Upload
hoangthuy
View
224
Download
0
Embed Size (px)
Citation preview
ANALYTICAL SOI MOSFET MODEL VALID FOR
GRADED-CHANNEL DEVICES
Benjamín Iñíguez1 , Marcelo Antonio Pavanello2,3, João AntonioMartino3 and Denis Flandre4
3Escola Tècnica Superior d`EngenyeriaUniversitat Rovira I Virgili, Tarragona, Spain
2Center for Semiconductor ComponentsState University of Campinas, Campinas, Brazil
3Laboratório de Sistemas IntegráveisEscola Politécnica da Universidade de São Paulo
4Laboratoire de MicroéletroniqueUniversité Catholique de Louvain, Louvain-la-Neuve, Belgium
Outline:Outline:
• Introduction
• Analytical Model Formulation
• Results and Discussion
• Fabrication Process and Experimental Results
• Conclusion
The advantages of Fully-depleted SOI nMOSFETs overPartially-Depleted (PD) and bulk MOSFETs are well known
1) Lower subthreshold ideality factor2) Reduced short-channel effects3) Better analog performance4) Better microwave performance (gain, speed and cutofffrequency)5) Lower 1/f and thermal noise
As a consequence, FD SOI MOSFET is a very goodcandidate for low-power and low-noise microwavecircuits
Introduction - 1
However, Fully-Depleted SOI nMOSFETs suffer from someParasitic Bipolar Effects due to the Floating Body
Premature Drain Breakdown Parasitic Bipolar Transistor Action: Emitter – Source Base – Body (floating) Collector – Drain
Some of the proposed solutions:
� Bandgap Engineering (Ge implanted) [M. Yoshimi et al., IEDM 94, p. 429]
� Source-Body Contact [IEDM 94, p. 657]
� Asymmetric Channel Profile using tilt implantation [B. Cheng et al.,
IEEE Int. SOI Conference, p. 113, 1998]
Introduction - 2
⇒ THE GRADED-CHANNEL (GC) SOI nMOSFET
N+
LLD
L
PolyField
Oxide
FieldOxide
Substrate
naturaldopingNA
Buried Oxide
N+
Source Drain
Introduction - 3
� FD SOI CMOS technology fully compatible processing
� One photolitographic step is used to adjust the threshold voltage ionicimplantation position
� This photolithographic step is the same used to mask the p-type transistor NO additional photolithographic step has to be included in the CMOS processing.
� The effective channel length is about L-LLD
Introduction - 4
Reported results indicate:Reported results indicate:
Increase in the breakdown voltage
Tremendous reduction in the drain output conductance
Increase in the transconductance
The Graded-Channel SOI MOSFET is a great candidateThe Graded-Channel SOI MOSFET is a great candidatefor analog circuits resulting in improved amplifiers andfor analog circuits resulting in improved amplifiers and
current mirrorscurrent mirrors
Our goal: Development of a continuos model for GCtransistor to allow reliable simulation of analog
circuits
Introduction - 5
y directionLyHD yLD
log(NA)
NAF, HD
NAF, LD
Schematic doping profile along channel
( )
−−=
4-
2HD
HD,AFA10 2
yyexpNN
In the interval yHD≤y ≤ yLD
Introduction - 6
Based on physical principles, an analytical FD SOI MOSFET model was developed by B. Iñiguez et al.
For a conventional fully-depleted SOI nMOSFET:
−
φµ−=
dy
dQv
dy
dQWI nf
Tsf
nfDS
There is a linear relation between the surface potential, φsf,and the inversion charge density, Qnf:
−−−=
oxf
bFBfGFoxfnf C2
QVVCQ
φ−
+−− SF
oxb
bFBbGB
oxf
bb nC2
QVV
C
C
Introduction - 7
An explicit and unified expression for Qnf is used in the IDS equation(proposed by Iñiguez et al): .
NTToxfnf SnvCQ −= ( )
−−−
+T
FiGF
NTT
oxf0
v n
ynVVthVexp
S v n
CQ
1log( )
−−+
NTT
FGF
S v n
ynVVthVexp
This model is continuous, valid from weak to strongThis model is continuous, valid from weak to stronginversion regimesinversion regimes
Integrating the inversion charges along channel results:
( )[ S,nfD,nfTDS QQvL
WI −µ=
−−
oxf
2S,nf
2D,nf
Cn 2
Qnf,D and Qnf,S are the inversion charge densities at the drain and source,respectively
Introduction - 8
1- Short-channel effects includedDIBL and charge sharingVelocity saturationChannel length modulation
2- Complete charge model3- Scalability down to 0.16 µm4- Temperature dependencies included. Model validated upto 300 oC :5-Macro-model developed to extend the model to themicrowave range. Accuracy demonstrated up to 40 GHz..
These physical principles allow to develop a complete CAD model and to adapt it to many conditions
Analytical Model Formulation - 1
For long-channel transistors, the transition region into the channelFor long-channel transistors, the transition region into the channel(from (from yyHDHD≤≤y y ≤≤ yyLDLD) can be considered negligible:) can be considered negligible:
With this assumption, there will be two explicit equations for theinversion charges in the low doped (Qnf, LD) and conventionally
doped (Qnf, HD) parts of the channel
NTToxfLD,nf SnvCQ −=( )
−−−
+T
LD,FiGF
NTT
oxfLD,0
v n
ynVVthVexp
S v n
CQ
1log( )
−−+
NTT
LD,FGF
S v n
ynVVthVexp
( )
−−−
+T
HD,FiGF
NTT
oxfHD,0
v n
ynVVthVexp
S v n
CQ
1logNTToxfHD,nf SnvCQ −=( )
−−+
NTT
HD,FGF
S v n
ynVVthVexp
V(y) is the potential variation along channel
Analytical Model Formulation - 2
Integrating the inversion charges along channel results:
−µ−= ∫
− LDLL
0
HD,nfT
oxf
HD nf,HD,nfHDDS dy
dQv
dy Cn
dQ QWI
−µ−= ∫
−
L
LL
LD,nfT
oxf
LD nf,LD,nfLD
LD
dy
dQv
dy Cn
dQ QW
( )[0LDLL HD,nfHD,nfT
LD
HDDS QQv
LL
WI −
−µ
=−
−− −
oxf
2HD,nf
2HD,nf
Cn 2
QQ0LDLL
( )[LDLLL LD,nfLD,nfT
LD
LDDS QQv
L
WI
−−
µ=
−− −
oxf
2LD,nf
2LD,nf
Cn 2
QQLDLLL
The GC SOI The GC SOI nMOSFET nMOSFET is therefore interpreted as ais therefore interpreted as aseries association of two conventional transistors, eachseries association of two conventional transistors, each
one representing a part of the channel regionone representing a part of the channel region
Results and Discussion - 1
Comparison between MEDICI numerical two-dimensionalComparison between MEDICI numerical two-dimensionalsimulations and model equations solutionsimulations and model equations solution
toxf=30 nm, tSi=80 nm, toxb=400 nm, Qox1/q=Qox2/q=5 x 1010 cm-2,NAF, HD=1017 cm-3 and NAF, LD=1015 cm-3. L=10 µm
Device characteristics:
Results and Discussion - 2
0.0 0.5 1.0 1.5 2.0 2.5 3.00.0
0.2
0.4
0.6
0.8
1.0
1.2
LLD/L=0.50
VDS =0.1 V
LLD/L=0.25
MEDICI Model
I DS
[A/ µ
m]
VGF [V]
Gate characteristics in the linear regime
Results and Discussion - 3
0.5 1.0 1.5 2.0 2.5 3.010-12
10-11
10-10
10-9
10-8
10-7
10-6
10-5
LLD/L=0.50
LLD/L=0.25
VDS=0.1 V
MEDICI Model
I DS
[A/ µ
m]
VGF [V]
Gate characteristics in linear regime
Results and Discussion - 4
Drain characteristics
0.0 0.5 1.0 1.5 2.0 2.5 3.00.00
0.05
0.10
0.15
0.20
0.25
LLD/L=0.25
LLD/L=0.5
VGT=200 mV
MEDICI Model
I DS
[A/ µ
m]
VDS [V]
Excellent agreement between the model and MEDICI simulationsExcellent agreement between the model and MEDICI simulations
The increase of IThe increase of IDSDS with L with LLDLD/L (/L (LLeffeff reduction) is greatly modeledreduction) is greatly modeled
Experimental Results - 1
Drain characteristics - L = 4µm
0 1 2 30
20
40
60
80
100
FD SOI
LLD/L=0.25
LLD/L=0.37
LLD/L=0.50
VGT=500 mV
Experimental Model
I DS
[ µA
]
VDS [V]The tremendous increase in the Early voltage is adequatelyThe tremendous increase in the Early voltage is adequately
reproduced by the modelreproduced by the model
Experimental Results - 2
Gate characteristics in saturation - L = 2µm
0.0 0.5 1.0 1.5 2.0
10-11
10-10
10-9
10-8
10-7
10-6
10-5
10-4
10-3
LLD/L=0.51
LLD/L=0.25
VDS=1.5 V
Model Experimental
I DS
[A]
VGF [V]
10-11 10-10 10-9 10-8 10-7 10-6 10-5 10-40
10
20
30
40
VDS=1.5 V
model experimental
g m/I
DS
[V-1
]
IDS/(W/Leff) [A]
Experimental Results - 3
Transconductance over drain current vs scaled drain currentL = 2µm
Conclusion - 1
Our unified FD SOI MOSFET model has been adapted toOur unified FD SOI MOSFET model has been adapted tolong-channel Graded-Channel SOI long-channel Graded-Channel SOI nMOSFET nMOSFET based on abased on a
series association of transistors to represent the channel dopingseries association of transistors to represent the channel dopingprofileprofile
The validity of the proposed model has been verified by bothThe validity of the proposed model has been verified by bothnumerical simulations and experimental resultsnumerical simulations and experimental results
An excellent agreement has been found in both casesAn excellent agreement has been found in both cases
AcknowledgementsAcknowledgements
Brazilian Federal Agency CNPq for the financial support
UCL Microelectronics Laboratory Staff for the device processing