Micro-diffractive nanostructures for cold-atom manipulation

(and other interesting stuff)Groupe optique atomique et applications aux nanostructures

Université Paul Sabatier—CNRSToulouse, France

http://www.nanocold.cict.fr http://www.fastnet.fr

FRISNO-8Ein-Bokek, Israel

February 20-25, 2005

The basic idea

incoming laser light

output optical field

structured surface

cold atoms

Subwavelength StructuresFIB Fabrication

Arrays of Holes in Metal Films

Light transmission spectrum

Is this evidence of surface plasmons?

Decorated Slits and Holesin Subwavelength Ag

Membranes

Light transmission througha slit flanked by periodic

groovesdetected far-field transmission

calculated profile

Garcia-Vidal, et al., Appl. Phys. Lett. 83, 4500 (2003)

Measuring the transmission profile–atomic fluorescence mapping of the

field intensity

Relation between atomic fluorescence and field intensityexcited atoms field intensity

100 nm slit flanked by 10 grooves each side

measured

calculated

How does the slit/groove structureproduce the observed field

distributions?

Composite Diffracted Evanescent Wave

Lezec and ThioOptics Express, 12 3629 (2004)

Composite Evanescent Wave-I

Consider diffraction at a slit of width d. The field along x is given by:

00 0( , 0) Si Si

2 2E d dE x z k x k x

0

sin( )Si( )l tl dtt

Kowarz, Appl. Optics. 34, 3055 (1995)

-2 -1 1 2

-1.5

-1

-0.5

0.5

1

1.5

--2 -1 1 2

-1.5

-1

-0.5

0.5

1

1.5

= -4 -2 2 4

-2

-1.5

-1

-0.5

0.5

1

1.5

2

Composite Evanescent Wave-II

The phase shift of pi/2 isa signature of the CEW.0

0( ) cos2

E dE x k xx

-4 -2 2 4

-2

-1.5

-1

-0.5

0.5

1

1.5

2

x/d

CEWs launched on the surface

incoming laser light

output optical field

structured surface

A pi/2 phase shift between the directly transmitted wave and the

CEW

100 nm830 nm

50 nm

variablejog

incoming light

We can “jog” the structures tocompensate for the phase shift

00( ) cos

2E dE x k xx

Jogged and unjogged slit structures

unjogged jogged inward by ¼ period

Distribution and number of grooves controls the output optical field

far-field intensity, no phase shift

far-field intensity, with phase shift

We can produce a “phased array” with constructive interference in

the forward direction. The result is a forward “flame” .

Flame divergence vs. grooves—expmt. and model for unjogged grooves

10 grooves

30 grooves

calculation

Flame intensity vs. groove number for jogged and unjogged grooves

unjogged jogged

Flame angular distributions

incoming light

outgoing light

coupling

f

CCDmicroscopeobjective

d

/d f

Angular distribution of light vs. groove-slit spacing (in units of period N) for 5

grooves

Emission diagram (expanded)0°

-6°

3° -3°

6°

N = 1 (0.83 µm shift) N = 6 (4.98 µm shift) N = 24 (19.92 µm shift)

Intensity angular distribution—jogged grooves output side, 3 selected distances

Emission diagram (expanded)0°

-6°

3° -3°

6°

N = 1 (0.83 µm shift) N = 6 (4.98 µm shift) N = 24 (19.92 µm shift)

-1 -0.5 0.5 1

0.5

1

1.5

2

2.5

3

CDEW model measured

Intensity angular distribution—output side, unjogged red, jogged black

Emission diagram (expanded)N = 16 for both types of devices

0°

-6°

3° -3°

6°

-2 -1 1 2

1

2

3

4

CDEW model:Phase shift: CDEW/2

Index: nCDEW=850/830=1.024Relative field amplitude x=2/x

Angular lobe spacing

0 5 10 15 20 25

1

2

3

4

5

6

7

8

9

Grooves Shift (µm)

Subs

eque

nt lo

bes

spac

ing

(deg

rees

)

jogged

unjogged

CDEW model

0,5

1,5

2,5

3,5

4,5

5,5

6,5

7,5

8,5

9,5

0 5 10 15 20 25

Experiment

Next step: mirror MOT to get cold atoms close to surface

Next step: mirror MOT with the optical funnel generated by a planar

nanostructured phased arrayMirror MOT

atom trajectories

phased arraystructure

optical funnel

plane waveexcitation

Toward integrated structures:cold atom sources and transport

on a chip

Interaction atomes neutres-lumière

Champs proches optiques : confinement sub-longueur

d’onde de la lumière.

nanolithographie optique atomique

cohérente : diffraction

interaction dipolaire

Diffraction and confinement

Coupled resonant rings:symmetric/antisymmetric modes

Funnel effect: optical potential above the rings

f

e0

0ik

iErk

rE

Proposé et réalisé : J.V. Hajnal & G.I Opat (1989).Théorie : R. Deutschmann (1993), C. Henkel (1994).Expérience : Villetaneuse (1996), Orsay (1998).

surfk

*surfksurfk

surfk surfk2

)2cos(1),( 020 xkeVzxV surf

z

Réseau à onde évanescente stationnaire

surfk2surfk2

+1-1

ikiE

Réseau à onde évanescente nanostructurée :

xm Lm eQ 2

x

nm300L

période orientation motif

20

22 kmm k m

izm

mm eez lkElE ),(

Diffraction de l’onde évanescente :

A. Roberts & J.E. Murphy (1996)

msurfm Qkk

Autre approche : potentiel nanostructuré

D. van Labeke & D. Barchiesi

Periodicity controllable through angle of optical coupling-1

50 nm above surface 250 nm above surface

Periodicity controllable through angle of optical coupling-2

50 nm above surface 250 nm above surface

dB=5nmLx=Ly=250nmdiff=20mrade100nm=40°=55°

n=2.1 (TiO2)zr=250nm

MOT

Experimental parameters

Tales from the future—nanophotonics, addressable atom manipulation with optical arrays

E=10V/200nm E= 0.5 108 V/m

BaTiO3: n0=2.4 r42=1300pm/V

n=0.45 (20%)

SrRu4O3

BaTiO3

Quartz

Electro-Optic Beaming Controlwith variable index of refraction

AgV

Variable Depth Focusing

SrRu4O3

BaTiO3

Quartz

Electro-Optic Beaming Control

AgV

Steering

Change n from 0.8 to 1.2

Future: arrays of optical traps loaded with one or more atoms

125µm

R. Dumke, et al. PRL 89, 097903 (2002)

Summary—evolution of conception

Yesterday: Today:

10 µm

Tomorrow:

The GroupGuillaume Gay Renaud Mathevet

Bruno Viaris

Olivier Alloschery

Colm O’Dwyer

Gaetan Leveque