Copper(ii) and triphenylphosphine copper(i) ethylene glycol carboxylates: Synthesis, characterisation and copper nanoparticle generation

Article abstract of DOI:10.1039/c3dt51913a

Ethylene glycol-functionalised copper(ii) carboxylates Cu[O ₂CCR₂(OC₂H₄)_nOCH ₃]₂ (n = 0-3; R = H, Me) (2a-e) have been prepared by the reaction of [Cu₂(OAc)₄·2H₂O] with CH₃O(C₂H₄O)_nCR₂CO ₂H (1a-e). Upon reduction of 2a-e with triphenylphosphine, the corresponding tris(triphenylphosphine)copper(i) complexes 3a-e were obtained, which could be converted to the bis(triphenylphosphine)copper(i) complexes 4a-e by removal of one phosphine ligand. Based on IR spectroscopy and single crystal X-ray structure analysis the binding motif of the carboxylato group on the copper ion is discussed. DSC, TG and TG-MS experiments were performed to analyse the thermal decomposition mechanism of 2-4. Complex 4c was used as a precursor for the generation of copper nanoparticles by thermal decomposition in hexadecylamine without the need of any further reactants. Depending on the precursor concentration, spherical copper nanoparticles with a mean diameter ranging from 10 to 85 nm as well as nanorods with a length of up to 1.3 μm (aspect ratios ranging between 2 and 32) were obtained. Electron diffraction analysis of the rods suggested that they consist of five domains which are arranged around a fivefold rotational axis. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3dt51913a

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Copper(II) and triphenylphosphine copper(I) ethylene
glycol carboxylates: synthesis, characterisation and
copper nanoparticle generation†‡
Cite this: Dalton Trans., 2013, 42, 15599
David Adner,a Stefan Möckel,a Marcus Korb,a Roy Buschbeck,a Tobias Rüffer,a
Steffen Schulze,b Lutz Mertens,c Michael Hietschold,b Michael Mehringc and
Heinrich Lang*a
Ethylene glycol-functionalised copper(II) carboxylates Cu[O2CCR2(OC2H4)nOCH3]2 (n = 0–3; R = H, Me)
(2a–e) have been prepared by the reaction of [Cu2(OAc)4·2H2O] with CH3O(C2H4O)nCR2CO2H (1a–e).
Upon reduction of 2a–e with triphenylphosphine, the corresponding tris(triphenylphosphine)copper(I)
complexes 3a–e were obtained, which could be converted to the bis(triphenylphosphine)copper(I) com-
plexes 4a–e by removal of one phosphine ligand. Based on IR spectroscopy and single crystal X-ray struc-
ture analysis the binding motif of the carboxylato group on the copper ion is discussed. DSC, TG and
TG-MS experiments were performed to analyse the thermal decomposition mechanism of 2–4. Complex
4c was used as a precursor for the generation of copper nanoparticles by thermal decomposition in
hexadecylamine without the need of any further reactants. Depending on the precursor concentration,
spherical copper nanoparticles with a mean diameter ranging from 10 to 85 nm as well as nanorods
with a length of up to 1.3 µm (aspect ratios ranging between 2 and 32) were obtained. Electron diffrac-
tion analysis of the rods suggested that they consist of five domains which are arranged around a
fivefold rotational axis.
Received 15th July 2013,
Accepted 27th August 2013
DOI: 10.1039/c3dt51913a
www.rsc.org/dalton
well as sonochemical6 and sonoelectrochemical methods.7
These techniques are alternatives to the traditional three-com-
1 Introduction
Persistent innovations in the field of nanotechnology led to ponent approach in which copper nanoparticles are prepared
the development of numerous new methods for the prepa- by adding a reducing agent to a solution of a copper com-
ration of copper nanoparticles, among which are the prepa- pound in the presence of a stabilising agent.8
ration in micelles,1 spray pyrolysis,2 sputtering approaches,3
A non-classical nanoparticle preparation method with
nanosphere lithography,4 chemical vapor deposition (CVD)5 as growing importance is the single-source precursor approach
which aims to simplify the particle generation process by com-
bining two or three of the traditional components (vide supra)
in one molecule.9 Usually, the compounds used as precursors
are metal–organic complexes which decompose thermally to
generate the respective copper(0), e.g. by the decomposition of
copper(II) acetylacetonate at 230 °C in oleylamine.9c, f,i
aFaculty of Natural Sciences, Institute of Chemistry, Inorganic Chemistry,
Technische Universität Chemnitz, D-09107 Chemnitz, Germany.
E-mail: heinrich.lang@chemie.tu-chemnitz.de; Fax: +49 (0) 371 531 21219;
Tel: +49 (0) 371 531 21210
bFaculty of Natural Sciences, Solid Surfaces Analyses, Technische Universität
Chemnitz, D-09107 Chemnitz, Germany. E-mail: hietschold@physik.tu-chemnitz.de;
Fax: +49 (0) 371 531 21639; Tel: +49 (0) 371 531 21630
cFaculty of Natural Sciences, Institute of Chemistry, Coordination Chemistry,
Technische Universität Chemnitz, D-09107 Chemnitz, Germany.
The thermal decomposition route for nanoparticle syn-
thesis has considerable advantages, e.g. it allows easy reaction
control because the system is much less complex than in
multi-component approaches. Usually the decomposition is a
reaction of first order and no injection rates or concentration
effects have to be considered, which provides a high reproduci-
bility.9b No addition of strong reducing agents such as hydra-
zine is necessary; this simplifies the application of the
particles in catalytic or biological processes and makes the
actual particle generation more secure and economical.10
Despite the low complexity of the reaction system systematic
E-mail: michael.mehring@chemie.tu-chemnitz.de; Fax: +49 (0) 371 531 21219;
Tel: +49 (0) 371 531 35128
†Dedicated to Prof. Dr Werner Uhl, Universität Münster, on the occasion of his
60th birthday.
‡Electronic supplementary information (ESI) available: IR and NMR (1H,
13C{1H}) spectra, crystal structures and data, TG and TG-MS measurements,
XRPD data, additional TEM images, electron diffractograms. CCDC
932855–932857 and 933501. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c3dt51913a
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 15599–15609 | 15599

View Article Online
Paper
Dalton Transactions
optimisation is possible, e.g. by variation of the chemical struc-
ture of the precursor.
2 Synthesis and characterisation
A common family of single-source precursors for the gene- The ethylene glycol-functionalised copper(II) carboxylates 2a–e
ration of copper is its carboxylates. In the absence of oxygen have been prepared by an anion exchange reaction of the car-
they are known to generate copper(0) by decarboxylation at boxylic acids CH3O(C2H4O)nCR2CO2H (n = 0–3; R = H, Me)
temperatures between 200 and 300 °C.11 For the solution syn- (1a–e) with copper(II) acetate hydrate (Scheme 1). Acetic acid
thesis of copper nanoparticles a variety of carboxylates such as and water were removed from the reaction mixture by an azeo-
copper(I) acetate,9f,g copper(II) formate,9d copper(II) oxalate9e tropic distillation with toluene as entrainer, thus shifting the
and copper(II) hydrazine carboxylate9a have been investigated. equilibrium to the product side. After removal of the solvent
Besides carboxylates, copper CVD precursors, such as copper(II) no further workup was necessary. The acids 1a–c were com-
acetylacetonate9c,f,i and copper(II) 1-dimethylamino-2-propa- mercially available, 1d–e have been synthesised by the reaction
nolate,9b were used.
of the corresponding methyl ethylene glycol with 2-bromo-
Despite the large number of investigated precursors, sys- 2-methylpropanoic acid or 2-bromoacetic acid (see the Experi-
tematic optimisation of the precursor properties is rarely per- mental section for details).17
formed. One of the most prevalent problems in all investigated
Copper(II) methoxyacetate 2a was obtained as a green solid,
precursors is reported to be the high oxidation sensitivity of whereas 2b–e are green oils which show a decrease of viscosity
the particle generation process, which often prevented the iso- with increasing chain length. They are highly soluble in protic
lation of pure copper nanoparticles9c,d, f or demanded and aprotic solvents, e.g. in water as well as in toluene. The
advanced working techniques such as the use of glove- identity of the paramagnetic copper(II) carboxylates was proven
boxes.9b,i This issue is due to the limited number of reducing by elemental analysis, IR spectroscopy and ESI-TOF mass
equivalents, which does not exceed one in all mentioned precur- spectrometry.
sors. An easy way to raise the number of reducing equivalents
The tris(triphenylphosphine)copper(I) carboxylates 3a–e are
and thus to lower the oxidation sensitivity of the particle accessible by the reaction of the copper(II) carboxylates 2a–e
preparation process is the use of phosphine copper(I) carboxy- with 3.5 equiv. of triphenylphosphine in the presence of 0.5
lates, since phosphine ligands possess reducing properties. equiv. of water, of which copper(II) is reduced to copper(I) by
Furthermore, triphenylphosphine copper(I) carboxylates are 0.5 equiv. of the phosphine.12 The reaction involves a nucleo-
stable species which are easy to handle and to synthesise.12
philic attack of the carboxylate on the phosphine and the
Recently, we reported about the use of ethylene glycol-func- hydrolysis of the resulting phosphonium ion by water to form
tionalised metal carboxylates for the straightforward gene- triphenylphosphine oxide and the corresponding carboxylic
ration of silver,13 gold14 and rhodium15 nanoparticles at low acid. Three molar equivalents of PPh3 act as ligands to stabi-
temperatures. Thermally initiated decarboxylation was used to lise the soft copper(I) ion. After recrystallisation from metha-
afford reactive metal(0) colloids, whereas the ethylene glycol nol, the products are obtained as colourless, air stable solids.
chains were sufficient to stabilise the nanoparticles of silver One of the three ligands is bound only weakly to the metal and
and gold even in non-coordinating solvents. First results con- can be removed by two-solvent recrystallisation, i.e. by solving
cerning the synthesis and characterisation of bis(triphenylpho- the compounds in dichloromethane and adding nhexane. In
sphine)copper(I) 2-[2-(2-methoxyethoxy)ethoxy]acetate (4c) and this manner the bis(triphenylphosphine)copper(I) carboxylates
its use as a single-source precursor for the generation of 4a–e were obtained as colourless solids. Analyses of 3–4
spherical copper nanoparticles were reported lately.16 Herein, included elemental analysis, IR and NMR spectroscopy (1H,
we give a full report of the synthesis and characterisation of a 13C{1H}, 31P{1H}), ESI-TOF mass spectrometry as well as single-
series of novel copper(I) and copper(II) ethylene glycol carboxy- crystal X-ray structure analysis.
lates. In addition, the application of 4c to the solution syn-
thesis of spherical as well as rod-like copper nanoparticles reliable method to investigate the binding motif of the carb-
without the addition of any reducing agents is described.
oxylato group to a metal.18 With reference to ionic carboxylates
IR-spectroscopy has proven to be a fast and in most cases
Scheme 1 Synthesis of the ethylene glycol-functionalised copper(II) and copper(I) carboxylates 2–4: (i) toluene, 111 °C, 5 h, azeotropic distillation; (ii) 3.5PPh3,
0.5H2O, MeOH, 65 °C, 3 h; (iii) two-solvent recrystallisation from dichloromethane/nhexane.
15600 | Dalton Trans., 2013, 42, 15599–15609
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
(e.g., their sodium or potassium salts), bridging and chelating can be assumed but no certain conclusions can be made
carboxylates show a smaller difference Δ˜νCO of the symmetric about 4a, c and e (166–190 cm−1), since these values are close
2
and asymmetric ˜νCO valence absorptions, whereas mono- to the values of the respective potassium salts (vide supra).
2
dentate bonded groups show a significantly increased Δ˜νCO
.
For a precise investigation of the questionable structures
2
As reference substances we prepared the potassium carboxy- 4a–e we performed single crystal X-ray structure analysis (Fig. 1),
lates of the acids 1c and 1d,19 their ˜νCO differences were found out of which we recently reported the structure of 4c.16 The
2
to be 185 cm−1 and 193 cm−1, respectively.
remaining complexes crystallise in monoclinic (P21/n, 4e), tri-
ˉ
The ˜νCO values of complexes 2 are indicative of monoden- clinic (P1; 4a and d) or orthorhombic (Pbca, 4b) space groups
2
tate coordination (Table 1). This corresponds well to the without any additional solvent molecules. Selected bond
crystal structures of the hydrates of 2a and 2b as determined lengths and angles are summarised in Table 2, while selected
by Rossotti et al.20 and Hawkins et al.21 In contrast to the well crystallographic and measurement data are given in the ESI.‡
known dimeric paddle-wheel structure of copper(II) acetate
Complexes 4a, b, d and e show a similar structure to the
hydrate, monomeric structures with η1-bonded carboxylato copper(I) ion situated in the expected distorted tetrahedral
groups were found in which one ether oxygen per chain com- arrangement formed by the two PPh3 ligands and the two
pletes the coordination sphere of the metal. For the tris(triphe- oxygen atoms of the bidentate carboxylato group. The P1–Cu1–P2
nylphosphine)copper(I) carboxylates 3a–e the Δ˜νCO values are angle is enlarged due to the spatially demanding triphenyl-
2
between 223 cm−1 (3b) and 230 cm−1 (3e). This indicates a phosphine units, whereas the O1–Cu1–O2 angle is signifi-
monodentate binding motif, which was confirmed by X-ray cantly smaller since it is part of a four-membered ring.
structure analysis for 3a, b and d (see ESI‡). For the bis(triphe- Complexes 4b and 4d evince quite similar Cu–O-bond lengths
nylphosphine) compounds 4a–e the Δ˜νCO values are generally which correlate well with the small differences of the Δ˜νCO
2
2
smaller than for 2 and 3. This resembles the expected trend to valence absorptions (Table 1). In complexes 4a and e the Cu–O-
η2-bonded carboxylato groups which compensate for the bond lengths differ more, but the carboxylato group is still
missing third ligand. However, the Δ˜νCO values show a dis- clearly η2-bond in both cases. With these results, the structures
2
tinct variation, ranging from 153 cm−1 to 190 cm−1. For 4b and of 4a, b, d and e are in contrast to the structure of 4c, for which
d (153 and 154 cm−1, resp.) η2-bounded carboxylato groups we found a η1-bonded carboxylato group.16 Furthermore, we
noted a distinct intermolecular van der Waals interaction
Table 1 ˜vCO absorptions and Δ˜vCO values for complexes 2–4
2
ΔvCO [cm−1
]
ΔvCO [cm−1
]
ΔvCO a [cm−1
]
˜
˜
˜
Compd.
2
2
2
Table 2 Selected bond lengths [Å] and angles [°] of 4a, b, d and ea
2a
2b
2c
2d
2e
3a
3b
3c
3d
3e
4a
4b
4c
4d
4e
1393
1415
1416
1412
1416
1373
1373
1381
1380
1379
1407
1410
1403
1402
1433
1602
1630
1635
1625
1635
1597
1596
1607
1599
1599
1582
1564
1593
1556
1599
209
215
219
213
219
224
223
226
229
230
175
153
190
154
166
4a
4b
4d
4e
Cu1–P1
Cu1–P2
Cu1–O1
Cu1–O2
C1–O1
C1–O2
P1–Cu1–P2
O1–Cu1–O2
O1–C1–O2
P1–Cu1–O1
P2–Cu1–O2
2.2410(8)
2.2507(7)
2.1126(18)
2.3575(19)
1.254(3)
1.242(3)
125.81(3)
58.58(6)
2.2074(6)
2.2504(6)
2.1388(15)
2.2556(16)
1.257(3)
1.250(3)
132.75(2)
60.30(6)
2.2218(7)
2.2533(7)
2.1959(19)
2.2193(19)
1.257(3)
1.2611(3)
129.50(3)
60.13(7)
2.2526(6)
2.2048(6)
2.1202(14)
2.3013(14)
1.256(2)
1.261(2)
131.72(2)
59.93(5)
123.4(2)
104.54(5)
116.64(5)
123.6(2)
102.23(5)
116.05(5)
123.0(2)
100.77(6)
109.88(5)
123.22(19)
98.76(4)
117.16(4)
a The estimated standard deviations of the last significant digits are
shown in parentheses.
a Δ˜νCO = ˜νCO , asym − Δ˜νCO , sym
.
2
2
2
Fig. 1 ORTEP diagrams (50% probability level) of 4a, 4b, 4d and 4e (from left to right). In the case of 4b, only one conformation of the disordered chain is shown.
All hydrogen atoms are omitted for clarity.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 15599–15609 | 15601

View Article Online
Paper
Dalton Transactions
between the ethylene glycol chains in 4c and d, which might Complexes 2a, c and e have the same decomposition tempera-
serve as an explanation for the aberrant structure of 4c.
ture, whereas the decomposition of the dimethyl-substituted
derivate 2d starts already at 190 °C. The significant lower
decomposition temperature of 2a–e in comparison with
copper(II) acetate (260 °C) can be explained by a lower energy
of the radical decarboxylation products due to the presence of
the stabilising aliphatic groups. The residual masses of all
samples fit well to the copper content of the compounds. The
identity of the residues was proven by X-ray powder diffraction
(XRPD) measurements, confirming the formation of elemental
copper (see ESI‡). The first derivative of the TG curve indicates
that the decomposition occurs in two steps. The DSC measure-
ment shows both to be endothermic processes.
The TG traces of the bis- and tris(triphenylphosphine) com-
plexes 3 and 4 show mass decays starting slowly at ca. 200 °C
and ending at 330 °C. No dependence of the decomposition
temperature on the degree of substitution at the α-position of
the carboxylato group is found. The residual masses of all
probes fit well to the exclusive formation of copper. The DSC
measurements show melting points ranging from 52 to 132 °C
(3e and 3a, resp.) for the tris(triphenylphosphine)copper(I) car-
boxylates and from 110 °C to 181 °C (4e and 4b, resp.) for the
bis(triphenylphosphine)copper(I) compounds. The thermal
decompositions give rise to single endothermic signals in the
DSC studies.
3 TG, DSC and TG-MS investigations
The thermal behaviour of 2–4 was studied by thermogravi-
metry (TG) and differential scanning calorimetry (DSC) to gain
information on the decomposition temperature and the degra-
dation mechanism of these species. Exemplarily, the TG and
DSC traces of 2b and 4b are shown in Fig. 2. Please notice that
the traces of the tris(triphenylphosphine)copper(I) complexes 3
are similar to that of the bis(triphenylphosphine) carboxylates
4 and hence are depicted in the ESI‡ along with further TG
traces of complexes 2 and 4.
The thermal decomposition of the monosubstituted carb-
oxylate 2b starts at an onset temperature of 220 °C (Fig. 2).
TG-MS studies have been performed to investigate the
decomposition behaviour of 2–4 in detail (2 and 4: Fig. 2; 3:
ESI‡). The two decomposition steps of the copper(II) carboxy-
late 2b can be clearly distinguished by the observation of
different characteristic fragments. The first step is characterised
+
by the mass fragments m/z = 44 (CO2 ) as well as m/z = 28 (CO)
and m/z = 22 (CO2++) (both not shown), indicating the step to be
a decarboxylation. It involves homolytic C–C- and C–O-bond
cleavage and is accompanied by the formation of copper(0). The
remaining organic radicals might decompose or recombine to
form longer ethylene glycol chains. The second decomposition
step is observed at higher temperatures and gives rise to frag-
ments such as m/z = 29 (C2H5 ), 31 (OMe+), 45 (C2H5O+) and
+
+
60 (C2H4O2 ) as well as to the ion characteristic for the intact
+
chain at m/z = 89 (CH3OC2H4OCH2 ).
Regarding 4b, the typical fragmentation products of triphe-
nylphosphine (m/z = 78, 108) are observed to begin at a temp-
erature of ca. 200 °C, which proves that the decomposition
starts with the loss of the phosphine ligands. A similar behav-
iour has been described by W. Reichle et al. for tris(triphenyl-
phosphine)copper(I) chloride22 and was also found for bis-
(trinbutylphosphine)copper(I) carboxylates.23 In
a second
decomposition step beginning at a temperature of 260 °C, de-
carboxylation and decomposition of the chain are observed as
simultaneous processes by fragment formation of m/z =
+
+
31 (OMe+), 44 (CO2 ) and 60 (C2H4O2 ).
The formation of metallic copper by thermolysis of 2–4 in
defined decomposition mechanisms makes the compounds
suitable precursors for the generation of copper(0) in various
applications, e.g. in the preparation of copper nanoparticles or
in metal–organic inks.
Fig. 2 Thermal decomposition of copper carboxylates 2b (top) and 4b
(bottom). Upper parts: TG, TG’ and DSC traces; lower parts: selected MS scans of
the decomposition products. Fragments: m/z
44 (CO2+, C2H4O+), 45 (C2H5O+), 60 (C2H4O2+), 78 (C6H6+), 89 (CH3OC2H4OCH2+),
=
29 (C2H2+), 31 (OCH3+),
108 (C6H5P+). Ar gas flow 60 mL min−1, heating rate 5 K min−1
.
15602 | Dalton Trans., 2013, 42, 15599–15609
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
4 Nanoparticle formation and
characterisation
The copper(I) and copper(II) carboxylates 2–4 were applied as
precursors in the solution synthesis of copper nanoparticles
without the addition of any reducing agents. As a high-boiling
solvent we have chosen 1-hexadecylamine (mp. 40 °C,
bp. 330 °C), which is available in a higher purity than the
common oleylamine. By coordination to the metal it acts as a
stabilising component and, by its long alkyl chain, it is able to
retard oxidation of the particles more effectively than short-
chained amines, e.g. 1-octylamine. After preliminary studies
we focused on the triphenylphosphine copper(I) carboxylates 3
and 4 because, due to the reducing properties of the phos-
phine ligands, the reaction system is much less sensitive to
oxygen than a system with the copper(II) carboxylates 2a–e. We
found no significant differences between the bis- and tris(tri-
phenylphosphine)copper(I) complexes and decided to use the
former due to their higher copper content.
For the nanoparticle synthesis, solutions of the bis(triphe-
nylphosphine)copper(I) carboxylate 4c in 1-hexadecylamine
were heated to reflux. While keeping the temperature at 330 °C
for 10 min the formation of copper nanoparticles was observed
by the sudden appearance of a red colour. After cooling the
reaction mixture to ambient temperature, the solidified reac-
tion mixtures were dissolved in nhexane. Addition of ethanol
led to the precipitation of the particles which were separated
by centrifugation, washed with ethanol and redispersed in
nhexane for further analysis.
Fig. 3 Characterisation of copper nanoparticles. Left: UV-vis spectra, measured
in nhexane; right: XRPD of 1 mM sample and reflex pattern of copper (ICDD 00-
04-0836), copper(I) oxide (ICDD 01-071-3645) and copper(II) oxide (ICDD 01-
073-6023) for comparison.
The copper nanoparticles obtained from the 1 mM batch
were investigated by XRPD (Fig. 3). Three reflexes were
observed which were assigned to the (111) plane at 43.3°, the
(200) plane at 50.4° and the (220) plane at 74.1° of the face-
centred cubic cell of copper (ICDD 00-004-0836). No reflexes of
copper(I) or copper(II) oxide were found, which indicates that
the amount of crystalline oxides is below the detection level.
The reflex broadening, which is characteristic for nanoparti-
cles, was used to estimate the crystallite size using the Scherrer
equation. By analysis of the (111) reflex, a size of about
31 2 nm was calculated. This result is in accordance with the
data obtained by microscopic measurements (vide infra).
The exact size, size distribution and shape of the copper
nanoparticles were determined by transmission electron
microscopy (TEM, Fig. 4 and 5). All particles in a selected area
on the TEM grid were counted (at least 200 particles), exclud-
ing only those in undistinguishable aggregates. For the
Different concentrations of the precursor solutions were
investigated to study the influence of concentration on the par-
ticle size, shape and size distribution (Table 3).
Copper nanoparticle dispersions are known to absorb elec-
tromagnetic radiation at wavelengths of 550–600 nm due to
excitation of the particles plasmon resonance, whereby the
resonance frequency depends on the size and shape of the par-
ticles.24 However, interpretation is more difficult for copper
than for its higher homologues due to further influences
specific for this metal, such as the degree of surface oxi-
dation.25 It was found that the plasmon absorption of the pre-
n
pared nanoparticles in hexane shifts from 570 nm to 580 nm
for precursor concentration rising from 0.5 mM to 2.0 mM
(Fig. 3). This indicates that larger particles were formed at
higher concentrations.
Table 3 Comparison of spherical copper nanoparticles obtained by thermal
decomposition of 4c in hexadecylamine (d: mean diameter, σ: standard
deviation)16
c [mM]
T [°C]
da [nm]
σa [nm]
0.5
1.0
2.0
330
330
330
9.8
28.2
85.6
1.7
2.6
15.0
Fig. 4 TEM images and size distributions of copper nanoparticles at different
precursor concentrations: (A) c = 0.5 mM; (B) c = 1.0 mM, inset: penta-twinned
particle.
a Obtained by TEM measurements (vide infra).
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 15599–15609 | 15603

View Article Online
Paper
Dalton Transactions
Fig. 5 TEM image and size distribution of copper nanoparticles at a precursor
concentration of c = 2.0 mM.
mathematical interpretation, the size distributions were fitted
by Gaussian functions.
For a precursor concentration of 0.5 mM copper particles
with a mean diameter of d = 9.8 nm and a standard deviation
of σ = 1.7 nm were generated (Fig. 4A), whereas a moderate
increase in precursor concentration to 1.0 mM resulted in the
formation of significantly larger particles which had a mean
diameter of d = 28.2 nm (σ = 2.6 nm, Fig. 4B).16 The particles
of both samples were spherical and had size variations cv =
d/σ × 100% of 17% (0.5 mM sample) and 9% (1.0 mM sample).
By this, they cannot be regarded as monodisperse by strict cri-
teria (cv < 5%), however, they are in the order of the also common
15% criterion.26 Due to their rather small size variation, the par-
Fig. 6 (A) TEM image of copper rods obtained from 10 mM solution; (B) sche-
matic representation of the rod structure and cross sections of the rods (gray:
contribution to the diffractogram, white: no contribution to the diffractogram);
(C) diffractogram of nanorod lying on a side as well as calculated <100> (solid
lines, bold numbers) and <112> (dashed lines, italic numbers) reciprocal lattices;
(D) diffractogram of nanorod lying on edge as well as calculated <110> (solid
lines, bold numbers) and <111> (dashed lines, italic numbers) reciprocal lattices.
ticles tend to form two- and three-dimensional close-packed 100 nm to 1.3 µm. Usually the short rods have slightly larger
arrangements. The interparticle distance is typical for particles diameters than the longer ones. The aspect ratio is between
stabilised by long alkyl chains. It varies between 1.5 and 3.5 nm, 2 and 32. Regarding the spherical particles, a size distribution
which is shorter than the length of two stretched hexadecyl maximum of 38.3 nm (σ = 13 nm) is found.
chains (2 × 2.1 nm) and indicates interpenetrating organic
A peculiar feature of copper nanorods is their optical pro-
matrices. An influence of the ethylene glycol chains is not visible. perties, as their UV-vis absorption is reported to shift to
Occasionally, simple and multiple twinning can be observed in shorter wavelengths with increasing rod lengths.30 However,
the 1 mM sample by different absorbance of the electron beam we were not able to distinguish the absorption resulting from
on different oriented lattices in a particle (inset Fig. 4B).
A precursor concentration of 2 mM generated an ensemble ones, most likely because of broad length variation of the rods.
of irregular particles. The size distribution is still monomodal,
To gain information about the internal structure of the
the elongated particles from the absorption of the spherical
but broad with a maximum of 85.6 nm (σ = 15 nm). Next to copper rods we performed selected area electron diffraction
spherical particles anisotropic shapes were found, e.g. oval (SAED) on single rods. Investigation of about 25 rods revealed
and triangular plates as well as short rods. The generation of two different diffraction patterns (Fig. 6C and D) in about
anisotropic particles in the isotropic crystal-system of copper equal proportions. The diffractograms are similar to the ones
can be explained as a consequence of symmetry breaking by found by investigation of copper nanorods prepared by growth
twinning, e.g. by the assumption of a twin plane parallel to the in micelles28c as well as to the patterns obtained from silver31
plates faces.27
and gold27a nanorods.
In addition, experiments at a higher precursor concen-
The two obtained diffraction patterns cannot be interpreted
tration of 10 mM were performed. Next to spherical nanoparti- as a single crystallographic zone but are found to be superposi-
cles significantly elongated nanoparticles (“nanorods”) were tions of two zones. In Fig. 6C the square reciprocal lattice
ˉ
observed, both in about equal proportions (Fig. 6A). The for- made up of the reflections of the (200), (020) and (220) planes
mation of copper nanorods is well investigated for template corresponds to the <100> zone. It is overlaid by the rectangular
assisted growth28 and for CVD processes.29 Furthermore, it has reciprocal lattice of the <112> zone consisting of the reflec-
ˉ
been observed for thermal decomposition at high precursor tions of the (111), (311) and (220) planes. Some reflections
concentrations, but no further investigations were carried out (marked by an asterisk) cannot be assigned to a single zone
for the latter.9b
but are double diffractions. The second diffractogram (Fig. 6D)
The rods of our samples have a narrow diameter distri- is made up of the rectangular reciprocal lattice of the <110>
ˉˉˉ
ˉ
ˉ
bution between 40 and 50 nm and a length ranging from zone with the reflections from the (111), (111), (220) and (002)
15604 | Dalton Trans., 2013, 42, 15599–15609
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
planes and the hexagonal reciprocal lattice of the <111> zone Cu[O2CCR2(OC2H4)nOCH3]2 (n = 0–3; R = H, Me) (2a–e) as well
ˉˉ
with (022), (202) and (220) reflections. These superpositions as the corresponding tris- and bis(triphenylphosphine)-
prove that the rods are not formed by a single domain. copper(I) carboxylates 3a–e and 4a–e is described. Investigation
However, they are consistent with a penta-twinned rod struc- of the thermal decomposition by TG-MS experiments showed
ture as shown in Fig. 6B. In this structure, the rods are built that the carboxylates decompose in two steps starting with a
up by five elongated tetrahedra which are radially arranged along decarboxylation in the case of the copper(II) carboxylates 2a–e
a common [110] axis. Neighbouring tetrahedra are connected by and with the loss of the ligand for the triphenylphosphine-
{110} boundaries. This arrangement is possible with only a little copper(I) complexes 3 and 4. It was shown that substitution at
strain, because the interfacial angle between the {111} planes of the carbon in α-position of the carboxylato group results in a
70.53° is close to 360°/5 = 72°. The basic area of the tetrahedra lower decomposition temperature for the copper(II) complexes.
and hence the surface of the rods consist of {100} planes. The In all thermally induced decompositions copper(0) was
rod ends might be made up of {111} planes which results in the formed. The straightforward synthesis, the defined decompo-
overall form of an elongated pentagonal bipyramide (Fig. 6B). sition mechanism and the possibility to influence the
However, rods can also be terminated by {110} planes.31
In the penta-twinned arrangement a rod cannot be oriented precursors for the generation of metallic copper.
in a way that allows the contemporary observation of the Bragg The application of the copper carboxylates was demon-
decomposition qualifies the respective complexes as excellent
reflections of all five domains, but only three of them (Fig. 6B). strated by employing complex 4c as a precursor for the gene-
If a rod lying on its side interacts with an electron beam per- ration of oxide-free copper nanoparticles by thermal
pendicular to this side the domains T1, T3 and T4 contribute decomposition in a long-chained amine, i.e. without the
to the diffractogram. T1 is oriented along <100> and T3 and T4 addition of any further reducing or stabilising reagents. A con-
along <112>. The interaction results in the diffractogram pre- siderable advantage of this single-source precursor approach is
sented in Fig. 5C. If the rod is tilted around its axis by 72°/4 = its reproducibility, since only few reaction parameters (as pre-
18° or if the rod lies on its edge in such a way that the electron cursor concentration and reaction temperature) have to be con-
beam is parallel to its side, domain T5 will cause diffractions trolled. Whereas known precursors such as copper(II)
of the <110> zone and T2 and T3 that of <111>, resulting in acetylacetonate9c,f,i often give rise to air sensitive reaction mix-
the diffractogram shown in Fig. 6D.
tures, complex 4c features a significantly lower oxygen sensibi-
The experimental results may be explained by the following lity, which is due to the presence of its reducing phosphine
approach: as soon as the reaction mixture reaches the necess- ligands, and hence allows to generate oxide free copper nano-
ary temperature, simultaneous decomposition of the precursor particles by the use of standard Schlenk techniques.
molecules leads to a high supersaturation, followed by nuclea-
Systematic variation of the precursor concentration between
tion and 3D particle growth. Twins are formed by stacking 0.5 mM and 2.0 mM allowed one to obtain nanoparticles with
faults or by collision of small metal clusters in this early stage size-distribution maxima between 10 nm and 85 nm and mod-
of growth. Both processes are more likely at higher growth erate size variations. To the best of our knowledge, the possi-
speed which explains the more frequent observation of twins bility to vary the particle size to
a similar extent is
at higher precursor concentration. At a low precursor concen- unprecedented for copper nanoparticle generation by the pre-
tration (c ≤ 2 mM) the growth stays in the 3D regime for cursor approach. Using a precursor concentration of 10 mM,
twinned as well as for untwinned particles, which has recently copper nanorods with aspect ratios of up to 32 have been
been explained under the assumption of diffusion controlled obtained. Electron diffraction studies showed that the rods are
growth.32 A high precursor concentration can lead to 1D not single domain crystals but consist of five domains. Com-
growth of twinned particles, resulting in the formation of pared with other rod preparation methods, the herein
copper rods. However, particles whose symmetry is not broken described single-source precursor approach provides access to
by twinning will grow homogeneously in all three dimensions copper nanorods by a much simpler procedure.
and cannot convert into rods.27a If one-dimensional growth
We are confident that the copper(I) and copper(II) com-
starts, the rods diameter corresponds to that of its spherical plexes can be applied successfully in further applications such
seed particle; that is why the short rods which are formed as the preparation of metal–organic inks for inkjet printing,
towards the end of the reaction time have larger diameters. spray pyrolysis or spin-coating of conductive copper patterns
The generation of monodisperse rods in solution is possible and layers as successfully shown for the respective silver and
by seed mediated growth, as shown for gold.27a However, to gold carboxylates.33
the best of our knowledge, seed mediated growth was not yet
observed for copper.
6 Experimental section
6.1. Instruments and materials
5 Summary and conclusions
The synthesis of the copper(I) and copper(II) complexes were
A straightforward synthetic methodology for the preparation performed under non-inert conditions in undried solvents.
of ethylene glycol functionalised copper(II) carboxylates Nanoparticle synthesis was performed with standard Schlenk
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 15599–15609 | 15605

View Article Online
Paper
Dalton Transactions
techniques under an argon atmosphere with solvents dried by residual solvent was removed in vacuo to give the products as
standard techniques34 and degassed by multiple freeze– green solids or viscous fluids (vide supra) in quantitative
pump–thaw cycles. All chemicals were commercially available yields. Copper(II) 2-methoxyacetate (2a): Mp. 219 °C. Anal.
and were used without further purification. For synthesis and calcd for C6H10O6Cu (241.68): C 29.82, H 4.17; found C 29.87,
spectroscopic characterisation of 2–4c please refer to ref. 16.
H 4.12. FTIR (ATR): ˜ν = 2944 (m), 1602(s), 1579(s), 1435(m),
Inert centrifugation was performed using a Labofuge 200 1393(m), 1340(m), 1244(w), 1115(m), 1072(m), 1072(m),
from Heraeus. Elemental analyses were performed using a 988(w), 966(m), 942(m), 731(w) cm−1. HRMS (ESI): m/z =
Thermo FLASHEA 112 Series instrument. Infrared spectra were 263.9666 (M + Na+). Copper(II) 2-(2-methoxyethoxy)acetate (2b):
recorded using a Biorad FTS-165 or using a Nicolet iS 10 from Anal. calcd for C10H18O8Cu (329.79): C 36.42, H 5.50, found
Thermo Scientific. 1H NMR spectra were recorded using a C 36.20, H 5.82. FTIR (ATR): ˜ν = 2883(m), 1756(w), 1630(s),
Bruker Avance III 500 spectrometer operating at 500.30 MHz in 1415(s), 1574(s), 1415(m), 1432(s), 1329(s), 1260(s), 1199(m),
the Fourier transform mode; 13C{1H} NMR spectra were 1093(s), 1027(m), 947(w), 894(w), 849(m), 753(m), 694(m)
recorded at 125.80 MHz; 31P{1H} NMR spectra at 202.53 MHz. cm−1. HRMS (ESI): m/z = 352.0190 (M + Na+). Copper(II)
Chemical shifts are given relative to the internal standard 2-[2-(methoxyethoxy)ethoxy]-2-methylpropanoate (2d): Anal.
1
tetramethylsilane for H and 13C{1H} NMR spectra and to the calcd for C18H34O10Cu (473.14): C 45.69, H 7.24, found
external standard 85% phosphoric acid for 31P{1H} NMR C 45.98, H 7.27. FTIR (ATR): ˜ν = 2978(m), 2929(m), 2875(m),
spectra. Mass spectra were recorded using a micrOTOF QII 1625(s), 1472(w), 1412(m), 1357(w), 1181(m), 1081(s), 979(w),
Bruker Daltonite workstation. TG, DSC and TG-MS experi- 848(w), 813(w), 779(w) cm−1. HRMS (ESI) m/z = 496.1363
ments were performed using a Mettler Toledo TGA/DSC1 1600 (M + Na+). Copper(II) 2-{-2-[2-(2-methoxyethoxy)ethoxy]ethoxy}-
system with an MX1 balance coupled with a Pfeiffer Vacuum acetate (2e): Anal. calcd for C18H30O12Cu (501.96): C 42.72, H
MS ThermoStar GSD 301 T2 mass spectrometer at a heating 6.77, found C 42.54, H 6.97. FTIR (ATR): ˜ν = 2871(m), 1635(s),
rate of 5 K min−1 under a flowing argon atmosphere (40 mL 1577(s), 1416(m), 1329(m), 1247(s), 1199(w), 1081(s), 934(w),
min−1). The DSC measurements were used to determine the 851(w), 724(w) cm−1. HRMS (ESI) m/z = 528.1238 (M + Na+).
melting point. UV-vis spectra were recorded using a Thermo
Tris(triphenylphosphine)copper(I) carboxylates (3a, b, d, e):
Genesys 6 spectrometer. XRPD was carried out on a STOE-S- Copper(II) carboxylates 2a, b, d, e, 3.5 equiv. of triphenylphos-
TAD IP device using CuKα (λ = 154.184 pm) radiation. TEM phine and 0.5 equiv. of water were dissolved in methanol
imaging was performed using a PHILIPS CM 20 FEG instru- (200 mL for a 20 mmol reaction). The mixture was heated to
ment operated at 200 kV.
reflux until it became colourless. Upon cooling to −20 °C the
respective products precipitated. Recrystallisation from metha-
nol and drying in vacuo gave the title complexes as colourless
6.2. Synthesis
2-[2-(2-Methoxyethoxy)ethoxy]-2-methylpropanoic acid (1d) was solids. Tris(triphenylphosphine)copper(I) 2-methoxyacetate
prepared following a protocol by J. A. Ragan et al.17a Yield: (3a): Yield: 88%. Mp. 132 °C. Anal. calcd for C57H50O3P3Cu
56%. Bp: 185 °C at 2 Pa. Anal. calcd for C9H18O5 (206.23): (939.44): C 72.87, H 5.36, found C 72.65, H 5.42. FTIR (ATR):
C 52.41, H 8.80, found C 51.82, H 8.96. FTIR (ATR): ˜ν = ˜ν = 3053(w), 1614(m), 1597(s), 1480(m), 1434(s), 1373(m),
1
3253(w), 2984(m), 2877(m), 1735(s), 1489(m), 1359(m), 1320(m), 1109(m), 1093(s), 997(m), 738(s), 691(s), 618(w) cm−1. H NMR
1133(s), 1083(s), 977(m), 845(m), 747(m) cm−1
.
1H NMR (CDCl3): δ = 3.26 (s, 3H, OCH3), 3.83 (s, 2H, CH2), 7.26–7.29
(CDCl3): δ = 1.47 (s, 6H, CCH3), 3.39 (s, 3H, OCH3), 3.56–3.60 (m, 18H, CH), 7.34–7.37 (m, 27H, CH) ppm. 13C{1H} NMR
(m, 4 H), 3.67–3.72 (m, 4 H) ppm. 13C{1H} NMR (CDCl3): δ = (CDCl3): δ = 58.7 (s, OCH3), 72.1 (s, CH2), 128.7 (d, mCH, JCP
24.5 (s, CCH3), 59.2 (s, OCH3), 64.7 (s, CH2), 70.1 (s, CH2), 71.0 8.2 Hz), 129.6 (s, CH), 133.9 (d, CH, JCP = 16.2 Hz), 134.1 (d,
=
3
p
o
2
1
(s, CH2), 71.9 (s, CH2), 79.0 (s, CCH2), 176.6 (s, CO2H) ppm.
iCH, JCP = 18.2 Hz), 176.3 (OCO) ppm. 31P{1H} NMR (CDCl3):
2-{2-[2-(2-Methoxyethoxy)ethoxy]ethoxy}acetatic acid (1e) δ = −3.3 ppm. HRMS (ESI) m/z = 739.0648 (M − PPh3 + Cu+).
was prepared following a protocol by W. Kunz et al.17b Yield: Tris(triphenylphosphine)copper(I) 2-(2-methoxyethoxy)acetate
58%. Bp: 198 °C at 2 Pa. Anal. calcd for C9H18O6 (222.28): (3b): Yield: 85%. Mp. 86 °C. Anal. calcd for C59H54O4P3Cu
C 48.63, H 8.16, found C 47.85, H 8.47. FTIR (ATR): ˜ν = 3201(w), (983.49): C 72.05, H 5.53, found C 71.90, H 5.53. FTIR (ATR):
2875(m), 1756(s), 1604(w), 1432(m), 1351(w), 1199(m), 1092(s), ˜ν = 3053(w), 2938(w), 2813(w), 1596(m), 1480(w), 1433(s),
1
849(m), 743(s), 692(s) cm−1. H NMR (CDCl3): δ = 3.37 (s, 3H, 1394(m), 1373(m), 1111(w), 1092(s), 1026(s), 744(s), 693(s) cm−1
.
OCH3), 3.55–3.56 (m, 2H, CH2), 3.62–3.65 (m, 4H, CH2), 1H NMR (CDCl3): δ = 3.24 (s, 3H, OCH3), 3.41–3.42 (m, 2H,
3.67–3.68 (m, 4H, CH2), 3.74–3.76 (m, 2H, CH2), 4.15 (s, 2H, CH2), 3.45–3.46 (m, 2H, CH2), 3.88 (s, 2H, OCOCH2), 7.16–7.19
OCOCH2). 13C{1H} NMR (CDCl3): δ = 58.9 (s, OCH3), 68.8 (s, (m, 18H, CH), 7.23–2.27 (m, 27H, CH) ppm. 13C{1H} NMR
CH2), 70.3 (s, CH2), 70.4 (s, CH2), 70.4 (s, CH2), 70.6 (s, CH2), (CDCl3): δ = 58.0 (s, OCH3), 69.0 (s, CH2), 69.7 (s, CH2), 71.0 (s,
3
p
71.5 (s, CH2), 72.0 (s, OCOCH2), 172.8 (s, CO2H) ppm.
OCOCH2), 127.3 (d, mCH, JCP = 8.3 Hz), 128.6 (s, CH), 132.9
Copper(II) carboxylates (2a, b, d, e): two equivalents of the (d, oCH, 2JCP = 16.3 Hz), 133.2 (d, iCH, 1JCP = 17.6 Hz), 175.5 (s,
carboxylic acid were added to a suspension of copper(II) OCO) ppm. 31P{1H} NMR (CDCl3): δ = −3.3 ppm. HRMS (ESI)
acetate monohydrate in a sufficient volume of toluene. Upon m/z = 783.0910 (M − PPh3 + Cu+). Tris(triphenylphosphine)-
boiling, a mixture of water, acetic acid and toluene was dis- copper(I)
2-[2-(2-methoxyethoxy)ethoxy]-2-methylpropanoate
tilled. When reaching the boiling point of pure toluene, the (3d): Yield: 53%. Mp. 94 °C. Anal. calcd for C63H62O5P3Cu
15606 | Dalton Trans., 2013, 42, 15599–15609
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
(1055.64): C 71.68, H 5.92, found C 71.71, H 5.94. FTIR (ATR): 783.0910 (M + Cu+). Bis(triphenylphosphine)copper(I) 2-[2-(2-
˜ν = 3052(w), 2979(w), 2927(w), 1599(s), 1585(w), 1479(m), methoxyethoxy)ethoxy]-2-methylpropanoate (4d): Yield: 89%.
1433(s), 1380(m), 1347(w), 1091(m), 1026(m), 842(w), 742(s), Mp. 122 °C. Anal. calcd for C45H47O5P2Cu (793.32): C 68.12,
1
692(s), 607(w) cm−1. H NMR (CDCl3): δ = 1.38 (s, 6H, CCH3), H 5.97, found C 68.03, H 5.93. FTIR (ATR) ˜ν = 2866(m), 1583(m),
3.30–3.32 (m, 2H, CH2), 3.24 (s, 3H, OCH3), 3.46–3.48 (m, 2H, 1556(s), 1479(m), 1434(s), 1402(m), 1355(w), 1093(s), 1026(m),
1
CH2), 3.50–3.52 (m, 2H, CH2), 3.54–3.56 (m, 2H, CH2), 847(m), 744(s), 692(s) cm−1. H NMR (CDCl3): δ = 1.37 (s, 6H,
7.22–7.25 (m, 18H, CH), 7.31–7.35 (m, 27H, CH) ppm. 13C{1H} CCH3), 3.31–3.32 (m, 2H, CH2), 3.32 (s, 3H, OCH3), 3.45–3.47
NMR (CDCl3): δ = 25.7 (s, CCH3), 59.1 (s, OCH3), 63.7 (s, CH2), (m, 2H, CH2), 3.50–3.52 (m, 2H, CH2), 3.54–3.55 (m, 2H, CH2),
70.4 (s, CH2), 71.3 (s, CH2), 72.1 (s, CH2), 78.8 (s, CCH3), 128.7 7.19–7.22 (m, 12H, CH), 7.30–7.32 (m, 18H, CH) ppm. 13C{1H}
3
i
1
(d, mCH, JCP = 8.9 Hz), 129.8 (s, pCH), 133.3 (d, CH, JCP
=
NMR (CDCl3): δ = 25.7 (s, CCH3), 59.0 (s, OCH3), 63.7 (s, CH2),
o
2
25.3 Hz), 134.0 (d, CH, JCP = 15.5 Hz), 181.7 (s, OCO) ppm. 70.3 (s, CH2), 71.2 (s, CH2), 72.0 (s, CH2), 78.7 (s, CCH3), 128.6
31P{1H} NMR (CDCl3): δ = −3.3 ppm. HRMS (ESI) m/z = (d, mCH, JCP = 6.8 Hz), 129.9 (s, pCH), 132.7 (d, oCH, JCP
=
3
2
857.1470 (M − PPh3 + Cu+). Tris(triphenylphosphine)copper(I) 31.5 Hz), 133.9 (d, CH, JCP = 13.5 Hz), 181.6 (s, OCO) ppm.
2-{2-[2-(2-methoxyethoxy)ethoxy]ethoxy}acetate (3e): Yield: 31P{1H} NMR (CDCl3): δ = −2.4 ppm. Bis(triphenylphosphine)-
43%. Mp. 52 °C. Anal. calcd for C63H62O5P3Cu (1055.59): copper(I) 2-{2-[2-(2-methoxyethoxy)ethoxy]ethoxy}acetate (4e):
C 70.61, H 5.83, found C 70.19, H 5.85. FTIR (ATR): ˜ν = 3047(w), Yield: 87%. Mp. 110 °C. Anal. calcd for C45H47O6P2Cu (809.32):
2890(w), 1606(s), 1599(s), 1480(m), 1432(s), 1379(m), 1092(s), C 66.78, H 5.85, found C 66.25, H 6.00. FTIR (ATR): ˜ν = 3051
1026(w), 850(w), 743(s), 692(s) cm−1. 1H NMR (CDCl3): δ = 3.28 (w), 2861(w), 1599(m), 1478(s), 1433(s), 1410(w), 1379(w), 1093
i
1
(s, 3H, OCH3), 3.44–3.47 (m, 4H, CH2), 3.51–3.53 (m, 8H, CH2), (s), 1026(m), 997(m), 850(w), 742(s), 692(s) cm−1 1H NMR
.
3.88 (s, 2H, OCOCH2), 7.17–7.20 (m, 18H, CH), 7.23–2.27 (m, (CDCl3): δ = 3.34 (s, 3H, OCH3), 3.50–3.53 (m, 4H, CH2),
27H, CH) ppm. 13C{1H} NMR (CDCl3): δ = 59.1 (s, OCH3), 70.1 3.57–3.61 (m, 8H, CH2), 3.94 (s, 2H, OCOCH2), 7.19–7.22 (m,
(s, CH2), 70.5 (m, CH2), 70.6 (s, CH2), 70.7 (s, CH2), 72.0 (s, 12H, CH), 7.30–7.33 (m, 18H, CH) ppm. 13C{1H} NMR (CDCl3):
3
p
OCOCH2), 128.6 (d, mCH, JCP = 8.4 Hz), 129.5 (s, CH), 133.9 δ = 59.1 (s, OCH3), 70.1 (s, CH2), 70.5 (m, CH2), 70.6 (s, CH2),
(d, oCH, 2JCP = 16.3 Hz), 134.2 (d, iCH, 1JCP = 17.8 Hz), 177.6 (s, 70.7 (s, CH2), 72.0 (s, OCOCHH2), 128.7 (d, mCH, 3JCP = 5.7 Hz),
OCO) ppm. 31P{1H} NMR (CDCl3): δ = −3.3 ppm.
129.9 (s, pCH), 132.6 (d, CH, JCP = 30.8 Hz), 133.8 (d, oCH,
i
1
Bis(triphenylphosphine)copper(I) carboxylates (4a, b, d, e): 2JCP = 12.9 Hz), 176.6 (s, OCO) ppm. 31P{1H} NMR (CDCl3): δ =
A concentrated solution of a tris(triphenylphosphine)copper(I) −2.34 ppm.
carboxylate 3a, b, d, e in dichloromethane was added dropwise
Generation of spherical and rod-like copper nanoparticles:
n
to hexane under vigorous stirring at ambient conditions (for Complex 4c was dissolved in 10 mL of 1-hexadecylamine to
a 20 mmol reaction about 10 mL dichloromethane and obtain the desired concentration (0.5 mM, 1.0 mM, 2.0 mM,
500 mL nhexane). The respective title complexes precipitated 10.0 mM). After preheating the solution to 250 °C it was
as colourless solids which were isolated by filtration, washed further heated to 330 °C at a constant heating rate of 25 K
with nhexane and dried in vacuo. Bis(triphenylphosphine)- min−1 and was kept at this temperature for 10 min. The gene-
copper(I) 2-methoxyacetate (4a): Yield: 98%. Mp. 145 °C. Anal. ration of nanoparticles was observed by the sudden appear-
calcd for C39H35O3P2Cu (677.19): C 69.17, H 5.21, found ance of red colour. After air-cooling the solution to ambient
C 69.25, H 5.26. FTIR (ATR): ˜ν = 3052(w), 2874(w), 2811(w), temperature, the solidified reaction mixture was dissolved
1582(s), 1479(s), 1434(s), 1407(m), 1315(m), 1194(w), 1123(s), in 10 mL of nhexane. The particles were precipitated by
1
1095(s), 1024(m), 996(m), 855(w), 735(s), 692(s) cm−1. H NMR addition of 20 mL ethanol. A sample of 5 mL was subjected to
(CDCl3): δ = 3.19 (s, 3H, OCH3), 3.75 (s, 2H, CH2), 7.16–7.19 centrifugation (30 min, 5300 min−1). The particles were redis-
(m, 12H, CH), 7.26–7.29 (m, 18H, CH) ppm. 13C{1H} NMR persed in 5 mL of ethanol, isolated by centrifugation and
3
(CDCl3): δ = 58.7 (s, OCH3), 72.1 (s, CH2), 128.7 (d, mCH, JCP
=
redispersed in 5 mL of nhexane by the use of ultrasound for
p
o
2
8.3 Hz), 129.6 (s, CH), 133.8 (d, CH, JCP = 16.0 Hz), 134.2 (d, further analysis.
iCH, JCP = 18.6 Hz), 176.3 (s, OCO) ppm. 31P{1H} NMR
For TEM measurement, one drop of the particle dispersions
1
(CDCl3): δ = −2.4 ppm. HRMS (ESI) m/z = 739.0648 (M + Cu+). was placed on a polymer-coated, copper-supported carbon
Bis(triphenylphosphine)copper(I) 2-(2-methoxyethoxy)acetate grid. After removing the drop by the use of filter paper the grid
(4b): Yield: 95%. Mp. 181 °C. Anal. calcd for C41H39O4P2Cu was dried in vacuo.
(721.22): C 68.28, H 5.45, found C 68.20, H 5.52. FTIR (ATR):
˜ν = 3064(w), 2895(w), 1564(s), 1478(m), 1433(s), 1410(m), b, d and e suitable for X-ray diffraction analysis were obtained
1321(m), 1120(s), 1094(s), 997(w), 853(w), 738(s), 691(s) cm−1 by diffusion of nhexane into a concentrated solution of the
Single-crystal X-ray diffraction analysis: Single crystals of 4a,
.
1H NMR (CDCl3): δ = 3.24 (s 3H, OCH3), 3.43–3.47 (m, 4H, respective complexes in dichloromethane. Data were collected
CH2), 3.89 (s, 2H, OCOCH2), 7.15–7.19 (m, 12 H, CH), using an Oxford Gemini S diffractometer using MoKα (λ =
7.25–7.28 (m, 18H, CH) ppm. 13C{1H} NMR (CDCl3): δ = 58.9 (s, 71.073 pm) or CuKα (λ = 154.184 pm) radiation. The structures
OCH3), 69.8 (s, CH2), 70.7 (s, CH2), 71.9 (s, OCOCH2), 128.6 (d, were solved by direct methods and refined by full matrix least
3
1
mCH, JCP = 4.3 Hz), 129.9 (s, pCH), 132.6 (d, iCH, JCP
=
squares procedures on F2.35 All non-hydrogen atoms were
o
2
30.1 Hz), 133.9 (d, CH, JCP = 12.6 Hz), 176.5 (s, OCO) ppm. refined anisotropically and a riding model was employed in
31P{1H} NMR (CDCl3): δ = −2.4 ppm. HRMS (ESI) m/z = the refinement of the hydrogen-atom positions.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 15599–15609 | 15607

View Article Online
Paper
Dalton Transactions
CCDC-932857 (4a), 933501 (4b), 932855 (4d) and 932856 13 M. Steffan, A. Jakob, P. Claus and H. Lang, Catal. Commun.,
(4e) contain the supplementary crystallographic data for this
2009, 10, 437.
paper.
14 (a) A. Tuchscherer, D. Schaarschmidt, S. Schulze,
M. Hietschold and H. Lang, Dalton Trans., 2012, 41, 2738;
(b) A. Tuchscherer, D. Schaarschmidt, S. Schulze,
M. Hietschold and H. Lang, Eur. J. Inorg. Chem., 2011, 28,
4421; (c) A. Tuchscherer, D. Schaarschmidt, S. Schulze,
M. Hietschold and H. Lang, Inorg. Chem. Commun., 2011,
14, 676.
Acknowledgements
We gratefully acknowledge generous financial support from
the
Deutsche
Forschungsmemeinschaft
(FOR
1713
SMINT). M. Korb thanks the Fond der Chemischen Industrie 15 A. Tuchscherer, R. Packheiser, T. Rüffer, H. Schletter,
(FCI) for a doctoral fellowship.
M. Hietschold and H. Lang, Eur. J. Chem., 2012, 13, 2251.
16 D. Adner, M. Korb, S. Schulze, M. Hietschold and H. Lang,
Chem. Commun., 2013, 49, 6855.
17 (a) J. A. Ragan, N. D. Ide, W. Cai, J. J. Cawley, R. Colon-
Cruz, R. Kumar, Z. Peng and B. C. Vanderplas, Org. Process
Res. Dev., 2010, 14, 1402; (b) O. Zech, M. Kellermeier,
S. Thomaier, E. Maurer, R. Klein, C. Schreiner and
W. Kunz, Chem.–Eur. J., 2009, 15, 1341.
Notes and references
1 (a) M. P. Pileni, T. Gulik-Krzywicki, J. Tanori,
A. Filankembo and J. C. Dedieu, Langmuir, 1998, 14, 7359;
(b) A. Filankembo and M. P. Pileni, Appl. Surf. Sci., 2000,
164, 260; (c) S.-H. Wu and D.-H. Chen, J. Colloid Interface 18 (a) G. B. Deacon and P. J. Phillips, Coord. Chem. Rev., 1980,
Sci., 2004, 273, 165.
2 J. H. Kim, T. A. Germer, G. W. Mulholland and
S. H. Ehrman, Adv. Mater., 2002, 14, 518.
3 D. B. Pedersen and S. Wang, J. Phys. Chem., 2007, 111,
17493.
4 G. H. Chan, J. Zhao, E. M. Hicks, G. C. Schatz and R. P. van
Duyne, Nano Lett., 2007, 7, 1947.
33, 227; (b) G. B. Deacon and F. Huber, Inorg. Chim. Acta,
1985, 104, 41.
19 Synthesis of the potassium salts of 1c and 1e: 0.25 mmol of
the acid 1c or 1e and 0.25 mmol KOtBu were stirred in
20 mL of methanol for 30 min. After removal of all volatile
components the residues were dried in vacuo. The products
were obtained as colourless substances.
5 P. X. Rutkowski and J. I. Zink, Inorg. Chem., 2009, 48, 1655.
6 R. A. Salkar, P. Jeevanandam, G. Kataby, S. T. Aruna,
20 J. G. Forrest, C. K. Prout and F. J. C. Rossotti, Chem.
Commun., 1966, 658.
Y. Koltypin, O. Palchik and A. Gedanken, J. Phys. Chem. B, 21 J. Fawcett, S. H. Laurie, C. Simpson, M. C. R. Symons,
2000, 104, 893.
F. A. Taiwo and I. Hawkins, Inorg. Chim. Acta, 2001, 312,
245.
22 W. T. Reichle, Inorg. Chim. Acta, 1971, 5, 325.
7 I. Haas, S. Shanmugam and A. Gedanken, J. Phys. Chem. B,
2006, 110, 16947.
8 (a) H. Hirai, H. Wakabayashi and M. Komiyama, Bull. 23 (a) A. Jakob, Y. Shen, T. Wächtler, S. Schulz, T. Gessner,
Chem. Soc. Jpn., 1986, 59, 367; (b) H. H. Huang, F. Q. Yan,
Y. M. Kek, C. H. Chew, G. Q. Xu, W. Ji, P. S. Oh and
S. H. Tang, Langmuir, 1997, 13, 172.
R. Riedel, C. Fasel and H. Lang, Z. Anorg. Allg. Chem., 2008,
634, 2226; (b) A. Jakob, T. Rüffer, P. Ecorchard, B. Walfort,
K. Körbitz, S. Frühauf, S. E. Schulz, T. Gessner and
H. Lang, Z. Anorg. Allg. Chem., 2010, 636, 1931.
9 (a) N. A. Dhas, C. P. Raj and A. Gedanken, Chem. Mater.,
1998, 10, 1446; (b) J. Hambrock, R. Becker, A. Birkner, 24 C. Salzemann, A. Brioude and M.-P. Pileni, J. Phys. Chem. B,
J. Weiß and R. A. Fischer, Chem. Commun., 2002, 68; 2006, 110, 7208.
(c) S. U. Sun, I. K. Park, J. Park and T. Hyeon, Chem. 25 J. H. Kim, S. H. Ehrman and T. A. Germer, Appl. Phys. Lett.,
Commun., 2004, 778; (d) X. Sun, Y.-W. Zhang, R. Si and 2004, 84, 1278.
C.-H. Yan, Small, 2005, 1, 1081; (e) M. Salavati-Niasari, 26 (a) Q. Zhang, J. Xie, Y. Yu and J. Y. Lee, Nanoscale, 2010, 2,
F. Davar and N. Mir, Polyhedron, 2008, 27, 3514;
(f) G. Cheng and A. R. H. Walker, Anal. Bioanal. Chem.,
1962; (b) C. B. Murray, C. R. Kagan and M. G. Bawendi,
Annu. Rev. Mater. Sci., 2000, 30, 545.
2010, 396, 1057; (g) L.-I. Hung, C.-K. Tsung, W. Huang and 27 (a) C. J. Johnson, E. Dujardin, S. A. Davis, C. J. Murphy and
P. Yang, Adv. Mater., 2010, 22, 1910; (h) M. H. Habibi,
R. Mokhtari, M. Mikhak, M. Amirnasr and A. Amiri, Spec-
S. Mann, J. Mater. Chem., 2002, 12, 1765; (b) C. Lofton and
W. Sigmund, Adv. Funct. Mater., 2005, 15, 1197.
trochim. Acta, Part A, 2011, 79, 1524; (i) M. Shi, H. S. Kwon, 28 (a) M. P. Pileni, T. Gulik-Krzywicki, J. Tanori,
Z. Peng, A. Elder and H. Yang, ACS Nano, 2012, 6, 2157.
10 K. S. Gayen, T. Sengupta, Y. Saima, A. Das, D. K. Maiti and
A. Mitra, Green Chem., 2012, 14, 1589.
11 (a) Z. Lin, D. Han and S. Li, J. Therm. Anal. Calorim., 2012,
107, 47; (b) E. Lamprecht, G. M. Watkins and M. E. Brown,
Thermochim. Acta, 2006, 446, 91.
A. Filankembo and J. C. Dedieu, Langmuir, 1998, 14, 7359;
(b) A. Filankembo and M.-P. Pileni, Appl. Surf. Sci., 2000,
164, 260; (c) I. Lisiecki, A. Filankembo, H. Sack-Kongehl,
K. Weiss, M.-P. Pileni and J. Urban, Phys. Rev. B: Condens.
Matter, 2000, 61, 4968.
29 J.-L. Duan, J. Liu, H.-J. Yao, D. Mo, M.-D. Hou,
Y.-M. Sun, Y.-F. Chen and L. Zhang, Mater. Sci. Eng., B,
2008, 147, 57.
12 B. Hammond, F. H. Jardine and A. G. Vohra, J. Inorg. Nucl.
Chem., 1971, 33, 1017.
15608 | Dalton Trans., 2013, 42, 15599–15609
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
30 (a) A. Azarian, A. I. Zad, A. Dolati and M. Ghorbani, J. Phys.:
Condens. Mater, 2007, 19, 446007; (b) X. Zhang, D. Zhang,
X. Ni and H. Zheng, Solid State Commun., 2006, 139, 412.
Mater., 2010, 22, 3067; C. Schoner, A. Tuchscherer,
T. Blaudeck, S. F. Jahn, R. R. Baumann and H. Lang, Thin
Solid Films, 2013, 531, 147.
31 S.-H. Zhang, Z.-Y. Jiang, Z.-X. Xie, X. Xu, R.-B. Huang and 34 Author collective, Organikum, Wiley-VCH, Weinheim, 22nd
L.-S. Zheng, J. Phys. Chem. B, 2005, 109, 9416. edn, 2004.
32 Z. A. Peng and X. Peng, J. Am. Chem. Soc., 2001, 123, 35 (a) G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crys-
1389.
tallogr., 1990, 46, 467; (b) G. M. Sheldrick, SHELXL-97,
Program for Crystal Structure Refinement, Universität Göttin-
gen, Germany, 1997.
33 S. F. Jahn, T. Blaudeck, R. R. Baumann, A. Jakob,
P. Ecorchard, T. Rüffer, H. Lang and P. Schmidt, Chem.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 15599–15609 | 15609