SEARCH · Search NASA

Results for “wet etching”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records

All sources NASA NTRS DOE OSTI

NASA NTRS2021

Anisotropic Wet Etching of Si as a Fabrication Tool Enabling 3-D Microphotonics Structures and Devices

It is shown that due to high speed and large volume of production, anisotropic wet etching of Si is a unique method of manufacturing high quality micropyramidal arrays for silicon photonics. Four types of arrays were fabricated: (i) square pyramids with the 54.7 slope of the sidewall surface, (ii) microcones with the 45 slope of sidewall surface, (iii) inverted square pyramids with 54.7 slope, and (iv) inverted triangular pyramids with different angles of the sidewalls. It is shown that the first type of arrays has a capability to produce tightly focused mid-wave infrared (MWIR) beams at the tips of the pyramids. Other arrays can be also used as light concentrators and various multifunctional microphotonic components.

Astratov, Vasily N.↗

NASA NTRS2014

Electrochemical Method of Making Porous Particles Using a Constant Current Density

Provided is a particle that includes a first porous region and a second porous region that differs from the first porous region. Also provided is a particle that has a wet etched porous region and that does have a nucleation layer associated with wet etching. Methods of making porous particles are also provided.

Ferrari, Mauro↗

NASA NTRS2006

Selective Plasma Deposition of Fluorocarbon Films on SAMs

A dry plasma process has been demonstrated to be useful for the selective modification of self-assembled monolayers (SAMs) of alkanethiolates. These SAMs are used, during the fabrication of semiconductor electronic devices, as etch masks on gold layers that are destined to be patterned and incorporated into the devices. The selective modification involves the formation of fluorocarbon films that render the SAMs more effective in protecting the masked areas of the gold against etching by a potassium iodide (KI) solution. This modification can be utilized, not only in the fabrication of single electronic devices but also in the fabrication of integrated circuits, microelectromechanical systems, and circuit boards. In the steps that precede the dry plasma process, a silicon mold in the desired pattern is fabricated by standard photolithographic techniques. A stamp is then made by casting polydimethylsiloxane (commonly known as silicone rubber) in the mold. The stamp is coated with an alkanethiol solution, then the stamp is pressed on the gold layer of a device to be fabricated in order to deposit the alkanethiol to form an alkanethiolate SAM in the desired pattern (see figure). Next, the workpiece is exposed to a radio-frequency plasma generated from a mixture of CF4 and H2 gases. After this plasma treatment, the SAM is found to be modified, while the exposed areas of gold remain unchanged. This dry plasma process offers the potential for forming masks superior to those formed in a prior wet etching process. Among the advantages over the wet etching process are greater selectivity, fewer pin holes in the masks, and less nonuniformity of the masks. The fluorocarbon films formed in this way may also be useful as intermediate layers for subsequent fabrication steps and as dielectric layers to be incorporated into finished products.

Crain, Mark M., III↗

NASA NTRS1980

Precise wet-chemical etching

Controlled amount of etchant applied to surface of rotating sample removes only few angstroms of material. Technique is suited to study of chemical and crystal structures. Rate can be varied through control of spin frequency, liquid viscosity, droplet size, total etchant volume, etchant concentration.

Grunthaner, F. J.↗

NASA NTRS1993

Investigation of etching techniques for superconductive Nb/Al-Al2O3/Nb fabrication processes

Wet etching, CF4 and SF6 reactive ion etching (RIE), RIE/wet hybrid etching, Cl-based RIE, ion milling, and liftoff techniques have been investigated for use in superconductive Nb/Al-Al2O3/Nb fabrication processes. High-quality superconductor-insulator-superconductor (SIS) junctions have been fabricated using a variety of these etching methods; however, each technique offers distinct tradeoffs for a given process an wafer design. In particular, it was shown that SF6 provides an excellent RIE chemistry for low-voltage anisotropic etching of Nb with high selectivity to Al. The SF6 tool has greatly improved the trilevel resist junction insulation process. Excellent repeatability, selectivity with respect to quartz, and submicron resolution make Cl2 + BCl3 + CHCl3 RIE a very attractive process for trilayer patterning.

Lichtenberger, A. W.↗

NASA NTRS2003

A Dry-Etch Process for Low Temperature Superconducting Transition Edge Sensors for Far Infrared Bolometer Arrays

The next generation of ultra-low power bolometer arrays, with applications in far infrared imaging, spectroscopy and polarimetry, utilizes a superconducting bilayer as the sensing element to enable SQUID multiplexed readout. Superconducting transition edge sensors (TES s) are being produced with dual metal systems of superconductinghormal bilayers. The transition temperature (Tc) is tuned by altering the relative thickness of the superconductor with respect to the normal layer. We are currently investigating MoAu and MoCu bilayers. We have developed a dry-etching process for MoAu TES s with integrated molybdenum leads, and are working on adapting the process to MoCu. Dry etching has the advantage over wet etching in the MoAu system in that one can achieve a high degree of selectivity, greater than 10, using argon ME, or argon ion milling, for patterning gold on molybdenum. Molybdenum leads are subsequently patterned using fluorine plasma.. The dry-etch technique results in a smooth, featureless TES with sharp sidewalls, no undercutting of the Mo beneath the normal metal, and Mo leads with high critical current. The effects of individual processing parameters on the characteristics of the transition will be reported.

Allen, Christine A.↗

NASA NTRS2013

Fabrication Methods for Adaptive Deformable Mirrors

Previously, it was difficult to fabricate deformable mirrors made by piezoelectric actuators. This is because numerous actuators need to be precisely assembled to control the surface shape of the mirror. Two approaches have been developed. Both approaches begin by depositing a stack of piezoelectric films and electrodes over a silicon wafer substrate. In the first approach, the silicon wafer is removed initially by plasmabased reactive ion etching (RIE), and non-plasma dry etching with xenon difluoride (XeF2). In the second approach, the actuator film stack is immersed in a liquid such as deionized water. The adhesion between the actuator film stack and the substrate is relatively weak. Simply by seeping liquid between the film and the substrate, the actuator film stack is gently released from the substrate. The deformable mirror contains multiple piezoelectric membrane layers as well as multiple electrode layers (some are patterned and some are unpatterned). At the piezolectric layer, polyvinylidene fluoride (PVDF), or its co-polymer, poly(vinylidene fluoride trifluoroethylene P(VDF-TrFE) is used. The surface of the mirror is coated with a reflective coating. The actuator film stack is fabricated on silicon, or silicon on insulator (SOI) substrate, by repeatedly spin-coating the PVDF or P(VDFTrFE) solution and patterned metal (electrode) deposition. In the first approach, the actuator film stack is prepared on SOI substrate. Then, the thick silicon (typically 500-micron thick and called handle silicon) of the SOI wafer is etched by a deep reactive ion etching process tool (SF6-based plasma etching). This deep RIE stops at the middle SiO2 layer. The middle SiO2 layer is etched by either HF-based wet etching or dry plasma etch. The thin silicon layer (generally called a device layer) of SOI is removed by XeF2 dry etch. This XeF2 etch is very gentle and extremely selective, so the released mirror membrane is not damaged. It is possible to replace SOI with silicon substrate, but this will require tighter DRIE process control as well as generally longer and less efficient XeF2 etch. In the second approach, the actuator film stack is first constructed on a silicon wafer. It helps to use a polyimide intermediate layer such as Kapton because the adhesion between the polyimide and silicon is generally weak. A mirror mount ring is attached by using adhesive. Then, the assembly is partially submerged in liquid water. The water tends to seep between the actuator film stack and silicon substrate. As a result, the actuator membrane can be gently released from the silicon substrate. The actuator membrane is very flat because it is fixed to the mirror mount prior to the release. Deformable mirrors require extremely good surface optical quality. In the technology described here, the deformable mirror is fabricated on pristine substrates such as prime-grade silicon wafers. The deformable mirror is released by selectively removing the substrate. Therefore, the released deformable mirror surface replicates the optical quality of the underlying pristine substrate.

Toda, Risaku↗

NASA NTRS2008

Patterning Techniques for Magnesium Diboride Films for Superconducting Transition Edge Sensor Bolometer Arrays

We present a comparison of several patterning techniques for magnesium diboride films with the goal of making membrane-supported superconducting transition edge sensor bolometer arrays suited for far infrared exploration of outer planets and icy moons. We compare etch properties such as etch rate, selectivity to resist, and edge roughness for wet etching, reactive ion etching, and ion-milling processes.

Travers, D. E.↗

NASA NTRS1982

Holographic fabrication of gratings in metal substrates

A program for investigating the grain enlargement resulting from the laser recrystallization of a thin gallium arsenide film on a patterned substrate, a technique known as graphoepitaxy was evaluated. More specifically, the effects of recrystallizing an uncapped gallium arsenide film using a continuous wave neodymium YAG laser operating at 1.06 microns were studied. In an effort to minimize arsenic loss from the film, the specimens were held in an arsine atmosphere during recrystallization. Two methods for fabricating patterned substrates were developed, one using reactive ion etching of a molybdenum film on both sapphire and silicon substates and another by preferential wet etching of a silicon substrate onto which a film of molybdenum was subsequently deposited.

Fletcher, R. M.↗

NASA NTRS1980

The development of a method of producing etch resistant wax patterns on solar cells

A potentially attractive technique for wax masking of solar cells prior to etching processes was studied. This technique made use of a reuseable wax composition which was applied to the solar cell in patterned form by means of a letterpress printing method. After standard wet etching was performed, wax removal by means of hot water was investigated. Application of the letterpress wax printing process to silicon was met with a number of difficulties. The most serious shortcoming of the process was its inability to produce consistently well-defined printed patterns on the hard silicon cell surface.

Pastirik, E.↗

NASA NTRS1998

160-190 GHz Monolithic Low Noise Amplifiers

This paper presents the results of two 160-190 GHz monolithic low noise amplifiers (LNAs) fabricated with 0.07-microns pseudomorphic (PM) InAlAs/InGaAs/InP HEMT technology using a reactive ion etch (RIE) via hole process. A peak small signal gain of 9 dB was measured at 188 GHz for the first LNA with a 3-dB bandwidth from 164 to 192 GHz while the second LNA has achieved over 6-dB gain from 142 to 180 GHz. The same design (second LNA) was also fabricated with 0.08-micron gate and a wet etch process, showing a small signal gain of 6 dB with noise figure 6 dB. All the measurement results were obtained via on-wafer probing. The LNA noise measurement at 170 GHz is also the first attempt at this frequency.

Kok, Y. L.↗

NASA NTRS1987

Surface characterization of InP using photoluminescence

Photoluminescence (PL) measurements have been performed on InP samples in situ during various surface treatments including chemical etching, wet anodization, and low-pressure chemical vapor deposition. It was found, in agreement with previously published results, that the magnitude of the PL signal varies markedly with surface treatment due presumably to changes in either surface-state density, and/or surface potential. In an attempt to assess the effectiveness of this noninvasive method as a tool for characterizing and monitoring the progressive development of a semiconductor surface during processing, a number of experiments on InP have been performed. The results indicate that although some uncertainty may exist in assigning a mechanism for the PL change in any given experiment, the general trend appears to be that surface degradation results in a reduced signal. As a result, process steps which enhance the PL intensity are likely to be beneficial in the preparation of a high-quality interface.

Chang, R. R.↗

NASA NTRS1987

Improved beta-SiC heteroepitaxial films using off-axis Si substrates

All beta-SiC films grown using on-axis (001) Si substrates that have been examined with transmission electron microscopy exhibit a high density of interfacial twins, stacking faults, and antiphase disorder. The antiphase boundaries can be decorated by chemical etching, sputter etching, wet oxidation, and beta-SiC growth in the presence of diborane. All traces of antiphase disorder are eliminated when the heteroepitaxial growth is carried out on vicinal (001) Si substrates that are tilted 2 deg about a 110-line axis. In addition, growth on the off-axis Si produces beta-SiC films that are significantly smoother than on-axis films. The density of stacking faults is apparently unaffected by growth on the off-axis substrates.

Powell, J. A.↗

NASA NTRS1992

Microwave properties of peeled HEMT devices sapphire substrates

The focus of this research is to demonstrate the first full radio frequency characterization of high electron mobility transistor (HEMT) device parameters. The results of this research are used in the design of circuits with peeled HEMT devices, e.g. 10 GHz amplifiers. Devices were fabricated using two HEMT structures grown by molecular beam epitaxy methods. A 500 A AlAs release layer for 'peel off' was included under the active layers of the structure. The structures are a homogeneously doped Al(0.3)GA(0.7)As/GaAs and a delta doped square well Al(.23)Ga(.77)As/GaAs HEMT structure. Devices were fabricated using a mesa isolation process. Contacts were done by sequentially evaporating Au/Ge/Au/Ni/Au followed by rapid thermal anneal at 400 C for 15 seconds. Gates were wet etch recessed and 1 to 1.4 micron Ti/Au gate metal was deposited. Devices were peeled off the GaAs substrate using Apiezon wax to support the active layer and a HF:DI (1:10) solution to remove the AlAs separation layer. Devices were then attached to sapphire substrates using van der Waals bonding.

Young, Paul G.↗

NASA NTRS2001

MEMS Incandescent Light Source

A MEMS-based, low-power, incandescent light source is being developed. This light source is fabricated using three bonded chips. The bottom chip consists of a reflector on Silicon, the middle chip contains a Tungsten filament bonded to silicon and the top layer is a transparent window. A 25-micrometer-thick spiral filament is fabricated in Tungsten using lithography and wet-etching. A proof-of-concept device has been fabricated and tested in a vacuum chamber. Results indicate that the filament is electrically heated to approximately 2650 K. The power required to drive the proof-of-concept spiral filament to incandescence is 1.25 W. The emitted optical power is expected to be approximately 1.0 W with the spectral peak at 1.1 microns. The micromachining techniques used to fabricate this light source can be applied to other MEMS devices.

Tuma, Margaret↗

NASA NTRS2001

MEMS Device Being Developed for Active Cooling and Temperature Control

High-capacity cooling options remain limited for many small-scale applications such as microelectronic components, miniature sensors, and microsystems. A microelectromechanical system (MEMS) is currently under development at the NASA Glenn Research Center to meet this need. It uses a thermodynamic cycle to provide cooling or heating directly to a thermally loaded surface. The device can be used strictly in the cooling mode, or it can be switched between cooling and heating modes in milliseconds for precise temperature control. Fabrication and assembly are accomplished by wet etching and wafer bonding techniques routinely used in the semiconductor processing industry. Benefits of the MEMS cooler include scalability to fractions of a millimeter, modularity for increased capacity and staging to low temperatures, simple interfaces and limited failure modes, and minimal induced vibration.

Moran, Matthew E.↗

NASA NTRS2007

Compact Packaging of Photonic Millimeter-Wave Receiver

A carrier structure made from a single silicon substrate is the basis of a compact, lightweight, relatively inexpensive package that holds the main optical/electronic coupling components of a photonic millimeter-wave receiver based on a lithium niobate resonator disk. The design of the package is simple and provides for precise relative placement of optical components, eliminating the need for complex, bulky positioning mechanisms like those commonly used to align optical components to optimize focus and coupling. Although a prototype of the package was fabricated as a discrete unit, the design is amenable to integration of the package into a larger photonic and/or electronic receiver system. The components (see figure) include a lithium niobate optical resonator disk of 5-mm diameter and .200- m thickness, positioned adjacent to a millimeter- wave resonator electrode. Other components include input and output coupling prisms and input and output optical fibers tipped with ball lenses for focusing and collimation, respectively. Laser light is introduced via the input optical fiber and focused into the input coupling prism. The input coupling prism is positioned near (but not in contact with) the resonator disk so that by means of evanescent-wave coupling, the input laser light in the prism gives rise to laser light propagating circumferentially in guided modes in the resonator disk. Similarly, a portion of the circumferentially propagating optical power is extracted from the disk by evanescent-wave coupling from the disk to the output coupling prism, from whence the light passes through the collimating ball lens into the output optical fiber. The lens-tipped optical fibers must be positioned at a specified focal distance from the prisms. The optical fibers and the prisms must be correctly positioned relative to the resonator disk and must be oriented to obtain the angle of incidence (55 in the prototype) required for evanescent-wave coupling of light into and out of the desired guided modes in the resonator disk. To satisfy all these requirements, precise alignment features are formed in the silicon substrate by use of a conventional wet-etching process. These features include a 5-mm-diameter, 50- m-deep cavity that holds the disk; two trapezoidal-cross-section recesses for the prisms; and two grooves that hold the optical fibers at the correct positions and angles relative to the prisms and disk. The fiber grooves contain abrupt tapers, near the prisms, that serve as hard stops for positioning the lenses at the focal distance from the prisms. There are also two grooves for prismadjusting rods. The design provides a little slack in the prism recesses for adjusting the positions of the prisms by means of these rods to optimize the optical coupling.

Nguyen, Hung↗

NASA NTRS2007

Microrectenna: A Terahertz Antenna and Rectifier on a Chip

A microrectenna that would operate at a frequency of 2.5 THz has been designed and partially fabricated. The circuit is intended to be a prototype of an extremely compact device that could be used to convert radio-beamed power to DC to drive microdevices (see Figure 1). The microrectenna (see Figure 2) circuit consists of an antenna, a diode rectifier and a DC output port. The antenna consists of a twin slot array in a conducting ground plane (denoted the antenna ground plane) over an enclosed quarter-wavelength-thick resonant cavity (denoted the reflecting ground plane). The circuit also contains a planar high-frequency low-parasitic Schottky-barrier diode, a low-impedance microstrip transmission line, capacitors, and contact beam leads. The entire 3-D circuit is fabricated monolithically from a single GaAs wafer. The resonant cavity renders the slot radiation pattern unidirectional with a half-power beam width of about 65. A unique metal mesh on the rear of the wafer forms the backplate for the cavity but allows the GaAs to be wet etched from the rear surface of the twin slot antennas and ground plane. The beam leads protrude past the edge of the chip and are used both to mount the microrectenna and to make the DC electrical connection with external circuitry. The antenna ground plane and the components on top of it are formed on a 2- m thick GaAs membrane that is grown in the initial wafer MBE (molecular beam epitaxy) process. The side walls of the antenna cavity are not metal coated and, hence, would cause some loss of power; however, the relatively high permittivity (epsilon=13) of the GaAs keeps the cavity modes well confined, without the usual surface-wave losses associated with thick dielectric substrates. The Schottky-barrier diode has the usual submicron dimensions associated with THz operation and is formed in a mesa process above the antenna ground plane. The diode is connected at the midpoint of a microstrip transmission line, which is formed on 1- m-thick SiO (permittivity of 5) laid down on top of the GaAs membrane. The twin slots are fed in phase by this structure. To prevent radio-frequency (RF) leakage past the slot antennas, low-loss capacitors are integrated into the microstrip transmission line at the edges of the slots. The DC current- carrying lines extend from the outer edges of the capacitors, widen approaching the edges of the chip, and continue past the edges of the chip to become the beam leads used in tacking down the devices. The structure provides a self-contained RF to DC converter that works in the THz range.

Siegel, Peter↗