Фазовая диаграмма системы H-Si
К оглавлению: Другие диаграммы (Others phase diargams)
H-Si (Hydrogen-Silicon)
A. San-Martin and F.D. Manchester
No equilibrium phase diagram was found in the literature for the Si-H system.
The assessed T-X diagram at atmospheric pressure is based on the limited
quantitative attempts to determine H solubility in Si [56Wie, 70Koz]. The
boundaries, except for those between solid and liquid Si, represent present
limits on the maximum observed H solubility and do not necessarily indicate
the range of existence of equilibrium phases.
The left side of the boundary shown in the assessed diagram for solid Si
represents absorption of H at low concentration, but the boundary itself does
not necessarily represent the limits of the a phase, or the terminal solid
solubility curve as it has been traditionally called. Nonetheless, this curve
is based on analytical approximations of [56Wie] and [70Koz] to yield the
maximum observed solubility at atmospheric pressure.
The solubility of H in solid (Si) is less than in liquid Si [70Koz]; thus, the
phase diagram should contain a eutectic-type reaction near the melting point
of Si (1414 C) [Melt].
The invariant temperature for the (L = (Si) + H2) reaction is unknown, as are
the (Si)/[(L) + (Si)] and [(L) + (Si)]/(L) boundaries. However, because the
dissolved quantities of H are small, the net effect of lowering the melting
temperature of Si should be barely measurable, and the invariant temperature
should be very close to that melting point. The compositions of the liquid and
the solid phase of the three-phase equilibrium were estimated as 8.7 x 10-2
and 1.20 x 10-2 at.% H, respectively.
In discussing the absorption of H by Si, it is necessary to distinguish
between absorption in crystalline Si and that in amorphous Si. Amorphous Si is
included because of the recent upsurge of technological interest in this
material.
Hydrogen absorption in crystalline Si was determined by [56Wie] in a single
crystal of Si at 967 to 1207 C at pressures between 10 and 100 kPa to yield a
solubility of 1.98 x 1015 hydrogens per cubic centimeter of Si at 1200 C, or
expressed as the atomic ratio (X = H/Si), an X value of 3.96 x 10-8 (~3.96 x
10-6 at.% of H). The term "hydrogens" is used to denote a screened proton
inside the Si.
The hydrogen solubility in crystalline Si can be increased beyond the limits
of a chemiadsorbed layer by hydrogen-ion implantation [63Bec]. Implanting
crystalline Si with H (or D) has revealed (by infrared absorption spectroscopy)
numerous H-associated bands [83Car]. The multiplicity of the observed lines
and their rearrangement with thermal annealing [75Ste] suggest that lattice
defects are associated with the implanted H.
The only demonstration that implanted H is stable in crystalline Si at room
temperature is from the comparison of a H profile (on a single-crystal wafer)
immediately after implantation of 3 x 1016 H+ ions at 7.5 keV with a profile
taken 30 days later [76Lan]. This result is likely to be strongly influenced
by the presence of lattice defects resulting from the implantation;
consequently, it does not provide a clear statement about the diffusive motion
of H in Si at room temperature.
The incorporation of H into amorphous Si substantially modifies the electrical
and optical properties of this material [75Spe], making the amorphous Si:H
alloys (also called silicon hydrides or hydrogenated amorphous silicon)
attractive for technological applications.
In films prepared either by decomposition [79Kni] or by sputtering [81Ros],
evidence of periodic density fluctuations was found.
The existence of molecular H (H2) clustered in the microvoids of the a-Si:H
films has been deduced from nuclear magnetic resonance [80Car, 81Con, 82Leo]
and calorimetric experiements [84Gra, 84Loh, 85Gra]. [84Cha] and [85Gra]
estimated the maximum and the minimum number of H2 molecules trapped in a void
from infrared spectra studies and calorimetric measurements, respectively, to
be between 100 and about 10 or 20 H2 molecules.
56Wie: A. Van Wieringen and N. Warmoltz, Physica, 22, 849-865 (1956).
63Bec: G.E. Becker and G.W. Gobeli, J. Chem. Phys., 38, 2942-2945 (1963).
70Koz: T.K. Kostina, B.A. Baum, and K.T. Kuroschkin, Akad. Nauk SSSR. Izv.
Neorg. Mater., 6, 117 (1970) in Russian.
75Spe: W.E. Spear and P.G. LeComber, Solid State Commun., 17, 1193-1196 (1975).
75Ste: H.J. Stein, J. Electron. Mater., 4, 159-173 (1975).
76Lan: W.A. Lanford, H.P. Trautvetter, J.F. Ziegler, and J. Keller, Appl. Phys.
Lett., 28, 566-568 (1976).
79Kni: J.C. Knights, G. Lucovsky, and R.J. Nemanich, J. Non-Cryst. Solids, 32,
393-403 (1979).
80Car: W.E. Carlos and P.C. Taylor, Phys. Rev. Lett., 45, 358-362 (1980).
80Min: S. Minomura and K. Tsuji, J. Non-Cryst. Solids, 35/36, 513-518 (1980).
81Con: M.S. Conradi and R.E. Norberg, Phys. Rev. B, 24, 2285-2288 (1981).
81Ros: R.C. Ross and R. Messier, J. Appl. Phys., 52, 5329-5339 (1981).
82Leo: D.J. Leopold, J.B. Boyce, P.A. Fedders, and R.E. Norberg, Phys. Reu B,
26, 6053-6066 (1982).
83Car: M. Cardona, Phys. Status Solidi (b), 118, 463-481 (1983).
84Cha: Y.J. Chabal and C.K.N. Patel, Phys. Rev. Lett., 53, 1771-1774 (1984).
84Gra: J.E. Graebner, B. Golding, L.C. Allen, D.K. Biegelson, and M. Stutzmann,
Phys. Rev. Lett., 52, 553-556.
84Loh: H. v. Lohneysen, H.J. Schink, and W. Beyer, Phys. Rev. Lett., 52, 549-
552 (1984).
85Gra: J.E. Graebner, L.C. Allen, and B. Golding, Phys. Rev. B, 31, 904-912 (
1985).
85Tat: J. Tatarkiewicz and A. Krol, Phys. Rev. B, 32, 8401-8404 (1985).
Submitted to the APD Program. Complete evaluation contains 2 figures, 2 tables,
and 48 references.
1
