Click chemistry by microcontact printing on self-assembled monolayers: A structure-reactivity study by fluorescence microscopy

Article abstract of DOI:10.1039/c1ob05187c

Microcontact printing (μCP) has developed into a powerful tool to functionalize surfaces with patterned molecular monolayers. μCP can also be used to induce a chemical reaction between a molecular ink and a self-assembled monolayer (SAM) in the nanoscale confinement between stamp and substrate. In this paper, we investigate the Huisgen 1,3-dipolar cycloaddition, the Diels-Alder cycloaddition and the thiol-ene/yne reaction induced by μCP. A range of fluorescent alkyne inks were printed on azide SAMs and fluorescence microscopy was used to monitor the extent of the 1,3-dipolar cycloaddition on a glass substrate. The rate of cycloaddition depends on the reactivity of the alkyne and on the presence of Cu(i). The cycloaddition is accelerated by Cu(i) but it also proceeds readily in the absence of Cu(i). In addition, a range of fluorescent diene inks were printed on alkene SAMs on glass. In this case, fluorescence microscopy was used to monitor the rate of the Diels-Alder cycloaddition as well as its retro-reaction. Finally, fluorescent thiol inks were printed on alkene SAMs on glass, and fluorescent alkenes and alkynes were printed on thiol SAMs. It is shown that reactions by μCP follow structure-reactivity relationships similar to solution reactions. Under optimized conditions all reactions lead to dense microarrays of addition products within minutes of printing time.

Full text of DOI:10.1039/c1ob05187c

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 4108
www.rsc.org/obc
PAPER
Click chemistry by microcontact printing on self-assembled monolayers:
A structure–reactivity study by fluorescence microscopy†
Jan Mehlich and Bart Jan Ravoo*
Received 2nd February 2011, Accepted 14th March 2011
DOI: 10.1039/c1ob05187c
Microcontact printing (mCP) has developed into a powerful tool to functionalize surfaces with
patterned molecular monolayers. mCP can also be used to induce a chemical reaction between a
molecular ink and a self-assembled monolayer (SAM) in the nanoscale confinement between stamp and
substrate. In this paper, we investigate the Huisgen 1,3-dipolar cycloaddition, the Diels–Alder
cycloaddition and the thiol-ene/yne reaction induced by mCP. A range of fluorescent alkyne inks were
printed on azide SAMs and fluorescence microscopy was used to monitor the extent of the 1,3-dipolar
cycloaddition on a glass substrate. The rate of cycloaddition depends on the reactivity of the alkyne and
on the presence of Cu(I). The cycloaddition is accelerated by Cu(I) but it also proceeds readily in the
absence of Cu(I). In addition, a range of fluorescent diene inks were printed on alkene SAMs on glass.
In this case, fluorescence microscopy was used to monitor the rate of the Diels–Alder cycloaddition as
well as its retro-reaction. Finally, fluorescent thiol inks were printed on alkene SAMs on glass, and
fluorescent alkenes and alkynes were printed on thiol SAMs. It is shown that reactions by mCP follow
structure–reactivity relationships similar to solution reactions. Under optimized conditions all
reactions lead to dense microarrays of addition products within minutes of printing time.
between an ink and a reactive SAM. For example, amides are
Introduction
formed when amines are printed on an anhydride SAM on gold.13
The modification of inorganic surfaces with monolayers of organic
It must be noted that this observation is fully consistent with the
molecules has found widespread application in nanofabrication,
rapid chemisorption in the contact area of stamp and substrate
as well as with the inherent reactivity of amines and anhydrides.
patterning of molecular monolayers is crucial to all of these
However, Huck and coworkers described the formation of peptides
sensing, diagnostics and molecular electronics.1–6 The microscale
applications. In recent years, microcontact printing (mCP) has
developed into a powerful tool to functionalize substrates with
spatially patterned molecular monolayers.7–10 mCP is based on
the selective transfer of ink molecules from a microstructured
elastomeric stamp to a substrate in the contact area between stamp
and substrate. mCP is commonly used to print monolayers of thiols
on gold surfaces.11,12 The sulfur atom binds to the gold surface and
a dense self-assembled monolayer (SAM) is formed within seconds
of contact time. However, mCP is not limited to printing thiols on
gold. It has been shown that – subject to a suitable modification of
stamp and substrate – virtually any type of ink can be patterned
by mCP. The transfer of ink molecules from stamp to substrate
is normally driven by a steep concentration gradient and it is
favoured by weak adhesion to the stamp and strong chemisorption
or physisorption to the substrate.
by printing N-protected amino acids onto an amine SAM on
gold.14 Of course, peptides do not spontaneously form from
carboxylic acids and amines under ambient conditions, and it
was proposed by Huck that “the nanoscale confinement of the ink
at the interface between the stamp and the SAM, in combination
with the pre-organization of the reactants in the SAM, facilitates
the formation of covalent bonds”. Subsequently, it was also shown
that imines can be formed under ambient conditions by printing
amines onto an aldehyde SAMs on gold.15 This useful variation
of mCP is also known as “reactive mCP”13 or “microcontact
chemistry”.16
Recently, it was shown that also the Huisgen 1,3-dipolar
cycloaddition of alkynes and azides can be induced by mCP.17
Using a range of sensitive surface analysis methods (including
infrared spectroscopy, X-ray photo-electron spectroscopy, atomic
force microscopy, and fluorescence microscopy), it was shown that
triazoles are formed within minutes when an alkyne is printed on
an azide SAM on a Si wafer or glass, even in the absence of
a Cu(I) catalyst that is normally used to accelerate this type of
“click chemistry”.18 Click chemistry by mCP was then applied to
print microarrays of alkyne-modified DNA19 and alkyne-modified
carbohydrates20,21 on azide SAMs on Si wafers and glass. It was
It was demonstrated in several early papers by Whitesides and
coworkers that mCP can also be used to induce a chemical reaction
Organic Chemistry Institute and CeNTech, Westfa¨lische Wilhelms Univer-
sita¨t Mu¨nster, Corrensstraße 40, 48149, Mu¨nster, Germany. E-mail: b.j.
ravoo@uni-muenster.de; Fax: +49 251 83 36557
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c1ob05187c
4108 | Org. Biomol. Chem., 2011, 9, 4108–4115
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
observed that although DNA could be printed without a Cu(I)
catalyst, the surface density of carbohydrates was low in the
absence of Cu(I) catalyst. Additionally, the heterogeneous catalysis
of the cycloaddition of alkynes and azides by mCP was reported: a
mCP stamp covered with a thin film of Cu (air-oxidized to Cu2O)
was used to induce the cycloaddition of alkynes on an azide SAM
on gold.22 It was demonstrated that the cycloaddition by mCP
proceeds to completion (i.e. until all reactive sites on the surface
are occupied) within a few hours when a Cu-coated stamp is used
or Cu(I) catalyst is added to the alkyne ink. Recently also the
Diels–Alder reaction as well as the photochemical thiol-ene/thiol
yne reaction have been exploited to fabricate patterned arrays of
carbohydrates by mCP.23,24
In this paper we report a structure–reactivity study of the
Huisgen 1,3-dipolar cycloaddition, the Diels–Alder cycloaddition
and the thiol-ene/yne reaction induced by mCP (Fig. 1). These
reactions have found widespread use for the bio-orthogonal
ligation of biomolecules to surfaces25–28 and their combination with
mCP results in an attractive method to prepare microarrays.23,24
Our method of choice to investigate these surface reactions
induced by mCP is fluorescence microscopy, since this is one of
very few methods that are sensitive enough to detect the monolayer
and sub-monolayer coverage of ink molecules on transparent sub-
strates under ambient conditions. The intensity contrast between
contact and non-contact regions serves as an internal reference
in each printing experiment. Since the microscopy settings can
be carefully controlled and reproduced, the intensity contrast can
be taken as a quantitative measure for the surface coverage. In
combination witha desorption experimentinwhichthe fluorescent
ink is removed from the surface, also the surface density (number
of molecules per unit area) can be determined. The overall aim of
this study is to gain insight into the kinetics of addition reactions
on surfaces induced by mCP. In the first part of this paper we
report a structure–reactivity study of the Huisgen 1,3-dipolar
cycloaddition induced by mCP. In the second part, we investigate
Diels–Alder reactions induced by mCP. Finally, in the third part
we investigate thiol-ene and thiol-yne reactions induced by mCP.
Results and discussion
1,3-Dipolar cycloaddition of alkynes and azides
In order to investigate the Huisgen 1,3-dipolar cycloaddition of
alkynes and azides induced by mCP, a set of alkyne inks was
synthesized (Scheme 1). The inks differ in the electron density of
Fig. 1 A: Schematic illustration of the principle of microcontact chem-
istry: A fluorescent ink is printed on a substrate SAM and a reaction occurs
exclusively in the area of contact. B: Huisgen 1,3-dipolar cycloaddition,
C: Diels–Alder cycloaddition, D and E: Thiol-ene reaction, F: Thiol-yne
reaction.
Scheme 1 Alkyne inks 1–5, fluorescent labels and azide SAM used for
the Huisgen 1,3-dipolar cycloaddition induced by mCP.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4108–4115 | 4109
©

View Article Online
the terminal alkyne, which results in different inherent reactivity in
the 1,3-dipolar cycloaddition with an aliphatic azide. According to
the Frontier Molecular Orbital (FMO) description of this reaction
an electron-rich alkyne (such as aliphatic alkyne 1) is less reactive
than an electron-poor alkyne conjugated to carbonyl groups such
as amide 2, esters 3 and 4, and ketone 5. Based on FMO arguments,
it is expected that the order of reactivity of alkynes 1–5 with
aliphatic azides is: 5 > 4 ~ 3 > 2 ꢀ 1.29 All inks were labelled with
a fluorophore so that they can be easily detected by fluorescence
microscopy. To verify if the structure of the fluorophore affects the
cycloaddition, inks 1–5 were labelled with the bulky, red-emitting
lissamine rhodamine B (1 – 5LR) as well as with the compact,
blue-emitting N-methylanthranyl amide (1 – 5AA). The synthesis
of inks 1–5 is described in the ESI.†
The alkyne inks were printed on an azide-terminated SAM
on glass. The azide SAM was prepared according to a literature
procedure.30 In short, a cleaned and activated microscopy cover
slide was coated with a SAM of 11-bromoundecyltrichlorosilane.
Formation of the 11-bromoundecylsilane SAM was followed by
a substitution reaction with sodium azide. Due to the short
immobilization time (20 min), the surface coverage of the 11-
bromoundecylsilane SAM is substantially lower than a densely
packed alkylsilane SAM. On the basis of literature data for the
formation of octadecyltrichlorosilane SAMs, we estimate that the
surface coverage of the 11-bromoundecylsilane SAM is around
1 molecule/nm2 or 1 ¥ 1014 molecules/cm2.31,32,33 The advantage
of a lower density of reactive groups is a reduced steric hindrance,
a better access for incoming reagents, and hence a more rapid
conversion than would be expected for a densely packed SAM.
Elastomeric polydimethylsiloxane (PDMS) stamps for mCP
were prepared according to established procedures (see experimen-
tal section for details). Stamps of 1 ¥ 1 cm2 were structured with
stripes in two different dimensions due to the design of the available
master (5 mm width and 3 mm interspace or 10 mm width and 5 mm
interspace). The microstructure of the stamp did not give rise to
significant differences during printing. In view of the polarity of
the ink molecules, the PDMSstamps were activated by oxidation in
an ozone generator prior to printing. Oxidation of the PDMS sur-
face results in significantly increased adsorption of the ethanolic
ink solution to the stamp surface, and hence enhanced transfer of
ink molecules from the stamp to the substrate during printing.8–10
All alkyne inks were printed and imaged under identical
conditions: inking time 1 min, ink concentration 1 mM in ethanol,
20 g weight during printing, identical rinsing procedure following
printing, identical microscopy and camera settings. The only
variable parameters were the structure of the ink and the printing
time. Fig. 2 shows a set of images obtained by printing alkyne
1LR on an azide SAM substrate for different printing times as
well as the brightness contrast between contact and non-contact
area observed for each printing time. The brightness of the stripes
in the contact area was taken as a measure for the amount of
ink that reacted on the substrate during printing. The dark area in
between the stripes (i.e. the non-contacted area) is taken as internal
reference. This experimental procedure was repeated several times
for all rhodamine inks 1 – 5LR in order to exclude experimental
errors. In addition, ink 1LR was also printed in the presence of
catalytic amounts of Cu(I) in the ethanolic ink solution. The results
are summarised in a plot of brightness contrast against printing
time in Fig. 3.
Fig. 2 Microscopy images and contrast profiles for printing fluorescent
alkyne 1LR on an azide SAM for increasing printing time.
Fig. 3 Alkyne inks 1-5LR and 1LR + Cu(I) printed on azide SAMs.
Printing time dependence of the fluorescence intensity. All printing
experiments were carried out under identical conditions. Trend lines serve
to guide the eye.
A clear trend is obvious from the curves collected in Fig. 3. First
of all, it is reasonable to assume that the fluorescence brightness
contrast scales with the surface density of fluorescent molecules
on the surface. In line with this assumption, all curves for inks
1 – 5LR display a similar maximum intensity contrast, which
may be taken as the maximum coverage of alkyne molecules that
can be immobilized on the azide SAM by mCP. Furthermore, it
is clear from Fig. 2 that all surface reactions induced by mCP
are complete within 15 min. However, a distinct reactivity trend
of the inks (1 < 2 < 3 ª 4 < 5) is also evident, demonstrating
that the cycloaddition induced by mCP follows the trend predicted
by FMO theory. Moreover, the reaction of 1LR in the presence
of Cu(I) catalyst is much faster than all other reactions, but
the maximum density of fluorescent molecules on the surface
is not significantly higher. Obviously the limiting factor also in
this “click” immobilization reaction is the availability of azide on
the substrate. To aid a straightforward comparison, the reactivity
curves in Fig. 3 were converted into an approximate half life of the
mCP induced cycloaddition, assuming pseudo first order kinetics
for the reaction of the azide SAM with an excess of alkyne ink
on the stamp. The half life of the cycloaddition for each ink is
4110 | Org. Biomol. Chem., 2011, 9, 4108–4115
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 1 Half life (min) of the reactions of alkyne inks 1–5LR and 1–5AA
on azide SAMs
Table 2 Half life (min) of the reactions of diene inks 6–8 on dienophile
SAMs 9 and 10
1
2
3
4
5
1 + Cu(I)
6
7
8
RL
AA
6.3
6.6
5.7
5.9
4.3
4.2
3.8
3.9
2.8
3.0
0.4
0.5
9
10
>20
16.3
9.4
6.0
5.5
4.7
listed in Table 1. From Table 1 it can be seen that for example
ketone 5 reacts approximately 2 times faster than aliphatic alkyne
1, and that the addition of Cu(I) catalyst to alkyne 1 leads to an
acceleration of the cycloaddition by a factor of 15. Finally, in order
to exclude effects of the bulky rhodamine fluorescent label on the
cycloaddition by mCP, the same experiments were carried out with
the inks 1–5AA, equipped with a much smaller fluorescent N-
methylanthranyl amide label. By using the experimental protocol
described above, the half life for the mCP induced cycloaddition
reactions of the blue alkyne inks can be obtained. They are also
shown in Table 1. The similarity of the results is obvious from the
half life. We conclude that the fluorescent labels do not affect the
surface reaction.
electron demand Diels–Alder reaction, furan 6 should have the
lowest reactivity, pentadiene 7 should have intermediate reactivity,
and cyclopentadiene 8 should have the highest reactivity towards
a given dienophile.34,35 As a substrate for mCP, SAMs terminated
with an alkene (9) and an acrylamide (10) were prepared (see
experimental section). Alkene SAM 9 was prepared directly from
7-octenyltriethoxysilane. Acrylamide SAM 10 was prepared in a
three step procedure by reduction of an azido SAM (see above)
followed by reaction with acryloylchloride. The surface coverage
of these SAMs is around 1 ¥ 1014 molecules/cm2.31,32,33
As described above for the 1,3-dipolar cycloaddition, all diene
inks were printed and imaged under identical conditions: inking
time 1 min, ink concentration 1 mM in ethanol, 20 g weight during
printing, identical rinsing procedure following printing, identical
microscopy and camera settings. The only variable parameter was
the structure of the ink, the substrate SAM, and the printing time.
After printing each ink 6–8 the substrate was carefully rinsed and
fluorescence images were evaluated. The brightness of the stripes
in the contact area was taken as a measure for the amount of
diene ink that reacted on the dienophile surface during printing.
The dark area in between the stripes (i.e. the non-contacted area) is
taken as internal reference. The observed brightness contrast as a
function of printing time is shown for mCP of dienes 6–8 on alkene
SAM 9 in Fig. 4. It can be seen from Fig. 4 that the Diels–Alder
reaction of dienes 6–8 with SAM 9 follows the reactivity trend
predicted by FMO theory (8 > 7 > 6). However, no maximum
contrast is reached within 20 min for any of the inks, indicating
that these Diels–Alder reactions proceed significantly slower than
the Huisgen 1,3-dipolar cycloadditions described above. Longer
printing times are not practical in view of ink diffusion into the
non-contacted area. Table 2 displays the results in terms of an
approximate half life for the Diels–Alder reactions by mCP. The
maximum contrast obtained for 8 printed on 9 was taken as
reference for complete conversion.
Diels–Alder reaction
We also performed a reactivity study of the Diels–Alder reaction
induced by mCP of fluorescent dienes on alkene SAMs using a simi-
lar procedure as described above for the 1,3-dipolar cycloaddition.
However, in this case not only the reactivity of the ink molecules
but also the reactivity of the substrates was varied systematically.
Depending on the nature of the diene and dienophile, the Diels–
Alder reaction can be reversible and much more temperature
dependent than the 1,3-dipolar cycloaddition of alkynes and
azides. Diene derivatives 6, 7 and 8 were used as molecular inks
(Scheme 2). These dienes were labelled with rhodamine lissamine
B. The synthesis of inks 6–8 is described in the ESI.† In a normal
Fig. 4 Diene inks 6–8 printed on dienophile SAM 9. Printing time
dependence of the fluorescence intensity. All printing experiments carried
out under identical conditions. Trend lines serve to guide the eye.
Scheme 2 Diene inks and dienophile SAMs used for Diels–Alder
reactions induced by mCP.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4108–4115 | 4111
©

View Article Online
A comparable result is obtained when the substrate instead of
the ink is varied in reactivity. Fig. 5 shows the increase in contrast
with time resulting from printing pentadiene 7 on substrate SAMs
9 and 10. Although the differences are modest, a significantly
higher reactivity can be observed for 10. These experiments are
also included in Table 2 in terms of an approximate half life for
the Diels–Alder reaction by mCP.
Fig. 6 Temperature dependence of a Diels–Alder reaction (diene ink 7
printed on dienophile SAM 9) and 1,3-dipolar cycloaddition (alkyne ink
1 printed on azide SAM), printed for 10 min in all cases. Trend lines serve
to guide the eye.
Fig. 5 Diene ink 7 printed on dienophile SAMs 9 and 10. Printing time
dependence of the fluorescence intensity. All printing experiments carried
out under identical conditions.
A comparison of each of the three diene inks printed on each of
the dienophile SAMs can be made on the basis of the approximate
half life to reach maximum surface coverage. To calculate the
half life, the maximum contrast obtained for diene 8 printed on
dienophile 10 was taken as reference since this was the combination
of ink and substrate that gave the highest fluorescence contrast,
i.e. the highest surface density of ink, as a result of a rapid Diels–
Alder reaction between electron-rich cyclopentadiene 8 printed on
electron-poor acrylamide SAM 10. It can be seen from Table 2 that
furan 6 reacts only slowly with any of the dienophile SAMs 9 and
10. We note that these findings are fully in line with FMO theory
as well as experimental rate constants of Diels–Alder reactions in
solution.34,35
Fig. 7 Fluorescence images and contrast profiles for diene 7 printed on
dienophile SAM 9 for 30 min followed by treatment with boiling ethanol
for 10, 30 and 60 min.
(i.e. cycloaddition product obtained by mCP) become less and
less fluorescent the longer they are treated with boiling ethanol.
This observation implies that the surface density of fluorescent ink
decreases with time. The desorption of ink in boiling ethanol can
be readily explained by a retro–Diels–Alder reaction. The rate of
the reverse reaction is different for the three diene inks (Fig. 8).
Whereas the Diels–Alder reaction is completely reversible within
10 min in boiling ethanol for furan 6 printed on SAM 9, it is hardly
reversible for cyclopentadiene 8. We note that also these findings
are fully in line with FMO theory.
Temperature dependence and retro-Diels–Alder reaction
The temperature dependence of the Huisgen 1,3-dipolar cycload-
dition as well as the Diels–Alder reaction induced by mCP was
investigated by printing alkyne 1 on an azide SAM and by
printing pentadiene 7 on alkene SAM 9 at a range of temperatures
(Fig. 6). The printing time was 10 min in each case. It can
clearly be observed that there is a strong temperature effect on
the Diels–Alder reaction whereas the mCP-induced 1,3-dipolar
cycloaddition is hardly influenced. This finding is in accordance
with the literature on the corresponding reactions in solution.29,34,35
The absence of a large temperature effect on the 1,3-dipolar
cycloaddition also demonstrates that the temperature effect on the
Diels–Alder reaction cannot be explained by temperature effects
on the mCP process, such as faster diffusion of the ink at higher
temperature.
Fig. 8 Retro–Diels–Alder reaction in boiling ethanol of dienes 6–8
printed on dienophile SAM 9. Trend lines serve to guide the eye.
The reversibility of the Diels–Alder reaction can be exploited to
perform a second type of reactivity study. To this end, diene inks
6–8 were printed for 30 min on alkene substrates 9, rinsed as usual
and then dipped into boiling ethanol. Fig. 7 shows the results
for 7 printed on 9 as an example. The bright fluorescent stripes
The reversibility of the Diels–Alder reaction was exploited in
order to determine the surface density of printed diene ink. After
printing diene 7 with a large, flat, unstructured stamp on an
4112 | Org. Biomol. Chem., 2011, 9, 4108–4115
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
alkene SAM 9 on a substrate with defined size (1.8 ¥ 1.8 cm,
3.24 cm2) and extensive rinsing with cold ethanol, the retro–
Diels–Alder reaction was induced by treating the samples with
boiling ethanol. The concentration of rhodamine desorbed from
the substrate was measured with a fluorescence spectrometer and
compared to a calibration curve for rhodamine solutions with
known concentrations. It was found that on average 4 nmol or 6 ¥
1013 fluorescent molecules desorbed from the substrate. Assuming
that this number equals the number of diene molecules that were
attached to the alkene surface as a result of mCP, the surface density
of immobilized ink was 1.8 ¥ 1013 molecules/cm2. As discussed
above, the surface density of the substrate SAM was around 1 ¥
1014 molecules/cm2. Hence, mCP results in the formation of a
rather dense layer of cycloaddition product on top of the substrate
SAM. It is likely that the density of the cycloaddition product is
limited by the size and polarity of the rhodamine inks rather than
the surface coverage of the reactive groups on the substrate SAM.
Unfortunately the desorption experiment cannot be performed
for the 1,3-dipolar cycloaddition because this reaction is not
thermally reversible. However, if one compares (under identical
microscopy and camera settings) the maximum brightness ob-
tained when printing alkynes on azide SAMs with the maximum
obtained when printing dienes on dienophile SAMs, it appears that
the surface coverage as a result of the 1,3-dipolar cycloaddition
is even higher than the surface coverage as a result of the DA
reaction. Hence, there is no doubt that mCP induced 1,3-dipolar
cycloadditions results in a dense microarray of cycloaddition
product, even in the absence of Cu(I) catalyst.22
Scheme 3 Inks and SAMs used for thiol-ene reactions induced by mCP.
Thiol-ene and thiol-yne reaction
The photo-induced thiol-ene addition reaction (and to a lesser
extent, the thiol-yne reaction) has had a major impact on chemistry
and materials science in the last few years.36 We have recently
shown that these reactions can also be used to prepare microarrays
by photochemical mCP.24 In this article, we report a structure–
reactivity study for a range of thiols and alkenes or alkynes
undergoing a photo-induced thiol-ene or thiol-yne reaction,
respectively. An interesting aspect of this particular addition
reaction is to verify whether it can be induced by mCP without UV
irradiation, similar to the rapid Huisgen 1,3-dipolar cycloaddition
without Cu(I) induced by mCP. To this end, fluorescent thiols were
printed on alkene SAMs and fluorescent alkenes were printed on
thiol SAMs. Scheme 3 shows an overview of all used inks and
SAMs investigated in thiol-ene reactions induced by mCP. The
synthesis of inks 7 and 11–14 is described in the ESI.†
Alkene inks 7, 11 and 12 were printed onto thiol SAM 15.
Furthermore, thiol inks 13 and 14 were printed onto alkene
SAM 16. The experimental procedure was essentially the same
as described for the cycloaddition reactions described above.
However, an UV diode with maximum emission at 365 nm was
placed in a small chamber with the printing setup and the interface
of stamp and substrate was irradiated during printing. Fig. 9 shows
the results of printing inks 7, 11, and 12 onto SAM 15. Allylic ink
11 is expected to be the least reactive ink, acrylic ester ink 12
is the most reactive, while reactivity of the conjugated diene 7 is
between the other two.37 This is found also in this study and can
be quantified by an approximate half life (2.0 min for 11, 0.9 min
for 12, 1.2 min for 7).
Fig. 9 Alkene inks 7, 11 and 12 printed on thiol SAM 15. Printing time
dependence of the fluorescence intensity. All printing experiments carried
out under UV irradiation. Trend lines serve to guide the eye.
Printing the thiol inks 13 and 14 onto alkene SAM 16 also
showed a structure–reactivity relationship as expected. No signif-
icant difference between the reaction rate of aminoethanethiol 13
and aminothiophenol 14 could be observed: the thiol-ene reaction
proceeds very fast in either case (Fig. 10). Ink 14 was printed
on alkene SAM 16 using various irradiation conditions. Under
irradiation with UV light the reaction rate is high as expected, at
ambient light conditions the yield is still remarkably high, while
in the dark the reaction proceeds only very slowly (Fig. 10).
Finally, the alkyne inks 1–5 used for the investigation of the
Huisgen 1,3-dipolar cycloaddition were used for the thiol-yne
reaction by printing them onto thiol SAM 15. Their relative
reactivity is the same in both reactions, i.e. electron-poor alkynes
react faster than electron-rich alkynes (5 > 4 ~ 3 > 2 ꢀ 1). The
results of the printing experiments are summarised in Fig. 11.
Although a trend confirming the expected structure–reactivity
relationship is visible, it was found that all reactions proceed
rapidly to completion. In every case, the half life is less than 1 min
and it is difficult to quantify the reactivity of the alkynes.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4108–4115 | 4113
©

View Article Online
Experimental section
Ink synthesis
Fluorescent alkyne, diene, alkene and thiol derivatives were
synthesised following literature methods. Experimental details and
spectroscopic data are provided in the ESI.†
Surface modification
The azide terminated SAM was obtained following a method
first described by Balachander and Sukenik.30 A glass substrate
R
(microscopy cover slides, servopraxꢀ, Wesel (Germany)) was
cleaned and activated by treatment with hot “piranha” solution
(conc. H2SO4/30% H2O2, 3 : 1) for at least 30 min. The substrate
was dipped into a solution of 11-bromoundecyltrichlorosilane
(ABCR) in toluene (0.1%) for 20 min at room temperature
Fig. 10 Thiol inks 13 and 14 printed onto alkene SAM 16. Printing time
dependence of the fluorescence intensity under varying light conditions.
Trend lines serve to guide the eye.
◦
followed by storing the dried substrates at 80 C for at least 2 h.
The samples are then brought into a saturated solution of NaN3
◦
in DMF at 70 C for 48 h. After sonication for 30 s and rinsing
with ethanol the substrates are dried and stored in air-tight boxes.
The quality of the monolayer formation and transformation was
verified by contact angle mea◦surements (adv. angle (lit.): 82◦
2
(77◦ 2),30 rec. angle (lit.): 74 2 (71◦ 2)30).
The alkene terminated SAMs 9 and 16 were formed by dipping
the piranha-treated glass substrates into a 0.1% solution of 7-
octenyltriethoxysilane (ABCR), or 10-undecenyltrichlorosilane,
respectively, in toluene for 2 h at 50 ◦C. Contact angle: adv. (lit.):
115◦ 3 (111◦ 4),23 rec. (lit.): 91◦ 2 (88◦ 3).23
To obtain the alkene terminated amide SAM 10, an azide
terminated SAM was reduced to an amine SAM with a solution of
PPh3 in dry THF (0.1 M) for 3 h and then dipped into a solution
of acrylic acid (0.02 M), DCC (9.5mM) and a catalytic amount of
DMAP in ethyl acetate for 48 h at RT. Contact angle: adv.: 117◦
4, rec.: 77◦ 4.
Fig. 11 Alkyne inks 1–5 printed on thiol SAM 15. Printing time
dependence of the fluorescence intensity. All printing experiments carried
out under UV irradiation. Trend lines serve to guide the eye.
Thiol terminated SAM 15 was prepared following a protocol by
Balachander and Sukenik.30 Piranha-activated glass substrates are
equipped with a monolayer of 11-bromoundecyltrichlorosilane by
dipping them into a solution of that substance in toluene (0.1%)
for 20 min at room temperature followed by storing the dried
substrates at 80 ◦C for at least 2 h. These substrates are dipped
into a solution of potassium thiocyanate in DMF (10 mg mL-1) for
20 h. The resulting SCN-terminated monolayer was transformed
to a thiol-SAM by treatment with LiAlH4 for 4 h. Contact angle:
adv. (lit.): 72◦ 2 (71◦ 3),30 rec. (lit.): 45◦ 4 (49◦ 3).30
Conclusion
The Huisgen 1,3-dipolar cycloaddition, Diels–Alder cycloaddi-
tion, and thiol-ene/yne reaction can be induced by mCP of alkynes
on an azide SAM, dienes on a dienophile SAM, and thiols on
an alkene or alkyne SAM (and vice versa), respectively. All three
addition reactions result in a dense surface coverage within a few
minutes, but reaction by mCP is fastest for the most reactive reagent
pairs. The Huisgen 1,3-dipolar cycloaddition can be significantly
accelerated by Cu(I) catalysis while the thiol-ene/yne addition is
particularly effective under UV irradiation. Under optimized con-
ditions, the Cu(I) catalyzed 1,3-dipolar cycloaddition of alkynes
and azides and the photo-induced thiol-ene/yne reactions are
substantially faster than the Cu(I)-free Huisgen cycloaddition and
the Diels–Alder reaction. The characteristic structure–reactivity
relationships of the addition reactions as well as the retro-Diels–
Alder reaction can be taken as evidence that in all cases a chemical
reaction (rather than physisorption) is induced by mCP. In many
respects, reactions induced by mCP follow the principles of click
chemistry (i.e. near-quantitative yield, mild reaction conditions,
and short reaction time) and thus provide a straightforward
method for the fabrication of (bio)molecular microarrays for a
range of applications.
PDMS stamps
The PDMS stamps were prepared by mixing PDMS (obtained
R
from Dow Corningꢀ as Sylgard 184) with a curing agent (10 : 1)
followed by degassing and pouring the bubble free mixture onto
a 4¢¢ silicon wafer carrying 24 8 ¥ 8 mm areas of the desired
pattern (produced by common photolithographic methods). The
prepolymer was hardened by baking it overnight at 70 ◦C and
the stamp was peeled off the master carefully. After cutting it
into the 24 single stamps these were oxidised in an UV/ozone
generator (Novascan PSD-UV) and stored in MilliQ water to
prevent hydrophobic recovery.
4114 | Org. Biomol. Chem., 2011, 9, 4108–4115
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Printing procedure
3 D. B. Weibel, P. Garstecki and G. M. Whitesides, Curr. Opin. Neurobiol.,
2005, 15, 560–567.
4 M. J. W. Ludden, J. K. Sinha, G. Wittstock, D. N. Reinhoudt and J.
Huskens, Org. Biomol. Chem., 2008, 6, 1553–1557.
5 K. Ichimura, S. K. Oh and M. Nakagawa, Science, 2000, 288, 1624–
1626.
6 M. Kleinert, T. Winkler, A. Terfort and T. K. Lindhorst, Org. Biomol.
Chem., 2008, 6, 2118–2132.
7 Y. Xia and G. M. Whitesides, Angew. Chem., Int. Ed., 1998, 37,
550.
8 S. A. Ruiz and C. S. Chen, Soft Matter, 2007, 3, 168.
9 A. Perl, D. N. Reinhoudt and J. Huskens, Adv. Mater., 2009, 21,
2257.
10 T. Kaufmann and B. J. Ravoo, Polym. Chem., 2010, 1, 371.
11 A. Kumar and G. M. Whitesides, Appl. Phys. Lett., 1993, 63,
2002–2004.
The dry stamp was inked for 1 min by placing drops of ink on
the patterned side of the stamp or by dipping the stamp into ink
solution. The stamp was dried with compressed air and placed
with the patterned side onto the cleaned and dried substrate. After
contact any further movement is avoided. A weight of 20 g is
put on top of the stamp to ensure conformal contact. After the
desired printing time the stamp plus weight was removed in one
move. The substrate was rinsed vigorously with ethanol, sonicated
for 30 s and rinsed again with ethanol to remove all physisorbed
ink and contamination. The printed substrates were investigated
with a fluorescence microscope (Olympus) and micrographs are
taken with a digital camera (Kappa). Identical parameter settings
(exposure time 990 ms, gain 180, Gamma 100) are used in order to
obtain images that can be quantitatively compared and evaluated
with ImageJ software. Printing at different temperatures was
carried out on a metal plate placed on a isopropanol/dry ice
mixture (-80 ◦C), on a ice/salt mixture (-15 ◦C) or on paraffin oil
heated to 70 ◦C. The substrate was placed on the plate for several
min to make sure it has the same temperature as the plate. For
12 J. L. Wilbur, A. Kumar, E. Kim and G. M. Whitesides, Adv. Mater.,
1994, 6, 600–604.
13 L. Yan, W. T. S. Huck, X. M. Zhao and G. M. Whitesides, Langmuir,
1999, 15, 1208–1214.
14 T. P. Sullivan, M. L. van Poll, P. Y. W. Dankers and W. T. S. Huck,
Angew. Chem., Int. Ed., 2004, 43, 4190–4193.
15 D. I. Rozkiewicz, B. J. Ravoo and D. N. Reinhoudt, Langmuir, 2005,
21, 6337.
16 B. J. Ravoo, J. Mater. Chem., 2009, 19, 8902–8906.
17 D. I. Rozkiewicz, D. Jancewski, W. Verboom, B. J. Ravoo and D. N.
Reinhoudt, Angew. Chem., Int. Ed., 2006, 45, 5292.
18 H. C. Kolb, M. G. Finn and K. B. Sharpless, Angew. Chem., Int. Ed.,
2001, 40, 2004–2021.
◦
temperatures below 0 C condensation of water was avoided by
placing the setup in dry argon atmosphere.
19 D. I. Rozkiewicz, J. Gierlich, G. A. Burley, K. Gutsmiedl, T. Carell, B.
J. Ravoo and D. N. Reinhoudt, ChemBioChem, 2007, 8, 1997–2002.
20 O. Michel and B. J. Ravoo, Langmuir, 2008, 24, 12116–12118.
21 K. Godula, D. Rabuka, K. Nam and C. R. Bertozzi, Angew. Chem.,
Int. Ed., 2009, 48, 4973–4976.
22 J. M. Spruell, B. A. Sheriff, D. I. Rozkiewicz, W. R. Dichtel, R. D.
Rohde, D. N. Reinhoudt, J. F. Stoddart and J. R. Heath, Angew. Chem.,
Int. Ed., 2008, 47, 9927–9932.
Retro-Diels–Alder reaction
After printing the inks on the SAMs according to the standard
procedure described above for 30 min and taking the fluorescence
images the samples are dipped into boiling ethanol. After 10, 30
and 60 min of treatment the samples are taken out, rinsed with
ethanol, dried and investigated with fluorescence microscopy.
23 C. Wendeln, A. Heile, H. F. Arlinghaus and B. J. Ravoo, Langmuir,
2010, 26, 4933–4940.
24 C. Wendeln, S. Rinnen, C. Schulz, H. F. Arlinghaus and B. J. Ravoo,
Langmuir, 2010, 26, 15966–15971.
Surface density determination
25 A. D. de Araujo, J. M. Palomo, J. Cramer, M. Kohn, H. Schroder, R.
Wacker, C. M. Niemeyer, K. Alexandrov and H. Waldmann, Angew.
Chem., Int. Ed., 2006, 45, 296–301.
26 P. Jonkheijm, D. Weinrich, M. Koehn, H. Engelkamp, P. C. M.
Christianen, J. Kuhlmann, J. C. Maan, D. Nuesse, H. Schroeder, R.
Wacker, R. Breinbauer, C. M. Niemeyer and H. Waldmann, Angew.
Chem., Int. Ed., 2008, 47, 4421–4424.
27 L. S. Wong, F. Khan and J. Micklefield, Chem. Rev., 2009, 109, 4025–
4053.
28 N. Gupta, B. F. Lin, L. Campos, M. D. Dimitriou, S. T. Hikita, N. D.
Treat, M. V. Tirrell, D. O. Clegg, E. J. Kramer and C. J. Hawker, Nat.
Chem., 2010, 2, 138–145.
Pentadiene ink 7 was printed on alkene SAM 9 (18 ¥ 18mm glass
slide) with a flat stamp bigger than the slide for 30 min. The
substrate was rinsed vigorously with ethanol and sonicated for 30 s.
The substrate is placed in approx. 15 mL boiling ethanol under
reflux for 2 h, then taken out and rinsed with a small amount of
ethanol which is added to the boiling ethanol. The exact volume
of this solution is determined and then the fluorescence intensity
at 585 nm (lmax of rhodamine lissamine B) is measured with a
fluorescence spectrometer and compared to a calibration curve
which is obtained from solutions with known concentrations of
lissamine rhodamine B.
29 R. Huisgen, Angew. Chem., 1963, 75, 742–754.
30 N. Balachander and C. N. Sukenik, Langmuir, 1990, 6, 1621–1627.
31 S. Semal, M. Voue’, M. J. de Ruijter, J. Dehuit and J. De Coninck, J.
Phys. Chem. B, 1999, 103, 4854–4861.
32 Y. Liu, L. K. Wolf and M. C. Messmer, Langmuir, 2001, 17, 4329–
Acknowledgement
4335.
33 S. Onclin, B. J. Ravoo and D. N. Reinhoudt, Angew. Chem., Int. Ed.,
2005, 44, 6282–6304.
The authors are grateful for financial support by the Deutsche
Forschungsgemeinschaft (IRTG 1143 Mu¨nster-Nagoya).
34 V. D. Kiselev and A. I. Konovalov, J. Phys. Org. Chem., 2009, 22,
466–483.
35 J. Sauer and R. Sustmann, Angew. Chem., Int. Ed. Engl., 1980, 19,
779–807.
References
1 R. J. Hamers, Annu. Rev. Anal. Chem., 2008, 1, 707–736.
2 A. Natan, L. Kronik, H. Haick and R. T. Tung, Adv. Mater., 2007, 19,
4103–4117.
36 C. E. Hoyle and C. N. Bowman, Angew. Chem., Int. Ed., 2010, 49,
1540–1573.
37 K. Griesbaum, Angew. Chem., 1970, 82, 276–287.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4108–4115 | 4115
©