Chemistry Reference
In-Depth Information
A hole needs to hit and stick to the surface to have a reaction at the active site. The
pore tip was considered to be oxidized to a varying degree depending on the anodiza-
tion conditions. The sticking factor is different along the surface of a pore tip which is
partially covered with an oxide. When conditions are such that the sticking probability
is much higher at the edge of the tip than in the middle, the dissolution process tends
to result in electropolishing rather than PS formation.
Although the model simulated some of the morphological features of PS, it was
too genera
to account for the different current conduction mechanisms for different
types of silicon substrates. For example, for
n
-Si in the dark, the current is by electron
injection into the conduction band from the surface, which cannot be explained by the
carrier diffusion model. Also, there is no physical foundation for the sticking factor of
carrier to the surface. Furthermore, the model did not consider the nature of the elec-
trochemical reactions at the interface, which must be an important part of the forma-
tion mechanism.
Formation Condition of PS.
A critical current density (the peak current of the
i-V
curve), below which PS forms and above which it does not, was identified by
Turner's early study. However, it was not clear how accurate this current characterizes
the formation condition and whether it is identical for different types of silicon sub-
strates. Zhang
et al.
2
in the late 1980s made a systematic study on the
i-V
curves of
different silicon types and doping concentrations, current densities, and HF concentra-
tions and on the electrode surface condition after anodization. The condition for PS for-
mation on different substrates in the entire continuum of current and HF concentration
was then established (Fig. 8.5). It was concluded that the conditions determining
whether PS formation or electropolishing occurs is largely independent of the electronic
properties of silicon such as doping type and concentration. It is the nature of the reac-
tions that is responsible for the occurrence of the different regions of potential. The
transition from PS formation to electropolishing was postulated to result from the two
competing reactions between a direct dissolution of silicon through reacting with
fluoride species and an indirect dissolution through oxide formation and dissolution.
This hypothesis has been adopted in many of the later models on the formation mech-
anisms of PS.
12,859,1087,1135
Quantum Confinement Model.
To account for the formation of micropores of
less than a few nanometers formed on
p
-Si, Lehmann and Gosele
71
in the early 1990s
postulated that instead of the depletion layer, which is involved in macropores,
quantum carrier confinement is responsible for the formation of the micropores on
p
-Si. The confinement occurs due to an increase in band-gap energy and energy
barrier caused by the quantum size porous structure, which prevents the carriers from
entering the wall regions of the PS as illustrated in Fig. 8.61.
25
Due to the quantum
confinement the pore walls are depleted of carriers and thus do not dissolve during the
anodization.
The quantum confinement model was extended by Frohnhoff
et al.
133
to account
for the wide distribution of pore diameters of the PS formed on
p
-Si. Tunneling of holes
through silicon crystallites was proposed as a process also involved in the formation of
the quantum size porous structure. The tunneling current oscillates with the crystallite
size which was considered to be responsible for the uneven pore size distribution and
for the stability of very small crystallites in the PS. The quantum confinement model
l
