Layer dependent thermal transport properties of one- to three-layer magnetic Fe:MoS2
Not provided.
Engineering topics
Publications and source records attributed to Yang, Eui-Hyeok.
Not provided.
Lightweight, flexible, large-aperture mirrors of a type being developed for use in outer space have unimorph structures that enable precise adjustment of their surface figures. A mirror of this type includes a reflective membrane layer bonded with an electrostrictive grafted elastomer (G-elastomer) layer, plus electrodes suitably positioned with respect to these layers. By virtue of the electrostrictive effect, an electric field applied to the G-elastomer membrane induces a strain along the membrane and thus causes a deflection of the mirror surface. Utilizing this effect, the mirror surface figure can be adjusted locally by individually addressing pairs of electrodes. G-elastomers, which were developed at NASA Langley Research Center, were chosen for this development in preference to other electroactive polymers partly because they offer superior electromechanical performance. Whereas other electroactive polymers offer, variously, large strains with low moduli of elasticity or small strains with high moduli of elasticity, G-elastomers offer both large strains (as large as 4 percent) and high moduli of elasticity (about 580 MPa). In addition, G-elastomer layers can be made by standard melt pressing or room-temperature solution casting.
A process for fabricating a planar array of dots having characteristic dimensions of the order of several nanometers to several hundred nanometers involves the formation and use of a thin alumina nanopore template on a semiconductor substrate. The dot material is deposited in the nanopores, then the template is lifted off the substrate after the dots have been formed. This process is expected to be a basis for development of other, similar nanofabrication processes for relatively inexpensive mass production of nanometerscale optical, optoelectronic, electronic, and magnetic devices. Alumina nanopore templates are self-organized structures that result from anodization of aluminum under appropriate conditions. Alumina nanopore templates have been regarded as attractive for use in fabricating the devices mentioned above, but prior efforts to use alumina nanopore templates for this purpose have not been successful. One reason for the lack of success is that the aspect ratios (ratios between depth and diameter) of the pores have been too large: large aspect ratios can result in blockage of deposition and/or can prevent successful lift-off. The development of the present process was motivated partly by a requirement to reduce aspect ratios to values (of the order of 10) for which there is little or no blockage of deposition and attempts at lift-off are more likely to be successful. The fabrication process is outlined.
An experimental inchworm-type linear microactuator is depicted. This microactuator is a successor to one described in "MEMS-Based Piezoelectric/Electrostatic Inchworm Actuator" (NPO-30672), NASA Tech Briefs, Vol. 27, No. 6 (June 2003), page 68. Both actuators are based on the principle of using a piezoelectric transducer operated in alternation with electrostatically actuated clutches to cause a slider to move in small increments. However, the design of the present actuator incorporates several improvements over that of the previous one. The most readily apparent improvement is in geometry and, consequently, in fabrication: In the previous actuator, the inchworm motion was perpendicular to the broad faces of a flat silicon wafer on which the actuator was fabricated, and fabrication involved complex processes to form complex three-dimensional shapes in and on the wafer. In the present actuator, the inchworm motion is parallel to the broad faces of a wafer on which it is fabricated. The components needed to produce the in-plane motion are nearly planar in character and, consequently, easier to fabricate. Other advantages of the present design are described, including that the previous actuator contained two clutches (denoted 'holders' in the cited prior article), the present actuator contains four clutches. The operational sequence of the previous two-clutch actuator is similar. However, the two-clutch configuration is susceptible to tilt of the slider and a consequent large increase in drag. Hence, the primary operational advantages of the present four-point-latching design over the prior two-point-latching design are less drag and greater control robustness arising from greater stability of the orientation of the slider.
A class of proposed thermionic cooling devices would incorporate precise arrays of metal nanowires as electron emitters. The proposed devices could be highly miniaturized, enabling removal of heat from locations, very close to electronic devices, that have previously been inaccessible for heat-removal purposes. The resulting enhancement of removal of heat would enable operation of the devices at higher power levels and higher clock speeds. Moreover, the mass, complexity, and bulk of electronic circuitry incorporating these highly miniaturized cooling devices could be considerably reduced, relative to otherwise equivalent circuitry cooled by conventional electromechanical, thermoelectric, and fluidic means. In thermionic cooling, one exploits the fact that because only the highest-energy electrons are thermionically emitted, collecting those electrons to prevent their return to the emitting electrode results in the net removal of heat from that electrode. Collection is effected by applying an appropriate positive bias potential to another electrode placed near the emitting electrode. The concept underlying the proposal is that the thermionic-emission current and, hence, the cooling effect attainable by use of an array of nanowires could be significantly greater than that attainable by use of a single emitting electrode or other electron- emitting surface. The wires in an array according to the proposal would protrude perpendicularly from a planar surface and their heights would be made uniform to within a sub-nanometer level of precision
A method of fabricating channels having widths of tens of nanometers in silicon substrates and burying the channels under overlying layers of dielectric materials has been demonstrated. With further refinement, the method might be useful for fabricating nanochannels for manipulation and analysis of large biomolecules at single-molecule resolution. Unlike in prior methods, burying the channels does not involve bonding of flat wafers to the silicon substrates to cover exposed channels in the substrates. Instead, the formation and burying of the channels are accomplished in a more sophisticated process that is less vulnerable to defects in the substrates and less likely to result in clogging of, or leakage from, the channels. In this method, the first step is to establish the channel pattern by forming an array of sacrificial metal nanowires on an SiO2-on-Si substrate. In particular, the wire pattern is made by use of focused-ion-beam (FIB) lithography and a subsequent metallization/lift-off process. The pattern of metal nanowires is then transferred onto the SiO2 layer by reactive-ion etching, which yields sacrificial SiO2 nanowires covered by metal. After removal of the metal covering the SiO2 nanowires, what remains are SiO2 nanowires on an Si substrate. Plasma-enhanced chemical vapor deposition (PECVD) is used to form a layer of a dielectric material over the Si substrate and over the SiO2 wires on the surface of the substrate. FIB milling is then performed to form trenches at both ends of each SiO2 wire. The trenches serve as openings for the entry of chemicals that etch SiO2 much faster than they etch Si. Provided that the nanowires are not so long that the diffusion of the etching chemicals is blocked, the sacrificial SiO2 nanowires become etched out from between the dielectric material and the Si substrate, leaving buried channels. At the time of reporting the information for this article, channels 3 m long, 20 nm deep, and 80 nm wide (see figure) had been fabricated by this method.
A document presents the concept of a curved telescope primary reflector structure, made mostly of silicon, that would have an areal mass density 1 kg/m2 and would be deployed in outer space, where it would be operated at a temperature in the cryogenic range. The concept provides for adjustment of the shape of the mirror to maintain the required precise optical surface figure despite the flexibility inherent in the ultra-lightweight design. The structure would include a thin mirror layer divided into hexagonal segments supported by flexure hinges on a lightweight two-layer backing structure. Each segment would also be supported at three points by sets of piezoelectric linear microactuators that could impose small displacements along the optical axis. The excitations applied to the aforementioned microactuators would be chosen to effect fine adjustments of the axial positions and the orientations of the segments relative to the supporting structure. Other piezoelectric linear microactuators embedded in the backing structure would enable control of the displacements of the segments along the hexagonal axes; they would also enable control of the curvature of the backing structure and, thus, additional control of the curvature of the reflector.
This paper describes an optimization of PZT unimorph membrane microactuators in view of their application to deformable mirrors (DMs).
Back actuators have been proposed as alternatives to edge actuators considered previously for use in aligning hexagonal segments of lightweight segmented astronomical mirrors planned for use in outer space. The proposed back actuators could also be useful on Earth as parts of wafer-conveyance systems in the semiconductor industry. Whereas the prior edge actuators were required to impose rotations and torques (in addition to forces and displacements) at joints between mirror segments, the proposed back actuators would be required to impose only forces and displacements (sometimes accompanied by small incidental torques and rotations). The advantages of the back-actuation approach, relative to the edge-actuation approach, are that the actuation mechanisms could be made simpler and a single overall actuation scheme could incorporate what were previously separate actuation schemes for (1) orienting the mirror segments at the required angles and (2) placing the mirror segments at the required distances along the optical axis from the focus. Each hexagonal mirror segment would be supported at three points by sets of linear actuators (see figure). The linear actuators at each support point would include one to impose displacement along the optical axis (the z axis in the figure) plus one or two to impose displacement along one or two of the hexagonal axes. The linear actuators could be, for example, shape-memory-alloy actuators or piezoelectric actuators that move in the manner of an inchworm like those described in several previous NASA Tech Briefs articles
For the first time, this paper reports on the piezoelectric unimorph-based deformable mirror (DM) with continuous single-crystal silicon deformable membrane.
A leak-tight, low-power, liquid-compatible, piezoelectric microvalve has been designed, fabricated, and fully characterized.
An alternative approach to the design and fabrication of vibratory gyroscopes is founded on the use of fabrication techniques that yield best results in the mesoscopic size range, which is characterized by overall device dimensions of the order of a centimeter. This approach stands in contradistinction to prior approaches in (1) the macroscopic size range (the size range of conventional design and fabrication, characterized by overall device dimensions of many centimeters) and (2) the microscopic size range [the size range of microelectromechanical systems (MEMS), characterized by overall device dimensions of the order of a millimeter or less]. The mesoscale approach offers some of the advantage of the MEMS approach (sizes and power demands smaller than those of the macroscale approach) and some of the advantage of the macroscale approach (the possibility of achieving relative dimensional precision greater than that of the MEMS approach). Relative dimensional precision is a major issue in the operation of a vibratory gyroscope. The heart of a vibratory gyroscope is a mechanical resonator that is required to have a specified symmetry in a plane orthogonal to the axis about which rotation is to be measured. If the resonator could be perfectly symmetrical, then in the absence of rotation, a free vibration of the resonator could remain fixed along any orientation relative to its housing; that is, the gyroscope could exhibit zero drift. In practice, manufacturing imprecision gives rise to some asymmetry in mass, flexural stiffness or dissipation, resulting in a slight drift or beating motion of an initial vibration pattern that cannot be distinguished from rotation. In the mesoscale approach, one exploits the following concepts: For a given amount of dimensional error generated in manufacturing, the asymmetry and hence the rate-of-rotation drift of the gyroscope can be reduced by increasing the scale. The decrease in asymmetry also reduces coupling of vibrations to the external environment. Mechanical thermal noise and electronic measurement noise and drift can also be reduced by increasing the size of the resonator and its associated sensors.
A method, now undergoing development, of forming nanochannels in planar substrates is intended to enable the fabrication of advanced fluidic devices that could be integrated with complementary metal oxide semiconductor (CMOS) electronic circuits. Such integral combinations of fluidic and electronic components ("laboratory-on-a-chip" devices) could be used, for example, to detect individual molecules of deoxyribonucleic acid (DNA) and proteins. The width of a channel in such a device would be chosen so that molecules of the species of interest would move along the channel in single file. In this method, the fabrication of channels includes the use of such CMOS-compatible processes as chemical-mechanical polishing and oxide deposition. The layout of the channels in the substrate plane is determined by a single photolithographic process, but it is not a nanoscale lithographic process, and this process is not relied upon to define the thickness and width of the channels. Stating it from a slightly different perspective, unlike in the prior fabrication of electronic and fluidic devices involving the use of lithography to define microscale features, this process does not include the use of lithography to define nanoscale features. It is this aspect of the method that enables simplification of the process and, hence, a decrease in cost.
Modifications have been proposed to effect further improvement of the device described in Improved Piezoelectrically Actuated Microvalve (NPO-30158), NASA Tech Briefs, Vol. 26, No. 1 (January 2002), page 29. To recapitulate: What is being developed is a prototype of valves for microfluidic systems and other microelectromechanical systems (MEMS). The version of the valve reported in the cited previous article included a base (which contained a seat, an inlet, and an outlet), a diaphragm, and a linear actuator. With the exception of the actuator, the parts were micromachined from silicon. The linear actuator consisted of a stack of piezoelectric disks in a rigid housing. To make the diaphragm apply a large sealing force on the inlet and outlet, the piezoelectric stack was compressed into a slightly contracted condition during assembly of the valve. Application of a voltage across the stack caused the stack to contract into an even more compressed condition, lifting the diaphragm away from the seat, thereby creating a narrow channel between the inlet and outlet. The positions of the inlet and outlet, relative to the diaphragm and seat, were such that the inlet flow and pressure contributed to sealing and thus to a desired normally-closed mode of operation.
A proposed normally-closed microvalve would contain a piezoelectric bending actuator instead of a piezoelectric linear actuator like that of the microvalve described in the preceding article. Whereas the stroke of the linear actuator of the preceding article would be limited to approximately equal to 6 micrometers, the stroke of the proposed bending actuator would lie in the approximate range of 10 to 15 micrometers-large enough to enable the microvalve to handle a variety of liquids containing suspended particles having sizes up to 10 m. Such particulate-laden liquids occur in a variety of microfluidic systems, one example being a system that sorts cells or large biomolecules for analysis. In comparison with the linear actuator of the preceding article, the bending actuator would be smaller and less massive. The combination of increased stroke, smaller mass, and smaller volume would be obtained at the cost of decreased actuation force: The proposed actuator would generate a force in the approximate range of 1 to 4 N, the exact amount depending on operating conditions and details of design. This level of actuation force would be too low to enable the valve to handle a fluid at the high pressure level mentioned in the preceding article. The proposal encompasses two alternative designs one featuring a miniature piezoelectric bimorph actuator and one featuring a thick-film unimorph piezoelectric actuator (see figure). In either version, the valve would consume a power of only 0.01 W when actuated at a frequency of 100 Hz. Also, in either version, it would be necessary to attach a soft elastomeric sealing ring to the valve seat so that any particles that settle on the seat would be pushed deep into the elastomeric material to prevent or reduce leakage. The overall dimensions of the bimorph version would be 7 by 7 by 1 mm. The actuator in this version would generate a force of 1 N and a stroke of 10 m at an applied potential of 150 V. The actuation force would be sufficient to enable the valve to handle a fluid pressurized up to about 50 psi (approximately equal to 0.35 MPa). The overall dimensions of the unimorph version would be 2 by 2 by 0.5 mm. In this version, an electric field across the piezoelectric film on a diaphragm would cause the film to pull on, and thereby bend, the diaphragm. At an applied potential of 20 V, the actuator in this version would generate a stroke of 10 micrometers and a force of 0.01 N. This force level would be too low to enable handling of fluids at pressures comparable to those of the bimorph version. This version would be useful primarily in microfluidic and nanofluidic applications that involve extremely low differential pressures and in which there are requirements for extreme miniaturization of valves. Examples of such applications include liquid chromatography and sequencing of deoxyribonucleic acid.
We have developed a new nanochannel fabrication technique using chemical-mechanical polishing (CMP) and thermal oxidation. With this technique, it is possible to control the width, length, and depth of the nanochannels without the need for nanolithography. The use of sacrificial SiO2 layers allows the fabrication of centimeter-long nanochannels. In addition, the fabrication process is CMOS compatible. We have successfully fabricated an array of extremely long and narrow nanochannels (i.e. 10 mm long, 25 nm wide and 100 nm deep) with smooth inner surfaces.
A proposed inchworm actuator, to be designed and fabricated according to the principles of microelectromechanical systems (MEMS), would effect linear motion characterized by steps as small as nanometers and an overall range of travel of hundreds of microns. Potential applications for actuators like this one include precise positioning of optical components and active suppression of noise and vibration in scientific instruments, conveyance of wafers in the semiconductor industry, precise positioning for machine tools, and positioning and actuation of micro-surgical instruments. The inchworm motion would be generated by a combination of piezoelectric driving and electrostatic clamping. The actuator (see figure), would include a pair of holders (used for electrostatic clamping), a slider (the part that would engage in the desired linear motion), a driver, a piezoelectric stack under the driver, and a pair of polymer beams centrally clamped to the flexure beam via a T bar. The holders would be held stationary. One end of the piezoelectric stack would be held stationary; the other end would be connected to the bottom of the driver, which would be free to move up and down. All of these components except the piezoelectric stack and the polymer beams would be micromachined from a 500- m-thick silicon wafer by deep reactive-ion etching. The inchworm motion would be perpendicular to the broad faces of the wafer (perpendicular to the plane of the figure). The combination of the polymer beams and the centrally clamped flexure beam would spring-bias the slider into a position such that, in the absence of electrostatic clamping, the gap between the slider on the one hand and both the driver and the holder on the other hand would be no more than a few microns. This arrangement would make it possible to electrostatically pull the slider into contact with either the holders or the driver at a clamping force of the order of 1 N by applying a reasonably small voltage (of the order of 100 V).
Deformable mirrors of a proposed type would be equipped with relatively-large-stroke microscopic piezoelectric actuators that would be used to maintain their reflective surfaces in precise shapes. These mirrors would be members of the class of MEMS-DM (for microelectromechanical system deformable mirror) devices, which offer potential for a precise optical control in adaptive-optics applications in such diverse fields as astronomy and vision science. The proposed mirror would be fabricated, in part, by use of a membrane-transfer technique. The actuator design would contain bimorph-type piezoelectric actuators.