Statistical Modelling of ft to Process Parameters in 30 NM Gate Length Finfets

International journal of VLSI design & Communication Systems (VLSICS) Vol.1, No.3, September 2010
DOI : 10.5121/vlsic.2010.1304 36
STATISTICAL MODELLING OF ft TO PROCESS
PARAMETERS IN
30 NM GATE LENGTH FINFETS
B. Lakshmi and R. Srinivasan
Department of Information Technology
SSN College of Engineering, Kalavakkam – 603 110, Chennai, India
laxmi.balu@yahoo.com
srinivasanr@ssn.edu.in
ABSTRACT
This paper investigates the effect of process variations on unity gain frequency (ft) in 30 nm gate length
FinFET by performing extensive TCAD simulations. Six different geometrical parameters, channel
doping, source/drain doping and gate electrode work function are studied for their sensitivity on ft. It is
found that ft is more sensitive to gate length, underlap, gate-oxide thickness, channel and Source/Drain
doping and less sensitive to source/drain width and length, and work function variations. Statistical
modelling has been performed for ft through design of experiment with respect to sensitive parameters.
The model has been validated through a comparison between random set of experimental data simulations
and predicted values obtained from the model.
KEYWORDS
ft , FinFET, process variations, Statistical modelling, Design of Experiments
1. INTRODUCTION
The progress in CMOS technology has made it well suited for RF and microwave operations at
high level of integration [1], and the continuous improvement of the device performance has
made it a contender for low-power and, eventually, low-cost radio front end. In the area of multi-
gate transistors, double-gate FinFETs are considered a serious contender for channel scaling [2],
[3] because of their quasi-planar structure and the compatibility with CMOS. Several authors
have already studied the low field mobility in fins of various widths [4]-[6]. RF performance of
FinFETs is reported in [7], [8].
Unity gain frequency (ft) is one of the important metric in RF applications. ft is defined as the
frequency at which the current gain of the device becomes unity. ft is calculated by
)
1
(
C
2
g
f
gg
m
t
π
=
where Cgg is the combination of Cgs, Cgd, overlap capacitance and any other fringing capacitance.
In this article, nine different geometrical parameters related to FinFET are varied to capture their
sensitivity on ft.
This paper analyzes double-gate FinFETs as a downscaling option for CMOS technology from
an RF perspective. The effect of various structural and doping parameters (9 parameters in total)
on ft is studied and the five most sensitive parameters are identified. Using these sensitive
parameters a 5 point DOE (design of experiments) is designed and the simulations are done. i.e.
this work is based on design and simulation of the nominal device, design of experiment and
running of the experiment, extraction of results, fitting the response surfaces (models) and
testing of the models. We have modelled ft in terms of the most sensitive parameters like gate

37
length, underlap, gate oxide thickness, channel and source/drain doping. The model that describe
these quantities have an approximated form [9] as,
)
2
(
x
x
b
x
x
b
x
x
b
x
x
b
x
x
b
x
x
b
x
x
b
x
x
b
x
x
b
x
x
b
x
b
x
b
x
b
x
b
x
b
x
b
x
b
x
b
x
b
x
b
b
y
5
4
45
5
3
35
4
3
34
5
2
25
4
2
24
3
2
23
5
1
15
4
1
14
3
1
13
2
1
12
2
5
55
2
4
44
2
3
33
2
2
22
2
1
11
5
5
4
4
3
3
2
2
1
1
0
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
=
where x1 is the gate length, x2 is the underlap, x3 is the gate oxide thickness, x4 is the channel
doping and x5 is the Source/Drain doping, y is the unity gain frequency, b’s are the fitting
parameters determined by the data obtained from the experiment. Next section deals with the
simulation methodology followed in this paper. Section III discusses the simulation results and
statistical modelling. Finally section IV provides conclusions.
2. SIMULATION METHODOOGY
Sentaurus TCAD simulator from Synopsys [10] is used to perform all the simulations. This
simulator has many modules and the following are used in this study.
• Sentaurus structure editor (SDE): To create the device structure, to define doping, to
define contacts, and to generate mesh for device simulation
• Sentaurus device simulator (SDEVICE): To perform all DC and AC simulations
• Inspect and Tecplot: To view the results.
The physics section of SDEVICE includes the appropriate models for band to band tunnelling,
quantization of inversion layer charge, doping dependency of mobility, effect of high and normal
electric fields on mobility, and velocity saturation. The structure generated from SDE is shown
in Fig 1. Doping and mesh information can also be observed in Fig. 1. Figure 2 shows the
schematic diagram of the device. Totally nine different parameters are considered in this study.
Out of them, six are geometrical parameters - gate length (Lg), underlap (Lun), fin width (W),
source/drain width (SW), source/drain length (SL), and Tox and these are shown in Fig. 2. Other
three parameters are channel doping (Nch), source/drain doping (NSD) and gate electrode work
function (WF). The various process parameters considered in this study and their range are given
in Table. 1. Table 1 also gives the dimensions of the nominal device. Standard AC simulations
are done in SDEVICE and ft is extracted from these results. ft is the frequency at which
|Y21/Y11| equals one, and it strongly depends on the gate bias. At various gate biases ft is
calculated and the maximum of them is taken as ft. Supply voltage (Vdd) used in this study is 0.8
V.
Figure 1. Structure of the Dual-Gate FinFET

38
Figure 2. Schematic view of Dual-Gate FinFET
Table 1. Characteristics of the device
Process parameter Nominal value Range
Gate length (Lg) 30 nm 20 nm to 40 nm
Underlap (Lun) 3 nm 1 nm to 8 nm
Fin Width (W) 4 nm 2 nm to 7 nm
Source Length (SL) 15 nm 10 nm to 20 nm
Source Width (SW) 8 nm 4 nm to 16 nm
Channel Doping (Nch) 1x1016
/cm3
1x1015
/cm3
to
1x1019
/cm3
Source Drain Doping
(NSD)
1x1020
/cm3
1x1018
/cm3
to
2x20
/cm3
Oxide Thickness (Tox) 1 nm 0.5 nm to 2 nm
Gate Work Function
(WF)
4.337 eV 4.13 eV to 4.9 eV

39
3. RESULTS AND DISCUSSION
3.1 Sensitivity Analysis of Process Parameters
The nine different process parameters are varied one at a time, according to the range
given in Table 1 and their sensitivity to ft is analysed in this section.
3.1.1 Variation in Gate Length
Figure 3 shows the variation of ft against Lg. It can be observed from Fig. 3 that ft initially
increases and then decreases. As per (1), ft is decided by both gm and Cgg. While gm degrades
with Lg, Cgg shows a different behaviour i.e. initially decreases with Lg and then increases (Fig
4).The initial decrease of Cgg can be attributed to the reduction of Cgd [11]. At some point, Cgs
starts dominating and calls for the increase in Cgg. The combined behaviour of gm and Cgg
contributes to the variation of ft w.r.t. Lg.
Figure 3. Variation of ft with respect to Lg
Figure 4. Variation of Cgs with respect to Lg

40
3.1.2. Variation in Underlap
Figure 5 depicts the plot between ft and Lun. It can be seen from Fig 5 that ft initially increases
and then decreases w.r.t. Lun. Increasing Lun reduces the fringing capacitance, and thereby
decreases Cgg [12], [13].The Cgg in DGMOS can be expressed as
( ) )
3
(
fringing
ov
si
ox
gg C
||
C
||
C
,
C
Series
C =
where Cox is the oxide capacitance, Csi is the silicon body capacitance, Cov is the gate to
source/drain overlap capacitance and Cfringing is the fringing capacitance and is given by
)
4
(
e
T
L
W
ln
WK
C ox
T
un
L
ox
T
un
L
2
ox
2
un
di
fringing
+
−
−
+
π
π
∈
=
When Lun increases current degrades and thereby gm monotonically decreases. The combined
behavior of gm and Cgg is responsible for the ft trend seen in Fig. 5.
Figure 5.Variation of ft with respect to Lun
3.1.3 .Variation in Fin Width
When W is varied we may either face volume inversion or may not, depending upon the channel
doping levels. When the channel doping is 1x1016
/cm3
volume inversion is not seen [14].
Therefore, the increase in W increases current and thereby gm and ft. Figure 6 shows this kind of
behaviour between ft and W. For the channel doping around 1.5x1018
/cm3
, volume inversion
effect is seen which causes ft to decrease initially and then to increase w.r.t W. This is depicted
in Fig. 7.

41
Figure 6. Variation of ft w.r.t W without volume inversion
Figure 7. Variation of ft w.r.t W with volume inversion
3.1.4. Variation in Source Length
Figure 8 shows the ft versus source length plot. Increasing source length increases the parasitic
resistance associated with the channel and degrades gm which in turn decreases ft.
Figure 8. Variation of ft with respect to SL

42
3.1.5 .Variation in Source Width
Figure 9 shows ft as a function of source width. It can be noticed from Fig. 9 that ft is almost
independent of SW.
Figure 9. Variation of ft with respect to SW
3.1.6 .Variation in Oxide Thickness
Figure 10 shows the variation of ft with Tox. ft increases initially w.r..t Tox. Both gm and Cgg
together control ft. Cgg always decreases when Tox increases whereas gm may go down or high
depending on whether we are driven by short channel effects or not. The combined effect of gm
and Cgg decides ft behaviour with respect to Tox.
Figure 10. Variation of ft with respect to Tox
3.1.7 .Variation in Channel Doping
Figure 11 shows the variation of ft against Nch. Threshold voltage of DGFET/FinFET is
insensitive up to 1x1017
/cm3
[15]. From which it can be reasoned out that ft is also insensitive at
lower channel doping levels. The same is seen in Fig. 11. At higher doping levels, ft decreases
due to gm degradation.

43
Figure 11. Variation of ft with respect to Nch.
3.1.8 .Variation in Source/drain Doping
Figure 12 shows the variation of ft with source/drain doping. When NSD increases Ion and gm
increase due to the lowered parasitic series resistance values, and thereby ft increases. This is
reflected in Fig.12
Figure 12. Variation of ft with respect to NSD
3.1.9 .Variation in Work Function
High frequency characteristics are less sensitive to work function variation [16]. Therefore, ft is
expected to be indifferent to gate electrode work function. Figure 13 shows ft versus gate work
function plot and it can be noticed that ft exhibits a flat behaviour w.r.t work function.

44
Figure 13. Variation of ft with respect to WF
3.2 Statistical modelling:
The sensitive parameters are chosen as Lg, Lun, Tox, Nch and NSD. These sensitive parameters are
made to undergo a variability study with the help of design of experiments. The range of these
sensitive parameters for the variability study is given in Table 2 which results in 2500
simulations. The model that has been obtained by the regression technique is generated with the
help of SPSS [17].The regression coefficients for the second order polynomial of the process
parameters are shown in Table 3.
Table 2 .Range of Sensitive Parameters
Process parameters Range Points in the range
Gate length (Lg) 20 nm to 40 nm 20, 25, 30, 35 and 40 (nm)
Underlap (Lun) 1nm to 9 nm 1,3, 5,7 and 9 (nm)
Oxide Thickness (Tox) 0.5 nm to 2.2 nm 0.5, 1,1.2,2 and 2.2 (nm)
Channel Doping (Nch) 1X1016
/cm3
to 1X1019
/cm3
1X1016
, 1X1017
, 1X1018
and
1X1019
(/cm3
)
Source Drain Doping (NSD) 5X1018
/cm3
to 2X1020
/cm3
5X1018
, 5X1019
, 9.5X1019
,
1.1X1020
and 2X1020
(/cm3
)

45
Table 3. Process parameters along with their corresponding
regression coefficients
Factor Coefficient Factor values
Constant bo 2.765E12
Lg b1 -1.103E20
Lun b2 -3.964E19
Tox b3 4.007E19
Nch b4 -.052
NSD b5 .009
Lg
2
b11 1.164E27
Lun
2
b22 -4.210E27
Tox
2
b33 -3.564E28
Nch
2
b44 9.698E-16
NSD
2
b55 -1.489E-17
LgLun b12 1.874E27
LgTox b13 1.639E27
LgNch b14 1266894.540
LgNSD b15 -100967.539
LunTox b23 7.137E27
LunNch b24 -3105744.884
LunNSD b25 -124191.069
ToxNch b34 -1672454.265
ToxNSD b35 23534.568
NchNSD b45 -4.254E-17
In order to study the statistical nature of the device output with respect to sensitive parameters,
we have generated 35
(=243) uniformly distributed pseudo-random numbers for each of these
sensitive parameters and ft values are predicted from the generated model. For the same set of
random numbers generated, TCAD simulations are carried out and the ft values are extracted. To
have a correlation plot TCAD values are plotted against model values and the same is depicted
in Fig. 14. The correlation coefficient is found out as r=0.992.

46
Fig. 14: Correlation plot between TCAD simulated data and model
predicted data for ft. Correlation coefficient r is depicted here
4. CONCLUSION
Six geometrical parameters (Lg, Lun, W, SW, SL, and Tox), and three non-geometrical parameters
(Nch, NSD, and WF) have been varied over a range and their effect on ft have been studied
through TCAD simulations. It was found that Lg, Lun, Tox, Nch and NSD are more sensitive
parameters whereas gate electrode work function, source/drain length and width are less
sensitive parameters. By running DOE using the sensitive parameters in TCAD, statistical
modelling has been performed. Correlation coefficient around 0.992 was got while running the
simulations with random set of values for sensitive parameters.
ACKNOWLEDGEMENT
This work is supported by Department of Science & Technology, Government of India under
SERC scheme
REFERENCES
[1] L. Larson, “Silicon technology tradeoffs for radio-frequency/mixed signal systems-on-a-chip,”
IEEE Trans. Electron Devices, vol. 50, no. 3, pp. 683–699, Mar. 2003.
[2] International Technology Roadmap for Semiconductor (ITRS). [Online]. Available:
www.itrs.net/Links/2005ITRS/PIDS2005.pdf
[3] D. J. Frank, R. H. Dennard, E. Nowak, P. M. Solomon, Y. Taur, and H. S. P. Wong, "Device
scaling limits of Si MOSFETs and their application dependencies," Proc. IEEE, vol. 89, no. 3, pp.
259-288, Mar. 2001.
[4] F. Gamiz, J. Roldan, J. Lopez-Villanueva, P. Cartujo-Cassinello, and J. E. Carceller, "Surface
roughness at the Si?SiO2 interfaces in fully depleted silicon-on-insulator inversion layers," J.
Appl. Phys., vol. 86, no. 12, pp. 6854-6863, Dec. 1999.

47
[5] A. Khakifirooz and D. Antoniadis, "On the electron mobility in ultrathin SOI and GOI," IEEE
Electron Device Lett., vol. 25, no. 2, pp. 80-82, Feb. 2004.
[6] K. Uchida, H. Watanabe, J. Koga, A. Kinoshita, and S. Takagi, “Experimental study on carrier
transport mechanism in ultrathinbody SOI MOSFETs,” in Proc. IEEE SISPAD, Sep. 2003, pp. 8–
13.
[7] S. Nuttinck, “Ultra-thin-body silicon-on-insulator as a CMOS downscaling option: An RF
perspective,” IEEE Trans. Electron Devices,vol. 53, no.5, pp. 1193–1199, May 2006.
[8] D. Lederer, B. Parvais, A. Mercha, N. Collaert, M. Jurczak, J.-P. Raskin, and S. Decoutere,
“Dependence of finFET RF performance on fin width,” in Proc. 6th Top. Meeting SiRF, San
Diego, CA, Jan. 18–20, 2006, pp. 4–6.
[9] Montgomery DC. Design and analysis of experiments. 5th ed. New York: John Wiley & Sons;
2001.
[10] Synopsys Sentaurus Device User Guide, 2008-09
[11] Han-Su Kim, Kangwook Park, Hansu Oh, and Eun Seung Jung” Importance of Vth and
Substrate Resistance Control for RF Performance Improvement in MOSFETs” IEEE Electron
Device letters, Vol. 30, No. 10, October 2009
[12] Fathipour Morteza, Nematian Hamed, Kohani Fatemeh “ The Impact of Structural Parameters on
the electrical characteristics of Nano scale DC-SOI MOSFETs in sub-thereshold region” SETIT
2007 4th International Conference: Sciences of Electronic, Technologies of Information and
Telecommunications March 25-29, 2007
[13] R. Shrivastava and K. Fitzpartick, “A simple model for the overlap capacitance of a VLSI MOS
device.” IEEE Trans.Electron Devices, Vol. ED-29, pp.1870-1875,1982
[14] G Curatola, S. Nuttinck, “The Role of Volume Inversion on the Intrinsic RF Performance of
Double-Gate FinFETs”, IEEE Transactions On Electron Devices, Vol. 54, No.1, Jan 2007.
[15] Shiying Xiong and Jeffrey Bokor,” Sensitivity of Double-Gate and FinFET Devices to Process
Variations” IEEE Transactions on Electron Devices, Vol. 50, No. 11, Nov 2000.
[16] Chih-Hong Hwang, Tien-Yeh Li1, Ming-Hung Han, Kuo-Fu Lee, Hui-Wen Cheng1, and
Yiming Li,” Statistical Analysis of Metal Gate Workfunction Variability, Process Variation, and
Random Dopant Fluctuation in Nano-CMOS Circuits”, Proceedings of 14th ¬International
Conference on Simulation of Semiconductor Processes and Devices,” (SISPAD), Pg No 99-102,
2009.
[17] SPSS Base 15.0 User’s Guide, 2006.

Statistical Modelling of ft to Process Parameters in 30 NM Gate Length Finfets

More Related Content

What's hot (8)

Similar to Statistical Modelling of ft to Process Parameters in 30 NM Gate Length Finfets (20)

Recently uploaded (20)

Statistical Modelling of ft to Process Parameters in 30 NM Gate Length Finfets