Profile

 

Alloy scattering of substitutional carbon in silicon: a first principles approach

 

Overview

The introduction of carbon into SiGe/Ge heterostructures is of technological interest due the dual effects of strain-compensation in the Si(1-y)Ge(y) layers [1] and suppression of the out-diffusion of p-type acceptors, in particular boron [2] and indium [3], during wafer fabrication. The transport properties of the material are then expected to be modified by two competing processes. On the one hand, in silicon, the induced strain will lift the degeneracy of the Δ valleys allowing the conduction electrons to see a smaller effective mass. On the other hand, the substitutional carbon is likely to introduce alloy scattering, acting to reduce the mobility.

A method is developed to obtain the alloy scattering matrix from density functional theory calculations based on a similar approach to previous first principles work [4, 5]. It is found that the scattering matrix can be decomposed into two additive components: a chemical part due to pseudopotential substitution and a part due to ionic relaxation that we attribute to the deformation potential. The method is then applied to find the intravalley and intervalley electron scattering matrix for substitutional carbon in silicon. It is found that intravalley scattering is the dominant process. However, we find that alloy scattering due to substitutional carbon is too weak to reduce the mobility to the degree observed in strained Si(1-x)C(x) layers [6] and therefore support the conclusion that this degradation is due to interstitial carbon complexes.

Related people

    Philip Murphy
    Stephen Fahy

Related publications

    First-principles investigation of the alloy scattering potential in dilute Si(1-x)C(x),
    M.P. Vaughan, F. Murphy-Armando and S. Fahy,
    Phys. Rev. B 85, 165209 (2012)
    doi: 10.1103/PhysRevB.85.165209

    Alloy scattering of substitutional carbon in silicon: A first principles approach,
    MP Vaughan, F Murphy-Armando and S Fahy,
    Ultimate Integration on Silicon (ULIS) (2011)
    doi: 10.1109/ULIS.2011.5757983

    First principles calculations of the scattering cross section of substitutional carbon in silicon,
    MP Vaughan and S Fahy,
    Journal of Physics: Conference Series 242, 012003 (2010)
    doi: 10.1088/1742-6596/242/1/012003

Related presentations

References

[1] S.C. Jain, H.J. Osten, B. Dietrich and H. Rücker, Semicond. Sci. Tech. 10, 1289 (1995)
[2] H.J. Osten, H. Rücker, J.P. Liu and B. Heinemann, Microelectron. Eng. 56, 209 (2001)
[3] C.F. Tan, E.F. Chor, J. Liu, H. Lee, E. Quek and L. Chan, Appl. Phys. Lett. 83, 4169 (2003)
[4] F. Murphy-Armando and S. Fahy, Phys. Rev. Lett. 97, 96606 (2006)
[5] F. Murphy-Armando and S. Fahy, Phys. Rev. B 78, 35202 (2008)
[6] H.J. Osten and P. Gaworzewski, J. Appl. Phys. 82, 4977 (1997)
[7] K. Eberl, K. Brunner and W. Winter, Thin Solid Films 294, 98 (1997)

 
Fig. 1: Experimental mobilities for strained Si1-xCx and the calculated alloy scattering-limited mobilities. The mobility for unstrained Si is shown as clear circles and clear triangles. For 0.4% C, the work of Eberl et al [7] (filled diamonds) shows an increase in the mobility. The calculated alloy scattering-limited mobility for this concentration is shown in comparison (solid line). The work of Osten et al [6] shows a degradation of the mobility, even for lower C concentrations. Here we show their results for 0.31% (filled circles) and 0.57% (filled squares). For comparison, the calculated alloy scattering-limited mobility for 0.31% is also shown (dashed line).
(Click to see larger image)