Sub-Micron InGaAs Esaki Diodes With Record High Peak Current Density D. Pawlik , M. Barth, P. Thomas, S. Kurinec, S. Rommel S. Mookerjea, D. Mohata, S. Datta S. Cohen, D. Ritter Device Research Conference: June 22, 2010 This work is partially supported by NSF (ECCS-0725760)
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Sub-Micron InGaAs Esaki Diodes With
Record High Peak Current Density
D. Pawlik, M. Barth, P. Thomas,
S. Kurinec, S. Rommel
S. Mookerjea, D. Mohata, S. Datta
S. Cohen, D. Ritter
Device Research Conference: June 22, 2010
This work is partially supported by NSF (ECCS-0725760)
D. J. Pawlik DRC: June 22, 2010 2/20
Key Metrics
Peak Current: JP = IP/Area
Valley Current: JV = IV/Area
Peak-to-Valley current ratio: PVCR
n+++ -ip++
-0.5 0.0 0.5
10-1
100
101
102
TD1
TD2
Cu
rre
nt
De
nsity (
mA
/m
2)
Voltage (V)
Record High
Current Density
JP
JV
JZ
VP VV
Esaki Tunnel Diode (ETD)
1
2
12
*
*
exp
exp
P d
A Dd
A D
P
J w
N Nw N
N N
J N
-
-
-
+
-
D. J. Pawlik DRC: June 22, 2010 3/20
Applications
Gated “p-i-n” Esaki diode
Operated in Zener direction
High JZ needed for drive current
Scaled to deep sub-micron dimensions
n++
Source
Drain
i
P++
Gat
eTunneling FET (TFET)
Multijunction Solar Cells Tunneling SRAM BJT & Contacts
website:www.sustainability.rit.edu/nanopower/
Word Line
VCC
Bit Line VSN
I
VCC
VSN
Word Line
VCC
Bit Line VSN
Word Line
VCC
Bit Line VSN
I
VCC
VSN
I
VCC
VSN
Tunnel
Diodes
E. Yalon, et Al., DRC 2010
Tunnel
Junction
D. J. Pawlik DRC: June 22, 2010 4/20
Scaling D. J. Pawlik, et al., ISDRS 2009
100
101
102
103
104
10
15
20
PV
CR
Mask Defined Area (m2)
PVCR(max)
PVCR(Ave)
Max. PVCR improved with area
What about current density?
D. J. Pawlik DRC: June 22, 2010 5/20
InGaAs Tunnel Diodes
MBE Grown at IQE
Higher level of doping than TD2
Contact layers: reduced doping
MOMBE Grown at Technion,
Haifa, Israel
Lower Doping than TD1
Contact layers; high doping
1. JTD1 > JTD2
2. RTD1 > RTD2
Predictions:
TD1 TD2
D. J. Pawlik DRC: June 22, 2010 6/20
200 nm
200 nm
Fabrication Process
Metal 1
Defined by E-Beam lithography
Au/Zn/Au (20nm/20nm/160nm)
Mesa Isolation Etch
Citric Acid : H2O2 (20:1)
~80 nm deep
~20 nm below the junction
D. J. Pawlik DRC: June 22, 2010 7/20
Device Area Characterization
Mask Defined Radius = 100nm
Extensive SEM analysis needed
Critical for accurate analysis
D. J. Pawlik DRC: June 22, 2010 8/20
(3) Contact pads Final Structure
(1) Metal and Mesa (2) BCB Etch back
200 nm
Junction
Location
Metal 1
(Gold)
40 m
Fabrication (Continued)
D. J. Pawlik DRC: June 22, 2010 9/20
-0.50 -0.25 0.00 0.25 0.50 0.75 1.00
10-5
10-4
10-3
10-2
10-1
TD2 (m2)
0.693
0.315
0.111
0.0461
TD1 (m2)
4.38
1.55
0.694
Cu
rre
nt
(A)
Voltage (V)
Electrical Characterization
Large area ETD as virtual ground
~25,200 m2
Current Scales with area
PVCRTD2 > PVCRTD1
JTD1 > JTD2
Ground Connection:
Preliminary Results
D. J. Pawlik DRC: June 22, 2010 10/20
64 128 256 512 10240.0
0.2
0.4
0.6
0.8
1.0
1.2
TD1
No
rmali
zed
Co
un
tJ
P (kA/cm
2)
TD2
Area Scaling & Current Density
Area scaling of IP JP Histogram
Current scales linearly with area
Deviations from trendline are
likely due to uncertainties in area
Devices measured: 200+ each
Gaussian fit used to determine JP
As predicted JTD1 > JTD2
10-2
10-1
100
101
10-5
10-4
10-3
10-2
TD1
P: 5 x 1019
cm-3
N: 5 x 1019
cm-3
TD2
P: 5 x 1019
cm-3
N: 5 x 1019
cm-3
slop
e = 2
.2 m
A/
m2
I P (
A)
Area (m2)
slop
e = 1
0 m
A/
m2
JP
(mA/m2)
s
(mA/m2)
s
(%)
TD1 9.75 90.2 9.3
TD2 2.10 17.5 8.3
Record high current density !!
D. J. Pawlik DRC: June 22, 2010 11/20
1 10 100
10-4
10-3
10-2
10-1
100
101
102
Franks, 1965
Eberl, 2001
Chung, 2006
Wang, 2
003Ismail, 1993
See, 2001
Rommel, 1998
Franks, 1965
Holonyak, 1960Richard, 1993
Smet, 1993
Day,1993
Tsai, 1994
Broekaert,1989
Cohen, 1995
Rommel 08
Rommel 08
Rommel 08
Rommel 08
This work
This work
Si Substrate
GaAs Substrate
InP Substrate
GaAs on Si
JP (
mA
/m
2)
PVCR
Richard, 1993
Holonyak, 1960
Area Scaling & Current Density
D. J. Pawlik DRC: June 22, 2010 12/20
Current Density vs. N*
*
*
exp
1exp
P d
P
A D
A D
J w
JN
N NN
N N
-
-
+
D. J. Pawlik DRC: June 22, 2010 13/20
-0.50 -0.25 0.00 0.25 0.50 0.75 1.0010
-1
100
101
Cu
rre
nt
(mA
/m
2)
Voltage (V)
4.38 m2
0.694 m2
~ 8
RS introduces voltage shift
Effects: Apparent (false) reduction in JZ
Smaller currents result in smaller shifts
Smaller junctions have are effected by RS less
Series Resistance (RS)
2.5x~0.3 V
D. J. Pawlik DRC: June 22, 2010 14/20
10-5
10-4
10-3
10-2
0.00
0.25
0.50
0.75
TD2
VP = 1.65() x I
P + 0.137(V)
TD1
TD2
VP (
V)
IP (A)
VP = 7.84() x I
P + 0.377(V)
TD1
Series Resistance: Constant
Types of RS:
(1) Constant (RC)
Provides linear change in
VP(Meas) with respect to
RC is the slope of VP vs. IP
(2) Area dependent (rArea)
( ) .
( .)
AreaP Measured P P Const
P Meas C P P A P
V V I RArea
V R I J V
r
r
+ +
+ +
y = m * x + b
D. J. Pawlik DRC: June 22, 2010 15/20
0 5 10 15 20
-150
-100
-50
0
50
100
V -1 (2
.5 m
A)
V -1
(-30 mA)
V -1(-5 mA)
V -1(-2
50
A)
V -
1 (800
A)
Slope = m (1/V-m2)
VI-1
Area (m2)
V -
1 (15
0
A)
-0.50 -0.25 0.00 0.25 0.50 0.75
10-5
10-4
10-3
10-2
10-1
84 nm
127 nm
189 nm
297 nm
518 nm
746 nm
1,053 nm
1,545 nm
2,029 nm
2,505 nm
V5
V4
V3
V2
Cu
rre
nt
(A)
Voltage (V)V1
-30 -20 -10 0 10
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
TD2
1/s
lop
es (
V*
m2)
Current (mA)
TD1
Series Resistance: Area Dependant
For a constant current (IC);
RC constant V shift
rA variable V shift
TD2: I-V Curves Constant Current Curves Specific Resistivity