90
4 Ultrathin Layers
spectroscopy (XPS) are both suitable to check the composition as long as the thickness is within the depth sensitivity range of the electron spectroscopic analysis. In
addition, electron spectroscopy is also sensitive to the chemical state of the surface,
and the desired layer chemistry can also be checked. Both AES and XPS have been
applied successfully to characterize EC-ALD layers since the introduction of the
EC-ALD concept [7, 20, 63], and even electrochemical cells directly connected to
the antchamber of the high-vacuum equipment have been elaborated [56].
The electrochemical stripping analysis method can also be made suitable for
the analysis of the deposit after a few EC-ALD cycles. The discussion of these
procedures below closely follows the description of Innocenti el al. [21, 29, 51,
73, 75]. The forthcoming discussion refers to binary deposits for sake of simplicity
but the procedure was shown to work also for ternary deposits with two metallic
components [76]. For analysing the charge corresponding to the stripping of either
of the components in a binary deposit, one needs two series of samples, and each
of them will be used for the quantitative assessment of one component only. As
a first step of the experiment, one of the components has to be stripped off from
the EC-ALD deposit. The metallic component can be stripped off with an anodic
treatment and vice versa, the chalcogenide component is stripped off as negative ions
produced with a cathodic treatment. It is obvious that the atoms remaining on the
surface have to be rearranged and will behave thereafter as a deposit formed from
a single element. Then, the solution containing the ions of the already dissolved
component is removed, and it may also be necessary that a solution with dissimilar
composition should be used for dissolving the component still on the surface of
the working electrode. During the rinsing/solution exchange process, the electrode
potential has to be well controlled so that the coating component still present remains
intact. Then, a linear sweep is run in the appropriate direction in which the charge
corresponding to the stripping peak is proportional to the amount of the deposit.
This study has to be carried out with two series of samples, each series containing
samples with different EC-ALD cycle numbers. A representative set of curves for
the stripping analysis of CuS layers is shown in Fig. 4.5.
Attention has to be drawn to a few features of the stripping curves of EC-ALD
deposits. When the chalcogenide component (mostly sulphur) is stripped off with a
linear potential scan, two peaks are observed. If the curves are regarded in the sweep
direction, the surface area of the first peak changes with EC-ALD cycle number but
the second does not, should the second be more negative [29, 51, 73, 76] or more
positive [21]. This is because the stripping treatment first removes the bulk of the
deposit, and the UPD layer of the deposit can be dissolved only later. This is in
good agreement with the fact that sulphur can form a UPD layer on the substrate
used in the above-mentioned studies. The curves shown in Fig. 4.5 have such a
character. However, stripping of the metallic component exhibits one stripping peak
in cases when this metal does not tend to form a UPD layer on the substrate used
(In [21], Pb [73] and Cu [29] on Ag(111), also the curves in Fig. 4.5). However, if
the metallic component of the EC-ALD deposit also exhibits UPD on the substrate,
a voltammogram with two peaks can be obtained for its oxidative stripping with the
4 Ultrathin Layers
spectroscopy (XPS) are both suitable to check the composition as long as the thickness is within the depth sensitivity range of the electron spectroscopic analysis. In
addition, electron spectroscopy is also sensitive to the chemical state of the surface,
and the desired layer chemistry can also be checked. Both AES and XPS have been
applied successfully to characterize EC-ALD layers since the introduction of the
EC-ALD concept [7, 20, 63], and even electrochemical cells directly connected to
the antchamber of the high-vacuum equipment have been elaborated [56].
The electrochemical stripping analysis method can also be made suitable for
the analysis of the deposit after a few EC-ALD cycles. The discussion of these
procedures below closely follows the description of Innocenti el al. [21, 29, 51,
73, 75]. The forthcoming discussion refers to binary deposits for sake of simplicity
but the procedure was shown to work also for ternary deposits with two metallic
components [76]. For analysing the charge corresponding to the stripping of either
of the components in a binary deposit, one needs two series of samples, and each
of them will be used for the quantitative assessment of one component only. As
a first step of the experiment, one of the components has to be stripped off from
the EC-ALD deposit. The metallic component can be stripped off with an anodic
treatment and vice versa, the chalcogenide component is stripped off as negative ions
produced with a cathodic treatment. It is obvious that the atoms remaining on the
surface have to be rearranged and will behave thereafter as a deposit formed from
a single element. Then, the solution containing the ions of the already dissolved
component is removed, and it may also be necessary that a solution with dissimilar
composition should be used for dissolving the component still on the surface of
the working electrode. During the rinsing/solution exchange process, the electrode
potential has to be well controlled so that the coating component still present remains
intact. Then, a linear sweep is run in the appropriate direction in which the charge
corresponding to the stripping peak is proportional to the amount of the deposit.
This study has to be carried out with two series of samples, each series containing
samples with different EC-ALD cycle numbers. A representative set of curves for
the stripping analysis of CuS layers is shown in Fig. 4.5.
Attention has to be drawn to a few features of the stripping curves of EC-ALD
deposits. When the chalcogenide component (mostly sulphur) is stripped off with a
linear potential scan, two peaks are observed. If the curves are regarded in the sweep
direction, the surface area of the first peak changes with EC-ALD cycle number but
the second does not, should the second be more negative [29, 51, 73, 76] or more
positive [21]. This is because the stripping treatment first removes the bulk of the
deposit, and the UPD layer of the deposit can be dissolved only later. This is in
good agreement with the fact that sulphur can form a UPD layer on the substrate
used in the above-mentioned studies. The curves shown in Fig. 4.5 have such a
character. However, stripping of the metallic component exhibits one stripping peak
in cases when this metal does not tend to form a UPD layer on the substrate used
(In [21], Pb [73] and Cu [29] on Ag(111), also the curves in Fig. 4.5). However, if
the metallic component of the EC-ALD deposit also exhibits UPD on the substrate,
a voltammogram with two peaks can be obtained for its oxidative stripping with the