Ion implants of 1 keV B+11 and 5 keV BF2+, to a dose of 1×1015/cm2 at a tilt angle of 0°, were implanted into preamorphized (Si+,70 keV, 1×1015/cm2) wafers. These samples were rapid thermal annealed in an ambient of 33 ppm of oxygen in N2 at very short times (<0.1 s spike anneals) at 1000 and 1050 °C to investigate the effects of the fluorine in BF2 implants on transient enhanced diffusion (TED). By using a relatively deep preamorphization of 1450 Å, any difference in damage between the typically amorphizing BF2 implants and the nonamorphizing B implants is eliminated because the entire profile (<800 Å after annealing) is well contained within the amorphous layer. Upon annealing, the backflow of interstitials from the end-of-range damage from the preamorphization implant produces TED of the B in the regrown layer. This allows the chemical effect of the fluorine on the TED of the B in the regrown Si to be studied independent of the damage. The secondary ion mass spectroscopy results show that upon annealing, the presence of fluorine reduces the amount of B diffusion by 30% for the 1000 °C spike anneal, and by 44% for the 1050 °C spike anneal. This clearly illustrates there is a dramatic effect of F on TED of B independent of the effects of implant damage. Analysis of the temperature dependence of the enhancement factors point to transient enhanced diffusion not boridation as the source of the interstitials.

Ion implants of 1.0 keV 11B+, 5 keV BF 2+, and 2.0 keV As+ at a dose of IeI5/cm2 were rapid thermal annealed (RTA) in a STEAG AST-2800µ with varying percents of oxygen in N2, ranging from 0-lppm to 50,000 ppm to investigate the effects of low concentrations of oxygen during anneal. Sheet resistance (Rs), ellipsometry, SIMS, Tapered Groove Profilometry (TGP), and Scanning Force Microscopy (SFM) were employed to characterize these layers. For each of these implant cases, an optimal RTA condition is established which maximizes retained dose while still producing shallow junctions. As a function of O2 content, anneal temperature and implant condition, three regimes are observed that affect after anneal retained dose. These regimes are: dopant loss to the ambient resulting from etching of Si, dopant loss by out-diffusion from evaporation/chemical reactions, a capping regime that minimizes out-diffusion. In this later regime the dopant loss results from consumption into the RTA grown oxide. In addition, this paper also discusses oxidation enhanced diffusion (OED) and identifies its extent as a function of temperature and O2 content of the anneal for the three implant conditions investigated. For example, a 1.0 keV 11B+wafer annealed at 1050°C lOs in a controlled 33 ppm of O2 in N2 yields a SIMS junction depth 320 Å shallower than previously reported by others.

For the formation of ultrashallow junctions, a controlled gaseous ambient during rapid thermal annealing is indispensible. To understand the diffusion/activation mechanism, the influencing and depending variables have to be clarified precisely. Ion implantations of 1 keV boron at a fluence of ⌽ Ϸ 1 ϫ 10 15 cm Ϫ2 are annealed isothermally for 10 s at 1000, 1050, and 1100ЊC in an AST2800⑀ rapid thermal processing system under controlled concentrations of oxygen in nitrogen ambient (0-1 ppm up to 1%). The concentration-depth profiles, measured by secondary ion mass spectroscopy, are analyzed within the framework of the kickout model involving diffusion enhancement via supersaturation of silicon self-interstitials and the Fermi-level effect. The validity of this interpretation is supported by the simulated results which are in good agreement with experimental data. Two input parameters for the SSUPREM IV simulator yield finite values of silicon self-interstitial supersaturation as a function of temperature and oxygen concentrations, values for the boron diffusion coefficient via neutral and positively charged silicon self-interstitials, and data for transient enhanced diffusion. After rapid thermal annealing for 10 s at 1050ЊC, the junctions vary within 800-1400 Å depending on the annealing ambient.

The effects of time, temperature, ramp-up, and ramp-down rates with rapid thermal annealing employing a STEAG AST SHS3000 were investigated on 1.0 and 2.0 keV 11 B + , 2.2, 5.0, and 8.9 keV 49 BF 2 + , and 2 keV 75 As + , 1E15/cm 2 samples implanted in a Varian VIISion-80 PLUS ion implanter at 0 o tilt angles. These annealed samples were analyzed by four-point probe, secondary ion mass spectrometry (SIMS), and in select cases by spreading resistance profiling (SRP) and transmission electron microscopy (TEM). To ensure reproducibility and to minimize oxidation enhanced diffusion as an uncontrolled variable, the O 2 background concentration in N 2 was maintained at a controlled low level. Under these conditions, ramp-rates alone were found not to be significant. Spike anneals (1050°C,~ 0 s) with fast ramp-rates (240°C/s) and fast cool down rates (86°C/s) provided the shallowest junctions, while still yielding good sheet resistance values. Post annealed samples were examined for extended defect levels (by TEM) and trapped interstitial concentrations. Fluorine concentration measurements were employed to qualitatively explain differences in the B diffusion from 11 B + and 49 BF 2 + ion implants at various energies. The 2.2 keV 49 BF 2 + "fast" spike annealed sample at 1050°C exhibited limited, if any, enhanced diffusion, yielding a SIMS junction depth of 490Å, an electrical junction of 386Å (by SRP) and a sheet resistance of 406 ohm/sq.

No abstract