Electrically Variable or Programmable Nonvolatile … 2. The Capacitance of the Electrically...
2
Figure 1. An Electrically Variable Capacitor of the type described in the text can be fabricated on a sil- icon or other substrate as part of an integrated circuit. literal function, that is the fundamental element in multi-valued logic.) Hence, the adjustable-threshold inverter can be made a basic building block of quater- nary logic circuits. The real-time-reconfigurable logic gate can be realized, in a circuit partly resembling the adjustable-threshold in- verter, by applying the logic input signals to JG1 and JG2 and connecting the input terminal of what would otherwise be the inverter to a constant reference voltage (that is, making V in a constant voltage). The number of transistors in this circuit is smaller than in a classical CMOS circuit that performs an equiva- lent logic function. The same hardware can be made to form any of three differ- ent functions: Depending on the value of V in, the function is disabled output ( V out = V DD or 0), the NOR of the logic levels represented by V JG1 and V JG2, or the NAND of the logic levels repre- sented by V JG1 and V JG2. In the DRAM cell, the lateral inver- sion-mode PMOSFET (a MOSFET hav- ing a p-doped channel and an n-doped Si substrate) inherent in the n-channel G4FET is used for writing data in the horizontal direction, while the p-chan- nel JFET serves to read the data in the vertical direction. When the WWL sig- nal turns on the PMOS switch, the po- tential of the storage node (SN) is mod- ulated by WBL. When writing is disabled, SN is isolated, and during the retention time, its depletion region is more or less extended toward the body, depending on value of the datum stored in it. As a result, the resistance of the JFET channel in the vertical direc- tion is affected, causing the sensing cur- rent ( I sense) to be a function of the stored data. The sensing-current char- acteristics can be optimized via the lay- out of the G4FET structure. This work was done by Mohammad Mojar- radi of Caltech; Kerem Akarvardar, Sorin Cristoleveanu, and Paul Gentil of Grenoble University; and Benjamin Blalock and Suhan Chen of University of Tennessee for NASA’s Jet Propulsion Laboratory. In accordance with Public Law 96-517, the contractor has elected to retain title to this invention. Inquiries concerning rights for its commercial use should be addressed to: Innovative Technology Assets Management JPL Mail Stop 202-233 4800 Oak Grove Drive Pasadena, CA 91109-8099 (818) 354-2240 E-mail: [email protected] Refer to NPO-44007, volume and number of this NASA Tech Briefs issue, and the page number. r Electrically Variable or Programmable Nonvolatile Capacitors Capacitances are measured using small AC signals or changed using larger pulses. Marshall Space Flight Center, Alabama Electrically variable or programmable capacitors based on the unique proper- ties of thin perovskite films are undergo- ing development. These capacitors show promise of overcoming two impor- tant deficiencies of prior electrically programmable capacitors: • Unlike in the case of varactors, it is not necessary to supply power continu- ously to make these capacitors retain their capacitance values. Hence, these capacitors may prove useful as compo- nents of nonvolatile analog and digital electronic memories. • Unlike in the case of ferroelectric ca- pacitors, it is possible to measure the capacitance values of these capacitors without changing the values. In other words, whereas readout of ferroelectric capacitors is destructive, readout of these capacitors can be nondestructive. A capacitor of this type is a simple two- terminal device. It includes a thin film of a suitable perovskite as the dielectric layer, sandwiched between two metal or metal oxide electrodes (for example, see Figure 1). The utility of this device as a variable capacitor is based on a phenom- enon, known as electrical-pulse-induced capacitance (EPIC), that is observed in thin perovskite films and especially in those thin perovskite films that exhibit the colossal magnetoresistive (CMR) ef- fect. In EPIC, the application of one or more electrical pulses that exceed a threshold magnitude (typically some- what less than 1 V) gives rise to a non- volatile change in capacitance. The change in capacitance depends on the magnitude duration, polarity, and num- ber of pulses. It is not necessary to apply a magnetic field or to cool the device below (or heat it above) room tempera- ture to obtain EPIC. Examples of suit- able CMR perovskites include Pr 1_ .,Ca.,MnO 3 , La 1_ .,Ca.,MnO 3 , La 1_ .,S- r.,MnO 3 , and Nb 1_ .,Ca.,MnO 3 . Figure 2 is a block diagram showing an EPIC capacitor connected to a circuit that can vary the capacitance, measure the capacitance, and/or measure the re- 10 NASA Tech Briefs, October 2009 https://ntrs.nasa.gov/search.jsp?R=20090035881 2018-05-28T11:25:16+00:00Z
Transcript of Electrically Variable or Programmable Nonvolatile … 2. The Capacitance of the Electrically...
Figure 1. An Electrically Variable Capacitor of the type described in the text can be fabricated on a sil-icon or other substrate as part of an integrated circuit.
literal function, that is the fundamentalelement in multi-valued logic.) Hence,the adjustable-threshold inverter can bemade a basic building block of quater-nary logic circuits.
The real-time-reconfigurable logicgate can be realized, in a circuit partlyresembling the adjustable-threshold in-verter, by applying the logic input signalsto JG1 and JG2 and connecting theinput terminal of what would otherwisebe the inverter to a constant referencevoltage (that is, making Vin a constantvoltage). The number of transistors inthis circuit is smaller than in a classicalCMOS circuit that performs an equiva-lent logic function. The same hardwarecan be made to form any of three differ-ent functions: Depending on the valueof Vin, the function is disabled output( Vout = VDD or 0), the NOR of the logiclevels represented by VJG1 and VJG2, or
the NAND of the logic levels repre-sented by VJG1 and VJG2.
In the DRAM cell, the lateral inver-sion-mode PMOSFET (a MOSFET hav-ing a p-doped channel and an n-dopedSi substrate) inherent in the n-channelG4FET is used for writing data in thehorizontal direction, while the p-chan-nel JFET serves to read the data in thevertical direction. When the WWL sig-nal turns on the PMOS switch, the po-tential of the storage node (SN) is mod-ulated by WBL. When writing isdisabled, SN is isolated, and during theretention time, its depletion region ismore or less extended toward the body,depending on value of the datumstored in it. As a result, the resistance ofthe JFET channel in the vertical direc-tion is affected, causing the sensing cur-rent (Isense) to be a function of thestored data. The sensing-current char-
acteristics can be optimized via the lay-out of the G4FET structure.
This work was done by Mohammad Mojar-radi of Caltech; Kerem Akarvardar, SorinCristoleveanu, and Paul Gentil of GrenobleUniversity; and Benjamin Blalock andSuhan Chen of University of Tennessee forNASA’s Jet Propulsion Laboratory.
In accordance with Public Law 96-517,the contractor has elected to retain title to thisinvention. Inquiries concerning rights for itscommercial use should be addressed to:
Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099(818) 354-2240E-mail: [email protected] to NPO-44007, volume and number
of this NASA Tech Briefs issue, and thepage number.
r Electrically Variable or Programmable Nonvolatile CapacitorsCapacitances are measured using small AC signals or changed using larger pulses.Marshall Space Flight Center, Alabama
Electrically variable or programmablecapacitors based on the unique proper-ties of thin perovskite films are undergo-ing development. These capacitorsshow promise of overcoming two impor-tant deficiencies of prior electricallyprogrammable capacitors:• Unlike in the case of varactors, it is not
necessary to supply power continu-ously to make these capacitors retaintheir capacitance values. Hence, thesecapacitors may prove useful as compo-nents of nonvolatile analog and digitalelectronic memories.
• Unlike in the case of ferroelectric ca-pacitors, it is possible to measure thecapacitance values of these capacitorswithout changing the values. In otherwords, whereas readout of ferroelectriccapacitors is destructive, readout ofthese capacitors can be nondestructive.A capacitor of this type is a simple two-
terminal device. It includes a thin film ofa suitable perovskite as the dielectriclayer, sandwiched between two metal ormetal oxide electrodes (for example, seeFigure 1). The utility of this device as avariable capacitor is based on a phenom-enon, known as electrical-pulse-inducedcapacitance (EPIC), that is observed inthin perovskite films and especially inthose thin perovskite films that exhibitthe colossal magnetoresistive (CMR) ef-
fect. In EPIC, the application of one ormore electrical pulses that exceed athreshold magnitude (typically some-what less than 1 V) gives rise to a non-volatile change in capacitance. Thechange in capacitance depends on themagnitude duration, polarity, and num-ber of pulses. It is not necessary to applya magnetic field or to cool the device
below (or heat it above) room tempera-ture to obtain EPIC. Examples of suit-able CMR perovskites includePr1_.,Ca.,MnO3, La1_.,Ca.,MnO3, La1_ .,S-r.,MnO3, and Nb1_.,Ca.,MnO3 .
Figure 2 is a block diagram showingan EPIC capacitor connected to a circuitthat can vary the capacitance, measurethe capacitance, and/or measure the re-
10 NASA Tech Briefs, October 2009
https://ntrs.nasa.gov/search.jsp?R=20090035881 2018-05-28T11:25:16+00:00Z
Figure 2. The Capacitance of the Electrically Variable Capacitor is changed or measured, depending on theposition of the switch and the nature of the applied signal.
sistance of the capacitor. A pulse genera-tor applies voltage pulses to change thecapacitance. If desired, after each pulse,the capacitance and resistance can bemeasured by use of an inductance-capac-itance-resistance multimeter or an im-pedance/gain analyzer. Also if desired,the DC resistance can be measured byapplying a current of =1 µA and measur-
ElectricallyVariableCapacitor
ing the resulting voltage drop betweenthe electrodes by use of a high-internal-resistance voltmeter. The magnitude ofthe AC test potential applied by the mul-timeter or analyzer and/or the magni-tude of the DC test potential is keptbelow 50 mV — well below the thresholdmagnitude — so as not to change the ca-pacitance unintentionally.
The threshold potential dependson a number of factors, including thecomposition and thickness of the per-ovskite film and the details of theprocess used to fabricate the device.The change in capacitance caused bya given pulse can be wholly or partlyreversed by reversing the polarity ofthe pulse: that is, a pulse with one po-larity causes the capacitance to de-crease, and a pulse of the oppositepolarity causes the capacitance to in-crease. The sign of the change in ca-pacitance in relation to polarity of apulse depends on the aforemen-tioned factors and on additional fac-tors, including the capacitance-change history of the device, theamplitude and duration of the pulse.After each change, the capacitancevalue is stable: It remains the sameafter repeated measurements using asignal much smaller than a capaci-
tance-changing pulse.This work was done by Shangqing Liu,
NaiJuan Wu, Alex Ignatiev, and Jianren Liof the University of Houston for MarshallSpace Flight Center. For more information,contact Sammy Nabors, MSFC Com -mercialization Assistance Lead, [email protected]. Refer to MFS-31960-1
Inductance-Capacitance-ResistanceMultimeter or
Impedance/Gain Analyzer
DC Current Source andVoltmeter
Pulse Generator
10 System for Automated Calibration of Vector ModulatorsThis test system helps create tabular or algorithmic functions to compensate for non-idealbehavior in vector modulators.NASA’s Jet Propulsion Laboratory, Pasadena, California
Vector modulators are used to imposebaseband modulation on RF signals, butnon-ideal behavior limits the overall per-formance. The non-ideal behavior of thevector modulator is compensated usingdata collected with the use of an auto-mated test system driven by a LabVIEW®
program that systematically appliesthousands of control-signal values to thedevice under test and collects RF meas-urement data.
The technology innovation automatesseveral steps in the process. First, an au-tomated test system, using computer-controlled digital-to-analog converters(DACs) and a computer-controlled vec-tor network analyzer (VNA) systemati-cally can apply different I and Q signals(which represent the complex numberby which the RF signal is multiplied) tothe vector modulator under test(VMUT), while measuring the RF per-
Vector Network Analyzer— -
jr PC Running Vector Modulator •• •Data Collection I^ Under Test I
r: Program 1^
- °- ^
RF In rRF
;iIEEE-488 GPIB
Digital to Analogi Converters J^
The Automated Test System uses computer-controlled digital-to-analog converters and a VNA to sys-tematically apply I and Q signals to the VMUT, while measuring the RF performance.
NASA Tech Briefs, October 2009 11
