Structural and magnetic properties of Mn-implanted Si

Structural and magnetic properties in Mn-implanted, $p$-type Si were investigated. High resolution structural analysis techniques such as synchrotron x-ray diffraction revealed the formation of $\mathrm{Mn}{\mathrm{Si}}_{1.7}$ nanoparticles already in the as-implanted samples. Depending on the Mn fluence, the size increases from $5\mathrm{nm}\text{to}20\mathrm{nm}$ upon rapid thermal annealing. No significant evidence is found for Mn substituting Si sites either in the as-implanted or annealed samples. The observed ferromagnetism yields a saturation moment of $0.21{\mu }_{\mathrm{B}}$ per implanted Mn at $10\mathrm{K}$, which could be assigned to $\mathrm{Mn}{\mathrm{Si}}_{1.7}$ nanoparticles as revealed by a temperature-dependent magnetization measurement.

Figure 1

RBS random and channeling along Si[001] spectra of as-implanted Si samples. Random (엯), channeling (+) for the fluence of $5\times {10}^{16}{\mathrm{cm}}^{-2}$; channeling (∎) for the fluence of $1\times {10}^{16}{\mathrm{cm}}^{-2}$; channeling (▵) for the fluence of $1\times {10}^{15}{\mathrm{cm}}^{-2}$; and channeling for virgin Si (—) which is slightly below triangle symbols.

Figure 2

RBS random and channeling along Si[001] spectra for sample implanted with a fluence of $5\times {10}^{16}{\mathrm{cm}}^{-2}$. Random (엯) and channeling (+) for as-implanted sample, channeling (—) for RTA sample, and channeling (▵) for virgin Si. Only a small fraction of damage was recovered by annealing. For Mn signals, no channeling was observed either for as-implanted or annealed samples, which means Mn does not substitute Si lattice site.

Figure 3

TEM-overview images show the implantation induced damage and nanosized precipitates after RTA: (a) $1\times {10}^{15}{\mathrm{cm}}^{-2}$, (b) $1\times {10}^{16}{\mathrm{cm}}^{-2}$, and (c) $5\times {10}^{16}{\mathrm{cm}}^{-2}$. The nanosized precipitates are visible as spotlike contrast, and they are growing with increasing fluence. The arrow marks the direction towards the sample surface.

Figure 4

High-resolution TEM images for representative $\mathrm{Mn}{\mathrm{Si}}_{\mathrm{x}}$ nanoparticles in (a) 1E16 RTA, (b) 5E16 RTA, and (c) the diffractogram of one nanoparticle in (b). The reflection spots marked by the arrows in (c) reveal a lattice spacing of around $1.7\mathrm{nm}$, which can be attributed to $\mathrm{Mn}{\mathrm{Si}}_{1.7}$.

Figure 5

(Color online) XRD grazing incidence scans of all investigated samples implanted with Mn. In some samples two peaks arise, which can be attributed to Mn-silicide nanoparticles $\left(\mathrm{Mn}{\mathrm{Si}}_{1.7}\right)$.

Figure 6

(Color online) $M\text{−}H$ hysteresis loops for Mn-implanted Si samples. Only the sample of 1E16 RTA exhibits a clear hysteretic behavior. Inset: ZFC (open circle) and FC (solid line) magnetization curves reveal a typical characteristic of a magnetic nanoparticle system for the sample of 1E16 RTA. Above the blocking temperature, the sample of 1E16 RTA shows superparamagnetic properties, where neither coercivity nor remanence was observed at $100\mathrm{K}$. None of the other samples (1E15 RTA, 1E15 as implanted, and 1E16 as implanted) exhibit ferromagnetism, even at temperatures down to $10\mathrm{K}$ (not shown).