Ligand effect on the growth and the digestion of Co nanocrystals

Article abstract of DOI:10.1021/ja044419r

The reaction product of cobalt carbonyl decomposition depends on the concentration of the oleic acid ligand. With a low concentration of ligand, nanocrystals nucleate and grow to large ferromagnetic particles through the process of Ostwald ripening and coalescence coarsening. With a high concentration of ligand, stable cluster complexes are formed. Addition or removal of ligand from the reaction products can interchange the formation of cluster complexes and nanocrystals. Copyright

Full text of DOI:10.1021/ja044419r

Published on Web 03/03/2005
Ligand Effect on the Growth and the Digestion of Co Nanocrystals
Anna C. S. Samia,† Kylee Hyzer,‡ John A. Schlueter,‡ Chang-Jin Qin,§ J. Samuel Jiang,‡
Samuel D. Bader,‡, and Xiao-Min Lin*,†,‡,
Chemistry DiVision, Materials Science DiVision, and Center for Nanoscale Materials, Argonne National Laboratory,
Argonne, Illinois 60439, and Department of Chemistry, UniVersity of Chicago, 5635 South Ellis AVenue,
Chicago, Illinois 60637
Received September 14, 2004; E-mail: xmlin@anl.gov
The growth and stability of nanocrystals in organic solvents rely
on the presence of stabilizing ligands.1 Under various kinetic and
thermodynamic conditions, monodispersed nanocrystals with con-
trollable sizes, shapes, and chemical compositions have been
synthesized recently.2 However, the effects of surface ligands,
especially how they regulate the growth of nanocrystals, have not
been studied thoroughly. The common notion is that Ostwald
ripening and coalescence coarsening control the particle growth
process after nucleation.3 Thus, a kinetic control process such as
“size focusing” is needed to maintain the monodispersity of a
growing colloid.4 Recent experimental evidences in Au, Ag, and
Figure 1. Schematic diagram illustrating the experimental procedures to
PbS nanocrystal synthesis have shown that the presence of excess
synthesize cobalt nanocrystals and cluster complexes using different
concentrations of oleic acid ligand (see Supporting Information for
experimental details.)
ligand in the system could result in a completely different ripening
process, namely the growth inhibition or even disintegration of large
nanocrystals in favor of smaller particles, a process named digestive
ripening.5 In this report, we demonstrate that using different
concentrations of oleic acid ligand during the cobalt carbonyl
decomposition results in the formation of either large nanocrystals
or small clusters. More dramatically, we show that by adding or
removing free oleic acid ligand from the final reaction product,
we can turn a nanocrystal colloid into a cluster complex solution
and vice versa.
A general procedure to synthesize cobalt nanocrystals is shown
in path (a) of Figure 1, which is a modification of the procedure
used by several other groups6 (see Supporting Information for
detailed description). Nanocrystals are obtained through thermal
decomposition of Co2(CO)8 in the presence of oleic acid (OA) and
a small amount of trioctylphosphine oxide (TOPO), with the
Co:OA molar ratio fixed at 5:1. It was found that after a quick
injection of precursor at 180 °C and then lowering the growth
temperature to 130 °C, nanocrystals will nucleate and grow
uniformly. The particle size, which is a function of reaction time,
typically reaches 6 nm in diameter after 15 min, and becomes
9.5 nm after 1 h. Figure 2a shows a monolayer of 9.5 nm cobalt
nanocrystals after being washed by methanol, redispersed in
anhydrous toluene, and then deposited onto a carbon grid. Powder
X-ray diffraction (inset of Figure 2a) shows that, different from
the face centered cubic (fcc) and hexagonal closed packed (hcp)
structures in the bulk cobalt, these particles adopt the ꢀ phase, a
distorted fcc structure that is only observed in nanocrystals.6
Magnetic measurements show the blocking temperature of these
large particles is close to room temperature; thus, they are
ferromagnetic near room temperature, with their easy axis randomly
orientated in the sample (Figure 2b and Figure S4 in Supporting
Information). The saturation magnetization was close to the bulk
value (Figure S4 in Supporting Information), and no shift of
Figure 2. (a) Transmission electron micrograph (TEM) of the as-prepared
9.5 nm Co nanocrystals. (Inset) X-ray powder diffraction showing that the
crystal structure of the cobalt is ꢀ-phase. (b) Zero-field-cooled (ZFC) and
field-cooled (FC) magnetic susceptibility of large particles. Small kink at
250 K of the ZFC curve is attributed to small size nonuniformity of the
sample.
hysteresis loop was detected, indicating the absence of a cobalt
oxide layer on the nanocrystal surface.
On the other hand, when the same reaction was conducted using
a higher concentration of OA (Co:OA ratio of 1:2.5), while keeping
all the other conditions the same (path (b) in Figure 1), the result
is drastically different. After the Co2(CO)8 injection, the color of
the reaction mixture turned to black briefly (indicating the initial
formation of cobalt nanocrystals) before changing to blue color after
1 h. The rate at which this transition takes place depends on the
temperature of the reaction. UV-vis absorption measurement of
this solution showed an absorption band in the range of 450-
650 nm (inset of Figure 3), with optical transitions between discrete
energy levels visible in the spectrum. Electrospray ionization mass
spectroscopy (ESI-MS) data were collected using methanol or
ethanol as a carrier solvent and under various fragmentation voltages
(Figure 3 and Figure S1 in supporting materials). The five groups
of predominant ions detected centered near 990 Da. The isotope
pattern indicates that these are all +1 charged ions with mass
corresponding to clusters of Co2 and Co3 coordinated with three
† Chemistry Division, Argonne National Laboratory.
‡ Materials Science Division, Argonne National Laboratory.
Center for Nanoscale Materials, Argonne National Laboratory.
§ Department of Chemistry, University of Chicago.
9
4126
J. AM. CHEM. SOC. 2005, 127, 4126-4127
10.1021/ja044419r CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
ligand bonds. The binding between the ligand and metal surface,
however, is a dynamic process. Reducing the concentration of free
ligand (in path (d) by reacting it with LAH) will shift the
equilibrium, which lowers the concentration of the surface ligand
molecules and allows clusters to sinter and eventually form large
particles through Ostwald ripening and coalescence coarsening. The
fact that Co2 and Co3 clusters are possible intermediate building
blocks for constructing large nanocrystals may explain why ꢀ phase
rather than fcc or hcp phase is observed in these syntheses.
The formation of clusters under high concentration of OA is not
limited to cobalt. Thermal decomposition of Fe(CO)5 under identical
conditions (Fe:OA molar ratio of 1:2.5) form Fe1, Fe2 clusters
(Figure S2 in Supporting Information). However, reducing platinum
acetylacetonate with 1,2-hexandecanediol under the same conditions
leads to the formation of bulk Pt precipitation. This was also
confirmed by results from Weller’s group.9 Therefore, Pt has a much
weaker interaction with OA than Co or Fe. Interestingly, when the
thermal decomposition of Fe(CO)5 occurs in the same solution
where platinum acetylacetonate is being reduced (metal:
OA ) 1:2.5), stable FePt nanocrystals with diameter about 5 nm
form.10 This indicates that it is the formation of alloys between
Fe-OA clusters and growing Pt particles that provides a stabiliza-
tion mechanism for FePt nanocrytals. The relative strength of the
ligand-metal and metal-metal interaction can thus regulate the
growth of metal and their alloy particles, such as Co, CoPt, and
FePt, and in some cases lead to conditions wherein the traditional
ripening processes do not occur.
Figure 3. ESI-MS spectrum of the cobalt cluster complex in
1,2-dichlorobenzene at the fragmentation voltage of 200 V, using methanol
as a carrier solvent. The mass of the peaks correspond to (a) [Co(II)2-
(C18H33O2)3]+, (b)[Co(II)2(C18H33O2)3O]+, (c) [Co(I)2(C18H33O2)3CO+2H]+
(d) [Co(I)2(C18H33O2)3OCO+2H]+, (e) [Co(I)3(C18H33O2)3+H]+. (Inset)
UV-vis absorbance of the cluster solution.
OA ligand molecules. The cobalt centers in these ions have two
different valence states. Since the electrospray is a soft ionization
technique that creates very little fragmentation,7 the clusters we
detected most likely exist in the solution.
Intrigued by the very different reaction products under different
amount of OA ligand, we carried out experiments shown in paths
(c) and (d) of Figure 1. In path (c), Co nanocrystals made from
reaction path a were precipitated and washed with anhydrous
methanol and redispersed in 1,2-dichlorobenzene. Elemental analy-
sis of the precipitated sample indicates there is a monolayer coating
of OA on the particle surface. Excess OA was then added to increase
the molar ratio of Co:OA to 1:2.5. The solution was then heated
under reflux at 180 °C. Following this step, the black color of the
colloidal solution gradually turned blue after 5 h, indicating the
disintegration of the large ferromagnetic nanocrystals. Mass
spectroscopy data confirmed the formation of the same cluster
complex, as shown in Figure 3.
Removal of the excess OA from the cluster complex solution
(Figure 1 path (d)) was achieved through reacting the complex
solution with lithium aluminum hydride (LAH), a standard reaction
that forms lithium aluminum carboxylate salt. LAH also reduce
the cobalt ions back to zero valence state. After the LAH injection,
the solution changed to black almost instantaneously. Upon removal
of the reaction side products through centrifugation, the cobalt
nanocrystals were isolated by precipitation with ethanol and
confirmed by TEM measurement.
The formation of either large nanocrystals or small cluster
complexes under different concentration of OA ligand, and the fact
that they can be reversibly interchanged by varying the free ligand
concentration, indicate that there is a strong and yet reversible
binding between the OA ligand and the Co nanocrystal surface.
With the presence of a high concentration of OA ligand, cluster
complexes are more stable than large nanocrystals. The dramatic
digestive ripening process (path (c)) in which excess OA ligands
literally pull Co atoms off the nanocrystal surface to form clusters
is a thermodynamically driven process. The similar ripening process
in gold-dodecanethiol colloids results in 5-7 nm particles at
120 °C5 and small clusters at 300 °C.8 In comparison, cluster
complexes form at a much lower temperature in the Co-OA system
due to the different strengths of the metal-metal and the metal-
Acknowledgment. We are grateful to Yuping Bao and Kannan
Krishnan for the help on initial experiments and Chris Sorensen
for many valuable discussions. This work is supported by DOE,
BES-Materials Sciences, under Contract W-31-109-ENG-38, and
the University of Chicago-ANL Consortium for Nanoscience
Research (CNR).
Supporting Information Available: Synthesis procedures, ad-
ditional TEM images, ESI-MS, and magnetic data. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1) Schmid, G., Eds. Clusters and Colloids: From Theory to Applications;
VCH: New York, 1994.
(2) (a) Murray, C. B.; Kagan, C. R.; Bawendi, M. G. Annu. ReV. Mater. Sci.
2000, 30, 545-610. (b) Lee, S.-M.; Jun, Y.-w.; Cho, S.-N.; Cheon, J. J.
Am. Chem. Soc. 2002, 124, 11244-11245. (c) Song, Q.; Zhang, Z. J. J.
Am. Chem. Soc. 2004, 126, 6164-6168. (d) Jana, N. R.; Gearheart, L.;
Murphy, C. J. J. Phys. Chem. B 2001, 105, 4065-4067. (e) Peng Z. A.
Peng, X. J. Am. Chem. Soc. 2001, 123, 1389-1395. (f) Manna, L. Scher,
E. C.; Alivisatos, A. P. J. Am. Chem. Soc. 2000, 122, 12700-12706.
(3) Ratke, L.; Voorhees, P. W. Growth and Coarsening; Springer-Verlag:
Berlin, 2002.
(4) Peng, X. G.; Wickham, J.; Alivisatos, A. P. J. Am. Chem. Soc. 1998,
120, 5343-5344.
(5) (a) Lin, X. M.; Sorensen; C. M. Klabunde, K. J. J. Nanopart. Res. 2000,
2, 157-164. (b) Prasad, B. L. V.; Stoeva, S. I.; Sorensen, C. M.; Klabunde,
K. J. Chem. Mater. 2003, 15, 935-942. (c) Stoeva, S.; Klabunde, K. J.;
Sorensen, C. M.; Dragieva, I. J. Am. Chem. Soc. 2002, 124, 2305-2311.
(d) Smetana, A. B.; Klabunde, K. J.; Sorensen, C. M. To be published.
(e) Hines, M. A.; Scholes, G. D. AdV. Mater. 2003, 15, 1844-1849.
(6) (a) Puntes, V. F.; Krishnan, K. M.; Alivisatos, A. P. Science 2001, 291,
2115-2117. (b) Sun, S. H.; Murray, C. B. J. Appl. Phys. 1999, 85, 4325-
4330. (c) Dinega, D. P.; Bawendi, M. G. Angew. Chem., Int. Ed. 1999,
38, 1788-1791.
(7) Johnson, B. F. G.; McIndoe, J. S. Coord. Chem. ReV. 2000, 200-202,
901-932.
(8) Jin, R. C.; Egusa, S.; Scherer, N. F. J. Am. Chem. Soc. 2004, 126, 9900-
9901.
(9) Shevchenko, E. V.; Talapin, D. V.; Schnablegger, H.; Kornowski, A.;
Festin, O.; Svedlindh, P.; Haase, M.; Weller, H. J. Am. Chem. Soc. 2003,
125, 9090-9101.
(10) Chen, M.; Liu, J. P.; Sun, S. J. Am. Chem. Soc. 2004, 126, 8394-8395.
JA044419R
9
J. AM. CHEM. SOC. VOL. 127, NO. 12, 2005 4127