Nonradiative recombination mechanisms in 2.37 m InGaAsSb GaSb

1
Non-radiative recombination mechanisms in 2.37 µm
InGaAsSb / GaSb
Kevin OBrien, Stephen Sweeney, Alf Adams,
Shirong Jin, Nasir Ahmad and Ben Murdin Advanced
Technology Institute , University of Surrey,
Guildford, GU2 7XH, UK A. Salhi, Y. Rouillard
and A. Joullié Centre dElectronique et de
Microoptoélectronique de Montpellier,
CEM2-Universite Montpellier II-UMR CNRS 5507,
case 067, 34095 Montpellier Cedex 05, France Y.
Cao, S. R. Johnson and Y.-H. Zhang MBE
optoelectronics group, Arizona State University,
Tempe, AZ 85287-5706, USA
2
Applications of MIR lasers

• Gas detection
• Pollution monitoring
• Medical
• Free-space . optical comms
• IR countermeasures

3
Laser Structure

• 3 x 10nm QWs
• 2.37µm emission
• 1.4 compressive strain
• Solid-Source MBE growth
• 1mm cavity length

4
Expected Threshold
In the absence of non-radiative processes, Jth
? as Eg? Jrad ? Eg2
1.5?m IAug/Ith 0.8
1.3?m IAug/Ith 0.55
Real devices Jth ? as Eg? Due to the presence
of a band gap dependent non-radiative process.
Auger Recombination
5
Expected Threshold
Trend of near-IR devices indicates that Jth of
the 2.37 µm lasers will be considerably higher
than Jth of the 1.5 µm devices
2.37?m
1.5?m IAug/Ith 0.8
1.3?m IAug/Ith 0.55
6
Actual Threshold Current
7
Actual Threshold Current

• Jth/QW _at_ RT for 2.37µm devices is
  42A/cm2
• Lower than for 1.5µm InGaAs/InGaAsP devices.
• Must have suppression of the band gap dependent
  process.

8
Measurements
Chalcogenide Optical Fibre
Pure Spontaneous Emission (SE)
Milled window in laser substrate
Facet Emission
Laser chip
I eV ( An Bn2 Cn3 )
9
Determination of dominant recombination process
I eV ( An Bn2 Cn3 )
Over a limited current range we may
write I ? nZ and the collected
spontaneous emission, LSE ? Bn2, hence n ? LSE1/2
? I ? (LSE1/2)Z a log-log plot
produces a graph with a gradient of Z ln
(I) Z ln (LSE1/2) So, with I ? nZ and Z
1,2,3 we can identify the dominant recombination
process at threshold.
10
Determination of dominant recombination process
Z 2
Lpin
T 102 K
11
Variation of Zth with temperature
Z 3 around RT and above indicating strong Auger
component
Majority of Ith at low T is radiative.
12
Radiative / Non-Radiative Contributions to Jth

• Jth of the 2.37µm InGaAsSb devices is
• 20 radiative
• 80 non-radiative _at_ RT
• Jth of the 1.5µm InGaAs devices is
  20 radiative .
• 80 non-radiative . _at_ RT

• No increase in proportion of Auger recombination
• from 1.5 µm to 2.37 µm devices

13
Auger Suppression
CHHL process still occurs!
CHSH process which is significant in 1.5µm
devices is not allowed!
Conduction
Conduction
Band
Band
Heavy Hole
Heavy Hole
Light Hole
Light Hole
Spin Orbit Split-off Band
Spin Orbit Split-off Band
CHHL
CHSH
14
Hydrostatic Pressure Measurements
15
Pressure Dependence

• CHSH increases as Eg ?so

• CHHL decreases as band gap increases

Data from Unipress group Adamiec et al, APL, 85,
4292 (2004)
16
Application of Pressure
CHHL is reduced as Eg increases and Auger
coefficient, C, decreases.
Heavy Hole
CHSH is activated as Eg increases with applied
pressure and approaches the spin orbit splitting
energy.
Light Hole
Spin Orbit Split-off Band
CHSH
17
Overview
Schematic of radiative current and dominant Auger
recombination current

• Jrad low
• Monomolecular recombination low
• CHSH suppressed
• CHHL reduced

CHSH
Moving to Sb-based system
J (current density)
CHHL
80
80
Jrad
20
20
2.37 µm
1.5 µm
?
18
Pressure dependence of Ith of 2.37 ?m and 2.18
?m lasers
Increase with pressure since Eg is larger. CHSH
occurring at atmospheric pressure
19
Summary

• Jth of 2.37 µm devices is lower than that of 1.5
  µm devices due to suppression of CHSH Auger
  process, since ?so gt Eg.
• However, the remaining Auger mechanisms continue
  and still dominate Jth at room temperature.
• Pressure dependence shows potential for QW laser
  approaching the ideal with this material system.