Heterogeneous catalysis through microcontact printing

Article abstract of DOI:10.1002/anie.200803480

(Figure Presented) Minting a Stamp: The preparation of copper metal-coated elastomeric stamps and their use in catalyzing the Cu-catalyzed azide-alkyne cycloaddition reaction heterogeneously through microcontact printing is described (see scheme). This StampCat process is compared to other conventional surface-functionalization techniques, including traditional microcontact printing and solution-surface-based reactions.

Full text of DOI:10.1002/anie.200803480

Angewandte
Chemie
DOI: 10.1002/anie.200803480
Surface Functionalization
Heterogeneous Catalysis through Microcontact Printing**
Jason M. Spruell, Bonnie A. Sheriff, Dorota I. Rozkiewicz, William R. Dichtel, Rosemary D.
Rohde, David N. Reinhoudt,* J. Fraser Stoddart,* and James R. Heath*
The chemical toolkit for the modification of inorganic
surfaces with organic molecules has grown enormously in
recent years. Much of the underlying science has been driven
by developing technologies[1] such as biomolecular and
chemical sensing,[2] molecular electronics,[3] light harvesting,[4]
and others. These applications all have certain requirements
that place unique demands on the chemistry, which often
impact more than just the choice of substrate or organic
modifier. For example, electronic communication between
surface-bound organic molecules and the underlying sub-
strate may be necessary. Other application-specific require-
ments include rapid and efficient surface attachment reac-
tions that provide 1) high surface coverage, 2) protection of
the underlying substrate against oxidation,[5] and/or 3) control
over molecular orientation.
The ability to pattern the organic monolayer spatially is a
common requirement. The technique of microcontact print-
ing (mCP) permits[6] the patterned transfer of a molecular ink
onto a surface using an elastomeric stamp. The transferred
molecules may form a self-assembled, high coverage mono-
layer, e.g., thiols on gold[6b,e] or silanes on silicon dioxide.[7]
Alternatively, mCP can promote the spatially selective reac-
tion between an already surface-bound molecule and a
molecular ink. This concept has been demonstrated for
1) the patterning of peptides using amide[8] coupling chemis-
try, 2) the formation of thermodynamically stable imine[9]
bonds, and 3) for the promotion of the Huisgen 1,3-dipolar
cycloaddition, yielding[10] a triazole link between a surface-
bound moiety and the ink molecule. Both the triazole- and
amide-coupling reactions typically require catalysts or addi-
tional stoichiometric reagents, such as carbodiimides, when
performed in solution. However, within the unique environ-
ment[11] created when the surface-bound and molecular ink
reagents are brought into close contact using a stamp, the
reactions have been reported[10,11] to proceed—at least to
some extent—without these additional components.
The organic functionalization of surfaces is heterogene-
ous, involving a solid-surface reacting with solution- or gas-
phase reagents. The two-dimensional nature of surfaces also
means that steric effects can either impose kinetic barriers
upon thermodynamically favored chemical processes, or they
can impart kinetic stability to certain non-equilibrium surface
structures.[12] The presence of kinetic barriers, in turn, implies
that different chemical approaches towards functionalizing
surfaces with organic components might lead to different
coverages and/or surface structures.
Here, we investigate four different chemical pathways
(Scheme 1a–d) relevant to the Cu-catalyzed azide–alkyne
cycloaddition (CuAAC) reaction.[13] Three of those pathways
lead to surfaces functionalized with organic molecules.[5,11,14]
At the outset, our practical goal was to identify surface-
functionalization protocols that are capable of attaining
1) spatial selectivity, 2) high surface coverage, and 3) rapid
reaction kinetics. Our ultimate goal is to achieve a funda-
mental understanding of how different reaction pathways
influence the chemical outcome as it applies to the organic
functionalization of surfaces.
The four pathways illustrated in Scheme 1a–d vary with
regard to how the reaction components (1–7 and the Cu
source) are introduced to one another. A Cu catalyst is
required for the CuAAC reaction to proceed to any
appreciable extent in each scheme. Probes 5–7, when attached
to surfaces, are respectively designed for specific redox, X-ray
photoelectron spectroscopy (XPS), or fluorescent signatures,
thus allowing for the full characterization of the functional-
ized surfaces using complementary techniques. Scheme 1a
summarizes a homogeneous, solution-based pathway in which
1-azidohexane (3), 1-octyne (4), and CuSO4·5H2O/ascorbic
acid[13b] react in solution, and provides a reference against
which the surface chemistries can be compared. Under
conditions chosen to simulate those relevant to surface
functionalization (20-fold excess of 4 relative to 3), the
[*] J. M. Spruell,[+] Dr. W. R. Dichtel, Prof. J. F. Stoddart
Department of Chemistry, Northwestern University, 2145 Sheridan
Road, Evanston, IL 60208 (USA)
Fax: (+1)847-491-1009
E-mail: stoddart@northwestern.edu
Homepage: http://stoddart.northwestern.edu
Dr. D. I. Rozkiewicz,[+] Prof. D. N. Reinhoudt
Laboratory of Supramolecular Chemistry and Technology, MESA+
Institute for Nanotechnology, University of Twente, P.O. Box 217,
7500 AE Enschede (The Netherlands)
Fax: (+31)53-489-4645
E-mail: d.n.reinhoudt@utwente.nl
Dr. B. A. Sheriff,[+] Dr. W. R. Dichtel, R. D. Rohde, Prof. J. R. Heath
Division of Chemistry and Chemical Engineering, California Insti-
tute of Technology, 1200 East California Boulevard, Pasadena, CA
91125 (USA)
E-mail: heath@caltech.edu
Homepage: http://www.its.caltech.edu/~heathgrp/
[+] These authors have contributed equally to the research described in
this Communication.
[**] The research was supported by the Microelectronics Advanced
Research Corporation (MARCO) and its Focus Center of Functional
Engineered NanoArchitectonics (FENA) and Materials, Structures,
and Devices (MSD) and by the NanoImpuls/NanoNed program of
the Dutch Ministry of Economic Affairs (grant TTF6329). J.M.S.
gratefully acknowledges the award of a Graduate Research Fellow-
ship from the National Science Foundation (NSF).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200803480.
Angew. Chem. Int. Ed. 2008, 47, 9927 –9932
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9927

Communications
CuAAC reaction proceeded to 95% conversion[15] during 6 h.
This finding is a useful baseline for comparison of the kinetics
of the reactions performed through each of the heterogeneous
surface reactions (Scheme 1b–d).
The substrates for the CuAAC reactions described in
Scheme 1b–d were composed of azide functionalities teth-
ered to undecyl alkyl chains as self-assembled monolayers
(SAMs) on gold and silicon oxide through terminal thiol and
silane groups, respectively. Following on the work of Collman
et al.,[14a,b] the monolayers on Au were deposited from a 1:1
mixture of two thiols (azidoundecanethiol and octanethiol) to
form mixed monolayers[16] in which the relatively long,
projecting azido functions are accessible for reaction. Glass
slides functionalized with azide moieties were created in a
two-step process[10] whereby a bromoundecylsilane SAM is
converted to the azido-terminated SAM through nucleophilic
displacement of the bromide by sodium azide. The Solution–
Surface approach (Scheme 1b) introduces both the alkyne-
terminated molecules and the CuI catalyst, dissolved in a
solution placed in contact with the surface. Similar chemistry
has been reported previously on Au[14a,b] and other[17,18]
substrates. The Reagent-Stamping method (Scheme 1c)
Scheme 1. Various routes to affect the CuI-catalyzed azide–alkyne
coupling (CuAAC) in different environments: a) Solution: A solution
reaction where azide, alkyne, and catalyst participate in a homoge-
neous reaction; b) Solution–Surface: A heterogeneous reaction where
dissolved alkyne and catalyst react with a surface bound azide;
c) Reagent-Stamping: A heterogeneous reaction where the alkyne and
catalyst are brought into contact with a surface-bound azide in the
condensed phase; d) StampCat: A heterogeneous reaction where
immobilized copper catalyzes the reaction of an alkyne with a surface
bound azide in the condensed phase.
employs
mCP
methods—using
polydimethylsiloxane
(PDMS) stamps—to introduce both the alkyne-containing
reactant and the CuI catalyst to the surfaces by the (solvent-
less) contacting of inked stamps onto the surfaces. Finally, the
StampCat procedure (Scheme 1d) utilizes a PDMS stamp
coated with a thin Cu metal film. The stamp itself catalyzes
the spatially-selective reaction between a small-molecule
alkyne and the azide surface, also under solvent-free con-
ditions.
Elastomeric stamps are often modified[19] to improve their
ability to transfer various inks for mCP, generally by increasing
the loading potential of the stamp. The CuAAC reaction may
be catalyzed in both the solution[13b,20] and solid phase[21] using
heterogeneous catalysis in the form of a copper metal source.
We hypothesized that a Cu-coated elastomer would similarly
catalyze the coupling between surface-bound azides and
terminal alkynes. Thus, we modified cured PDMS stamps via
electron-beam deposition of a 10 nm Ti adhesion layer
followed by a 50 nm Cu layer. The resulting metal-coated
flat or patterned stamps retained most of their elasticity and
were used in the StampCat process.
Surfaces functionalized according to Schemes 1b–d were
characterized by water contact angle, XPS, FTIR, and cyclic
voltammetry (CV) measurements. For Scheme 1c and d, flat,
non-patterned stamps were utilized. For the experiments on
surfaces following the CuAAC reaction with each molecular
probe (5–7), all measurements other than CV were carried
out after the relevant reaction steps had proceeded to
completion (see below). For the reaction of an Au surface
with 5, CV measurements were carried out to validate the
formation of a covalent linkage between 5 and the surface,
and to follow the reaction kinetics of Scheme 1b–d.
Cyclic voltammetry measurements were utilized to con-
firm the formation of a covalent bond between the surface
and 5, and to quantify the extent of the reaction. Figure 1
displays CV data, collected for a mixed monolayer of 1 and 2
that was functionalized by StampCat of 5. The linear increase
9928 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 9927 –9932

Angewandte
Chemie
Figure 1. Electrochemical measurements at various scan rates, indicat-
ing covalent functionalization of ferrocene-containing ink 3 through
StampCat. Proof of covalent attachment is supported by the linear plot
of current versus scan rate. Inset: current recorded at +0.51 V for
each scan rate. The electrolyte was 1m HClO4.
in current as a function of scan rate, and the identical values
for the peak potential for both oxidation and reduction of the
ferrocene, indicate covalent surface attachment of the ferro-
cenyl moieties. A square-root dependence of current on scan
rate would be expected for similar redox behavior if the
electroactive components were dissolved in solution. Similar
spectra were obtained for surfaces functionalized with 5
employing the Reagent-Stamping and Solution–Surface
schemes. Integration of these spectra confirmed that similar
efficiencies were obtained using all three schemes.
Further spectroscopic proof of surface functionalization
was obtained by XPS. The StampCat process using the
pentafluorophenolic ether 6 demonstrated a pronounced
change (Figure 2a,b) in the XPS spectra of the F1s region
as a consequence of the addition of a highly fluorinated
molecular entity onto the surface. Surfaces functionalized
with 5, using the Reagent-Stamping process, displayed similar
changes in their XPS spectra. Figure 2c presents high-
resolution XPS spectra of the N1s region, and shows two
peaks at 400 and 404 eV with an intensity ratio of 2:1, similar
to other reports for azide groups on gold[14b] and graphitic[17]
surfaces. After functionalization with 5, a single feature in the
N1s region is observed (Figure 2d). These changes for
CuAAC functionalized surfaces were common to every
situation we investigated, and are also consistent with
literature observations.[14a,b] For surfaces functionalized with
5, changes in the Fe2p region of the spectrum revealed
(Figure 2 f) the presence of an Fe-containing molecule.
External reflectance IR spectroscopy also revealed the
presence or absence of azide functions at the surface
(Figure 3) by measurement of a large peak centered on
2103 cmÀ1 that arises from the azide symmetric stretching
mode. Upon either StampCat or Reagent-Stamping of 5, the
integrated peak intensity is attenuated by ca. 90% while a
Solution–Surface reaction with 5 leads to its complete
disappearance. These results suggest azide consumption
resulting from triazole ring formation and hence the func-
tionalization of the surface in general. Water contact angle
measurements also lent qualitative support to the idea that all
three methods proceed with comparable reaction efficiency
(see Supporting Information).
Figure 2. XPS data demonstrating the functionalization of azide-termi-
nated Au substrates (a, c, and e) with pentafluorobenzene 6-containing
(b) and ferrocene 5-containing alkynes (d and f). The StampCat
CuAAC reaction is demonstrated in the F1s region before (a) and after
(b) the reaction with the pentafluoro alkyne 6. The Reagent-Stamping
CuAAC reaction is demonstrated in the N1s and Fe2p regions of the
XPS before (c, e) and after (d, f) the reaction with the ferrocene-
containing alkyne 5.
Figure 3. IR spectra demonstrating the reduction of the azide stretch
after functionalization with the ferrocene-containing ink 5 using the
Reagent-Stamping, StampCat, and Solution–Surface methods.
Angew. Chem. Int. Ed. 2008, 47, 9927 –9932
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9929

Communications
The solution-based homogeneous reaction (Scheme 1a)
benchmark for the kinetics of the different surface function-
alization approaches.
provided a baseline for comparison with the heterogeneous,
surface-based reactions of Schemes 1b–d. Under the condi-
tions employed to approximate[22] those expected for surface-
based reactions, the reaction proceeds (Figure 4c) to > 95%
conversion within 400 min with a half-life around 1 hour. The
homogeneous CuAAC reaction is known to demonstrate
complicated kinetics. Under catalytic conditions in copper,
Finn and co-workers[23] found the reaction to demonstrate
different rate orders in copper depending on concentration
and ligand choice. Thus, rather than elucidate the rate law for
the conditions described above, we used them as a rough
The measured rates (using cyclic voltammetry) of the
chemical processes summarized in Schemes 1b–d are shown
in Figure 4. For these experiments, the attachment of 5 to
azide-presenting Au surfaces was monitored as a function of
reaction (or stamping) time. When making the StampCat
measurements, a new Cu-coated PDMS stamp was used in the
microcontact printing of the surfaces produced for each time
point. Two different StampCat kinetic series were investi-
gated: one in which freshly deposited CuPDMS stamps were
employed (fresh) and another in which the stamps were
allowed to age for 24 h (aged). The integrated intensity of the
CV spectra increased (Figure 4a) as a function of reaction (or
stamping) time for all methods, and permitted calculation of
ferrocene coverage at the surfaces. That all three methods
produced coverages at a maximum of 2 ꢀ 1014 moleculescmÀ2
at extended reaction times indicates that the CuAAC
proceeds to completion, in agreement with previous measur-
ements[14b] made on the same mixed monolayer surfaces.
Plotting the number of functionalizing ferrocene molecules
per cm2 as a function of the reaction time provides (Fig-
ure 4b) a means to monitor the kinetics of each CuAAC
process.
The Reagent-Stamping process produced fully function-
alized surfaces at the highest rate, with complete conversion
being attained within 30 min. In comparison, the StampCat
process using aged stamps catalyzed the reaction at half the
rate, generating complete coverage after 1 hour. Interestingly,
a comparison of the two kinetic curves reveals different
operable rate orders, with the Reagent-Stamping displaying
what appears to be pseudo-first order exponential growth and
the StampCat process zero-order linear growth. Zero-order
reaction behavior was also observed when freshly deposited
Cu-coated PDMS stamps were used in the StampCat process,
but the rate was lower, producing ten-fold less coverage after
60 min of stamping time. The Solution–Surface reaction
proceeded at a very high rate initially, producing surfaces
functionalized with over 1.2 ꢀ 1014 ferrocene molecules per
cm2 after 5 min. However, the efficiency plummets in the later
stages of the reaction, only producing fully functionalized
surfaces (containing 2 ꢀ 1014 moleculescmÀ2) after 180 min[24]
of reaction. This non-linear kinetic behavior for the Solution–
Surface method is comparable to the pseudo-first order
exponential growth of the Reagent-Stamping method, a
feature that can be attributed to an abundance of CuI catalytic
centers available for reaction when employing both of these
methods. The Solution–Surface process produced fully func-
tionalized surfaces slower than either Reagent-Stamping or
StampCat (using aged stamps).
Figure 4. The kinetics of CuAAC catalysis for Schemes 1a–d. a) Cyclic
voltammograms, collected (at 300 mVsÀ1) as a function of stamping
time, for the StampCat process. Coverage increases with longer
stamping times (using aged stamps). b) Plot demonstrating the
conversion of azide surfaces to surfaces functionalized with 5, as a
function of time, for all three surface-based reactions (Schemes 1b–d).
For the StampCat process, reaction kinetics for stamps that were used
within an hour of metal deposition, and for stamps that were aged in
air for 24 h, are shown. Both exhibit linear response functions on this
plot, as indicated by the straight lines that are drawn to guide the eye.
c) Reaction conversion versus reaction time for the Solution reaction
(Scheme 1a) between 1-azidohexane (3) and 1-octyne (4), as mea-
sured by 1H NMR spectroscopy.
The StampCat process involving the CuAAC reaction is
dependent on the catalytic Cu metal coating being present.
When bare PDMS or platinum-coated stamps were used for
the functionalization of the azide-coated gold surfaces, there
was no electrochemical evidence (Figure 5) of the CuAAC
reaction having occurred with ferrocene alkyne 5. This was
unexpected based upon a previous report[10] of quantitative
copper-free click reaction onto glass surfaces. An electro-
chemical analysis of surfaces functionalized with 5 provides a
9930 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 9927 –9932

Angewandte
Chemie
bed water layer at the stamp or wafer surface. Thus, we used
the StampCat method to react the azide-presenting glass
surface with a micro-patterned stamp containing the fluores-
cent alkyne 7. We reasoned that, for long stamping periods,
the finer features of the stamp would be washed out of the
stamped pattern, if the catalyst originated from diffusing[27]
CuI ions. Otherwise, the stamped pattern would reflect the
fine structure of the stamp itself. The stamped patterns were
measured using confocal fluorescence microscopy, and com-
pared (Figure 6) alongside atomic force microscopy (AFM)
images of the surface of the Cu-coated PDMS stamp.
AFM of the Cu-coated PDMS stamp reflected the molded
pattern of 3 ꢀ 5 mm lines (Figure 6a), but also revealed a finer
structure of submicrometer width lines at a pitch of about
1 mm. This pattern likely arises from contraction of the PDMS
stamp when it is removed from the vacuum of the metal
deposition chamber. In any case, both the patterned and the
fine-structure features are observed in the fluorescence
micrographs of the glass surface. This observation, coupled
with the above described results that PDMS-only stamps, or
Pt-coated PDMS stamps do not promote the surface click
reaction to any appreciable extent, confirms that the catalyst
for StampCat is likely the (oxidized), Cu-coated surface of the
PDMS stamp. The StampCat process can clearly be har-
nessed[28] to generate impressively small feature sizes for mCP.
We have explored multiple pathways for achieving Cu-
catalyzed, surface functionalization of azide-modified surface.
One particularly novel pathway is the StampCat method, in
which a Cu-coated elastomeric stamp is utilized to catalyze
the reaction heterogeneously, and with spatial selectivity,
between molecules with pendant alkynes and surface-bound
azides. The StampCat method enables solvent-free, fast and
efficient surface functionalization. The mechanism and kinet-
ics of the StampCat pathway are contrasted with three other
schemes that vary primarily in terms of the increasing range of
reactant and reagent immobilization for the CuAAC reaction.
Figure 5. No ferrocene functionalization was observed for stamping
with Pt-coated PDMS or bare PDMS for 1 hour in comparison to the
high functionalization observed for stamping using Cu-coated PDMS.
quantitative assessment of reaction efficiencies. The earlier
work employed highly sensitive but non-quantitative surface
characterization methods. We appreciate now that the
reaction without a catalyst results in very limited conversion,
even over prolonged reaction times.
In an effort to interrogate the nature of the StampCat
process, the reaction efficiencies achieved from stamps with
freshly deposited Cu films, and those with Cu films that had
been aged for 24 h (Figure 4b) were compared. Both
situations exhibit linear response functions in terms of
ferrocene functionalization versus time, indicating a similar
chemical mechanism. However, the aged stamps yielded five-
fold more ferrocene coverage at 60 min than the freshly
coated stamps, implying that the aging process increases the
concentration of the active catalytic species on the Cu-coated
stamps. While the homogeneous CuAAC reaction is cata-
lyzed by free CuI ions, heterogeneous versions of the reaction
have been reported[20,21] in which surface-bound CuO and
Cu2O species are thought[20b,25] to be respon-
sible for the catalysis.
Studies of the oxidation of Cu surfaces[26]
indicate that Cu2O is the predominant initial
surface species formed under conditions that
are most similar to those for the stamps aged in
air for 24 h. It is this Cu2O coating which most
likely acts as the active heterogeneous catalyst
in the StampCat process. The stamps with
freshly-deposited Cu films are less catalytically
active because they contain fewer catalytically
active Cu2O surface sites. The aged Cu-coated
PDMS stamps also lose their activity over
repeated stampings (see Supporting Informa-
tion), implying that the surface Cu2O film is
somehow disturbed or depleted during the first
stamping event.
The ability to produce spatial surface
Figure 6. a) AFM investigation of a CuPDMS patterned stamp topography where both
the large line patterns (present before metal deposition) and metal ripples (as a
consequence of metal deposition) are observed. b) StampCat of the fluorescent alkyne
7 onto azide-functionalized glass slides using the patterned CuPDMS stamp produces
lines of stripes that are observed by fluorescent microscopy. The ripple pattern transfer
is consistent with heterogeneous catalysis.
patterns is a key advantage of mCP methods.
It is possible that the active catalyst from the
StampCat scheme is either the solid surface of
the stamp itself, or potentially CuI ions that
leave the stamp and diffuse through an adsor-
Angew. Chem. Int. Ed. 2008, 47, 9927 –9932
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9931

Communications
Angew. Chem. Int. Ed. 2007, 46, 1018; d) H. Nandivada, X. Jiang,
J. Lahann, Adv. Mater. 2007, 19, 2197.
A linear rate order is observed for metallic Cu film based
catalysis while pseudo-first order behavior is exhibited by
freely diffusing CuI catalytic processes. The CuAAC reaction
is becoming an increasingly important approach towards
chemically functionalizing surfaces, and these studies should
provide a firmer understanding of how that reaction can be
harnessed for application-specific uses.
[14] a) J. P. Collman, N. K. Devaraj, C. E. D. Chidsey, Langmuir
2004, 20, 1051; b) J. P. Collman, N. K. Devaraj, T. P. A. Eber-
spacher, C. E. D. Chidsey, Langmuir 2006, 22, 2457; c) N. K.
Devaraj, P. H. Dinolfo, C. E. D. Chidsey, J. P. Collman, J. Am.
Chem. Soc. 2006, 128, 1794; d) S. T. Seo, X. Bai, H. Ruparel, Z.
Li, N. J. Turro, J. Ju, Proc. Natl. Acad. Sci. USA 2004, 101, 5488;
e) M. J. Joralemon, R. K. OꢃReilly, C. J. Hawker, K. L. Wooley,
J. Am. Chem. Soc. 2005, 127, 16892; f) N. J. Agard, J. A.
Prescher, C. R. Bertozzi, J. Am. Chem. Soc. 2004, 126, 15046.
[15] The reaction was performed in an NMR tube using deuterated
DMF solvent such that reaction conversion could be determined
simply by comparing the integration of the signals due to the
appearing 1,2,3-triazole product with those due to disappearing 3
Received: July 17, 2008
Revised: September 11, 2008
Published online: November 14, 2008
Keywords: azide–alkynecycloaddition·heterogeneouscatalysis·
.
microcontact printing · StampCat · surface chemistry
1
in the H NMR spectrum.
[16] The mixed monolayer surfaces were reported to contain 45%
azide moieties when deposited from a 1:1 molar ratio solution of
azidoundecanethiol and octanethiol. See Ref. [14b].
[1] a) Z. H. Nie, E. Kumacheva, Nat. Mater. 2008, 7, 277; b) J.
Nakanishi, T. Takarada, K. Yamaguchi, M. Maeda, Anal. Sci.
2008, 24, 67; c) L. J. Guo, Adv. Mater. 2007, 19, 495; d) X. D.
Chen, S. Lenhert, M. Hirtz, N. Lu, H. Fuchs, L. F. Chi, Acc.
Chem. Res. 2007, 40, 393; e) J. Henzie, J. E. Barton, C. L.
Stender, T. W. Odom, Acc. Chem. Res. 2006, 39, 249.
[2] a) A. S. Anderson, A. M. Dattelbaum, G. A. Montaꢁo, D. N.
Price, J. G. Schmidt, J. S. Martinez, W. K. Grace, K. M. Grace,
B. I. Swanson, Langmuir 2008, 24, 2240; b) E. Katz, I. Willner, J.
Wang, Electroanalysis 2004, 16, 19.
[17] A. G. Marrani, E. A. Dalchiele, R. Zanoni, F. Decker, F. Cattar-
uzza, D. Bonifazi, M. Prato, Electrochim. Acta 2008, 53, 3903.
[18] A. Devadoss, C. E. D. Chidsey, J. Am. Chem. Soc. 2007, 129,
5370.
[19] a) D. I. Rozkiewicz, J. Gierlich, G. A. Burley, K. Gutsmiedl, T.
Carell, B. J. Ravoo, D. N. Reinhoudt, ChemBioChem 2007, 8,
1997; b) D. I. Rozkiewicz, W. Brugman, R. M. Kerkhoven, B. J.
Ravoo, D. N. Reinhoudt, J. Am. Chem. Soc. 2007, 129, 11593;
c) X. X. Duan, V. B. Sadhu, A. Perl, M. Peter, D. N. Reinhoudt,
J. Huskens, Langmuir 2008, 24, 3621.
[3] a) J. E. Green, J. W. Choi, A. Boukai, Y. Bunimovich, E.
Johnston-Halperin, E. DeIonno, Y. Luo, B. A. Sheriff, K. Xu,
Y. S. Shin, H.-R. Tseng, J. F. Stoddart, J. R. Heath, Nature 2007,
445, 414; b) W. R. Dichtel, J. R. Heath, J. F. Stoddart, Philos.
Trans. R. Soc. London Ser. A 2007, 365, 1607.
[4] A. Hagfeldt, M. Grꢂtzel, Acc. Chem. Res. 2000, 33, 269.
[5] R. D. Rohde, H. D. Agnew, W. S. Yeo, R. C. Bailey, J. R. Heath,
J. Am. Chem. Soc. 2006, 128, 9518.
[6] a) N. Balachander, C. N. Sukenik, Langmuir 1990, 6, 1621; b) A.
Kumar, G. M. Whitesides, Appl. Phys. Lett. 1993, 63, 2002; c) L.
Yan, X. M. Zhao, G. M. Whitesides, J. Am. Chem. Soc. 1998, 120,
6179; d) J. Lahiri, E. Ostuni, G. M. Whitesides, Langmuir 1999,
15, 2055; e) B. Michel, A. Bernard, A. Bietsch, E. Delamarche,
M. Geissler, D. Juncker, H. Kind, J. P. Renault, H. Rothuizen, H.
Schmid, P. Schmidt-Winkel, R. Stutz, H. Wolf, IBM J. Res. Dev.
2001, 45, 697.
[7] a) N. L. Jeon, K. Finnie, K. Branshaw, R. G. Nuzzo, Langmuir
1997, 13, 3382; b) Y. N. Xia, G. M. Whitesides, Angew. Chem.
1998, 110, 568; Angew. Chem. Int. Ed. 1998, 37, 550.
[8] a) L. Yan, X. M. Zhao, G. M. Whitesides, J. Am. Chem. Soc.
1998, 120, 6179; b) J. Lahiri, E. Ostuni, G. M. Whitesides,
Langmuir 1999, 15, 2055; c) L. Yan, W. T. S. Huck, X. M. Zhao,
G. M. Whitesides, Langmuir 1999, 15, 1208.
[9] D. I. Rozkiewicz, B. J. Ravoo, D. N. Reinhoudt, Langmuir 2005,
21, 6337.
[10] D. I. Rozkiewicz, D. Janczewski, W. Verboom, B. J. Ravoo, D. N.
Reinhoudt, Angew. Chem. 2006, 118, 5418; Angew. Chem. Int.
Ed. 2006, 45, 5292.
[20] a) C. N. Urbani, C. A. Bell, M. R. Whittaker, M. J. Monteiro,
Macromolecules 2008, 41, 1057; b) I. S. Park, M. S. Kwon, Y.
Kim, J. S. Lee, J. Park, Org. Lett. 2008, 10, 497; c) J. M. Spruell,
W. R. Dichtel, J. R. Heath, J. F. Stoddart, Chem. Eur. J. 2008, 14,
4168; d) B. H. Lipshutz, B. R. Taft, Angew. Chem. 2006, 118,
8415; Angew. Chem. Int. Ed. 2006, 45, 8235; e) L. V. Lee, M. L.
Mitchell, S.-J. Huang, V. V. Fokin, K. B. Sharpless, C.-H. Wong, J.
Am. Chem. Soc. 2003, 125, 9588; f) Q. Wang, T. R. Chan, R.
Hilgraf, V. V. Fokin, K. B. Sharpless, M. G. Finn, J. Am. Chem.
Soc. 2003, 125, 3192.
[21] D. D. Diaz, S. Punna, P. Holzer, A. K. McPherson, K. B. Sharpless,
V. V. Fokin, M. G. Finn, J. Polym. Sci. A 2004, 42, 4392.
[22] While the azide function is the limiting reagent for surface-based
reactions, it begins the reaction in vast excess at the surface in
comparison to the added alkyne and catalyst. Throughout the
reaction, the azide moieties are depleted while excess of the
alkyne molecules diffuse freely, presumably maintaining a
constant concentration. While these concentration effects are
difficult to mimic in traditional homogeneous reaction strategies,
the decreased reagent degrees of freedom affecting the potential
energy landscape for the surface reactions in comparison to
solution reactions introduces additional significant differences.
However, the flooding conditions in alkyne were adapted as a
reasonably close representation of the concentration conditions
at the surface.
[23] V. O. Rodionov, V. V. Fokin, M. G. Finn, Angew. Chem. 2005,
117, 2250; Angew. Chem. Int. Ed. 2005, 44, 2210.
[24] A kinetic plot similar to Figure 4b showing the full reaction time
for the Solution–Surface reaction is included in the Supporting
Information.
[25] T. Tsoncheva, S. Vankova, D. Mehandjiev, Fuel 2003, 82, 755.
[26] J. Iijima, J.-W. Lim, S.-H. Hong, S. Suzuki, K. Mimura, M. Isshiki,
Appl. Surf. Sci. 2006, 253, 2825.
[27] A freely diffusing particle will move about 1 micrometer in
1 millisecond.
[11] T. P. Sullivan, M. L. van Poll, P. Y. W. Dankers, W. T. S. Huck,
Angew. Chem. 2004, 116, 4286; Angew. Chem. Int. Ed. 2004, 43,
4190.
[12] a) M. D. Johnson, C. Orme, A. W. Hunt, D. Graff, J. Sudijono,
L. M. Sander, B. G. Orr, Phys. Rev. Lett. 1994, 72, 116; b) H. C.
Jeong, E. D. Williams, Surf. Sci. Rep. 1999, 34, 171 – 294; c) G. S.
Bales, D. C. Chrzan, Phys. Rev. B 1994, 50, 6057.
[13] a) C. W. Tornoe, C. Christensen, M. Meldal, J. Org. Chem. 2002,
67, 3057; b) V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B.
Sharpless, Angew. Chem. 2002, 114, 2708; Angew. Chem. Int. Ed.
2002, 41, 2596; c) J.-F. Lutz, Angew. Chem. 2007, 119, 1036;
[28] The resolution achieved using the Reagent-Stamping protocol
was similar to that of traditional microcontact printing for
spatially selective reactions (see refs. [8–10]).
9932 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 9927 –9932