Ultra Wideband DHBTs using a Graded CarbonDoped InGaAs Base

1
Ultra Wideband DHBTs using a Graded Carbon-Doped
InGaAs Base
Mattias Dahlström, Miguel Urteaga,Sundararajan
Krishnan, Navin Parthasarathy, Mark
Rodwell Department of Electrical and Computer
Engineering, University of California, Santa
Barbara
mattias_at_ece.ucsb.edu 805-893-8044, 805-893-3262
fax
2
UCSB
Wideband InP/InGaAs/InP Mesa DHBT
Mattias Dahlström

• Objectivesfast HBTs ? mm-wave power, 160 Gb
  fiber opticsdesired 440 GHz ft fmax, 10
  mA/mm2, Ccb/Iclt0.5 ps/Vbetter manufacturability
  than transferred-substrate HBTs
• Approach
• narrow base mesa ? moderately low Ccbvery low
  base contact resistance required
• ? carbon base doping, good base contact process
• high ft through high current density, thin layers
• Bandgap engineering small device transit time
  with wide bandgap emitter and collector

3
UCSB
DHBT Layer Structure and Band Diagram
M Dahlstrom
Emitter
InGaAs 3E19 Si 400 Å
Collector
InP 3E19 Si 800 Å
InP 8E17 Si 100 Å
InP 3E17 Si 300 Å
InGaAs graded doping 300 Å
Setback 2E16 Si 200 Å
Vbe 0.8 VVce 1.5 V
Base
Grade 2E16 Si 240 Å
InP 3E18 Si 30 Å
InP 2E16 Si 1700 Å

• 300 A doping graded base
• Carbon doped 81019?5 1019 cm-2
• 200 Å n-InGaAs setback
• 240 Å InAlAs-InGaAs SL grade
• Thin InGaAs in subcollector

InP 1.5E19 Si 500 Å
InGaAs 2E19 Si 500 Å
InP 3E19 Si 2000 Å
SI-InP substrate
4
InP/InGaAs/InP Mesa DHBTBase contact resistance
UCSB
Mattias Dahlström

• Carbon doping 6E19 cm-3
• Pd-based p-contacts
• Careful ashing and oxide etch
• RTP _at_ 300 C, 1 minute

The size of the base contacts must be minimized
due to Ccb
Pc is immeasurably low below 10 7
?cm-2 Critical for narrow base mesa HBT
5
 
 
InP/InGaAs/InP Mesa DHBTDevice Results
UCSB
Mattias Dahlström
b20-25
No evidence of current blocking or heating
J3.5 mA/um2
BVCEO7.5 V
 
6
UCSB
Accurate Transistor Measurements Are Not Easy
Miguel Urteaga Mattias Dahlstrom

• Submicron HBTs have very low Ccb
  Characterization requires accurate measure of
  very small S12
• Standard 12-term VNA calibrations do not correct
  S12 background error due to probe-to-probe
  coupling
• Solution
• Embed transistors in sufficient length of
  transmission line to reduce coupling
• Place calibration reference planes at transistor
  terminals
• Line-Reflect-Line Calibration
• Standards easily realized on-wafer
• Does not require accurate characterization of
  reflect standards
• CPW lines suffer from substrate TE, TM mode
  coupling thin wafer, use Fe absorber !lateral
  TEM mode on CPW ground plane present above
  150 GHz , must use narrower CPW grounds

Transistor in Embedded in LRL Test Structure
Corrupted 75-110 GHz measurements due
to excessive probe-to-probe coupling
7
InP/InGaAs/InP Mesa DHBTDevice Results
UCSB
Mattias Dahlström

• 2.7 mm base mesa,
• 0.54 mm emitter junction
• 0.7 mm emitter contact
• Vce1.7 V
• J3.7E5 A/cm2

ft 282 GHz fmax480 GHz b 25 BVCEO 7.5
V
8
InP/InGaAs/InP Mesa DHBTDevice Results
UCSB
Mattias Dahlström
Aej3.4 um2 J4.4 mA/?m2
Vcb0.9 V
fmax measurement above 500 GHzcurrently not
reliable in CPW environment

• Emitter contact sizes 0.5-2.0 um, 8 um
  long.
• Base extends 0.25-1.0 um on each side of
  the contact
• Maximum current density gt10 mA/um2

• Vce gt1.5 V for best performance
• Best ft found at current density of 3-5 mA/?m2

9
InP/InGaAs/InP Mesa DHBTConclusions
UCSB
Mattias Dahlström

• Doping-graded base InGaAs/InP Mesa DHBT
• High current density Operates up to
  10 mA/?m2 without destruction Kirk
  threshold 4.4 mA/?m2 at 1.5 V
  
• ft of 280 GHz with a 220 nm collector
• fmax is 450 GHz or higher
• Rbb is no longer a major factor - excellent base
  ohmics
• fmax no longer a good measure of Ccb or circuit
  performance
• Ccb reduction a priority
• 87 GHz static frequency divider circuit already
  demonstrated

10
UCSB
Narrow-mesa DHBTbase design
Mattias Dahlström
Many approximate methods for determining Ef such
as Boltzmann, Joyce-Dixon are insufficient
Energy (eV)
Base doping (cm-3)
Doping graded base At degenerate doping levels
(gt1E19) the variation of the Fermi level in the
base is very rapid Exponential doping roll-off
not needed, linear roll-off good enough!
11
UCSB
Narrow-mesa DHBTbase design
Mattias Dahlström

• Base transit time calculation
• Bandgap narrowing
• Fermi-Dirac statistics
• doping and bandgap dependent mobility

Transit time (ps)
Base width (A)
The exit term (electron velocity in top of
collector) important for thin bases use InGaAs,
not InP, close to base