Plasmonic Metasurface Integrated Black Phosphorus‐Based Mid‐Infrared Photodetector with High Responsivity and Speed

Black Phosphorus (BP), a van der Waals (vdW) semiconducting material, is studied intensively due to its unique structural and anisotropic optical properties. BP exhibits optical anisotropy with thickness tunable direct bandgap from 0.3–2.0 eV. Many BP‐based photodetectors operating in the mid‐infrared (MIR) band are proposed, but their performance is still limited due to the small optical absorption cross section at room temperature. In this work, a plasmonic metasurface that has localized surface plasmon resonance (LSPR) at 3.7 µm is designed, matching well with the band edge of a BP flake and field confinement around the edge of Au‐disk as confirmed by finite difference time domain simulation and Fourier‐transform infrared spectroscopy measurement results. In addition, a BP flake is integrated on the plasmonic metasurface and a significant quenching (twelve‐fold) is observed in the photoluminescence owing to the Förster resonance energy transfer effect induced by dipole–dipole coupling. Following that, a BP‐based MIR photodetector is fabricated on the plasmonic metasurface. Here, the demonstrated BP/plasmonic metasurface‐based photodetector achieves the peak responsivity of 495.85 mAW−1 and ultrahigh operation speed (>10 MHz) at the power of 8.55 µW under 3.7 µm incident wavelength. This demonstrated photodetector opens a new opportunity for optoelectronic applications in the MIR region.

photodetectors operating in the MIR region with high responsivity have not been realized yet.
In this work, we have designed and fabricated an Au disk-based plasmonic metasurface with LSPR in the MIR region (3.7 µm), matching the band edge of BP, as indicated by our numerical simulation and optical experiments. Further, we have integrated BP on the plasmonic metasurface and discovered that the photoluminescence (PL) of BP gets quenched approximately twelvefold due to the Förster resonance energy transfer (FRET) effect. Finally, a vertically stacked Gr t /BP/Gr b /plasmonic metasurfacebased photodetector has been fabricated, demonstrating optimal optoelectronic performance. By integrating BP-based vdW heterostructures photodetector with MIR plasmonic metasurface, we observed that the photocurrent of BP has been enhanced by approximately fourfold (309%) due to enhanced light-matter interaction and strong coupling between LSPR-induced near field in plasmonic metasurface and the BP flake. The photoresponsivity of the hybrid detector can reach up to 495.85 mAW −1 at an illumination wavelength of 3.7 µm, and its operating speed can reach up to an order of ns (>10 MHz) at room temperature, showing great promise for practical applications.

Results and Discussion
The schematic of the designed Au disk-based plasmonic metasurface on the SiO 2 /Si substrate is shown in Figure 1a. To design the plasmonic metasurface with LSPR in the MIR, near the band edge of BP, we optimized the dimension of the Au disk by exciting LSPR and achieving maximum resonance intensity. The parameter sweep was performed by using the finite difference time domain (FDTD) methods (refer to Experimental Section for simulation details), to optimize the diameter of the Au disk for keeping a fixed periodicity of 2.2 µm. As shown in Figure 1b, the reflectance intensity reaches its maximum value at 3.7 µm for a disk diameter of 1.7 µm, corresponding to LSPR energy. The thickness of Au disk and periodicity were optimized by calculating reflectance spectra for various Au disk thicknesses and periodicity as shown in Figure S1, Supporting Information. The optimized periodicity (P), diameter (D), and height (H) of the Au disk-based plasmonic metasurface were found to be 2.2, 1.7, and 0.1 µm, respectively. The electric field distribution |E| 2 at the LSPR wavelength of 3.7 µm in the x-y and x-z planes is shown in Figures 1c and 1d, respectively. Figure 1c,d illustrates that the maximum field is localized along the x-axis and at the edge of the disk due to the induced dipoles at the resonant wavelength of incident light. The E z component of the field in the x-y and x-z plane is illustrated in Figure S2, Supporting Information, indicating the dipole formation at the LSPR wavelength. This unique feature of induced dipoles and field localization at the band edge of BP could improve the optical absorption and dipole-dipole coupling between the metasurface and BP.
Next, the plasmonic metasurface was fabricated using the e-beam lithography technique (refer to Figure S3, Supporting Information) on the SiO 2 /Si substrate. The field emission Figure 1. Structural parameter optimization of plasmonic metasurface. a) Schematic illustration of Au disk-based plasmonic metasurface. b) Parameter sweep for resonance wavelength with a radius of the Au disk with black dotted line show resonance intensity peak. c,d) Electric field |E| 2 distribution of a single Au disk in the x-y and x-z plane at 3.7 µm. The color bar in (b), and (c,d) represents reflectance, and electric field |E| 2 distribution, respectively. e) FESEM image of the fabricated plasmonic metasurface at 45° tilted angle with the normal axis and inset image shows single Au-disk. f) Measured FTIR reflectance spectra in red line and calculated reflectance spectra of the fabricated and designed plasmonic metasurface, respectively. The inset shows the structural parameter (height of disk (H), diameter (D), and periodicity (P)) of the designed metasurface. www.advmatinterfaces.de scanning electron microscopy (FESEM) was performed to observe the structure parameter of the fabricated plasmonic metasurface and single Au-disk as shown in Figure 1e and Figure S4, Supporting Information, and the inset image of Figure 1e, respectively. From FESEM images, it is evident that the structural parameters of the fabricated sample correspond with the designed parameters. The reflectance spectra of the fabricated metasurface were measured using a micro-Fourier transform infrared spectrometer (µ-FTIR) and compared with FDTD calculated reflectance spectra, shown in Figure 1f. According to our measurements, the reflectance peak is ≈3.7 µm, which is in good agreement with the calculated reflectance spectrum.
To understand the optical properties of the BP and hybrid BP/plasmonic metasurface configuration, the dry transfer technique was used to transfer the BP flake on the designed metasurface, following the standard mechanical exfoliation of the vdW flake. [50] The sample was prepared in such a way that ≈50% of the BP flake was hybridized with the metasurface, whereas the remaining portion was placed on the SiO 2 /Si substrate as a control, as shown in Figure 2a. The phonon modes of the BP flake can be characterized by Raman spectroscopy and are depicted in Figure 2b, showing A 2 g, B 2g, and A 1 g peaks at 465.5, 437.8, and 361.3 cm −1 , respectively, in agreement with the literature. [33,51] To identify the crystal orientation of the transferred BP flake, we performed a polarization-dependent Raman spectroscopy, and the corresponding spectrum of A 2 g is shown in Figure 2c. From here, we can see that the maximum intensity of the A 2 g peak falls along the x-axis, corresponding to the armchair orientation, whereas zig-zag orientation is along the y-axis with the lowest intensity. [33,52,53] The crystal orientations are also labeled in the optical microscope (OM) image of the device, as displayed in Figure 2a, along with the schematic for crystal orientation in the inset of the figure.
Next, we investigated the PL spectra of BP on SiO 2 and plasmonic metasurface, respectively, by using an excitation wavelength of 2.5 µm. As shown in Figure 2d, both PL spectra peak at λ = 3.7 µm. As the PL peak wavelength is close to the resonant wavelength of our LSPR, the dipole-dipole coupling between BP and the metasurface occurs, causing the PL intensity measured from BP on the metasurface (red curve) to be lower by approximately twelvefold than BP on SiO 2 (black curve). Such PL quenching behavior can be explained by the FRET effect. [54] In the FRET effect, when the semiconductor is in close contact (≈<10 nm) with the metal, the energy transfer occurs from excited dipoles in the semiconductor to metal nonradiatively and plasmon in metasurface extracted carriers from BP, resulting in quenching in the emission. [55][56][57][58] Moreover, Figure 2. Optical characterization. a) Optical microscope (OM) image of fabricated BP/plasmonic metasurface device with red arrow indicates the orientation along the x-axis (AR-Armchair) and y-axis (ZZ-zig-zag), inset image shows a schematic of molecular orientation with arms chair along the x-axis and zig-zag along the y-axis, the white dash rectangle shows area of metasurface and solid purple line shows area of BP flake. b) Normalized Raman spectroscopy spectra of BP. c) Normalized polarization-dependent Raman for A g 2 peak of BP. d) Normalized steady-state photoluminescence spectra for BP/SiO 2 /Si in black line and BP/metasurface in the red line. e,f) Normalized polarization-dependent photoluminescence for BP/SiO 2 /Si and BP/metasurface, respectively, measured with 2.5 µm incident light wavelength at 10 mW power. The measured data (black dots) is fitted using asin 2 θ + bcos 2 θ function represent by red line curve. PL intensity is maximum when θ = 90° which is along the x-axis of BP.

www.advmatinterfaces.de
to investigate the polarization-dependent excitonic emission behavior of pristine BP and BP/metasurface hybrid structure, the polarization-dependent PL was measured for BP/SiO 2 /Si and BP/metasurface and are shown in Figures 2e,f, respectively. For both cases, the anisotropic PL response of BP is evident, in which the PL for BP shows maximum intensity along the x-axis (armchair direction) and minimum along the y-axis (zigzag direction), which is consistent with our optical absorption measurement. [33] Additionally, to evaluate the polarization efficiency of PL emission from BP and BP/metasurface, the degree of polarizability (DOP) is calculated by [59] Degree of polarizibility DOP Using Equation (1), the DOP for the BP/SiO 2 /Si sample is 87.0% and for the BP/metasurface decreases to 49.4%. The reduced polarizability of the latter is attributed to the dipoledipole coupling between BP and metasurface. The powerdependent PL for these two configurations with two different polarization orientations is shown in Figure S5, Supporting Information, demonstrating that while intensity reduces significantly along the armchair orientation, there is a slight enhancement along the zig-zag orientation for BP/metasurface. In the case of polarization along the armchair direction, the emitted excitons resonate with the Au-disk, causing significant reabsorption than that along the zig-zag direction. This happens due to the symmetry criteria, which let on emission and absorption only parallel to the armchair orientation. [33,34] Therefore, PL will be enhanced along the zig-zag orientation in contrast to the armchair orientation. [59] The enhancement and the quenching of PL along the zig-zag and armchair orientation, respectively, resulted in the low DOP on BP/metasurface.
In order to gain further insight, we conducted polarizationresolved PL scanning of the BP/metasurface device using incident light of 2.5 µm wavelength and the results are shown in Figure 3. In this setup, incident polarization was fixed along the x-axis (armchair direction), and the PL emission scanning was obtained by keeping the detector analyzer along the x-axis and the y-axis, as illustrated in Figures 3a,b, respectively. The corresponding normalized PL intensity line plots extracted along the horizontal dashed line are exhibit in Figure 3c. As shown in Figure 3a, the emission peak is highly localized in the BP/ SiO 2 area, and PL intensity is highest when both incident polarizer and the detection analyzer are along the x-axis, also evident in Figure 3c (black line). The peak intensity reduces significantly when the detection analyzer turns along the y-axis with the highest PL intensity (emission) localized at BP/metasurface region, which is evident in Figure 3b,c (red line). This clarifies that PL emission from BP is the highest due to the higher optical absorption and dipole recombination along the x-axis, as expected. Additionally, emitted PL by pristine BP is highly polarized along the x-axis due to the anisotropic behavior of BP on SiO 2 /Si substrate in contrast to the BP/metasurface hybridization, as shown in Figure 3b.
To take advantage of the exciton-plasmon coupling between BP and plasmonic metasurface, a vertically stacked photodetector device was designed to scrutinize photodetection performance. First, a BP flake with a thickness of 50.6 nm ( Figure S7, Supporting Information) was sandwiched between the bottom and top graphene (Gr b and Gr t ) electrodes with a thickness of 7.2 and 8.5 nm, respectively ( Figure S8a,b, Supporting Information) on two Au electrode patterned SiO 2 /Si substrates, as exhibited in Figure 4a, and the corresponding OM image of the fabricated device is shown in Figure 4b. In this photodetector, the BP flake serves as a photoactive layer, and the optically transparent and highly conductive Gr b (Gr t ) serves as a drain (source). Next, the photo-sensing performance of the designed device was probed by applying biased voltage in the range of +0.7 to −0.7 volts across the drain and source electrodes. The incident light was polarized along the maximum absorption orientation (armchair orientation) of the BP. When a bias voltage was applied across these two (Gr b and Gr t ) electrodes, the photogenerated excitons were dissociated due to strong vertical fields. These disassociated charge carriers drift quickly toward the Gr electrodes in the opposite direction because of the large outof-plane carrier mobility of BP. [24,60,61] The dark current(I dark ) (without light illumination) and the light current(I light ) (with light illumination) responses were measured at different www.advmatinterfaces.de powers for the same 3.7 µm wavelength. The measured drain current (I ds ) is shown in Figure S9a, Supporting Information, manifesting a linear response with applied bias.
The photocurrent (I photo ) was calculated by I photo = |I light − I dark | and I photo with V ds are plotted with different illumination power as shown in Figure 4c, demonstrating higher photocurrent with negative bias (−V ds ) due to asymmetric Schottky junction formed between Gr t -BP and BP-Gr b interface. [38] The photocurrent increases with illumination power due to enhanced populations of photoexcited charge carriers. Figure 4d illustrates the power-dependent photocurrent with V ds fixed at 0, −0.1, −0.3, −0.5, and −0.7 V, showing a sublinear relationship with power due to saturation in the absorption with a higher power. In addition, the 2D photocurrent mapping (measurement setup schematic in Figure S10, Supporting Information) with fixed applied biased voltages of −0.1, 0, and +0.1 volts were also measured and are shown in Figure S11a-c, Supporting Information, demonstrating an efficient photocurrent generation under the reverse bias. The polarization-dependent normalized photocurrent is shown in Figure 4e demonstrating highly polarized (DOP ≈ 84.5%) photocurrent generation with BP flake. To explain the charge carrier drift mechanism, the energy band diagram of BP with two semi-metallic Gr electrodes without and with applied bias voltage (V ds ≠ 0 V) is shown in Figures 4f,g, respectively. When light is incident at BP photodetector, for V ds = 0 V, short-circuit photocurrent is created at the junction interfaces, and under a reverse bias voltage (V ds < 0 V), photocurrent increases with the bias voltage. This implies that bias-induced band-bending facilitates the drift of the photocarrier and the separation of electron-hole pairs. [38] Afterward, the same BP flake with graphene electrode was transferred to the designed plasmonic metasurface with Au electrodes to investigate the advantage of the exciton-plasmon coupling between BP and plasmonic metasurface. The schematic of the designed Gr t /BP/Gr b /plasmonic metasurface photodetector is shown in Figure 5a. The corresponding OM image of the fabricated photodetector is shown in Figure 5b. In this photodetector, the BP flake is a photoactive layer, the optically transparent and highly conductive Gr b (Gr t ) serves as a drain . e) Polarization-dependent normalized photocurrent for BP/SiO 2 /Si device with the degree of polarizability (DOP) of 84.5%, Band diagram for Gr t /BP/ Gr b /SiO 2 /Si: f) at V ds = 0 volt, short-circuit photocurrent is produced at the junction interfaces, and g) under a reverse bias voltage (V ds < 0 V) with MIR (3.7 µm wavelength) illumination, due to the band bending caused by the field, the whole BP channel can produce photocurrent.

www.advmatinterfaces.de
(source), and the plasmonic metasurface enhances light-matter interaction. Following that, the photo-sensing performance of the designed device was probed by applying biased voltage in the range of +0.7 to −0.7 volts across the drain and source electrodes. The linear polarized light was incident with polarization along the armchair direction of the BP. The light/dark current responses were measured at different powers for the same 3.7 µm wavelength. The measured drain current (I ds ) with applied bias V ds is shown in Figure S9b, Supporting Information, demonstrating a linear response with applied bias.
The photocurrent (I photo ) was calculated by the previously discussed formula and is shown in Figure 5c with applied bias (V ds ) of −7 to +0.7 volts. The photocurrent with the incident power at various fixed applied biased voltages V ds as displayed in Figure 5d, demonstrates enhanced photocurrent as compared with BP/SiO 2 /Si photodetector. The 2D photocurrent mapping with fixed applied biased voltages of −0.1, 0, and +0.1 V were also measured and are shown in Figure S11d-f, Supporting Information, demonstrating an efficient (approximately twofold) photocurrent generation under the reverse bias. In addition, the polarization-dependent photocurrent was also probed and is shown in Figure 5e, illustrating its DOP was reduced as compared with the BP-based photodetector. The DOP of normalized photocurrent for BP/metasurface was found to be 47.6%, which is in accordance with our polarization-dependent PL measurement result illustrated in Figure 2f. Moreover, to estimate the photodetector response speed, the transient photocurrent response at fixed incident power of 8.5 µW and fixed bias of V ds = −0.7 V was measured and is displayed in Figure S12, Supporting Information. Furthermore, to examine the photo response speed, we extracted the rise time ( r ) and fall time ( f ), as shown in Figures 5f,g, respectively. r is defined by the time it takes for the maximum photocurrent to reach from 10% to 90%, whereas the f is the duration when the maximum photocurrent reduces from 90% to 10%. [62] The calculated r and f were found to be 64 and 167 ns, respectively, which are comparable with the other photonic-based photodetectors. [41] Finally, to evaluate the photodetection performance of the vertically stacked photodetector device, the Equations (2)-(4) were used to calculate the figure of merits, for example, responsivity (R), the external quantum efficiency (EQE), and specific www.advmatinterfaces.de detectivity (D*) of pristine BP/SiO 2 /Si and BP/metasurface photodetector, [62][63][64][65] photo R I P = (2) here I photo , P, R λ , q, h, c, λ, A, B, and NEP are photocurrent, illumination power, responsivity at wavelength λ, charge, plank constant, speed of the light, wavelength, active area of photodetector, bandwidth, and noise equivalent power, respectively. NEP can be expressed as Equation (5); here, I N is the noise current, and R is the responsivity of the photodetector. Further, I N is defined as where, e is the electron's charge, I D is the dark current, and B is the bandwidth.
The computed I photo , R, and EQE with illumination power for both BP/SiO 2 /Si and BP/metasurface are displayed in Figure 6 (refer to CS1, Supporting Information for calculation details). From Figure 6a, a significant photocurrent enhancement is observed in BP with metasurface in contrast to only BP-based photodetector. This increase in photocurrent is, of course, the result of increased optical absorption, intensified light-matter interaction, and strengthened near-field coupling between the BP, and plasmonic metasurface, which in total increase the population of excited excitons in the BP. The enhancement in the photocurrent of BP/metasurface compared with BP/ SiO 2 /Si substrate can also be understood by the PL quenching due to the FRET effect induced by plasmon-excitons coupling, [55][56][57] demonstrated in Figure 2d. As a result, we obtained the maximum R of 495.85 mAW −1 for BP with a plasmonic metasurface-based photodetector, in contrast to 121 mAW −1 only for BP/SiO 2 /Si-based photodetector, as shown in Figure 6b. The calculated R with illumination power at various fixed applied biases (V ds ) for BP/SiO 2 /Si and BP/plasmonic metasurface is shown in Figure S13a,b, Supporting Information, respectively, manifesting the enhanced R with larger V ds .
Another key parameter, the EQEs for BP/SiO 2 /Si and BP/ metasurface photodetector were found to be 4.0% and 16.61%, respectively, at the illumination power of 8.55 µW. It is also observed that the responsivity for both devices decreased with increasing the illumination power, as shown in Figure 6b. The decrease in the responsivity with an increase in the incident illumination power is caused by the saturation of optical absorption, the screening of the field by the photoexcited carriers, and enhanced the carrier scattering rate. [38,66,67] A similar trend is also seen in EQE (Figure 6c). Next, we have calculated the specific detectivity (D*) for BP/plasmonic metasurface photodetector at the illumination power of 8.55 µW at −0.7 V and found to be 6.714 × 10 7 cm Hz 1/2 W −1 (Jones). The plasmonic metasurface also passivates the trap state of the semiconductor, which forbids the charge carrier trapping and contributes to photocurrent enhancement. [54,68] In the last, to compare the photoelectric performance of our designed photodetector with other reported BP-based MIR photodetectors, a comparison table was prepared as shown in the Table 1, and observed that our BP/plasmonic metasurface-based hybrid photodetector shows highest responsivity and speed as compared to the other BP/plasmonic-based photodetector.

Conclusion
In conclusion, we demonstrated a highly sensitive 2D BP photo detector functionalized by a plasmonic metasurface. In this hybrid photodetector, we have compared the optical properties, for example, photo luminescence emission, of the pristine BP/SiO 2 /Si with the BP/plasmonic metasurface. The maximum responsivity and EQE of 495.85 mAW −1 and 16.61% for the BP/plasmonic metasurface were achieved, while 121 mAW −1 , and 4% for the pristine BP/SiO 2 /Si photodetector, respectively, at an applied bias voltage of −0.7 V for 3.7 µm laser of 8.55 µW Figure 6. a-c) Photocurrent, responsivity, and EQE for BP/SiO 2 /Si and BP/metasurface with varied illumination power, at a fixed applied biased voltage of −0.7 volts for the incident wavelength of 3.7 µm. www.advmatinterfaces.de illumination power. This BP/plasmonic metasurface-based hybrid photodetector shows a four-fold enhancement in each responsivity and EQE, respectively, in contrast to the BP/SiO 2 /Si photodetector. The BP/plasmonic metasurface hybrid photodetector performs better than the pristine BP/SiO 2 /Si photodetector due to the strong light-matter interaction between BP and the plasmonic metasurface induced by exciting LSPR. This BP/plasmonic metasurface-based photodetector also shows ultrafast operation speed up to the order of nanoseconds. This demonstrated that the BP/plasmonic metasurface hybrid photodetector opens a new opportunity for optoelectronic applications in the MIR region.

Experimental Section
Plasmonic Metasurface Fabrication: The electron beam lithography technique was performed to fabricate a Au disk-based plasmonic metasurface structure. The flow chart of the fabrication process is illustrated in Figure S3, Supporting Information. First, SiO 2 of 285 nm was thermally grown on p+ doped silicon (Si). SiO 2 /p + -Si substrate was cleaned with acetone and isopropyl alcohol (IPA) for 5 min, rinsed with DI water for 1 min, then dried with N 2 gas and baked for 5 min on a hot plate at 110 °C for 5 min (Step I). 2. Spin-coating of a layer of photoresist at 1000 and 3000 rpm for 10 and 60 s, respectively then annealing at a hot plate at 180 °C for 2 min (step 2). 3. A mature and sophisticated e-beam lithography consisting of electron beam writing for 110 × 110 µm 2 area was conducted, followed by development using developer solution (step 3). 4. Au/Cr (100/10 nm) was deposited using the e-gun evaporation technique, and finally lift-off process with acetone was conducted (step 4). 5. Spin-coating of a layer of photoresist at 1000 and 5000 rpm for 11 and 35 s, respectively then annealing at a hot plate at 90 °C for 2 min (step 5). 6. A mature and sophisticated UV lithography was conducted using Digital Lithography Projection (DLP) for a 200 × 100 µm 2 area of electrodes and developed for 25 s and rinsed with DI followed by drying with N 2 (step 6). 7. Au/Cr (100/10 nm) was deposited using the e-gun evaporation technique, and finally, the lift-off process was performed using acetone. The OM image of the as-fabricated device is shown in Figure S6, Supporting Information.
Graphene and BP Exfoliation and Transfer Process: The scotch tape exfoliation technique was used to exfoliate high-quality multi-layer graphene (MLG) and BP film. The dry transfer technique [50] was used to transfer exfoliated BP sandwich between the bottom and top graphene electrodes on the fabricated plasmonic metasurface. This transfer process was performed in ambient conditions at room temperature. The fabricated BP-based photodetector was packed into a vacuum chamber for further optical and electrical measurement at room temperature.
Numerical Simulation: The reflectance spectra and near electric field distributions were calculated using the 3D FDTD method with a commercially available Lumerical software package. A periodic boundary condition with a mesh size of 5 nm along the x-y direction and a perfect match layer boundary condition with a mesh size of 2 nm along the z-direction were applied over all the fields, including the gold disk. The plane wave source was used with the normal incidents over 2.5 to 5.0 µm. The frequency-domain field profile was used to calculate electric field distribution along the x-y and x-z planes. The refractive index of Si, SiO 2 , and Au was 3.4699, 1.40, and Ciesielski [69][70][71] was used from reported data about the MIR frequency region.
Structural, Optical, and Optoelectronic Characterization: An FESEM Hitachi SU8010 was used to investigate the structure parameter of the fabricated plasmonic metasurface. To measure the reflectance spectra of the fabricated plasmonic metasurface, a µ-FTIR spectrometer (Vertex 80 V) equipped with an infrared microscope (Bruker Hyperion 2000) in the wavenumber range of 400-8600 cm −1 was used. The scanning reflectance, photoluminescence, Raman spectroscopy, and polarization-dependent photoluminescence of the fabricated photodetectors were performed by the methods used in refs. [30], [38], [41], and [51]. The steady and temporal photoresponse and scanning photocurrent characterization (for setup refer to Figure S10, Supporting Information) of the photodetector was performed by the methods used in reference. [38,41]

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.