Z. Li Brookhaven National Laboratory, Upton, NY , USA E. Verbitskaya, V. Eremin, A. Ivanov Ioffe Physico-Technical Institute of Russian Academy of Sciences St. Petersburg, Russia J. Härkönen, E.Tuovinen, P. Luukka Helsinki Institute of Physics, CERN/PH, 1211 Geneva, Switzerland Initial data on the DRIVE approach: results and analysis This work has been supported in part by the US Department of Energy, contract No.: DE-AC02-98CH10886, by the grant of the President of Russian Federation # NS , by Academy of Finland, and by the EU Intas contract; and it is within the framework CERN RD39 and RD50 collaborations

Introduction Current/Conventional Approaches Defect/Impurity/Material Engineering (DIME) Device Structure Engineering (DSE) Device Operational Mode Engineering (DOME) New Approaches Device Recovery/Improvement Via Elevated temp annealing (DRIVE) Experimental Methods Results and Discussions Summary OUTLINE

For SLHC, with cm -2 s -1 luminosity, total radiation fluence can be as high as n eq /cm 2 At this highest fluence, detector degrades in many ways: oHigh leakage current (I L ): 4x10 -1 A/cm 3 at RT! oHigh full depletion bias (V fd ): 6400 V for 200  m thick STD FZ detector! oLow charge collection distance due to trapping (CCE(  tr )): 20 to 40  m regardless of detector detector thickness and depletion depth! Radiation hard/tolerant Si detectors have to improve with regard to all three problems listed above Most conventional approaches may solve only one to two problems Need a new, simple/easy approach Introduction

Name of the approach R&D groupGood forTechnology difficulty Defect/Impurity/Materi al Engineering (DIME) Oxygenation/MCZ SiCERN RD50V fd Easy Diamond/SiC/GaN, etc.CERN RD42, RD50 V fd Difficult Device Structure Engineering (DSE) 3d detectorsCERN RD50 CCE(  tr ), V fd Very difficult Semi-3d detectorsCERN RD50V fd Easy Thin detectorsCERN RD50V fd Difficult Device Operational Mode Engineering (DOME) Cryogenic Si detectors (LN 2 T) CERN RD39I L, V fd Difficult Cryogenic Si detectors (LHe T) CERN RD39I L, V fd, and CCE(  tr )? Very difficult Current/Conventional Approaches

Device Recovery/Improvement Via Elevated temperature annealing (DRIVE)  The DRIVE was first proposed by Z. Li and J. Harkonen on the 5 th RD50 Workshop in Florence, Italy, October 14-16, 2004  Preliminary data presented by E. Verbitskaya et al, at the 6th RD50 Workshop, Helsinki, June 2-4, 2005  Thermal annealing of radiation damage has been a conventional way for material/device recovery  It has not been used so far in HE physics detector field for standard FZ and oxygenated FZ Si detectors because :  If the annealing temperature is too high (>450 °C), it will destroy the detector and/or electronics  If the annealing temperature is too low (< 450 °C), the reverse annealing (generation of more negative space charges) will make the detector worse New Approach

Device Recovery/Improvement Via Elevated temperature annealing (DRIVE) (continues)  However, for high resistivity MCZ Si material/detectors, thermal annealing in the temperature range from 200 °C to 450 °C will generate thermal donor (TD, positive space charge) due to high oxygen concentration [O]  By playing the annealing T and time, and [O] (n-type or p-type MCZ Si), one may adjust the TD creation rate in such a way to cancel the reverse annealing effect, and even better, to compensate the original (as-irradiated) negative space charges, which may bring the full depletion voltage down to a manageable range  Elevated temperature annealing (ETA) will also anneal out/down the detector leakage current, and improve detector CCE by annealing out shallow trapping centers  The DRIVE approach may therefore offer a technology to improve all three areas:  I L, V fd, CCE (  tr )  The degree of difficulty may vary depending on the annealing techniques New Approach

p-type MCZ n-type MCZ n-type oxygenated FZ TD generation in different Si with different [O]

One can reach >10 14 /cm 3 in TD concentration in one hour! Might be enough to cancel – SC generated in irradiation and during reverse anneal TD generation at 450 °C in p-type MCZ

Experimental  Detectors: p + -n-n + pad structures with multiple GRs from n-type MCZ Si with a resistivity of 1 kOhm  cm,  p + -p-n + pad structures with multiple GRs, from p-type MCZ Si with a resistivity of 3 kOhm  cm all processed at HIP  Irradiation :

Annealings  multistep process with a variable time  T:  C  t ann : variable, 10 min up to 120 min/step  nitrogen flow  fast cooling (~10 min) All as-irradiated detectors (p or n-type) have negative space charges before annealing This means, for n-type detector, the space charge sign inversion (SCSI) occurred

Experimental techniques Measurements: After each annealing step (at BNL):  I-V and C-V dependences  current pulse response using TCT with a laser pulse generation of non-equilibrium carriers After final detector annealing (at Ioffe Institute): Spectra of deep levels (C-DLTS)

Experimental results Three stages of the changes of detector characteristics irrespective to E p and F p : 1.Reverse annealing: I rev , V fd , negative N eff , W  TCT signal from the detector p + side disappears 2.Recovery of reverse current I rev  and the signal from p + side 3. SCSI from negative to positive

Annealing of detectors irradiated by 24 GeV/c protons Compensation of acceptor-type radiation induced defects: Thermal Donors are introduced at  C Reverse annealing

Recovery of reverse current 24 GeV protons, P352-18, F p = 9  cm  C – I rev is still saturated Leakage current decreased by more than one order of magnitude.

24 GeV protons, P352-48, F p = 5  cm -2 T = 450  C Bulk current decreases after 1 st anneal, but the leakage arises at high biases t ann = 305 min: I rev becomes saturated After final anneal, leakage current decreased by more than 2 orders of magnitude

= Changes of TCT signal: F p = 9  cm -2, P SC (-) p + /p/n + structure High field on the n + side

SCSI: SCSI: t ann (450  C) = 210 min SC (+) p + /n/n + structure High field on the p + side

N eff evolution N eff derived from TCT measurements Reverse annealing N eff recovery up to SCSI

N eff evolution Range of reverse annealingRange of TD introduction Finally All positive space charge Since these are all initially p-type wafers with less than /cm 3 donors, the positive space charge has to be thermal donor +SC -SC

Annealing of detectors irradiated by 20 MeV protons Compensation of acceptor-type radiation induced defects: Thermal Donors are introduced at  C n-type detectors before irradiation, initially (+SC) All positive space charge after annealing (+SC) After irradiation, space charge sign inverted to negative (“p”) (-SC)

Recovery of reverse current Leakage is observed in I-V curves

Annealing time at 450  C required for SCSI

Defect spectra after SCSI with injection C-DLTS spectra Reference spectra: P204-I-A3-7: 24 GeV/c F p = 1.5  cm -2  Continuous defect spectra  Increase of t ann is favorable (for P t ann is maximal)

Electron traps P GeV/c, 9  cm -2 Hole traps

Defect parameters Electron traps Hole traps P GeV/c, 9  cm -2  Defects detected after annealing are presumably products of RDs decay  Concentrations of defects after annealing are <10% of as-irradiated RD concentrations  Minimal concentrations correspond to the maximal annealing time (P352-48)

Ways of ETA May be different: 1) localized laser anneal; 2) localized anneal using a lamp; 3) annealing using pre-built-in external heating resistors; 4) annealing using leakage current of the detector itself; 5) RT annealing using ultrasonic power 4) and 5) may be the easiest and most practical method Multiple annealing steps at smaller doses maybe favorable Fine tuning of the annealing time is required for precise manipulation of thermal donor introduction and resulting N eff – further studies are needed

o The conventional approaches may not solved all the radiation induced problems o The new approach, DRIVE, could offer a perfect solution to radiation induced problems, especially at high fluences for SLHC o Recovery of the positive space charge and detector reverse current is realized by ETA at 450  C o By changing annealing temp, time, one may get a perfect receipt for different fluences o P-type MCZ Si may be the best material due to its higher [O] than that of n-type MCZ o Different annealing techniques may be used for DRIVE depending on the experiment conditions o Further studies on detector CCE, on FZ Si, and effect of radiation/annealing cycles are underway Conclusions