Reactive silver and cobalt nanoparticles modified with fatty acid ligands functionalized by imidazole derivatives

Article abstract of DOI:10.1021/cm1018508

Novel nucleophilic ligands were synthesized from oleic acid conjugated with 2-mercaptoimidazole via a UV-initiated thiol-ene reaction, and by a condensation of trans-9,10-epoxystearic acid with 4(5)-imidazoledithiocarboxylic acid. Nanoparticles (NPs) with

Full text of DOI:10.1021/cm1018508

Chem. Mater. 2010, 22, 5383–5391 5383
DOI:10.1021/cm1018508
Reactive Silver and Cobalt Nanoparticles Modified with Fatty Acid
Ligands Functionalized by Imidazole Derivatives
Lev Bromberg, Liang Chen, Emily P. Chang, Sa Wang, and T. Alan Hatton*
Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge,
Massachusetts 02139
Received July 2, 2010. Revised Manuscript Received August 6, 2010
Novel nucleophilic ligands were synthesized from oleic acid conjugated with 2-mercaptoimidazole
via a UV-initiated thiol-ene reaction, and by a condensation of trans-9,10-epoxystearic acid with
4(5)-imidazoledithiocarboxylic acid. Nanoparticles (NPs) with 10-100 nm Ag(0) or Co(0) cores were
obtained by reduction of Agþ or Co2þ by borohydride in N,N-dimethylformamide or aqueous solu-
tions respectively. The NPs capped with oleic acid or 9,10-epoxystearic acids were further covalently
bound to 2-mercaptoimidazole or 4(5)-imidazoledithiocarboxylic acid. The ligands and NPs
functionalized with nucleophilic imidazole moieties enabled facile hydrolysis of paraoxon (O,O-
diethyl O-(p-nitrophenyl) phosphate) by NPs in their aqueous media. The NPs acted as recoverable
semiheterogeneous catalysts. The paraoxon hydrolysis was accelerated 10- to 50-fold by the form-
ation of complexes between the imidazole-containing ligands or NPs with Co2þ
.
Introduction
our NPs are not only capable of carrying out catalytic
functions but can also bind strongly to bacterial mem-
branes and kill various germs, thus acting as efficient
disinfectants.6,7 Previously, the majority of the ligands we
employed to impart esterolytic function to the NP and
other colloids have been oximes, hydroxamic acids, io-
dosobenzoates, and other R-nucleophiles.3,4,8,9 Although
capable of facile hydrolysis of organophosphorous (OP)
pesticides and chemical warfare agents, R-nucleophiles
are quite sensitive to the presence of compounds that are
not targeted and can also undergo transformations ren-
dering them unreactive, such as Beckmann and Lossen
rearrangements. Therefore, in the present work, our goal
was to create NPs functionalized with imidazole derivatives.
Polymers and nanosized assemblies functionalized with
imidazole catalyze esterolytic and proteolytic reactions.10-19
Transition metal and metal oxide nanoparticles have
generated significant interest because of their large sur-
face-to-volume ratios, functional properties, and impor-
tant applications, ranging from electronics and catalysis
to biomedicine.1,2 Our specific area of interest has been
nanoparticles (NPs) with a metal or metal oxide core that
is functionalized with an organic ligand, either a small-
molecular-weight species or a polymer, tailored for the
destruction of toxins.3-5 Our NP design is based on the
reactivity of the ligand, and not on the catalytic properties
of a transition metal per se. The core of the ligand-deco-
rated NP is nonreactive, but determines the NP shape and
size, while facilitating the particle recoverability. The NPs
can be readily recovered from the reaction medium by
centrifugation or filtration. Large surface area, small
thickness (up to 10% of the total hydrodynamic diameter
of a given NP) and compatibility of the ligand shell with
the continuous phases ensure accessibility of the reactive
ligands to the reagents.
(6) Bromberg, L.; Raduyk, S.; Hatton, T. A. Anal. Chem. 2009, 81,
5637–5645.
(7) Bromberg, L.; Chang, E. P.; Alvarez-Lorenzo, C.; Magarinos, B.;
Concheiro, A.; Hatton, T. A. Langmuir 2010, 26, 8829–8835.
(8) Bromberg, L.; Zhang, H.; Hatton, T. A. Chem. Mater. 2008, 20,
2001–2008.
(9) Bromberg, L.; Schreuder-Gibson, H.; Creasy, W. R.; McGarvey,
D. J.; Fry, R. A.; Hatton, T. A. Ind. Eng. Chem. Res. 2009, 48,
1650–1659.
Our catalytic NP is a hybrid between the supported
transition metal particles and functional polymeric colloids.
Because of the multifunctionality of the organic ligands,
(10) Overberger, C. G.; Salamone, J. C. Acc. Chem. Res. 1969, 2,
217–224.
(11) Overberger, C. G.; Kwon, B.-D. In Initiation of Polymerization;
Bailey, F. E., Jr., Vandeberg, E. J., Blumstein,A., Bowden, M. J., Arthur,
J. C., Lal, J., Ottenbrite, R. M., Eds.; ACS Symposium Series; American
Chemical Society: Washington, D.C., 1983; Vol. 212, Chapter 6,
pp 65-74.
*Corresponding author. E-mail: tahatton@mit.edu.
(1) Fedlheim, D. L. ; Foss, C. A. Metal Nanoparticles: Synthesis,
Characterization, and Applications; CRC Press: Boca Raton, FL,
2001; 352 pp.
(2) Rotello, V. Nanoparticles: Building Blocks for Nanotechnology;
Springer: New York, 2004. Nanoparticles and Catalysis; Astruc, D.,
Ed.; Wiley-VCH: Weinheim, Germany, 2008.
(3) Bromberg, L.; Hatton, T. A. Ind. Eng. Chem. Res. 2005, 44, 7991–
7998.
(4) Bromberg, L.; Hatton, T. A. Ind. Eng. Chem. Res. 2007, 46, 3296–
3303.
(5) Hatton; T. A., Bromberg; L. E. Catalytic nanoparticles for nerve-
(12) Overberger, C. G.; Tomko, R. In Initiation of Polymerization;
Bailey, F. E., Jr., Vandeberg, E. J., Blumstein,A., Bowden, M. J., Arthur,
J. C., Lal, J., Ottenbrite, R. M., Eds.; ACS Symposium Series; American
Chemical Society: Washington, D.C., 1983; Vol. 212, Chapter 2, pp
13-21.
(13) Guler, M. O.; Stupp, S. I. J. Am. Chem. Soc. 2007, 129, 12082–
12083.
agent destruction. U.S. Patent 7 598 199, October 6, 2009.
(14) Kitano, H.; Sun, Z.; Ise, N. Macromolecules 1983, 16, 1823–1827.
r
2010 American Chemical Society
Published on Web 08/20/2010
pubs.acs.org/cm

5384 Chem. Mater., Vol. 22, No. 18, 2010
Bromberg et al.
The mechanism of the ester hydrolysis catalysis by low-
molecular-weight imidazole derivatives is a function of the
ester (substrate) structure.20-26 Esters that possess poor
leaving groups are subject to classical general base catalysis
by imidazole, whereas esters with good leaving groups are
subject to nucleophilic catalysis.20 In polymers and colloidal
assemblies, the activity of the imidazole group (much like in
enzymes), depends on its microenvironment such as the
presence of hydrophobic or hydrophilic moieties, as well as
other imidazole groups, in its proximity.19 In that regard, we
were interested in potential neighboring effects of the imida-
zole groups concentrated in the organic shell, on esterolytic
reactions. It has been shown that self-assembling monolayers
of various ligands form a multitude of patterns on the NP
surfaces, which affect the microenvironment of the ligands
and hence the NP solvation and its interaction with
substrates.27 The confinement of the catalytic units to
the monolayer covering the NP triggers a cooperative,
pH-dependent hydrolytic mechanism in which an imida-
zolium ion acts as a catalyst.28-30
Noting the widespread use of fatty acids as steric and
electrostatic stabilizers for the colloidal stability of
NPs,31,32 in the present work, we created functional fatty
acid-imidazole conjugates for the NP surface modifica-
tion. Because oleic acid is strongly surface-active and pos-
sesses a reactive unsaturated bond, we concentrated on its
use for conjugating with imidazole derivatives, along with
its epoxidized derivative, trans-9,10-epoxystearic acid.
Thiol-ene coupling reactions involving the double bond
of oleic acid and its esters are known.33 Commercially
available derivatives of imidazole such as 2-mercaptoimi-
dazole and 4(5)-imidazoledithiocarboxylic acid were cho-
sen for conjugation with oleic acid and its derivative.
2-Mercaptoimidazole is a reactive, neutral nucleophile
(pKa of the thiol group ionization, 11.6)34,35 capable of
participating in thiol-ene reactions,35,36 which are facili-
tated by the relatively low S-H bond dissociation energy,
and even more readily, in conjugation reactions with
epoxy groups.37 2-Mercaptoimidazole is also a bioactive
metal-complexing agent.38-43 Likewise, the reactivity of
the dithio group and metal-chelating capability of the
dithio and imidazole functionalities in 4(5)-imidazole-
dithiocarboxylic acid have made it a building block for
synthesis of various metal-organic compounds, func-
tional polymers, and metal ion adsorbents.44-48
Testing of the hydrolytic activity of the ligands and
nanoparticles was performed using the insecticide para-
oxon as a substrate. Hydrolysis of paraoxon and analo-
gous OP pesticides catalyzed by metal ions,49 micelles of
nucleophilic surfactants, and metallomicelles50-52 and
enzymes53,54 has been studied in detail and such hydro-
lysis represents a convenient reaction reflecting upon the
activity of the catalytic agent used. Although the hydro-
lysis of paraoxon is of practical importance in its own
right from an environmental standpoint,49 it can also
serve as a mechanistic model of degradation of more toxic
warfare agents. We chose cobalt and silver to be the cores
of our NPs, as their synthesis and detection are well-
established,55,56 yet fatty acids are capable of binding
(35) Carlsson, J.; Kierstan, M. P. J.; Brocklienhurst, K. Biochem. J.
1974, 139, 221–235.
(36) Persson, J. C.; Jannasch, P. Macromolecules 2005, 38, 3283–3289.
(37) Schwarz, A.; Wilchek, M. U.S. Patent 5 719 269, February 17,
1998.
(38) Tweit, R. C.; Kreider, E. M.; Muir, R. D. J. Med. Chem. 1973, 16,
1161–1169.
(39) Roser, K. S.; Brookes, P. S.; Wojtovich, A. P.; Olson, L. P.; Shojaie,
J.; Parton, R. L.; Anders, M. W. Bioorg. Med. Chem. 2010, 18,
1441–1448.
(40) Van Lommen, G.; Doyon, J.; Coesemans, E.; Boeckx, S.; Cools,
M.; Buntinx, M.; Hermans, B.; VanWauwe, J. Bioorg. Med. Chem.
Lett. 2005, 15, 497–500.
ꢀ
ꢀ
(41) Alvarez-Bustamante, R.; Negron-Silva, G.; Abreu-Quijano, M.;
(15) Overberger, C. G.; Morimoto, M. J. Am. Chem. Soc. 1971, 93,
3222–3228.
(16) Kitano, H.; Sun, Z. H.; Isa, N. Macromolecules 1983, 16, 1306–
1310.
(17) Kunitake, T.; Shinkai, S. J. Am. Chem. Soc. 1971, 93, 4256–4263.
(18) Kunitake, T.; Shimada, F.; Aso, C. J. Am. Chem. Soc. 1969, 91,
2716–2723.
(19) Suh, J.; Oh, S. J. Org. Chem. 2000, 65, 7534–7540.
(20) Kirsch, J. F.; Jencks, W. P. J. Am. Chem. Soc. 1964, 86, 837.
(21) Rogers, G. A.; Bruice, B. C. J. Am. Chem. Soc. 1974, 96, 2463–2473.
(22) Pollack, R. M.; Dumsha, T. C. J. Am. Chem. Soc. 1975, 97, 377.
(23) Kunitake, T.; Ihara, H.; Okahata, Y. J. Am. Chem. Soc. 1983, 105,
6070.
(24) Mallick, I. M.; D’Souza, V. T.; Yamaguchi, M.; Lee, T.; Chalabi,
P.; Gadwood, R. C.; Bencer, M. L. J. Am. Chem. Soc. 1984, 106,
7252.
(25) Ohkubo, K.; Uranbe, K.; Yamamoto, J.; Sagawa, T.; Usui, S. J.
Chem. Soc., Perkin Tran. 11995, 2957.
(26) Broo, K. S.; Nilsson, H.; Flodberg, A.; Baltzer, L. J. Am. Chem.
Soc. 1998, 120, 4063.
(27) Centrone, A.; Penzo, E.; Sharma, M.; Myerson, J. W.; Jackson,
A. M.; Marzari, N.; Stellacci, F. Proc. Natl. Acad. Sci. U.S.A. 2008,
105(29), 9886–9891.
ꢀ
ꢀ
Herrera-Hernandez, H.; Romero-Romo, M.; Cuan, A.; Palomar-
ꢀ
Pardave, M. Electrochim. Acta 2009, 54(30), 5393–5399.
(42) Ryan, N. T.; Richert, D. A.; Westerfeld, W. W.; Fujishima, M.;
Matsuo, Y.; Takatori, H.; Uchida, K. Electrochem. Commun. 2008,
10, 1482–1485.
(43) Dias Filho, N. L.; Gushikem, Y. Sep. Sci. Technol. 1997, 32, 2535–
2545.
(44) Sawa, N., Saeki, T. Imidazole 4(5)-dithiocarboxylic acids or salts,
U.S. Patent 4 394 511, Jul 19, 1983.
(45) Ma, C.; Han, Y.; Zhang, R. Inorg. Chim. Acta 2005, 358, 3084–
3092.
(46) Carter, S.; Hunt, B.; Rimmer, S. Macromolecules 2005, 38, 4595–
4603.
(47) Carter, S.; Rimmer, S.; Rutkaite, R.; Swanson, L.; Fairclough,
J. P. A. Biomacromolecules 2006, 7, 1124–1130.
(48) Grigoropoulou, G.; Stathi, P.; Karakassides, M. A.; Louloudi, M.;
Deligiannakis, Y. Colloids Surf., A: Physicochem. Eng. Aspects 2008,
320, 25–35.
(49) Smolen, J. M; Stone, A. T. Environ. Sci. Technol. 1997, 31, 1664–
1673.
(50) Moss, R. A.; Kanamathareddy, S.; Vijayaraghavan, S. Langmuir
2001, 17, 6108–6111.
(51) Morales-Rojas, H.; Moss, R. A. Chem. Rev. 2002, 102, 2497–2522.
(52) Fubin, J.; Fubin, J.; Bingying, J.; Bingying, J.; Xiaoqi, Y.; Xiancheng,
Z. Langmuir 2002, 18, 6769–6774.
(53) Lei, Y.; Mulchandani, A.; Chen, W. Biotechnol. Prog. 2005, 21,
678–681.
(54) Richter, R. J.; Jarvik, G. P.; Furlong, C. E. Toxicol. Appl. Phar-
(28) Lucarini, M.; Franchi, P.; Pedulli, G. F.; Pengo, P.; Scrimin, P.;
Pasquato, L. J. Am. Chem. Soc. 2004, 126, 9326–9329.
(29) Pasquato, L.; Rancan, F.; Scrimin, P.; Mancin, F.; Frigeri, C.
Chem. Commun. 2000, 2253–2254.
(30) Pengo, P.; Polizzi, S.; Pasquato, L.; Scrimin, P. J. Am. Chem. Soc.
2005, 127, 1616–1617.
(31) Yin, Y.; Alivisatos, A. P. Nature 2004, 437, 664–670.
(32) Harada, T.; Hatton, T. A. Langmuir 2009, 25, 6407–6412.
(33) Samuelsson, J.; Jonsson, M.; Brinck, T.; Johansson, M. J. Polym.
Sci., Part A: Polym. Chem. 2004, 42, 6346–6352.
(34) Stanovnik, B.; Tisler, M. Anal. Biochem. 1964, 9, 68–74.
macol. 2009, 235, 1–9.
(55) Mulfinger, L.; Solomon, S. D.; Bahadory, M.; Jeyarajasingam,
A. V.; Rutkowsky, S. A.; Boritz, C. J. Chem. Educ. 2007, 84, 322.
(56) Puntes, V. F.; Krishnan, K. M.; Alivisatos, A. P. Science 2001, 291
(5511), 2115–2117.

Article
Chem. Mater., Vol. 22, No. 18, 2010 5385
Figure 2. Synthesis of conjugate of trans-9,10-epoxystearic and 4(5)-
imidazoledithiocarboxylic acids (2).
Conjugate of trans-9,10-Epoxystearic and Imidazoledithiocarbo-
xylic Acids ((9s,10s)-1,1,10-Trihydroxyoctadecan-9-Yl 1H-Imidazole-
5-carbodithioate) (2) (Figure 2). trans-9,10-Epoxystearic acid
(30 mg, 0.1 mmol) and 4(5)-imidazoledithiocarboxylic acid
(15 mg, 0.1 mmol) were dissolved in 1 mL of N,N-dimethylfor-
mamide and the reaction was conducted in a sealed vial at 70 °C
for 24 h. The initial reaction solution was dark-red and became
brown at the end of the reaction. The products were purified as
follows. The reaction mixture was allowed to cool to ambient
temperature and excess deionized water (15 mL) was added.
Dark-brown precipitates were formed. The precipitates were
centrifuged (12 900 g, 15 min) and separated. The solids were re-
dissolved in cold acetone (4 °C, 15 mL). Upon heating to 70 °C,
dark-red precipitates were formed in the acetone. The solids
were separated from the hot acetone by centrifugation and
residual solvent was evaporated under vacuum. Yield, 83%.
The TLC in ethanol/acetone (1:1) mixture showed one product
Figure 1. Synthesis of oleic acid and 2-mercaptoimidazole conjugate (1)
by thiol-ene reaction.
through their carboxylic groups with the surfaces of these
metals. On the basis of the above-described rationale, in
the present work we designed and synthesized water-dis-
persible NPs with silver or cobalt cores functionalized by
shells of fatty acids modified with imidazole derivatives.
The NPs are reactive toward paraoxon, especially in the
presence of cobalt(II) ions, as described below.
Experimental Section
Materials. 2-Mercaptoimidazole (98%), trans-9,10-epoxys-
tearic acid (99% by TLC), 4(5)-imidazoledithiocarboxylic acid
(70%), oleic acid (g99%), (4-(2-hydroxyethyl)-1-piperazineeth-
anesulfonic acid) (HEPES, 99%), paraoxon-ethyl (O,O-diethyl
O-(4-nitrophenyl) phosphate, 98%), cobalt(II) chloride hexahy-
drate (98%), and silver nitrate (99%) were all obtained from
Sigma-Aldrich Chemical Co. and used as received. All other
chemicals and solvents were obtained from commercial sources
and were of the highest purity available.
1
(Rf ≈ 0.5). H NMR (DMF-d7, 400 MHz, S2), δ (ppm): 12.9
(1H, imidazole), 12.1 (1H, H-O-C(dO)), 7.88 (1H, imidazole),
3.48 (1H, methane), 2.24, 151 (2H, methylene), 1.25-1.44 (24H,
methylene),0.88 (3H, methyl). Found (calcd): C, 60.1 (59.69); H,
8.59 (8.65); N, 6.05 (6.33); S, 13.9(14.49). Compound 2 was
water-soluble at pH >8.5 and dissolved in acetone, DMF, and
DMSO at 2.5 mM and room temperature.
Synthesis of Co(0) Nanoparticles without Capping Agents58.
Cobalt chloride hexahydrate (238 mg, 1.0 mmol) dissolved in
15 mL ethylene glycol was mixed with an aqueous solution of
NaOH (2 mmol in 20 mL). Hydrazine (1.6 g, 50 mmol) was
slowly added to the resulting solution via syringe and the
mixture was briefly sonicated and kept at 80 °C for 5 h, with
intermittent vortexing. The formed particles were then sepa-
rated from the supernatant by centrifugation (12 900 g, 5 min)
and washed with excess ethanol, followed by centrifugation and
repeated washing. The wash-outs were diluted 3-fold by deio-
nized water and the pH of the resulting solution was measured to
be 7.6, indicating that the excess hydrazine and sodium hydro-
xide was removed. The resulting particles were separated by
centrifugation, dried under vacuum and stored in a desiccator in
oxygen-free conditions, to prevent cobalt oxidation. For DLS
measurements, the particles were dispersed in deionized water
by brief sonication and then the number-average hydrodynamic
diameter of the particles was measured within 1-2 min to be
7.1 ( 1.3 nm.
Synthesis of Co(0) Nanoparticles Modified by 9,10-trans-
Epoxystearic Acid. Cobalt nanoparticles prepared without cap-
ping agents (20 mg) were suspended in a solution of 9,10-trans-
epoxystearic acid(30mg, 0.1 mmol, dissolved in 1.0 mL methanol)
and briefly sonicated. To the suspension, 5 mL of deionized water
was added and the suspension was vortexed. The particles were
then separated from the reaction mixture by centrifugation at
12,900 g, suspended in 20 mL deionized water with brief sonica-
tion and the procedure of washing and separation was repeated
Syntheses. Conjugate of Oleic Acid and 2-Mercaptoimidazole
(10-((1H-Imidazol-2-yl)thio)octadecanoic acid) (1) (Figure 1). A
solution of oleic acid (1.0 g, 3.54 mmol), 2-mercaptoimidazole
(0.354 g, 3.54 mmol), and benzophenone (50 mg, 0.27 mmol) in
10 mL dry methanol was deaerated and placed in a glass vial,
which was subsequently sealed and irradiated using a UV-vis
5000EC flood lamp (spectral output, 300-500 nm; maximum
intensity, 250 mW/cm2 at 365 nm; Dymax Corp, Torrington,
CT) for 10 min. The resulting yellow solution was equilibrated at
ambient temperature and vacuum-evaporated. The products
were separated by TLC (silica gel, aluminum oxide sheets Si
60 F254, 5 ꢀ 10 cm, EMD Chemicals, Gibbstown, NJ) using
hexane-ethyl acetate (80:20 v/v).57,58 Yield, 75%. The initial
compounds moved with the solvent front; the product observed
under UV (Rf ≈ 0.5) was separated and dissolved in 250 μL of
1
CD3OD. H NMR (CD3OD, 400 MHz, see the Supporting
Information, S1), δ (ppm): 6.8 (2H, imidazole), 3.2 (1H,
methine, R to C-S), 2.3 (2H, methylene, R to COOH), 1.55
(2H, methylene), 1.1-1.3 (22H, methylene), 0.87 (3H, methyl).
Found (calcd): C, 65.1 (64.82); H, 9.26 (10.34); N, 6.93 (7.56); S,
8.29 (8.65). The solubility of compound 1 was tested at room
temperature and at 2.5 mM concentration level. Compound 1
formed clear solutions in water at pH >5, methanol, ethanol,
DMF, and DMSO.
(57) Newman, J. W.; Hammock, B. D. J. Chromatogr., A 2001, 925,
223–240.
(58) Balela, M. D. L.; Lockman, Z.; Azizan, A.; Matsubara, E.;
Amorsolo., A. V. J.Phys. Sci. 2008, 9, 1–11.

5386 Chem. Mater., Vol. 22, No. 18, 2010
Bromberg et al.
three times. The resulting Co(0) particles were water- and
organic solvent-dispersible. The particles were characterized
by DLS in water (pH 7), thermogravimetric analysis, and
SQUID. The particles were found to have a number-average
hydrodynamic diameter of 26 ( 11 nm, a saturation magnetiza-
tion of ∼25 emu/g of cobalt, and a fatty acid content of 22 wt %.
Synthesis of Co(0) Nanoparticles Modified by Oleic Acid. A
solution of oleic acid (30 mg, 106 μmol, dissolved in 1.0 mL
methanol) was suspended in 8 mL deionized water using brief
sonication. To the suspension, a solution of cobalt chloride
hexahydrate (238 mg, 1.0 mmol, dissolved in 2 mL water) was
added, resulting in a pink-colored, opaque suspension, to which
a solution of sodium borohydride (100 mg, 2.6 mmol in 3 mL
water) was added under stirring. Foaming and formation of
black-colored particles was observed. The reaction was allowed
to proceed in a sealed flask for 16 h at room temperature with
shaking. In a separate set of experiments, a freshly prepared
solution of 2-mercaptoimidazole (150 mg, 1.5 mmol in 10 mL
water) was added first, followed by addition of sodium borohy-
dride (100 mg, 2.6 mmol in 3 mL water) to cobalt chloride
hexahydrate and oleic acid suspension as above. The reaction
was allowed to proceed identically to the one without 2-mer-
captoimidazole. The particles were separated from the reaction
mixture by centrifugation at 12 900 g and suspended in 50 mL of
deionized water with brief sonication and the procedure of
washing and separation was repeated three times. The resulting
Co(0) particles were characterized by TEM, DLS in water
(pH 7), thermogravimetric analysis and SQUID. After washing,
the Co and S composition (wt%) of the cubical cobalt nano-
particles was: Co, 83.0; S, 0.75. No sulfur was detected in the
particles prepared without 2-mercaptoimidazole. SQUIDanalysis
of the cobalt NP modified with oleic acid yielded a saturation
magnetization of approximately 25 emu/g of cobalt (S3).
Synthesis of Silver Nanoparticles in DMF. Silver nanoparti-
cles were prepared by the reduction of AgNO3 in N,N-dimethyl-
formamide (DMF).59 In a typical experiment, silver nitrate (2 g)
and sodium borohydride (200 mg) were dissolved separately,
each in 25 mL of DMF. Then, the borohydride solution was
added dropwise into the silver nitrate solution in DMF. The
color of the solution became black. The suspension was stirred
and kept at 80 °C for 1 h to complete the reaction. At the end, a
black precipitate was formed, which was centrifuged at 12 900 g
and purified by washing with acetone, which was repeated
three times to remove the excess DMF. The black precipitate
was dried in a vacuum oven at 50 °C.
Ag or Co particles (Co particles were prepared with the use of
2-mercaptoimidazole, see above) were further modified by
conjugating 2-mercaptoimidazole with oleic acid on the particle
surface using the thiol-ene reaction. Namely, 25 mg of particles
modified with oleic acid were suspended in 2 mL of dry
methanol with brief sonication, a solution of 2-mercaptoimida-
zole (4 mg, 40 μmol) and benzophenone (2 mg, 11 μmol) in 2 mL
dry methanol was added, and the mixture was deaerated and
placed in a glass vial, which was subsequently sealed and irradi-
ated using a UV-vis 5000EC flood lamp (spectral output,
300-500 nm; maximum intensity, 250 mW/cm2 at 365 nm;
Dymax Corp, Torrington, CT) for 10 min. The resulting opaque
suspension was equilibrated at ambient temperature, vacuum-
evaporated, and resuspended in methanol (20 mL), following
which the particles were separated by centrifugation (12 900 g,
5 min). The procedure of washing with methanol was repeated
three times and the particles were vacuum-dried and kept at 4 °C
prior to the use. The particles were visualized by TEM (S4).
Silver particles modified with conjugate (1) (designated Silver
NP-(1)) appeared to be aggregates of spherical particles of ∼10 nm
diameter, with the aggregate size of approximately 200 nm (S4),
whereas modified Co particles (designated Cobalt NP-(1)) were
cubical or octahedral, with an average size of ∼100 nm (S5).
These observations were corroborated by DLS measurements in
water (pH adjusted to 7), yielding number-average hydrody-
namic diameters of 190 and 130 nm for Silver NP-(1) and Cobalt
NP-(1), respectively. TGA yielded the contents of Ag and Co
to be 77 and 80 wt % in Silver NP-(1) and Cobalt NP-(1),
respectively. SQUID measurements yielded a saturation mag-
netization of Cobalt NP-(1) to be 6.7 emu/g of cobalt (S3).
Elemental analysis found, for silver NP-(1): C, 15.2; H, 2.51; Ag,
76.9; N, 1.68; S, 1.04; found, for cobalt NP-(1): C, 13.4; H, 2.31;
Co, 77.4; N, 2.44; S, 2.95.
Modification of Ag or Co Nanoparticles with Epoxystearic and
4(5)-Imidazoledithiocarboxylic Acids Conjugate (2). The silver or
cobalt NPs capped with 9,10-trans-epoxystearic acid (25 mg)
were suspended in a solution of 4(5)-imidazoledithiocarboxylic
acid (14 mg, 0.1 mmol, dissolved in 1.0 mL DMF) and the
resulting suspension was kept at 70 °C for 8 h with intermittent
vortexing. The resulting particles were separated by centrifuga-
tion (12 900 g, 5 min), diluted with 20 mL ethanol, briefly soni-
cated, and again separated by centrifugation. The procedure
was repeated three times. The resulting particles were dried
under vacuum and characterized by TGA, TEM (S6), and ele-
mental analysis. SQUID measurements yielded a saturation
magnetization of Cobalt NP-(2) to be approximately 2.0 emu/g
of cobalt (S3). TGA measured the contents of Ag and Co to be
82 and 79 wt % in silver NP-(2) and cobalt NP-(2), respectively.
Elemental analysis found, for Silver NP-(2): C, 11.8; H, 1.58; Ag,
81.0; N, 1.18; S, 2.69; found, for Cobalt NP-(2): C, 13.1; H, 1.91;
Co, 78.7; N, 1.43; S, 3.11.
Complexes of Ligands and Nanoparticles with Co2þ. Com-
plexes of ligands and nanoparticles with Co2þ were prepared by
suspending the corresponding species in deionized water at 3-4
mg/mL concentration and adding an aqueous solution of
cobalt(II) chloride hexahydrate (5 mL) resulting in a cobalt
concentration that was equimolar to the corresponding ligand.
The resulting suspension was separated by centrifugation
(12 900 g, 20 min) and the remaining blue paste or solids were
lyophilized and stored at -20 °C prior to the use.
Synthesis of Silver Nanoparticles Modified by Oleic or 9,10-
trans-Epoxystearic Acid. The black powder of silver particles
prepared in DMF was dispersed in 15 mL of oleic acid (initial
particle amount, 220 mg) or a solution of 9,10-trans-epoxystea-
ric acid (50 mg acid 1 mL of DMF; initial particle amount,
30 mg), which was vortexed and vigorously shaken for 24 h at room
temperature. The capped silver nanoparticles were centrifuged
at 12 900 g and purified by washing with absolute ethanol. This
process was repeated three times in order to remove the excess
capping acid. Further removal of ethanol by vaporizing and
drying in a vacuum oven gave a bright-black, dry powder of
surface-modified silver nanoparticles. The particles capped with
oleic and epoxystearic acid contained 24 and 16 wt % corre-
sponding fatty acids, respectively.
Synthesis of Ag and Co Nanoparticles Modified by 2-Mercap-
toimidazole and Oleic Acid Conjugate (1). The oleic-acid-capped
Characterization Methods. NMR Spectroscopy and Kinetics
of Paraoxon Hydrolysis. 1H and 31P NMR spectra were collected
at 25 ( 0.5 °C using a Bruker Avance-400 spectrometer operat-
(59) Chen, M.; Ding, W. H.; Kong, Y.; Diao, G. W. Langmuir 2008, 24,
3471–3478.
1
ing at 400.01 and 161.98 MHz, respectively. H NMR spectra

Article
Chem. Mater., Vol. 22, No. 18, 2010 5387
were measured with solutions of ligands in deuterated solvents.
The kinetics of paraoxon degradation were determined using 31
the magnetization of the particles in an applied magnetic field.
All SQUID measurements were performed at 300 K over a
0 to þ50 kOe range on dried particles weighing 5-7 mg.
ζ-Potential Measurements. All ζ-potential measurements
were performed using a Brookhaven ZetaPALS ζ-potential
analyzer (Brookhaven Instruments Corporation). The Smolu-
chowski equation was used to calculate the ζ-potential from the
electrophoretic mobility. The reported ζ-potential values are an
average of 5 measurements, each of which was obtained more
than 25 electrode cycles. The particles were dispersed in 5 mM
KCl aqueous solution at approximately 0.05 wt % concentra-
tion and the pH of the nanoparticle suspensions was adjusted by
adding 1 M HCl or NaOH aqueous solutions.
P
NMR. Spectra (1500 scans for cobalt chloride solutions and
64 scans for all other samples) were recorded using an 85%
phosphoric acid solution in D2O as an external reference
(0 ppm). The reaction milieu consisted of 50 mM HEPES buffer
in D2O to keep the solution pH constant at 9. The reaction time
was taken to be the midpoint of the acquisition period. For
kinetic measurements with nonmagnetic materials, samples of
2-3 mg/mL of ligand or silver nanoparticle-based suspension in
HEPES/D2O solution (50 mM, pH 9.0) and paraoxon solution
in acetonitrile (50 mM) were prepared. Basic pH accelerated the
spontaneous hydrolysis of paraoxon to acceptable levels versus
neutral pH, and buffering allowed for maintenance of constant
reaction rate up to high degrees of conversion. Acetonitrile was
added to improve the miscibility of paraoxon (aqueous solubi-
lity at 20 °C, up to 3.6 mg/mL)60 and aqueous buffer as well as to
act as a convenient reactant diluent that did not trigger the
reaction prior tocontactwiththe aqueous phase. Att = 0, 630 μL
of ligand or nonmagnetic particle suspension and 70 μL of
paraoxon solution were mixed by vortexing and placed into an
NMR tube for kinetic measurements. Paramagnetic cobalt NP-
based suspensions (3 mg/mL) in HEPES/D2O buffer (50 mM,
pH 9.0) were prepared as a 5 mL stock sample, which was mixed
with 556 μL of paraoxon in acetonitrile at t = 0, resulting in the
initial paraoxon concentration of 5 mM. The final suspension
was kept at room temperature while shaking. At given time
intervals, the suspension was centrifuged (6,500 g, 5 min), and
the top clear solution (0.7 mL) was carefully withdrawn by a
pipet and placed into an NMR tube. The solution was placed
back into the suspension immediately after each NMR measure-
ment. The degree of paraoxon conversion was expressed as9
Dynamic Light Scattering (DLS). Dynamic light scattering
(DLS) experiments were performed with a Brookhaven BI-
200SM light scattering system (Brookhaven Instruments
Corporation) at a measurement angle of 90°. Number-average
particle size distributions were obtained using the built-in soft-
ware and the reported particle mean hydrodynamic diameters
are the average of five measurements, each conducted for 3 min.
All samples dispersed in aqueous 10 mM KCl (pH adjusted by 1
M NaOH or HCl) were filtered with a 0.45 μm syringe filter prior
to the DLS tests.
Molecular weight and other structural parameters of ligands
and nanoparticles were calculated using ChemBioDraw Ultra
version 12.0 software (CambridgeSoft Co.).
Results and Discussion
Ligand and Particle Syntheses and Characterization. As
stated in the Introduction, we wished to develop a straight-
forward synthetic route toward fatty acids that are modi-
fied by a reactive moiety capable of chelating with
transition metal ions and accelerating the hydrolysis of
organophosphorous esters. Neither 9-((1H-imidazol-2-yl)-
thio)octadecanoic acid (1) nor 9-((1H-imidazole-5-carbo-
nothioyl) thio)-8-hydroxyoctadecanoic acid (2) and their
syntheses via the thiol-ene reaction of oleic acid and
2-mercaptoimidazole (Figure 1), and trans-9,10-epoxys-
tearic and 4(5)-imidazoledithiocarboxylic acids (Figure 2),
respectively, have been previously reported. The one-step
synthetic routes described herein are facile and produce
acceptable yields, limited only by the product purification
processes, as is common for fine fatty acid derivatives.
The targeted fatty acid derivatives 1 and 2 are multi-
dentate ligands with thiol, amine, and carboxylic groups
capable of complexation with metal ions. Thiol sulfur,
dithiocarboxylic moieties and imidazole nitrogens pos-
sess inherently stronger coordination and chelation cap-
ability with metals than do carboxyl groups.62-65 There-
fore, in order to maintain the NP complexation by fatty
acids through metal-carboxyl binding, we synthesized
oleic acid- or epoxystearic acid-modified particles, which
were subsequently modified by 2-mercaptoimidazole or
(4)5-imidazoledithiocarboxylic acid (Figures 1 and 2). If
Ft ¼ ΣIp=ðΣIr þ ΣIpÞ
ð1Þ
where ΣIr and ΣIp are the sums of the integrations of the signals
corresponding to the reactant (paraoxon, -6.11 ppm) and the
products (diethyl phosphate, 1.22 ppm, and diethyl phosphate-
HEPES adduct61), respectively (S7). The observed rate con-
stant, kobs, is found from the slope of the ln(1 - Ft) vs t plot:
lnð1 - FtÞ ¼ - kobs
t
ð2Þ
Transmission Electron Microscopy (TEM). TEM was per-
formed on a JEOL 200-CX transmission electron microscope.
Samples were prepared by placing drops of the nanoparticle
dispersion on lacey carbon-coated 200 Mesh copper grids
(Structure Probe, Inc.) and imaged at an accelerating voltage
of 200 kV.
Fourier Transform Infrared Spectroscopy (FTIR). FTIR spec-
troscopy was performed on a NEXUS 870 FTIR spectrometer
(Thermo Nicolet Inc.). Spectra were recorded over the wave-
number range between 4000 and 400 cm-1 at a resolution of 2
cm-1 and are reported as the average of 64 spectral scans. All
samples were dried under a vacuum to constant weight, ground
and blended with KBr, and pressed to form the pellets used in
the measurements.
Superconducting Quantum Interference Device (SQUID) ex-
periments were conducted using a magnetic property measure-
ment system model MPMS-5S (Quantum Design) to determine
(62) Ceretti Mazza, M. T.; De Cicco, L.; De Rosa, G.; Caramazza, R. J.
Biol. Res.-Boll. Soc. It. Biol. Sper. 1996, 72, 79–86.
(63) Sakurai, H. Transition Met. Chem. 1977, 2, 103–106.
(64) Ahrland, S.; Chatt, J.; Davies, N. R. Q. Rev. Chem. Soc. 1958, 12,
265–276.
(60) Bowman, B. T.; Sans, W. W. J. Environ. Sci. Health 1979, B14, 625–
634.
€
(61) Gaba, J.; John, H.; Melzer, M.; Blum, M.-M. J. Chromatogt., B
(65) Watt, J.; Cheong, S.; Toney, M. F.; Ingham, B.; Cookson, J.;
Bishop, P. T.; Tilley, R. D. ACS Nano 2010, 4, 396–402.
2010, in press.

5388 Chem. Mater., Vol. 22, No. 18, 2010
Bromberg et al.
Figure 3. Schematic of nanoparticle synthesis by metal ion reduction
followed by complexation with fatty acids.
we were to bind multidentate compounds 1 and 2 to
cobalt or silver NP directly, the metal-thiol or metal-
imidazole chelation would have been more likely than
metal-carboxyl. The synthetic routes toward fatty-acid-
chelated NP in the present study are shown in Figure 3.
Notably, in our synthetic route, the nanoparticles were
first prepared by reduction of the corresponding cobalt or
silver salts by sodium borohydride, purified, and then
complexed with fatty acids. Reduction of silver and
cobalt salts by sodium borohydride in the presence of
oxygen is well-known55,65-67
AgNO3 þ NaBH4 f Ag þ 0:5H2 þ 0:5B2H6 þ NaNO3
2CoCl2 þ 4NaBH4 þ 9H2O f Co2B þ 4NaCl þ 12:5H2
þ 3BðOHÞ3
4Co2B þ 3O2 f 8Co þ 2B2O3
Figure 4. (a) TEM image ofCo nanoparticlessynthesizedbyreduction of
cobalt(II) chloride hexahydrate by sodium borohydride in the presence of
oleic acid only. (b) TEM image of Co(0) nanoparticles synthesized by
reduction of cobalt(II) chloride hexahydrate by sodium borohydride in
the presence of 2-mercaptoimidazole and oleic acid. For synthesis details,
see Experimental Section.
In the case of Co(II), when the mixture is exposed to
oxygen, a sacrificial reaction takes place whereby boron is
oxidized while cobalt is reduced, resulting in the conver-
sion of Co2B to Co(metal).67
Interestingly, the structures of the Co NPs prepared
with and without 2-mercaptoimidazole appeared to be
quite different (Figure.4). Particles prepared with oleic
acid as the only capping agent aggregated into ∼200 nm
clusters composed of ∼10 nm spherical primary particles
(Figure.4a), whereas the particles prepared in the pre-
sence of both 2-mercaptoimidazole and oleic acid were
cubical or octahedral in shape, with one side of the cube
sized ∼50 nm (Figure.4b). Although the formation of
cubical particles in the presence of both mercaptoimida-
zole and oleic acid is a novel observation of the present
study, it has been reported previously that simultaneous
presence of several agents strongly binding to the nano-
crystal surface such as oleic acid, trioctylphosphine oxide,
and others affects the shape of the cobalt crystals.56
It has been argued68 that the reduction of silver sulfate
by borohydride directly in the presence of oleic acid
results in the binding of the silver nanoparticle surface
to the oleic acid through the double bond. We believe that
in our synthesis (Figure 3), because the reduction was
completed prior to the contact of fatty acids with the
resulting NP, the adsorption of the fatty acids on the Ag
or Co surface is through the carboxylate groups. Small
angle neutron scattering, magnetization analysis, and
other techniques have been previously used to demon-
strate that metal or metal oxide nanoparticles modified
with excess monocarboxylic acids form double layers of
the acids on the particle surface in water, which helps to
stabilize the particles colloidally.69-73 The ζ-potential
measured in 5 mM KCl at pH 7.0 was -22 ( 3 mV
(68) Bala, T.; Swami, A.; Prasad, B. L. V.; Sastry, M. J. Colloid Interface
Sci. 2005, 283, 422–431.
(69) Shen, L.; Laibinis, P. E.; Hatton, T. A. Langmuir 1999, 15, 447–453.
(70) Shen, L.; Stachowiak, A.; Fateen, S.-E. K.; Laibinis, P. E.; Hatton,
T. A. Langmuir 2001, 17(2), 288–299.
(66) Glavee, G. N.; Klabunde, K. J.; Sorensen, C. M.; Hadjipanayis,
G. C. Langmuir 1993, 9, 162–169.
(67) Glavee, G. N.; Klabunde, K. J.; Sorensen, C. M.; Hadjipanayis,
G. C. Langmuir 1992, 8, 771–773.

Article
Chem. Mater., Vol. 22, No. 18, 2010 5389
Figure 6. Representative kinetics of paraoxon decomposition, both
spontaneous and catalyzed by silver NP-(1) and the same nanoparticles
complexed with Co2þ
.
these analyses, we estimate that the average NP modified
with conjugate 1 contained approximately one molecule
of 1 per 25 silver and 27 cobalt atoms, respectively,
whereas NP modified with conjugate 2 contained ap-
proximately one molecule of 2 per 18 silver and 27 cobalt
atoms, respectively. On the basis of the elemental analysis,
and assuming the average diameter of a single particle to
be 10 nm, we estimate that the area occupied by a single
ligand on Ag and Co nanoparticle surface in our experi-
ments is in the range 0.08-0.15 nm2. For comparison, the
mean molecular area occupied by a single oleic acid mole-
cule is about 0.42nm2/molecule.75,76 Given that the ligands
1 and 2 are significantly larger than the oleic acid molecule
because of the attachment of the imidazole moieties, we
can conclude that the ligands were attached to the NPs in
multilayers. The calculations were performed assuming
conjugate structures on the NP surface to be of those
depicted in Figures 1 and 2.
Ligand and Nanoparticle Performance in Organopho-
sphate Hydrolysis. We studied the performance of the
ligands attached to the nanoparticles in the hydrolysis of
the organophosphate insecticide, paraoxon. The kinetics
of the insecticide hydrolysis was studied in aqueous
solutions containing 10% acetonitrile. The pH was kept
constant at 9.0 throughout the study. The basic pH
accelerated the spontaneous hydrolysis of paraoxon to
acceptable levels versus neutral pH, and buffering al-
lowed for maintenance of constant reaction rate up to
high degrees of conversion. The reaction was followed by
31P NMR as described in the Experimental Section and
typical kinetics data are shown in Figure 6.
Figure 5. FTIR spectra of cobalt NP-(1), silver NP-(2), and their repre-
sentative precursors: 2-mercaptoimidazole, 4(5)-imidazoledithiocar-
boxylic acid, and trans-9,10- epoxystearic acid.
(silver-epoxystearic acid NP) and -28 ( 2 mV (cobalt-
oleic acid NP), indicating that the particles are stabilized
by the negative charge of the carboxylate groups exposed
to the outer layer of the water-bilayer interface.
Modification of particles capped by oleic or epoxys-
tearic acid by the thiol-ene reaction with mercaptoimi-
dazole (Figure.1) or condensation with imidazoledithio-
carboxylic acid (Figure.2), respectively, was evident from
the spectroscopic data (Figure.5). Thus, FTIR spectra of
the silver and cobalt NP-(1) and NP-(2) featured two
strong bands at about 2920 and 2850 cm-l, typical of the
antisymmetric and symmetric CH stretching vibrations,
respectively, because of the fatty acid components.74 The
spectra also showed bands characteristic of vibrations of
the imidazole ring (amidine bands around 1640-1650
cm-1), and C-C-S stretching bands at 686 cm-1 (cobalt
NP-(1)). The strong band characteristic of the free car-
boxylate group at 1710 cm-1 in the fatty acid as well as
free conjugates 1 and 2 (not shown) was present only as a
shoulder of a broader signal centered in the 1640 cm-1
area in the NP spectra. Overall, the FTIR spectra support
the structure wherein imidazole-containing conjugates
are adsorbed onto the NP metal surfaces via carboxy-
late-metal complexation.
When the kinetics results were expressed in terms of
eq 2, the curve fits were linear (R2 > 0.98 in all cases),
indicating a pseudo-first-order reaction order. The ob-
served rate constants afforded calculation of the reaction
half-life (τ1/2 = ln(2)/kobs) and the apparent second-order
rate constant k00 = kobs/Ccat, where Ccat is the initial
effective molar concentration of the catalytic groups. The
results of all kinetic measurements are collected in Table 1.
As is seen, weakly nucleophilic 2-mercaptoimidazole
Further quantitative information on the ratio of metal
to organic ligands comes from the TGA and elemental
analysis (see the Experimental Section). On the basis of
(71) Avdeev, M. V.; Aksenov, V. L.; Balasoiu, M.; Garamus, V. M.;
Schreyer, A.; Toeroek, Gy.; Rosta, L.; Bica, D.; Vekas, L. J. Colloid
Interface Sci. 2006, 295, 100–107.
(72) Yanga, K.; Penga, H.; Wen, Y.; Li, N. Appl. Surf. Sci. 2010, 256,
3093–3097.
ꢀ ꢀ
(73) Avdeev, M. V.; Mucha, B.; Lamszus, K.; Vekas, L.; Garamus,
(75) Lundgren, S. M.; Persson, K.; Mueller, G.; Kronberg, B.; Clarke,
J.; Chtaib, M.; Claesson, P. M. Langmuir 2007, 23, 10598–10602.
(76) Pichon, B. P.; Demortiere, A.; Pauly, M.; Mougin, K.; Derory, A.;
Begin-Colin, S. J. Phys. Chem. C 2010, 114, 9041–9048.
V. M.; Feoktystov, A. V.; Marinica, O.; Turcu, R.; Willumeit, R.
Langmuir 2010, 26, 8503–8509.
(74) Kimura, F.; Umemura, J.; Takenaka, T. Langmuir 1986, 2, 96–101.

5390 Chem. Mater., Vol. 22, No. 18, 2010
Bromberg et al.
Table 1. Half-Life (τ1/2) and Apparent Second-Order Rate Constant (k00)
of Paraoxon Hydrolysis in the Presence of Ligands and Nanoparticles of
the Present Study at pH 9.0
ligands and cobalt ions was observed with nanoparticles
modified with conjugates 1 and 2, where ligand comple-
xation with Co2þ resulted in 10- to almost 50-fold hydro-
lysis acceleration. Mechanistically, such a phenomenon
half-life,
τ1/2
(days)
second-order
rate constant,
can be explained49 by the ability of the Co2þ or Co2þ
-
systema
k00 (ꢀ 104 M-1 s-1 b
)
imidazole ligand to coordinate the oxonate oxygen, with-
drawing electron density from the phosphorus atom and
generating a more reactive electrophile capable of effi-
cient hydrolysis even by moderately nucleophilic ligands
such as imidazole. Schematics of both imidazole ligand-
and ligand-Co2þ mechanisms of hydrolysis are presented
in Figure 7. Similarly, acceleration of the paraoxon hydro-
lysis has been observed with vinylimidazole-containing
spontaneous hydrolysis
2-mercaptoimidazole
stoichiometric complex
2-mercaptoimidazole/Co2þ
4(5)-imidazoledithiocarboxylic acid
stoichiometric complex
4(5)-imidazoledithiocarboxylic
acid/Co2þ
>140
35
2.6
1.2
16
11
6.6
3.6
6.1
silver NP-(1)
silver NP-(2)
20
24
1.9
4.9
1.5
53
polymers chelated (“molecularly imprinted”) with Co2þ 77
.
stoichiometric complex
Silver NP-(1)/Co2þ
Co2þ
An analogy between such an active imidazole-Co2þ com-
plex and the active site of phosphotriesterase with Co2þ
ions coordinated to histidine78,79 can be seen. To the best
of our knowledge, ours is the first encounter of such a
“biomimetic”catalytic activityofnanoparticlescomplexed
with metal ions.
21
5.8
32
0.30
2.1
0.38
18
cobalt NP-(1)c
cobalt NP-(2)
stoichiometric complex
Cobalt NP-(2)/Co2þ
2.3
a Each system comprised 10 vol % acetonitrile solution in 50 mM
HEPES/D2O (pD 9.0). bApparent second-order rate constants were
calculated from k00 = ro/CsCcat, where ro = kobsCs is the initial reaction
rate, Cs and Ccat are initial effective concentrations of the substrate
(paraoxon) and catalyst, respectively. The Ccat values were calculated as
effective molar concentrations of the catalytic groups. c Synthesized in
the presence of 2-mercaptoimidazole and oleic acid (see Experimental
Section).
Conclusions
For the first time, we described a facile synthesis of
imidazole-containing ligands that are conjugated to fatty
acid chains via a thiol-ene reaction between oleic acid
and 2-mercaptoimidazole and a condensation between
(4)5-imidazoledithiocarboxylic and 9,10-epoxystearic
acids. To obtain the ligands bound to the metal surface
by the carboxyl groups, we first synthesized NPs of Ag(0)
and Co(0) by conventional reduction of corresponding
salts by sodium borohydride. Then the adsorbed oleic or
9,10-epoxystearic acids were further modified by 2-mer-
captoimidazole and (4)5-imidazoledithiocarboxylic acid,
respectively. As a variation of the “fatty acid adsorption
first” synthesis, we were able to obtain Co(0)-based NPs
by reduction of Co2þ in the presence of both the fatty acid
and 2-mercaptoimidazole, which unexpectedly resulted in
the NPs of cubical or octahedral shape, as opposed to
spherical particles formed in the presence of oleic acid
only. The thio- and dithio- imidazole ligands and NPs
demonstrated significant, albeit moderate activity in cat-
alyzing esterolysis of paraoxon, but their complexes with
Co2þ were 10- to 50-fold more reactive. The uncovered
synergistic effect of the Co2þ (but not Co(0) in the
nanoparticle core) and imidazole derivatives mimics that
of the metal-histidine moieties in the active site of en-
zymes in that the metal-ligand complex is far more
reactive than the ligand itself. The described NPs were
readily recovered from the reaction medium by filtering
or centrifugation and in that sense were advantageous as
catalysts over their organic ligands without the metal
core. In addition, the metal cores resulted in intense
coloring of the particles, which can be exploited for
particle detection in our ongoing studies.
Figure 7. Schematic of paraoxon hydrolysis acceleration by nucleophile
such as imidazole and its derivatives and nucleophile-Co2þ complex.
afforded only approximately 4-fold faster paraoxon de-
gradation compared to the spontaneous hydrolysis, but
4(5)-imidazoledithiocarboxylic acid was approximately
12-fold more active because of its higher nucleophilicity.
Cobalt chloride solution per se was not particularly re-
active. However, stoichiometric complexation of either of
2-mercaptoimidazole or 4(5)-imidazoledithiocarboxylic
acid with cobalt ions resulted in a significant, up to 13-fold
acceleration of the paraoxon hydrolysis. The same syner-
gistic effect of the presence of both imidazole-containing
(77) Yamazaki, T.; Yilmaz, E.; Mosbach, K.; Sode, K. Anal. Chim. Acta
2001, 435, 209–214.
(78) Omburo, G. A.; Kuo, J. M.; Mullins, L. S.; Raushel, F. M. J. Biol.
Chem. 1992, 267, 13278–13283.
(79) Kuo, S.J. M.; Raushel, F. M. Biochemistry 1994, 33, 4265–4272.

Article
Acknowledgment. This work was sponsored by the Defense
Threat Reduction Agency grant HDTR1-09-1-0012 and
supported via the Department of the Army, U.S. Army Research
Office, under Grant W911NF-07-1-0139. Any opinions,
findings, conclusions, and recommendations expressed in
this paper are those of the authors and do not necessarily
reflect theviewof the U.S. ArmyResearch Office. The authors
Chem. Mater., Vol. 22, No. 18, 2010 5391
are grateful to Dr. Takuya Harada for help with TEM and
SQUID measurements.
Supporting Information Available: 1H NMR spectra of func-
tional ligands, TEM images of modified Ag and Co nanoparticles,
SQUID particle characterization, and 31P NMR of paraoxon
degradation (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.