1. Introduction

Quantum cascade detectors (QCDs) are an appealing alternative to quantum well (QW) infrared photoconductors (QWIP) because they operate at zero bias, and therefore do not suffer from dark current [1–3]. Specifically, they are very promising for high-speed thermal imaging in the 8-12μm range with small-pixel and large-area focal plane arrays [3], and thus they are experiencing fast progresses [4]. In this context the GaN/Al(Ga)N material system [5] has an important merit, due to the large conduction band offset, which provides room for QCDs operating in a record broad spectral range from near-infrared wavelengths as short as 1 μm [6] up to the THz domain (λ = 22-100 μm) [7]. In addition, GaN-based QCDs offer ultrafast operation as demonstrated in the spectral range used for fiber-optics telecommunications [8, 9]. An additional important practical merit of the GaN/Al(Ga)N material system is that it represents available compound semiconductor grown on silicon substrate – opening the possibility of monolithic integration with focal plane array signal processing.

One major drawback of the QCD in applications based on normal light incidence is related to the polarization selection rule of QW inter-subband (ISB) transitions, allowing absorption only for an electric field polarized perpendicular to the QW layers (so called TM polarization) [10]. For most technical applications, planar device geometry and normal incidence illumination are essential. Another deficiency is the relatively small absorption cross section of each QWs (namely ~0.1 to 0.2% per QW) [10, 11], which necessitates a multiple QW structure. Enhancing the QW quantum efficiency by increasing the photonic density of states – potentially achievable by plasmonic nanostructures, is one appealing route to boost the QCD responsivity. Thus, designing and fabricating a dedicated nanophotonic structure embedded in GaN-based QCDs is of tremendous importance for application perspectives.

Normal-incidence operation of ISB detectors is usually achieved by patterning a top grating on the device surface [10, 12]. The function of the grating is to backscatter at a large angle the radiation entering from the rear side (back-side illumination), aligning the polarization more favorably with the required polarization for ISB absorption. Alternative techniques including V-grooves on the mesa facet [10], photonic crystal-QWIP and enhanced QWIP utilizing photonic crystal slab resonator [13, 14] have been demonstrated.

A different effective approach has emerged recently. It makes use of two dimensional metallic nano-holes arrays (MHAs) and generation of surface plasmon polaritons (SPPs) [15–18]. The SPP is bound to the metal/dielectric interface with an evanescent field that decays exponentially away from the interface. Plasmon resonance is observed when a wave vector of the array provides phase-matching between the incident light and the SPP at either metal/dielectric interface. SPP is a TM mode and has the major electric field component normal to the surface. In addition, a carefully designed plasmonic array forms standing waves and produces a lateral cavity effect, which leads to an enhanced transverse plasmonic mode and effective light slowing (in the normal direction). Therefore, if properly coupled, surface plasmons can resonate with electron ISB transitions, and efficiently excite carriers in the QWs to generate a strong photocurrent.

In previous reports [15–19] research was focused on quantum dot infrared photodetectors where the quantum dot symmetry enables normal incidence absorption and the photoresponse enhancement was analyzed only qualitatively. In the work of Bonakdar and Wu et al. [20, 21] on InP based QWIP structure and Zhai et al. [22] on GaAs based QCD structure, the polarization rotation due to MHA is demonstrated experimentally but there is no demonstration of an absolute enhancement of the absorption due to interaction between the plasmonic near field and the QW ISB absorption resonance.

In this work, we demonstrate that incorporation of MHA technology provides a substantial improvement to the responsivity of GaN based alloy QCD device, while allowing for normal incidence illumination. This demonstration paves the way for pixelated imagers operating in a wide spectral range based on nitride materials. The first role of the nanoplasmonic structure is to convert the incoming light into a TM-polarized SPP, which excites the ISB absorption in the QCD device. In addition, we experimentally demonstrate an absolute enhancement of the absorption cross-section per QW attributed to the near-field optical intensity enhancement by the nanoplasmonic structure.

2. Design and device fabrication

SPPs are bound electromagnetic excitations propagating at the interface between a metal and (in our case) a semiconductor [23]. The dispersion relation for a SPP is described by the following relation

To demonstrate the ISB photodetector with enhanced responsivity operating at normal incidence, we have applied the above-described nanoplasmonic structure to a nitride QCD based on a simplified design of the electron extractor [11]. In this QCD structure the 'conventional' multiple QW extractor region is replaced by an AlGaN thick layer, whose composition (kept constant in this layer) is chosen to engineer the internal field and achieve a graded potential thus improving carrier collection, and device performance. This simplified design is robust against thickness fluctuations in the extractor region and offers prospects for ultrafast detectors [11]. In addition, the QCD operates at the important short wave infra-red (SWIR) wavelength range. The details of the structure and QCD performance are described in in Ref. 11. The sample comprising 40 active periods sandwiched between 100 nm thick top and 700 nm thick bottom Si-doped Al_0.4Ga_0.6N contact layers was grown by ammonia molecular beam epitaxy. The 2 nm thick GaN QW with 2 and 1 nm (in growth order) AlN barriers is followed by an extraction region formed by an undoped Al_0.58Ga_0.42N layer. The schematic cross section and conduction band diagram of the structure is shown in Figs. 1(a) and 1(b), respectively.

 

Fig. 1 (a) Schematic cross-section of the alloyed-QCD with MHA on top. (b) Conduction band diagram of two periods of the alloy extractor QCD and squared envelope functions of bound states. (c) Top-view scanning electron microscopy image of the MHA.

Download Full Size | PDF

The operation of the device is as follows. Upon ISB absorption, the photoexcited electrons dominantly tunnel to the right extractor (in growth order). This is permitted by the thinner barrier on the right side of the active QW. Electrons transferred in the extractor stage experience multiple parallel relaxation channels via LO-phonon emission between the numerous bound states towards the ground state of the extractor. They are finally scattered to the ground state of the next period active QW, giving rise to the charge transfer and the photovoltage. In contrast to standard III-V QCD designs based on sequential electron scattering in the extractor region, the fast parallel extraction channels and the presence of a quasi-continuum of states in the extractor region in line with the excited state of the active QW are expected to improve the electron extraction efficiency while minimizing the back-scattering of electrons towards the initial active QW [11]. The devices were first processed in the form of 700x700 μm² mesas on a wafer having a pre-polished 45° wedge facet. Ti/Al/Ti/Au layers were deposited to form the top and bottom contact layers. A transparent window (550 × 550 μm²) at the center of the top surface allows front illumination as shown in Fig. 1(a). The photocurrent spectra were collected by illuminating with a broadband light source coupled to a Bruker Equinox 55 Fourier transform IR (FTIR) spectrometer operated in the step-scan mode.

After full characterization (photo response and I-V measurements) of the mesa QCDs, MHA was fabricated on the top mesa surface using electron-beam lithography, followed by evaporation of Ti (5 nm)/Au (200 nm) and subsequent lift-off process. A passive reference MHA structure was also fabricated to assess the SPP resonance wavelengths. An SEM image of the MHA is shown in Fig. 1(c). The measured diameter of the holes is 580 nm and the period of the array is 835 nm. The diameter and period of the hole array were designed to deliberately achieve a plasmon peak resonance at slightly lower energy than the ISB peak resonance – in order to single-out the net plasmonic effect.

The plasmon electric field distribution through the active region of the device was calculated by a three dimensional finite difference time domain (FDTD) commercial software (Lumerical). Figure 2(a) shows schematic view of a unit cell of the simulated structure. The distribution of ${\left|E_z\right|}^2$ at the peak transmission wavelength, 1.82 μm, of the (1,0) SPP mode in a unit cell of the MHA is illustrated in Figs. 2(b), 2(c) and 2(d), at different planes across the active region of the device. The values on the color bar of Figs. 2(b), 2(c) and 2(d) code the ratio of ${\left|E_z\right|}^2$ compared with ${\left|E\right|}^2=\left|E_x^2+E_y^2\right|$ for a structure without MHA, where E_x and E_y are the in-plane impinging electric field components. The x-y plane in this simulation is at the metal/semiconductor interface (z = 0) in Fig. 2(b) and at the center of the active region (z = 0.5 μm) in Fig. 2(c). The circle in Figs. 2(b) and 2(c) represents the position of hole perforated in the Au film. The x, y values in these figures are relative coordinates to the hole center, as schematically shown in Fig. 2(a) (MHA unit cell) in the x-y plane. Clear enhancement of ${\left|E_z\right|}^2$ in the active region due to SPP’s generation at the vicinity of the metal hole can be seen in Figs. 2(b) and 2(c). In the vicinity of the metal/semiconductor interface, the electric field intensity is enhanced almost by a factor of 50 compared to the incident peak field strength. The depth distribution of ${\left|E_z\right|}^2$ shown in Fig. 2(d) was simulated and monitored starting from the QCD top contact layer/active layer interface (z = 0.1 μm) down through the active region in the x-z plane located at y = 0.21 µm. As the field decays exponentially with the distance from the metal/semiconductor interface, the active QWs region has to be as close as possible to the surface (the Al_0.4Ga_0.6N cap layer thickness is only 100 nm). As illustrated in Fig. 2(d), the z component of the electric field is beamed over more than 1 μm below the metallic film, i.e., covering completely the active region. These results predict that the active region of the QCD strongly overlaps with the longitudinally polarized component of the incident electric field.

 

Fig. 2 FDTD simulated ${\left|E_z\right|}^2$ profile for normal incidence, at the peak wavelength of 1.82 μm. z = 0 is at the interface between the metallic film and the sample. The ratio between ${\left|E_z\right|}^2$ with MHA to ${\left|E_z\right|}^2$, the intensity without the surface plasmonic coupling at normal incidence, is coded by the color bars; (a) A schematic illustration of the simulated structure unit cell; (b) ${\left|E_z\right|}^2$ in the x-y plane at z = 0 (metal/semiconductor interface) at the peak wavelength of 1.82 μm; (c) ${\left|E_z\right|}^2$ in the x-y plane at the center of the active region (z = 0.5 µm). The dashed white line shows the position of the metal hole; (d) ${\left|E_z\right|}^2$ in the x-z plane at y = 0.21 μm and at z = 0.1 μm (contact layer/active region interface) to 0.9 μm for the peak transmission wavelength.

Download Full Size | PDF

3. Experimental results

First we have measured the MHA transmission characteristics on a passive element. Figure 3 depicts the far-field, normal incidence, transmission spectrum of a passive reference sample consisting of Sapphire (300 μm) / AlN (2 μm) covered by the Ti (5 nm)/Au (200 nm) MHA pattern. The measured transmission resonances demonstrate good agreement with the theoretically simulated normal incidence near-field (at the GaN/MHA interface, z = 0) transmission spectrum of a 200 nm Ti/Au MHA over 1μm GaN sample, as shown in the inset of Fig. 3. The three first plasmon resonances are clearly evidenced from both experimental and simulated spectra. Indeed, the measured peak resonance of SPP mode (1,0) is at 1.82 μm, which perfectly matches the designed wavelength. The dips and asymmetric shape of the transmission spectrum of the reference MHA sample are similar to previously reported spectra and are interpreted as the Fano-type interference between the discrete plasmonic resonance and the radiative damping of plasmons due to scattering on the hole arrays [18, 24]. We attribute the background transmission in the far-field measurement of Fig. 3 to non-resonant scattered field through the sub-wavelength holes according to Bethe's solution [24]. The relatively low transmission in the far field measurement is related mainly to Fresnel reflections from both the MHA/air and the Sapphire/air interfaces due to impedance mismatch. In addition and to a lesser extent, part of the losses are due to absorption in the metal hole mask and in the Sapphire/AlN substrate.

 

Fig. 3 The measured (far field) transmission spectrum of a 200 nm Ti/Au MHA on AlN (1 μm)/Sapphire (300 μm) reference sample. (i, j) are the plasmon mode indices. Inset: Simulated near field transmission spectrum (at the interface GaN/MHA) of 200 nm Ti/Au MHA over 1 μm GaN sample.

Download Full Size | PDF

The spectral response of the QCD device with and without the MHA shown in Fig. 4 was measured at room temperature and zero bias. First, we have measured the photoresponse spectra of the QCD without MHA at normal incidence, and at 45° polished wedge configuration. The spectra were normalized with respect to the effective exposed detector area (700 × 700 μm² for wedge illumination and 550 × 550 μm² due to top mesa metal contact layer for front illumination as shown in Fig. 1(a)). The spectra, without MHA, are peaked at 1.87 μm, as predicted by simulation [11], with a full width at half maximum (FWHM) of 0.3 μm and a relative broadening of Δλ/λ = 0.16. The weak normal incidence response observed in Fig. 4, which is theoretically not allowed, is attributed mainly to slight polarization rotation due to scattering at the mesa walls. The photoresponse spectrum of the same QCD after MHA processing for normal incidence illumination is peaked at 1.82 μm, matching the designed plasmon peak resonance wavelength (SPP mode (1,0) of Fig. 3). It shows a reduced FWHM of 0.16 μm (Δλ/λ = 0.09). This normal incidence photoresponse of the QCD-MHA device represents a huge enhancement compared to the normal incidence photoresponse of the bare device (by a factor of ~30) and a significant enhancement (by a factor of 3) when compared to the bare QCD with 45° light coupling through the wedged facet. This observation is a clear demonstration of the efficient coupling between the SPP E_z electric field and the ISB transition due to the dipole selection rules.

 

Fig. 4 Normalized photocurrent spectral intensities of the alloyed-QCD device with and without MHA at various illumination configurations. The vertical dashed lines indicate the shift in peak position of the device; bare QCD (1.87 µm) and plasmonic enhanced QCD (1.82 µm). The spectra are normalized by the illumination exposed effective area. All measurements were done at zero bias and at room temperature.

Download Full Size | PDF

For obtaining the absolute values of the responsivity and detectivity, we have measured spectra and DC dark current-voltage characteristics in a wide range of temperatures using a closed cycle He cryostat and HP semiconductor analyzer. The peak responsivity of the QCD was estimated using a super continuum tunable laser source which was focused at normal incidence on the top surface. It was found that the absolute normal incidence responsivity of the MHA-QCD at the peak wavelength of 1.82 μm is ~2.6 mA/W at room temperature. The measured value at 45° multipass waveguide geometry, for absolute responsivity of the QCD without MHA at its peak wavelength of 1.87 μm, is 0.7 mA/W. These numbers are consistent with the peak ratio of the photoresponse spectra shown in Fig. 3. The performance of photodetectors is usually expressed by the specific detectivity D*. For a QCD at temperatures above the so-called background limited infrared performance temperature T_BLIP, the noise current $i_n=i_n^J$ and D* = $D_J^{*}$ are dominated by Johnson noise. Taking into account the measured R₀A vs. temperature for the QCD and the theoretical D* (BLIP) for 1.8 µm detector (~5*10¹³ Johns) the T_BLIP for this device is below 50K and we get for the detectivity the following equation:

4. Discussion and analysis

The objective of the following discussion is to demonstrate that the polarization rotation induced by the MHA is by far insufficient to account for the observed strong responsivity increase in MHA-QCDs. We attribute this extra-enhancement to the plasmonic induced local field enhancement increasing the actual QW absorption cross section. The photoresponse, R_i, of a QCD can be expressed as [10, 11]:

where hυ is the ISB energy expressed in electronvolts, Φ is the incident photon number per unit time, a – the absorption per one QW, η is the electron extraction efficiency, i.e., the probability of photo-excited electrons to transit into the extractor towards the next active QW. The absorption per one QW is related to the carrier concentration in the ground state and to the light coupling geometry by the following equations:

(5)$$a={\sigma }_{TM}\frac{{\sin }^{2}r}{\cos r}{n}_{2d}$$

(6)$${\sigma }_{TM}=\frac{I_{photo}}{\text{Φ}{n}_{2d}\eta \left(\frac{{\sin }^{2}r}{\cos r}\right)}$$

where σ_TM is the absorption cross section for normally-polarized light, r is the refraction angle inside the active region (34.5° in the Sapphire/AlN multipass geometry) and n_2D is the electron surface concentration in the ground state in the active QW. While the extraction efficiency η and the incident radiation intensity Φ are independent on the illumination geometry, the amount of photons interacting with the absorbing layers is configuration dependent due to the different transmission, coupling efficiency and light path length in the layers. We examine the performance enhancement by the MHA-QCD structure compared to the 45° wedge reference QCD at their peak wavelengths by:

(7)$$\frac{{\sigma }_{TM}\left(MHA\right)}{{\sigma }_{TM}\left(45°\right)}=\frac{I_{photo}\left(MHA\right)T\left(45°\right)\left[{\sin }^2\left(34.5°\right)/\cos \left(34.5°\right)\right]\times 2}{I_{photo}\left(45°\right)T\left(MHA\right)}$$

where I_photo(MHA) and I_photo(45°) are the normalized photoresponse signals respectively, T(45°) is the transmittance through the interface air/Sapphire in 45° wedge device (≈70%, from Fresnel formula), ${\sin }^2\left(34.5\right)/\cos \left(34.5\right)$ is the light coupling efficiency to the ISB at the Sapphire/AlN interface. For the MHA-QCD device at normal incidence we assume 100% coupling efficiency (complete polarization conversion). The transmittance through the MHA, T(MHA), was estimated from the simulations and from the experimental far-field transmission measurements to be in the range of 8% to 25% (Fig. 3). The factor of 2 at the nominator compensates the double pass of incident light in the 45° wedge configuration. Evaluating Eq. (7) we get an additional (on the top of polarization rotation) cross-section enhancement by a factor from 6 to 17 (depending on the estimated transmittance value of the MHA). The underlying physical mechanism explaining this enhancement is the surface nanoplasmonic effective light slowing [25] in the normal direction, or equivalently the enhanced density of states that results in a larger absorption cross section of the QWs [26].

To further validate this result we performed additional direct measurement of the absorption coefficient by measuring the transmission characteristics of the QCD sample. Due to the lack of material, the QCD sample was regrown following exactly the same design, however some minor variations of the layer thickness cannot be excluded leading to a small peak wavelength shift. The sample consisting of a 40 period alloy QCD structure was measured using a FTIR spectrometer in a multipass geometry. The number of passes within the active region is 18. Figure 5 shows the normalized absorbance per pass spectrum, with an absorption peaked at 1.76 μm an FWHM of 0.3 μm (T_S and T_P are the measured multi-pass transmission spectra for P and S polarizations, [log₁₀(T_S/T_P)]). Following the wedge transmission measurement, MHA Ti/Au (560 nm diameter and 800 nm period) was deposited on top surface. The transmission of the sample with MHA was measured under normal incident illumination and normalized by the transmission of same sample without MHA. The absorption peak at 1.8 μm with (0.16 μm FWHM) is depicted in Fig. 5.

 

Fig. 5 Measured absorbance spectra per pass of 40 periods QCD sample at room temperature in 45° wedge waveguide configuration (blue line) and in normal incident illumination on metal hole array (green dashed).

Download Full Size | PDF

The measurements lead to an absorption coefficient of α_45° = 448 cm⁻¹ for the wedge experiment and α_MHA = 1957 cm⁻¹ at normal incidence with MHA, which is a factor of 4.4 larger than α_45°. Since we use the same sample in both measurements, the maximum absorption enhancement ratio that we can expect due to the TM selection rule is$a_{MHA}/a_{45°}=\left[\cos \left(34.5°\right)/{\sin }^2\left(34.5°\right)\right]=2.6$. This simple calculation proves that the polarization rotation taken alone cannot explain the measured absolute absorption enhancement by a factor of 4.4 times compared to wedge transmission measurements without MHA.

We finally validated the absorption enhancement by simulating the absorption coefficient with the FDTD software, where the imaginary part of the refractive index (extinction coefficient) for the ISB transition resonance was extracted from the measured absorption coefficient in wedge configuration. The calculated percentage of the absorbed electromagnetic power in normal incidence, with and without MHA, is 2.9% and 6.2% respectively. It turns out that the MHA antenna increases effectively the absorption in the active region by a factor of 2.1, which confirms qualitatively the experimental results on enhancement of absorption due to the localized plasmon field.

In conclusion, we demonstrated a normal-incident quantum cascade detector excited by surface plasmons resonance using Ti/Au metallic nano-hole arrays integrated on top surface of the detector active region. The MHA rotates the polarization of the incoming radiation in the near field and makes it compatible with the dipole selection rules present in the conduction band of III-nitrides quantum-wells. Light is only detected if the surface plasmon wave mode overlaps energetically with the QWs transitions. We report on a resonant enhancement of the responsivity in comparison to a 45° wedge illumination due to the combined effect of polarization rotation and substantial increase of the absorption quantum efficiency due to enhanced density of photon states induced by the localized plasmon electric field.

Acknowledgments

The Authors acknowledge partial finance support by the EC FET OPEN “Unitride” Grant # 233950” and by the FTA program on Nanophotonics for Detection and Sensing.

References and links

1. L. Gendron, M. Carras, A. Huynh, V. Ortiz, C. Koeniguer, and V. Berger, “Quantum cascade photodetector,” Appl. Phys. Lett. 85(14), 2824–2826 (2004). [CrossRef]  

2. D. Hoffstetter, F. R. Giorgetta, E. Baumann, Q. Yang, C. Manz, and K. Köhler, “Experimental investigation on line width compression of stimulated Brillouin scattering in water,” Appl. Phys. Lett. 93, 221106 (2008).

3. A. Delga, L. Doyennette, V. Berger, M. Carras, V. Trinit, and A. Nedelcu, “Performances of quantum cascade detectors,” Infrared Phys. Technol. 59, 100–107 (2013). [CrossRef]  

4. For a review on III–V QCDs, seeF. R. Giorgetta, E. Baumann, M. Graf, Q. Yang, C. Manz, K. Köhler, H. E. Beere, D. A. Ritchie, E. Linfield, A. G. Davies, Y. Fedoryshyn, H. Jäckel, M. Fischer, J. Faist, and D. Hofstetter, “Quantum cascade detectors,” IEEE J. Quantum Electron. 45, 1039–1052 (2009).

5. M. Tchernycheva, L. Nevou, L. Doyennette, F. H. Julien, E. Warde, F. Guillot, E. Monroy, E. Bellet-Amalric, T. Remmele, and M. Albrecht, “Systematic experimental and theoretical investigation of intersubband absorption in GaN/AlN quantum wells,” Phys. Rev. B 73(12), 125347 (2006). [CrossRef]  

6. S. Sakr, E. Giraud, A. Dussaigne, M. Tchernycheva, N. Grandjean, and F. H. Julien, “Two-color GaN/AlGaN quantum cascade detector at short infrared wavelengths of 1 and 1.7 μm,” Appl. Phys. Lett. 100(18), 181103 (2012). [CrossRef]  

7. F. F. Sudradjat, W. Zhang, J. Woodward, H. Durmaz, T. D. Moustakas, and R. Paiella, “Far-infrared intersubband photodetectors based on double-step III-nitride quantum wells,” Appl. Phys. Lett. 100(24), 241113 (2012). [CrossRef]  

8. A. Vardi, S. Sakr, J. Mangeney, K. W. Kandaswamy, E. Monroy, M. Tchernycheva, S. E. Schacham, F. H. Julien, and G. Bahir, “Femto-second electron transit time characterization in GaN/AlGaN quantum cascade detector at 1.5 micron,” Appl. Phys. Lett. 99(20), 202111 (2011). [CrossRef]  

9. S. Sakr, P. Crozat, D. Gacemi, Y. Kotsar, A. Pesach, P. Quach, N. Isac, M. Tchernycheva, L. Vivien, G. Bahir, E. Monroy, and F. H. Julien, “GaN/AlGaN waveguide quantum cascade photodetectors at ≈ 1.55 μm with enhanced responsivity and ~ 40 GHz frequency bandwidth,” Appl. Phys. Lett. 102(1), 011135 (2013). [CrossRef]  

10. H. C. Liu and F. Capasso, “Intersubband Transitions in Quantum Wells: Physics and Device Applications I”, in Semiconductors and Semimetals (Academic, 2000), Vol. 62, Chaps. 1 and 4.

11. S. Sakr, E. Giraud, M. Tchernycheva, N. Isac, P. Quach, E. Warde, N. Grandjean, and F. H. Julien, “A simplified GaN/AGaN quantum cascade detector with an alloy extractor,” Appl. Phys. Lett. 101(25), 251101 (2012). [CrossRef]  

12. J. Y. Andersson, L. Lundqvist, and Z. F. Paska, “Quantum efficiency enhancement of AlGaAs/GaAs quantum well infrared detectors using a waveguide with a grating coupler,” Appl. Phys. Lett. 58(20), 2264–2266 (1991). [CrossRef]  

13. S. Kalchmair, H. Detz, G. D. Cole, A. M. Andrews, P. Klang, M. Nobile, R. Gansch, C. Ostermaier, W. Schrenk, and G. Strasser, “Photonic crystal slab quantum well infrared photodetector,” Appl. Phys. Lett. 98(1), 011105 (2011). [CrossRef]  

14. S. Kalchmair, R. Gansch, S. I. Ahn, A. M. Andrews, H. Detz, T. Zederbauer, E. Mujagić, P. Reininger, G. Lasser, W. Schrenk, and G. Strasser, “Detectivity enhancement in quantum well infrared photodetectors utilizing a photonic crystal slab resonator,” Opt. Express 20(5), 5622–5628 (2012). [CrossRef]   [PubMed]  

15. S. C. Lee, S. Krishna, and S. R. J. Brueck, “Quantum dot infrared photodetector enhanced by surface plasma wave excitation,” Opt. Express 17(25), 23160–23168 (2009). [CrossRef]   [PubMed]  

16. S. Hellström, Z.-H. Chen, Y. Fu, M. Qiu, R. Soltanmoradi, Q. Wang, and J. Y. Andersson, “Increased photocurrent in quantum dot infrared photodetector by subwavelength hole array in metal thin film,” Appl. Phys. Lett. 96(23), 231110 (2010). [CrossRef]  

17. C.-C. Chang, Y. D. Sharma, Y.-S. Kim, J. A. Bur, R. V. Shenoi, S. Krishna, D. Huang, and S.-Y. Lin, “A Surface Plasmon Enhanced Infrared Photodetector Based on InAs Quantum Dots,” Nano Lett. 10(5), 1704–1709 (2010). [CrossRef]   [PubMed]  

18. P. Vasinajindakaw, J. Vaillancourt, G. Gu, R. Liu, Y. Ling, and X. Lu, “A Fano-type interference enhanced quantum dot infrared photodetector,” Appl. Phys. Lett. 98(21), 211111 (2011). [CrossRef]  

19. A. Vardi, G. Bahir, S. E. Schacham, P. K. Kandaswamy, and E. Monroy, “Photocurrent spectroscopy of bound-to-bound intraband transitions in GaN/AlN quantum dots,” Phys. Rev. B 80(15), 155439 (2009). [CrossRef]  

20. W. Wu, A. Bonakdar, and H. Mohseni, “Plasmonic enhanced quantum well infrared photodetector with high detectivity,” Appl. Phys. Lett. 96(16), 161107 (2010). [CrossRef]  

21. A. Bonakdar and H. Mohseni, “Impact of optical antenna and plasmonics on infrared imagers,” Infrared Phys. Technol. 59, 142–145 (2013). [CrossRef]  

22. S.-Q. Zhai, J.-Q. Liu, F.-Q. Liu, and Z.-G. Wang, “A normal incident quantum cascade detector enhanced by surface plasmons,” Appl. Phys. Lett. 100(18), 181104 (2012). [CrossRef]  

23. For review see, C. Genet and T. W. Ebbesen, “Light in tiny holes,” Nature 445(7123), 39–46 (2007). [CrossRef]   [PubMed]  

24. C. Genet, M. P. van Exter, and J. P. Woerdman, “Fano-type interpretation of red shifts and red tails in hole array transmission spectra,” Opt. Commun. 225(4-6), 331–336 (2003). [CrossRef]  

25. G. Rosenblatt, E. Feigenbaum, and M. Orenstein, “Circular motion of electromagnetic power shaping the dispersion of Surface Plasmon Polaritons,” Opt. Express 18(25), 25861–25872 (2010). [CrossRef]   [PubMed]  

26. L. Novotny and N. van Hulst, “Antennas for light,” Nat. Photonics 5(2), 83–90 (2011). [CrossRef]