39
Ties Behnke: The TESLA project American Linear Collider Workshop, Baltimore, March 2001 1 1 The TESLA Project TESLA: The machine " status of cavity development " RF system " FEL at TESLA " accelerator layout A Detector for particle physics at TESLA Ties Behnke, DESY

The TESLA Project

  • Upload
    vomien

  • View
    216

  • Download
    0

Embed Size (px)

Citation preview

Page 1: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 11

The TESLA Project

TESLA: The machine" status of cavity development" RF system" FEL at TESLA" accelerator layout

A Detector for particle physics at TESLA

Ties Behnke, DESY

Page 2: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 22

TESLA

TESLA

a machine concept: superconducting acceleration modules

a collaboration: build and operate a test accelerator TTF

a proposal to build such a machine

The TESLA Collaboration

Uni HamburgUni RostokBESSY Berlin Yerevan Physics Institut

INR Troitsk

MEPhI Moscow

INFN Legnaro

Page 3: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 33

TESLA Basic Concept

superconducting solid Nb cavitiesE(acc) ~ 25 MV/m, T=2K

Long RF pulses ( ~ 1 ms)low RF peak power (200 kW/m)long bunch train with large interbunch spacing

Low RF frequency (1.3 GHz)small wakefields

Overall design compatible with E(cms) = 91 .... 800 GeV

baseline design and parameters for 500 GeV

modulegeometry

module length V(acc) Fill factor RF/module

9−cell structure 1.04 23.40 78.00% 219

4x7superstructure 3.23 22.00 89.00% 675

The TESLA acceleration structures:

Page 4: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 44

TESLA Parameters

TESLA 500 GeV parameters TESLA 800 GeV parameters

Page 5: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 55

TESLA Bunch Structure

Main characteristics: long bunch trains, even longer times between bunch trains

500 GeV

800 GeV

5 Hz x 2820 x 2.0 1010

3 Hz x 4568 x 1.4 1010

possibility of orbit corrections within single bunch train (fast feedback system)

Head on collisions are possible Bunch collisions are well separated

in detector

Page 6: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 66

Status of Cavities Development

TESLA Test Facility (TTF) Goals: Phase I:

development of acceleration modulesproof of principle of operation of SC linacat high (> 22.5 GeV) gradientproof of principle for SASE FEL in the VUV (60 nm) cavity performance

per production series

Tesla500

Page 7: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 77

Page 8: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 88

RF Power: Klystrons

TH 1801 multi beam Klystron

High power (10 MW peak) Low voltage (117 kV) High efficiency (65 %) Long pulse (1.5 ms)

System has been fabricated in industry

Is now being used at the TTF LINAC

Page 9: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 99

Lorentz Force Deformation

Problem: Cavity deform under the Lorentz force at high gradientCavity changes its shapecavity is detuned

first successful test on cavity C45 at 20 MV/m solution:

active compensation using piezo−crystal

piezo actuator

l = 39mmV(max)= 150 Vf(max) = 500 Hz

Page 10: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 1010

The Free Electron Laser at TTF

TTF LINAC is used to drive a SASE FELGoal I: Proof of Principle for VUV FELGoal II: Operation of user facility after 2003

Page 11: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 1111

The TTF FEL

February 2000: observe first lasing at <100 nm Since then: systematic studies

very reliable and reproducible behaviourcontinuous reduction of the frequencyMain radiation characteristics have been found

CCD image of the FEL beam: Signal development

Page 12: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 1212

The TTF FEL

Development of X−ray energy

Since observation of first lasing: continuous further development of the system towards:

Smaller wavelengthbetter reproducibilityhigher brilliance

FEL operation: brilliance vs energy

Page 13: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 1313

Overall TESLA Layout

Overall collider layout:

DESY Westerhorn

TESLA tunnel: diameter 5.50 m

Page 14: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 1414

Collider Layout: Injector

TESLA injector complex:

Laser driven electron guns Three separate guns for

UnpolarisedPolarisedFEL beam

Electron polarisation is part of the baseline program

Page 15: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 1515

Collider Layout: Positron Source

Positron source: use incoming electron beam as a source of photons produce positrons

Small degradation of quality of beam is acceptable

Allows very high positron currents Possibility of positron polarisation

SLC TESLA

No of positron per pulse 4.00E+010 5.60E+013No of bunches per pulse 1 2820Pulse duration 3 ps 0.95 msBunch spacing 8.3 ms 337 ns

Repetition frequency 120 Hz 5 Hz

Expected positron polarisation: between 45 and 60% at (nearly) full intensity

Need to build a helical undulator (technologically challenging)

Positron Polarisation is not part of the baseline design

Page 16: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 1616

The Interaction Region

Conceptual layout of the interaction region(s):

IR for gamma gamma electron gamma electron electron electron positron34 mrad crossing angle

IR for primary electron positron program (or electron electron)no crossing angle

2. IR not part of baseline design

Page 17: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 1717

Fast Feedback at the IP

Long bunch trains, long times between bunches: Feedback system within bunch train possible to stabilise the luminosity

Act on angle Act on offset

After about 90 bunches: reduction by factor 1000

Train to train tolerance of final doublet limiting the luminosity loss to 10%: 200nm

Page 18: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 1818

A TESLA Site near Hamburg

Page 19: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 1919

Cost

Project will be presented to the public at the TESLA Colloquium on March 23/24including cost

Page 20: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 2020

A Detector for TESLA

Large detector for Optimal trackingOptimal energy flow

High central magnetic field (4T)

High granularity ECAL High granularity HCAL

Both inside the coil!

Instrumentation down to very small angles: hermeticity!

Iron return yoke instrumented as muon system

Page 21: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 2121

Detector Overview

Enlarge view of the inner tracking system:

Overall detector view:

Page 22: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 2222

The VTX Detector

High precision detector close to the beam pipe (R(min) = 1.5 cm) Several technologies are under discussion

Active pixel sensors (a la LHC technology)CCD based sensors (SLD technology)CMOS based sensors (new development) SI ladders are "stretched"

The CCD version:

Page 23: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 2323

Central Tracking Detector: TPC

Sideview of the TPC

Fieldcage / TPC vessel:Light composite wallsModelled after ALEPH / ALICE fieldcageMax Voltage: 100 kV

Page 24: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 2424

Readout Technology: GEMs

Use GEM (or similar) system for signal amplification and readout:

True 2−D readout possible Compact, thin endplates Mechanically "simple" Gains > 10000 have been observed

Measured gainin a 2−GEMstructure

Page 25: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 2525

Performance of the Tracking System

Overall tracking performance:

Efficiency > 98.5%

Full pattern recognition in TPC, VXT, FTD, FCH

Sophisticated merging of different subdetectors

Final clean−up step Based on LEP software with

further developments

Secondary vertex finding

Based on SLD ZVTOP Combined with OPAL NN approach Tuned on Z−data

Page 26: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 2626

Calorimetry: ECAL

Both ECAL and HCAL are inside the coil

ECAL: fine grained SI−W calorimeter

Module length: 160 cmtransverse: 1x1 cmlongitudinal:

30 x 0.4 X W12 x 1.2 X W

About 30 million channels

Page 27: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 2727

Calorimetery: HCAL

HCAL: "tile calorimeter" with iron as absorber, scintillator tiles as active medium

16 fold symmetry in Phi9 longitudinal samplestransverse sampling > 5x5 cm

Readout of tiles with clear fibres to a place outside the barrel

Energy resolution for tileor alternative digital option

Page 28: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 2828

Calorimeter Performance

Thorough evaluation difficult and needs significant software developmentSome preliminary results:

The calorimeter is optimisedfor the measurement of the energy flow in the event:

Need exellent separation of charged and neutral particles

Excellent connection to the tracker informationExcellent measurement of the longitudinal shower shapeGoal: Energy flow resolutionof 30%

Mass resolution of the visible mass of the Z in hadronic Z decays

HHZ events: separation of signal and background for different energy flow resolution

a) LEPtype resolution 60%(1+|cos θ|)b) 30% resolution

Page 29: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 2929

Calorimeter Performance

Separation of WWZ:

" Standard performance " High resolution calorimetera) resolution 30%b) resolution 60%

Performance

Page 30: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 3030

Backgrounds

10 simulated pair particles

Per BX: 129000 pairs (360 TeV total energy)

Hits from pairs/ BX on the vertex detector layers

Beam Beam backgrounds:PairsHadronic backgroundneutrons

Synchroton Radiation induced backgroundsBeam gas backgroundsMuon induced backgrounds

Some numbers: SIT/FTD: O(20) hits / detectorPhotons into TPC: O(1300)

Occupancy O(0.1%)Particles with E>3 MeV intoECAL: O(200)

Page 31: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 3131

Backgrounds: Neutrons

Simulation of neutron backgrounds:FLUKA2000 & PythiaFull detector model in FLUKACross checked with Pythia

Some fluxes in the detector:

VTX <1E09FTD <2E09SIT <7E08

TPC 15000/BX neutrons/BXECAL <10000/BXHCAL <10000/BXYoke ~55000/BX

(1MeV)/year/cm2

Fluxes are not expected to be a problemfor detector components

Page 32: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 3232

Overall Detector Performance

Two particularly challenging examples: determine the detailed properties of the Higgs

Reconstruction of the Higgs branching ratio into different flavours, as a function of the Higgs mass

Reconstruction of the Higgs Potentialvia ZHH events

Combination of high luminosity and high precision detector allows reconstruction of complete picture eg of the Higgs.

Page 33: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 3333

Detector Mechanics

Open the endcap Yoke Retract the endcap calorimeters Move the TPC along z Acces the inner detectors

Proposed cable routes outof the detector

First conceptual version of detectormoving and installation:

Page 34: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 3434

The TESLA Research Campus

Central laboratory site at km15

HEP experiment(s)XFEL laboratory

Aerial view of Ellerhoop

Artists drawing of the HEP hall

Page 35: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 3535

TESLA: Goals and Milestones

Goals:Develop superconducting technology Use LINAC as driver for X−FEL

Milestones reached:Routine production of cavities with > 25MV/mCavities with >40MV/m as single cell cavitiesConstruction and operation of TTF I

Stable operation for > 8600 hDemonstrate SASE principle at <100 nm

Successful development of klystrons, RF couplers, etc

Development of the Physics Case2 ECFA/DESY workshops with large and international

attendance (total >10 workshop meetings)Milestone reached: TESLA TDR Part III (physics),PartIV (detector), Part VI (other research options)Continuation for two more years to

Develop the physics studies furtherReact to new developmentsContinue work on the detector (R&D efforts are starting)Continue the work on machine/ detector interface

Page 36: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 3636

TESLA: The Next Steps

Presentation to the public: TESLA ColloquiumIntegrated HEP and FEL laboratory with site close to HamburgIncluding cost estimate

Detailed costing of accelerator costsBased on industrial studies and the experience gained at TTF

Cryostat INFNDamping Ring INFNInput Coupler IN2P3SC Magnets SpainCryogenics TU DresdenKlystrons DESYModulators DESY The next steps:

Operation of TTF I, upgrade to TTF II (2003)Formal proposal (TDR) March 2001Evaluation of proposal by German Wissenschaftsrat during 2001ECFA/DESY study on long term perspectives of particle physicsin Europe (2000/2001) with similar studies in US and AsiaICFA study of the Global Accelerator Network concept (2000/2001)

Laboratory

Global accelerator Network

Operation, R&D Operation, R&D Operation, R&D

Page 37: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 3737

Page 38: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 3838

Conclusions

TESLA: a proposal for a new large interdisciplinary research center

Most technical problem are solved 500 GeV baseline design is "conservative" Energy upgrade potential is real

HEP experimentation at TESLA is challenging

Needs serious and significant Detector R&D

Combination of HEP and FEL offers exciting new perspectives

Plans: TESLA TDR now German Wissenschaftsrat: 2002 International technical review?

Page 39: The TESLA Project

Ties Behnke: The TESLA projectAmerican Linear Collider Workshop, Baltimore, March 2001 3939

A Global Accelerator Network

Construct and operate future large accelerators in the framework of a global network

Make projects part of the national programs of the participating countries Maintain the scientific and technical culture and know how in home labes, remain

attractive for young people, yet contribute to and participate in large, unique projects Maintain and run accelerators to a large extend from participating labs Pull together world−wide competence, ideas, resources

Capital investment is done at home Site selections becomes a less critical issue Put accelerator close to an existing laboratory:

Make optimal use of existing experience, manpower, and infrastructureSpecific financial obligations for the host country

ICFA study findings:Global considerations:

Need laboratory structureHost nation is essentialWill bear a major fraction of the cost

Technical considerations:Project requires central managementHost lab will have safety responsibilityRemote operation is in principle feasibleLocal staff of approx. 200 is needed