Patterning of a surface immobilized ATRP initiator with an inkjet printer

Article abstract of DOI:10.1021/ma902836n

A new technique for patterning polymer brushes on the micrometer scale has been developed in which an inkjet printer was used to deposit droplets of acid on a surface-immobilized initiator for atom transfer radical polymerization (ATRP). The acid cleaved an ester bond in the ATRP initiator in a saponification reaction. As a result, the ATRP initiator was rendered inactive. To control the degree of defunctionalization, a new initiator containing a weak ester bond was derived from a tertiary alcohol. Comparison to an established ATRP initiator, derived from a primary alcohol, showed that the novel initiator was defunctionalized with a higher efficiency. Control of the reaction time allowed to partially defunctionalize the initiator molecules, leading to control of grafting densities within the written patterns.

Full text of DOI:10.1021/ma902836n

Macromolecules 2010, 43, 5033–5042 5033
DOI: 10.1021/ma902836n
Patterning of a Surface Immobilized ATRP Initiator with an Inkjet Printer
†
,‡
†
Sebastian G. J. Emmerling, Laura B. N. Langer, Sascha A. Pihan, Philipp Lellig, and
Jochen S. Gutmann*
†
†
†,‡
,‡
Max Planck Institute for Polymer Research, Ackermannweg 10, D-55128 Mainz, Germany, and
Institute of Physical Chemistry, Johannes Gutenberg University Mainz, Jakob-Welder-Weg 11,
‡
D-55128 Mainz, Germany
Received December 23, 2009; Revised Manuscript Received March 28, 2010
ABSTRACT: A new technique for patterning polymer brushes on the micrometer scale has been developed
in which an inkjet printer was used to deposit droplets of acid on a surface-immobilized initiator for atom
transfer radical polymerization (ATRP). The acid cleaved an ester bond in the ATRP initiator in a
saponification reaction. As a result, the ATRP initiator was rendered inactive. To control the degree of
defunctionalization, a new initiator containing a weak ester bond was derived from a tertiary alcohol.
Comparison to an established ATRP initiator, derived from a primary alcohol, showed that the novel
initiator was defunctionalized with a higher efficiency. Control of the reaction time allowed to partially
defunctionalize the initiator molecules, leading to control of grafting densities within the written patterns.
Introduction
Soon after the initial interest in polymer brushes, researchers
no longer solely focused on covering entire surfaces with polymer
Fabrication of polymer brush patterns over large areas re-
mains a demanding task, despite recent achievements using
photolithography and microcontact printing (μCP). A com-
bination of templating techniques and polymer brush synthesis
techniques to amplify the patterns is essential. In this publication
we will introduce a versatile fabrication method to pattern
polymer brushes on the micrometer scale. This process focuses
on flexibility regarding choice of substrates, control of brush
length, and the possibility for automation. Among the existing
techniques, μ-CP would be the most suitable, but its automation
is not easily achieved. We resorted to inkjet printing, circumvent-
ing the use of masters and stamps, instead directly writing our
patterns on the substrate. Additionally, the new technique
enabled us to create gradients in grafting densities over large
areas.
brushes. In 1995 R u€ he and co-workers published three different
2
0
1
2,3
methods for achieving lateral patterning of a polymer coating.
R u€ he’s group recognized from the beginning the potential
4
application for these patterned surfaces that control the adhesion
2
1
of biological cells to solid substrates. Research is still focusing
22
on biological applications in this area. Photolithography has
2
3
also been investigated for patterning by the Hawker group. In
1
addition to deep and near-UV irradiation, electron beams have
been deployed. Both these approaches require the use of shadow-
ing masks if larger areas are to be patterned. This is a drawback to
keeping production costs low. The major advantage of using
shadowing masks, however, is the resolution in the nanometer
range. Alongside the development of photolithography and
e-beam (chemical) lithography (EBL), various other lithography
techniques for pattern formation have been developed. This is
Access to brushes with high grafting density has become avail-
able through surface-initiated polymerization (SIP) techniques,
also commonly referred to as “grafting from”. Controlled free-
24
5
highlighted in a recent review. Under the term scanning probe
2
5
6
lithography (SPL) are grouped techniques like nanoshaving
26
and dip-pen nanolithography. As the name suggests, these
techniques rely on the use of atomic force microscopes (AFM)
to manipulate SAMs. The popularity of SPL-techniques stem
from the fact that they are technologically simpler compared to
EBL, while maintaining a resolution close to that of EBL. When
it comes to patterning of large areas, SPL techniques are at
disadvantage. This could potentially be mitigated by the use of
cantilever arrays. Soft lithography has attracted by far the most
interest among all techniques available for creating laterally
structured brushes. Developed in the 1990s by Whitesides and
radical polymerization techniques, such as nitroxide-mediated
7
radical polymerization (NMP), reversible addition-fragmenta-
8,9
tion chain transfer polymerization, (RAFT) and atom transfer
10,11
radical polymerization
(ATRP), are frequently employed. All
these techniques have a wide range of polymerizable monomers in
common. Even monomers carrying functional groups can be
polymerized. Another advantage is the possibility of creating block
1
2
copolymer brushes by the reactivation of dormant chain ends in
the presence of an additional monomer. Further, transformation of
the functional end groups is feasible. ATRP is set apart from the
other techniques in a number of ways. Not only has surface-
initiated ATRP (SI-ATRP) been reported on a broad range of
2
7,28
co-workers,
soft lithography using polymer stamps to deliver
patterns was utilized to create initiator functionalized micropat-
terns for SIP. This technique is also known as microcontact
printing (μCP). A thiol containing an initiator moiety for ring-
1
3,14
15,16
substrates, viz., silicon wafers,
polymer films, and others.
silica (nano)particles,
flexible
2
17,18
In addition this technique can be
2
9
30-33
19
opening polymerization (ROP) or ATRP
is delivered with
performed under mild conditions at low temperatures, and most
of the reagents are commercially available. A common approach is
the covalent self-assembly of initiator species on a substrate.
a stamp onto a gold-covered substrate. Then the uncovered
regions are backfilled with a nonfunctional thiol or vice versa.
3
4
Huck and co-workers developed this approach further by
repeating cycles of printing initiator thiols, ATRP thereof, and
passivating the “living” chain ends by nucleophilic substitution
*Corresponding author: e-mail gutmann@mpip-mainz.mpg.de; Ph
þ49-6131-379117; Fax þ49-6131-379100.
r 2010 American Chemical Society
Published on Web 05/14/2010
pubs.acs.org/Macromolecules

5
034 Macromolecules, Vol. 43, No. 11, 2010
Emmerling et al.
Scheme 1. Structures of 4-(Chlorodimethylsilyl)-2-methylbutan-2-yl
-bromo-2-methylpropanoate (1) and 3-(Chlorodimethylsilyl)propyl
contains a bromoester functionality but is derived from a tertiary
alcohol. In this initiator the ester bond was much more labile and
easier to cleave, allowing us to suppress brush growth completely.
Two aspects of the saponification reaction of the SI-ATRP
initiator are discussed in this study. First, we investigated the
usefulness of hydrochloric acid or sulfuric acid in saponification
of the initiator monolayer and its concentration dependence. To
obtain results independent of initial SAM preparation, we used
the inkjet printer to screen these parameters on one substrate.
Second, we investigated the time dependence of the saponifica-
tion reaction by controlling the contact time between acid and
initiator. In this way we achieved a gradual variation of grafting
2
a
2
-bromo-2-methylpropanoate (2)
a
SI-ATRP initiators deployed for “grafting-from” of MMA on
silicon wafers in this study. The ester bond (bold) in both initiators
can be cleaved under acidic conditions.
4
8
density from the brush regime to the mushroom regime over the
area of a silicon piece the size of 2.3 ꢀ 3 cm using a dip-coater.
with NaN . After performing three repeat cycles and one back-
3
filling step, they obtained laterally distinctive multicomponent
polymer brushes. Inherent difficulties of this approach include
the preparation and handling of the gold films, a potential edge
Experimental Section
Materials. Allymagnesium bromide (1.0 M in diethyl ether,
Aldrich), 2-bromoisobutyryl bromide (97%, Alfa), pyridine (99%,
Aldrich), chlorodimethylsilane (95%, Fluka), dihydrogen hexa-
chloroplatinate(IV) hexahydrate (99.9% metals basis, Alfa), ethyl
R-bromoisobutyrate (97%, Fluka), toluene (anhydrous 99,8%,
Aldrich), sulfuric acid (purum p.a. 95-97%, Fluka), and hydro-
chloric acid (37%, Aldrich) were used as received.
Copper bromide (98%, Aldrich) was purified by boiling it for
a short time in a mixture of 50:50 v/v Millipore water/glacial
acetic acid and thereafter filtered off. This was followed by
rinsing the precipitate with water, ethanol, and diethyl ether,
35
roughness of the written patterns, and delamination. Working
with oxidized silicon layers can circumvent some of these pro-
blems. First of all, the Si-O bond dissociation energy of 96-133
kcal/mol is higher than the Au-S bond energy at 30-40 kcal/mol.
Hence, by using silane/silica chemistry, the layers stay anchored
36
even above 70 °C where Au-S bonds become unstable. How-
ever, few publications on the subject of μCP and SIP on SiO can
2
37-39
be found.
We attribute this to difficulties in the reproducible
formation of organosilane monolayer patterns by μCP for SIP.
Indeed, all cited procedures require several steps to create patterns
of “initiator” silanes instead of direct deposition of “initiator”
silanes.
4
9
and finally it was dried under reduced pressure for 24 h.
Methyl methacrylate (MMA) (99%, Acros) was purified by
passing it through an alumina column, distilled under reduced
pressure from CaH , and stored under argon at -20 °C.
2
This publication describes a route to circumvent the above-
stated problems with μCP and SIP on silicon surfaces by a direct
and easy one-step approach. We do not create patterns by μCP,
but by the use of a computer-assisted drop-on-demand inkjet
system and amplify these by ATRP of MMA. The major
advantage offered by inkjet printing is the ability to write patterns
directly onto substrates, eliminating the use of masks or masters,
0
0
00
N,N,N ,N ,N -Pentamethyldiethylenetriamine (PMDETA)
99%, Aldrich) was purified by distillation under reduced
(
pressure and triethylamine by distillation from CaH in an
argon atmosphere.
2
Instrumentation. A Nano-Plotter NP 2.0 (GeSIM GmbH,
Germany) was used to dispense droplets of acid onto the silicon
surface covered with a monolayer of ATRP initiator. The
plotter consists of a mobile pipet dispenser and a work plate
that can be cooled. During time-dependent saponification ex-
periments the temperature was set to 10 °C. All other printing
experiments were conducted at room temperature (20 °C). The
dispenser is equipped with a piezo-driven microdosage head that
works in a noncontact regime. Its positioning can be fully
controlled along the plane parallel to the substrate with a speed
of 50 cm/s.
40-45
combined with a resolution of hundreds of nanometers.
Furthermore, the computer control of inkjet printers enables
fully automated and high throughput manufacturing of pat-
4
6
terned materials. This prompted Sankhe et al. to combine inkjet
printing with ATRP to produce patterned polymer surfaces.
Their work, however, is limited to gold/thiol chemistry, as they
print a bromine-containing thiol for ATRP and backfill with a
non-ATRP active thiol (or vice versa). Controlling the reaction
parameters of the backfilling step proved to be critical, since the
printed thiol monolayer was otherwise replaced by the backfilling
thiol.
An imaging ellipsometer (EP3 Imaging Spectroscopic Ellipso-
meter, Nanofilm, Goettingen, Germany) was used to determine
the thickness of the polymer brush layer. It was operated at
In this article we report on a one-step approach to create
patterns with a focus on silicon/silane chemistry, since silane
layers on silicon are thermally more stable compared to thiol
layers on gold. Nevertheless, our approach is also applicable to
gold/thiol chemistry. Contrary to Sankhe et al., we covered the
entire surface of the wafer with a silane monolayer of an ATRP
initiator and used the inkjet printer to chemically alter this
monolayer. The ATRP initiator contains a hydrolytically labile
bromoester functionality (see Scheme 1, bold bond). We printed
acids to cleave the molecule at this point and pattern the initiator
SAM in this way. After cleavage, the surface contains fewer
bromo functionalities in the contact areas that participate in
ATRP. The thickness of PMMA brushes that grew in the contact
areas was thus reduced. The observed changes in film thickness in
the contact areas are related to changes in grafting density as an
effect of the cleavage of a fraction of ATRP initiators. The results
a wavelength λ = 532 nm, an angle of incidence R = 60°, and
i
a field of view of 0.39 ꢀ 0.3 mm (horizontal ꢀ vertical).
The calculation of PMMA thickness was conducted using a
multilayer model (Air, PMMA, initiator monolayer, SiO ,
2
5
0
amorphous Si). The SiO
Initiator = 1.4603) were treated as one layer due to the small
difference in refractive indices and set to n = 1.46. Furthermore,
for the refractive index of the dry PMMA brush film nPMMA
.49 was used. With the above stated values, the error for
reported layer thickness was calculated as (1 nm.
2
(nSiO2 = 1.46) and initiator layer
(
n
=
1
Analysis of the bulk PMMA was carried out at 20 °C with a
GPC setup consisting of the following components: Waters 515
HPLC pump, Waters WISP717 autosampler, three Polymers
Standard Service columns (SDV) in series (dimension: 300 ꢀ 8
5
6
˚
mm, porosity: 500, 10 , and 10 A, respectively), filled with poly-
(
styrene/divinylbenzene), and an ERC refractive index detector
along with a UV-Soma detector. Tetrahydrofuran, at a flow
4
7
-1
obtained with a common ATRP initiator (depicted in Scheme 1,
structure 2) led us to the synthesis of a more labile ATRP initiator
rate of 1 mL min , was used as an eluent. Calibration was
done applying low-polydispersity poly(methyl methacrylate)
standards.
(
depicted in Scheme 1, structure 1). The novel initiator also

Article
Scheme 2. Synthesis of the Tertiary SI-ATRP Initiator
Macromolecules, Vol. 43, No. 11, 2010 5035
pressure-equalizing dropping funnel, a condenser, and a mag-
netic stir bar. While cooling the flask with an ice bath, 2-bro-
moisobutyryl bromide (16 mL, 0.126 mol) was added slowly
(
25 min), forming a yellow precipitate. After heating in a water
bath for 15 min, agitation was continued overnight under
ambient conditions. The suspension was poured into ice water,
and the pH was lowered from 6 to 2 by slow addition of
concentrated HCl. The precipitate was dissolved, leaving a clear
yellow solution. For further processing, the solution was trans-
ferred to a separatory funnel, and the crude product extracted by
washing three times with dichloromethane. This slightly acidic
organic phase was neutralized by washing with an aqueous
solution of hydrogen carbonate (2ꢀ). Residual salt was removed
by washing with water, and then the solution was dried over
4
MgSO . The solvent was removed under reduced pressure, and
subsequently vacuum distillation (0.1 mbar) of the yellow oil
yielded the ester (23 g, 73.3%) with a boiling point of 80 °C as a
1
colorless liquid. H NMR δ (ppm): 5.85-5.71 (m, 1H, =CH-);
5
6
.1-5.03 (m, 2H, =CH
2
); 2.54-2.51 (d, 2H, -CH -); 1.85 (s,
2
H, -C(CH ) -Br); 1.45 (s, 6H, -C(CH ) -O-).
3
2
3 2
Third Step: Synthesis of 4-(Chlorodimethylsilyl)-2-methylbu-
tan-2-yl 2-bromo-2-methylpropanoate by Hydrosilylation Using
5
3
Hydrogen-Hexachloroplatinate(IV) Hexahydrate . In this step
moisture-free conditions are of uttermost importance, since the
chlorosilane group in the final product is easily hydrolyzed. To
ensure this, the experiment was conducted in a dry argon
atmosphere. The reaction was carried out in a Schlenk tube
with a condenser and a magnetic stir bar. First a catalytic
amount of hydrogen-hexachloroplatinate(IV) hexahydrate
was placed in the tube and melted with the use of a blow-drier.
The released water was purged by dry argon. Subsequently,
2-methylpent-4-en-2-yl 2-bromo-2-methylpropanoate (8 g, 0.03
mol) as well as chlorodimethylsilane (30 mL, 0.27 mol) was
added, forming a white residue. After refluxing overnight, a
clear red liquid had formed. By distillation under normal
pressure the excess chlorosilane was removed. Finally, vacuum
distillation (0.1 mbar) allowed us to obtain the clean product
(9.54 g, 96.4%) as a colorless oil (bp 110 °C). Degradation by
moisture during storage was prevented by placing the initiator
1
All H NMR spectra were recorded with a Bruker DPX250,
operating at 250 MHz in deuterated chloroform, and the spectra
were calibrated on this solvent.
Contact angle measurements were conducted using a Kruess
DSA10-MK2 tensiometer. Static contact angles were obtained
in air at room temperature with Millipore water as solution.
AFM under ambient conditions was carried out with a
Dimension 3100 setup connected to a NanoScopeV controller
(
Dimension 3100, Veeco). Silicon cantilevers (OMCL-AC 160
1
TS-W2, Olympus, Japan) with a tip radius of ∼10 nm were used
in the tapping mode with a resonance-frequency near 300 kHz.
Images were recorded in retrace line direction.
Synthesis of SI-ATRP Initiator Derived from the Tertiary
Alcohol 2-Methyl-4-penten-2-ol (Depicted in Scheme 2). First
Step: Synthesis of 2-Methyl-4-penten-2-ol Using the Barbier-
Grignard Reaction . The reaction was carried out in a two-neck
round-bottom flask equipped with a pressure-equalizing drop-
ping funnel, a condenser, and a magnetic stir bar. Allylmagne-
under an argon atmosphere over silica gel in a desiccator. H
NMR δ (ppm): 1.87 (s, 6H, -C(CH ) -Br); 1.53-1.41 (m, 10H,
3
2
-O-C(CH ) -CH -CH -); 0.85-0.78 (m, 2H, -Si-CH -);
3
2
2
2
2
0.4 (s, 6H, -Si(CH ) -Cl).
3
2
Synthesis of SI-ATRP Initiator Derived from the Primary
Alcohol Prop-2-en-1-ol (Structure 2 in Scheme 1). The initiator
3-(chlorodimethylsilyl)propyl 2-bromo-2-methylpropanoate
51
5
4
was synthesized according to Ramakrishnan et al. However,
we opted to do the hydrosilylation step following the above
described approach for the tertiary initiator with hydrogen-
hexachloroplatinate(IV) hexahydrate. Distillation under reduced
pressure was carried out to obtain the pure product, which was also
stored in an argon atmosphere over silica gel in a desiccator.
Cleaning of Silicon Wafers. Dust was blown off the 2.3 ꢀ 3 cm
big wafer in a nitrogen stream. Organic impurities were removed
by immersing the wafer in dichloromethane and placing it in an
ultrasound bath for 10 min followed by base cleaning. The wafer
was placed in a mixture of NH (4 mL, 32%), H O (4 mL,
sium bromide (in Et O; 200 mL, 0.2 mol) was placed in the flask
2
in a protective argon atmosphere and cooled using an ice bath.
Under stirring, acetone (14.7 mL, 0.2 mol) was added dropwise
(
30 min) to the bromide using a dropping funnel while main-
taining the temperature at 0 °C. After the addition was com-
pleted, the solution was refluxed for 1 h. While cooling, 70 mL of
Milli-Q water was added dropwise (20 min), generating a white
solid residue. The temperature was kept around 0 °C, and 60 mL
of half-concentrated HCl was added dropwise (5 min). The
white residue was dissolved and the pH set to 2. For further
work-up the solution was transferred to a separatory funnel.
The aqueous phase was separated and washed twice with diethyl
ether. These ether extracts were combined with the organic
3
2
2
35%), and Milli-Q water (50 mL) at 80-85 °C for 20 min.
Finally, the wafer was washed twice with Milli-Q water, desic-
cated in nitrogen current, and used immediately.
Immobilization of ATRP Initiators. Moisture-free conditions
and dry chemicals are of importance in this step since both
initiators are easily hydrolyzed. The wafer was mounted in a
holder; toluene (25 mL), triethylamine (0.2 mL, 1.435 mmol),
and one of the initiators (0.2 mL) were added in a dry argon
counter flow. After stirring for 20 h, a suspension formed. The
4
phase, washed with little water, dried (MgSO ), and finally
evaporated under reduced pressure to yield the crude alcohol.
The final product (12.58 g, 62.8%) was attained by fractional
destillation and had a boiling point of 117 °C.
Second Step: Synthesis of 2-Methylpent-4-en-2-yl 2-bromo-2-
5
methylpropanoate Using the Pyridine Method of Einhorn .
2
formation of the insoluble salt (TEA HCl) indicated the
3
Pyridine (102 mL, 1.26 mol) and 2-methyl-4-penten-2-ol
12.58 g, 0.126 mol) were placed in a protective argon atmo-
sphere in a two-neck round-bottom flask equipped with a
successful functionalization of the wafer surface. The result-
ing silanated wafer was washed with Milli-Q water (2ꢀ) and
then placed in a Soxhlet apparatus with dichloromethane
(

5
036 Macromolecules, Vol. 43, No. 11, 2010
Emmerling et al.
Scheme 3. Sample Preparation by (a) Printing the Acid with a Commercially Available Inkjet Printer, (b) Washing, (c) Drying the Sample in a Stream
of Nitrogen, (d) ATRP of MMA, and (e) Top View of Sample Prepared by Printing Various Concentrations of H SO
2
4
wafer was taken out of the printer and washed with plenty of
Milli-Q water (Scheme 3b). This was done to remove the applied
acid and the generated bromoisobutyric acid. Subsequently, the
wafer was blown dry in a stream of nitrogen (Scheme 3c). In the
final step, the PMMA brushes were grown on the surface of the
wafer. Reaction details are given in the next paragraph and in
Scheme 4. In the areas, where the acid previously had been in
contact with the surface, grafting density was lowered. For some
concentrations of acid brush growth was completely suppressed.
The red areas in Scheme 3d represent these contact areas, and
the green dots represent the monomer. Scheme 3e illustrates the
top view of the samples prepared for an in-depth study of
different concentrations of H SO . For this purpose two lines
2
4
of 10, 4, 3, and 1 mol/L H
, 6, and 2 mol/L. Data of samples prepared in this fashion are
displayed in Figure 4.
Not shown in Scheme 3 are the samples that we prepared in
order to compare the effect of concentrated HCl and different
2 4
SO were printed, along with a line of
Figure 1. Typical image of data acquisition with the imaging ellips-
ometer on a sample of patterned PMMA brushes. The white rectangles
mark the regions of interest, where the film thickness was acquired. The
spherical drop is imaged as elliptical in shape because the image was
taken under a viewing angle of 60°. The white arrow indicates viewing
direction of the optical system.
9
2 4
concentrations of H SO on the ester bond in the initiator
molecules. Overall, droplets of concentrated HCl were arranged
overnight to remove any traces of salt and physisorbed
initiator molecules.
in three lines. In addition, two lines of 10 and 1 mol/L H
each were printed, along with one line of 6 and 3 mol/L H
2
SO
SO
4
4
2
Application of Acids with the Inkjet Printer. In a typical
dispense experiment, a silicon wafer covered with a monolayer
of ATRP initiator was mounted onto the working plate of the
inkjet printer. The printing nozzle was loaded with either
concentrated HCl or one of the different concentrations of
and a line of Milli-Q water. The data obtained from these
samples are represented in Figure 3.
Application of Acids for Investigation of Time Dependence of
Saponification Reaction. Two different approaches were taken
2 4
to investigate the effect of contact time between H SO and both
H
2
SO . As depicted in Scheme 3a, single droplets (volume
4
of the initiator SAMs. First of all, we created samples with the
inkjet printer that allowed us to compare the effect of the contact
time between three different concentrations of H SO (1, 0.5,
2 4
and 0.1 mol/L) on the initiator molecules. To slow the saponi-
fication reaction, we cooled the samples to 10 °C. For each
concentration we printed a line of droplets every 4 min. 32 min
after deposition of the first line, the last line was printed, and
0
.35-0.4 nL) were deposited and arranged in lines. This was
completed within seconds. After spreading, the typical droplet
size was below 100 μm (see Figure 1). Individual droplets
were placed around 300 μm apart from each other. The distance
between lines was about 1 mm. The average contact time
between the surface and the acid was 20 min, after which the

Article
Macromolecules, Vol. 43, No. 11, 2010 5037
Figure 2. Saponification of surface-immobilized tertiary ATRP initiator with catalytic amount of acid. The error of the contact angles is (10°.
a
Scheme 4. Reaction Scheme of SI-ATRP of MMA with Tertiary Initiator Covalently Anchored on a Silicon Surface
a
The sacrificial initiator 2-EiBBr generated free polymer.
2 4
Figure 4. Effect of various concentrations of H SO on the film thick-
ness of PMMA grown through SI-ATRP.
Figure 3. Effect of concentrated HCl and various concentrations of
SO on the film thickness of PMMA brushes.
H
2
4
over the area of a whole wafer. Additionally, evaporation plays
no role in dip-coating, whereas the small droplets in inkjet
printing evaporate rapidly. Two wafers, one coated with the
tertiary initiator and one with the primary, were fixed on the dip-
coater. These wafers were totally immersed in a solution of
after an additional 4 min the wafer was washed and the
polymerization started.
In the second approach we utilized a dip-coater. The use of a
dip-coater allowed us to create a gradient in grafting density

5
038 Macromolecules, Vol. 43, No. 11, 2010
mol/L H SO , except for an area 3 mm in width. Starting after
min, the wafers were pulled out of the solution with a speed of
.5 mm/min.
SI-ATRP of MMA. Surface-initiated ATRP was carried out
Emmerling et al.
1
2
1
2
4
in a modified Schlenk tube containing the freshly functionalized
wafers and a magnetic stir bar. Oxygen-free conditions were
ensured by working in an argon atmosphere and by flushing
anisole (10 mL) and MMA (10 mL, 0.09 mol) with argon prior to
use. Under argon counter flow, CuBr (13.4 mg, 0.09 mmol),
anisole, MMA, and PMDETA (19.6 μL, 0.09 mmol) were added
to the tube. The active copper complex formed during stirring
for 10 min. The slightly green tint of the solution is caused by
traces of Cu(II). Subsequent to the addition of the free initiator
Figure 5. The logo of our institute is shown on the left. On the right side
is the pattern after polymerization with PMMA shown. The light
regions correspond to no polymer growth, while the regions where
polymer was grown appear darker. The image was taken with an optical
microscope.
2
-EiBBr (13.8 μL, 0.09 mmol), the holder was capped with a
stopper, and three freeze-pump-thaw cycles were executed.
During the course of heating overnight at 30 °C the polymeri-
zation proceeded and finally was stopped by exposing the
solution to air. The wafers were taken out, washed with dichlor-
omethane, and then extracted with dichloromethane overnight
in a Soxhlet apparatus to remove the free polymer. The free
polymer generated by the sacrificial initiator was recovered by
precipitation into methanol, drying, and subjected to GPC
measurement. Molecular weights, grafting densities, and a plot
showing the linear dependency of film thickness on molecular
weight of the free polymer can be found in the Supporting
Information. From the linear dependency of film thickness on
molecular weight we concluded that the polymerization pro-
ceeds in a controlled way.
as for the comparison between the different concentrations of
H
Similarly, mean values for the contact area and nearby are
SO (Figure 4).
2 4
plotted in Figures 3 and 4. On average, four single drops were
measured and the mean value was calculated. Once again the
margin of error from the modeling of the ellipsometric angles
was bigger than the standard deviation from the mean value, so
the error of the modeling ((1 nm) was chosen as error bars.
In addition to plotting the data against absolute film thick-
ness, we also chose to plot the data as relative thickness with
respect to the reference thickness. Trends in the data are at times
more articulated using the relative thickness, since the absolute
thickness varies from sample to sample. Several factors, among
them the polymerization time and the conversion, are the reason
for this. This motivated us to plot the data in Figures 6 and 7
exclusively as relative thickness.
If the data of wafers covered with tertiary initiator are
compared to the data of wafers covered with primary initiator,
the PMMA brushes on both wafers were typically grown in the
same solution, ensuring the same molecular weight of polymer
molecules.
Data Acquisition and Treatment. PMMA film thicknesses
given throughout this publication were acquired with an ima-
ging ellipsometer. This technique allowed us to select certain
regions of interest, where we wanted to measure the film
thickness. This is sketched in Figure 1. All objects were imaged
distorted along the incoming beam (indicated by the white
arrow) due to a viewing angle of 60°. A spherical droplet is thus
imaged as an ellipse. For information on the calculation of the
feature size on the sample surface in vertical direction, see the
Supporting Information.
The dark 60 μm wide (horizontal direction) feature was
generated by spotting a droplet of acid on the initiator layer
prior to polymerization. Little or no polymer grew in this
contact area. Therefore, the change in polarization of these
areas with thinner films is clearly distinguishable from the
polarization of the light induced by the surrounding thicker
films.
Results and Discussion
Contact Angles: Determination of Surface Properties after
Contact with Acid. The basic principle underlying this re-
search is the saponification reaction of surface-immobilized
ATRP initiator molecules. Ordinarily, saponification is done
on mobile molecules in solution and not on molecules
immobilized on a surface. Our first step was therefore to
prove that saponification of surface-immobilized molecules
is feasible. We performed static contact angle measurements
to examine the change in surface energy after the saponifica-
tion. As depicted in the upper part of Figure 2, quantitative
saponification of the tertiary initiator leaves a tertiary alco-
hol on the surface. The free bromoisobutyric acid can be
washed away in the cleaning step. Two wafers were covered
with the initiator molecules and contact angles of 70 ( 10°
for the primary initiator (compared to 77° from an earlier
By contact area (34 ꢀ 27 μm/horizontal ꢀ vertical size) we
refer to the region of interest that was located within one of the
patterns. Within the distance of 50-100 μm of the pattern we
placed a region of interest labeled nearby (292 ꢀ 274 μm/
horizontal ꢀ vertical size).
5
6
publication of our group ) and 65 ( 10° for the tertiary
initiator on the freshly prepared samples measured. The
samples were processed further by immersing them halfway
into 1 mol/L H SO for 40 min to allow the saponification to
2
4
A representative value taken as a reference thickness (390 ꢀ
proceed. A cleaning step;washing with Milli-Q water and
extraction with DCM overnight;was performed to remove
any residual free bromoisobutyric acid from the surface. No
changes in contact angles were observed on the parts of the
samples that had not been in contact with the acid. In the
contact area we observed a decrease to 48 ( 10° in the case of
the primary initiator and 50 ( 10° for the tertiary one.
3
00 μm/horizontal ꢀ vertical size) was obtained from positions
on the wafer that were at least 1 mm away from any dispensed
droplet. In these areas the recorded data were not affected by
acid. Spraying due to the formation of satellite droplets may
occur in inkjet printing as well as spreading of the liquid on the
55
surface. Both these effects could have potentially influenced
the nearby film thickness.
5
7
Miyake et al. reported a contact angle for a hydroxyl-
terminated SAM of 55°, which is within the error of our
observation. This indicates that saponification reactions on
surface-immobilized molecules are possible.
Comparison of Effectiveness of Hydrochloric Acid and
Sulfuric Acid in Saponification. In general, saponification is
an equilibrium reaction. The reaction conditions must be
On average, we measured eight points on the sample away
from patterns to calculate the mean value and plotted it as the
reference thickness. The standard deviation was in all cases
lower than the error that resulted from the modeling of the
ellipsometric angles. Therefore, we chose the error of the
modeling ((1 nm) as error bars. We proceeded like this when
comparing the data for the two kinds of acids (Figure 3) as well

Article
Macromolecules, Vol. 43, No. 11, 2010 5039
Similar observations concerning the effect of concentrated
HCl can be made in the case of the primary initiator
(Figure 3b). Film thickness after treatment with concentratd
HCl is only reduced to 30 nm (75%), whereas on the other
hand 6, 3, and 1 mol/L H SO reduce the film thickness to
2
4
9
nm. This corresponds to 25% of the reference film thick-
ness. 10 mol/L H SO shows a significant discrepancy com-
2
4
pared to the other H SO concentrations. In this case the film
2
4
thickness was only reduced to 45% (16 nm).
Our motivation for the synthesis of the tertiary initiator
was the hydrolytically more labile ester bond in this mole-
cule, which should be easier to cleave. Indeed, for all con-
centrations of H SO and for the HCl, the comparison of
2
4
relative film thickness of the two initiators show a decrease in
relative thickness by 20% or even up to 40% for the 10 mol/L
H SO in the case of the tertiary initiator. This demonstrates
2
4
the increased lability of the ester bond in the tertiary in-
itiator. The tertiary initiator combined with diluted H SO of
2
4
6
, 3, or 1 mol/L concentration suppresses polymer growth
completely within the margin of error. Yet the primary
initiator is interesting to use, when tuning of grafting density
is desired.
An explanation for the difference between concentrated
HCl and all of the chosen H SO is the high volatility of
2
4
58
concentrated HCl. HCl has a boiling point of 45 °C (38%),
unlike concentrated sulfuric acid, which is nonvolatile with a
5
9
boiling point of 300-320 °C (96%). The high surface-to-
volume ratio of inkjet-printed microdroplets leads to extre-
mely rapid evaporation, even for liquids with boiling points
-
1
of 200 °C (vapor pressure around 10 mmHg). Microdroplets
of liquids with high boiling point around 200 °C evaporate
4
3
within ∼1 min. Our more quantitative observations are
consistent with this argument of volatility. Immediately after
deposition the drops of Milli-Q water or concentrated HCl
were evaporated. This behavior of HCl is expected, since it is
a gas dissolved in water. By contrast, the drops of diluted
H SO were still present after 20 min, although they seemed
2
4
to shrink to some extent after deposition due to the loss of
water. Video clips of evaporating droplets can be found in
the Supporting Information.
Figure 6. Time dependence of the saponification reaction from inkjet
printing. For each concentration a line of droplets was deposited every
4
min. Error bars with long end-caps belong to the data of the 0.1 mol/L
Comparison of Effectiveness of Different Concentrations of
Sulfuric Acid in Saponification. Next we studied in more
detail why 10 mol/L H SO was not as effective as lower
sulfuric acid. Error bars with medium end-caps belong to 0.5 mol/L.
Error bars with no end-caps indicate data related to 1 mol/L.
2
4
concentrations, at least in the case of the primary initiator.
Our hypothesis is that the water content is solely responsible
for the position of the equilibrium of the saponification
reaction.
adjusted in such a way that the equilibrium is located on the
side of the product. In our case the product is the alcohol on
the surface and the free bromoisobutyric acid (see Figure 2).
The parameters determining the position of the equilibrium
are the type of acid used and the amount of water available.
With the use of the inkjet printer we were able to screen these
parameters simultaneously. In a first attempt we compared
two different types of acid, concentrated HCl, and four
concentrations of H SO (Figure 3). Direct techniques to
measure the density of active initiator molecules in an area as
small as 60 μm were not available. We therefore measured
the film thickness and related this data to changes in grafting
density. The more cleaved ester bonds were generated with
the acid, the less molecules remained on the surface that still
carried bromo functionalities. In turn, this means a reduc-
tion in grafting density, resulting in a lower film thickness for
a given molecular weight.
Regardless of the concentration of H SO plotted, on the
4
2
tertiary initiator the remaining film thickness is below 5% of
the original thickness (Figure 4a). With all concentrations of
10 to 1 mol/L of H SO and a contact time of 20 min, brush
2
4
growth from the tertiary initiator can be suppressed within
the margin of error. For the primary initiator (Figure 4b) the
trend to lower thickness with lower concentration of de-
ployed H SO is clear-cut. In none of the used concentrations
2
4
2
4
does the thickness go below 20%. So the comparison of the
relative film thickness in the contact area between the two
initiators for all the different concentrations of H SO sup-
2
4
ports the conclusion that we have drawn from Figure 3. The
ester bond is more labile in the tertiary initiator. This effect is
independently observed from the value of absolute film
thickness, as a comparison of Figures 3 and 4 shows.
The data for both initiators is congruent regarding the
effectiveness in saponification of 10 mol/L H SO . It is lower
We will begin with a discussion of the effect of the various
acids on the tertiary initiator (Figure 3a). After treatment with
any of the four concentrations of H SO , no polymer is found
2
4
2
4
within the margin of error (> 1.2 ( 1 nm in every case). In
contrast, the film thickness after treatment with concentrated
HCl is still 13 nm, which is ∼55% of the reference thickness.
than all of the more dilute concentrations. This effect is more
pronounced for the primary initiator but is in principle also
observed for the tertiary initiator. The equilibrium in the

5
040 Macromolecules, Vol. 43, No. 11, 2010
Emmerling et al.
saponification reaction (Figure 2) can be shifted solely by
supplying water; protons are just needed as a catalyst. With
higher water content, the equilibrium is shifted more and
more to the desired side of the alcohol on the surface and the
free bromoisobutyric acid. For the tertiary initiator the water
content does not seem to be as crucial. This is due to the fact
that the ester bond is more labile and breaks easier from the
start.
The compliance between the reference thickness (gray)
and the nearby (red) again indicates that the droplets did not
splash upon contact with the surface (Figure 4).
Polymerized Patterned Surface. After finding the optimum
conditions for suppressing polymer growth completely, we
wanted to test the potential of our method for printing more
complex designs. Since imaging a pattern of several hundred
micrometers in size was not possible with the use of the
imaging ellipsometer, we printed a pattern with 1 mm in size.
With the use of an optical microscope we were able to image
this pattern. In Figure 5 the pattern is shown on the right
side. It resembles the logo of our institute, which is shown on
the left. The pattern was formed by printing 1 mol/L sulfuric
acid on the tertiary starter. The film thickness is 8 ( 1 nm as
measured by ellipsometry. Polymerization did not occur in
the regions where sulfuric acid had been printed. In these
regions the film thickness is >1.2 ( 1 nm. The former
contact areas appear lighter in the optical microscope image
because we measured under moist conditions. The moisture
generated a contact angle contrast between the more hydro-
phobic regions where polymer was grown and the more
hydrophilic regions where the initiator was destroyed.
Time Dependence of Saponification Reaction from Inkjet
Printing. It would be much more convenient to tune the
grafting density by controlling the contact time between the
acid and ATRP initiator. Hence, we investigated the time
dependence of the saponification reaction. 1 mol/L H SO in
Figure 7. Time dependence of the saponification reaction from dip-
coating. The wafers were pulled out of a solution of 1 mol/L H SO with
2 4
a speed of 1.5 mm/min using a dip-coater. Solid lines are guides to the
eye. Error bars of the primary starter have no end-caps.
saponification reaction to completion. Because of the cool-
ing of the substrate, we changed the evaporation of the water
from the microdroplet. As a result, a lower fraction of ester
molecules was cleaved on the time scale under investigation.
Time Dependence from Dip-Coating/Creation of a Gradi-
ent in Grafting Density over the Area of a Whole Wafer. To
study the efficiency of the saponification reaction indepen-
dent of droplet evaporation, we used a dip-coater and gra-
dually pulled out samples from a solution of 1 mol/L H SO .
2 4
This approach allowed us to tune the grafting density from
brush regime to mushroom regime over larger areas, which is
desirable when studying the dependence of brush height on
2
4
contact with the surface leads to almost complete defunctio-
nalization of the tertiary initiator. To limit water evapora-
tion, we chose to slow the reaction by cooling the samples
to 10 °C and to include lower acid concentrations of 0.5 and
6
0
grafting density.
The data in Figure 7 are plotted as relative thickness
against contact time. As the point of reference we elected
for each initiator (100% relative thickness) the 3 mm wide
area that had not been immersed in the H SO . Measure-
0
.1 mol/L H SO .
2 4
The data in Figure 6 are plotted as relative thickness
2
4
against contact time. We measured the film thickness in four
contact areas, plotted the mean value of those, and chose the
error of the modeling as error bars ((1 nm).
ments of the film thickness were taken at four different areas;
we plotted the mean value of this and chose the error of the
modeling as error bars ((1 nm).
In Figure 6a there is a gradual decrease in relative brush
thickness with contact time for the tertiary initiator. The
same holds true in the case of the primary initiator in
Figure 6b, but the curve levels off after 20 min within the
margin of error. In addition, the data for both initiators show
no difference regarding the three concentrations. Given that
the water content matters, and not the amount of protons,
any concentration below 10 mol/L H SO has enough water
With longer contact time the grafting density was reduced
for both initiators. In analogy to the data in Figure 6 the film
thickness in the case of the primary initiator reaches a steady
state after 20 min. No steady state is reached for the tertiary
initiator, at least not within the contact times under investi-
gation. As is noticeable, the film thickness can be tuned all
the way down to 55% for the primary initiator and 30% for
the tertiary initiator. Compared to the data from inkjet
printing of 1 mol/L H SO where we reached levels of 25%
2
4
to shift the equilibrium of the saponification to the product
side before the water is evaporated.
When comparing the final thickness for 1 mol/L H SO
2
4
for the primary initiator and 5% for the tertiary initiator, the
values here are much higher. When using the dip-coater
and 20 mL of acid, evaporation plays no role. Thus, the
reaction proceeds in a different way, and therefore the
saponification reaction did not reach the same conversion
as in printing.
AFM Images To Investigate the Effect of the Acid on Areas
Surrounding the Droplets. After testing the effect of acid on
the SI-ATRP initiators our last step was to study in detail the
“ink behavior” on the substrate. A crucial step in ink-
printing is the moment a droplet hits the solid substrate
and comes to rest. Thus, we turned to AFM to check the
features that we created for uniformity and roughness. The
AFM images in Figure 8 were taken on the rim of the former
2
4
from Figure 4 to Figure 6 for both initiators it is clear that
they do not reach the same level, even though the contact
time was twice as long. We have altered the sample prepara-
tion in the case of the time dependency by cooling the
samples to 10 °C. Cooling of the samples slows down the
evaporation of water out of the microdroplets, while heat-
4
1
ing enhances evaporation. This supports our hypothesis
that the initial water content of the microdroplet matters.
That was the reason why the film thickness decreased with
decreasing concentration of H SO . Then the evaporation
2
4
of water from the microdroplet set in, and the concentra-
tion of H SO started to increase. This seems to drive the
2
4

Article
Macromolecules, Vol. 43, No. 11, 2010 5041
5
.4 nm for the primary one. Ramakrishnan et al. published
4
0
a roughness of 0.5 nm for the PMMA layer from X-ray
reflectivity measurements in the case of the primary initiator.
The agreement between the roughness of the PMMA brushes
around the pattern between the two tested ATRP initiators
and the roughness from the literature demonstrates that the
acid did not have any effect on the surrounding areas and in
turn catalyzed the saponification reaction only in the printed
areas.
Conclusions and Outlook
We have demonstrated that chemistry with an inkjet printer is
feasible. The industrial relevance of inkjet printing is related to
the potential for automation and high throughput. In our case the
inkjet printer was used to deposit microdroplets of acid. Within
these microdroplets we carried out the saponification of surface-
immobilized ATRP initiators containing an ester bond. The
changes in the monolayer of ester molecules due to saponification
were amplified by SI-ATRP. The polymer brush thickness was
then related to effectiveness of saponification. The use of an inkjet
printer allowed us to screen parameters such as type of acid,
concentration of acid, and contact time between acid and surface
simultaneously. Additionally, we utilized a dip-coater to test the
saponification independent of droplet evaporation.
The advantage of the developed process is that it is applicable
to all surface-immobilized initiators containing ester bonds.
In future research we plan to use our process to defunctionalize
the initiator molecules covering a microcantilever. If we do this
selectively on one side of a cantilever, we can generate an asym-
metric coating of the cantilever with polymer brushes. An asym-
metric coating allows the use of a microcantilever for sensing
4
9
applications.
Figure 8. AFM height images taken on the rim of the contact area after
the polymerization. Cross sections indicate the structure and the height
of the rim.
Acknowledgment. S. G. J. Emmerling (GU 771/3) and
S. A. Pihan (BE 3286/1) gratefully acknowledge financial support
of the Deutsche Forschungsgemeinschaft (DFG) under the
SPP1369 Priority Program (Polymer-Solid Contacts: Interfaces
and Interphases). P. Lellig (GU 771/2) gratefully acknowledges
the DFG-SPP1181 for financial support. Special thanks are given
to Gunnar Kircher, Sebastian Nett, and E. Bryan Coughlin for
helpful discussions.
contact area between 1 mol/L H SO and (a) the tertiary
2
4
initiator and (b) the primary initiator. In both images a sharp
step is clearly distinguishable. The step separates the former
contact between initiator molecules and the acid from the
surrounding PMMA brushes. The cross sections in (c) sub-
stantiate this. They clearly show a difference in slope between
the two initiators. Within 0.6 μm the transition from the
former contact area to unaffected areas is complete in the
case of the tertiary initiator. The transition for the primary
initiator is completed within 0.4 μm. In the transition areas
the change in film thickness is a result of the change in the
grafting density of brushes. The difference in slope that we
have observed has its origin in the combination of droplet
evaporation and ester bond lability. For the tertiary starter
the evaporation sets in while the saponification proceeds and
the droplet shrinks in size. In the case of the primary starter
saponification is slower, and therefore the greater part of the
saponification takes place after the shrinking of the droplet
has proceeded.
Supporting Information Available: Correction of optical
distortion along the beam explaining the calculation of feature
size on the sample, correlation of film thickness with M of
n
polymer grown in solution showing the control of the polymer-
ization process, tables with molecular weight and polydispersity
of polymer grown in solution, tables with grafting density of
polymer brushes, video clips demonstrating the evaporation of
single droplets of HCl and H SO . This material is available free
2
4
of charge via the Internet at http://pubs.acs.org.
References and Notes
(1) Schmelmer, U.; Jordan, R.; Geyer, W.; Eck, W.; Golzhauser, A.;
Grunze, M.; Ulman, A. Angew. Chem., Int. Ed. 2003, 42 (5), 559.
The heights of the rim indicated in the cross sections from
AFM (Figure 8c) are now compared to the ellipsometry data.
Given that AFM can only measure relative height differ-
ences, we chose the reference thickness from ellipsometry
measurement as the highest point of the cross section.
For the tertiary initiator (filled squares) AFM gave a
height of the rim of 24 nm, which is in good comparison to
(
(
2) Farhan, T.; Huck, W. T. S. Eur. Polym. J. 2004, 40 (8), 1599–1604.
3) Li, H. W.; Muir, B. V. O.; Fichet, G.; Huck, W. T. S. Langmuir
2
003, 19 (6), 1963–1965.
(
4) Advincula, R. C.; Brittain, W. J.; Baster, K. C.; R u€ he, J. Polymer
Brushes: Synthesis, Characterization, Applications; Wiley-VCH:
Weinheim, 2004.
(5) Prucker, O.; R
(
u€ he, J. Macromolecules 1998, 31 (3), 592–601.
6) Zhao, B.; Brittain, W. J. Prog. Polym. Sci. 2000, 25 (5), 677–710.
23 ( 1 nm calculated from ellipsometry data. In the case of
the primary initiator (stars) ellipsometry yielded a height of
(
7) Husseman, M.; Malmstrom, E. E.; McNamara, M.; Mate, M.;
Mecerreyes, D.; Benoit, D. G.; Hedrick, J. L.; Mansky, P.; Huang,
E.; Russell, T. P.; Hawker, C. J. Macromolecules 1999, 32 (5), 1424–
3
1 ( 1 nm compared to 33 nm from AFM.
Concerning the rms roughness of the surrounding PMMA
layers, we obtained 0.5 nm for the tertiary initiator and
1
431.
(8) Baum, M.; Brittain, W. J. Macromolecules 2002, 35, 610–615.

5
042 Macromolecules, Vol. 43, No. 11, 2010
Emmerling et al.
(
9) Barner-Kowollik, C. Handbook of RAFT; Wiley-VCH: Weinheim,
(35) Love, J. C.; Wolfe, D. B.; Chabinyc, M. L.; Paul, K. E.; Whitesides,
G. M. J. Am. Chem. Soc. 2002, 124 (8), 1576–1577.
(36) Dyer, D. J. Adv. Funct. Mater. 2003, 13 (9), 667–670.
(37) Jeon, N. L.; Choi, I. S.; Whitesides, G. M.; Kim, N. Y.; Laibinis,
P. E.; Harada, Y.; Finnie, K. R.; Girolami, G. S.; Nuzzo, R. G.
Appl. Phys. Lett. 1999, 75 (26), 4201–4203.
Germany, 2008.
(
(
10) Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921–2990.
11) Pyun, J.; Kowalewski, T.; Matyjaszewski, K. Macromol. Rapid
Commun. 2003, 24 (18), 1043–1059.
(
(
12) Boyes, S. G.; Granville, A. M.; Baum, M.; Akgun, B.; Mirous,
B. K.; Brittain, W. J. Surf. Sci. 2004, 570 (1-2), 1–12.
(38) Kim, N. Y.; Jeon, N. L.; Choi, I. S.; Takami, S.; Harada, Y.; Finnie,
K. R.; Girolami, G. S.; Nuzzo, R. G.; Whitesides, G. M.; Laibinis,
P. E. Macromolecules 2000, 33 (8), 2793–2795.
(39) Tu, H.; Heitzman, C. E.; Braun, P. V. Langmuir 2004, 20 (19),
8313–8320.
(40) Calvert, P. Chem. Mater. 2001, 13 (10), 3299–3305.
(41) Tekin, E.; Smith, P. J.; Schubert, U. S. Soft Matter 2008, 4 (4), 703–
713.
(42) Derby, B.; Reis, N. MRS Bull. 2003, 28 (11), 815–818.
(43) Shimoda, T.; Morii, K.; Seki, S.; Kiguchi, H. MRS Bull. 2003,
28 (11), 821–827.
(44) Burns, S. E.; Cain, P.; Mills, J.; Wang, J. Z.; Sirringhaus, H. MRS
Bull. 2003, 28 (11), 829–834.
(45) Zaugg, F. G.; Wagner, P. MRS Bull. 2003, 28 (11), 837–842.
(46) Sankhe, A. Y.; Booth, B. D.; Wiker, N. J.; Kilbey, S. M. Langmuir
2005, 21 (12), 5332–5336.
(47) Ramakrishnan, A.; Dhamodharan, R.; R u€ he, J. J. Polym. Sci.,
Part A: Polym. Chem. 2006, 44 (5), 1758–1769.
(48) deGennes, P. G. Macromolecules 1980, 13 (5), 1069–1075.
(49) Bumbu, G. G.; Kircher, G.; Wolkenhauer, M.; Berger, R.;
Gutmann, J. S. Macromol. Chem. Phys. 2004, 205 (13), 1713–
1720.
13) Matyjaszewski, K.; Miller, P. J.; Shukla, N.; Immaraporn, B.;
Gelman, A.; Luokala, B. B.; Siclovan, T. M.; Kickelbick, G.;
Vallant, T.; Hoffmann, H.; Pakula, T. Macromolecules 1999,
3
2 (26), 8716–8724.
14) Kong, X. X.; Kawai, T.; Abe, J.; Iyoda, T. Macromolecules 2001,
4 (6), 1837–1844.
15) von Werne, T.; Patten, T. E. J. Am. Chem. Soc. 1999, 121 (32),
409–7410.
16) von Werne, T.; Patten, T. E. J. Am. Chem. Soc. 2001, 123 (31),
497–7505.
17) Carlmark, A.; Malmstrom, E. J. Am. Chem. Soc. 2002, 124 (6),
00–901.
18) Zhao, H. Y.; Farrell, B. P.; Shipp, D. A. Polymer 2004, 45 (13),
473–4481.
19) Edmondson, S.; Osborne, V. L.; Huck, W. T. S. Chem. Soc. Rev.
004, 33 (1), 14–22.
20) Tovar, G.; Paul, S.; Knoll, W.; Prucker, O.; R u€ he, J. Supramol. Sci.
995, 2 (2), 89–98.
(
(
(
(
(
(
(
(
3
7
7
9
4
2
1
21) R u€ he, J. In Polymers Grafted from Solid Surfaces, 6th Dresden
Polymer Discussion on Surface Modification, Meissen, Germany, Apr
1
4-17; Jacobasch, H. J., Ed.; Huthig & Wepf Verlag: Meissen,
Germany, 1997; pp 215-222.
(50) Azzam, N. M.; Bashara, R. M. A. Ellipsometry and Polarized Light;
North-Holland: Amsterdam, 1977.
(51) Sormunen, G. J.; Lewis, D. E. Synth. Commun. 2004, 34 (19), 3473–
3480.
(52) Einhorn, A.; Hollandt, F. Justus Liebigs Ann. Chem. 1898, 301 (1),
95–115.
(
22) Tsai, I. Y.; Crosby, A. J.; Russell, T. P. Surface patterning. In Cell
Mechanics; Academic Press: San Diego, 2007; Vol. 83, pp 67-87.
23) Husemann, M.; Morrison, M.; Benoit, D.; Frommer, K. J.; Mate,
C. M.; Hinsberg, W. D.; Hedrick, J. L.; Hawker, C. J. J. Am. Chem.
Soc. 2000, 122 (8), 1844–1845.
(
(
(
(
(
(
(
24) Ducker, R.; Garcia, A.; Zhang, J. M.; Chen, T.; Zauscher, S. Soft
Matter 2008, 4 (9), 1774–1786.
25) Kaholek, M.; Lee, W. K.; LaMattina, B.; Caster, K. C.; Zauscher,
S. Nano Lett. 2004, 4 (2), 373–376.
(53) Marciniec, B.; Gulinski, J.; Urbaniak, W.; Kornetka, Z. W.
Comprehensive Handbook on Hydrosilylation; Pergamon Press:
Elmsford, NY, 1993.
(54) Ramakrishnan, A.; Dhamodharan, R.; R u€ he, J. Macromol. Rapid
Commun. 2002, 23, 612–616.
26) Zapotoczny, S.; Benetti, E. M.; Vancso, G. J. J. Mater. Chem. 2007,
1
7 (31), 3293–3296.
27) Kumar, A.; Whitesides, G. M. Appl. Phys. Lett. 1993, 63 (14),
002–2004.
28) Xia, Y. N.; Whitesides, G. M. Angew. Chem., Int. Ed. 1998, 37 (5),
51–575.
(55) Kawase, T.; Shimoda, T.; Newsome, C.; Sirringhaus, H.; Friend,
R. H. In Inkjet Printing of Polymer Thin Film Transistors, 5th Inter-
national Conference on Nano-Molecular Electronics (ICNME2002),
Kobe, Japan, Dec 10-12; Elsevier Science: Kobe, Japan, 2002;
pp 279-287.
2
5
29) Husemann, M.; Mecerreyes, D.; Hawker, C. J.; Hedrick, J. L.;
Shah, R.; Abbott, N. L. Angew. Chem., Int. Ed. 1999, 38 (5), 647–
(56) Nett, S. K.; Kircher, G.; Gutmann, J. S. Macromol. Chem. Phys.
2009, 210 (11), 971–976.
6
49.
(57) Miyake, T.; Tanii, T.; Kato, K.; Hosaka, T.; Kanari, Y.; Sonobe,
H.; Ohdomari, I. Chem. Phys. Lett. 2006, 426 (4-6), 361–364.
(58) Hydrochloric acid - boiling point. http://www.solvaychlorinatedin-
organics.com/docroot/chlo_inorg/static_files/attachments/pch_
1300_0005_w_en_ww.pdf.
(59) Sulfuric acid. http://www.inorganics.basf.com/p02/CAPortal/en_GB/
portal/Anorganische_Basen_/content/Produktgruppen/Anorganische_
Basen_(Laugen)/Produktinformationen/Schwefelsaeure.
(60) Wu, T.; Efimenko, K.; Genzer, J. J. Am. Chem. Soc. 2002, 124 (32),
9394–9395.
(
30) Shah, R. R.; Merreceyes, D.; Husemann, M.; Rees, I.; Abbott,
N. L.; Hawker, C. J.; Hedrick, J. L. Macromolecules 2000, 33 (2),
5
97–605.
(
(
31) Jones, D. M.; Huck, W. T. S. Adv. Mater. 2001, 13 (16), 1256–1259.
32) Jones, D. M.; Smith, J. R.; Huck, W. T. S.; Alexander, C. Adv.
Mater. 2002, 14 (16), 1130–1134.
33) Zhou, F.; Huck, W. T. S. Chem. Commun. 2005, 48, 5999–6001.
34) Zhou, F.; Zheng, Z. J.; Yu, B.; Liu, W. M.; Huck, W. T. S. J. Am.
Chem. Soc. 2006, 128 (50), 16253–16258.
(
(