Charge Collection in CVD Diamond

download Charge Collection in CVD Diamond

of 3

  • date post

  • Category


  • view

  • download


Embed Size (px)


Charge Collection in CVD Diamond. Paris – Nov. 22, 2012. R. Kass. 35. Radiation Damage - Basics. Charge trapping the only relevant radiation damage effect NIEL scaling questionable a priori E gap in diamond 5 times larger than in Si Many processes freeze out - PowerPoint PPT Presentation

Transcript of Charge Collection in CVD Diamond

  • R. KassParis Nov. 22, 2012**Richard KassOhio State University

    November 22, 2012The Diamond Beam Monitor for Luminosity Upgrade of ATLASOutline of TalkMotivationThe ATLAS DBM ConceptDBM DesignDBM StatusSummary

  • R. KassParis Nov. 22, 2012Diamond Beam Monitor Motivation*.And L will continue to grow! L= (nbfr)/inelProblems occur when can not be reliably measured =average # of inelastic interactions per bunch crossing nb= # of colliding bunch pairs fr= machine revolution frequency inel= inelastic cross sectionLuminosity is a counting issue: requires good segmentation in space or timeLHC parametersnowMany precision measurements limited by luminosity determination (e. g. absolute cross section measurements, ppZX, WZX,..)Rapid increase in LHC Luminosity now >7x1033cm-2s-1

  • R. KassParis Nov. 22, 2012In order to make a precise determination of the luminosity at the highest instantaneous luminosity & energy we need a detector that is:radiation hard: accumulate 2x1015 neq/cm2 fast: ~nsec resolution, resolve beam bunchesstable/reliable: last for ~ 10 yearssensitive to charged particles: prefer min. ionizing**Diamond Beam Monitor MotivationDiamond is a good sensor candidate...

  • R. KassParis Nov. 22, 2012*Diamond as sensor material

    Single-crystal CVD & poly CVD fall along the same damage curveProton damage well understoodAt all energies diamond is >3x more radiation tolerant than siliconRadiation StudiesThe properties of diamond make it interesting/useful for particle detectors:radiation hard, short collection time, low leakage current, high thermal conductivityChemical Vapor Deposition Diamond

    PropertyDiamondSiliconBand gap [eV] Low leakage5.51.12Breakdown field [V/cm]1073x105Intrinsic resistivity @ R.T. [ cm]> 10112.3x105Intrinsic carrier density [cm-3]< 1031.5x1010Electron mobility [cm2/Vs]19001350Hole mobility [cm2/Vs]2300480Saturation velocity [cm/s]0.9(e)-1.4(h)x 1070.82x 107Density [g/cm3]3.522.33Atomic number - Z614Dielectric constant Low cap5.711.9Displacement energy [eV/atom] Rad hard4313-20Thermal conductivity [W/m.K] Heat spreader~2000150Energy to create e-h pair [eV]133.61Radiation length [cm]12.29.36Interaction length [cm]24.545.5Spec. Ionization Loss [MeV/cm]6.073.21Aver. Signal Created / 100 m [e0] Low Noise, Low signal36028892Aver. Signal Created / 0.1 X0 [e0]44018323

  • R. KassParis Nov. 22, 2012ATLAS Already has 2 Diamond Based Systems **The Beam Condition Monitor (BCM)The Beam Loss Monitor (BLM)Both of these systems monitor the LHC beams: can abort the LHC beams useful for determining luminosity

  • 2 diamonds/moduleR. KassParis Nov. 22, 2012*ATLAS Beam Conditions MonitorThe ATLAS BCM plays a key role in data taking luminosity measurement + protectionBCM uses 16 pCVD diamonds ( 1 x 1 cm2)z=184 cm r=5.5 cm ||=4.2

  • R. KassParis Nov. 22, 2012*Data taking with the ATLAS BCMBCM achieved 687ps timingBCM can distinguish real collisions from background

    BCM stores a record of 1177+100 orbitsBCM sensitive enough to abort the beamsIncreasing background

  • R. KassParis Nov. 22, 2012*ATLAS Beam Loss MonitorBLM extends the range of the BCM and adds redundancyBLM running sums give count rate for a given time interval (ms to s)Both BCM and BLM operate continuouslyAborts beam when necessaryBLMBCMBLM uses 12 pCVD diamonds, 6 per side z=345cm, r=6.5 cm

  • R. KassParis Nov. 22, 2012DBM Motivation: lessons learned**Luminosity measurement with the ATLAS diamond BCMSingle Particle Counting: in-time Luminosity & out-of-time BackgroundFlexible: Can run as OR, AND, horizontal, vertical=0.7nsSpeed, robustness, stability required for good luminosity measurement

  • R. KassParis Nov. 22, 2012DBM Motivation: lessons learned**The BCM rate (speed) is Beam Crossing ID aware

    Provides robust rate measurements, ~ 10-3 backgrounds ORANDside A ONLYside C ONLY

  • R. KassParis Nov. 22, 2012DBM Motivation: lessons learned**Two independent luminosity measurements BCMH & BCMV: (H=horizontal, V=vertical)In 2011 BCM achieved a 1.9% luminosity measurement! (BUT issues with vDM scans increased systematic error to ~3.5%) Stable over monthsStable against pile-up

  • R. KassParis Nov. 22, 2012DBM Motivation: lessons learned**The BCM is preferred ATLAS luminosity device since early 2011:

    Calibrated in Van der Meer scansOperates when other systems are not active!

  • R. KassParis Nov. 22, 2012DBM Motivation**The BCM will begin to saturate at ~1034 cm-2s-1:More segmentation Diamond Beam Monitor (DBM)Increase number of channels by ~ 105now

  • R. KassParis Nov. 22, 2012The ATLAS DBM Concept** Build on success of BCM pixelate the sensorsUse IBL diamond pixel demonstrator module use many of the same pieces (FEI4, etc) as IBL to save time/money same segmentation as IBL sensors, 50 x 250 m2Install during new Service Quarter Panel (nSQP) replacementFour 3-plane stations on each side of the IR

  • R. KassParis Nov. 22, 2012The ATLAS DBM Concept**24 diamond pixel modules arranged in 8 telescopes provideBunch by bunch luminosity monitoringBunch by bunch beam spot monitoringInstallation in summer 2013

  • R. KassParis Nov. 22, 2012The ATLAS DBM Specs and Collaboration**Specs:Bunch by bunch luminosity monitoring (
  • R. KassParis Nov. 22, 2012The ATLAS DBM Tracking Simulation*50m in direction50m in r directionz= ~4cmz=~0.6cmSimulate DBM to find best orientation and resolution Focus on z vertex resolution (momentum resolution bad) 3 layers of tracking with 50 x 250 m2 pixel cell Two choices of pixel orientation, pick the one that gives the best z-vertex resolutionGEANT

  • R. KassParis Nov. 22, 2012The ATLAS DBM Concept**Mechanics finalized use as many IBL parts as possibleMechanical simulations complete Al, AlN, PeekFour Telescopes on Cruciform

  • R. KassParis Nov. 22, 2012The ATLAS DBM Concept*Thermal & mechanical simulations completeSingle Telescope w/ Cooling ChannelResult: Max Displacement = 26.2 mMechanical SimulationResult: Max T = 14.2CThermal Simulation

  • R. KassParis Nov. 22, 2012*DBM Diamond Sensor PlanTwo diamond suppliers involved: Diamond Detectors Limited (DDL)/E6 (UK based) II-VI (US based)Diamond Sensors for DBM: Type: polycrystalline CVD diamond Charge collection distance > 250 m (as measured with Sr90 source) Size: 21 x 18 mm2, 525 25 mm thickness Number: 24 for DBM modules + spares 5 for Irradiation studies21 x 18 mm2 pCVD diamond

  • R. KassParis Nov. 22, 2012*How much Diamond is in the DBM?Each DBM sensor is ~ 3.2 carats (1 carat= 200 mg)Entire DBM: 24 sensors ~ 76 carats!Price of all DBM diamond < $150kAbout as much as in thediamond Richard Burtonbought for Liz Taylor in 1969 ~ 70 caratsAuctioned in 1978 for $5M

  • R. KassParis Nov. 22, 2012*DBM Diamond Sensor QualificationA wafer from DDLAt OSU we put conducting contacts (gold dots) on the diamond & measurethe charge collection properties in the region of each dot. Use Sr90 as a source of particles for charge collection measurementsWe also make a map of the current draw as we measure CCD Good regions have I < 5 nA at 1000V in air

  • R. KassParis Nov. 22, 2012*CCD measurement with Sr90 @ OSU1mC Sr90 sourcetrigger scint.diamond holder and preamppower suppliespostampScope:use as digitizersee individual pulsesNIMtriggerI/Vmeters

  • R. KassParis Nov. 22, 2012*DBM Diamond Sensor QualificationMap of chargecollection distancesCut mapThis wafer will becut into 11 sensorsMap of currentdrawBack to manufacturerfor dicing & thinningBack to OSUre-test with Sr90,metalize backplaneTo IZMfor bumpbonding, etc

  • R. KassParis Nov. 22, 2012DBM Module Production*HV Problems with first modules from IZM Backside metalization goes to the edge of diamond and breaks down Fixed by changing back metalization procedure no longer performed by IZM, performed at Ohio State14 sensors currently at IZM put pixel pattern on diamond bump bond diamond to FEI4IZMOSU

  • R. KassParis Nov. 22, 2012DBM Module Testbeam**Three Testbeam campaigns Oct 11, Mar 12, Jun 12Learning about FE-I4 performanceCalibration/tunings for low threshold performance

  • R. KassParis Nov. 22, 2012DBM Module Testbeam**Prototype Modules Tested:21mmx 18mm pCVD diamond w/FE-I4A336 x 80 = 26880 channels50 x 250 m2 pixel cellResultsNoise map uniformEfficiency >95%

    Hit Map

  • R. KassParis Nov. 22, 2012DBM Module Testbeam**Prototype Modules Tested:21mmx 18mm pCVD diamond w/FE-I4A336 x 80 = 26880 channels50 x 250 m2 pixel cellResultsSpatial resolution looks digital

  • R. KassParis Nov. 22, 2012*OSUs DBM Hitbus Chip OverviewASIC developed to provide L1A trigger for the DBM by using the FEI4 Hitbus outputs Each DBM hitbus chip services2 telescopes so 4 chips neededfor entire DBM

    Use IBM 130nm 8RF CMOSSize: 4.59 mm x 1.06mm

    Incorporates shared circuit blocks from FEI4 collaboration& custom blocks designed at OSU

    PLL (OSU)programmabledelay cell (OSU)LVDS RX (Bonn)LVDS TX (Bonn)FEI4 commanddecoderHit processorlogic core (OSU)Power onreset (OSU)

  • R. KassParis Nov. 22, 2012*Half DBM System Block Diagram


    2 Support PCBs/side

  • R. KassParis Nov. 22, 2012**Hitbus Chip IrradiationNeed to certify that Hitbus Chip is rad hardIrradiated 2 Hitbus chips to 4.3 x 1015 p/cm2 (115 Mrad) Used 24 GeV proton beam at CERN

    Both chips survived Slight increase in supply current consumption

    All functionality active during irradiationTest system checked for SEU in the rece