Toward the syntheses of universal ligands for metal oxide surfaces: Controlling surface functionality through click chemistry

Article abstract of DOI:10.1021/ja064041s

A new means for functionalizing metal oxide surfaces, specifically nanoparticles, is demostrated. This process involves the design of stable ligands that bind strongly to the surface of metal oxides and can undergo further chemical modification via click chemistry, with both small molecules as well as polymers, to yield metal oxide surfaces with tailored functionality. The nanoparticles are stable and easily dispersed in both polar and nonpolar solvents, a property that is controlled by the ligand. The resultant nanoparticles were characterized by TEM, TGA, FTIR, and NMR. Copyright

Full text of DOI:10.1021/ja064041s

Published on Web 08/16/2006
Toward the Syntheses of Universal Ligands for Metal Oxide Surfaces:
Controlling Surface Functionality through Click Chemistry
†
†
‡
,†,‡
Meghann A. White, Jeremiah A. Johnson, Jeffrey T. Koberstein, and Nicholas J. Turro*
Department of Chemistry, Columbia UniVersity, 3000 Broadway, MC 3157, New York, New York 10027, and
Department of Chemical Engineering, Columbia UniVersity, 500 West 120th Street, New York, New York 10027
Received June 8, 2006; E-mail: njt3@columbia.edu; jk1191@columbia.edu
Metal oxide nanoparticles, in particular magnetic metal oxides,
have recently attracted a great deal of attention for their potential
wide range of applications including use as cellular delivery
Scheme 1. General Scheme for Functionalizing the Surface of
Iron Oxide Nanoparticles
1
2
3a,b
carriers, magnetic storage media, and MRI contrast agents. In
many practical applications, nanoparticle cores must be provided
with surface ligands which both prevent aggregation and also
provide a handle that conveniently allows functionalization of the
final periphery of the system. The harsh conditions generally
required for nanoparticle preparation severely limit the use of
functionalized ligands during the synthesis. As a result, a common
method of functionalizing the nanoparticle surface involves stripping
off nonfunctional ligands used for synthesis and then redispersing
the particles with functionalized ligands.
In this report we demonstrate a strategy for the synthesis of
surface functionalized metal oxide nanoparticles through the design
of versatile ligands whose structures include the following fea-
tures: (1) a robust anchor that can bind generally to a variety of
metal oxide surfaces, (2) tailored surface groups that act as spacers
or branches from the metal oxide surface, and (3) a general method
for covalently attaching a functional perimeter to the spacers through
“
click” chemistry. Ligands which possess the flexibility and
synthetic generality of features 1-3 possess the characteristic of
universal” ligands which allow the construction of a broad range
“
of functionalities for the periphery of nanoparticles with good yields
and synthetic facility.
The copper(I) catalyzed azide-alkyne cycloaddition (CuAAC)
has received broad attention because of its unique “click” nature:
an azide or alkyne group at the other terminus, providing orthogonal
functionality for further chemical modification of the nanoparticle-
ligand complex surface. Because organo-phosphates and carboxy-
lates bind well to most metal oxides, this method also has the
potential of applying to a range of metal oxide surfaces.
namely, the reaction proceeds with high yields and no byproducts
and exhibits functional group orthogonality.4
a-e
As a result, the
CuAAC reaction has been employed in a variety of materials
synthesis applications including functionalization of polymers and
2 3
γ-Fe O nanoparticles were synthesized by a previously reported
13
method employing oleic acid as a ligand. XRD and TEM images
show that the nanoparticles are crystalline and well dispersed.
Within one reaction vessel the nanoparticles vary in size only
slightly (<5% rms). The oleic acid was stripped from the particles
and exchanged with either the phosphonic acid ligand or 5-hexynoic
acid (1 and 2 in Scheme 1). The particles were then washed with
hexanes and methanol for 1 and 2, respectively, to remove excess
ligand followed by dispersion in chloroform. The TEM images of
the newly coated nanoparticles indicated that they had not formed
aggregates and that their size did not change upon ligand exchange
within the limits of TEM accuracy. FTIR spectra of the nanopar-
dendrimers,5
a-c
polysaccharides, DNA, and a variety of bulk
6,7
8
9
10
surfaces including gold and silica, as well as gold nanoparticle
surfaces.11 In the design of universal ligands which allow simple
and efficient functionalization of nanoparticle-ligand complexes
with a desired moeity, we found the CuAAC reaction to be ideal
because of its click nature. Herein we demonstrate the potential
2 3
effectiveness of CuAAC for the modification of γ-Fe O nanopar-
ticles with a diverse array of functional species including small
molecules and polymers.
In the design of functional ligands for nanoparticle surfaces, one
must consider both the binding properties as well as the stability
of the ligand and, importantly, prevention of particle aggregation.
14
ticles were compared to those of the free ligands. For 1 there
-1
was a very strong absorbance at 2114 cm owing to the azide
It has been established that organo-phosphates as well as carboxy-
NdNdN antisymmetric stretch, as well as a strong absorbance at
lates bind strongly to the surface of metal oxides.1
2a-g
For our
-1
1
742 cm owing to the CdO stretch of the ester. There was also
studies we chose ligands containing either a phosphonic acid group
or a carboxylic acid group at one terminus, to serve as anchors
-1
a series of stretching bands from 1250 to 990 cm assigned to the
P-O and P-O-Fe stretches of the phosphonic acid group. For 2
which can bind strongly to the surface of the γ-Fe
2 3
O , and either
-1
there was a very weak absorbance at 2119 cm owing to the alkyne
-
1
†
as well as a strong absorbance at 1710 cm owing to the CdO
Department of Chemistry.
Department of Chemical Engineering.
‡
stretch of the carboxylic acid. It was estimated from thermogravi-
11356
9
J. AM. CHEM. SOC. 2006, 128, 11356-11357
10.1021/ja064041s CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
14
proton at ∼8 ppm was detected. TEM images of the nanoparticles
revealed that they were well-dispersed and that no aggregation had
occurred (Figure 1). These results demonstrate that the CuAAC
reaction successfully coupled 5 to 1 yielding a novel nanoparticle-
polymer complex, 6.
2 3
In conclusion, ligand exchange on γ-Fe O nanoparticles was
performed with two types of ligands: (1) phosphonic acid-azide
and (2) carboxylic acid-alkyne. The resultant particles were
submitted to CuAAC reactions with organic substrates, the products
of which were well-dispersed in a range of solvents, a property
which is dependent upon the ligand of choice. These results establish
a “universal ligand” approach for the surface functionalization of
2 3
γ-Fe O nanoparticles and demonstrate that the nanoparticles can
be dispersed in a variety of media by control of the ligand. Not
only is this method applicable to iron oxide nanoparticles, but it is
also presumably applicable to any other metal oxide surface
Figure 1. TEM image of γ-Fe2O3 nanoparticles that have undergone 1,3-
dipolar cycloaddition with poly(tert-butylacrylate) (6 in Scheme 1).
(nanoparticle or bulk).
metric analysis (TGA) data that the surface coverage of the particles
Acknowledgment. Funding for this work was provided by the
2
was ∼1 ligand/nm for 1, and independent of particle size. For 2
National Science Foundation (NSF) under Grant No. IGERT-02-
21589 and No. NSF-04-15516. Facilities were provided by
Columbia University’s Materials Research Science and Engineering
Center (MRSEC) under NSF Grant No. DMR-0214363.
2
we estimated the surface coverage to be ∼11 ligands/nm .
With 1 and 2 in hand, a CuAAC reaction using the complemen-
tary click functional molecule (5-chloropentyne for 1 and benzyl
azide for 2) was performed to prepare 3 and 4 (Scheme 1). The
reactions were allowed to proceed overnight, after which time the
Supporting Information Available: Detailed experimental pro-
cedures as well as control experiments and spectroscopic data; complete
ref 5c. This material is available free of charge via the Internet at http://
pubs.acs.org.
2 3
γ-Fe O particles were recovered via extraction with organic
solvents. The FTIR spectrum of 3 showed a loss of the NdNdN
stretching band, indicating a high yield for the CuAAC reaction,
-
1
and a peak at 1554 cm agreeing with the literature value for a
,2,3-triazole.15 The absorbance bands due to the phosphonic acid
References
1
-
1
(1) Xu, Z. P.; Zeng, Q. H.; Lu, G. Q.; Yu A. B. Chem. Eng. Sci. 2006, 61,
group (1250-990 cm ) are still present, implying that the ligand
is still attached to the particles and the phosphonate is still intact.
The FTIR spectrum of 4 also has a characteristic band at 1551
cm , and has no band at 2100 cm , indicating conversion of the
alkyne group to a triazole. As in previous work using NMR to study
1
027-1040.
(
2) Huber, D. L. Small 2005, 1, 482-501.
(3) (a) Huh, Y. M.; Jun, Y. W.; Song, H. T.; Kim, S.; Choi, J. S.; Lee, J. H.;
Yoon, S.; Kim, K. S.; Shin, J. S.; Suh, J. S.; Cheon, J. J. Am. Chem. Soc.
-
1
-1
2
005, 127, 12387-12391. (b) Song, H. T.; Choi, J. S.; Huh, Y. M.; Kim,
S.; Jun, Y. W.; Suh, J. S.; Cheon, J. J. Am. Chem. Soc. 2005, 127, 9992-
9
993.
16
surface groups on γ-Fe
3
2 3
O nanoparticles, very dilute samples of
(
4) (a) Rodionov, V. O.; Fokin, V. V.; Finn, M. G. Angew. Chem., Int. Ed.
2005, 44, 2210-2215. (b) Narayan, S.; Muldoon, J.; Finn, M. G.; Fokin,
V. V.; Kolb, H. C.; Sharpless, K. B. Angew. Chem., Int. Ed. 2005, 44,
1
and 4 were examined by H NMR. Although the peaks were
slightly broadened, the characteristic peak at ∼8.0 ppm owing to
the triazole proton was clearly present. Both the FTIR and NMR
spectra correspond well to those of the control reactions. This
spectroscopic evidence, coupled with the TEM image showing
dispersed particles, suggests that the 1,3-dipolar cycloaddition was
successful and that the particles are stabilized from aggregation by
the new ligand system.14
To demonstrate that complex functionality could be introduced
on the periphery of the nanoparticle surface by the universal ligand
strategy, a polymeric ligand was attached to the nanoparticle by
CuAAC. An R-acetylene-poly(tert-butyl acrylate) (ptBA) polymer
3
275-3279. (c) Li, Z. M.; Seo, T. S.; Ju, J. Tetrahedron Lett. 2004, 45,
3143-3146. (d) Kolb, H. C.; Sharpless, K. B.; Drug DiscoVery Today
2
003, 8, 1128-1137. (e) Kolb, H. C.; Finn, M. G.; Sharpless, K. B.;
Angew. Chem., Int. Ed. 2001, 40, 2004-2021.
(
5) (a) Hawker, C. J.; Wooley, K. L. Science 2005, 309, 1200-1205. (b)
Helms, B.; Mynar, J. L.; Hawker, C. J.; Frechet, J. M. J. J. Am. Chem.
Soc. 2004, 126, 15020-15021. (c) for complete list of references see
Supporting Information.
(6) Liebert, T.; Hansch, C.; Heinze, T. Macromol. Rapid Commun. 2006, 27,
2
08-213.
(
7) Punna, S.; Kaltgrad, E.; Finn, M. G. Bioconjugate Chem. 2005, 16, 1536-
1541.
(
(
8) Seo, T. S.; Li Z.; Ruparel, H.; Ju J. J. Org. Chem. 2003, 68, 609-612.
9) Collman, J. P.; Devaraj, N. K.; Chidsey, C. E. D. Langmuir 2004, 20,
1051-1053.
(
(
(
10) Lummerstorfer, T.; Hoffmann, H. J. Phys. Chem. B 2004, 108, 3963-
(5) was prepared by atom transfer radical polymerization (ATRP)
3966.
of tert-butyl acrylate in the presence of an acetylene-functional
initiator, CuBr catalyst, N,N,N′,N′′,N′′-pentamethyldiethylenetri-
amine ligand, and toluene solvent. Treatment of 1 with 5 under
CuAAC conditions yielded ptBA coated particles (6) as shown in
Figure 1. On the basis of the estimated surface coverage of 1 on
the nanoparticles a 1:1 molar ratio of alkyne to azide was used in
the CuAAC reaction. The resultant particles were extracted with
organic solvents and then precipitated to remove excess unbound
ptBA. The nanoparticles were characterized using TEM, FTIR, and
NMR. The FTIR spectrum of the polymer coated particles showed
11) Fleming, D. A.; Thode, C. J.; Williams, M. E. Chem. Mater. 2006, 18,
2327-2334.
12) (a) Mutin, P. H.; Guerrero, G.; Vioux, A. J. Mater. Chem. 2005, 15, 3761-
3
768. (b) Matsuno, R.; Yamamoto, K.; Otsuka, H.; Takahara, A.
Macromolecules 2004, 37, 2203-2209. (c) Song, Y. N.; Zavalij P. Y.;
Chernova, N. A.; Suzuku, M.; Whittingham, M. S. J. Solid State Chem.
2
003, 175, 63-71. (d) Kreller, D. I.; Gibson, G.; Novak, W.; van Loon,
G. W.; Horton, J. H. Colloids Surf., A 2003, 212, 249-264. (e) Kreller,
D. I.; Gibson, G.; van Loon, G. W.; Horton, J. H. J. Colloid Interface
Sci. 2002, 254, 205-213. (f) Sahoo, Y.; Pizem, H.; Fried, T; Golodnitsky,
D.; Burstein, L.; Sukenik, C. N.; Markovich, G. Langmuir 2001, 17, 7907-
7
911. (g) Hofer, R.; Textor, M.; Spencer, N. D. Langmuir 2001, 17, 4014-
4
020.
(
13) Yin, M.; Willis, A.; Redl, F.; Turro, N. J.; O’Brien, S. P. J. Mater. Res.
-
1
2004, 19, 1208-1215.
the disappearance of the azide peak at 2114 cm , indicating high
yield of the CuAAC reaction. The FTIR for the polymer coated
particles also contained characteristic peaks due to the ptBA, as
(
(
14) See Supporting Information for further details.
15) Billes, F.; Endredi, H.; Keresztury, G. J. Mol. Struct. 2000, 530, 183-
200.
-
1
(16) Willis, A. L.; Turro, N. J.; O’Brien, S. Chem. Mater. 2005, 17, 5970-5975.
well as the peaks ∼1100 cm due to the phosphonic acid.
1
Furthermore, in the H NMR spectrum, the presence of the triazole
JA064041S
J. AM. CHEM. SOC.
9
VOL. 128, NO. 35, 2006 11357