Search

Desire to integrate electronic devices more closely into the everyday life of the consumer led to the emergence of flexible devices, which can conform to any shape surface. These flexible devices are lighter, thinner, and more resilient than their rigid counterparts. Since consumers these days are heavily concerned about the visual look of any product along with its technological features, the flexible devices also pave the way for innovation and creativity from the aesthetic point of view. As more and more electronics including wearable devices and displays are shifting towards the flexible regime, their basic components like sensors and integrated circuits are also required to be flexible and conformal. This paper reviews the development and evolution of several types of flexible MEMS sensors, accelerometers, and energy harvesters from the design, fabrication, and performance perspective.

We investigated the effect of channel hot carrier (CHC) stressing on the creation of gate oxide defect centers causing random telegraph signals (RTSs) in pMOSFETs. To identify these stress-induced hole traps in the oxide and fully characterize their trapping properties, variable temperature RTS measurements were performed from room temperature down to 215 K and repeated after each stressing interval, up to 1200 s. The trapping properties of the stress-induced defect sites and those present before stress are quantified, including the hole capture and emission activation energies, structural relaxation in the oxide network due to trapping/detrapping, and entropy change as well as the energy level and position of the trap in the oxide. Based on this information, disassociated III-Si and hydrogen bridge defects are identified as the hole trapping centers in SiO2. Difference was observed between the process- and stress-induced traps in the entropy change upon hole emission from the oxide trap to Si channel (electron capture from the Si valence band), implying the possibility of stress-induced structural defects being different than the native ones.

Although it is the worst degradation mechanism, the effect of channel hot carrier (CHC) stressing on random telegraph signals (RTS) has not been given enough attention in pMOSFETs. We report on the effect of CHC stressing on different RTS trap parameters namely screened scattering coefficient which controls the amount of charge carrier mobility fluctuations due to remote Coulomb scattering by the trap, RTS fluctuation amplitude, average capture time and capture cross-section. The generation of positive fixed oxide charge with stressing influences the screened Coulomb scattering of the channel carriers and therefore their mobility, in addition to the commonly accepted self-screening of the channel carriers. The two-dimensional mobility fluctuations model is adopted for analyzing the CHC effect on pMOSFETs. The comparison between theoretically and experimentally obtained screened scattering coefficients points toward the impact of the newly generated positive fixed oxide charge due to stressing. The decrease of the screened scattering coefficient results in increased RTS amplitude and decreased contribution of mobility fluctuations to RTS amplitude. Capture time and capture cross-section also change as the capture cross-section pre-factor has been impacted by the reduction of relaxation energy with stressing.

In recent years, material printing on flexible and disposable substrates has received substantial interest for realizing low cost electronic and sensing systems on large-areas. As a result, new methods such as printing and additive manufacturing, etc. have received significant attention to develop a wide range of easily deployable systems such as displays, sensors, and RFID smart tags. The technologies developed for silicon-based planar electronics and solid-state sensors is also being repurposed to meet the demands in the emerging field of flexible and printed sensors. This trend will continue as all market indicators point towards significant growth in this area. The exponential growth in flexible, printed and disposable sensor technology, the development of novel printable and solution processable nanomaterials, and the development of novel printing techniques have brought about a new wave of novel printed sensor technologies and applications. Considering this growth, it became necessary to provide a forum to discuss latest developments in the field and to bring together academia and industry to lay the future roadmap. The IEEE International Conference on Flexible and Printable Sensors and Systems (IEEE FLEPS) aims to be such a forum. The 2019 conference was the first edition, started by IEEE Sensors Council, with an aim to be a premier conference in the world. The high standards of the conference will be ensured by the steering committee and the track chairs formed by world-leading experts in the field.

We report on the investigation of random telegraph signals (RTSs) in pMOSFETs at variable temperatures to identify and characterize the hole traps residing at the oxide–semiconductor interface. Attractive center defects as well as a repulsive center defect are identified with trap location, capture cross section, remote Coulomb scattering coefficient, capture barrier energy, trap binding enthalpy, relaxation energy, and change in entropy with hole emission. The attractive centers responsible for RTS were a pair of dissociated three-coordinated silicon (D-III-Si) defects while the repulsive center was identified as a puckered/back-projected oxygen vacancy defect, [a pair of under-coordinated silicon (III-Si) and over-coordinated oxygen (III-O) defects]. Both contain ${E}\gamma ^{\prime }$ centers. This is the first time when such detailed analysis is reported on the hole traps in SiO2 which are responsible for RTS.

This paper presents a wafer-level vacuum-packaged z-axis capacitive micro accelerometer fabricated on a flexible polyimide substrate and encapsulated by another polyimide superstrate enabling the flexibility, bendability, and portability of both the device and the package. The novel low-cost and low-temperature encapsulation procedure requires a fewer fabrication steps than the processes, where a thin epitaxial Si layer or a full Si wafer with cavity is used as a superstrate. It also provides a better pliability. The fabricated micro accelerometers demonstrated excellent integrity when bowed down to 2-cm radius of curvature with the sensitivity varying between 139 and 174 fF/g, depending on the design, under the application of ±4 g of acceleration. The resonant frequencies varied from 800 Hz for the largest device to 1050 Hz for the smallest device. The results reported here represent the best performing, wafer-level-packaged flexible micro accelerometers to date.

Investigating random telegraph signals (RTS) observed in MOS devices is important for studying the gate-oxide defect characteristics and developing simulation and modeling tools in highly scaled devices. In this paper, we are presenting a comprehensive, variable-temperature, single-to-multitrap scalable RTS model and a simulation tool (RTSSIM) based on the first principles, and supported by experimental data. The physical and electrical characteristics of the actual oxide defects are considered, such as trapping barrier energy, capture cross section, trap-binding enthalpy, entropy change with emission, Coulombic scattering effect, and the trap location. Experimentally obtained time-domain switching data and RTS traces reconstructed by the developed simulation tool are compared for assessing the accuracy of the modeling. In this paper, we identified the trap species responsible for the RTS as an unrelaxed neutral oxygen deficiency center ( $V^{0}$ ODC II) with measured and theoretically verified relaxation energy of 1.11, 1.25, and 0.60 eV. RTSSIM is a tool used to mimic or reproduce a given noise measurement on a particular device. The simulation tool shows over 92% accuracy in replicating the RTS and trap characteristics. The model also represents the first realistic simulation of multilevel RTS.

The fabrication and characterization of micromachined piezoresistive absolute pressure sensors in a flexible substrate is presented. A suspended aluminum oxide diaphragm containing nichrome (Ni-80%/Cr-20%) piezoresistive sensors backed by a vacuum cavity was utilized to form the sensor. The piezoresistors were placed in a half-Wheatstone bridge geometry to provide a linear response and thermal stability. The average value of the gauge factor of nichrome was measured to be 1.95. The average normalized Hooge coefficient $\text{K}_{1/f}$ was found to be $4.64\times 10^{-11}$ for the nichrome piezoresistors. The pressure sensors displayed an average sensitivity of 1.25 nV/Pa and average value of noise equivalent pressure (NEPr) of 7.44 kPa in the bandwidth of 1–10 Hz in the 1/ $f$ -noise limited regime. In the Johnson noise-limited regime, the NEPr was found to be 10 Pa in a 1-Hz bandwidth. [2014-0028]

Flexible, bendable tactile pressure sensors are developed with crystalline ZnO nanorods as a robotic skin or as part of a multimodal sensing garment. The sensors are arranged in addressable arrays for detection of spatial tactile pressure variation with a resolution of 1 mm or better. The sensors are fully conformable to nonplanar surfaces. A low-temperature hydrothermal process in conjunction with thermal nanoimprint lithography is demonstrated to achieve ordered growth of ZnO nanorods placed vertically on the flexible substrate. The nanorods are fully packaged in inert polyimide for protection against dust, moisture, and mechanical shock. The sensor response to applied pressure is observed to be nearly linear and with negligible hysteresis. A peak sensitivity of 21.2 μV/kPa has been observed for tactile pressure in the 10-200 kPa range.

MEMS force sensors embedded in flexible polyimide substrates are reported, motivated by the need to monitor force and pressure on nonplanar surfaces for structural health monitoring. Details of the fabrication, measured preliminary results, and figures of merit are described. The sensors showed an average piezoresistive gauge factor of 1.75. The average value of normalized Hooge noise coefficient K1/f was found to be 1.89×10-10. The sensitivity was measured to the range from 0.266 to 2.248 V/N with an average sensitivity of 1.25 V/N. The noise equivalent force was calculated to the range from 8.83 to 46.56 μN with an average value of 20.47 μN for the force sensors in a noise frequency bandwidth from 1 to 8 Hz.

Piezoresistive NiCr tactile sensors have been developed on surface micromachined aluminum oxide membranes, embedded between two polyimide layers, i.e., one serving as a substrate, another as a superstrate. A novel method to bond a flexible superstrate polyimide layer onto a microelectromechanical system tactile sensor array is presented. The piezoresistors were connected in a half-Wheatstone bridge configuration to minimize the effects of thermal drift. Three different types of sensor designs were fabricated and characterized to obtain the nichrome thin-film gauge factor. The experimental results were compared with those simulated for the same conditions of membrane deflection. The gauge factors range between 2.2 and 7.9 for sensors with a superstrate and between 1.5 and 3.2 without a superstrate.

A physics-based model has been implemented to describe the low-frequency noise behavior in differently processed reduced-surface-field lateral double-diffused MOS devices. The developed model is based upon the correlated carrier number and the mobility fluctuation theory known as the unified model but has been modified to account for the fluctuations in the extended drain and the channel. Unlike the unified 1/f noise model, nonuniform trap distribution has been taken into account with respect to position in the gate oxide and band-gap energy. The effect of stress on dc and noise characteristics has been investigated. Individual resistance and noise components in the channel and in the extended drain regions under the gate and field oxides are evaluated as a function of stress duration. The model is experimentally verified to identify the physical mechanisms for degradation due to stressing.

This paper presents the fabrication and measured performance of temperature sensors embedded in flexible polyimide substrates. The sensing material used for the temperature sensor was undoped amorphous silicon which was fabricated on a 35-m-thick layer of polyimide that serves as the flexible substrate. Another flexible polyimide superstrate layer of 35-40 m was spin-coated on top of the sensors to make sure the temperature sensors lie on a zero stress plane. The temperature coefficient of resistance was measured over the temperature range of 200- 360 K. The maximum temperature coefficient of resistance at 30 C was measured to be 0.0288 K-1. The effect of the flicker noise and voltage dependence of the voltage noise power spectral density exhibited by the sensor was evaluated. The normalized flicker noise coefficient was found to be 1.2 × 10-11.

Capacitive accelerometers were designed, fabricated and characterized on Si and flexible polyimide substrates with sensitivities ranging between 8.9-23 fF/g and 2.5-9.2 fF/g for z-axis and lateral axes, respectively. UV-LIGA process in conjunction with Ni electroplating was utilized to form the proof mass. Measured sensitivities on the flexible accelerometers were higher than the predictions of the simulations, ranging between 22-27 fF/g in the z axis and 12-18 fF/g in the lateral axes, respectively.

Nanocrystalline piezoresistive polysilicon films were obtained at low temperatures by aluminum-induced crystallization (AIC). The films exhibited granular structure with good polycrystalline properties. A piezoresistive pressure sensor was fabricated on a polyimide substrate, in a Wheatstone bridge configuration comprising two passive resistors and two active piezoresistors made of polysilicon films obtained by AIC. The resistors showed linear I-V characteristics with typical resistance values between 15 and 30 kΩ . Atomic force microscopy was used in contact mode to study the response of the pressure sensor with applied pressure in the 2-19 kPa range. For the higher range of 450 kPa-2 MPa, a load-cell with a nanopositioner was utilized. The pressure-sensor sensitivity was measured to be 41.12 and 5.02 mV/MPa, respectively, for these ranges, when the Wheatstone bridge was bias at 1 V.

For specific RF applications, where the use of MEMS is highly attractive, cost-effective reliable packaging is one of the primary barriers to commercialization. Many RF MEMS devices require a hermetic or vacuum operation environment. This paper presents a post-CMOS-compatible method for vacuum packaging of RF MEMS devices by growing an encapsulation layer during the device fabrication. The resulting MEMS devices are surrounded by a vacuum cavity and can then be placed in a conventional low-cost circuit package. This is a low-temperature area-efficient device-level encapsulation for MEMS devices. RF MEMS resonators in a fixed-fixed configuration were used as the test bed since their quality factor can be used as a measure of the package quality. The encapsulation process is based on a double-sacrificial-layer surface micromachining technique, which is used to create a cavity under and above the resonator. Polyimide was used as the sacrificial layer, followed by the deposition of a packaging layer with trench cuts, which facilitate the sacrificial layer removal. The trench cuts were then sealed at a low pressure, forming a cavity around the device at the sealant layer deposition pressure. Extensive RF characterization and reliability tests were performed on the packaged resonators.

Understanding and minimization of low-frequency noise (LFN) originating from high- k (HK) gate dielectrics in new generation MOSFETs are of critical importance to applications in RF, analog, and digital circuits. To understand the effect of stress conditions on noise, nMOSFETs were subjected to accelerated hot-carrier stress (HCS) and positive constant-voltage stress (CVS). The additional LFN introduced through stressing was evaluated on nMOSFETs with TiN metal gate and HfSiON gate dielectric. Nitridation of HfSiO gate-dielectric MOSFETs was achieved by either a high-temperature NH3 anneal or a lower temperature plasma anneal. Influence of different dielectric nitridation procedures on the stress-induced degradation of transconductance, threshold properties, and LFN was studied. Worst degradation conditions, i.e., Vg = Vd, were used for HCS, whereas for CVS, the vertical field was fixed at 10 MV/cm for all transistors to achieve comparable stressing conditions. Plasma-nitrided devices showed less increase in their noise in the linear operation region than the thermally nitrided devices. This difference in noise behavior is attributed to the nitrogen profile across the HK/Si interface and in the bulk of the HK oxide caused by different nitridation techniques. The dielectric defect profile resultant from different annealing techniques was consistent with the spectral form of the observed drain-voltage LFN.

This paper reports on the design, fabrication, and characterization of device-level vacuum-packaged microbolometers on rigid Si wafers and flexible polyimide substrates. Semiconducting yttrium barium copper oxide (commonly, referred to as YBCO) serves as the bolometric material. Operating micromachined bolometers in vacuum reduces the thermal conductance G(th) from the detector to the substrate. If flexibility of the substrate is not to be sacrificed, then the vacuum packaging needs to be done at the device level. Here, the microbolometers are fabricated on a silicon nitride support membrane, isolated from the substrate using surface micromachining. Suitable materials as well as various dimensions in the vacuum cavity are determined using finite-element method (FEM)-based CoventorWARE. A vacuum cavity made of Al2O3 has been designed. The thermal conductance G(th) of bolometers with the geometry implemented in this work is the same for devices on rigid and flexible substrates. The theoretical value of G(th) was calculated to be 4.0 x 10(-6) W/K for devices operating in vacuum and 1.4 x 10(-4) W/K for devices operating at atmospheric pressure. Device-level vacuum-packaged microbolometers on both rigid Si and flexible polyimide substrates have been fabricated and characterized for optical and electrical properties. A low thermal conductance of 1.1 x 10(-6) W/K has been measured six months after fabrication, which implies an intact vacuum cavity.

Low-frequency noise measurements and analysis were performed on n-channel MOSFETs with HfSiON as the gate-dielectric material. The role of SiON interfacial-layer thickness was investigated. It was observed that these fluctuations can be described by the unified flicker-noise model that attributes noise to correlated carrier-number/mobility fluctuations due to trapping states in the gate dielectric. The model was modified to include the effect of different gate stack layers on the observed noise. The carrier-number fluctuations were found to dominate over the correlated mobility fluctuations in the measured bias range and more so at the lower gate overdrives. The noise magnitude showed a decrease with increasing SiON interfacial-layer thickness. Furthermore, an inverse-proportionality relationship was revealed between the effective oxide trap density and SiON thickness.

Low-frequency noise characteristics are reported for TaSiN-gated n-channel MOSFETs with atomic-layer deposited HfO/sub 2/ on thermal SiO/sub 2/ with stress-relieved preoxide (SRPO) pretreatment. For comparison, control devices were also included with chemical SiO/sub 2/ resulting from standard Radio Corporation of America clean process. The normalized noise spectral density values for these devices are found to be lower when compared to reference poly Si gate stack with similar HfO/sub 2/ dielectric. Consequently, a lower oxide trap density of /spl sim/4/spl times/10/sup 17/ cm/sup -3/eV/sup -1/ is extracted compared to over 3/spl times/10/sup 18/ cm/sup -3/eV/sup -1/ values reported for poly Si devices indicating an improvement in the high-/spl kappa/ and interfacial layer quality. In fact, this represents the lowest trap density values reported to date on HfO/sub 2/ MOSFETs. The peak electron mobility measured on the SRPO devices is over 330 cm/sup 2//V/spl middot/s, much higher than those for equivalent poly Si or metal gate stacks. In addition, the devices with SRPO SiO/sub 2/ are found to exhibit at least /spl sim/10% higher effective mobility than RCA devices, notwithstanding the differences in the high-/spl kappa/ and interfacial layer thicknesses. The lower Coulomb scattering coefficient obtained from the noise data for the SRPO devices imply that channel carriers are better screened due to the presence of SRPO SiO/sub 2/, which, in part, contributes to the mobility improvement.

This special issue focuses on the sensors and sensing systems having features like flexibility, bendability, conformability, stretchability, and presence over large areas and 3D surfaces. Like conventional sensors, flexible sensors are also used to measure parameters and agents like pH, temperature, humidity, force, gasses, velocity, flow etc. In addition, they possess the ability to bend. This requires special front-end processing techniques compatible with the flexible materials that these sensors are made of. In response to the call for papers for this special issue, 114 manuscripts were received. Of those, 39 were found not to fit to the scope of the special issue and were separated to be handled by the permanent editors of the Journal. The peer review process resulted in 29 innovative manuscripts to be published here submitted from over 45 institutions, with a few trailing manuscripts to be published at a later issue due to time constraints. Of the work presented here, nine papers describe sensors for health and medical applications. Five papers detail sensors for the general fields of environmental and structural health monitoring. Four are on robotic applications. Although majority of the manuscripts focus on a single sensing function such as fluidic flow, force, pH, temperature or presence of a specific gas, four papers present multifunctional flexible sensors.

The role of the interfacial oxide (IFO) between the polysilicon and monosilicon emitter regions on the noise behavior of n-p-n poly-emitter bipolar transistors was investigated through 1/f noise measurements. Bipolar junction transistors with different IFO thickness, and emitter geometry were utilized. Measurements with variable external base bias resistance (R/sub S/) were used to investigate the relative contribution of each individual noise source from the base current (S/sub IB/), the collector current (S/sub IC/) and, the internal emitter and base series resistances (S/sub Vr/). When the voltage noise power spectral densities S/sub VC/ and S/sub VB/ were measured across resistances in series with the collector and base, respectively, using a relatively large R/sub S/ (/spl sim/1 M/spl Omega/), S/sub IB/ was found to have the dominant noise contribution at lower bias currents. On the other hand, when the voltage noise power spectral densities S/sub VC/ and S/sub VE/ were measured across resistances in series with the collector and emitter, respectively, in a different experimental setup with a low R/sub S/ value, S/sub Vr/ was found to have the dominant noise contribution at higher bias currents. IFO was found to increase S/sub IB/, S/sub IC/, and S/sub Vr/. S/sub IB/ was modeled as a combination of tunneling and diffusion fluctuations of the minority carriers in the emitter; whereas S/sub IC/ was modeled as a combination of number and diffusion fluctuations of the minority carriers in the base. S/sub Vr/ was attributed to the internal emitter resistance noise originating from the fluctuation in the majority carrier flow through the IFO.

Low-frequency noise measurements were performed on p- and n-channel MOSFETs with HfO/sub 2/, HfAlO/sub x/ and HfO/sub 2//Al/sub 2/O/sub 3/ as the gate dielectric materials. The gate length varied from 0.135 to 0.36 /spl mu/m with 10.02 /spl mu/m gate width. The equivalent oxide thicknesses were: HfO/sub 2/ 23 /spl Aring/, HfAlO/sub x/ 28.5 /spl Aring/ and HfO/sub 2//Al/sub 2/O/sub 3/ 33 /spl Aring/. In addition to the core structures with only about 10 /spl Aring/ of oxide between the high-/spl kappa/ dielectric and silicon substrate, there were "double-gate oxide" structures where an interfacial oxide layer of 40 /spl Aring/ was grown between the high-/spl kappa/ dielectric and Si. DC analysis showed low gate leakage currents in the order of 10/sup -12/A(2-5/spl times/10/sup -5/ A/cm/sup 2/) for the devices and, in general, yielded higher threshold voltages and lower mobility values when compared to the corresponding SiO/sub 2/ devices. The unified number-mobility fluctuation model was used to account for the observed 1/f noise and to extract the oxide trap density, which ranged from 1.8/spl times/10/sup 17/ cm/sup -3/eV/sup -1/ to 1.3/spl times/10/sup 19/ cm/sup -3/eV/sup -1/, somewhat higher compared to conventional SiO/sub 2/ MOSFETs with the similar device dimensions. There was no evidence of single electron switching events or random telegraph signals. The aim of this paper is to present a general discussion on low-frequency noise characteristics of the three different high-/spl kappa//gate stacks, relative comparison among them and to the Si--SiO/sub 2/ system.

Uncooled semiconducting YBaCuO infrared microbolometers have been fabricated on a flexible polyimide substrate formed by spin-coating a silicon wafer with a release layer. The wafer was used as a carrier for the flexible substrate during fabrication. The finished microbolometers on the flexible substrate showed a temperature coefficient of resistance (TCR) TCR =(1/R)(dR/dT) of -3.03% K/sup -1/, at room temperature, which is comparable to the TCR values observed for semiconducting YBaCuO microbolometers fabricated directly on Si. In order to provide protection and better mechanical integrity, some of the devices were encapsulated. The microbolometers attained a responsivity and detectivity as high as 3.5/spl times/10/sup 3/ V/W and 1/spl times/10/sup 7/ cm/spl middot/Hz/sup 1/2//W, respectively, at 2.88 /spl mu/A of current bias. The responsivity and detectivity of the encapsulated microbolometers, on the other hand, were 1.6/spl times/10/sup 3/ V/W and 4.9/spl times/10/sup 6/ cm/spl middot/Hz/sup 1/2//W, respectively at 1 /spl mu/A of current bias. Spin-coated liquid polyimide solved two major problems previously encountered with the solid polyimide sheets when used as a flexible substrate. First, flatness of the flexible substrate was maintained with no air bubbles. Second, the thermal expansion of the flexible substrate during the fabrication process due to thermal cycling was minimal. All measurements reported in this paper, were taken prior to releasing the flexible substrate from the Si wafer containing the finished microbolometers.

Low-frequency noise in double reduced-surface-field lateral-double-diffused-MOS devices has been measured, and the effect of dc stressing is analyzed. The noise components contributing from the extended drain regions under the gate and field oxides were differentiated from the channel noise by a series of experimental and analytical techniques. The effect of voltage stressing on each noise component was investigated. Trapped-charge carrier fluctuations due to Si/SiO2 interface traps in the overlap region in the extended drain as well as in the channel were found to be the dominant source of noise. The bulk resistance fluctuations in the extended drain region under the field oxide were found to be insignificant. High-voltage stressing caused an increase in the interface traps, thus increasing both the extended drain overlap resistance and the noise.

1/f noise was measured on polysilicon-emitter bipolar n-p-n and p-n-p transistors over a temperature range of 173K

A new generation of microbolometers were designed, fabricated and tested for the NASA CERES (Clouds and the Earth's Radiant Energy System) instrument to measure the radiation flux at the Earth's surface and the radiant energy now within the atmosphere. These detectors are designed to measure the earth radiances in three spectral channels consisting of a short wave channel of 0.3 to 5 /spl mu/m, a wide-band channel of 0.3 to 100 /spl mu/m and a window channel from 8 to 12 /spl mu/m each housing a 1.5 mm x 1.5 mm microbolometers or alternatively 400 /spl mu/m x 400 mm microbolometers in a 1 /spl times/ 4 array of detectors in each of the three wavelength bands, thus yielding a total of 12 channels. The microbolometers were fabricated by radio frequency (RF) magnetron sputtering at ambient temperature, using polyimide sacrificial layers and standard micromachining techniques. A semiconducting YBaCuO thermometer was employed. A double micromirror structure with multiple resonance cavities was designed to achieve a relatively uniform absorption from 0.3 to 100 /spl mu/m wavelength. Surface micromachining techniques in conjunction with a polyimide sacrificial layer were utilized to create a gap underneath the detector and the Si/sub 3/N/sub 4/ bridge layer. The temperature coefficient of resistance was measured to be -2.8%/K. The voltage responsivities were over 10/sup 3/ V/W, detectivities above 10/sup 8/ cm Hz/sup 1/2//W, noise equivalent power less than 4 /spl times/ 10/sup -10/W/Hz/sup 1/2/ and thermal time constant less than 15 ms.

Microbolometer infrared detectors based on semiconducting yttrium-barium-copper-oxide (YBCO) have been fabricated on a 125-/spl mu/m-thick flexible polyimide substrate, Kapton(R). During the fabrication process, the highest temperature reached was 270/spl deg/C, used to cure a polyimide insulation layer. The 60 /spl mu/m/spl times/60 /spl mu/m microbolometers displayed a temperature coefficient of resistance of -3.07%/K at room temperature. The thermal conductivity of the structure was measured to be 4.18/spl times/10/sup -5/ W/K. The bolometer response showed a linear dependence on bias current with a responsivity of 2/spl times/10/sup 3/ V/W and detectivity of 3/spl times/10/sup 7/ cm Hz/sup 1/2//W at 1.13 Hz and 1 /spl mu/A bias.

A new micromachined microbolometer array structure is presented that utilizes a self-supporting semiconducting yttrium barium copper oxide (Y-Ba-Cu-O) thin film thermometer. The Y-Ba-Cu-O thermometer is held above the substrate only by the electrode arms without the need of any underlying supporting membrane. This represents a significant improvement in the state-of-the-art for microbolometers by eliminating the thermal mass associated with the supporting membrane. The reduced thermal mass permits lowering the thermal conductance to the substrate to obtain increased responsivity or having a shorter thermal time constant to allow for higher frame rate camera. The simple structure does not suffer from warping problems associated with stress imbalances in multilayer microbolometer structures that utilize a supporting membrane such as Si/sub 3/N/sub 4/. Devices were fabricated by growing Y-Ba-Cu-O films on a conventional polyimide sacrificial layer mesa. Subsequent etching of the sacrificial layer provides the air gap that thermally isolates the microbolometer. Y-Ba-Cu-O possesses a relatively high temperature coefficient of resistance of 3.1%/K at room temperature. The 400-nm-thick Y-Ba-Cu-O film exhibited absorptivity of about 30%. The responsivity and detectivity approached 10/sup 4/ V/W and 10/sup 8/ cm Hz/sup 1/2//W to filtered blackbody infrared (IR) radiation covering the 2.5 to 13.5 /spl mu/m band. This extrapolates to noise equivalent temperature difference (NETD) less than 100 mK. The micromachining techniques employed are post-complementary metal-oxide-semiconductor (CMOS) compatible, allowing for the fabrication of focal plane arrays for IR cameras.

This paper describes a new micromachining technique for fabrication of semiconducting yttrium barium copper oxide (YBCO) microbolometers using magnesium oxide (MgO) as the sacrificial layer. This type of bolometer can be operated at room temperature, providing a low-cost alternative for more expensive cryogenically cooled thermal detectors used for infrared (IR) imaging. The new micromachining techniques described here would enable the fabrication of YBCO IR focal plane array (FPA) with CMOS signal processing circuitry. Devices were fabricated by growing YBCO films on 4000-/spl Aring/-thick suspended Si/sub 3/N/sub 4/ membranes measuring 40/spl times/40 /spl mu/m/sup 2/ in area and extended over micromachined air gaps, which provide the low thermal conductivity that is required for high responsivity. The gap was created by etching an MgO sacrificial layer. This is the first example of using MgO in this type of application. The MgO sacrificial layer technique is fully CMOS compatible and presents no major fabrication challenges. Thermal conductivities achieved in vacuum with the Si/sub 3/N/sub 4/ suspended structures were on the order of 10/sup -7/ W/K, yielding an effective thermal isolation for bolometer operation. Optical characterization has shown responsivity up to 60 kV/W and detectivity over 10/sup 8/ cm.Hz/sup 1/2//W to black-body IR radiation, indicating that this technology is a suitable candidate for low-cost thermal imaging.

The characteristics of infrared (IR) microbolometer arrays utilizing semiconducting YBaCuO and operating at room temperature are presented. Surface-micromachined structures in the form of 1/spl times/10 arrays of pixel size 40 /spl mu/m/spl times/40 /spl mu/m and 60 /spl mu/m/spl times/60 /spl mu/m as well as single pixels of various geometries were constructed. Using the chopped radiation from a broad-band IR source, the responsivity R/sub V/ of the sensor was measured to be as high as 10/sup 4/ V/W and detectivity D* to be /spl sim/2/spl times/10/sup 7/ cm Hz/sup 1/2//W for a thermal conductance G/spl sim/10/sup -5/ W/K between the detector and the substrate. The spectral response was found to be uniform over a range of 1-12 /spl mu/m. Silicon micromachining and ambient-temperature processing were employed to ensure compatibility and, therefore, potential integration with CMOS-based signal processing circuitry. Methods of enhancing the figures of merit are discussed.

Thermal infrared detectors require thermal isolation to permit the infrared-sensitive material to integrate the incident photon energy and thereby obtain high responsivity and detectivity. This paper describes the fabrication of semiconducting YBaCuO microbolometer arrays into thermal isolation structures by employing Si surface-micromachining techniques. An isotropic HF:HNO/sub 3/ etch was used to remove the underlying Si substrate from the front-side of the wafer and suspend SiO/sub 2/ membranes into 1/spl times/10 pixel-array structures. The infrared-sensitive material, YBaCuO, was subsequently deposited onto the thermal isolation structures and patterned to form the detector arrays. The high-temperature coefficient of resistance and low noise of semiconducting YBaCuO at room temperature is attractive for uncooled infrared detection. The fabrication process was conducted entirely at room temperature. In this manner, infrared detectors are fabricated in a process that is compatible with CMOS technology to allow for the integration with on-chip signal processing circuitry. The end result is low-cost infrared-detector arrays for night vision in a variety of applications including transportation and security. Preliminary results show a temperature coefficient of resistance above 3%, voltage responsivity close to 10/sup 4/ V/W, and detectivity over 10/sup 7/ cm/spl middot/Hz/sup 1/2//W at a bias current of 0.79 /spl mu/A.

We report 1/f noise and Hall effect measurements performed on c-axis oriented YBa/sub 2/Cu/sub 3/O/sub 7-/spl delta// thin films at superconducting transition and in normal state. Magnetic fields up to 6 kG were applied parallel to c-axis. The voltage noise spectral density was normalized using Hall carrier density as well as atomic concentration. Close to the superconductive state, normalized voltage noise spectral density had a sharp rise. The noise did not seem to originate from local thermal fluctuations. We interpreted the data in terms of the classic percolation model near superconductive transition, according to which normalized noise is proportional to R/sup -k'/s/, where R is the macroscopic sample resistance and k' and s are critical exponents. The ratio k'/s, which can be regarded as an index of the rise of normalized noise, steadily increased with higher magnetic field. The zero-Gauss value of k'/s was calculated as 1.06; a result that agrees with the prediction of the classical percolation theory for a 2-D network. >

We measured 1/f noise on Hg/sub 0.71/Cd/sub 0.29/Te Metal-Insulator-Semiconductor (MIS) infrared detectors operated over the temperature range of 40 K to 90 K under 300 K Infrared (IR) radiation. The purpose of the study was to identify the sources of 1/f noise, especially in relation to the dark current. The devices were operated in the correlated double sampling mode where the voltage across the MIS capacitor was sampled at empty potential well and right after the accumulation of minority carriers in the well due to IR radiation generation. The noise power spectral density for the charge integrated in the MIS well was investigated in relation to the dominant component of dark current. At lower temperatures T/spl les/65 K, the charge noise power spectral density was found to depend quadratically on the dark current. At higher temperatures, this quadratic dependence did not exist. We attribute the dark current to a mixture of tunneling and depletion-region-originated minority carrier generation which seems to be responsible for 1/f fluctuations in these structures for temperatures below 65 K. >

Microwave mixing and parametric conversion in YBa/sub 2/Cu/sub 3/O/sub 7/ microbridges were investigated. The devices have been operated as down-converters, converting a 20-GHz radio frequency to a 1.5-GHz intermediate frequency under the action of an 18.5-GHz local oscillator. The microbridges were fabricated both as granular films on MgO substrates possessing grain boundary weak links and as epitaxial films deposited over substrate steps in LaAlO/sub 3/ substrates. The authors have measured conversion gains as high as -8+or-4 dB when mixing on the nonlinear resistive (superconducting-normal) transition and as high as -30 to -40 dB when parametrically converting via the nonlinear inductance of the bridge at bias currents less than the critical current. Output referred noise powers as low as -206 dB/Hz (3*10/sup -21/ W/Hz) have been measured. In some samples, noise switching events were observed. >

Random telegraph signals (RTS) have been measured in the drain to source voltage of W/spl times/L=0.97/spl times/0.15 /spl mu/m/sup 2/ medium-doped drain (MDD) n-MOSFET's. The depth of the trapping center in the oxide is found from the gate voltage dependence of the emission and capture times. The difference in the drain voltage dependence of the capture and emission times between the forward and reverse modes is utilized to find the position of the trap in the channel with respect to the source.

The concentration of silicon-silicon dioxide interface traps with energies down to within 20 meV of the majority-carrier band edge was determined using 1/f noise measurements. p-type diffused resistors of four-probe geometry were fabricated in a metal-oxide-silicon structure. Flicker noise measurements were performed on these devices at cryogenic temperatures of 20 to 280 K. Using A. L. McWhorter's (1957) 1/f noise model and the calculated position of the Fermi level with respect to the valence band edge at each temperature, the density of interface traps was calculated at energy levels corresponding to the position of the Fermi level at that temperature. This technique is proposed as an alternate method to measure the oxide trap and slow interface-state densities with energies close to the band edges. >