degradation in gate dielectrics and the effects on simple ...€¦ · kumar, santosh and william b....
TRANSCRIPT
1
Degradation in gate dielectrics and the effects on Degradation in gate dielectrics and the effects on simple integrated circuit building blockssimple integrated circuit building blocks
((SICBBsSICBBs))
DoDDoD MURI MURI –– Program ReviewProgram Review
Richard Southwick III1, Bill Knowlton1,2 and Jake Baker11Department of Electrical and Computer Engineering
2Department of Materials Science and EngineeringBoise State University, Boise, ID
July 14, 2006
2
AcknowledgmentsAcknowledgments
University InvolvementSTUDENTS:◘ Betsy Cheek (G), ECE◘ Mike Ogas (UG,G) ECE◘ Nate Stutzke (UG), ECE◘ Windy Wilson (UG) ECE◘ Carrie Lawrence (UG,G), ECE◘ Dave Whelchel (UG) ECE◘ Patrick Nagler (UG) ECE◘ David Jenkins (UG) ECE◘ Mark Elgin (UG) ECE◘ Dave Estrada (UG) ECE◘ Terry Gorseth (UG) ECE◘ Josh Kiepert (UG) ECE◘ Patrick Price (UG) MSEFACULTY:◘ Amy Moll, ME Faculty
Industry Involvement◘ Santosh Kumar, Cypress
Semiconductor◘ Dr. Amr Haggag, Motorola/Freescale◘ Dr. Gennadi Bersuker, SEMATECH◘ Dr. Rino Choi, SEMATECH
Funding◘ 2001 DOD Multidisciplinary Research
Initiative (MURI)◘ NSF-Idaho EPSCoR Program ◘ Cypress Semiconductor◘ NSF Major Research Instrumentation
Grant◘ DARPA Grant◘ 2000 Governor’s Higher Education
Initiative - Idaho State Board of Education◘ 2003 Micron Campus Engineering
Research Program◘ NIH INBRE #P20RR16454
3
MURI Supported Journal & Conference PublicationsMURI Supported Journal & Conference Publications((*denotes student authors)*denotes student authors)
Southwick III*, Richard G. and William B. Knowlton, Stacked Dual Oxide MOS Energy Band Diagram Visual Representation Program, IEEE Transactions on Device and Materials Reliability, (2006) - accepted for publication.T. L. Gorseth*, D. Estrada*, J. Kiepert*, M. L. Ogas*, B. J. Cheek*, P.M. Price, R. J. Baker, G. Bersuker, W.B. Knowlton, Preliminary Study of NOR Digital Response to Single pMOSFET Dielectric Degradation, presented at the Workshop on Microelectronic Devices (Boise, Idaho; April 14, 2006)M. L. Ogas*, P. M. Price*, J. Kiepert*, R. J. Baker, G. Bersuker, and W. B. Knowlton, Degradation of Rise Time in NAND Gates Using 2.0 nm Gate Dielectrics, oral presentation and publication at the 2005 IEEE Integrated Reliability Workshop, (October 2005) p. 63-66.Ogas* M. L., R. G. Southwick III* B. J. Cheek* R. J. Baker, G. Bersuker, W. B. Knowlton, Survey of Oxide Degradation in Inverter Circuits Using 2.0nm MOS Devices, in proceedings of the 2004 IEEE International Integrated Reliability Workshop, (Oct. 2004), pp. 32-36.Cheek* Betsy J., Stutzke* Nate, Santosh Kumar, R. Jacob Baker, Amy J. Moll and William B. Knowlton, Investigation of Circuit-Level Oxide Degradation and its Effect on CMOS Inverter Operation Performance and MOSFET Characteristics, in proceedings of the 2004 IEEE International Reliability Physics Symposium (April, 25-29, 2004) pp. 110-116.Kumar, Santosh, William B. Knowlton, Sridhar Kasichainula and Cesar Payan, SRAM Subthreshold Current Recovery after unipolar AC stressing, in proceedings of the 2004 IEEE International Reliability Physics Symposium (April, 25-29, 2004) pp. 46-48.Stutzke*, N., B.J. Cheek*, S. Kumar, R.J. Baker, A.J. Moll and W.B. Knowlton, Effects of Circuit-Level Stress on Inverter Performance and MOSFET Characteristics, in proceedings 2003 IEEE International Integrated Reliability Workshop, (Oct, 20-23, 2003) pp. 71-79.Payan, C., S. Kumar, A. Thupil, S. Kasichainula and W.B. Knowlton, Leakage Current Recovery in SRAM after AC stressing, in proceedings 2003 IEEE International Integrated Reliability Workshop, (Oct, 20-23, 2003) pp. 67-70.Lawrence*, C.E., B.J. Cheek*, T.E. Lawrence*, Santosh Kumar, A. Haggag, R.J. Baker, and W.B. Knowlton, Gate Dielectric Degradation Effects on nMOS Devices Using a Noise Model Approach, in Proceedings of the 15th Biennial IEEE University/Government/Industry Microelectronics (UGIM) Symposium, June 30 - July 2, 2003, pp.263-266.Kumar, Santosh and William B. Knowlton, Alternate method of TDDB study for aluminum oxide using magneto-resistance. in IEEE International Integrated Reliability Workshop, (2002) pp. 180-183.Knowlton, W.B., T. Caldwell*, J.J. Gomez*, and S. Kumar. On the nature of ultrathin gate oxide degradation during pulse stressing of nMOSCAPs in accumulation. in IEEE International Integrated Reliability Workshop, (2001) pp. 87-88.
4
MURI Supported Conference Presentations, Posters and MURI Supported Conference Presentations, Posters and Extended Abstracts Extended Abstracts (to show inclusion of UG and Grad students as authors*)(to show inclusion of UG and Grad students as authors*)
Richard Southwick III*, Michael Ogas* and William B. Knowlton, Interactive dual oxide MOS energy band diagram program, poster presentation at the 2005 IEEE International Integrated Reliability Workshop, (October 2005).Richard G. Southwick III*, Vaughn Johnson*, Joe Lindsey*, Tim Lawrence*, Jim Jozwiak, Amy Moll, William B. Knowlton, Through Wafer Interconnects: Preliminary Investigation Of Plasma Induced Damage In pMOSFETs By BOSCH DRIE, poster presentation at 2004 IEEE International Integrated Reliability Workshop, (Oct. 2004).Ogas*, Michael, Richard Southwick III*, Betsy Cheek*, Carrie Lawrence*, BSU, Santosh Kumar, Amr Haggag, Jacob Baker, and William Knowlton, Multiple Waveform Pulse Voltage Stress Technique for Modeling Noise in Ultra Thin Oxides, poster presentation at Workshop on Microelectronics and Electron Devices, (Boise, Idaho, April 16, 2004).Ogas, Michael*, Dorian Kiri*, Ouahid Salhi*, Richard Southwick III*, Betsy Cheek*, William B. Knowlton, Investigation of Ultra Thin Gate Oxide Reliability in MOS Devices and Simple ICs, poster presentation at Workshop on Microelectronics and Electron Devices, (Boise, Idaho, April 16, 2004) and at 2004 IEEE International Integrated Reliability Workshop, (Oct. 2004).Ogas*, M. L., R. G. Southwick III, B. J. Cheek*, C. E. Lawrence*, S. Kumar, A. Haggag, R. J. Baker, W. B. Knowlton, Investigation of Multiple waveform pulse voltage stress (MWPVS) technique in ultrathin oxides, poster presentation at the 2003 IEEE International Integrated Reliability Workshop (Oct, 20-23, 2003).Stutzke*, Nate, Betsy J. Cheek*, and William B. Knowlton, Circuit-level stress and gate dielectric degradation in MOSFETs, oral presentation and poster at the 2003 Micron Foundation Summer Technical Conference, Boise, ID (August 8, 2003). [Awarded Best Poster Award]Southwick III, R.G. *, Betsy Cheek* and William B. Knowlton, Charge pumping techniques for MOS devices, oral presentation and poster at the 2003 Micron Foundation Summer Technical Conference, Boise, ID (August 8, 2003).Cheek, Betsy*, Carrie Lawrence*, Tim Lawrence*, Jose Gomez*, Theodora Caldwell*, Dorian Kiri*, Santosh Kumar, Jake Baker, Amy J. Moll and William B. Knowlton, Gate dielectric degradation effects on nMOS devices and simple IC building blocks (SICBBs), extended abstract accepted for poster session at IEEE/EDS Workshop on Microelectronics and Electron Devices, (October 25, 2002).Lawrence*, C., B. Cheek*, T. Caldwell*, T. Lawrence*, D. Kiri*, S. Kumar, J. Baker, A. J. Moll and W. B. Knowlton, Pulse voltage stressing of ultrathin gate oxides in NMOS devices, poster session at the IEEE International Integrated Reliability Workshop, (October 21-24, 2002).Cheek*, B., C. Lawrence, T. Lawrence*, T. Caldwell*, D. Kiri*, S. Kumar, J. Baker, A. J. Moll and W. B. Knowlton, Circuit level reliability of ultrathin gate oxides for SICBBs: Preliminary study concentrated on the effect of stress on the NMOSFET of an inverter, poster session at the IEEE International Integrated Reliability Workshop, (October 21-24, 2002).Knowlton, William B., Research efforts in materials science related areas at Boise State University, invited talk at Idaho State University-Idaho Accelerator Center Workshop on "Application of Novel X-ray sources to Biological and Materials Science" (September 5 & 6, 2002).
5
OutlineOutline
Statement of Work for Gate Oxide Reliability
Measurement Techniques Developed
Circuits Studied Relative to Gate Oxide Thickness
Alternative Gate Dielectrics
Future Work
6
Statement of WorkStatement of Work
Focus: Gate dielectric/oxide (SiO2) degradation by EM radiation in MOSFETs & Effects on ICs ◘ Determine degradation mechanisms in gate dielectric due to
EM radiation◘ Mimic EM radiation-induced degradation in gate oxides
using DC techniques◘ Determine how gate oxide degradation mechanisms affect
simple integrated circuit building blocksInvertersLogic gates: NAND & NOR
To do this: Developed 2 - test and measurement techniques◘◘ 1) Multi1) Multi--Waveform Pulse Voltage StressingWaveform Pulse Voltage Stressing◘◘ 2) Switch Matrix Technique2) Switch Matrix Technique
7
Techniques for Examining EM Radiation Effects Techniques for Examining EM Radiation Effects on Devices and Circuitson Devices and Circuits
MultiMulti--Waveform Pulse Voltage Stressing (MWPVS)Waveform Pulse Voltage Stressing (MWPVS)
Switch Matrix TechniqueSwitch Matrix Technique
8
Continued size reduction in ICs, leads to:◘ Very close interconnect
proximity
◘ EM radiation will Capacitive Couple to Interconnects
◘ Cause noise spiking
◘ Increase voltage:From: Vcarrier
To: Vcarrier + VEM = Vnoise
V
+ =Vcarrier V EM
t
MultiMulti--waveform Pulse Voltage Stressing (MWPVS)waveform Pulse Voltage Stressing (MWPVS)-- Mimics EM RadiationMimics EM Radiation--IC CouplingIC Coupling
9
Time
Vol
tage
Time
Vol
tage
RVS (ramped voltage stress)◘ DC – difficult to extract time dependency
CVS (constant voltage stress)◘ DC - NOT typical for digital circuit operation
PVS (pulse voltage stress)◘ Better mimics digital device behavior
MWPVSMWPVS◘ Represents circuit operation with noise source
such as EM radiation
Reliability Test MethodsReliability Test Methods
Agilent4156C
O’scope
Probe DeviceDevice
Agilent 81110APulse
Pattern Generator
Ch 1
Ch 2
Time
Vol
tage
10
MWPVS MWPVS -- Experimental ResultsExperimental Results
0.0 -0.5 -1.0 -1.5 -2.010-15
10-13
10-11
10-9
10-7
10-5
10-3
10-1
10-10
10-8
10-6
10-4
10-2
100
102
SILC
Softer SBD
SBD
LHBD
| IG |
(A)
VG (V)
tox = 3.2 nm
Aox = 2.1 x 10-4 cm2
Accumulation
Fresh
| JG | (A
/cm2)
MWPVS
CVS
Finding: Degradation mechanisms induced by either EM-like pulse voltage stressing or DC stress the same.
Pre- and post- MWPVS IGATE-VGATE results:◘ Degradation mechanisms observed
SILC (Stress Induced Leakage Current)SBD and Softer SBD (Soft Breakdown) LHBD (Limited Hard Breakdown)HBD (Hard Breakdown)
CVS
11
MWPVS MWPVS vsvs CVSCVS
Weibull plots indicate device lifetime decreases by orders of magnitudes when compared to preliminary CVS data
MWPVS Weibull Plots
Similar results for lower frequencies
12
Preliminary Noise Model for MWPVSPreliminary Noise Model for MWPVS
Initial data indicates that increasing the noise signal decreases device lifetime exponentially†
◘ d, constant proportional to DCBASE of carrier signal
◘ d’, constant proportional to DCSPIKEof noise signal
◘ c, voltage accelerator factor◘ dV, noise amplitude
Preliminary noise model for a spike voltage with a DCSPIKE of 20%
†Lawrence, C.E., et al, “Gate Dielectric Degradation Effects on nMOS Devices Using a Noise Model Approach”, in Proc. of the 15th Biennial IEEE UGIM Symposium, June 30 - July 2, 2003, pp.263-266.
( )
,
1 ' c V dVcV
bd noise
d e d et
+≈ ⋅ + ⋅
,
,
bd bd noisebd
bd noise
bd BASE bd
t tt
tt T DC P
−Δ =
= ⋅ ⋅
0 1 2 3 4 5 60
200400600800
1000120014001600
Δ t bd
/ t bd
, noi
se
| Noise Voltage | (V)
EM Radiation can cause significant reduction in lifetime (over 3 orders of magnitude)
13
Designed a MWPVS technique to simulate effects of EM radiation on MOSFETs
Reliability IssuesConstructive Interference occurs due Superposition of waveforms
o Electromagnetic radiation
o Capacitive Coupling
o Mixed Signals
Device lifetime shorter for EM-like radiation than PVS or CVS
ConclusionsConclusions
Data corresponds to the noise modelData corresponds to the noise model: Device lifetime : Device lifetime exponentially decreases with increase in noise voltageexponentially decreases with increase in noise voltage
14
Techniques for Examining EM Radiation Effects Techniques for Examining EM Radiation Effects on Devices and Circuitson Devices and Circuits
MultiMulti--Waveform Pulse Voltage Stressing (MWPVS)Waveform Pulse Voltage Stressing (MWPVS)
Switch Matrix Technique (SMT)Switch Matrix Technique (SMT)
15
Gate Oxide Reliability in Simple IC Building BlocksGate Oxide Reliability in Simple IC Building Blocks((SICBBsSICBBs))
11 J.H. Stathis, R. Rodriguez, B.P. Linder, J.H. Stathis, R. Rodriguez, B.P. Linder, ““Circuit Implications of Gate Oxide Circuit Implications of Gate Oxide Breakdown,Breakdown,”” Proceedings Proceedings WoDIMWoDIM, 2002., 2002.
Only one other group examining SICBB reliabilityThey can only perform VTC – a DC technique
• They did not examine time domain• Cannot determine oxide degradation mechanism• Why? Because they cannot examine individual MOSFETS
Inverter
16
Switch Matrix Technique (SMT)Switch Matrix Technique (SMT)
Reliability studies - focus mainly on MOSFETS and large-scale ICs.
Degradation effects in MOSFETs cannot be directly correlated to gate oxide degradation in ICs
ANSWER: We developed a method that can isolate MOSFET from IC to:◘ examine EM-radiation-like oxide degradation in individual MOSFETs◘ Induce EM-radiation-like oxide degradation in individual MOSFETs or circuit
METHOD: Switch Matrix Technique (SMT) enables reliability studies at the simple integrated circuit building block (SICBBSICBB) level.1
Using SMT, reliability studies have focused on the following SICBBsSICBBs:1,2
◘ Inverter (tox: 3.2 and 2 nm)◘ Transmission Gate (TG not shown)◘ NAND (tox: 2 nm)◘ NOR (tox: 2 nm)
1B. Cheek, et al., "Investigation of circuit-level oxide degradation and its effect on CMOS inverter operation and MOSFET characteristics," in 2004 Proc. IEEE IRPS, pp. 110-116.2M. L. Ogas, et al., "Degradation of risetime in NAND gates using 2nm gate dielectrics," in 2005 Proc. IEEE IIRW, pp. 63-66.
17
Switch Matrix Technique (SMT) Switch Matrix Technique (SMT) -- SystemSystem
Agilent 41501B SMU and Pulse Generator
Unit
Agilent 4156C Precision Semiconductor Parameter
Analyzer
Agilent E5250A Low Leakage Switch Matrix
HP 4284A LCR Meter
Agilent Infinium 54832D 1GHz Oscilloscope
Micromanipulator Probe Station:•8 Cascade MicrotechMicromanipulators•Faraday Cage
Metrics ICS Software
Micromanipulator Probe Station:•8 Cascade MicrotechMicromanipulators•Faraday Cage
High Resolution: High Resolution: femtoamperefemtoampere and microvoltand microvolt
18
Switch Matrix Technique (SMT) Switch Matrix Technique (SMT) -- Experimental Experimental SetupSetup
Station #1Station #1 Station #2Station #2
Switch Matrix TechniqueSwitch Matrix Technique
SICBB CircuitVDD
Output
Input A
Input B
N
P
N
P
19
Switch Matrix Technique (SMT) Switch Matrix Technique (SMT) --Addressing Load Capacitance Addressing Load Capacitance
Δtr is within standard deviation when comparing wire bonded NAND circuit and SMT NAND circuit.
0 250.0p 500.0p 750.0p 1.0n0
20
40
60
80
100150 nm CMOS Process
Δ t r P
erce
nt
Load Capacitance (F)
tr NOR
Wire Bonded SMT
Simulation of NOR circuit shows when Cload is decreased, Δtrpercent remains constant.
SMT is a viable technique
Experimental Simulation
0
0.2
0.4
0.6
0.8
1.0
Input A Input B Fresh LLDA
90%
10%
OutputV OU
T (V)
Time (a.u.)
IncreasedDegradation
tr
1R. J. Baker, "CMOS: Circuit design, layout, and simulation," 2 ed: IEEE Press Wiley-Interscience, 2005, pp 278, 154-158.
20
SMT Experimental ProcedureSMT Experimental Procedure
Transistor-levelStress applied
to the gate terminal of the device
Before stress measurementsCircuit: VTC, V-t, IIN-VIN
MOSFET: IG-VG, ID-VD, ID-VG
Device Or
Circuit?
After stress measurementsSICBB Circuit: VTC, V-t, IIN-VIN
MOSFET: IG-VG, ID-VD, ID-VG
Circuit-level+RVS applied to the input
node of the circuit
Data Analysis & Parameter ExtractionCircuit: IIN, VOH, VOL, VSP, VOut,MAX, VOut,MIN, tr, tf
MOSFET: IG, IOn, IOff, Vth, gm, S
21
SMT SMT –– SICBBsSICBBs StudiedStudied
- +VIN VOUT
VDD
SICBB: 3.2nm InverterCircuit Level Degradation
GND
SICBB: 2nm InverterDevice Level Degradation
SICBB: 2nm NAND GateDevice Level Degradation
SICBB: 2nm NOR GateDevice Level Degradation
- +VIN VOUT
VDD
GND
Input A
Input B
VDD
Output
OUT
Input A =
Input B = VDD
VDD VDD
22
0
0.3
0.6
0.9
1.2
1.5
1.8
tf increase
tr increase
Time (a.u.)
VO
UT (
V)
ΔVOut,MIN
ΔVOut,MAXdecrease
63%
increase ~ 24%
I II
3.2nm Inverter VTC and V3.2nm Inverter VTC and V--t Responset Response
0 0.3 0.6 0.9 1.2 1.5 1.80
0.3
0.6
0.9
1.2
1.5
1.8
decrease ~ 1%
ΔVOL
VIN (V)
ΔVSP ~ 11 %
increase ~ 24 % VO
UT (
V)
Fresh
ΔVOH Fresh+ LHBD (P,N )
HBD (P,N)
x LHBD (P,N )
Increasing Degradation
x HBD (P), LHBD (N)
Time-domain degradationmore significant!ΔVOH << ΔVOut,MAX
ΔVOL = ΔVOut,MIN
23
2nm pMOSFET Gate Oxide Degradation (I2nm pMOSFET Gate Oxide Degradation (IGG--VVGG)
Observed gate leakage current increase (2.0 nm)◘ Accumulation mode ~ 2 to 3 orders of magnitude◘ Inversion mode < 1 order of magnitude
CVS IGate vs VGate
24
2nm Inverter VTC and V2nm Inverter VTC and V--t Responset Response
1R. J. Baker, H. W. Li, and D. E. Boyce, "CMOS: Circuit design, layout, and simulation," IEEE Press, pp. 201-228, 1998.
Fresh E level degradation:◙ Δ VSP ~8% shift left1
◙ Output behavior transitions from 1 to 0
Fresh E level degradation Δ rise time1 ~36% to 62%
DC Operation AC Operation
Time (a.u.)
25
• Significant reduction in NAND circuit performance
OUTInput A =
Input B = VDD
VDD VDDDegraded PMOS Position
┼R. J. Baker, "CMOS: Circuit design, layout, and simulation," 2 ed: IEEE Press Wiley-Interscience, pp 317-318, 2005.
ΔVSP ≈ 9.0% left, ± 1.2%Δtr ≈ +64.2%, ± 17.5%
%10%90 tttr −=┼
0 0.2 0.4 0.6 0.8 1.0
0
0.2
0.4
0.6
0.8
1.0
Knowlton's GroupBoise State University
August 2, 2005W3C4R2 W4C4R7
NAND1BVDD
V OU
T (V)
VIN (V)
Inc.Degradation
Fresh
VSP
DC Operation AC Operation
Setting 1-3
2nm NAND Gate VTC and V2nm NAND Gate VTC and V--t Responset Response
26
2nm NOR Gate V2nm NOR Gate V--t Responset Response
• Approximately ½ Δtr,NAND• Significant reduction in NOR circuit performance
0
0.2
0.4
0.6
0.8
1.0
Input A Input B Fresh LLDA
IncreasedDegradation
V OU
T (V)
Time (a.u.)
tr
Output10%
90%
Δtr ≈ +31% ± 6%
VDD
OUTPUT
0V
AC Operation
27
NAND Gate NAND Gate -- Potential Circuit FailurePotential Circuit Failure
Want: tdelay < tsetup
If tdelay ↑ then output becomes:
Idealoutput
Degradedoutput
Resulting problem:• Does data get clocked?• Too much delay fail
Clk
DataIdeal output Degraded output
D
ClkVDD
Q
HighLow
28
SICBB SICBB -- ConclusionsConclusions
Switch Matrix Technique – viable technique◘ Determine degradation in individual devices◘ Ability to connect device degradation to circuit degradation
Gate Oxides: 2.0 nm more susceptible than 3.2 nm to EM-radiation-like degradation◘ SICBB failure may result at a fairly low level degradation
VTC measurements may show negligible inverter degradation ◘ Suggests Oxide degradation effects in SICBBs are not a reliability issue
Decrease in logic gate performance, particularly in time domain, directly related to:
Observed degradation in Δtr of the NOR gate is about half of that observed in NAND gates
CHr Rt Δ∝ΔCH
DRIVE RI 1
∝DRIVEIΔ
29
High k Dielectrics to Replace SiOHigh k Dielectrics to Replace SiO22
Motivation
M. Houssa et al., Materials Science and Engineering R, vol. 51, pp. 37-85, 2006.
Leading high k candidate: HfO2
Dielectric Constant: ~25
30
High k Dielectrics High k Dielectrics –– HfOHfO22
Many Issues with HfO2: ◘ Prompted by SiOx interfacial layer (IL)◘ Larger number of defects◘ Reliability Assessment of IL needed
R. G. Southwick III and W. B. Knowlton, "Stacked dual oxide MOS energy band diagram visual representation program (IRW Student Paper)," IEEE
Transactions on Device and Materials Reliability, accepted for publication, 2006.
31
HfOHfO22 –– Preliminary RVS Reliability TestingPreliminary RVS Reliability Testing
Time-zero dielectric breakdown studies
-2.5 -3.0 -3.5 -4.0 -4.5 -5.0 -5.5 -6.0 -6.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
tox,HfO2
3nm 5nm 7nm 10nm 12nm 15nm
ln[-
ln[1
-F]]
EHfO2 (MV/cm)
solid = thermal SiO2 dashed/open = chemical SiO2nMOSFETW/L=10μm/0.2μmtox,SiO2
= 1.1nm
HfO2
Thickness
-0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5 -4.0 -4.5 -5.0 -5.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
tox,HfO2
3nm 5nm 7nm 10nm 12nm 15nm
ln[-
ln[1
-F]]
VHfO2 (V)
HfO2
Thickness
nMOSFETW/L=10μm/0.2μmtox,SiO
2
= 1.1nm
solid = thermal SiO2 dashed/open = chemical SiO
2
-8 -10 -12 -14
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
-16 -18 -20 -22
ln[-
ln[1
-F]]
ESiO2 (MV/cm)
tox,HfO2
3nm 5nm 7nm 10nm 12nm 15nm
n = chemical SiO2nMOSFETW/L=10μm/0.2μmtox,SiO2
= 1.1nm
solid = thermal SiO2 dashed/ope
HfO2
Thickness
-8 -10 -12 -14 -16 -18 -20 -22
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
ln[-
ln[1
-F]]
ESiO2 (MV/cm)
tox,HfO2
3nm 5nm 7nm 10nm 12nm 15nm
nMOSFETW/L=10μm/0.2μmtox,SiO
2
= 1.1nm
solid = thermal SiO2 dashed/open = chemical SiO2
HfO2
Thickness
32
Future WorkFuture Work: : Will useWill useVariable Temperature Variable Temperature ProbestationProbestation
• Custom Design (3+ years in design & development)
• Temp. RangeTemp. Range:: 5.5 – 450 K;• System PressureSystem Pressure: ~10-6 Torr
Keithley 1 SystemKeithley 1 System
High Resolution: High Resolution: attoampereattoampere and microvoltand microvolt
Keithley 4200 SCSKeithley 4200 SCSKeithley 595 QSCVKeithley 595 QSCV
Agilent 81110A Agilent 81110A 22--channel PPG channel PPG
Keithley707A Keithley707A Switch matrixSwitch matrix
Agilent 4284A LCR meter
33
Future WorkFuture Work: : Will useWill useVariable Temperature Variable Temperature ProbestationProbestation
Cryogenic Temperature Studies◘ SiO2 gate oxides
Device reliabilityCircuit reliability
◘ High dielectric constant reliability
High temp measurements ◘ Thermal stability◘ Reliability: thermal &
voltage acceleration
-1.0 -0.5 0 0.5 1.010-16
10-14
10-12
10-10
VG (V)
| IG |(
A)
Temperature: 5.5 K 50 K 75 K 100 K 150 K 250 K 300 K 350 K
pMOSFET: 10 μm/125 nm tOX = 2.0 nm
Knowlton's GroupBoise State University[LTPS 4.19.06]
inc. temp.inc. temp.
inc. temp.inc. temp.
inc. temp.inc. temp.
0 -0.2 -0.4 -0.6 -0.8 -1.010-13
10-11
10-9
10-7
10-5
10-3
10-1
Temperature: 5.5 K 50 K 75 K 100 K 150 K 250 K 300 K 350 K
Knowlton's GroupBoise State University[LTPS 4.19.06] VD (V)
| ID |
(A)
pMOSFET: 10 μm/125 nm tOX = 2.0 nm VD = -1.0 V
inc.
temp.
inc.
temp.
34
Thank youThank you
Questions?