We report Al 2 O 3 /In 0.53 Ga 0.47 As MOSFETs having both self-aligned in situ Mo source/drain ohmic contacts and self-aligned InAs source/drain n + regions formed by MBE regrowth. The device epitaxial dimensions are small, as is required for 22-nm gate length MOSFETs; a 5-nm In 0.53 Ga 0.47 As channel with an In 0.48 Al 0.52 As back confinement layer and the n ++ source/drain junctions do not extend below the 5-nm channel. A device with 200-nm gate length showed I D = 0.95 mA/μm current density at V GS = 4.0 V and g m = 0.45 mS/μm peak transconductance at V DS = 2.0 V. Index Terms-InAs source/drain, InGaAs MOSFET, migration-enhanced epitaxial regrowth, source/drain regrowth, III-V MOSFET.

The authors report ultralow specific contact resistivity (ρc) in nonalloyed, in situ Ohmic contacts to heavily doped n-type In0.53Ga0.47As:Si layers with 6×1019cm−3 active carrier concentration, lattice matched to InP. The contacts were formed by depositing molybdenum (Mo) immediately after the In0.53Ga0.47As growth without breaking vacuum. Transmission line model measurements showed a contact resistivity of (1.1±0.6)×10−8Ωcm2 for the Mo∕InGaAs interface. The contacts show no observable degradation in resistivity after annealing at 300 and 400°C for 1min duration.

We calculate the minimum feasible contact resistivity to n-type and p-type In 0.53 Ga 0.47 As, InAs, GaAs, GaSb, InP, and InSb. The calculations consider image force lowering and assume either parabolic or non-parabolic energy dispersion in the semiconductor; their results are compared with recent experimental data. Among significant results, the measured contact resistivity to n-In 0.53 Ga 0.47 As at a carrier concentration of 5 Â 10 19 cm À3 is only 2.3:1 higher than that calculated assuming a 0.2 eV barrier potential, and the measured contact resistivity is only 9.0:1 larger than the Landauer quantum conductivity limit at this carrier concentration. These results indicate that, with the surface preparation procedures presently employed, surface contamination does not markedly increase the interface resistance, and that the transmission coefficient for carriers crossing the interface exceeds 10%. V

FeCo alloy nanoparticles have been prepared by using a two step modified polyol process using Fe͑II͒ chloride and Co acetate tetrahydrate as Fe and Co metal precursors. Tetraethyl silicate, aluminum isopropoxide, and zirconium͑IV͒ acetyl acetonate were used to make amorphous SiO 2 , Al 2 O 3 , and ZrO 2 coatings, respectively. X-ray diffraction studies showed that there are no crystalline peaks corresponding to SiO 2 , Al 2 O 3 , and ZrO 2 because the oxide coatings of the FeCo core are amorphous in nature. The scanning electron micrograph analysis depicted the cubic nature of the particles with mean particle size of about 45 nm. The maximum saturation magnetization of 205 emu/ g was achieved at 300 and 4 K. FeCo nanocomposites were screen printed as films and aligned by using an external magnetic field of 10 kOe. The microwave properties measured by in-plane ferromagnetic resonance at various frequencies indicate a minimum linewidth of Ϸ3700 Oe.

The intrinsic lower limit of contact resistivity (q LL c) for InAs, In 0:53 Ga 0:47 As, GaSb, and Si is calculated using a full band ballistic quantum transport approach. Surprisingly, our results show that q LL c is almost independent of the semiconductor. An analytical model, derived for 1D, 2D, and 3D, correctly reproduces the numerical results and explains why q LL c is very similar in all cases. Our analysis sets a minimal carrier density required to meet the International Technology Roadmap for Semiconductors call for q c ¼ 10 À9 X-cm 2 by 2023. Comparison with experiments shows there is room for improvement, which will come from optimizing interfacial properties.