Thin Active Region Research Articles

Optimization of Hybrid Organic−Inorganic Interdigitated Photovoltaic Device Structure Using a 2D Diffusion Model

To improve the efficiency of organic photovoltaic devices the inclusion of semiconducting nanoparticles such as PbS has been used to enhance near-infrared absorption. Additionally the use of interdigitated heterojunctions has been explored as a means of improving charge extraction. In this paper we provide a two-dimensional model taking into account these approaches with the aim of predicting an optimized device geometry to maximize the efficiency. The steady-state exciton population has been calculated in each of the active regions taking into account the full optical response based on using a finite difference approach to obtain approximate numerical solutions to the 2D exciton diffusion equation. On the basis of this we calculate the contribution of each active material to the device short circuit current and power conversion efficiency. We show that optimized structures can lead to power conversions efficiencies of ∼50% compared to a maximum of ∼17% for planar heterojunction devices. To achieve this the interdigitated region thickness should be ∼800 nm with PbS and C(60) widths of ∼60 and 20 nm, respectively. Even modest nanopatterning using much thinner active regions provides improvements in efficiency and may be approached using a variety of methods including nanoimprinting lithography, nanotemplating, or the incorporation of presynthesized nanorod structures.

Plasmon-based photosensors comprising a very thin semiconducting region

We theoretically investigate an elementary subwavelength plasmonic sensor comprising a very thin active region. High quantum efficiency (QE), broad spectral band and nearly no sensitivity to the incidence angle and polarization can be achieved. We particularly discuss different examples based on HgCdTe for infrared detection: QEs of 75% are obtained for active layers λ/(8n) thin, corresponding to a tenfold absorption enhancement.

Mid-Infrared Tunable Resonant Cavity Enhanced Detectors.

Mid-infrared detectors that are sensitive only in a tunable narrow spectral band are presented. They are based on the Resonant Cavity Enhanced Detector (RCED) principle and employing a thin active region using IV-VI narrow gap semiconductor layers. A Fabry-Pérot cavity is formed by two mirrors. The active layer is grown onto one mirror, while the second mirror can be displaced. This changes the cavity length thus shifting the resonances where the detector is sensitive. Using electrostatically actuated MEMS micromirrors, a very compact tunable detector system has been fabricated. Mirror movements of more than 3 μm at 30V are obtained. With these mirrors, detectors with a wavelength tuning range of about 0.7 μm have been realized. Single detectors can be used in mid-infrared micro spectrometers, while a detector arrangement in an array makes it possible to realize Adaptive Focal Plane Arrays (AFPA).

Bottom-gated metal-induced laterally crystallized silicon thin-film transistor with self-aligned raised source/drain

A bottom-gated metal-induced laterally crystallized silicon thin-film transistor with self-aligned raised source/drain has been fabricated and characterized. In order to achieve complete alignment of the thick source/drain to the gate, a simple process of back surface exposure was employed twice. It features a thin active channel region with fewer grains, resulting in a low grain boundary trap density and a thick source/drain region with a low lateral electric field as well as a low resistance. The proposed device simultaneously improves on the turn-on and -off currents, showing good saturation characteristics and good short-channel characteristics.

A Novel Four-Mask-Step Low-Temperature Polysilicon Thin-Film Transistor With Self-Aligned Raised Source/Drain (SARSD)

In this letter, a novel structure of polycrystalline-silicon thin-film transistors (TFTs) with self-aligned raised source/drain (SARSD) and a thin channel has been developed and investigated. In the proposed structure, a thick SD and a thin active region could be achieved with only four mask steps, which are less than that in conventional raised SD TFTs. The proposed SARSD TFT has a higher on-state current and a lower off-state leakage current. Moreover, the on/off current ratio of the proposed SARSD TFT is also higher than that of a conventional coplanar TFT

On the Conduction Mechanism in Polycrystalline Silicon Thin-Film Transistors

Physical and electrical analyses were carried out on n-channel polycrystalline silicon thin-film transistors (nTFTs) with active regions as thin as approximately 6 nm. Such thin active regions extinguish the dominating effects of anomalous leakage and allow the conduction energy barrier height to be analyzed as a function of gate voltage in the femtoampere source/drain current regime. Grain size statistics were determined using plan view transmission electron microscopy. It is shown for the first time that in the absence of anomalous leakage, the barrier height does not decrease with decreasing gate voltage. In addition, the maximum measured barrier height is almost independent of active region thickness and grain size statistics. The source/drain current at low lateral field and high vertical field is also independent of channel length for all devices with length varied over an order of magnitude. These important discrepancies with existing TFT conduction theory are discussed within a physics-based model that addresses the effects of disorder-induced localized electron states in the bandgap. Besides describing existing data and well-known TFT behavioral trends, the model predicts a previously unknown relationship between threshold voltage variation, average threshold voltage and the number of grains in the channel. Analysis of data gathered from hundreds of devices of different dimensions across three different grain size distributions not only leads to agreement with the model but also to a remarkable universal behavior linking electrical and physical properties. This study shows the physics of polycrystalline silicon TFT conduction to be of the same form as amorphous and single crystal devices with the degree of disorder as the sliding scale between the two extremes.

Optical modelling of cone concentrators for positive and negative IR emitters

InSb optical cone concentrators are modelled to assess their optical performance in positive and negative emission modes. The output distribution from the active layer is shown to be isotropic for thin active regions, tending towards Lambertian distribution as the layer thickness increases. Refraction from the emitting surface is shown to make the distribution more Lambertian. Optical efficiencies of straight-sided and Winston cone concentrators are modelled using a ray-tracing program, and those of straight-sided cones also determined using an analytical approximation, which allows use of the distribution derived earlier. The effect on the active layer thickness on the positive and negative emission is also determined. Normal light emitting diode structures are found to be poor positive emitters and to draw large currents when used as negative emitters. A Winston cone of area gain equal to n 2 is found to be the best option for devices required to work in both positive and negative modes. Intermediate structures are also considered. The optimum active layer thickness is derived for three emitter configurations.

1.4× efficiency improvement in transparent-substrate (AlxGa1−x)0.5In0.5P light-emitting diodes with thin (⩽2000 Å) active regions

Improvement of 1.4× in the external quantum efficiency and luminous efficiency (lm/W) of transparent-substrate (AlxGa1−x)0.5In0.5P/GaP light-emitting diodes is demonstrated. The improvement is accomplished by reducing the thickness of the active layer to ⩽2000 Å and increasing the internal quantum efficiency by using multiple thin (⩽500 Å) active layers. The maximum luminous efficiency achieved is 73.7 lm/W at λp∼615 nm and the maximum external quantum efficiency is 32.0% at λp∼632 nm.

A combined finite difference and analytic expression approach to crossover capacitance in a multilayer dielectric environment

The capacitances in the crossover regions of orthogonal transmission lines of finite thickness, fabricated on structures of the approximate dimensions of an integrated circuit, are evaluated in this paper by means of a combined finite difference and analytic expression method. The three-dimensional (3-D) finite difference method (FDM), using very fine mesh grids, is applied to an artificially defined region only a few microns in thickness, where orthogonal transmission lines on different metal layers of an integrated circuit cross one another. In the 600-/spl mu/m thick dielectric region above the ground plane on the lower surface of an integrated circuit, analytic expressions of the solution to the Laplace equation are formulated. An artificial boundary is assumed to separate the substrate of the 600-/spl mu/m thick integrated circuit from the thin active region on its uppermost surface where the active circuits and interconnects are actually fabricated, creating two separate regions which are treated using different approaches. An iterative procedure is employed to create a continuous interface between solutions across the boundary of the two regions generated by the two methods. The algorithm converges rapidly to very accurate solutions. The errors between this simulation method and laboratory measurements are within 8% for very small absolute values in the femtoFarad (fF) range. The field solutions are then converted into the equivalent circuit parameters. Finally, the waveshapes of propagating signal pulses are simulated by a networking program in a general electromagnetic modeling tool suite referred to as the Mayo Graphical Integrated Computer Aided Design Suite, or MagiCAD, in which the equivalent circuit model and capacitance values of the crossover problem are integrated.

Room temperature spontaneous emission in five-micron-long Fabry–Pérot vertical cavities

Five-micron long AlGaAs/GaAs Fabry–Pérot vertical cavities with thin GaAs active regions of two different thicknesses are analyzed both theoretically and experimentally in terms of their optical transmission characteristics, spontaneous radiation patterns, and spontaneous spectral emission characteristics. The effects on spontaneous emission of the precise placement of the GaAs active region and also the thickness of this region, as compared to the emitted wavelength, are demonstrated. The measured results are compared with theoretical calculations which are based on the first-order perturbation field theory presented in an earlier publication. Good agreement is found in the comparison between the theoretical predictions and the experimental results, and indicates the controllable spontaneous emission in these long vertical cavities.

Narrow-gap semiconductor magnetic-field sensors and applications

Narrow-gap semiconductors have been used for decades in the fabrication of magnetic field sensors, such as magnetoresistors and Hall sensors. Magnetic field sensors are, in turn, used in conjunction with permanent magnets to make contactless potentiometers and rotary encoders. This sensing technology offers the most reliable way to convert a mechanical movement into an electrical signal, and is widespread in automotive applications. Recent developments in the growth of thin epitaxial layers of InAs and InSb on semiinsulating GaAs or InP substrates have resulted in the development of magnetoresistors with excellent sensitivity and operating temperatures up to 285 degrees C. Magnetoresistors and Hall sensors require a very thin active semiconductor region, a high carrier density and a high room-temperature mobility. The best materials are narrow-gap III-V compounds. 2DEG layers in InSb and InAs would be ideally suited for these devices. The accumulation layer at the surface of InAs has been used to make magnetoresistors, Hall sensors and magnetotransistors. n-type doped thin InSb films are used to make magnetoresistors that outperform Si-based Hall sensors, even with integrated amplification. The authors describe device design criteria, materials requirements and a direct comparison of the three types of galvanomagnetic devices, magnetoresistors, Hall sensors and magnetotransistors, made from the same material. They compare the output of different magnetic field sensing technologies, such as Si and GaAs Hall sensors, and NiFe-based magnetoresistors, with InSb magnetoresistors.

High peak power picosecond pulse generation from AlGaAs external cavity mode-locked semiconductor laser and traveling-wave amplifier

Thin active region single stripe aluminum gallium arsenide traveling-wave amplifiers have been used to generate and amplify 15 ps optical pulses to nearly 50 mW average powers and to peak powers in excess of 3 W. No appreciable temporal pulse distortion is observed after amplification. These peak power levels should prove to be sufficient to induce nonlinear optical effects in fibers.

Noise enhancement in laser amplifiers caused by gain nonuniformity

We have calculated the noise-enhancement factor of a semiconductor laser amplifier with a very thin active region. This (somewhat academic) model exhibits large Petermann's K-factors. However, the calculated noise-enhancement factor K´ is only about half the K-value. The reason for the discrepancy is that in evaluating K´ the radiation modes of the guiding structure have been taken into account. Similar conclusions have been reached for realistic laser models. These theoretical considerations show that gain-guided laser amplifiers are not as noisy as was originally thought.

Three-layer waveguide InGaAsP/lnP injection lasers

Injection lasers with a three-layer waveguide emitting at the wavelength of about 1.3 μ were made and investigated. The liquid phase epitaxy method was used to form heterostructures on n- and p-type InP substrates and these heterostructures had lower threshold currents than conventional double-sided heterojunction lasers. A heterostructure with a three-layer waveguide had a thin active region (0.05-0.2 μ) consisting of a quaternary solid solution with adjoining layers of a transparent solid solution of intermediate composition which, together with the active layer, formed a waveguide for the laser radiation traveling in the medium with the wider band gap (InP). In the pulsed regime the minimum threshold current density was 930 A/cm2 at room temperature.

Modal analysis of separate-confinement heterojunction lasers with inhomogeneous cladding layers

Separate-confinement heterostructure lasers with thin active regions are analyzed. For four different interiorcladding-region refractive-index spatial variations, i.e., step, triangular, parabolic, and inverted parabolic, waveguidemodes are calculated and thresholds are compared. Based on optical considerations alone, the step-indexprofile has the lowest threshold; however, for optimum cladding thicknesses the differences are not great.

Reemission phenomenon in heterostructures with a thin active region

Reemission phenomenon in heterostructures with a thin active region

Electron-beam-pumped semiconductor laser operating at low acceleration voltages

An investigation was made of the threshold characteristics of a laser pumped transversely by an electron beam accelerated by voltages below 25 kV. The target was a three-layer heterostructure with a thin active region between two wide-gap layers. The lowest threshold energy of the electron beam was 10 keV at T≈90 °K (the corresponding threshold value of the current density was jth = 0.45 A/cm2) and 12 keV at T≈300 °K (jth≈0.9A/cm2).

High-power constricted double-heterojunction AlGaAs diode lasers for optical recording

In this letter we describe a new type of constricted double-heterojunction injection laser especially designed for optical disk recording. The laser is capable of steady operation in a stable fundamental spatial mode at powers of 80 mW/facet at 20% duty cycle (25-MHz repetition rate, 8-nsec-wide pulses) and 50 mW/facet at 50% duty cycle (10-MHz repetition rate, 50-nsec-wide pulses). Such devices have been used in high-data-rate optical recording and preliminary results indicate operating life in excess of 103 h at 40 mW/facet and 50% duty cycle. The laser employs a very thin active region (500–800 Å) and a nearly planar structure to provide as large a source size as possible while remaining in fundamental spatial mode operation. Single longitudinal mode cw operationwas recorded up to 21 mW/facet. It is also found that an antireflection coating on the emitting facet not only raises the overall efficiency and facet damage limit, but also can simplify the modal pattern.

A condition of single longitudinal mode operation in injection lasers with index-guiding structure

A condition for single longitudinal mode operation (SMO, for short) of index-guided injection lasers is given theoretically and supported by experiment. For SMO, the transverse higher modes must be cut off by using a narrow-width index-guiding waveguide. Inclusion of the spontaneous emission into the lasing field must be reduced by using a thinner active region. In terms of the impurity concentration of the active region, the undoped case is the most stable for temperature variation. A heavily doped active region may also produce SMO. The thermal resistance must be reduced to increase temperature stability. MO with a fixed lasing wavelength is experimentally obtained by temperature control up to an injection current of twice threshold.

A high-efficiency GaAlAs double-heterostructure photovoltaic detector

An antireflection-coated GaAlAs double-heterostructure photovoltaic detector is described whose extrinsic power conversion efficiency is 56% when used with a focused 8075-μm wavelength laser beam. This, to our knowledge, is the highest photovoltaic efficiency yet reported. A partially reflective rear contact provides extrinsic quantum efficiencies exceeding 0.90 with a relatively thin active region. By restricting the diameter of this contact to 50 μm and using high dopings to reduce series resistance, open-circuit voltages of 1.15 V and fill factors up to 0.84 are obtained. The series resistance was 3.0 Ω, with 1.0 Ω being attributable to the contacts.