Wide direct band gap III nitride materials have opened up many new optoelectronic applications because they allow access to a very wide spectral range, from 200 nm to1.77μm, from a single material system. In addition, III Nitrides exhibit exceptional resistance to mechanical, chemical, and temperature loads, as well as superior radiation hardness. Photodetectors with advanced properties can be achieved through tailoring layered III nitride structures with various layer orders, chemical compositions, conductivities, and thicknesses. In addition, substrate selection plays an important role in the fabrication of these devices as well as in achieving desired device properties. Along with the efficiency of detection, in most cases it is important to provide selectivity of the optical radiation sensed by the photovoltaic device.
Wavelength-Selective and High Temperature Photodetectors Based on III-Nitrides
The objective of the first project was to develop an inexpensive miniature solid-state high temperature sensor for false alarm-free flame/fire detection. Flames and fires produce emissions ranging from ultraviolet to IR. Such emissions can be detected over the wide-range of ambient light background only by fast multi-range optical detectors allowing time resolved measurements in particular optical regions. As a result, not only the spectral range, but also the detector speed and spatial resolution and alignment become critical for fast fire detection and avoiding costly false alarms. Traditional photomultiplier tubes (PMTs) have high sensitivity, but are bulky, require high voltage operation, and have low mechanical and temperature strength. Currently used flame detectors that are composed of discrete UV and IR solid-state components are even bulkier, since they combine multiple components in one housing, sustain temperatures only up to 125°C, and are not capable of detecting the multi –band optical signal with high spatial resolution.
Employment of a miniature, chip-based dual-color high-temperature visible -blind optical sensor would allow for more efficient detection of the fire by placing the sensor closer to the fire source, and
providing a fast and reliable response in separated UV and IR bands with high spatial resolution. Moreover, development of such sensors can promote fabrication of multi-pixel focal arrays for dual-band visible -blind cameras, which can be used not only for fire/flame detection and imaging, but also for various space- and military-related applications, that involve object/target recognition.
The following diagrams represent the device structures targeted in this project:
In the course of this project, the RF MBE III-Nitride growth was developed and optimized. task has:
- Fabricated GaN-based layers for prototype development of visible blind detectors.
- Developed AlGaN-based layers for use in solar blind prototype detectors.
- Identified and solved issue with impurity incorporation in the III-Nitride layers.
- Addressed the issues that thin Si substrates have on III-Nitride layer deposition.
- Performed theoretical analysis of InGaN layers for infrared photodiode detection.
A substantial effort was taken for the optimization of the Schottky barrier diode fabrication on both III Nitride layers grown on Si wafers and Si wafers themselves. We have thus successfully realized visible blind photodiodes and solar blind photodiode structures. The achieved peak responsivities of the dual-band photodetectors were 3.8 and 55 mA∕W at wavelengths of 349 and 1000 nm, respectively .
A typical multi-pixel photodetector responsivity graph is shown in the picture below:
The development of the UV/IR detectors considered in this project resulted in two US patents [2,3].
Another project was dedicated to the development of the III-Nitride material growth for polarization sensitive devices [4-6]. While the majority of group III-nitride devices are grown along the  c axis of the wurtzite crystal structure, the active layers of the opto- electronic devices grown along this orientation suffer from undesirable spontaneous and piezoelectric polarization effects, which generate electrostatic fields. It has been demonstrated that such electrostatic fields can be avoided by growing heterostructures along non-polar directions ,such as [1-100] and [11-20], referred to as M- and A-plane respectively. It also has been shown that M-plane group III nitrides exhibit highly anisotropic optical polarization properties that should allow for the polarization sensitive detection across the III-nitride alloy range.
In this project, the growth front of M-plane GaN on LiAlO substrates was developed by using plasma-assisted molecular beam epitaxy with simultaneous study of the Ga-adsoption and desorption kinetics using in-situ ellipsometry and RHEED. Using these GaN growth conditions, M-plane InGaN layers were grown and fabricated into polarization-sensitive photodetectors.
The picture below shows a photoresponse of an InGaN Schottky photodiode with 10% Indium, measured as a function of polarization of light. The inset shows the transmittance of the sample for light polarized at angle of 0o (E||c), 45o and 90o (E ⊥ c).
High-Speed Radiation Tolerant Avalanche Photodiodes
High-performance, radiation-tolerant detectors are required for time-of-flight laser based rangefinders. Avalanche photodiodes (APDs) are conventionally chosen as detectors for standard laser rangefinder systems. However, the performance of currently used APDs degrades significantly after exposure to high levels of radiation. Integrated Micro Sensors Inc (IMS, Houston, TX)) proposes novel intrinsically radiation-tolerant III nitrides based high-speed APDs superior for use in space-based laser-altimeter systems.
The Indium Gallium Nitride (InGaN) alloy has the potential of forming photosensitive devices covering a range of 0.7 eV (InN) to 3.4 eV (GaN). This energy range provides a perfect match to the 1.06µm wavelength (~1.17 eV) of the lasers used in time-of-flight range finders. The III-Nitrides exhibit inherent chemical and thermal ruggedness, which makes them suitable for several space and military applications. It has recently been determined that these Nitride materials can also offer exceptional radiation tolerance that is well beyond what can be achieved with conventional materials that are currently flown in space.
Although InGaN as an APD would be a less mature technology than other III-V semiconductors, the important factor is that it will degrade far less over the life of the mission. The InGaN APDs to be developed in this project will be targeted for operating conditions up to 250 ºC, and up to 2 MeV proton irradiation, which are substantially higher than those for currently used materials, such as Si or GaAs. IMS envisions that devices developed in this project would be especially beneficial to Europa Jupiter System Mission (EJSM) that requires high performance sensors and detectors that can operate with low noise under the severe radiation environment.
Avalanche photodiodes (APDs) based on group III-Nitride materials are of interest due to their potential capability for low dark current densities, high sensitivities and high optical gains in the ultraviolet (UV) spectral region. Wide-bandgap GaN-based APDs, such as AlGaN and GaN-based p-i-n diodes, are excellent candidates for short wavelength photodetectors because they have the capability for cut-off wavelengths in the ultraviolet (UV) spectral region less than λ = 290 nm. A high density of defects and polarization effects are usually introduced during the growth of GaN-based heteroepitaxial layers on lattice-mismatched substrates; thereby, causing a device failure by premature microplasma breakdown before the electric field reaches the level of the bulk avalanche breakdown field, which has been a major issue for GaN-based APDs.
The ultimate goal of this project was to develop high-speed, radiation-tolerant visible-blind APDs responding to laser beams of 1.06 µm wavelength for rangefinder applications. In order to reach this goal, in this project has targeted the following milestones:
- Improvement in growth of InxGa1-xN with Indium fractions of x>0.75 using a novel lattice-matched buffer layer technology.
- Modeling, optimization, and design of back-illuminated APD structures based on improved quality InxGa1-xN layers.
- Fabrication and testing of the InGaN based experimental APD structures.
The main concept of the InGaN APDs in this project was based on employment of novel lattice-matched buffer layers growth of high quality InGaN using RF Plasma Assisted Molecular Beam Epitaxy (RF MBE).
The principle of an APD operation is shown in Figure 1below:
The selected buffer layers planned to be deposited on commercial Si and sapphire wafers used as substrates. Employment of Si and sapphire substrates for the InGaN growth provides several benefits, such as relatively low cost, efficient thermal management during the growth and device operation, large area, and high compatibility with conventional semiconductor processing systems. However, one of the most important benefits provided by Si wafers is also their utilization as passive optical filters for absorbing the visible light in a backside illumination configuration shown in Figure 2.
As was mentioned above, depending on the In fraction (x) in the InxGa1-xN layer, photodiodes with spectral sensitivities from 364 to 1771 nm can be achieved. Accordingly, wavelengths longer than 1 µm, can be detected at In fractions x > ~0.75. Our preliminary results (that will be detailed in the following sections) on the development of visible blind UV/IR photodetectors indicate that moderately doped (~1017 – 1018 cm-3) 175 µm thick Si wafers allow for complete elimination of the visible light sensitivity, when backside illumination configuration is used.
The approach for high quality InGaN layer growth is based on employment of InN-InGaN lattice matched buffer layers of selected materials deposited on sacrificial substrates (Si or sapphire) by various methods prior to APD structure growth. Among several examples, employment of buffer layers based on rare earth metal nitrides has been recently attempted.
For example, (001) InN films have been grown on lattice matched (111) EuN buffer layers by pulsed laser deposition (PLD). The InN films grown at relatively low temperature of 490 °C with an in-plane epitaxial relationship of: [112¯0]InN[11¯0]EuN[11¯0]MgO, which minimized the lattice mismatch between InN and EuN.
Another example of a lattice matched layer that can be used on (111) Si substrates is an ultra-thin silicon nitride grown by Nitrogen Plasma Assisted MBE. Growth of a (0001)-oriented single crystalline high-quality wurtzite–InN layer was confirmed by high resolution X-ray diffraction and Raman characterization. It is found that a double-buffer technique (InN/β–Si3N4) insures improved crystallinity, smooth surface, and good optical properties. High potential for employment as InN/InGaN lattice
Figure 2 below shows the proposed configuration of a backside illuminated InGaN APD structure grown on Si or Ge substrates used for visible light rejection:
Overall, the results of the experimental InGaN test devices designed and fabricated during the project were extremely positive, as evidenced by the characterization of the working laboratory APD device structures based on the band-engineered InGaN materials. High purity single phase InxGa1-xN (0.64 < x < 1.0) layers with very high indium compositions as verified by XRD (Figure 3), spectroscopic ellipsometry and optical absorption measurements, have been grown. The growth conditions developed for these high Indium InxGa1-xN layers were used to fabricate p-n junction devices with XRD showing high crystalline quality (Figure 4).
The p-n junction InxGa1-xN layers were processed using standard semiconductor processing techniques to develop experimental prototypes which were tested to show significant responsivities in a narrow band at the wavelength of interest . Figure 5 shows the responsivity of fabricated various p-n junction InxGa1-xN layers. Experimental prototypes with a narrow band responsivity greater than 0.8 mA/W at 1.06 μm (Figure 3), corresponding to the wavelength employed in current laser altimeter systems have been successfully fabricated. The efficiency and electrical properties of these device structures were affected by non-optimal device designs and material non-idealities. The causes are well understood, and can be addressed in the future.
Figure 3 below shows Θ-2Θ x-ray data for 400-500 nm thick layers of In-polar InN on thick GaN-sapphire template and N-polar InxGa1-xN on thin GaN buffers on sapphire with 0.68 < x < 0.96.
Figure 4 below shows Θ-2Θ x-ray data for p-n junction InGaN devices with thickness 425 – 450 nm and various indium compositions, along with a 570 nm In0.84Ga0.16N single layer for reference. The shoulders on the GaN-facing side of the InGaN (0002) peaks for the p-n junctions are evidence of intentional compositional grading to higher bandgap (lower indium fraction) material to aid in making contact to the devices.
Figure 5 below shows the responsivity of InGaN APD devices fabricated in this project.
 R. Pillai, D. Starikov, P. Misra, C. Boney, and A. Bensaoula. Radiation Selective Photodetectors Based on III-N. J. Vac. Sci. & Technol. A, (2008).
 D. Starikov and A. Bensaoula. US Patent 7381966: Single-chip monolithic dual-band visible- or solar-blind photodetector (2008). https://www.google.com/patents/US7381966
 D. Starikov and A. Bensaoula. US Patent 7566875: Single-chip monolithic dual-band visible- or solar-blind photodetector (2009). https://www.google.ch/patents/US7566875
 C. Boney, P. Misra, R. Pillai, D. Starikov, and A. Bensaoula, “Molecular beam epitaxy III-nitride growth for polarization sensitive devices based on M-plane films with in situ real time analysis by spectroscopic ellipsometry”, J. Vac. Sci. Technol. B 26 (3) (2008). http://scitation.aip.org/content/avs/journal/jvstb/26/3/10.1116/1.2830628
 P. Misra, C. Boney, D. Starikov, and A. Bensaoula. M-Plane III-Nitride Materials for Polarization Sensitive Devices Grown by PAMBE with Real Time Analysis by Spectroscopic Ellipsometry. Phys. Stat. Sol. (c) 5, No. 6, 2286–2289 (2008).
 P. Misra, C. Boney, D. Starikov, A. Bensaoula. Gallium adlayer adsorption and desorption studies with real-time analysis by spectroscopic ellipsometry and RHEED on A-, M-, and C-plane GaN grown by PAMBE. Journal of Crystal Growth, 311 2033–2038, (2009).
 “InGaN/Silicon Heterojunction Based Narrow Band Near-Infrared Detector”. R. Pillai, D. Starikov, C. Boney, J. Gandhi, A. Debnath, R. Li, and A. Bensaoula. Journal of Vacuum Science and Technology (2014).