B Abi OSU R Boyd OU P Skubic
B. Abi( OSU), R. Boyd (OU), P. Skubic (OU), F. Rizatdinova (OSU), K. K. Gan (Ohio State U. ) Status of the Pi. N diodes irradiation tests
Oklahoma responsibility • Our goal is to identify radiation hard high speed Pi. N diodes and measure their lifetime. • High speed Pi. N arrays available on the market are Ga. As or In. Ga. As. • Radiation tests performed by KK in August 2007 have revealed that most of these devices are not radiation hard. 2
Pi. N diodes in our tests • Contacted Hamamatsu to get the Si Pi. N diodes. • Got two types of single Pi. N diodes to test them this spring/summer – S 9055 -01 – S 5973 -01 • In case of success have an agreement that Hamamatsu will produce arrays for us. • In addition, purchased Hamamatsu Ga. As Pi. N diode, G 8255 – Actually there are 3 varieties of this Pi. N: G 8255 -01, G 8255 -02, G 8255 -03 – Difference is in the size of chip and in the speed 3
Characteristics of selected Pi. Ns • S 9055 -01: – 2 GHz; – Active area d = 0. 1 mm; – Spectral response range 320 – 1000 nm; Peak sensitivity wavelength 700 nm; – Reverse voltage 2 V; • S 5973 – 1. 5 GHz; – Active area d = 0. 4 mm; – Reverse voltage 3. 3 V • G 8522 (3 types) – for all of them reverse voltage is 2 V; – Spectral response range from 570 to 870 nm; Peak at 850 nm – G 8522 -01: • 3 GHz; Active area d=40 mm – G 8522 -02: • 1. 9 GHz; Active area d=80 mm – G 8522 -03: • 1. 5 GHz; Active area d=120 mm This information is from data sheets, available online from Hamamatsu web page 4
Performed irradiation tests • TID test at BNL – Total dose 10 Mrad; – Passive test; • 2 tests with protons at IUCF – 200 Me. V protons; – Total dose 80 Mrad (two times of 40 Mrads); – Online readout • current of Pi. N diodes • Temperature was within the range 22. 5 -23 C° 5
TID test • TID test with gamma rays at BNL together with SMU performed on April 10 -11. – Tested three G 8255 and one S 9055 -01 diodes, all have been biased. – Responsivity was measured offline at 0 Mrad, 5. 6 Mrad and 9. 6 Mrad • Homogenous IR LED source biased by constant current was used to illuminate the Pi. N diodes for the responsivity measurements. – For each dose made 5 measurements, varying the optical power of the IR source from 2 to 10 m. W. – Measurements were averaged to get better understanding of the true responsivity. 6
TID results Absolute responsivity of G 8255 -02 at different Optical Power R@5 u. W 0, 55 R@7. 5 u. W R@10 u. W 0, 5 R@12. 5 u. W R@17 u. W 0, 45 0, 4 0 5 Gamma Ray /MRad Conclusion: No degradation has been observed for any type of tested Pi. Ns in TID test with 10 Mrad Relative Responsivity I/I 0 Responsivity I/W 0, 6 10 1 0, 99 0, 98 S 9055 -1 0, 97 G 8522 -3 G 8522 -1 0, 96 G 8522 -2 0, 95 0 5 10 Gamma ray /MRad 7
Test with protons at IUCF • Performed in May and June at IUCF: 200 Me. V protons – Have neutron source too, can be used; – Charge $550/hour, min number of hours to purchase is 12. • Used 10 Pi. Ns, two of each available Pi. N diodes • Did not package them – Did not have a possibility to do that at Oklahoma; • Used IR LED sources to illuminate Pi. Ns • One of each type of diodes was coated with radiation hard resin, another was a bare chip • Started irradiation with very low flux, increased it gradually • Got 40 Mrad in May and 40 Mrad in June. • Still have not done full analysis of results obtained in June – will have in ~ two weeks. 8
Setup Board with Pi. N diodes IR Source together with the board Setup installed in front of the proton beam exit 9
Results of irradiaton test with protons • Example: Responsivity vs dose for all types of the detectors in the region between 16 and 29 Mrad. • Responsivity of all diodes coated with resin is lower compared to bare ones • The thickness of the resin varies for each type of Pi. N (added by hand at Oklahoma State U. ) 1 G 8522 -01 1, 05 8522 -01 wr 0, 95 0, 9 0, 85 0, 8 0, 75 Relative Responsivity 1, 05 G 8522 -02 wr 8522 -02 1 0, 95 0, 9 0, 85 0, 8 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Accumulated dose /MRad 10
Resin vs no resin (cont’d) S 5973 wr 1, 05 S 5973 1 Relative Responsivity 1, 05 0, 9 0, 85 0, 8 1 S 9055 wr S 9055 0, 9 0, 85 0, 8 0, 75 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Accumulated dose /MRad Before irradiation 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Accumulated dose /MRad After 40 Mrad: dark are Pi. Ns with resin 11
Relative responsivity • Compare all Pi. Ns without resin in the same dose region, from 16 to 29 Mrad • S 9055 -01 and G 8522 -02 have very similar behavior 12
Annealing effect • We had a 5 hours break in data taking – cyclotron was off for maintenance. • The total dose before the break was 3. 7 Mrad. 13
Responsivity vs dose • Si S 9055 -01 and Ga. As G 8522 -02 look very promising. After 40 Mrad, the responsivity went down by ~ 25%. 14
Preliminary results from June • In June, irradiated the same Pi. N diodes as in May. • Measured responsivity with the same LED before irradiation, and found that the annealing did not improve the responsivity compared to that observed after 12 hours of annealing in May. • All Pi. Ns were biased and readout online. • Observed response from all Pi. Ns, including Ga. As Pi. Ns. • Estimated degradation of Si 9055 -01 Pi. Ns only: – Observed ~50% responsivity after 80 MRads compared to the initial (0 Mrads) responsivity. – Did not take into account degradation of LED due to residual irradiation – preliminary estimation is that optical power of LED went down by ~10% 15
Lifetime test • Plan to start the lifetime test of irradiated Pi. Ns at the end of June, as soon as irradiated Pi. Ns will be released from IUCF. • Currently have a running test for single channel. Also have ready-to-use setup to test up to 64 channels simultaneously. • It is online at http: //139. 78. 127. 190/Acc. Life. M 1. 01. html • Monitor Pi. Ns current and temperature. 16
Summary on tests • We developed a simple test stand for Pi. N responsivity studies, that allows us to avoid packaging. • TID passive test shows no degradation in Pi. N responsivity up to 10 Mrad; • Test with protons shows that both S 9055 -01 and G 8522 -02 are good candidates. They demonstrate 75% of their initial responsivity after 40 Mrad and 50% after 80 Mrad. • It would be good to test them at CERN this August with 24 Ge. V protons and make a detailed comparison. • If Ga. As Pi. N diodes survive there will be no need to order a special production of the Si Pi. N arrays made of S 9055 -01 chips. 17
- Slides: 17