This PDF file contains the front matter associated with SPIE Proceedings Volume 11407, including the Title Page, Copyright information, and Table of Contents.

Innovation is the lifeblood of national defense and the Aerospace and Defense (A&D) industry. A&D employs a number of tools in the pursuit of innovation. Internal research and development, IRAD, is the principal tool and IRAD funding is the primary fuel for innovation. Contracted research and development, CRAD, contributes as well. A&D interacts with the academic, research and development, and small business communities to advance their innovation programs. As importantly, A&D engages academia to satisfy its need for top technical talent, the ultimate engine of innovation. In this talk, I’ll discuss the aforementioned elements of A&D innovation and academic relations and highlight key motivations and constraints. My aim is to inform and enable better and more mutually beneficial relations between the R&D and A&D communities.

Applications for short-wave infrared sensors continue to increase in commercial and military applications. Reducing pixel pitch decreases the size and cost expanding the ability to deploy sensors on smaller platforms including handheld and autonomous vehicles. Small pixel pitch cameras allow the user to shrink the lens thereby shrinking the total system size. Attollo Engineering will present the development of a VGA SWIR camera with 5 micron pixel pitch. Attollo will discuss designs of the components including detector, ROIC, and camera electronics. Attollo will show imagery and characteristics of the sensor.

Extremely low level dark current values are required for SWIR detection during the night when there is no active illumination due to weak sources and the lack of self emission. InGaAs detectors with planar pixel structures are the popular choice in SWIR range, even though there are some shortcomings such as the incompatability with dual color applications and the not tunable cut-off wavelength. To address these points, two nBn structured photodetector designs covering SWIR and extended SWIR in InGaAs material system will be discussed here.

Heterojunction nanohybrids based on low-dimension semiconductors, including colloidal quantum dots (QDs) and 2D atomic materials (graphene, transition metal chalcogenides, etc) provide a fascinating platform to design of new photonic and optoelectronic devices that take advantages of the enhanced light-solid interaction attributed to their strong quantum confinement and superior charge mobility for uncooled photodetectors with a high gain up to 1010. In these heterojunction nanohybrids, the van der Waals (vdW) interface plays a critical role in controlling the optoelectronic process including exciton dissociation by the interface built-in field that drives the follow-up charge injection and transport to graphene. In this paper, we present our recent progress in development of such heterostructures nanohybrids for uncooled infrared detectors including PbS and FeS₂ QDs/graphene and 2D vdW heterostructures MoTe₂/Graphene/SnS₂ and GaTe/InSe. We have found that nonstoichiometric Fe_1–xS₂ QDs (x = 0.01–0.107) with strong localized surface plasmonic resonance (LSPR) can have much enhanced absorption in broadband from ultraviolet to short-wave infrared (SWIR, 1–3 μm). Consequently, the LSPR Fe1–xS2 QDs/graphene heterostructure photodetectors exhibit extraordinary photoresponsivity in exceeding 4.32 ×106 A/W and figure-of-merit detectivity D* < 7.50 ×10¹² Jones in the broadband of UV–Vis–SWIR at room temperature. The 2D vdW heterostructures allows novel designs of interface band alignments with uncooled NIR-SWIR D* up to 10¹² Jones. These results illustrate that the heterostructure nanohybrids provide a promising pathway for low-cost, printable and flexible infrared detectors and imaging systems.

We present, for the first time, solution processed colloidal quantum dot extended short wavelength infrared (eSWIR) 1920 x 1080 format cameras sensitive from 300 nm to 2100 nm wavelengths at a pitch of 15 microns. We describe the performance of this high-definition imager using EMVA1288 machine vision testing and demonstrate pixel detectivities of 1 x10¹² Jones and 32 fA of dark current when operating at room temperature [1]. These results show this monolithically integrated detector technology has sensitivity superior to most traditional epitaxial IR sensors employed in the 1700 to 2100 nm band [2]. The straightforward and scalable fabrication processes employed have enabled the first commercially available room temperature 2.1 MP camera with sensitivity out to 2.1 um.

CMOS image sensors for visible wavelength range have been receiving much attention over the last two decades, offering ultra-low power and camera-on-chip integration. Imagers are now able to extract additional information from the scene thanks to infrared sensing for recognition or Time-of-Flight for 3D imaging. Such capabilities enable an unlimited amount of applications in several businesses, i.e. automotive, industrial, life science, security, agricultural or consumer. Imec has been continuously developing advanced technologies together with innovative pixel and circuit architectures to realize prototypes for various scientific applications. Thanks to state-of-the-art IIIV and thin-film (organics or quantum dots) material integration experience combined with imager design and manufacturing, imec is proposing a set of research activities which ambition is to innovate in the field of low cost and high resolution NIR/SWIR uncooled sensors as well as 3D sensing in NIR with Silicon-based Time-of-Flight pixels. This work will present the recent integration achievements with demonstration examples as well as development prospects in this research framework.

An InGaAs/GaAsSb Type-II superlattice is explored as an absorber material for extended short-wave infrared detection. A 10.5 nm period was grown with an InGaAs/GaAsSb thickness ratio of 2 with a target In composition of 46% and target Sb composition of 62%. Cutoff wavelengths near 2.8 μm were achieved with responsivity beyond 3 μm. Demonstrated dark current densities were as low as 1.4 mA/cm² at 295K and 13 μA/cm² at 235K at -1V bias. A significant barrier to hole extraction was identified in the detector design that severely limited the external quantum efficiency (EQE) of the detectors. A redesign of the detector that removes that barrier could make InGaAs/GaAsSb very competitive with current commercial HgCdTe and extended InGaAs technology.

Next-generation multi-mode tracking (MMT) technology developed by Sensors Unlimited Inc. (SUI), a Collins Aerospace Company, is presented. The technology provides focal plane arays (FPAs) with pulse detection capability in addition to traditional passive imaging, enabling more compact electro-optical (EO) systems with significantly lower power consumption. SUI’s latest MMT device is a short wave infrared (SWIR) FPA that features low-noise imaging, asynchronous laser pulse detection (ALPD), and time-of-arrival (TOA) capability in every pixel in the 12 μm pitch, 1280 x 1024 array. Simulated and measured results from the new FPA’s readout integrated circuit (ROIC), the 1280MMT, are given along with a comparison to SUI’s first multi-mode tracking ROIC, the 640MMT.

Sensors Unlimited Inc. (SUI), a Collins Aerospace Company, has developed a large-area, high-speed, short-wave infrared (SWIR) focal plane array (FPA) to meet the field-of-view (FOV) and bandwidth requirements of LiDAR applications. Modifications to SUI’s standard InGaAs photodiode array (PDA), include junction shape, dielectric thickness, and contact metallization. These changes allow for a reduction in the effective capacitance seen by the hybridized FPA’s readout integrated circuit (ROIC) while preserving the epitaxial structure that ensures the company’s industry-leading dark current. Compared to SUI’s standard device, significant capacitance reductions have been demonstrated. Enhancements of laser pulse detection performance arising from the capacitance improvement, and suitability of the resulting device for implementation in LiDAR systems, will be discussed.

Three-dimensional read-out integrated circuit (3D-ROIC) based advanced focal-plane-array (FPA) architectures are driven by the requirements of pixel level analog-to-digital conversion, on-chip digital-signal-processing, high framerates, low read noise, and extended dynamic range. These 3D-ROIC FPAs require hybridization of analog technology ROIC silicon wafers with digital technology ROIC silicon wafers and detectors. Small pitch vias are needed both at the topmost and bottommost surfaces of the analog wafers to facilitate these 3D-ROIC architectures. Although it is easy to build top surface vias, integration of bottom surface vias into the CMOS process flows is uncommon. In this paper, we describe the integration of tungsten-filled isolated deep-silicon-vias (iDSV) extending from metal1 to the handle wafer of the substrate in thick-film SOI based 130nm and 180nm mixed-signal CMOS processes. Major advantages of the iDSV based integration scheme reported here include ease of porting of mixed-signal designs from bulk CMOS to SOI, high density 3D-ROIC interconnect formation with via-middle integration, and an inbuilt buried-oxide etch stop layer for post wafer process thinning for iDSV reveal. These iDSVs have resistances of 2.5Ω/via, breakdown voltages exceeding 50V, and leakage currents to adjacent wells of less than 10fA/via. We report electrical results and process flow of the more complex face-to-back hybrid wafer bonding utilizing the top vias in the daughter wafer and iDSV in the mother wafer with copper direct-bond-interconnect (DBI) hybridization at 4μm pitch. We outline various 3D-ROIC integration architectures for large-format and small pixel FPAs using iDSV and DBI bonding, along with their advantages and disadvantages.

As GaSb based LWIR nBn detector structures have progressed from development into production, the standard substrates have been 76.2 and 100 mm in diameter. Additionally, production growths on 125 mm substrates are gaining popularity. To meet demands for larger focal plane arrays (FPA), improve throughput, volume, and yield, IQE has made the next step in this progression. Our first demonstration of multi-wafer growth of 150 mm GaSb exhibited uniform, high quality MWIR epitaxy. In this work, we will share our results for the more challenging GaSb-based LWIR nBn detector structures. These growths were carried out on a large format platen (7 × 150 mm) by Molecular Beam Epitaxy (MBE). The material properties as measured by AFM, HRXRD, Nomarski microscope, and PL, along with diode performance (turn-on, QE, cutoff wavelength), will be presented. Using this data, we will analyze the wafer-to-wafer cross-platen uniformity. Additionally, we will compare the characterization and device results to similar structures grown on smaller diameter substrates, and consider the viability of utilizing 150 mm GaSb substrates for LWIR nBn detector structures.

Cd_1-yZnyTe single crystal is the current material of choice to be used as substrate for the growth of lattice-matched Hg_1-x Cd_xTe epilayers with cutoffs wavelengths in the SWIR range (~2,9 μm (x=0,4) to 2μm (x=0,54)). For the manufacturing of large 2k² IR focal plane arrays with a 15 μm pitch, large diameter Cd_1-yZnyTe ingots with a state-of-the-art material quality are required. Crystal growth method from the melt; like Vertical Gradient Freeze technique; enables us to get close to 5 inches in diameter, high quality single crystals, after decades of developments. As the growth of high-quality Cd_1-yZnyTe single crystal ingots remains a big technological challenge, we present some recent technical achievements in this field, got within the frame of the H2020 ASTEROID project. Some requirements regarding material specification, like Cd1-yZnyTe substrate size, geometrical perfection (TTV, faces parallelism), material quality (crystallinity, dislocations) have imposed many new process updates and developments in our elaboration scheme. State-Of-the-Art 72x73 mm² Cd1-yZnyTe substrates with epiready surface preparations are now available for the Front-End-Of-Line of 2k² IR Focal Plane Array Processing at CEA-LETI / LYNRED (previously named SOFRADIR).

We present a method of determining the background doping type in semiconductors using capacitance-voltage measurements on overetched double mesa p-i-n or n-i-p structures. Unlike Hall measurements, this method is not limited by the conductivity of the substrate. By measuring the capacitance of devices with varying top and bottom mesa sizes, we were able to conclusively determine which mesa contained the p-n junction, revealing the polarity of the intrinsic layer. This method, when demonstrated on GaSb p-i-n and n-i-p structures, determined that the material is residually doped p-type, which is well established by other sources. The method was then applied on a 10 monolayer InAs/10 monolayer AlSb superlattice, for which the doping polarity was unknown, and indicated that this material is also p-type.

III-V GaSb based strained layer superlattice (SLS) can provide broadband detection from 3 to 15 microns and beyond. However, the longer wavelength material has smaller external quantum efficiency (EQE) because of the smaller absorption coefficient and carrier diffusion length. The usual approach of using thicker material for higher EQE is not viable because of the limited carrier collection. In this work, we adopt a metasurface on top of a thin detector material to increase its absorption without increasing the thickness, thereby increasing EQE. The metasurface consists of an array of rings. These rings excite different optical responses in different wavelength regimes, thus yielding a broad detection spectrum. In the MWIR regime, they diffract light collectively, and in the LWIR and VLWIR regimes, they focus light as Fresnel lenses and excite localized surface plasmon polaritons (SPPs). Through these different optical interactions, the incident radiation is redirected and trapped in the detector volume. When the trapped light interferes constructively, the optical intensity builds up to a large value, leading to a large absorption in a thin material. A properly designed metastructure is thus able to achieve a large EQE in a broad spectrum. The thin absorbing layer also allows several of them stacking together to form a multi-color detector. Narrowband detection can be obtained in each color using circular posts instead of rings. By exploiting different optical field profiles under different optical responses, a voltage-switchable detection scheme can be implemented to detect four discrete wavelengths ranging from MWIR to VLWIR. The broad and narrow band designs thus show the flexibility of metastructures. To compare the present metastructure approach with the more conventional approaches using microlenses and Fresnel lenses that can also reduce dark current while maintaining a sizable QE, we optimize the respective structures and compare the merits under different approaches.

The DARPA Microsytems Technology Office develops components meant to advance state of the art systems. The electro-optical/infrared (EO/IR) portfolio has ongoing programs to advance imaging capability in all areas, from advancing read-out integrated circuit technology to improving resolution to improved processing. This talk will cover those efforts as well as discuss new opportunities.

With the increasing use of resonant structures in optical devices, broadband optical characterization of the refractive index and extinction coefficient is necessary for accurate simulation and device design. For resonance-enhanced photodetectors, the complex refractive index is necessary to impedance match not only the resonator to air, minimizing the reflection, but also the resonator to the detector element, ensuring absorption occurs in the photodiode. To work towards better resonator-detector coupling, we present the complex refractive index for GaSb and an InAs/GaSb strained layer superlattice designed to be the absorber layer for a long-wave infrared photodetector. The optical properties were extracted using spectroscopic ellipsometry. Several modeling methods will be discussed for both the superlattice and the single-side polished bulk GaSb. Comparison to transmission and reflection values as well as absorption coefficients from literature provide additional confidence in the extraction process. Future work will incorporate these values into a resonance-enhanced photodetector.

Resonant cavity-enhanced photodetectors (RCE _{PDs) present a compelling alternative to broadband detection techniques in the field of gas detection and environmental sensing, due to the distinctive narrow-band absorption fingerprints of gases such as N2O} (at 4.5 μm) or CO (4.6 μm). This characteristic aligns well with the operational mode of an RCE PD, whose VCSEL-like architecture results in a tuneable narrow-band spectral response with a significantly enhanced quantum efficiency. Additionally, unlike broadband detectors, RCE PDs are not subject to the broadband BLIP limit due to their high spectral selectivity, while the substantially reduced absorber volume offers commensurately reduced Auger and generation-recombination dark current densities. In this work, we present efforts to extend the operability of these structures beyond 4.0 μm wavelength by employing the type-II InAs/InAsSb superlattice as the absorber material. The tuneable bandgap of this structure allows to achieve and demonstrate a MWIR RCE PD with a highly thermally stable resonant response at ~ 4.45 μm, a Q factor of 85-95, full-width-at-half-maximum of ~ 50 nm and a peak quantum efficiency of 84% at 240 K - features which are promising for detection of gases such as CO and N₂O. The broadband BLIP is also achieved at 180 K, a result which could potentially enable thermoelectrically cooled operation in the future. Finally, thanks to the inherent bandgap tunability of the InAs/InAsSb superlattice, extension of resonant response into the LWIR range is achievable with relatively straightforward changes to the already existing RCE PD structure.

In this paper, the performance of small, lightweight, and low power (SWaP) midwave infrared (MWIR) type-II superlattice detectors have been studied. The detectors cover the MWIR range from 3.7 μm to 5.1 μm with quantum efficiency higher than 60% in the entire range. Statistics from the focal plane array (FPA) production show excellent reproducibility with average temporal NETD of 20 mK, spatial NETD of 5 mK and operability values typically higher than 99.85%. Good uniformity across the arrays is demonstrated with narrow NETD histograms and highly uniform gain-correction maps. Temperature studies of the FPA performance show that the low NETD values, good uniformity and high operability are maintained up to 110 K. After integration of these FPAs in SWaP integrated detector dewar cooler assemblies (IDDCAs) with F/5.5, the high FPA performance is maintained with good imaging properties.

The III-V material system has proven to be a solid base for building infrared detector focal plane arrays (FPAs), enabling well-established designs such as bulk InSb-photodiodes, GaAs-based quantum well infrared photodetectors (QWIP) and GaSb-based type-II superlattices (T2SL). The remarkable advancement in the sensitivity and stability of such FPAs seen in the past decades calls for revision of the merits and the acceptance criteria used for the performance evaluation of infrared imaging arrays. Specifically, the early skepticism toward QWIP has largely been focused on their low quantum efficiency, and consequently low specific detectivity of the FPA pixels, as compared to the counterparts with bulk design. It was later demonstrated that unrivalled spatial uniformity of QWIP arrays (typically with a residual non-uniformity <0.05 %) not only compensated for this drawback, but rendered them superior in terms of the sensitivity, with typical noise equivalent temperature difference (NETD) <30 mK (when staring at 30 °C source through F/2.24 aperture, FPA on 15 μm pitch operated at 70 K and 60 Hz frame rate, peak absorption at 8.6 μm). This introduced QWIPs to the demanding applications, such as tactical or gas sensing, in spite of the design’s inherently low detectivity. In this paper, we propose the tightened criteria for marking the defective pixels and demonstrate how IRnova’s III-V based long-wave QWIP and mid-wave T2SL FPAs with 15 μm pitch live up to them. High spatial uniformity is reflected in symmetric and almost tail-free NETD histograms for both detector types. Only 0.13 % of the pixels exhibit NETD above ×1.5 of the median value for QWIP and 0.14 % - for T2SL detectors. At the same time, there are virtually no pixels falling outside less stringent limit of ×3 the median value. For QWIPs, this observation is consistent with flat noise spectra, which are completely free from the low-frequency “1/f”-component, as measured on the test cells. Finally, our attesting procedure allows tracing even solitary pixels exhibiting deviating types of noisy behavior, including random telegraph signal (RTS)-like, whenever they are present in the array. Low number of RTS-like pixels is a key advantage of III-V based FPAs that enables them to be used for high-operating-temperature (HOT) applications. Another characteristic feature of III-V based detectors is the outstanding stability of their performance. For IRnova’s mid-wave T2SL FPA on 30 μm pitch, using a 3-year-old non-uniformity correction map even without refreshing the offset is penalized by a mere 100 % increase in the residual fixed pattern noise relative to its temporal NETD. In the paper, we complement this with a study on the stability of the non-uniformity correction of IRnova’s QWIP and T2SL detectors on 15 μm pitch.

Sb-based III-V semiconductors are a promising alternative to HgCdTe. They can be produced with a similar bandgap to HgCdTe, but take advantage of the strong bonding between group III and group V elements which leads to very stable materials, good radiation hardness, and high uniformity. In this paper, we will discuss the recent progress of our research and present the main contributions of the Center for Quantum Devices to the Sb-based 3th generation imagers.

In any sensing application, the most universal performance parameter is the signal-to-noise ratio (SNR). Over time, application needs push high-performance thermal imagers toward their ultimate SNR limit – background limited performance (BLIP) – the best any imager can do. BLIP SNR is determined by random shot noise inherent in the scene itself. The noise (and thus the SNR) magnitude is the square root of the number of signal photons collected, requiring the best sensors to collect all available scene light and add no additional noise. Quantum photodiodes made from infrared semiconductors have proved to be the highest sensitivity thermal detectors, but their dark currents (which add noise but no signal) increase exponentially with temperature and so still require significant cooling to attain BLIP. Recent breakthrough results from Teledyne show that HgCdTe can attain much lower dark currents than previously reported leading to higher BLIP operating temperatures and approaching fundamental limits. This talk will focus on HgCdTe photodiodes which have the best demonstrated performance to date for devices either fielded or in development and will compare these devices to other semiconductors past, present, and (likely) future.

In the past decade, there has been significant progress in development of the colloidal quantum dot (CQD) photodetectors. The QCD’s potential advantages include: cheap and easy fabrications, size-tunable across wide infrared spectral region, and direct coating on silicon electronics for imaging, what potentially reduces array cost and offers new modifications like flexible infrared detectors. The performance of CQD high operating temperature (HOT) photodetectors is lower in comparison with traditionally detectors existing on the global market (InGaAs, HgCdTe and type-II superlattices). In several papers their performance is compared with the semiempirical rule, “Rule 07” (specified in 2007) for P-on-n HgCdTe photodiodes. However, at present stage of technology, the fully-depleted background limited HgCdTe photodiodes can achieve the level of roomtemperature dark current considerably lower than predicted by Rule 07. In this paper, the performance of HOT CQD photodetectors is compared with that predicted for depleted P-i-N HgCdTe photodiodes. Theoretical estimations are collated with experimental data for both HgCdTe photodiodes and CQD detectors. The presented estimates provide further encouragement for achieving low-cost and high performance MWIR and LWIR HgCdTe focal plane arrays operating in HOT conditions.

In the last decade, IR imaging detector trend has gone for smaller pixels and larger formats. Most of the time, this scaling is carried out at given total sensitive area for a single focal plane array (FPA). As an example, QVGA 30µm pitch and VGA 15µm pitch exhibit the exact same sensitive area. SXGA 10µm pitch tends to be very similar as well. This increase in format is beneficial to image resolution. However, this scaling to even smaller pixels raises questions because the pixel size becomes similar to the IR wavelength but also to typical transport dimensions in the absorbing material. Hence, maintaining resolution for such small pixel pitches requires a good control of the modulation transfer function (MTF) and quantum efficiency (QE) of the array, while decreasing the pixel size. This might not be obtained just scaling the pixel dimensions. As an example, bulk planar structures suffer from excessive lateral diffusion length inducing pixel-to-pixel cross talk and thus degrading MTF. Non-bulk semiconductor materials such as colloidal quantum dots might exhibit much smaller cross talk due to weak transport properties, but it usually strongly degrades the accessible QE. On the other side, mesa structures might minimize cross-talk physically separating pixels, but also tends to degrade the QE due to non-negligible pixel fill factor shrinking down the pixel size. This paper intend to discuss those issues, taking into account different material systems and structures, in the perspective of expected future pixel pitch IR FPAs. This paper also introduces an important issue in this context: how to reliably measure the MTF of those small pitch detectors. As an answer to this question we will share our first investigations of MTF measurement using the electron beam of an SEM instead of a photon beam (EBIC measurement)

In 2007, Teledyne presented and subsequently published an empirically derived formula, known as “Rule 07”, for the dark current performance of Mercury Cadmium Telluride (HgCdTe) P-on-n diodes. The Rule 07 metric has become widely popular within the infrared community, not only as a basis for predicting HgCdTe detector and system performance, but as the “number to beat” for other technologies, notably III-V nBn and strained-layer superlattice (SLS) devices. For materials that have sufficiently long recombination lifetimes, HgCdTe being one of the few such widely used materials, internal currents within the detector can be suppressed and the detector becomes limited by the background radiation from the surrounding environment. These currents can be orders of magnitude below Rule 07 and even further orders of magnitude below the Auger-limit. The ability to suppress Auger currents and operate at the radiative limit allows for significantly higher operating temperature and provides several significant advantages, including: 

Reduced size, weight, power, cost, and improved reliability associated with reduced cooler requirements



Lower dark current when operating at conventional temperatures, permitting improved sensitivity from lower shot noise and longer achievable integration times

Because background radiatively-limited performance is both fundamental and physics-driven, in 2019 we proposed replacing Rule 07 with “Law 19” and provided a comparison of this fundamental limit with Rule 07. In this paper, we review the theoretical performance of Teledyne’s fully-depleted HgCdTe P-υ-N detectors and provide performance data on dark current, dynamic impedance and quantum efficiency (QE) for mid-wavelength infrared (MWIR) and long-wavelength infrared (LWIR) detectors both at high operating temperatures (up to 300K) and as a function of temperature

The HIgh-Resolution Mid-infrarEd Spectrometer (HIRMES) is the 3rd Generation Instrument for the Stratospheric Observatory For Infrared Astronomy (SOFIA). The HIRMES cryostat is comprised of several cooling stages (300K, 65K, 4K, 800mK, 400mK, and 70mK), which are essential to achieve the required sensitivity on the mid-infrared waveband in which HIRMES observes (25 um-122um). The science application of the HIRMES instrument is the study of proto-planetary systems that contain water-vapor, water-ice, deuterated hydrogen, and neutral oxygen. In this paper, we discuss the use of a He3 stage coupled to an Adiabatic Demagnetization Refrigerator (ADR) stage in order to achieve the required temperature of 70mK on the Focal Plane Assembly (FPA). HIRMES' FPA contains two Transitioning Edge Sensor (TES) bolometric detector arrays that can provide a combination of four primary observing modes; High Resolution (R ~ 100,000), Mid-Resolution (R ~ 10,000), Low-Resolution (R ~ 600) spectroscopy, and Spectral Imaging (R ~ 2000).¹ Moreover, we discuss the operations of the detector readout, in combination of a grating-dispersive spectroscopy mechanism, and a host of Fabry-Perot tunable narrow-band filters to achieve HIRMES' unmatched sensitivity in all four modes.

Molecular dynamics (MD) simulations of nerve-agent-sorbent binding provide, in principle, interpretation of measured infrared (IR) spectra obtained using molecular binding spectroscopy. Comparison of IR absorption spectra for nerve-agent-sorbent binding obtained using MD and those measured experimentally, however, indicate inconsistencies with respect to interpretation of underlying molecular interactions. Accordingly, there is a need to examine physical assumptions underlying these MD simulations and potential functions representing the molecular dynamics. This study examines aspects of MD simulated nerve-agent-sorbent binding for more quantitative interpretation of ATR spectra associated with nerve-agent detection.

Copious Imaging is commercializing a digital-readout integrated circuit (DROIC) technology that not only digitizes, but also performs computations on the signal at each pixel. When the DROIC is mated to a photodiode detector array, the device forms a Computational Pixel Imager (CPI). The technology was in development for many years at MIT Lincoln Laboratory, and now also at Copious Imaging for new application areas. CPI technology is fundamental to the operation of WISP – the Wide-Area Infrared System with Persistence. WISP is a 500-Mpix longwave infrared (LWIR) motion imaging sensor that produces imagery covering the complete surroundings, more than 2π steradians, every 2 seconds. WISP utilizes a fast scanner to cover the scene quickly. The WISP sensor is coupled with a real-time processing system that stiches the scene together from scanned swaths, performs a non-uniformity compensation, stabilizes the imagery to sub-pixel accuracy, detects motion, and tracks all moving objects of interest. The system has machine learning built-in to aid in identifying objects of interest while ignoring clutter. WISP is in use for many applications and in development for several more. We are currently evaluating WISP for use in screening for fevers.

RPA (Remotely Piloted Aircraft) also named UAV (Unmanned Air Vehicle) are today limited by the lack of adapted regulation and equipment. Above all, current UAV are not able to assume the “see and avoid” principle required for all manned aircraft flying in non-segregated airspace though strong interest exists for long range beyond line of sight mission such as infrastructure surveillance. In order to allow integration in all airspaces, detect and avoid function shall deal with cooperative intruders, such as aircraft equipped with transponders and non-cooperative ones. This paper addresses the use of Electro Optical sensors for the purpose of a non-cooperative sense function. Electro optical or infrared sensors suite commonly provides accurate angular and angular rates measurements, while having low SWAP. Therefore, it makes them attractive for medium and small size UAV as a primary non-cooperative sensor or in addition, with a range finding device such as low size, weight and power radar. In that way, SAFRAN has developed an Eo/IR sensor demonstrator dedicated to the Detect and Avoid function. The demonstrator includes sets of cameras, built-in multi-scale image processing and tracking processing to detect and track intruders in cluttered environments. It transmits typical outputs to the fusion or avoidance function. In this paper, experimental results obtained from flight trials are provided in order to illustrate demonstrator performances when facing different scenarios that stress on the “see and avoid” Requirements. In addition, most advanced developments are presented such as time to go estimation and clutter suppression by machine learning.

Joining their technological skills and strengths, SOFRADIR and ULIS merged in a new High Tech Company named LYNRED. This new company will be oriented towards excellence in II-VI, III-V and bolometers technologies, covering all Society’s needs in term of infrared detection. Nowadays one noticeable trend in infrared market sensors is the optimization of two key metrics: sensor range and image quality. These metrics depend mostly on three criteria: modulation transfer function (MTF), Noise Equivalent Temperature Difference (NETD) and Residual Fixed Pattern Noise (RFPN). MTF improvement was previously achieved by reducing pixel pitch to 30μm [1] to 10μm [2]. NETD and RFPN were improving thanks to a better material quality. The improvement of NETD and RFPN can also be used to increase the operating temperature. The Modulation Transfer Function performance is directly linked to the ability to distinguish objects details and sharpness of the imaging system. MTF value determines, with NETD and RFPN, the system range via the so -called Minimum Resolvable Temperature Range (MRTD). We need to keep in mind that not only MTF has to be optimized but also MRTD, which is very challenging in the context of High Operating Temperature (HOT) and pixel pitch reduction. There are many challenges to be addressed for future small pitch, large format and HOT+ (> 160K) detectors. Electrical and optical crosstalks are one of the prime concerns for detectors with sub-10μm pixel pitch. We will compare several existing materials and their optical and electronic transport properties, elementary sensors designs, such as III-V superlattice, II-VI materials, MESA, loophole, planar and depleted photodiodes. This discussion will be based on measurements on improved HgCdTe 10μm and 7.5μm pitch technology completed by simulation using Finite Element Modelization (FEM) coupled with Finite Difference Time Domain (FDTD).

Graphene has unique optoelectronic properties and potential applications in improved infrared (IR) photodetectors. Due to its Dirac cone structure, graphene exhibits broadband light absorption and rapid responsivity. In addition, unlike conventional quantum photomaterials, graphene can be synthesized inexpensively via a non-toxic process. Although graphene has advantages in IR photodetector applications, graphene photodetectors have shown low responsivity due to their minimal IR absorption (just 2.3%) and also require cooling. Therefore, there is considerable interest in enhancing the responsivity of graphene photodetectors operating at room temperature so that their advantages can be employed in IR applications. The present work demonstrates room temperature, high-responsivity, long-wavelength infrared (LWIR) graphene photodetectors. These devices operate on the photogating effect, using a lithium niobate (LiNbO₃) substrate with enhancement of the photogating via a pyroelectric effect in the substrate in conjunction with a SiN layer. This effect significantly modulates the back-gate voltage to increase the photoresponse by a factor of approximately 600 compared to that for a conventional graphene photodetector. This work also found a change in the type of charge carrier with variations in temperature, which was attributed to a large shift in the Dirac point owing to the strong photogating effect. The results of this study are expected to contribute to the future realization of high-responsivity, low-cost LWIR photodetectors for applications such as thermal imaging, medical care and gas analysis.

Graphene infrared (IR) photodetectors are promising devices that take advantage of the unique optoelectronic properties of graphene, such as broadband light absorption, rapid response, and high chemical stability. Despite its advantages, graphene has a low absorbance of 2.3%, which limits its photoresponsivity. We have previously reported the responsivity enhancement of graphene middle wavelength IR (MWIR) photodetectors using the photogating effect. The photogating effect is induced by photosensitizers located around the graphene channel that generate a large electrical change. The MWIR photoresponse with the photogating effect was enhanced by 100-fold relative to conventional graphene field-effect transistors (FETs). Although our graphene FETs using photogating exhibited ultrahigh responsivity, the dark current was extremely high, as in the case of conventional graphene FETs, because the normally-OFF operation cannot be realized in graphene. Therefore, reducing the high dark current is essential for applying graphene photodetectors to IR applications. We demonstrate dark current reduction and high responsivity MWIR light detection in graphene MWIR photodetectors. The devices consist of graphene FETs with a carrier injection region. The dark current is reduced by applying a bias voltage. The photocarriers injected into the graphene are amplified by the photogating effect induced in the graphene/insulator region. The dark current of the devices was significantly suppressed compared with that of conventional graphene FETs. The photoresponse characteristics were investigated for devices of different structure sizes. The results obtained in this study will contribute to the development of high-performance graphene-based IR image sensors.

Microbolometer detectors have drawn attention due to their advantages such as small dimensions, light weight, low power consumption and no cooling requirement. In recent years, studies regarding the microbolometers have accelerated. Research on uncooled infrared microbolometer detectors was initiated at ASELSAN in 2014. As result of these activities, microbolometers with high TCR and low 1/f noise level have been achieved based on vanadium oxide active material. Pixel structures with high fill factor ratios are accomplished using double layer pixel architecture and high responsivity values were obtained by FPA optimizations. 640 x 480 format and 17 μm pixel pitch microbolometer FPAs along with VO_x sensing layer were produced. These detectors have demonstrated state of the art NETD values (< 30 mK @ f/1) and with such properties that could be used in various applications which require high frame rates due to their low time constant values (< 12 ms). FPA fabrication has been performed on 8” CMOS ROIC wafers which are also designed by ASELSAN. 17 μm pixel pitch 640 x 480 format ROIC has been developed for resistive microbolometer detector arrays. ROIC wafers were fabricated using 0.18 μm CMOS process where 3.3 V analog and 1.8 V digital supplies are used. The ROIC has 4 analog video outputs and 2 analog reference outputs. It can be operated at 1, 2 or 4 output modes depending on the frame rate requirement. In 2-output mode of operation power consumption of the ROIC is less than 150 mW. In this paper, the details of the microbolometer detectors including the ROIC design, vacuum packaging and detector performances are presented.

We developed silicon-on-insulator (SOI) diode-based uncooled infrared focal plane arrays (IRFPAs), in which single-crystal pn junction diodes formed in an SOI layer are used as temperature sensors. These diodes, based on silicon large-scale integration technology, offer excellent uniformity, and have led to the use of high-performance uncooled IRFPAs in a wide variety of applications. In order to extend the pitch to less than 12 μm, a scalable new pixel structure has been developed to reduce the pixel size, based on a novel thermally isolated structure, which is fabricated above a CMOS processed wafer. The pn junction diodes used as a temperature sensor are separated from the underlying substrate by supporting legs made from thin metal wire, forming a cavity. To reduce the pixel size, we are developing a new diode structure by optimizing the ion implantation condition, thinning the SOI layer, and redesigning the supporting legs, achieving a smaller pixel size even with ten serially connected diodes. We also evaluated a new readout circuit architecture that enables an increase in sensitivity by generating a larger change in the diode forward voltage at a given temperature with no change in the number of diodes in the SOI layer. The effectiveness of the proposed readout circuit architecture was verified using a fabricated test element. The sensitivity of the test element was 128% of that for existing circuit structures, and further increases are expected with circuit structure optimization. These techniques have greatly enhanced the performance of our SOI diode based uncooled IRFPAs.

Veoneer is the single largest supplier of thermal sensing systems in the automotive marketplace. As the leader, Veoneer continues to innovate smaller cameras with increased resolution, while adding critical safety features such as pedestrian detection, animal detection and automatic emergency braking. Far infrared cameras are rapidly expanding into the autonomous vehicle space by providing an essential added modality for sensing objects in complex driving situations. In this keynote, Mr. Stuart Klapper will discuss where automotive infrared systems are headed in the future. Next generation night vision cameras are anticipated to be at least 50 percent smaller with VGA resolution. Some concepts include wider field of view cameras and implementation of multiple cameras for 360-degree coverage surrounding the vehicle. Night vision technology is viewed as unique for handling complex situations where other technologies like LiDAR, radar and vision cameras may have limitations.

Cross-model heterogeneous face recognition (HFR) has been one of the most challenging areas of research in biometrics and computer vision. The main goal of HFR is to accurately recognize/identify the visible face image with probe images captured in alternative sensing modalities, such as thermal spectrum. The polarization state information of thermal faces contains the missing textural and geometrics details in the conventional thermal face imagery, which facilitate the development of cross-spectrum face recognition. In this paper, we propose a coupled dictionary learning architecture to find a common embedding space between the visible and sensing domains, where we model the learning problem as a bilevel optimization. The learned coupled dictionaries are used to transform visible and polarimetric thermal face images into the common embedding feature space via patch-wise sparse recovery. Experiments conducted with the polarimetric thermal facial datasets which contains face images that has been taken at three different ranges and with different face expressions. The results show that our proposed coupled learning method could fuse polarimetric and thermal features in a way to outperform the conventional methods and enhance the performance of a thermal-to-visible face recognition system.

Manmade objects typically exhibit polarized signatures that can be utilized for target detection more effectively than thermal signatures alone. IR polarimetric imaging can also suppress clutter to highlight manmade objects in thermal equilibrium. However, quantifying the contrast improvement of polarimetric signatures over that of thermal signatures of the same scene has proved problematic due to the nature and dynamic range of the data. The Pyxis camera, a microbolometer-based imaging polarimeter that produces live polarimetric video of conventional, polarimetric, and fused image products and is small enough to be mounted on commercial drones, has collected datasets that demonstrate the improvement. In this paper, we describe the Pyxis camera, show representative data, and present the metrics used to characterize the improvement of polarimetric imaging over that of conventional thermal imaging.

In this paper we present a new class of high-performance infrared imagers utilizing near-critical angle spatial filters. The integrated photonic device design and nanofabricated thin-film optical coating are relatively simple but promise significant performance and image quality improvements, as well as reduction of cost, Size, Weight, and Power (cSWaP). Many of the common questions related to this technological approach are addressed and answered, including thermal isolation, spectral dependence, ambient temperature sensitivity, and sub-diffraction limit pixel size. Calculations are provided to access the expected performance, which indicate significant advantages to this technological approach.

Advanced uncooled infrared (IR) sensors with wavelength- or polarization-selectivity are advantageous in applications such as fire detection, gas analysis, hazardous material recognition, biological analysis, and polarimetric imaging. Plasmonic metasurfaces (PM) are potential candidates to realize such functionality over a wide range of wavelength, i.e., middle- to long-wavelength IR regions. In particular, metal-insulator-metal (MIM) PMs are most suitable for image sensor applications because of their small pixel size and simple surface configuration. However, such PMs produce various absorption modes, of which some degrade the wavelength or polarization selectivity. In this study, therefore, control of the absorption modes of PMs was investigated for wavelength- or polarization-selective uncooled IR sensor applications. The PMs produce various absorption modes, namely, localized surface plasmon resonance (LSPR), waveguide, Fabry-Pérot, and plasmonic surface lattice resonance (PSLR) modes. These modes can be controlled by engineering the surface configuration or optimizing the dielectric material. The LSPR mode is appropriate for wide-angle detection because it does not depend on the incident angle, whereas the PSLR mode is a potential candidate for narrowband uncooled IR sensors. The waveguide mode, which can degrade the wavelength and polarization selectivity, can be eliminated using an appropriate dielectric material, such as SiO2 or SiN, for high-performance wavelength- or polarization-selective uncooled IR sensors.

Graphene, an atomically thin carbon sheet, has drawn significant attention in many fields because of its unique electronic and optical properties. Graphene is a potential candidate for plasmonic metamaterial absorbers and emitters because of its optical tunability and extreme thinness. We have previously demonstrated graphene Salisbury screen metasurfaces. Although the absorption wavelength of such metasurfaces can be controlled by varying the graphene patch size, the absorbance is insufficient for practical applications. In this study, therefore, multilayer graphene metamaterial absorbers (MGMAs) were theoretically investigated in the middle- to long-wavelength infrared (IR) region. The MGMAs consist of graphene layers alternating with insulator layers formed on a bottom reflector. The spectral absorbance was calculated using the rigorous coupled-wave analysis method. The calculation results demonstrated that a high absorption of ~100% can be achieved because of the multiple plasmonic resonance between each graphene layer and the bottom reflector. The absorption wavelength can be controlled by regulating the graphene pattern size because of the plasmonic resonance of graphene. Furthermore, the absorption wavelength can be tuned by controlling the chemical potential of graphene, which allows for the development of electrically tunable wavelength-selective IR absorbers and emitters. These results will contribute to the development of high-performance wavelength-tunable graphene-based IR detectors and emitters.

Disorderly stacked multilayer graphene, called turbostratic graphene, is a promising candidate for highly responsive infrared detectors due to its higher carrier mobility than well-ordered multilayer graphene, and facility to suppress the Coulomb scattering from the substrate. Such properties are expected to enhance photogating for high-responsivity infrared detection. The electronic structure of turbostratic graphene was investigated using first-principles calculations. The turbostratic graphene was modeled by introducing disorder to bilayer graphene in terms of the distance and the rotation angle between the graphene layers. The calculation results show that an increase in these parameters leads to linear band dispersion and a structure similar to monolayer graphene.

We describe a new system for broadband characterization of mid-infrared (MIR) photodetectors, manufactured by the VIGO System and encapsulated in TO-8 can. The system comprises a high repetition rate pulsed quantum cascade laser source, stimulating a detector in a special test fixture that is connected to ports of a vector network analyzer (VNA) with a pair of cables. The system operates in a mode of nonlinear VNA, i.e. it is capable of measuring magnitude and phase spectra of harmonic signals up to 5 GHz of bandwidth. This approach enables recovering sub-nanosecond differential pulses of amplitudes below 1 mV and speed up performance evaluation of MIR detectors tested.