Quality Factors of Ionizing Radiation in CMOS Transistor Gates

The absorbed dose (D) or total ionizing dose (TID) is used in radiation hardening (RadHard) studies for evaluating an electronic component’s susceptibility against ionizing radiation damages. While D is measured as the energy absorbed in a unit volume, some of the arisen questions are: 1) Can the TID, in relation to the energy deposition (defined as stopping power S, or linear energy transfer LET or L), be solely used to determine radiation damage for various types of radiation particles? 2) Are there other alternative quantities and/or qualities for RadHard analysis? 3) Is the effect (such as single event effect, SEE) of TID depending on the total irradiation time, i.e. the irradiation dose rate? To understand some of these questions, published data from the literatures are analyzed with parameters of radiation physics, the specific ionization number I
spec
and mean free path δx of radiation particles. As effects of ionizing radiation interaction with target materials are either directly through primary particles, or indirectly through secondary particles and even tertiary particles, the S and L for electrons are calculated as the reference. For transistor gate materials of CMOS, the radiation quality factors (Q(L)) vs LET, or weighting factors (W
R
) vs energy are assessed for their suitability. Hence, the I
spec
and δx values are calculated and used as ‘alternative quality factors’. The present and past calculation results of LET, I
spec
and δx have been used to correlate with the thickness t
g
(10–30nm of control and floating gates) in SLC NAND Flash in 51-nm IC node. Preliminary good correlation of δx w.r.t. gate thickness of SiO
2
is obtained though continued studies on the reliable IC and IP and new semiconductor device designs is suggested.