Naval Research LaboratoryPlasma Physics DivisionWashington, DC 20375

11th HAPL MeetingNaval Research Laboratory

Washington, DC March 3-4, 2005

Work supported by DOE/NNSA/DP

KrF lasers

NRLJ. SethianM. MyersJ. GiulianiR. LehmbergS. Obenschain

Commonwealth TechF. HegelerM. Friedman

T. Albert

SAIC M. Wolford

RSIP. Burns

TITAN S. Swanekamp

MRC Albuquerque D. Rose Titan PSD, Inc

D. Weidenheimer

Opti Switch, Inc D. Giorgi

The goal for the Electra KrF Laser Program is to develop the technologies that can meet fusion energy requirements

for durability, efficiency, and cost

• Meet target physics and IFE requirements for laser beam uniformity and wavelength

• Wall plug efficiency > 6% (preferably > 7%)

• Operate at repetition rates up to 5 Hz

• Produce 400- 700 J of laser light/shot per pulse

• Run continuously for tens of thousands of shots

• Technologies developed must scale to IFE size system

KrF laser wavelength is 248 nm (1)

Target gain………………………………..

Imprint (standoff, bandwidth)………....

Laser-plasma interactions……………..

Target stability…………………………...

Laser efficiency (wallplug target)….

Optics damage/lifetime…………………

Integration complexity………………….

Systems optimum (COE?)……………..

?

?

Demonstrated beam uniformity on NIKE (0.2% spatial non-uniformity)

Straightforward zooming

Pulsed power based (low cost, industrial technology)

à laJohn Perkins

Electra title page

Electra’s main amplifier

Two-sided e-beam pumping: 500 kV, 100 kA, 140 ns FWHM

Pulsedpowersystem

Pulsedforming

line Diode

Laser gas recirculator

Operation of Electra’s main amplifier (oscillator)

Capacitor charging to +/- 43 kV(>160 ms)

Discharging through the1:12 step-up transformer

Charging of thePFL to 1 MV

(3-4 s)

Lasertriggered

switch

Matched diodeprovides

500 kV, 110 kA140 ns pulse

Diode

Magnet

Diode

Operation of Electra’s main amplifier (oscillator)

Velvet strip cathode

Ceramiccathode

still needsto be

partitionedinto

strips

Cathode generateselectron beam

e-beam

Laser GasKr + F2 + Ar

1 mil SSpressure foil

e-beam

e-beam

Electron beam ionizes laser gas

hibachi

Operation of Electra’s main amplifier (oscillator)

Laser cell

Laser gas recirculates(provides cooling

and quiescent flow)

e + Kr Kr + e

Kr + F KrF + F

KrF + h(248 nm) Kr + F + 2 h(248 nm)2

*- -

**

*

Electra produced between 620 - 710 J of laser energy using high efficient velvet strip cathode

Avg.: 690 J

Velvet strip cathode

e-beam

Hibachi

2004 results

intrinsic eff.: 9.7%

Electra’s oscillator achieved an intrinsic efficiency of 9.7% ...expect ~11% as an amplifier

As an IFE amplifier: KrF grade windows with AR coatingAmplification from input laser

Pulsed Power Advanced Switch 87%

Hibachi Structure No Anode, Pattern Beam 82%

KrF Based on Electra expt’s 11%

Optical train to target Estimate 95%

Ancillaries Pumps, recirculator 95%

Global efficiency 7.1%

> 6 % is adequate for fusion target gains > 100......and latest designs have 2D gains ~ 160

Based on our research, an IFE-sized KrF system is projected to have a wall plug efficiency of ~7% (meets goal)

Durability tests – more than 16,000 shots in 24 hours

• Thursday afternoon, February 24, 2005– 3740 shots @ 1 Hz (1 hour of continuous operation)

– Laser output: 300 J per shot» Used “low quality” windows with 83% transmission

• Friday morning, February 25, 2005– 1,352 shots @ 1 Hz

– 10,015 shots @ 1 Hz (2.5 hours of continuous operation)

– Laser output: 300 J per shot

– 200 shots @ 2.5 Hz

– 200 shots @ 5 Hz

– 500 shots @ 5 Hz (limited by radiation safety permission)

1 mil stainless steel pressure foils and ceramic cathodes are doing fine after 16,0000 shots; window transmission remained constant

Electra operated continuously for 10,000 shots (2.5 hours)

0

50

100

150

200

250

300

350

0 2000 4000 6000 8000 10000

shot number

lase

r en

erg

y (J

)

300 W laser output for 10,0000 shots @ 1 Hz

Data was recorded in 1000 shot intervalsand then processed.

Laser operated continuously for 10,015 shots

Lower laser energy caused by “misfires” and “mistiming”of the pulsed power prime and output switches.

This low “mistiming rate” of ~1% is unprecedented for aspark gap based pulsed power system.

Spark gaps require sufficient gas flowfor reliable operation

1 prime power switch for each side 2 output switches in parallel for each side

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0 50 100 150 200 250 300

time (ns)

ph

oto

dio

de

sig

nal

(V

) shot 1

shot 10015

Laser energy shows slight global decrease during the 10,000 shot run

to be investigated

0

50

100

150

200

250

300

350

0 100 200 300 400 500

shot number

sin

gle

sh

ot

ener

gy

(J)

Laser energy is stable for 500 shot run @ 5 Hz

“Misfires” and “mistiming”increase during 5 Hz operation because SF6 cart

does not have enough gas flow capacity.New SF6 cart should be operational late March 2005.

Last major challenge for the KrF laser system is durability.We are closing in on three technologies to address this

1. Foil durability: Primarily an issue of thermal management

3. Pulsed power: Achievable with new advanced solid state switch

2. Cathodes: Demonstrate patterned emission from our durable ceramic cathode leading to high efficiency

We are developing three techniques to cool the foil

• Convection cooling by the laser gasPro: Successful operation @ 1 Hz (foil

temp < 200°C)

Con: Foil temperature rises to 350-500°C @ 5 Hz operation

Foils

Rib Ribt = 0 ms

e-beam e-beam

gasflow

Louvers

Gas Velocity

t = 25 msFoils

Rib Ribt = 0 ms

e-beam e-beam

t = 75 mst = 125 mst = 0 ms(200 ms)

e-beam e-beam

t = 25 ms

• Mist cooling of the foil (developed by Georgia Tech)Pro: Successful demonstration @5 Hz (foil temp < 140°C)

Con: More complex system, lower e-beam transport efficiency

ElectronEmitter

RibRib

PressureFoil

e-beam

Laser GasKr + Ar + F2

AnodeFoil

He (air) watermist + film on

foils

ElectronEmitter

RibRib

PressureFoil

e-beam

Laser GasKr + Ar + F2

AnodeFoil

AnodeFoil

He (air) watermist + film on

foils

• Conduction cooling to hibachi ribs

(see poster by John Giuliani)Pro: Simple cooling system

Con: Requires close rib spacing, challenging e-beam transport through hibachi (may be reduced with advanced materials)

b =.300" a = .394"

X = 1.088"

P = 30 psi

H = 0.9-2.3W/cm2

.204

a f=.428"

.750"

.375"

f = 40º

r = .6

13

.187 dia

.234 r

f

.150.057 wall

b =.300" a = .394"

X = 1.088"

P = 30 psi

H = 0.9-2.3W/cm2

.204

a f=.428"

.750"

.375"

f = 40º

r = .6

13

.187 dia

.234 r

f

.150.057 wall

Convection cooling: Old louvers were replaced by “V”plate

Foils

Rib Ribt = 0 ms

e-beam e-beam

gasflow

Louvers

Gas Velocity

t = 25 msFoils

Rib Ribt = 0 ms

e-beam e-beam

t = 75 mst = 125 mst = 0 ms(200 ms)

e-beam e-beam

t = 25 ms

Old louvers were bending decreased gas flow velocity at foil

Foil temp reached 500°C(too high for long life operation)

Radiation cooling may play a role

Used in 2004

Stiffer louvers are under development

Currently in use

“V” plate mimics “ideal” closed louvers optimum gas flow velocity at foil

Rep-rated oscillator laser output was not influenced by fixed “V” plate

gas flow

Rib Rib

Gas Velocity

Foils

Developed a durable cathode using ceramic honeycomb

Cathode has numerous examples for durability

sides Hz shots target

1 1 5x10,000 anode plate

2 1 10,000 laser gas

1 5 7,800 Ar gas

1 5 2x1,700 laser gas

2 5 500 laser gas

Ceramic cathode must be patterned into strips for high efficiency

soft iron

e-beam

anode

Electron beam is compressed

Particle plot

Achieve higher current densitywithout changing the A-K gap

B-field is altered locally at the cathode

p

n+

n-

n++

p++

Diode Laser

D Laser

SiliconThyrsitor

CONCEPT:• All solid state• Diode lasers flood entire thyristor with photons• ultra fast switching times (< 100 nsec)• Continuous laser pumping reduces losses

PROGRESS:• > 1,200,000 shots (multiple runs)• Required specs (5 Hz, 16 kV, 1 kA/cm2)• Has run @ 50 Hz

shot 220,000vs

shot 1,000,000

topview

Newly demonstrated laser gated and pumped thyristor (LGPT) will be the basis for durable, efficient pulsed power

Power and Energy

Advanced Pulsed Power System based on LGPT

Electron beam

FastSolid State Switch

Marx

PulseForming

Line

MagneticSwitch

TransitTime

Isolator

This pulsed power architecture is being evaluated with thenew Electra front end

p

n+

n-

n++

p++

Diode Laser

D Laser

SiliconThyrsitor

The Electra Front Enda) Platform to develop advanced solid state pulsed power systemb) Provide 30-50 J of laser light input to the main amplifier

Has operated in e-beam mode @ 5 Hz, with very low jitter (<1000 ps)Pulsed power runs > 100,000 shots without maintenance

Pleasesee poster by

Matt Myers

The Electra Front EndVoltage and current diode waveforms

-200

20406080

100120140

160180

0 50 100 150 200 250 300

time (ns)

volt

age

(kV

)

-100

10203040

506070

8090

curr

ent

(kA

)

Next generation KrF laser

IOUT(MW/cm2)

10 x IIN(MW/cm2)

10 x PEB(MW/cm3)

I(right)

I(left)IASE

ISAT

Inte

nsity

(M

W/c

m2 )

Distance from rear mirror (cm)

15

10

5

0

t1 = 50 t2 = 450

Time (ns)2000 400 500100 300 2000 400100 300

15

10

5

0

t = 200 ns

e-beam

50 50 50 50mirror

3.4 meters

10 40 40 1040

Option 1

laser cell window100 cm

x100 cm

independentpulsed power

systems

laser light

flat-toplaser

energy:53.2 kJ

Higher e-beam pump power increases the intrinsiclaser efficiency as well as the foil heat loading

Diodes: 1 MV, 1.5 MA (total of both sides), 400 ns flat-top, 60 ns rise and fallPressure foil: 1 mil stainless steel

Diodecurrentdensity

Flat-toplaser

energy

Flat-topintrinsic

efficiency

E-beampumppower

Foilloadingfactor

Comparedto

Electra

Option 1 37.6 A/cm253.2 kJ 10.6% 627 kW/cc 9.2 W/cm2 2.2 times

Option 3 48.3 kJ 9.6% 314 kW/cc 4.6 W/cm2 1.1 times18.8 A/cm2

Option 2 55.3 kJ 11% 836 kW/cc 12.3 W/cm2 3.0 times50.2 A/cm2

All options have the same total e-beam energy deposition

Next generation KrF laser

Option 1ae-beam

50 50 50 50 mirror

2.2 meters

10 10

independentpulsed power

systems

laser light

IOUT(MW/cm2)

10 x IIN(MW/cm2)

10 x PEB(MW/cm3)

15

10

5

0

20

Time (ns)2000 400 500100 300

t1 = 50 t2 = 450

Distance from rear mirror (cm)1000 25050 150 200

t = 200 ns

I(right)

I(left)IASE

ISAT

15

10

5

0

20

Inte

nsity

(M

W/c

m2 )

flat-toplaser

energy:60.3 kJ

intrinsiclaser

efficiency:12%

Shrinking the unpumpedregions increases the laser output

(and intrinsic efficiency)

Shrinking the unpumpedregions increases the laser output

(and intrinsic efficiency)

Main Amplifier for the next generation KrF laser:50 kJ IFE beam line

Estimated cost of Pulsed power system:$30M (conventional)$50M (advanced switch)

Prime power tank

Pulse forming line

Magnetic Switch Gas Recirculator

Mirror

Laser Cell100 cm x 100 cm x 340 cm

Four electron beam generators per sideeach: 1 MV, 188 kA, 400 ns flat-top

Phase I goals are in sight with Electra

• Electra has produced 700 J/shot @ 1 Hz (700 W) and 400 J/pulse @ 5 Hz (2,000 W) in 100 second bursts

• Electra operated continuously and consistently for more than 2.5 hours: 10,000 shots @ 1 Hz (300 W)

• Oscillator results predict KrF intrinsic of efficiency ~11% for the Electra main amplifier

• Developed high transmission hibachi (>75% e-beam energy into gas)

• Demonstrated solid state switch, will be basis for durable, efficient (87%), cost effective pulsed power system

• Based on above, predict ~7% wall plug efficiency for IFE systems