Open Access
Add to BibSonomy
Abstract
Complex lightwave manipulation such as broadband absorption has been realized with metasurfaces based on laterally arranged metal-dielectric-metal cavities with different geometries. However, application of these metasurfaces for optoelectronic devices by incorporating functional dielectrics remains challenging. Here, we integrate a quantum well infrared photodetector (QWIP) with a metasurface made of a patchwork of square cavities with different dimensions arranged in a subwavelength unit cell. Our detector realizes wideband photoresponse approaching the entire responsivity spectrum of the QWIP—single-sized square cavities can utilize only 60% of the possible bandwidth—and external quantum efficiencies of up to 78% at 6.8 µm. Our highly flexible design scheme enables integration of photodetectors and metasurfaces with arbitrary arrangements of cavities selectively responding to incidence with a specific wavefront.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Metal-dielectric-metal (MDM) plasmon cavities, with two metallic layers sandwiching a dielectric layer, are precisely controllable building blocks for metasurfaces [1,2]; a cavity’s optical response is determined by its critical dimensions and the refractive index of the dielectric layer [3]. Anomalous reflection [4 –6], focusing [7,8], and broadband absorption [9 –11] have been realized in metasurfaces by laterally arranging multiple subwavelength MDM cavities with different dimensions in a single unit cell. However, integrating such light control into MDM devices sandwiching functional dielectrics [12,13] remains a challenge.
Metasurface quantum well infrared photodetectors, engineered MDM cavities with a quantum well infrared photodetector (QWIP) as the dielectric layer [14 –17], are promising alternatives to toxic HgCdTe detectors presently dominant at wavelengths longer than 5 µm [18]. QWIPs operate using intersubband transitions (ISBT) in the quantum well requiring a vertical electric field due to quantum mechanical selection rules [19,20]. Thus, QWIPs cannot directly absorb normally incident light with a horizontal electric field—fundamentally limiting their quantum efficiency. A vertically stacked MDM cavity rotates the horizontal electric field of normal incidence to vertical by magnetic coupling and enhances its intensity [3,12,21], greatly improving the QWIP’s ISBT absorption. With enhanced absorption, thinner QWIP layers can maintain high responsivity (R, signal) with drastically reduced dark current (I D, thus reduced noise) [12,22]. Metasurface QWIPs have realized signal to noise ratios, detectivity (D*), as high as 3.9×1010 cm Hz1/2/W at 6.7 µm, approaching a theoretical limit of ∼ 5.0×1010 cm Hz1/2/W, room temperature operation, and GHz response [16 –18,23].
Previously reported metasurface QWIPs primarily used etched square cavities connected by thin (100–150 nm) wires; the exposed QWIP layer outside the square cavities was removed by dry etching, and thin conductive wires connected the otherwise isolated cavities [16 –18,23,24]. While square patches ensure polarization-independent absorption for maximizing R [17,24 –26], the etched, wired design minimizes the total semiconductor area, further reducing I D to maximize D* [14,16,17,23,27]. However, the thin wires, indispensable for electron collection from isolated cavities, present an unfortunate limitation to etched designs. The wires can induce unwanted polarization-dependence [16] and phase mismatch [17], gravely hindering design flexibility, normally a hallmark of metasurfaces, and indicating a need for alternative device structures.
A recent systematic study of metasurface QWIPs hinted at a route to eliminate wires and recover design flexibility [27]. For metasurface QWIPs using stripe cavities, stripe detectors with a continuous, unetched QWIP layer showed higher R (2.5 A/W) than etched detectors with the QWIP layer outside the stripe cavities removed (2.3 A/W), contrary to theoretical predictions. Etching-induced electron depletion in the exposed sidewalls of the QWIP layers was suggested to degrade R, reducing the advantage of the etched detectors. Thus, despite having higher I D, the unetched detectors had a competitive, only slightly reduced D* (I D = 4.0×10−7 A, D* = 2.7×1010 cm Hz1/2/W) compared to etched detectors (I D = 2.7×10−7 A, D* = 3.3×1010 cm Hz1/2/W), because higher R compensated for higher I D [27]. Additionally, the lateral conductivity of the QWIP layer was considered to be sufficient for electron collection without wires [15,25,27].
Here we first demonstrate that a metasurface QWIP using a square patch and a continuous, unetched QWIP layer is a competitive alternative to state-of-the-art etched patch detectors. D* of 3.7×1010 cm Hz1/2/W for the unetched detector can be realized, comparable to D* of etched detectors (3.9×1010 cm Hz1/2/W) [17], because higher R compensates for elevated I D. While unetched patch structures have been previously reported [25], until now, their detailed performance and merits relative to etched patch detectors [17] have not been clarified.
An unetched patch structure further eliminates the need for thin connecting wires, restoring the design flexibility lost in an etched patch design, and allowing us to freely arrange multiple cavities in a single subwavelength unit cell. Here, using a patchwork of 2 or 4 square cavities with different dimensions, we demonstrate metasurface QWIPs with broadened photoresponse. Each cavity selectively absorbs at its respective resonance wavelength with an absorption cross-section much larger than its geometrical area [28 –30]. Therefore, incident photons are sorted by wavelength into the corresponding resonant cavity within a near field—a phenomenon unique to metasurfaces, known as resonant photon sorting [10,13,31]. All photocurrent is summed up, enabling sensitive detection covering the responsivity spectra of the underlying QWIP [9 –11].
Our 4 Patch detector shows a broadened photoresponse utilizing nearly the entire responsivity spectrum of the QWIP—a one patch (1 Patch) detector can only use ∼ 60% of the possible bandwidth—and a maximum R of 4.3 A/W at 6.8 µm, corresponding to a 78% external quantum efficiency (EQE). Numerical simulation confirms incident light is absorbed by different resonant MDM cavities at different bands across the spectrum of the QWIP, indicative of resonant photon sorting. Demonstrated high performance in 2 and 4 Patch detectors paves the way for the development of functional detectors with wavefront-selective responses based on previously developed multi-cavity designs.
2. Fundamental properties of unetched, one patch detectors
2.1 Design of one patch detectors
The photoconductive QWIP layer used in this study consists of a single 4 nm quantum well layer (Si doping density: 3×1018 /cm3) sandwiched by two 50 nm Al0.3Ga0.7As barrier layers and two 48 nm GaAs contact layers [17,27,32], with a total thickness T = 200 nm, optimized for maximum absorption at a peak wavelength of 6.7 µm [17,27]. The continuous QWIP layer is sandwiched by square Au patches with t m1 = 100 nm, and a thick Au substrate (650 nm, t m2 = 200 nm for calculation) (Fig. 1(a)).
Fig. 1. Design, fabrication, and absorption behavior of square patch detectors. a) Schematic of square patch structure sandwiching QWIP layers made of contact, barrier, and quantum well layers. E x and k indicate the electric field polarization and wave vector of incident light. b) Colorized scanning electron microscope (SEM) image of Au patch array (gold) on top of QWIP layer (blue). c) Schematic of device operation. The yellow lines illustrate electron transport through the device. d) Experimental and simulated total absorption (A TOT) determined from s-polarized reflection (incidence angle θ = 26° in the y-z plane) for a device with L = 0.92 µm and P = 1.8 µm with a resonance peak nearly matching the quantum well absorption peak at 6.7 µm (red line). e) Simulated absorption in each layer of the detector (θ = 0°). A ISBT is the ISBT absorption of the quantum well.
Optical properties of the square patch detectors were obtained by numerical simulation using finite element analysis. The QWIP layer was treated as a wavelength-dependent five-layer multilayer [17,27,33], and the quantum well was assumed as a uniaxial material with ISBT absorption in the vertical direction and free-carrier absorption in the lateral directions. More details on this model can be found in our earlier papers [17,27]. Square cavities with length L = 0.92 µm and period P = 1.8 µm were found to give maximum total absorption (A TOT) at a position nearly matching the quantum well absorption peak of 6.7 µm (Fig. 1(d)).
2.2 Fabrication of one patch detectors
The QWIP layer grown by molecular beam epitaxy on a GaAs substrate was transferred to an Au substrate by wafer bonding. For the patterning of the unetched square cavities, electron beam lithography was used. Several detectors with patches having L values around 0.92 µm were fabricated (Fig. 1(b)). Each detector is a 100 µm × 100 µm square containing thousands of arrayed patches.
In devices with etched cavities, electron conduction occurs through the Au patches and thin connecting wires [17]; for our unetched design in the present paper, lateral transport occurs through the contact layer of the QWIP instead, and the Au patches simply act as antennas [15], not as electrodes (Fig. 1(c)). For efficient systematic evaluation, seven detectors with different parameters were integrated on a chip and assembled on an 8-pin ceramic package. Another pin is for the common ground. The signals were extracted by bonding an Au wire at the root of each Au electrode.
2.3 Optical characterization
To confirm the absorption peak positions behavior in our detectors, experimental A TOT was determined from reflection spectra (r TOT) taken with s-polarized incidence at an incidence angle θ = 26° using an infrared microscope equipped with a Cassegrain objective lens connected to an FTIR. A relatively small incidence angle distribution centered at 26° was realized with an aperture. Given the thickness of our Au substrate, we consider transmission to be negligible, so A TOT = 1 – r TOT. Reflection measurement for an oblique incidence was due to the experimental difficulty of vertical reflection measurement. However, because square cavities do not show remarkable incidence-angle dependence [17], there is no essential influence on the discussion here.
Fundamental absorption behavior of our one patch detectors can be seen in Fig. 1(d) and (e); experimental and simulated absorption for a fabricated detector with L = 0.92 µm and P = 1.8 µm reaches ∼90% (Fig. 1(d)). Figure 1(e) illustrates the share of A TOT contributed by each layer: Au, contact, and quantum well layers. Absorption in the Al0.3Ga0.7As barrier layers is negligible.
For responsivity measurements, the devices were installed in a cryostat with ZnSe windows, and their responsivity spectra were measured with FTIR by feeding the amplified current signal to the external port. The spectral responsivity was quantified based on a calibrated HgCdTe detector. The intrinsic properties of the QWIP layer were evaluated by a Brewster-angle detector at θ = 65° [17].
For detectors with increasing L, the peak position of R systematically shifts to longer wavelengths (Fig. 2(a)); the spectra exhibit an envelope analogous to the responsivity of the quantum well (Fig. 2(b)). A maximum R peak of 3.9 A/W, corresponding to a maximum external quantum efficiency EQEpeak of 71%, and nearly 20% larger than previously reported for etched patch antennas (Table 1) [17], is observed for the detector with L = 0.99 µm at a bias voltage V peak = 0.45 V. The higher R for the unetched patch detectors is consistent with previous reports on unetched and etched stripe detectors [27]. Characteristic of square patch detectors [17,25], R peak is polarization-independent (Fig. 2(c)) for both unetched and etched detectors. The peak wavelength λpeak = 7.05 µm of the detector with maximum R peak in Fig. 2(a) is red-shifted from the 6.7 µm peak of the underlying quantum well (Fig. 2(b))—similar to behavior previously observed and attributed to strain effects during metasurface fabrication [27,33].
Fig. 2. Device characterization of single patch detectors. a) Systematic unpolarized responsivity (R) spectra covering the breadth of the quantum well. b) Brewster-angle detector spectrum. c) Polarization angle-dependent peak responsivity (R peak) for L = 0.99 µm unetched patch detector and etched patch detector from [17], patch length L = 1.19 µm, P = 2.0 µm, and thin Z-shape wires with width W = 100 nm and folding length S = 0.29 µm. Inset schematic defines the polarization angle. E and k indicate the electric field polarization and wave vector of incident light.
Table 1. Electrical Properties of Unetched and Etched Patch Detectorsa
2.4 Electrical characterization
In discussing the electrical properties of infrared photodetectors, we need to use several specific terms; all are defined here. Dark current, I D, is the current present without incident radiation in a detector under bias due to its inherent transport properties. Background current, I BG, is the baseline current inevitable during normal operation in a detector under bias due to incident radiation from a room temperature (298 K) environment. The background photocurrent (I P, BG) [27] is the contribution of photoexcited electrons to I BG: I P, BG = I BG - I D. Photoconductive gain, g, is the number of electrons circulating in a photoconductive detector per photoexcited electron generated in the quantum well [16,19,20]. Finally, background-limited detectivity D*BG, a practical detectivity for real-world application where the detector is exposed to environmental illumination, is calculated as:
in relation to R peak, I BG, and g, where A is the device area and e is the electron charge. At 78 K, our detectors are in a background-current limited regime [17,27], thus D*BG is the relevant detectivity. From here, we simply refer to D*BG as D*.To measure the dark current I D, the detector was covered by a cold shield with a blackbody coating cooled at the same temperature as the detector (78 K). Since the blackbody radiation at 78 K is sufficiently small, we can consider current measured under these conditions to be inherent to the detector. For background current (I BG) measurements, the cold shield is removed, and a background light at 298 K was incident on the detector from an area with an effective field of view of 102°. The current–voltage relationships for I D and I BG were measured with a source meter. The value of photoconductive gain g was determined from I D and noise spectral density i n/Δf 1/2 (Δf: unit frequency bandwidth) measured by a fast Fourier transform analyzer, by the relationship i n/Δf 1/2 = (4egI D)1/2. The value at 1 kHz was used [34].
From Table 1, we can clearly compare the properties of unetched and etched square patch detectors. The unetched patch detector with the largest R peak (L = 0.99 µm) has D* = 3.7×1010 cm Hz1/2/W, very close to the previously reported maximum D* of etched patch detectors [17] (Table 1). Although R peak is greater for the unetched patch detector than the etched patch detector, I D and I BG are also higher due to the greater semiconductor area, leading to the slightly lower D*. Higher I P, BG for the unetched patch detector (2.3×10−7 A, compared to 2.1×10−7 A for the etched patch detector) indicates more efficient light absorption than the etched patch detector, in good agreement with the observed R peak values for unetched and etched detectors.
The values of D* and R suggest unetched patch detectors are viable alternatives to etched designs. Since our unetched design does not need connective wires between cavities, we can freely arrange multiple cavities across the surface of our QWIP layer, taking advantage of the inherent design flexibility of metasurfaces. In the next section, we apply this design flexibility to broaden the photoresponse of our detectors.
3. Patchwork detectors for broadband absorption via resonant photon sorting
3.1 Design and fabrication of patchwork detectors
Although the unetched patch detectors have D* comparable to an etched patch detector, the bandwidth of the best unetched detector (bandwidth ∼ 0.8 µm) only utilizes ∼ 60% of the total bandwidth of the QWIP (bandwidth∼ 1.3 µm). By combining multiple cavities with different resonance wavelengths in a subwavelength unit cell, we can realize absorption across the QWIP’s full bandwidth [10,11], with all photocurrent adding together. Note that in this manuscript, full width half maximum is applied as the index describing the bandwidth of our spectra.
Following the above design and fabrication process for the single cavity detector, cavity dimensions and periods for detectors with 2 patches, called 2 Patch detectors from here, designed for absorption at λ1 = 6.2 µm and λ2 = 7.2 µm (L 1 = 0.90 µm, L 2 = 1.06 µm, P = 2.5 µm) and 4 patches, called 4 Patch detectors, for absorption at λ1 = 6.0 µm, λ2 = 6.5 µm, λ3 = 7.0 µm, and λ4 = 7.5 µm (L 1 = 0.87 µm, L 2 = 0.94 µm, L 3 = 1.01 µm, L 4 = 1.08 µm, P = 3.3 µm) were fabricated. Within a unit cell extending from (x, y) = (0, 0) to (P, P), patches are centered at (P/4, P/4) and (3P/4, 3P/4) for a 2 Patch detector and (P/4, P/4), (3P/4, P/4), (P/4, 3P/4), and (3P/4, 3P/4) for a 4 Patch detector (Fig. 3(a)). The unetched detector with a single square patch per unit cell with L = 0.99 µm and P = 1.8 µm from earlier, called a 1 Patch detector from here, is shown for comparison.
Fig. 3. (a) SEM images of 1 Patch (orange, L = 0.99 µm, P = 1.8 µm), 2 Patch (red, L 1 = 0.90 µm, L 2 = 1.06 µm, P = 2.5 µm), and 4 Patch (green, L 1 = 0.87 µm, L 2 = 0.94 µm, L 3 = 1.01 µm, L 4 = 1.08 µm, P = 3.3 µm) detectors. One unit cell of each structure extending from (0, 0) to (P, P) is enclosed by the dashed lines. b) Experimental (left) and simulated (right) total absorption (A TOT) determined from s-polarized reflection (incidence angle θ = 26° in the y-z plane), and c) experimental (left) and simulated (right) responsivity (R) (incidence angle θ = 0°) for 1 Patch, 2 Patch, and 4 Patch detectors.
3.2 Optical characterization
A TOT of the metasurface QWIPs [17], which includes absorption by the quantum well, Au, and contact layers, and R, were determined experimentally and through numerical simulation (Fig. 3(b,) (c)). For the 2 and 4 Patch detectors, spectra for both A TOT and R show remarkable broadening compared to the original 1 Patch detector (Table 2). Note A TOT is consistently broader than R. The bandwidth and device performance for all detectors are summarized in Tables 2 and 3. Most importantly, the 4 Patch detector demonstrates a bandwidth for R (1.23 µm) 1.5 times larger than the 1 Patch detector (0.83 µm), utilizing nearly the entire bandwidth of the underlying QWIP (bandwidth 1.32 µm). The 2 and 4 Patch detectors show high R peak values as well; R peak of 4.3 A/W at 6.8 µm for the 4 Patch detector is larger than the 3.9 A/W at 7.05 µm for the 1 Patch detector (Table 2).
Table 2. Bandwidths of experimental and simulated total absorption (ATOT) and responsivity (R)a
Table 3. Experimental device performance of single and multi-cavity detectors
As seen in Fig. 3(b), (c) and Table 2, bandwidth of simulated A TOT and R spectra are similar to their experimental counterparts. Simulated R was determined as:
where A ISBT is the ISBT absorption of the quantum well, g = 2.7 (Table 1), λ the wavelength, h Planck’s constant, and c the speed of light. Internal quantum efficiencies are equivalent to A ISBT [27], and term ${A_{ISBT}} \times g$ describes the EQE of the detector [19,20]. Simulated R in this manuscript was based on experimental R of the Brewster-angle detector (Fig. 2(b)). In the simulated spectra in Fig. 3(c), the overestimation of R at shorter wavelengths and underestimation of R at longer wavelengths are generally observed; these features are consistent with previous reports [27], arising due to effects of metasurface fabrication as noted in section 2.Simulated absorption in quantum well, Au and contact layers in 1 Patch, 2 Patch, and 4 Patch detectors (Fig. S1) demonstrates why bandwidth is smaller in simulated and experimental R spectra compared to A TOT. Monotonous absorption from the Au and contact layers due to free electrons contributes a larger fraction of A TOT over a broader absorption range than A ISBT in the quantum well. Similar behavior is observed in the detector in Fig. 1(e). Internal quantum efficiencies (A ISBT) have maxima of ∼20–25% for our detectors, which is not necessarily high. However, the high g arising from the single-quantum-well photoconductive QWIP design [17] leads to greatly enhanced EQE (Eq. (2), Table 3) and thus high R.
3.3 Resonant photon sorting
In detectors using a patchwork of cavities, each cavity selectively absorbs at its resonance wavelength. To visualize selective absorption in the 2 and 4 Patch detector we calculated the distribution of A ISBT in the 2 and 4 Patch detectors at wavelengths corresponding to the resonance peaks of different cavities. In Fig. 4, a ISBT represents the local ISBT absorbance within the quantum well layer in units of 1/µm2. Integration of a ISBT over the total P×P area gives A ISBT at corresponding wavelengths (dark blue traces, Fig. S1). For both 2 and 4 Patch detectors (Fig. 4(a)–(b), Fig. 4(c)–(f) respectively), we can see how a ISBT is concentrated within the corresponding resonant cavity at different wavelengths—demonstrating clear resonant photon sorting [10,13,31,33].
Fig. 4. Local ISBT absorbance within the quantum well layer a ISBT (1/ µm2) in one unit cell of a, b) 2 Patch detectors and c–f) 4 Patch detectors at the resonance wavelengths of individual patches in the detectors. White outlines indicate the positions of the Au patches. The wavelength used for simulation is indicated in the bottom of each figure.
In our previous study using electrically isolated, paired metasurface QWIPs [33], resonant photon sorting was applied to route incident photons with different wavelengths into their corresponding resonant cavity to generate two independent photocurrents. For detectors with a patchwork of cavities and a continuous QWIP layer in this study, all ISBT absorption sums up to one total photocurrent. Different patches promote enhanced ISBT absorption at different wavelengths, so sensitive photodetection can be realized across nearly the entire responsivity spectrum of the underlying QWIP.
3.4 3 + 1 Patch detector
Attempting to maximize bandwidth in R by enhancing A ISBT at short wavelengths, we also fabricated another detector with four cavities, called a 3 + 1 Patch detector. Three cavities were designed for resonance at 6.0 µm and one cavity for resonance at 7.5 µm (L 1 = L 2 = L 3 = 0.85 µm, L 4 = 1.06 µm, P = 2.8 µm) (Fig. S2). Compared to the 4 Patch detector, the 3 + 1 Patch detector indeed realized broader A TOT and R (bandwidth of R: 1.25 µm, Fig. S2, Table 2). Unexpectedly however, this broadening arose not primarily from enhanced absorption near 6.0 µm, but from a new strong resonant absorption peak at an intermediate wavelength close to 6.6 µm. This behavior is attributed to near-field coupling between two of the three L 1 cavities aligned along the y axis around λ = 6.6 µm (Fig. S3). More details are contained in the Supplement 1.
4. Conclusion
In summary, we have demonstrated metasurface QWIPs using square patches and a continuous, unetched QWIP layer. We found that the square patches provide strong, polarization-independent absorption, leading to high R up to 3.9 A/W at 7.05 µm (EQE = 71%). Although I D is higher than in detectors using etched square patches, a detectivity of D* = 3.7×1010 cm Hz1/2/W can be achieved, close to the reported maximum for state-of-the-art etched patch detectors. Because thin wires connecting individual patches are unnecessary, unetched detectors recover the hallmark design flexibility of metasurfaces; we can freely arrange a patchwork of cavities in a single unit cell atop a continuous QWIP layer. Using 2 Patch and 4 Patch cavity arrays, we apply resonant photon sorting to realize sensitive photodetection covering nearly the entire responsivity spectrum of our underlying QWIP, compared to only ∼ 60% used by than a 1 Patch detector, and R up to 4.3 A/W (EQE = 78%).
Metasurfaces combining MDM cavities with different geometries on a continuous dielectric layer have previously demonstrated light manipulation such as anomalous reflection [4 –6], focusing [7,8] and circular polarization discrimination [35]. Applying an unetched design scheme as demonstrated in this manuscript, these metasurfaces can be directly integrated with a photodetective dielectric layer. Thus, already existing multi-cavity metasurfaces can be transformed into functional detectors, selectively responsive to incidence with specific angle [5], convergence, divergence [17], and even vortex [36,37].
Funding
Japan Society for the Promotion of Science (JP15H02011, JP17H01275, JP19H00875); Iketani Science and Technology Foundation; Center for Functional Sensors and Actuators; National Institute for Materials Science.
Acknowledgments
The authors acknowledge helpful discussion with an anonymous company, Y. Sakuma, T. Noda, A. Ohtake, D. Tsuya, N. Ikeda, E. Watanabe, K. Miyano, M. Iwanaga, and K. Sakoda, and the technical assistance of SIJTechnology, Inc. This work was further supported the NIMS Nanofabrication Platform in Nanotechnology Platform Project sponsored by the Ministry of Education, Culture, Sports, Science and Technology, Japan.
Disclosures
The authors declare no conflicts of interest.
See Supplement 1 for supporting content.
References
1. N. Yu and F. Capasso, “Flat optics with designer metasurfaces,” Nat. Mater. 13(2), 139–150 (2014). [CrossRef]
2. N. Meinzer, W. L. Barnes, and I. R. Hooper, “Plasmonic meta-atoms and metasurfaces,” Nat. Photonics 8(12), 889–898 (2014). [CrossRef]
3. G. Lévêque and O. J. F. Martin, “Tunable composite nanoparticle for plasmonics,” Opt. Lett. 31(18), 2750–2752 (2006). [CrossRef]
4. D. M. Pozar, S. D. Targonski, and H. D. Syrigos, “Design of millimeter wave microstrip reflectarrays,” IEEE Trans. Antennas Propag. 45(2), 287–296 (1997). [CrossRef]
5. S. Sun, K. Y. Yang, C. M. Wang, T. K. Juan, W. T. Chen, C. Y. Liao, Q. He, S. Xiao, W. T. Kung, G. Y. Guo, L. Zhou, and D. P. Tsai, “High-efficiency broadband anomalous reflection by gradient meta-surfaces,” Nano Lett. 12(12), 6223–6229 (2012). [CrossRef]
6. A. Pors and S. I. Bozhevolnyi, “Plasmonic metasurfaces for efficient phase control in reflection,” Opt. Express 21(22), 27438–27451 (2013). [CrossRef]
7. A. Pors, M. G. Nielsen, R. L. Eriksen, and S. I. Bozhevolnyi, “Broadband focusing flat mirrors based on plasmonic gradient metasurfaces,” Nano Lett. 13(2), 829–834 (2013). [CrossRef]
8. S. Zhang, M.-H. Kim, F. Aieta, A. She, T. Mansuripur, I. Gabay, M. Khorasaninejad, D. Rousso, X. Wang, M. Troccoli, N. Yu, and F. Capasso, “High efficiency near diffraction-limited mid-infrared flat lenses based on metasurface reflectarrays,” Opt. Express 24(16), 18024–18034 (2016). [CrossRef]
9. X. Liu, T. Tyler, T. Starr, A. F. Starr, N. M. Jokerst, and W. J. Padilla, “Taming the blackbody with infrared metamaterials as selective thermal emitters,” Phys. Rev. Lett. 107(4), 045901 (2011). [CrossRef]
10. P. Bouchon, C. Koechlin, F. Pardo, R. Haïdar, and J.-L. Pelouard, “Wideband omnidirectional infrared absorber with a patchwork of plasmonic nanoantennas,” Opt. Lett. 37(6), 1038–1040 (2012). [CrossRef]
11. C.-W. Cheng, M. N. Abbas, C.-W. Chiu, K.-T. Lai, M.-H. Shih, and Y.-C. Chang, “Wide-angle polarization independent infrared broadband absorbers based on metallic multi-sized disk arrays,” Opt. Express 20(9), 10376–10381 (2012). [CrossRef]
12. J. Le Perchec, Y. Desieres, and R. Espiau De Lamaestre, “Plasmon-based photosensors comprising a very thin semiconducting region,” Appl. Phys. Lett. 94(18), 181104 (2009). [CrossRef]
13. J. Le Perchec, Y. Desieres, N. Rochat, and R. Espiau De Lamaestre, “Subwavelength optical absorber with an integrated photon sorter,” Appl. Phys. Lett. 100(11), 113305 (2012). [CrossRef]
14. Y. Nga Chen, Y. Todorov, B. Askenazi, A. Vasanelli, G. Biasiol, R. Colombelli, and C. Sirtori, “Antenna-coupled microcavities for enhanced infrared photo-detection,” Appl. Phys. Lett. 104(3), 031113 (2014). [CrossRef]
15. Q. Li, Z. Li, N. Li, X. Chen, P. Chen, X. Shen, and W. Lu, “High-polarization-discriminating infrared detection using a single quantum well sandwiched in plasmonic micro-cavity,” Sci. Rep. 4(1), 6332 (2015). [CrossRef]
16. D. Palaferri, Y. Todorov, A. Bigioli, A. Mottaghizadeh, D. Gacemi, A. Calabrese, A. Vasanelli, L. Li, A. G. Davies, E. H. Linfield, F. Kapsalidis, M. Beck, J. Faist, and C. Sirtori, “Room-temperature nine-µm-wavelength photodetectors and GHz-frequency heterodyne receivers,” Nature 556(7699), 85–88 (2018). [CrossRef]
17. H. T. Miyazaki, T. Mano, T. Kasaya, H. Osato, K. Watanabe, Y. Sugimoto, T. Kawazu, Y. Arai, A. Shigetou, T. Ochiai, Y. Jimba, and H. Miyazaki, “Synchronously wired infrared antennas for resonant single-quantum-well photodetection up to room temperature,” Nat. Commun. 11(1), 565 (2020). [CrossRef]
18. A. Rogalski, “History of infrared detectors,” Opto-Electron. Rev. 20(3), 279–308 (2012). [CrossRef]
19. B. F. Levine, “Quantum-well infrared photodetectors,” J. Appl. Phys. 74(8), R1–R81 (1993). [CrossRef]
20. H. Schneider and H. C. Liu, Quantum Well Infrared Photodetectors (Springer, 2007).
21. G. Dolling, C. Enkrich, M. Wegener, J. F. Zhou, C. M. Soukoulis, and S. Linden, “Cut-wire pairs and plate pairs as magnetic atoms for optical metamaterials,” Opt. Lett. 30(23), 3198–3200 (2005). [CrossRef]
22. A. Rogalski, “Infrared detectors: status and trends,” Prog. Quantum Electron. 27(2-3), 59–210 (2003). [CrossRef]
23. A. Bigioli, G. Armaroli, A. Vasanelli, D. Gacemi, Y. Todorov, D. Palaferri, L. Li, A. G. Davies, E. H. Linfield, and C. Sirtori, “Long-wavelength infrared photovoltaic heterodyne receivers using patch-antenna quantum cascade detectors,” Appl. Phys. Lett. 116(16), 161101 (2020). [CrossRef]
24. J. Deng, Y. Zheng, J. Zhou, Z. Li, S. Guo, X. Dai, Y. Yu, Z. Ji, Z. Chu, X. Chen, and W. Lu, “Absorption enhancement in all-semiconductor plasmonic cavity integrated THz quantum well infrared photodetectors,” Opt. Express 28(11), 16427–16436 (2020). [CrossRef]
25. Y. L. Jing, Z. F. Li, Q. Li, X. S. Chen, P. P. Chen, H. Wang, M. Y. Li, N. Li, and W. Lu, “Pixel-level plasmonic microcavity infrared photodetector,” Sci. Rep. 6(1), 25849 (2016). [CrossRef]
26. J. A. Montoya, Z.-B. Tian, S. Krishna, and W. J. Padilla, “Ultra-thin infrared metamaterial detector for multicolor imaging applications,” Opt. Express 25(19), 23343–23355 (2017). [CrossRef]
27. M. F. Hainey, T. Mano, T. Kasaya, T. Ochiai, H. Osato, K. Watanabe, Y. Sugimoto, T. Kawazu, Y. Arai, A. Shigetou, and H. T. Miyazaki, “Systematic studies for improving device performance of quantum well infrared stripe photodetectors,” Nanophotonics 9(10), 3373–3384 (2020). [CrossRef]
28. C. F. Bohren and D. R. Huffman, Absorption and Scattering of Light by Small Particles (Wiley, 1983).
29. H. T. Miyazaki and Y. Kurokawa, “How can a resonant nanogap enhance optical fields by many orders of magnitude?” IEEE J. Sel. Top. Quantum Electron. 14(6), 1565–1576 (2008). [CrossRef]
30. D. Palaferri, Y. Todorov, A. Mottaghizadeh, G. Frucci, G. Biasiol, and C. Sirtori, “Ultra-subwavelength resonators for high temperature high performance quantum detectors,” New J. Phys. 18(11), 113016 (2016). [CrossRef]
31. C. Koechlin, P. Bouchon, F. Pardo, J. Jaeck, X. Lafosse, J. L. Pelouard, and R. Haïdar, “Total routing and absorption of photons in dual color plasmonic antennas,” Appl. Phys. Lett. 99(24), 241104 (2011). [CrossRef]
32. T. Mano, H. T. Miyazaki, T. Kasaya, T. Noda, and Y. Sakuma, “Double-sided nonalloyed ohmic contacts to Si-doped GaAs for plasmoelectronic devices,” ACS Omega 4(4), 7300–7307 (2019). [CrossRef]
33. M. F. Hainey, T. Mano, T. Kasaya, T. Ochiai, H. Osato, K. Watanabe, Y. Sugimoto, T. Kawazu, Y. Arai, A. Shigetou, and H. T. Miyazaki, “Near-field resonant photon sorting applied: dual-band metasurface quantum well infrared photodetectors for gas sensing,” Nanophotonics 9(16), 4775–4784 (2020). [CrossRef]
34. B. F. Levine, C. G. Bethea, G. Hasnain, V. O. Shen, E. Pelve, R. R. Abbott, and S. J. Hsieh, “High sensitivity low dark current 10 µm GaAs quantum well infrared photodetectors,” Appl. Phys. Lett. 56(9), 851–853 (1990). [CrossRef]
35. Q. Jiang, B. Du, M. Jiang, D. Liu, Z. Liu, B. Li, Z. Liu, F. Lin, X. Zhu, and Z. Fang, “Ultrathin circular polarimeter based on chiral plasmonic metasurface and monolayer MoSe2,” Nanoscale 12(10), 5906–5913 (2020). [CrossRef]
36. F. Yue, D. Wen, J. Xin, B. D. Gerardot, J. Li, and X. Chen, “Vector vortex beam generation with a single plasmonic metasurface,” ACS Photonics 3(9), 1558–1563 (2016). [CrossRef]
37. S. Tang and F. Ding, “High-efficiency focused optical vortex generation with geometric gap-surface plasmon metalenses,” Appl. Phys. Lett. 117(1), 011103 (2020). [CrossRef]
