Controlling Mechanical Sensors with Light Research Program - - PowerPoint PPT Presentation

Controlling Mechanical Sensors with Light

Research Program

• Optically-Controlled MEMS
• Force Sensing
• Quantum Transduction
• MEMS Materials Research

McGill Optomechanics Lab Christoph Reinhardt, Simon Bernard Alexandre Bourassa, Xin Yuan Zhang Julian Self, Chris McNally Jack Sankey

Measuring Forces with Flexible Materials

Grütter Lab: forces from individual atoms (McGill)

Goal al: high-Q

• Q & tunab

able.

Rugar Lab: non-contact forces from virus nuclei (IBM)

One Pursuit: Replace Mechanical Materials w/ Light

Very weak restoring force from material Very strong restoring force from light

• laser-tuned frequency
• laser-tuned damping
• optically supported disc predicted

to ring for a month 0.75 mm ~1 um Fabrication at McGill

Implications for Sensing

Grütter Lab: forces from individual atoms (McGill)

Current technology ringdown wn ~ second nds

Rugar Lab: non-contact forces from virus nuclei (IBM)

High Mechanical Q = Better Force Sensor

Quantum Properties of Massive Objects?

Penrose: Gravity might ruin quantum superpositions of heavy objects.

• R. Penrose, Gen. Rel. Grav. 28, 5, 581-600 (1996)

For me: For me:

• Quantum state transfer and

entanglement at a distance?

• Fundamental studies of quantum

behavior in solid objects

photon

information transfer

“superposition”

Encouraging Result: Optically Trapped Pendulum

Kimble Group at Caltech:

• Optically trapped disc, Q ~ f 2
• Mechanical modes mix, limiting Q
• Q x f ~ 1011 (“low” initial Q)

K.-K. Ni et. al., PRL 108, 214302 (2012)

MIT ENS Vienna ENS

Solid Objects Controlled by Photons

kg g μg ng pg

nanotubes, BEC's, atoms... NIST JILA, Caltech Max Planck Vienna LIGO Yale NIST-JILA Caltech Laussane Oregon UCSB, Leiden Laussane-LMU Yale (this talk) UMichigan UCSB (not photons) n ~ 0.4 n ~ 0.01

One goal: make solid objects behave quantum mechanically.

n ~ 0.9 Caltech

Mechanical Displacement Frequency cavity resonances laser

Most Devices:

Optomechanical Systems Are Guitars: Same Physics

Electric Field ~ Zero (i.e. “clamped”) String Motion ~ Zero (clamped) Single Frequency: 200,000,000,000,000 Hz Speed of Light: 300,000,000 m/s Speed of Sound: ~500 m/s Single Frequency: 440 Hz Radiation Force Acoustic Force

When the Input Frequency Matches the Cavity Frequency

Energy Inside Cavity Mirror Position combined force of 1,000,000,000,000,000,000,000 photons striking the mirror each second ~milliwatt input (weak laser pointer) ~ micronewton ~ weight of 10 grains of salt ~ push a paper clip 1 mm in a few seconds (in space) ~ micronewton ~ weight of 10 grains of salt ~ push a paper clip 1 mm in a few seconds (in space) more light = more force less light = less force a spring! (plus “wind”) ~ 100 Watts circulating

Laser Engines

Energy in the Cavity (i.e. Radiation Force) Mirror Position Takes time for light to leak in and out cavity light pumps mirror motion

∫ F dx

work =

“Damping” for this Optical Spring

Radiation Force Mirror Position Power takes time to ramp up and down mirror motion pumps cavity light “Laser Cooling” Motion can be cooled to its quantum ground state

Our Device: “Linear” and “Quadratic” Coupling

• Couple laser to TEM00 cavity mode
• Cavity resonance frequencies vary

with membrane position

• Can choose linear or quadratic
• ptomechanical coupling
• Linear coupling within factor of 3 of

maximum possible

linear quadratic

• ptical node

antinode

At Yale: Laser Cooling in Cryogenic Environment

motorized membrane mount cavity

3He fridge

membrane free-space laser

• 50 nm thick membrane, 1.5 x 1.5 mm2,

261 kHz drumhead, Q = 5 Million

• System at 400 mK (~30,000 phonons)
• Free-space laser coupling
• Heterodyne readout of reflected light

Test: Laser Cooling to Low Phonon Numbers

• “Raw” laser cooling from 30,000 to ~60-

80 phonons, limited by classical laser noise

• Filtered laser cooling to ~20 phonons
• Next: monolithic, vibration isolation,

double-filtered laser, smaller membrane (should achieve < 0.01 phonons*)

laser filter cavity Unfiltered Cooling Laser

Theory: Kjetil Børkje alternative fitting method sideband ratio theory: filtered laser theory: unfiltered laser

now also commercially available

~30,000 phonons (i.e. 0.4K) with no lasers

linear

Quadratic Coupling: QND Phonon Readout

• Quadratic coupling: detuning measures

displacement squared

• Enables nondemolition readout of

phonon number state quadratic

Quadratic Coupling: QND Phonon Readout

Cavity Frequency Membrane Position Membrane Wavefunction average ground state average

• ne phonon

average two phonons

• If cavity response is slow, it will time

average the membrane's motion

• Each membrane quantum state has

a different average cavity shift

• Want: sharper curvature, more zero-

point motion, ... motionless

Quadratic Coupling: QND Phonon Readout

• Quadratic coupling: detuning measures

displacement squared

• Enables nondemolition readout of

phonon number state

• This coupling is too weak to measure

mechanical energy quantization. (ground state motion ~ fm) quadratic Totally symmetric coupling:

• cavity shift per phonon
• me

mechan anical shift per photon (enh nhanc anced optical al trap ap)

cavity photons membrane phonons

quadratic

Increasing Quadratic Readout Sensitivity

• Cavity supports many transverse
• ptical modes
• Transverse modes cross each other

Avoided Crossings Generate Stronger Quadratic Coupling

membrane aligned membrane tilted 0.4 mrad

• Membrane lifts degeneracies, couples modes
• Quadratic coupling > 100 x stronger
• Strong enough to resolve phonon shot noise of a

driven membrane

• Coupling is highly tunable in situ

laser cooling QND

4.5 MHz/nm2

30 kHz/nm2

St Strong cav avity

• ptical

al trap ap!

Optomechanics Lab at McGill

200 Hz linewidth!?

6 mo months 12 mo months right now Tables here, please.

New Types of Devices and Theory

SiN Fabrication (varied shapes, sizes, tethers, ...) Christoph Reinhardt Transfer matrix formalism

UHV Interferometer

Design & Fabrication Alexandre Bourassa

• 2” platter
• 1.5” travel
• UHV
• Bakeable

Technology stolen from Peter Grütter's lab (inside job): Yoichi Miyahara, Will Paul, Jeff Bates

Additional Directions

Cryogenic System

• 100x fewer thermal phonons
• Higher mechanical quality (generally)

Fiber Cavities

• compact, monolithic optomechanics
• no laser alignment required
• stronger coupling to smaller MEMS

cleaved, coated optical fiber 250-micron membrane

Harris Lab

death ray

One Goal at McGill: Improved Mechanics

COMSOL Mechanical Simulation Materials and Geometry

• improve untrapped Q x f
• Decrease optically-mediated mode

coupling

• Create geometries useful for AFM

Xinyuan Zhang

Another Goal: Stronger Per-Photon Impact

MEEP Simulation (cross-section of 3D model) Normalized Laser Frequency Normalized Transmission 99% reflective = 10 x more restoring force holey dielectric Julian Self Incident light 99% reflective Solgaard group

Summary

Optomechanics with a SiN Membrane

• laser cooling to ~20 phonons in a

He-3 cryostat

• Strong, tunable nonlinear coupling

for QND readout and trapping Building Stuff at McGill:

• Optically-Supported MEMS
• Force Sensing
• Hybrid quantum systems

McGill Optomechanics Group

Graduate Students Christoph Reinhardt Simon Bernard Undergrads Alexandre Bourassa Xin Yuan Zhang Julian Self Perry Philippopoulos Chris McNally

Acknowledgments (Yale)

Experiment Jack Harris (P.I.) Andrew Jayich Benjamin Zwickl Cheng Yang Donghun Lee Nathan Flowers-Jacobs Scott Hoch Woody Underwood Lily Childress Anna Kashkanova Andrei Petrenko Theory Steve Girvin (P.I.) Kjetil Børkje Andreas Nunnenkamp