High-k Dielectric Process to Minimize Mobile Ionic Penetration David W. Parent, Eric Basham, Janet Davis

basketontarioElectronics - Devices

Nov 2, 2013 (3 years and 9 months ago)

62 views

TEMPLATE DESIGN © 2008

www.PosterPresentations.com

High
-
k Dielectric Process to Minimize Mobile Ionic Penetration

David W. Parent, Eric Basham, Janet Davis

Department of Electrical Engineering, San Jose State university, San Jose, California 95192.

Introduction

The nMOS field effect transistor configured as a high gain
amplifier can be used to amplify microvolt signals from
neurons
1
. To increase the gain of the amplifier, the use of
a high
-
k dialectic material, such as hafnium oxide (HfO) is
explored.


Integration of the MOSFET with the neuron provides the
ability to record neural signals. This can further
understanding of how they transmit and process data
efficiently. Research for interfacing neurons to transistors
has been conducted in the past, however, open gate
transistors using standard SiO
2

as the gate dielectric
suffer from drift due to mobile species ionic penetration
2
.
Furthermore, SiO
2

degrades over time when introduced in
a biological environment. Because HfO is non reactive
and has good biocompatibility, it is a good candidate for
use in biological applications
3
. In addition, it is a high
-
k
dielectric material which has the benefit of increasing the
gain of the transistor. This structure is shown below.



Methodology

Results

Summary

Key References

Acknowledgements

A process for fabricating HfO films was developed and
tested. A 333Å HfO film was successfully deposited and
its C
-
V plots were extracted. The dielectric constant was
lower than expected. This could be due to the formation of
an interfacial layer. With the fabrication process verified,
HfO and SiO
2

films can be fabricated and characterized.
The high
-
k of hafnium will allow for fabrication of a high
gain transistor that is more resistant to ionic penetration.
These conditions make the transistor suitable for
interfacing with a neuron to record neural activity.


[1] D. Parent and E. Basham, "Hafnium transistor design
for neural interfacing," in Engineering in Medicine and
Biology Society, 2008. EMBS 2008. 30th Annual
International Conference of the IEEE, 2008, pp. 3356
-
3359.

[2] Miremadi, B.K., S.R. Morrison, and K. Colbow,
Stabilization of silicon
-
based devices in ion
-
containing
media using thin Al2O3 platelets and MoS2 oriented thin
films. Applied Physics A: Materials Science & Processing,
1995. 62(1): p. 39
-
42.

[3] Matsuno, H., et al., Biocompatibility and osteogenesis
of refractory metal implants, titanium, hafnium, niobium,
tantalum and rhenium. Biomaterials, 2001. 22(11): p.
1253
-
1262.

[4] Blanchin, M.G., et al., Structure and dielectric
properties of HfO2 films prepared by a sol

gel route.
Journal of Sol
-
Gel Science and Technology, 2008. 47(2):
p. 165
-
172.

The authors wish to thank Craig Stauffer for his assistance
in machining the components for the evaporator. Thanks to
Neil

Peters for support in the
Microelectronics Process
Engineering Laboratory (MPEL).

This work was supported
by Defense Microelectronics Activity Cooperative
Agreement # H94003
-
08
-
2
-
0806
-
SJSU.




HfO films for MOS capacitor shave been previously
fabricated but were damaged before characterization
4
.
This work seeks to fabricate HfO films for MOS capacitors
using improved equipment for better film quality. In
addition, film characterization via C
-
V, C
-
T, and TVS will
be performed.

Film

Growth


Synopsis

TCAD

software

was

used

to

model

the

physical

behavior

of

the

MOS

capacitor
.

The

target

oxide

thickness

for

both

the

hafnium

and

silicon

films

was

400
Å
.

A

1
-
D

TCAD

simulation

was

used

to

extract

the

N
A
,

V
T
,

and

Q
ss
.













Source

V
T

(V)

SiO
2

HfO

Hand Calculations

2.97

0.53

TCAD

2.76

0.69

To correct this, prior to evaporation, the boats were
pretreated by placing a small amount of hafnium in the
boat and bringing the temperature up to the melting point
of hafnium for a short period. This resulted in a hafnium
coating that had partially diffused into the tungsten and
reacted to form a barrier to preventing addition hafnium
from diffusing through during evaporation.




To deposit films a separate evaporator was built with
added components to increase the control over the growth
of the film:


Quartz Crystal Microbalance (QCM) to monitor the
thickness


Residual Gas Analyzer (RGA) to monitor gas
composition


Shutter control to control the amount of material being
deposited. This provides additional control of the thickness
of the film


Cooling jacket to act as a heat sink and protect the
gaskets in the system



The film was then annealed at 800
°
C. The resulting C
-
V
plot is shown below and the dielectric constant was found
to be 10.5.

To verify the procedure for thermally evaporating HfO in
the evaporator previously shown, a control sample was
created. A HfO film was grown on top of a <100>
orientation p
-
type substrate and a film of 333Å was
achieved. The EDAX spectrum was collected as shown
below.

The films were grown using thermal evaporation.
Tungsten boats were used for evaporating the hafnium.
However, the hafnium diffused through the tungsten and
caused breakage.

The total capacitance is made up of the series
capacitance of the HfO and the SiO
2
. Since the
ε
siO2

value
of the SiO
2

interfacial layer is small, it can reduce the
overall capacitance which in turn reduces the extracted
value for the dielectric constant. If an interfacial layer is
assumed, the dielectric constant can be calculated using
the equations below. Assuming an interfacial layer of 10Å
and an
ε
ox
of 3.9, the dielectric constant of the film was
found to be 18.





Capacitance versus Voltage (C
-
V): A plot of the
capacitance versus voltage will be used to extract the V
FB
,
V
T
,
t
ox
, and
Q
ss

parameters.


Methodology

To characterize the SiO
2

and HfO

MOS capacitors, the
following test will be performed:



Triangular Voltage Sweep (TVS): The TVS test is used
to determine the amount of sodium contamination present
in the MOS capacitor.



Capacitance versus Time (C
-
T): This test will monitor
the MOS capacitor’s ability to maintain its capacitance over
time in the presence of sodium .


The
ε
ox
value from the TCAD simulation was based upon
the previous work. The fact that the
ε
ox
of this experiment
is lower can be attributed to an interfacial layer
4
.



The film was then characterized using the C
-
T test to see
how well the capacitance is maintain over time when
exposed to a 0.5M saline solution.



The capacitance remained fairly constant over time for
different bias voltages. In addition, there appeared to be
no significant shift in the C
-
V curve before and after
exposure to the saline

solution. The conductance

of the film before and after

the C
-
T tests was also

plotted and it was noted that

there was no significant shift.