Molecular beam epitaxy, because of its low growth rate and control (i.e., modulation) over numero... more Molecular beam epitaxy, because of its low growth rate and control (i.e., modulation) over numerous molecular beams, provides tremendous accuracy and flexibility in growing the most complex structures. Lamella doping is used to tailor doping profiles within an active region to nearly any concentration gradient and profile required. The technique doping requires the momentary suspension of growth, during which a predetermined amount of dopant (lamella) is deposited on the surface of the grown layer. Growth is then resumed and the process repeated with subsequent lamella concentrations altered to provide correct doping profiles. Another potentially important technique involves the simultaneous deposition of single crystal and polycrystalline GaAs. This planar process of selective molecular beam epitaxy was used to define and isolate single crystal GaAs regions.
... Modulated S-Parameter Measurements for Isothermal Microwave Device Characterization George Me... more ... Modulated S-Parameter Measurements for Isothermal Microwave Device Characterization George Metze, Senior Member, IEEE, Mark Calcatera, Member, IEEE, Chris Eppers, Bob Neidhard, and James Whalen, Senior Member, IEEE 531 ...
An RF energy scavenging circuit implementing a power matched Villard voltage doubler followed by ... more An RF energy scavenging circuit implementing a power matched Villard voltage doubler followed by a switched capacitor DC-DC converter for scavenging ultra-low RF power levels (,220 dBm) is presented. Measurement results for the circuit, fabricated in a 130 nm CMOS process, show that 1V can be generated across a 5 MV load from as little as 225.5 dBm of input RF energy at 2.2 GHz. This represents a 9.5 dB improvement, over the measured sensitivity of our RF energy scavenging circuit without the use of a switched capacitor DC-DC converter stage.
This paper treats the problem of automatic fault diaging several hundred active components. While... more This paper treats the problem of automatic fault diaging several hundred active components. While being nosis for systems with multiple faults. The system is decomposed undecomposable, these units would have the computainto n units ul, u2, , u,IU where a unit is a well-identifiable portion of the system which cannot be further decomposed for the purpose ofI diagnosis. By means of a given arrangement of testing links (connec-
Recent developments in mm-wave MMICs foundry and the complex MMICs that can be produced using new... more Recent developments in mm-wave MMICs foundry and the complex MMICs that can be produced using new processes are discussed. Attention is given to the design of mm-wave HEMT MMICs and to material and processing considerations. Some of the trade-offs that need to be made in order to realize a multifunctional MMIC are discussed, showing that it is possible to overcome
Semiconductor Device Research International Symposium, 2001
Asymmetrically doped CMOS inverters have been designed. A mixed mode device/circuit simulator has... more Asymmetrically doped CMOS inverters have been designed. A mixed mode device/circuit simulator has been developed to analyze the new designs. Results show that the asymmetrically doped devices can give rise to inverter switching speeds which are several times faster than those composed of conventional CMOS. Such an increase should translate directly to similar increases in computer processor speeds. The asymmetrical
Accurate thermal modeling is critical to address the heating-related problems in integrated circu... more Accurate thermal modeling is critical to address the heating-related problems in integrated circuits, as well as to ensure as-designed operation or to obtain performance predictions in a range of different operating conditions. We present our experimental and theoretical work on modeling thermal behavior in integrated circuits (ICs) at the resolution of single material layers. Through measurements and simulations we find
We report on a novel method for predicting the temperature profile of complex integrated circuits... more We report on a novel method for predicting the temperature profile of complex integrated circuits at the resolution of a single device. The proposed new modeling method establishes the necessary link between full-chip heating and non-isothermal device operation for resolving effects of the individual devices on the overall full-chip heating. The technique accounts for the application specific activity levels and the layout replacements of individual devices. We use a lumped full-chip heating model that has thermal resistances and capacitances determined by the layout design, and heat sources that are set according to the operation statistics of devices on the chip. To embed the effects of operation statistics for a given application, we use a Monte Carlo type methodology. We analyzed a Pentium III [http://www.intel.com] chip considering a realistic layout geometry and averaged activity statistics. Our analysis shows 43 and 33 K increases above the ambient for the peak and median temperatures, respectively.
The DC and transient operations of asymmetrically doped CMOS inverters are investigated. The asym... more The DC and transient operations of asymmetrically doped CMOS inverters are investigated. The asymmetrical doping consists of a halo around either the source or the drain. We investigate the DC and transient characteristics of devices and CMOS inverters that contain these halos. To facilitate the investigation, a new algorithm for mixed mode device/circuit simulation is developed. We find that the utilization of the source halo provides significant improvement in the DC and switching characteristics of both devices and CMOS inverters. The drain halo improves performance but gives rise to increased source-drain leakage currents. The effects of narrowing the halo doping is also investigated.
A method for predicting full chip temperature heating resulting from device operation is presente... more A method for predicting full chip temperature heating resulting from device operation is presented. The method couples distributed device simulation with lumped thermal analysis. Predictions show sixty degree Kelvin temperature increases for 0.5 cm IC's. A method for reducing chip temperature is also presented.
2011 International Semiconductor Device Research Symposium (ISDRS), 2011
RF energy harvesting has been widely investigated as a promising method of providing power to pas... more RF energy harvesting has been widely investigated as a promising method of providing power to passive wireless devices or recharging the battery used in such systems. The AC input signal which is the ambient RF radiation from the available ISM band is received through the coupled antenna and converted into DC output voltage through the energy harvester chip. The energy harvester, which is a novel resonant boost rectifier circuit, typically consists of multiple stages of voltage doubler stacked in series to achieve a desired output voltage level. In order to achieve maximum rectifier conversion efficiency, it is imperative to have impedance matching between the receiving antenna and the harvester circuit. The input of the harvester is frequently modeled as a resistance in parallel with a capacitance [1]. Therefore, a physical model-based derivation of the equivalent input resistance and capacitance of the rectifier is useful to design an appropriate matching network to resonate out the input capacitance and to match the input resistance to the radiation resistance of the antenna. This paper presents an analytical approach to derive the equivalent input impedance of a multistage voltage doubler circuit designed using short-channel diode-connected MOSFETs.
2011 International Semiconductor Device Research Symposium (ISDRS), 2011
Past work on radio frequency (RF) energy harvester design mostly focused on harvesting RF energy ... more Past work on radio frequency (RF) energy harvester design mostly focused on harvesting RF energy in a single RF band [1, 2,]. While a significant portion of RF energy is concentrated in the communication band around 900 MHz, there is also abundant RF energy present in the band covering 1800 MHz to 2100 MHz. For example. Li [2011] demonstrated the
2011 International Semiconductor Device Research Symposium (ISDRS), 2011
Radio Frequency (RF) energy harvesting aims at collecting and converting ambient RF wave energy i... more Radio Frequency (RF) energy harvesting aims at collecting and converting ambient RF wave energy into storable electrical energy to power electronics. The reported RF energy density in urban areas can be as high as 0.5μW/cm2 which corresponds to an input power level of 16.6 μW (−17.6 dBm) at 1800MHz [1]. It is thus appealing to convert the RF energy in
2006 International Conference on Simulation of Semiconductor Processes and Devices, 2006
We present a methodology for investigating the response of a complex on-chip interconnect network... more We present a methodology for investigating the response of a complex on-chip interconnect network to external and internal noise, through the development of a 3-D solver based on creating a lumped element model of an interconnect network and solving for its impulse responses. Our method exhibits a lower computational cost than SPICE and allows the user the flexibility to work
International Semiconductor Device Research Symposium, 2003, 2003
The numerical modeling of on-chip inductor structures, applied to the comparison of inductor geom... more The numerical modeling of on-chip inductor structures, applied to the comparison of inductor geometries was studied. The frequency-dependence of inductor characteristics depends on skin effect in the conductors and induced currents in the semiconductor substrates.
ABSTRACT Initial studies on the fabrication of a new double-layer capacitor based on hydrated rut... more ABSTRACT Initial studies on the fabrication of a new double-layer capacitor based on hydrated ruthenium oxide (RuO2.xH2O) are presented. The electrochemical capacitor cell is based on three designs approaches: active surface enhancement via corrugations, the use of an ultra-thin self assembled dielectric implemented with hydrated Ruthenium oxide and activated carbon powders in combination with sulfuric acid and the fusion of both processes. We address the key components of the electrochemical cell (i) packaging,; (ii) the methodology of incorporating sulfuric acid and the various powders, and (iii) the deposition technique for the powder. The powder composition was varied to establish its effect on the overall capacitance. Mixing RuO2.xH2O and AC yielded devices with the greatest capacitance value, whereas devices containing only one of the two components had capacitance values lower by a factor of 10-50. The devices have capacitance values as large as 50 mF/cm2 per plate pair.
Molecular beam epitaxy, because of its low growth rate and control (i.e., modulation) over numero... more Molecular beam epitaxy, because of its low growth rate and control (i.e., modulation) over numerous molecular beams, provides tremendous accuracy and flexibility in growing the most complex structures. Lamella doping is used to tailor doping profiles within an active region to nearly any concentration gradient and profile required. The technique doping requires the momentary suspension of growth, during which a predetermined amount of dopant (lamella) is deposited on the surface of the grown layer. Growth is then resumed and the process repeated with subsequent lamella concentrations altered to provide correct doping profiles. Another potentially important technique involves the simultaneous deposition of single crystal and polycrystalline GaAs. This planar process of selective molecular beam epitaxy was used to define and isolate single crystal GaAs regions.
... Modulated S-Parameter Measurements for Isothermal Microwave Device Characterization George Me... more ... Modulated S-Parameter Measurements for Isothermal Microwave Device Characterization George Metze, Senior Member, IEEE, Mark Calcatera, Member, IEEE, Chris Eppers, Bob Neidhard, and James Whalen, Senior Member, IEEE 531 ...
An RF energy scavenging circuit implementing a power matched Villard voltage doubler followed by ... more An RF energy scavenging circuit implementing a power matched Villard voltage doubler followed by a switched capacitor DC-DC converter for scavenging ultra-low RF power levels (,220 dBm) is presented. Measurement results for the circuit, fabricated in a 130 nm CMOS process, show that 1V can be generated across a 5 MV load from as little as 225.5 dBm of input RF energy at 2.2 GHz. This represents a 9.5 dB improvement, over the measured sensitivity of our RF energy scavenging circuit without the use of a switched capacitor DC-DC converter stage.
This paper treats the problem of automatic fault diaging several hundred active components. While... more This paper treats the problem of automatic fault diaging several hundred active components. While being nosis for systems with multiple faults. The system is decomposed undecomposable, these units would have the computainto n units ul, u2, , u,IU where a unit is a well-identifiable portion of the system which cannot be further decomposed for the purpose ofI diagnosis. By means of a given arrangement of testing links (connec-
Recent developments in mm-wave MMICs foundry and the complex MMICs that can be produced using new... more Recent developments in mm-wave MMICs foundry and the complex MMICs that can be produced using new processes are discussed. Attention is given to the design of mm-wave HEMT MMICs and to material and processing considerations. Some of the trade-offs that need to be made in order to realize a multifunctional MMIC are discussed, showing that it is possible to overcome
Semiconductor Device Research International Symposium, 2001
Asymmetrically doped CMOS inverters have been designed. A mixed mode device/circuit simulator has... more Asymmetrically doped CMOS inverters have been designed. A mixed mode device/circuit simulator has been developed to analyze the new designs. Results show that the asymmetrically doped devices can give rise to inverter switching speeds which are several times faster than those composed of conventional CMOS. Such an increase should translate directly to similar increases in computer processor speeds. The asymmetrical
Accurate thermal modeling is critical to address the heating-related problems in integrated circu... more Accurate thermal modeling is critical to address the heating-related problems in integrated circuits, as well as to ensure as-designed operation or to obtain performance predictions in a range of different operating conditions. We present our experimental and theoretical work on modeling thermal behavior in integrated circuits (ICs) at the resolution of single material layers. Through measurements and simulations we find
We report on a novel method for predicting the temperature profile of complex integrated circuits... more We report on a novel method for predicting the temperature profile of complex integrated circuits at the resolution of a single device. The proposed new modeling method establishes the necessary link between full-chip heating and non-isothermal device operation for resolving effects of the individual devices on the overall full-chip heating. The technique accounts for the application specific activity levels and the layout replacements of individual devices. We use a lumped full-chip heating model that has thermal resistances and capacitances determined by the layout design, and heat sources that are set according to the operation statistics of devices on the chip. To embed the effects of operation statistics for a given application, we use a Monte Carlo type methodology. We analyzed a Pentium III [http://www.intel.com] chip considering a realistic layout geometry and averaged activity statistics. Our analysis shows 43 and 33 K increases above the ambient for the peak and median temperatures, respectively.
The DC and transient operations of asymmetrically doped CMOS inverters are investigated. The asym... more The DC and transient operations of asymmetrically doped CMOS inverters are investigated. The asymmetrical doping consists of a halo around either the source or the drain. We investigate the DC and transient characteristics of devices and CMOS inverters that contain these halos. To facilitate the investigation, a new algorithm for mixed mode device/circuit simulation is developed. We find that the utilization of the source halo provides significant improvement in the DC and switching characteristics of both devices and CMOS inverters. The drain halo improves performance but gives rise to increased source-drain leakage currents. The effects of narrowing the halo doping is also investigated.
A method for predicting full chip temperature heating resulting from device operation is presente... more A method for predicting full chip temperature heating resulting from device operation is presented. The method couples distributed device simulation with lumped thermal analysis. Predictions show sixty degree Kelvin temperature increases for 0.5 cm IC's. A method for reducing chip temperature is also presented.
2011 International Semiconductor Device Research Symposium (ISDRS), 2011
RF energy harvesting has been widely investigated as a promising method of providing power to pas... more RF energy harvesting has been widely investigated as a promising method of providing power to passive wireless devices or recharging the battery used in such systems. The AC input signal which is the ambient RF radiation from the available ISM band is received through the coupled antenna and converted into DC output voltage through the energy harvester chip. The energy harvester, which is a novel resonant boost rectifier circuit, typically consists of multiple stages of voltage doubler stacked in series to achieve a desired output voltage level. In order to achieve maximum rectifier conversion efficiency, it is imperative to have impedance matching between the receiving antenna and the harvester circuit. The input of the harvester is frequently modeled as a resistance in parallel with a capacitance [1]. Therefore, a physical model-based derivation of the equivalent input resistance and capacitance of the rectifier is useful to design an appropriate matching network to resonate out the input capacitance and to match the input resistance to the radiation resistance of the antenna. This paper presents an analytical approach to derive the equivalent input impedance of a multistage voltage doubler circuit designed using short-channel diode-connected MOSFETs.
2011 International Semiconductor Device Research Symposium (ISDRS), 2011
Past work on radio frequency (RF) energy harvester design mostly focused on harvesting RF energy ... more Past work on radio frequency (RF) energy harvester design mostly focused on harvesting RF energy in a single RF band [1, 2,]. While a significant portion of RF energy is concentrated in the communication band around 900 MHz, there is also abundant RF energy present in the band covering 1800 MHz to 2100 MHz. For example. Li [2011] demonstrated the
2011 International Semiconductor Device Research Symposium (ISDRS), 2011
Radio Frequency (RF) energy harvesting aims at collecting and converting ambient RF wave energy i... more Radio Frequency (RF) energy harvesting aims at collecting and converting ambient RF wave energy into storable electrical energy to power electronics. The reported RF energy density in urban areas can be as high as 0.5μW/cm2 which corresponds to an input power level of 16.6 μW (−17.6 dBm) at 1800MHz [1]. It is thus appealing to convert the RF energy in
2006 International Conference on Simulation of Semiconductor Processes and Devices, 2006
We present a methodology for investigating the response of a complex on-chip interconnect network... more We present a methodology for investigating the response of a complex on-chip interconnect network to external and internal noise, through the development of a 3-D solver based on creating a lumped element model of an interconnect network and solving for its impulse responses. Our method exhibits a lower computational cost than SPICE and allows the user the flexibility to work
International Semiconductor Device Research Symposium, 2003, 2003
The numerical modeling of on-chip inductor structures, applied to the comparison of inductor geom... more The numerical modeling of on-chip inductor structures, applied to the comparison of inductor geometries was studied. The frequency-dependence of inductor characteristics depends on skin effect in the conductors and induced currents in the semiconductor substrates.
ABSTRACT Initial studies on the fabrication of a new double-layer capacitor based on hydrated rut... more ABSTRACT Initial studies on the fabrication of a new double-layer capacitor based on hydrated ruthenium oxide (RuO2.xH2O) are presented. The electrochemical capacitor cell is based on three designs approaches: active surface enhancement via corrugations, the use of an ultra-thin self assembled dielectric implemented with hydrated Ruthenium oxide and activated carbon powders in combination with sulfuric acid and the fusion of both processes. We address the key components of the electrochemical cell (i) packaging,; (ii) the methodology of incorporating sulfuric acid and the various powders, and (iii) the deposition technique for the powder. The powder composition was varied to establish its effect on the overall capacitance. Mixing RuO2.xH2O and AC yielded devices with the greatest capacitance value, whereas devices containing only one of the two components had capacitance values lower by a factor of 10-50. The devices have capacitance values as large as 50 mF/cm2 per plate pair.
Uploads
Papers by George Metze