published_or_final_version / Electrical and Electronic Engineering / Doctoral / Doctor of Philosophy

Monolayer organic thin films and heterostructures are of great interest for their optical and electronic properties and as systems which allow the interplay between the structural and functional properties of organic molecules to be investigated. In the first experimental section of this thesis, sub-monolayer coverages of perylene tetracarboxylic diimide (PTCDI) were grown on hBN substrates and found to form needle-like monolayer islands at room temperature, while higher growth temperatures gave larger monolayer islands. The molecular packing of monolayer PTCDI was confirmed, using AFM, to correspond to the canted phase. The 0-0 fluorescence peak of this structure was found to occur at 2.208 ± 0.002 eV. The fluorescence of multi-layer PTCDI samples was mapped, with additional peaks measured at 2.135 ± 0.002 eV (580.7 ± 0.5 nm) and 2.118 ± 0.002 eV (585.4 ± 0.5 nm). Relating the morphology and fluorescence of such films using AFM and fluorescence microscopy is a promising way to investigate structural effects on the optical properties of multi-layer organic systems. Using solution deposition techniques, the PTCDI-melamine supramolecular network and the canted phase of PTCDI were deposited on hBN. The molecular packing of both structures was confirmed using AFM and the 0-0 fluorescence peaks were measured to be 2.245 ± 0.002 eV and 2.214 ± 0.002 eV for the PTCDI-melamine network and PTCDI respectively. The fluorescence of sublimed PTCDI, solution deposited PTCDI, PTCDI-melamine and measurements of Me-PTCDI doped helium nano droplets (HND) were compared. A 0.031 ± 0.002 eV red shift was measured from PTCDI-melamine to PTCDI while a 0.346 ± 0.002 eV red shift was measured from doped HND to PTCDI on hBN. A second perylene derivative, perylene tetracarboxylic dianhidride (PTCDA), was also deposited on hBN. Comparing the fluorescence of PTCDA monolayers on various dielectric substrates suggested a large shift due to the coupling of transition dipole moments and image dipoles beneath the dielectric surface. The shift between PTCDI and PTCDI-melamine was attributed to the coupling of transition dipole moments, for which the exciton bandstructure of both phases has been calculated with and without screening. The growth of sublimed C60 was also investigated, with monolayer islands observed for growth at room temperature and faceted bi-layer islands observed at 212 °C. The growth of PTCDI/C60 ¬heterostructures was also investigated, with C60 found to form monolayer islands on monolayer PTCDI at room temperature. At higher growth temperatures, C60 was found to form multilayers, with a reduced island density at PTCDI island edges, suggesting upward and downward hopping from the PTCDI surface to the second C60 layer and hBN respectively. C60 was found to quench the fluorescence of PTCDI and led to a 0.032 ± 0.02 eV blue shift. Finally, the growth of cyanuric acid-melamine (CA.M) on CVD graphene and CA.M/PTCDI heterostructures on hBN was investigated. Cyanuric acid-melamine was found to form monolayers with a honeycomb packing structure on CVD graphene. On monolayers of CA.M, PTCDI was found to form needle-like monolayer islands, the row direction of PTCDI is thought to have an on-axis registry with the substrate. Finally, the fluorescence of CA.M/PTCDI heterostructures on hBN was measured, with a 0.045 ± 0.002 eV blue shift from PTCDI on hBN.

Quantum Well Infrared Photodetectors (QWIPs) have the advantages of excellent uniformity and mature material properties. Thanks to these properties, large format and low cost QWIP focal plane arrays (FPAs) can be fabricated. The standard material system used for QWIP FPAs is AlGaAs/GaAs in the long wavelength infrared (LWIR) band. AlGaAs/GaAs material system has some disadvantages such as low quantum and conversion efficiencies under high frame rate and/or low background conditions. These limitations of the standard material system give rise to research on alternative material systems for QWIPs. InP/InGaAs material system is an alternative to AlGaAs/GaAs for LWIR QWIPs. This thesis focuses on the development of InP/InGaAs QWIP FPAs. A large format (640x512) LWIR QWIP FPA constructed with strained InP/InGaAs system is demonstrated with high quantum and conversion efficiencies. The FPA fabricated with the 40-well epilayer structure yielded a peak quantum efficiency as high as 20% with a broad spectral response (15%). The responsivity peak and the cut-off wavelengths of the FPA are 8.5 and ~9 um, respectively. The peak responsivity of the FPA pixels is larger than 1 A/W with a conversion efficiency as high as ~17 % in the bias region where the detectivity is reasonably high. The FPA provides a background limited performance (BLIP) temperature higher than 65 K (f/1.5) and satisfies the requirements of most low integration time/low background applications. Noise equivalent temperature difference (NETD) of the FPA is as low as 25 mK with integration times as short as 2 ms (f/1.5, 68 K).

Quantum well infrared photodetectors (QWIP) utilize quantum wells of large bandgap materials to detect infrared radiation. When compared to conventional low bandgap LWIR photodetectors, the QWIP technology offers largest format thermal imagers with much better uniformity. The theoretical part of this study includes the development of a QWIP ensemble Monte Carlo simulator. Capture paths of electrons to quantum wells are simulated in detail. For standard AlGaAs/GaAs QWIPs, at medium and high E-fields L valley quantum well (QW) is a trap for electrons which causes higher capture probability when compared with InP/InGaAs and GaAs/InGaAs QWIPs. The results suggest that high photoconductive gain observed in InP/InGaAs and GaAs/InGaAs QWIPs is not due to good transport properties of binary barrier material but due to higher &amp / #61511 / -L valley energy separation. The experimental part of the study includes the fabrication and characterization of InP/InGaAs and InP/InGaAsP QWIPs and 640x512 FPAs with the main objective of investigating the feasibility of these material systems for QWIPs. The InP/InGaAs and InP/InGaAsP QWIP detectors showed specific detectivity values above 1x1010 cm.Hz1/2/W (70K, f/2, background limited). The devices offer higher allowable system noise floor when compared with the standard AlGaAs/GaAs QWIP technology. It is also experimentally shown that for strategic applications LWIR InP based QWIPs have advantages over the standard QWIP technology. The InP/InGaAs 640x512 QWIP FPA reached 36 mK average NETD value at 70 K with f/1.5 optics and 10 ms integration time. The InP/InGaAsP QWIP on the other hand yielded 38 mK NETD histogram peak at 70 K with f/1.5 optics and 5 ms integration time on 320x256 window of the 640x512 FPA.

Quantum Well Infrared Photodetectors (QWIPs) are strong competitors to other detector technologies for future third generation thermal imagers. QWIPs have inherent advantages of mature III-V material system and well settled fabrication technology, as well as narrow band photo-response which is an important property facilitating the development of dual-band imagers with low crosstalk. This thesis focuses on the development of long/mid wavelength dual band QWIP focal plane arrays (FPAs) based on the AlGaAs/GaAs material system. Apart from traditional single band QWIPs, the dual-band operation is achieved by proper design of a bias tunable quantum well structure which has two responsivity peaks at 4.8 and 8.4 um for midwave infrared (MWIR) and longwave infrared (LWIR) atmospheric windows, respectively. The fabricated large format (640x512) FPA has MWIR and LWIR cut-off wavelengths of 5.1 and 8.9 um, and it provides noise equivalent temperature differences (NETDs) of ~ 20 and 32 mK (f/1.5 at 65 K) in these bands, respectively. The employed bias tuning approach for the dual-band operation requires the same fabrication steps established for single band QWIP FPAs, which is an important advantage of the selected method resulting in high-yield, high-uniformity and low-cost. Results are encouraging for fabrication of low cost, large format, and high performance dual band FPAs, making QWIP a stronger candidate in the competition for third generation thermal imagers

This thesis presents a detailed investigation of the performance limiting factors of long wavelength infrared (LWIR) and very long wavelength infrared (VLWIR) p on n HgCdTe detectors through numerical simulations at 77 K incorporating all considerable generation-recombination mechanisms including trap assisted tunneling (TAT), Shockley-Read-Hall (SRH), Auger and radiative processes. Numerical simulations under dark and illuminated conditions were performed with different absorber layer thicknesses, material compositions (cut-off wavelengths), trap density, and trap energy level. The results identify the relative strength of the dark current generation mechanisms by numerically extracting the contribution of each G-R mechanism on the detector characteristics with various cut off wavelengths and practically achievable material parameters. While the provided information can be used as a guide for optimizing the device processing conditions and detector structure, it also enlights the importance of various intrinsic mechanisms on the detector sensitivity. The results show that the dominant sensitivity degrading trap level depends on the detector cut-off wavelength being about 0.7Eg for LWIR HgCdTe sensors (cut-off wavelength=10 &micro / m) instead of 0.5Eg which is generally believed to be the most efficient R-G level. TAT related 1/f noise dominates the sensor noise even under small reverse bias voltages at a trap density as low as 1E14 cm-3 for sensors with cut-off wavelength &gt / 11 &micro / m. Considering the fact that trap densities below this level are rarely reported for HgCdTe material, exceptionally trap-free material is required to achieve desirable imaging performance with these sensors. Simulation results show that Auger mechanism has twofold effect on the sensitivity of the sensor by increasing the dark current and decreasing the photo current of the detector. As to our knowledge, this work is one of the most comprehensive simulation based investigations of the HgCdTe detector performance providing important results that can be used as a guide for optimization of the detector performance in order to meet the demanding requirements of the third generation thermal imagers.

InAsxSb1-x (Indium Arsenide Antimonide) is an important low bandgap semiconductor whose high quality growth on GaAs or Si substrates is indispensible for low cost, large format infrared focal plane arrays (FPAs). Quantum well infrared photodetector (QWIP) technology, relying on mature semiconductors, is also promising for the above purpose. While AlGaAs/GaAs has been the standard material system for QWIPs, the search for alternative materials is needed for better performance. This thesis reports a detailed investigation of molecular beam epitaxy grown mid-wavelength infrared InAsxSb1-x photodiodes on alternative substrates, and long wavelength infrared InP/InGaAs QWIPs. In the first part of the study, InSb and InAs0.8Sb0.2 photodiodes grown on Si and GaAs substrates are investigated to reveal the performance degrading mechanisms due to large lattice mismatch. InAs0.8Sb0.2/GaAs photodiodes yield peak detectivities of 1.4&times / 1010 and 7.5&times / 108 cmHz&frac12 / /W at 77 K and 240 K, respectively, showing that the alloy is promising for both cooled and near room temperature detectors. Under moderate reverse bias, 80 K RoA product limiting mechanism is trap assisted tunneling, which introduces considerable 1/f noise. InSb/Si photodiodes display peak 77 K detectivity as high as ~1&times / 1010 cmHz 1/2/W and reasonably high peak quantum efficiency in spite of large lattice mismatch. RoA product of detectors at 80 K is limited by Ohmic leakage with small activation energy (25 meV). Bias and temperature dependence of 1/f noise is in reasonable agreement with Kleinpenning&rsquo / s mobility fluctuation model, confirming the validity of this approach. The second part of the study concentrates on InP/In0.53Ga0.47As QWIPs, and 640&times / 512 FPA, which to our knowledge, is the largest format InP/InGaAs QWIP FPA reported. InP/InGaAs QWIPs yield quantum efficiency-gain product as high as 0.46 under moderate bias. At 70 K, detector performance is background limited with f/2 aperture up to ~3 V bias where peak responsivity (2.9 A/W) is thirty times higher than that of the Al0.275Ga0.725As/GaAs QWIP with similar spectral response. Impact ionization in InP/InGaAs QWIPs does not start until the average electric-field reaches 25 kV/cm, maintaining high detectivity under moderate bias. The 640&times / 512 InP/InGaAs QWIP FPA yields noise equivalent temperature difference of ~40 mK at an FPA temperature as high as 77 K and reasonably low NETD even with short integration times (t). 70 K NETD values of the FPA with f/1.5 optics are 36 and 64 mK under &ndash / 0.5 V (t=11 ms) and &ndash / 2 V (t=650 Rs) bias, respectively. The results clearly show the potential of InP/InGaAs QWIPs for thermal imaging applications requiring short integration times. Keywords: Cooled infrared detectors, InAsSb, QWIP, focal plane array.

This thesis reports the fabrication and characterization of long wavelength infrared mercury cadmium telluride (Hg1-xCdxTe) photodiodes and 128x128 focal plane arrays grown on lattice matched cadmium zinc telluride (Cd1-yZnyTe) substrates by metal organic vapor phase epitaxy (MOVPE). The dark current modeling of 33x33 mm2 Hg1-xCdxTe photodiodes has shown the dark current is dominated by trap assisted tunneling under small reverse bias voltages typically used to bias these detectors. The dominant dark current mechanisms under high reverse bias and low forward bias are band&ndash / to&ndash / band tunneling and generation&ndash / recombination, respectively. The photodiodes have yielded a peak 77 K detectivity of 3.2x1010 cm&amp / #8730 / Hz/W with a cut-off wavelength (50%) of 10.92 mm. It has also been found that the 1/f noise current of the detectors at 1 Hz is related to the trap-assisted tunneling current through the empirical relation in=&amp / #945 / TAT(ITAT)&amp / #946 / with &amp / #945 / TAT=7.0 x 10-5 and &amp / #946 / =0.65. In the course of the focal plane array (FPA) fabrication process development work, ohmic contact formation on p-type Hg1-xCdxTe and mesa wet etch were studied in detail. Contacts with chromium, gold, platinum and copper on p-type Hg1-xCdxTe resulted in bad ohmic contacts, which did not seem to improve with annealing. On the other hand a HgTe cap layer on p-type Hg1-xCdxTe resulted in good ohmic contact with acceptably low resistance. Among the etchants studied for mesa etching of the diode structures, Br2/HBr solution yielded the best performance. After developing all of the steps of FPA processing, 128x128 Hg1-xCdxTe FPAs were successfully fabricated and tested in a thermal imager. While thermal imaging was performed with the FPAs, high nonuniformity of the material and low R0A product of the pixels did not allow high sensitivity imaging.

In this thesis study, fabrication and doping of silicon thin films prepared by electron beam evaporation equipped with effusion cells for solar cell applications have been investigated. Thin film amorphous Si (a-Si) layers have been fabricated by the electron beam evaporator and simultaneously doped with boron (B) and phosphorous (P) using effusion cells. Samples were prepared on glass substrates for the future solar cell operations. Following the deposition of a-Si thin film, crystallization of the films has been carried out. Solid Phase Crystallization (SPC) and Metal Induced Crystallization (MIC) have been employed to obtain thin film crystalline Si. Crystallization was performed in a conventional tube furnaces and Rapid Thermal annealing systems (RTA) as a function of process parameters such as annealing temperature and duration. Produced films have been characterized using chemical and structural characterization techniques such as Raman Spectroscopy, X-Ray Diffractometer and Secondary Ion Mass Spectrometer (SIMS). The electrical properties of the films have been studied using Hall Effect and I-V measurements as a function of doping. We have demonstrated successful crystallization of a-Si by SPC at temperatures above 600 &deg / C. The crystallization occurred at lower temperatures in the case of MIC. For doping, P was evaporated from the effusion cell at a temperature between 600 &deg / C and 800 &deg / C. For B, the evaporation temperature was 1700 &deg / C and 1900 &deg / C. The thickness and the band gap of the Si films were determined by ellipsometry method and the results were compared for different evaporation temperatures. The effect of doping was monitored by the I-V and Hall Effect measurements. We have seen that the doping was accomplished in most of the cases. For the samples annealed at relatively high temperatures, the measured doping type was inconsistent with the expected results. This was attributed to the contamination from the glass substrate. To understand the origin of this contamination, we analyzed the chemical structure of the film and glass by X-ray Fluorescence (XRF) and seen that the glass is the main source of contamination. In order to prevent this contamination we have suggested covering the glass substrate with Si3N4 (Silicon Nitride) which act as a good diffusion barrier for impurities.

Processing of microwave signals in electrical domain introduces many difficulties especially when the frequency of the signal is increased beyond several GHz. Electromagnetic interference (EMI) and frequency depended losses can be given as examples to these difficulties. Photonic processing of microwave signals, however, is immune to these problems since optical components such as fiber cables, lasers, optical modulators and photodetectors are both immune to EMI and have wide bandwidths. This thesis deals with down conversion of a microwave signal using a Mach-Zender modulator and filtering unwanted harmonics using a photonic filter.