VOL. 11, NO. 5, MARCH 2016 ISSN 1819-6608
ARPN Journal of Engineering and Applied Sciences
© 2006-2016 Asian Research Publishing Network (ARPN). All rights reserved.
www.arpnjournals.com
3137
VARIABILITY ANALYSIS OF PROCESS PARAMETERS ON
SUBTHRESHOLD SWING IN VERTICAL DG-MOSFET DEVICE
K. E. Kaharudin
1
, F. Salehuddin
1
, A. H. Hamidon
1
, A. S. M. Zain
1
, M. N. I. Abd Aziz
1
and I. Ahmad
2
1
Centre for Telecommunication Research and Innovation (CeTRI), Faculty of Electronics and Computer Engineering, Universiti
Teknikal Malaysia Melaka (UTeM), Hang Tuah Jaya, Durian Tunggal, Melaka, Malaysia
2
Centre for Micro and Nano Engineering (CeMNE), College of Engineering, Universiti Tenaga Nasional (UNITEN), Kajang, Selangor,
Malaysia
Email ID: khairilezwan@yahoo.com.my
ABSTRACT
As the MOSFET’s size is expected to be shrunk every year, it is difficult to mitigate the short channel effect
(SCE) issues arising in the device. The conventional MOSFET’s structure is no longer practical to apprehend these types of
issues, especially for a device with a very small gate length (Lg). The SCE issues happen due to the reduction of the gate
length (Lg), which causes the distance between the source and the drain region to become too close to each other. As a
consequence, it causes the charge sharing effects between source and drain region that eventually leads to higher
subthreshold swing (SS). A steep SS value around 55 to 65 mV/dec is desired in MOSFET device for faster switching
operation. Therefore, a new architecture of Vertical Double Gate (DG) MOSFET device is proposed to circumvent these
issues. Besides that, the process parameter variations in the device are also considered as one of the important factors that
significantly affect the SS value. In this paper, an attempt to analyze the variability of multiple process parameters towards
the SS value in 12nm gate length (Lg) vertical DG-MOSFET device has been made. At the end of the experiments, it was
found that the most dominant process parameter that contributed a large effect on SS value was halo implantation tilt angle.
The lowest possible value of SS was observed to be 62.52 mV/dec with signal-to-noise ratio (SNR) of -35.83 dB.
Keywords: ANOVA, DG-MOSFET, subthreshold swing, taguchi method.
INTRODUCTION
The Metal Oxide Semiconductor Field Effect
Transistor (MOSFET) is widely applied for amplifying
and switching electronics signals. In MOSFET devices, a
voltage potential that appears on the oxide-insulated gate
electrode is capable of creating a conducting channel
between the source and the drain region. The channel can
be either n-type or p-type depending on the type of dopant
used in the substrate. It is commonly recognized as a
NMOS and a PMOS device. The source region is the
source of charge carrier, either electrons or holes flow
through the channel [1]. Meanwhile, the drain region is
where the charge carrier leaves the channel.
Nowadays, MOSFET devices are actively being
scaled down for miniaturization of integrated circuit (IC).
By reducing the size of transistors and interconnectors,
more circuits can be fabricated on each silicon wafer.
Thus, the price of each circuit will become cheaper and
affordable. However, miniaturization in MOSFET device
has its side effect, which is known as short channel effects
[2]. Many electronics scientists and researchers have
conducted research on suppressing the SCEs in the nano-
scale MOSFET devices. They tend to come out with new
ideas of architecture and approach on how to mitigate the
SCE issues.
Through the implementation of vertical DG-
MOSFET architecture, the effective channel length (Lc) is
no longer dependent on the gate length (Lg). The height of
silicon pillars acts as a control factor to vary the size of
effective channel length [3]. Therefore, various SCE
problems are still possible to be mitigated even at small
gate length (Lg) such 12nm of Lg. However, there is an
important aspect that has to be considered, which is known
as process parameter variations. Process parameter
variations occur in every type of MOSFET device. These
variations would affect the output responses (electrical
characteristics) of the MOSFET devices. They may cause
the degradation of overall MOSFET’s electrical
performance [4, 5].
In this paper, the 12nm gate length of vertical
DG-MOSFET was successfully designed by using
software SILVACO Technology Computer Aided Design
(TCAD). The input process parameters that contribute the
most significant impact on the SS value were investigated
by utilizing L27 orthogonal array Taguchi method. The
subthreshold swing is an important response which
indicates the scalability limit of the MOSFET. It also
shows how much change in the gate voltage (VG) is
required to change drain current (ID) by one decade [6].
Vertical Dg-MOSFET design using Silvaco TCAD
A P-type silicon with <100> orientation was used
as the main substrate for this experiment. A different
silicon orientation will give different substrate properties
such as device mobility. Normally, P-type silicon with
<100> orientation is used for MOSFET fabrication
process due to a significant improvement in the effective
maximum mobility for the n-MOSFETs with a channel
along the 100 directions and the channel direction had no
effect on the noise level and the performances of the
devices [7]. The silicon was etched in order to form a
pillar or ridge that separated the two gates. The
combination of buried oxide (BOX) and the polysilicon
layer formed an enhanced channel that was capable of