Crystal Technologies for LC - The California Institute of Technology

Crystal Technologies for LC

Ren-yuan ZhuCaltech

International Workshop on Linear ColliderSitges, Barcelona, Spain

May 3, 1999

� Design Considerations for Calorimeter at LC.

� What Can Crystal Calorimetry Offer?

� Possible Crystal Technologies for LC:

– Oxides: from BGO to PbWO � ;– Halides: from CsI to PbF � .

Design Considerations for Calorimetr y at LC

� Precision measurement of electrons and photons: e/ � re-lated physics.

– Electromagnetic energy resolutions;

– Position and photon angular resolutions;

– e/ � identification and reconstruction efficiency.

� Good missing energy resolutions: � /SUSY related physics.

– Hermeticity.

� Good jet energy resolution: jet related physics.

– Achievable jet resolution and intrinsic limitation;

– Dead material (coil?) in the middle of a calorimeter;

– Jet resolution improvement by using other detector com-ponents;

– Jet resolution improvement by using kinetic constraints:Z mass and center of mass.

� Dense absorber: a compact and cost effective calorimeter.

Disco very Power of Precision e & �� Study quarkonium system through inclusive photons by Crys-

tal Ball and CLEO.

� � � � � �� �

���

��� � � � ��� � � � � �

��� � � � �� � � ! ! !" # #$ % & &'(*) + +

,

-. /

021435

687 9 : ; η < ′ = > ? @ A η B

CDEFGHI JKMLNOPQ RS

Tγ U VXWZY\[

]_ _acbdef

gihjckl_m n oprq s�t

u8v w x y_z {�|}_~ � ��_� � �

�

�������

��������� �� ���� ����

��������� �2¡¢¤£ ¢�¥ ¦¤§ ¦©¨ ª¤« ª©¬ ­¤® ¯°­ ±¤² ³°´ µ¤¶ ·°¸ ¹¤º »½¼� Searches for excited leptons in composite models and a

SUSY breaking model with gravitino ¾¿ as LSP at LEP II.ÀÂÁÃÀÂÄÆÅ ÇÉÈÊÇÉÈÌËÎÍÏÇÉÈ\ÇÎÐÌÇÉÈÑÅ ÇÓÒÔÐ ÀÂÁÃÀÕÄÖÅ ×ØÔÙ Ú ×ØÌÙ Ú ËÎÍÛ×ØÌÙ Ú ×Ü ÐÝ×ØÌÙ Ú Å ×Ü Ò

L3

Photon Energy / Ebeam

Eve

nts

/ 2 G

eV

DATA = 343

∫ Þ

L = 102.4 / pb

MC (νν−γ) = 353.6 (350.8)

(14o<θß

γà <166o)

0

50

0 0.5 1

Disco very Power of Precision PhotonsH á �â� Searches with PWO ECAL by CMS at LHC

4000ã

5000ä

6000å

7000æ

8000ç

110 120 130 140

Eve

nts/

500

MeV

for

100

fb–1

0

200

400

600

110 120 130 140

mγγ (GeV)

Eve

nts/

500

MeV

for

100

fb–1

a) b)mγγ (GeV)

Higgs Searches by L3 at LEP II

The Higgs background is suppressed by b-tagging and neuralnetwork analysis. Kinetic constraints improve mass resolution toè 3% for 87 GeV Higgs.

éÌêëéÔì á íïîcðÛñòíóîÛá ôöõôø÷úùüûýðþîÛá ÿ õÿ

DataHZ→qqqqBG

a) L3

Btag

Eve

nts/

0.6

b) L3

NNout

Eve

nts/

0.05

c) L3

MrecH (GeV)

Eve

nts/

2.5

GeV

d) L3

PM

Eve

nts/

0.05

10-2

1

10 2

0 5 1010

-2

1

10 2

0 0.25 0.5 0.75 1

10-1

1

10

60 80

10-1

10

10 3

0 0.25 0.5 0.75 1

What Can Crystal Calorimetr y Offer?

� Good electr omagnetic energy resolution because of totalabsorption: 0.6% is achievable for isolated e or � , � ���� ��� � ��� �� �����

.

� Good position resolution because of its fine segmenta-tion: 0.3 mm is achievable for cell size and Moliere radiusof 2 cm, � � ����� � ������

mm.

� Good photon angular resolution because of no ambiguityin primary event vertex in LC — bunch length 0.2 mm.

� Good e and � identification and reconstruction efficien-cy because of fine granularity and pointing geometry: e/ �discrimination better than 10 � is achievable for e ID effi-ciency of 95%.

� Good missing energy resolution together with HCAL be-cause of hermeticity.

� Good jet energy resolution by using information from oth-er detector components: L3 achieved 7% for hadronic Zdecays.

� Can be rather compact by using heavy crystals of less than1 cm radiation length (PbWO � and PbF � ).

KTeV CsI Calorimeter & Measured Resolution

Bhabha Electr on Energy Resolution with L3 BGO0.5% Calibration Achieved in situ with RFQ

Contribution “Radiative”+Intrinsic Temperature Calibration Overall

Barrel 0.8% 0.5% 0.5% 1.07%

Endcaps 0.6% 0.5% 0.4% 0.88%

45.6 GeV �

<� EBeam <� 94.3 GeV

Barrel

σ� = 1.06%

Endcaps

σ� = 0.86%

BGO energy / Beam energy

Eve

nts

/ 0.0

025

0�

500�

1000

0.9 1 1.1

0�

10000

0.9 1 1.1

Energy (GeV)

σ(E

) / E

(%

)

Barrel

Endcaps

0

1

2

3

4�

10-1

1 10 102

RFQ Installation in L3 Experiment

8235�5425 4010

2530!

BGOTEC"

Coil

Magnet Yoke

Muon Chambers

Muon Filter

Hadron Calorimeter Barrel

HC2HC3

14 180 mm

Luminosity Monitore- e+#

L3

y863col$

HC1

Forward-BackwardMuon Chambers

SMD%

Support Tube%

RFQ

Impr ovement of L3 Jet Mass ResolutionUsing Information from other Detector Components

L3 Data 2 Jet Eventsσ(EJet

& ) about 6.3 GeV'(was 9.3 GeV)

new algorithmold algorithm

Visible Energy [ GeV ]

Eve

nts

/ GeV

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

0 20 40 60 80 100 120 140

( )*( ),+CsI(Tl) ECAL and Resolution

10- -1

1-

10-

10-2

10-1

1

Noise

Intrinsic

All.

Energy (GeV)

σ/E

(%

)

/

CMS PbWO 0 ECAL and Resolution

0.11

1

10

1 10 100 1000

σ/E

[%]

Intrinsic

All

Noise

Photo

E[GeV]

Crystal Position Resolution��� � � ������mm for R 2436587:9<;=9 = 2 cm

>@?BA

CEDBF

GEHBI

JEKBL

MENBMPOQO R@S

TBU V@WEXBW Y[ZE\B] ^[_@_E`B_

Possib le Choices of Crystal Technology

� Oxides:

– BGO is a mature and dense crystal ( a = 7.13 g/cc, X î =1.12 cm, R 2b365c7ed9P;=9 = 2.3 cm), but has a slow scintillation(300 ns) and not cost effective ($7/cc) due to expensiveraw material (GeO � ).

– PbWO � is a mature and dense crystal ( a = 8.28 g/cc,X î = 0.89 cm, R 2436587ed9<;=9 = 2.0 cm). It is a fast and costeffective crystal ($2.5/cc). Its low light yield is overcomeby using Si avalanche photodiode. � � 0 �gfh� � � � ���i�j�� kl�mfn ����

has been achieved with 25 mm � APDreadout in beam test. It is possible to develop a brighterPbWO � crystal.

� Halides:

– CsI is a mature and cost effective crystal ($2/cc), buthas low density ( a = 4.5 g/cc, X î = 1.85 cm, R 2b36587ed9<;=9 =3.5 cm). In addition, CsI(Tl or Na) is too slow ( è 1 o s)and CsI is less bright.

– PbF � is a mature and dense crystal ( a = 7.77 g/cc, X î= 0.93 cm, R 2436587ed9p;=9 = 2.1 cm). It is also cost effective(less than PbWO � ). However, it is not yet a scintillator,but being used as a Cerenkov radiator. A scintillatingPbF � crystal may be developed by selected doping.

CMS PbWO 0 Energy Resolution in Beam TestStoc hastic & Constant Terms

q �� � r �� � st� l�mfn ����u õr = 4.1% v 1.7 p.e./MeV in 25 mm � APD.

CMS ECAL TDR uses 50 mm � APD w õr = 2.9%.

u õs = 0.37% v a better understanding of the consequenceof light response uniformity.

Effect of Light Response Unif ormityD. Graham & C. Seez, CMS Note 1996-002

u Minimize contributions to the constant term of energy reso-lution, caused by light response non-uniformity.

CMS PbWO 0 ECAL Beam TestResolution of 280 GeV Electr onsq �� � xy�{zn�

� � |���}�~�� ��mfn ��E� � l� 0 ��

PbWO 0 Scintillation Light OutputMeasured with R2059 PMT

Typical Full Size���mf ��� ��i � �l�:i � PWO

0

5

10

15

20

0 1000 2000�

3000�

4000�

Time (ns)

Lig

ht O

utpu

t (p

.e./M

eV)

PWO �

F+S�

PWO �

F�

PbWO � Scintillation Light OutputSize �l� f�� � f�� � ��� f�� PWO

56 p.e./MeV Measured with with R2059 PMT

0

10

20

30

40

50

60

70

80

0 1000 2000 3000 4000�

Time (ns)

Ligh

t Out

put (

p.e.

/MeV

)

Gate (ns): 50 100 200 1000 2000

10.2 14.8 22.3 49.2 55.4

10.0 14.4 21.9 48.7�

55.2

SIC-S025

red - as received

green - tested before sent to Woody�

Status of PbF � Crystal as a Scintillator

� PbF � has been studied in details as a Cerenkov materialby D. Anderson and C. Woody et al., NIM A290 (1990) 385and IEEE Trans. Nucl. Sci. NS-40 (1993) 546.

� Attempt has been made to produce scintillating PbF � throughphase transition (cubic to orthorhomic). Positive result re-ported by N. Klassen et al. in Crystal 2000 (1992) 587 doesnot agree with observations by S. Derenzo et al. IEEE Tran-s. Nucl.Sci. NS-37 (1990) 206 and D. Anderson et al. NIMA342 (1994) 473.

� Observation of fast scintillation in PbF � (Gd) and PbF � (Eu)was reported by D. Shen et al. (SIC) Jour. Inor. Mater. Vol101 (1995) 11. The scintillation emission of PbF � (Gd) wasconfirmed by C. Woody et al. in Delft Conference (1995),and 6.5 p.e./MeV was observed for a PbF � (Gd) sample of� ���m��� �l��� cm from SIC by using R2059 PMT.

� About 1,000 PbF � crystals of � � � � ���l�:� cm (a total of0.167 m � ) are being produced by SIC in 1998 for an exper-iment at Mainzer Microtron, Germany. They are used asCerenkov radiator.

X-ray Excited Emission Spectra of PbF � (Gd)D. Shen et al., Jour. Inor. Mater. Vol 101 (1995) 11.

X-ray Excited Emission Spectra of PbF � (Eu)D. Shen et al., Jour. Inor. Mater. Vol 101 (1995) 11.

� -ray Excited Emission Spectra of PbF � (Gd)C. Wood y et al., Delft Conference (1995)

PbF � (Gd) ( ���¡ £¢ ¤ �¥ ¦� cm) Pulse HeightMeasured at AGS with 1 GeV/c MIPS by C. Wood y et al.

6.5 p.e./MeV Obser ved by R2059 PMT

Longitudinal Transmittance of PbF �Measured with Hitac hi U-3210 Photospectr ometer

� � � � �n�l�§� cm Sample from SIC

0

20

40

60

80

100

200¨

300©

400ª

500 600 700«

800

Wavelength (nm)

Tra

nsm

itta

nce

(%)

PbF2 SIC-1

Radiation Damage (10 krad) and Annealing� � � � ���l�:� cm PbF � Sample from SIC

Measured by F. Maas

Summar y

¬ To maximize physics reach, calorimetry for LCshould have good measurement on electrons, pho-tons and jets.

¬ A crystal calorimeter provides the best achievableresolutions for electrons and photons, a good miss-ing energy resolution and an adequate jet resolu-tion.

¬ Recently developed low cost, heavy crystals of-fers a cost effective crystal calorimeter solution.

¬ Feasible crystal technologies:

– A PbWO � calorimeter;

– A PbF � calorimeter following successful R&D.