Molecular-weight determination of polymer brushes generated by SI-ATRP on flat surfaces

Article abstract of DOI:10.1021/ma401951w

A new type of photolabile, surface-initiated, atom-transfer radical polymerization (SI-ATRP) initiator 3-(2-bromo-2-methylpropanamido)-3-(2- nitrophenyl)propanoic acid was synthesized, and immobilized via an aminosilane linker onto a flat silicon surface. Poly(lauryl methacrylate) and polystyrene brushes were grown from the surface via SI-ATRP, and the surface-tethered polymer chains cleaved off under UV irradiation. The kinetics of the cleavage process were investigated, and an apparent effect of osmotic forces within the polymer brush could be observed. The molecular weight of the cleaved polymers could be readily determined by means of size-exclusion chromatography.

Full text of DOI:10.1021/ma401951w

Article
pubs.acs.org/Macromolecules
Molecular-Weight Determination of Polymer Brushes Generated by
SI-ATRP on Flat Surfaces
Chengjun Kang,† Rowena M. Crockett,‡ and Nicholas D. Spencer*,†
†Laboratory for Surface Science and Technology, Department of Materials, ETH Zurich, Wolfgang-Pauli-Strasse 10, CH 8093 Zurich,
Switzerland
‡Swiss Federal Laboratories for Materials Science and Technology, Empa, Ueberlandstrasse 129, CH 8600 Duebendorf, Switzerland
S
* Supporting Information
ABSTRACT: A new type of photolabile, surface-initiated,
atom-transfer radical polymerization (SI-ATRP) initiator 3-(2-
bromo-2-methylpropanamido)-3-(2-nitrophenyl)propanoic
acid was synthesized, and immobilized via an aminosilane
linker onto a flat silicon surface. Poly(lauryl methacrylate) and
polystyrene brushes were grown from the surface via SI-ATRP,
and the surface-tethered polymer chains cleaved off under UV
irradiation. The kinetics of the cleavage process were
investigated, and an apparent effect of osmotic forces within
the polymer brush could be observed. The molecular weight of
the cleaved polymers could be readily determined by means of
size-exclusion chromatography.
determine these parameters is to cleave polymer brushes
from the substrates and measure them directly via size-
exclusion chromatography (SEC). Several publications have
reported, for example, the cleaving of poly(styrene) (PS) from
silica nanoparticles by dissolving the silica substrates in
selectively corrosive media,10−12 such as hydrofluoric acid, or
the growth of polymer brushes from substrates initiated by an
acid-labile linker that could be cleaved off with, for example, p-
toluenesulfonic acid.13 The inherent curvature of nanoparticle
substrates can potentially lead to significant differences in the
polymer-growth kinetics compared to brushes formed on
planar surfaces. Furthermore, these methods cannot be applied
to corrosion-resistant substrates or to HF-sensitive polymers.
Moreover, intricate separation procedures are involved,
introducing the risk of incorporating impurities in the small
amounts of polymers harvested. Since analyzing polymers from
flat substrates is more challenging, because a much smaller
amount of polymer is generated than on nanoparticles, the
determination of molecular weight and PDI of surface-tethered
polymers has often been carried out on polymers grown
simultaneously in the bulk solution while the polymer grows on
the surface.14 This approach relies on the assumption that the
polymers generated in solution and on the surface have the
same molecular weight and PDI. Experimental data suggest that
this is not always the case,5,13 and the assumption has been
convincingly challenged by numerical simulations.15 Compared
to acidic or basic cleavage techniques, the photocleavage
INTRODUCTION
■
The chemical and physical properties of materials have
frequently been tailored by coating with densely grafted
polymer chains, composed of homo- or copolymers with one
end attached to the substrate. These surface-tethered polymer
chains stretch out to form brush-like structures in good
solvents,1 in order to minimize chain−chain interactions.
Polymer brushes can be fabricated by grafting from the
substrates through surface-initiated controlled radical polymer-
ization (SI-CRP) techniques,2 such as surface-initiated atom-
transfer radical polymerization (SI-ATRP),3 surface-initiated
reversible addition−fragmentation chain-transfer polymeriza-
tion (SI-RAFT),4 and surface-initiated nitroxide-mediated
polymerization (SI-NMP).5 Compared to “grafting to”
methods, these techniques allow much more flexibility in
designing the architecture of polymer brushes and can lead to a
much higher chain-grafting density. SI-ATRP, first reported by
Huang and Wirth in 1997,3 is the most frequently used
controlled radical polymerization method because it is
chemically versatile, compatible with a large variety of
monomers, tolerant to a relatively high concentration of
impurities, including oxygen, and the commercial availability of
most of the components necessary for ATRP synthesis.
Substrates modified with polymer brushes have been
extensively used in a variety of applications, such as responsive
surfaces,6 nonfouling surfaces,7 cell-adhesive surfaces,8 and low-
friction surfaces.9 However, the characterization of polymer
brushes tethered to surfaces remains challenging, as crucial
properties including the grafting density, molecular weight, and
polydispersity index (PDI) of surface-tethered polymer brushes
have been difficult to determine. One possible way to
Received: September 20, 2013
Revised: November 25, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ma401951w | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 1. Schematic of the Growth and UV-Induced Cleavage of ATRP-Generated Polymers from an Oxidized Silicon Surface
Chemicals, Germany) were used as received. All solvents were
purchased from Sigma-Aldrich and used as received.
technique is a mild, clean, and easily controlled procedure.
Photolabile linkers have been frequently used in the biomedical
field,16 particularly in drug discovery and development,17 and
controlled cell, peptide, and protein delivery.18−20
We report an efficient photochemical method for cleaving
polymer brushes from planar silicon substrates. We have
utilized this approach to facilitate characterization of the
molecular weight and PDI of the surface-grown polymers by
means of SEC. To this end, an o-nitrobenzyl type21
photocleavable ATRP initiator, 3-(2-bromo-2-methylpropana-
mido)-3-(2-nitrophenyl)propanoic acid (BMNP) (Scheme 1),
has been developed.
Synthesis of 3-(2-Bromo-2-methylpropanamido)-3-(2-
nitrophenyl)propanoic Acid (BMNP). 12.0 mL of triethylamine
(NEt3) was added to a suspension of active ester 2,5-dioxopyrrolidine-
1-yl 2-bromo-2-methylpropanoate (DBMP) (11.0 g, 41.6 mmol), and
3-amino-3-(2-nitrophenyl)propanoic acid (ANPA) (7.3 g, 34.7 mmol)
in 80 mL of DMSO over 15 min with gentle stirring, after which a
clear brown solution was obtained. After the mixture had reacted for
24 h, DMSO was removed under reduced pressure to yield a
semisolid, which was dissolved in a mixture of 250 mL of H2O and
19.2 mL of NEt3 (pH ≈ 9.0). Next, ethyl acetate (EtOAc, 180 mL)
was added under vigorous stirring, and the aqueous layer was
separated, acidified with 12 N HCl to pH ≈ 2.0, and extracted with
EtOAc (3 × 250 mL). The organic layers were combined and washed
with 1 N HCl (1 × 160), H2O (2 × 160), dried over MgSO4, and then
filtered and concentrated under vacuum to give 11.2 g of a light yellow
EXPERIMENTAL SECTION
■
Materials. 3-Amino-3-(2-nitrophenyl)propanoic acid (98%, ABCR-
Chemicals, Germany), lauryl methacrylate (96%, Acros Organics
Belgium), styrene (99%, Sigma-Aldrich, Germany), 1,4,8,11-tetra-
methyl-1,4,8,11-tetraazacyclotetradecane (Me4Cyclam, 97%, Sigma-
Aldrich, Germany), anisole (99%, Sigma-Aldrich, Germany), 4,4′-
dinonyl-2,2′-bipyridine (97%, Sigma-Aldrich, Germany), n-nonane
(99%, ABCR-Chemicals, Germany), N,N-dimethylformamide (DMF,
99.8%, anhydrous, Acros, Germany), N,N′-dicyclohexylcarbodiimide,
(DCC, 99%, Sigma-Aldrich, Germany), 2-bromo-2-methylpropanoyl
bromide (98%, Sigma-Aldrich, Germany), tetrahydrofuran (THF,
99.7%, Alfa-Aesar, Germany), triethylamine (NEt3, 99.5%, Sigma-
Aldrich, Germany), copper(II) bromide (99.9%, Sigma-Aldrich,
Germany), copper(I) bromide (CuBr, 99.9%, Sigma-Aldrich, Ger-
many), (3-aminopropyl)triethoxysilane (APTES, 98%, Sigma-Aldrich,
Germany), and ethyl 2-bromoisobutyrate (EBiB, 98%, ABCR-
1
solid (yield, 89%). H NMR (300 MHz, d-DMSO): δ (ppm) 7.55−
8.03 (m, 4H), 5.28 (m, 1H), 2.62−2.97 (m, 2H), 2.05 (s, 6H). 13C
NMR (300 MHz, d-DMSO): δ (ppm) 171.68, 170.66, 148.59, 137.39,
129.05, 128.89, 124.48, 120.12, 60.29, 46.84, 39.15, 31.28. Elemental
analysis: calcd: C 43.47%, H 4.21%, N 7.80%. Found: C 43.63%, H
4.26%, N 7.76%.
If not specified otherwise, all experiments described in this paper
were carried out under the exclusion of light.
Fabrication of APTES-Modified Silicon Surfaces. Silicon wafers
(P/B(100), Si-Mat Silicon Wafers, Germany) were cut to size,
sonicated three times (10 min each) in 2-isopropanol, dried with
nitrogen gas, and then cleaned in a UV-ozone cleaner (UV/Ozone
ProCleaner and ProCleaner Plus, BioForce, Ames, IA) for 30 min. The
cleaned wafers were immediately placed into an oven at 150 °C for 0.5
B
dx.doi.org/10.1021/ma401951w | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
h and subsequently put into a desiccator with one drop of (3-
aminopropyl)trimethoxysilane (APTES) at the bottom. The desiccator
was evacuated with a vacuum pump for 1 h, after which the pump was
valved off and the desiccator pressure held constant for another 1.5 h
(i.e., at the vapor pressure of the APTES). Finally, APTES-modified
silicon wafers were sonicated in toluene (3 × 10 min) and dried with
nitrogen gas.
Fabrication of BMNP-Modified Silicon Surfaces. In a typical
procedure, BMNP (0.6 g, 1.6 mmol) and DCC (1.7 g, 8 mmol) were
dissolved in 40 mL of DMF over 5 min to yield a clear, light-brown
solution, into which the freshly prepared APTES-modified silicon
wafer (2.5 × 4 cm2) was transferred under a nitrogen atmosphere.
After 48 h, the wafer was removed and rinsed by DMF three times,
followed by sonication in toluene for 5 min to remove physisorbed
BMNP.
spectra of dried samples were recorded on a Bruker infrared
spectrometer (IFS 66 V), equipped with a liquid-nitrogen-cooled
detector. Before measuring the samples, background spectra were
obtained by using freshly cleaned bare silicon wafers (P/B (100), Si-
Mat Silicon wafers, Germany). UV−vis measurements of photo-
cleavable initiators were performed on a V-660 spectrometer (JASCO,
Japan), the measurement range being 400−170 nm and the scanning
speed 100 nm/min. Analytical SEC measurements were carried out on
a Viscotek SEC-system equipped with a pump, a degasser (SEC max
VE2001), a detector module (Viscotek 302 TDA), a UV detector
(Viscotek 2500, λ = 254 nm), and two columns (PLGel Mix-B, PLGel
Mix-C), using chloroform as eluent (flow rate 1.0 mL/min). The
molecular weights of samples were calibrated by universal calibration
with polystyrene as standards in the range of Mp 1480−4 340 000.
SI-ATRP of Lauryl Methacrylate. All polymerization reactions
were carried out in a Schlenk line under a nitrogen atmosphere. In a
typical experiment, dNbpy (1.9 mmol, 0.8 g) and CuBr2/(dNbpy)2
complexes (0.05 mmol) in 550 μL of DMF were first dissolved in
lauryl methacrylate (50 mL, 0.17 mol), after which the solution
underwent three freeze−pump−thaw circles (10 min each) to remove
the dissolved oxygen. Then the mixture was transferred to another
flask containing CuBr (0.95 mmol, 0.14 g). After stirring for 30 min at
room temperature, 75 μL of EBiB/monomer solution (0.1% v/v) was
added to the clear dark solution, and the mixture was immediately
transferred to freshly prepared, initiator-modified samples. Polymer-
ization was carried out at 110 °C for various lengths of time (see
Results and Discussion), and subsequently the reaction was quenched
by precipitation in methanol, the wafer being subsequently sonicated
in toluene to remove physisorbed polymers.
SI-ATRP of Styrene. In a typical procedure, dNbpy (0.008 mmol,
3.2 mg) and Me4cyclam (0.08 mmol, 10.4 mg) were first dissolved in a
10 mL mixture of styrene, DMF, and anisole (v:v:v = 3:1:1), after
which the solution underwent three freeze−pump−thaw circles (10
min each) to remove the dissolved oxygen, before being transferred to
another flask containing CuBr (0.04 mmol, 6.0 mg) and CuBr2 (0.004
mmol, 1.0 mg). After stirring for 30 min at 50 °C, 1.5 μL of EBiB was
added to yield a clear, light-green solution, into which the freshly
prepared initiator-modified samples were immediately transferred.
Polymerization was carried out at 90 °C for various lengths of time
(see Results and Discussion). The polymerization solution was
quenched by precipitation with methanol, the wafer being
subsequently sonicated in toluene to remove physisorbed polymers.
Cleavage and Collection of Polymer Brushes from Silicon
Surfaces. In order to remove impurities from the brush-covered
surfaces, all substrates were extracted in chloroform for 24−48 h
before cleaving off the polymer brushes. In a typical procedure, a PLM-
modified silicon wafer was immersed in 3 mL of nonane in a glass dish,
which was illuminated under a UV lamp (254 nm) for various lengths
of time (see Results and Discussion). Each sample was fixed at a
distance of 2.5 cm away from the UV light source, which had a
measured intensity of 1.5 mW/cm2. PLM cleaved from silicon surfaces
was gathered by removing the nonane under reduced pressure at room
temperature. The cleavage of PS from substrates involved the same
cleaving procedures as above, except that THF was used as the solvent
instead of nonane.
RESULTS AND DISCUSSION
■
Synthesis of Photocleavable SI-ATRP Initiator BNMP.
The photocleavable moiety 2-nitrobenzyl has previously been
shown to be a relatively stable linker, yet can be efficiently
cleaved under UV-irradiation.22,23 This moiety was employed in
the design of the photocleavable ATRP initiator BMNP, which
was synthesized by reacting 2,5-dioxopyrrolidine-1-yl 2-bromo-
2-methylpropanoate (DBMP)24 with 3-amino-3-(2-
nitrophenyl)propanoic acid (ANPA) in DMSO at room
temperature in the presence of triethylamine (NEt3) (Scheme
1
1). The successful synthesis of BMNP was confirmed by H
NMR, 13C NMR, and elemental analysis. The strong absorption
peak in the UV−vis spectrum of BMNP in THF (see
Supporting Information for details) at 256 nm indicates the
possibility of cleavage of BMNP at this wavelength.
Tethering BMNP onto Silicon Substrate. In order to
immobilize BMNP onto the surface of a silicon wafer, surface
modification was required to provide covalently bound amine
groups. Aminosilanes have been widely used for modifying
silicon-based surfaces due to their bifunctional nature, and (3-
aminopropyl)triethoxysilane (APTES) is the most frequently
used of these. However, the polymerization of APTES in
solution is difficult to avoid.25,26 Thus, vapor-phase deposition
of the APTES was carried out in order to generate a
monolayer.27 By controlling the deposition time, it was possible
to obtain an ATPES coating with a thickness of 0.5 0.1 nm
and with a static water contact angle of 59 2°. The formation
of a monolayer of APTES and the absence of APTES polymers
on the silicon substrate were confirmed by the smooth surface
morphology observed with atomic force microscopy (AFM)
(see Supporting Information for details).
BMNP was then tethered to the amine groups on the silicon
surface by amino-dehydroxylation. The successfully reaction of
BMNP with APTES was confirmed with ellipsometrythe
thickness of the organic film on the silicon substrate increased
to 0.8
0.1 nm. Furthermore, following the reaction with
BMNP, the multiple-transmission−reflection infrared (MTR-
IR)28 spectrum showed an additional absorption peak at 1663
cm−1, assigned to the amide bond (see Supporting Information
for details).
Growing Polymer Brushes from Silicon Surfaces. SI-
ATRP was carried out according to the procedure previously
reported by Bielecki et al.,29 with minor modifications. A kinetic
study with ellipsometry showed an approximately linear
increase in dry polymer film thickness with reaction time,
suggesting that the polymer chains were growing from the
surface in a controlled manner (see Supporting Information for
details). MTR-IR measurements after polymer growth showed
a strong absorption peak at 1730 cm−1 (top spectrum of Figure
Characterization. The chemical structure of BMNP was
1
determined using a H NMR Bruker Avance 300 spectrometer, the
signal of DMSO-d6 (1H 2.54 ppm) being used as internal standard for
the determination of chemical shift. The dry thicknesses of APTES and
the surfaces modified with polymer brushes were measured with a
variable-angle spectroscopic ellipsometer (VASE, M-2000F, LOT
Oriel GmbH, Darmstadt, Germany) at an incident angle of 70°, using
a three-layer model (software WVASE32, LOT Oriel GmbH,
Darmstadt, Germany), each sample being measured at three different
spots. Morphologies of APTES-modified silicon surfaces were
determined at room temperature in TappingMode using a Multimode
AFM with a NanoScope IIIa controller (Veeco, Santa Barbara, CA)
equipped with a silicon cantilever (Olympus, Japan), with resonant
frequency 300 kHz and a spring constant of 26.1 N/m. MTR-IR
C
dx.doi.org/10.1021/ma401951w | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
1), which was assigned to the ester groups of PLM. The surface
morphology of the silicon substrate coated with polymer
Figure 1. MTR-IR spectrum of a PLM-functionalized silicon substrate
as the polymer thickness (measured dry) is reduced by UV-induced
cleavage from the surface. Top spectrum represents the surface prior
to irradiation. Spectra are normalized to the C−H stretch at 2929
cm−1.
brushes was also characterized by AFM (see Supporting
Information for details).
Figure 2. Kinetics of cleavage by UV irradiation (A) from the PLM
modified substrates in dry state and (B) from the PLM-modified
substrates immersed in nonane and the PS-modified substrates under
THF. Dashed lines are intended as a guide to the eye. (a)−(d) indicate
linear regions of the kinetic plot, referred to in detail in the text.
Cleavage of Polymer Brushes from Silicon Substrates.
Before cleaving the polymer brushes from the substrates, the
silicon wafers were extracted in Soxhlet sets for 24−48 h with
chloroform, in order to remove surface-attached impurities,
such as adsorbed polymers generated by the free initiator. The
cleaving processes were carried out by exposing poly(lauryl
methacrylate) (PLM)-modified substrates, both dry and under
nonane, in which PLM dissolves, to UV irradiation with a
wavelength of 254 nm. The cleavage reaction is shown in
Scheme 2.
The progress of the cleavage reaction was monitored by
determining the change in thickness (measured dry) of the
polymer films on the substrates (Figure 2). Cleavage of PLM
was also monitored by MTR-IR. As the polymer brushes were
cleaved from the surface, new absorption peaks at 1709 and
1435 cm−1 appeared, which could be assigned to ketone and
nitroso groups, respectively,30 both of these being present in
the photolysis products of BMNP17 (Figure 1). The entire
process was also repeated, for comparison purposes, for the
synthesis of polystyrene (PS) brushes.
As can be seen from Figure 2, when dry, polymer-coated
substrates were directly exposed to UV light, it took over 10 h
to achieve a 95% reduction of total polymer film thickness. The
cleavage kinetics exhibited first-order behavior, as indicated by
the logarithmic decrease in film thickness with UV-illumination
time (Figure 2A). When PLM was removed with UV under
nonane, much faster cleavage was observed. A sharp decrease in
film thickness from 219.6 11.8 to 21.0 1.5 nm over the first
30 min was followed by a slower reduction, with the thickness
being reduced by more than 99% over 2 h. The cleavage
kinetics appeared to be composed of two distinct first-order
processes, which were connected by a turning point at a dry
polymer thickness of approximately 20 nm (Figure 2B). This
phenomenon could possibly be explained by the brush−
mushroom transition as the reaction proceeds; however, the
UV-exposure used to obtain polymer for SEC measurements
involved was restricted to short periods of time to avoid
decomposition of the PLM or PS. The length of the polymers
at the turning point in the reaction rate was, therefore, not
determined. In the initial stage at high coverage (Figure 2B(a),
slope k = 0.044), the relatively higher grafting density resulted
in a brush configuration, with strong repulsion forces between
the solvated, surface-tethered polymer chains leading to high
osmotic pressure and facilitating the photocleavage process by
the faster “ejection” of cleaved chains. The crowding in the
brush state presumably lowered the energy of the transition
state, accelerating the cleaving reaction by about 15-fold,
compared to the collapsed conformation of dry polymer chains
on substrates (Figure 2A, slope k = 0.003). As the grafting
density decreased, due to the cleavage reaction, the solvated,
surface-tethered polymers formed the mushroom-like con-
formation (Figure 2 B(b), slope k = 0.008), and the rate of
cleavage under solvated conditions was thus reduced to only
about 2 times greater than that of the polymer chains under dry
conditions. The osmotically accelerated detachment of polymer
brushes from substrates has been previously observed by
Paripovic et al.,31 who investigated the influence of SI-ATRP
initiator chemical structures on the detachment of anchored
polymer brushes, when incubated in deionized water and cell-
culture medium. It was proposed that those reactions involved
Scheme 2. Schematic of the Cleavage Reaction of Polymer Brushes Tethered on Substrates under UV Illumination
D
dx.doi.org/10.1021/ma401951w | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
the breakage of the Si−O bond that connects the brush and
substrate32 due to steric-crowding-induced tension along the
polymer backbone. This mechanism is consistent with two
further reports showing that noncovalent interfacial interactions
can result in the breakage of a covalent bond.33,34
Aromatic groups strongly absorb UV-light and may therefore
block its penetration through the polymer film, preventing the
cleavage process. To examine this possibility, poly(styrene)
(PS) was generated on the surface and then removed following
the same cleaving procedure as for PLM, in THF rather than in
nonane. The transition in cleavage rate observed for PLM was
also observed for PS. The cleavage kinetics exhibited a slower
rate compared to PLM, which may have been due to the partial
absorption of UV-light by PS chains (Figure 2B(c) slope k =
0.023, Figure 2B(d) slope k = 0.005). The thickness of the PS
coating decreased from 57 2.7 to 6.9 0.5 nm within 30 min
and was reduced by 95% within 2 h (Figure 2B).
UV light with a wavelength of 256 nm leads to the scission of
polymer chains and therefore to a reduction in polymer film
thickness, even in the absence of a photocleavable linker,
potentially confounding the measurements of cleavage kinetics.
To explore this possibility, the nonoptically sensitive initiator 2-
bromo-2-methylpropanoic acid (BMPA) was immobilized on
silicon substrates by reacting DBMP directly with amino groups
on APTES-modified substrates. PLM brushes were then
generated and subsequently illuminated with UV light, as
described for the photocleavable initiator. Figure 3 shows the
Furthermore, in order to quantitatively determine the
reduction of PLM molecular weight due to UV-induced
polymer chain decomposition, PLM (Mw = 4.69 × 105 Da)
was dissolved in nonane (1.0 mg/mL) and illuminated under
the same conditions as those of polymer brushes on the
modified substrate. As can be seen from Figure 3, the total
molecular weight of free PLM was reduced by about 40% after
2 h UV illumination. However, during the first 20 min, the
reduction of molecular weight could not be significantly
distinguished by SEC. Therefore, in order to avoid the
influence of UV-induced decomposition, the polymer brushes
attached to modified substrates were exposed to UV for a
maximum of 10 min (see dotted line in Figure 3), 2 min being
most frequently used for substrates with a polymer film
thickness above 100 nm.
SEC Measurement of Cleaved Polymers. Polymer-
izations to produce PLM were carried out for three different
lengths of time, each being carried out three times, in order to
determine the total experimental error (A, B, and C series).
Cleaved chains from samples with approximately 210 nm
(PLM-A series), 120 nm (PLM-B series), or 60 nm (PLM-C
series) dry thickness of tethered polymers were collected by
removing the solvent under reduced pressure at room
temperature and then redissolving in 0.2 mL of chloroform
prior to characterization by SEC. For a planar substrate with an
area of 20 cm2 and a reduction in dry polymer film thickness of
50 nm, the concentration of polymer in 0.2 mL of chloroform
was approximately 0.5 mg/mL. As PS yielded a strong signal in
the UV-detector in SEC, this could be detected at a lower
concentration than PLM (Figure 4).
Figure 4. Typical SEC traces of PLM and PS.
Figure 3. (A) Kinetics of UV-decomposition of polymer brushes that
had been grown from the nonphotocleavable BMPA, (B) UV-induced
molecular weight reduction of free PLM in nonane solution, and (C)
UV-cleavage of polymer brushes initiated by the photocleavable
initiator, BNMP. The vertical line indicates the maximum UV-
exposure time after which polymers were harvested from the surface
prior to SEC analysis in order to minimize polymer decomposition.
Detailed information about the molecular weight of the
polymer brushes is presented in Table 1.
The surface coverage Γ (mg/m2), grafting density Σ (chain/
nm2), and mean distance between polymer chains D (nm) were
calculated. The surface coverage Γ (mg/m2) can be calculated
from the dry thickness of polymer brushes h (nm) by means of
eq 1:13
reduction in dry film thickness as measured by ellipsometry
against illumination time. The rate of reduction in thickness for
the BMPA-initiated polymer brushes was far lower than that of
the polymer brushes initiated by BNMP. For example, the
thickness of BMPA-initiated polymer brushes decreased from
201.8 to 195.7 nm in the first 20 min, while during the same
time period the thickness of BNMP-initiated brushes reduced
from 219.6 to 31.8 nm. Thus, for similar thicknesses, UV-light-
induced polymer-chain decomposition contributes only ap-
proximately 4% to total thickness reduction.
Γ = ρ × h
(1)
where ρ (1.05 g/cm3) and ρ (0.897 g/cm3)35 are the densities
of polystyrene and poly(lauryl methacrylate), respectively. The
grafting density Σ (chains/nm2) can be determined from eq 2:
Σ = ΓN × 10−21/Mn = (6.023Γ × 100)/Mn
(2)
A
where NA is Avogadro’s number and Mn (g/mol) is the
number-average molecular weight of the grafted polymer. The
E
dx.doi.org/10.1021/ma401951w | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
a
Table 1. SEC Data with Calculated Grafting Density and Mean Chain−Chain Spacing of Polymers Cleaved from Surface
polymer brushes
dry thickness (nm)
Mn × 105 (Da)
Mw × 105 (Da)
PDI
grafting density (chains/nm2)
mean chain−chain spacing (nm)
PLM-A-1
PLM-A-2
PLM-A-3
PLM-B-1
PLM-B-2
PLM-B-3
PLM-C-1
PLM-C-2
PLM-C-3
PS-A
217.1 3.5
215.0 2.6
220.3 5.2
120.5 6.3
125.4 4.4
118.7 6.4
57.8 4.9
55.1 3.4
60.1 4.7
41.5 1.2
47.1 1.9
6.3
7.6
9.0
4.6
4.2
4.5
2.1
1.9
2.6
0.5
0.7
11.9
15.9
18.0
7.8
1.9
2.1
2.0
1.7
1.6
1.7
1.6
1.5
1.6
1.6
1.9
0.19
0.15
0.13
0.14
0.17
0.14
0.15
0.16
0.13
0.50
0.44
2.5
2.8
3.0
2.9
2.6
2.9
2.8
2.7
3.0
1.5
1.6
6.7
7.6
3.4
2.9
4.2
0.8
PS-B
1.3
a
Thickness results are averages of three measurements at different locations on the sample.
distance between grafting sites was obtained with eq 3,
assuming that the packing is hexagonal.
polymerization. Finally, the much lower grafting density for
PLM than for PS could be attributed to the longer side chains
of PLM, which result in significant steric interactions that act to
separate neighboring chains.
2
D =
∑ × 3
(3)
ASSOCIATED CONTENT
* Supporting Information
■
The results of the calculation of grafting density and average
grafting site distance are shown in Table 1. From Table 1, it can
be seen that the grafting density did not change significantly as
the thickness of the polymer brushes increased from
approximately 60 to 210 nm. This indicates that the number
of growing chains remained constant and negligible chain
termination occurred during the polymerization process, which
is consistent with the approximately linear polymerization
kinetics (see Supporting Information for details). The PLM
brushes are separated by an average distance of about 2.8 nm,
which is much larger than PS chain average distance 1.6 nm
(Table 1). This difference could be attributed to the longer side
chains of PLM compared to PS, which lead to greater steric
repulsion.
S
Detailed information about the AFM images of APTES- and
PLM-modified silicon surfaces, growing kinetics of polymer
brushes, MTR-IR spectra of silane- and initiator-modified
surfaces, and UV−vis spectra of initiator in THF. This material
is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Fax +41 44 633 10 27; Tel +41 44 632 58 50; e-mail
nspencer@ethz.ch (N.D.S).
■
Notes
The authors declare no competing financial interest.
The molecular weight of PLM was also estimated from the
ellipsometric thickness, assuming a swelling ratio of 10 to 15 in
nonane.28 This yields for PLM-A-1, for example, an average
height of 2173−3260 nm for the solvated polymer brush.
Taking a repeat distance of 0.295 nm per monomer and
assuming, for the purposes of establishing a lower limit of Mw,
that the brush is completely stretched, this yields an average
molecular weight of (1.9−2.8) × 106, which is consistent with
the values of Mw measured by SEC and slightly higher than the
values of Mn (Table 1).
ACKNOWLEDGMENTS
■
The authors express their gratitude to Dr. Thomas Schweizer
(ETH Zurich, Switzerland) for his help with the SEC
measurements. Funding from the ETH Research Commission
is gratefully acknowledged.
REFERENCES
■
(1) Gennes, D. P. G. Macromolecules 1980, 13, 1069−1075.
(2) Barbey, R.; Lavanant, L.; Paripovic, D.; Schuwer, N.; Sugnaux, C.;
̈
Tugulu, S.; Klok, H. A. Chem. Rev. 2009, 109, 5437−5527.
(3) Huang, X.; Wirth, M. J. Anal. Chem. 1997, 69, 4577−4580.
(4) Baum, M.; Brittain, W. J. Macromolecules 2002, 35, 610−615.
CONCLUSION
■
It could be shown that the cleavage of polymer brushes from
planar silicon surfaces by means of UV-irradiation of a
photocleavable SI-ATRP initiator is a convenient method for
harvesting grafted polymer chains for molecular weight and
polydispersity analysis. This approach allowed 99% of the
polymer brushes on the silicon surface to be cleaved off within
2 h. The cleavage kinetics of polymer brushes exhibited linear
first-order behavior, composed of two distinct regions that may
correspond to a transition in the polymer conformation from
“brush” to “mushroom” regimes. Cleaved polymers were
collected by removing the solvent under vacuum and analyzing
by SEC. The polymer chain grafting density as well as average
chain distance remained almost constant, within experimental
error, during the polymerization process, which indicates that
the number of growing chains remained constant during
(5) 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, 1424−1431.
(6) Chen, J. K.; Hsieh, C. Y.; Huang, C.-F.; Li, P. M.; Kuo, S. W.;
Chang, F. C. Macromolecules 2008, 41, 8729−8736.
(7) Li, L.; Chen, S.; Zheng, J.; Ratner, B. D.; Jiang, S. J. Phys. Chem. B
2005, 109, 2934−2941.
(8) Petrie, T. A.; Raynor, J. E.; Reyes, C. D.; Burns, K. L.; Collard, D.
M.; García, A. J. Biomaterials 2008, 29, 2849−2857.
(9) Kobayashi, M.; Terayama, Y.; Hosaka, N.; Kaido, M.; Suzuki, A.;
Yamada, N.; Torikai, N.; Ishihara, K.; Takahara, A. Soft Matter 2007, 3,
740−746.
̀
(10) Pasetto, P.; Blas, H.; Audouin, F.; Boissiere, C.; Sanchez, C.;
Save, M.; Charleux, B. Macromolecules 2009, 42, 5983−5995.
(11) Kruk, M.; Dufour, B.; Celer, E. B.; Kowalewski, T.; Jaroniec, M.;
Matyjaszewski, K. Macromolecules 2008, 41, 8584−8591.
F
dx.doi.org/10.1021/ma401951w | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(12) Xiang, P.; Petrie, K.; Kontopoulou, M.; Ye, Z.; Subramanian, R.
Polym. Chem. 2013, 4, 1381−1395.
(13) Koylu, D.; Carter, K. R. Macromolecules 2009, 42, 8655−8660.
(14) Feng, W.; Brash, J.; Zhu, S. Biomaterials 2006, 27, 847−855.
(15) Turgman, C. S.; Genzer, J. J. Am. Chem. Soc. 2011, 133, 17567−
17569.
(16) Yu, H.; Li, J.; Wu, D.; Qiu, Z.; Zhang, Y. Chem. Soc. Rev. 2010,
39, 464−473.
(17) Dorman
77.
́
, G.; Prestwich, G. D. Trends Biotechnol. 2000, 18, 64−
(18) Parker, L. L.; Brueggemeier, S. B.; Rhee, W. J.; Wu, D.; Kent, S.
B. H.; Kron, S. J.; Palecek, S. P. Analyst 2006, 131, 1097−1104.
(19) Yan, F.; Chen, L.; Tang, Q.; Wang, R. Bioconjugate Chem. 2004,
15, 1030−1036.
(20) Wirkner, M.; Alonso, J. M.; Maus, V.; Salierno, M.; Lee, T. T.;
García, A. J.; Campo, D. A. Adv. Mater. 2011, 23, 3907−3910.
(21) Brown, B. B.; Wagner, D. S.; Geysen, H. M. Mol. Diversity 1995,
1, 4−12.
(22) Li, Z.; Bai, X.; Ruparel, H.; Kim, S.; Turro, N. J.; Ju, J. Proc. Natl.
Acad. Sci. U. S. A. 2003, 100, 414−419.
(23) Kim, S.; Shi, S.; Bonome, T.; Ulz, M. E.; Edwards, J. R.; Fodstad,
H.; Russo, J. J.; Ju, J. Anal. Biochem. 2003, 316, 251−258.
(24) Lecolley, F. O.; Tao, L.; Mantovani, G.; Durkin, I.; Lautru, S.;
Haddleton, D. M. Chem. Commun. 2004, 2026−2027.
(25) Howarter, J. A.; Youngblood, J. P. Langmuir 2006, 22, 11142−
11147.
(26) Zhu, M.; Lerum, M. Z.; Chen, W. Langmuir 2011, 28, 416−423.
(27) Jonsson, U.; Olofsson, G.; Malmqvist, M.; Ronnberg, I. Thin
̈
̈
Solid Films 1985, 124, 117−123.
(28) Liu, H. B.; Venkataraman, N. V.; Spencer, N. D.; Textor, M.;
Xiao, S. J. ChemPhysChem 2008, 9, 1979−1981.
(29) Bielecki, R. M.; Benetti, E. M.; Kumar, D.; Spencer, N. D. Tribol
Lett. 2012, 45, 477−487.
(30) Bradley, G. M.; Strauss, H. L. J. Phys. Chem. 1975, 79, 1953−
1957.
(31) Paripovic, D.; Klok, H. A. Macromol. Chem. Phys. 2011, 212,
950−958.
(32) Onclin, S.; Ravoo, B. J.; Reinhoudt, D. N. Angew. Chem., Int. Ed.
2005, 44, 6282−6304.
(33) Sheiko, S. S.; Sun, F. C.; Randall, A.; Shirvanyants, D.;
Rubinstein, M.; Lee, H. I.; Matyjaszewski, K. Nature 2006, 440, 191−
194.
(34) Deng, Y.; Zhu, X. Y. J. Am. Chem. Soc. 2007, 129, 7557−7561.
(35) Zhao, J.; Xiao, C.; Xu, N.; Ma, X. Polym. Bull. 2012, 69, 733−
746.
G
dx.doi.org/10.1021/ma401951w | Macromolecules XXXX, XXX, XXX−XXX