Precise control of light–matter interactions at the nanoscale lies at the heart of nanophotonics. However, experimental examination at this length scale is challenging since the corresponding electromagnetic near-field is often confined within volumes below the resolution of conventional optical microscopy. In semiconductor nanophotonics, electromagnetic fields are further restricted within the confines of individual subwavelength resonators, limiting access to critical light–matter interactions in these structures. In this work, we demonstrate that photoelectron emission microscopy (PEEM) can be used for polarization-resolved near-field spectroscopy and imaging of electromagnetic resonances supported by broken-symmetry silicon metasurfaces. We find that the photoemission results, enabled through an in situ potassium surface layer, are consistent with full-wave simulations and far-field reflectance measurements across visible and near-infrared wavelengths. In addition, we uncover a polarization-dependent evolution of collective resonances near the metasurface array edge taking advantage of the far-field excitation and full-field imaging of PEEM. Here, we deduce that coupling between eight resonators or more establishes the collective excitations of this metasurface. All told, we demonstrate that the high-spatial resolution hyperspectral imaging and far-field illumination of PEEM can be leveraged for the metrology of collective, non-local, optical resonances in semiconductor nanophotonic structures.

A recurring desire within nanophotonics is to engineer structures that confine light within a volume smaller than the wavelength of light. Such tight confinement of electromagnetic fields, along with the Purcell effect enhancing the photonic density of states, can greatly intensify light–matter interactions. Resonant structures, such as plasmonic antennas,1–3 metasurfaces,4–6 or photonic crystals,7,8 are known for their ability to manipulate electromagnetic fields efficiently within ultrasmall light–matter interaction volumes, leading to applications in sensing,9 imaging,10 holography,11 nonlinear,12 and quantum optics.13,14 While miniaturization has enabled control of the light–matter interactions with unprecedented precision, it has made metrology of such photonic structures more challenging, as the spatial resolution of conventional optical microscopy is inadequate for examination of the spatial field distribution.

In recent years, the development of semiconductor nanophotonic structures has granted access to enhanced light–matter interactions strong enough to elicit non-local (i.e., collective) phenomena and nonlinear optical responses.15,16 Light–matter interaction is, in general, polarization-dependent due to the anisotropic nature of the material response or dielectric permittivity, and in semiconductor metasurfaces, these characteristics are frequently enhanced through utilization of asymmetric structures. In this context, it is crucial that nanophotonics systems are investigated using optical illumination with well-defined resolutions in momentum and frequency. Conventional near-field microscopy using electron beam excitations (e.g., cathodoluminescence microscopy and electron energy loss microscopy17–21) are limited to indirect examinations of polarization-dependent optical phenomena. Whereas for scanning probe-type near-field microscopy,22–24 insertion of a proximal probe in the process of light illumination, or detection, obscures the well-defined polarization of light, in addition to inadvertently impacting the near-field interaction by the proximal probe itself. An approach to overcome such shortcomings is to employ far-field illumination, in which the basic characteristics of light (e.g., polarization, energy, and incident geometry) can be controlled within the scope of nanoscale imaging. One such technique that meets these criteria is photoemission electron microscopy (PEEM).

For more than a decade, PEEM has been employed to image the near-field distributions of metallic plasmonic nanostructures25–31 and, to a smaller extent, photonic resonances in a conductive oxide32 and excited states in organic molecules.33,34 However, one must acknowledge that fundamental differences between high-index semiconductor metasurfaces and their plasmonic counterparts make near-field imaging of the former non-trivial. Namely, that semiconductor metasurfaces typically confine electromagnetic fields within the volume of the individual resonators as opposed to plasmonic nanostructures, where electromagnetic fields are evanescent and concentrated at the metal–vacuum or metal–dielectric interface. Nonetheless, PEEM has been utilized in a few semiconductor nanostructure studies to examine localized absorption,35 multiphoton electron excitation, and hot electron dynamics.36,37 Altogether, the geometric constraints on electromagnetic fields, compounded with additional technical challenges that arise from narrower resonance linewidths (higher Q-factor), generally lower energy resonances (longer excitation wavelengths), and higher likelihood of sample charging have posed a significant barrier toward the utilization of PEEM for semiconductor metasurfaces. Each of these challenges is addressed throughout the work presented here.

In this work, we demonstrate polarization resolved spectroscopy and high-resolution imaging of the near-field distribution of optical resonances in a silicon (Si) metasurface using PEEM. The Si metasurface is designed with broken rotational symmetry granting access to quasi-bound states in the continuum (quasi-BICs) that exhibit tight confinement of electromagnetic fields within the body of the resonator, strong polarization dependence, and non-local resonances, rendering it an ideal archetype to examine the near-field imaging and spectroscopy of semiconductor nanophotonic structures.38–40 Two-photon photoemission under near-infrared excitation, achieved via a sub-monolayer of potassium (K) deposited in situ, enables nonlinear mapping of the near-field distribution of optical resonances by capturing photoelectron images under different wavelengths and polarizations. We show how the optical resonances evolve at the nanoscale as a function of spatial location within the metasurface, providing a detailed account of both the individual optical resonances within a given resonator and the emergence of their collective behavior in aggregate.15 These findings establish that the versatile hyperspectral near-field imaging capabilities of PEEM can facilitate detailed characterization of both local and non-local (i.e., collective) photonic resonances in semiconductor photonic structures.

Figure 1(a) shows the working concept of photoelectron emission microscopy (PEEM), in which a far-field tunable light source optically excites a sample generating photoelectrons that are collected and redirected onto a 2D imaging detector while preserving their spatial origin. Photoelectron imaging can capture near-field distributions of photonic resonances because photoelectron yield scales with electromagnetic field intensity, meaning that regions where electric fields are more concentrated will produce more photoelectrons, thus generating an image contrast.41 A photoelectron intensity image, therefore, reveals the spatial distribution of electromagnetic fields within the sample. This “photon in, electron out” approach offers numerous advantages: far-field excitation with light, akin to typical optical experimental conditions, provides fine control over the polarization, wavelength, incident angle of light, and is able to excite large nanostructures simultaneously,42,43 while electron imaging maintains a high spatial resolution given the small de Broglie wavelength of the photoelectrons.44 Multiple photoelectron intensity images can be sequentially acquired while scanning one of the far-field excitation conditions, which are then compiled into a “spectral hypercube” dataset. When the varied condition is the excitation wavelength, for example, the photoelectron yield spectra, analogous to far-field optical absorption spectra, can be extracted at each pixel of the PEEM spectral hypercube post eventum45 (see the Methods section and Supplementary note in the supplementary material for more information about the data acquisition and processing).

FIG. 1.

Schematic of the photoelectron emission microscope (PEEM). (a) Ti:sapphire oscillator (for two-photon photoelectron excitation) and deep UV laser (for topographical imaging, not depicted) are coupled to a photoelectron microscope. The wavelength tunability of the Ti:sapphire oscillator allows for acquiring a two-dimensional photoelectron yield map as a function of the incident light wavelength. (b) Energy diagram of various multiphoton processes for the photoemission and materials’ work functions involved in this work. (c) Scanning electron micrograph showing a few individual Si resonators that make up the larger metasurface; the scale bar is 250 nm.

FIG. 1.

Schematic of the photoelectron emission microscope (PEEM). (a) Ti:sapphire oscillator (for two-photon photoelectron excitation) and deep UV laser (for topographical imaging, not depicted) are coupled to a photoelectron microscope. The wavelength tunability of the Ti:sapphire oscillator allows for acquiring a two-dimensional photoelectron yield map as a function of the incident light wavelength. (b) Energy diagram of various multiphoton processes for the photoemission and materials’ work functions involved in this work. (c) Scanning electron micrograph showing a few individual Si resonators that make up the larger metasurface; the scale bar is 250 nm.

Close modal

Photoemission requires that the energy gained by an electron from absorbed photons exceeds the work function of the electron’s host material, as shown in Fig. 1(b). However, in semiconductor metasurfaces designed for optical frequencies [from 3 eV (400 nm) to 1.2 eV (1 µm)],46 the energies of the optical resonances are well below the work function of the resonator material. For this reason, here we rely on multiphoton absorption processes at the optical resonance energies to generate photoelectrons for PEEM imaging, which has been utilized previously with plasmonic nanostructures.28,42,43,47 In order to increase the efficiency of this nonlinear process, we deposit a sub-monolayer of potassium (K) onto the metasurface in situ to reduce the work function to ∼2.7 eV [see Fig. 1(b) and supplementary material, Fig. S1 for detail], thereby shifting the threshold for two-photon photoemission into the near-IR [∼920 nm (∼1.35 eV)]. Photoelectron intensity images generated in this regime can be regarded as nonlinear maps of the near-field distribution of the optical resonances because the intensity of two-photon photoemission is proportional to the square of the electromagnetic field intensity.25,28,47 The utilization of polarized far-field optical excitation and a large illumination spot (tens of micrometers, see supplementary material) with high spatial resolution and full field imaging enables PEEM to uniquely access collective and delocalized photonic resonances supported by a metasurface by simultaneously imaging many resonators.

To showcase this approach, we designed and fabricated an Si metasurface with broken symmetry nanoresonators arranged in periodic arrays [Fig. 1(c)]. The metasurface is isolated from a gold-coated Si substrate by a 50 nm dielectric layer stack and coated by a thin (10 nm) TiO2 layer to prevent sample charging during the PEEM measurements (see sample fabrication in the supplementary material and Fig. S2). This type of metasurface design exhibits narrowband quasi-BIC optical modes,38–40 which are enabled through symmetry breaking operations that permit outcoupling of symmetry-protected BICs to the far-field.48,49 Spectrally, they appear as sharp resonance peaks in the far-field reflectance measurements. We note that the TiO2 layer and sub-monolayer of K do perturb the optical properties of the metasurfaces to a minor extent, which will be described later. However, the quasi-BIC optical resonances supported by the metasurfaces are preserved. Due to their high-quality factors at normal incidence excitation, resulting in significant electromagnetic field concentrations,50 quasi-BICs are expected to result in significant enhancements of the photoemission process. Moreover, these modes are sensitive to input polarization51 and their resonances are collective modes: i.e., they depend on the location of the resonator within the entire resonator array that is being excited.15 Thus, these types of quasi-BICs supported by the Si-metasurface are ideally suited for investigation with PEEM.

To determine the near-field and far-field optical response of the Si metasurface, we first performed full-wave numerical simulations using far-field plane wave excitation. The simulated far-field reflectance spectra predict a series of modes spanning the visible to near-IR range for the fabricated Si metasurfaces that exhibit strong polarization dependencies. Shown as 1-reflectance (dashed line) in Fig. 2(b), these spectra show the different responses for normal incidence excitation at the four polarization orientations (0°, 45°, 90°, and 135°) shown in Fig. 2(a). Within these spectra, the two lowest energy modes correspond to quasi-BICs originating from out-of-plane magnetic and electric dipoles, respectively, while additional higher order modes extend toward larger energies. Importantly, each of the four incident polarization orientations is predicted to elicit a distinct spectral profile, where the relative intensity of the modes varies greatly while retaining roughly the same spectral position, providing an ideal system for comparison across far-field and PEEM techniques.

FIG. 2.

Comparison of metasurface absorption and photoelectron yield spectra at normal incident optical excitation. (a) Schematic of the polarization orientation of the far-field excitation source with respect to the resonator geometry. (b) 1-reflectance spectra of the fabricated Si metasurface calculated using full-wave electromagnetic simulations (COMSOL). The dashed spectra depict the initial simulated spectra, while the solid spectra are the result of broadening with a Gaussian convolution (FWHM 17.5 meV or ∼9 nm). The vertical colored dashed lines indicate the general position of each broadened resonance and are labeled A to E to correlate with the far-field reflectance as discussed in the main text. (c) The measured 1-reflectance spectra of the Si metasurface measured with a home-built reflectance set-up. The sample used for this measurement was the same Si metasurface coated with TiO2 and exposed to K during the PEEM measurements. Before the reflectance measurements, the sample was cleaned by annealing under UHV conditions to remove K. (d) Photoelectron yield spectra averaged over many unit-cells measured with PEEM. The spectra in panels (b), (c), and (d) are presented for the four incident polarization orientations depicted in (a).

FIG. 2.

Comparison of metasurface absorption and photoelectron yield spectra at normal incident optical excitation. (a) Schematic of the polarization orientation of the far-field excitation source with respect to the resonator geometry. (b) 1-reflectance spectra of the fabricated Si metasurface calculated using full-wave electromagnetic simulations (COMSOL). The dashed spectra depict the initial simulated spectra, while the solid spectra are the result of broadening with a Gaussian convolution (FWHM 17.5 meV or ∼9 nm). The vertical colored dashed lines indicate the general position of each broadened resonance and are labeled A to E to correlate with the far-field reflectance as discussed in the main text. (c) The measured 1-reflectance spectra of the Si metasurface measured with a home-built reflectance set-up. The sample used for this measurement was the same Si metasurface coated with TiO2 and exposed to K during the PEEM measurements. Before the reflectance measurements, the sample was cleaned by annealing under UHV conditions to remove K. (d) Photoelectron yield spectra averaged over many unit-cells measured with PEEM. The spectra in panels (b), (c), and (d) are presented for the four incident polarization orientations depicted in (a).

Close modal

The experimental far-field reflectance spectra shown in Fig. 2(c) exhibit five resonances spanning the visible to near-IR spectral range, which can be mapped to the modes predicted from simulations. These five resonance peaks are labeled from left to right with letters A–E and are fit with a Lorentzian peak shape using least-squares fitting for easier visualization. We note that the reflectance spectra are acquired from an Si metasurface with a 10 nm TiO2 layer, but without K. To map these resonance peaks to the modes predicted in the simulated spectra, the latter is convoluted with a Gaussian function to represent experimental broadening that originates from imperfect dimensionality of individual resonators, slanted sidewalls, and dimensional variation between the resonator units, as well as instrumental broadening from off-normal incidence optical illumination. The full width at half maximum (FWHM) of the Gaussian function was set to the minimum value (FWHM = 17.5 meV, ∼9 nm) required to achieve the same number of final peaks (five), and the resulting broadened simulated spectra are shown as solid lines in Fig. 2(b). From here, the five broadened simulated peaks were assigned one-to-one, with the experimental resonance peaks starting from the lowest energy peak using the A–E labels.

Following this scheme, we find good matching between the simulation and experiment for the energy, spacing, and polarization response of the resonance peaks. In the experimental reflectance spectra, resonances are blueshifted by an average of 50 meV (∼25 nm across this wavelength range); however, the total energy separation (A–E) remains constant (245 vs 275 meV for simulated and far-field reflectance, respectively) supporting these mode designations. Further corroboration of these assignments is found in the polarization response, where the spectral profiles for the experimentally acquired reflectance spectra exhibit clear polarization dependencies that match well with the predictions. For example, the A-resonance peak, which is mapped to the lowest energy peak from the simulation, consisting of a single, well-isolated mode (out-of-plane magnetic dipole), exhibits the precise polarization dependency predicted from the simulation with strong coupling to the far-field at 45° and 90° polarization orientations, but significant suppression at 0° and 135°. Additional polarization responses predicted from simulation can be observed for the other resonance peaks; however, they are less apparent as the B and C resonances consist of multiple overlapping modes, and the highest energy (D- and E-) resonances, due to their larger quality factors and, therefore, field enhancements, are more susceptible to losses. One clear discrepancy is the suppression of the B resonance at 45°, which is not predicted from the simulated spectra; as this same discrepancy is noted in the photoelectron yield spectra, less attention is paid in the following discussions.

Photoelectron yield spectra spanning the same visible to near-IR wavelength range captures the four lowest energy resonances (A–D) while retaining the wavelength and polarization responses noted in the simulated and far-field reflectance spectra. Photoelectron yield spectra are extracted after in situ K deposition from a series of photoelectron images collected sequentially while varying the near-IR excitation wavelength at a fixed polarization orientation [Fig. 2(d)]. Photoelectron yield, or intensity, is integrated over all electron kinetic energies. More information on PEEM data processing is provided in the supplementary material, note analysis of PEEM data. We find that the photoelectron yield spectra correlate well with the simulated and experimental spectra, exhibiting the majority of anticipated wavelength and polarization dependences. For instance, at a 135° polarization orientation, we observe an enhancement of the B and D resonances and suppression of the A and C resonances, consistent with spectral profiles for the far-field reflectance [Fig. 2(c)] and simulation [Fig. 2(b)]. Similarly, at a 90° incident, polarization orientation all resonances are strongly expressed, which is again expected. Similar to the far-field reflectance measurements, discrepancies from the simulated spectra are primarily attributed to limitations in fabrication tolerance and a component of the optical excitation that is off-normal incidence, in this case arising from the tilt of the sample for alignment with the electron optics (<1°). Interestingly, at a 45° polarization orientation, the B resonance is much weaker than the neighboring A and C resonances, as noted previously in the far-field reflectance measurements, but a clear deviation from the simulated spectra. This consistency between the experimental methods suggests that photoelectron yield imaging appropriately captures the Si metasurface response under an equivalent far-field optical excitation.

There are some notable differences between the photoelectron yield spectra and reflectance spectra. First, the peak positions in the photoelectron yield spectra are blueshifted by ∼35 meV, or ∼2% of the excitation energy. This slight blueshift is expected due to the static charge accumulation at the surface of the TiO2 layer from K deposition. An atomic layer of K is known to transfer electrons to a substrate on the order of 1013 cm−2 (corresponding to 1019−20 cm−3), which explains the significant reduction in the work function achieved.52 Such high near-surface electron accumulation would also lower the refractive index and hence, induce a sizable blueshift of the resonant mode frequencies.53 This concept is verified through simulation with and without a K surface layer (see the supplementary material, Fig. S3). This blueshift also explains the absence of the E-resonance in the photoelectron yield spectra as it now lies outside the attainable wavelength range of the light source. We also note some qualitative differences in the peak shape of the photoelectron spectra with respect to the measured reflectance spectra, namely, broadening of the respective linewidths (from a 20–30 meV FWHM in reflectance to a 20–50 meV in PEEM) and an asymmetric peak shape tailing toward lower energies. We postulate that the surface charge accumulation also contributes to these changes in the peak shape, as the electrons accumulated near the surface would induce band bending in the TiO2 film leading to a gradual reduction of the blueshift as a function of distance from the surface. Such asymmetry is captured in the spectral fitting by employing two Lorentzian peaks in the resonances that show obvious tailing. The larger photoelectron linewidths may also, in part, come from the broad spectral width of the Ti:sapphire laser (8–14 meV across the excitation wavelength range used) and the variability in the size and shape of the resonators. Overall, we find a reasonable agreement between the measured far-field absorption and the photoelectron yield spectra, affirming the polarization resolved near-field spectroscopy capabilities of PEEM for semiconductor metasurfaces.

Next we demonstrate real-space imaging of the resonances based on a PEEM imaging scheme. Figures 3(a) and 3(b) show large scale photoelectron intensity images containing hundreds of individual resonators located near the center of the fabricated Si metasurface array acquired under optical excitation with deep UV (5.82 eV) and near-IR (1.52 eV) light. Under deep UV laser excitation [Fig. 3(a)], the shapes of resonators are rendered well (compare to the SEM image shown in Fig. 1(c) and resonator design in Fig. S2). However, when the excitation is switched to the near-IR, the photoelectron intensity image instead reveals the location of enhanced electric fields for this resonance (A-resonance at 135° polarization orientation). Using the deep UV PEEM image and the measured resonator dimensions from SEM, the borders of a 5 × 5 subsection of resonators are outlined in white [Fig. 3(a)]. Overlaying these resonator positions within the NIR image [Fig. 3(b)] reveals that the enhanced photoemission and, therefore, near-field electric fields, are confined within the resonator volume as expected for these semiconducting metasurfaces. This field confinement is further verified through the co-illumination experiments provided in the supplementary material, Fig. S4.

FIG. 3.

Photoelectron intensity confinement via energy-dependent nanoscale mode imaging. (a) and (b) Large field view of photoelectron intensity images of the metasurface array imaged using deep ultraviolet (5.82 eV), (a) and A-resonance (1.52 eV), and (b) excitation energies. The boundaries for a 5 × 5 subset of the array are outlined in white in panel (a) and transposed to panel (b) to show the confinement of photoelectron intensity within the resonator boundaries. Yellow circles in panel (a) identify defective resonators from fabrication, which are found in panel (b) to be inactive. Abnormal high intensity spots in panel (b) are identified with red circles that may arise from additional fabrication defects or particle formation. (c)–(f) Resonator averaged photoelectron intensity images at resonant excitation energies identified for 135° polarization orientation, 1.52 eV (c), 1.62 eV (d), 1.69 eV (e), and 1.73 eV (f). Polarization orientation is indicated by the white double arrows in (b)–(f).

FIG. 3.

Photoelectron intensity confinement via energy-dependent nanoscale mode imaging. (a) and (b) Large field view of photoelectron intensity images of the metasurface array imaged using deep ultraviolet (5.82 eV), (a) and A-resonance (1.52 eV), and (b) excitation energies. The boundaries for a 5 × 5 subset of the array are outlined in white in panel (a) and transposed to panel (b) to show the confinement of photoelectron intensity within the resonator boundaries. Yellow circles in panel (a) identify defective resonators from fabrication, which are found in panel (b) to be inactive. Abnormal high intensity spots in panel (b) are identified with red circles that may arise from additional fabrication defects or particle formation. (c)–(f) Resonator averaged photoelectron intensity images at resonant excitation energies identified for 135° polarization orientation, 1.52 eV (c), 1.62 eV (d), 1.69 eV (e), and 1.73 eV (f). Polarization orientation is indicated by the white double arrows in (b)–(f).

Close modal

We also find significant resonator to resonator variations in the photoelectron intensity image appearing as additional high intensity spots and “inactive” resonators. A few examples are highlighted by the red and yellow circles, respectively, shown in Figs. 3(a) and 3(b). The shape of the high intensity spots shown in Fig. 3(b) resembles those of the silver nanoparticles imaged with PEEM under resonant conditions in Ref. 25 that are uncontrollably agglomerated through the metal deposition process. This observation leads us to speculate that the high intensity resonance spots arise from imperfect side walls of the resonators due to the limited fabrication precision or due to clustered TiO2 on resonator surfaces. As such, PEEM can also serve as a tool to diagnose heterogeneities and defective resonators within a metasurface through direct light–matter interactions and improve the engineering aspect of metasurface fabrication.

High-resolution imaging of the photoelectron intensity distributions at different excitation wavelengths demonstrates the sensitivity of this approach to the near-field electric field profiles of the different modes. Figures 3(c)3(f) display the averaged photoelectron intensity images for a single resonator unit under excitation at the four resonance wavelengths shown in Fig. 2(d) for 135° polarization orientation. These images are created by superimposing 25 resonators within a single image to account for fabrication variations and improve statistics. Here, we observe that the near-field photoelectron intensity distribution is sensitive to excitation energy. For the A and B resonances, the near-field distributions are highly symmetric as expected for the out-of-plane magnetic and electronic dipole modes, respectively. Meanwhile, the C and D resonances display non-symmetric and highly structured near-field distributions, as expected given the higher order of the modes that make up these resonances. In addition to excitation wavelength, photoelectron intensity images also capture the polarization dependence of the optical modes and match well with full-wave simulations, which will be discussed in the following paragraphs.

Enabled by the far-field optical excitation, polarization resolved near-field photoelectron images are acquired for the resonances supported by this Si metasurface and found to match well with simulated profiles of the square of the electric field intensity (|E|4). Near-field photoelectron intensity images obtained at the four resonant wavelengths (A–D) for a 0° polarization orientation are shown in Figs. 4(a)4(d), averaged over at least 20 individual resonators. In addition, Video S1 in the supplementary material displays the complete PEEM spectral hypercube for a 0° polarization. Here, each resonant wavelength generates a distinct spatial distribution with an enhancement in the photoelectron intensity of 3–5 times higher than the area outside the resonator boundaries, demonstrating the near-field enhancement of two-photon photoemission that arises from the optical resonances. Importantly, these spatial distributions are visually dissimilar from those obtained at a 135° polarization orientation [Figs. 3(c)3(f)] exemplifying the capability to achieve polarization resolved near-field imaging through far-field excitation.

FIG. 4.

Excitation energy dependent and polarization resolved nanoscale mode imaging and simulated electromagnetic field profiles of resonators. (a)–(d) Resonator averaged photoelectron intensity images at resonant excitation energies for a 0° polarization orientation, 1.52 eV (a), 1.62 eV (b), 1.67 eV (c), and 1.73 eV (d). (e)–(h) Simulated profiles of the square of the electric field intensity (|E|4) for the corresponding modes at a 0° polarization orientation. (i)–(l) Resonator averaged photoelectron intensity images at resonant excitation energies for a 45° polarization orientation, 1.53 eV (i), 1.61 eV (j), 1.67 eV (k), and 1.75 eV (l). (m)–(p) Simulated profiles of the square of the electric field intensity (|E|4) for the corresponding modes at 45° polarization orientation. Polarization orientations are also indicated by the white double arrows.

FIG. 4.

Excitation energy dependent and polarization resolved nanoscale mode imaging and simulated electromagnetic field profiles of resonators. (a)–(d) Resonator averaged photoelectron intensity images at resonant excitation energies for a 0° polarization orientation, 1.52 eV (a), 1.62 eV (b), 1.67 eV (c), and 1.73 eV (d). (e)–(h) Simulated profiles of the square of the electric field intensity (|E|4) for the corresponding modes at a 0° polarization orientation. (i)–(l) Resonator averaged photoelectron intensity images at resonant excitation energies for a 45° polarization orientation, 1.53 eV (i), 1.61 eV (j), 1.67 eV (k), and 1.75 eV (l). (m)–(p) Simulated profiles of the square of the electric field intensity (|E|4) for the corresponding modes at 45° polarization orientation. Polarization orientations are also indicated by the white double arrows.

Close modal

Moreover, at a 0° polarization orientation, each resonance observed experimentally for the fabricated Si metasurface can be mapped to a singular mode from simulation (Fig. 2). As a result, the photoelectron intensity distributions shown in Figs. 4(a)4(d) are found to recreate the simulated profiles of the square of the electric field intensity for each of these individual modes exceptionally well [Figs. 4(e)4(h)]. The simulated field profiles shown in Figs. 4(e)4(h) are the 2D planes extracted from the middle height of a resonator in an infinite array under excitation at a 0° polarization orientation; regions outside the resonator boundaries are set to zero. Photoelectron intensity images for the quasi-BIC A and B resonances [Figs. 4(a) and 4(b)] exhibit spherical distributions that closely resemble the fields expected from the simulation [Figs. 4(e) and 4(f)], which places these enhancements near the center of the resonator unit. Extending to the higher order modes [C and D resonances, Figs. 4(c) and 4(d)], we observe an extension and then splitting of the photoelectron intensity toward the upper right portion of the resonator, a trend that is well captured in the corresponding simulated field distributions [Figs. 4(g) and 4(h)]. Generally, the photoelectron intensity distributions appear spatially broadened relative to the simulated field enhancements, and some portions of fine structure are not captured, such as the splitting into three high intensity regions for the highest energy D-mode. Such discrepancies are likely due to imperfections in the fabricated resonator geometries and array spacing, in addition to the broader linewidth of the excitation source in the experiment. Nonetheless, we capture high-resolution near-field images through photoelectron imaging that are sensitive to wavelength and are well correlated with the expected electric field enhancements from full-wave simulations.

Near-field photoelectron intensity imaging at a 45° polarization orientation for the same four resonances (A–D) reveals further distinct distributions that well recreate the expected field enhancements, despite the overlapping modes of some of the resonances. As shown in Figs. 4(i) and 4(m), the photoelectron intensity distribution at the lowest energy A resonance is spherical in nature and located near the center of the resonator unit. However, when increasing the excitation energy, there is a clear elongation and splitting of the photoelectron intensity along the polarization axis [Figs. 4(j)4(l)]. This evolution in field distribution is consistent with the simulated fields for the modes corresponding to these resonances [Fig. 4(m)4(p)]. We note that the B and C resonances at a 45° polarization orientation [Figs. 4(j) and 4(k)] consist of two overlapping modes (Fig. 2). The simulated field distributions shown in Figs. 4(n) and 4(o) for these resonances are the single mode profiles that best match each photoelectron image, although the latter is likely a mixture of the two overlapping modes. Photoelectron images and the corresponding simulated profiles of the square of the electric field intensity for all the resonance and modes are given in the supplementary material, Figs. S5–S8, sorted by incident polarization orientation. Comparison of the photoelectron images and simulated field profiles at 90° and 135° polarization angles (Figs. S7 and S8) reveals poorer matching than for 0° and 45° (Figs. 4, S5, and S6) due to an increase in the number of excited modes. In the former, the poorer matching results from the fact that the field distribution will depend on the coherent superposition, resulting from interference between these simultaneously excited modes. Therefore, the overall fields will depend on the relative amplitudes, Q-factors, and relative phases of the modes being excited, which is extremely difficult to predict using simulations. Altogether, we demonstrate high-resolution polarization resolved near-field imaging through photoelectron microscopy and find reasonable agreement with the electric field enhancements from full-wave simulations.

Photoemission processes are typically regarded as surface sensitive since the inelastic mean free path (IMFP) of photoelectrons in most materials is expected to be on the order of a few nanometers.54,55 Although, in the case of low energy electrons (<10 eV), as in this work, IMFPs of 10–100 nm have been obtained experimentally for Au and Al2O3, and predicted from theory for many more elements.56–59 Nevertheless, the number of escaping electrons is expected to decay exponentially as a function of depth, and thus, we expect the PEEM signal to originate primarily from the TiO2 layer or sub-monolayer K surface layer. Even so, the photoelectron intensity images shown in Fig. 4 correlate well with the simulated profiles of the square of the electric field intensity extracted as 2D cross sections at the middle height of a Si resonator. A degree of sensitivity along this z direction may arise from photoelectron emission from the sidewalls of the resonators, as these electrons could escape to vacuum without traversing the full resonator height. A more probable interpretation is that the field enhancements within the resonator can excite photoelectrons from the surface layers, thereby eliciting photoelectron contrasts from light–matter interactions occurring at depths well beneath where the photoelectrons themselves are emitted. Previous studies have demonstrated this depth sensitivity beyond typical photoelectron escape lengths arising from similar phenomena in buried dielectric interfaces.60,61 Finally, we note that the nanostructures in this study are ill-suited for detailed examination of z sensitivity given that the resonator height is on the order of half the excitation wavelength and that future studies on taller nanostructures (height > λ) designed for more structured variation of the fields along the z axis would be more appropriate.

Thus far, all photoelectron spectra and images have originated from regions within the resonator array distant from array edges as the photonic modes studied in this work are collective modes (i.e., involve neighboring resonators). However, we anticipate that at the boundaries of the array, the confinement of the electromagnetic field will vary as a function of distance from the edge until the resonances are fully established.15 To examine this collective nature, we investigate the evolution of the photoelectron yield spectra over the first few resonators near the array edge.

Within this regime, we observe an enhancement of the quasi-BIC containing B resonance in the outermost resonators, followed by a polarization-dependent quenching of the second column/row of the array. Figure 5(a) shows a photoelectron image collected at the upper-left corner of a large resonator array under excitation at the B resonance wavelength with a 90° polarization angle [supplementary material, Fig. S9 shows the same region imaged under DUV (5.82 eV) excitation]. The resonators making up the outermost border of the array (R1 and C1) exhibit significantly enhanced photoelectron intensity relative to their interior counterparts. We interpret this enhancement as a weakening of the quasi-BIC confinement due to lack of neighboring elements resulting in better in-coupling of light to these edge resonators. Moreover, we note a strong suppression of the B-resonance photoelectron intensity in the second column (C2) that is quickly recovered in the third column (C3). This suppression is not a consequence of defective fabrication, but rather a unique interaction occurring within the near-edge region of the resonator array parallel to the incident polarization. For when the incident light polarization is rotated parallel to the top edge of the array (0° polarization), the resonators in C2 exhibit comparable photoelectron intensity to their interior counterparts, and instead the resonators in R2 become suppressed, as shown in Fig. S10(a). The enhancement, suppression, and then recovery of the B-resonance photoelectron intensity across the near edge of the array (C1–3) can also be visualized by plotting the photoelectron yield spectra as a function of the column number (distance from the left edge), as shown in Fig. 5(b). Through fits to these spectra, we ascertain that while the described near-edge evolution is unique to the B resonance, the remaining resonances do undergo more gradual changes, with all trending toward bulk as the distance from the array edge increases.

FIG. 5.

Evolution of the coherent interactions from the edges to the middle of the resonator array. (a) Photoelectron image of the upper-left corner of a large metasurface array acquired at 1.62 eV excitation (B resonance) and a 90° polarization orientation (indicated by the white double arrow). The rows and columns are designated as R1–R12 and C1–C10, respectively, within the image. The supplementary material, Fig. S9, shows the same region imaged under DUV (5.82 eV) excitation for better viewing of the array corner. (b) The area averaged spectra of the resonators as a function of the distance from the left array edge (column number). The spectra are averaged for four unit cells along the vertical direction beginning at R8, as indicated by the yellow, red, and gray doted rectangles. (c) and (d) The relative peak area for each resonance as a function of the column (c) or row (d) number (i.e., distance from left or top array edge, respectively). The relative peak areas in panel (d) come from fits to the spectra averaged over three unit-cells along the horizontal direction beginning at C8, as indicated in panel (a). Relative peak areas from bulk come from the spectral fits shown in Fig. 2(d). The error bars correspond to the standard deviation observed for each resonance across the bulk region averaged for Fig. 2(d).

FIG. 5.

Evolution of the coherent interactions from the edges to the middle of the resonator array. (a) Photoelectron image of the upper-left corner of a large metasurface array acquired at 1.62 eV excitation (B resonance) and a 90° polarization orientation (indicated by the white double arrow). The rows and columns are designated as R1–R12 and C1–C10, respectively, within the image. The supplementary material, Fig. S9, shows the same region imaged under DUV (5.82 eV) excitation for better viewing of the array corner. (b) The area averaged spectra of the resonators as a function of the distance from the left array edge (column number). The spectra are averaged for four unit cells along the vertical direction beginning at R8, as indicated by the yellow, red, and gray doted rectangles. (c) and (d) The relative peak area for each resonance as a function of the column (c) or row (d) number (i.e., distance from left or top array edge, respectively). The relative peak areas in panel (d) come from fits to the spectra averaged over three unit-cells along the horizontal direction beginning at C8, as indicated in panel (a). Relative peak areas from bulk come from the spectral fits shown in Fig. 2(d). The error bars correspond to the standard deviation observed for each resonance across the bulk region averaged for Fig. 2(d).

Close modal

From the evolution of the photoelectron yield spectra and relative peak areas across this range, we conclude that approximately eight resonators, in both propagation directions, are necessary to fully establish the collective array modes, corroborating the result of previous studies. 15,20 This is shown in Fig. 5(b) when the photoelectron yield spectra (and corresponding fits) continue to evolve beyond C3 before converging after about eight resonators (C8), where they closely resemble the bulk spectra [Fig. 2(d)]. This trend is more clearly visualized when the relative peak areas for each resonance are plotted as a function of the column number [Fig. 5(c)]. Here, the dramatic changes across the near-edge region (C1–3) quickly give way to more consistent trends spanning C4–7 (termed the “transition” region) before stabilizing at C8 and beyond (termed the “bulk” region) at comparable values to the bulk shown in Fig. 2(d). Propagating away from the top array edge (i.e., row number) shown in Fig. 5(d) demonstrates the same general trends, that is, significant variation across the near-edge region, more gradual changes within the transition region, and stabilization after eight resonators. Evolution of the photoelectron yield spectra and relative peak areas for the same region under illumination at a 0° polarization orientation also demonstrate the trends noted here (Fig. S10). These data clearly support the notion that the excited photonic modes of the resonators near the array edges are different from the photonic modes in the bulk of the metasurface (i.e., the center of the array). By capturing photoelectron yield spectra at each point of the image, PEEM enables us to draw such conclusions and study how these collective optical resonances form within the metasurfaces under far-field optical excitations.

In summary, we have presented polarization resolved concurrent spectroscopy and high-resolution imaging of the near-field distribution of photonic resonances in an Si metasurface using photoelectron emission microscopy (PEEM). By capitalizing on far-field optical excitation, we gain insights into the anisotropic nature of light–matter interactions in semiconductor nanostructures in the form of polarization-dependent collective modes in broken-symmetry Si metasurfaces. Upon simultaneously exciting and imaging many resonators, we uncover an unanticipated polarization-dependent field enhancement switching near the edges of a larger metasurface array and that coupling between eight resonators is needed to establish quasi-BIC resonances in this Si metasurface. In expanding the application space of PEEM beyond plasmonic nanostructures, we establish a powerful route for the metrology of local and non-local photonic resonances in semiconductor nanostructures.

We thank P. Mantos, A. Jarzembski, and G. Copeland for their measurement support and A. Cerjan for discussions. A.B., T.E.B., and T.O. acknowledge support from the Laboratory Directed Research and Development program at Sandia National Laboratories. S.D.G., C.F.D., M.B.S., I.B., and R.S. acknowledge support from the U.S. Department of Energy, Office of Basic Energy Sciences, Division of Materials Sciences and Engineering. This work was performed in part at the Center for Integrated Nanotechnologies, an Office of Science User Facility operated for the U.S. Department of Energy (DOE) Office of Science. Sandia National Laboratories is a multimission laboratory managed and operated by the National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under Contract No. DE-NA0003525. This paper describes objective technical results and analysis. Any subjective views or opinions that might be expressed in the paper do not necessarily represent the views of the U.S. Department of Energy or the United States Government.

This article has been authored by an employee of National Technology and Engineering Solutions of Sandia, LLC under Contract No. DE-NA0003525 with the U.S. Department of Energy (DOE). The employee owns all right, title, and interest in and to the article and is solely responsible for its contents. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this article or allow others to do so, for United States Government purposes. The DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan https://www.energy.gov/downloads/doe-public-access-plan.

The authors have no conflicts to disclose.

Alex Boehm: Data curation (lead); Formal analysis (lead); Investigation (lead); Methodology (equal); Software (supporting); Visualization (lead); Writing – original draft (lead); Writing – review & editing (lead). Sylvain D. Gennaro: Conceptualization (supporting); Formal analysis (supporting); Investigation (supporting); Methodology (supporting); Resources (equal); Software (supporting); Writing – original draft (supporting). Chloe F. Doiron: Conceptualization (supporting); Data curation (supporting); Formal analysis (equal); Investigation (equal); Methodology (equal); Resources (supporting); Software (lead); Writing – original draft (supporting); Writing – review & editing (supporting). Thomas E. Beechem: Conceptualization (equal); Funding acquisition (equal); Project administration (supporting); Supervision (supporting); Writing – original draft (supporting); Writing – review & editing (supporting). Michael B. Sinclair: Conceptualization (equal); Funding acquisition (equal); Project administration (supporting); Resources (supporting); Supervision (supporting). Igal Brener: Conceptualization (supporting); Funding acquisition (equal); Project administration (supporting); Resources (supporting); Supervision (supporting); Writing – original draft (supporting); Writing – review & editing (supporting). Raktim Sarma: Conceptualization (equal); Funding acquisition (supporting); Project administration (supporting); Software (supporting); Writing – original draft (equal); Writing – review & editing (supporting). Taisuke Ohta: Conceptualization (lead); Data curation (supporting); Formal analysis (equal); Funding acquisition (lead); Investigation (supporting); Methodology (equal); Project administration (lead); Resources (equal); Software (equal); Supervision (lead); Writing – original draft (equal); Writing – review & editing (equal).

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

1.
L.
Novotny
,
E. J.
Sánchez
, and
X.
Sunney Xie
, “
Near-field optical imaging using metal tips illuminated by higher-order Hermite–Gaussian beams
,”
Ultramicroscopy
71
(
1–4
),
21
29
(
1998
).
2.
P.
Mühlschlegel
,
H.-J.
Eisler
,
O. J. F.
Martin
,
B.
Hecht
, and
D. W.
Pohl
, “
Resonant optical antennas
,”
Science
308
(
5728
),
1607
1609
(
2005
).
3.
P.
Bharadwaj
,
B.
Deutsch
, and
L.
Novotny
, “
Optical antennas
,”
Adv. Opt. Photonics
1
(
3
),
438
483
(
2009
).
4.
C. L.
Holloway
,
E. F.
Kuester
,
J. A.
Gordon
,
J.
O’Hara
,
J.
Booth
, and
D. R.
Smith
, “
An overview of the theory and applications of metasurfaces: The two-dimensional equivalents of metamaterials
,”
IEEE Antennas Propag. Mag.
54
(
2
),
10
35
(
2012
).
5.
J. C.
Ginn
,
I.
Brener
,
D. W.
Peters
,
J. R.
Wendt
,
J. O.
Stevens
,
P. F.
Hines
,
L. I.
Basilio
,
L. K.
Warne
,
J. F.
Ihlefeld
,
P. G.
Clem
, and
M. B.
Sinclair
, “
Realizing optical magnetism from dielectric metamaterials
,”
Phys. Rev. Lett.
108
(
9
),
097402
(
2012
).
6.
I.
Brener
,
S.
Liu
,
I.
Staude
,
J.
Valentine
, and
C.
Holloway
,
Dielectric Metamaterials: Fundamentals, Designs and Applications
(
Woodhead Publishing
,
2019
).
7.
E.
Yablonovitch
, “
Inhibited spontaneous emission in solid-state physics and electronics
,”
Phys. Rev. Lett.
58
,
2059
2062
(
1987
).
8.
E.
Yablonovitch
,
T. J.
Gmitter
, and
K. M.
Leung
, “
Photonic band structure: The face-centered-cubic case employing nonspherical atoms
,”
Phys. Rev. Lett.
67
(
17
),
2295
2298
(
1991
).
9.
X.
Chen
,
Y.
Zhang
,
G.
Cai
,
J.
Zhuo
,
K.
Lai
, and
L.
Ye
, “
All-dielectric metasurfaces with high Q-factor Fano resonances enabling multi-scenario sensing
,”
Nanophotonics
11
(
20
),
4537
4549
(
2022
).
10.
S. M.
Kamali
,
E.
Arbabi
,
A.
Arbabi
, and
A.
Faraon
, “
A review of dielectric optical metasurfaces for wavefront control
,”
Nanophotonics
7
(
6
),
1041
1068
(
2018
).
11.
P.
Genevet
and
F.
Capasso
, “
Holographic optical metasurfaces: A review of current progress
,”
Rep. Prog. Phys.
78
,
024401
(
2015
).
12.
A.
Krasnok
,
M.
Tymchenko
, and
A.
Alu
, “
Nonlinear metasurfaces: A paradigm shift in nonlinear optics
,”
Mater. Today
21
(
1
),
8
21
(
2018
).
13.
A. S.
Solntsev
,
G. S.
Agarwal
, and
Y. S.
Kivshar
, “
Metasurfaces for quantum photonics
,”
Nat. Photonics
15
,
327
336
(
2021
).
14.
T.
Santiago-Cruz
,
S. D.
Gennaro
,
O.
Mitrofanov
,
S.
Addamane
,
J.
Reno
,
I.
Brener
, and
M. V.
Chekhova
, “
Resonant metasurfaces for generating complex quantum states
,”
Science
377
(
6609
),
991
995
(
2022
).
15.
S.
Campione
,
S.
Liu
,
L. I.
Basilio
,
L. K.
Warne
,
W. L.
Langston
,
T. S.
Luk
,
J. R.
Wendt
,
J. L.
Reno
,
G. A.
Keeler
,
I.
Brener
, and
M. B.
Sinclair
, “
Broken symmetry dielectric resonators for high quality factor Fano metasurfaces
,”
ACS Photonics
3
(
12
),
2362
2367
(
2016
).
16.
Z.
Liu
,
Y.
Xu
,
Y.
Lin
,
J.
Xiang
,
T.
Feng
,
Q.
Cao
,
J.
Li
,
S.
Lan
, and
J.
Liu
, “
High-Q quasibound states in the continuum for nonlinear metasurfaces
,”
Phys. Rev. Lett.
123
(
25
),
253901
(
2019
).
17.
F. J.
García de Abajo
, “
Optical excitations in electron microscopy
,”
Rev. Mod. Phys.
82
(
1
),
209
275
(
2010
).
18.
T.
Coenen
,
J.
van de Groep
, and
A.
Polman
, “
Resonant modes of single silicon nanocavities excited by electron irradiation
,”
ACS Nano
7
,
1689
1698
(
2013
).
19.
M.
Kociak
and
O.
Stéphan
, “
Mapping plasmons at the nanometer scale in an electron microscope
,”
Chem. Soc. Rev.
43
,
3865
3883
(
2014
).
20.
Z.
Dong
,
Z.
Mahfoud
,
R.
Paniagua-Domínguez
,
H.
Wang
,
A. I.
Fernández-Domínguez
,
S.
Gorelik
,
S. T.
Ha
,
F.
Tjiptoharsono
,
A. I.
Kuznetsov
,
M.
Bosman
, and
J. K. W.
Yang
, “
Nanoscale mapping of optically inaccessible bound-states-in-the-continuum
,”
Light: Sci. Appl.
11
,
20
(
2022
).
21.
A.
Polman
,
M.
Kociak
, and
F. J.
García de Abajo
, “
Electron-beam spectroscopy for nanophotonics
,”
Nat. Mater.
18
,
1158
1171
(
2019
).
22.
T. G.
Habteyes
,
I.
Staude
,
K. E.
Chong
,
J.
Dominguez
,
M.
Decker
,
A.
Miroshnichenko
,
Y.
Kivshar
, and
I.
Brener
, “
Near-field mapping of optical modes on all-dielectric silicon nanodisks
,”
ACS Photonics
1
(
9
),
794
798
(
2014
).
23.
M.
Tamagnone
,
A.
Ambrosio
,
K.
Chaudhary
,
L. A.
Jauregui
,
P.
Kim
,
W. L.
Wilson
, and
F.
Capasso
, “
Ultra-confined mid-infrared resonant phonon polaritons in van der Waals nanostructures
,”
Sci. Adv.
4
,
eaat7189
(
2018
).
24.
L.
Yin
,
V. K.
Vlasko-Vlasov
,
A.
Rydh
,
J.
Pearson
,
U.
Welp
,
S.-H.
Chang
,
S. K.
Gray
,
G. C.
Schatz
,
D. B.
Brown
, and
C. W.
Kimball
, “
Surface plasmons at single nanoholes in Au films
,”
Appl. Phys. Lett.
85
,
467
(
2004
).
25.
A.
Kubo
,
K.
Onda
,
H.
Petek
,
Z.
Sun
,
Y. S.
Jung
, and
H. K.
Kim
, “
Femtosecond imaging of surface plasmon dynamics in a nanostructured silver film
,”
Nano Lett.
5
(
6
),
1123
1127
(
2005
).
26.
M.
Cinchetti
,
A.
Gloskovskii
,
S. A.
Nepjiko
,
G.
Schönhense
,
H.
Rochholz
, and
M.
Kreiter
, “
Photoemission electron microscopy as a tool for the investigation of optical near fields
,”
Phys. Rev. Lett.
95
,
047601
(
2005
).
27.
B.
Frank
,
P.
Kahl
,
D.
Podbiel
,
G.
Spektor
,
M.
Orenstein
,
L.
Fu
,
T.
Weiss
,
M.
Horn-von Hoegen
,
T. J.
Davis
,
F.-J.
Meyer zu Heringdorf
, and
H.
Giessen
, “
Short-range surface plasmonics: Localized electron emission dynamics from a 60-nm spot on an atomically flat single-crystalline gold surface
,”
Sci. Adv.
3
(
7
),
e1700721
(
2017
).
28.
Q.
Sun
,
S.
Zu
, and
H.
Misawa
, “
Ultrafast photoemission electron microscopy: Capability and potential in probing plasmonic nanostructures from multiple domains
,”
J. Chem. Phys.
153
(
12
),
120902
(
2020
).
29.
L.
Zhang
,
A.
Kubo
,
L.
Wang
,
H.
Petek
, and
T.
Seideman
, “
Imaging of surface plasmon polariton fields excited at a nanometer-scale slit
,”
Phys. Rev. B
84
,
245442
(
2011
).
30.
P.
Kahl
,
S.
Wall
,
C.
Witt
,
C.
Schneider
,
D.
Bayer
,
A.
Fischer
,
P.
Melchior
,
M.
Horn-von Hoegen
,
M.
Aeschlimann
, and
F. J.
Meyer zu Heringdorf
, “
Normal-incidence photoemission electron microscopy (NI-PEEM) for imaging surface plasmon polaritons
,”
Plasmonics
9
,
1401
1407
(
2014
).
31.
Y.
Gong
,
A. G.
Joly
,
P. Z.
El-Khoury
, and
W. P.
Hess
, “
Nonlinear photoemission electron micrographs of plasmonic nanoholes in gold thin films
,”
J. Phys. Chem. C
118
,
25671
25676
(
2014
).
32.
J. P. S.
Fitzgerald
,
R. C.
Word
,
S. D.
Saliba
, and
R.
Könenkamp
, “
Photonic near-field imaging in multiphoton photoemission electron microscopy
,”
Phys. Rev. B
87
,
205419
(
2013
).
33.
H.
Hartmann
,
I.
Barke
,
A.
Friedrich
,
P.-A.
Plötz
,
O. S.
Bokareva
,
M.
Bahrami
,
K.
Oldenburg
,
J. A. A. W.
Elemans
,
R.
Irsig
,
K.-H.
Meiwes-Broer
,
O.
Kühn
,
S.
Lochbrunner
, and
S.
Speller
, “
Mapping long-lived dark states in copper porphyrin nanostructures
,”
J. Phys. Chem. C
120
(
30
),
16977
16984
(
2016
).
34.
K.
Stallberg
,
G.
Lilienkamp
, and
W.
Daum
, “
Plasmon–exciton coupling at individual porphyrin-covered silver clusters
,”
J. Phys. Chem. C
121
(
25
),
13833
13839
(
2017
).
35.
M.
Aeschlimann
,
T.
Brixner
,
D.
Differt
,
U.
Heinzmann
,
M.
Hensen
,
C.
Kramer
,
F.
Lükermann
,
P.
Melchior
,
W.
Pfeiffer
,
M.
Piecuch
,
C.
Schneider
,
H.
Stiebig
,
C.
Strüber
, and
P.
Thielen
, “
Perfect absorption in nanotextured thin films via Anderson-localized photon modes
,”
Nat. Photonics
9
(
10
),
663
668
(
2015
).
36.
E.
Mårsell
,
E.
Boström
,
A.
Harth
,
A.
Losquin
,
C.
Guo
,
Y.-C.
Cheng
,
E.
Lorek
,
S.
Lehmann
,
G.
Nylund
,
M.
Stankovski
,
C. L.
Arnold
,
M.
Miranda
,
K. A.
Dick
,
J.
Mauritsson
,
C.
Verdozzi
,
A.
L’Huillier
, and
A.
Mikkelsen
, “
Spatial control of multiphoton electron excitations in InAs nanowires by varying crystal phase and light polarization
,”
Nano Lett.
18
(
2
),
907
915
(
2018
).
37.
C.
Xu
,
H. W.
Yong
,
J.
He
,
R.
Long
,
A. R.
Cadore
,
I.
Paradisanos
,
A. K.
Ott
,
G.
Soavi
,
S.
Tongay
,
G.
Cerullo
,
A. C.
Ferrari
,
O. V.
Prezhdo
, and
Z.-H.
Loh
, “
Weak distance dependence of hot-electron-transfer rates at the interface between monolayer MoS2 and gold
,”
ACS Nano
15
(
1
),
819
828
(
2021
).
38.
C. W.
Hsu
,
B.
Zhen
,
A. D.
Stone
,
J. D.
Joannopoulos
, and
M.
Soljačić
, “
Bound states in the continuum
,”
Nat. Rev. Mater.
1
,
16048
(
2016
).
39.
K.
Koshelev
,
S.
Lepeshov
,
M.
Liu
,
A.
Bogdanov
, and
Y.
Kivshar
, “
Asymmetric metasurfaces with high-Q resonances governed by bound states in the continuum
,”
Phys. Rev. Lett.
121
,
193903
(
2018
).
40.
K.
Koshelev
,
A.
Bogdanov
, and
Y.
Kivshar
, “
Engineering with bound states in the continuum
,”
Opt. Photonics News
31
,
38
45
(
2020
).
41.
S.
Hüfner
,
Photoelectron Spectroscopy—Principles and Applications
(
Springer-Verlag
,
Berlin, Heidelberg
,
2003
).
42.
P.
Melchior
,
D.
Bayer
,
C.
Schneider
,
A.
Fischer
,
M.
Rohmer
,
W.
Pfeiffer
, and
M.
Aeschlimann
, “
Optical near-field interference in the excitation of a bowtie nanoantenna
,”
Phys. Rev. B
83
(
23
),
235407
(
2011
).
43.
H.
Yu
,
Q.
Sun
,
K.
Ueno
,
T.
Oshikiri
,
A.
Kubo
,
Y.
Matsuo
, and
H.
Misawa
, “
Exploring coupled plasmonic nanostructures in the near field by photoemission electron microscopy
,”
ACS Nano
10
(
11
),
10373
10381
(
2016
).
44.
E.
Bauer
,
Surface Microscopy with Low Energy Electrons
(
Springer
,
New York
,
2014
).
45.
A.
Genco
,
C.
Cruciano
,
M.
Corti
,
K. E.
McGhee
,
B.
Ardini
,
L.
Sortino
,
L.
Huttenhofer
,
T.
Virgili
,
D. G.
Lidzey
,
S. A.
Maier
,
A.
Bassi
,
G.
Valentini
,
G.
Cerullo
, and
C.
Manzoni
, “
k-space hyperspectral imaging by a birefringent common-path interferometer
,”
ACS Photonics
9
,
3563
3572
(
2022
).
46.
I.
Staude
and
J.
Schilling
, “
Metamaterial-inspired silicon nanophotonics
,”
Nat. Photonics
11
,
274
284
(
2017
).
47.
L.
Douillard
and
F.
Charra
, “
Photoemission electron microscopy, a tool for plasmonics
,”
J. Electron Spectrosc. Relat. Phenom.
189
,
24
29
(
2013
).
48.
Y.
Plotnik
,
O.
Peleg
,
F.
Dreisow
,
M.
Heinrich
,
S.
Nolte
,
A.
Szameit
, and
M.
Segev
, “
Experimental observation of optical bound states in the continuum
,”
Phys. Rev. Lett.
107
,
183901
(
2011
).
49.
A.
Cerjan
,
C.
Jörg
,
S.
Vaidya
,
S.
Augustine
,
W. A.
Benalcazar
,
C. W.
Hsu
,
G.
von Freymann
, and
M. C.
Rechtsman
, “
Observation of bound states in the continuum embedded in symmetry bandgaps
,”
Sci. Adv.
7
,
eabk1117
(
2021
).
50.
S.
Liu
,
A.
Vaskin
,
S.
Addamane
,
B.
Leung
,
M. C.
Tsai
,
Y.
Yang
,
P. P.
Vabishchevich
,
G. A.
Keeler
,
G.
Wang
,
X.
He
,
Y.
Kim
,
N. F.
Hartmann
,
H.
Htoon
,
S. K.
Doorn
,
M.
Zilk
,
T.
Pertsch
,
G.
Balakrishnan
,
M. B.
Sinclair
,
I.
Staude
, and
I.
Brener
, “
Light-emitting metasurfaces: Simultaneous control of spontaneous emission and far-field radiation
,”
Nano Lett.
18
(
11
),
6906
6914
(
2018
).
51.
A. C.
Overvig
,
S. C.
Malek
,
M. J.
Carter
,
S.
Shrestha
, and
N.
Yu
, “
Selection rules for quasi-bound states in the continuum
,”
Phys. Rev. B
102
,
035434
(
2020
).
52.
A.
Bostwick
,
T.
Ohta
,
T.
Seyller
,
K.
Horn
, and
E.
Rotenberg
, “
Quasiparticle dynamics in graphene
,”
Nat. Phys.
3
,
36
40
(
2007
).
53.
R.
Sarma
,
S.
Campione
,
M.
Goldflam
,
J.
Shank
,
J.
Noh
,
S.
Smith
,
P. D.
Ye
,
M.
Sinclair
,
J.
Klem
,
J.
Wendt
,
I.
Ruiz
,
S. W.
Howell
, and
I.
Brener
, “
Low dissipation spectral filtering using a field-effect tunable III–V hybrid metasurface
,”
Appl. Phys. Lett.
113
,
061108
(
2018
).
54.
S.
Tanuma
,
C. J.
Powell
, and
D. R.
Penn
, “
Calculations of electorn inelastic mean free paths. II. Data for 27 elements over the 50–2000 eV range
,”
Surf. Interface Anal.
17
(
13
),
911
926
(
1991
).
55.
J. C.
Kuhr
and
H. J.
Fitting
, “
Monte Carlo simulation of electron emission from solids
,”
J. Electron Spectrosc. Relat. Phenom.
105
(
2–3
),
257
273
(
1999
).
56.
Z.-J.
Ding
and
R.
Shimizu
, “
Inelastic collisions of kV electrons in solids
,”
Surf. Sci.
222
(
2–3
),
313
331
(
1989
).
57.
W.
Pong
, “
Photoemission from Al–Al2O3 films in the vacuum ultraviolet region
,”
J. Appl. Phys.
40
(
4
),
1733
1739
(
1969
).
58.
S. M.
Sze
,
J. L.
Moll
, and
T.
Sugano
, “
Range-energy relation of hot electrons in gold
,”
Solid-State Electron.
7
(
7
),
509
523
(
1964
).
59.
O. Yu.
Ridzel
,
V.
Astašauskas
, and
W. S. M.
Werner
, “
Low energy (1–100 eV) electron inelastic mean free path (IMFP) values determined from analysis of secondary electron yields (SEY) in the incident energy range of 0.1–10 keV
,”
J. Electron Spectrosc. Relat. Phenom.
241
,
146824
(
2020
).
60.
M.
Berg
,
F.
Liu
,
S.
Smith
,
R. G.
Copeland
,
C. K.
Chan
,
A. D.
Mohite
,
T. E.
Beechem
, and
T.
Ohta
, “
Imaging atomically thin semiconductors beneath dielectrics via deep ultraviolet photoemission electron microscopy
,”
Phys. Rev. Appl.
12
,
064064
(
2019
).
61.
T. E.
Beechem
,
S. W.
Smith
,
R. G.
Copeland
,
F.
Liu
, and
T.
Ohta
, “
Spectral and polarization based imaging in deep-ultraviolet excited photoelectron microscopy
,”
Rev. Sci. Instrum.
93
,
053701
(
2022
).