Liquid Nickel Salts: Synthesis, Crystal Structure Determination, and Electrochemical Synthesis of Nickel Nanoparticles

Article abstract of DOI:10.1002/chem.201504123

New nickel-containing ionic liquids were synthesized, characterized and their electrochemistry was investigated. In addition, a mechanism for the electrochemical synthesis of nanoparticles from these compounds is proposed. In these so-called liquid metal salts, the nickel(II) cation is octahedrally coordinated by six N-alkylimidazole ligands. The different counter anions that were used are bis(trifluoromethanesulfonyl)imide (Tf₂N^-), trifluoromethanesulfonate (OTf^-) and methanesulfonate (OMs^-). Several different N-alkylimidazoles were considered, with the alkyl sidechain ranging in length from methyl to dodecyl. The newly synthesized liquid metal salts were characterized by CHN analysis, FTIR, DSC, TGA and viscosity measurements. An odd-even effect was observed for the melting temperatures and viscosities of the ionic liquids, with the complexes with an even number of carbon atoms in the alkyl chain of the imidazole having a higher melting temperature and a lower viscosity than the complexes with an odd number of carbons. The crystal structures of several of the nickel(II) complexes that are not liquid at room temperature were determined. The electrochemistry of the compounds with the lowest viscosities was investigated. The nickel(II) cation could be reduced but surprisingly no nickel deposits were obtained on the electrode. Instead, nickel nanoparticles were formed at 100 % selectivity, as confirmed by TEM. The magnetic properties of these nanoparticles were investigated by SQUID measurements.

Full text of DOI:10.1002/chem.201504123

DOI: 10.1002/chem.201504123
Full Paper
&
Electrochemistry
Liquid Nickel Salts: Synthesis, Crystal Structure Determination,
and Electrochemical Synthesis of Nickel Nanoparticles
Jeroen Sniekers,[a] Ken Verguts,[a] Neil R. Brooks,[a] Stijn Schaltin,[b] Thanh Hai Phan,[a]
Thi Mien Trung Huynh,[a] Luc Van Meervelt,[a] Steven De Feyter,[a] Jin Won Seo,[b]
Jan Fransaer,[b] and Koen Binnemans*[a]
Abstract: New nickel-containing ionic liquids were
synthesized, characterized and their electrochemistry was in-
vestigated. In addition, a mechanism for the electrochemical
synthesis of nanoparticles from these compounds is pro-
posed. In these so-called liquid metal salts, the nickel(II)
cation is octahedrally coordinated by six N-alkylimidazole
ligands. The different counter anions that were used are
bis(trifluoromethanesulfonyl)imide (Tf2NÀ), trifluoromethane-
sulfonate (OTfÀ) and methanesulfonate (OMsÀ). Several dif-
ferent N-alkylimidazoles were considered, with the alkyl side-
chain ranging in length from methyl to dodecyl. The newly
synthesized liquid metal salts were characterized by CHN
analysis, FTIR, DSC, TGA and viscosity measurements. An
odd-even effect was observed for the melting temperatures
and viscosities of the ionic liquids, with the complexes with
an even number of carbon atoms in the alkyl chain of the
imidazole having a higher melting temperature and a lower
viscosity than the complexes with an odd number of
carbons. The crystal structures of several of the nickel(II)
complexes that are not liquid at room temperature were de-
termined. The electrochemistry of the compounds with the
lowest viscosities was investigated. The nickel(II) cation
could be reduced but surprisingly no nickel deposits were
obtained on the electrode. Instead, nickel nanoparticles
were formed at 100% selectivity, as confirmed by TEM. The
magnetic properties of these nanoparticles were investi-
gated by SQUID measurements.
Introduction
the ionic liquid can be very high. Moreover, as the metal is
part of the cation of the ionic liquid, it is electrostatically at-
tracted to the negatively charged cathode during electrodepo-
sition, increasing the mass transport significantly. These metal-
containing ionic liquids are called liquid metal salts (LMS) and
examples of such salts have been described before.[8–13] Several
examples of electrodeposition of metals from liquid metal salts
can be found in the literature.[14–22] Our research group has ob-
tained very promising results for the electrodeposition of
copper,[23–25] silver,[26–30] palladium[31] and lithium.[32] Metal coat-
ings of a good quality were achieved at very high current den-
sities, even without the use of additives. The electrodeposition
of nickel is an important industrial process for protection of
metals against wear, for electrocatalysis and magnetic applica-
tions.[33–36] Although nickel is mostly electroplated from aque-
ous electrolyte baths, the use of non-aqueous electrolytes has
been described before, for example nickel electrodeposition
from haloaluminate ionic liquids,[37,38] from conventional ionic
liquids,[39–41] and from deep-eutectic solvents.[42] Non-aqueous
electrolytes offer advantages in certain applications, such as
the synthesis of composite nickel coatings with co-deposited
luminescent particles[43] and the deposition of nickel–zinc
alloys as precursors for supported Raney catalysts.[44]
Ionic liquids consist entirely of ions and typically they are
made of bulky organic cations combined with weakly
coordinating organic or inorganic anions. They offer several ad-
vantages over conventional solvents such as a wide electro-
chemical window, a wide liquidus range and a negligible vapor
pressure.[1–4] Due to this wide electrochemical window, ionic
liquids are useful electrolytes for electrodeposition since sever-
al metals that cannot be deposited from aqueous solutions,
can be deposited from ionic liquids.[5–7] Disadvantages of most
ionic liquids used for the electrodeposition of metals are their
rather high viscosity and their poor solvent properties for
simple metal salts. These disadvantages can be overcome by
incorporating a metal into the cation of the molecular struc-
ture of the ionic liquid. In this way, the metal concentration in
[a] J. Sniekers, K. Verguts, Dr. N. R. Brooks, Dr. T. H. Phan,
Dr. T. M. Trung Huynh, Prof. L. Van Meervelt, Prof. S. De Feyter,
Prof. K. Binnemans
Department of Chemistry, KU Leuven
Celestijnenlaan 200 F - P.O. Box 2404, 3001 Heverlee (Belgium)
E-mail: Koen.Binnemans@chem.kuleuven.be
[b] Dr. S. Schaltin, Prof. J. W. Seo, Prof. J. Fransaer
Department of Materials Engineering, KU Leuven
Here, we report the synthesis of new nickel(II)-containing
ionic liquids with N-alkylimidazole ligands and different coun-
ter anions. A general structure of these compounds is present-
ed in Figure 1. The low-melting nickel(II) salts have been used
Kasteelpark Arenberg 44 - P.O. Box 2450, 3001 Heverlee (Belgium)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201504123; containing additional images of
crystal structures, crystal structure information and cyclic voltammograms.
Chem. Eur. J. 2016, 22, 1010 – 1020
1010
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 1. View of the general structure of the liquid nickel salts, in this case
with two bis(trifluoromethanesulfonyl)imide anions.
for the electrodeposition of nickel. It is shown that nickel(0)
nanoparticles were obtained in the electrolyte instead of
a nickel coating on the electrode.
Figure 2. TGA trace of [Ni(HeIm)6][Tf2N]2 in a nitrogen atmosphere. The
heating rate was 58CminÀ1
.
Results and Discussion
compared, the general trend that can be observed is a decrease
in melting point with increasing alkyl chain length on the N-al-
kylimidazole ligands. This trend is due to increasing steric hin-
drance from the longer alkyl chains, causing a less optimal
crystal packing and hence a lower melting temperature. For
very long alkyl chains, there is an additional stabilization due
to the van der Waals forces between the hydrocarbon chains,
causing an increase in melting temperature. For [Ni(DoIm)6]
[Tf2N]2 however, a melting point below room temperature was
observed. It is possible that at these long chain lengths, disor-
der again takes over and causes a poor crystal packing, result-
ing in a liquid at room temperature. In general, the melting
points and the viscosities of these liquid nickel salts are higher
than those of previously reported copper(I) and silver(I)
salts.[23–30] This is due to the fact that most of the liquid copper
and silver salts are composed of monovalent cations and one
anion, whereas nickel has a +II valency state. This results in
a higher Coulombic attraction between cations and anions and
therefore, a higher viscosity and melting point.
Synthesis and thermal properties
The starting reagents [Ni(H2O)6][Tf2N]2·2H2O, [Ni(H2O)6]
[OTf]2·2H2O and [Ni(H2O)6][OMs]2 were synthesized from nick-
el(II) oxide and the conjugated acid of the anion. To these
starting reagents, the N-alkylimidazole ligands were added in
a 1:6 metal-to-ligand ratio. These imidazole ligands replace the
water ligands from the starting reagents. Different N-alkyl-
imidazoles with a varying number of carbon atoms in the side-
chain were used, ranging from N-methylimidazole (MeIm) to N-
dodecylimidazole (DoIm). The N-alkylimidazoles coordinate to
the nickel(II) center through their free nitrogen atom. Different
counter anions were considered: bis(trifluoromethanesulfonyl)-
imide (bistriflimide, Tf2NÀ), trifluoromethanesulfonate (triflate,
OTfÀ) and methanesulfonate (mesylate, OMsÀ). Most com-
pounds were obtained at room temperature as blue solids,
whereas the starting products were green solids. Due to its
low melting temperature and low viscosity, [Ni(HeIm)6][Tf2N]2
was selected as the best candidate for electrochemical experi-
ments. The thermal stability of this model compound was de-
termined by thermogravimetric analysis (TGA). From the TGA
trace, it is evident that this compound is thermally stable up to
around 1308C, and only starts losing its ligands and anions sig-
nificantly at higher temperatures (Figure 2). All electrochemical
measurements were performed at 1208C, but care was taken
not to leave the compound at these temperatures too long in
between measurements to ensure optimal stability. At these
temperatures, the viscosity is significantly lowered. Since the
viscosity could not be measured at 1208C due to the use of
a water bath as heating device for the viscometer, it was
measured every five degrees between 258C and 808C. Extra-
polation gave a kinematic viscosity of 9”10À6 m2 sÀ1 at 1208C.
Also, a distinct odd-even effect can be observed for the
melting points and for the viscosities. Except for the N-methyl-
imidazole complex, the melting points of complexes contain-
ing N-alkylimidazoles with an even number of carbon atoms in
the alkyl chain are slightly higher than those of complexes
containing N-alkylimidazoles with an odd number of carbon
atoms. This effect has already been described in the literature,
but its origin is still a subject of debate.[45,46] It was not until re-
cently that attention was paid to the odd-even effect in the
melting points of ionic liquids.[47,48] For the viscosities, an oppo-
site trend is found: the compounds with an even number of
atoms in the sidechain have a lower viscosity than those with
an odd number of carbon atoms. Nickel(II) complexes with the
trifluoromethanesulfonate anion or the non-fluorinated
methanesulfonate anion have higher melting temperatures
and are less useful for electrodeposition experiments.
Melting temperatures
The melting points of all compounds were measured by
differential scanning calorimetry (DSC; Table 1). When the com-
pounds with the bis(trifluoromethanesulfonyl)imide anion are
Chem. Eur. J. 2016, 22, 1010 – 1020
www.chemeurj.org
1011
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 1. Melting points, glass transition temperatures and viscosities of
nickel(II) N-alkylimidazole complexes.
Compound
Melting
point [8C][a]
Glass
transition [8C]
Viscosity at
808C [mPas]
[Ni(MeIm)6][Tf2N]2
[Ni(EtIm)6][Tf2N]2
[Ni(PrIm)6][Tf2N]2
[Ni(BuIm)6][Tf2N]2
[Ni(PeIm)6][Tf2N]2
[Ni(HeIm)6][Tf2N]2
[Ni(HpIm)6][Tf2N]2
[Ni(OcIm)6][Tf2N]2
[Ni(NoIm)6][Tf2N]2
[Ni(DeIm)6][Tf2N]2
[Ni(DoIm)6][Tf2N]2
[Ni(EtIm)6][OTf]2
[Ni(PrIm)6][OTf]2
[Ni(BuIm)6][OTf]2
[Ni(MeIm)6][OMs]2
[Ni(EtIm)6][OMs]2
[Ni(PrIm)6][OMs]2
[Ni(BuIm)6][OMs]2
149
81
48
68
62
<RT
<RT
28
27
42
<RT
158
190
117
205
150
148
149
/
/
/
/
/
/
221
182
166
108
131
111
204
106
126
/
/
/
/
/
/
/
/
À53
À55
/
/
/
À52
/
/
/
/
/
/
/
[a]<RT indicates that the compound is a liquid below room temperature
and no melting point could be determined with DSC. In most cases,
a glass transition could be measured. Me=methyl, Et=ethyl, Pr=propyl,
Bu=butyl, Pe=pentyl, He=hexyl, Hp=heptyl, Oc=octyl, No=nonyl,
De=decyl, Do=dodecyl.
Figure 3. View of the crystal structure of [Ni(EtIm)6][Tf2N]2.
Crystal structures of nickel(II) complexes with the trifluoro-
methanesulfonate anion were also determined. In Figure S4 in
the Supporting Information, a view of the crystal structure of
[Ni(EtIm)6][OTf]2 is shown. Nickel(II) is coordinated by six N-eth-
ylimidazole ligands in an octahedral configuration. The com-
Crystal structures
The crystal structures of several nickel(II) complexes were
determined by single crystal X-ray diffraction. A view of the
crystal structure of [Ni(EtIm)6][Tf2N]2 is presented in Figure 3.
The nickel(II) center is coordinated by six N-ethylimidazole li-
gands in an octahedral geometry. The two bis(trifluorometha-
nesulfonyl)imide anions are non-coordinating. The asymmetric
unit consists of three ligands, one anion and half of a nickel(II)
ion. The other three ligands are generated by an inversion
center present in the space group P21/c. The nickelÀnitrogen
distances range from 2.11953(9) to 2.1321(5) Š, so there is
a slight distortion from the perfectly symmetrical octahedron
configuration. There are four CÀH···O interactions present,
ranging from 3.243(2) to 3.373(2) Š. CÀH···p interactions range
from 3.3571(2) to 3.5608(2) Š.
¯
pound crystallizes in the rhombohedral space group R3. All
nickelÀnitrogen distances are the same (2.12352(14) Š) since
all ligands are symmetry related, indicating that the coordina-
tion polyhedron in this compound is an undistorted octa-
hedron. The two trifluoromethanesulfonate anions in
[Ni(EtIm)6][OTf]2 are non-coordinating and there are no CÀH···X
or CÀH···p interactions present. The structure of [Ni(BuIm)6]
[OTf]2 is very similar to that of the compound with N-ethylimi-
dazole ligands (Figure S5 in the Supporting Information). The
crystal structure of [Ni(PrIm)6][OTf]2, differs from the two previ-
¯
ous structures and crystallizes in the cubic space group Ia3
The crystal structure of [Ni(PrIm)6][Tf2N]2 is very similar to
that of [Ni(EtIm)6][Tf2N]2, although they are not isomorphous
since the placement of the bis(trifluoromethanesulfonyl)imide
anion is different. Views of all the crystal structures, including
the numbering of the atoms, can be found in the Supporting
Information, together with tables containing further crystallo-
graphic information, CÀH···O, CÀH···N and CÀH···p interactions
(Figures S1–S9 and Tables S1–S3 in the Supporting Informa-
(Figure S6 in the Supporting Information). There are a few
CÀH···X interactions present in the N-propyl- and N-butyl-
imidazole analogues (Table S3 in the Supporting Information).
Several compounds with the non-fluorinated methanesulfo-
nate anion were examined by single crystal X-ray diffraction:
[Ni(EtIm)6][OMs]2, [Ni(PrIm)6][OMs]2 and [Ni(BuIm)6][OMs]2. The
crystal structure of [Ni(EtIm)6][OMs]2 is shown in Figure S7 in
the Supporting Information. The compound crystallizes in the
space group P21/c with the nickelÀnitrogen distances varying
between 2.11100(11) and 2.13722(7) Š. The asymmetric unit
¯
tion). [Ni(BuIm)6][Tf2N]2 crystallizes in the space group P1 with
the asymmetric unit containing two half [Ni(BuIm)3] moieties
and two bis(trifluoromethanesulfonyl)imide anions. The nickelÀ consists of 1.5 cations and 3 anions. Six N-ethylimidazole
nitrogen distances range between 2.0978(14) and 2.1383(14) Š.
CÀH···X interactions (X=O, N or F) range from 3.0806(2) to
3.510(2) Š in [Ni(PrIm)6][Tf2N]2 and from 3.060(2) to 3.383(2) Š
in the N-butylimidazole complex, with CÀH···p interactions
being at slightly longer distances in both compounds.
ligands are in an octahedral geometry around the nickel(II)
center, with additionally two non-coordinating methanesulfo-
nate anions. [Ni(PrIm)6][OMs]2 (Figure S8 in the Supporting In-
formation) and [Ni(BuIm)6][OMs]2 (Figure S9 in the Supporting
¯
¯
Information) crystallize in the space groups Ia3 and R3, respec-
Chem. Eur. J. 2016, 22, 1010 – 1020
www.chemeurj.org
1012 ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tively. All compounds have six ligands coordinated to the nick-
el(II) center with non-coordinating anions. Several CÀH···X and
CÀH···p interactions were observed (Table S3 in the Supporting
Information).
The crystal structures of [Ni(PrIm)6][OTf]2 and [Ni(PrIm)6]
[OMs]2 have many similarities. They crystallize in the same
¯
cubic space group Ia3. Their unit cells are almost identical,
except the unit cell of the trifluoromethanesulfonate analogue
is slightly larger by 217 Š3. There are a few additional CÀH···X
interactions in [Ni(PrIm)6][OTf]2 due to three more hydrogen
atoms present on the anion, but the CÀH···p interaction is very
similar in both compounds. Similar observations can be made
for the crystal structures of [Ni(BuIm)6][OTf]2 and [Ni(BuIm)6]
¯
[OMs]2. They both crystallize in the space group R3 and the
unit cell of the trifluoromethanesulfonate analogue is only
3
Figure 4. Cyclic voltammogram of [Ni(HeIm)6][Tf2N]2 versus Ni pseudo-refer-
ence electrode with different negative vertex potentials. Scan rate 50 mVsÀ1
temperature 1208C on a platinum working electrode with a nickel plate as
counter electrode.
228 Š larger in volume than that of the non-fluorinated
,
compound. There were no similarities observed between the
crystal structures of [Ni(EtIm)6][OTf]2 and [Ni(EtIm)6][OMs]2.
In none of the abovementioned crystal structures was any
p–p interactions below 4 Š observed. Some of the alkyl side-
chains in [Ni(EtIm)6][OTf]2, [Ni(PrIm)6][OTf]2, [Ni(PrIm)6][OMs]2
and [Ni(BuIm)6][OMs]2, as well as the methanesulfonate anion
in [Ni(EtIm)6][OMs]2 were modelled as disordered. This way, all
structures were refined to an R-value below 6.5%.
Electrochemistry
[Ni(HeIm)6][Tf2N]2 is a room temperature ionic liquid and has
a rather low viscosity at 1208C (9”10À6 m2 sÀ1). It was therefore
selected for electrochemical studies. A cyclic voltammogram
(CV) of [Ni(HeIm)6][Tf2N]2 is presented in Figure 4. The CV was
measured at 1208C at a scan rate of 50 mVsÀ1. The working
electrode was a platinum-sputtered (100 nm) silicon substrate
that was cleaned with HCl solution and then rinsed with water
and acetone before use. A nickel wire was used as pseudo-
reference electrode and a nickel plate as counter electrode.
The dotted line in the CV started at a potential of +0.50 V
versus Ni2+/Ni where Ni2+ was the stable species. Upon lower-
ing the potential, there is a very small negative current around
À0.50 V versus Ni2+/Ni while the bulk reduction starts at
À0.80 V versus Ni2+/Ni, so there is a large overpotential for the
reduction of nickel. The cathodic process shows no limiting
current up to the negative vertex potential of À1.50 V versus
Ni2+/Ni, although current densities here are still rather small,
especially compared to previous results from liquid metal salts.
Upon reversal of the scan direction, an anodic current is
measured with a peak around +0.40 V versus Ni2+/Ni. This
anodic current is smaller than the cathodic, indicating that the
nickel electrodeposition is not fully reversible.
Figure 5. Linear potential scan of an unstirred [Ni(HeIm)6][Tf2N]2 solution,
measured versus a Ni pseudo-reference electrode to À5.0 V versus Ni2+/Ni.
Scan rate 50 mVsÀ1, temperature 1208C on a platinum working electrode
with a nickel plate as counter electrode.
Ni2+ to Ni0 is not mass-transfer controlled. This phenomenon
was also found for previous liquid metal salts with copper and
silver.[24–31]
Upon removal of the working electrode, a black precipitate
formed around the cathode. Rinsing with acetone and water
completely rinsed off the precipitate and no nickel deposit was
found. Even after reduction from an unstirred solution at
À1.0 AdmÀ2 for 5 min (which should result in a nickel coating
with a theoretical thickness of 2.0 mm), the black precipitate
was completely removed when the substrate was rinsed with
acetone. When the substrates were investigated by scanning
electron microscopy (SEM), the surface of the electrode was
found to be unchanged. The sample was also analyzed with
energy dispersive X-ray analysis (EDX) which showed no traces
of nickel metal on the electrode, only of the platinum–silicon
substrate (Figure S10 in the Supporting Information). An accel-
eration voltage of 10 keV was used to avoid deep penetration
of the electrons in the sample and to only scan the surface.
When a different negative vertex potential was used, as in
the full line and the dotted-dashed line in Figure 4, it became
apparent that the anodic peak must be the reoxidation of Ni0,
since it is not present in the full and dotted-dashed curves
which have a smaller negative vertex potential, so there is
much less formation of Ni0.
When a very negative vertex potential is used, as in Figure 5,
there is still no limiting current, indicating the reduction of
Chem. Eur. J. 2016, 22, 1010 – 1020
www.chemeurj.org
1013
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Because of its reasonably low viscosity, [Ni(DoIm)6][Tf2N]2
was also tested for the electrodeposition of nickel. Similar
results as with [Ni(HeIm)6][Tf2N]2 were found. This CV can be
found in the Supporting Information (Figure S11).
Characterization of nanoparticles
After plating for 30 min at a current density of À5.0 AdmÀ2, it
was noted that the color of the entire solution darkened. This
color change and the fact that no nickel deposition was found
on the substrate led us to the assumption that nickel was not
deposited, but that nickel nanoparticles were formed instead.
To investigate this assumption, a galvanostatic experiment was
performed in which nanoparticles were formed at a current
density of À1.0 AdmÀ2 for 20 min (which should result in
a nickel layer of 8.0 mm). The working electrode was removed
from solution and rinsed with acetone. This acetone was
Figure 7. TEM image of nickel nanoparticles in pure [Ni(DoIm)6][Tf2N]2.
deposition whatsoever was obtained, indicating a 100% selec-
tivity for the synthesis of nanoparticles! This might explain
why the anodic peak current in the cyclic voltammogram in
Figure 4 is smaller than the cathodic one. It is more difficult to
reoxidize these nickel nanoparticles that have diffused from
the electrode than to reoxidize a nickel coating attached to
the electrode.
collected for
a transmission electron microscope (TEM)
measurement (Figure 6).
It was also found that after performing a galvanostatic ex-
periment on [Ni(HeIm)6][Tf2N]2 at À0.5 AdmÀ2 for 5, 15 and
45 min under stirring, there were no nickel deposits on the
electrode after the working electrodes were placed in a vial
containing acetone and left overnight. So even in an unstirred
acetone solution, overnight, all of the particles disperse into
solution from the electrode, indicating a 100% selectivity for
the formation of nanoparticles. Also when a different working
electrode was used (copper or gold), similar results were
obtained.
To investigate the composition of the nanoparticles a galva-
nostatic experiment in an unstirred solution was executed in
which the current was kept at À1.0 AdmÀ2 for 30 min. After
the synthesis of the nanoparticles, the electrode was removed
from the liquid nickel salt and the liquid was allowed to drip
off the electrode as much as possible. The remainder of the
particles that were still on the surface of the electrode were
rinsed off with acetone and collected in a vial. This solution
contained acetone, nickel nanoparticles and only a very small
remainder of the liquid nickel salt. In order to remove the
small fraction of liquid metal salt, ultracentrifugation was tried
and after 24 h at 248800 g, a large fraction of the particles
formed a sediment at the bottom of the vial, while above the
sediment, there was a clear acetone layer and at the very top
was another cloudy layer containing the smallest particles. The
bottom fraction of precipitated nanoparticles was redissolved
in ethanol and examined using total reflection X-ray fluores-
cence (TXRF). The obtained TXRF spectrum is presented in Fig-
ure S13 in the Supporting Information. It shows two large fluo-
rescent peaks (Ka and Kb) from Ni and some small peaks arising
from the molybdenum source, silicon from the TXRF sample
carrier and argon from the atmosphere. No other metals were
found, indicating the nanoparticles consist of pure nickel.
It is known that very small clusters of metals can have differ-
ent magnetic properties than the bulk material. The size and
geometry of nickel nanoparticles has a big influence on their
magnetic behavior.[49–56] To investigate the magnetism of the
nickel nanoparticles synthesized from [Ni(HeIm)6][Tf2N]2, nano-
Figure 6. Transmission electron microscopy image of nickel nanoparticles in
acetone, formed after electroplating from [Ni(HeIm)6][Tf2N]2 for 20 min at
À1.0 AdmÀ2
.
The average diameter of the particles was analyzed by
ImageJ software and was determined to be 5.4 nm.
A sample of the liquid metal salt [Ni(DoIm)6][Tf2N]2 was also
kept at À1.0 AdmÀ2 for 20 min. Afterwards, a TEM grid was
submersed in the pure liquid nickel salt adjacent to the work-
ing electrode. The grid was covered with blotting paper in
order to absorb the excess ionic liquid and was placed in the
transmission electron microscope. The TEM image of the nano-
particles from this liquid is presented in Figure 7. It was ob-
served that the nickel particles in pure [Ni(DoIm)6][Tf2N]2 are
formed of primary particles of a few nanometers in diameter
that coagulate to form larger spherical agglomerates with a
diameter of around 60 nm. This method of TEM measurement
was also tried for nanoparticles synthesized from the model
compound [Ni(HeIm)6][Tf2N]2, but due to the very high back-
ground of nickel in the pure liquid metal salt, this resulted in
poor resolution.
The presence of nanoparticles could also be visualized by
shining a laser beam through the acetone that was used to
rinse the working electrode. A picture of this can be found in
the Supporting Information (Figure S12).
The preferred formation of nanoparticles to the deposition
of nickel was a fascinating surprise, especially since no nickel
Chem. Eur. J. 2016, 22, 1010 – 1020
www.chemeurj.org
1014
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
are synthesized by electroreduction within a template.[7–72] and
in one case they were even synthesized by an anodic pro-
cess.[73] Electrochemical synthesis of nickel nanoparticles with-
out a template has been described before by Reetz et al.[74,75]
although their systems are highly dependent on metal concen-
tration and the presence of stabilizers, which does not seem to
be the case for our liquid nickel salts. Katayama et al.[76–81]
reported the electrochemical synthesis of nanoparticles of
silver,[76] nickel,[77,78] iron,[79] cobalt,[80] and palladium.[81] They dis-
solved transition metal salts with bis(trifluoromethanesulfony-
l)imide anions in the ionic liquid [BMP][Tf2N] (BMP=1-butyl-1-
methylpyrrolidinium) and reduced the metals electrochemical-
ly. In the case of nickel, they obtained nickel depositions from
Ni(Tf2N)2 in [BMP][Tf2N] at shorter plating experiments,[77,78] but
they found that at longer plating times and larger overpoten-
tials the color of the ionic liquid darkened and nickel nanopar-
ticles were dispersed in the liquid.[79] The cyclic voltammo-
grams they measured were similar to the ones presented in
this work (i.e., large overpotential for the reduction of the
metal and a process that is not fully reversible). Since they
have deposits at shorter plating times, the anodic dissolution
is higher in their case since these deposits are more easily re-
oxidized than nanoparticles dispersed in the liquid. Their pro-
posed mechanism for the formation of nanoparticles was
based on the presence of a double layer on top of the elec-
trode. Due to the negative overpotential during deposition,
the bulky BMP+ cations are attracted to the electrode and
form a blocking layer, inhibiting attachment of the nickel layer
to the electrode. Since the growth of nickel nuclei on top of
the electrode is hindered by this accumulation of BMP cations,
the nuclei will detach from the electrode. Supporting this
theory, is the fact that it is known that nanoparticles can
be dispersed in ionic liquids,[82] and it is also proven that
Figure 8. Experimental M–H curve for nickel(0) nanoparticles from
[Ni(HeIm)6][Tf2N]2.
particles were synthesized electrochemically at À1.0 AdmÀ2 for
20 min. A bulk sample of the Ni LMS after electroreduction
was collected and a SQUID hysteresis loop measurement at
room temperature (quantum design MPMS-XL SQUID) was
performed. The results are presented in Figure 8.
As can be seen from this Figure, the curve does not show
a hysteresis and goes through the origin. It is therefore Lange-
vin shaped, which means the nickel nanoparticles are super-
paramagnetic. In 1946, Kittel proved that one single magnetic
domain would be more stable for particles that are smaller
than a certain size, since for such small particles, thermal agita-
tion will hinder the existence of stable magnetization, which
leads to a superparamagnetic state.[57] The existence of super-
paramagnetic nickel nanoparticles has been measured by
SQUID before.[58,59] Ramírez-Meneses et al.[59] calculated that
the superparamagnetic limit of nickel is around 44 nm. Al-
though, according to a different calculation by Respaud
et al.,[60] when the thermal activation is taken into consider-
ation, the superparamagnetic limit of nickel is 9.57 nm. This is
in good agreement with the TEM measurement, from which it
was determined that the particles in acetone are on average
5.4 nm and should therefore be superparamagnetic. The ad-
vantage of the SQUID measurement is that this could be per-
formed on the nanoparticles in the liquid metal salt, whereas
the TEM image from the nickel nanoparticles is dispersed in
acetone.
bis(trifluoromethanesulfonyl)imide
anions
can
stabilize
certain metals in their zero valence state.[83] This stabilization
by bistriflimide also inhibits further growth of the particles
according to Katayama et al.[76–81]
It has been known for quite some time that organic moieties
can adsorb to a surface, as was determined by thin-layer
coulometry,[84] and surface enhanced Raman spectroscopy
(SERS).[85,86] Only since quite recently, there has been interest in
the surface structure at the ionic liquid–metal electrode inter-
face. Baldelli[87] investigated the double layer behavior of sever-
al ionic liquids with a 1-butyl-3-methylimidazolium cation and
BF4À, PF6À, DCAÀ and Tf2NÀ anions by electrochemical impe-
dance techniques and sum frequency generation (SFG). He
states that the double layer in those ionic liquids consists of
one layer of cations and one layer of anions and the orienta-
tion of the layers is potential dependent. At negative poten-
tials the cation of the ionic liquid adsorbs quite strongly to the
surface, and at positive potentials the anion. Zhou et al.[88] in-
vestigated the double layer behavior of [bmim][OTf] using SFG
and found that the adsorption of this ionic liquid to an elec-
trode is also potential dependent and there is an adsorption
hysteresis when switched from a negative to a positive poten-
tial and vice versa. This indicates that there is a certain binding
energy and Madelung potential that needs to be overcome
Nickel nanoparticles are used in a wide variety of different
applications such as electrocatalysts in fuel cells, in magnetic
materials and in various other fields such as microelectronics,
catalysis in organic synthesis, medical diagnostics and magnet-
ic storage devices.[61–63] They can be synthesized in many differ-
ent ways, such as pyrolysis[64] or sputtering,[65] but the most
common way is chemical reduction of NiCl2 (sometimes micro-
wave assisted) with reducing agents such as hydrogen, hydra-
zine and NaBH4, in the presence of stabilizers such as hexa-
decylamine (HAD) and trioctylphosphine oxide (TOPO).[66–69] In
this work, Ni nanoparticles were synthesized electrochemically.
The electrochemical synthesis of transition metal nanoparticles
has been described in the literature before. Most often they
Chem. Eur. J. 2016, 22, 1010 – 1020
www.chemeurj.org
1015
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
when an adsorbed 1-butyl-3-methylimidazolium cation at neg-
ative potentials is interchanged with an adsorbed triflate anion
at positive potentials. The groups of Atkin,[89] Hayes[90] and
Endres[91,92] investigated the adsorption of several different
ionic liquids with atomic force microscopy (AFM) and found
that usually, the solvation layer consists of multiple (3–5) layers
of ionic liquid adsorbed to the surface. The species that ad-
sorbed to the electrode was also potential dependent and at
higher overpotentials, the force to break through one ad-
sorbed layer, as measured by AFM, was larger, indicating
a stronger adsorption. This adsorption of [Py1,4][Tf2N] and
[emim][Tf2N] to the electrode had an influence for example on
the crystallinity of aluminium deposits from AlCl3 dissolved in
those liquids.[91] The ionic liquid interface in [bmim][PF6] was
also investigated by Gnahm et al.[93–96] using impedance
measurements and in situ STM measurements. A potential
dependent adsorption of cations and anions was observed.
Also Tulodziecki et al.[97] found that it is possible in ionic liquid
media containing imidazolium cations that these cations are
restructured in the double layer next to the electrode. The
electrons can tunnel through this imidazolium layer and nickel
is reduced at the edge of this layer, which is at a certain
distance from the electrode. They observed that this leads to
poorly adhesive layers of Co, but this applies also to Mn, Ni
and Zn.
measurements, compounds with the general formula
[Ni(MeIm)x(HeIm)6–x][Tf2N]2 and [Ni(EtIm)x(HeIm)6–x][Tf2N]2 were
synthesized. It was found that the melting temperature of the
series with mixed methyl- and hexylimidazole ligands was too
high (Tm of [Ni(MeIm)4(HeIm)2][Tf2N]2 is 1318C and Tm of
[Ni(MeIm)5(HeIm)1][Tf2N]2 is 1568C). Therefore, the series with
mixed N-ethyl- and N-hexylimidazole ligands was investigated
(Figure S14 in the Supporting Information).
The electrochemical results are very similar to what was
found for [Ni(HeIm)6][Tf2N]2. For each LMS, there is a compara-
ble reduction of nickel(II) at negative potentials versus the
nickel reference electrode. The reversibility of the Ni deposition
in the liquid [Ni(EtIm)1(HeIm)5][Tf2N]2 (black curve) however, is
the lowest of all the investigated liquids. There is almost no
visible reoxidation of nickel(0) at positive potentials versus Ni2+
/Ni. For the liquid [Ni(EtIm)5(HeIm)1][Tf2N]2 on the other hand,
on the reverse scan there is a clearly visible anodic peak start-
ing already around À0.25 V versus Ni2+/Ni and with its peak
current at +0.25 V versus Ni2+/Ni, although this anodic current
is still much smaller than the cathodic current. For all
compounds, current densities are roughly the same order of
magnitude, although currents increase slightly when less
N-hexylimidazole is present.
For each of the liquids, a galvanostatic deposition experi-
ment was performed by keeping the current at À1.0 AdmÀ2
for 15 min (no stirring). Upon removal of the working elec-
trode, the Pt-coated wafer was gently rinsed with acetone and
dried. For [Ni(EtIm)1(HeIm)5][Tf2N]2, no nickel deposition was
detected with SEM and EDX after this rinsing. For liquids with
a larger fraction of N-ethylimidazole on the other hand, from
[Ni(EtIm)3(HeIm)3][Tf2N]2 to [Ni(EtIm)5(HeIm)1][Tf2N]2, a clearly
visible black deposit remained attached to the electrode, al-
though a large fraction of the reduced nickel was also rinsed
off. All working electrodes were investigated using SEM and
EDX. An SEM image of the deposits from [Ni(EtIm)5(HeIm)1]
[Tf2N]2 is presented in Figure 9. The depositions are not uni-
form: they look like a collection of particles welded together.
This cauliflower-like deposition was observed before by Kataya-
ma et al.[76–81] for nickel deposits from Ni(Tf2N)2 in [BMP][Tf2N]
and also by Sun and co-workers, who deposited Ni2+ from the
ionic liquid [emim][DCA].[41]
Therefore, based on the available literature, a possible mech-
anism for the electrochemical synthesis of nickel nanoparticles
from liquid metal salts can be proposed. The solvation layer in
[Ni(HeIm)6][Tf2N]2 covers the surface of the platinum working
electrode, thereby hindering the nickel ions to reach the sur-
face during the reduction process. The theory of extended
electron transfer states that it is possible for electrons to
tunnel through such a layer[98] and they can reduce nickel(II) at
the edge of this solvation layer, forming nickel(0) adatoms at
a certain distance of the electrode. This tunneling of electrons
requires a larger energy, which explains the large overpoten-
tials for the reduction of nickel(II) from [Ni(HeIm)6][Tf2N]2. These
adatoms have a very high surface energy and will cluster to-
gether in order to reduce this surface energy, hence they form
nanoparticles. It is known from literature that tetraalkylammo-
nium and imidazolium cations with a sidechain longer than
four carbon atoms have surfactant properties and can stabilize
nanoparticles.[99,100] Therefore, once these Ni adatoms, formed
at the edge of the solvation layer, cluster to form nanoparticles
of a certain size, they can be stabilized by the N-hexylimidazole
ligands and possibly by the bis(trifluoromethanesulfonyl)imide
anions.[76–81,83] Thereafter, the stabilized nanoparticles are
gradually dispersed into the bulk liquid metal salt.
To visualize the solvation layer, electrochemical scanning
tunneling microscope (EC-STM) analysis was performed. The
measurement was performed at room temperature on the
pure [Ni(HeIm)6][Tf2N]2 liquid. An Au(111) working electrode
was used since this usually yields good STM images and very
similar CVs to those on Pt were also obtained on a gold elec-
trode. The Au(111) single crystal (MaTeck GmbH, Germany) was
cleaned by flame-annealing for 5 min before being transferred
into the electrochemical cell. The tunneling tips used in the ex-
periments were electrochemically etched from 0.25 mm tung-
sten wire in 2 mm KOH solution and subsequently isolated by
passing the tip through a hot melt glue film in order to mini-
mize residual Faradaic currents. During the measurement, the
potential was varied and the adsorbed species on top of the
Au(111) electrode was examined. In Figure 10, the potential
was varied from +200 mV versus Ni2+/Ni (Figure 10a) to
To verify this proposed mechanism, the effect of different
cations was investigated. If the solvation layer blocks the
surface of the electrode, smaller cations would cause for the
formation of Ni adatoms at a shorter distance to the electrode
and perhaps lead to a nickel layer that adheres to the
electrode. Therefore, new liquid nickel salts were synthesized
with imidazole ligands with shorter sidechains. Since the com-
pounds with N-methylimidazole and N-ethylimidazole ligands
have a melting point too high to be useful for electrochemical
Chem. Eur. J. 2016, 22, 1010 – 1020
www.chemeurj.org
1016
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 9. SEM image of a nickel layer deposited from [Ni(EtIm)5(HeIm)1]
[Tf2N]2 at À1.0 AdmÀ2 for 15 min.
Figure 11. EC-STM image of [Ni(EtIm)6][Tf2N]2 on an Au(111) working
electrode at different potentials versus a nickel wire. a) E= +10 mV,
b) E= +100 mV, c) E= +200 mV, d) E= +300 mV, e) E= +400 mV, and
f) E= +550 mV versus Ni2+/Ni. The total area is 125.3 nm2, I=0.2 nA,
U=À300 mV.
melting temperature was found for N-octylimidazole to N-do-
decylimidazole. In addition to this general trend, an odd-even
effect was observed in which the complexes with ligands with
an even number of carbon atoms in the alkyl sidechain had
a higher melting temperature than complexes with an odd
number of carbon atoms in the alkyl chain. Two of the liquid
metal salts, [Ni(HeIm)6][Tf2N]2 and [Ni(DoIm)6][Tf2N]2 had a rea-
sonably low viscosity and were tested for the electrodeposition
of nickel. No nickel deposits were found but nanoparticles
were formed instead, at 100% selectivity. The nanoparticles
were investigated using TEM, TXRF and their superparamag-
netic nature was measured with SQUID. When a sufficient
amount of shorter sidechain ligands such as N-methylimidazole
and N-ethylimidazole were added to complex nickel, it was
found that deposits could be obtained. From this result and
from EC-STM measurements, we proposed a mechanism for
the electrochemical synthesis of nickel nanoparticles, based on
the presence of a solvation layer, adsorbed to the surface of
the working electrode, that blocks the electrode, necessitating
electrons to tunnel through this layer in order to reduce nickel
at a certain distance of the working electrode. The longer this
distance (as was the case for N-hexylimidazole ligands or
longer sidechains) the more nanoparticle formation. When this
imidazole layer is short, as for N-ethylimidazole, Ni deposits
were found on the electrode.
Figure 10. EC-STM image of [Ni(EtIm)6][Tf2N]2 on an Au(111) working
electrode at different potentials versus a nickel wire. a) E= +200 mV,
b) E= +100 mV, c) E=0.0 mV, d) E=À50 mV, e) E=À100 mV, and
f) E=À150 mV versus Ni2+/Ni. The total area is 125.3 nm2, I=0.2 nA,
U=À300 mV.
+100, 0.0, À50, À100 and À150 mV (Figure 10b to f). As can
be seen from these images, the surface of the electrode gets
increasingly more blocked by adsorbing layers. The adsorbant
cannot be nickel deposition since the deposition only starts at
À0.8 V versus Ni2+/Ni.
When the same surface is scanned in the positive direction,
starting at +10 mV versus Ni2+/Ni, to +100, +200, +300,
+400 and +550 mV (Figure 11 a to f), it is seen that there is
also adsorption of a species onto the Au(111) electrode, but
these adsorbants look different from the solvation layer at
negative potentials. From these measurements, it is not clear
which species adsorbs to the surface at each potential, but it is
very clear that the electrode is blocked by an adsorbed moiety,
thereby inhibiting the attachment of nickel adatoms onto the
surface to form nickel deposits, providing evidence for our pro-
posed mechanism of electrochemical nanoparticle synthesis.
Experimental Section
General
All commercially available N-alkylimidazoles were ordered from
Sigma–Aldrich (Diegem, Belgium) or IoLiTec (Heilbronn, Germany),
except for N-propylimidazole, N-heptylimidazole and N-nonylimida-
zole, which were not commercially available and were synthesized
according to a literature procedure.[101] The infrared spectra were
recorded at a resolution of 4 cmÀ1, using an attenuated total reflec-
tance (ATR) technique on a Bruker Vertex 70 FTIR spectrometer,
with a Platinum ATR extension. Assignment of the peaks was
based on the available literature.[102,103] Melting points were deter-
Conclusions
Several new liquid nickel(II) salts with N-alkylimidazole ligands
with different alkyl chain lengths were synthesized and charac-
terized. The melting temperatures decreased from N-methyl-
imidazole to N-hexylimidazole ligands, while an increase in
Chem. Eur. J. 2016, 22, 1010 – 1020
www.chemeurj.org
1017
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
mined on a Mettler-Toledo 822 DSC instrument at a heating rate of
108CminÀ1 under a helium atmosphere. The viscosities were mea-
sured on a Brookfield cone plate viscometer (LVDV-II+ program-
mable viscometer) with a cone spindle CPE-40. The sample holder
was heated to 808C by a circulating water bath. Elemental analyses
(CHN) were carried out using a CE Instruments EA-1110 elemental
analyzer. Characterization of the nanoparticles was performed on
a CM 200 FEG Custom transmission electron microscope, equipped
with a GATAN magnetic imaging filter. For the EC-STM measure-
ments, the electrolyte was employed without any further modifica-
tion and the experiments were carried out at room temperature
using a Bonn EC-STM system.[104] Separation of the nickel nanopar-
ticles from the ionic liquid solutions was carried out with a prepara-
tive ultracentrifuge (Beckman Coulter, model Optima LE-80K, rotor
SW 65), at 50000 rpm (248800 g).
ture. The solution was refluxed at 658C, overnight. The sodium
bromide precipitate was filtered off and excess of tetrahydrofuran
was removed on a rotary evaporator. The crude product was dis-
solved in dichloromethane. Activated carbon and MgSO4 were
added. The solution was filtered off and excess of dichloromethane
was removed on a rotary evaporator, yielding a yellow liquid at
room temperature. Yield: 4.79 g (87%). 1H NMR (300 MHz, CDCl3):
d=7.45 (s, 1H), 7.03 (s, 1H), 6.91 (s, 1H), 3.91 (t, J=7.2 Hz, 2H),
1.76 (quin, J=7.3 Hz, 2H), 1.43–1.17 (m, 4H), 0.89 ppm (t, J=
7.2 Hz, 3H); 13C NMR (75 MHz, CDCl3): d=136.91 (s), 129.07 (s),
118.73 (s), 46.89 (s), 30.63 (s), 28.52 (s), 22.02 (s), 13.76 ppm (s).
Synthesis of 1-heptylimidazole (HpIm): To sodium imidazolate
(4.00; 44.46 mmol) in tetrahydrofuran (20 mL) was added 1-bromo-
heptane (6.65 g; 37.12 mmol) and stirred for 3 h at room tempera-
ture. The solution was refluxed at 658C overnight. The sodium bro-
mide precipitate was filtered off and excess of tetrahydrofuran was
removed on a rotary evaporator. The crude product was dissolved
in dichloromethane. Activated carbon and MgSO4 were added. The
solution was filtered off and excess of dichloromethane was re-
moved on a rotary evaporator, yielding a yellow liquid at room
Synthesis of starting reagents
Synthesis [Ni(H2O)6][Tf2N]2·2H2O: To nickel(II) oxide (6.30 g;
84.32 mmol) in water (50 mL) was added a 80 wt% HTf2N (55.54 g;
197.6 mmol) solution and stirred for 3 h at 808C. The excess of
nickel(II) oxide was filtered off. The solvent was removed on
a rotary evaporator and the product was dried, overnight, on
a high vacuum line, yielding a green solid at room temperature.
Yield: 60.26 g (100%). CHN found (calcd) for C4H16N2O16F12S4Ni: C
6.14 (6.30), H 2.12 (2.11), N 3.62 (3.67)%; IR (ATR): nmax =3524
n(OH), 3363, 1639, 1357 na(SO2), 1337 na(SO2), 1197 na(CF3), 1136
ns(SO2), 1038 na(SNS), 797 n(CS), 743 ds(CF3), 610 da(SO2), 569
da(CF3), 509 cmÀ1 da(CF3); m.p. 768C.
1
temperature. Yield: 5.80 g (94%). NMR: H NMR (300 MHz, CDCl3):
d=7.45 (s, 1H), 7.04 (s, 1H), 6.91 (s, 1H), 3.92 (t, J=7.1 Hz, 2H),
1.76 (quin, J=7.0 Hz, 2H), 1.43–1.19 (m, 8H), 0.88 ppm (t, J=
6.8 Hz, 3H); 13C NMR (75 MHz, [D6]DMSO): d=136.95 (s), 129.15 (s),
118.76 (s), 46.97 (s), 31.54 (s), 31.00 (s), 28.65 (s), 26.42 (s), 22.46 (s),
13.95 ppm (s).
Synthesis of 1-nonylimidazole (NoIm): To sodium imidazolate
(3.51; 38.97 mmol) in tetrahydrofuran (20 mL) was added 1-bromo-
nonane (7.27 g; 35.07 mmol) and stirred for 3 h at room tempera-
ture. The solution was refluxed at 658C, overnight. The sodium
bromide precipitate was filtered off and excess of tetrahydrofuran
was removed on a rotary evaporator. The crude product was dis-
solved in dichloromethane. Activated carbon and MgSO4 were
added. The solution was filtered off and excess of dichloromethane
was removed on a rotary evaporator, yielding a yellow liquid at
room temperature. Yield: 6.06 g (89%). 1H NMR (300 MHz,
[D6]DMSO): d=7.46 (s, 1H), 7.05 (s, 1H), 6.90 (s, 1H), 3.92 (t, J=
7.2 Hz, 2H), 1.77 (quin, J=7.0 Hz, 2H), 1.37–1.19 (m, 12H),
0.88 ppm (t, J=6.8 Hz, 3H); 13C NMR (75 MHz, [D6]DMSO): d=
137.12 (s), 128.24 (s), 119.11 (s), 45.85 (s), 31.22 (s), 30.56 (s), 28.86
(s), 28.58 (s), 28.45 (s), 25.87 (s), 22.06 (s), 13.88 ppm (s).
Synthesis of [Ni(H2O)6][OTf]2·2H2O: To nickel(II) oxide (4.11 g;
55.03 mmol) in water (50 mL) was added a triflic acid (15.16 g;
101.0 mmol) solution and stirred for 3 h at room temperature. The
excess of nickel(II) oxide was filtered off. The solvent was removed
on a rotary evaporator and the product was dried, overnight, on
a high vacuum line, yielding a dark green solid at room tempera-
ture. Yield: 23.64 g (93%). CHN found (calcd) for C2H16O14F6S2Ni: C
4.55 (4.80), H 2.98 (3.22)%; IR (ATR): nmax =2927, 2857, 1464, 1413,
1349 na(SO2), 1185 na(CF3), 1136 ns(SO2), 1057, 789 n(CS), 762, 739
ds(CF3), 720, 654, 616 da(SO2), 570 da(CF3), 512 cmÀ1 da(CF3); m.p.
1778C.
Synthesis of [Ni(H2O)6][OMs]2: To nickel(II) oxide (5.27 g;
70.60 mmol) in water (50 mL) was added a methanesulfonic acid
(12.33 g; 128.3 mmol) solution and stirred for 3 h at room tempera-
ture. The excess of nickel(II) oxide was filtered off. The solvent was
removed on a rotary evaporator and the product was dried, over-
night, on a high vacuum line, yielding a green solid at room tem-
perature. Yield: 20.15 g (88%). CHN found (calcd) for C2H18O12S2Ni:
C 6.86 (6.73), H 5.15 (5.08)%; IR (ATR): nmax =3328 n(OH), 3030
ns(CH3), 3014, 2941, 1653, 1423 da(CH3), 1340 ds(CH3), 1168 ns(SO2),
1053, 962 1(CH3), 825 n(SO), 779, 729, 596, 556, 515, 473 cmÀ1; m.p.
2368C.
Synthesis of liquid nickel salts
Synthesis of [Ni(MeIm)6][Tf2N]2: To [Ni(H2O)6][Tf2N]2·2H2O (1.50;
1.97 mmol) in acetonitrile (30 mL) was added 1-methylimidazole
(0.98 g; 11.90 mmol) and stirred for 30 min at room temperature.
Excess of acetonitrile was removed on a rotary evaporator and the
product was dried, overnight, on a high vacuum line, yielding
a purple/blue solid at room temperature. Yield: 2.09 g (95%). CHN
found (calcd) for C28H36N14O8F12S4Ni: C 30.32 (30.25), H 3.54 (3.26),
N 17.49 (17.64)%; IR (ATR): nmax =3141 n(CH), 1533 n(ring), 1520
n(ring), 1350 na(SO2), 1331 na(SO2), 1287, 1235 ns(CF3), 1225 na(CF3),
1194 na(CF3), 1175, 1140 ns(SO2) and n(ring), 1107 d(CH), 1089 d(CH),
1053 na(SNS), 935 d(ring) and g(CH), 831 g(CH), 755, 744 ds(CF3)
and g(CH), 664 g(ring), 610 g(ring) and da(SO2), 569 da(CF3),
514 cmÀ1 da(CF3); m.p. 1498C.
Synthesis of sodium imidazolate (NaIm): To 1H-imidazole
(15.00 g; 220.29 mmol) was added sodium hydroxide (9.26 g;
231.42 mmol). The solution was stirred for 4 h at 110 8C. A small
amount of water (<10 mL) was added to rinse the flask. Water was
removed on a rotary evaporator and the product was dried for
48 h on a high vacuum line at 110 8C, yielding a yellow solid at
room temperature. Yield: 18.87 g (95%). 1H NMR (300 MHz,
[D6]DMSO): d=7.12 (s, 1H), 6.69 ppm (d, J=0.8 Hz, 2H); 13C NMR
(75 MHz, [D6]DMSO): d=142.52 (s), 124.62 ppm (s).
Synthesis of [Ni(EtIm)6][OTf]2: To [Ni(H2O)6][OTf]2·2H2O (1.01 g;
2.17 mmol) in acetonitrile (30 mL) was added 1-ethylimidazole
(1.26 g; 13.15 mmol) and stirred for 30 min at room temperature.
Excess of acetonitrile was removed on a rotary evaporator and the
product was dried overnight on a high vacuum line, yielding
a purple/blue solid at room temperature. Yield: 1.96 g (97%). CHN
Synthesis of 1-pentylimidazole (PeIm): To sodium imidazolate
(4.00; 44.45 mmol) in tetrahydrofuran (20 mL) was added 1-bromo-
pentane (6.04 g; 40.01 mmol) and stirred for 3 h at room tempera-
Chem. Eur. J. 2016, 22, 1010 – 1020
www.chemeurj.org
1018
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
found (calcd) for C32H48N12O6F6S2Ni: C 41.17 (41.17), H 4.82 (5.18), N
17.92 (18.00)%. IR (ATR): nmax =3137 n(CH), 2989 n(CH), 1603, 1521
n(ring), 1467 n(ring), 1450, 1354 na(SO2), 1258, 1224 na(CF3), 1163,
1106 d(CH), 1086 d(CH), 1026, 966, 936 d(ring) and g(CH), 841
g(CH), 745 ds(CF3) and g(CH), 668 g(ring), 633, 623 da(SO2), 572
da(CF3), 516 da(CF3), 422 cmÀ1; m.p. 1588C.
is greatly appreciated, as well as the help of Dr. Ward Brullot
and Johanna Jochum for the SQUID measurement, Dr. Tom
Vander Hoogerstraete with the TXRF measurement and Dirk
Henot for the CHN measurements. The authors also thank the
Hercules Foundation for supporting the purchase of an X-ray
diffractometer through project AKUL/09/0035 and support by
IoLiTec (Heilbronn, Germany).
Synthesis of [Ni(MeIm)6][OMs]2·2H2O: To [Ni(H2O)6][OMs]2 (1.50 g;
4.21 mmol) in ethanol (30 mL) was added 1-methylimidazole
(2.08 g; 25.37 mmol) and stirred for 30 min at room temperature.
Excess of ethanol was removed on a rotary evaporator and the
product was dried, overnight, on a high vacuum line, yielding
a blue solid at room temperature. Yield: 3.08 g (94%). CHN found
(calcd) for C26H46N12O8S2Ni: C 39.71 (40.16), H 6.26 (5.96), N 21.01
(21.62)%; IR (ATR): nmax =3446 n(OH), 3120 n(CH), 3012 n(CH), 2935
n(CH), 1635, 1530 n(ring), 1519, 1422 n(ring) and da(CH3), 1374
na(SO2), 1319 n(ring), 1288, 1235, 1195 ns(SO2), 1090 d(CH), 1036,
939 d(ring) and g(CH), 828 n(SO), 761, 662 g(ring), 619 g(ring), 550,
522 cmÀ1 d(SO2); m.p. 2058C.
Keywords: electrochemistry
·
ionic liquids
·
magnetic
properties · nanoparticles · X-ray diffraction
[1] T. Welton, Chem. Rev. 1999, 99, 2071–2083.
[2] K. R. Seddon, J. Chem. Technol. Biotechnol. 1997, 68, 351–356.
[3] N. V. Plechkova, K. R. Seddon, Chem. Soc. Rev. 2008, 37, 123–150.
[4] P. Wasserscheid, W. Keim, Angew. Chem. Int. Ed. 2000, 39, 3772–3789;
Angew. Chem. 2000, 112, 3926–3945.
[5] S. Zein El Abedin, E. M. Moustafa, R. Hempelmann, H. Natter, F. Endres,
ChemPhysChem 2006, 7, 1535–1543.
[6] F. Endres, ChemPhysChem 2002, 3, 144–154.
[7] A. P. Abbott, K. J. McKenzie, Phys. Chem. Chem. Phys. 2006, 8, 4265–
4279.
Synthetic procedures for all other liquid nickel salts can be found
in the Supporting Information.
[8] J. A. Welleman, F. B. Hulsbergen, J. Verbiest, J. Reedijk, J. Inorg. Nucl.
Chem. 1978, 40, 143–147.
Electrochemistry
Platinum (100 nm) covered silicon wafers were used as working
electrodes for electrodeposition. The substrates were washed with
an HCl solution and rinsed with water and acetone and dried
before use. The electrochemical experiments were conducted in
a glovebox (water and oxygen concentrations below 1 ppm) at
1208C. The experiments were performed using a Solartron Instru-
ments SI1287 electrochemical interface controlled by a computer
with Corrware software in an unstirred solution, unless mentioned
otherwise. All potential values in this study are relative to a nickel
wire directly immersed in solution (pseudo-reference electrode). All
experiments were performed at least three times on three different
batches of liquid nickel salt in order to ensure reproducibility.
[9] A. C. AlbØniz, J. Barbera, P. Espinet, C. M. Lequerica, A. M. Levelut, F. J.
Lopez-Marcos, J. L. Serrano, Eur. J. Inorg. Chem. 2000, 133–138.
[10] T. Vander Hoogerstraete, N. R. Brooks, B. Norberg, J. Wouters, K. Van
Hecke, L. Van Meervelt, K. Binnemans, CrystEngComm 2012, 14, 4902–
4911.
[11] S. Mori, T. Mochida, Organometallics 2013, 32, 780–787.
[12] Y. Funasako, T. Inagaki, T. Mochida, T. Sakurai, H. Ohta, T. Nakamura,
Dalton Trans. 2013, 42, 8317–8327.
[13] W. Wang, D. Lee, A. M. Leone, R. W. Murray, Chem. Phys. 2005, 319,
126–135.
[14] M. Iida, C. Baba, M. Inoue, H. Yoshida, E. Taguchi, H. Furusho, Chem.
Eur. J. 2008, 14, 5047–5056.
[15] M. Iida, S. Kawakami, E. Syouno, H. Er, E. Taguchi, J. Colloid Interface
Sci. 2011, 356, 630–638.
[16] A. P. Abbott, J. C. Barron, K. S. Ryder, D. Wilson, Chem. Eur. J. 2007, 13,
6495–6501.
Crystallography
[17] J. Huang, H. Luo, S. Dai, J. Electrochem. Soc. 2006, 153, J9–J13.
[18] A. P. Abbott, G. Capper, D. L. Davies, R. K. Rasheed, Chem. Eur. J. 2004,
10, 3769–3774.
Single crystals that were grown from the compounds [Ni(EtIm)6]
[Tf2N]2,
[Ni(PrIm)6][Tf2N]2,
[Ni(BuIm)6][Tf2N]2,
[Ni(EtIm)6][OTf]2,
[Ni(PrIm)6][OTf]2, [Ni(BuIm)6][OTf]2, [Ni(EtIm)6][OMs]2, [Ni(PrIm)6]
[OMs]2 and [Ni(BuIm)6][OMs]2 by recrystallization from ethanol and
which were suitable for single crystal X-ray diffraction were mount-
ed on a nylon loop attached to a copper pin and placed on an Agi-
lent SuperNova diffractometer into the cold stream of an Oxford
Cryostream 700 at 100(2) K using MoKa radiation (l=0.71073 Š).
The absorption corrections were applied using CrysAlisPro (Agilent
Technologies, Yarnton, UK, 2011). All structures were solved using
direct methods and refined by the full-matrix least-squares proce-
dure in SHELXL,[105] using the graphical interface OLEX2[106] which
was also used for the preparation of the figures. All hydrogen
atoms were placed in calculated positions and refined using
a riding model. CCDC 1439002–1439010 contain the supplementa-
ry crystallographic data for this paper. These data are provided free
of charge by The Cambridge Crystallographic Data Centre.
[19] H. M. A. Abood, A. P. Abbott, D. Ballantyne, K. S. Ryder, Chem.
Commun. 2011, 47, 3523–3525.
[20] T. M. Anderson, D. Ingersoll, A. J. Rose, C. L. Staiger, J. C. Leonard,
Dalton Trans. 2010, 39, 8609–8612.
[21] H. D. Pratt III, A. J. Rose, C. L. Staiger, D. Ingersoll, T. M. Anderson,
Dalton Trans. 2011, 40, 11396–11401.
[22] H. D. Pratt, D. Ingersoll, N. S. Hudak, B. B. McKenzie, T. M. Anderson, J.
Electroanal. Chem. 2013, 704, 153–158.
[23] N. R. Brooks, S. Schaltin, K. Van Hecke, L. Van Meervelt, K. Binnemans, J.
Fransaer, Chem. Eur. J. 2011, 17, 5054–5059.
[24] S. Schaltin, N. R. Brooks, K. Binnemans, J. Fransaer, J. Electrochem. Soc.
2011, 158, D21–D27.
[25] S. Schaltin, N. R. Brooks, L. Stappers, L. D’Urzo, H. Plank, C. Gspan, K.
Binnemans, J. Fransaer, J. Electrochem. Soc. 2011, 158, D647–D650.
[26] N. R. Brooks, S. Schaltin, K. Van Hecke, L. Van Meervelt, J. Fransaer, K.
Binnemans, Dalton Trans. 2012, 41, 6902–6905.
[27] S. Schaltin, N. R. Brooks, L. Stappers, K. Van Hecke, L. Van Meervelt, K.
Binnemans, J. Fransaer, Phys. Chem. Chem. Phys. 2012, 14, 1706–1715.
[28] D. Depuydt, N. R. Brooks, S. Schaltin, L. Van Meervelt, J. Fransaer, K.
Binnemans, ChemPlusChem 2013, 78, 578–588.
Acknowledgements
[29] S. Schaltin, N. R. Brooks, J. Sniekers, D. Depuydt, L. Van Meervelt, K. Bin-
nemans, J. Fransaer, Phys. Chem. Chem. Phys. 2013, 15, 18934–18943.
[30] J. Sniekers, N. R. Brooks, S. Schaltin, L. Van Meervelt, J. Fransaer, K. Bin-
nemans, Dalton Trans. 2014, 43, 1589–1598.
[31] S. Schaltin, N. R. Brooks, J. Sniekers, L. Van Meervelt, K. Binnemans, J.
Fransaer, Chem. Commun. 2014, 50, 10248–10250.
The authors acknowledge financial support by the FWO-Flan-
ders (research project G.0B9613.N and research community
“Ionic Liquids”). The help of Prof. Johan Billen, Dr. Maarten
Bloemen and Tom van der Donck with the TEM measurements
Chem. Eur. J. 2016, 22, 1010 – 1020
www.chemeurj.org
1019
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[32] G. Vanhoutte, N. R. Brooks, S. Schaltin, B. Opperdoes, L. Van Meervelt,
J.-P. Locquet, P. M. Vereecken, J. Fransaer, K. Binnemans, J. Phys. Chem.
C 2014, 118, 20152–20162.
[33] G. Roventi, R. Fratesi, R. A. Della Guardia, G. Barucca, J. Appl. Electro-
chem. 2000, 30, 173–179.
[71] P. N. Bartlett, P. R. Birkin, M. a. Ghanem, Chem. Commun. 2000, 1671–
1672.
[72] L. Xu, W. L. Zhou, C. Frommen, R. H. Baughman, A. A. Zakhidov, L. Mal-
kinkski, J.-Q. Wang, J. B. Wiley, Chem. Commun. 2000, 1, 997–998.
[73] A. I. Yanson, P. Rodriguez, N. Garcia-Araez, R. V. Mom, F. D. Tichelaar,
M. T. M. Koper, Angew. Chem. Int. Ed. 2011, 50, 6346–6350; Angew.
Chem. 2011, 50, 6346–6350.
[34] S. Chouchane, a. Levesque, J. Douglade, R. Rehamnia, J. P. Chopart,
Surf. Coat. Technol. 2007, 201, 6212–6216.
[35] F. J. Fabri Miranda, O. E. Barcia, O. R. Mattos, R. Wiart, J. Electrochem.
Soc. 1997, 144, 3441–3449.
[36] M. M. Abou-Krisha, Appl. Surf. Sci. 2005, 252, 1035–1048.
[37] T. P. Moffat, J. Electrochem. Soc. 1994, 141, 3059–3070.
[38] M. R. Ali, A. Nishikata, T. Tsuru, J. Electroanal. Chem. 2001, 513, 111 –
118.
[39] M. Li, Z. Wang, R. G. Reddy, J. Electrochem. Soc. 2014, 161, D150–D153.
[40] S. P. Gou, I. W. Sun, Electrochim. Acta 2008, 53, 2538–2544.
[41] M. J. Deng, I. W. Sun, P. Y. Chen, J. K. Chang, W. T. Tsai, Electrochim. Acta
2008, 53, 5812–5818.
[74] M. T. Reetz, M. Maase, Adv. Mater. 1999, 11, 773–777.
[75] M. T. Reetz, M. Winter, R. Breinbauer, T. Thurn-Albrecht, W. Vogel,
Chem. Eur. J. 2001, 7, 1084–1094.
[76] R. Fukui, Y. Katayama, T. Miura, J. Electrochem. Soc. 2011, 158, D567–
D572.
[77] Y. L. Zhu, Y. Kozuma, Y. Katayama, T. Miura, Electrochim. Acta 2009, 54,
7502–7506.
[78] Y. L. Zhu, Y. Katayama, T. Miura, Electrochim. Acta 2010, 55, 9019–9023.
[79] Y.-L. Zhu, Y. Katayama, T. Miura, Electrochem. Solid-State Lett. 2011, 14,
D110–D115.
[42] H. Y. Yang, X. W. Guo, X. B. Chen, S. H. Wang, G. H. Wu, W. J. Ding, N.
Birbilis, Electrochim. Acta 2012, 63, 131–138.
[43] M. Ganapathi, S. V. Eliseeva, N. R. Brooks, D. Soccol, J. Fransaer, K. Bin-
nemans, J. Mater. Chem. 2012, 22, 5514–5522.
[44] K. L. Fow, M. Ganapathi, I. Stassen, J. Fransaer, K. Binnemans, D. E.
De Vos, Chem. Commun. 2013, 49, 8498–8500.
[45] R. Boese, H.-C. Weiss, D. Bläser, Angew. Chem. Int. Ed. 1999, 38, 988–
992; Angew. Chem. 1999, 111, 1042–1045.
[46] V. R. Thalladi, R. Boese, H.-C. Weiss, Angew. Chem. Int. Ed. 2000, 39,
918–922; Angew. Chem. 2000, 112, 942–946.
[47] G. Adamovµ, J. N. Canongia Lopes, L. P. N. Rebelo, L. M. N. B. Santos,
K. R. Seddon, K. Shimizu, Phys. Chem. Chem. Phys. 2014, 16, 4033–
4038.
[80] Y. Katayama, R. Fukui, T. Miura, Electrochemistry 2013, 81, 532–534.
[81] Y. Katayama, Y. Oshino, N. Ichihashi, N. Tachikawa, K. Yoshii, K. Toshima,
Electrochim. Acta 2015, 183, 37–41.
[82] T. Torimoto, K. I. Okazaki, T. Kiyama, K. Hirahara, N. Tanaka, S. Kuwabata,
Appl. Phys. Lett. 2006, 89, 243117.
[83] D. B. Williams, M. E. Stoll, B. L. Scott, D. A. Costa, W. J. Oldham, Chem.
Commun. 2005, 1438–1440.
[84] M. P. Soriaga, J. H. White, D. Song, A. T. Hubbard, J. Electroanal. Chem.
1984, 171, 359–363.
[85] B. Ren, Q. J. Huang, W. B. Cai, B. W. Mao, F. M. Liu, Z. Q. Tian, J. Electroa-
nal. Chem. 1996, 415, 175–178.
[86] B. H. Loo, Y. Tse, K. Parsons, C. Adelman, A. El-Hage, Y. G. Lee, J. Raman
Spectrosc. 2006, 37, 299–304.
[48] J. Leys, C. S. P. Tripathi, C. Glorieux, S. Zahn, B. Kirchner, S. Longuemart,
K. C. Lethesh, P. Nockemann, W. Dehaen, K. Binnemans, Phys. Chem.
Chem. Phys. 2014, 16, 10548–10557.
[49] S. Bouarab, A. Vega, M. J. López, M. P. IÇiguez, J. A. Alonso, Phys. Rev. B
1997, 55, 13279–13282.
[50] S. Apsel, J. Emmert, J. Deng, L. Bloomfield, Phys. Rev. Lett. 1996, 76,
1441–1444.
[51] I. M. Billas, A. Châtelain, W. A. de Heer, Science 1994, 265, 1682–1684.
[52] R. H. Kodama, J. Magn. Magn. Mater. 1999, 200, 359–372.
[53] T. Hyeon, Chem. Commun. 2003, 927–934.
[54] F. A. Reuse, S. N. Khanna, Chem. Phys. Lett. 1995, 234, 77–81.
[55] M. Castro, C. Jamorski, D. R. Salahub, Chem. Phys. Lett. 1997, 271, 133–
142.
[56] N. Cordente, C. Amiens, B. Chaudret, M. Respaud, F. Senocq, M. J. Casa-
nove, J. Appl. Phys. 2003, 94, 6358–6365.
[57] C. Kittel, Phys. Rev. 1946, 70, 965–971.
[58] F. Fonseca, G. Goya, R. Jardim, R. Muccillo, N. CarreÇo, E. Longo, E.
Leite, Phys. Rev. B 2002, 66, 1–5.
[87] S. Baldelli, Acc. Chem. Res. 2008, 41, 421–431.
[88] W. Zhou, S. Inoue, T. Iwahashi, K. Kanai, K. Seki, T. Miyamae, D. Kim, Y.
Katayama, Y. Ouchi, Electrochem. Commun. 2010, 12, 672–675.
[89] R. Atkin, G. G. Warr, J. Phys. Chem. C 2007, 111, 5162–5168.
[90] R. Hayes, N. Borisenko, M. K. Tam, P. C. Howlett, F. Endres, R. Atkin, J.
Phys. Chem. C 2011, 115, 6855–6863.
[91] F. Endres, O. Hçfft, N. Borisenko, L. H. Gasparotto, A. Prowald, R. Al-
Salman, T. Carstens, R. Atkin, A. Bund, S. Z. El Abedin, Phys. Chem.
Chem. Phys. 2010, 12, 1724–1732.
[92] F. Endres, Phys. Chem. Chem. Phys. 2012, 14, 5008–5009.
[93] M. Gnahm, T. Pajkossy, D. M. Kolb, Electrochim. Acta 2010, 55, 6212–
6217.
[94] M. Gnahm, D. M. Kolb, J. Electroanal. Chem. 2011, 651, 250–252.
[95] T. Pajkossy, D. M. Kolb, Electrochem. Commun. 2011, 13, 284–286.
[96] M. Gnahm, C. Berger, M. Arkhipova, H. Kunkel, T. Pajkossy, G. Maas,
D. M. Kolb, Phys. Chem. Chem. Phys. 2012, 14, 10647–10652.
[97] M. Tułodziecki, J.-M. Tarascon, P. L. Taberna, C. GuØry, Electrochim. Acta
2014, 134, 55–66.
[59] E. Ramírez-Meneses, I. Betancourt, F. Morales, V. Montiel-Palma, C. C.
Villanueva-Alvarado, M. E. Hernµndez-Rojas, J. Nanopart. Res. 2011, 13,
365–374.
[60] M. Respaud, J. M. Broto, H. Rakoto, A. Fert, L. Thomas, B. Barbara, M.
Verelst, E. Snoeck, P. Lecante, A. Mosset, J. Osuna, T. Ely, C. Amiens, B.
Chaudret, Phys. Rev. B 1998, 57, 2925–2935.
[61] G. Schmid, L. F. Chi, Adv. Mater. 1998, 10, 515–526.
[62] Z. Wang, J. Phys. Chem. B 2000, 104, 1153–1175.
[63] I. Geukens, E. Plessers, J. W. Seo, D. E. De Vos, Eur. J. Inorg. Chem. 2013,
2623–2628.
[64] S. L. Che, K. Takada, K. Takashima, O. Sakurai, K. Shinozaki, N. Mizutani,
J. Mater. Sci. 1999, 34, 1313–1318.
[98] A. Ponce, H. B. Gray, J. R. Winkler, J. Am. Chem. Soc. 2000, 122, 8187–
8191.
[99] J. Huang, T. Jiang, B. Han, H. Gao, Y. Chang, G. Zhao, W. Wu, Chem.
Commun. 2003, 1654–1655.
[100] F. Fernµndez, B. Cordero, J. Durand, G. Muller, F. Malbosc, Y. Kihn, E.
Teuma, M. Gomez, Dalton Trans. 2007, 5572–5581.
[101] J. E. Bara, Ind. Eng. Chem. Res. 2011, 50, 13614–13619.
[102] J. B. Hodgson, G. C. Percy, D. A. Thornton, J. Mol. Struct. 1980, 66, 81–
92.
[103] I. Rey, P. Johansson, J. Lindgren, J. C. Lass›gues, J. Grondin, L. Servant,
J. Phys. Chem. A 1998, 102, 3249–3258.
[104] M. Wilms, M. Kruft, G. Bermes, K. Wandelt, Rev. Sci. Instrum. 1999, 70,
3641–3650.
[105] G. M. Sheldrick, Acta Crystallogr. Sect. A 2015, 71, 3–8.
[106] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. K. Howard, H. Puschmann,
H. Puschmann, J. Appl. Crystallogr. 2009, 42, 339–341.
[65] G. B. Thompson, R. Banerjee, X. D. Zhang, P. M. Anderson, H. L. Fraser,
Acta Mater. 2002, 50, 643–651.
[66] W. Xu, K. Y. Liew, H. Liu, T. Huang, C. Sun, Y. Zhao, Mater. Lett. 2008, 62,
2571–2573.
[67] L. Bai, F. Yuan, Q. Tang, Mater. Lett. 2008, 62, 2267–2270.
[68] Z. G. Wu, M. Munoz, O. Montero, Adv. Powder Technol. 2010, 21, 165–
168.
[69] Y. Hou, H. Kondoh, T. Ohta, S. Gao, Appl. Surf. Sci. 2005, 241, 218–222.
[70] J. Zoval, R. Stinger, P. Biernacki, R. Penner, J. Phys. Chem. 1996, 100,
837–844.
Received: October 14, 2015
Published online on December 8, 2015
Chem. Eur. J. 2016, 22, 1010 – 1020
www.chemeurj.org
1020
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim