Radiation Effects in MOS Capacitors With Very Thin Oxides at 80°K

IEEE Transactions on Nuclear Science

Volumn 31, Issue 6, 1984, Pages 1249-1255

  Ancona, M G ^b   Modolo, J A ^a  

^a NAVAL RESEARCH LABORATORY   (United States)

^b SFA INC   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

SEMICONDUCTOR DEVICES, MOS - RADIATION EFFECTS;

FLATBAND VOLTAGE SHIFTS; MOS CAPACITORS; TRAPPED HOLES; TUNNELING;

CAPACITORS;

EID: 0021599338     PISSN: 00189499     EISSN: 15581578     Source Type: Journal    
DOI: 10.1109/TNS.1984.4333491     Document Type: Article

References (20)

1. H.H. Sander and B.L. Gregory, IEEE Trans. Nucl. Sci. NS-22, 2157 (1975).
2. H.E. Boesch, Jr., and J.M. McGarrity, IEEE Trans. Nucl. Sci. NS-23, 1520 (1976).
3. R.C. Hughes, Appl. Phys. Letts., 26, 436 (1975).
4. N.S. Saks, IEEE Trans. Nucl. Sci. NS-25, 1226 (1978).
5. N.S. Saks and J.M. Modolo, IEEE Trans. Nucl. Sci. NS-30, 4197 (1983).
6. H.E. Boesch, Jr., IEEE Trans. Nucl. Sci., NS-24, 2135 (1977).
7. H.E. Boesch, Jr., F.B. McLean, J.M. McGarrity, and P.S. Winokur, IEEE Trans. Nucl. Sci. NS-25, 1239 (1978).
8. J.R. Srour and K.Y. Chiu, IEEE Trans. Nucl. Sci. NS-24, 2140, (1977).
9. C.R. Viswanathan and J. Maserjian, IEEE Trans. Nucl. Sci. NS-23, 1540 (1976).
10. P.S. Winokur and H.E. Boesch, Jr., IEEE Trans. Nucl. Sci., NS-27, 1647 (1980).
11. H.E. Boesch, Jr., F.B. McLean, J.M. McGarrity and G.A. Ausman, IEEE Trans. Nucl. Sci. NS-22, 2163 (1975).
12. Another possible mechanism relates to diffusion of the generated holes within the oxide. This will be primarily an edge effect (because of the hole concentration gradients present near the edges) and will act to decrease yield in the thinner oxides. However, according to calculations of R.C. Hughes (private communication) the effect is minimal.
13. This formula may be deduced as in S. Manzini and A. Modelli, in Insulating Films on Semiconductors, ed. J. Verweij and D. Wolters, Elsevier (North-Holland) p. 112 (1983) with the additional assumption that Fowler-Nordheim tunneling is negligible, i.e., that 2xEox/ET<< 1 where ET is a trap depth. Taking Eox= MV/cm, x=5nm and [from E. Harari, S. Wang and B. Royce, J. Appl. Phys. 46, 1310 (1975)] ET>1.8 eV we compute qxEox/ET < 0.28 which indicates our assumption is not unreasonable.
14. The tunneling of trapped holes has also been discussed by Manzini and Modelli [cited in Ref. 13] in connection with tunneling of avalanche injected holes from oxides at room temperature.
15. Boesch and McGarrity [2] deduced their value of the hole transport distance d under the assumption that it was identical for hole motion toward the gate (negative bias) and away from the gate (positive bias). If this assumption is not made and if the positive and negative bias data are analyzed separately, then different shift distances are deduced (d+ and d-) as has been noticed previously by Srour and Chiu [8]. Several reasons for this may be suggested with perhaps the most plausible being based on the supposition of a higher trap density in the vicinity of the Si-SiO 2 interface. Preliminary experiments designed to test this involving implanted oxides appear to be consistent with such an hypothesis.
16. O.L. Curtis, Jr., J.R. Srour and K.Y. Chiu, J. Appl. Phys. 45, 4506 (1974).
17. As one might expect from our discussion of the limitations of Boesch and McGarrity's uniform shift model, the drift model, when expanded for large tox/d, reduces to the uniform shift model. Thus in the thick oxide regime d is approximately equal to d and the values obtained by Ref 2 using the uniform shift model are applicable to the drift model.
18. R.C. Hughes and C.H. Seager, IEEE Trans. Nucl. Sci. NS-30, 4049 (1983).
19. R.C. Hughes, Phys. Rev. Lett. 30, 1333 (1973).
20. K.M. Schlesier and C.W. Benyon, IEEE Trans. Nucl. Sci. NS-23, 1599 (1976).