Insights into the thermal decomposition of Co(II) oleate for the shape-controlled synthesis of wurtzite-type CoO nanocrystals

Article abstract of DOI:10.1021/cm4041055

Fatty acid salts of transition metals are known to undergo thermal decomposition in high-boiling organic solvents. Although it is a straightforward, promising approach for generating colloidal metal oxide nanocrystals in high yield, its widespread implementation is hindered by irreproducibility. Subtle structural variations and impurities are often introduced during preparation of the carboxylate precursors, which exhibit strong influence on thermal decomposition, and the resulting nucleation and growth of oxide nanocrystals. Here, we studied the colloidal synthesis of wurtzite-type CoO (wz-CoO) nanostructures via thermal decomposition of Co(II) oleate complex [Co(OL)]₂. Using Fourier transform infrared spectroscopy, we discovered that the conventional method for preparing [Co(OL)]₂ gives rise to an isolable impurity containing a free hydroxide moiety. Furthermore, [Co(OL)]₂ did not thermally decompose within the expected temperature range when the impurity was removed. In contrast, pencil-shaped wz-CoO nanorods were synthesized when no measures were taken to remove the impurity, suggesting that the hydroxide functionality may facilitate the thermal decomposition. These insights enabled us to prepare size-and shape-controlled wz-CoO nanocrystals using purified [Co(OL)] ₂ solutions, by incorporating additives that mimic the action of the hydroxide impurity. We also demonstrate a simplified pathway to wz-CoO nanocrystals that does not require chemical additives for thermolysis.

Full text of DOI:10.1021/cm4041055

Article
pubs.acs.org/cm
Insights into the Thermal Decomposition of Co(II) Oleate for the
Shape-Controlled Synthesis of Wurtzite-Type CoO Nanocrystals
Matthew R. Buck, Adam J. Biacchi, and Raymond E. Schaak*
Department of Chemistry and Materials Research Institute, The Pennsylvania State University, University Park, Pennsylvania 16802,
United States
ABSTRACT: Fatty acid salts of transition metals are known
to undergo thermal decomposition in high-boiling organic
solvents. Although it is a straightforward, promising approach
for generating colloidal metal oxide nanocrystals in high yield,
its widespread implementation is hindered by irreproducibility.
Subtle structural variations and impurities are often introduced
during preparation of the carboxylate precursors, which exhibit
strong influence on thermal decomposition, and the resulting
nucleation and growth of oxide nanocrystals. Here, we studied
the colloidal synthesis of wurtzite-type CoO (wz-CoO)
nanostructures via thermal decomposition of Co(II) oleate
complex [Co(OL)]2. Using Fourier transform infrared spectroscopy, we discovered that the conventional method for preparing
[Co(OL)]2 gives rise to an isolable impurity containing a free hydroxide moiety. Furthermore, [Co(OL)]2 did not thermally
decompose within the expected temperature range when the impurity was removed. In contrast, pencil-shaped wz-CoO nanorods
were synthesized when no measures were taken to remove the impurity, suggesting that the hydroxide functionality may facilitate
the thermal decomposition. These insights enabled us to prepare size-and shape-controlled wz-CoO nanocrystals using purified
[Co(OL)]2 solutions, by incorporating additives that mimic the action of the hydroxide impurity. We also demonstrate a
simplified pathway to wz-CoO nanocrystals that does not require chemical additives for thermolysis.
subsequent reports indicated that body-centered cubic Fe and
wustite-type FeO were the dominant products of this
reaction.22−25 The thermolysis of Fe(III) oleate solutions was
later studied using mass spectrometry, and many volatile
byproducts, including H2, CO, CO2, and various hydrocarbons,
were detected during decomposition.26 The reducing power of
dissolved H2 or CO was likely responsible for the generation of
Fe and lower-valent iron oxides. This series of reports
demonstrates the unpredictable influence that side products
can have on the formation of nanocrystals. Although the
conditions that may have regulated the generation of such side
products include the solvent and the heating rate, the
underlying cause for the discrepancy remains unclear.
The temperature at which thermal decomposition occurs is
also known to be sensitive to subtle, structural variability in the
metal−carboxylate complexes. For example, the presence of
coordinated solvent or water molecules influences the
coordination geometry of the carboxylate ligand and the
stability of the complex. Bronstein and co-workers have shown
that Fe(III) oleate exhibits decomposition temperatures
ranging from 300 to 380 °C, and these decomposition
temperatures depend on whether the complex is purified by
washing with water, methanol, or acetone.25 They reported a
range of decomposition temperatures that also depended on
INTRODUCTION
■
The thermal decomposition of metal carboxylates in high-
boiling organic solvents represents an increasingly common
pathway for synthesizing high-quality colloidal metal oxide
nanocrystals in high yield. Gram-scale quantities of uniformly
sized Fe3O4 nanocrystals, for example, can be produced in a
facile, one-step reaction by thermolysis of Fe(III) cis-9-
octadecanoate (Fe-oleate) in noncoordinating solvents.1
Metal−carboxylate decomposition has also been applied
successfully to the synthesis of Mn(II), Fe(II), Fe(III), Co(II),
Co(II,III), and Zn(II) oxide nanocrystals.2−7 Several fatty acid
salts can be decomposed together in the same solution to yield
mixed-metal oxide nanocrystals such as CoFe2O4, MnFe2O4,
and NiFe2O4.8,9 In several cases, the reaction conditions,
including the incorporation of surface stabilizers and other
chemical additives, have enabled researchers to rigorously
control the size, shape, and uniformity of the nanocrystals.10,11
This is important, because nanostructured transition metal
oxides, including colloidal nanocrystals, have size- and shape-
dependent magnetic, optical, and electrical properties that are
promising for applications in biology,12−14 catalysis,15−18 and
energy.19−21
Despite its widespread use, the thermal decomposition of
metal−carboxylate solutions is often sensitive to subtle
differences in experimental conditions, which can have
deleterious effects on reproducibility and utility. Although
scalable quantities of monodisperse, inverse-spinel Fe3O4
nanocrystals were prepared using Fe(III) oleate,1 for example,
Received: December 13, 2013
Revised: January 14, 2014
Published: January 17, 2014
© 2014 American Chemical Society
1492
dx.doi.org/10.1021/cm4041055 | Chem. Mater. 2014, 26, 1492−1499

Chemistry of Materials
Article
how the complex was dried. When decomposition occurs at
higher temperatures, side reactions may also play a dominant
role in the processes that impact nanocrystal growth, such as
reduction, oxidation, and etching.
and can be reproducibly implemented, and this may also be
applicable to other related metal oxide nanocrystal precursors.
EXPERIMENTAL SECTION
■
Peng and co-workers described the synthesis of several 3d
transition metal oxide nanocrystals and identified a significant
difference in reactivity between carboxylate salts of different
metals.2 Stearate and oleate salts of Mn(II), Fe(II), Fe(III), and
Zn(II) decomposed readily in noncoordinating solvents at
300−320 °C. Fatty carboxylate salts of Cr(III), Co(II), and
Ni(II), however, were sluggish to react at similar temperatures
in the same solvents. Metal oxide nanocrystals were produced
from the latter precursors only when the decomposition
pathway was changed by the addition of “activating” agents.
Nucleophilic amines and alcohols, for example, destabilize the
metal−carboxylate complexes by attacking the ligand carbonyl
group and eliminate either amides or esters.2,4,6 In contrast to
this report, it was later found that solutions of Co(II) oleate
[Co(OL)2] straightforwardly decomposed into colloidal,
pencil-shaped, CoO nanocrystals that had a metastable,
wurtzite-type crystal structure [wz-CoO].5 This called into
question the “inert” behavior of those [Co(OL)2] solutions and
the need for activating additives.
Nanostructured cobalt oxides are promising materials for
applications in energy storage,19 sensing,27−30 catalysis,31−33
and magnetism.34 Cobalt(II) oxide typically crystallizes in the
cubic, NaCl-type structure, but hexagonal wz-CoO has been
synthesized by several groups.5,35−38 Interest in wz-CoO stems
from the discovery of room-temperature ferromagnetism in
dilute, solid solutions of wz-CoO in a wz-ZnO host lattice.39
Studies that have highlighted the magnetic and optical
properties of nanostructured wz-CoO have followed.5,36,40,41
Recently, wz-CoO nanorods were demonstrated as a suitable
host for Cu2+ dopants, which remarkably increased the stability
of the (metastable) wz-CoO phase.42 Preliminary studies
showed enhanced electrochemical capacitance and cycling
performance for Cu2+-doped wz-CoO, which is promising for
lithium-ion battery applications.42
In this work, we have investigated the colloidal synthesis of
wz-CoO nanostructures via thermal decomposition of the
[Co(OL)2] complex. Importantly, we have discovered an
isolable impurity that is inherent to the conventional synthesis
of [Co(OL)2]. Furthermore, we have discovered that [Co-
(OL)2] does not thermally decompose within the expected
temperature range, when it is purified of the impurity. In
contrast, we have synthesized pencil-shaped wz-CoO nanorods
when no measures were taken to remove the impurity.
Comparative analysis of purified [Co(OL)2] and the separated
impurity, using Fourier transform infrared (FTIR) spectrosco-
py, suggests that the impurity contains a free hydroxide
functionality, which may be acting as an activating agent. We
have successfully synthesized size- and shape-controlled wz-
CoO nanocrystals, using purified [Co(OL)2] solutions, by
incorporating additives that mimic the action of the hydroxide
impurity.
By employing a two-stage heating process under controlled
atmosphere, we have also determined that the distillation of
volatile side products is a critical step in the thermal
decomposition of purified [Co(OL)2]. This has enabled us to
prepare colloidal wz-CoO nanorods from pure [Co(OL)2]
solutions, without the need for any chemical additives. These
insights provide a greater understanding of how the thermal
decomposition of an important metal carboxylate salt occurs
Materials. All chemicals were used as received. Cobalt(II) chloride
hexahydrate (CoCl2·6H2O, 98%), 1-octadecene (ODE, tech. 90%),
oleic acid (OLAC, tech. 90%), and tetramethylammonium hydroxide
pentahydrate (TMAOH, ≥97%) were purchased from Sigma-Aldrich.
Sodium oleate [Na(OL)] was purchased from TCI America. All
syntheses were carried out under Ar using standard Schlenk
techniques, and work-ups were performed in air.
Preparation of Crude Co(OL)2. CoCl2·6H2O (4.76 g, 20 mmol)
and Na(OL) (12.18 g, 40 mmol) were loaded into a one-neck, 250 mL
round-bottom flask, equipped with a reflux condenser and magnetic
stir bar. At room temperature, 15 mL of ethanol, 20 mL of distilled
deionized water, and 35 mL of hexanes were added to the mixture of
powders. The reaction vessel was sealed under a blanket of Ar, placed
in a water bath preheated to 70 °C, and allowed to react for 4 h while
stirring vigorously. The cooled product mixture was transferred to a
separatory funnel, the purple-colored organic layer was washed three
times with distilled, deionized water, and the aqueous phase was
discarded. The washing procedure resulted in a cloudy, purple, organic
phase. The solvent was evaporated at room temperature under
vacuum, yielding Co(OL)2 as a waxy, dark-purple solid, which was
immediately transferred to a refrigerated (8−12 °C) Ar-filled glovebox
for storage.
Purification of Crude Co(OL)2. In certain iterations of our
experiments, the cloudy, purple, organic phase was further purified by
centrifugation. An off-white colored solid was separated from a clear,
purple solution upon centrifugation. The solvent was evaporated at
room temperature under vacuum, yielding Co(OL)2 as a waxy, dark-
purple solid, which was immediately transferred to an Ar-filled
glovebox for storage (also refrigerated, 8−12 °C). The off-white
colored solid was also dried under vacuum and stored in an inert
atmosphere for analysis.
Synthesis of Pencil-Shaped wz-CoO Nanocrystals. Crude
Co(OL)2 (250 mg, 0.4 mmol) was added to 6 mL of degassed ODE in
a 50 mL round-bottom flask, equipped with a reflux condenser and
thermometer adapter. Under an Ar blanket, the mixture was heated
from room temperature to reflux (∼310 °C) at approximately 2 °C/
min. Between 100 and 250 °C, the increasing temperature caused a
gradual color change from purple to blue to dark blue. Beginning at
roughly 200 °C, significant amounts of a white-gray vapor were
evolved from the blue solution, which subsided at about 250 °C. At the
reflux temperature, the color gradually changed from dark blue to dark
green over the course of about 15 min, indicating the nucleation of wz-
CoO nanocrystals. The reaction was cooled after 30 min of heating,
and 2-propanol was added to precipitate the nanocrystals.
Centrifugation (7000 rpm) was applied to collect the green-colored
precipitate, and several redispersion/centrifugation cycles in hexanes/
ethanol were used to purify the product.
Synthesis of Cone- and Golf Tee-Shaped wz-CoO Nano-
crystals. For cone-shaped nanocrystals, 250 mg of purified Co(OL)2
(0.4 mmol) and 9−18 mg (0.05−0.10 mmol) of TMAOH were added
to 6 mL of degassed ODE in a 50 mL round-bottom flask, equipped
with a reflux condenser and thermometer adapter. The heating process
and isolation procedure are identical to those used to synthesize
pencil-shaped wz-CoO. For golf tee-shaped nanocrystals, we replaced
the TMAOH with 30−60 mg (0.10−0.20 mmol) of Na(OL) and kept
other variables the same. In cases where Na(OL) was added, we
noticed that the color change from dark blue to green occurred more
rapidly, in 3 to 5 min. We also found it useful to perform at least two
redispersion/centrifugation cycles in chloroform/acetone to purify the
nanocrystals, when Na(OL) was used.
Synthesis of Bullet-Shaped wz-CoO Nanocrystals. Purified
Co(OL)2 (250 mg, 0.4 mmol) was added to 6 mL of a degassed ODE
solution in a 50 mL round-bottom flask, equipped with a reflux
condenser, thermometer adapter, and flow-control adapter. The
mixture was heated to 240 °C under flowing Ar, and the temperature
1493
dx.doi.org/10.1021/cm4041055 | Chem. Mater. 2014, 26, 1492−1499

Chemistry of Materials
Article
was maintained there for 2 h. During this time, clear, colorless distillate
could be collected in an external vial. After 2 h, an Ar blanket was
applied, and the dark blue solution was heated at a rate of 2 °C/min to
the reflux temperature. The reaction was allowed to proceed for an
additional 20 min, during which time the color changed from dark blue
to dark green. The isolation procedure was identical to that used for
pencil-shaped wz-CoO nanocrystals.
Characterization. Powder X-ray diffraction (XRD) data were
collected using a Bruker D8 Advance X-ray diffractometer equipped
with Cu Kα radiation. Transmission electron microscopy (TEM)
images and selected area electron diffraction (SAED) patterns were
obtained from a JEOL 1200 EX II operating at 80 kV. High resolution
TEM (HRTEM) images were obtained from a JEOL 2010F field
emission microscope operating at 200 kV. Samples were prepared by
suspending the washed nanocrystals in hexanes or toluene and drop-
casting onto Formvar-coated copper TEM grids. Fourier transform
infrared (FTIR) spectra were collected using a Bruker Vertex V70
Spectrometer (Bruker Optics, Billerica MA) equipped with an MVP-
Pro diamond single reflection ATR accessory (Harrick Scientific,
Pleasantville NY). The 400 scans at 6 cm−1 resolution were averaged
for each sample using a DTGS detector and scan frequency of 5 kHz.
In all cases, the spectrum of the clean diamond crystal was used as the
reference spectrum. All spectral manipulations were performed using
OPUS 5.5 (Bruker Optics, Billerica, MA).
extracted into the organic phase, and NaCl remains dissolved in
the aqueous phase. Following the established procedure5 for
synthesizing pencil-shaped wz-CoO nanorods that was used
above, we reacted CoCl2·6H2O and Na(OL) in a mixture of
hexanes, ethanol, and water at 70 °C. After the product was
cooled, a purple-colored organic layer separated from a clear,
colorless aqueous layer. The colorless liquid was discarded, and
the purple organic layer was washed several times with water in
a separatory funnel. Although the organic and aqueous phases
clearly separated after each washing, we noticed that this
process caused the organic layer to become noticeably cloudy.
At first, we thought that the cloudiness was due to slow or
incomplete phase separation, likely caused by a Co(OL)2-
stabilized, water-in-oil emulsion. After several hours, however, a
light-colored precipitate settled at the bottom of the organic
phase, leaving behind a clear, purple solution. In subsequent
iterations of the synthesis of Co(OL)2, the precipitated
“impurity” was removed from the purple Co(OL)2 solution
using centrifugation. After the solution was decanted and
hexanes evaporated, the product was vacuum-dried overnight,
which yielded Co(OL)2 as the expected waxy, purple solid.
Although the precipitated impurity appeared an off-white color
after centrifugation, a tan-colored, waxy solid remained after
drying.
RESULTS AND DISCUSSION
■
Synthesis of wz-CoO Nanocrystals. Thermolysis of
“crude” Co(OL)2 in refluxing ODE resulted in the formation
of pencil-shaped nanorods of wz-CoO, as evidenced by the
powder XRD pattern and representative TEM image in Figure
1. However, this result, which is consistent with a previous
The FTIR spectra of Co(OL)2 and the impurity are
presented in Figure 2c,d, respectively, and the spectra for free
Figure 2. FTIR absorbance spectra of (a) OLAC, (b) Na(OL), (c)
purified Co(OL)2, and (d) an impurity that was isolated by
centrifugation during preparation of Co(OL)2.
Figure 1. Main panel: Experimental powder XRD pattern for colloidal
wz-CoO nanocrystals, along with a simulated pattern for reference.
Inset: Representative TEM image of the nanocrystal product, which
consists primarily of uniform nanorods with pencil-point tips. These
results were obtained by thermally decomposing “crude” Co(OL)2.
OLAC and Na(OL) (Figure 2a,b) are included as a basis for
comparison. All four spectra exhibit similar absorbance features
in the 2800−3000 cm−1 range, characteristic of CH3 and CH2
stretching bands, including the CH stretch at 3025 cm−1,
which are due to the cis-9-octadecenoic (oleic) chains present
in each molecule. In the 1400−1750 cm−1 range, the FTIR
spectrum of OLAC is dominated by a strong peak at 1710
cm−1, which is characteristic of CO stretching. Instead of a
strong CO stretch, Na(OL) exhibits peaks at 1420 and 1560
cm−1, which correspond to the symmetric [νsym(COO−)] and
asymmetric [νasym(COO−)] stretching modes of carboxylate
ion, respectively.43 Importantly, these features of the Na(OL)
spectrum are representative of metal−carboxylate bonding that
is ionic. Peaks in the Na(OL) spectrum at 1447 and 1465 cm−1
are consistent with asymmetric CH3 bending and symmetric
CH2 bending.
report,5 was achievable only very sporadically, even when using
the same Co(OL)2 precursor sample from experiment to
experiment and under identical reaction conditions [250 mg of
Co(OL)2 in 6 mL of ODE, ramp rate 2 °C/min]. This caused
us to examine, in more detail, the formation of the Co(OL)2
precursor and the pathway by which it thermally decomposes
to form wz-CoO nanocrystals.
Synthesis and Characterization of Metal Oleate
Complexes. The conventional preparation of transition
metal oleates involves the reaction of Na(OL) with the
appropriate metal chloride in a biphasic (aqueous/organic)
solvent mixture. During the reaction, which is usually carried
out at reflux (60−70 °C), metal−oleate complex is rapidly
Carboxylate ligands can coordinate to ionic metal centers in a
variety of ways, which include unidentate, chelating, and
1494
dx.doi.org/10.1021/cm4041055 | Chem. Mater. 2014, 26, 1492−1499

Chemistry of Materials
Article
bridging modes. Each mode of coordination causes variance in
the equivalence of the two oxygen atoms, which is observable in
FTIR spectra. It is possible to infer the coordination mode of a
carboxylate complex by comparing the frequency difference
observed in FTIR spectra, sometimes, albeit infrequently,
decomposed into wz-CoO nanocrystals. This led us to
hypothesize that the function of Co(OL)2−x(OH)x could be
mimickedand the thermal decomposition of Co(OL)2 to
form wz-CoO nanocrystals could be made more reprodu-
cibleby including controlled amounts of other hydroxide-
containing additives along with purified Co(OL)2. Indeed,
when judicious quantities of TMAOH were added to the
reaction mixture, we consistently obtained wz-CoO nanocrytals
(Figure 3) via thermolysis of purified Co(OL)2 that did not
(Δ) between νsym(COO−) and νasym(COO−) to the case of
43,44
ionic bonding, in this case ΔNa(OL)
.
The FTIR spectrum of
Co(OL)2 (Figure 2c) is very similar to that of Na(OL),
however, exhibiting strong signals for νsym(COO−) at 1415
cm−1 and νasym(COO−) at 1562 cm−1. These similarities
suggest that metal−ligand bonding in Co(OL)2 is largely ionic
in character, which is consistent with conclusions drawn by
other researchers for the chemically related, cobalt(II) stearate
complex.45 Unlike Na(OL), Co(OL)2 absorbs weakly at 1715
cm−1, which suggests the presence of either free or weakly
bound OLAC. It is not uncommon for the synthesis of metal
oleates to result in acidic salts, unless measures are taken to
remove the free OLAC, such as Sohxlet extraction or washing
with organic solvents.25,46
The FTIR spectrum of the Co(OL)2 impurity (Figure 2d) is
nearly identical to that of Co(OL)2, with two important
differences. First, a sharp doublet centered at ∼3560 cm−1 is
observed that is not present in the spectrum of Co(OL)2, and
this is characteristic of “free” (not hydrogen-bonded) O−H
stretching found in metal-hydroxide complexes.43,47,48 Second,
the relative intensities of the peaks in the 2800−3000 cm−1
range, when compared to the intensities of signals in the 1400−
1750 cm−1 range, are diminished significantly. When Figures
2c,d are compared, this second observation suggests that there
is simply less of the cis-9-octadecenoic moiety in the impurity
than in Co(OL)2. The emergence of hydroxide groups and the
decreased presence of oleyl chains imply strongly that the
impurity is structurally related to Co(II) mono-oleate
hydroxide complex [Co(OL)2−x(OH)x] (referred to as “Co-
(OL)2−x(OH)x impurity” hereafter). Mono-oleate hydroxide
complexes have been observed in the preparation of other
transition metal oleates (MnII, FeIII) using the described
approach and could be derived from the in situ hydrolysis of
metal−oleate bonds during synthesis.7,25
Thermal Decomposition of Co(OL)2 in 1-Octadecene.
We found that the thermal decomposition of Co(OL)2, which
was reported to occur in refluxing 1-octadecene (ODE) solvent
(∼320 °C),5 depended critically upon whether or not the
putative Co(OL)2−x(OH)x impurity was removed during
preparation of Co(OL)2. In cases where the Co(OL)2−x(OH)x
impurity was removed via centrifugation, thermolysis of
Co(OL)2 did not take place in refluxing ODE. Rather, the
solution remained a stable blue color, even after 90 min of
reflux. Between 90 and 120 min at this temperature, partial
decomposition of Co(OL)2 was evidenced by a slow color
change from blue to blue/gray-brown, which resulted in an
insoluble, tan precipitate upon cooling. In contrast, when no
measures were taken to remove the Co(OL)2−x(OH)x impurity
or to purify Co(OL)2 beyond washing with water, we sometimes
observed thermal decomposition of Co(OL)2 into a green-
colored colloidal dispersion. In the absence of the Co-
(OL)2−x(OH)x impurity, Co(OL)2 was sluggish to react at
elevated temperatures, suggesting that the impurity was
necessary for Co(OL)2 thermolysis. Decomposition to wz-
CoO was not consistently reproducible, implying that the
Co(OL)2−x(OH)x impurity may be distributed inhomoge-
neously within the crude Co(OL)2 precursor.
Figure 3. (a) Low-magnification and (b) higher-magnification TEM
images of stout, cone-shaped nanocrystals that were formed by
thermolysis of Co(OL)2 in the presence of TMAOH. The nanocones
appeared to assemble into pairs, with the bases adjacent to each other.
(c) SAED pattern of the nanocrystals in (a) and (b), showing
diffraction rings that can be assigned to wz-CoO. (d) Representation
of the wz-CoO crystal structure highlighting the nonequivalent (001)
and (001) planes and the polar attractions that may have caused the
̅
nanocones to assemble pairwise. Oxygen atoms are colored red and
cobalt atoms are colored blue.
contain Co(OL)2−x(OH)x. Figure 3a−c shows TEM images
and a SAED pattern of wz-CoO nanocrytals that have a stout,
conical shape, where the average distance from the apex of the
particle to the base is 23 2 nm and the basal facet measures
29 2 nm. Notably, the particles tended to assemble into pairs
with the bases in contact with one another, which has been
observed previously for pyramidal- and conically shaped wz-
ZnO nanocrystals.6 This may be due to polar attractions that
arise from the nonequivalence of (001) and (001) planes in
̅
wurtzite-structured crystals, which are Co2+ and O2− termi-
nated, respectively (Figure 3d). These results were highly
reproducible when either a 2:1 or 4:1 Co(OL)2:TMAOH molar
ratio were used. When larger quantities of TMAOH were
Co(OL)2 precursor that had not been purified by
centrifugation, wherein the Co(OL)2−x(OH)x impurity was
1495
dx.doi.org/10.1021/cm4041055 | Chem. Mater. 2014, 26, 1492−1499

Chemistry of Materials
Article
added, insoluble precipitates often resulted, and these were
identified as primarily NaCl-type CoO by XRD.
When we attempted to modify the nanocrystal shape by
injecting a substoichiometric volume of OLACa frequently
used surface stabilizing agent that can lead to the preferential
exposure of different crystal facetsafter 20 min at reflux, the
dark-green, colloidal solution changed (back) to a vivid blue
color in ∼1 min, indicating the wz-CoO nanocrystals are
unstable relative to dissolution and that they dissolve to reform
Co(OL)2. Peng and co-workers reported previously that wz-
ZnO nanocrystals also readily dissolve in OLAC.6 This
observation may be critical to understanding the solution-
mediated thermolysis of Co(OL)2, although it makes the design
of a reproducible procedure for generating nanocrystalline wz-
CoO more complex. For example, the FTIR spectrum of
Co(OL)2 in Figure 2c shows a significant CO signal at 1710
cm−1, implying the presence of free or weakly bound OLAC,
which may hinder the formation of wz-CoO. It is possible that
the hydroxide-containing additives, either Co(OL)2−x(OH)x or
TMAOH, play a role by deprotonating and neutralizing excess
OLAC. This pathway may be complicated by the water of
hydration that is associated with purple-colored Co(OL)2,
because amphoteric water can also react with either OLAC,
TMAOH, or Co(OL)2−x(OH)x. The preceding observation-
s(a) that wz-CoO dissolves in the presence of (the weak
acid) OLAC to form Co(OL)2 and (b) that strongly basic,
hydroxide-containing additives are necessary to induce Co-
(OL)2 decomposition into wz-CoOsuggest that complex
acid−base equilibria are involved in the thermolysis of
Co(OL)2. Moreover, it is reasonable to assume that the
formation of wz-CoO is favored in a basic environment. We
therefore continued studying the influence of basic additives to
the reaction mixture.
Influence of Basic Additives on the Shape of wz-CoO
Nanocrystals. Adding controlled amounts of sodium oleate
[Na(OL)], which contains the conjugate base of OLAC,
promoted decomposition of Co(OL)2 in ODE. A powder XRD
pattern of the green-colored colloidal product (Figure 4a)
shows broad reflections, with peak positions and relative
intensities that are consistent with wz-CoO. TEM images
revealed the formation of nanocrystals that consist of high
aspect ratio nanorod domains protruding from a conically
shaped end, similar to a golf tee (Figure 4b). The similarity of
the conical ends to the stout, wz-CoO nanocones shown in
Figure 3 suggests that the conical portion forms at early stages
of nanocrystal growth and that the rod-shaped domains grow
from the apex of each cone. When a 4:1 molar ratio of
Co(OL)2:Na(OL) was used, the average distance from the
basal facet to the tip of the golf tee shaped nanostructures was
35 4 nm, the conical ends measured 18 2 nm in diameter,
and the nanorod domains had diameters of 7.9 0.7 nm. Using
a Co(OL)2:Na(OL) ratio of 2, the nanorod portion of the
particles was significantly longer, resulting in golf tee
nanostructures with an average length of 62 8 nm (Figure
4c). The differences in the diameters of the conical and
nanorod regions between the two samples were negligible.
Interestingly, the color change from blue to green occurred
gradually, over the course of ∼15 min, when a Co(OL)2:Na-
(OL) ratio of 4 was used. When Na(OL) was present in higher
concentration [Co(OL)2:Na(OL) = 2], the color change
happened much faster, in 3−5 min. It seems likely that the
addition of Na(OL) plays a key role in the formation of reactive
monomers that ultimately nucleate wz-CoO and that the
Figure 4. (a) Powder XRD patterns (experimental and simulated) and
(b) SAED pattern for the wz-CoO nanocrystals shown in (c). (c)
TEM image of the nanocrystals, which have rod-shaped domains
protruding from a conical base. The nanocrystals in (c) were formed
by thermolysis of Co(OL)2 in the presence of Na(OL), at a
Co(OL)2:Na(OL) molar ratio of 4. When Co(OL)2:Na(OL) of 2
was used, the rod-shaped domains were longer, which can be seen in
the TEM image in (d). The inset in (d) shows a region where some
wz-CoO nanocrystals did not contain a rod-shaped domain and others
had multiple rods protruding from a single cone.
resulting monomer production is proportional to the amount of
Na(OL) present. Thus, relatively high concentrations of
Na(OL) allow the rod portions of the wz-CoO nanostructures
to grow longer. In those cases where higher concentrations of
Na(OL) were used, however, a significant fraction of the wz-
CoO nanocrystals appeared only as short cones (or triangles),
without a rod-shaped domain (Figure 4d, inset). Others in the
same sample had multiple rods protruding from a single cone
(Figure 4d, inset). The mechanistic role of Na(OL) in the
generation of Co−O monomers, which presumably occurs by a
reaction with Co(OL)2, is currently under investigation. A fatty
acid−conjugate base pair can react with one another via ketonic
decarbonylation, to form a symmetrically substituted ketone
and CO2.49 It is possible that Na(OL) and excess OLAC react
and are consumed in this way, promoting the thermolysis of
Co(OL)2 by preventing the dissolution of wz-CoO by OLAC.
Additive-Free Synthesis of wz-CoO from Purified
Co(OL)2. Thermal decomposition of purified Co(OL)2 into
wz-CoO was also achievable without any additives, when a two-
stage heating process was used. Heating Co(OL)2 dissolved in
ODE at a constant rate (2 °C/min) to reflux in Ar atmosphere,
as described earlier, did not reproducibly result in thermal
decomposition. Under these conditions, however, we con-
1496
dx.doi.org/10.1021/cm4041055 | Chem. Mater. 2014, 26, 1492−1499

Chemistry of Materials
Article
sistently observed significant quantities of white vapor being
liberated from the solution (filling the vessel headspace)
between 210 and 250 °C. This observation gave us the idea that
an equilibrium concentration of those volatiles may be
dissolved in the reaction mixture and that a component of
those dissolved gases may be preventing thermolysis of
Co(OL)2.
We therefore designed a two-stage heating profile, wherein
flowing Ar was used in the temperature range 60−240 °C. The
temperature was held at 240 °C under Ar flow for 2 h, with the
intention of removing those volatiles from the reaction mixture
by distillation. Indeed, we collected 0.5−1.0 mL of clear,
colorless liquid distillate. The temperature of the reaction was
then increased to reflux (310 °C) at a controlled rate, and the
blue solution gradually changed to dark green, indicating
thermal decomposition of Co(OL)2. Figure 5 shows TEM
For transition metal carboxylates that contain coordinated
water, such as the tetrahydrates of Co(II) acetate [Co-
(CH3COO)2·4H2O] and Ni(II) acetate [Ni(CH3COO)2·
4H2O], a key step in solid-state thermal decomposition is
dehydration.50−52 [Co(CH3COO)2·4H2O] and [Ni-
(CH3COO)2·4H2O] are isostructural, having two axial,
unidentate acetate ligands and four coordinated equatorial
water molecules.53,54 [Co(CH3COO)2·4H2O] is pink and turns
blue upon loss of its water, like many hydrated Co(II)
complexes. Co(OL)2 is purple, notably, and represents a case of
partial hydration.46 Interestingly, the FTIR spectra (Figure 6a)
of [Co(CH3COO)2·4H2O] and Co(OL)2 are similar with
respect to νsym(COO−), νasym(COO−), and Δ [ΔCo(OL)2
=
(1567 cm−1−1415 cm−1) = 152 cm−1, ΔCo(OAc) = (1554 cm−1−
1392 cm−1) = 162 cm−1], suggesting that unidentate oleate
ligand may be found in Co(OL)2. Thermal displacement of
water from either [Co(CH3COO)2·4H2O] or [Ni(CH3COO)2·
4H2O] is known to cause concomitant evolution of acetic acid
by hydrolysis of surface acetate groups.51,52 At elevated
temperatures, this process gives rise to dehydrated, basic
metal acetate, M(CH3COO)2−x(OH)x (M = Co, Ni) which has
been identified as the key intermediate that decomposes into
oxides. During decomposition of M(CH3COO)2−x(OH)x,
acetone was also a major product, presumably formed via
ketonic decarboxylation.
Assuming that the coordination sphere of Co(OL)2
resembles that of [Co(CH3COO)2·4H2O], we thought that
Co(OL)2 must also be dehydrated for thermolysis to occur. At
210 °C, the vapors that evolve from Co(OL)2 in ODE may be
related to thermal displacement of water molecules and/or side
products of that process. We collected a substantial volume of
distillate by flowing Ar over our reaction at 240 °C, but we
could not unambiguously identify any new species in the liquid
using FTIR. The spectrum in Figure 6b is identical to that of
the ODE solvent, with the exception of peaks at 1740, 1130,
and 1075 cm−1. The signal at 1740 cm−1 corresponds well to
the CO stretch in esters, and the three bands in the 1000−
1300 cm−1 range could be characteristic of ester CO
stretches.43 Further investigation is necessary to propose a
pathway that results in ester formation.
We also heated Co(OL)2 in the solid state to 240 °C,
maintained that temperature for two hours, and analyzed the
before-and-after FTIR spectra (Figure 6c). Disappearance of
the CO peak at 1710 cm−1 and emergence of a new peak at
1725 cm−1 indicate the consumption of free OLAC and the
formation of ketones. This result suggests that Co(OL)2, like
many other metal carboxylates in the solid state, may
decompose by pathways that involve ketonic decarboxylation.
The thermal decomposition of Co(OL)2 in solution likely
follows a different pathway, one that is influenced by the
presence of solvent, volatile and soluble intermediates, the
practical handling of atmosphere, and complex acid−base
equilibria.
Figure 5. (a) Large-area and (b) higher-magnification TEM images of
wz-CoO nanocrystals with a bullet-like shape, along with (c) the
corresponding SAED pattern. (d) Representative HRTEM image,
highlighting the (002) and (110) planes of wz-CoO.
images of the product, which consists of nanorods with average
dimensions of 49 7 nm in length and 6.9 0.7 in width.
Many have a bullet-like shape, consisting of two distinct
domains that have slightly different diameters (the average
diameter was calculated from measurements that were made at
the length-bisector of each nanorod) and a pointed tip. The
SAED pattern shown in Figure 5c is consistent with the crystal
structure of wz-CoO. The HRTEM image in Figure 5d shows
that the nanorods are single crystalline, and the observed lattice
planes are consistent with the (002) and (110) planes of wz-
CoO.
CONCLUSIONS
■
The solution-phase, thermal decomposition of transition metal
carboxylates is a promising chemical transformation for
producing scalable quantities of high-quality metal oxide
nanocrystals. Unfortunately, subtle experimental details can
profoundly impact the outcome of such reactions,55 including
the choice of solvent, the presence of impurities, precursor
preparation, and atmosphere handling. Moreover, metal
carboxylate thermolysis is likely propagated by the formation
1497
dx.doi.org/10.1021/cm4041055 | Chem. Mater. 2014, 26, 1492−1499

Chemistry of Materials
Article
Figure 6. FTIR absorbance spectra (a) [Co(CH3COO)2·4H2O] and Co(OL)2, highlighting the νsym(COO−) and νasym(COO−) stretches in each
compound. (b) A distillate (top), collected from an ODE solution of Co(OL)2 solution over a heating period of 2 h at 240 °C. The FTIR spectrum
of ODE solvent (bottom) is included as a reference. Peaks marked with an asterisk (*), in the distillate spectrum, do not appear in the spectrum for
ODE. (c) Co(OL)2, before (bottom) and after (top) heating in the solid-state for 2 h at 240 °C. The spectra highlight the disappearance of a peak at
1705 cm−1, and the appearance of a new peak at 1725 cm−1. The spectral changes suggest that free carboxylic acid is consumed and a ketone is
generated, as a result of heating.
of reactive radical species, which makes the pathways by which
decomposition occurs difficult to analyze. Reports on this
subject have been contradictory, and synthetic protocols are
often irreproducible.
ACKNOWLEDGMENTS
■
This work was supported by the U.S. National Science
Foundation under Grants CHE-1213475 (M.R.B.) and DMR-
1305564 (A.J.B., R.E.S.). TEM imaging was performed in the
Penn State Microscopy and Cytometry Facility (University
Park, PA), and HRTEM imaging and FTIR spectra were
acquired at the Materials Characterization Laboratory of the
Penn State Materials Research Institute.
We were motivated in particular by the case of Co(OL)2
thermolysis into wz-CoO nanocrystals, which feature Co2+ in a
metastable, tetrahedral geometry. First of all, the conventional,
biphasic synthesis of Co(OL)2 produces a side product, which
we have proposed to be Co(OL)2−x(OH)x. We have also found
that colloidal wz-CoO nanocrystals dissolve quickly in the
presence of the weak acid, OLAC. The tendency of Co(OL)2 to
decompose into wz-CoO in the presence of the basic
Co(OL)2−x(OH)x impurity but not from purified Co(OL)2
precursoralong with the tendency of protogenic OLAC to
impede the thermolysisled us to hypothesize that acid−base
interactions may play a key role in the reaction. Indeed, basic
additives such as TMAOH and Na(OL) allowed us to
reproducibly synthesize wz-CoO nanocrystals with interesting
shapes, including nanoscale analogues of cones and golf tees.
Simplified Co(OL)2 thermolysis reactions, without any
additives, could also generate uniform, elongated wz-CoO
nanocrystals. We achieved this result by designing a two-stage
heating process, whereby volatile byproducts were removed via
distillation, prior to reaching the decomposition temperature.
The molecular processes that lead to nucleation of colloidal
inorganic nanostructures are often complex. Reproducible
synthetic routes to large-scale quantities of morphology-
controlled nanocrystals are highly desirable, however, and
efforts to understand the mechanisms are ongoing. The thermal
decomposition of metal carboxylate solutions is an especially
simple case of transformation, requiring only one reactant,
solvent, and heat, and is therefore a worthwhile process to
study in depth. Our studies of Co(OL)2 decomposition have
allowed us to gain several insights which may be extendable to
other carboxylate precursors that yield important nanocrystal
targets, including inorganic solids that feature metal ions in
metastable geometries,56−60 as does wz-CoO.
REFERENCES
■
(1) Park, J.; An, K.; Hwang, Y.; Park, J.-G.; Noh, H.-J.; Kim, J.-Y.;
Park, J.-H.; Hwang, N.-M.; Hyeon, T. Nat. Mater. 2004, 3, 891.
(2) Jana, N. R.; Chen, Y.; Peng, X. Chem. Mater. 2004, 16, 3931.
(3) Choi, S.-H.; Kim, E.-G.; Park, J.; An, K.; Lee, N.; Kim, S. C.;
Hyeon, T. J. Phys. Chem. B 2005, 109, 14792.
(4) Chen, Y.; Johnson, E.; Peng, X. J. Am. Chem. Soc. 2007, 129,
10937.
(5) An, K.; Lee, N.; Park, J.; Kim, S. C.; Hwang, Y.; Park, J.-G.; Kim,
J.-Y.; Park, J.-H.; Han, M. J.; Yu, J.; Hyeon, T. J. Am. Chem. Soc. 2006,
128, 9753.
(6) Chen, Y.; Kim, M.; Lian, G.; Johnson, M. B.; Peng, X. J. Am.
Chem. Soc. 2005, 127, 13331.
(7) Puglisi, A.; Mondini, S.; Cenedese, S.; Ferretti, A. M.; Santo, N.;
Ponti, A. Chem. Mater. 2010, 22, 2805.
(8) Bao, N.; Shen, L.; Wang, Y.; Padhan, P.; Gupta, A. J. Am. Chem.
Soc. 2007, 129, 12374.
(9) Bao, N.; Shen, L.; An, W.; Padhan, P.; Turner, H.; Gupta, A .
Chem. Mater. 2009, 21, 3458.
(10) Park, J.; Lee, E.; Hwnag, N.-M.; Kang, M.; Kim, S. C.; Hwang,
Y.; Park, J.-G.; Noh, H.-J.; Kim, J.-Y.; Park, J.-H.; Hyeon, T. Angew.
Chem., Int. Ed. 2005, 44, 2872.
(11) Kovalenko, M. V.; Bodnarchuk, M. I.; Lechner, R. T.; Hesser,
G.; Schaffler, F.; Heiss, W. J. Am. Chem. Soc. 2007, 129, 6352.
(12) Lee, J. H.; Huh, Y. M.; Jun, Y.; Seo, J.; Jang, J.; Song, H. T.; Kim,
S.; Cho, E. J.; Yoon, H. G.; Suh, J. S.; Cheon, J. Nat. Med. 2007, 13, 95.
(13) Ho, D.; Sun, X. L.; Sun, S. Acc. Chem. Res. 2011, 44, 875.
(14) Xie, J.; Liu, G.; Eden, H. S.; Ai, H.; Chen, X. Y. Acc. Chem. Res.
2011, 44, 883.
(15) Han, X. G.; Jin, M. S.; Xie, S. F.; Kuang, Q.; Jiang, Z. Y.; Jiang, Y.
Q.; Xie, Z. X.; Zheng, L. S. Angew. Chem., Int. Ed. 2009, 48, 9180.
(16) Yang, H. G.; Sun, C. H.; Qiao, S. Z.; Zou, J.; Liu, G.; Smith, S.
C.; Cheng, H. M.; Lu, G. Q. Nature 2008, 453, 638.
(17) Yang, H. G.; Liu, G.; Qiao, S. Z.; Sun, C. H.; Jin, Y. G.; Smith, S.
C.; Zou, J.; Cheng, H. M.; Lu, G. Q. J. Am. Chem. Soc. 2009, 131, 4078.
(18) Huang, W. C.; Lyu, L. M.; Yang, Y. C.; Huang, M. H. J. Am.
Chem. Soc. 2012, 134, 1261.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: schaak@chem.psu.edu.
Notes
(19) Poizot, P.; Laruelle, S.; Grugeon, S.; Dupont, L.; Tarascon, J.-M.
The authors declare no competing financial interest.
Nature 2000, 407, 496.
1498
dx.doi.org/10.1021/cm4041055 | Chem. Mater. 2014, 26, 1492−1499

Chemistry of Materials
Article
(20) O’Regan, B.; Gratzel, M. Nature 1991, 353, 737.
(58) Singh, A.; Geaney, H.; Laffir, F.; Ryan, K. M. J. Am. Chem. Soc.
2012, 134, 2910.
(59) Soriano, R. B.; Malliakas, C. D.; Wu, J.; Kanatzidis, M. G. J. Am.
Chem. Soc. 2012, 134, 3228.
(60) Soriano, R. B.; Arachchige, I. U.; Malliakas, C. D.; Wu, J.;
Kanatzidis, M. G. J. Am. Chem. Soc. 2013, 135, 738.
̈
(21) Chen, X.; Mao, S. S. Chem. Rev. 2007, 107, 2891.
(22) Bodnarchuk, M. I.; Kovalenko, M. V.; Groiss, H.; Resel, R.;
Reissner, M.; Hesser, G.; Lechner, R. T.; Steiner, W.; Schaffler, F.;
Heiss, W. Small 2009, 5, 2247.
(23) Shavel, A.; Rodriguez-Gonzales, B.; Spasova, M.; Farle, M.; Liz-
Marzan, L. M. Adv. Funct. Mater. 2007, 17, 3870.
(24) Pinchon, B. P.; Gerber, O.; Lefevre, C.; Florea, I.; Fleutot, S.;
Baaziz, W.; Pauly, M.; Ohlmann, M.; Ulhaq, C.; Ersen, O.; et al. Chem.
Mater. 2011, 23, 2886.
(25) Bronstein, L. M.; Huang, X.; Retrum, J.; Schmucker, A.; Pink,
M.; Stein, B. D.; Dragnea, B. Chem. Mater. 2007, 19, 3624.
(26) Kwon, S. G.; Piao, Y.; Park, J.; Angappane, S.; Jo, Y.; Hwang, N.-
M.; Park, J.-G.; Hyeon, T. J. Am. Chem. Soc. 2007, 129, 12571.
(27) Cao, A.-M.; Hu, J.-S.; Liang, H.-P.; Song, W.-G.; Wan, L.-J.; He,
X.-L.; Gao, X.-G.; Xia, S.-H. J. Phys. Chem. B 2006, 110, 15858.
(28) Choia, K.; Kima, H. R.; Kim, K. M.; Liu, D.; Cao, G. Z.; Lee, J.
H. Sens. Actuators, B 2010, 146, 183.
(29) Sun, C.; Su, X. T.; Xiao, F.; Niu, C.; Wang, J. Sens. Actuators, B
2011, 157, 681.
(30) Jiao, Q.; Fu, M.; You, C.; Zhao, Y.; Li, H. Inorg. Chem. 2012, 51,
11513.
(31) Hu, L.; Peng, Q.; Li, Y. J. Am. Chem. Soc. 2008, 130, 16136.
(32) Koza, J. A.; He, Z.; Miller, A. S.; Switzer, J. A. Chem. Mater.
2012, 24, 3567.
(33) Xie, X.; Li, Y.; Liu, Z.-Q.; Haruta, M.; Shen, W. Nature 2009,
458, 746.
(34) Skumryev, V.; Stoyanov, S.; Zhang, Y.; Hadjipanayis, G.; Givord,
D.; Nogues, J. Nature 2003, 423, 850.
(35) Redman, M. J.; Steward, E. G. Nature 1962, 193, 867.
(36) Risbud, A. S.; Snedeker, L. P.; Elcombe, M. M.; Cheetham, A.
K.; Seshadri, R. Chem. Mater. 2005, 17, 834.
(37) Grimes, R. W.; Lagerlof, K. P. D. J. Am. Ceram. Soc. 1991, 74,
270.
(38) Nam, K. M.; Shim, J. H.; Han, D.-W.; Kwon, H. S.; Kang, Y.-M.;
Li, Y.; Song, H.; Seo, W. S.; Park, J. T. Chem. Mater. 2010, 22, 4446.
(39) Ueda, K.; Tabata, H.; Kawai, T. Appl. Phys. Lett. 2001, 79, 988.
(40) Seo, W. S.; Shim, J. H.; Oh, S. J.; Lee, E. K.; Hur, N. H.; Park, J.
T. J. Am. Chem. Soc. 2005, 127, 6188.
(41) White, M. A.; Ochsenbein, S. T.; Gamelin, D. R. Chem. Mater.
2008, 20, 7107.
(42) Nam, K. M.; Choi, Y. C.; Jung, S. C.; Kim, Y.-I.; Jo, M. R.; Park,
S. H.; Kang, Y.-M.; Han, Y.-K.; Park, J. T. Nanoscale 2012, 4, 473.
(43) Nakamoto, K. Infrared and Raman Spectra of Inorganic and
Coordination Compounds, Part B: Applications in Coordination,
Organometallic, and Bioinorganic Chemistry; John Wiley and Sons,
Inc.: Hoboken, 2009.
(44) Deacon, G. B.; Phillips, R. J. Coord. Chem. Rev. 1980, 33, 227.
(45) Kambe, H.; Mita, I. Bull. Chem. Soc. Jpn. 1961, 34, 1797.
(46) Kambe, H.; Mita, I. Bull. Chem. Soc. Jpn. 1961, 34, 1790.
(47) Shieh, S. R.; Duffy, T. S. Phys. Rev. B 2002, 66, 134301.
(48) Wang, X.; Andrews, L. J. Phys. Chem. A 2006, 110, 10035.
(49) Renz, M. Eur. J. Org. Chem. 2005, 6, 979.
(50) Grimes, W. R.; Fitch, A. N. J. Mater. Chem. 1991, 1, 461.
(51) Wanjun, T.; Donghua, C. Chem. Pap. 2007, 61, 329.
(52) De Jesus, J. C.; Gonzalez, I.; Quevedo, A.; Puerta, T. J. Mol.
Catal. A: Chem. 2005, 228, 283.
(53) van Niekirk, J. N.; Schoening, F. R. L. Acta Crystallogr. 1953, 6,
609.
(54) Nicolai, B.; Kearley, G. J.; Johnson, M. R.; Fillaux, F.; Suard, E. J.
Chem. Phys. 1998, 109, 9062.
(55) Liakakos, N.; Cormary, B.; Li, X. J.; Lecante, P.; Respaud, M.;
Maron, L.; Falqui, A.; Genovese, A.; Vendier, L.; Koinis, S.; Chaudret,
B.; Soulantica, K. J. Am. Chem. Soc. 2012, 134, 17922.
(56) Sines, I. T.; Misra, R.; Schiffer, P.; Schaak, R. E. Angew. Chem.,
Int. Ed. 2010, 49, 4638.
(57) Norako, M. E.; Greaney, M. J.; Brutchey, R. L. J. Am. Chem. Soc.
2012, 134, 23.
1499
dx.doi.org/10.1021/cm4041055 | Chem. Mater. 2014, 26, 1492−1499