Fig. 4 XPS spectra of Fe nanocrystals deposited on Au(111) at 0.4 V from
an ionic melt {AlCl3–[MBIm]1Cl2} 1 5 mM Fe(II); (– – –) before
sputtering, (—) after 5 min sputtering.
acknowledged. C. L. A thanks the Alexander-von-Humboldt
foundation for a fellowship.
Fig. 3 STM image (160 6 160 nm2) of monodispersed Fe nanocrystals.
E ~ 0.4 V, Etip ~ 0.7 V, Itunn ~ 1 nA.
Notes and references
{ The method used to synthesize [MBIm]1Cl2 from 1-ethylimidazole and
1-chlorobutane (Merck) and the purification by recrystallization have been
described earlier.11 Anhydrous AlCl3 (Fluka, w99%) was purified by
adding 2 mass% of NaCl and 0.2 mass% of Al and heating this mixture in
an evacuated quartz ampoule up to 200 uC for several hours followed by
the distillation of AlCl3. The AlCl3 crystals were slowly added to
[MBIm]1Cl2 in a molar ratio of 58:42. Fe(II) was introduced into the
{AlCl3–[MBIm]1Cl2} melt by controlled potential-coulometric dissolution
of an Fe wire (Alfa, w99.99%, 0.25 mm) up to a concentration of 5 mM. All
the above processes were carried out under high purity argon atmosphere.
§ A detailed description of the specially designed home-built electro-
chemical STM setup used in the present investigations is given in ref. 12. A
pre-cleaned Teflon electrochemical cell (effective area of 0.36 cm2) with
three-electrode assembly was used for all the studies. In every experiment a
fresh gold sample (12 6 12 mm2, Berliner Glas KG, Germany) was
annealed in an H2 flame and cooled slowly in a N2 stream prior to use as a
working electrode. An Fe ring and Al wire (w99.99%, Alfa Acer,
Germany) dipped in to the melt served as the counter and reference
electrode, respectively. The STM tips were freshly prepared by etching
tungsten wires (250 mm diameter, w99.98%, Alfa) in a NaOH solution
(2 mol l21). To avoid the effect of Faraday currents, tips were coated via
electrophoresis with an electropaint (BASF ZQ 84-3225 0201, Germany)
and cured at 150 uC for 2 h and subsequently at 200 uC for 10 min. The
whole setup for the EC-STM studies was assembled in an argon-filled
glovebox (O2 and H2O v 1 ppm) and mounted in a clean and Ar-filled
airtight stainless steel container to ensure relatively long measurement
times.
monolayers. The islands continue to grow with time but do not
coalesce, instead they grow rapidly in length rather than in width to
form individual nanocrystal domains (Fig. 2(c)). Each nanocrystal
reaches a uniform width of about 50 nm and a length of about
120 nm (Fig. 2(d)). The average height of such nanocrystals is
y2 nm. Typically grown monodispersed Fe nanocrystals on
Au(111) by stepping the potential directly from OCP to 0.4 V are
shown in Fig. 3. Some of the crystals align in boomerang shapes
with an angle of about 120–140u like self-ordered Fe nanostructures
on Si(111).9 The latter have been prepared by rf-sputtering and so
the formation mechanism of the Fe nanostructures may be
different. The nanocrystals prepared under these conditions are
remarkably stable up to an anodic potential of 0.8 V.
In order to confirm the deposition of Fe, XPS spectra for a
sample deposited on Au(111) potentiostatically at 0.4 V for 10 min
were recorded using Al–Ka radiation (1486.6 eV) in a chamber
housing the OMICRON EA 125 multichannel hemispherical
analyzer. In the spectra (Fig. 4) of an as-prepared sample a weak
peak at 707 eV and a broad hump at 711.1 eV due to Fe and FeOx
are observed. Upon sputtering the sample for 5 min with an Ar1
ion beam, the Fe 2p3/2 peak becomes more prominent. There is no
shift in the binding energy of the Fe 2p3/2 peak which confirms the
deposition of pure Fe.10 The peak seen for FeOx is due to surface
oxidation of the sample while transferring from the glovebox to the
XPS spectrometer. Trace signatures of Al and chloride were also
seen in the spectra of as-prepared samples which almost vanish
upon sputtering. This could originate from spurious adsorbed
electrolyte.
1 Frontiers in Surface and Interface Science, ed. C. B. Duke and
E. W. Plummer, Elsevier, Amsterdam, 2002.
2 F. A. Moller, O. M. Magnussen and R. J. Behm, Phys. Rev. Lett., 1996,
77, 5249.
3 H. Hagenstrom, M. A. Schneeweiss and D. M. Kolb, Langmuir, 1999,
15, 7802.
4 Y. Gimeno, A. Hernandez, P. Carro, S. Gonzalez, R. C. Salvarezza and
A. J. Arvia, J. Phys. Chem. B, 2002, 106, 4232.
5 F. Endres, M. Bukowski and R. Hempelmann, Angew. Chem., Int. Ed.,
2003, 42, 3428.
6 C. A. Zell, F. Endres and W. Freyland, Phys. Chem. Chem. Phys., 1999,
1, 697.
7 C. L. Aravinda, I. Mukhopadhyay and W. Freyland, Phys. Chem.
Chem. Phys., in press.
8 M. Lipsztajn and R. A. Osteryoung, Inorg. Chem., 1985, 24, 716.
9 M. Cougo dos Santos, J. Geshev, D. K. Silva, J. E. Schmidt,
L. G. Pereira, R. Huber and P. Allongue, J. Appl. Phys., 2003, 94, 1490.
10 A. Velon and I. Olefjord, Oxid. Met., 2001, 56, 425.
11 F. Endres and W. Freyland, J. Phys. Chem. B, 1998, 102, 10229.
12 A. Shkurankov, F. Endres and W. Freyland, Rev. Sci. Instrum., 2002, 73,
102.
Results of our in-situ EC-STM studies show, for the first time,
that monodispersed Fe nanocrystals can be electrodeposited from
an ionic melt AlCl3–[MBIm]1Cl2 on Au(111). Such nanocrystals
with well defined shape open further perspectives for nanotechno-
logical applications. The acidic molten salt electrolyte AlCl3–
[MBIm]1Cl2 used in the present studies has a large electrochemical
window of greater than 2 V vs. Al/Al(III) on Au(111) and certainly
opens an arena for electrochemical synthesis of a new class of
nanostructured materials. In comparison to aqueous electrolytes it
has a specific advantage that hydrogen evolution does not interfere
with the nucleation and growth of deposits. In-depth studies on the
potential dependent growth kinetics and efforts to tailor the crystal
size are underway.
Financial support of this study by the DFG Center of
Functional Nanostructures, University of Karlsruhe, is
C h e m . C o m m u n . , 2 0 0 4 , 2 7 5 4 – 2 7 5 5
2 7 5 5