Addressable carbene anchors for gold surfaces

Article abstract of DOI:10.1021/ja401965d

New strategies to access functional monolayers could augment current surface modification methods. Here we present addressable N-heterocyclic carbene (ANHC) anchors for gold surfaces. A suite of experimental and theoretical methods was used to characteriz

Full text of DOI:10.1021/ja401965d

Communication
pubs.acs.org/JACS
Addressable Carbene Anchors for Gold Surfaces
Aleksandr V. Zhukhovitskiy, Michael G. Mavros, Troy Van Voorhis, and Jeremiah A. Johnson*
Department of Chemistry, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, Massachusetts 02139,
United States
S
* Supporting Information
bond strength, rate of adsorption, layer density, or electronic
ABSTRACT: New strategies to access functional mono-
layers could augment current surface modification
methods. Here we present addressable N-heterocyclic
carbene (ANHC) anchors for gold surfaces. A suite of
experimental and theoretical methods was used to
characterize ANHC monolayers. We demonstrate grafting
of highly fluorinated polymers from surface-bound
ANHCs. This work establishes ANHCs as viable anchors
for gold surfaces.
structure. Moreover, the exclusively N,N-dialkyl NHCs
described rendered the surfaces inert toward further function-
alization.
To study NHC-gold surface binding and facilitate NHC
monolayer functionalization, we prepared (Scheme S1) two
“addressable NHCs” (ANHCs) that possess aryl bromide21 (1,
Figure 2A) and β-methylstyrene (2, Figure 2A) functional
ince its discovery in 1983,1 the chemisorption of thiols on
S
gold surfaces has enabled countless technological advances
in the fields of electronics,2 sensing,3 microfabrication,4 and
nanotechnology.5 Despite this broad utility, S−Au monolayers
have limitations. For example, the relatively weak S−Au bond
(∼45 kcal/mol)6 can lead to monolayer desorption at moderate
temperatures (∼100−150 °C).6,7 Furthermore, S−Au mono-
layers often have ill-defined binding geometries.8 Finally, S−Au
bonds typically have low conductance, which could limit their
use in molecular electronics applications.9
Figure 2. (A) ANHC structures. (B) Crystal structure of complex 3.
groups. Both ANHCs form mono- and bis-Au(I) complexes
(e.g., 3, Figure 2B) upon exposure to (Ph3P)AuCl in THF
(Scheme S2). The crystal structure of 3 features a C−Au bond
length of 1.98 Å, which is consistent with reported values for
IMes- and SIMes-Au(I) complexes (2.00 and 1.98 Å,
respectively).22 Of note, this bond length is much shorter
than the Au−S bond length (2.2−2.6 Å) observed in crystal
structures of thiolate-stabilized gold nanoparticles.23
Quartz crystal microbalance with dissipation (QCM-D) was
used to study binding of 1 and commercially available IMes to
gold surfaces. For all QCM-D experiments, a THF solution of
free carbene was flowed over a gold-coated sensor; binding was
characterized via changes in frequency (F) and dissipation (D)
of the sensor. The carbene solutions were prepared as follows
(see SI for details):
Other anchor groups have been explored for binding to gold
surfaces. Though some of these, such as alkyl10,11 and
dithiocarbamate,9 display increased conductance or improved
binding strength, there is still need for a general, synthetically
versatile complement to Au−S monolayer formation.
We were drawn to N-heterocyclic carbenes (NHCs)12 as a
potentially useful class of reagents for binding to inorganic
surfaces (Figure 1). NHCs offer a combination of exceptional
(a) For 1 and 2, a THF suspension of imidazolium salt
ANHC precursor (IS1 or IS2, respectively, Scheme S1) was
exposed to potassium hexamethyldisilazide (KHMDS, 1.0
equiv) under N2. The resulting solution was filtered through
a 0.25 μm syringe filter.
Figure 1. Functionalization of gold with ANHCs.
(b) For IMes, IMes was dissolved in THF under N2. The
solution was filtered through a 0.25 μm filter.
σ-donating and moderate π-backbonding ability,13 which has
made them ligands of choice for late transition metals like
Ru(II)14,15 and Au(I).16 We envisioned that these same
characteristics could lead to strong, partially conjugated surface
bonds. Furthermore, the synthetic flexibility of NHCs could
facilitate their general use for surface functionalization.
Both carbene solutions initiated a rapid frequency change
upon introduction to the QCM-D sensor; saturation was
approached within 15 min (Figure 3A, B). As expected for rigid
monolayers,24 the surfaces were characterized by small ratios of
ΔD:ΔF (≪4 × 10−7 Hz−1). The areal mass density (AMD) of
bound species was estimated using the Sauerbrey method.24,25
To date, one published work17 explored the use of NHCs for
coating planar gold surfaces; three reports18−20 entertained
gold nanoparticle stabilization with NHCs. While these studies
are encouraging, they provide no details for such parameters as
Received: February 26, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja401965d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Figure 3. QCM-D traces for carbene solutions. (A) Solution (a): ANHC 1 generated via exposure of IS1 to KHMDS in THF. (B) Solution (b):
Commercially available NHC IMes in THF. (C) THF solution of ANHC 1 generated via decarboxylation of CO2-1 (no KHMDS present). The
measured areal mass densities (AMDs) for each sample are 210 80, 56 6, and 63 14 ng/cm2, respectively.
Average AMD values for 1 and IMes taken from three
measurements were 210 80 and 56 6 ng/cm2, respectively.
Control experiments with HMDS amine or amide in the
absence of carbene showed little binding of the former, but
significant binding of the latter (Figure S1). Thus, we
hypothesized that binding of residual HMDS amide led to
the larger AMD, and increased deviation, for 1 compared to
IMes.
To test this hypothesis, a solution of 1 in THF was prepared
via thermal decarboxylation of an independently synthesized
CO2-1 adduct (Scheme S3). The average AMD value for this
solution of 1 without HMDS was 63 14 ng/cm2 (Figure 3C),
which agrees well with the value for IMes.
Surfaces treated with IMes showed no detectable Br (Figure
4B).
In order to gain further insight into the nature of the NHC-
Au surface interaction, density functional theory (DFT) was
used to model the binding of 1 to a charge-neutral gold adatom
above a fixed gold lattice. The calculated structure (Figure 4C)
possesses a C−Au bond length of 2.03 Å, which agrees well
with that of the crystal structure for 3. Furthermore, the
calculated homolytic Au−C bond dissociation energy (BDE)
was found to be 67 kcal/mol, which is more than 20 kcal/mol
larger than a typical Au−S bond. Calculations performed using
either a single gold atom or gold clusters produced similar 1-Au
σ-bonding orbitals, which suggests that the bonding is highly
localized to a single gold atom.
Based on the dimensions of 1 obtained via crystallography
(Figure 2B), we estimate the maximum possible AMD for a
monolayer of 1 on a perfectly flat surface to be 85 ng/cm2. This
limit would be higher for a rough surface. Given the steric bulk
of 1 and IMes, the measured AMDs (∼63 and ∼56 ng/cm2,
respectively) are reasonable.
Monolayers of 1 and IMes prepared via immersion of gold-
coated silicon wafers in solution (a) or (b), respectively, were
characterized by narrow-scan X-ray photoelectron spectroscopy
(XPS). The spectra were normalized to the transmission-
corrected area of the carbon peaks (Figure 4A). The surface
exposed to 1 showed a significant Br signal (Figure 4B). The
measured Br/N ratio of 0.16:1 corresponds to a mixed
monolayer with 21% 1 and 79% HMDS by mass (Table S1).
We next studied the electronic structure of the 1-Au bond via
DFT using the B3LYP functional; the basis set was LANL2DZ
+ effective core potential for gold, and 6-31g* for all other
atoms. For clarity, we present results (Figures 4D and S2) from
the simplest (one gold atom) model. The electron density in
the HOMO (Figure 4D) is delocalized over the gold atom and
the carbene carbon; this delocalization extends to the nitrogen
atoms in the HOMO. These results suggest that ANHCs could
form conductive surface linkages.
Next we sought to demonstrate chemical modification of an
ANHC-gold surface. Attempts to perform various metal-
catalyzed cross-coupling reactions on 1-Au surfaces were met
with difficulty due to nonspecific adsorption. Thus, we focused
on modification of the olefins of 2-Au monolayers. We
envisioned that treatment of 2-Au surfaces with third-
generation Grubbs catalyst26 (Ru, Figure 5A) would generate
surface-bound initiators for ring-opening metathesis polymer-
ization (ROMP).27,28
A series of model experiments using an isolated bis-2-Au
complex demonstrated that cross-metathesis between Ru and 2
was efficient in solution (Figures S3−S5, Table S2).
Encouraged by these results, we performed the sequence of
reactions depicted in Figure 5A on a gold-coated QCM-D
sensor; relevant steps are labeled i−vi. First, exposure of the
sensor to a 0.21 mM solution of 2 (prepared via method (a))
for 15 min (region i) followed by a wash with fresh THF
(region ii) resulted in an AMD of 230 ng/cm2. If we assume
that 2 binds to the surface with equal affinity to IMes, then ∼61
ng/cm2 of this AMD value corresponds to 2.
The surface was then exposed to a 5.80 mM solution of Ru in
THF for 5 h (region iii). Another THF wash was performed
(region iv). At this stage, the surface consisted of putative Ru-
benzylidene complexes bound via the 2-Au linkage (Ru-2-Au
Figure 4. (A) C 1s and (B) Br 3p regions of the XPS spectra for 1 and
IMes bound to planar gold surfaces. (C) DFT model of 1 bound to a
gold surface. (D) HOMO of the 1-Au(0) complex.
B
dx.doi.org/10.1021/ja401965d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Figure 5. (A) Scheme for ROMP from 2-Au surfaces and QCM-D data for the surface functionalization process. Red and blue curves correspond to
various frequency (Fj, inset legend) and dissipation (Dj, inset legend) overtones, respectively. (B) F 1s and Ru 3p3/2 regions of the XPS spectrum of
poly(4)-2-Au.
surface). The 2-Au to Ru-2-Au process coincided with a ∼2.6
Hz frequency change, and a significant change in dissipation
(from ∼0.2 × 10−6 to ∼0.7 × 10−6). The Voigt model29,30 was
used to calculate an AMD of 60 ng/cm2 of bound Ru (see SI
for details). This AMD combined with the estimated AMD of 2
(61 ng/cm2) corresponds to ∼39% olefin conversion, which is
consistent with the percentage of cross metathesis observed in
solution studies (Figure S5).
Subsequent exposure of the surface to pentafluorophenyl
exo-norbornene derivative 4 (0.121 M in THF) for 2 h (region
v) resulted in drastically altered frequency and dissipation
values along with an observed dispersion in (1/j)Δf for
different overtones j (Figure 5A). These results are consistent
with growth of a soft polymer brush from the surface (poly(4)-
2-Au).24 The ΔAMD from polymerization was 1520 ng/cm2,
which, if polymer solvation is neglected, translates to an average
degree of polymerization (DP) of 35.
No polymerization was observed when the same sequence of
events was carried out using 1 rather than 2 (Figure S6A),
which confirms the role of the olefinic groups of 2. Finally,
exposure of a 2-Au surface to monomer 4 in the absence of Ru
gave no change in dissipation and a ΔAMD of ∼53 ng/cm2
from nonspecific adsorption (Figure S7A); no polymerization
occurred.
XPS analysis was performed on the same surfaces used for
QCM-D experiments (Figures 5B, S6B, and S7B; Table S1). As
expected, the poly(4)-2-Au surface exhibited high fluorine
content along with Ru (Figure 5B). Control samples with 1 or
no Ru showed much lower fluorine signal from adsorbed 4.
The Ru/F ratio for poly(4)-2-Au (see footnote for Table S1)
was used to calculate an average brush DP of 18 (assuming one
Ru atom per polymer chain and five F atoms per polymer
repeat unit). The difference in DP compared to QCM-D is
likely due to polymer solvation.31
Figure 6. AFM characterization of (A) poly(4)-2-Au surface and (B)
control surface with no Ru catalyst added.
6B) and 2.0 nm (1-Au monolayer, Figure S8), which matches
that reported for bare sensors (3 nm).32 The elongated features
present only in the AFM image of poly(4)-2-Au (Figure 6A)
resemble those reported for other poly-norbornene grafted
surfaces.27
In this report we have described the first examples of gold
surface functionalization with addressable N-heterocyclic
carbenes. We describe details of the ANHC-surface binding
interaction and demonstrate the growth of novel polymer
brushes from robust ANHC monolayers. We expect that these
results will spark interest in the use of ANHCs and other stable
carbenes as general surface anchors.
ASSOCIATED CONTENT
* Supporting Information
Supplemental figures, schemes, tables, calculations, specifica-
Tapping-mode atomic force microscopy (AFM) of these
surfaces revealed a marked difference in roughness. Poly(4)-2-
Au had a roughness (RMS) of 5.6 nm (Figure 6A). In contrast,
the roughness of control sensors was 1.4 nm (no Ru, Figure
■
S
tions of DFT analysis, detailed synthetic procedures, and
C
dx.doi.org/10.1021/ja401965d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(26) Love, J. A.; Morgan, J. P.; Trnka, T. M.; Grubbs, R. H. Angew.
Chem., Int. Ed. 2002, 41, 4035.
(27) Kong, B.; Lee, J. K.; Choi, I. S. Langmuir 2007, 23, 6761.
(28) Weck, M.; Jackiw, J. J.; Rossi, R. R.; Weiss, P. S.; Grubbs, R. H. J.
Am. Chem. Soc. 1999, 121, 4088.
(29) Johannsmann, D. Phys. Chem. Chem. Phys. 2008, 10, 4516.
(30) Voinova, M. V.; Rodahl, M.; Jonson, M.; Kasemo, B. Phys. Scr.
1999, 59, 391.
spectral and crystallographic data. This material is available free
of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
jaj2109@mit.edu
■
Notes
(31) Muller, M. T.; Yan, X.; Lee, S.; Perry, S. S.; Spencer, N. D.
̈
The authors declare no competing financial interest.
Macromolecules 2005, 38, 5706.
(32) Reported by QSense on www.q-sense.com/sensors-and-
coatings-1.
ACKNOWLEDGMENTS
■
We thank the Deshpande Center for Technological Innovation,
the MIT Lincoln Laboratories, and the DoD NDSEG program
for support of this work. This work made use of the MRSEC
Shared Experimental Facilities at MIT (NSF, DMR-08-19762)
and the MIT X-Ray Facility (NSF, CHE-0946721). We thank J.
Im, R. Faust, and L. Ren for help with QCM-D experiments, P.
Muller for X-ray crystallography, and J. E. Whitten and Y. Wang
̈
for supplemental XPS characterization.
REFERENCES
■
(1) Nuzzo, R. G.; Allara, D. L. J. Am. Chem. Soc. 1983, 105, 4481.
(2) DiBenedetto, S. A.; Facchetti, A.; Ratner, M. A.; Marks, T. J. Adv.
Mater. 2009, 21, 1407.
(3) Ranieri, A.; Monari, S.; Sola, M.; Borsari, M.; Battistuzzi, G.;
Ringhieri, P.; Nastri, F.; Pavone, V.; Lombardi, A. Langmuir 2010, 26,
17831.
(4) Koh, S. Nanoscale Res. Lett. 2007, 2, 519.
(5) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides,
G. M. Chem. Rev. 2005, 105, 1103.
(6) Nuzzo, R. G.; Zegarski, B. R.; Dubois, L. H. J. Am. Chem. Soc.
1987, 109, 733.
(7) Cometto, F. P.; Patrito, E. M.; Paredes Olivera, P.; Zampieri, G.;
Ascolani, H. Langmuir 2012, 28, 13624.
(8) Pensa, E.; Cortes, E.; Corthey, G.; Carro, P.; Vericat, C.;
́
Fonticelli, M. H.; Benítez, G.; Rubert, A. A.; Salvarezza, R. C. Acc.
Chem. Res. 2012, 45, 1183.
(9) von Wrochem, F.; Gao, D.; Scholz, F.; Nothofer, H.-G.; Nelles,
G.; Wessels, J. M. Nat. Nanotechnol. 2010, 5, 618.
(10) Cheng, Z. L.; Skouta, R.; Vazquez, H.; Widawsky, J. R.;
Schneebeli, S.; Chen, W.; Hybertsen, M. S.; Breslow, R.;
Venkataraman, L. Nat. Nanotechnol. 2011, 6, 353.
́
(11) Chen, W.; Widawsky, J. R.; Vazquez, H.; Schneebeli, S. T.;
Hybertsen, M. S.; Breslow, R.; Venkataraman, L. J. Am. Chem. Soc.
2011, 133, 17160.
(12) Kirmse, W. Angew. Chem., Int. Ed. 2010, 49, 8798.
(13) Jacobsen, H.; Correa, A.; Poater, A.; Costabile, C.; Cavallo, L.
Coord. Chem. Rev. 2009, 253, 687.
(14) Huang, J.; Stevens, E. D.; Nolan, S. P.; Petersen, J. J. Am. Chem.
Soc. 1999, 121, 2674.
(15) Scholl, M.; Trnka, T. M.; Morgan, J. P.; Grubbs, R. H.
Tetrahedron Lett. 1999, 40, 2247.
(16) Nolan, S. P. Acc. Chem. Res. 2010, 44, 91.
(17) Weidner, T.; Baio, J. E.; Mundstock, A.; Grosse, C.; Karthaeuser,
S.; Bruhn, C.; Siemeling, U. Aust. J. Chem. 2011, 64, 1177.
(18) Serpell, C. J.; Cookson, J.; Thompson, A. L.; Brown, C. M.;
Beer, P. D. Dalton Trans. 2013, 42, 1385.
(19) Vignolle, J.; Tilley, T. D. Chem. Commun. 2009, 7230.
(20) Hurst, E. C.; Wilson, K.; Fairlamb, I. J. S.; Chechik, V. New J.
Chem. 2009, 33, 1837.
(21) Sußner, M.; Plenio, H. Chem. Commun. 2005, 5417.
̈
́
(22) de Fremont, P.; Scott, N. M.; Stevens, E. D.; Nolan, S. P.
Organometallics 2005, 24, 2411.
(23) Jadzinsky, P. D.; Calero, G.; Ackerson, C. J.; Bushnell, D. A.;
Kornberg, R. D. Science 2007, 318, 430.
(24) Reviakine, I.; Johannsmann, D.; Richter, R. P. Anal. Chem. 2011,
83, 8838.
(25) Sauerbrey, G. Z. Phys. 1959, 155, 206.
D
dx.doi.org/10.1021/ja401965d | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX