Transcript 2009_6_june_barkashar_EMC - University of California, Santa

High Doping Effects on In-situ and
Ex-situ Ohmic Contacts to n-InGaAs
Ashish Baraskar*, Mark A. Wistey, Vibhor Jain, Uttam Singisetti, Greg Burek,
Brian J. Thibeault, Arthur C. Gossard and Mark J. W. Rodwell
ECE and Materials Departments, University of California, Santa Barbara
Yong J. Lee
Intel Corporation, Technology Manufacturing Group, Santa Clara, CA
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
1
Outline
• Motivation
– Low resistance contacts for high speed HBTs
– Approach
• Experimental details
– Contact formation
– Fabrication of Transmission Line Model structures
• Results
– Doping characteristics
– Effect of doping on contact resistivity
– Effect of annealing
• Conclusion
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
2
Outline
• Motivation
– Low resistance contacts for high speed HBTs
– Approach
• Experimental details
– Contact formation
– Fabrication of Transmission Line Model structures
• Results
– Doping characteristics
– Effect of doping on contact resistivity
– Effect of annealing
• Conclusion
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
3
Device Bandwidth Scaling Laws for HBT
To double device bandwidth:
• Cut transit time 2x:
– Reduce thickness 2:1 ☺
– Capacitance increases 2:1 ☹
• Cut RC delay 2x
– Scale contact resistivities by 4:1
We
Wbc
Tb
Tc
1
  in  RC
2f
HBT: Heterojunction Bipolar Transistor
f
max

f
8    Rbb  Ccb eff
Uttam Singisetti, DRC 2007
*M.J.W. Rodwell, IEEE Trans. Electron. Dev., 2001
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
4
InP Bipolar Transistor Scaling Roadmap
Emitter: 512
16
256
8
128
4
64
2
32
1
width (nm)
access ρ, (m2)
Base:
300
20
175
10
120
5
60
2.5
30
1.25
contact width (nm)
contact ρ (m2)
ft:
fmax:
370
490
520
850
730
1300
1000
2000
1400
2800
GHz
GHz
- Contact resistance serious barrier to THz technology
Less than 2 Ω-µm2 contact resistivity required for
simultaneous THz ft and fmax*
*M.J.W. Rodwell, CSICS 2008
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
5
Approach
To achieve low resistance, stable ohmic contacts
• Higher number of active carriers
- Reduced depletion width
- Enhanced tunneling across metal-semiconductor
interface
• Better surface preparation techniques
- Ex-situ contacts: treatment with UV-O3, HCl etch
- In-situ contacts: no air exposure before metal
deposition
• Use of refractory metal for thermal stability
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
6
Outline
• Motivation
– Low resistance contacts for high speed HBTs and FETs
– Approach
• Experimental details
– Contact formation
– Fabrication of Transmission Line Model structures
• Results
– Doping characteristics
– Effect of doping on contact resistivity
– Effect of annealing
• Conclusion
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
7
Epilayer Growth
Semiconductor epilayer growth by Solid Source Molecular Beam
Epitaxy (SS-MBE)– n-InGaAs/InAlAs
- Semi insulating InP (100) substrate
- Unintentionally doped InAlAs buffer
- Electron concentration determined by Hall measurements
100 nm In0.53Ga0.47As: Si (n-type)
150 nm In0.52Al0.48As: NID buffer
Semi-insulating InP Substrate
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
8
Two Types of Contacts Investigated
 In-situ contacts: Mo
- Samples transferred under vacuum for contact metal deposition
- no air exposure
 Ex-situ contacts: Ti/Ti0.1W0.9
- exposed to air
- surface treatment before contact metal deposition
Contact metal
100 nm In0.53Ga0.47As: Si (n-type)
150 nm In0.52Al0.48As: NID buffer
Semi-insulating InP Substrate
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
9
In-situ contacts
In-situ Molybdenum (Mo) deposition
- E-beam chamber connected to MBE chamber
Why Mo?
- Refractory metal (melting point ~ 2623 C)
- Work function ~ 4.6 (± 0.15) eV, close to the conduction band
edge of InGaAs
- Easy to deposit by e-beam technique
- Easy to process and integrate in HBT process flow
20 nm in-situ Mo
100 nm In0.53Ga0.47As: Si (n-type)
150 nm In0.52Al0.48As: NID buffer
Semi-insulating InP Substrate
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
10
Ex-situ contacts
Ex-situ Ti/Ti0.1W0.9 contacts on InGaAs
• Surface preparation
- Oxidized with UV-ozone for 10 min
- Dilute HCl (1:10) etch and DI rinse for 1 min each
• Immediate transfer to sputter unit for contact metal deposition
• Ti: Oxygen gettering property, forms good ohmic contacts*
*G. Stareev, H. Künzel, and G. Dortmann, J. Appl. Phys., 74, 7344 (1993).
100 nm ex-situ Ti0.1W0.9
in-situ
5 nm ex-situ Ti
100 nm In0.53Ga0.47As: Si (n-type)
ex-situ
150 nm In0.52Al0.48As: NID buffer
Semi-insulating InP Substrate
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
11
TLM (Transmission Line Model) fabrication
• E-beam deposition of Ti, Au and Ni layers
• Samples processed into TLM structures by photolithography and liftoff
• Mo and Ti/TiW dry etched in SF6/Ar with Ni as etch mask, isolated by
wet etch
50 nm ex-situ Ni
500 nm ex-situ Au
20 nm ex-situ Ti
Mo or Ti/TiW
100 nm In0.53Ga0.47As: Si (n-type)
150 nm In0.52Al0.48As: NID buffer
Semi-insulating InP Substrate
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
12
Resistance Measurement
• Resistance measured by Agilent
4155C semiconductor parameter
analyzer
• TLM pad spacing varied from
0.6-26 µm; verified from scanning
electron microscope
• TLM Width ~ 10 µm
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
13
Error Analysis
3.5
3
Resistance ()
• Error due to extrapolation*
– 4-point probe resistance measurements
on Agilent 4155C
– For the smallest TLM gap, Rc is 40% of
total measured resistance
• Metal Resistance
– Minimized using thick metal stack
– Minimized using small contact widths
– Correction included in data
• Overlap Resistance
– Higher for small contact widths
2.5
2
dd
1.5
dR
dRc
1
0.5
0
0
1
2
3
4
5
Gap Spacing (m)
6
*Haw-Jye Ueng, IEEE TED 2001
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
14
Outline
• Motivation
– Low resistance contacts for high speed HBTs and FETs
– Approach
• Experimental details
– Contact formation
– Fabrication of Transmission Line Model structures
• Results
– Doping characteristics
– Effect of doping on contact resistivity
– Effect of annealing
• Conclusion
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
15
Results: Doping Characteristics
-3
Active Carriers (x 10 cm )
20
19
-3
Active carriers (cm )
10
10
10
19
Tsub : 440 oC
18
10
18
10
19
-3
10
Total Si atoms (cm )
n saturates at high
dopant concentration
2009 Electronic Materials
Conference
20
7.5
[Si]: 1.5 x1020 cm-3
7
6.5
6
5.5
5
4.5
400
410
420
430
Substrate Temperature, T
sub
o
440
( C)
Enhanced n for colder growths
-hypothesis: As-rich surface drives Si
onto group-III sites
June 24-26, 2009 – University Park, PA
Ashish Baraskar
16
Results: Contact Resistivity
Metal Contact
Active Carriers (cm-3)
ρc (Ω-µm2)
In-situ Mo
6 x1019
1.1±0.6
In-situ Mo
4.2 x1019
2.0±1.1
Ex-situ Ti/Ti0.1W0.9
4.2 x1019
2.1±1.2
• Mo contacts: in-situ deposition;
clean interface
• Ti: oxygen gettering property
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
17
-3
4
active carriers
12
10
Tsub: 440 oC
3
8
in-situ Mo
ex-situ Ti/Ti W
6
0.1
2
0.9
4
1
2
0
2
4
6
8
10
12
19
-3
14
19
14
0
Active Carriers (x10 cm )
5
16
2
Contact Resistivity (-m )
Results: Effect of doping-I
0
Total Si atoms (x10 cm )
• Contact resistivity ( c) ↓ with ↑ in electron concentration
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
18
Results: Effect of doping-II
c
Contact Resistivity (log )
100
10
Tunneling 
 1 

C  exp 
 N 
d 

Thermionic
Emission 
c ~ constant*
*
Data suggests tunneling.
1 -10
1 10
-10
-10
2 10
3 10
-1/2
-3/2
[N ] , cm
-10
4 10
d
High active carrier concentration is the key to low resistance contacts
* Physics of Semiconductor Devices, SM Sze
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
19
2
Contact Resistivity (-m )
Results: Thermal Stability
Contacts annealed under N2
flow for 60 secs
• Mo contacts stable to at
least 400 C
• Ti/Ti0.1W0.9 contacts degrade
on annealing*
3
2.5
ex-situ Ti/Ti W
0.1
0.9
2
in-situ Mo
1.5
0
100
200
300
Anneal Temperature (C)
400
*T. Nittono, H. Ito, O. Nakajima, and T. Ishibashi, Jpn. J. Appl. Phys., Part 1 27, 1718 (1988).
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
20
Conclusion
• Extreme Si doping improves contact resistance
• In-situ Mo and ex-situ Ti/Ti0.1W0.9 give low contact resistance
- Mo contacts are thermally stable
- Ti/Ti0.1W0.9 contacts degrade
• ρc ~ (1.1 ± 0.6) Ω-µm2 for in-situ Mo contacts
- less than 2 Ω-µm2 required for simultaneous THz ft and fmax
 Contacts suitable for THz transistors
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
21
Thank You !
Questions?
Acknowledgements
ONR, DARPA-TFAST, DARPA-FLARE
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
22
Extra Slides
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
23
Correction for Metal Resistance in 4-Point Test Structure
W
Rmetal
(  sheet  contact )1 / 2 / W
 sheet L / W
L
(  sheet  contact )1 / 2 / W   sheet L / W  Rmetal / 3
From hand analysis & finite element simulation
Rmetal / 2
Error term (-Rmetal/3) from metal resistance
Effect changes measured c by ~40% (@1.3 -m2)
All data presented corrects for this effect
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
24
5
19
Doping density (x10 )cm
-3
Doping Vs As flux
4
3
Tsub: 440 C
2
2
2009 Electronic Materials
Conference
4
6
8
10
12
-6
As flux (x10 ) torr
June 24-26, 2009 – University Park, PA
14
Ashish Baraskar
25
Active Carrier, Mobility Vs Total Si
2200
2000
Active Carrier
4
1800
3
1600
1400
2
1200
1
Mobility
0
0
2009 Electronic Materials
Conference
5
10
Total Si atoms (x1019cm-3)
June 24-26, 2009 – University Park, PA
Mobility (cm2/Vs)
Active Carriers (x1019 cm-3)
5
1000
800
15
Ashish Baraskar
26
Strain Effects
[Si]=1.5 x1020 cm-3, n=6 x1019 cm-3
(asub – aepi)/asub = 5.1 x10-4
Van de Walle, C. G., Phys. Rev. B 39, 3 (1989) 1871
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
27
Random and Offset Error in 4155C
0.6647
• Random Error in
resistance
measurement ~ 0.5
m
• Offset Error < 5 m*
Resistance ()
0.6646
0.6645
0.6644
0.6643
0.6642
0
5
10
15
20
25
Current (mA)
*4155C datasheet
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
28
Accuracy Limits
• Error Calculations
– dR = 50 mΩ (Safe estimate)
– dW = 1 µm
– dGap = 20 nm
• Error in ρc ~ 40% at 1.1 Ω-µm2
2009 Electronic Materials
Conference
June 24-26, 2009 – University Park, PA
Ashish Baraskar
29