F. Hamadache et al.: Electrodeposition of Fe–Co alloys into nanoporous p-type silicon: Influence of the electrolyte composition
On the whole, our results indicate that the transport
mechanisms in porous p-type silicon are completely dif-
ferent depending on the kind of electrolyte used to fill the
pores: pure iron-based electrolyte or cobalt-based solu-
tions. This probably comes to the fact that the cathodic
standard potentials of these two metals are different and
might be discussed in light of a detailed description of the
charge transport within PS band structure at PS/electrolyte
and Si/electrolyte interfaces. This requires further inves-
tigations on the band structure of porous silicon formed
from a p-type silicon substrate.
hydride species. On the other hand, the metal deposition
process oxidizes the structure, and the oxygen detected
by AES in metallized PS is localized mainly at the pore
walls as seen by FTIR measurements. The present ex-
perimental results suggest that, in contact with pure Co
electrolyte, the porous skeleton is electrically isolated
from the bulk silicon, at least at the beginning of the
deposition process. On the contrary, PS is always
conductive when contacted by pure Fe electrolyte. Hence
our study provides further experimental data to analyze
the mechanism responsible for the incorporation of met-
als into the pores. A complete understanding of the depo-
sition process requires, however, further knowledge
about the transport mechanisms in the porous silicon and
the band bending at Si/electrolyte and PS/electrolyte in-
terfaces. These new results involving the deposition of
iron-group materials into cylindrical nanoporous p-type
silicon might be useful for future silicon technologies.
Some applications of metallized porous silicon net-
work, such as the synthesis of metal silicides, require
oxide-free systems. From the AES sputter depth profile it
has also been noticed that oxygen was present in our
samples up to a concentration of 5 at.%. FTIR analyses
allowed us to see that a considerable part of oxygen
content was localized at the pore walls, introduced dur-
ing the deposition process and intimately related to the pres-
2+
2+
ence of Fe and Co species in the solution. The presence
of silicon oxide between the electrodeposit and the pore
walls was not expected, since as-formed PS was found by
FTIR and AES analyses to be practically oxide-free and
is normally electrically protected even during cathodic
metal deposition. It has been moreover established, in the
literature, that a binary Si/metal system can be built with-
ACKNOWLEDGMENTS
This work performed at the PCPM laboratory was sup-
ported in part by the Secretariat a` la Coop e´ ration Inter-
nationale, Universit e´ Catholique de Louvain. Additional
experiments involving FTIR measurements were per-
formed at the LPC Laboratory, Institut des Matériaux de
Nantes. The authors gratefully acknowledge the supply
of implanted silicon wafers by V. Bayot, C. Renaux, and
B. Katschmarskyj. We thank R. Morlat and E. Ferain for
assistance with SEM and EDS analyses. We are obliged
to J-L. Delplancke, for useful discussions about the elec-
trochemical aspect of this work, and to Y. Baltog, for
fruitful discussions about porous silicon and FTIR char-
acterizations. F.H. also gratefully acknowledges encour-
agements from J-P. Michenaud.
6,40
out oxidation of PS.
In our samples, the oxidation
process of silicon may be associated either with the con-
finement by the deposit of oxygen present in the electro-
lyte that was not deaerated or possibly with electroless
deposition which occurred certainly at the beginning of
4,6,41
the deposition.
V. SUMMARY AND CONCLUSION
This work demonstrates that it is possible to achieve
homogeneous filling of porous p-type silicon with elec-
trodeposited pure Co metal and Fe–Co alloys. The de-
termination by EDS of the average alloy composition
versus the related bath composition shows an unexpected
deficit in iron. This deficit instead of an excess in less
noble metal is explained by the fact that Co deposition is
electrically promoted into the pores as revealed by the
AES measurements. Auger study shows that pure Co
deposition starts at the pore bottom, while the nucleation
of pure Fe occurs all over the pore walls leading to a
preferential deposition on the top surface of the porous
layer. In the mixed solutions, the reaction rate of iron
inside the pore is catalyzed by cobalt and accounts for a
filling of the pores with Fe–Co alloys. Indeed, a small
amount (5 at.%) of cobalt in the solution is enough to
cause iron nucleation to start at the pore tips.
REFERENCES
1. R. Hérino, Mater. Sci. Eng., B 69–70, 70 (2000).
2
3
4
5
6
7
8
9
. C. Renaux, V. Scheuren, and D. Flandre, Micrcelectron. Reliab.
0, 877 (2000).
. X.J. Li, De L. Zhu, Q.W. Chen, and Y.H. Zhang, Appl. Phys. Lett.
4, 389 (1999).
. F. Ronkel, J.W. Schultze, and R. Arens-Fischer, Thin Solid Films
276, 40 (1996).
. B.J. Aylett, I.S. Harding, L.G. Earwaker, K. Forcey, and T. Giaddui,
Thin Solid Films 276, 253 (1996).
. M. Jeske, J.W. Schultze, M. Thönissen, and H. Münder, Thin
Solid Films 255, 63 (1995).
4
7
. T. Ito, T. Yoneda, K. Furuta, A. Hatta, and A. Hiraki, Jpn. J. Appl.
Phys. 34, L649 (1995).
. P. Steiner, F. Kozlowski, and W. Lang, Thin Solid Films 255, 49
(1995).
. S.O. Izidinov, V.N. Nazarov, and O.E. Shcheglov, Russ. J. Appl.
Furthermore, it is deduced from both AES and FTIR
studies that almost no trace of oxygen is found in the
as-formed PS and the pore walls are covered with
Chem. 68 (Part 1), 519 (1995).
1
0. S.A. Gusev, N.A. Korotkova, D.B. Rozenstein, and A.A. Fraerman,
J. Appl. Phys. 76, 6671 (1994).
J. Mater. Res., Vol. 17, No. 5, May 2002
1083
Downloaded: 04 Feb 2015
IP address: 194.27.18.18