Temperature Stabilized Measurements of Laser Spectra

Temperature Stabilized Measurements of Laser SpectraT. Flick, Wuppertal UniversityMini Opto Workshop4.-5. March 2010CERN

T. Flick, Temperature Stabilized Measurements of Laser Spectra 2

Overview•Introduction

▫Measurement purpose▫Measurement principle

•Setup•Performed measurements

▫Temperature behavior▫Spectra

•Status and future plans

05.03.2010


Introduction• In the innermost of the existing HEP detectors VCSEL need to

stand severe radiation environments• This will get worse with future experiments• Main damaging effects for lasers:

▫ Radiation Damages▫ Temperature effects inside the semiconductor material (at the

junction)• Mostly both effects come along together, but heat can be cooled

away.• This study is investigating the possibility to quantify a measure

and prepare an improvement possibility for the cooling.• Similar work has been investigated by Markus Axer (Jan,

Francois) for the CMS experiment and we inherit a lot from this work.▫ I will use several slides from him to explain the principle

05.03.2010


Wavelength Spectrum• The wavelength spectrum emitted by a laser diode is a perfect indicator of the device’s

internal temperature – the junction temperature Tj

• The wavelength spectrum is red-shifted when the device is heated by increasing the ambient temperature or the input power

• If a given cavity mode remains at the same wavelength, the junction temperature Tj must be constant

• The change in junction temperature due to varying the input power Pin to the laser can be cancelled by a change in the heat sink temperature, so as to keep the selected mode fixed in wavelength (nulling method Paoli method)

• The thermal resistance is found from the ratio of the change in heat sink temperature to the change in input power.

-60

-50

-40

-30

-20O

ptic

al O

utpu

t Pow

er [d

Bm

]

1295.81295.61295.41295.2

Wavelength [nm]

Data Gaussian Fit Peak Wavelength

-60

-50

-40

-30

-20

Opt

ical

Out

put P

ower

[dB

m]

130012961292

Wavelength [nm]

Typical wavelength spectrum of a Fabry-Perot type laser measured with an Optical Spectrum Analyzer

05.03.2010


L-I Characteristic Light-Current (L-I) characteristic of a non-irradiated laser at Tamb=20°C

IthDL

DI Eff=DL/DI

Thermal rollover

Threshold current Ith laser starts to emit coherent light

Efficiency Eff slope of L-I curve in linear part

Thermal rollover non-linear part of L-I curve where non-radiative recombination mechanisms (Auger) become dominant due to internal temperature

05.03.2010


Spectral Behavior during Irradiation

optjths

ambj

optin

ambj

diss

ambjth PVIRI

TTPPTT

PTT

R

=

=

=2

1298.4

1298.2

1298.0

1297.8

1297.6

1297.4

1297.2

1297.0

Pea

k M

ode

Wav

elen

gth

[nm

]

543210

20 MeV Neutron Fluence [1014n/cm2]

XP439D03, CD/DFluence = 0.24656D/DFluence = 0.090094D/DFluence = 0.061354D/DFluence = 0.049894D/DFluence = 0.042622

The behavior of certain mode peaks is unique for all LDs:

• “Slight” red-shift with increasing fluence at the same input current level

• “Large” red-shift when increasing the input current

10mA

25mA

45mA

55mA

•Rs is constant during irradiation term is mainly affected by I

•Ith increases during irradiation term is mainly affected by irradiation

Popt is affected by I and by irradiation

05.03.2010


Paoli Method

T0

T1

100%1%

-60

-50

-40

-30

-20

Opt

ical

Out

put P

ower

[dB

m]

1295.81295.61295.41295.2

Specific Mode Peak Wavelength [nm]

T0,DC0

-60

-50

-40

-30

-20

Opt

ical

Out

put P

ower

[dB

m]

1295.81295.61295.41295.2


T0,DC0

T0,DC1

-60

-50

-40

-30

-20

Opt

ical

Out

put P

ower

[dB

m]

1295.81295.61295.41295.2


T0,DC0

TX,DC1

-60

-50

-40

-30

-20O

ptic

al O

utpu

t Pow

er [d

Bm

]

1295.81295.61295.41295.2


T0,DC0

T1,DC1

Am

bien

t Tem

pera

ture

Input Pulse Duration

internal heating

exte

rnal

coo

ling

T1,DC1 = T0,DC0

T0,DC0

05.03.2010

1,0,, == xDCVIPwith xxxxin

10

00,11,

0,1,

)()()(

TT

TTTTPPR

jj

ininth

=

=


The Paoli MethodT0

T1

100%1%

-60

-50

-40

-30

-20

Opt

ical

Out

put P

ower

[dB

m]

1295.81295.61295.41295.2


T0,DC0

-60

-50

-40

-30

-20

Opt

ical

Out

put P

ower

[dB

m]

1295.81295.61295.41295.2


T0,DC0

T0,DC1

-60

-50

-40

-30

-20

Opt

ical

Out

put P

ower

[dB

m]

1295.81295.61295.41295.2


T0,DC0

TX,DC1

-60

-50

-40

-30

-20

Opt

ical

Out

put P

ower

[dB

m]

1295.81295.61295.41295.2


T0,DC0

T1,DC1

Am

bien

t Tem

pera

ture

Input Pulse Duration

internal heating

exte

rnal

coo

ling

T1,DC1 = T0,DC0

T0,DC0

05.03.2010

€

Rth =Tj,0 −T0Pin,0 − Popt ,0

≈Tj,0 −T0Pin,0

0,1,

1,

11, , jj

in

jth TT

PTT

R =

1000,11,0,1, )()()( TTTTTTPPR jjininth ==1,0,, == xDCVIPwith xxxxin


-60

-50

-40

-30

-20

Opt

ical

Pow

er [d

Bm

]

1310130813061304130213001298

Wavelength [nm]

Temp=30.04(°C)Current=-10(mA)Total Power=-10.42(dBm)

• Extraction of spectrum properties

-60

-50

-40

-30

-20

Opt

ical

Pow

er [d

Bm

]

1304.61304.41304.2

Wavelength [nm]

'Peak Maximum' 'Fit'

Temp=30.04(°C)

Cavity Mode-40

-35

-30

-25

-20

Gai

n E

nvel

ope

Opt

ical

Pow

er [d

Bm

]

1310130813061304130213001298

Wavelength [nm]

'Peak Maximum' 'Fit'

Temp=30.04°CMean=1304.3nmWidth=2.7274nm

Gain

The Paoli Method Step by Step05.03.2010

T. Flick, Temperature Stabilized Measurements of Laser Spectra

Thermal Effects during Irradiation

57

56

55

54

53Inpu

t Pow

er [m

W]

1310.21310.11310.0

Wavelength [nm]

-60

-55

-50

-45

-40

-35

Opt

ical

Out

put P

ower

[dB

m]

1310.21310.11310.0

Pin=52.8mWPin=55.0mWPin=56.9mW

24.8

24.6

24.4

24.2Am

bien

t Tem

p T a

mb

[°C

]

1295.71295.61295.51295.41295.3

Wavelength [nm]

-60

-50

-40

-30

-20

Opt

ical

Out

put P

ower

[dB

m]

1295.71295.61295.51295.41295.3

T=24.2°C T=24.5°C T=24.9°C

//

j amb inth

diss in opt amb

T T PTRP P P T

D DD= =

D D

• A parameter that describes the device’s efficiency to release heat generated inside the laser is called Thermal Resistance Rth

/ 0.09 /ambT nm CD D =

• D/DPin monitored during irradiation:

/0.0153 /

inPnm mW

D D=

0.0153 1700.09

C CmW W

= =

• D/DTamb measured in an oven:


Measurement Setup•In Wuppertal a similar setup as used by

Markus has been realized:▫DUT is kept in a thermally isolated box▫Cooling and heating capabilities are realized

using a Peltier element and a temperature control / regulation circuit

▫Optical fibres connected to an OSA (Yokogawa A6319)

▫Laser driving using external pulser / waveform generator and current source.

05.03.2010


Setup Schematic

05.03.2010

Current Source

Waveform Generator Temperature

Regulation

PCLabView Control Program

OSA

Thermal Enclosure

Peltier

DUT

Data Stream

Cooling

Spectra Mesaurement


Setup Pictures

05.03.2010

ThermalEnclosure

Waveform Generator

Spectrum Analyser

Current Source for

Laser


Temperature Studies• Different regulation

algorithms have been studied

• PID algorithm has been chosen to control the Peltier element

• Temperature regulation is very fast▫ O(few mins)

• Temperature remains very stable ▫ < ±0.05 °C

05.03.2010

22 min


Optical Measurements• OSA measurement time is

depending on the resulotion and span:▫ 1.5 - 25 s per measurement

(10 pm resolution)• Different analysis possible,

directly in the OSA or offline on the raw data

• Scan of temperature dependent spectra shows the wished behavior

• Red shift of the spectrum while warming the laser

05.03.2010


Red Shift vs Temperature• Monitoring 3 peaks from the

spectrum under temperature change

• Temperature range 10-30°C in 1°C steps

• Spectrum peaks change by ▫ 0.0780 nm/K▫ 0.0779 nm/K▫ 0.0786 nm/K

• Zooming into the range of 16-18°C measured in 0.1°C steps shows a jump

• It is not yet fully understood and needs further investigation

05.03.2010

Temperature [°C]

Temperature [°C]

Wav

elen

gth

[nm

]W

avel

engt

h [n

m]


Interesting Topic to Look at• Peak does not shift, but

more transforms into another

• Polarization effect?

05.03.2010

Wavelength [nm]

Inte

nsity

[dB

m]


Duty Cycle Dependency• First DC measurements

performed• Increase of wavelength

with introduced power• Error is RMS of the peak• Careful handling of the

peak error needed• Inclusion of this

measurement into the Paoli Method to be done

• This measurement shows the working principle only

05.03.2010


Status of the Setup and Further Plans• The measurement itself (Paoli Method) is automized• Temperature depending spectra and duty cycle (power) depending

spectra are taken automatically • Analysis tools are under investigation:

▫ Evaluate peaks▫ Fit the Gaussian▫ Extract the l shift and the gain curve▫ Conclude for thermal resistance▫ …

• Different types of optical components (simple diode, transmitter board, …) need to be implemented, but this is prepared already.

• Planned:▫ Laser package optimization studies▫ Test several different laser diodes (different materials, speed,

wavelength … compare properties)▫ Package optimization studies (heat coupling)▫ Irradiation

05.03.2010


Summary and Outlook• The setup used at CERN for CMS studies has been

reproduced in Wuppertal• First measurements have been taken• Spectra measured in dependence on temperature

and power have been performed• Measurement can be run automatically

• Analysis software is under way• More devices will be tested and the setup will be

qualified further• Will be used to qualify lasers afterwards

05.03.2010

Temperature Stabilized Measurements of Laser Spectra

Documents

Transcript of Temperature Stabilized Measurements of Laser Spectra