art galleries, this becomes particularly important for the accurate display of artwork.[25]However, it is not the only concern in this application. Light can degrade pigments, leading to color fading and discoloration, and therefore, it also plays a significant role in the preservation of artworks.[26] Data storage Among the objectives aimed to be achieved in the field of data storage, there are several elements to be overcome while maintaining the compatibility with the existing technology: increasing capacity, achieving fast data access, low energy consumption, etc. There are multiple efforts in the field of optical storage that undoubtedly differ significantly from the old CDs and DVDs, and they are truly promising.For example, optical multiplexing or multilevel with RE elements is under study.[27,28] |5 Chapter 1 Figure 1.3:Illustrative diagrams of applications where having control over emission properties is relevant Anti-counterfeiting & Security We also find that phosphors, due to their unique photoluminescent properties, have a place in the field of generating elements that ensure authenticity and prevent counterfeiting. Luminescent anti-counterfeiting consists of elements with different elements based on a series of increasingly complex security layers that result in a distinctive luminescent response to different types of stimuli such as temperature, time, photoexcitation, or mechanical stress.[29,30] Detectors & Sensing Another application for these materials is in detection and sensing, where the controlled dependence of the optical response on external stimuli can be effectively exploited.[7]Two quite different examples are their use in the detection of latent fingerprints and their use in scintillators.[31,32]However, today, one of the current hot topics is the development of upconverting phosphor-based temperature sensing, which challenges the established paradigm of resistive thermometry. It relies on the temperature dependence of the luminescence of phosphors with various response parameters such as band shape, peak energy, or intensity, as well as excited state lifetimes and risetimes.[33] Bioimaging & Theranostics The use of phosphor nanoparticles (NPs) for imaging and therapy has gained significant attention in the field of medical research and diagnostics due to their unique optical properties, that allow precise and versatile applications in both medical imaging, therapeutics and diagnostics (theranostics). Upconverting nanophosphors stand 6| Introduction out playing different roles: imaging contrast agents, delivery agents, energy transducers.[34] High-Power illumination High-power LEDs and lasers are used with phosphors in various applications due to their efficiency, durability and brightness. They are widely used in automotive lighting, street lighting, and industrial lighting, providing high-intensity illumination with lower energy consumption compared to traditional lighting technologies.[35,36] As mentioned, there are many applications for phosphors, ranging from lighting and displays to medical imaging and sensors. This increase in the number of applications goes hand in hand with higher demands on the materials, requiring them to meet more stringent performance criteria. Some of the requirements are related to miniaturization of optical components, where traditional materials are constrained by their micrometer scale. In addition, the tools and techniques we use to achieve precise control over the properties of these materials are becoming increasingly sophisticated. Advances in synthesis methods, characterization techniques, and computational modeling are all essential to developing phosphors, or more generally, color-converting materials, that can meet the evolving needs of these applications. Photonics, understood as a tool for manipulating light, is discussed in the following section. Later, the requirements that materials must meet to be compatible with these approaches will be addressed. 1.2 The role of photonics in light emitting materials The field of photonics encompasses approaches to generate, control and detect light. Photonics has demonstrated its utility across a broad range of fields, from telecommunications and medical diagnostics and therapy to renewable energy.[37,38,39,40]The widespread application of photonics across diverse fields highlights its versatility and crucial role in advancing modern technology. In terms of managing light emission, advancements in the chemical engineering of novel emitters, combined with developments in nanotechnology, opens novel routes for new emitting devices or materials with enhanced properties and functionalities, as well as a deeper understanding of physical processes.[41,42,43,44] Before discussing particular photonic structures to modify the photoluminescence of phosphors, it is worthwhile to first generically explore the process of spontaneous light emission and its main characteristics to understand what can be modified and |7 Chapter 1 how. If we focus on a photoluminescence process, it involves converting one photon into another, typically of lower energy. Initially, the material absorbs a specific amount of energy from incident photons, which excites electrons to higher energy states. After absorption, the excited electrons relax back to their original states, releasing energy in the form of new photons. Throughout this conversion process, we must pay attention to key subprocesses and magnitudes that will allow us to understand the fundamental characteristics of the emitting materials and their applicability or performance. Thus, in this section we focus in the absorption, the generation of photons and the extraction of light from a material or device, together with magnitudes that are key to understand theses processes and to theoretically and experimentally study them, i.e. absorptance, electric field intensity, local density of states, radiative and total decay rates, photoluminescence quantum yield and light extraction. Absorption process If we assume an isolated atom, the absorption of a photon gives rise to the promotion of an electron to an excited atomic level. The necessary energy is determined by its electronic level structure while the probability is influenced by the dipole transition moment. When we consider the case of a solid material distributed spatially, the geometry of the material and the surrounding materials become critical. To quantify absorption in optical architectures, a useful quantity is the absorptance (A), which can be expressed as A= 1 −R−T, (1.1) typically as a function of wavelength, where Rand Tare the reflectance and transmittance respectively. Thus, Aprovides absolute information about the percentage of photons absorbed for different incident energies and can be estimated from R and T measurements with UV-Vis spectrophotometry. Figure 1.4a–c show three scheme of a film surrounded by air and different types of light-matter interaction mechanisms within the material. If the sample is homogeneous and has a small roughness compared to the wavelength, a part of the incident light will be specularly reflected and another part will be ballistically transmitted (Figure 1.4a). If the sample is additionally absorbing, the sum of R and T will be less than 100% of the incident light (Figure 1.4b). Otherwise, if the sample has refractive index variations on a scale comparable to the wavelength, the multiple scattering phenomena caused by these refractive index inhomogeneities will lead to the observation of a portion of the reflected and incident light diffusely scattered in different directions (Figure 1.4c). For the case of absorbing materials, this can lead to an increase in the number of 8| Introduction Figure 1.4:Schematic of the interaction of incident light with different types of thin films showing the processes of reflection, transmission, absorption, and emission. (a) Nonabsorbing flat sample. (b) Absorbing flat sample where red arrows indicate emission. (c) Sample with roughness where the short arrows show the contribution of diffuse processes found in scattering materials. absorbed photons if the mean free path of the photons is substantially increased, resulting in a higher probability of being absorbed.[45]The extreme case is the use of disordered materials, where it is possible to achieve engineering of such complexity and find systems that allow us to have high control over light propagation.[46] The refractive index is a measure of the speed of light through a material. It is critical in the design of optical materials because it determines how light interacts with materials, affecting the efficiency and performance of optical devices. The Aof the system can be estimated knowing the refractive index and geometrical parameters and obtaining R and T in the system, following Equation (1.1) or, by knowing the distribution of the electric field and obtaining the absorbed power by integrating the electric field in the absorbing layers, that is, those with a refractive index with a non-zero imaginary part, by following the next equation:[47] Amat(λ) = λϵ0Zmat |E(z, λ)|2nmat(λ)kmat(λ)dz. (1.2) Here, Amat is the power absorbed in the material mat,λis the wavelength, ϵ0is the vacuum permittivity, Ethe electric field intensity normalized to the incident, nand kthe real and imaginary part of the refractive index and zthe spatial coordinate. Notice that this example considers a radial symmetry along z-axis (1D structure). The calculations can be carried out analytically following the transfer matrix method based on the values of thicknesses and refractive indices (see Abeles Transfer Matrix Method in Section 4.3). When the structures are not stratified, numerical methods such as the finite-difference time-domain (FDTD) method must be used for studying complex geometries. |9 Chapter 1 Equation (1.2) indicates that the modification of the electric field intensity distribution in the volume where the absorber is placed and, at a wavelength where the material absorbs, leads to a modification of the absorption. This results in different types of approaches that can be focused in the enhancement of the external efficiency of the process by either (i) increasing the number of photons entering the system or (ii) increasing the probability of absorption when photons strike a semi-transparent system. These methods are useful for reducing the use of expensive or toxic materials, modifying the spectral shape of A and enhancing the external efficiency of a photonic device.[48]Some of the straightforward procedures include the use of antireflective coatings, which reduce R at the first interface, promoting total internal reflection or using disorder media, which enhance the absorption inside the materials. The fabrication of resonant photonic structures that intensely concentrate the field, enabled by advances in material processing methods over recent decades, allows for precise control of A. This results in significantly increased absorption and behaviors that can depend on polarization, energy, and/or directionality.[49,50,51] A one-dimensional example is discussed above to illustrate the absorption modification of a emitting material. The resonant structure is shown in Figure 1.5a-e and it consists of a Fabry-Perot cavity made of a Rhodamine 6G (Rh6G) dye dispersed in polysterine (PS) sandwiched between two silver mirrors is analyzed as an example of absorption modification. This system was studied assuming a multilayered Figure 1.5:(a-e) Theoretical example of a metallic cavity with a dye (Rhodamine 6G) diluted in polysterine (PS) that is used to modify the A of the dye.(a) Scheme of the simulated Fabry-Perot cavity. (b) Real (n, dashed curves) and imaginary (k, solid curves) part of the refractive index of Ag (gray) and the effective refractive index of Rhodamine 6G at a concentration of 20% in polysterine (magenta), extracted from Refs.[52,53,54]. (c) Reflectance spectrum showing a cavity mode at 595 nm. (d) Spatial and spectral distribution of the electric field intensity showing the nature of the optical mode, i.e. two 2nd-order mode. (e) The A of the cavity is shown as a black line while and the contribution of each layers is shown as shaded curves. Notice the spectral change of the A in the cavity compared to a bare film outside the cavity atop a glass (magenta curve). 10 | Introduction structured with the thicknesses indicated in the Figure 1.5a (20 nm Ag + 300 nm Rh6G-PS + 150 nm Ag) and using reported refractive indexes for Ag, Rh6G and PS (Figure 1.5b).[52,53,54]The calculated R is shown in Figure 1.5c with one dip that correspond to a 2nd-order resonance whose electric field intensity distribution is depicted in (Figure 1.5d) showing two nodes. The resulting optical response and the contributions of each material are detailed Figure 1.5e. In order to highlight the modification of the A, the calculations for a film of Rh6G diluted in PS with the same thickness on top of a glass is displayed in Figure 1.5e in magenta while the A of the film sandwiched between Ag is black, both calculated from Equation (1.1). In order to understand where the absorption takes place along the structure, the spatial integral for each layer are shown as filled areas to discriminate the absorption in the metallic films compared to the dye. A huge modification of the Rh6G absorption leads to a significant change of the spectral shape. Furthermore, this behavior depends on the angle of incidence of the light. This could be interesting for spectrally amplify the absorption of materials. To summarize, controlling the interaction of the incident light with a material that absorbs and then emits is of interest to improve its performance in terms of light incoupling efficiency, increasing the ratio of total absorbed photons to incident photons in our system. In addition, device efficiency can be improved by not only increasing absorption in our target material, but also minimizing absorption in materials we do not want to absorb, such as parasitic absorption in metals that might act as contacts.[47]. Photon generation The emission of a photon from a material in its excited state is a probabilistic phenomenon that depends on the material’s levels structure, and also on its optical environment. Specifically, Fermi’s golden rule provides a basis for this discussion, expressing the dependence of the radiative rate of a spontaneous decay (i) on the intrinsic properties, determined by the set of energy levels, as well as, (ii) on the local density of states (LDOS), which measures the availability of electromagnetic modes at a given point in space.[55]The expression of the Fermi’s Golden rule for a randomly oriented two-level emitter under the dipole approximation turns:[56,55] Γrad =πω|p|2 3ℏε0n2LDOS(r, ω) (1.3) where pis the dipole moment of the quantum emitter,ωis the angular frequency of the emitted photon, ε0is the dielectric permittivity of vacuum, nis the refractive |11 Chapter 1 and/or sintering of materials. TThis section emphasizes the importance of transparency to excite photonic modes and modify the emission. The preparation of thin films (<1µm) of materials such as phosphors, which meet interesting requirements as light-emitting elements in devices, is not only limited from the perspective of obtaining a transparent, flat, homogeneous material, but their optical properties can also be compromised. A clear example is the discussed issue of light absorption dependence with the optical path. For instance, YAG:Ce3+ is a material with a low absorption cross section due to the electronic structure of Ce3+. This is not a problem to its application in LED bulbs, where plates have a thickness on the order of 0.1–1 mm, as the path is sufficient for the absorption of the blue light used for excitation. However, if we make a mechanical and coarse size reduction, a 1µm film will absorb 2–3 orders of magnitude less unless the optical design focuses on increasing absorption. For example, if material savings are desired, a composition that includes other elements that increase absorption or efficiency, such as the use of sensitizers, may be helpful.[89,87,90]There are studies that start from thick transparent YAG plates to extract a greater amount of light with directional properties.[91,80]. The modification of thin phosphor films has also been studied, in some works using Eu3+ complexes,[92,93]and using photonic elements deposited on top of the transparent films. This approach is useful when the processing of materials in a different way than as a film is complicated, which is generally the case of phosphors due to practical incompatibilities with nanostructure processing techniques. On the other hand, for 1D systems, there are some precedents in the MOM group where NPs can be integrated into multilayered system. Nevertheless, photonic approaches are not restricted to materials that exhibit downshifting. Many current studies focus on phosphors that exhibit upconversion (UC), i.e., due to their intrinsic efficiency limitations, offer a significant margin of improvement through the use of photonics.[94,95]In addition, the MOM group has started to work with persistent phosphors with the goal of not only improving performance but also understanding the physical phenomena underlying this process.[45]When this Ph.D. work started, the MOM group was developing thin films of nanophosphors, with most of the experimental work done by Dr. Dongling Geng et. al..[96]. The liquid phase deposition was then optimized for integration into photonic systems, focusing on 1D photonic architectures as shown in Figure 1.8. In this work, a structure based on a one-dimensional photonic crystal (1DPC) was used, as shown in Figure 1.8a–b. This type of structure is based on the periodic alternation of films of two dielectric materials with a refractive index contrast, resulting in a so-called Bragg peak due to the interference phenomenon. A film of europium-doped vana- 18 | Introduction Figure 1.8:(a) Scheme and (b) scanning electron micrograph of the cross-sectional view of the 1D photonic crystal. (c) Reflectance spectra of a defect-free photonic crystal (blue) and the optical resonator (red) at θ= 6°. Analytical calculations are shown as dashed lines. (d) Scanning electron micrograph of the cross section and (e) scheme of the resonator based on a 1D photonic crystal. (f) Spatial and spectral distribution of the electric field for the designed structure. (g) Normal direction photoluminescence of the nanophosphors integrated into an optical resonator (black curve) and the reference (gray, solid curve). Reflectance from (c) is plotted again in red. (h–i) Angular emission of the nanophosphors integrated into the optical resonator and the reference respectively. dates was introduced as a defect in the middle of the system, creating a resonator (Figure 1.8d–e). The optical design of the system allowed for the spectral and spatial localization of the resonance at the europium emission wavelength and in the region where the NPs were located (Figure 1.8f). The achieved effect was, on the one hand, the spectral modification leading to an increased color purity and, on the other hand, the angular control of the emission dictated by the angular dispersion of the dielectric cavity mode. The spectral change is shown in Figure 1.8g, while a map for different emission angles is shown in Figure 1.8h–i for the photonic structure and the reference, respectively. This thesis begins with these precedents, a project in which I participated in the design and analysis of optical properties prior to starting my Ph.D. work. However, in this thesis, as we will see later, we will focus on 2D photonic structures for emission modification of nanophosphors. |19 Chapter 1 1.4 Motivation and objectives The motivation of this thesis is to find photonic solutions for phosphors. Starting from a series of limitations or intrinsic characteristics of phosphors, the aim is to propose alternatives in the processing and optical design of systems to achieve emission control. The aim of the work is to explore new ways to control light emission using dielectric and metallic photonic nanostructures, and thus to develop light sources based on the integration of nanoemitters in customized photonic environments. From a fundamental perspective, the goals are explained above: •Develop efficient nanophosphors compatible with integration into photonic nanostructures. The goal is to create high-performance nanophosphors that can be seamlessly integrated into various nanophotonic configurations. •Design and fabrication of resonant photonic architectures at the emission and/or excitation frequencies of the nanophosphors. This involves the creation of structures that allow precise control of light emission properties. Proposed configurations include integration of nanophosphor films into multilayer systems, combination of emitting films with two-dimensional periodic arrays of scattering elements, and surface texturing of nanophosphor films. •Investigate the origin of photonic and hybrid photonic-plasmonic effects arising from the interaction between resonant systems and nanoscale light sources. This objective focuses on understanding the fundamental mechanisms behind the interaction of resonant photonic structures with nano-emitters and the resulting effects. From a technological perspective the goal is •to develop photonic architectures that can be integrated into light-emitting devices to achieve full control over the brightness, chromaticity and emission directionality of color converters used in LEDs. This involves the creation of advanced photonic structures that can be integrated into LED devices to enhance their performance by precisely controlling their light emission characteristics. 20 | Introduction 1.5 This Thesis This work, entitled ”Nanophosphor-based Photonics”, focuses on the control of light emission from photonic structures made of NPs doped with RE cations, namely nanophosphors. The work is mainly based on four journal articles, which are listed at the end of the document. In particular, the thesis is not organized as a chapter per article, but an effort has been made to synthesize the contents into a comprehensive whole that includes, first, a chapter on novel methods, another on the optical modification of nanophosphors by integrating them into complex structures, and finally, another chapter on more conventional methods. Thus, the results are divided into three chapters. Chapter 2 introduces the downshifting nanoemitters used and their basic properties. In addition, three novel processing methods implemented during this thesis are described in detail: ultrafast processing of nanophosphors, soft lithography of nanophosphors, and printable ink formulation based on nanophosphors. Furthermore, the results obtained for the upconversion of nanoparticles and their thermal processing and photophysical properties are discussed, although the integration of these nanoparticles into photonic structures is an ongoing process currently being carried out by other members of the group. Chapter 3 presents the results derived from the implemented methods with the down-shifting NPs, focusing on the optical properties of the photonic architectures. It is shown how the light emission is tuned using different photonic approaches. Chapter 4 describes in detail the Methods used. This chapter is placed at the end of the document to facilitate reading, since some of them may be known and/or not necessary for a smooth reading. In particular, the chapter is divided into three parts. First, the synthesis and processing of nanophosphors are discussed. Then, the techniques used for optical characterization are described. Finally, the computational methods used to study the samples in this thesis are briefly explained. |21 Chapter 1 22 | 2|Processing of nanophosphor thin films compatible with photonic integration In this chapter, novel processing routes of nanophosphors are addressed to obtain photonic solutions based on phosphors. There is an interest in optimizing the structural properties and photophysics of nanophosphors simultaneously. Hightemperature processing is often an intrinsic step in the treatment of these materials, which can potentially compromise their final properties. Therefore, we pay special attention to the compatibility of the processing methods and the maintenance and enhancement of the optical properties of the systems. Three different approaches are described in this chapter. First, rapid thermal annealing (RTA) is employed to treat thermally transparent and efficient thin films. Second, a brief introduction on soft-lithographed phosphors is discussed although the effects on the photoluminescent properties are discussed in the next chapter. Third, inkjet printing is presented as a tool to obtain transparent arbitrary motifs made from phosphors that can be combined with soft lithography to achieve versatile systems. Whereas the synthesis and optical and structural characterization have been fully carried out at ICMS, this chapter compiles 3 different collaborations that contributed in significant ways to the main results of this thesis. First, the RTA of phosphors was carried out in the Laboratory of Dr. Shunsuke Murai at Kyoto University.[97] Secondly, a brief stay in the Laboratory of Dr. Agustin Mihi at ICMAB resulted in the first monolithic patterned structures made of phosphor NPs.[98]And, finally, a secondment at the research centre PROFACTOR where it was possible to successfully combine soft lithography and inkjet printing techniques with phosphors.[99] Before addressing the phosphor processing with the aforementioned techniques, we introduce vanadate-based phosphors that are used with such techniques, discussing 23 Chapter 2 their emission properties and providing fundamental concepts for addressing the thin film processing approaches. Finally, a section focuses on the processing of versatile oxyfluoride films that exhibit photoluminescence upconversion. Although these materials were not later integrated into photonic structures as part of my thesis, it is a line that a colleague of the MOM group has continued to develop. However, it is relevant the work we carried out in the sense of optimizing a more novel system with different characteristics and that is why it is introduced in this section. 2.1 Downshifting vanadate nanophosphors This section provides a brief introduction to the nanophosphors used throughout this Thesis. Vanadate matrixes doped with RE cations stand out as a workhorse due to their efficiency, colloidal stability and simple synthesis. Yet, these findings are not limited to this particular material but may also extend to other phosphor-based NPs with varying characteristics. Before initiating my thesis, systematic studies were conducted in the group on synthesizing various particles based on different inorganic matrices that could be processed into thin, transparent films. A comparative analysis of tungstate, molybdate, or vanadate compounds revealed, through combined structural and optical analysis, that the tetragonal GdVO4matrix exhibited the highest efficiency among the different transparent nanophosphor films studied.[87] Nanophosphors are synthesized following a previously reported recipe using the solvothermal method.[100]The concentration of the dopant cations Eu3+ (Dy3+) was fixed at 10% (2%) molar since exceeding these values reduces the luminescence due to the creation of non-radiative de-excitation, i.e. through concentration quenching. The procedure followed for the synthesis of Eu0.1Gd0.9VO4and Dy0.02Gd0.98VO4is the same, except for the RE precursor salt and the adjustment of quantities based on the introduced dopant percentage and the molecular weight of the reactants. The specific procedure is not detailed here as it has not been modified for obtaining NPs but can be found in the Chapter 4. The resulting NPs are redispersed in methanol, where they exhibit high colloidal stability. Figure 2.1a shows an image taken with the transmission electron microscope (TEM) of the Eudoped NPs synthesized following the method described. The NPs doped with Dy3+ and Eu3+ have an average size around 35nm. The size distribution of a representative sample can be observed in the inset in Figure 2.1a. NPs dispersed in methanol can be directly deposited by spin or dip coating, forming thin layers. Figure 2.1b–c 24 | Processing of nanophosphor thin films compatible with photonic integration Figure 2.1:(a) TEM images of the GdVO4:Eu3+. The inset displays the size distribution. (b–c) SEM images of the top and cross-section views, respectively, of a thin film deposited with spin-coating. (d) Excitation (left side) and emission spectra (right side) of the GdVO4NPs doped with Dy3+(blue curve) and Eu3+(red curve). (e) Digital photographs of GdVO4:Eu3+ thin films excited with a xenon lamp using a monochromator at 285 nm. The samples are cut in half, with the right side thermally treated . shows the top view and cross-sectional view of a film made of GdVO4NPs using spin-coating taken with the scanning electron microscope (SEM). The emission of these phosphors is determined by the cation, while the absorption takes place in the matrix, which subsequently transfers energy from the vanadate group. Figure 2.1d shows the excitation spectrum taken for the most intense Eu3+ peak around 620 nm, but its shape does not change for the rest of the transitions for either Eu3+ or Dy3+. It can be excited over a wide range centered at 275nm. The emission spectra for Dy-doped and Eu-doped are also displayed in the right side of Figure 2.1d. In the particular case of Dy3+, it can be seen that the two intense transitions in the blue (485 nm) and yellow (570 nm) regions correspond to the 4D9/2→6H15/2and 4D9/2→6H13/2transitions, respectively. In the case of Eu3+, there is one transition that is notably more intense than the others in the red region of the spectrum (620 nm), corresponding to the 5D0→7F2transition. It is worth noting that these types of f–f levels transitions are quite rigid due to be internal levels. Therefore, their positions are quite fixed, and through chemical engineering, we cannot modify these positions but rather the relative intensity between peaks depending on the chemical environment of the cations. Thermal processing plays a role in the efficiency of phosphors for three reasons. First, as temperature increases, crystallinity improves and phase transitions can occur. Note that phase transitions may be undesirable for certain systems. In the case of our vanadate nanophosphors, the phase remains throughout the heating process in a zircon-type tetragonal phase and there is an increase in crystallinity. Second, adsorbed impurities on the surface of the nanocrystals synthesized, such as OH−or polyacrylic acid from the synthesis, are removed. These impurities cre- |25 Chapter 2 ate non-radiative pathways for de-excitation, thus decreasing PLQY. Furthermore, the emission of RE-based NPs is affected by the surface-to-volume ratio, which decreases with annealing due to nanocrystal growth. However, it is worth noting that at temperatures above 750°C, thermal treatment may affect the optical quality of the films, reducing their transparency. In a previous study, the dependence of quantum efficiency and optical quality of dip-coated films on processing temperature was investigated, and up to ≈800°C, an increase in the performance was observed.[87] However, these high temperatures are often incompatible with the other materials that may comprise a complex system, such as the one found in a multilayered device. This can initially limit the use of nanophotonics due to the sintering process that occurs at such high temperatures, in addition to the change in effective refractive index caused by changes in the porosity of the films. Therefore, in this thesis, thermal treatment is essential and is being carefully controlled in all works to find a balance between performance and efficiency in combining these NPs with photonic structures that modify their emission properties. In Figure 2.1e, a digital photograph illustrates how the thermal treatment affects a film with europium NPs. The sample is illuminated using a xenon arc lamp monochromated at 280 nm, and the emission of the material is photographed after only half of the sample has been calcined at 400°C. We can see a significant increase in intensity. 2.2 Rapid thermal processing As mentioned in the previous section, nanophosphors are typically synthesized at mild temperatures. Thus, high temperature processing is usually needed to improve their efficiency and unleash their full potential. Conventional heat treatments can lead to nanocrystal growth, resulting in increased scattering and loss of film transparency. RTA is a manufacturing technique that allows the heating of samples to temperatures above 1000 °C using ultrafast heating rates (several degrees per second) attained by high intensity lamps and lasers.[101]RTA is commonly used in very large-scale integration (VLSI) technologies to improve electrical properties of semiconductors for the fabrication of integrated circuits. This technique has been recently employed in the context of nanophosphor synthesis, with the aim of enhancing the PLQY, while simultaneously avoiding the growth of the grains.[102]However, transparent films have not yet been achieved with this technique. To carry out the use of this processing technique, we followed the steps sketched in Figure 2.2a–b. Thin films were prepared by spin coating the alcoholic disper- 26 | Processing of nanophosphor thin films compatible with photonic integration Figure 2.2:Scheme of the procedure: (a) spin-coating of the dispersion of NPs and (b) rapid annealing of the film. Time profile of thermal processing for conventional heating (red) and rapid thermal annealing (blue). (d–i) Scanning electron micrographs of the top view of nanophosphor films as-deposited (d–e), annealed at 1000°C in a conventional oven (f–g) and annealed at 1000°C using RTA (h–i). (j) X-ray diffraction patterns of samples as-deposited (black), annealed at 1000°C in a conventional oven (red) and 1000°C with RTA (blue). Diffractogram of standard PDF-86-0996 (ICDD 2014) is also shown as grey lines for comparison. Circles highlight reflections associated to cristobalite structure. (k) Ballistic transmittance measured at normal incidence from a GdVO4:Eu3+ nanophosphor thin film as-deposited (grey), and after annealing at 1000 °C in a conventional oven (red) or using RTA (blue). sion of nanophosphors (GdVO4:Eu3+ in methanol 5%) over quartz substrates that withstand the thermal treatment and are UV-transparent. Three depositions were conducted to ensure a uniform film with a total thickness of ∼250 nm. Then, we studied films treated with RTA, not annealed (as-deposited) and annealed with a typical rate in a common furnace (conventional). The thermal curves of the processing are represented as time profiles of the temperature in Figure 2.2c. In particular, it takes 16 hours for a film to go from room temperature to 1000 °C with a heating rate of 1 °C per minute and less than 7 minutes with RTA with a rate of 10 °C per second. However, it encloses more than a shortening of the process; as discussed below, there is an influence on both the structural and optical properties. We employ SEM to inspect the structural changes in the annealed films as a function of the heating profile. Specifically, we show images of the top view of films as-deposited — Figure 2.2d,e—, and annealed in a conventional furnace — Figure 2.2f,g—, or using RTA —Figure 2.2h,i—. SEM pictures clearly show that conventional processing leads to uncontrolled crystal growth and the emergence of |27 Chapter 2 Figure 2.8:(a) Image of a photonic QR code made of nanopatterned nanophosphors with defects caused by swelling of the stamp before the evaporation of the solvent. The size of the substrate is 5×5cm2. (b) Detail of the image (a). (c) Image of a PDMS stamp with the printed phosphor NPs on top after the solvent has evaporated and before the photopolymer is deposited. The size of the printed QR code is 2.4×2.4cm2. 34 | Processing of nanophosphor thin films compatible with photonic integration 2.4 Inkjet-printed nanophosphors Inkjet printing is a versatile deposition technique that ease a controlled deposition of several types of materials, referred to as inks, with high spatial resolution onto different substrates.[116,117]. It proves interesting for processing functional nanophosphor films since inkjet printing provides practical advantages such as material savings if it is compared with other conventional wet-deposition technique (spin or dip-coating). Moreover, controlled position in the deposition is accomplished, which yields an interesting prospect for fields of applications like anti-counterfeiting (AC) or microdevices.[118,119,120,121,122]Among the inks developed with phosphors, the applications are focused on AC, and typically the combination of different light emission processes is sought: down-shifting, upconversion, and persistent PL. This is because systems are sought that respond to multiple stimuli simultaneously to provide additional layers of security.[117,123,124] The physics underlying the inkjet printing process involves different interactions. A comprehensive understanding of different steps such us droplet formation dynamics, substrate interaction upon impact, or solvent evaporation kinetics, is crucial to implement this technique.[125]The requirements of rheological magnitudes depend on the printing mechanism and the specific printer and cartridge. For instance, there are differences between direct printing, i.e. a continuous jet, and drop-on-demand printing.[126]In the framework of my research stage in PROFACTOR, I used a Fujifilm Dimatix DMP-2850 printer which is designed for ink development in the field of materials science, because it has a set of tools that allow you to perform many tests and verify the behavior and viability of the ink. For example, it is possible to change the resolution (spacing between dots), heat the substrate, heat the ink, it contains a camera to observe droplet ejection, and the cartridge has a capacity of 1.5 mL ensuring material savings. The electronics behind the piezoelectric actuator, i.e. the waveform, can be also modified, although our focus remained on the material properties due to the numerous parameters to optimize during the research stay. The final objective is combining inkjet-printed samples with soft lithography and tune the optical properties of the phosphors. We followed a simple route as proof of concept focusing on the modifications of the rheological parameters to fit between the boundaries of type of printer (drop-on-demand print-head with a piezoelectric actuation mechanism). We part from the above-descripted paste based on RE-doped GdVO4phosphor NPs which uses ethyl cellulose as an organic binder and α-terpineol as solvent. The paste is further diluted in α-terpineol, reducing the initial particle |35 Chapter 2 Figure 2.9:(a) Schematic of the ink formulation and printing mechanism. (b) Contact angle measurements at room temperature between ink or water and the different substrates (glass, flat PDMS, PDMS Stamp). concentration to 3.3 wt.%. The surface tension of the inks was within the printable range at room temperature (28.3 mN m−1); however, due to the high viscosity of α-terpineol (37.9 mPa.s at 25°C, value extracted from manufacturer, SAFC≤96% ) and the role of ethyl cellulose in increasing the viscosity, the ink significantly exceeded the maximum viscosity required for droplet ejection. A common approach is to combine solvents with different boiling point and/or viscosity to achieve the desired parameters and control coffee-stain effect.[127]Initially, we attempted to combine it with ethanol (22.39 mN m−1at 20°C surface tension). However, the printing was unstable because the cartridge got clogged as ethanol evaporated at the nozzle exit. Due to the limited number of organic solvents that are soluble in α-terpineol, which is oily, to address this issue, we raised the printing temperature by heating the ink in the cartridge, and we carried out the printing at 60°C lowering the viscosity by an order of magnitude to ∼18 mPa s. Figure 2.9a shows the scheme of the final formulation. Another crucial aspect of the printing process is the interaction of the generated droplet with the substrate. On one hand, good wetting is necessary to ensure that the material adheres where it is deposited. On the other hand, excessive wetting can lead to a loss of resolution if the droplet spreads too much. Ultimately, wetting also determines the thickness of the deposition. Wetting depends on the surface tension not only of the ink but also of the surface on which it is deposited. To study this interaction, we measured the contact angle (θc) between the ink and different substrates as depicted in Figure 2.9b. The measurements with water are also included. The wetting of the ink on the glass substrate is high (θc= 22.4°), while it is substantially reduced on flat PDMS (θc= 61.0°), making the printing more challenging. However, for nanostructured PDMS, the contact angle is slightly lower (θc= 52.9°). To evaluate the performance of the inks, we printed different shapes onto glass while 36 | Processing of nanophosphor thin films compatible with photonic integration Figure 2.10:Pictures of emitting patterns under UV excitation. Eu-inklines between 300 mm and 50 mm in width with intervals of 50 mm, and regular hexagons with an apothem of 2mm. An image of the PL of similar line patterns printed with an EuDy-ink is displayed at the bottom of the panel. (b–d) Low-magnification SEM images of red-emitting stripes with widths of 250, 150 and 50 µm. The nominal value of the print width is indicated by a dashed line. The scale bar corresponds to 1mm. (e) Low-magnification SEM image of a red-emitting printed hexagon. The nominal design of the print is indicated by a dashed line. (f) High-magnification SEM image of a top view of the hexagon shown in (e). (g) PL spectra of an Eu-ink (top) and an EuDy-ink print (bottom). (h) logos of the European Research Council and the project Nanophom printed on glass under UV light. varying the resolution, which was ultimately fixed at 1270 dpi. To achieve this, we used two inks doped with different cations. The first ink (Eu-ink) exclusively contains particles doped with Eu3+, while the second ink consists of a mixture of NPs doped with Eu3+ and Dy3+.Figure 2.10a displays the appearance of the prints under UV light for diverse geometries, in particular, they are lines between 300 mm and 50 mm in width with intervals of 50 mm, and regular hexagons with an apothem of 2mm. Low magnification SEM images are shown in Figure 2.10 b–d for different width. Certain instabilities become discernible, particularly when printing motifs with smaller dimensions. However, they are imperceptible to the naked eye when dealing with larger dimensions. Figures 2.10e–f illustrate a hexagon at low and high magnifications, respectively. The photoluminescence spectra are similar to those shown previously, revealing the characteristic peaks of Eu3+ and Dy3+. Finally, the red amitting logos of the European Research Council and the project that partially funded this work, Nanophom, are shown under UV light excitation. Other RE-based NPs have been processed into a paste following a similar recipe, which in principle opens the possibility of printing various nanomaterials using inkjet printing,[128,115,45]thus expanding the range of applications, for example to those where persistent or upconverting luminescent materials stands out, such as signaling or thermometry. |37 Chapter 2 2.5 Upconverting oxyfluoride nanophosphors UC phosphors are materials that can absorb two or more low-energy photons through sequential absorption or energy transfer, and then emit a single high-energy photon from the deexcitation to a lower energy state or ground state. This process involves metastable excited states, typically found in RE ions with f-electron levels, which are critical for enabling upconversion UC. There are three major UC processes: Excited State Absorption (ESA), which is a two-step absorption process with relatively low efficiency; Energy Transfer Upconversion (ETU), which is the most efficient mechanism, where energy transfer from another RE cation supports absorption in metastable states; and Photon Avalanche, where electrons in metastable states are further excited to higher energy states by resonant pump light absorption, exhibiting a distinct power threshold in power-dependent measurements.[129]These upconversion processes are of interest for applications such as optoelectronics, solar cells, displays, biolabeling, fluorescence microscopy, sensing, and security labelling. In the canonical material, Yb3+ is introduced as the sensitizer and excited in the near infrared (NIR), around 980 nm, and, Er3+ is the active cation that emits green and red light from two-photon processes. Yb3+ can also be used with Ho3+ to produce red and green emission or Tm3+ to produce a blue color, processes involving two and three NIR photons, respectively. In this section, we will explore the optimization of thin films of UC NPs based on oxyfluoride hosts.[128]The optimization of the performance of UC nanoparticles (UCNP) is significant because the major challenge in developing applications based on UC nanotechnology is related to efficiency and brightness.[130,131]Several strategies have been designed to overcome inefficient channels of deexcitation and low absorption, including surface passivation through core–shell architectures, engineering the distribution of dopants, or selecting the appropriate host for the nanophosphor.[132]The most efficient UC nanophosphors consist of fluoride-based hosts, which have very low phonon energies, reducing the probability of nonradiative pathways through multiphonon relaxation. Specifically, sodium yttrium fluoride (NaYF4) stands out as the most efficient host developed to date, despite its low thermal stability, which limits its applicability.[133,134]On the other hand, oxide hosts generally offer better thermal stability, although they tend to have higher phonon energies, which can lead to lower UC efficiencies.[135,136]It is therefore attractive to develop host nanomaterials that combine the robustness of oxides with the high conversion efficiency of fluorides. In this context, oxyfluorides represent a promising host family that has attracted interest in recent years.[137,138] 38 | Processing of nanophosphor thin films compatible with photonic integration Figure 2.11:(a) TEM micrograph of the starting YF3:Yb3+Er3+ NPs. The inset shows the size distribution of the long side of the NPs. (b) Schematic processing of nanophosphor film preparation by doctor blading method from a nanophosphor paste. (c) SEM micrographs of cross section (left) and top view (right) of a nanophosphor film annealed at 450 °C for 6h. (e) Picture of a film same film under sunlight. These NPs are obtained following previously reported methods.[139]The synthesis can be found in Section 4.1 from which we obtain YF3:Yb3+,Er3+. The NPs have a lentil-like appearance, with some size dispersion. A TEM as well as the long dimension distribution histogram of YF3:Yb3+,Er3+ NPs is shown in the inset. The mean long size is 134 nm (σ= 19 nm). The processing of the NPs as thin films is performed in two ways. The first one is directly by spin-coating, obtaining films with thicknesses >1µm that can be found in the Supporting Information of the article (Ref. [128]). However, these films are not very efficient from an absorption point of view, as we will see later, so efforts have been made to process thicker films by using viscous pastes and blade coating. (Figure 2.11b). In Figure 2.11b we can see how the films look like after thermal processing (450°C, 6h), both the cross-section, where we observe a 14.7µm-thick film, and the top view where we see the particles forming the layer. Finally, an image of the sample under sunlight is shown, showing a square in the center of the substrate. Samples are rather transparent but with a certain degree of scattering, typically associated to surface roughness. After the preparation of the films by blade-coating, thermal annealing is used to remove organic impurities from the NP synthesis and paste and to provide mechanical stability to the film. Annealing also affects the composition and structural properties of the crystalline NPs and has an effect on the photophysical properties, so the study of the optical and structural properties allows the understanding and optimization of this photoluminescent system. Figure 2.12a shows 4 curves of UC photoluminescence (UCPL) of (Yb3+, Er3+)-doped nanophosphor films with a thickness of ∼15 µm annealed at temperatures ranging from 400°C to 550 °C for 6 h. The excitation takes place at the Yb3+ cations which absorb a 980 nm laser and the energy is transferred to Er3+.Figure 2.12b shows the schematic of the Yb3+ and |39 Chapter 2 Figure 2.12:(a) UCPL spectra of nanophosphor films annealed for 6 h at different temperatures (blue for 400 °C, red for 450 °C, green for 500 °C, and yellow for 550 °C). The inset contains the spectrally integrated ([400, 750] nm) UCPL intensity of nanophosphor films annealed for 6 h at different temperatures. (b) Energy-level diagram of Yb3+ and Er3+ ions and energy transfer processes. (c) Total T with a zoom at the excitation range and (d) absorptance spectra of the excitation range measured from films annealed at different temperatures, as labeled in panel (a). Er3+ levels and the different possibilities that give rise to the UCPL. The absorption of two successive photons gives rise to red (4F9/2→4I15/2) and green (2H11/2→ 4I15/2&4S3/2→4I15/2) emission. The ratio between the transitions depend on the host and the percentage of cations. In our case, the samples are red emitting since the majority of the emitted photons come from the 4F9/2→4I15/2transition, while in the conventional NaYF4:Er,Yb NPs, the red and green emission peaks are comparable. As a result, NaYF4:Er,Yb NPs are green-emitting to the human eye because our eyes are more sensitive to green. In Figure 2.12a we observe a significant change of the emission intensity with annealing temperature. A change in intensity may be related to changes in absorption and changes in the LDOS. Normally the origin is a combination of the different effects. These may be attributed to: (i) removal of organics, (ii) changes of nonradiative pathways, (iii) compositional changes during the oxidation process and crystallographic arrangement (geometrical distortions) that have an influence on the oscillator strength and on the chemical and optical environment of the cations. To investigate the observed PL trend, we conducted optical and structural characterizations of samples processed at different temperatures. The total T for various temperatures is presented in Figure 2.12c. There are variations in intensity at the higher energy range, which could be linked to increased scattering with temperature. Examining the excitation region (inset), we observe some discrepancies. To clarify these, by measuring the total R, we can calculate A and identify changes in absorption along the samples processed at different temperatures. The spectra of A are displayed in figure Figure 2.12d. To make a quantitative comparison of 40 | Processing of nanophosphor thin films compatible with photonic integration the absorption and determine whether they contribute to the increase in PL for the sample with the highest intensity, which was treated at 450 °C for 6 hours, we highlighted the excitation wavelength used in the absorption measurements—a laser with a peak intensity at 976 nm. Indeed, A is higher (around 4.5%) for the sample with the strongest PL emission. However, the effect cannot be solely attributed to changes in absorption when examining the A to PL ratios. The A spectrum reveals not only a variation in the amount of absorption but also a shift in the spectral shape. Yb3+ ions exhibit a single transition in this spectral region, but the sublevel structure becomes evident, similar to what we observe in Er3+ emission around 660 nm or the doublet in the most intense peak of Eu3+ (Figure 2.1d). This sublevel structure reflects changes in the crystal environment, which in turn influence electronic energy levels. Specifically, samples treated at 500 °C and 550 °C display higher absorption in the high-energy sublevel, along with a more pronounced sublevel structure in emission (see the green and yellow curves in Figure 2.12a). In summary, the annealing process induces a significant change in the chemical environment of the absorbing and emitting cations, resulting in alterations in both the shape and intensity of absorptance. However, a more detailed structural analysis is necessary to fully understand these effects. In order to analyze the structural properties of (Yb3+, Er3+)-doped nanophosphor films, we performed XRD measurements. Figure 2.13a displays XRD patterns for ∼15 µm-thick films annealed at 400, 450, 500, and 550 °C. For comparison, we show the diffractogram of a thin film (∼1µm) of (Yb3+, Er3+)-doped nanophosphors (labeled as “as-deposited”). Paste-based samples require thermal processing to mechanically stabilize the film, making it impossible to perform structural or optical characterization on as-prepared films. For this reason, we use a film deposited from a suspension of the same YF3:Yb3+, Er3+ NPs, which does not require thermal processing to achieve mechanical stability, to compare the XRD patterns of upconverting films annealed at different temperatures with that of an as-prepared sample, i.e. without any thermal processing. The as-deposited sample features the orthorhombic YF3phase (PDF 00-032-1431).[139]Annealing induces a phase transformation of the nanophosphor matrix, because of the inherent instability of the fluoride host. Indeed, O atoms replace F ones in the lattice, increasing the oxygen content with temperature and time, until complete oxidation of YF3into Y2O3is eventually fulfilled.[140,141]Specifically, films annealed at 400 °C show a two-phase mixture of orthorhombic YF3and orthorhombic Y7O6F9(PDF 01-070- 0867). Increasing the annealing temperature to 450 °C turns the film into a pure orthorhombic Y7O6F9phase. In turn, samples annealed at 500 °C possess a pure |41 Chapter 2 Figure 2.13:(a) XRD patterns of (Yb3+, Er3+)-doped nanophosphor films annealed at different temperatures for 6 h. (b) Zoom of the angular range showing the most intense peak displayed in panel (a). PDF reference codes of the patterns included are 00-032-1431 for YF3, 01-070-0867 for Y7O6F9, 00-025-1012 for YOF, and 00-041-1105 for Y2O3. (c) Time-dependent UCPL of films annealed at different temperatures (blue circles for 400 °C, red for 450 °C, green for 500 °C, and yellow for 550 °C), along with their fittings (shown in black). A 980 nm pulsed laser was used as an excitation source. Emission is collected at 670 nm. rhombohedral YOF phase (PDF 00-025-1012), whereas at 550 °C, a two-phase mixture of rhombohedral YOF and cubic Y2O3phase (see PDF 00-041-1105) is found. Figure 2.13b shows a zoom of the most intense reflection to visualize the shift to higher angles following the trend of the oxidation process (see PDFs at bottom of the figures). Notice that experimental patterns appear to be displaced to higher angles, with respect to the theoretical ones for all cases, which can be attributed to the effect of doping cations on the size of the unit cell. Substitution of Y3+ ions for smaller Er3+ and Yb3+ in the lattice results in unit-cell contraction, causing a shift of the reflections to higher diffraction angles. In addition to the phase transformation, at 450 °C, a significant increase in the XRD intensity of the peaks is clearly observed, along with a reduction in their full width at half-maximum, which indicates crystallinity was improved as a result of the annealing. Besides healing crystal lattice defects, thermal processing enables the removal of organics that act as quenchers of the emission. All together, these effects result in the rise of the UCPL for samples annealed between 400 and 450 °C shown in Figure 2.12a. However, further increase in the annealing temperature leads to nanophosphor matrices with higher oxygen content and higher phonon energy, which cause a rapid reduction in the UCPL,[142,143]as displayed in Figure 2.12a. Our results bring out an inherent trade-off in the quest of efficient oxyfluoride materials: annealing temperature must be high enough to remove organic quenchers and heal lattice defects but at the same time temperature must be kept as low as possible to limit the oxidation of the fluoride host. Understanding these effects at the microscopic level is essential, as they directly influence the local environment of Er3+ 42 | Processing of nanophosphor thin films compatible with photonic integration emitters and, consequently, the dynamics of the UCPL.To further explore this relationship, we analyze the time-dependent UCPL of films annealed at temperatures ranging from 400 °C to 550 °C, establishing a clear link between lifetime and crystal structure. We monitor the most intense Er3+ transition. Results are shown in Figure 2.13c. A two-exponential model A1e−t/τ1+A2e−t/τ2(2.1) is employed to describe the PL dynamics of nanophosphors to account for the two different decay rates (τ−1 1and τ−1 2) expected for Er3+ cations embedded in different crystal lattices. A1and A2are fitting constants associated with the relative weight wiof each contribution to the sum. A similar model has been proven useful to account for the different rates expected for cations with different crystalline environment in nanophosphors.[103]Our analysis reveals two distinct components for films annealed below 450 °C, in which a two-phase mixture of Y7O6F9and YF3was identified, with a high decay rate component τ−1 1∼(20 µs)−1associated with the Y7O6F9phase, which is predominant (relative weight above 90%), and a low decay component τ−1 2∼(67 µs)−1associated with Er3+ cations that sit in the YF3lattice. We associate the low decay component with the fluoride phase, because lifetimes connected to fluoride lattices are generally longer than those of oxyfluorides.[144,145] Small variations around these average values are observed (see Table 2.1). These are expected considering that photoluminescence dynamics is extremely sensitive to the local environment of the emitters, which can be slightly different, depending on the processing conditions. Interestingly, the relative contribution of the low decay rate component decrease as the annealing temperature increases, which is in agree- Tanneal.(°C) τ1(µs) w1(%) τ2(µs) w2(%) Host crystal phase 400 15.6 93.7 68.3 6.3 YF3+ Y7O6F9 430 21.2 95.0 64.9 5.0 YF3+ Y7O6F9 450 21.2 100 – – Y7O6F9 470 19.5 100 – – Y7O6F9+ YOF 485 20.2 100 – – Y7O6F9+ YOF 500 30.5 100 – – YOF 550 34.4 100 – – YOF + Y2O3 Table 2.1:Fitting parameters of the time-dependent UCPL measured from nanophosphor films annealed at different temperatures along with the corresponding crystal phase of the host in each case, as extracted from the XRD analysis. |43 Chapter 3 400 nm, two extinction maxima are found at λ= [412, 492], while, for the 500 and 600 nm pitches, maxima were detected at λ= [515, 595] nm and λ= [615, 700] nm, respectively. Here, we fit the observed angular dispersion of the transmittance at θin = 0°, to that of the estimated Rayleigh Anomalies (RAs) observed, which are the conditions under which a diffracted order propagates in the plane of the array. This relationship can be predicted by considering the conservation of the parallel component of the wavevector, which is given by the following formula  ksc,|| = kin,|| + G(mx, my) (3.1) with ksc,|| and kin,|| being the parallel component of the scattered and incoming wavevectors, respectively, and G the reciprocal lattice vector,  G(mx, my) = 2π(mx/a, my/a), where ais the lattice constant of the square array and the pair of integers (mx, my) defines the diffraction order. We identified the first diffraction orders reflected into air that matched the experimental values as the ⟨±1, 0⟩and ⟨0, ±1⟩for the two sets of RAs observed (solid and dashed lines in Figure 3.1g–i). The fitting for the high energy RAs (dashed lines) results in an air-like refractive index (nef f = 1.03), while the lower energy RAs (solid lines) can be associated to diffraction at the top of the nanophosphor structure, with the values of the effective index found around 1.2, between the refractive index of the nanophosphor layer and that of air, namely nef f = 1.23, 1.19, or 1.17 for a = 400, 500, and 600 nm, respectively. Slight variations in the hole size or depth for different imprints may bring about small field intensity distribution variations, which result in the different effective refractive indices reported. The RA curves are plotted over the PL measurements in Figure 3.1g–i to analyze the directionality of the emission. The panels inFigure 3.1g–i shows the ratio between the PL of the patterned samples and the reference between θemi = –5°and θemi = 30°. All the features observed are correlated with those shown in the extinction measurements and are associated to the diffracted orders calculated previously. The emission spectra collected along the direction normal to the sample surface (θemi = 0°) for the flat and patterned samples is shown in Figure 3.2a together with a scheme (Figure 3.2b). The radiative coupling to free space for the most intense peak of the Eu3+ cation (619 nm) is enhanced twofold for the patterned film with the larger pitch (a = 600 nm). In fact, the intensity enhancement observed should be interpreted only as directional emission enhancement because patterning does not alter the fraction of light absorbed by the films at the excitation wavelength as has been verified by the measurement of absorptance which is plotted in Figure 3.2c 50 | Nanophosphor-based photonic architectures for tuned light emission and is roughly the same (see Section 4.2). The directionality of the emission was analyzed thoroughly imaging the focal plane. The PL images of the unpolarized red emission are shown in Figure 3.2b. The samples were mounted in a microscopebased setup and excited through the substrate with the UV source (see Section 4.2). The numerical aperture NA = 0.75 determined the maximum detectable angle (θemi ≈48.6°) and a bandpass filter was used to block the yellow emission and target the most intense red peak (associated to the 5D0→7F2transition). Please note that the previous measurements, shown in Figure 3.2g–i correspond to a horizontal (or vertical due to symmetry) cut in the BFP images displayed in Figure 3.2b. The flat sample (Figure 3.2e) shows no angular features, as expected for a Lambertian-like emitter. However, patterned films exhibit bands of improved directional emission that follow the dispersion of the RAs. In particular, 400 nm-pitch patterns result in directional sideward emission whereas 600 nm allows beaming of the emission in the normal direction. This analysis highlights the potential of surface patterning to tailor the directionality of the emission of thin phosphor films, being the angle at which the maximum PL directional enhancement is reached controlled with the lattice constant of the pattern. |51 Chapter 3 Figure 3.2:(a) Normal direction emission of the reference (shaded gray) and patterned samples: a = 400 nm (red), a = 500 nm (blue), and a = 600 nm (dark gray). The inset shows a picture taken from a sample under the illumination of a UV lamp. Notice that only the 1 cm ×1 cm central area of the sample is patterned. An scheme of the collection configuration is displayed above the graph in (b). (c) Absorptance of the reference (gray) and patterned samples following the same color code. (d) Fourier images of the unpolarized red emission (λ¿ 619 nm) of the flat sample, and nanostructured samples (a = 400, 500, and 600 nm, respectively). Curves correspond to the RAs calculated for the beams diffracted. Solid and dashed lines represent the diffraction curves corresponding to the higher and lower effective refractive indexes, respectively We measured the PL decay dynamics by MCS (Section 4.2). The excited state lifetime of the emitters is not expected to be modified by the patterning, since no significant change in the local density of photon states occurs as a result of the 2D grating imprinted onto the surface of the nanophosphor film. A two-exponential model is employed to describe the PL dynamics of nanophosphors to account for the different decay rates expected for RE cations located in different crystal environments. The analysis shows that the average lifetime of the Eu3+ transitions is found to be similar to that of the flat sample (1.39 ±0.02 ms), see Figure 3.3 . Besides, our results indicate that the main contribution to the average lifetime (around 90%) comes from the shorter rate associated to cations that sit in the bulk. This confirms that (1) our patterning process does not degrade the emission of the phosphor; and (2) the directional emission enhancement displayed in Figure 3.2 is only due to the efficient funneling of the emitted light in particular directions. Consistently, the absolute PLQY remains constant, within experimental error, with respect to the value attained for the reference samples. Flat films show emitted-to-absorbed photon ef- 52 | Nanophosphor-based photonic architectures for tuned light emission ficiencies of 55%, while the yields of patterned ones are 56%, 56%, and 54% for the lattice parameters 400, 500, and 600 nm, respectively. These results indicate that 600 nm-pitch patterning provides a redirection of the emitted light that results in a higher fraction of the emitted light in a solid angle close to the normal to the surface respect to a flat reference layer whose emission is Lambertian. Therefore, optical modes supported by the patterned film render the nanophosphor layer into a directional light source, being such directionality controlled by the lattice parameter of the periodic pattern. Figure 3.3:Time dependent photoluminescence at λ=619 nm for the flat (a) and patterned samples (b, c and d for a=400, 500 and 600 nm respectively). Solid curves represent the fitting as biexponential functions (N(t) = A1exp(–t/τ1) + A2exp(–t/τ2)). 3.1.2 Inkjet-printed photonic nanophosphors Combining inkjet printing with soft lithography allows for material savings and scalability, as well as expanding the range of potential applications. In this section, we study the results of modifying the optical properties resulting from such a combination. First, we nanoimprinted patternn from inkjet-printed substrates with prior partial solvent evaporation. Figure 3.4a shows images of the resulting samples under daylight and UV light; note that one of the hexagons was nanoimprinted over its entire surface, while the second was partially imprinted with a flat area of the mold. As a result, only half of the hexagon was textured for comparison, which clearly shows that the half of the untextured hexagon is transparent, i.e., the fraction of incident light that is diffusely scattered is only a few percent. In addition, the textured hexagons exhibit a characteristic diffraction pattern associated with the symmetry of the printed lattice, which appears as six bright spots when projected onto a screen |53 Chapter 3 Figure 3.4:(a) Image under daylight of the hexagons printed with the Eu-ink. An image of the PL of the same prints under UV excitation is shown as an inset. (b) Image of the diffraction pattern of transmitted beams projected on a screen when blue light illuminates the nanopatterned hexagon. (c) Top view SEM image of the transition area between patterned and flat print. (d) Top view SEM image of the patterned area with higher magnification. (e) Angular dependence of the PL intensity measured from a nanopatterned motif printed with the Eu-ink normalized to that of a reference flat motif. A bandpass filter at 620 nm (FWHM = 14 nm) is used to collect the most intense emission peak of Eu3+. Excitation is through the substrate using a UV lamp (254 nm). a) Angular dependence of the unpolarized directional emission enhancement. (f–h) Angular dependence of the polarized directional emission enhancement. The polarizer is aligned with an angle αrespect to the grating as shown in the sketch below (see Figure 3.4b). In particular , the PDMS mold consisted of a hexagonal array of pillars (radius = 150 nm, height = 250 nm) with a lattice parameter of 600 nm. The SEM micrograph in Figure 3.4c illustrates the transition in the sample that features the pattern in the center. The result is a periodic array of holes in which the phosphor NPs act as building blocks as shown in the higher magnification image (Figure 3.4d). The photonic patterning does not affect the structural properties of the films, which remain the same as for the untextured films. To characterize the emission of the prints, we performed an analysis of the directionality of the PL using back-focus plane imaging with a 0.75 numerical aperture objective. Figure 3.4e shows the unpolarized emission at λ= 620 nm of a nanopatterned print normalized to the PL of the reference, which has a homogeneous emission distribution. The nanopatterned print shows a C6 symmetry emission distribution associated with the triangular array and approximately a twofold increase in PL. In addition, when the polarized emission is collected, a notable effect on the directional features is observed. 54 | Nanophosphor-based photonic architectures for tuned light emission Figure 3.4f–h displays the directionality of the PL when a polarizer is rotated by α= 0°, 60°, and 120°with respect to the Y-axis of the grating, according to the sketches shown. As a result, the most intense directional features depend on the relative orientation between the polarizer and the surface texture. To gain insight into the physical origin of the observed enhanced directional emission and polarization dependence, we performed FDTD numerical simulations. To validate our model, we first measured the ballistic transmission (T0) of the nanopatterned print for s- and p-polarized light (see Multiangle UV-Vis Spectroscopy in Section 4.2) and reproduced these measurements with our simulations. The experimental data are shown in Figure 3.5a–b. Excellent agreement is found between the experimental and theoretical extinction spectra (defined as 1–T0), as shown in the Figure 3.5c–d. High extinction bands follow the dispersion of RAs. According to the reciprocity theorem, we study the luminescence properties of the nanopatterned print by considering the electric field intensity distribution inside the film under plane wave illumination.[152,153]In particular, we calculate the integrated intensity enhancement (IIE), i.e., the electric field intensity integrated inside the print |Epatt|2 |E0|2normalized to that of the reference film |Eref|2 |E0|2with E0the incident electric field, for a given λand an illumination direction given by the angle pair (θin,ϕin) of the spherical coordinates and the angle of polarization χrespect to the plane of incidence (see the sketch shown in Figure 3.5e). IIE(λ, θin, φin, χ) = RV|Epatt|2(r, λ, θin, φin, χ)dr RV|Eref |2(r, λ, θin, φin, χ)dr(3.2) Therefore, the PL intensity at λwill be favored at emission angles θem =θin and ϕem =ϕin where IIE>1. Figure 3.5f shows the IIE as a function of θin and ϕin = 0°for p-polarized (χ= 0°) and s-polarized (χ= 90°) incident light. The IIE peaks close to 1 for θin ∼5°and ∼10°for p- and s-polarized light, respectively due to the different angular dispersion of the diffractive modes for each polarization as revealed by the transmittance data in Figure Figure 3.5a–b. The maximum IIE for unpolarized light occurs at θin∼7°, as a result of the average between s- and p-polarization. Figure 3.5g shows the spatial dependence of the field intensity enhancement within a nanopatterned print in a unit cell of the array of p-polarized incident light (depicted in the sketch in Figure 3.5e), while the field intensity profile within a flat reference film is shown in Figure 3.5h. Bright regions of high electric field intensity (∼1.8) inside the film appear at 620 nm for θin ∼7°due to the nanopatterning. According to the reciprocity theorem, emitters placed at such regions of enhanced field will radiate |55 Chapter 3 Figure 3.5:(a,b) Angular experimental extinction (1–T0, where T0is the ballistic transmittance) for p and s polarizations. (c,d) Angular simulated extinction for p and s polarizations. (e) Schematic of a unit cell (top view and a cross section) of the nanopatterned and flat film used for the numerical simulations. (f) IIE for the nanopatterned print at λ= 620 nm for both polarizations. (g,h) Electric field intensity profile for p-polarized incident light at θin = 7°and λ= 620 nm for the nanopatterned (e) and flat film (f) more p-polarized light to the far field at 620 nm and θem∼7°than a flat unpatterned print, confirming the physical origin of the dependence of the PL on polarized light. By making a connection between the angle χand the polarization angle αof the analyzer (see Section 4.2), we also computed the full angular dependence of the IIE for unpolarized light and for incident light polarized along the Y-axis (α= 0°), which are shown in Figures 3.6a and b, respectively. The computed angular patterns show good agreement with the experimental maps shown in Figure 3.5d providing theoretical support for the optical phenomena observed experimentally. Figure 3.6:The angular dependence of the IIE for unpolarized light (a) and light polarized along the Y-axis (α= 0°) (b). 56 | Nanophosphor-based photonic architectures for tuned light emission Self-standing and inkjet-printed nanopatterned phosphors Photonic texturing thus provides our prints with an additional feature for authentication, as the polarization dependence developed allows any counterfeit copy of the prints to be precisely identified. To demonstrate the potential and the versatility of our approach, we attempted the preparation of intricate nanopatterned prints. As we discussed in Section 2.3, we took an alternative approach, in which the printing is performed directly on the photonic PDMS mold. Figure 3.7a simplifies the process. Specifically, we print our Eu-ink onto the mold to form a 2.4 ×2.4 cm2nanopat- terned QR code. The sample was then heated at 120 °C for 12 min to evaporate the solvent. The next step was to demold the nanopatterned print. A droplet (∼1.5 mL) of OrmoComp resist was deposited on top of the print (Figure 3.7a, left side). A flexible polyethylene terephthalate (PET) film was used to spread and planarize the resist droplet; this step is not shown to simplify the scheme. The resist was then cured for 2 min using a UV lamp (LED, 365 nm, 110 mW cm−2at a distance of 1.5 cm). Finally, the PET sheet was removed and the sample was flipped and the PDMS was demolded (Figure 3.7a, right side). It should be noted that the photonic corrugation of the substrate plays a key role in the printing process as it increases the wettability of the ink by reducing the contact angle —see Section 2.4—, thus preventing the coalescence of the designs that occurs when printing on a flat PDMS film. At the same time, such an increase in the contact angle results in thicker prints with the same Eu ink, with a thickness of ∼800 nm. Figure 3.7b,c shows images of the self-standing flexible nanopatterned QR code under daylight and its PL under UV light, respectively. Note that scanning the QR code with mobile device or special reader will take you to the Multifunctional Optical Materials Group website. Interestingly, the periodic nanopattern is not only transferred to the print but also to the resist which is glass-like after being cured. It is possible to identify the QR code in the glass-like resist due to the refractive index difference between the phosphor NPs and the polymer matrix. Indeed, SEM mages of both regions-signaled in Figure 3.7b with dashed and solid squares, are displayed in Figure 3.7d–g. Finally, we carried out a thorough characterization of the angular dependence of the PL at λ= 620 nm of the nanopatterned QR code printed with the Eu-ink. The numerical simulations displayed in Figure 3.8a for unpolarized light confirm the reliability of our calculations, where we have considered a thicker film, as observed in Figure 3.7f, which allows the excitation of an extended photonic mode in the print as simulations of electric field intensity spatial distribution, and the infiltration of the phosphor NP film with the resist, leading to an increase in the effective refractive |57 Chapter 3 Figure 3.7:Simplified scheme of the fabrication process for fabricating a self-standing photonic QR code. The image on the left depicts the materials before curing the polymer while the image on the right represents the demoulding of the sample. Images of the nanopatterned QR print under daylight (b) and UV excitation (c). (d–g) SEM images of a top view and a cross section of the nanopatterned phosphor NPs (d,f) and the cured polymer (e,g). index of the print from 1.30 to 1.65 in the visible range. Such an increase in the index contrast also results in optical resonances that are spectrally and angularly narrower. In fact, a good cualitative agreement is observed between the angular distribution of the IIE (Figure 3.8a) and the PL (Figure 3.8b–d) of the print, which show a strong directional behaviour of the PL. In addition, the PL intensity is maximum at angles where two RAs overlap with a C6 symmetry inherited from the grating geometry: e.g. maximum PL is observed at θem = 40°and ϕem = 0°, where the (0,–1) and (–1,1) diffracted orders meet. Interestingly, the nanopatterning yields large values of field intensity enhancement in the region where phosphor NPs distribute as shown in the simulations. Such field distribution leads to IIE values as high as ∼15 for the directions of maximum outcoupling. However, the limitations of printing in an unpatterned comparable PDMS substrate hinders the production of a flat reference and thus the experimental confirmation of the theoretical emission enhancement.Figure 3.8c,d show angle dependent PL measurements for the nanopatterned QR for two values of α, 0°and 60°. Rotation of the analyzer results in a shift in the value of ϕem at which most of the PL is beamed. This is from θem ≈40°and ϕem = 0°for α= 0°to θem ≈ 40°and φem = 60°for α= 60°. All these evidences demonstrate the feasibility of the proposed approach to achieve a unique optical response from complex yet versatile nanostructured motifs made by combining inkjet printing and soft lithography. Our 58 | Nanophosphor-based photonic architectures for tuned light emission Figure 3.8:(a) Calculated Angular dependence of the integrated intensity enhancement for unpolarized light. b) Angular dependence of the PL intensity for unpolarized light. (c,d) Angular dependence of the PL intensity measured with an analyser α= 0°(c) and α= 60°(d). PL measurements are collected using a bandpass filter at 620 nm with a FWHM of 14 nm to record the most intense emission peak of Eu3+. A UV lamp (254 nm) is used as the excitation source. design will prevent unauthorized tampering because: (i) the user can visually identify authenticity, and (ii) the authorities can prove it. In fact, our tag is only visible to the user when illuminated at the correct wavelength. In addition, nanopatterning introduces a subtle marking embedded in the tag itself, which may be visually detected as a specific diffraction pattern, like the one shown in Figure 3.4b, using a low intensity laser pointer. Finally, it is also possible to assess the encryption capacity of our tag according to cp,[154]where cis the number of possible responses that the label can exhibit and pis the number of pixels of which it is composed. In our case, the response depends on the colour of the ink and the geometry of the nanopattern, while the number of pixels was set to 625 (25 ×25). Although in our specific demonstration the nanopatterning does not increase the encoding capacity of a conventional QR code (2p = 1.39 ×10188 when pis 625) because the pattern is applied to all printed pixels, it would be easy to expand the encryption capacity of our system by printing pixels with inks that emit different colors and texturing them individually with patterns of different geometries. For example, if each pixel could be printed with three different inks (e.g. Eu3+, Dy3+ and Eu3+-Dy3+) and textured with two different nanopatterns (e.g., hexagonal and square), the encoding capacity would increase up to 7p (1.54 ×10528 when pis 625). As a result, our approach allows for an encryption capacity that is virtually infinite considering the wide variety of nanopatterns that could be used, and makes counterfeiting extremely difficult since it is unlikely that two identical labels could be produced. |59 Chapter 3 66 | 4|Methods 4.1 Nanophosphor synthesis and processing 4.1.1 GdVO4:Eu3+ nanophosphor synthesis The synthesis procedure is carried out following a recipe optimized by the Colloidal Materials group at the Institute of Materials Science of Seville.[100]A solvothermal method is used to synthesize the GdVO4:Eu3+ and GdVO4:Dy3+ nanophosphors in the present thesis. The concentration of the Eu3+ (Dy3+) dopant cations was set at 10% (2% molar), as above these values, luminescence decreases due to the creation of non-radiative de-excitation pathways (concentration quenching). The procedure followed for the synthesis of Eu0.1Gd0.9VO4and Dy0.02Gd0.98VO4is the same, except for the RE precursor salt and the readjustment of quantities given by the introduced dopant percentage and the molecular weight of the reagents. In particular, two independent solutions of the same volume are prepared. In the first, the relevant lanthanides are mixed, whose precursors are the hydrated nitrates Gd(NO3)3·6H2O and Eu(NO3)3·5H2O or Dy(NO3)3·5H2O in ethylene glycol (EG) with a total lanthanide concentration of 0.02M. To facilitate dissolution in the viscous solvent, it is stirred with a magnetic stirrer at 80°C. In a separate vial, with a concentration of 0.1M, Na3VO4salt is dissolved in a mixture of EG and Milli-Q water at a ratio of 2:3. Additionally, polyacrylic acid (2 mg mL−1) is added as a stabilizing agent, whose functional groups interact with the surface of the NPs during synthesis, promoting control over their size and shape, as well as their colloidal stability. Both solutions are mixed and transferred into hermetically sealed test tubes and aged in a preheated oven at 120°C for 3 hours. Once cooled, the NP dispersion is centrifuged at least twice in ethanol and once in water, removing the supernatants 67 Chapter 4 on each occasion. Finally, the precipitates were redispersed in methanol with a concentration of 1-5% wt. depending on the deposition method. 4.1.2 Oxyfluoride nanophosphor synthesis Yttrium fluoride NPs containing 20% ytterbium and 2% of erbium were synthesized following a procedure reported elsewhere.[139]Briefly, 1.872 mmol of YCl3and 0.48 mmol of YbCl3were dissolved together in 105.6 mL of Diethylene glycol (DEG) under magnetic stirring and heating at 70 °C, while 0.048 mmol of ErCl3 was dissolved in 12 mL of DEG in another vial at the same condition. After dissolving, ErCl3was added to a solution containing YCl3and YbCl3precursors, then the solution was cooled to room temperature. Consequently, 2.4 mL of 1-Butyl-3-methylimidazolium tetrafluoroborate was mixed keeping the magnetic stirring for few minutes at room temperature to obtain a homogeneous mixture. The final solution was introduced in an oven at 120 °C and heated at this temperature for 15 h. After aging, the resulting dispersion was cooled to room temperature. The NPs were centrifuged and cleaned three times with absolute ethanol, then dispersed in methanol for the preparation of colloidal suspensions and pastes. The yttrium fluoride NPs doped with of 20% ytterbium and 0.5% of holmium or 0.5% of thulium were synthesized adjusting the amounts of HoCl3or TmCl3precursors and following the same procedure. 4.1.3 Nanophosphor based paste preparation It is possible to stack NPs through wet deposition methods such as spin or dip coating starting from alcoholic dispersions. As a result, layers with high optical quality can be obtained with a thickness of up to one micron. However, we found it necessary to obtain dispersed NPs in a stable form in other solvents with different physicalchemical properties to (i) the application of soft lithography and (ii) to change their rheological properties for processing through inkjet printing and, (iii) the obtention of thick films (>1µm). Some of the key properties are the volatility, the rheological properties and the compatibility with the polymeric stamps, preventing them from degradation. To achieve this, we prepare NPs-based pastes. The procedure was reported previously for the vanadate-based NPs presented here with the objective of obtaining thick (>1µm) layers. The method was suited from the recipe of TiO2 NPs for electrodes in dye sensitized solar cells.[170]It involves a solvent change to α-terpineol and the addition of some organic ligands, ethyl cellulose, which prevent the agglomeration of the NPs. To do so, we start from the dispersion of the NPs 68 | Methods in methanol obtained from the synthesis. The organic binders and new solvent are incorporated with tip sonication with a relation in weight of 0.3 and 4.0 times the weight of the nanophosphors respectively. Finally, the methanol is remove via distillation. This procedure was apply for downshifting GdVO4:Eu3+ and upconverting YF3:Yb3+,Er3+/Ho3+/Tm3+ nanophoshors. 4.1.4 Flexible upconverting nanophosphor coating PMMA was selected as support material for the flexible upconverting nanophosphor coatings. PMMA solutions with concentrations of 5 and 8 wt% were obtained by dissolving PMMA in anisole. First, a sacrificial SiO2-dense layer with a thickness of ∼150 nm was deposited over the substrate via a two-step spin coating process at 500 rpm for 20 s and at 2000 rpm for 60 s, followed by the annealing at 500 °C in a hot plate for 30 min. Then, we deposited the nanophosphor film following the method described above. This step also includes thermal processing if needed. For the polymer infiltration, 5 wt% PMMA solution was first infiltrated onto annealed nanophosphor films. A 8 wt% PMMA solution was next deposited on the resulted films with the same spin coating parameters. After infiltrating PMMA, samples were dried at 60 °C for at least 1 h. Dried samples were immersed in a 1% hydrofluoric acid (HF, Fluka, 48%) solution in Milli-Q water (Millipore, Bedford, MA) for etching the sacrificial SiO2dense layer. After 60 min, a flexible coating was detached from substrate and washed abundantly in water to remove residual HF. |69 Chapter 4 4.2 Optical Characterization 4.2.1 UV-Vis Spectroscopy The use of UV-Vis spectroscopy adapted for solid samples allows us to characterize how light propagates when it impinges on our samples. Specifically, in this thesis, the Cary 5000 UV-Vis-NIR spectrophotometer is used with two types of integrated accessories that provide complementary information. One is the Universal Measurement Accessory, which consists of a rotating sample holder and a detector that also rotates around the axis of the sample. This setup allows for polarized measurements of specular reflectance and ballistic transmittance at different angles, with control of the angular position of the sample and detector to 0.02°, although it is defined by the use of slits and the solid angle reaching the Si/InGaAs detector (controllable between 1-6°by apertures). A schematic of the type of measurements is shown in Figure 4.1a. On the other hand, to estimate the total reflectance and transmittance values, as well as the diffuse components, the configuration shown in Figure 4.1b is used, where an integrating sphere with an internal detector is used. By using different sample configurations and a diffuser, it is possible to measure total and diffuse transmittance (Figure 4.1b,i–ii) as well as total and diffuse reflectance (Figure 4.1b,iii–iv). This approach allows us to extract the absorptance value and the diffuse contributions that appear in the samples of this thesis when the thermal processes are not optimized as required to maintain the transparency of the samples. Figure 4.1:Simplified scheme of the UV-Vis-NIR spectrophotometer used (Cary 5000, Agilent). (a) Multi-angle configuration (Universal Measurement Accessory) (b) Integrating sphere configuration (UV-Vis Internal Diffuse Reflectance Accessory). 70 | Methods 4.2.2 Photoluminescence Spectrometry The commercial spectrometer FLS 1000 from Edinburgh Instruments was used for the majority of the photolumminescent characterization. The spectrometer is depicted in Figure 4.2 highlighting its main components. This equipment is equipped with two monochromators and various light sources that can be coupled, although in this thesis, we exclusively employed the white light sources (continuous and pulsed) and a 980 nm laser of 2W operating at continuous mode. The excitation monochromator is used to select a specific wavelength for exciting the sample, whereas the collection monochromator is used to select a specific wavelength of the light emitted or scattered by the sample. The different calibrations of the equipment (background, detector sensitivity and lamps calibration) enable the acquisition of emission and excitation spectra, as well as the quantification of the PLQY. Figure 4.2:Simplified scheme of the spectrometer used for the majority of the luminescent characterization |71 Chapter 4 Steady-state measurements In order to measure PL spectra, the excitation is fixed in a region where the sample absorbs while a scan is performed to collect the counts along the rage of emission. Figure 4.3a shows the spectrum of emission of a GdVO4doped with Eu3+ under UV excitation at 285 nm (illustrated as a vertical line). Depending on the type of material, absorption will occur in a spectral range and with different intensity; therefore, excitation spectrum measurements can be useful for determining the range in which our sample is brighter and/or reaching a compromise regarding the energy shift, understood as something that contributes to inefficiency in conversion, and absorption intensity. To do so, we set our emission monochromator at the most intense peak of the emission (∼619 nm for Eu3+ cations as illustrated in Figure 4.3b as a vertical lines) and measure the PL intensity at this wavelength while scanning the excitation wavelength. We can also measure the absolute PLQY of photoluminescent materials by using an integrating sphere. The method consist of measuring the total number of photons that are both absorbed and emitted which allows the direct calculations following the formula PLQY = emitted photons absorbed photons (4.1) To achieve this value, we calculate the absorbed photons by comparing a reference sample consisting of a scattering plate (non-absorbing material) and the sample of interest. We use wide slits, typically 8–10 nm, in the excitation arm to ease the integration of the scatted light at the range of absorption. Through precise spectral calibration of the equipment, from the two measurements that cover the full range (excitation and emission) as depicted in Figure 4.3c, we are able to quantify the fraction between the absorbed photons and the emitted. 72 | Methods Figure 4.3:Schematic representation of the static measurements carried out in the spectrometer. (a) Example of an emission measurement performed by maintaining the excitation fixed, at 285 nm in this case, and scanning the emission wavelength in the visible range. (b) Example of an excitation spectrum measurement where the emission monochromator is held constant, at the maximum of emission, while the excitation wavelength is scanned. (c) Calculation of the absolute quantum yield through the fraction of the emitted photons over the absorbed photons. The absorbed photons are calculated from the difference of the excitation scattering peak between the two configurations displayed in (e) and (d) for the sample and the scattering plate respectively. On the other hand, the emitted photons are calculated by integrating the emission range (configuration (e)). |73 Chapter 4 Time-resolved measurements The dynamic of the emission of nanophosphors is studied by using the photon counting method called Multichannel Scaling which is widely used for lifetimes above hundreds of nanometers. This method involves the use of a pulsed excitation source, a flashlamp in our case, to excite the sample, followed by the recording of the resulting photoluminescence decay. After the excitation pulse, the arrival of the photons is organized in channels within the time sweep generating a histogram of the photoluminescence decay after several pulses. Figure 4.4a depicts two-time sweeps divided into channels and an illustrative histogram. A real measurement after the optimization of number of channels, pulse frequency and slit width is shown in Figure 4.4b. In particular the decay curve of the GdVO4doped with Eu3+ is displayed. Figure 4.4:(a) Schematic representation of Multichannel Scaling method for dynamic measurements. (b) Example of an PL dynamic measurement performed with the microsecond lamp operating at 285 nm with a frequency rate of 50 Hz while the recolection is set at 620 nm. 74 | Methods 4.2.3 Angular photoluminescence with excitation control Commercial instruments offer high metrological quality, but they can be rigid and limited for certain measurements. To address this, optical stage setups can adapt to specific systems. For my Ph.D. thesis, we built a double goniometer with independent control of excitation and emission. Commercial fluorimeters used to fix excitation and emission at 90°, so changing the excitation angle alters the emission angle. The systems studied in this thesis focus on the angular-dependance of emission, but variations in absorption may also occur. A double goniometer allows fixing the excitation angle while varying the emission angle, enabling comparable intensity measurements.Figure 4.5a shows the configuration used to measure angular photoluminescence. The fixed excitation source is outside the rotating stage and the angle of incidence is controlled by fixing the goniometer of the sample (dark gray, G1). The samples are excited with a 285 nm LED (FWHM = 13 nm, power = 50 mW) and the excitation is collected by varying the collection arm that is mounted in the other goniometer (light gray, G2), through an optical fiber coupled to a spectrophotometer (Ocean Optics USB2000+VIS-NIR-ES). The resolution (<2°) is controlled by an iris diaphragm. The same setup can be modified and used to measure the ballistic transmittance of the system and obtain its extinction Figure 4.5b. Using a white halogen light source (Ocean Optics HL-2000-HP FHSA), this setup would allow us to obtain measurements similar to those obtained with the commercial spectrophotometer mentioned in the previous section, but much faster due to the detection used. In the thesis, measurements are made with both systems, with the commercial equipment proving to be more sensitive in the UV range. Figure 4.5:Scheme of the optical table setup based on a double goniometer. (a) Configuration for angular measurement of photoluminescence. (b) Configuration for angular measurement of ballistic transmission. |75 Chapter 4 82 | General conclusions Solution-based methods have been used and developed to process nanophosphors that are compatible with integration into various photonic structures and allow precise control of emission properties. Regarding processing techniques, two novel approaches stand out. First, rapid thermal annealing has been used for the first time to thermally treat thin films of downshifting nanophosphors while maintaining high transparency. Second, these nanophosphors were used as building blocks for structures by optimizing nanoparticle dispersions to make them compatible with soft lithography processing. Additionally, inkjet printing was employed to achieve greater control over deposition and to reduce material waste compared to other liquid-phase techniques such as spin or dip coating. These techniques broadened the potential applications for nanophosphors. In addition, work has also been done on upconversion emitters, optimizing the thermal processing of the thin films taking into account the photophysical properties and increasing the brightness of the emission. After this optimization, we have obtained flexible films that emit different colors, highlighting the versatility of nanoparticlebased systems. The integration of downshifting nanophosphors into photonic structures using the aforementioned processing techniques resulted in different configurations with interesting properties in terms of efficiency and light decoupling, exhibiting strong spectral and directional dependence. The developed systems comprise: •A highly efficient film coupled with an array of aluminum nanodiscs forming a square lattice with a parameter of 600 nm significantly increased the directional emission of light, funneling it in the normal direction (<10°). 83 Chapter 4 •Monolithic systems composed entirely of nanophosphors were created, where periodic surface texturing allowed the extraction of emitted light in different directions using different lattice parameters. This is the first example of nanoimprinting of phosphors, as opposed to other approaches that involve nanoimprinting additional materials on top of them. •Inkjet printing has been used to make thin film deposition more scalable and save a significant amount of material. For the first time, the combination of inkjet printing and soft lithography was applied to emitting materials. The result was the fabrication of arbitrary shapes with directional and polarized emission, including QR codes for anti-counterfeiting applications. These results highlight the successful development and integration of nanophosphors into photonic structures, achieving precise control over light emission properties and expanding their potential applications in various advanced technologies. 84 | List of publications This thesis is based on the following publications: Enhanced Directional Light Extraction from Patterned Rare-Earth Phosphor Films Elena Cabello-Olmo, Pau Molet, Agust´ın Mihi, Gabriel Lozano and Hern´an M´ıguez. Advanced Optical Materials, 2021, 9, 2001611. DOI: 10.1002/adom.202001611. Collective plasmonic resonances enhance the photoluminescence of rare-earth nanocrystal films processed by ultrafast annealing Elena Cabello-Olmo, Makoto Higashino, Shunsuke Murai, Katsuhisa Tanaka, Gabriel Lozano, Hern´an M´ıguez. Chemical Communications, 2023, 59, 1289. DOI: 10.1039/D2CC04779A. Inkjet-Printed and Nanopatterned Photonic Phosphor Motifs with Strongly Polarized and Directional Light-Emission Elena Cabello-Olmo, Manuel Romero, Michael Kainz, Anna Bernroitner, Sonja Kopp, Michael M¨uhlberger, Gabriel Lozano and Hern´an M´ıguez. Advanced Functional Materials, 2023, 33, 2305907. DOI: 10.1002/adfm.202305907. Highly Versatile Upconverting Oxyfluoride-Based Nanophosphor Films Thi Tuyen Ngo*,Elena Cabello-Olmo*, Encarnaci´on Arroyo, Ana I. Becerro, Manuel Oca˜na, Gabriel Lozano and Hern´an M´ıguez. ACS Applied Materials & Interfaces 2021, 13(25), 30051. DOI: 10.1021/acsami.1c07012. 85 Other related publications in which different results of the thesis research work are used: Optical Responses of Localized and Extended Modes in a Mesoporous Layer on Plasmonic Array to Isopropanol Vapor Shunsuke Murai*,Elena Cabello- Olmo*, Ryosuke Kamakura, Mauricio E. Calvo, Gabriel Lozano, Taisuke Atsumi, Hern´an M´ıguez, and Katsuhisa Tanaka. Journal of Physical Chemistry C, 2020, 124(10), 5772. DOI: 10.1021/acs.jpcc.9b10999. Photonic structuring improves the colour purity of rare-earth nanophosphors Dongling Geng, Elena Cabello-Olmo, Gabriel Lozano and Hern´an M´ıguez. Materials Horizon, 2018, 5, 661. DOI: 10.1039/c8mh00123e. Tamm Plasmons Directionally Enhance Rare-Earth Nanophosphor Emission Dongling Geng*,Elena Cabello-Olmo*, Gabriel Lozano and Hern´an M´ıguez. ACS Photonics, 2019, 6, 634. DOI: 10.1021/acsphotonics.8b01407. Asterisk indicates first co-authorship. 86 | Bibliography [1] N. C. George, K. A. Denault, and R. Seshadri, “Phosphors for solid-state white lighting,” Annual Review of Materials Research, vol. 43, pp. 481–501, 2013. (Cited in pages 1 and 3.) [2] S. Hariyani, M. S´ojka, A. Setlur, and J. Brgoch, “A guide to comprehensive phosphor discovery for solid-state lighting,” Nature Reviews Materials, vol. 8, no. 11, pp. 759–775, 2023. (Cited in page 1.) [3] M. Grohol and C. Veeh, Study on the Critical Raw Materials for the EU 2023. Publications Office of the European Union, 2023. (Cited in page 1.) [4] L. Alcaraz, B. Sotillo, C. Iglesias, F. A. L´opez, P. Fern´andez, C. Belviso, and A. Urbieta, “Lta zeolite particles functionalized with nanomagnetite for effective recovery of dysprosium from liquid solutions,” Microporous and Mesoporous Materials, vol. 363, p. 112843, 2024. (Cited in page 2.) [5] R. Lacal-Ar´antegui, “Materials use in electricity generators in wind turbines– state-of-the-art and future specifications,” Journal of Cleaner Production, vol. 87, pp. 275–283, 2015. (Cited in page 2.) [6] J. Cai, D. Li, L. Jiang, J. Yuan, Z. Li, and K. Li, “Review on CeO2-based photocatalysts for photocatalytic reduction of CO2: progresses and perspectives,” Energy & Fuels, vol. 37, no. 7, pp. 4878–4897, 2023. (Cited in page 2.) [7] Z. Liao, M. S´ojka, J. Zhong, and J. Brgoch, “Developing the Na2CaZr2Ge3O12: Cr3+ garnet phosphor for advanced NIR pc-LEDs in night vision and bioimaging,” Chemistry of Materials, vol. 36, no. 9, pp. 4654–4663, 2024. (Cited in pages 2 and 6.) [8] X. Yuan, C. Chen, H. Lei, and Z. Zhang, “Synthesis, characterization of CeO2@ ZIF-8 composite abrasives and their chemical mechanical polishing behavior on glass substrate,” Ceramics International, vol. 49, no. 3, pp. 5189– 5198, 2023. (Cited in page 2.) 87 [9] M. Laguna, N. O. Nu˜nez, A. I. Becerro, G. Lozano, M. Moros, J. M. de la Fuente, A. Corral, M. Balcerzyk, and M. Oca˜na, “Synthesis, functionalization and properties of uniform europium-doped sodium lanthanum tungstate and molybdate (NaLa (XO4)2, X= Mo, W) probes for luminescent and x-ray computed tomography bioimaging,” Journal of Colloid and Interface Science, vol. 554, pp. 520–530, 2019. (Cited in page 2.) [10] S. V. Eliseeva and J.-C. G. B¨unzli, “Rare earths: jewels for functional materials of the future,” New Journal of Chemistry, vol. 35, no. 6, pp. 1165–1176, 2011. (Cited in page 2.) [11] R. P. Reboredo, Las tierras raras. No. 99 in Colecci´on ¿Qu´e sabemos de?, Madrid, Espa˜na: Editorial CSIC y Catarata, 2022. (Cited in page 2.) [12] J.-C. G. Buenzli, “On the design of highly luminescent lanthanide complexes,” Coordination Chemistry Reviews, vol. 293, pp. 19–47, 2015. (Cited in pages 2 and 4.) [13] J.-C. G. Bunzli and V. K. Pecharsky, Handbook on the physics and chemistry of rare earths: including actinides. Elsevier, 2016. (Cited in page 2.) [14] J. S. Pimputkar, Siddha Speck, S. P. DenBaars, and S. Nakamura, “Prospects for LED lighting,” Nature Photonics, vol. 3, no. 4, pp. 180–182, 2009. (Cited in pages 2 and 3.) [15] F. Falchi and S. Bar´a, “Light pollution is skyrocketing,” Science, vol. 379, no. 6629, pp. 234–235, 2023. (Cited in pages 2 and 5.) [16] P. Dorenbos, “Crystal field splitting of lanthanide 4fn−15d-levels in inorganic compounds,” Journal of Alloys and Compounds, vol. 341, no. 1-2, pp. 156– 159, 2002. (Cited in page 3.) [17] P. Pattison, J. Tsao, G. Brainard, and B. Bugbee, “LEDs for photons, physiology and food,” Nature, vol. 563, no. 7732, pp. 493–500, 2018. (Cited in page 4.) [18] J. Y. Tsao, M. H. Crawford, M. E. Coltrin, A. J. Fischer, D. D. Koleske, G. S. Subramania, G. T. Wang, J. J. Wierer, and R. F. Karlicek Jr, “Toward smart and ultra-efficient solid-state lighting,” Advanced Optical Materials, vol. 2, no. 9, pp. 809–836, 2014. (Cited in page 4.) [19] M.-H. Fang, G. N. A. De Guzman, Z. Bao, N. Majewska, S. Mahlik, M. Grinberg, G. Leniec, S. M. Kaczmarek, C.-W. Yang, K.-M. Lu, et al., “Ultrahigh-efficiency near-infrared Ga2O3: Cr3+ phosphor and controlling of phytochrome,” Journal of Materials Chemistry C, vol. 8, no. 32, pp. 11013– 11017, 2020. (Cited in page 4.) 88 | [20] U. Ekim, E. ˙ Ilter, E. ¨ Ozan, Y. Tem¨urhan, M. C¸. Ersundu, and A. E. Ersundu, “Glass-based LED system for indoor horticulture: enhanced plant growth through Sm3+ and Tm3+ co-doped luminescent glasses,” Physical Chemistry Chemical Physics, vol. 25, no. 34, pp. 23150–23163, 2023. (Cited in page 4.) [21] K. J. Gaston and A. S´anchez de Miguel, “Environmental impacts of artificial light at night,” Annual Review of Environment and Resources, vol. 47, pp. 373–398, 2022. (Cited in page 5.) [22] S. Hariyani and J. Brgoch, “Advancing human-centric LED lighting using Na2MgPO4F: Eu2+,” ACS Applied Materials & Interfaces, vol. 13, no. 14, pp. 16669–16676, 2021. (Cited in page 5.) [23] B. Zandi, O. Stefani, A. Herzog, L. J. Schlangen, Q. V. Trinh, and T. Q. Khanh, “Optimising metameric spectra for integrative lighting to modulate the circadian system without affecting visual appearance,” Scientific Reports, vol. 11, no. 1, p. 23188, 2021. (Cited in page 5.) [24] C. I. de l’´ Eclairage, “What to document and report in studies of ipRGC- influenced responses to light, cie tn 011: 2020,” CIE Vienna, Austria, 2020. (Cited in page 5.) [25] F. Feltrin, F. Leccese, P. Hanselaer, and K. A. Smet, “Impact of illumination correlated color temperature, background lightness, and painting color content on color appearance and appreciation of paintings,” Leukos, 2019. (Cited in page 5.) [26] R. Dang, B. Wang, X. Song, F. Zhang, and G. Liu, “The mathematical expression of damage law of museum lighting on dyed artworks,” Scientific Reports, vol. 11, no. 1, p. 10951, 2021. (Cited in page 5.) [27] M. Pan, Y. Zhong, H. Lin, H. Bao, L. Zheng, R. Hong, B. Dai, D. Yu, D. Zhang, and S. Zhuang, “Multilevel optical data storage in a Eu2+/Ho3+ doped Ba2SiO4phosphor with linear mapping between ultraviolet excitation and a thermoluminescence/photostimulated luminescence response,” Journal of Materials Chemistry C, vol. 10, no. 2, pp. 496–505, 2022. (Cited in page 5.) [28] M. Gu, Q. Zhang, and S. Lamon, “Nanomaterials for optical data storage,” Nature Reviews Materials, vol. 1, no. 12, pp. 1–14, 2016. (Cited in page 5.) [29] N. Katumo, K. Li, B. S. Richards, and I. A. Howard, “Dual-color dynamic anticounterfeiting labels with persistent emission after visible excitation allowing |89 smartphone authentication,” Scientific Reports, vol. 12, no. 1, p. 2100, 2022. (Cited in page 6.) [30] X. Yu, H. Zhang, and J. Yu, “Luminescence anti-counterfeiting: from elementary to advanced,” Aggregate, vol. 2, no. 1, pp. 20–34, 2021. (Cited in page 6.) [31] N. M. Maalej, A. Qurashi, I. Bennour, L. R. Haddada, M. N. Shaikh, M. Ilyas, N. E. B. Amara, R. Maalej, and M. A. Gondal, “Green emitting rare earth Gd2O3: Tb3+ nanoparticles for rapid imaging of latent fingerprint,” Methods and Applications in Fluorescence, vol. 9, no. 2, p. 025002, 2021. (Cited in page 6.) [32] Z. Liu, T. Lin, N. Chen, W. Meng, P. Cai, R. Liu, J. Zhang, X. Fu, and H. Zhang, “From scintillator to phosphor: X-ray induced long afterglow red luminescence from bi4ge3o12: Eu3+,” Journal of Luminescence, vol. 263, p. 120002, 2023. (Cited in page 6.) [33] C. D. Brites, R. Marin, M. Suta, A. N. Carneiro Neto, E. Ximendes, D. Jaque, and L. D. Carlos, “Spotlight on luminescence thermometry: basics, challenges, and cutting-edge applications,” Advanced Materials, vol. 35, no. 36, p. 2302749, 2023. (Cited in page 6.) [34] N. Chowdhury, N. Riesen, and H. Riesen, “Yb3+ and Er3+ codoped BaLiF3 nanocrystals for X-ray dosimetry and imaging by upconversion luminescence,” ACS Applied Nano Materials, vol. 4, no. 7, pp. 6659–6667, 2021. (Cited in page 7.) [35] Y. H. Kim, N. S. Viswanath, S. Unithrattil, H. J. Kim, and W. B. Im, “phosphor plates for high-power LED applications: challenges and opportunities toward perfect lighting,” ECS Journal of Solid State Science and Technology, vol. 7, no. 1, p. R3134, 2017. (Cited in page 7.) [36] C. Ma and Y. Cao, “Phosphor converters for laser driven light sources,” Applied Physics Letters, vol. 118, no. 21, 2021. (Cited in page 7.) [37] L. Wang, Z. Wei, C.-J. Chen, L. Wang, H. Fu, L. Zhang, K.-C. Chen, M.-C. Wu, Y. Dong, Z. Hao, et al., “1.3 GHz EO bandwidth GaN-based micro-LED for multi-gigabit visible light communication,” Photonics Research, vol. 9, no. 5, pp. 792–802, 2021. (Cited in page 7.) [38] X. Hu, Z. Chen, H. Ao, Q. Fan, Z. Yang, and W. Huang, “Rational molecular engineering of organic semiconducting nanoplatforms for advancing nirii fluorescence theranostics,” Advanced Optical Materials, vol. 10, no. 20, p. 2201067, 2022. (Cited in page 7.) 90 | [39] H. Li, Z. Zhou, Y. He, W. Sun, Y. Li, I. Liberal, and N. Engheta, “Geometryindependent antenna based on epsilon-near-zero medium,” Nature communications, vol. 13, no. 1, p. 3568, 2022. (Cited in page 7.) [40] P. Tockhorn, J. Sutter, A. Cruz, P. Wagner, K. J¨ager, D. Yoo, F. Lang, M. Grischek, B. Li, J. Li, et al., “Nano-optical designs for high-efficiency monolithic perovskite–silicon tandem solar cells,” Nature Nanotechnology, vol. 17, no. 11, pp. 1214–1221, 2022. (Cited in page 7.) [41] P. K. Sahoo, S. Memaran, Y. Xin, L. Balicas, and H. R. Guti´errez, “Onepot growth of two-dimensional lateral heterostructures via sequential edgeepitaxy,” Nature, vol. 553, no. 7686, pp. 63–67, 2018. (Cited in pages 7 and 17.) [42] C. Zhang, H. Dong, C. Zhang, Y. Fan, J. Yao, and Y. S. Zhao, “Photonic skins based on flexible organic microlaser arrays,” Science Advances, vol. 7, no. 31, p. eabh3530, 2021. (Cited in page 7.) [43] K. Ji, W. Lin, Y. Sun, L.-S. Cui, J. Shamsi, Y.-H. Chiang, J. Chen, E. M. Tennyson, L. Dai, Q. Li, et al., “Self-supervised deep learning for tracking degradation of perovskite light-emitting diodes with multispectral imaging,” Nature Machine Intelligence, vol. 5, no. 11, pp. 1225–1235, 2023. (Cited in page 7.) [44] C. Bujalance, L. Cali`o, D. N. Dirin, D. O. Tiede, J. F. Galisteo-L´opez, J. Feist, F. J. Garc´ıa-Vidal, M. V. Kovalenko, and H. M´ıguez, “Strong light–matter coupling in lead halide perovskite quantum dot solids,” ACS nano, vol. 18, no. 6, pp. 4922–4931, 2024. (Cited in page 7.) [45] V. Castaing, M. Romero, J. Torres, G. Lozano, and H. M´ıguez, “Scattering spheres boost afterglow: A mie glass approach to go beyond the limits set by persistent phosphor composition,” Advanced Optical Materials, vol. 12, no. 3, p. 2301565, 2024. (Cited in pages 9, 18, and 37.) [46] J. Bertolotti, “Designing disorder,” Nature Photonics, vol. 12, no. 2, pp. 59– 60, 2018. (Cited in page 9.) [47] K. Guo, G. Lozano, M. A. Verschuuren, and J. G´omez Rivas, “Control of the external photoluminescent quantum yield of emitters coupled to nanoantenna phased arrays,” Journal of Applied Physics, vol. 118, no. 7, 2015. (Cited in pages 9 and 11.) [48] A. Jim´enez-Solano, J. F. Galisteo-L´opez, and H. M´ıguez, “Absorption and emission of light in optoelectronic nanomaterials: The role of the local optical environment,” The Journal of Physical Chemistry Letters, vol. 9, no. 8, pp. 2077–2084, 2018. (Cited in page 10.) |91 [107] D. Qin, Y. Xia, and G. M. Whitesides, “Soft lithography for micro-and nanoscale patterning,” Nature Protocols, vol. 5, no. 3, p. 491, 2010. (Cited in page 31.) [108] M. A. Verschuuren, M. Megens, Y. Ni, H. van Sprang, and A. Polman, “Large area nanoimprint by substrate conformal imprint lithography (SCIL),” Advanced Optical Technologies, vol. 6, no. 3-4, pp. 243–264, 2017. (Cited in page 31.) [109] S. Checcucci, T. Bottein, M. Gurioli, L. Favre, D. Grosso, and M. Abbarchi, “Multifunctional metasurfaces based on direct nanoimprint of titania sol–gel coatings,” Advanced Optical Materials, vol. 7, no. 10, p. 1801406, 2019. (Cited in page 31.) [110] M. Modaresialam, Z. Chehadi, T. Bottein, M. Abbarchi, and D. Grosso, “Nanoimprint lithography processing of inorganic-based materials,” Chemistry of Materials, vol. 33, no. 14, pp. 5464–5482, 2021. (Cited in page 31.) [111] P. Molet, L. K. Gil-Herrera, J. L. Garcia-Pomar, N. Caselli, ´ A. Blanco, C. L´opez, and A. Mihi, “Large area metasurfaces made with spherical silicon resonators,” Nanophotonics, vol. 9, no. 4, pp. 943–951, 2020. (Cited in page 31.) [112] S.-W. Baek, P. Molet, M.-J. Choi, M. Biondi, O. Ouellette, J. Fan, S. Hoogland, F. P. Garc´ıa de Arquer, A. Mihi, and E. H. Sargent, “Nanostructured back reflectors for efficient colloidal quantum-dot infrared optoelectronics,” Advanced Materials, vol. 31, no. 33, p. 1901745, 2019. (Cited in page 31.) [113] L. Scarabelli, D. Vila-Liarte, A. Mihi, and L. M. Liz-Marz´an, “Templated colloidal self-assembly for lattice plasmon engineering,” Accounts of Materials Research, vol. 2, no. 9, pp. 816–827, 2021. (Cited in page 31.) [114] Y. Conti, N. Passarelli, J. Mendoza-Carre˜no, L. Scarabelli, and A. Mihi, “Colloidal silver nanoparticle plasmonic arrays for versatile lasing architectures via template-assisted self-assembly,” Advanced Optical Materials, vol. 11, no. 23, p. 2300983, 2023. (Cited in page 31.) [115] J. M. Miranda-Mu˜noz, D. Geng, M. E. Calvo, G. Lozano, and H. M´ıguez, “Flexible nanophosphor films doped with mie resonators for enhanced outcoupling of the emission,” Journal of Materials Chemistry C, vol. 7, no. 2, pp. 267–274, 2019. (Cited in pages 31 and 37.) [116] M. Singh, H. M. Haverinen, P. Dhagat, and G. E. Jabbour, “Inkjet printingprocess and its applications,” Advanced Materials, vol. 22, no. 6, pp. 673–685, 2010. (Cited in page 35.) 98 | [117] M. You, J. Zhong, Y. Hong, Z. Duan, M. Lin, and F. Xu, “Inkjet printing of upconversion nanoparticles for anti-counterfeit applications,” Nanoscale, vol. 7, no. 10, pp. 4423–4431, 2015. (Cited in page 35.) [118] C. Sun, S. Su, Z. Gao, H. Liu, H. Wu, X. Shen, and W. Bi, “Stimuli-responsive inks based on perovskite quantum dots for advanced full-color information encryption and decryption,” ACS Applied Materials & Interfaces, vol. 11, no. 8, pp. 8210–8216, 2019. (Cited in page 35.) [119] K. Li, T. Li, T. Zhang, H. Li, A. Li, Z. Li, X. Lai, X. Hou, Y. Wang, L. Shi, et al., “Facile full-color printing with a single transparent ink,” Science Advances, vol. 7, no. 39, p. eabh1992, 2021. (Cited in page 35.) [120] Q. Zhang, Q. Jin, A. Mertens, C. Rainer, R. Huber, J. Fessler, G. Hernandez- Sosa, and U. Lemmer, “Fabrication of bragg mirrors by multilayer inkjet printing,” Advanced Materials, vol. 34, no. 33, p. 2201348, 2022. (Cited in page 35.) [121] Y. Liu, Q. Zhang, A. Huang, K. Zhang, S. Wan, H. Chen, Y. Fu, W. Zuo, Y. Wang, X. Cao, et al., “Fully inkjet-printed Ag2Se flexible thermoelectric devices for sustainable power generation,” Nature Communications, vol. 15, no. 1, p. 2141, 2024. (Cited in page 35.) [122] Q. Jin, Q. Zhang, C. Rainer, H. Hu, J. Chen, T. Gehring, J. Dycke, R. Singh, U. W. Paetzold, G. Hern´andez-Sosa, et al., “Inkjet-printed optical interference filters,” Nature Communications, vol. 15, no. 1, p. 3372, 2024. (Cited in page 35.) [123] P. Pei, R. Wei, B. Wang, J. Su, Z. Zhang, and W. Liu, “An advanced tunable multimodal luminescent La4GeO8: Eu2+, Er3+ phosphor for multicolor anticounterfeiting,” Advanced Functional Materials, vol. 31, no. 31, p. 2102479, 2021. (Cited in page 35.) [124] J. Xiahou, Q. Zhu, L. Zhu, S. Huang, T. Zhang, X. Sun, and J.-G. Li, “Lattice-site engineering in ZnGa2O4: Cr3+ through Li+doping for dynamic luminescence and advanced optical anti-counterfeiting,” Journal of Materials Chemistry C, vol. 10, no. 20, pp. 7935–7948, 2022. (Cited in page 35.) [125] S. D. Hoath, Fundamentals of inkjet printing: the science of inkjet and droplets. John Wiley & Sons, 2016. (Cited in page 35.) [126] G. J. Cordonier and K. A. Sierros, “Unconventional application of direct ink writing: Surface force-driven patterning of low viscosity inks,” ACS Applied Materials & Interfaces, vol. 12, no. 13, pp. 15875–15884, 2020. (Cited in page 35.) |99 [127] J. A. Lim, W. H. Lee, H. S. Lee, J. H. Lee, Y. D. Park, and K. Cho, “Self-organization of ink-jet-printed triisopropylsilylethynyl pentacene via evaporation-induced flows in a drying droplet,” Advanced Functional Materials, vol. 18, no. 2, pp. 229–234, 2008. (Cited in page 36.) [128] T. T. Ngo, E. Cabello-Olmo, E. Arroyo, A. I. Becerro, M. Oca˜na, G. Lozano, and H. M´ıguez, “Highly versatile upconverting oxyfluoride-based nanophosphor films,” ACS Applied Materials & Interfaces, vol. 13, no. 25, pp. 30051– 30060, 2021. (Cited in pages 37, 38, and 39.) [129] F. Auzel, “Upconversion and anti-stokes processes with f and d ions in solids,” Chemical reviews, vol. 104, no. 1, pp. 139–174, 2004. (Cited in page 38.) [130] S. Wen, J. Zhou, K. Zheng, A. Bednarkiewicz, X. Liu, and D. Jin, “Advances in highly doped upconversion nanoparticles,” Nature communications, vol. 9, no. 1, p. 2415, 2018. (Cited in page 38.) [131] S. Wilhelm, “Perspectives for upconverting nanoparticles,” ACS nano, vol. 11, no. 11, pp. 10644–10653, 2017. (Cited in page 38.) [132] J. Zhou, Q. Liu, W. Feng, Y. Sun, and F. Li, “Upconversion luminescent materials: advances and applications,” Chemical reviews, vol. 115, no. 1, pp. 395–465, 2015. (Cited in page 38.) [133] M. Gunaseelan, S. Yamini, G. Kumar, D. Sardar, and J. Senthilselvan, “Reverse microemulsion synthesis of mixed αand βphase NaYF4: Yb, Er nanoparticles: calcination induced phase formation, morphology, and upconversion emission,” Journal of Sol-Gel Science and Technology, vol. 96, pp. 550–563, 2020. (Cited in page 38.) [134] S. H. Nannuri, S. D. Kulkarni, C. Subash, S. Chidangil, and S. D. George, “Post annealing induced manipulation of phase and upconversion luminescence of Cr3+ doped NaYF4: Yb, Er crystals,” RSC advances, vol. 9, no. 17, pp. 9364–9372, 2019. (Cited in page 38.) [135] H. J. Ma, W. K. Jung, Y. Park, and D. K. Kim, “A novel approach of an infrared transparent Er: Y2O3-MgO nanocomposite for eye-safe laser ceramics,” Journal of Materials Chemistry C, vol. 6, no. 41, pp. 11096–11103, 2018. (Cited in page 38.) [136] E. E. Brown, U. H¨ommerich, A. Bluiett, C. Kucera, J. Ballato, and S. Trivedi, “Near-Infrared and upconversion luminescence in Er: Y2O3ceramics under 1.5 µm excitation,” Journal of the American Ceramic Society, vol. 97, no. 7, pp. 2105–2110, 2014. (Cited in page 38.) 100 | [137] S.-T. Zhang, M. Modreanu, H. Roussel, C. Jim´enez, and J.-L. Deschanvres, “Exploring the optical properties of Vernier phase yttrium oxyfluoride thin films grown by pulsed liquid injection MOCVD,” Dalton Transactions, vol. 47, no. 8, pp. 2655–2661, 2018. (Cited in page 38.) [138] L. Tao, W. Xu, Y. Zhu, L. Xu, H. Zhu, Y. Liu, S. Xu, P. Zhou, and H. Song, “Modulation of upconversion luminescence in Er3+, Yb3+-codoped lanthanide oxyfluoride (YOF, GdOF, LaOF) inverse opals,” Journal of Materials Chemistry C, vol. 2, no. 21, pp. 4186–4195, 2014. (Cited in pages 38 and 44.) [139] N. O. Nu˜nez, M. Quintanilla, E. Cantelar, F. Cuss´o, and M. Oca˜na, “Uniform YF3: Yb, Er up-conversion nanophosphors of various morphologies synthesized in polyol media through an ionic liquid,” Journal of Nanoparticle Research, vol. 12, pp. 2553–2565, 2010. (Cited in pages 39, 41, and 68.) [140] L. Sun, J. Pan, X. Zhang, H. Wang, L. Li, and Y. Yu, “Synthesis, morphology and spectroscopic properties of red-luminescent rhombohedral YOF: Yb3+, Er3+ powders,” RSC advances, vol. 5, no. 95, pp. 77673–77681, 2015. (Cited in pages 41 and 44.) [141] G. Chai, G. Dong, J. Qiu, Q. Zhang, and Z. Yang, “Phase transformation and intense 2.7 µm emission from Er3+ doped YF3/YOF submicron-crystals,” Scientific Reports, vol. 3, no. 1, p. 1598, 2013. (Cited in page 41.) [142] L. Tian, Y. Shang, S. Hao, Q. Han, T. Chen, W. Lv, and C. Yang, “Constructing a “native” oxyfluoride layer on fluoride particles for enhanced upconversion luminescence,” Advanced Functional Materials, vol. 28, no. 48, p. 1803946, 2018. (Cited in page 42.) [143] N. Rakov and G. S. Maciel, “Comparative study of Er3+ and Tm3+ co-doped YOF and Y2O3powders as red spectrally pure upconverters,” Optical Materials, vol. 35, no. 12, pp. 2372–2375, 2013. (Cited in page 42.) [144] J. Bergstrand, Q. Liu, B. Huang, X. Peng, C. W¨urth, U. Resch-Genger, Q. Zhan, J. Widengren, H. ˚ Agren, and H. Liu, “On the decay time of upconversion luminescence,” Nanoscale, vol. 11, no. 11, pp. 4959–4969, 2019. (Cited in page 43.) [145] H. Qin, D. Wu, J. Sathian, X. Xie, M. Ryan, and F. Xie, “Tuning the upconversion photoluminescence lifetimes of NaYF4: Yb3+, Er3+ through lanthanide Gd3+ doping,” Scientific reports, vol. 8, no. 1, p. 12683, 2018. (Cited in page 43.) [146] X. Wang, Y. Bu, S. Xiao, X. Yang, and J. Ding, “Upconversion in Ho3+- doped YbF3particle prepared by coprecipitation method,” Applied Physics B, vol. 93, pp. 801–807, 2008. (Cited in page 44.) |101 [147] R. Arppe, I. Hypp¨anen, N. Per¨al¨a, R. Peltomaa, M. Kaiser, C. W¨urth, S. Christ, U. Resch-Genger, M. Sch¨aferling, and T. Soukka, “Quenching of the upconversion luminescence of NaYF4: Yb3+, Er3+ and NaYF4: Yb3+, Tm3+ nanophosphors by water: the role of the sensitizer Yb3+ in non-radiative relaxation,” Nanoscale, vol. 7, no. 27, pp. 11746–11757, 2015. (Cited in page 45.) [148] G. Lozano, G. Grzela, M. Verschuuren, M. Ramezani, and J. G´omez Rivas, “Tailor-made directional emission in nanoimprinted plasmonic-based lightemitting devices,” Nanoscale, vol. 2014, pp. 9223–9229, 2014. (Cited in pages 48 and 60.) [149] M. Khaywah, A. Potdevin, F. R´everet, R. Mahiou, Y. Ouerdane, A. D´esert, S. Parola, G. Chadeyron, E. Centeno, R. Smaali, et al., “Large and versatile plasmonic enhancement of photoluminescence using colloidal metallic nanocubes,” The Journal of Physical Chemistry C, vol. 125, no. 14, pp. 7780– 7790, 2021. (Cited in page 48.) [150] H. Sun, A. Piquette, M. Raukas, and T. D. Moustakas, “Enhancement of yellow light extraction efficiency of Y3Al5O12:Ce3+ ceramic converters using a 2-D TiO2hexagonal-lattice nanocylinder photonic crystal layer,” IEEE Photonics Journal, vol. 8, no. 1, pp. 1–10, 2016. (Cited in page 48.) [151] B. Leng, Y. Zhang, D. P. Tsai, and S. Xiao, “Meta-device: advanced manufacturing,” Light: Advanced Manufacturing, vol. 5, no. 1, pp. 117–132, 2024. (Cited in page 48.) [152] S. Zhang, E. R. Martins, A. G. Diyaf, J. I. Wilson, G. A. Turnbull, and I. D. Samuel, “Calculation of the emission power distribution of microstructured OLEDs using the reciprocity theorem,” Synthetic Metals, vol. 205, pp. 127– 133, 2015. (Cited in pages 55 and 81.) [153] A. Vaskin, R. Kolkowski, A. F. Koenderink, and I. Staude, “Light-emitting metasurfaces,” Nanophotonics, vol. 8, no. 7, pp. 1151–1198, 2019. (Cited in page 55.) [154] M. R. Carro-Temboury, R. Arppe, T. Vosch, and T. J. Sørensen, “An optical authentication system based on imaging of excitation-selected lanthanide luminescence,” Science Advances, vol. 4, no. 1, p. e1701384, 2018. (Cited in page 59.) [155] A. David, H. Benisty, and C. Weisbuch, “Photonic crystal light-emitting sources,” Reports on Progress in Physics, vol. 75, no. 12, p. 126501, 2012. (Cited in page 60.) 102 | [156] G. Lozano, S. R. Rodriguez, M. A. Verschuuren, and J. G´omez Rivas, “Metallic nanostructures for efficient LED lighting,” Light: Science & Applications, vol. 5, no. 6, pp. e16080–e16080, 2016. (Cited in page 60.) [157] S. Zou, N. Janel, and G. C. Schatz, “Silver nanoparticle array structures that produce remarkably narrow plasmon lineshapes,” The Journal of Chemical Physics, vol. 120, no. 23, pp. 10871–10875, 2004. (Cited in page 60.) [158] V. Kravets, F. Schedin, and A. Grigorenko, “Extremely narrow plasmon resonances based on diffraction coupling of localized plasmons in arrays of metallic nanoparticles,” Physical Review Letters, vol. 101, no. 8, p. 087403, 2008. (Cited in page 60.) [159] B. Augui´e and W. L. Barnes, “Collective resonances in gold nanoparticle arrays,” Physical Review Letters, vol. 101, no. 14, p. 143902, 2008. (Cited in page 60.) [160] Y. Chu, E. Schonbrun, T. Yang, and K. B. Crozier, “Experimental observation of narrow surface plasmon resonances in gold nanoparticle arrays,” Applied Physics Letters, vol. 93, no. 18, 2008. (Cited in page 60.) [161] G. Lozano, D. Louwers, S. Rodr´ıguez, S. Murai, O. Jansen, M. Verschuuren, and J. G´omez Rivas, “Plasmonics for solid-state lighting: enhanced excitation and directional emission of highly efficient light sources,” Light: Science & Applications, vol. 2, p. e66, 2013. (Cited in page 60.) [162] J. Schuller, E. Barnard, W. Cai, Y. Jun, J. White, and M. Brongersma, “Plasmonics for extreme light concentration and manipulation,” Nature Materials, vol. 9, pp. 193–204, 2010. (Cited in page 60.) [163] M. Ramezani, G. Lozano, M. Verschuuren, and J. G´omez-Rivas, “Modified emission of extended light-emitting layers by selective coupling to collective lattice resonances,” Physical Review B, vol. 94, p. 125406, 2016. (Cited in page 60.) [164] G. Vecchi, V. Giannini, and J. G´omez-Rivas, “Shaping the fluorescent emission by lattice resonances in plasmonic crystals of nanoantennas,” Physical Review Letters, vol. 102, p. 146807, 2009. (Cited in page 60.) [165] L. Langguth, D. Punj, J. Wenger, and A. Koenderink, “Plasmonic band structure controls single-molecule fluorescence,” ACS Nano, vol. 7, pp. 8840–8848, 2013. (Cited in page 60.) [166] S. Rodriguez, G. Lozano, M. Verschuuren, R. Gomes, K. Lambert, B. De Geyter, A. Hassinen, D. Van Thourhout, Z. Hens, and J. G´omez Rivas, |103 “Quantum rod emission coupled to plasmonic lattice resonances: a collective directional source of polarized light,” Applied Physics Letters, vol. 100, p. 111103, 2012. (Cited in page 60.) [167] S. Murai, M. Verschuuren, G. Lozano, G. Pirruccio, S. Rodriguez, and J. G´omez Rivas, “Hybrid plasmonic-photonic modes in diffractive arrays of nanoparticles coupled to light-emitting optical waveguides,” Optics Express, vol. 21, pp. 4250–4262, 2013. (Cited in page 60.) [168] S. Murai, M. Saito, H. Sakamoto, M. Yamamoto, R. Kamakura, T. Nakanishi, K. Fujita, M. Verschuuren, Y. Hasegawa, and K. Tanaka, “Directional outcoupling of photoluminescence from Eu(III)-complex thin films by plasmonic array,” APL Photonics, vol. 2, p. 026104, 2017. (Cited in page 60.) [169] K. Agata, S. Murai, and K. Tanaka, “Stick-and-play metasurfaces for directional light outcoupling,” Applied Physics Letters, vol. 118, p. 021110, 2021. (Cited in page 60.) [170] Y. J. Kim, M. H. Lee, H. J. Kim, G. Lim, Y. S. Choi, N.-G. Park, K. Kim, and W. I. Lee, “Formation of highly efficient dye-sensitized solar cells by hierarchical pore generation with nanoporous TiO2spheres,” Advanced Materials, vol. 21, no. 36, pp. 3668–3673, 2009. (Cited in page 68.) [171] P. Yeh and M. Hendry, Optical waves in layered media. American Institute of Physics, 1990. (Cited in page 77.) [172] H. A. Macleod, Thin-film optical filters. CRC press, 2010. (Cited in page 77.) [173] R. Azzam and N. Bashara, Ellipsometry and Polarized Light. North-Holland Publishing Company, 1977. (Cited in page 77.) [174] J. D. Jackson, Classical electrodynamics. John Wiley & Sons, 2021. (Cited in page 77.) [175] F. Abeles, “Sur la propagation des ondes ´electromagn´etiques dans les milieux sratifi´es,” Annales de Physique, vol. 12, no. 3, pp. 504–520, 1948. (Cited in page 77.) [176] K. Ohta and H. Ishida, “Matrix formalism for calculation of electric field intensity of light in stratified multilayered films,” Applied Optics, vol. 29, no. 13, pp. 1952–1959, 1990. (Cited in pages 77 and 79.) [177] “Multifunctional Optical Materials Group GitHub.” https://github.com/ Multifunctional-Optical-Materials-Group. [Online; accessed July 2024]. (Cited in page 79.) 104 | [178] I. Moreels, K. Lambert, D. Smeets, D. De Muynck, T. Nollet, J. C. Martins, F. Vanhaecke, A. Vantomme, C. Delerue, G. Allan, et al., “Size-dependent optical properties of colloidal PbS quantum dots,” ACS Nano, vol. 3, no. 10, pp. 3023–3030, 2009. (Cited in page 80.) [179] D. D. Grinolds, P. R. Brown, D. K. Harris, V. Bulovic, and M. G. Bawendi, “Quantum-dot size and thin-film dielectric constant: precision measurement and disparity with simple models,” Nano Letters, vol. 15, no. 1, pp. 21–26, 2015. (Cited in page 80.) [180] A. Rubino, G. Lozano, M. E. Calvo, and H. M´ıguez, “Determination of the optical constants of ligand-free organic lead halide perovskite quantum dots,” Nanoscale, vol. 15, no. 6, pp. 2553–2560, 2023. (Cited in page 80.) [181] K. Yee, “Numerical solution of initial boundary value problems involving Maxwell’s equations in isotropic media,” IEEE Transactions on antennas and propagation, vol. 14, no. 3, pp. 302–307, 1966. (Cited in page 81.) |105 106 | Agradecimientos Quiero empezar agradeciendo a mis directores, Hern´an y Gabri, por darme la oportunidad de desarrollar esta tesis en el MOM. He aprendido mucho de vosotros, entre otras cosas, a disfrutar del no siempre f´acil camino que he recorrido hasta el final de esta tesis. Gracias tambi´en por apoyarme en las decisiones sobre qu´e hacer despu´es. Tambi´en quiero agradecer a mi tutor, Felipe Guti´errez, por la disponibilidad y atenci´on. Throughout these years I had the opportunity to collaborate with different researchers who have improved the work of my thesis, contributing in a considerable way to all the works. I would like to thank Agust´ın Mihi and Pau Molet for introducing me to soft lithography during the brief visit I made to ICMAB at the beginning of my thesis. Being able to apply those techniques with our materials has largely determined the development of my thesis and has allowed me to put it into practice later in more complex systems and with other materials. I would like to thank Elena Guill´en from PROFACTOR GmbH for helping me to organize my 3-month stay there, in Austria, under the supervision of Michael M¨uhlberger and Sonja Kopp. It was a great experience to be in a scientific-technical environment working on wet processing techniques with a lot of potential for scaling and functionalizing materials. Thanks to Michael Kainz and Anna Bernroitner for accompanying me until the last minute of the last day in the lab and helping me to get those photonic QRs out. Light bulb moments come from experience and the knowledge of the material properties. Also, thanks to Shunsuke Murai and his group from the Kyoto University, for allowing me to get closer to plasmonics with his metal arrays, first with the specular reflectance based porosimetry project, and then for providing me with his flexible arrays and facilitating the use of his RTA furnace. Finally, I would like to thank Miguel Anaya for hosting me at the University of Cambridge for 2 months and for teaching me how to use the hyperspectral microscope with our materials, it was a very enriching experience, even if those GBs of data did not appear in this thesis. 107