MADMAX: introduction and status · 3rd Workshop on Microwave cavities and detectors for Axion...

MADMAX: introduction and status

Chang Lee on behalf of the MADMAX collaborationMPI for Physics

3rd Workshop on Microwave cavities and detectors for Axion Research Aug 24th, Livermore, CA, USA

�1

MADMAX: Introduction and status C. Lee

3rd Workshop on Microwave cavities and detectors for Axion Research

Post-inflation axion

!2

Our Universe:OurUniverse

PQ symmetry breaking after inflationma ~ 100 μeV

ma < 10 meV



High-mass challenge

• 40—400μeV (10—100 GHz): Challenging for traditional resonant cavity due to smaller volume, lower Q

!3

preferred by post-inflation scenarios



• Inverse-Primakoff in matter

• Discontinuity of ε or Be generates propagating EM field.

Axion signal

!4

Ea = −gaγBe

εa

E∥,1 = E∥,2 H∥,1 = H∥,2 g H1

g E1

E1 a

k1

Region 1 e1 = 4

Be

Region 2 e2 = 1

k2

g E2

E2 a

Be

x

y

z

g H2

A. J. Millar et al JCAP01(2017)061

Ea2

Ea1

Eγ1

Eγ2

Be



signal power• Simplest: a metallic mirror (ε = ∞)

~12 photons / day / m2 (@ 25 GHz)

• Mirror + dielectric:Leaky resonator,boost factor:

• Add more dielectric disks…

!5

E0 = 1.3 × 10−12 [V/m] × ( Be

10T )

Q ∝ εβ = E/E0

=λ

4n

n = 5

β ~10

standing wave

propagating wave

P ∝ E2 ∝ β2


3rd Workshop on Microwave cavities and detectors for Axion Research !6

Mirror Dielectric Disks Receiver

Be

• more sources +coherent interferences

• Boost

β2

ν

more N!

Dielectric haloscope


3rd Workshop on Microwave cavities and detectors for Axion Research !7

Boost• β2 > 60,000 possible

with ~80 disks.

• Disk spacings tune β.

• Broad band for coverage

• higher β for confirmation

80 discs LnAlO3

50 MHzΔνν

= 2 × 10−3



Scan strategy• Area Law:

Psig x Δν is independent of disk spacings.

•

• Narrower peak leads to faster scan.

• In practice, ttot_adj ≈ ttot_scan.

!8

24.8 25.0 25.2 25.4 25.6

0

20

40

60

80

100

120

140

tscan

Δν= ( S

N )2

(kBTsys

Psig)

2

Psig ∝ B20 , β2, n, A, N,

1Δν

same area



collaboration

!9

MAgnetized Disk-and-Mirror Axion eXperiment DESY, Univ. of Hamburg, CEA-IRFU, MPI for Radioastronomy,

RWTH Aachen, Univ. of Zaragoza, MPI for Physics



Instrument

!10

4K

mirror

Cover the QCD axion @ 10—100 GHz



Simulation

!11

• Maxwell-Axion equation solved by analytic, FEM, ray tracing, and other methods.

• 1D calculations confirmed.

• Latest topics:3D effects, boundary loss, diffraction

dielectricdisk



Boost measurement• Boost factor is indirectly

confirmed by the reflectivity and group delay measurements.

!12

-0.5

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

0

0.05

0.1

0.15

0.2

0.25

0.3

14 16 18 20 22 24

Gro

up d

elay

[ns]

Refle

ctiv

ity

Frequency [GHz]

Reflectivity MeasurementReflectivity SimulationGroup Delay MeasuredGroup Delay Simulated

Boost Factor

Reflectivity

Group Delay(scaled)

0

80 discs1

0.5

5 disks

measurementby S. Knirck & J. Egge



receiver chain

!13

FFT

(signal generator)

• 10-23 W detected in a week



System temperature

!14

• Thermal emissivity of the internal parts to be measured.goal: ΔT < 0.5 K

• Bottleneck: ~5 K noise temperature from HEMT

• Quantum noise: ~0.48K @ 20 GHz

• Latest options (JPA, TWPA, …) to be considered

tscan

Δν∝ T2

sys, Tsys = Tbg + Tamp< 4K 5~6K



Mechanical design• Baseline design by

M. Matysek & D. Kittlinger

• 80 disks at cryogenic temp.

• Each moves 1.5—30mm w/< 60μm precision for β ~1000.

• Piezo motor for moving disks

• test @ cryogenic + high magnetic field

• Hysteresis, accuracy measurement at varying loads

!15

image by M. Matysek

μm



Dielectric study

!16

singlecrystal

dielectrics ε = n2 tanδ

Al2O3 10 10-5

LaAlO3 24 3 x 10-5

TiO2 100 3 x 10-5

1 Sapphire + mirror

for 1 mirror + 1 disk setup at resonance

βmax = 2n −1n



Magnets

!17

• Prototype magnets for R&D (2~3 years)

• 3-4 T, ~400mm bore

• Used magnets at Saclay under survey.

• Final setup (> 5 years)

• 10T dipole, B2A: ~100 T2m2, ~400 MJ stored in 2m x 1m2

• Two independent design studies by CEA Saclay & Bilfinger Noell GmbH

Noell Bilfinger

CEA Saclay

Preliminary



Site & infrastructure

• DESY offered HERA north hall (H1)

• Large supply of LHe for magnets

• Support for magnets’ weight (~150 tons)

!18

image from http://h1.desy.de



Prototype• Feasibility test & First scientific data

• starts in ~3 yrs

!19

Preliminary design

20 x Φ30 cm disks Mirror + motors

3~4 T



Future perspective

!20

Prototype

Full scale: 80 x 1m2 disks,10T magnet in 202x?



Summary• 40—400 μeV is an

interesting target for post-inflation QCD axion search.

• Dielectric haloscope is a promising technique.

• MADMAX is developing, aiming to cover the parameter space.

• PRL 118 091801JCAP01 (2017) 061MADMAX white paper

!21



Back-up slides



Boost• Disks + mirror boosts signal by

β = E / E0

• Transparent mode: δ = n x d x ν = π, 3π, 5π… constructive interference.

• Resonant mode: δ = π/2, 3π/2, …disks + mirror forms a leaky resonator.

• Combined boost from both contributions.

!23

mirror disk




Transfer matrix

!24

• Transfer matrix formalism w/ boundary conditions


E∥,1 = E∥,2 H∥,1 = H∥,2



list of show-stoppers

• Unexpected losses

• High tanδ, tilt, 3D loss, diffraction

• Components incompatible with high B field

• Mechanical precision too difficult.

!25

Incident Wave

Reflected / Transmitted Wave



Dielectric width

• For a set of dielectric disk with width d, how wide frequency can we cover?

• 1mm 15—30GHz

• 3mm 5—15GHz (LaAlO3)

!26



Fine-tuning search

!27

MADMAX: introduction and status · 3rd Workshop on Microwave cavities and detectors for Axion...

Documents

Transcript of MADMAX: introduction and status · 3rd Workshop on Microwave cavities and detectors for Axion...