The silicon detector of the muon g-2

experiment at J-PARC

Vertex 2011, Rust

June 24, 2011

Tsutomu Mibe (KEK)

for the J-PARC muon g-2/EDM collaboration

1

Particle dipole moments

Magnetic dipole momentg = 2 from Dirac equation, in general g≠2 due to quantum-loop effects

＝ ＋ ＋ ＋ …

Example : electron

Hamiltonian of spin 1/2 particle includes

2

Magnetic dipole moment

Electric dipole moment

a “anomalous magnetic moment”

Anomalous magnetic moment : g-2• Standard model can predict g-2 with ultra high precision

• Useful in searching for new particles and/or interactions

• Experiment has reached the sensitivity to see such effects...

0.24ppb

0.54ppm

Dal(SM)/al

4.5 ppb

0.41ppm

Dal(exp)/al

3

DHMZ, Tau 2010 workshop

aexp – a

SM = (296 ± 81) 10–11

(259 ± 81) 10–11

3.2~3.6 ”standard deviations“ To be confirmed by

new experiments

HLMNT,Tau 2010 workshop

Muon anomalous spin precession

in B and E-field

• Muon spin rotates “ahead” of momentum due to g-2 >0.

• Precession frequency

• BNL E821

– Focusing electric field to confine muons.

– At the magic momentum

g = 29.3, p = 3.094 GeV/c (a -1/(g2-1) ) = 0

4

c

EB

c

EaBa

m

e

h

g

21

12

c

EB

c

EaBa

m

e

h

g

21

12

Safely be neglected with current

upper limit on EDM

Continuation of the experiment at FNAL is planned.

Compact storage ring

– Suited for precision control of B-field• Example : MRI magnet , 1ppm local uniformity

• Completely different systematics than the BNL E821 or FNAL

Our approach

5

80 cm

Hitachi co.

14m

BNL E821 (FNAL ) J-PARC g-2

P= 3.1 GeV/c , B=1.45 T P= 0.3 GeV/c , B=3.0 T

Our approach (cont’)

6

c

EB

c

EaBa

m

e

h

g

21

12

Zero Focusing Electric field (E = 0 )

Ultra-cold muon beam (pT/p < 10-5) by utilizing the

laser resonant ionization of muonium makes it

possible to realize such experimental condition.

Equations of spin motion is as simple as at the magic momentum

BBa

m

e

h

2

BNL, FNAL, and J-PARC

7

BNL-E821 Fermilab J-PARC

Muon momentum 3.09 GeV/c 0.3 GeV/c

gamma 29.3 3

Storage field B=1.45 T 3.0 T

Focusing field Electric quad None

# of detected + decays 5.0E9 1.8E11 1.5E12

# of detected - decays 3.6E9 - -

Precision (stat) 0.46 ppm 0.1 ppm 0.1 ppm

Bird’s eye photo in Feb. 2008 8

9

10

Resonant Laser Ionization of Muonium

(~106 +/s)

Graphite target

(20 mm)

3 GeV proton beam

( 333 uA)

Surface muon beam

(28 MeV/c, 4x108/s)

Muonium Production

(300 K ~ 25 meV⇒2.3 keV/c)

11

Resonant Laser Ionization of Muonium

(~106 +/s)

Graphite target

(20 mm)

3 GeV proton beam

( 333 uA)

Surface muon beam

(28 MeV/c, 4x108/s)

Muonium Production

(300 K ~ 25 meV⇒2.3 keV/c)

Silicon Tracker

66 cm diameter

Super Precision Magnetic Field

(3T, ~1ppm local precision)

Injection, kicker and positron detector

12

Muon beam is injected here

mm

mm

Magnet coil

detectorkicker

Expected time spectrum of

e+nn decay

13

d=2E-20 e・cm

Up-d

ow

n a

sym

metr

y

g-2 precession spectrumParasitic EDM search

in up-down asymmetry

Time

High energy positron tends to be emitted in the direction of muon spin.

a

∝ED

M

Requirements

Detector should be efficient for Positron track with p = 200 - 300

MeV/c in 3T solenoidal B-field

Immune to early-to-late effect The decay positron rate changes by

two orders of magnitude.

1.6 MHz/strip 10kHz/strip for 200 um pich Silicon strip.

The positron detector must be stable over the measurements.

Zero E-field (

g-2 silicon tracker

• Tracking vanes made of Double-sided Silicon strip sensor– Anticipating excellent stability

and high granularity

• Number of sensors– 384 for 24 vanes*

• Number of channels– 0.2 mm pitch

– 288k for 24 vanes

• Detector area– 0.12 * number of vanes [m2]

– 2.9 m2 for 24 vanes

– * design studies in progress to determine these parameters

15

g-2 silicon tracker

576 mm

58

0 m

m

g-2 silicon vane

front back

The detector model• A GEANT4 model made of DSSD sensors (300m thick) has been

developed.– Dynamical Si response yet to be implemented (as discussed by Zbynek

Drasal on Wed)

• Track-finding performance is a key in the tracker design– Maximum ~10 tracks/10 ns

– Algorithm based on the Hough transform in “zf” plane is being explored.

16

Example event display

Signal e+ (>150MeV)

BG e+ (

Evaluation of DSSD sensor

17

HPK’s Belle-II DSSD sensor (discussed by Markus Fridel on Tue)

was used to evaluate timing response of the sensor.

A fast shaping ASD was wire-bonded to a part of strips (3x16 strip)

Bias

XY stage

ASD

ASD

p-side

Special thanks to

Toru Tsuboyama (KEK)

and Belle-II SVD group

Belle II Silicon Vertex Detector

Sensors from HPK

• Technical details (layers 4,5,6):

• Dimensions: 59.6 x 124.88 mm2

• p-side:

Readout pitch: 75 µm

768 strips

• n-side:

Readout pitch: 240 µm

512 readout strips n-side

Atoll p-stop scheme

21 June 2011 18Markus Friedl

First look at signal from sensor

• Full depletion above 60 V

• Well identified signals from 90Sr as well

as IR laser with a rise time of 10 nsec

• Plan to investigate timing response as a

function of bias voltage, instantaneous

rate, and temperature.

• Plan to perform a beam test at CERN in

collaboration with the SiLC.19

20 mV/div

40 ns/div

Test pulse (7fC)

IR laser (1060nm), n-side

Front-end electronics• Muon spill comes in every 40 msec.

We measure decay positrons for first 33 sec.

• Data acquisition sequence resembles to that of LC. The SiLC collaboration led by Aurore Navarro-Savoy (Paris) has been developing FEE for LC.

• R&D started to adopt the SiLC front-end technology to this experiment (French-Japan collaborative research program, 2011-2012).

20

Summary

• A new muon g-2/EDM experiment at J-PARC:– Off magic momentum

– Ultra-slow muon beam + compact g-2 ring

– Start in 2016

– Complementary to FNAL g-2

• Silicon tracker for g-2– Not quite a vertex detector, but a tracker for

incoming low energy positrons

– Stringent requirements on early-to-late effect, E-field and B-field

– Conceptual design and R&D are in progress21