Mark N. Kawaguchi

Layout impact on the performance of a locally strained PMOSFET

We present a study on the layout dependence of a SiGe S/D PMOSFET technology. While 65% increase in drive current is obtained for 45nm gate length transistors with large active areas, measurements and simulations show that this improvement may be seriously degraded when transistor dimensions, such as the source-drain length (L/sub s/d/) and the device width are further scaled. TDDB and NBTI measurements show that the oxide reliability is not degraded for this technology.

A systematic study of trade-offs in engineering a locally strained pMOSFET

We present the results of a study on the impact of process parameters on the performance of strain enhanced pMOSFETs with recessed SiGe S/D. Recess depth, channel length, layout sensitivity, and their subsequent impact on strain and hole mobility are explored. Micro-Raman spectroscopy (/spl mu/RS), process simulations, device simulations, and electrical results are presented. A 30% improvement in drive current is demonstrated.

Surface characterization of ion-enhanced implanted photoresist removal

We characterize the chemical constitutents of high dose implanted deep ultraviolet photoresist before and after dual-mode oxygen plasma processing, where a remote rf-plasma source is operated simultaneously with rf bias. Raman spectroscopy indicates that the organic composition of the crust comprises a mixture of sp2 graphite and sp3 diamondlike carbon structures. High dose ion implantation reduces the hydrogen content by about 50 at. % as measured by hydrogen forward scattering and explains the reduced optical emission signal intensity observed during crust removal. The crust thicknesses extracted from the secondary-ion-mass spectroscopy profile correspond well to prior scanning electron microscopy characterization [Kawaguchi et al., J. Vac. Sci. Technol. B (submitted)] and support the existence of a transitional layer between the hardened crust and the underlying photoresist. Angle-resolved x-ray photoelectron spectroscopy analysis of arsenic implanted photoresist shows that dual-mode plasma processing ...

Low temperature, ion-enhanced, implanted photoresist removal

We investigate the synergistic effects of a remote rf oxygen plasma source operated simultaneously with rf bias at low temperature and find that the photoresist removal rates substantially exceed those observed with either plasma source alone. When applied to high dose ion-implanted photoresist, optical emission curves show a transition between the hydrogen-devoid crust, where oxygen radical consumption is reduced, and the underlying bulk photoresist, where oxygen radical consumption is substantial. Differences between the reactant and product optical emission traces are indicative of complex mechanisms during the crust removal, which are likely related to crust inhomogeneities observed from scanning electron microscopy (SEM) analysis. Crust removal times are as short as 13–16 s, crust removal rates are 1–1.3μm∕min or about half the removal rate of unimplanted photoresist, and crust density estimates are similar to the density of graphite. Because photoresist removal rates are comparable to conventional h...

Surface modification and contamination characterization of ion-enhanced, implanted photoresist removal

The authors study the dopant contamination removal effectiveness of blanket high-dose implanted photoresist wafers exposed to dual mode oxygen plasma, where a remote rf-plasma source is operated simultaneously with rf bias at low temperature. Based on total x-ray reflectance x-ray fluorescence spectrometry and secondary ion mass spectrometry analyses, sputtering effects are minimal during typical process times with most of the dopant remaining on the substrate after dual mode oxygen plasma processing. SC1-based (NH4OH:H2O2:H2O Standard Clean 1) wet clean treatments reduce dopant levels but 2×1013–4×1013at∕cm2 still remain unless the substrate is processed with hydrofluoric acid (HF) wet cleanings where the contamination level is less than 2×1011at∕cm2. Remote plasma low temperature fluorine post-treatments volatilize dopant residues but are less effective than HF wet cleanings. Dual mode forming gas (4%H2∕96%N2) plasma post-treatments are ineffective for dopant contamination removal. Low temperature dual ...