Majeed A. Foad

Optimizing boron junctions through point defect and stress engineering using carbon and germanium co-implants

We report the fabrication of p+∕n junctions using Ge+, C+, and B+ co-implantation and a spike anneal. The best junction exhibits a depth of 26nm, vertical abruptness of 3nm∕decade, and sheet resistance of 520Ohm∕square. The junction location is defined by where the boron concentration drops to 1018cm−3. These junctions are close to the International Technology Roadmap specifications for the 65nm technology node and are achieved by careful engineering of amorphization, stresses, and point defects. Advanced simulation of boron diffusion is used to understand and optimize the process window. The simulations show that the optimum process completely suppresses the transient-enhanced diffusion of boron and the formation of boron-interstitial clusters. This increases the boron solubility to 20% above the equilibrium solid-state solubility.

Deactivation of ultrashallow boron implants in preamorphized silicon after nonmelt laser annealing with multiple scans

Electrical activation and redistribution of 500eV boron implants in preamorphized silicon after nonmelt laser annealing at 1150°C and isochronal rapid thermal postannealing are reported. Under the thermal conditions used for a nonmelt laser at 1150°C, a substantial residue of end-of-range defects remained after one laser scan but these were mainly dissolved within ten scans. The authors find dramatic boron deactivation and transient enhanced diffusion after postannealing the one-scan samples, but very little in the five- and ten-scan samples. The results show that end-of-range defect removal during nonmelt laser annealing is an achievable method for the stabilization of highly activated boron profiles in preamorphized silicon.

Large enhancement of boron solubility in silicon due to biaxial stress

One of the important challenges to the semiconductor industry today is to enhance the solid solubility of several dopants, boron in particular, in silicon. We calculate the equilibrium solid solubility of boron in Si from first principles and examine the effect of biaxial stress. We find an unexpectedly large enhancement, on the order of 150%, for only 1% strain primarily due to the charge of the substitutional boron impurity in Si. We point out that this effect is an intrinsic property of Si and is expected to be important for other dopants as well.

Ultrashallow junctions in Si using decaborane? A molecular dynamics simulation study

The feasibility of using decaborane B10H14, for the manufacture of shallow junctions in Si is investigated by means of molecular dynamics simulations. Bombardment energies of 1, 2, and 4 keV are investigated and the simulations run for up to 7 ps in order to ascertain the implantation profiles of the B atoms, the whereabouts of the H from the impacted molecule and the damage to the lattice. The simulations show that if a small binding energy of the B atom in the Si lattice is assumed then most of the B from the cluster is implanted. The implantation distributions are flatter with depth than those for single B interactions and the surface layers undergo damage and amorphisation in the proximity of the impact.

200 eV–10 keV boron implantation and rapid thermal annealing: Secondary ion mass spectroscopy and transmission electron microscopy study

Atomic profiles (secondary ion mass spectroscopy) and cross-section transmission electron microscopy (TEM) images of selectively etched, annealed profiles were studied for boron energies from 200 eV to 10 keV and rapid thermal processing anneals at 900, 975, and 1050 °C. Consistent variations of dopant depth were obtained over this process range. TEM images showed evidence of lateral dopant variation near the edges of poly-Si gate structures, perhaps an effect of lateral straggling and reflection of ions from the polymask.

Conformal Doping of FINFETs: a Fabrication and Metrology Challenge

This article deals with the developments in the measurement and identification of conformality which is a key function in conformal doping. For this purpose this paper extensively uses SSRM to characterize the vertical/lateral junction depths, concentration levels and degree of conformality. As a complement to the SSRM technique this paper developes a concept based on resistance measurements of fins which allows to map the sidewall doping across the wafers and provides fast feedback on conformality. The concept uses the reduction of the sheet resistance of a fin which was covered with a hardmask during the implantation, as a measure for the degree of side wall doing. The concept is supported by theoretical simulations and verified using tilted implants.

Elemental B distributions and clustering in low-energy B+ ion-implanted Si

A detailed study is presented of characteristic elemental B distributions in Si produced by low-energy B+ ion implantation and annealing. Implant concentration profiles have been determined with approximately nanometer spatial resolution using energy-filtered imaging in the transmission electron microscope, for a B+ ion dose close to those relevant to electronic device processing. It is demonstrated that, for as-implanted Si, the near-surface B distribution shows a smooth concentration peak which correlates well with theoretical simulation and shows no anomalous surface buildup of the type generally indicated by secondary ion mass spectrometry measurements. After annealing of the layers, the present direct observations reveal that the final B distribution is characterized by residual nanometer-scale elemental clusters which comprise disordered zones within the restructured Si lattice.

Experimental evidence of B clustering in amorphous Si during ultrashallow junction formation

The authors have investigated ultrashallow p+∕n-junction formation by solid-phase epitaxy, by using x-ray absorption near-edge spectroscopy measurements on the B K edge. A clear fingerprint of B–B clusters is detected in the spectra. The authors demonstrate that B clustering occurs during the very early stages of annealing-induced Si recrystallization, i.e., when B is still in an amorphous matrix. After complete regrowth the local structure around B remains the same as in the amorphous phase, implying that B clusters are transferred to the crystalline structure.

A study of low energy phosphorus implantation and annealing in Si:C epitaxial films

The effect of phosphorus implantation and thermal annealing on properties of Si:C epitaxial films was investigated. High resolution x-ray diffraction analysis and secondary ion mass spectroscopy indicated that spike annealing only causes slight loss of substitutional carbon. Phosphorus implantation, even with low energy, could cause surface damages and loss of substitutional carbon. Although spike annealing effectively activates implanted phosphorus, it also results in significant substitutional carbon loss (from 1.2% to less than 0.5%) within the phosphorus diffused layer. The interaction of carbon and phosphorus resulted in a junction profile as abrupt as with 3 nm/decade.

Process interactions between low-energy ion implantation and rapid-thermal annealing for optimized ultrashallow junction formation

The shallow doping requirements for the next 2–3 device generations can be satisfied by a combination of low-energy ion implantation and rapid-thermal anneal. However, the differing requirements of distinct types of devices preclude the definition of a single optimized process. To tailor the junction properties according to device type and geometry, requires an understanding of the effects of process parameters in both implant and anneal steps. In describing the interactions and mechanisms behind this optimization, a number of tradeoffs are highlighted: (i) The choice of implant energy and dose may be traded off against the anneal time–temperature profile. (ii) The benefits of preamorphization to reduce ion channeling are offset by the detrimental increase in transient-enhanced diffusion and dopant segregation. (iii) The use of oxygen in the anneal ambient is discussed in terms of its effects on diffusion versus dopant loss at the surface.