Eye safe solid state lasers for remote sensing and coherent laser radar

Jesper Munch, Matthew Heintze, Murray Hamilton, Sean Manning,Y. Mao, Damien Mudge and Peter Veitch

Department of PhysicsThe University of Adelaide

Adelaide SA 5005Australia

jesper.munch@adelaide.edu.au

How to make a laser sensor “Eye-safe”

Keep energy/power low

Low power laser,Large transmittedbeam

Low DutyCycle

Select wavelengthfor maximum allowedpulse energy

Content of talk

• Coherent eye-safe laser radar: Review of current work in Er:Yb:glass slab lasers

• Planned work in Er:Yb:YAG

• New composite slab laser design

• Eye-safe sensing at low power

Our chosen Eyesafe laser species is Erbium

• Erbium lases at 1.5 – 1.6 μm, where laser safety allows:• 10× the energy per pulse allowed at 2 μm• 100x the energy per pulse allowed at 10 μm

• Allows better spatial resolution (for otherwise similar conditions)

• Can make use of available telecommunications photonic components: eg Master fiber oscillator

• BUT: it is a 3-level laser, normally in a phosphate glass host

Er:Yb energy level diagram

8 ms

95-99%

(1130 nm)

(1100 nm)

2H11/24S3/2

4F9/2

4I9/2

4I11/2

4I13/2

4I15/2

Laser(1535 nm)

Erbium

500 sμ2F5/2

2 ms

976nm PUMP

Ytterbium

2F7/2

1% ThermalPopulation

Limiting Processes:Ytterbium Bleaching11/2 Upconversion13/2 Upconversion15/2 Depletion

Energy Transfer

Summary of Early work in Adelaide*

• Demonstrated first injection seeding of single frequency Er:glasslaser at 1.5μm

• Demonstrated successful transform limited coherent Doppler measurement at 1.5μm

• Initial wind sensing measurements

*A. McGrath, J. Munch, G. Smith, P. Veitch,Appl. Opt. 37 (29), 5706-5709 1998.

Coherent Laser Radar

Transmitted Pulse &Reflected Signal

SLAVE LASER

DETECTOR 2

DETECTOR 1: Transmitted frequencyDETECTOR 2: Received power/frequency ADC & Signal

Processing

First injection seeded Er:glass at 1.5μm

Er:glass rodintra-cavitytelescope

AOM

Masteroscillator outcoupler

Littrowgrating

heterodynedetector

Q-switch

Expanded output pulse

Local oscillator beam

Tx/Rx telescope

PBS

λ/2 plate

λ/4 plate

Faraday isolators

λ/2 plate

The injection seeded, Q-switched laser produced a transform limited linewidth

0

0.5

1

1.5

2

2.5

3

3.5

0 0.2 0.4 0.6 0.8 1 1.2

time (µs)

Out

put m

onito

r sig

nal (

arb.

uni

ts)

0

5

10

15

20

25

30

15 20 25 30 35 40 45 50 55

Frequency (MHz)

Pow

er (a

rb. u

nits

) FWHM = 1.5 MHz

We used the Er:Glass laser to make a Doppler velocity measurement of moving hard-targets

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

15 20 25 30 35 40

Frequency (MHz)

Pow

er (a

rb. u

nits

)

00.020.040.060.08

0.10.120.140.160.18

15 20 25 30 35 40

Frequency (MHz)

Pow

er (a

rb. u

nits

)Frequency shift = -4.5MHzReceding velocity 3.5m/s

Frequency shift = -8.5MHzReceding velocity 6.5m/s

Second Generation Er:Yb:Glass Slab

• Robust laser design

• Folded, total internal reflection, zig-zag slab

• Diode laser side-pumping (Q-CW)

• Injection seeded, Q-switched ring

• Long output pulse, using new resonator design with efficient out-coupling via throttled Q-switch

Standing-wave Er:Yb:Glass slab laser

0

Collimating lens

0

Laser slab

Diode laser0

Output 2

Output 1

R = 97.8 %

R = 97.4 %Flat Mirror

Flat Mirror

0 100 200 300 400 500 600

0

10

20

30

40

50

60

Tot

al m

ultim

ode

outp

ut p

ower

(mJ)

Input pump power (mJ)

Setup Output power

Side-pumped laser head

Heat Sink

TEC

Diode Array

Collimating lens

Pumped regionof slab

Slab

Injection seeded ring resonator

0

0

0

Laseroutput pulse

Pockels cell(Q-switch)

Image relay lenses

Max R mirrors

PZT

Slab

Diode laser

Master laserin

Ring Oscillator Q-switch results

Pump

OutputVolta

ge o

n Ph

otod

iode

outp

ut p

ulse

(V)

Volta

ge o

n P

hoto

diod

epu

mp

puls

e (V

)

Time (sec)

-4

-3

-2

-1

0

1

2

3

4

5

6

0.00 0.02 0.04 0.06 0.08 0.10-1

0

1

2

3

4

5

Q-switched pulse (E=3mJ/pulse) Q-switch pulse – expandedscale: ms

Gain switched lasing(E=8mJ/pulse)

0 2 4 6 8 10-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

-4

-2

0

2

4

6

8

0 2 4 6 8 10-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

-4

-2

0

2

4

6

8

Time (sec)

Volta

ge o

n Ph

otod

iode

outp

ut p

ulse

(V)

Volta

ge o

n P

hoto

diod

epu

mp

puls

e (V

)

Pump

Output

0 2 4 6 8 10-4

-2

0

2

4

6

8

0 2 4 6 8 10-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

0 2 4 6 8 10

-4

-2

0

2

4

6

8

0 2 4 6 8 10-0.5

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Time (sec)

Volta

ge o

n P

hoto

diod

eou

tput

pul

se (V

)

Volta

ge o

n P

hoto

diod

epu

mp

puls

e (V

)

Pump

Output

Current results with Er:Glass

• Good long pulse energy in standing-wave oscillator, near TEMoo(50mJ)

• Q-switched ring oscillator demonstrated

• Injection seeding demonstrated

However :However :

Problems with current Er:Glass slab laser

• Energy output limited by Er bleaching (measured)• High intra-cavity losses in ring oscillator (Pockels cell)• Serious thermal lensing limitations • Optical damage of glass host• Currently max energy per pulse Q-switched is 10mJ/pulse,

but need 20-50mJ/pulse raw laser output for scalable systems (eg: larger aperture, system losses)

• Pulse repetition rate will be limited by thermal effects• Pumping limited by frequency chirp in diode-lasers used

Continuing effort in Erbium

Two parallel approaches:

1. Improve and optimize Er:Yb:glass subject to its inherent thermal limitations.• Experiments using different Er, Yb concentrations for optimum

pumping• Reduce resonator losses• Complete injection seeding characterization as laser radar

2. Investigate third generation Er:Yb:YAG

Third generation: Er:Yb:YAG at 1.645μm

• Greatly improved thermal properties of YAG host• Better control of thermal lens• Better efficiency (lower level has 2% population)• Scalable to higher power, rep. rate• Manufacture as ceramic YAG material• Permits use of our new end-pumped composite slab geometry• Experience from our successful Nd:YAG designs directly

relevant• But requires a new, single frequency master oscillator• Recently demonstrated in bulk Er:Yb:YAG*

* Georgiou & Kiriakidi, Opt. Eng., 44 Jun. 200580mJ output, pumped by 4.7J

High pump intensities and necessary cooling of the gain medium leads to strong thermal gradients which cause undesirable effects.

Issues• strong thermal lensing

- change from top/bottom cooling to side cooling

• thermally induced birefringence- use specialized pump distribution

Scaling to higher power slabs

Effect of pump profile on depolarization loss in Nd:YAG

-4.0

-2.0

0.0

2.0

4.0

1.0 0.9 0.8 0.7 0.6

Tophat pump profileGaussian pump profile

Intensity Transmission

x [m

m]

-4.0

-2.0

0.0

2.0

4.0

1.0 0.9 0.8 0.7 0.6

Tophat pump profile Gaussian pump profile

Intensity Transmission

x [m

m] Pump

region

(Birefringence modeling: M. Ostermeyer)

Effect of pump profile on depolarization loss in Nd:YAG

-4.0

-2.0

0.0

2.0

4.0

1.0 0.9 0.8 0.7 0.6

Tophat pump profile Gaussian pump profile

Intensity Transmission

x [m

m]

-4.0

-2.0

0.0

2.0

4.0

1.0 0.9 0.8 0.7 0.6

Tophat pump profile Gaussian pump profile

Intensity Transmission

x [m

m] Pump

region

(Birefringence modeling: M. Ostermeyer)

Zones ofstrong depolarization

Composite end-pumped, side-cooled folded zigzag Nd:YAG slab

Silicon Dioxide

Brewsterangledwindow

Laser mode

TOP VIEW

SIDE VIEW

GlassUndoped:YAG

Nd:YAG (0.6 at.%)

Undoped:YAGGlass

Silicon Dioxide

PumpingCooling

AR 0.808 mHR 1.064 m

μμ

Off-axis zigzag pumping

• Rectilinear zigzag duct allows pumping at normal incidence andmixes pump light prior to slab entry

• Can pump using fibers by collimated bar-stack-array, and use non-imaging lens duct

• Scalable by increasing pump power, height of doped and undopedregion (mode volume)

Opticalfibres(2D array)

rectilinearzigzag duct

Optic axis of pump sourceOptic axis +θOptic axis -θ

• Tophat pump distribution – minimum birefringence

• Good absorption efficiency due to quasi end-pumping

• More uniform power loading within slab due to double-clad structure transporting pump light along slab before absorption

• No hard-edged apertures in vertical direction

• Large pump input aperture and acceptance angle accommodates real divergent pump sources

• Insensitive to pump beam-quality due to mixing of pump light in slab

• Undoped YAG layers produce reduced thermally induced stress

• Conduction-cooled

Composite slab advantages

End view of conduction-cooled laser head

Coolant

TECHeatsink

Heatsink

Cu

Cu

Indiumcontact

Nd:YAG

YAG

YAG

40 60 80 100 120 140 160 18010

15

20

25

30

35

40

45

Slope efficiency 29.5%

Out

put P

ower

[W]

Pump Power [W]

Initial Laser Performance in Nd:YAG

Approximately 90% pump light absorption in end-pumped slab

Composite slab design for Er:Yb:YAG

• Ceramic

• Doped and undoped Er:Yb:YAG

• Doping concentrations easily changed

• Slab configuration based on success with Nd:YAG

Er:Yb:YAG laser radar system

• New master oscillator under development– NPRO (non-planar monolithic ring oscillator)– Ceramic Er:Yb:YAG– To be developed in collaboration with Innolight

• Injection seeded slave ring oscillator

• Ceramic composite slab slave as described

The DIAL program(DIAL = differential absorption lidar)

• Aim: Low-Cost profiling of water vapour up to top of boundary layer

• Provide water vapour concentrations for

– Quantitative precipitation forecasting, Bushfire danger assessment, fog prediction

– current technique - radiosondes, high recurrent cost, infrequent data

• 830nm GaAs diode lasers (mature technology)

– Single mode limited to ca 0.5W (Average power ca 0.5mW - eyesafe!)

– Detector technology well developed (low-noise single photon)

• Wavelength control

– On-line laser (master oscillator) stabilised to peak of water resonance

– Off-line/ On-line difference frequency stabilised to 15GHz

– Water resonance ~ 6GHz width @ sea level ~ 1GHz width @ 4km altitude

– Freq. stability of ~ 20MHz adequate

Setup for DIAL

Spectral properties of amplifier

Wavelength control of master lasers

On-line laser stabilisation– BLUE LOOP

Wavelength difference stabilisation- GREEN LOOP

Water resonances near 829nm

• accessible for diode lasers

• appropriate line intensity (10-23cm-1)

• sufficiently isolated from other resonances

• other lines at 832nm

Stabilization to water resonance (832nm)- error signal at lock-in output

V

wavelength

Conclusion

• Er:glass at 1.53 μm is a useful approach for a simple, low average power eye-safe coherent laser radar, but is limited by thermal effects and damage in glass.

• Er:Yb:YAG is a promising new, preferred option at 1.6μmDesign experience form Nd:YAG directly transferable

• Low cost alternatives to eye-safe incoherent sensing for short range (<4km) applications using shorter wavelengths are feasible.

Producing a tophat pump distribution

• How? – Use a composite slab (doped & undoped YAG layers)– End-pumped for good efficiency– Side-cooled zigzag slab

Pump absorption is a tophat profile, thus minimizing thermally induced birefringence loss (even though diode-laser pump profiles typically produce Gaussian transverse profiles)

Thermal lensing minimized by using a zigzag mode-path in the plane of cooling, and by controlling the heat flow in the orthogonal plane

Small-signal gain measurement proves bleaching of Erbium

Gai

n

Proportional to Pump Energy (mJ)

Gai

n

Proportional to Pump Energy (mJ)

1

1.2

0.8

0 65 260 578