Ge_1-xSn_x thin films are capable of detecting mid- to long IR wavelengths. However, due to the size difference between Ge and Sn atoms, homogeneous alloys are difficult to manufacture in thicknesses greater than a couple hundred nanometers. To address the difficulty in manufacturing these IR detecting alloys, we developed an in situ-capable x-ray fluorescence (XRF) system to characterize and monitor thin film composition and concentration information as a function of position during thin film growth. The inclusion of a focusing polycapillary optic enabled the x-ray source and detector distances from the sample to be large enough for placement outside the deposition chamber. RF sputtered, metastable Ge_1-xSn_x thin films (~ 4 μm, ~ 8 μm) were compositionally mapped with the XRF system. Since Sn phase separation destroys the IR capabilities of the Ge_1-xSn_x sensor, it is imperative to prevent this during deposition. Additionally, to produce an IR sensing array, more gradual changes in composition are desired. With our system, both Sn phase separation and more gradual changes in alloyed Sn concentrations were observed in 1000 s. However, the signal-to-noise ratio was such that 100 s would have been sufficient, meaning this system could be used in situ to characterize the alloy composition during deposition. A spatial resolution of ± 25 μm was obtained by oversampling with the focusing optic (100 μm spot size). From these measurements, the minimum detectable limits were on the order of nanograms using the Sn-L_a signal, which corresponds to picograms from a Sn-K_a signal. Such low levels are usually only possible with a rotating anode source 10³ times more powerful than the low power sealed tube source used in this experiment. Additional ultra thin samples (< 100 nm) made by ion implantation were also analyzed with this XRF system. Sn was ion implanted into single crystal Ge, resulting in a sample representative of early stage thin film growth. In 300 s, a detectable signal was obtained, indicating the viability of this system for in situ, thin film, composition monitoring and characterization of uniformly alloyed metastable semiconductors for IR sensors.

Results of comparative studies of opto-electrical properties of photodiode arrays built on 30-um, 75-um, and 100-um thick single Silicon dies are presented. The size of the square pixels varied from 1.5 mm to 250-um for different arrays with the gaps between adjacent elements as small as 20 um. The internal quantum efficiency was close to 100%, DC and AC cross talks were smaller than 0.01% within the spectral range 400 to 800 nm. The arrays were characterized with very low leakage currents and high shunt resistance - above 1 GΩhm. The features of the array structure are discussed for the first time.

Near-infrared detection has a lot of application in the fields of telecommunication, bio-sensing, environmental gas detection and hyper-spectral spectrometer. Phototransistor is one of the most promising photodetection devices that can provide integrated photo current gain even at high-speed operating conditions. A double heterojunction phototransistor was grown by molecular beam epitaxy (MBE) and fabricated on semi-insulating InP substrate. The collector-up device structure consists of InP emitter, InGaAs base, InAlAs collector and subcollector, InGaAs/InAlAs superlattice graded emitter-base (EB) and base-collector (BC) junctions. A solid GaP source was used to clean the InP substrate and grow InP buffer and emitter layer. In-situ Reflective High-Energy Electron Diffraction (RHEED) and ex-situ X-Ray Diffraction (XRD) were utilized to monitor the material growth for lattice match condition of InGaAs and InAlAs to InP substrate. A digital-alloy MBE growth technique was applied to implement the graded interfacial layers of EB junction and BC junction. The performance of the device was characterized, including current-voltage characteristics, breakdown and responsivity at the wavelength of 1.55um (telecommunication application). Simple model of phototransistor current gain with different device parameters and various operating conditions was developed and applied to the device design for hyper-spectral spectrometer application.

The dilute-nitride GaInNAs shows great promise in becoming the next choice for long-wavelength (0.9 to 1.6 μm) photodetector applications due to the ability for it to be grown lattice-matched on GaAs substrates. GaAs-based devices have several advantages over InP-based devices, such as substrate cost, convenience of processing, and optoelectronic band parameters. This paper will present results from the first high-quality thick GaInNAs films grown by solid state molecular beam epitaxy with a nitrogen plasma source and the first high efficiency photodetectors which have been fabricated from those materials. GaInNAs films up to 2 microns thick have been grown coherently on GaAs substrates. These films exhibit reasonable photoluminescence intensities at peak wavelengths of 1.22 to 1.13 μm before and after a rapid thermal anneal at a series of temperatures. PIN photodiodes with these thick GaInNAs films in the intrinsic regions show responsivity (better than 0.5 A/W at 1.064 μm), dark current (200 nA at -2 V), and signal-to-noise ratio (greater than 10⁵) approaching those of commercially available InGaAs/InP devices. Furthermore, it will be shown that these devices show significantly lower dark current and higher signal-to-noise ratio than similar metamorphic InGaAs/GaAs structures.

Quantum well infrared photodetectors (QWIPs) have gained maturity for large focal plane arrays (FPA) with excellent thermal resolution, low 1/f noise, low fixed-pattern noise, and high pixel operability. Due to their spectrally narrow absorption, QWIPs are particularly suitable for thermal imaging applications involving several atmospheric transmission bands or several colors within the same band. We report on our progress on dual-band QWIP FPAs with pixel-registered, simultaneous integration in both bands. The arrays with 384x288 pixels and 40 μm pitch are based on a photoconductive QWIP for the 3-5 μm regime (MWIR) and a photovoltaic "low-noise" QWIP for 8-12 μm (LWIR). Excellent noise-equivalent temperature differences of only 20.6 mK (LWIR) and 26.7 mK (MWIR) have been achieved at 6.8 ms integration time and f/2 aperture. In addition, we have investigated test devices with different gratings, and discuss their dual-band coupling efficiencies.

We have designed and developed a new family of photodetectors with Internal Discrete Amplification (IDA) mechanism. These photodetectors can operate in linear (analog) detection mode with gain-bandwidth product up to 5^.10¹⁴ and few-photon sensitivity as well as in the photon counting mode with count rates up to 10⁸ cps. Some of their key performance characteristics exceed those of photomultiplier tube (PMT) and avalanche photodiode (APD) devices. The measured parameters of the detectors are gain > 10⁵, excess noise factor as low as 1.02, maximum count rate > 10⁸ counts/s, and rise/fall time < 300 ps.

Practical, packaged photodetectors (PDs) must be interfaced to bias and transmission lines, which introduce parasitics. These parasitics (resistance, capacitance and inductance) can be used to shape the temporal and frequency response of packaged photodetectors. Thus, the bias circuitry, external passives, and high speed interconnections must be carefully designed to produce the desired response in a packaged photodetector. Applications dictate the desired PD characteristics, which are generally either a flat frequency response, or a fast, ring-free impulse response. In this paper, the effects of the parasitic resistance, capacitance, and inductance are studied to affect the intrinsic response of photodetectors for a flat frequency response or a fast ring-free impulse response. For the optical transmission of microwave and millimeter wave RF signals, such as remote antennas or radar arrays, a flat frequency response is critical. A flat frequency response can be obtained from controlled ringing in the temporal domain. This paper explores the control of ringing in the temporal domain using varied external loads. A fast fall time, ring-free pulse is useful for digital communications applications where ringing can degrade the bit error rate. Fourier transforms show that a ring-free impulse response has a characteristic fall-off at high frequencies. However, this fall-off is detrimental for frequency domain applications, so the optimization condition for the inductance and capacitance is different for these applications. This paper explores the suppression of the impulse response tail by varying the external loads.

High speed optical interconnections offer an attractive alternative to electrical interconnections, particularly when they can be integrated into electrical systems. In particular, waveguide signal distribution and optical to electrical (O/E) conversion are critical to the integration of optical signals into electrical systems. The integration and interfaces between waveguides and O/E devices is a topic under intensive study. One approach to the integration of optical interconnections into electrical systems is to use fully embedded thin film optoelectronic (OE) devices in planar lightwave components on electrical interconnection substrates. In this approach, the propagating optical signal from the optical waveguide can be evanescently or directly coupled into the embedded thin film OE devices based on the embedded structure. Efficient and high speed optical signal distribution and O/E conversion, such as those using planar channel polymer waveguides with embedded thin film photodetectors, are examples of optical interconnection critical functions that are optimally implemented in electrical systems. In this paper, a 1 by 4 thin film metal semiconductor metal (MSM) photodetector (PD) array is embedded in a 1 by 4 photoimageable polymer multimode interference (MMI) coupler. This optical distribution and E/O system was fabricated and experimentally characterized at a wavelength of 1.3 μm. The measured overall loss, including the propagation loss and splitting loss of the MMI coupler was -0.18 dB at λ = 1.3 μm.

Recent advances in silicon based photon counting detectors have lead to the development of a new photon counting module. The detector for the module is fabricated in standard bulk silicon using complementary metal oxide semiconductor (CMOS) processing steps. High quantum efficiency across the visible spectrum combinedwith low timing jitter (150ps) provide ideal characteristics for photon counting applications. Through careful selection of components and operating conditions, the time-walk, which degrades the timing resolution of silicon photon counting detectors is minimised. The detector module is designed with a microprocessor interface which allows the internal operating characteristics to be optimised depending on the application. This work seeks to provide a overview of current photon counting detectors demonstrate the tradeoffs associated with each detector and present the new detector module demonstrating operation with zero time-walk affect.

This paper discusses the selection of parameters and the design of a CMOS detector for use in a structured illumination technique implemented with a 4M device (Miniaturized Multi-Modal Microscope) for precancer detection. To obtain real time sectioning the framing rate was set on the order of 500 frames / sec. 500 images allow us to obtain 8 to 16 sections / sec reconstructed from 16 to 64 images. The reconstruction technique is a sine approximation algorithm. To obtain a 1 μm spatial resolution, the required pixel size is 4 μm with a magnification 4:1. The field of extent of 250um required approximately 350 x 350 array size.

We report on the recent production release of our 320x240 pixel InGaAs/InP focal plane array and camera for visible and short-wavelength infrared light imaging. For this camera, we have fabricated a substrate-removed backside-illuminated InGaAs/InP photodiode array hybridized to a silicon read out integrated circuit (ROIC). Removing the InP substrate from the focal plane array allows visible wavelengths, which would otherwise be absorbed by the InP substrate due to its 920 nm wavelength cut-off, to reach the pixels’ active region. Quantum efficiency is approximately 15% at 500 nm, 70% at 850 nm, 85% at 1310 nm and 80% at 1550 nm. This focal plane array is useable for visible imaging as well as imaging eye-safe lasers and is of particular interest for day and low light level imaging as well as hyperspectral imaging.

Hyperspectral imaging has been receiving much attention for its potential for high-resolution imaging and target recognition, chemical analysis and spectroscopy. In target recognition, identifying targets in cluttered and partially obscured environments requires the analysis of spectral content of the scenery. Spectroscopy type of applications can benefit from the real-time data collection of spatial and spectral content in a single image capture. We report on the design, simulation and fabrication of integrating MEMs tunable Fabry-Perot etalon filters with 2 dimensional InGaAs focal plane arrays for simultaneous spectral and spatial imaging. By tuning the transmission wavelength of the MEMs based filter, the spectral information is provided at each pixel of the photodiode array. The MEMs device is based on two InP/air-gap DBR reflectors, and a single wavelength air cavity that separates them. The selective etching of InGaAs forms the air gaps that suspend the quarter wavelength InP reflector layers. The top mirror reflectivity as well as the cavity air-gap is tuned by deflecting the suspended InP layer through a reverse biased p-i-n junction. Due to the high refractive index contrast of InP and air, the spectral width of the DBR reflectors is wide enough to block transmitted light from 1000nm to 1700nm, allowing the InGaAs absorber layer to detect only the MEMs filtered spectral content. A theoretical study on wide tuning range designs and the expected FWHM will be presented.

We are reporting on research and development in the field of thin-layer planar silicon avalanche photodiodes operated as photon counters in a Geiger mode. We have developed and tested a technique, which permits an estimation of the photon number initiated a detection process. It can be applied in a time correlated photon counting experiment simultaneously with originally required time interval estimation. The principal limitation is a using of laser pulse with width below 30 ps to achieve detection concurrent in compare with carrier multiplication speed. The number of photons which triggered the avalanche is estimated on the basis of the effective rise-time difference of the avalanche current. The active quenching and gating circuit provides two uniform electrical pulses, and the time interval between them is related to the number of photons detected. The strong temporal correlation between avalanche start and one of pulses is preserved. Employing the picosecond event timing device, the photon number can be estimated within the dynamical range from 1 up to 1000 photons with the resolution better than a factor of three. The avalanche structure is operated on temperature achievable by thermo-electrical cooling. The applications of presented technique are in any time correlated photon counting (TCPC) measurement where the additional information about signal strength, i.e. statistical number of photons in laser pulse, is interesting. Other applications in the testing of quantum-well-based single photon light sources or squeezed light sources are expected.

A focal plane array detector sensitive from 2.0-2.5 μm and consisting of 32, 1.0 mm x 50 μm pixels, all functional, is demonstrated. Mean room-temperature R₀A is found to be 1.0 Ωcm², limited by sidewall leakage. The focal plane array is fabricated from an MBE-grown homojunction p-i-n GaInAsSb grown on an n-type GaSb substrate. Back-illumination geometry is compared to front-illumination geometry and is found to be favorable, particularly the improved responsivity (1.3 A/W at 2.35 μm corresponding to 68% quantum efficiency) due to reflection of light off the metal contact. Further, back-illumination is the most convenient geometry for mounting the array onto a compact blood glucose sensor because both contacts can be mounted on one side, while detector illumination occurs on the other.

Avalanche photodiode (APD) arrays fabricated by using complementary metal-oxide-semiconductor (CMOS) fabrication technology offer the possibility of combining these high sensitivity detectors with cost effective, on-board, complementary circuitry. Using CMOS techniques, Radiation Monitoring Devices has developed prototype pixels with active diameters ranging from 5 to 60 microns and with measured quantum efficiencies of up to 65%. The prototype CMOS APD pixel designs support both proportional and Geiger modes of photo-detection. When operating in Geiger mode, these APD’s act as single-optical-photon-counting detectors that can be used for time-resolved measurements under signal-starved conditions. We have also designed and fabricated CMOS chips that contain not only the APD pixels, but also associated circuitry for both actively and passively quenching the self-propagating Geiger avalanche. This report presents the noise and timing performance for the prototype CMOS APD pixels in both the proportional and Geiger modes of operation. It compares the quantum efficiency and dark-count rate of different pixel designs as a function of the applied bias and presents a discussion of the maximum count rates that is obtained with each of the two types of quenching circuits for operating the pixel in Geiger mode. Preliminary data on the application of the APD pixels to laser ranging and fluorescent lifetime measurement is also presented.

4×288 MCT LWIR linear arrays with 28X25 μm diodes and silicon ROICs were designed, manufactured and tested. MCT layers were grown by MBE technology on (013) GaAs substrates with CdTe/ZnTe buffer layers and λ_co = 11.2±0.15 μm at T = 78 K. CCD and CMOS “hybrid” technology for design and manufacture of silicon ROICs was used. The design rules 2.5 μm for CCD technology and 2.0 μm design rules for CMOS technology happened to be sufficient to realize most of the functions for 288×4 MCT TDI array. Analog functions were realized by CCD elements. An amplification of the output signals is realized by CMOS buffer amplifier. Decoding and deselection code storing functions are accomplished by digital CMOS elements. 288 information channels were attached to 4 analog outputs operating in the frequency range f≤4 MHz clock. Total consumption power measured is 50 mW at T = 298 K and 70 mW at T = 78 K. Before hybridization the parameters of MCT linear arrays and Si readouts were tested separately. With aperture 28⁰×64⁰ the detectivities D_λ ≈ 1.8^.10¹¹ cm^.Hz^1/2/W were achieved (λ_co ≈ 11.2 μm, λ_max 10.0 μm) with standard deviation about 15 % and operability close to 100 %.

Silicon ROIC for MCT LWIR 576x6 diode matrix arrays was designed. It includes 4 blocks of 144x6 arrays with 56x43 micron pixels. Diodes shift perpendicular to scanning direction is 0.25 of pixel size. ROICs were designed for their manufacturing by 0.6 micron design rules CMOS technology with 2 polysilicon levels and 2 metal levels. Six elements TDI function is used with bidirectional scanning, “dead” elements deselection, gain trim control, image data format and integration time selection, 8 levels input capacity programming, direct testing of the PV sensitive elements, etc. Max input capacity is 2.7 pC, the capacity at the TDI output register is about 2.0 pC, the output signal amplitude is not less than 2.8 V, the dynamic range is 77 dB. There are 8 video outputs, and the frequency range is 5 MHz.

We have investigated the photocurrent spectra of lateral conduction self-assembled Ge/Si quantum dots (QDs) infrared photodetector structure. We have observed a broad mid-infrared photocurrent spectrum in photon energy range of 120-400 meV (λ~3-10 μm) due to bound-to-bound as well as bound-to-continuum intersubband transition of normal incidence radiation in the valence band of self-assembled Ge QDs and subsequent lateral transport of photoexcited carriers in the Si/SiGe two-dimensional channel. The peak responsivity was as high as 134 mA/W at photon energy of 240 meV (λ~5.2 μm) at T=10 K and V_b=8 V.

Interest in eye-safe range-finding and lidar applications at 1060nm and 1550nm has increased dramatically in the last couple of years. However, APD receiver module performance has remained constant. This paper will present results from the characterization of an eye-safe module based on novel ultra -low excess noise InGaAs APD’s. The design basics of the APD and circuit will be discussed, with key performance characteristics highlighted. The principle of APD excess noise will be reviewed, and the effect it has on receiver module performance will be illustrated. A comparison of module performance between different receiver modules will be summarized.

An overview of photon counting detection using CMOS compatible Single Photon Avalanche Diodes (SPAD) will be presented. These SPADs have a planar structure, and are processed using CMOS technology. The most promising aspect of this technology is the potential for building large area arrays that can be operated in photon counting mode - without the read-out noise and bulkiness associated with low noise CCD cameras. Using the iAQC (integrated Active Quenching Circuit) produced by Micro-Photonics Devices, a low noise InGaAs/InAlAs APD will be characterized for photon counting. Finally, Characterization data from a photon counting module using Intevac’s IPD’s (Tube+APD hybrd) will be presented for photon counting at 1064nm.

The purpose of this paper is to describe the conception of the detectors developed for a system dedicated to quantum keys distribution on optical fiber network. Because there is still no commercial single photon sources quantum keys distribution is usually based on faint laser pulses transmission. Detection of such signals is very difficult but it can be improved by the use of a clock signal. Detectors based on APD can be designed to use this clock signal. In order to have a very good synchronization of both quantum keys and clock signals it is necessary to minimize chromatic dispersion in the optical fibers. The innovation presented in this paper is to use modulators effects leading to a very small change in the wavelength of the modulated signal. So it is possible to get two very close wavelengths dedicated to both signals, avoiding dispersion in the optical fiber. The system is designed at the 1.55 μm and is based on an acousto-optic modulator and an optical filter to separate both wavelengths at the reception. The detector is realized with an InGaAs APD, working in gated mode at liquid nitrogen temperature in order to reduce dark counts.