Ex-situ Ohmic Contacts to n-ingaas

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 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 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 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 T b W e W bc T c f max 1 = τ in + RC 2πf τ = 8 π f τ ( Rbb Ccb)eff HBT: Heterojunction Bipolar Transistor Uttam Singisetti, DRC 2007 *M.J.W. Rodwell, IEEE Trans. Electron. Dev., 2001 June 24-26, 2009 University Park, PA Ashish Baraskar 4

InP Bipolar Transistor Scaling Roadmap Emitter: 512 256 128 64 32 width (nm) 16 8 4 2 1 access ρ, (Ω μm 2 ) Base: 300 175 120 60 30 contact width (nm) 20 10 5 2.5 1.25 contact ρ (Ω μm 2 ) f t : 370 520 730 1000 1400 GHz f max : 490 850 1300 2000 2800 GHz max - Contact resistance serious barrier to THz technology Less than 2 Ω-µm 2 contact resistivity required for simultaneous THz f t and f max * *M.J.W. Rodwell, CSICS 2008 June 24-26, 2009 University Park, PA Ashish Baraskar 5

Approach To achieve low resistance, stable ohmic contacts Higher number of active carriers - Reduced d depletion width - Enhanced tunneling across metal-semiconductor it interface Better surface preparation techniques - Ex-situ contacts: treatment with UV-O 3, HCl etch - In-situ contacts: no air exposure before metal deposition Use of refractory metal for thermal stability 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 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 In 0.53 Ga 0.47 As: Si (n-type) 150 nm In 0.52 Al 0.48 As: NID buffer Semi-insulating InP Substrate 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/Ti 0.1 W 0.9 - exposed dto air - surface treatment before contact metal deposition Contact metal 100 nm In 0.53 Ga 0.47 As: Si (n-type) 150 nm In 0.52 Al 0.48 As: NID buffer Semi-insulating InP Substrate 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 ~ 46(± 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 In 0.53 Ga 0.47 As: Si (n-type) 150 nm In 0.52 Al 0.48 As: NID buffer Semi-insulating InP Substrate June 24-26, 2009 University Park, PA Ashish Baraskar 10

Ex-situ contacts Ex-situ Ti/Ti 0.1 W 0.9 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* t *G. Stareev, H. Künzel, and G. Dortmann, J. Appl. Phys., 74, 7344 (1993). in-situ 100 nm ex-situ Ti 0.1 W 0.9 5 nm ex-situ Ti 100 nm In 0.53 Ga 0.47 As: Si (n-type) 150 nm In 0.52 Al 0.48 As: NID buffer Semi-insulating InP Substrate ex-situ 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 SF 6 /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 In 0.53 Ga 0.47 As: Si (n-type) 150 nm In 0.52 Al 0.48 As: NID buffer Semi-insulating InP Substrate 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 June 24-26, 2009 University Park, PA Ashish Baraskar 13

Error Analysis * 3.5 Error due to extrapolation 4-point probe resistance measurements 2.5 on Agilent 4155C 2 For the smallest TLM gap, R c is 40% of 1.5 total measured resistance 1 Metal Resistance 0.5 Minimized using thick metal stack Minimized using small contact widths Correction included in data Overlap Resistance Higher for small contact widths Resistan nce (Ω) 3 δr c δd δr 0 0 1 2 3 4 5 6 Gap Spacing (μm) *Haw-Jye Ueng, IEEE TED 2001 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 June 24-26, 2009 University Park, PA Ashish Baraskar 15

Results: Doping Characteristics Ac ctive carrier rs (cm -3 ) 10 20 7.5 10 19 T sub : 440 o C 10 18 10 18 10 19 10 20 Total Si atoms (cm -3 ) n saturates at high dopant concentration 3 ) Active Carriers (x 10 19 cm -3 7 6.5 6 5.5 5 [Si]: 1.5 x10 20 cm -3 4.5 400 410 420 430 440 Substrate Temperature, T ( o C) sub Enhanced n for colder growths -hypothesis: As-rich ihsurface drives di Si onto group-iii sites June 24-26, 2009 University Park, PA Ashish Baraskar 16

Results: Contact Resistivity Metal Contact Active Carriers (cm -3 ) ρ c (Ω-µm 2 ) In-situ Mo 6 x10 19 1.1±0.6 In-situ Mo 4.2 x10 19 20±11 2.0±1.1 Ex-situ Ti/Ti 0.1 W 0.9 4.2 x10 19 2.1±1.2 Mo contacts: in-situ deposition; clean interface Ti: oxygen gettering property June 24-26, 2009 University Park, PA Ashish Baraskar 17

Results: Effect of doping-i Con ntact Resi istivity (Ω Ω-μm 2 ) 16 5 14 12 active carriers 10 T sub : 440 o C 3 8 6 in-situ it Mo ex-situ Ti/Ti W 0.1 0.9 2 4 1 2 0 0 2 4 6 8 10 12 14 0 Total Si atoms (x10 19 cm -3 ) 4 19-3 Act tive Carri iers (x10 cm ) Contact resistivity ( c ) with in electron concentration June 24-26, 2009 University Park, PA Ashish Baraskar 18

Results: Effect of doping-ii esistivity (logρ ) c 100 10 Tunneling ρ C exp Thermionic Emission ρ c ~ constant * 1 N d * Contact R 1 1 10-10 2 10-10 3 10-10 4 10-10 [N d ] -1/2, cm -3/2 Data suggests tunneling. High active carrier concentration is the key to low resistance contacts *Ph Physics of fsemiconductor Devices, SMSze June 24-26, 2009 University Park, PA Ashish Baraskar 19

Results: Thermal Stability Conta act Resis stivity (Ω Ω-μm 2 ) Contacts annealed under N 2 3 flow for 60 secs Mo contacts stable to at least 400 C Ti/Ti 0.1 W 0.9 contacts degrade on annealing* 2.5 ex-situ Ti/Ti W 0.1 0.9 2 in-situ Mo 1.5 0 100 200 300 400 Anneal Temperature (C) *T. Nittono, H. Ito, O. Nakajima, and T. Ishibashi, Jpn. J. Appl. Phys., Part 1 27, 1718 (1988). June 24-26, 2009 University Park, PA Ashish Baraskar 20

Conclusion Extreme Si doping improves contact resistance In-situ Mo and ex-situ Ti/Ti 0.1 W 0.9 give low contact resistance - Mo contacts are thermally stable - Ti/Ti 01 0.1W 09 0.9 contacts degrade ρ c ~ (1.1 ± 0.6) Ω-µm 2 for in-situ Mo contacts - less than 2 Ω-µm 2 required for simultaneous THz f t and f max Contacts suitable for THz transistors June 24-26, 2009 University Park, PA Ashish Baraskar 21

Thank You! Questions? Acknowledgements ONR, DARPA-TFAST, DARPA-FLARE June 24-26, 2009 University Park, PA Ashish Baraskar 22

Extra Slides June 24-26, 2009 University Park, PA Ashish Baraskar 23

Correction for Metal Resistance in 4-Point Test Structure W R metal 1/ 2 ( ρsheet ρcontact ) / W ρ sheet L /W/ W L ( ρ sheet ρ contact ) 1/ 2 / W + ρ sheet L / W R metal / 3 From hand analysis & finite element simulation R metal / 2 Error term (-R metal /3) from metal resistance Effect changes measured ρ c by ~40% (@1.3 Ω-μm 2 ) All data presented corrects for this effect June 24-26, 2009 University Park, PA Ashish Baraskar 24

Doping Vs As flux cm -3 Dopi ing densit ty (x10 19 ) 5 4 3 2 T sub : 440 C 2 4 6 8 10 12 14 As flux (x10-6 ) torr June 24-26, 2009 University Park, PA Ashish Baraskar 25

Active Carrier, Mobility Vs Total Si 5 2200 (x10 19 cm -3 ) 4 3 Active Carrier 2000 1800 1600 Mobilit Active Carriers 2 1 Mobility 1400 1200 1000 ty (cm 2 /Vs s) 0 800 0 5 10 15 Total Si atoms (x10 19 cm -3 ) June 24-26, 2009 University Park, PA Ashish Baraskar 26

Strain Effects [Si]=1.5 x10 20 cm -3, n=6 x10 19 cm -3 (a =51x10-4 sub a epi )/a sub 5.1 Van de Walle, C. G., Phys. Rev. B 39, 3 (1989) 1871 June 24-26, 2009 University Park, PA Ashish Baraskar 27

Random and Offset Error in 4155C Resistan nce (Ω) 0.6647 0.6646 0.6645 0.6644 0.6643 Random Error in resistance measurement ~ 0.5 mω Offset Error < 5 mω * 0.6642 0 5 10 15 20 25 Current (ma) *4155C datasheet 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 ρ c ~ 40% at 1.1 Ω-µm 2 June 24-26, 2009 University Park, PA Ashish Baraskar 29