CVD Diamond Sensors for the Very Forward Calorimeter

  • Slides: 14
Download presentation
CVD Diamond Sensors for the Very Forward Calorimeter of a Linear Collider Detector K.

CVD Diamond Sensors for the Very Forward Calorimeter of a Linear Collider Detector K. Afanaciev, E. Kouznetsova, W. Lange, W. Lohmann

Diamond samples • Fraunhofer Institute (Freiburg) : – CVD diamond 12 x 12 mm

Diamond samples • Fraunhofer Institute (Freiburg) : – CVD diamond 12 x 12 mm – 300 and 200 um thickness – Different surface treatment : • • #1 – substrate side polished; 300 um #2 – substrate removed; 200 um #3 – growth side polished; 300 um #4 – both sides polished; 300 um – Metallization: • • 10 nm Ti + 400 nm Au Area 10 X 10 mm

I(V) dependence – setup HV Diamond Extremely low currents => N 2 atmosphere EM

I(V) dependence – setup HV Diamond Extremely low currents => N 2 atmosphere EM shielding Keithley 487 Measurements were done with Keithly 487 picoammeter • N 2 Usual I(V) curve Non-ohmic curve ü Average resistance ~(1013 -1014) Ohm (ohmic behavior) ü 3 samples from different groups have “non-ohmic” behavior and lower resistance (~1011 Ohm)

Charge Collection Distance (CCD) Qmeas. = Qcreated x ccd / L Qcreated(mm) = 36

Charge Collection Distance (CCD) Qmeas. = Qcreated x ccd / L Qcreated(mm) = 36 e-h pairs Sr 90 diam. üThe samples haven’t been irradiated Scint. before these measurements üAll data was taken 2 minutes after bias voltage applied PA PM 1 discr PM 2 discr delay & Gate ADC

CCD measurements results CCD, mu Sample# R(average), Ohm 11 4. 60 E+13 35 12

CCD measurements results CCD, mu Sample# R(average), Ohm 11 4. 60 E+13 35 12 3. 20 E+14 35 43 27 13 8. 88 E+11 25 33 21 21 5. 92 E+14 24 22 7. 39 E+14 9 11 9 10 (700 V) 23 3. 93 E+14 9 12 10 13 31 1. 04 E+11 28 32 5. 12 E+13 50 57 50 33 4. 63 E+13 49 54 52 41 5. 12 E+11 45 (400 V) 48 42 4. 35 E+14 43 5. 24 E+13 54 60 (+500 V) (+800 V) (-500 V) (-800 V) 37 30 17 33 (400 V) 58 65

CCD – irradiation studies ü The samples were irradiated with Sr-source with estimated doserate

CCD – irradiation studies ü The samples were irradiated with Sr-source with estimated doserate of about 0. 45 Gray per hour ü The total absorbed dose for all the samples was at least 5 Gy. ü Bias field was set to 1 V/ m ü Irradiation was homogeneous over the sample area ü Parameters monitored during the irradiation: - Sr-spectrum peak position - width of the peak (->noise) - current in HV-circuit - test pulse from a generator (-> electronics stability)

CCD – irradiation studies – results Group #2 (substrate side removed). HV = 200

CCD – irradiation studies – results Group #2 (substrate side removed). HV = 200 V Group #3 (growth side polished). = 300 V HV

CCD – irradiation studies – results Group #1 (substrate side polished). HV = 300

CCD – irradiation studies – results Group #1 (substrate side polished). HV = 300 V Group #2 (substrate side removed). HV = 200 V

CCD – irradiation studies – results Group #3 (growth side polished). = 300 V

CCD – irradiation studies – results Group #3 (growth side polished). = 300 V HV Group #4 (both sides polished). = 300 V HV

Photoluminescence analysis -> no nitrogen, no silicon He. Cd Laser FAP 2_1 LO Phonon

Photoluminescence analysis -> no nitrogen, no silicon He. Cd Laser FAP 2_1 LO Phonon N (575) N (637) Si (770) FAP 4_2 Reference spectra

Raman spectroscopy Resolution ~ 1 cm-1 Result = S(diam)/S(graphite)*1000

Raman spectroscopy Resolution ~ 1 cm-1 Result = S(diam)/S(graphite)*1000

Raman spectroscopy results FAP 2_1 R_ave Ccd_Down 500 V ‘ 1_1 ‘ 1_2 ‘

Raman spectroscopy results FAP 2_1 R_ave Ccd_Down 500 V ‘ 1_1 ‘ 1_2 ‘ 1_3 4. 60 E 13 3. 2 E 14 8. 88 E 11 35 35 25 ‘ 2_1 ‘ 2_2 ‘ 2_3 5. 92 E 14 7. 39 E 14 3. 93 E 14 ‘ 3_1 ‘ 3_2 ‘ 3_3 1. 04 E 11 5. 12 E 13 4. 63 E 13 FAP 4_1 Raman FWHM Halbw. Breite PL Si. Signa l 24 9 9 117. 7 4. 9568 No 102. 0 5. 4254 No 28 50 49 329. 3 182. 4 265. 3 4. 4672 5. 4132 4. 4851 No

Removed substrate ü Group#3 – removed substrate (300 m -> 240 m)

Removed substrate ü Group#3 – removed substrate (300 m -> 240 m)

Results and further studies ü Group#2 in general can work as a detector ü

Results and further studies ü Group#2 in general can work as a detector ü Raman spectroscopy + photoluminescence analysis -> no nitrogen, no silicon ü Next steps: - Influence of the substrate side on CCD and stability - Homogeneity and linearity required for the application -Test beam (May 2004)