What could you do with a particle accelerator on-a-chip?

What could you do with a particle accelerator on-a-chip?Advances in Dielectric Laser Accelerators and Prospects for Applications

Gil TravishUCLA Department of Physics & AstronomyParticle Beam Physics Laboratory&Radius Health

Presented at Varian Medical on 12SEP11

> 10 million new cancer cases each year> 50% of cases require radiation therapy

Current treatment issues:Immediate and long term side effects

Targeting of tumorsHandling and containment of radiation

High capital costs

what happens if we shrink this all down by

1000 ?

Our long term goal is to develop a mm-scale, laser-powered, disposable, relativistic particle source

>10m

<10mm

x1000

Large Application Space:

Industrial

• Petroleum Exploration

• Non-Destructive Testing (NDT)

• X-ray Photolithography

Medical

• Cardiology

• Veterinary

• Medical Imaging

Defense

• Package screening

• Stand-off detection

The conceptual design encapsulates the accelerator as a fiber optic tip

Fiber

(to

lase

r)

End

osco

pe e

nd

Enc

losu

re

MAP

Tum

or

Par

ticles

MAP technology irradiates the tumor and avoids healthy tissue

external beam

minimally invasive

vs

Bringing the source closer to the site allows electrons to deposit energy over a tumor depth

Electron Range and Stopping Power in Soft Tissue

soft tissue density = 1 g/cm3

1-6 MeV electrons have ~1-3 cm rangeminimal stopping power = minimal surrounding tissue damage

Reducing a therapy machine’s size may alter the economics of oncology treatment

BigHeavyExpensive ($3-5M)Microwave powered

Microchip sizedEndoscope tipDisposableLaser powered ($200k)

EBRT Machine MAP

What

magic scalingare we using to make this happen?

Breakdown limits scale favorably with wavelength and dielectric materials support high fields

10-15

10-13

10-11

10-9

10-7

10-5

10-3

10-1

100

102

104

106

108

1010

1012

1014

Pu

lse

Le

ng

th [

s]

Frequency [Hz]

GHz THz IR-VISfs

ps

ns

us

Conventional RF

DWA

L

A

S

E

R

Du (1996)~GV/m

T-481

Breakdown LimitsConventional Structure

Eacc ~ Prf/λ

in metals...

Of available power sources at wavelengths shorter than microwaves, lasers are the most capable

lack of sources, materials and fabrication technology force us to make a leap from Microwave to Optical

300MHz 3GHz 30GHz 300GHz 3THz 30THz 300THz

Visible  and  UV…Radio  Frequency Terahertz

Mobile  phones

Satellite  TV Medical  &  industrial  lasers

Terahertz  gapGround  to  satellite

The choice of accelerator technology impacts the size and nature of the beam produced...

RF Optical

Gradient

Energy gain per period

Repetition Rate

Charge per Bunch

Bunch Length

10-100 MeV/m 1-10 GeV/m

1 MeV 1 keV

100 Hz 10-100 MHz

0.1 - 1+ nC 0.01-1 pC

1-100 ps 1-100 fs

key: charge and time scale; not gradient

McGuinness

Eacc ~ Prf/λBreakdown limits metal:

Optical structures naturally have sub-fs time structures and favor high rep. rate operation

3.3 fs charge capture< 1 fs

//

Fill Time ~ 1 ps Fill Time ~ 1-5 ps

Optical Cycles

Laser Pulse

femtosec

picosec

//Emitter Pulse

nanosecEmission Time ~1 ns

100-1000 ns(1-10 MHz)

Macropulse

Micropulse

Laser Cycle

A variety of optical-scale dielectric structures are under consideration

MAP Logpile Grating

~2.5 GV/m

PBG-fiber-based structures afford large apertures and length-scalability

Planar structures offer beam dynamics advantages as well as ease of coupling power

The MAP structure consists of a diffractive optic coupling structure and a partial reflector resonator

laser light

Gun

sub-relativistic relativistic

ShortBraggStack

Coupler

TallBraggStack

VacuumGap

electron beam

MicroAcceleratorPlatform

The three functional parts of the MAP—gun, low energy section and relativistic section—form one system

➊ Integrated “gun”

➌ Relativistic structure

➋ Low beta structure

Field Emitter

Bragg FieldDeflector

relativistic structure ➌

The design of the relativistic structure is mature and includes realistic material properties.

Ez = E0 cos(! z c) !

laser

gap (1 optical wavelength)

cot kz ! "1 b " a( )#

$%& = kza ! "1 ! !

For gap a anddielectric b-a

idealizedresonance:

Tuning: control “matching” layer (b-a).

The MAP is a moderate-to-low Q structure which matches well with existing laser technology

Parameter Value

Laser wavelength 800nm

Cell length 800nm

Effective gradient 1.5 GeV/m

Quality factor Q 800

Effective shunt impedance R 2000 ohms

Effective shunt impedance per unit 2.0833 ohms

R/Q 2.5 ohms

R/Q per unit 0.0026 ohms

Transit factor 0.86

Stored energy 0.9 mJ

Power dissipation <1% (0.75 MW)

Fill time 0.5 ps

Laser intensity 100 MW

Laser pulse length 1.8ps

Energy gain per unit cell ~2.5keV

Simulations show energy spectrum clean-up; and, include acceleration, beam dynamics and material properties.

Resonant Fields (@ t = 7 ps)Incident laser

y(m)

x(m)

Ex (V/m)

t(s)

t(s)

Ex (V/m)

Ex (V/m)

Input laser sourcecan correspond to actual Ti:Al2O3 laser

Energy Distributions Energy Gain

low energy structure ➋

at 1 GeV/m, each period only produces 1KeV1000 periods only yields 1 MeV

Creating a sub-relativistic MAP is hard:the coupling and periodicity are one and the same

tapered structure

periodicity variation

two-color operation

DTL-like Solutions

periodicity skipping

Thick Glass Substrate

!!/"

!! 2!

laser light

β

z  (mm)

0.3

1

0.65

0 0.5 1

rapid change in velocity

The accelerating field may die off before the

particle fullly dephases

2.4 µm = 10βλ

β=0.3e-beam

λ=800 nm incident laser

Electrons’ energy gains outweigh losses, resulting in net acceleration

We have a viable design for the low energy accelerating structure.

0

-0.5

+1.0

0 1 2 µm

100 keV energy gain over~250 microns

0.4 GeV/m Accelerating Gradient

We have convincing evidence of energy gain for low-beta electrons in simulations

25 ke

V

125 k

eV

25 ke

V

125 k

eV

deph

asing

&

wall lo

sses

particle source (gun) ➊

A micro patterned Ferroelectric crystalcan be used as an integrated gun

Metalized, Grounded Scintilator: Electron (Camera) & X-ray Detection

Metalized LiNbO3 Crystal:Micro-Emitter(s) in Center

Grounded Heat Dissipater

TEC Heater

Additional Copper Heat Sink

Test Rig for Field-Enhanced Emissionfrom LiNbO3 Crystals

Before After Brightness - Background

Field overlay plot showing the magnitude of the electric field

An isolation region is able to control the fields inside the emitting region

FE Control Scheme

Integration of a gun into the low beta structure completes the source.

cont

rol

field

isol

ator

emitt

er

With field isolation, we can use any emitter

These pyroelectric emitters are pretty nifty so we decided to try to use them on their own.

We have spun off this pyroelectric + emitter technology to produce a flat panel x-ray source

!"#$%&'%()*'%*+,*

-!%."')/),'&+(')(,#$$',

'&#**#+%)#%.#0!(+,

1!2',3)"'4'"

'&#**#+%)!,'!

MAX: Microemitter Array X-rays

...with this.

Tube alone = US$ 25k (CRP) Target Price to OEM = US$20k

By replacing this …

Replacing vacuum tubes reduces manufacturing and maintenance costs.

We have generated arrays of prototype scale & small area images

Shaped Copper Mask

Cockroach

Where are we at with dielectric laser accelerators?

Thousand period structures—mm long and ~1 MeV gain—are now being produced.

Full scale coupler DBR

96.2nm

96.2nm

92.4nm

130.8nm

134.6nm

Structure  Dimension:  300nmX250μmX1000

287.6nm

+

DBR+Coupler

Fusion bonding of the top and bottom slabs seems to be the most promising integration method.

1. DBR deposition on optical substrate 2. Patterned SiO2 deposition for gap support

4. Build matching layers and DBR

3. Release layer Cr and coupling layer fabricated on Si substrate

5. Bond the two pieces together by fusion bonding

+

We have begun a ß=1 MAP beam de/acceleration experiment at SLAC’s E163

E-163 PMQs

structure causes energy loss (dE/dx) to misaligned particles

no structure at IP

IPspectrometer

J. England

1’’ 1mm250 µm

800 nm

Glass dummy structure

1 mm

Slot in dummy structure (not to scale)

Aluminum holder for glass structure

e-beam

A “dummy structure” and mount was designed for beam transmission studies.

Bunches from NLCTA Beamline

Spot size = 96 x 83 µm2

εx = 43 µm-radεy = 24 µm-rad

electrons that lost energy while traveling through glass

electrons that made it through slot

Spectrometer Image (higher energy to the left)

‣Theoretically, we expect peaks to be separated by 0.5 MeV

‣With calibration of 1.776 KeV/pixel, we find separation of 0.337 MeV

For the first time, beam was transmitted through the optical-scale structure!

Data analysis is ongoing

Just one more thing...

an all optical light source

A MAP-based undulator structure has been designed

For E=3 GV/m,Beqv=10 Tesla

Undulator Period = Laser Phase Flip

E-field

…………

λu >> λlaser

waveplate

It is possible to have an all-laser-powered x-ray source using optical accelerator structures...

... but compromises must be made

low energy+

optical undulator=

QFEL

high energy+

conventional undulator=

FEL but long

predictionA particle accelerator “on a chip”, capable of

producing intense pulses of relativistic electrons and x-rays will be widely available

in 10 5 years

Funding:NNSADTRAUCLADOE

Acknowledgments

Team:Rodney YoderJianyun Zhou (Postdoc - Fabrication)Josh McNeur (Grad - Simulations)Esperanza Arab (Staff - Engineering)Several past and present students...