1. Introduction

Infrared imaging technology has made steady technological progress over the last forty years [1,2] and many scientifically relevant applications such as thermal medical diagnostics [3] and remote sensing technologies [4] have emerged. In the late 1970s the first generation of infrared detectors entered production and consisted of scanning linear arrays [1], while the second-generation of infrared imagers consisted of two-dimensional arrays and reached full-scale production in the late 1980s and early 1990s [1]. Much of the successful growth of infrared imaging technology can be attributed to dramatic advances in the analysis and processing of infrared sensor manufacturing [1]. Researchers are now developing third-generation infrared systems which call for a larger number of pixels, faster frame-rates, higher sensitivity, elevated operating temperatures and multicolor functionality [2]. Whereas increasing the number of pixels is more-or-less straightforward, achieving higher frame-rates, enhancements to the sensitivity, and multicolor detection are more challenging.

In the past few years, there has been a concerted effort to create a new generation of infrared detectors with significant improvement in dark current and noise performance using antimonide-based III-V materials to compete with currently deployed state-of-the-art mercury cadmium telluride (MCT) detectors [2]. Various type-II superlattice (T2SL) architectures such as nBn [5–7], CBIRD [8], W-structure [9], M-structure [10], N-structure [11] and pBiBn [12] detectors have been demonstrated. Recently, the T2SL interband cascade design has been used to demonstrate room temperature single pixel detectors [13–18] and high operating temperature focal plane arrays (FPAs) in the mid-wave infrared [19]. However, the epitaxial layers are very thick, requiring long growth times, and require many epitaxial growth calibration steps. In addition, there is a trade-off between thick absorbers needed for high quantum efficiency and thin absorbers needed for low dark current. Beyond some critical photodetector thickness, the signal-to-noise ratio for a photodetector decreases because dark current begins to dominate [20].

Materials utilized in third-generation infrared imagers are typically responsive over a broad spectral range (2-20$\mu$m) and realize low dark current [1]. Recently, bias-selectable multicolor infrared detectors, formed by co-located broadband detectors, have been demonstrated [21–24]. However, narrowband spectral information is typically required to accurately identify objects in the infrared spectrum. Imagers therefore use filter wheels or dispersive elements to acquire spectral information [1,20]. These additional components make a multicolor third-generation focal plane array (FPA) more complex, heavier and larger, limiting their applicability to portable applications. These additional components also introduce temporal and/or spatial registration inaccuracy and reduce system sensitivity [2,20]. Eliminating the need for these components would improve sensitivity and frame rate.

Though the challenges of achieving a third-generation multicolor infrared imaging system are substantial, there are many potential applications of great interest for medical, scientific and military applications. One example is imaging in the thermal infrared, which is plagued with difficulties in measuring and eliminating emissivity effects to obtain the absolute temperature of an object [25]. The advantage of a multispectral detector is the ability to indirectly calculate the absolute temperature of an object according to the principles of a two-color pyrometer. A two-color pyrometer compares two closely spaced, narrowband intensities where the permittivity is approximately constant [25]. Moreover, a multicolor system may have the potential to process image data at the pixel level – with the aid of onsite analog electronics – thereby yielding compressed and classified images directly with similar function to the retina of the human eye [26].

One approach to narrowband infrared detectors is to couple incoming radiation to the detection material using a resonant structure. Recently, a thin metal film perforated with a two-dimensional hole array has been combined with quantum well and quantum dot infrared photodetectors to provide pixel-level optical filtering and signal enhancement [27–32]. Resonances are observed when diffracted wave vector from a metallic photonic crystal provides phase-matching between the incident photon and surface plasmon waves at the metal/dielectric interface [33–35]. The emphasis in much of this work has been on coupling normally incident radiation into surface plasmon waves with quantum well and quantum dot infrared photodetectors that require in-plane radiation due to their absorption selection rules. However, metallic photonic crystals require low loss dielectric layers which significantly limits the types of detectors that can be utilized. This requirement rules out MCT and antimonide-based III-V infrared detectors that possess relatively large absorption coefficients. In addition, quantum well and quantum dot detectors are typically a few microns thick to take advantage of the evanescent field bound to the metal/dielectric interface [27–35].

In detector applications, it is advantageous to focus all incoming radiation into a deep subwavelength volume of the detector material to achieve high quantum efficiency and low dark current. We have built and tested an ultra-thin ($~\lambda /15)$ multispectral infrared metamaterial detector with a broadband state-of-the-art antimonide-based III-V infrared photodetector. Although some progress has been made in device fabrication and simulations [36-37], to the best of our knowledge, this is the first demonstration of this type of photodetector. In our approach, a metamaterial absorber concentrates incoming radiation over a region of the spectrum where Fabry-Perot resonance would normally dominate the scattering and necessesitate the use of anti-reflective layers. Because the metamaterial design can be adjusted to match any target infrared wavelength, this work demonstrates the ability of metamaterials to operate in any part of the infrared spectrum, with a broad range of detector materials and thicknesses, for enhanced infrared imaging applications.

2. Design and optical properties

The metamaterial design methodology enables a novel approach to engineer the scattering of electromagnetic radiation through modification of the real and imaginary parts of the effective permittivity $\text{ε}\left(\text{ω}\right)$ and permeability $\text{μ}\left(\text{ω}\right)$ [38–40]. For example, the metamaterial absorber permits construction of optical materials that are impedance matched to free space, thus minimizing the reflectance, and drive the transmission to zero [41,42]. Metamaterial absorbers typically consist of a metallic ground plane, a deep subwavelength lossy dielectric layer and a geometrically patterned metallic top layer that is designed to strongly couple to a uniform incident electric field [43]. Altering the geometry of the metallic pattern and dielectric thickness enables tuning of the effective material parameters allowing for strong absorption at nearly any desired frequency with an incredibly high degree of spectral selectivity with nearly any dielectric material [42].

Our metamaterial detector, with an integrated interband cascade photodetector, is depicted in Fig. 1. The key innovation in this effort is coupling the metamaterial absorber to an ultra-thin state-of-the-art infrared photodetector. The metamaterial absorber structure strongly focuses incident IR radiation into the active region of the photodetector, and thus the signal-to-noise ratio can be dramatically increased at any desired wavelength. The enhanced signal-to-noise is empowered by several key features: (1) reducing the electrical volume, that could reduce the detector dark current down to a fraction of that for a conventional photodetector, (2) a deep sub-wavelength metamaterial pattern array on a pixel significantly improves photodetector absorption efficiency and (3) spectral filtering is achieved for any desired wavelength with a single photolithography step. Through this effort, an ultra-thin multispectral infrared metamaterial detector can achieve higher frame rates, increased sensitivity, and multicolor detection.

 

Fig. 1 Metamaterial interband cascade detector bonded to a silicon fan-out chip with an indium bump and an illustration of the heterostructure of the interband cascade detector above a metallic ground plane and below a metallic array of squares for resonant quantum efficiency enhancement.

Download Full Size | PDF

There are several challenges associated with this effort. The first is to match the absorptive properties of the detector with the metamaterial geometry for maximum optical absorption. The second is to minimize the dissipative metallic losses and to ensure most absorption occurs in the active region of the infrared photodetector. Although metal loss in the infrared is much lower than in the visible, simulations indicate it may be as high as 20%, thus limiting the quantum efficiency (QE) of the detector to 80%. Lastly, we desire an ultra-thin detector to maximize performance, and thus strive for a thickness that is 15 times smaller than the detection free-space wavelength. Here we investigate a multispectral infrared detector array fabricated in a similar process to that used for commercial infrared focal plane arrays (Appendix A). Our metamaterial pattern is 100$\mu$m × 100$\mu$m with a pixel size of 120$\mu$m × 120$\mu$m. The pixel array is flip-chip bonded to a silicon fan-out used to provide an electrical connection to each pixel. A range of different metamaterial patterns are fabricated for each pixel in the array to experimentally tune the absorptive properties of the metamaterial structure.

The dielectrics in our metamaterial detector fall into three groups: an InAs/GaSb T2SL active detector region, non-active contact layers, and unipolar barrier layers. The relatively thin unipolar barriers are implemented to block excess dark current. In the ultra-thin T2SL photodetector, an ultra-fast carrier relaxation channel is introduced for efficient photo-generated carrier extraction [13–19]. The InAs/GaSb T2SL material is characterized by a broken-gap type-II alignment, with electrons and holes localized in InAs and GaSb layers, respectively. The overlap of electron (hole) wave functions between adjacent InAs (GaSb) layers results in the formation of an electron (hole) mini-bands in the conduction (valence) band with a high, broadband absorption coefficient [44,45]. In our study, we used molecular beam epitaxy to fashion a single-stage interband cascade detector (Fig. 1) with a 260nm T2SL active region sandwiched between a hole barrier (59nm) and an electron barrier (34nm). A long wavelength cutoff of 9.5$\mu$m in our T2SL detector material is determined by the highest hole (heavy-hole) and the lowest conduction bands, which can be tuned by varying the thickness of the InAs and GaSb layers, respectively.

We first modeled our infrared pixels with CST Microwave Studio to determine the relationship between the metamaterial detector geometry and its effect on total absorption and – perhaps more importantly – the spatial dependence of absorbed power. To simulate the interband cascade photodetector, we modeled the doped contact and cascade semiconductor layers with a Drude model, which can accurately describe the optical properties of a semiconductor, so long as incoming radiation is below the bandgap [46-47]. The complex dielectric constant of the T2SL in the active region of the photodetector is estimated based on prior refractive index measurement and absorption coefficient measurement work [48] and adjusted based on our experimental measurements. Multiple metamaterial absorber designs were chosen for both the mid-wave infrared (MWIR) and long-wave infrared (LWIR) regions of the electromagnetic spectrum. In all structures studied, we kept the metamaterial pitch (p) constant for the mid-wave and long-wave designs, (2.0µm and 2.5µm respectively) while the width of the element (a) was varied from 1.65µm to 1.80µm and from 2.10µm to 2.35µm in 0.05µm steps, respectively. Simulations indicate that the main absorption is obtained from a narrow second order resonant peak for all designs.

In Figs. 2(a)-2(b) we show the mid-wave and long-wave simulated spectral response from 4$\mu$m to 10.5$\mu$m of the metamaterial detector design. A maximum absorptance of 79% at 6.1$\mu$m is obtained for the mid-wave design, and 78% at 7.2$\mu$m for the long-wave (red curves). The wavelength dependent absorptance was experimentally determined (Figs. 2(c)-2(d)) using an infrared microscope and Schwarzschild objective with a numerical aperture of 0.4 (15x magnification). The reflectance was measured for each sample and the absorptance was calculated by normalizing the spectra with respect to a reference gold mirror, (due to the thick gold ground plane, zero transmission was assumed). We find excellent agreement between the simulated and experimental measurements as is evident from Fig. 2.

 

Fig. 2 Numerical simulation (a), (b) and experimental measurement (c), (d) of the absorptance for the mid-wave (a), (c) and long-wave (b), (d) infrared metamaterial detector designs with the pitch for the individual metamaterial elements in each design kept constant (2.0 µm and 2.5 µm respectively) while the width of the square pattern was varied in 0.05 µm steps.

Download Full Size | PDF

3. Quantum efficiency measurement

We next turn toward mid-wave and long-wave characterization of the detector material both with and without the metamaterial absorber structure. The quantum efficiency of the metamaterial detectors is shown in Fig. 3 as narrowband absorption curves with varying geometry and was measured at a temperature of 77K with an applied bias of −100mV. A peak QE of 18.5% is obtained at a wavelength of 6.14μm for dimensions of p = 2.0μm and a = 1.65μm. We find excellent agreement of the wavelength dependent QE for the metamaterial detectors, compared to the spectral absorptance measurements shown in Fig. 2. In addition, the peak QE for both mid-wave and long-wave designs gradually increases as the width of the metamaterial element is tuned, in accord with the gradual increase of the detector materials increase in QE for shorter wavelengths. In order to underscore advantages provided by the metamaterial, we also characterized two different reference detector materials – a conventional relatively large (400μm x 400μm) pixel (here termed Reference 1), and an unpatterned detector without a top metamaterial metal layer (Reference 2). Reference 1, shown as the dashed gray curve in Fig. 3, verifies that our T2SL material achieves relatively broadband detector response across much of the range measured, with a cutoff near 10.5μm. The other detector materials characterized (Reference 2 shown as the black curve in Fig. 3) demonstrates that we are operating in a regime of wavelength and thickness, in which a strong Fabry-Perot resonance would normally dominate the spectral and QE responses.

 

Fig. 3 Experimental measurement of the quantum efficiency for the mid-wave and long-wave infrared metamaterial detector designs with the pitch for the individual metamaterial elements in each design kept constant (2.0 µm and 2.5 µm respectively) while the width of the square pattern was varied in 0.05 µm steps. Also shown are two reference detectors, the black (Fabry-Perot) and dashed gray (Substrate Pixel) curves, discussed in detail in the text.

Download Full Size | PDF

In order to quantify the increased performance the metamaterial absorber structure has on the T2SL detector material, we define the enhancement factor (EF) as the peak metamaterial detector QE, (max[QE_META(λ₀)]) divided by the QE of the Reference 1 detector at the same wavelength QE_Ref(λ₀), i.e. EF = max[QE_META(λ₀)]/QE_Ref(λ₀). For example, the MWIR design (red curve in Fig. 3) realizes a peak QE of 18.5% at a wavelength of 6.14μm, in contrast the Reference 1 detector realizes a QE = 6.1% at the same wavelength. We thus determine that our metamaterial design boosts the QE by a factor of EF_6.14 = 3.05. We find similar improvements for the LWIR devices where the QE increases from 4.1% to 14.7%, for the design peaked at 7.13μm yielding EF_7.13 = 3.6. In Table 1 we provide the enhancement factor for each metamaterial detector characterized as a function of metamaterial geometry. We note the appearance of secondary absorption peaks at longer wavelengths not present in the quantum efficiency measurement. We attribute these peaks to angle dependent modes of the structure, which would not be observed in the normally incident quantum efficiency measurements.

Table 1. Percentage of power absorbed in various portions of the metamaterial structure and detector efficiencies. A_total, QE and EF are obtained from measurements. The use of models to determine the other efficiencies is discussed in the text.

View Table

It is important to note that although there have been numerous experimental studies that have verified the possibility of near unity absorption with metamaterials [41–43], this is a necessary but not sufficient condition for “perfect detection,” i.e., a QE of 100%. For perfect detection, incident energy should be entirely dissipated only in the detector region, as opposed to in the metallic or doped contact portions of the metamaterial. We performed simulations to estimate the fraction of power absorbed in various portions of the metamaterial structure and the results are summarized in Table 1. If we consider the MWIR metamaterial detector which has a peak spectral absorbance of 85.6% at 7.13$\mu$m, we find that 48.1% of the incoming radiation is dissipated in the T2SL active detector layer, with the Ohmic and dielectric loss in the non-active regions accounting for 36.3% and 1.22%, respectively. Our direct measurement of the QE then allows us to estimate the conversion efficiency, which we calculate as ${\text{η}}_{\text{conv}}=\text{QE}/{\text{A}}_{\text{active}}$. We thus find that the T2SL active detector layer has a conversion efficiency of ${\text{η}}_{\text{conv}}=$38.7% and a recombination rate of ${\text{η}}_{\text{recomb}}=$61.3% at a wavelength of 7.13$\text{μ}$m. We summarize the absorption percentages and T2SL efficiencies for all infrared metamaterial detectors studied here in Table 1.

Even though only 38.7% of the absorbed energy is detected, the specific detectivity (D*) of our metamaterial detector realizes an increase over the value obtained from the reference detector. We calculated D* using the QE measurement results shown in Fig. 3, where both noise and shot noise were taken into consideration. The peak detectivity calculated for the metamaterial detector pixel at a wavelength of 6.14$\mu$m had a value of 6.06 × 10¹⁰ cmHz^1/2/W for the MWIR designs and 7.13µm with a value of 5.60 × 10¹⁰ cmHz^1/2/W for the LWIR designs. The calculated detectivity for the reference detector is 1.99 × 10¹⁰ cmHz^1/2/W at 6.14$\mu$m and 1.56 × 10¹⁰ cmHz^1/2/W at 7.13µm.

4. Discussion

We next highlight advantages of the metamaterial detector over a more conventional infrared detector. Conventional infrared detectors utilize color filter wheels to obtain narrowband spectral information, which introduces additional insertion loss. However, our approach shown here permits the response of the detector to be enhanced, since the metamaterial provides efficient spectral information, i.e. only out of band radiation is reflected away. Further the metamaterial enables field concentration and detection all with the same structure. We expect further gains in the enhancement factor are possible through refinements in detector and metamaterial design, as well as more precise device fabrication. The multispectral infrared metamaterial detector architecture presented here eliminates the complex optics network, with moving filters or dispersive elements, utilized by other approaches [1,2], while avoiding spatial and temporal registration problems inherent in fabrication, thereby decreasing the size, weight, power, and cost of the imaging system. Additionally, the spectral response of an individual pixel can be tuned to any part of the infrared spectrum for pixel-level image processing at the sensor, and within the read-out integrated circuit (ROIC), to enable the sensor to output “compressed and important information” directly. Lastly, we note that the optical properties of our detector material would normally be dominated by the Fabry-Perot peak at 6.75μm – leading to a significant increase in the materials effective absorption coefficient. However, the flexibility of metamaterials allows us to compensate for this drastic change in the material properties and to achieve relatively narrow-band absorption peaks at discrete wavelengths across both the mid-wave and long-wave infrared regimes.

5. Conclusion

In conclusion, we have demonstrated an ultra-thin infrared detector using a metamaterial absorber design for multicolor imaging applications. Varying the metamaterial geometries enabled demonstration of a detector chip with multicolor capability on an ultrathin broadband InAs/GaSb T2SL detector. Instead of an attenuation of the signal (as in the case of conventional filters), the QE was enhanced compared to the reference detector. At T = 77K, the QE was increased from 6.1% to 18.6% (3.1x) at a peak operating wavelength of 6.14μm and from 4.1% to 14.7% (3.6x) at a peak operating wavelength of 7.13μm. We achieved a specific detectivity at 6.14μm of 6.06 × 10¹⁰ cmHz^1/2/W for the MWIR design and a value of 5.60 × 10¹⁰ cmHz^1/2/W at 7.13 µm. We note that it is straightforward to design the metamaterial for higher absorption—absorption above 90% has been demonstrated [41–43]—but designing an absorber that optimally transfers that energy to the detector will be necessary to improve detectivity. This study also confirms that higher quality production of the T2SL detector material, to reduce electron and hole recombination losses, will also improve detectivity.

Appendix A metamaterial detector fabrication

The fabrication process of the infrared detectors begins by first defining a 500nm beryllium-doped (2e18cm⁻³) InAsSb etch stop layer and then the single stage interband cascade detector heterostructure (illustrated in Fig. 1) on a tellurium doped 2” GaSb substrate in a solid source molecular beam epitaxy (MBE) system. The device is composed of a 15nm silicon-doped (2e18cm⁻³) n-type InAs top contact layer, a beryllium-doped (1e16cm⁻³) 14ML InAs / 7ML GaSb T2SL absorber (260nm-thick) sandwiched between two unipolar barriers, and a 50nm beryllium-doped (2e18cm⁻³) GaSb bottom contact. The electron unipolar barrier (e-barrier) is undoped and composed of a 34nm AlSb / GaSb T2SL and the hole unipolar barrier (h-barrier) is undoped and composed of a 59nm InAs / AlSb T2SL. Prior to detector growth, the MBE system was calibrated for each detector region separately to ensure the growth conditions are close to optimum and to maintain lattice matching throughout the whole structure. During the MBE growth, the oxides desorption and buffer growth processes were monitored by their reflection high-energy electron diffraction (RHEED) patterns. For a detailed analysis of these detector architectures, see references [12–18].

To verify the longwave infrared interband cascade detector design and MBE growth, a small piece from this wafer is processed into large 400μm x 400μm single-pixel devices with a circular 300μm diameter optical aperture in the top contact. Conventional photolithography and phosphoric acid based wet chemical etching were used to define the device mesa; 100nm thick silicon dioxide layer was used for the side-wall passivation and isolation, and a 5nm adhesion layer of titanium and a 300nm contact layer of gold was deposited for both top and bottom Ohmic contacts. Wire bonding to the top and bottom contact layers was used to provide an external electrical connection for device characterization. Fig. 4 is a schematic representation of the fabrication process of substrate pixels. The detectors were then mounted into close-cycle helium-cooled cryostat for low-temperature characterizations.

 

Fig. 4 Schematic representation of the fabrication process a large 400μm x 400μm substrate pixel.

Download Full Size | PDF

The metamaterial detector test chip consists of a 6 x 6-pixel array with a similar fabrication process to an infrared focal plane array (FPA) to define 120μm x 120μm pixels with an additional step to define a metamaterial pattern on the illumination side of each pixel. This pixel array is bonded to a silicon fanout to provide electrical access to each pixel in the array. This process begins by defining an array of pixel mesas using standard photolithography and phosphoric acid wet etching down to the etch stop layer. To help reduce the density of surface traps that contribute to detector dark current, the pixels are rinsed in a dilute solution of hydrochloric acid (1 HCl: 1 H₂O) and immediately coated with a 100nm layer of silicon dioxide. A square 110μm x 110μm window is opened in the silicon dioxide layer with an ammonium hydroxide and hydroflouric acid etch solution (6 NH₄F: 1 49% HF in water). A square 115μm x 115μm contact metal containing a 5nm adhesion layer of titanium and a 300nm contact layer of gold is deposited over the opened silicon dioxide window (see Fig. 5). Because this metal contact layer also serves as an electromagnetic ground plane to the metamaterial photodetector, the titanium adhesion layer is kept as thin as possible to reduce the amount of radiation dissipated in this layer.

 

Fig. 5 Schematic representation of the fabrication procedure of the metamaterial detector pixel array.

Download Full Size | PDF

A silicon “fanout” chip, as illustrated in Fig. 6, is fabricated in parallel to the fabrication of the 6 x 6-pixel array. The silicon fanout provides electrical contact to each pixel in the array with metal trace lines with an indium bump on one end connected to a pixel and an external wire bonding pad on the other. The fabrication process for this fanout begins with a silicon substrate with a one micron thick layer of silicon dioxide for electrical passivation. Trace lines are defined on this substrate with a 10nm adhesion layer of titanium and a 200nm capping layer of gold. Next, a 300nm layer of photosensitive SU-8 polymer is deposited with openings for the wire bonding and indium bump bond pads and cured with heat. A layer of capping metal consisting of 100nm of titanium and 100nm of gold is deposited to cap the small opening in the SU-8 layer over the indium bump bond pads. A large indium pad (2$\mu$m thick) is deposited on the capping metal and heated with indium flux to reflow indium into a sphere-like shape. Indium flux reduces surface tension forces so that the attractive forces of indium dominate to form an indium bump. These indium bumps are used to make an electrical contact to each pixel in the array.

 

Fig. 6 Schematic representation of the fabrication procedure of the silicon fanout chip.

Download Full Size | PDF

Now that the pixel array and fanout are defined, their respective wafers are diced, cleaned and prepared for flip-chip bonding. Flip-chip bonding is a process where a chip is flipped, aligned and bonded to another chip to form an electrical connection. Next, the pixel array is flip-chip bonded to the fanout and permanently glued together with epoxy as illustrated in Fig. 7. The capillary force is employed to feed epoxy between the pixel array and the fanout. The GaSb substrate, on the backside of the pixel array, is removed using a chemical mechanical polishing process and the InAsSb etch stop layer is removed with a selective citric acid to hydrogen peroxide (5:1) solution. After substrate and etch stop layer removal, a common metal contact layer is formed on the illumination side of the pixel array with an optical aperture to each pixel. A conventional pixel array typically utilizes a thick (~500nm), highly doped semiconductor layer to form a common contact layer to the pixel array. By utilizing a metal layer with an optical aperture to each pixel, we can design ultra-thin photodetectors with an active region that accounts for most of the detectors volume. This new ultra-thin design allows radiation to be concentrated in the active region of our infrared photodetector.

 

Fig. 7 Schematic representation of the fabrication procedure for final device assembly.

Download Full Size | PDF

Once this common contact layer is placed onto the backside of the pixel array, electron-beam lithography is undertaken to define a metamaterial pattern in the optical aperture of each pixel. The backside of each pixel was patterned with a metamaterial design using an Elionix ELS-7500 EX E-Beam Lithography System. After exposure and development of the e-beam resist, a 60nm layer of gold is deposited followed by a metal lift-off process. Figure 8 shows a typical scanning electron microscope image of the metamaterial pattern on the illumination side of a pixel.

 

Fig. 8 Schematic representation of the e-beam lithography procedure for metamaterial patterning.

Download Full Size | PDF

With design lessons learned from this prototype metamaterial detector, a new FPA architecture is theorized and illustrated in reference [49] to allow high resolution pattern definition, low cost substrate removal, and the option for additional fabrication steps on the wafer level. It is important to note that many of the fabrication steps illustrated here (selective substrate removal, backside metal deposition, metamaterial patterning with e-beam lithography) are slow, expensive, and on the die level. Through this proposed process, a significantly faster fabrication time and lower system cost should result.

Funding

This work was supported by the US Government and performed in part at the Duke University Shared Materials Instrumentation Facility (SMIF), a member of the North Carolina Research Triangle Nanotechnology Network (RTNN), which is supported by the National Science Foundation (NSF) (Grant ECCS-1542015) as part of the National Nanotechnology Coordinated Infrastructure (NNCI).

References and links

1. A. Rogalski, J. Antoszewski, and L. Faraone, “Third-generation infrared photodetector arrays,” J. Appl. Phys. 105(9), 091101 (2009). [CrossRef]  

2. A. Rogalski, “Recent progress in infrared detector technologies,” Infrared Phys. Technol. 54(3), 136–154 (2011). [CrossRef]  

3. G. Lu and B. Fei, “Medical hyperspectral imaging: a review,” J. Biomed. Opt. 19(1), 010901 (2014). [CrossRef]   [PubMed]  

4. M. L. Althouse and C. I. Chang, “Chemical vapor detection with a multispectral thermal imager,” Opt. Eng. 30(11), 1725 (1991). [CrossRef]  

5. S. Maimon and G. W. Wicks, “nBn detector, an infrared detector with reduced dark current and higher operating temperature,” Appl. Phys. Lett. 89(15), 151109 (2006). [CrossRef]  

6. J. F. Klem, J. K. Kim, M. J. Cich, S. D. Hawkins, T. R. Fortune, and J. L. Rienstra, “Comparison of nBn and nBp mid-wave barrier infrared photodetectors,” Proc. SPIE 7608, 76081P (2010). [CrossRef]  

7. J. B. Rodriguez, E. Plis, G. Bishop, Y. D. Sharma, H. Kim, L. R. Dawson, and S. Krishna, “nBn structure based on InAs/GaSb type-II strained layer superlattices,” Appl. Phys. Lett. 91(4), 043514 (2007). [CrossRef]  

8. SirB. Rafol, S. Soibel, A. Khoshakhlagh, J. Nguyen, J. K. Liu, J. M. Mumolo, S. A. Keo, L. Hoglund, D. Z. Ting, and S. D. Gunapala, “Performance of a 1/4 VGA format long-wavelength infrared antimonides-based superlattice focal plane array,” IEEE J. Quantum Electron. 48(7), 878–884 (2012).

9. I. Vurgaftman, C. L. Canedy, E. M. Jackson, J. A. Nolde, C. A. Affouda, E. H. Aifer, J. R. Meyer, A. Hood, A. J. Evans, and W. T. Tennant, “Analysis and performance of type-II superlattice detectors,” Opt. Eng. 50(6), 061007 (2011). [CrossRef]  

10. B.-M. Nguyen, D. Hoffman, P.-Y. Delaunay, and M. Razeghi, “Background limited long wavelength infrared type-II InAs/GaSb superlattice photodiodes operating at 110K,” Phys. Lett. 91(12), 163511 (2007).

11. O. Salihoglu, A. Muti, K. Kutluer, T. Tansel, R. Turan, Y. Ergun, and A. Aydinli, “N-structure for type-II superlattice photodetectors,” Appl. Phys. Lett. 101(7), 073505 (2012). [CrossRef]  

12. N. Gautam, H. S. Kim, M. N. Kutty, E. Plis, L. R. Dawson, and S. Krishna, “Performance improvement of longwave infrared photodetector based on type-II InAs/GaSb superlattices using unipolar current blocking layers,” Appl. Phys. Lett. 96(23), 231107 (2010). [CrossRef]  

13. J. V. Li, R. Q. Yang, C. J. Hill, and S. L. Chuang, “Interband cascade detectors with room temperature photovoltaic operation,” Appl. Phys. Lett. 86(10), 101102 (2005). [CrossRef]  

14. R. Q. Yang, Z. Tian, Z. Cai, J. F. Klem, M. B. Johnson, and H. C. Liu, “Interband-cascade infrared photodetectors with superlattice absorbers,” J. Appl. Phys. 107(5), 054514 (2010). [CrossRef]  

15. N. Gautam, S. Myers, A. Barve, B. Klein, E. Smith, D. Rhiger, L. Dawson, and S. Krishna, “High operating temperature interband cascade midwave detector based on type-II InAs/GaSb strained layer superlattice,” Appl. Phys. Lett. 101(2), 021106 (2012). [CrossRef]  

16. W. Pusz, A. Kowalweski, P. Martyniuk, W. Gawron, E. Plis, S. Krishna, and A. Rogalksi, “Mid-wavelength infrared type-II InAs/GaSb superlattice interband cascade photodetectors,” Opt. Eng. 53(4), 043107 (2014). [CrossRef]  

17. H. Lofti, L. Lei, L. Li, R. Q. Yang, J. C. Keay, M. B. Johnson, Y. Qiu, D. Lubyshev, J. M. Fastenau, and A. W. K. Liu, “High-temperature operation of interband cascade infrared photodetectors with cutoff wavelengths near 8um,” Opt. Eng. 54(6), 063103 (2015). [CrossRef]  

18. Z.-B. Tian and S. Krishna, “Mid-infrared interband cascade photodetectors with different absorber designs,” IEEE J. Quantum Electron. 51(4), 4300105 (2015). [CrossRef]  

19. Z.-B. Tian, S. Godoy, H.-S. Kim, T. Schuler-Sandy, J. Montoya, and S. Krishna, “High operating temperature interband cascade focal plane arrays,” Appl. Phys. Lett. 105(5), 051109 (2014). [CrossRef]  

20. E. L. Dereniak and G. D. Boreman, Infrared Detectors and Systems (John Wiley, New York, 200–202 1996).

21. L. G. Hipwood, C. L. Jones, C. D. Maxey, H. W. Lau, J. Fitzmaurice, R. A. Catchpole, and M. Ordish, “Three-color MOVPE MCT diodes,” Proc. SPIE 6206, 620612 (2006). [CrossRef]  

22. A. Khoshakhlagh, J. B. Rodriguez, E. Plis, G. D. Bishop, Y. D. Sharma, H. S. Kim, L. R. Dawson, and S. Krishna, “Bias dependent dual band response from InAs/Ga(In)Sb type II strain layer super lattice detectors,” Appl. Phys. Lett. 91(26), 263504 (2007). [CrossRef]  

23. N. Gautam, M. Naydenkov, S. Myers, A. V. Barve, E. Plis, T. Rotter, L. R. Dawson, and S. Krishna, “Three color infrared detector using InAs/GaSb superlattices with unipolor barriers,” Appl. Phys. Lett. 98(12), 121106 (2011). [CrossRef]  

24. A. M. Hoang, A. Dehzangi, S. Adhikary, and M. Razeghi, “High performance bias-selectable three-color short-wave/mid-wave/long-wave infrared photodetectors based on type-II InAs/GaSb/AlSb superlattices,” Sci. Rep. 6(1), 24144 (2016). [CrossRef]   [PubMed]  

25. W. L. Wolfe and G. J. Zissis, The Infrared Handbook. Washington, DC: Office of Naval Research, Department of the Navy. (1978).

26. S. Krishna, “The Infrared Retina,” J. Phys. D Appl. Phys. 42(23), 234005 (2009). [CrossRef]  

27. 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]  

28. J. A. Montoya, A. V. Barve, R. V. Shenoi, M. Naydenkov, H. Kim, Z. Ku, S. R. J. Brueck, S. Krishna, S. J. Lee, and S. K. Noh, “Backside illuminated infrared detectors with plasmonic resonators,” Proc. SPIE 7660, 76603P (2010). [CrossRef]  

29. S.-J. Lee, Z. Ku, A. Barve, J. Montoya, W.-Y. Jang, S. R. J. Brueck, M. Sundaram, A. Reisinger, S. Krishna, and S. K. Noh, “A monolithically integrated plasmonic infrared quantum dot camera,” Nat. Commun. 2, 286 (2011). [CrossRef]   [PubMed]  

30. 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]   [PubMed]  

31. S. C. Lee, S. Krishna, and S. R. J. Brueck, “Light direction-dependent plasmonic enhancement in quantum dot infrared photodetectors,” Appl. Phys. Lett. 97(2), 021112 (2010). [CrossRef]  

32. 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]  

33. M. G. Moharam and T. K. Gaylord, “Rigorous coupled-wave analysis of planar-grating diffraction,” J. Opt. Soc. Am. 71(7), 811–818 (1981). [CrossRef]  

34. B. K. Minhas, W. Fan, K. Agi, S. R. J. Brueck, and K. J. Malloy, “Metallic inductive and capacitive grids: theory and experiment,” J. Opt. Soc. Am. A 19(7), 1352–1359 (2002). [CrossRef]   [PubMed]  

35. L. Martín-Moreno, F. J. García-Vidal, H. J. Lezec, K. M. Pellerin, T. Thio, J. B. Pendry, and T. W. Ebbesen, “Theory of extraordinary optical transmission through subwavelength hole arrays,” Phys. Rev. Lett. 86(6), 1114–1117 (2001). [CrossRef]   [PubMed]  

36. D. W. Peters, P. Davids, J. R. Wendt, A. A. Cruz-Cabrera, S. A. Kemme, and S. Samora, “Metamaterial-inspired high-absorption surfaces for thermal infrared applications,” Proc. SPIE 7609, 76091C (2010). [CrossRef]  

37. D. W. Peters, P. Davids, J. K. Kim, D. Leonhardt, T. E. Beecham, S. W. Howell, T. Ohta, J. R. Wendt, and J. A. Montoya, “Application of plasmonic subwavelength structuring to enhance infrared detection,” Proc. SPIE 8994, 899419 (2014). [CrossRef]  

38. D. R. Smith, W. J. Padilla, D. C. Vier, S. C. Nemat-Nasser, and S. Schultz, “Composite medium with simultaneously negative permeability and permittivity,” Phys. Rev. Lett. 84(18), 4184–4187 (2000). [CrossRef]   [PubMed]  

39. D. Schurig, J. J. Mock, B. J. Justice, S. A. Cummer, J. B. Pendry, A. F. Starr, and D. R. Smith, “Metamaterial electromagnetic cloak at microwave frequencies,” Science 314(5801), 977–980 (2006). [CrossRef]   [PubMed]  

40. J. B. Pendry, “Negative refraction makes a perfect lens,” Phys. Rev. Lett. 85(18), 3966–3969 (2000). [CrossRef]   [PubMed]  

41. N. I. Landy, S. Sajuyigbe, J. J. Mock, D. R. Smith, and W. J. Padilla, “Perfect metamaterial absorber,” Phys. Rev. Lett. 100(20), 207402 (2008). [CrossRef]   [PubMed]  

42. C. M. Watts, X. Liu, and W. J. Padilla, “Metamaterial electromagnetic wave absorbers,” Adv. Mater. 24(23), OP98–OP120 (2012). [PubMed]  

43. X. Liu, T. Starr, A. F. Starr, and W. J. Padilla, “Infrared spatial and frequency selective metamaterial with near-unity absorbance,” Phys. Rev. Lett. 104(20), 207403 (2010). [CrossRef]   [PubMed]  

44. R. Tsu, L. Esaki, and B. A. Sai-Halasz, “A new semiconductor superlattice,” Appl. Phys. Lett. 30(12), 651–653 (1977). [CrossRef]  

45. D. L. Smith and C. Mailhiot, “Theory of semiconductor superlattice electronic structure,” Rev. Mod. Phys. 62(1), 173–234 (1990). [CrossRef]  

46. S. Law, D. C. Adams, A. M. Taylor, and D. Wasserman, “Mid-infrared designer metals,” Opt. Express 20(11), 12155–12165 (2012). [CrossRef]   [PubMed]  

47. S. Adachi, “Model dielectric constants of GaP, GaAs, GaSb, InP, InAs, and InSb,” Phys. Rev. B Condens. Matter 35(14), 7454–7463 (1987). [CrossRef]   [PubMed]  

48. Z.-B. Tian, E. Plis, R. Hinkey, and S. Krishna, “The influence of composition in InAs/GaSb type-II superlattices on their optical properties,” Electron. Lett. 50(23), 1733–1734 (2014). [CrossRef]  

49. J. Montoya, S. Myers, and S. Krishna, inventor; STC.UNM, assignee. Wafer level integration of focal plane arrays having a flexible conductive layer to provide an optical aperture for each pixel. US patent 9,318,517 B1. April 19, 2016.