A straightforward approach to oxide-free copper nanoparticles by thermal decomposition of a copper(i) precursor

Article abstract of DOI:10.1039/c3cc42914h

The synthesis of the novel copper(i) precursor [Cu(PPh₃) ₂(O₂CCH₂OC₂H₄OC ₂H₄OCH₃)] and its application in the straightforward solution synthesis of oxide-free copper nanoparticles by mere thermal decomposition are reported; depending on the precursor concentration particles of sizes of 10 nm or 30 nm are obtained in narrow size distributions.

Full text of DOI:10.1039/c3cc42914h

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A straightforward approach to oxide-free copper
nanoparticles by thermal decomposition of a copper(I)
precursor†
David Adner,a Marcus Korb,a Steffen Schulze,b Michael Hietscholdb and
Heinrich Lang*a
Cite this: Chem. Commun., 2013,
49, 6855
Received 19th April 2013,
Accepted 10th June 2013
DOI: 10.1039/c3cc42914h
www.rsc.org/chemcomm
The synthesis of the novel copper(I) precursor [Cu(PPh3)2(O2CCH2- 2-[2-(2-methoxyethoxy)ethoxy]acetate (3) as a new precursor for
OC2H4OC2H4OCH3)] and its application in the straightforward copper nanoparticle formation. Due to the presence of two
solution synthesis of oxide-free copper nanoparticles by mere phosphine ligands with reducing properties this metal–organic
thermal decomposition are reported; depending on the precursor complex features a significantly lower oxidation sensitivity of the
concentration particles of sizes of 10 nm or 30 nm are obtained in particle generation process and facilitates the preparation of
narrow size distributions.
oxide-free copper nanoparticles by the use of standard Schlenk
techniques. Furthermore, to the best of our knowledge, it is the
Among various chemical1 and physical2 methods for the prepara- first copper(I) compound successfully applied in the solution
tion of copper nanoparticles, the precursor approach3 has received synthesis of copper nanoparticles next to the highly air and
continuous attention during the last few years. It can be under- moisture sensitive copper(I) acetate.3h,i
stood as a variation of the traditional chemical approach in which
Metal–organic complex 3 was prepared from commercially
all necessary reactants (i.e. copper sources, stabilisers and reducing available starting materials using copper(II) 2-[2-(2-methoxy-
agents) are combined into one molecule. The particle generation is ethoxy)ethoxy]acetate, which itself was synthesised by an anion
performed in solution by thermal decomposition of the precursor. exchange reaction of the acid CH3O(C2H4O)2CH2CO2H (1) with
The convenient processing and the high reproducibility of the first copper(II) acetate hydrate (Scheme 1, for experimental details see
order reaction make the precursor approach an interesting alter- ESI†). Acetic acid and water were removed by azeotropic distilla-
native, e.g. for industrial applications.
tion with toluene as an entrainer. Copper(II) carboxylate was
To date, the main drawback of the precursor approach has reduced by triphenylphosphine which was added in an excess,
been considered to be the high oxygen sensitivity of the particle sufficient to stabilise the resulting copper(I) carboxylate, yielding
preparation process, which arises from the absence of an excess of tris(triphenylphosphine) copper(I) 2-[2-(2-methoxyethoxy)ethoxy]-
reducing equivalents. Hence, often only oxidised copper nano- acetate (2).6 One of the three phosphine ligands in 2 could be
particles could be isolated.3c,d,f,h In the past, the problem had to removed by two-solvent recrystallisation, i.e. by dissolving 2 in
be addressed by using advanced working procedures, such as the dichloromethane and adding nhexane. This methodology allowed
use of dry boxes3b,k or reducing solvents and atmospheres.3a,e,g,j
the isolation of 3 as a colourless air stable solid with an overall
Completing our work on the use of gold4 and silver5 ethylene yield of 73% based on copper(II) acetate hydrate. The complex was
glycol carboxylates as precursor compounds for metal nano- characterised using elemental analysis, IR and NMR (1H, 13C{1H})
particle generation, we herein report the synthesis and applica- spectroscopy, mass spectrometry and single-crystal X-ray structure
tion of the air-stable complex bis(triphenylphosphine)copper(I) determination (see ESI†).
%
Complex 3 crystallises in the triclinic space group P1 without
a
any solvent molecules (Fig. 1).7 The carboxylato group is
monodentate-bonded to copper(I), resulting in the formulation
of a 16-valence-electron complex. The copper(I) ion is located
almost exactly in the plane defined by the three surrounding
atoms P1, P2 and O1, protruding only 0.1135 Å. Regarding
the ethylene glycol chains, the lowest energy conformation
(gauche C–C bonds and anti C–O bonds)8 is found to be one
of the two conformations, which are about equally populated.
The thermal behaviour of 3 was studied by thermogravimetric
(TG) and differential scanning calorimetric (DSC) investigations
¨
Technische Universitat Chemnitz, Faculty of Natural Sciences, Institute of
Chemistry, Inorganic Chemistry, D-09107 Chemnitz, Germany.
E-mail: heinrich.lang@chemie.tu-chemnitz.de; Fax: +49 (0)371 531 21219;
Tel: +49 (0)371 531 21210
b
¨
Technische Universitat Chemnitz, Faculty of Natural Sciences, Institute of Physics,
Solid Surfaces Analysis, D-09107 Chemnitz, Germany.
E-mail: hietschold@physik.tu-chemnitz.de; Fax: +49 (0)371 531 21639;
Tel: +49 (0)371 531 21630
† Electronic supplementary information (ESI) available: Experimental details, IR
and NMR spectra, additional TEM images, X-ray and electron diffractograms.
CCDC 932859. For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c3cc42914h
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 6855--6857 6855

View Article Online
Communication
ChemComm
Scheme 1 Synthesis of 3: (i) 2CH3O(C2H4O)2CH2CO2H, toluene, 111 1C, 5 h, azeotropic distillation; (ii): 3.5PPh3, 0.5H2O, MeOH, 65 1C, 3 h; (iii): two-solvent
recrystallisation from dichloromethane–nhexane.
Furthermore, presuming the amine coordinates to the metal, the
long alkyl chain should be able to avoid oxidation more effectively
than a short one, e.g. that of 1-octylamine.
A typical nanoparticle preparation protocol proceeded as
follows: a solution of 3 in 1-hexadecylamine (0.5 mM or 1.0 mM)
was prepared and preheated to 250 1C, followed by heating it to
reflux at a constant heating rate of 25 1C minÀ1. While keeping the
temperature at 330 1C for 10 min the formation of copper
nanoparticles was observed by the sudden reddening of the reac-
Fig. 1 ORTEP diagram (50% probability level) of 3. All disordered atoms and all
tion mixture. After cooling it to ambient temperature the solidified
hydrogen atoms are omitted for clarity. Selected bond distances [Å] and angles
[1]: Cu1–P1 2.2118(8), Cu1–P2 2.2547(7), Cu1–O1 2.063(2), Cu1–O2 2.539(3),
C1–O1 1.275(4), C1–O2 1.230(3), P1–Cu–P2 130.17(3), O1–C1–O2 124.7(3),
P1–Cu1–O1 109.31(6), P2–Cu1–O1 119.69(6).
n
mixture was dissolved in hexane. The copper nanoparticles were
precipitated by addition of ethanol, separated by centrifugation,
washed once with ethanol and dried or redispersed in nhexane for
analysis (see ESI† for details). In contrast to the nanoparticle
dispersions the dried particles are stable ad infinitum under inert
conditions, whereas they slowly form copper oxides when stored in
air for weeks. By gravimetric measurements and assuming the
organic shell to make up only a minor weight fraction we found
virtually quantitative conversion of the precursor to copper(0).
The obtained nanoparticles were investigated using UV-vis,
XRPD and electron transmission microscopy (TEM). The UV-vis
spectra of the copper nanoparticles in nhexane showed the expected
plasmon absorption which is due to the metallic properties of the
particles.9 It shifts from 570 cmÀ1 to 575 cmÀ1 for a precursor
concentration rising from 0.5 mM to 1.0 mM, indicating the
formation of larger nanoparticles at higher concentrations (Fig. 3).
The XRPD pattern of the nanoparticles (Fig. 3) shows the
reflections of the (111), (200) and (220) planes of the face-centered
cubic cell of copper (ICDD 00-004-0836). No reflections of copper(I)
oxide or copper(II) oxide could be detected, meaning that the
amount of crystalline impurities is below the detection limit. The
peak broadening, which is characteristic of nanoparticles, can be
used to estimate the crystallite size using the Scherrer equation.10
By interpretation of the (111) peak, the size was calculated to be
to gain information about the decomposition behaviour of the
complex (Fig. 2). It was found that the decomposition appears in
two steps, of which the first resembled a slow mass decay
starting at ca. 200 1C. In TG-MS coupling experiments performed
for the complex bis(triphenylphosphine)copper(I) 2-(2-methoxy-
ethoxy)acetate, the typical fragments of triphenylphosphine
could be detected, which indicates that the decomposition of
the complex starts with the loss of the phosphine ligands. In a
second decomposition step at ca. 300 1C decarboxylation occurs,
leading to the formation of copper(0). The process is overlaid
with the decomposition of the ethylene glycol chain. The resi-
dual mass of 8.6% fits well to the formation of elemental copper
(8.3% theoretical), whose presence was proven by X-ray powder
diffraction (XRPD) measurements (see ESI†). The DSC measure-
ment shows a melting point of 140 1C and portrays the thermal
decomposition as a single endothermic process.
The decomposition temperature of 3 requires the application of
high-boiling solvents for the nanoparticle formation (vide supra).
We choose 1-hexadecylamine (mp 40 1C, bp 330 1C) because it is
available in higher purity than the common oleylamine.
Fig. 3 UV-vis spectra of copper nanoparticles dispersed in nhexane (left); the
XRPD pattern of copper nanoparticles (1.0 mM sample) with reflections of copper
(ICDD 00-04-0836), copper(I) oxide (ICDD 01-071-3645) and copper(II) oxide
(ICDD 01-073-6023) for comparison (right).
Fig. 2 TG and DSC traces of
3
(gas flow N2 60 mL minÀ1 (TG) and N2
20 mL minÀ1 (DSC), heating rate 10 K minÀ1).
c
6856 Chem. Commun., 2013, 49, 6855--6857
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
the oxidation sensitivity of the reaction mixtures due to the
presence of its reducing triphenylphosphine ligands and allows
the use of standard Schlenk techniques rather than that of dry
boxes. Nanoparticles of different sizes, 10 nm and 30 nm, were
obtained in narrow size distributions by changing the precur-
sor concentration from 0.1 mM to 1.0 mM. We are confident
that 3, the respective tris(triphenylphosphine) copper(I) complex 2,
as well as the copper(II) carboxylate 1 can be applied successfully in
further applications such as the preparation of metal–organic inks
for inkjet printing, spray pyrolysis or spin-coating of conductive
copper patterns and layers as successfully shown for the respective
silver and gold carboxylates.12
We thank Prof. Dr Michael Mehring and Dr Maik Schlesinger
for performing the XRPD measurements and we gratefully
acknowledge generous financial support from the Deutsche
Forschungsgemeinschaft (FOR1713 SMINT) and the Fonds der
Chemischen Industrie (FCI). M. Korb thanks the FCI for a
doctoral fellowship.
Fig. 4 TEM images and size distribution of copper nanoparticles obtained
at different precursor concentrations. (A) c = 0.5 mM, d = 9.8 nm, s = 1.7 nm;
(B) c = 1.0 mM, d = 28.2 nm, s = 2.6 nm, inset: multiply twinned particles.
Notes and references
1 (a) H. Hirai, H. Wakabayashi and M. Komiyama, Bull. Chem. Soc.
Jpn., 1986, 59, 367; (b) R. A. Salkar, P. Jeevanandam, G. Kataby,
S. T. Aruna, Y. Koltypin, O. Palchik and A. Gedanken, J. Phys.
Chem. B, 2000, 104, 893; (c) J. H. Kim, T. A. Germer,
G. W. Mulholland and S. H. Ehrman, Adv. Mater., 2002, 14, 518;
(d) S.-H. Wu and D.-H. Chen, J. Colloid Interface Sci., 2004, 273, 165;
(e) P. X. Rutkowski and J. I. Zink, Inorg. Chem., 2009, 48, 1655.
2 (a) D. B. Pedersen and S. Wang, J. Phys. Chem. C, 2007, 111, 17493;
(b) G. H. Chan, J. Zhao, E. M. Hicks, G. C. Schatz and R. P. van
Duyne, Nano Lett., 2007, 7, 1947.
31 Æ 2 nm, which fits well to the size distributions obtained using
electron microscopy (vide infra).
The copper nanoparticle size and size distribution were
determined using TEM. At least 200 particles were counted
for each sample. For the mathematical interpretation the size
distributions were fitted by Gaussian functions.
3 (a) N. A. Dhas, C. P. Raj and A. Gedanken, Chem. Mater., 1998,
10, 1446; (b) J. Hambrock, R. Becker, A. Birkner, J. Weiß and
R. A. Fischer, Chem. Commun., 2002, 68; (c) S. U. Son, I. K. Park,
J. Park and T. Hyeon, Chem. Commun., 2004, 778; (d) X. Sun,
Y.-W. Zang, R. Si and C.-H. Yan, Small, 2005, 11, 1081; (e) M. Salavati-
Niasari, F. Davar and N. Mir, Polyhedron, 2008, 27, 3514; ( f ) M. Salavati-
Niasari and F. Davar, Mater. Lett., 2009, 63, 441; (g) M. Salavati-Niasari,
Z. Fereshteh and F. Davar, Polyhedron, 2009, 28, 126; (h) G. Cheng and
A. R. H. Walker, Anal. Bioanal. Chem., 2010, 396, 1057; (i) L.-I. Hung,
C.-K. Tsung, W. Huang and P. Yang, Adv. Mater., 2010, 22, 1910;
( j) M. H. Habibi, R. Mokhtari, M. Amirnasr and A. Amiri, Spectrochim.
Acta, Part A, 2011, 79, 1524; (k) M. Shi, H. S. Kwon, Z. Peng, A. Elder and
H. Yang, ACS Nano, 2012, 6, 2157.
4 (a) A. Tuchscherer, D. Schaarschmidt, S. Schulze, M. Hietschold and
H. Lang, Inorg. Chem. Commun., 2011, 14, 676; (b) A. Tuchscherer,
D. Schaarschmidt, S. Schulze, M. Hietschold and H. Lang, Dalton
Trans., 2012, 41, 2738.
5 M. Steffan, A. Jakob, P. Claus and H. Lang, Catal. Commun., 2009, 10, 437.
6 B. Hammond, F. H. Jardine and A. G. Vohra, J. Inorg. Nucl. Chem.,
1971, 33, 1017.
For the 0.5 mM sample spherical particles with a mean
diameter of d = 9.8 nm and a standard derivation of s = 1.7 nm
(size variation cv = 17%) were found (Fig. 4A). The 1.0 mM sample
yielded larger particles with d = 28.2 nm in a narrower size
distribution (s = 2.6 nm, cv = 9%, Fig. 4B). Rounding these results
to a reasonable accuracy we regard the particles to have sizes of
10 and 30 nm. The particles in both probes tend to form two- and
three-dimensional close-packed arrangements with short inter-
particle distances of about 2 nm which are likely due to the
presence of the long-chained amine (vide supra).3b In the 1 mM
sample, simple and multiple twinning can be observed in about
5% of the particles (see the inset in Fig. 4B), which is a known
feature of copper nanoparticles with a sufficient size.11 Electron
diffraction studies did not show any mark of the formation of
either copper(I) or copper(II) oxides, even though the probe pre-
paration was performed under non-inert conditions (see ESI†).
In terms of their microscopic images the particles prepared in
this work are similar to the particles obtained in other precursor
approach studies.3b,c The size variations are slightly larger than in
comparable systems. However, the accessibility of a range of particle
sizes by changing the copper precursor concentration seems to be
unique to our system and will be part of future investigations.
In conclusion, the synthesis of the new copper(I) precursor
[Cu(PPh3)2(O2CCH2OC2H4OC2H4OCH3)] (3) is discussed.
This metal–organic complex allows the easy and straightfor-
ward solution synthesis of oxide-free copper nanoparticles by
mere thermal decomposition, i.e. without the addition of any
reducing agents. Compared with other copper precursors such
as copper(II) carboxylates,3d,e,h complex 3 significantly reduces
7 Crystal data for [Cu(PPh3)2(OOCCH2OC2H4OC2H4OCH3)] (3):
C43H43CuO5P2, Mr = 765.25 g molÀ1, crystal dimensions 0.20 Â 0.10 Â
%
0.05 mm, T = 110 K, l = 154.184 pm, triclinic, P1, a = 10.7844(5) Å, b =
13.4639(6) Å, c = 14.1917(6) Å, a = 71.163(4)1, b = 77.783(4)1, g = 84.067(4)1,
V = 1904.74(15) Å3, Z = 2, rcalcd = 1.344 g cmÀ3, m = 1.958 mmÀ1
,
y range = 3.35–65.481, reflections collected: 12 298, independent: 6090
(Rint = 0.0380), R1 = 0.0390, wR2 = 0.0805 [I > 2s(I)]. CDCC 932859†.
8 P. Johansson, J. Grondin and J.-C. Lassegues, J. Phys. Chem. A, 2010,
`
114, 10700.
9 J. A. Creighton and D. G. Eadon, J. Chem. Soc., Faraday Trans., 1991,
87, 3881.
¨
10 (a) A. L. Patterson, Phys. Rev., 1939, 56, 978; (b) A. Kochendorfer,
Z. Kristallogr. Kristallgeom. Kristallphys. Kristallchem., 1944, 105, 393.
11 S. I. Cha, C. B. Mo, K. T. Kim, Y. J. Jeong and S. H. Hong, J. Mater.
Res., 2006, 21, 2371.
12 S. F. Jahn, T. Blaudeck, R. R. Baumann, A. Jakob, P. Ecorchard,
T. Ru¨ffer, H. Lang and P. Schmidt, Chem. 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.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 6855--6857 6857