Light shining through wall experiments - BCTPbctp.uni-bonn.de/bethe-forum/2016/axions/talks/Poeld.pdf · Light shining through wall experiments . Jan H. Põld ... > lock of both cavities

Jan H. Põld DESY

Bethe Forum: Axions and the Low Energy Frontier 9. March 2016

Light shining through wall experiments

Jan H. Põld | LSW experiments | 9.March 2016 | Page

Outline

Overview LSW experiments

> key elements

> sensitivity

> source, magnets, detector

> cavities as an improvement for LSW experiments

The ALPS experiment

> history

> layout

> current status and performance

> future efforts

2


What are the key elements of an LSW experiments?

> Source, magnet, wall, detector

> production and detection lab based

3Primakov effect


Sensitivity for ALP photon coupling

> probability for photon-ALP-photon conversion

> most effective improvements by increase of B and/or L

> coherent sources can be combined with resonators to raise the number of photons in front of the wall

4


Sources

> Sources: sub THz (microwave), optical (532nm and 1064nm lasers), xray sources, proton beam (talk by Babette on Friday about beam dump experiments)

5

CROWS STAX

ALPS I OSQAR

experiments at SPRING-8 ESRF

ALPS II

concluded under construction proposed


Magnets

> pulsed

> non-destructive: up to 60T and 15mm bore (MagLab)

> permanent

> e.g. 2.5T and 20mm bore (PVLAS)

> superconducting electromagnets

> 9 T and 50mm bore (LHC)

6


Detectors

> single photon detection

> high conversion efficiency

> low background rates

> Transition edge sensor (subTHZ,visible,IR)

> CCD (visible)

> heterodyne readout scheme (visible, IR)

> Germanium (Xray)

7

TES

Jan H. Põld | LSW experiments | 9.March 2016 | Page 8

Cavities as improvement for LSW

> resonator: arrangement of optical components which allows a beam of light to circulate

> resonator modes: field distributions which reproduce themselves after one roundtrip

> high power lasers are expensive and hard to get with requirements for a certain beam quality

8


0

0.2

0.4

0.6

0.8

1

-0.5 0 0.5 1 1.5

Nor

mal

ized

Inte

nsity

FSR

Finesse=10Finesse=50

How to characterize an optical resonator?

> Finesse describes the quality (losses) of a cavity

> free spectral range

> full width at half maximum

> Intensity

(circulating)


Pound-Drever-Hall locking

> published in 1983

> modulation/demodulation scheme to generate an error signal with a linear slope at the resonance

> feedback to the length of the cavity or the laser frequency possible


Production cavity

> increase number of photons in front of the wall

> high power buildup/ high Finesse

> in two mirror case a standing wave is produced inside the cavity

> only efficient with coherent source

> single frequency, single spatial mode source required

11


Regeneration cavity

> waist needs to be in the middle to have efficient coupling

> the physics is the same it is just harder to imagine

> higher mode density ==> higher reconversion probability

> similar to Purcell effect for atoms (predicted 1946, experimental verification 1989)

> three independent publications about this technique (Hoogeveen and Ziegenhagen (1990), Fukuda et al. (1996), Sikivie (2007))

12


Summary LSW experiments

> for a very sensitive LSW experiment you need

> strong magnets in a long string configuration

> low background rate detectors

> high power lasers and cavities

> sophisticated control systems

13


ALPS@DESY

14

> ALPS I (concluded 2010)

ALPS: Any light particle search

> ALPS IIa (currently being set up)

> ALPS IIc (2019)

> Best effort with existing magnets and facilities


ALPS I

> LSW experiment at DESY

> most sensitive LSW experiment at its time (recently surpassed by OSQAR)

> first experiment with production cavity

> used frequency doubled 1064nm laser

15

3.5·1021 1/s < 10-3 1/s

PLB Vol. 689 (2010), 149, or http://arxiv.org/abs/1004.1313

http://arxiv.org/abs/1004.1313


ALPS II

> LSW experiment with long baseline cavities in the HERA tunnel

> three orders of magnitude better than ALPS I

> surpassing CAST by a factor of three

> ALPS IIa is a pathfinder experiment to test the optical concept of ALPS IIc

16ALPS IIa (20m lab)

ALPS IIc in HERA tunnel


> Optics: high power cw 35W laser; light enhancement by high Finesse cavities

> Detector: Transition edge sensor ▪ very sensitive at 1064nm (in comparison to a CCD camera)

▪ Tungsten film kept at the transition to superconductivity at 80 mK; Sensor size 25µm x 25µm x 20nm

▪ dark background: 10-4 counts/second

> Magnets: 20 straightened, superconducting HERA dipole magnets (B=5.3 Tesla, LMagnet=8.8m)

Subsystems overview of ALPS II


Expected sensitivity

18

ALPS II


Cavity Parameters FSR≈7.5 MHz linewidth ≈ 0.9 kHz ROCin = 20 m ROCout= 20 m beam radius = 1.8 mm Tin = 750 ppm Tout= 750 ppm anticipated PB ≈ 1300 anticipated Finesse ≈ 4100

ALPS IIa

> current setup in the lab at DESY

> to maintain a high power buildup PDH lock needs to be stable and the input beam has to overlap with the cavity eigenmode

> cavity mirrors on two separate optical tables

> 10mW send into the cavity

PDH

DWS

EOMDWS

Piezo actuated mirror

20m PD trans

quadrant photo detector


Results of the cavity characterization measurements

20

> control signal displays the frequency noise of the cavity at low Fourier frequencies

> seismic noise is the dominant noise source

> common movement of optical tables at low frequencies

100 101 102 103 104 10510−4

10−2

100

102

104

106

Fourier frequency [Hz]

Freq

uenc

y N

oise

[Hz/

rtHz]

NPROerror signalcontrol signalseismic noise (geophone)


-60

-40

-20

0

10-3 10-2 10-1 100

mag

nitu

de (d

B)

frequency (FSR)

exact transfer functionfirst-order lowpass filter

Performance of the cavity: Pole frequency

> resonator characteristics: acting like a low pass with pole frequency at the half width at half maximum for power fluctuations on the input beam

> low pass approximation is good for a high finesse cavity


> power buildup: 891

> verified by independent ringdown measurement

> 230 ppm unexplained losses

22

frequency region of this measurement disappeared. This was presumably caused by thermaleffects and resonant higher order spatial modes. Considering the mirror transmission mea-surement we end up with 230 ppm of unexplained losses and a power buildup of 891. Thetheoretical throughput with 230 ppm of losses due to impedance mismatch is 98.66%. There-fore the higher order spatial mode content of the beam injected to the cavity is less than 4.16%.

-20

0

100 1k 10k

ma

gn

itud

e (

dB

)

Power modulation transfer function from upstream to downstream of the 20m cavity

measurementfit (f0 = 1184 Hz, F = 3166, FSR = 7.5 MHz)

-90

0

100 1k 10k

ph

ase

(d

eg

)

frequency (Hz)

Figure 1: Pole frequency measurement

5 Polarization dependence of cavity parameters

Turning the half waveplate in front of the cavity did not have any effect on the performance.

6 Length noise of cavity (spectra and time series)

The frequency noise measurement (see Fig. 2) is similar to what Reza and Marian measured.It is dominated by seismic noise for lower Fourier frequencies and by electronics noise forhigher frequencies. We will try to improve the signal-to-noise ratio and not be electronicsnoise limited any more.

2

Quality of 20m cavity


Extra losses due to scattering in cavity?


Auto alignment of input beam

24

101 102 103

−60

−40

−20

0

20

Mag

nitu

de [d

B]

Auto−alignment control loops

1x1y2x2y

101 102 103−180−135−90−45

04590

135180

Phas

e [d

eg]

Frequency [Hz]

PDH

DWS

EOMDWS


20m PD trans


> long term alignment of input beam with cavity eigenmode

> speed of the control loop is limited by Piezo resonances


Performance of automatic alignment

> alignment jitter on the input of a cavity produces power fluctuation in transmission of it

25

100 101 102

10−5

10−4

10−3

Frequency (Hz)

RIN

/ rtH

z

Auto−AlignmentNo Auto−AlignmentLaser Power Noise

100 101 102

10−4

10−3

Frequency (Hz)

RIN

/ rtH

z

Auto−Alignment EngagedNo Auto−Alingment EngagedNPRO

aligned not optimized


Status and next steps for ALPS IIa

> stable lock of the 20m cavity over the course of days -done

> control loops are running robustly -done

> characterization in vacuum -in progress

> reaching high power buildup and high circulating power in production cavity

> coherence measurements and noise mitigation

> implementation of central breadboard with aligned optics -in progress

> stable lock of production and regeneration cavity

> implementation and test of light connections and shutter concept

> integration of detector


Optical layout of ALPS II

27

Figure12.Layoutofthe

ALPS-IIb

andA

LPS-IIcopticaltables.

differentialwavefrontsensing

(DW

S)andPD

Hsensing

schemes

togenerate

errorsignalsforthe

–27

–


Optical layout of ALPS II

28

Figure12.Layoutofthe

ALPS-IIb

andA

LPS-IIcopticaltables.

differentialwavefrontsensing

(DW

S)andPD

Hsensing

schemes

togenerate

errorsignalsforthe

–27

–

> we have shown that our subsystems work independently, but we have not implemented the central part to ALPS IIa


How to control a cavity “without” light?

29

150kWPC RC

single photondetector

SHG

PDH

DWS

EOM

EOM

DWS

PDH

DWS

DWS



common optical cavity axis1064nm laser beam532nm laser beam

100m 100m

> different transmission coefficient for green light ==> less power buildup

> lock of both cavities has been tested in a 1m experiment at Hannover

> PDH lock and automatic alignment techniques the similar for PC and RC


• Positions of cavity eigenmodes defined by central optics

• Positioning optics with autocollimator • Must be better than 13.1µrad for ALPSIIa and

8.8µrad for ALPSIIc • Long term measurements look very promising

and meet the requirements

Accurate positioning of central breadboard mirrors


Seismic noise in HERA hall

31

• experiment should rest on the fundament of the HERA hall

• high seismic noise and long term drifts reduce effective free aperture of the beam pipe and eventually degrade the power buildup due to aperture losses


Limitations

> aperture size

> beam divergence angle

> intensity on mirrors

> wavelength

> actuator bandwidth

32

101 102 103 104

−20

0

20

Mag

nitu

de [d

B]

TF PZT z−axis ==> control signal of PDH control loop

101 102 103 104−180−135−90−45

04590

135180

Phas

e [d

eg]

Frequency [Hz]

measurement


Is it feasible what we are aiming for?

> scaling from 20m to 200m: reducing the aperture and moving the mirrors further apart

33

ALPS IIa 20m cavitySeptember 2015

ALPS IIa 20m cavityFebruary 2016

ALPS IIadesign sensitivity

ALPS IIcdesign sensitivity

plot from LIGO T-1400226-v6


Major challenges (for optics)

> thermal effects when going to higher circulating power (e.g. spatial higher order modes are resonant, change in ROC)

> lock both cavities with high bandwidth

> simultaneous, robust lock of production and regeneration cavity

> alignment of the beam in long baseline magnet string

34

150kWPC RC

single photondetector

SHG

PDH

DWS

EOM

EOM

DWS

PDH

DWS

DWS



common optical cavity axis1064nm laser beam532nm laser beam

100m 100m


Current timeline

> Formal approval by DESY in summer 2016?

> First results on hidden photons from a prototype experiment without magnets in spring 2017.

> Construction could start 2017.

> ALPS II would be finished in 2020.

35


Possible improvements beyond ALPS II

> ALPS III/ JURA

> requires dedicated magnet development

36

table by Axel Lindner


Conclusion

> ALPS II is the best approach for a next generation LSW experiment

> most vital subsystems have already been tested

> ALPS IIa is progressing well

> we are eager to see results of the first hidden photon run with ALPS IIa

37


(growing) collaboration

3838

ALPS II is a joint effort of

> DESY,

> Hamburg University,

> AEI Hannover (MPG & Hannover Uni.),

> Mainz University,

> University of Florida(Gainesville)

with strong support from

> neoLASE, PTB Berlin, NIST (Boulder), AIST (Japan).


Thank you for your attention

39


• 35W cw output power • 2W non-planar-ring-oscillator

(NPRO) and four stage Nd:YVO4 amplifier

• high fundamental mode content • low frequency noise due to NPRO • very reliable

High power laser


Differential wavefront sensing

• deviation of a beam with respect to a reference beam (e.g. resonator eigenmode)

ε =δxω0

$

%&

'

()

2

+δαΘD

$

%&

'

()

2


Performance of the cavity: Storage time

> cavity storage time via ring down measurements

42

−100 −50 0 50 100 150 200 250 300 350

−0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

time (µs)

Ampl

itude

(a.u

.)

Transmisson SignalExponential Fit τ = 134 µs

> fitted Finesse 3158

> F=t/(pi*FSR)


Accurate positioning of central breadboard mirrors

4 bewill/talks/15Alps_Cbfab_Aug

Autocollimator

� TRIOPTICS TA300-57 (accuray 0.75 arcsec)


32 day trend pitch (M9 mounted on grinded surface)


32 day trend yaw – with data gap projections


32 day trend pitch (M9 mounted on grinded surface)


32 day trend yaw – with data gap projections


shutter box

44


ALPs: Astrophysical hints

45

> Improve sensitivity for g, the ALPS-photon-photon coupling strength,by > 1,000.

The experiment measures a rate ~ g4:

> The experimental sensitivity is to increased by a factor 1012!

slide from Axel Lindner’s presentation


Sensitivity to hidden photons

46


Problems with old cavity mirrors

47

4

center

4

center

• 1000 ppm Losses in long cavity

• < 300 ppm Losses in short cavity

• Features size > 0.25 mm

• Rms surface fluctuations ~ 2.7 nm

> Insufficient power buildup in the PC

‣ 10 cm cavity with smaller eigenmode achieve full power build up

‣ Possible scattering from large features on substrate

> Surface profile measurement


Sensitivity of CROWS and x-ray experiments

48

Inada (2015) Betz (2014)

Light shining through wall experiments - BCTPbctp.uni-bonn.de/bethe-forum/2016/axions/talks/Poeld.pdf · Light shining through wall experiments . Jan H. Põld ... > lock of both cavities

Documents