Transformation of gold(I)-cyclo[-Met-Met-] complex supramolecular fibers into aligned gold nanoparticles

Article abstract of DOI:10.1246/cl.130106

A cyclic dipeptide cyclo[L-methionylL- methionyl] and gold(I) cation were found to form a complex in a 1:2 ratio. Pouring a DMF solution of the complex into ethyl acetate afforded fibers with lengths of more than 20 μm and 100400nm in width, which consisted of fibrous nanostructures 50100nm in diameter. Treatment of this supramolecular fiber with catechol gave linearly aligned Au nanoparticles on the original fibers.

Full text of DOI:10.1246/cl.130106

doi:10.1246/cl.130106
Published on the web May 15, 2013
601
Transformation of Gold(I)-cyclo[-Met-Met-] Complex Supramolecular Fibers
into Aligned Gold Nanoparticles
Masahiro Furutani and Kazuaki Kudo*
Institute of Industrial Science, The University of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo 153-8505
(Received February 9, 2013; CL-130106; E-mail: kkudo@iis.u-tokyo.ac.jp)
AuCl
A cyclic dipeptide cyclo[-L-methionyl-L-methionyl-] and
O
O
O
gold(I) cation were found to form a complex in a 1:2 ratio.
Pouring a DMF solution of the complex into ethyl acetate
afforded fibers with lengths of more than 20 ¯m and 100-400 nm
in width, which consisted of fibrous nanostructures 50-100 nm
in diameter. Treatment of this supramolecular fiber with catechol
gave linearly aligned Au nanoparticles on the original fibers.
S
S
NH
NH
HN
HN
S
S
O
AuCl
1
2 (proposed structure)
Figure 1. Chemical structure of compounds used in this study.
Supramolecular polymers are one-dimensionally (1D)
aligned small molecules, formed through noncovalent intermo-
lecular interactions such as hydrogen bonding, electrostatic
interactions, and ³-³ interactions. Supramolecular polymers are
interesting soft materials because of their flexibility, reversibil-
ity, and directionality.1 They are known to form nanostructures
such as fibers, tubes, or wires, which can behave as scaffolds
or templates for nanoparticles (NPs). This unique feature of
supramolecular polymers is of significance because 1D-aligned
metal NPs, especially Ag and Au NPs, can be applied in optical
and electronic devices where the particle size, shape, and
aggregation state of the arrays influence their physical proper-
ties.2
Two main methods have appeared for the introduction of
metallic components to fabricate 1D-aligned precious metal
NPs. One is the deposition of the NPs on supramolecular
polymer scaffolds,3-6 and the other involves the fixation of metal
cations to supramolecular polymer templates and subsequent
electroless-plating-like reduction in the presence of externally
added metal cations to form NPs.7-12 In the former case, because
NPs are generally not stable and tend to aggregate, they must
either be used immediately after preparation or be chemically
modified by organic molecules. Solutions of noble cations used
in the latter case are also unstable, and the complexation
conditions need to be controlled carefully.
On the other hand, if the supramolecular polymer can be
prepared from metal-ion-containing monomers, incorporation of
the metal species in every building unit is ensured, and this
would make the formation of 1D-aligned metal cations easier.
Concerning this, Nishi et al. reported that nanowire-shaped
crystals of silver(I) phenylacetylide could be converted into 1D-
aligned Ag NP arrays via photoreduction of this organosilver
species.13 In this example, the organometallic monomer was
insoluble, and potentially pyrophoric PMe3 was needed as a
temporal ligand to dissolve the monomers. Lecante et al.
reported the formation of quasi-1D-aligned Au NPs from rod-
like crystals of AuCl-n-octylamine complexes via autoreduction
of Au(I) species in air.14 The transformation is spontaneous and
not controllable. Moreover, in both these cases, the construction
of 1D-aligned NPs was based on the nanorod-like crystal
structure of the metal-containing species, the formation of which
is hardly predictable. In this letter, we report the formation of
1D-aligned Au NPs from a supramolecular polymer by utilizing
the known ability of cyclic dipeptides to make supramolecular
polymers and the Au-S interaction. Sulfides are known as
temporal ligands for the Au+ cation,15 but there are no examples
in which sulfide moieties are utilized as binding sites in
supramolecular fibers. Cyclic dipeptides are known to form a
tape or layer supramolecular structure owing to the multiple
intermolecular hydrogen bonds between rigid diketopiperazine
rings;16 however, they have not been utilized as building blocks
for functional materials.17 Thus, we designed cyclo[-L-methio-
nyl-L-methionyl-] (cyclo[-Met-Met-]) 1 (Figure 1), which is
considered to interact with Au(I) cations and to be transformable
into a 1D-aligned structure.
Cyclo[-Met-Met-] 1 was synthesized in three steps starting
from N-tert-butoxycarbonyl-L-methionine, through carboxy
group activation, amino group deprotection, and cyclization.18
The Au+ stock solution was prepared through the reduction of
Na[AuCl4] with 2,2¤-thiodiethanol (TDE) in methanol according
to the method described in the literature.19 Addition of the Au+
solution into a suspension of 1 in chloroform or methanol
followed by stirring for 25 h at room temperature in the dark
afforded a white solid that was soluble in dimethylformamide
(DMF) and dimethyl sulfoxide (DMSO). The 1H NMR spectrum
of a DMSO-d6 solution of this solid showed peaks similar to
those of 1 but shifted downfield in the range of less than 0.1 ppm
for the diketopiperazine rings and 0.12-0.39 ppm for the side
chains of the Met residues (see Supporting Information).18,20
Elemental analysis of the precipitate indicated that it consists
of AuCl and the ligand 1 in 2:1 stoichiometry.21 Furthermore,
the positive-ion fast-atom bombardment mass spectrum of the
precipitate showed a peak with m/z = 329.1, which could
correspond to the monocation of the digold complex of 1. From
these observations, it is most likely that the precipitate is a
2Au+-1 complex, of which the two sulfide moieties on the side
chains of 1 interact evenly with the Au+ ions; hereafter, we
denote this compound as complex 2. Next, thermogravimetric
analysis was performed (Figure 2). Compound 1 showed a
residual weight of 13% at 500 °C. On the basis of this value, the
expected residual weights for 1:1 and 1:2 complexes of 1 and
AuCl are calculated to be 47% and 59%, respectively. The
Chem. Lett. 2013, 42, 601-603
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

602
100
80
a)
2
1
60
40
20
0
200 nm
500
0
200 300
Temperature / °C
100
400
b)
Figure 2. TGA thermograms of 1 and 2. N2 flow (50
¹1
mL min¹1), 20 °C min
.
experimental value for 2 was 66%. Although this is not exactly
the same as any of the calculated values, the 1:2 stoichiometry is
considered to be more plausible. The deviation might be due to
the effect of coexisting Au species on the thermal decomposition
behavior of 1.22
200 nm
30 nm
0
The attempted preparation of Au+ complexes with other
cyclic dipeptides, cyclo[-Nle-Nle-] or cyclo[-His-His-] (Nle:
norleucine; His: histidine), was not successful. In the case of
cyclo[-Nle-Nle-], the chloroform solution was transparent.
After mixing the Au+ methanol solution, the mixture became
turbid immediately. A small amount of brown solid was
precipitated after 14 h. In the case of cyclo[-His-His-], the
methanol suspension became transparent after mixing. Stirring
for 18 h resulted in gold deposition on the wall. It seemed that
Au+ cations stabilized by TDE aggregated without continuous
interaction with the cyclic dipeptides. These results indicated
that the interaction of Au+ cations with the S atoms of the sulfide
moieties of 1 was important for the preparation of a stable
complex.
0
100
200 nm
Figure 3. (a) SEM micrograph of the fibers of 2. (b) AFM
micrograph of the fibers of 2. The profile shows that the height
and the width are 30 and 100 nm, respectively.
× 4
2
2
1
When 20 ¯L of a DMF solution of complex 2 (28 mM) was
poured into 80 ¯L of ethyl acetate,23 a white precipitate was
formed. Scanning electron microscopic observation revealed that
the precipitate consisted of fibers more than 20 ¯m long and
100-400 nm in width (Figure 3a).24 In the atomic force micro-
scopic analysis, thinner fibrous nanostructures of 50-100 nm in
width were also observed (Figure 3b). The use of methanol or
chloroform as a poor solvent instead of ethyl acetate resulted in
the formation of nonfibrous aggregates of 2 (Figure S1).18 In situ
formation of supramolecular fibers was also attempted with the
cyclic dipeptide 1, Au+ methanol stock solution, and ethyl
acetate. However, long fibers could not be formed in the case of
the DMF-ethyl acetate system, which indicated that temporal
dissolution of complex 2 toward DMF was important.
1
3500
3000 1700
Wavenumber / cm-1
1600
Figure 4. FT-IR spectra of the fibers of 1 and 2.
were different for 1 and 2, which might reflect a difference in
the preference of the two types of tape structures. The N-H
stretching regions of the spectra (Figure 4, left) were similar for
1 and 2, which again indicates the resemblance of the hydrogen-
bonding state between these two species.
Then, the supramolecular fibers of 2 were treated with
catechol.27 Catechol is a mild reductant with moderate hydro-
phobicity, and its oxidized compound, quinone, is known to
have a weak binding affinity for metal surfaces.28 When the
fibrous nanostructures were immersed into an ethyl acetate
solution of 30 mM catechol, small NPs of 10-20 nm in diameter
were formed on their surfaces. After 4 h, all the fibers were
finally transformed into the aligned NPs (Figure 5). X-ray
photoelectron spectroscopic analysis proved that the NPs were
composed of Au0 and that the sulfide (thioether) moieties were
not oxidized to sulfoxides during the conversion of Au+ to Au0
(Figure S3).18,29,30 A qualitative assay with energy-dispersive
spectroscopy showed the presence of all elements other than Cl
FT-IR analysis was performed to obtain structural informa-
tion on the fibers of 2. In the amide I region of the spectrum,
¹1
two peaks were observed at 1680 and 1651 cm (Figure 4,
right).17b,17f,25 The corresponding peaks of the fibers of 1
appeared at 1678 and 1653 cm¹1. This indicates that complex-
ation of 1 with AuCl did not significantly affect the mode of
hydrogen bonding (Figure S2).18 It is interesting to note that a
¹1
crystal of 1 gives only one amide I peak at 1675 cm due to a
single mode of hydrogen bonding.26 The fact that the fibers of 1
and 2 afforded two amide I peaks should imply that there are two
kinds of 1D hydrogen-bonded tapes of these diketopiperazines
in the fibrous state. The intensity ratios of the two amide I peaks
Chem. Lett. 2013, 42, 601-603
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

603
9
a) H.-G. Liu, F. Xiao, C.-W. Wang, Q. Xue, X. Chen, Y.-I. Lee, J.
Hao, J. Jiang, J. Colloid Interface Sci. 2007, 314, 297. b) T. Ogoshi,
S. Hiramitsu, T.-a. Yamagishi, Y. Nakamoto, Macromolecules 2009,
42, 3042. c) R. Chhabra, J. G. Moralez, J. Raez, T. Yamazaki, J.-Y.
Cho, A. J. Myles, A. Kovalenko, H. Fenniri, J. Am. Chem. Soc.
2010, 132, 32.
10 I. Díez, H. Hahn, O. Ikkala, H. G. Börner, R. H. A. Ras, Soft Matter
2010, 6, 3160.
11 P. Koley, A. Pramanik, Adv. Funct. Mater. 2011, 21, 4126.
12 Y. Song, S. R. Challa, C. J. Medforth, Y. Qiu, R. K. Watt, D. Peña,
J. E. Miller, F. van Swol, J. A. Shelnutt, Chem. Commun. 2004, 1044.
13 J. Nishijo, O. Oishi, K. Judai, N. Nishi, Chem. Mater. 2007, 19,
4627.
200 nm
14 S. Gomez, K. Philippot, V. Collière, B. Chaudret, F. Senocq, P.
Lecante, Chem. Commun. 2000, 1945.
15 a) P. G. Jones, J. Lautner, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun. 1988, 44, 2089. b) D. B. Pedersen, S. Duncan, J. Phys.
Chem. A 2005, 109, 11172.
16 J. C. MacDonald, G. M. Whitesides, Chem. Rev. 1994, 94, 2383.
17 a) Y. Ishida, T. Aida, J. Am. Chem. Soc. 2002, 124, 14017. b) K. B.
Joshi, S. Verma, Tetrahedron Lett. 2008, 49, 4231. c) T. Govindaraju,
Supramol. Chem. 2011, 23, 759. d) M. Jaworska, A. Jeziorna, E.
Drabik, M. J. Potrzebowski, J. Phys. Chem. C 2012, 116, 12330. e)
S. Manchineella, T. Govindaraju, RSC Adv. 2012, 2, 5539. f) J.-A.
Kim, Y.-H. Jeong, W.-D. Jang, Chem.®Eur. J. 2010, 16, 13955.
18 Supporting Information is available electronically on the CSJ-
Journal Web site, http://www.csj.jp/journals/chem-lett/index.html.
19 T. Yang, L. Campbell, D. J. Dixon, J. Am. Chem. Soc. 2007, 129,
12070.
20 This result agreed with a previous report: R. Ettorre, V. Guantieri,
A. M. Tamburro, L. De Nardo, Inorg. Chim. Acta 1979, 32, L39.
21 Elemental analysis: Calcd for two-to-one (Au+/1 = 2/1) complex,
C10H18O2N2S2Au2Cl2: C, 16.51; H, 2.49; N, 3.85%; Calcd for one-
to-one (Au+/1 = 1/1) complex, C10H18O2N2S2AuCl: C, 24.27; H,
3.67; N, 5.66%; (found) C, 16.76; H, 2.48; N, 3.61%.
200 nm
Figure 5. SEM micrographs of a reduced sample of the fibers
of 2 by 30 mM of catechol ethyl acetate solution for 4 h.
after the reduction, indicating that the Au+ cations were reduced
successfully by the catechol (Figure S4).18,31 Concerning the
mechanism of the Au NP formation, the UV spectral change of
the catechol solution phase revealed that oxidation of catechol
occurred within 2 h (Figure S5).18 At the same time, small Au
NPs began to appear on the fibers, and the number and size of
Au NPs increased gradually with concomitant thinning of the
fibers (Figure S6).18
In conclusion, the interaction between a Au+ cation and
a bissulfide, cyclo[-Met-Met-] 1, led to the selective and
quantitative placement of Au+ cations on the surface. The
complex as a simple monomeric unit successfully formed a
supramolecular nanofiber. Subsequently, Au+ cations were
readily reduced to Au0 atoms by catechol, and the fiber-like
template was transformed into aligned NPs.
22 The first weight loss for the complex 2 at around 200 °C might be
due to redox reaction of Au complex to give atomic Au and
molecular or atomic chlorine which then reacted with the sulfide
moiety of the ligand, as analogized from the thermal decomposition
of AuCl-amine complex: X. Lu, H.-Y. Tuan, B. A. Korgel, Y. Xia,
Chem.®Eur. J. 2008, 14, 1584.
23 The complex 2 was only soluble with aprotic polar solvents, such as
DMF and DMSO, and was insoluble with water, alcohols, acetone,
chloroform, and ethyl acetate, which was worse solubility than the
case of 1. Interestingly, when 2 was immersed into water at room
temperature for a few days in dark, the color changed gradually from
white into red.
References and Notes
1
L. Brunsveld, B. J. B. Folmer, E. W. Meijer, R. P. Sijbesma, Chem.
Rev. 2001, 101, 4071.
24 The white solid of 2 in the methanol-chloroform suspension
mentioned above was found to be an aggregate of rod-like
nanostructures, observed by scanning electron microscope.
25 G. T. R. Palmore, T.-J. M. Luo, M. T. McBride-Wieser, E. A.
Picciotto, C. M. Reynoso-Paz, Chem. Mater. 1999, 11, 3315.
26 A. P. Mendham, B. S. Potter, R. A. Palmer, T. J. Dines, J. C.
Mitchell, R. Withnall, B. Z. Chowdhry, J. Raman Spectrosc. 2010,
41, 148.
27 a) Y. Lee, S. H. Lee, J. S. Kim, A. Maruyama, X. Chen, T. G. Park,
J. Controlled Release 2011, 155, 3. b) Y. Lee, T. G. Park, Langmuir
2011, 27, 2965. c) K. C. L. Black, Z. Liu, P. B. Messersmith, Chem.
Mater. 2011, 23, 1130. d) M. Raula, D. Maity, Md. H. Rashid, T. K.
Mandal, J. Mater. Chem. 2012, 22, 18335.
2
a) Z. Nie, A. Petukhova, E. Kumacheva, Nat. Nanotechnol. 2010, 5,
15. b) H. Nakao, Anal. Sci. 2009, 25, 1387. c) S. Srivastava, N. A.
Kotov, Soft Matter 2009, 5, 1146. d) Z. Tang, N. A. Kotov, Adv.
Mater. 2005, 17, 951.
a) M. Endo, Y. Yang, T. Emura, K. Hidaka, H. Sugiyama, Chem.
Commun. 2011, 47, 10743. b) M. G. Warner, J. E. Hutchison, Nat.
Mater. 2003, 2, 272. c) H. Nakao, H. Shiigi, Y. Yamamoto, S.
Tokonami, T. Nagaoka, S. Sugiyama, T. Ohtani, Nano Lett. 2003, 3,
1391. d) Z. Deng, Y. Tian, S.-H. Lee, A. E. Ribbe, C. Mao, Angew.
Chem., Int. Ed. 2005, 44, 3582.
3
4
a) X. Fu, Y. Wang, L. Huang, Y. Sha, L. Gui, L. Lai, Y. Tang, Adv.
Mater. 2003, 15, 902. b) N. Sharma, A. Top, K. L. Kiick, D. J.
Pochan, Angew. Chem., Int. Ed. 2009, 48, 7078.
28 H. Lee, N. F. Scherer, P. B. Messersmith, Proc. Natl. Acad. Sci.
U.S.A. 2006, 103, 12999.
29 Y. Mikhlin, M. Likhatski, A. Karacharov, V. Zaikovski, A. Krylov,
Phys. Chem. Chem. Phys. 2009, 11, 5445.
5
6
H. Qi, T. Hegmann, J. Mater. Chem. 2006, 16, 4197.
a) S. Guha, M. G. B. Drew, A. Banerjee, Small 2008, 4, 1993. b) S.
Guha, A. Banerjee, Macromol. Chem. Phys. 2009, 210, 1422. c) S.
Guha, A. Banerjee, Adv. Funct. Mater. 2009, 19, 1949.
a) C. T. Wirges, J. Timper, M. Fischler, A. S. Sologubenko, J. Mayer,
U. Simon, T. Carell, Angew. Chem., Int. Ed. 2008, 48, 219. b) L. A.
Stearns, R. Chhabra, J. Sharma, Y. Liu, W. T. Petuskey, H. Yan, J. C.
Chaput, Angew. Chem., Int. Ed. 2009, 48, 8494. c) D. Sarkar, M.
Mandal, R. Das, K. Mandal, IEEE Trans. Magn. 2011, 47, 3163.
a) E. Dujardin, C. Peet, G. Stubbs, J. N. Culver, S. Mann, Nano Lett.
2003, 3, 413. b) M. Ł. Górzny, A. S. Walton, S. D. Evans, Adv.
Funct. Mater. 2010, 20, 1295.
30 B. J. Lindberg, K. Hamrin, G. Johansson, U. Gelius, A. Fahlman, C.
Nordling, K. Siegbahn, Phys. Scr. 1970, 1, 286.
7
8
31 Given that the particles (100 nm in diameter) consisted of Au0 atoms
and cyclic dipeptides
1 (volume occupied by one molecule;
319.2 ¡3), ca. 160 nm long fibrous nanostructure (100 nm in width,
30 nm in height from Figure 3b) could be used to make a particle.
This calculated result could be consistent with the observation in
Figure 5.
Chem. Lett. 2013, 42, 601-603
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/